Prophylactic valproic acid treatment prevents schizophrenia-related behaviour in Disc1-L100P mutant mice

Abstract

Background: Schizophrenia is a neurodevelopmental disorder with onset early in adulthood. Disrupted-In-Schizophrenia-1 (DISC1) is a susceptibility gene for schizophrenia and other psychiatric disorders. Disc1-L100P mutant mice show behaviors relevant to schizophrenia at 12 weeks, but not at 8 weeks of age, and may be useful for investigating the onset of schizophrenia in early adulthood.

Methods: We investigated whether early valproic acid treatment would prevent behavioral, cellular and gene expression abnormalities in Disc1-L100P mutants.

Results: Valproic acid prevented hyperactivity and deficits in prepulse inhibition and latent inhibition in Disc1-L100P mice. Genome-wide transcription profiling identified Lcn2 (lipocalin2) transcripts as being elevated by the Disc1 mutation and corrected by valproate. Disc1-L100P mice also had increased glial cell numbers in the subventricular zone, which was normalized by valproate. Genetic deletion of Lcn2 normalized glial cell numbers and behavior in Disc1-L100P mutants.

Conclusions: Pharmacological treatments are a feasible way of preventing abnormal behaviour in a genetic model of schizophrenia. Lcn2 is a potential novel drug target for early intervention in schizophrenia.

Figure 1. Disc1-L100P mutant mice show schizophrenia-related behavior at 12 but not 8 weeks of age.

(a) 12 week-old Disc1-L100P mutant mice were hyperactive (n = 7−19 per group) as measured by total distance traveled during 30 minutes in the open field [F_1,22 = 4.5, p<0.05 effect of genotype; F_1,22 = 17.1, p<0.001 effect of age; F_1,22 = 9.4, p<0.01 gene × age interaction]. (b) 12 week-old Disc1-L100P mutant mice have impaired PPI (n = 7−10 per group) [F_1,49 = 58.8, p<0.001 effect of genotype; F_1,49 = 4.97, p≤0.05 age; F_2,98 = 150.2, p<0.001 prepulse intensity; F_1,49 = 9.3, p<0.01 gene × age interaction]. (c) 12 week-old Disc1-L100P mutant mice have reduced Acoustic Startle Response (ASR) (n = 7−10 per group) [F_1,49 = 19.9, p<0.001 genotype; F_1,49 = 8.6, p<0.01 gene × age interactions]. *p<0.05, **p<0.01, ***p<0.001 in comparison with WT mice within each age group. (d) 12 week-old Disc1-L100P mutant mice have disrupted LI (n = 6−8 per group) [F_1,47 = 19.1, p<0.001 genotype; F_1,47 = 15.4, p<0.001 age; F_1,47 = 49.1 p<0.001 effect of pre-exposure; F_1,47 = 4.5, p<0.05 gene × age × pre-exposure interactions]. The eight experimental groups did not differ in A periods (all p>0.05, overall mean A period = 6.8 sec). ***p<0.001 non-pre-exposed (NPE) in comparison with pre-exposed (PE) animals to the conditioned stimulus (CS) within each genotype and age group. See also Supplementary Table S1.

Figure 2. Chronic treatment with valproic acid between 10 and 12 weeks of age prevented the emergence of hyperactivity and PPI deficits in Disc1-L100P mutants and had a delayed effect on LI.

