Blocking dPerk in the intestine suppresses neurodegeneration in a Drosophila model of Parkinson's disease

Abstract

Parkinson's disease (PD) is characterised by selective death of dopaminergic (DA) neurons in the midbrain and motor function impairment. Gastrointestinal issues often precede motor deficits in PD, indicating that the gut-brain axis is involved in the pathogenesis of this disease. The features of PD include both mitochondrial dysfunction and activation of the unfolded protein response (UPR) in the endoplasmic reticulum (ER). PINK1 is a mitochondrial kinase involved in the recycling of defective mitochondria, and PINK1 mutations cause early-onset PD. Like PD patients, pink1 mutant Drosophila show degeneration of DA neurons and intestinal dysfunction. These mutant flies also lack vital proteins due to sustained activation of the kinase R-like endoplasmic reticulum kinase (dPerk), a kinase that induces the UPR. Here, we investigated the role of dPerk in intestinal dysfunction. We showed that intestinal expression of dPerk impairs mitochondrial function, induces cell death, and decreases lifespan. We found that suppressing dPerk in the intestine of pink1-mutant flies rescues intestinal cell death and is neuroprotective. We conclude that in a fly model of PD, blocking gut-brain transmission of UPR-mediated toxicity, is neuroprotective.

Fig. 1. dPerk expression in the gut compromises mitochondrial function and induces cell death.

a An illustration of the imaged region of the adult fly gut. b, c Expression of dPerk leads to a reduction in the Δψm in the fly intestine. b Representative confocal images showing whole-mounted intestine loaded with TMRM. The intensity of TMRM is visualised using a five-tone heatmap. c Quantitative data (asterisks, one-way ANOVA with Dunnett’s multiple comparison test, comparison of control flies with dPerk and dPerk-KD flies, n indicates the number of guts). The data are shown as the average of 8 ROIs per gut. d, e dPerk expression in the intestine causes caspase (Dcp-1) activation. d Representative confocal images showing Dcp-1-positive cells in the midgut. Both dPerk constructs (dPerk and dPerk-KD) were tagged with HA. Green, Dcp-1; red, HA; blue, cell nuclei. e Quantification of Dcp-1-positive cells in the midgut (asterisks, Kruskal‒Wallis with Dunn’s multiple comparison test, comparison of control flies with dPerk and dPerk-KD flies, n indicates the number of guts). Data from 10-day-old male flies were analysed. Genotypes: control: w; NP3048Gal4/ + ; + , dPerk: w; NP3048Gal4/UASdPerk; +, dPerk-KD: w; NP3048Gal4/UASdPerk-K671R; +.

Fig. 2. dPerk expression increases cell proliferation in the gut.

a, b Cell proliferation was measured by counting the number of PH3-positive cells. a Representative confocal images showing PH3-positive cells in the midgut. Red, PH3; blue, cell nuclei. b Quantification of PH3-positive cells in the midgut (asterisks, Kruskal‒Wallis with Dunn’s multiple comparison test, comparison of control flies with dPerk and dPerk-KD flies, n indicates the number of guts). The number of PH3-positive cells in the anterior and posterior midgut (from the proventriculus region to the posterior midgut-hindgut junction) was determined. Data from 10-day-old male flies were analysed. Genotypes: control: w; NP3048Gal4/ + ; + , dPerk: w; NP3048Gal4/UASdPerk; +, dPerk-KD: w; NP3048Gal4/UASdPerk-K671R; +.

Fig. 3. Expression of dPerk in the gut compromises intestinal function.

a Representative images of fly faeces over a 20-hour feeding period. Data from 10-day-old male flies fed on food supplemented with blue dye to aid the visualisation of the faeces were analysed. b Expression of dPerk reduces the number of faecal deposits. The data are shown as the number of deposits per plate (means ± sem, asterisks, Kruskal‒Wallis with Dunn’s multiple comparison test, comparison of control flies with dPerk and dPerk-KD flies). c Expression of dPerk alters the shape of faecal deposits (asterisks, Kruskal‒Wallis with Dunn’s multiple comparison test, comparison of control flies with dPerk and dPerk-KD flies, n indicates the number of faecal deposits). Genotypes: control: w; NP3048Gal4/ + ; + , dPerk: w; NP3048Gal4/UASdPerk; +, dPerk-KD: w; NP3048Gal4/UASdPerk-K671R; +.

Fig. 4. Intestinal dysfunction driven by dPerk decreases lifespan.

a Lifespan analysis showed that the intestinal expression of dPerk in flies reduces lifespan (asterisks, log-rank test) but b does not affect the development of the flies (asterisks Chi-square two-tailed, 95% confidence intervals). Temporal control expression of dPerk using Gal80^ts in the flies c reduces lifespan (asterisks, log-rank test) and in 13 day old males, d decreases Δψm (asterisks, Mann-Whitney two-tailed test), e increase caspase (Dcp-1) activation (asterisks, two-tailed Student’s t-test) and f increases the number of PH3-positive cells (asterisks, Mann-Whitney two-tailed test) in the fly intestine. Genotypes: a, b control: w; NP3048Gal4/ + ; + , dPerk: w; NP3048Gal4/UASdPerk; +, c–f control: w; NP3048Gal4/ + ; Gal80^ts/+, dPerk: w; NP3048Gal4/UASdPerk; Gal80^ts/+.

