MitoPark transgenic mouse model recapitulates the gastrointestinal dysfunction and gut-microbiome changes of Parkinson's disease

Abstract

Gastrointestinal (GI) disturbances are one of the earliest symptoms affecting most patients with Parkinson's disease (PD). In many cases, these symptoms are observed years before motor impairments become apparent. Hence, the molecular and cellular underpinnings that contribute to this early GI dysfunction in PD have actively been explored using a relevant animal model. The MitoPark model is a chronic, progressive mouse model recapitulating several key pathophysiological aspects of PD. However, GI dysfunction and gut microbiome changes have not been categorized in this model. Herein, we show that decreased GI motility was one of the first non-motor symptoms to develop, evident as early as 8 weeks with significantly different transit times from 12 weeks onwards. These symptoms were observed well before motor symptoms developed, thereby paralleling PD progression in humans. At age 24 weeks, we observed increased colon transit time and reduced fecal water content, indicative of constipation. Intestinal inflammation was evidenced with increased expression of iNOS and TNFα in the small and large intestine. Specifically, iNOS was observed mainly in the enteric plexi, indicating enteric glial cell activation. A pronounced loss of tyrosine hydroxylase-positive neurons occurred at 24 weeks both in the mid-brain region as well as the gut, leading to a corresponding decrease in dopamine (DA) production. We also observed decreased DARPP-32 expression in the colon, validating the loss of DAergic neurons in the gut. However, the total number of enteric neurons did not significantly differ between the two groups. Metabolomic gas chromatography-mass spectrometry analysis of fecal samples showed increased sterol, glycerol, and tocopherol production in MitoPark mice compared to age-matched littermate controls at 20 weeks of age while 16 s microbiome sequencing showed a transient temporal increase in the genus Prevotella. Altogether, the data shed more light on the role of the gut dopaminergic system in maintaining intestinal health. Importantly, this model recapitulates the chronology and development of GI dysfunction along with other non-motor symptoms and can become an attractive translational animal model for pre-clinical assessment of the efficacy of new anti-Parkinsonian drugs that can alleviate GI dysfunction in PD.

Keywords: Gastrointestinal motility; Intestinal inflammation; Microbiome; Non-motor symptoms; Parkinson’s disease; Translational research.

Figure 1:. Age-dependent weight loss in MitoPark mice.

(A) Treatment paradigm. (B) Body weights measured over the duration of the study showing progressive weight loss in MP mice. (C) Food intake of both groups at age 24 weeks. Food intake was measured by weighing food pellets before and after a 1-hour feeding time. No difference in pellet weight (mg), representing food consumed in 1 hour, was observed between groups. (D) qRT-PCR analysis of NpY mRNA transcript, normalized to 18S rRNA expression. Data represented as the group mean ± SEM from n=6-7. For (A), data were analyzed by two-way ANOVA followed by Sidak’s post hoc test. For (B) and (C), data were analyzed by two-tailed t-test. Asterisks (*p<0.05, **p<0.01 and ***p<0.001) indicate significant differences between MP and LCs; n.s = no statistical significance; MitoP, MitoPark; LC, littermate control.

Figure 2:. Region-specific gastrointestinal motility in MitoPark mice.

(A) Whole gut transit time. From age 16 weeks onward, MP mice required significantly less time to expel carmine red that had been administered via intragastric gavage. (B) Gastric emptying. At age 24 weeks, MP and LC mice took similar times for ingested food to leave the stomach into the small intestine. (C) Colon transit time and (D) colonic motility rate. At age 24 weeks, MP mice took significantly more time to expel the glass bead inserted 2 cm into the distal colon than did age-matched LCs, thus displaying a constipated behavior. (E) Stool frequency. At age 24 weeks, MP mice expelled significantly fewer fecal pellets than LCs. (F) Fecal water content. MP mice progressively show lower water content in their fecal pellets, indicating constipation. (G) Colon length in centimeters. Representative photographs exhibiting smaller colon length in 24-week-old MP mice compared to LCs. Data represented as the group mean ± SEM from n=6-10. For (A), (F), and (G), data were analyzed by two-way ANOVA followed by Sidak’s post hoc test. For (B-E), data were analyzed by two-tailed t-test. Asterisks (*p<0.05, **p<0.01, and ***p<0.001) indicate significant differences between MP and LC. MitoP, MitoPark; LC, littermate control.