(a) Valproic acid prevented hyperactivity in Disc1-L100P mice, but had no effect on locomotion of WT mice (n = 8−15 per group) [F_1,41 = 8.4, p<0.001 genotype; F_1,41 = 12.5, p<0.001 drug treatment; F_1,41 = 5.6, p<0.05 gene × drug interaction. (b) Valproic acid prevented PPI deficits in Disc1-L100P mice. ANOVA showed a significant main effect of genotype [F_1,41 = 7.6, p<0.01 genotype; F_1,41 = 5.5, p<0.05 drug treatment; F_2,82 = 4.4, p<0.05 pre-pulse intensity; F_1,41 = 9.6, p<0.01 gene × drug interaction], #p<0.001 in comparison with vehicle-treated WT mice; *p<0.05, **p<0.01, ***p<0.001 in comparison with vehicle-treated Disc1-L100P. (c) Valproic acid did not prevent disrupted LI in Disc1-L100P mice [F_1,43 = 51.1, p<0.001 genotype; F_1,43 = 43.0 p<0.001 pre-exposure, and F_1,43 = 33.5, p<0.001 genotype × pre-exposure]. The eight experimental groups did not differ in A periods (all p>0.05, overall mean A period  = 7.2 sec). #p<0.001 non-pre-exposed (NPE) in comparison with pre-exposed (PE) vehicle-treated WT to the conditioned stimulus (CS); **p<0.01; ***p<0.001 NPE in comparison with PE animals within each genotype and drug treatment. (d-g) Valproic acid had lasting effects on behavior in Disc1-L100P mutants 3 weeks after the last dose. (d) Valproic acid-treated Disc1-L100P and WT mice had similar locomotion, whereas vehicle-treated Disc1-L100P mutants were hyperactive in comparison with vehicle-treated WT mice; *p<0.05 vs. vehicle-treated WT mice; *p<0.05 vs. vehicle-treated Disc1-L100P mice. (e) Valproic acid prevention of PPI deficits in Disc1-L100P mice persisted 3 weeks after drug treatment ended [F_1,40 = 5.7, p<0.05] as compared with vehicle-treated Disc1-L100P mutants (p<0.05 at 69 dB and 81 dB; p<0.01 at 73 dB). In contrast, vehicle-treated Disc1-L100P mice showed PPI deficits [F_1,40 = 5.54, p<0.05] in comparison with vehicle-treated WT animals at all three pre-pulses (p<0.05 at 69 and 73 dB, p<0.01 at 81 dB). #p<0.05, ##p<0.01 in comparison with vehicle-treated WT mice; *p<0.05; **p<0.01 in comparison with vehicle-treated Disc1-L100P mice. (f) Disc1-L100P mice treated with valproic acid developed LI three weeks after treatment had stopped. Valproic acid-treated Disc1-L100P mice showed LI (p<0.01), similar to vehicle- and valproic acid-treated WT mice (p<0.001 and p<0.01, respectively), but no LI in vehicle-treated Disc1-L100P mice (p>0.05). #p<0.001 in comparison with vehicle-treated WT mice; **p<0.01; ***p<0.001 NPE in comparison with PE animals within each genotype and drug treatment. ANOVA detected a significant effect of genotype [F_1,43 = 12.8, p<0.001], drug treatment [F_1,43 = 15.2 p<0.001], pre-exposure [F_1,43 = 47.9 p<0.001] and gene × drug × pre-exposure interactions [F_1,43 = 6.1, p<0.05]. There was no difference in A periods among the eight experimental groups (all p>0.05, overall mean A period = 8.1 sec).

Figure 3. Gene expression changes induced by the Disc1-L100P mutation and by treatment with valproic acid.

