Increasing N-acetylaspartate in the Brain during Postnatal Myelination Does Not Cause the CNS Pathologies of Canavan Disease

Abstract

Canavan disease is caused by mutations in the gene encoding aspartoacylase (ASPA), a deacetylase that catabolizes N-acetylaspartate (NAA). The precise involvement of elevated NAA in the pathogenesis of Canavan disease is an ongoing debate. In the present study, we tested the effects of elevated NAA in the brain during postnatal development. Mice were administered high doses of the hydrophobic methyl ester of NAA (M-NAA) twice daily starting on day 7 after birth. This treatment increased NAA levels in the brain to those observed in the brains of Nur7 mice, an established model of Canavan disease. We evaluated various serological parameters, oxidative stress, inflammatory and neurodegeneration markers and the results showed that there were no pathological alterations in any measure with increased brain NAA levels. We examined oxidative stress markers, malondialdehyde content (indicator of lipid peroxidation), expression of NADPH oxidase and nuclear translocation of the stress-responsive transcription factor nuclear factor (erythroid-derived 2)-like 2 (NRF-2) in brain. We also examined additional pathological markers by immunohistochemistry and the expression of activated caspase-3 and interleukin-6 by Western blot. None of the markers were increased in the brains of M-NAA treated mice, and no vacuoles were observed in any brain region. These results show that ASPA expression prevents the pathologies associated with excessive NAA concentrations in the brain during postnatal myelination. We hypothesize that the pathogenesis of Canavan disease involves not only disrupted NAA metabolism, but also excessive NAA related signaling processes in oligodendrocytes that have not been fully determined and we discuss some of the potential mechanisms.

Keywords: Acss1; Acss2; NAA; NAAG; Nat8L; aspartoacylase; vacuoles; vacuolization.

Figure 1

Liquid chromatography-mass spectrometry (LC-MS/MS) analysis of N-acetylaspartate (NAA) in mouse brain after acute dosing with methyl ester of NAA (M-NAA). Typical LC-MS/MS scan results for the four groups (blue = control, red = 1 g/kg, green = 3 g/kg, black = 10 g/kg). Graph insert shows the peak areas for each group (means ± SD, n = 3 per group). Brain NAA levels in the 3 g/kg group were approximately 2.4 times those of controls which is similar to the levels found in aspartoacylase (ASPA) deficient mice (*p = 0.01, **p = 0.0004, Student’s t-test, two tailed, two sample equal variance).

Figure 2

Total antioxidants, reactive oxygen species (ROS) and MDA levels in brain; Values are expressed as means ± SD (n = six mice per group). No significant differences were observed in any measure among the three groups.

Figure 3

Western blot analyses of inflammation markers. Western blots were used to assess expression levels of several markers of inflammation and stress responses including interleukin-6, cleaved caspase-3 and NADPH oxidase. No significant differences were observed among the groups. We also evaluated expression of the cytoprotective transcription factor nuclear factor (erythroid-derived 2)-like 2 (NRF-2) in both cytoplasmic and nuclear fractions and no translocation of NRF-2 to the nucleus was observed. Values indicate means ± SD with n = 6 per group.

Figure 4

Histopathology. Representative hematoxylin staining in the dorsal thalamus and hippocampus of control (A,B), M-NAA treated (C,D) and Nur7 mice (E,F). No vacuoles or other histopathological changes were observed after 30 day treatment with either M-NAA (5 g/kg twice per day) or methanol (0.84 mg/kg twice per day; data not shown). Extensive vacuolization was observed in many brain regions of Nur7 mice, including the thalamus and hippocampus (E,F). Abbreviations: CA1 and CA3, CA1 and CA3 pyramidal cell layers of hippocampus; DG, dentate gyrus of hippocampus; Hb, habenula; V3, third ventricle. Micron bar = 200 μm.

Figure 5

Glial fibrillary acidic protein (GFAP) immunohistochemistry. Representative sections stained for GFAP in the hippocampus and globus pallidus (GP) of control (A,B) and treated mice. No changes in immunoreactivity were observed after 30 days of treatment with either M-NAA (C,D; 5 g/kg twice per day) or methanol (0.84 mg/kg twice per day; data not shown). GFAP immunoreactivity was substantially increased in Nur7 mice (E,F). Abbreviations: CA1, CA1 pyramidal cell layer of hippocampus; DG, dentate gyrus of hippocampus; GP, globus pallidus. Micron bar = 200 μm.