Figure 3:. Reduced dopamine levels in the colon of 24-week-old MitoPark mice.

(A) HPLC analysis of monoamine neurotransmitter levels (ng/mg tissue) in the colon. Colonic tissue levels of DA were lower in 24-week-old MP mice than in LCs. (B) DA receptor mRNA transcript levels. No significant group differences in D1R and D2R levels in the colon. Data represented as the group mean ± SEM from n=10 or n=6 and analyzed by two-tailed t-test. Asterisks (*p<0.05) indicate significant differences between MP and LC. MitoP, MitoPark; LC, littermate control.

Figure 4:. Mild ileitis in older MitoPark mice.

(A) Western blot of ileum samples showing a trend of higher iNOS and TNFα expression in MP mice compared to LCs. Blot intensities were normalized to β-actin and expressed as arbitrary units. (B) Representative 60× images of ileum sections stained with GFAP showing decreased enteric glial numbers in the villi. Scale bar = 11 μm. Data represented as the group mean ± SEM from n=10 or n=6 and analyzed by two-tailed t-test. MitoP, MitoPark; LC, littermate control.

Figure 5:. Colonic inflammation in 24-week-old MitoPark mice.

(A) Western blot of colon samples showing increased GFAP and TNFα and decreased DARPP-32 expression in 24-week-old MP mice compared to LCs. Expression of iNOS did not differ significantly between groups. (B) Representative 60× image showing increased expression of iNOS in the myenteric (LMMP) and submucosal plexus (SMP) in 24-week-old MP mice compared to LC. (C) Oligomeric protein detection. Dot blot analysis of oligomeric protein content in the colon of 24-week-old MPs and LCs showed no significant differences between groups. (D) Dot blot analysis of aggregated α-synuclein expression in the colon showed no significant difference between groups. Data represented as the group mean ± SEM from n=6 or 7 and analyzed by two-tailed t-test. Asterisks (*p<0.05 and ***p<0.001) indicate significant differences between MP and LC. MitoP, MitoPark; LC, littermate control.

Figure 6:. Decreased neuronal expression of tyrosine hydroxylase without neuron loss in MitoPark mice.

(A) Representative 60× image showing fewer tyrosine hydroxylase (TH)-positive neurons in the myenteric plexus of the colon in MP mice. Scale bar = 7 μm. (B) qRT-PCR analysis of TH mRNA transcript, normalized to 18S rRNA expression (n=6). (C) Representative 40× images of PGP9.5-positive enteric neurons in the colon of LC and MP mice. No significant differences were observed between groups. Scale bar = 35 μm. Data represented as the group mean ± SEM from n=6 or 7 and analyzed by two-tailed t-test. Asterisks (*p<0.05) indicate significant differences between MP and LC. MitoP, MitoPark; LC, littermate control.

Figure 7:. Temporal changes in fecal metabolome track disease progression in MitoPark mice.

(A) Heat map and (B) paneled plots of metabolites that have a significant temporal variation. (C) Heat-map and (D) paneled plots of metabolites that differ significantly between 20-week-old MP and LC groups (LC n=6; MP n= 3-6).

Figure 7:. Temporal changes in fecal metabolome track disease progression in MitoPark mice.

(A) Heat map and (B) paneled plots of metabolites that have a significant temporal variation. (C) Heat-map and (D) paneled plots of metabolites that differ significantly between 20-week-old MP and LC groups (LC n=6; MP n= 3-6).

Figure 8:. 16s metagenomics analysis of MP and LC fecal samples.

(A) α-diversity measured by Chaol. (B) β-diversity as shown by principal component analysis (PCA) plot showing no significant diversity in microbial communities between age-matched LC and MP groups. (C) Relative abundances of bacterial Phyla among groups over time. (D) Relative abundances of bacterial genera among groups over time. MP, MitoPark; LC, littermate control.