Surprising behavioral and neurochemical enhancements in mice with combined mutations linked to Parkinson's disease

Abstract

Parkinson's disease (PD) is the second most common neurodegenerative disorder behind Alzheimer's disease. There are currently no therapies proven to halt or slow the progressive neuronal cell loss in PD. A better understanding of the molecular and cellular causes of PD is needed to develop disease-modifying therapies. PD is an age-dependent disease that causes the progressive death of dopamine-producing neurons in the brain. Loss of substantia nigra dopaminergic neurons results in locomotor symptoms such as slowness of movement, tremor, rigidity and postural instability. Abnormalities in other neurotransmitters, such as serotonin, may also be involved in both the motor and non-motor symptoms of PD. Most cases of PD are sporadic but many families show a Mendelian pattern of inherited Parkinsonism and causative mutations have been identified in genes such as Parkin, DJ-1, PINK1, alpha-synuclein and leucine rich repeat kinase 2 (LRRK2). Although the definitive causes of idiopathic PD remain uncertain, the activity of the antioxidant enzyme glutathione peroxidase 1 (Gpx1) is reduced in PD brains and has been shown to be a key determinant of vulnerability to dopaminergic neuron loss in PD animal models. Furthermore, Gpx1 activity decreases with age in human substantia nigra but not rodent substantia nigra. Therefore, we crossed mice deficient for both Parkin and DJ-1 with mice deficient for Gpx1 to test the hypothesis that loss-of-function mutations in Parkin and DJ-1 cause PD by increasing vulnerability to Gpx1 deficiency. Surprisingly, mice lacking Parkin, DJ-1 and Gpx1 have increased striatal dopamine levels in the absence of nigral cell loss compared to wild type, Gpx1(-/-), and Parkin(-/-)DJ-1(-/-) mutant mice. Additionally, Parkin(-/-)DJ-1(-/-) mice exhibit improved rotarod performance and have increased serotonin in the striatum and hippocampus. Stereological analysis indicated that the increased serotonin levels were not due to increased serotonergic projections. The results of our behavioral, neurochemical and immunohistochemical analyses reveal that PD-linked mutations in Parkin and DJ-1 cause dysregulation of neurotransmitter systems beyond the nigrostriatal dopaminergic circuit and that loss-of-function mutations in Parkin and DJ-1 lead to adaptive changes in dopamine and serotonin especially in the context of Gpx1 deficiency.

Keywords: DJ-1; Glutathione peroxidase; Knockout mice; Parkin; Parkinson's disease.

Figure 1

Parkin^−/−DJ-1^−/− and Parkin^−/−DJ-1^−/−Gpx1^−/− mice have normal substantia nigra cell numbers. Bars show the mean ± SEM bilateral number of tyrosine hydroxylase (TH)-positive neurons in the substantia nigra estimated by unbiased stereology. Separate cohorts of mice were analyzed at ages 6 (A), 12 (B) and 18 (C) months, n=4-6 mice per genotype at each age. The groups were not significantly different from each other (one-way ANOVA).

Figure 2

Striatal dopamine is increased in Parkin^−/−DJ-1^−/−Gpx1^−/− mice. Levels of striatal dopamine (DA) measured by HPLC with electrochemical detection. Separate cohorts of mice were analyzed at ages 6 (A), 12 (B) and 18 (C) months, n=6-11 mice per genotype at each age. Bars show the mean ± SEM of the level of dopamine measured from microdissected striatum. Asterisks indicate significant differences compared to wild-type mice at the same age (*p<0.05, t-test). Parkin^−/−DJ-1^−/−Gpx1^−/− mice have increased DA levels compared to wild type (**p<0.01, one-way ANOVA, Tukey's post-hoc) at age 18 months but not at 6 or 12 months.

Figure 3

Striatal serotonin is increased in Parkin^−/−DJ-1^−/− and Parkin^−/−DJ-1^−/−Gpx1^−/− mice. Levels of striatal serotonin (5-HT) measured by HPLC with electrochemical detection. Separate cohorts of mice were analyzed at ages 6 (A), 12 (B) and 18 (C) months, n=6-11 mice per genotype at each age. Bars show the mean ± SEM of the level of serotonin measured from microdissected striatum. *p<0.05, one-way ANOVA compared to wild-type mice at the same age.

