DJ-1 and SOD1 Act Independently in the Protection against Anoxia in Drosophila melanogaster

Abstract

Redox homeostasis is a vital process the maintenance of which is assured by the presence of numerous antioxidant small molecules and enzymes and the alteration of which is involved in many pathologies, including several neurodegenerative disorders. Among the different enzymes involved in the antioxidant response, SOD1 and DJ-1 have both been associated with the pathogenesis of amyotrophic lateral sclerosis and Parkinson's disease, suggesting a possible interplay in their mechanism of action. Copper deficiency in the SOD1-active site has been proposed as a central determinant in SOD1-related neurodegeneration. SOD1 maturation mainly relies on the presence of the protein copper chaperone for SOD1 (CCS), but a CCS-independent alternative pathway also exists and functions under anaerobic conditions. To explore the possible involvement of DJ-1 in such a pathway in vivo, we exposed Drosophila melanogaster to anoxia and evaluated the effect of DJ-1 on fly survival and SOD1 levels, in the presence or absence of CCS. Loss of DJ-1 negatively affects the fly response to the anoxic treatment, but our data indicate that the protective activity of DJ-1 is independent of SOD1 in Drosophila, indicating that the two proteins may act in different pathways.

Keywords: DJ-1; Drosophila melanogaster; Parkinson’s disease; SOD1; amyotrophic lateral sclerosis.

Figure 1

Ccs^n29E null flies show milder phenotypes than Sod1ⁿ¹ mutants: (A) Survival analysis. Ccs^n29E mutants lived significantly less than w¹¹¹⁸ controls but longer than Sod1ⁿ¹ flies (Mantel–Cox log-rank test: w¹¹¹⁸ vs. Ccs^n29E (****), w¹¹¹⁸ vs. Sod1ⁿ¹ (****), Ccs^n29E vs. Sod1ⁿ¹ (####), p < 0.0001 for all comparisons. N: 162 Ccs^n29E, 288 Sod1ⁿ¹, and 260 w¹¹¹⁸); (B) climbing activity (mean ± SEM). Analysis of variance and post-hoc tests indicate significant locomotor impairments in Ccs^n29E and Sod1ⁿ¹ flies as compared with w¹¹¹⁸ controls, with Sod1ⁿ¹ mutants showing the strongest phenotype (one-way ANOVA F_{2, 516} = 123.7 p < 0.0001; Tukey’s multiple comparisons test: w¹¹¹⁸ vs. Ccs^n29E (****) and w¹¹¹⁸ vs. Sod1ⁿ¹ (****), p < 0.001; Ccs^n29E vs. Sod1ⁿ¹ (**), p = 0.0075. N: 178 Ccs^n29E, 164 Sod1ⁿ¹, and 177 w¹¹¹⁸); (C) survival analysis under mild oxidative stress conditions (1 mM paraquat). Ccs^n29E flies were significantly more and less sensitive as compared with w¹¹¹⁸ and Sod1ⁿ¹ flies, respectively (Mantel–Cox log-rank test: w¹¹¹⁸ vs. Ccs^n29E (****), w¹¹¹⁸ vs. Sod1ⁿ¹ (****), Ccs^n29E vs. Sod1ⁿ¹ (####), p < 0.0001 for all comparisons. N: 172 Ccs^n29E, 56 Sod1ⁿ¹, and 99 w¹¹¹⁸); (D) representative TEM images of the mitochondrial morphology of Ccs^n29E, Sod1ⁿ¹ mutants, and w¹¹¹⁸ controls. Asterisks in the pictures indicate representative mitochondria; (E) representative Western blot; (F) relative quantification of Sod1 levels (mean ± SEM) in Ccs^n29E, Sod1^n1,, and w¹¹¹⁸ flies; in (F), Sod1 levels are reported as the Sod1/Actin ratio, with Actin signal used as a loading control. Ccs^n29E and Sod1ⁿ¹ flies showed a significantly lower Sod1 amount with respect to controls (one-way ANOVA: F_{2, 6} = 11.51, p = 0.0088; Tukey’s multiple comparisons test: w¹¹¹⁸ vs. Ccs^n29E (*) and w¹¹¹⁸ vs. Sod1ⁿ¹ (*) p < 0.05; Ccs^n29E vs. Sod1ⁿ¹ (ns, non-significant) p = 0.97. N = 3).

