The kynurenine pathway modulates neurodegeneration in a Drosophila model of Huntington's disease

Abstract

Neuroactive metabolites of the kynurenine pathway (KP) of tryptophan degradation have been implicated in the pathophysiology of neurodegenerative disorders, including Huntington's disease (HD) [1]. A central hallmark of HD is neurodegeneration caused by a polyglutamine expansion in the huntingtin (htt) protein [2]. Here we exploit a transgenic Drosophila melanogaster model of HD to interrogate the therapeutic potential of KP manipulation. We observe that genetic and pharmacological inhibition of kynurenine 3-monooxygenase (KMO) increases levels of the neuroprotective metabolite kynurenic acid (KYNA) relative to the neurotoxic metabolite 3-hydroxykynurenine (3-HK) and ameliorates neurodegeneration. We also find that genetic inhibition of tryptophan 2,3-dioxygenase (TDO), the first and rate-limiting step in the pathway, leads to a similar neuroprotective shift toward KYNA synthesis. Importantly, we demonstrate that the feeding of KYNA and 3-HK to HD model flies directly modulates neurodegeneration, underscoring the causative nature of these metabolites. This study provides the first genetic evidence that inhibition of KMO and TDO activity protects against neurodegenerative disease in an animal model, indicating that strategies targeted at two key points within the KP may have therapeutic relevance in HD, and possibly other neurodegenerative disorders.

Figure 1. Schematic Representation of the Kynurenine Pathway in Drosophila melanogaster

(A) Kynurenine pathway metabolites are essential for formation of the ommochromes, biological pigments required for wild-type eye color in the fly. Unlike yeast and mammalian counterparts, the neurotoxic metabolite quinolinic acid (QUIN) is not present in Drosophila, making the fly a unique system to model the effects of 3-hydroxykynurenine (3-HK) and kynurenic acid (KYNA) on neurodegeneration. (B) Flies expressing Htt93Q exhibit increased 3-HK/KYNA ratios at day 1 and day 7 relative to UAS controls. (C and D) cn³/cn³ (C) and v^36f/v^36f (D) flies show a significant decrease in 3-HK/KYNA at day 1 versus +/+ genotypes. (E) elav-Gal4 RNAi knockdown of cn and v KK lines leads to a decrease in 3-HK/KYNA at day 1 versus +/+ genotypes. RNAi knockdown of cn and v was confirmed by qPCR analysis (Figure S2). Statistical comparisons by analysis of variance (ANOVA) with Newman-Keuls post hoc tests. n = 4–8 samples per genotype. (*p < 0.05; **p < 0.01; ***p < 0.001; ns = not significant). Data are shown as means ± standard errors of the mean (SEMs).

Figure 2. Inhibition of cn and v Ameliorates Neurodegeneration in the Fly

(A) Pseudopupil images from wild-type flies, HD flies, and HD flies carrying either cn³ or v^36f amorphic alleles at day 1. In wild-type and in cn and v flies not expressing Htt93Q, seven rhabdomeres are visible per ommatidium. (B and D) Quantification of mean rhabdomeres (± SEM) per ommatidium in Htt93Q, Htt93Q cn³/cn³, Htt93Q cn³/+, Htt93Q v^36f/v^36f, Htt93Q v^36f/+, Htt93Q cn RNAi, and Htt93Q v RNAi flies at day 1 and day 7 after eclosion. Statistical comparisons by ANOVA and post hoc tests versus Htt93Q flies. n = 11–29 flies per genotype (*p<0.05; **p < 0.01; ***p < 0.001). ns = not significant. (C and E) Percent rescue of neurodegeneration in Htt93Q cn³/cn³, Htt93Q cn³/+, Htt93Q v^36f/v^36f, Htt93Q v^36f/+, Htt93Q cn RNAi, and Htt93Q v RNAi flies at day 1 and day 7. (F) Rhabdomere neurodegeneration at day 7 is ameliorated by treatment with the KMO inhibitor UPF 648 (doses in mM) relative to PBS control. n = 10–13 flies per treatment (***p < 0.001, ANOVA and post hoc tests). (G) Treatment with 100 µM UPF 648 shifts KP pathway flux toward synthesis of KYNA (**p < 0.01, Student’s t test). Flies were tested at day 7. n = 5 samples per treatment. Data are shown as means ± standard errors of the mean (SEMs).

Figure 3. 3-HK and KYNA Directly Modulate Neurodegeneration in Htt93Q Flies

(A) Representative images of newly emerged (day 0) cn³ flies fed 3-HK, showing restoration of wild-type (WT) eye color. Doses of 0.4, 0.6, 0.8, 1.1, or 1.4 mg/ml 3-HK were fed to cn³ and control flies (0.6 and 1.1 mg/ml not shown). (B) 3-HK abrogates cn³ protection in day 0 Htt93Q flies (ns = not significant, p = 0.15). n = 8–12 flies per treatment. Comparisons were made between genotypes with the same treatment. (C) 3-HK feeding leads to significantly increased 3-HK/KYNA ratios in newly emerged Htt93Q and Htt93Q cn³ flies versus untreated controls. n = 3–6 samples per treatment. (D) KYNA feeding at 1 mg/ml (**p < 0.01) and 5 mg/ml (*p < 0.05) increases rhabdomere number at day 0. n = 12 flies per treatment. Statistical comparisons by ANOVA and post hoc tests for (B), (C), and (D) (*p < 0.05; **p < 0.01; ***p < 0.001). (E) Feeding of 5 mg/ml KYNA decreases the 3-HK/KYNA ratio in Htt93Q flies at day 0 (*p = 0.05, Student’s t test). n = 4–5 samples per condition. Data are shown as means ± standard errors of the mean (SEMs).

Figure 4. Htt93Q Flies Carrying the cd¹ Mutation Exhibit Reduced Neurodegeneration

(A) Day 1 cd¹/cd¹ flies in the control elav-GAL4 background exhibit a significant increase in 3-HK, whereas 3-HK levels in Htt93Q cd¹/cd¹ flies are unchanged (n = 4–5). (B) Rhabdomeres in Htt93Q cd¹/+ and Htt93Q cd¹/cd¹ flies are significantly increased versus Htt93Q flies at day 1 and day 7 after eclosion (n = 12–13). In cd control flies, seven rhabdomeres are visible per ommatidium (see Figure S4D). (C) Day 1 Htt93Q cd¹/cd¹ flies exhibit an ~3-fold increase in KYNA levels versus Htt93Q flies, whereas KYNA levels in the elav-GAL4 background are unchanged (n = 4–5 samples per genotype). (D) cd¹/cd¹ flies in the control elav-GAL4 background exhibit a significant increase in 3-HK/ KYNA ratios, whereas 3-HK/KYNA ratios in Htt93Q cd¹/cd¹ flies are significantly reduced compared to Htt93Q flies (n = 4–5). ANOVA with Newman-Keuls post hoc tests, except for (C), Student’s t test (*p < 0.05; **p < 0.01; ***p < 0.001). Data are shown as means ± standard errors of the mean (SEMs).