Figure 4

Serotonin levels are increased in the hippocampus of 15- month Parkin^−/−DJ-1^−/− mice. (A) Levels of serotonin (5-HT) measured by HPLC with electrochemical detection. Bars show the mean ± SEM (n=10 per genotype). *p<0.05, one-way ANOVA compared to wild-type mice at the same age. (B) Length of serotonergic fibers in the hippocampus measured by unbiased stereology (n=4 mice per genotype).

Figure 5

Parkin^−/−DJ-1^−/− and Parkin^−/−DJ-1^−/−Gpx1^−/− mice have normal locomotor activity in a novel environment. Spontaneous locomotor activity was measured for separate cohorts of mice at age 6 (A), 12 (B) and 18 months (C), n=6-8 mice per genotype, 8-11 mice per genotype and 9-11 mice per genotype, respectively. For all ages, the mice acclimated to the novel environment of the testing chamber over time, but there are no significant differences between genotypes (one-way ANOVA). Symbols represent the mean ± SEM number of infrared beam breaks in each 5-minute period of the 2-hour test.

Figure 6

Parkin^−/−DJ-1^−/− and Parkin^−/−DJ-1^−/−Gpx1^−/− mice have improved rotarod performance. The latency to fall off an accelerating rotating rod was measured for separate cohorts of mice at age 6 (A), 12 (B) and 18 months (C), n=6-8 mice per genotype, 10-11 mice per genotype and 9-11 mice per genotype, respectively. Symbols represent the mean ± SEM time (seconds) before falling off the rod for each of 8 trials. While all genotypes learned the task over multiple trials, Parkin^−/−DJ-1^−/− and Parkin^−/−DJ- 1^−/−Gpx1^−/− mice showed a significant increase in the latency to fall compared to wild type at ages 12 and 18 months (**p<0.01, two-way ANOVA, Tukey's post-hoc for both Parkin^−/−DJ-1^−/− and Parkin^−/−DJ-1^−/−Gpx1^−/−), with genotype and trial as factors.

Figure 7

Rotarod behavior of single and double knockout mice. The rotarod performance of three separate cohorts of mice is shown in (A), (B) and (C). 2-way ANOVA with genotype and trial as factors showed a significant effect of trial as all genotypes learned the task and a significant increase in the latency to fall for double knockout mice compared to wild-type mice at all ages (p<0.0001). Asterisks indicate trials with significant differences compared to wild-type mice according to post-hoc analysis. DJ-1^−/− mice were significantly different from wild-type mice at age 6 months but not at age 16 months. Parkin^−/− mice were not significantly different from wild-type mice at any age.

Figure 8

Improved motor and non-motor skills in Parkin^−/−, DJ-1^−/− and Parkin^−/−DJ-1^−/− mice. A cohort of mice at age 8 months was fully trained to perform the rotarod test and then tested at fixed rotarod speeds of 5 (A), 10 (B), 15 (C) and 20 rotations per minute (RPM) (D). Bars represent the mean ± SEM latency to fall off the rotating rod. The single and double knock-out mice showed increased latency to fall compared to wild type (*p<0.0001, one-way ANOVA, Tukey's post-hoc). Parkin^−/−DJ-1^−/− mice also showed increased latency to fall compared to DJ-1^−/− at the lowest speed (A).

Figure 9

Parkin^−/−, DJ-1^−/− and Parkin^−/−DJ-1^−/− mice display less “distraction” behavior compared to wild type mice. Video of the fixed-speed rotarod behavior (Figure 8) was analyzed to measure the percent of time each mouse was not facing straight forward on the rotarod apparatus as a surrogate measure of “distraction” (A-C). Bars show mean ± SEM percent time on the rotarod not facing forward. Parkin^−/−DJ-1^−/− mice spent more time facing forward at 10 (A), 15 (B), and 20 (C) rotations per minute (RPM) compared to wild type (*p<0.05, **p<0.01, and ***p<0.001, one-way ANOVA, Tukey's post-hoc). (D) Mice were tested for the amount of time spent investigating novel odors of vanilla and urine from unfamiliar mice. Bars represent mean ± SEM time spent investigating the odor during the 3-minute trial. Parkin^−/−DJ-1^−/− and wild type mice showed comparable olfactory function (one-way ANOVA).