Fig. 5. Suppression of dPerk in the midgut of pink1 mutants rescues mitochondrial dysfunction.

a An illustration of the imaged region in the posterior midgut of the adult fly. c pink1 mutant flies show loss of the Δψm in the midgut (mean ± sem; asterisks, unpaired t test). d–e Recovery of the Δψm upon intestinal RNAi-mediated suppression of dPerk, with two independent RNAi lines, in pink1 mutant flies. b Representative confocal images showing the midguts of adult flies loaded with TMRM. The intensity of TMRM is visualised using a five-tone heatmap. Quantitative data (mean ± sem; asterisks, unpaired t test) with RNAi#1 d and RNAi#2 e. The flies were 10 days old in b and 20 days old in c–e. The data are shown as the average of 8 ROIs per gut. Genotypes: control: w; NP3048Gal4/ + ; + , Pink1: pink1B9; NP3048Gal4/ + ; + , Pink1, dPerk-RNAi#1: pink1B9; NP3048Gal4/UAS dPerk RNAi#1; +, Pink1, dPerk-RNAi#2: pink1B9; NP3048Gal4/UAS dPerk RNAi#2; +.

Fig. 6. Suppression of dPerk in the midguts of pink1 mutants rescue intestinal dysfunction.

a, b Caspase activation in the intestine of pink1 mutant flies was blocked by dPerk RNAi. a Representative confocal images showing Dcp-1-positive cells in the posterior midgut region. b Quantitation of Dcp-1-positive cells in the posterior midgut showing that silencing dPerk reduced apoptosis of enterocytes of pink1 mutant flies (asterisks, one-way ANOVA with Dunnett’s multiple comparison test, all genotypes are compared to the control, n indicates the number of guts). c The number of PH3-positive cells in pink1 mutants was not affected by RNAi-mediated suppression of dPerk (asterisks, Kruskal‒Wallis with Dunn’s multiple comparison test, all genotypes are compared to the control, n indicates the number of guts). Data from 30-day-old male flies were analysed. Genotypes: control: w; NP3048Gal4/ + ; + , Pink1: pink1B9; NP3048Gal4/ + ; + , Pink1, dPerk-RNAi#1: pink1B9; NP3048Gal4/UAS dPerk RNAi#1; +, Pink1, dPerk-RNAi#2: pink1B9; NP3048Gal4/UAS dPerk RNAi#2; +.

Fig. 7. Neuronal defects in pink1 mutants are rescued by suppression of dPerk in the midgut.

a Intestinal suppression of dPerk rescued activity defects in pink1 mutants (asterisks, Kruskal‒Wallis with Dunn’s multiple comparison test, comparison of pink1 mutant flies with control flies and dPerk-RNAi-treated pink1 mutant flies, n indicates the number of flies). b, c Intestinal suppression of dPerk prevented loss of the PPL1 cluster of DA neurons in pink1 mutants. b Top: schematic representation of the PPL1 anatomical location. Bottom: representative confocal images showing loss (and rescue) of PPL1 neurons in pink1 mutant and dPerk-RNAi-treated pink1 mutant flies. c RNAi-mediated suppression of dPerk in the intestine prevented loss of the PPL1 cluster of DA neurons in pink1 mutants (asterisks, Kruskal‒Wallis with Dunn’s multiple comparison test, comparison of pink1 mutant flies with control flies and dPerk-RNAi-treated pink1 mutant flies, n indicates the number of brains). d RNAi-mediated suppression of dPerk using a gut-specific driver reduced the mRNA levels of dPerk in the intestine (left, means ± sem; asterisks, unpaired t test, n indicates the number of biological replicates) but not in the heads (right) of pink1 mutants (means ± sem; NS, unpaired t test, n indicates the number of biological replicates). Data from 30-day-old male flies were analysed. Genotypes: a–c control: w;NP3084Gal4/ + ; + , Pink1: pink1B9; NP3048Gal4/ + ; + , Pink1, dPerk-RNAi#1: pink1B9; NP3048Gal4/UAS dPerk RNAi#1; +, Pink1, dPerk-RNAi#2: pink1B9; NP3048Gal4/UAS dPerk RNAi#2; +.

Fig. 8. Upregulation of relish in pink1 mutants are reversed by suppression of dPerk in the midgut.

a A representative image showing NP3084Gal4 drives GFP expression throughout the adult fly midgut. b NP3084Gal4 does not drive expression in the PPL1 cluster region of the adult fly brain. We used either the elav-Gal4 (top) or the NP3084Gal4 (bottom) to assay GFP fluorescence in adult fly brains in the PPL1 cluster of DA neurons. c RNAi-mediated suppression of dPerk in the intestine, using the mex1Gal4 driver prevented loss of the PPL1 cluster of DA neurons in pink1 mutants (asterisks, Mann-Whitney two-tailed test, n indicates the number of brains). d, e Analysis of the effect of the RNAi-mediated suppression of dPerk on the transcript levels of Relish in the intestine of pink1 mutants (asterisks, one-way ANOVA with Dunnett’s multiple comparison test, comparison of pink1 mutant with control and each dPerk-RNAi line. Genotypes: a w; NP3048Gal4/UASGFP; + , b top: w; elavGal4/UASGFP and bottom: w; NP3048Gal4/UASGFP c Pink1: pink1B9; mex1Gal4/+; +, Pink1, dPerk-RNAi#1: pink1B9; mex1Gal4/UAS dPerk RNAi#1 d control: w; NP3048Gal4/ + ; + , Pink1: pink1B9; NP3048Gal4/ + ; +, Pink1, dPerk-RNAi#1: pink1B9; NP3048Gal4/UAS dPerk RNAi#1; +, Pink1, dPerk-RNAi#2: pink1B9; NP3048Gal4/UAS dPerk RNAi#2; +.