(a) Heatmap showing expression microarray results. Columns indicate individual genes, rows individual tuples (brain region, genotype, treatment). The color legend along the bottom indicates the scaled signal intensity, with blue indicating high abundance. The colored boxes at right indicate, from left to right: the brain region (red - striatum, blue - hippocampus, green - frontal cortex, yellow - brain stem), treatment (purple - valproic acid, white - vehicle), and genotype (orange - Disc1-L100P, white - WT). BS - brain stem, FC - frontal cortex, HIP – hippocampus, STR - striatum. d1v - valproate-treated Disc1-L100P mutants; d1c - vehicle-treated Disc1-L100P mutants; wtv - valproate-treated WT; wtc - vehicle-treated WT. (b) Validation of array results: Arc, Egr2, Dusp1, Purb, Slc40a1, Adar, Mrpl39, Slc6a12, Lcn2, Igf1, EiF4ebp2 and Cyr61 transcripts were analyzed by RT-qPCR. Coefficients refer to either the genotype-drug interaction term or the genotype term in ANOVA analyses. For microarrays, this analysis was performed using the fold-change magnitude in log 2-space, for RT-qPCR, RNE values (log 2-based fold changes in mRNA quantity, normalized to Gapdh and β-actin levels). We obtained an overall validation rate of 0.60 (P = 0.037). r, Pearson correlation factor. (c-e) Elevated Lcn2 (Lipocalin 2) expression in Disc1-L100P mutant mice normalized by valproic acid treatment. (c) Lcn2 mRNA levels assessed by qRT-PCR (n = 5−6 per group). RNE values are relative to GAPDH and β-actin. ANOVA detected a significant effect of genotype [F_1,15 = 18.4, p<0.001], drug [F_1,15 = 20.6, p<0.001] and gene × drug interaction [F_1,15 = 18.2, p<0.001] on Lcn2 expression. (d) Protein levels in representative Western blots are shown, probed with an antibodies against LCN2 and β-actin as a loading control in extracts of brain stem isolated from vehicle- and valproate-treated Disc1-L100P and WT mice (n = 6−7 per group). (e) Densitometric analysis to quantify the relative intensity of LCN2-immunoreactive bands relative to β-actin. ANOVA found a significant effect of genotype [F_1,20 = 31.4, p<0.001], drug [F_1,20 = 21.9, p<0.001] and their interaction [F_1,20 = 25.9, p<0.001] on LCN2 protein levels. *p<0.05 in comparison with vehicle-treated WT; #p<0.05 in comparison with vehicle-treated Disc1-L100P mutants. See also Supplementary Tables S2–6.

Figure 4. Increased GFAP+ cell proliferation in Disc1-L100P mutants normalized by valproic acid.

(a) Representative examples of brain sections from olfactory bulb (OB), rostral migratory stream (RMS), subventricular zone (SVZ), and subgranular zone (SGZ) of the hippocampus of vehicle-treated (top) or valproate-treated (bottom) WT and Disc1-L100P mice, immunostained for ki67 (dark brown). All images were taken at 10× magnification. (b) Quantitative analysis of the number of ki67⁺ nuclei in OB, RMS, SVZ and SGZ in all gene and drug conditions. OB: There was a significant effect of genotype [F_1,62 = 18.23; p<0.01]; drug [F_1,62 = 18.9; p<0.01] and gene × drug interactions [F_1,62 = 19,14; p<0.01]. RMS: There was an effect of genotype [F_1,62 = 17.38, p<0.01]; drug [F_1,62 = 28.47, p<0.001] and their interaction [F_1,62 = 15.71, p<0.05]. SVZ: There was a significant effect of genotype [F_1,62 = 28.54, p<0.01]; drug [F_1,62 = 16.28; p<0.05] and their interactions [F_1,62 = 16.54, p<0.01]. ANOVA did not detect an effect of genotype and/or drug treatment on ki67⁺ cells in SGZ (all p>0.05). The number of ki67⁺ cells were expressed as the number of cells per 1 mm² using Aperio Image Scope. *p<0.05; **p<0.01 in comparison with vehicle-treated WT mice; #p<0.01 in comparison with vehicle-treated Disc1-L100P mice; Veh – vehicle; Val – valproate; n = 4 sections per animal from 4–6 mice; (c) Increased GFAP⁺ cells in OB, RMS and SVZ but not in SGZ of vehicle-treated Disc1-L100P mice as compared to WT, assessed by GFAP immunostaining (brown). All images were taken at 20× magnification. (d) Quantitative analysis of the number of GFAP⁺ nuclei in OB, RMS, SVZ and SGZ. ANOVA detected a significant effect of genotype in OB [F_1,30 = 7.71; p<0.01], RMS [F_1,30 = 8.18, p<0.01], SVZ [F_1,30 = 8.81, p<0.05] but not in SGZ [F_1,30 = 0.39, p>0.05]. The number of GFAP⁺ cells were expressed as the number of cells per 1 mm² using Aperio Image Scope. *p<0.05; **p<0.01 in comparison with WT mice. n = 4 sections per animal from 4–6 mice. See also Supplementary Table S7.

