Folic Acid Improves Parkin-Null Drosophila Phenotypes and Transiently Reduces Vulnerable Dopaminergic Neuron Mitochondrial Hydrogen Peroxide Levels and Glutathione Redox Equilibrium

Abstract

Loss-of-function parkin mutations cause oxidative stress and degeneration of dopaminergic neurons in the substantia nigra. Several consequences of parkin mutations have been described; to what degree they contribute to selective neurodegeneration remains unclear. Specific factors initiating excessive reactive oxygen species production, inefficient antioxidant capacity, or a combination are elusive. Identifying key oxidative stress contributors could inform targeted therapy. The absence of Drosophila parkin causes selective degeneration of a dopaminergic neuron cluster that is functionally homologous to the substantia nigra. By comparing observations in these to similar non-degenerating neurons, we may begin to understand mechanisms by which parkin loss of function causes selective degeneration. Using mitochondrially targeted redox-sensitive GFP2 fused with redox enzymes, we observed a sustained increased mitochondrial hydrogen peroxide levels in vulnerable dopaminergic neurons of parkin-null flies. Only transient increases in hydrogen peroxide were observed in similar but non-degenerating neurons. Glutathione redox equilibrium is preferentially dysregulated in vulnerable neuron mitochondria. To shed light on whether dysregulated glutathione redox equilibrium primarily contributes to oxidative stress, we supplemented food with folic acid, which can increase cysteine and glutathione levels. Folic acid improved survival, climbing, and transiently decreased hydrogen peroxide and glutathione redox equilibrium but did not mitigate whole-brain oxidative stress.

Keywords: Drosophila; antioxidants; dopaminergic neuron; folic acid; glutathione; hydrogen peroxide; mitochondria; oxidative stress; parkin; roGFP.

Figure 1

Parkin-null Drosophila have decreased climbing behavior and increased mortality and PPL1 mitochondrial hydrogen peroxide levels. (A) Individual fly climbing attempts and average height climbed were recorded for 20 min in the MB5 Multibeam Drosophila Activity Monitoring System. Each data point represents data from one control (park^+/+) or parkin-null (park^−/−) fly (n ≥ 21). A Welch’s t test and an unpaired t test were performed to determine effect of genotype on attempts, and height climbed, respectively. (B) For the survival study, flies were monitored every 2–3 days, and percent surviving were reported until all flies expired (n ≥ 27). Gehan–Breslow–Wilcoxon test was performed to determine the effect of genotype. (C) Brains were dissected from park^+/+ and park^−/− flies expressing mito-roGFP2-Orp1 on days 5 and 20 post-eclosion, and ratios of total volumes of oxidized to non-oxidized fluorophore emissions were calculated for one PPL1 region per brain. Each data point represents the ratio from one PPL1 region (n ≥ 13). An unpaired t test and a Welch’s t test were performed to determine the effect of genotype on hydrogen peroxide levels on day 5 and 20, respectively. Error bars represent standard error of the mean. p values are reported for each comparison in (A–C). Representative panels of PPL1 images on (D) day 5 and (E) day 20. Raw images are shown in the top row, where blue indicates tyrosine hydroxylase antibody labeling, and red and green indicate oxidized and non-oxidized roGFP2, respectively. Selected volume “isosurfaces” of oxidized (red, middle row) and non-oxidized (green, bottom row) mito-roGFP2-Orp1 above threshold are also shown. Scale bar, 10 µm.

Figure 2

Parkin-null Drosophila have increased PPL1 mitochondrial glutathione redox equilibrium. Brains were dissected from control (park^+/+) or parkin-null (park^−/−) flies expressing mito-roGFP2-Grx1 on days 5 and 20 post-eclosion, and ratios of total volumes of oxidized to non-oxidized fluorophore emissions were calculated for one PPL1 region per brain. Representative panels of days (A) 5 and (B) 20 showing raw PPL1 images (top row, where blue indicates tyrosine hydroxylase antibody labeling, and red and green indicate oxidized and non-oxidized roGFP2, respectively). Selected volume “isosurfaces” of oxidized (red, middle row) and non-oxidized (green, bottom row) mito-roGFP2-Grx1 above threshold are also shown. Scale bar, 10 µm. (C,D) Each data point represents the ratio from one PPL1 region (n ≥ 17). Unpaired t tests were performed to determine the effect of genotype on days 5 and 20, respectively. Error bars represent standard error of the mean, and p values are reported for each comparison.