Figure 2

dj-1β protects against oxygen deprivation without affecting Sod1 expression: (A) Percentage of survivors (mean ± SEM) in dj-1β^Δ93, Ccs^n29E, and w¹¹¹⁸ control flies, after 3, 4, 5, and 6 h of anoxic treatment. dj-1β^Δ93 and Ccs^n29E flies showed significantly higher mortality as compared with w¹¹¹⁸ controls after 4 and 5 h of anoxia. Anoxia induced similar effects in both dj-1β^Δ93 and Ccs^n29E flies, at all time points (two-way ANOVA (time of treatment X genotype) F_{6, 1578} = 1.28, p = 0.263; time of treatment effect: F_{3, 1578} = 222.4, p < 0.0001; genotype effect: F_{2, 1578} = 13.83, p < 0.0001; in the graph ****, ***, **, *, and ns indicate p < 0.0001, < 0.001, < 0.01, < 0.05, and non-significant, in Tukey’s multiple comparisons post-hoc tests, see text for details); (B) representative Western blot and (C) relative quantification of Sod1 protein levels (mean ± SEM) under basal conditions, in dj-1β KO and dj-1β-overexpressing (daGal4 > UAS-dj-1β) flies as compared with controls (w¹¹¹⁸, daGal4/+ and UAS-dj-1β/+); in (C), Sod1 levels are reported as the Sod1/Actin ratio, with Actin signal used as a loading control. No significant differences in Sod1 protein signals were detected among genotypes (one-way ANOVA F_{4, 23} = 0.90, p = 0.483, N > 4 per genotype); (D) representative Western blot and (E) relative quantification of Sod1 protein levels (mean ± SEM) in dj-1β KO and dj-1β-overexpressing (daGal4 > UAS-dj-1β) flies as compared with controls (w¹¹¹⁸, daGal4/+, and UAS-dj-1β/+), after 7 days of exposure to 1 mM paraquat; in (E), Sod1 protein amounts are reported as the Sod1/Actin ratio, with Actin signal used as a loading control. No significant differences in Sod1 protein amounts were detected among genotypes (one-way ANOVA F_{4, 17} = 2.29 p = 0.101; N > 3 per genotype).

Figure 3

dj-1β overexpression in a Ccs^n29E background does not rescue the effects induced by a Ccs depletion: (A) Percentage of survivors (mean ± SEM) in Ccs^n29E, daGal4 > UAS-dj-1β, and relative controls (Ccs^n29E, daGal4/+ and Ccs^n29E, UAS-dj-1β/+), after 3, 4, 5, and 6 h of anoxic treatment. Time of anoxic treatment significantly affected survival in all fly strains, with no differences among genotypes (two-way ANOVA (time of treatment X genotype) F_{6, 798} = 2.78, p = 0.011; time of treatment effect: F_{3, 798} = 144.1, p < 0.0001; genotype effect: F_{2, 798} = 0.76, p = 0.47, non-significant). For each time of treatment, 50–90 males per genotype were analyzed; (B) survival analysis under mild oxidative stress conditions (1 mM paraquat) in Ccs^n29E; daGal4 > UAS-dj-1β, and relative controls (Ccs^n29E, daGal4/+ and Ccs^n29E, UAS-dj-1β/+). The Ccs^n29E; daGal4 > UAS-dj-1β survival profile was intermediate between the two controls, indicating the differences were not due to a dj-1β overexpression in a Ccs null background (Mantel–Cox log-rank test: Ccs^n29E; daGal4 > UAS-dj-1β vs. Ccs^n29E; daGal4/+: p = 0.8; Ccs^n29E; daGal4 > UAS-dj-1β vs. Ccs^n29E; UAS-dj-1β/+: p = <0.0001; Ccs^n29E; daGal4/+ vs. Ccs^n29E; UAS-dj-1β/+: p <0.0001); (C) Representative Western blot and (D) relative quantification of Sod1 levels (mean ± SEM) in Ccs^n29E; daGal4 > UAS-dj-1β and control flies; in (D), Sod1 levels are reported as the Sod1/Actin ratio, with Actin signal used as a loading control. Sod1 levels in Ccs^n29E; daGal4 > UAS-dj-1β flies were significantly lower from that of w¹¹¹⁸ controls but similar to those of Ccs^n29E null flies and Ccs^n29E; daGal4/+ or Ccs^n29E; UAS-dj-1β/+ controls (one-way ANOVA: F_{4, 10} = 46.1, p < 0.0001; Tukey’s multiple comparisons test: w¹¹¹⁸ vs. all other genotypes p < 0.0001 (****), Ccs^n29E, all other comparisons, p ≥ 0.4, ns. N = 3).

Figure 4

Proposed model describing the independent actions of Sod1 and dj-1β in the protection against anoxia. In Drosophila melanogaster, Ccs is responsible for copper loading into the Sod1 active site through the so-called Ccs-dependent Sod1 maturation pathway. The loss of Ccs increases the susceptibility of flies to oxidative stress. dj-1β does not participate in the Ccs-independent Sod1 maturation pathway, as in a Ccs null genetic background, dj-1β overexpression does not induce any protection. The loss of dj-1β increases the susceptibility to oxidative stress without affecting the Ccs-related Sod1 maturation pathway, as Sod1 levels in dj-1β knock-out flies are similar to controls. Moreover, dj-1β protection appears independent from Sod1 as the modulation of dj-1β expression does not affect Sod1 levels either under basal conditions or in the presence of oxidative stress (created with BioRender.com, accessed on 7 July 2022).