Figure 5. LCN2 is co-expressed with GFAP in the brain and levels are associated with behavior in Disc1-L100P mice.

(a) Co-expression of LCN2 with GFAP in the mouse brain shown in confocal microscopic images of cortical slices stained with LCN2 and GFAP antibodies (20× magnification). (b) Correlation of LCN2 and GFAP levels in different brain areas, including olfactory bulbs, rostral migratory stream, subventricular zone (SVZ), subgranular zone of hippocampus and cortex (n = 21). Levels of GFAP⁺ and LCN2⁺ staining were expressed as the intensity per 1 um² using a computerized image analyzer (EZ-C1, gold v3.9, Nikon Corporation). (c-d) Genetic inactivation of Lcn2 corrected the increased number of GFAP⁺ cells in SVZ of Disc1-L100P mutant mice. (c). Confocal images of slices of the SVZ probed with LCN2 and GFAP in 4 genotypes: wild-type (WT); Disc1-L100P, Lcn2-KO and Disc1-L100P × Lcn2-KO (n = 4 slices per mouse; 3–5 mice per genotype; 20× magnification). (d). Quantitative analysis of LCN2 and GFAP expression in SVZ in mice of 4 genotypes. *p<0.05; **p<0.01 in comparison with WT; unpaired t-test. (e-g). Genetic inhibition of Lcn2 corrected behavior in Disc1-L100P mutant mice, including hyperactivity, pre-pulse inhibition (PPI) and latent inhibition (LI). (e) Locomotor activity in the open field in WT or Disc1-L100P mice carrying both Lcn2 alleles (WT) or missing both Lcn2 alleles (LCN2-KO). ANOVA with repeated measures found a main effect of genotype [F_3,43 = 11.4, p<0.001], testing interval [F_5,215 = 137.4, p<0.001] and genotype × testing interval interaction [F_15,215 = 3.2, p<0.001]. Disc1-L100P mutants were hyperactive at all tested intervals (p<0.001 during the first 15 minutes, p<0.01 at 15–20 minutes and p<0.05 at 20–30 minutes), whereas genetic ablation of Lcn2 rendered their activity the same as WT mice. n = 6−17 mice per genotype. *p<0.05; **p<0.01; ***p<0.001 in comparison with WT mice. (f) PPI deficit in Disc1-L100P mutants was normalized at all three pre-pulses by genetic ablation of Lcn2. ANOVA with repeated measures found a main effect of genotype [F_3,31 = 14.1, p<0.001], and pre-pulse intensities [F_2,62 = 6.6, p<0.05]. n = 6−11 mice per genotype. ***p<0.001 in comparison with WT mice (g) Deletion of Lcn2 in Disc1-L100P mutants also restored LI. ANOVA detected a significant effect of genotype [F_3,47 = 13.5; p<0.001], pre-exposure [F_1,47 = 62.4; p<0.001] and their interaction [F_3,47 = 9.9, p<0.001]. n = 6−8 per experimental group. ***p<0.001 non-pre-exposed (NPE) in comparison with pre-exposed (PE) animals to the conditioned stimulus (CS) within each genotype.

Figure 6. Summary of results.

Lcn2 levels correlate with GFAP in subventricular zone (SVZ) and these are in turn associated with abnormal PPI, an endophenotype for schizophrenia in Disc1-L100P mice. The 3D quadratic surface graph illustrates surfaces fitted by a smoothing technique to the average PPI, GFAP and Lcn2 intensity in SVZ data. The color spectrum (from green to brown) represents the GFAP and Lcn2 intensity data (from low to high, respectively). Pearson correlation coefficients are: r = −0.73; p<0.001 – for PPI and Lcn2; r = −0.56; p<0.05 – for PPI and GFAP and r = 0.77, p<0.001– for Lcn2 and GFAP (N = 16).