Figure 6

ASPA immunohistochemistry. Aspa immunoreactivity was very strong in oligodendrocytes throughout the brain in untreated wild type mice (A,B). Minor animal to animal staining variability was observed with antibodies to Aspa, but no consistent changes in immunoreactive intensity were observed after 30 days of treatment with either M-NAA (C,D) or methanol (data not shown) treatment. Aspa immunoreactivity was completely absent in Nur7 mice (E,F). Abbreviations: cc, corpus callosum; CP, caudate-putamen; Ctx, cortex; GP, globus pallidus; fim, fimbria; Hip, hippocampus; LV, lateral ventricle. Micron bar = 200 μm.

Figure 7

Acyl-CoA short chain synthetase family member 2 (Acss2) immunohistochemistry. Acss2 immunoreactivity was sparse in control mice (A), occurring almost exclusively in scattered cell nuclei throughout the brain. Immunoreactivity was slightly increased (arrows) in some cell nuclei in the M-NAA (B) and methanol groups (C). In contrast, Acss2 immunoreactivity was drastically reduced in the untreated Nur7 mouse brain (D). All images are from layers II–V of neocortex. Micron bar = 100 μm.

Figure 8

Immunoreactivity for claudin-11 (also known as oligodendrocyte specific protein) in the forebrain. Staining for claudin-11 was moderate to strong in fiber pathways in control mice (A,B), and there was no change in immunoreactivity with M-NAA treatment (C,D). In contrast, Claudin-11 immunoreactivity was highly elevated in all fiber pathways in untreated Nur7 mice (E,F). Major fiber pathways such as the fimbria of the hippocampus and internal capsule were reduced in thickness in Nur7 mice, suggesting axonal and oligodendrocyte loss. Oligodendrocyte processes were enlarged with swollen segments in Nur7 mice (H) as compared with controls (G). Abbreviations: cc, corpus callosum; CP, caudate/putamen; fim, fimbria; ic, internal capsule; GP, globus pallidus; st, stria terminalis. Micron bar = 200 μm in (A–F), 10 μm in (G,H).

Figure 9

Staining for claudin-11 (oligodendrocytes; A,B), GFAP (astrocytes; C,D) and parvalbumin (neurons; E,F) showed that all three cell types remained in heavily vacuolated areas such as the GP. Higher magnification images in (B,D,F) are from heavily vacuolated areas in the GP (extended depth of field). Micron bar = 200 μm (A,C,E), 50 μm (B,D,F).

Figure 10

Schematic depicting the effects of Nat8L and Aspa gene knockouts and associated pathophysiological changes. Single gene deletion effects for Nat8L/Asp-NAT (blue) and Aspa (red) are shown in (A), whereas changes associated with the double knockout mice (green) are shown in (B). Only the effects associated with the single Aspa gene knockout (red) result in brain vacuolization. Two major differences between the single ASPA knockout mice and the dual Nat8L+Aspa knockout are that in the double knockout mice acetyl-CoA is not depleted and trapped as NAA, and excess NAA signaling or other potentially toxic actions from NAA are not possible. Further, no NAAG signaling occurs in the dual knockout mice, whereas excess NAAG is present in the single Aspa deficient mice. Acss1 and Acss2 convert acetate, including acetate derived from Aspa-mediated deacetylation of NAA, into acetyl-CoA for energy derivation, lipid synthesis and protein acetylation reactions. In Aspa deficiency alone, acetyl-CoA is removed from the system, and cannot be reclaimed. This affects downstream utilization of acetate by Acss1 and Acss2. Acetyl-CoA is not depleted in the Nat8L+Aspa double knockout mice because no acetyl-CoA is used to synthesize NAA. Abbreviations: Asp, aspartate; Asp-NAT, aspartate N-acetyltransferase (Nat8L); ASPA, aspartoacylase; Acss1 and 2, acyl-CoA short chain synthetase 1 and 2; Glu, glutamate; GCPII, glutamate carboxypeptidase-2; NAAG, N-acetylaspartylglutamate; NS, NAAG synthetase.