Figure 3

Parkin-null Drosophila have transiently increased PPM3 mitochondrial hydrogen peroxide levels and control levels of glutathione redox equilibrium. Brains were dissected from control (park^+/+) or parkin-null (park^−/−) flies expressing mito-roGFP2-Orp1 or mito-roGFP2-Grx1 on days 5 and 20 post-eclosion, and ratios of total volumes of oxidized to non-oxidized fluorophore emissions were calculated for one PPM3 region per brain. Representative panels of PPM3 images with mito-roGFP2-Orp1 on days (A) 5 and (B) 20. Raw images are shown in the top row, where blue indicates tyrosine hydroxylase antibody labeling, and red and green indicate oxidized and non-oxidized roGFP2, respectively. Selected volume “isosurfaces” of oxidized (red, middle row) and non-oxidized (green, bottom row) mito-roGFP2-Orp1 above threshold are also shown. Mito-roGFP2-Grx1 images are not shown. Scale bar, 10 µm. (C,D) Each data point represents the ratio from one PPM3 region (n ≥ 12). Welch’s tests and an unpaired t tests were run to determine the effect of genotype for days 5 and day 20, respectively. Error bars represent standard error of the mean, and p values are reported for each comparison.

Figure 4

Folic acid administration increases parkin-null Drosophila climbing behavior and median survival. (A) Park^−/− and park^+/+ flies were raised on 50 µM folic acid and activity was recorded in the MB5 Multibeam Drosophila Activity Monitor for 20 min on day 10. Folic acid administration improved park^−/− but not park^+/+ climbing attempts and average height climbed. Each data point represents data from one fly (n ≥ 21). Unpaired t tests were run to determine the effect of folic acid on climbing attempts and height climbed. (B) For the survival study, flies maintained on folic acid-supplemented food were monitored every 2–3 days, and percent surviving were reported until all flies expired (n ≥ 47). Gehan–Breslow–Wilcoxon test was performed to determine the effect of folic acid. p values are reported for each comparison.

Figure 5

Folic acid administration decreases PPL1 mitochondrial hydrogen peroxide levels and glutathione redox equilibrium in parkin-null flies on day 5. Brains were dissected from folic acid-treated park^−/− flies expressing (A) mito-roGFP2-Orp1 or (B) mito-roGFP2-Grx1 on day 5 post-eclosion, and (C,D) ratios of total volumes of oxidized to non-oxidized fluorophore emissions were calculated for one PPL1 region per brain. (A) Representative raw images are shown in the top row, where blue indicates tyrosine hydroxylase antibody labeling, and red and green indicate oxidized and non-oxidized roGFP2, respectively. Selected volume “isosurfaces” of oxidized (red, middle row) and non-oxidized (green, bottom row) (A) mito-roGFP2-Orp1 and (B) mito-roGFP2-Grx1 above threshold are also shown. Scale bar, 10 µm. (C,D) Unpaired t tests were run to determine the effect of folic acid administration on hydrogen peroxide levels and glutathione redox equilibrium. Each data point represents the ratio from one PPL1 region (n ≥ 23). Error bars represent standard error of the mean, and p values are reported for each comparison.

Figure 6

Folic acid administration does not affect PPL1 mitochondrial hydrogen peroxide levels or glutathione redox equilibrium in parkin-null flies on day 20 post-eclosion. Brains were dissected from folic acid-treated park^−/− flies expressing (A) mito-roGFP2-Orp1 or (B) mito-roGFP2-Grx1 on day 20 post-eclosion, and (C, D) ratios of total volumes of oxidized to non-oxidized fluorophore emissions were calculated for one PPL1 region per brain. (A) Representative raw images are shown in the top row, where blue indicates tyrosine hydroxylase antibody labeling, and red and green indicate oxidized and non-oxidized roGFP2, respectively. Selected volume “isosurfaces” of oxidized (red, middle row) and non-oxidized (green, bottom row) (A) mito-roGFP2-Orp1 and (B) mito-roGFP2-Grx1 above threshold are also shown. Scale bar, 10 µm. (C,D) Each data point represents the ratio from one PPL1 region (n ≥ 23). A Welch’s t test and a Mann–Whitney test were run to determine the effect of folic acid on hydrogen peroxide levels and glutathione redox equilibrium, respectively. Error bars represent standard error of the mean, and p values are reported for each comparison.

Figure 7

Folic acid administration does not affect oxidative stress marker elevation in parkin-null fly heads. Heads of folic acid-treated park^+/+ and park^−/− flies were collected and frozen on days 5 (top row) and 20 (bottom row) post eclosion. LC-MS/MS was performed to detect reduced glutathione (GSH), cystine, and cysteine. The increased ratio of cystine to cysteine in parkin-null flies indicates oxidative stress, which is unaffected by folic acid administration. Each data point represents one tube of lysate or the average of up to three lysate tubes collected on the same day. Each tube contained seven to thirty-four heads. Effects of genotype and folic acid were determined using two-way repeated measures ANOVA followed by Šídák’s multiple comparison test (day 5) or two-way ANOVA followed by Tukey’s multiple comparisons test (day 20). Graph title asterisk indicates significant effect of genotype (p < 0.05). p values for post hoc comparisons are shown (n = 5 for day 5 and 7 for day 20).