Improvement of neuropathology and transcriptional deficits in CAG 140 knock-in mice supports a beneficial effect of dietary curcumin in Huntington's disease

Abstract

Background: No disease modifying treatment currently exists for Huntington's disease (HD), a fatal neurodegenerative disorder characterized by the formation of amyloid-like aggregates of the mutated huntingtin protein. Curcumin is a naturally occurring polyphenolic compound with Congo red-like amyloid binding properties and the ability to cross the blood brain barrier. CAG140 mice, a knock-in (KI) mouse model of HD, display abnormal aggregates of mutant huntingtin and striatal transcriptional deficits, as well as early motor, cognitive and affective abnormalities, many months prior to exhibiting spontaneous gait deficits, decreased striatal volume, and neuronal loss. We have examined the ability of life-long dietary curcumin to improve the early pathological phenotype of CAG140 mice.

Results: KI mice fed a curcumin-containing diet since conception showed decreased huntingtin aggregates and increased striatal DARPP-32 and D1 receptor mRNAs, as well as an amelioration of rearing deficits. However, similar to other antioxidants, curcumin impaired rotarod behavior in both WT and KI mice and climbing in WT mice. These behavioral effects were also noted in WT C57Bl/6 J mice exposed to the same curcumin regime as adults. However, neither locomotor function, behavioral despair, muscle strength or food utilization were affected by curcumin in this latter study. The clinical significance of curcumin's impairment of motor performance in mice remains unclear because curcumin has an excellent blood chemistry and adverse event safety profile, even in the elderly and in patients with Alzheimer's disease.

Conclusion: Together with this clinical experience, the improvement in several transgene-dependent parameters by curcumin in our study supports a net beneficial effect of dietary curcumin in HD.

Figure 1

Body weight profiles of WT and KI mice fed curcumin or control chow. No difference was detected between curcumin-fed and control-fed male mice, and while an overall effect of genotype was detected in female mice, no effect of food, or interaction of food and genotype, or food and genotype and age were found in either gender, indicating, as expected, no deleterious effect of curcumin. Data shown are mean ± sem. WT male n = 13-15; WT female n = 15-19; KI male n = 7-10; KI female n = 7-9. Data analyzed using GLM ANOVA followed by post hoc Bonferroni t-tests.

Figure 2

Chow utilized by control-fed and curcumin-fed adult C57Bl/6 J mice. No difference was noted between the groups. Data are mean ± sem. N = 9-10 per group.

Figure 3

At nanomolar concentrations, curcumin slightly reduces aggregate size and at micromolar concentrations, curcumin increases aggregate size in PC12 cells inducibly expressing exon 1 of mutant htt [42]. Cells were induced with 0.1 μM tebufenozide and treated with curcumin (5 nM, 50 nM, 500 nM, 5 μM, 10 μM or 20 μM) or vehicle (DMSO) at the same time. Uninduced cells (treated with EtOH (vehicle) were used as a control for expression of the mutant protein) are not shown. At 48 h, cells treated with 5 nM curcumin show an 8% reduction in size of aggregates compared to vehicle-treated induced cells (no curcumin). At 48 h and at 72 h, 10 or 20 μM increase aggregate size markedly, possibly reflecting curcumin's effect on the UPS at micromolar concentrations [43]. Data shown are of the mean ± sem of n = 4 independent experiments.

Figure 4

Curcumin reduces aggregates in KI mice. Mice were fed curcumin from conception and were analyzed for huntingtin aggregates, at 4.5 m of age. The mean number of stained nuclei (data not shown), stained nuclei containing microaggregates or inclusions, and neuropil aggregates per 20 μm² were counted over the entire striatum of each of two sections per mouse (1.32 mm and 2.28 mm lateral of the midline [47], data from medial section shown). A) control-treated KI mouse, B) curcumin-treated KI mouse. Arrowheads indicate stained nuclei containing inclusions. Small line arrows indicate neuropil aggregates and arrows indicate stained nuclei with microaggregates. C) Quantification of neuropathological analysis. Curcumin reduces several forms of aggregated huntingtin in 4.5 m old KI mice, data from medial section shown. Although the reduction in number of inclusions is not significant, a strong trend towards reduction was observed (effect of treatment F(1,20) = 4.3, p = 0.052). Data are shown as mean ± sem and were analyzed using ANOVA followed by Fishers LSD post hoc tests. N = 6 per group. * p < 0.05, **p < 0.01, compared to control-treated KI. Arrows are as for A) and B). Scale bar = 20 μm for A) and B).

Figure 5

Correlation and linear regression analysis of levels of striatal DARPP-32 mRNA. Using a separate group of mice to those used in our curcumin preclinical trial (n = 23 in total), we correlated levels of DARPP-32 mRNA normalized using HPRT or Atp5b and Eif4a [53]. We found excellent correlation between the results obtained with HPRT and with Atp5b and Eif4a (r² = 0.862 and slope of 1.08, p < 0.0001).

Figure 6

Curcumin rescues rearing deficit in KI mice. Control KI mice rear less than their littermate WT controls and treatment with curcumin abolished this deficit. Data are mean ± sem of first 5 min in open field. ** p < 0.01 compared to control-fed WTs, ΔΔ p < 0.01 compared to control-fed KI group. Groups were composed of balanced mixed gender groups since there was no significant effect of gender (see text). Data were analyzed using ANOVA followed by Fishers LSD post hoc tests. WT n = 29-32, KI n = 16 per group.

Figure 7

Although at this age no rotorod deficits were detected in the KI mice relative to WT, curcumin lowered performance scores. At the end of the trial, when tested on the rotarod, both WT (left) and KI (right) mice treated with curcumin showed impaired performance. Data shown are mean ± sem. Groups were composed of balanced mixed gender groups since there was no significant effect of gender (see text). Data were analyzed using ANOVA followed by Fishers LSD post hoc tests. WT n = 28-30, KI n = 15-16 per group. *p < 0.05, ** p < 0.01 compared to genotype-matched controls on the same day.

Figure 8

Curcumin impairs rotarod performance in normal C57Bl/6 J mice, but does not affect forelimb muscle strength, or locomoter activity. Adult mice were fed from 2.7 m until 8 m of age with curcumin, to approximate the duration that CAG140 mice were fed (gestation + neonatal + adult). Curcumin impaired performance in the rotorod in a gender-independent manner (A). At 8 m age curcumin treatment of males by not females was associated with climbing deficits. Overall, females climbed less than males possibly obscuring treatment effects (B). Curcumin improved grip strength at 4.5 m age, but did not affect grip at other ages tested (C). D) Activity in the open field was normal. Data are shown as mean ± sem, n = 10 per group (n = 9 in curcumin-fed females). Data were analyzed using ANOVA followed by Fishers LSD post hoc tests. *p < 0.05, **p < 0.01 compared to control-fed gender-matched mice.