Glial swip-10 controls systemic mitochondrial function, oxidative stress, and neuronal viability via copper ion homeostasis

Abstract

Cuprous copper [Cu(I)] is an essential cofactor for enzymes that support many fundamental cellular functions including mitochondrial respiration and suppression of oxidative stress. Neurons are particularly reliant on mitochondrial production of ATP, with many neurodegenerative diseases, including Parkinson's disease, associated with diminished mitochondrial function. The gene MBLAC1 encodes a ribonuclease that targets pre-mRNA of replication-dependent histones, proteins recently found in yeast to reduce Cu(II) to Cu(I), and when mutated disrupt ATP production, elevates oxidative stress, and severely impacts cell growth. Whether this process supports neuronal and/or systemic physiology in higher eukaryotes is unknown. Previously, we identified swip-10, the putative Caenorhabditis elegans ortholog of MBLAC1, establishing a role for glial swip-10 in limiting dopamine (DA) neuron excitability and sustaining DA neuron viability. Here, we provide evidence from computational modeling that SWIP-10 protein structure mirrors that of MBLAC1 and locates a loss of function coding mutation at a site expected to disrupt histone RNA hydrolysis. Moreover, we find through genetic, biochemical, and pharmacological studies that deletion of swip-10 in worms negatively impacts systemic Cu(I) levels, leading to deficits in mitochondrial respiration and ATP production, increased oxidative stress, and neurodegeneration. These phenotypes can be offset in swip-10 mutants by the Cu(I) enhancing molecule elesclomol and through glial expression of wildtype swip-10. Together, these studies reveal a glial-expressed pathway that supports systemic mitochondrial function and neuronal health via regulation of Cu(I) homeostasis, a mechanism that may lend itself to therapeutic strategies to treat devastating neurodegenerative diseases.

Keywords: C. elegans; copper; glia; neurodegeneration; swip-10.

Fig. 1.

Structural comparison of modeled SWIP-10 metallo b-lactamase domain (MBD) with the MBD of human MBLAC1. (A) Structural representation of the MBDs of crystalized human MBLAC1 [PDB id: 4V0H (27)] and modeled SWIP-10 obtained from the AlphaFold protein structure database, code: AF-Q20700-F1 (36) (Top view of the active sites). The percentage of residues of SWIP-10 and MBLAC overlapped after structural superimposition of both domains [using STAMP (34)] indicate that they share a similar fold (90% of residues in SWIP-10 MBD overlapped with homologous residues in MBLAC1). Loops surrounding the active site in both enzymes are colored in blue (Loop 1), green (Loop 2), magenta (Loop 3), orange (Loop 4), and red (Loop 5). Loop color is maintained in SI Appendix, Supplemental Figures. (B) Close view of the active site of both domains before and after their structural superimposition. Side chains located within 10Å of the catalytic Asp residue (D428 in SWIP-10 and 195 in MBLAC1) are shown as sticks and colored according to the loop in which are located. The presence of two Fe ions in the active site of the solved structure of MBLAC1, instead of the preferred Zn, are attributed by the authors to the expression system used for expression (E. coli instead of human cells) (27). A cyan sphere designates the C-α atom of G427 in MBLAC1 and is the site of a G427D mutation found in the swip-10 mutant line vt33 (14). Residue numbering for MBLAC1 derives from the numbering used for the X-ray structure (27).

Fig. 2.

Systemic Cu(I) levels are sustained by expression of swip-10. (A and B) Representative images of CF4 labeled worms obtained via 60× confocal microscopy, using excitation laser at 560 nm. Insets are higher magnification representation of CF4 puncta in comparable regions of WT and swip-10 mutant. Scale bar is 50 microns. (C) Number of puncta were quantified using an automated analysis in FIJI (ImageJ) by setting a threshold held constant for every image where particle counts were then measured. Confocal images from 25 animals in each genotype were used as individuals for quantitation. Analysis performed from maximum projection confocal images. A total of 25 animals per condition were analyzed. Statistical analyses were performed using one-sided Student’s t test. Puncta are found mainly in the intestinal section of the animal. Whole worm images were captured and used for analysis. (D) CF4 fluorescence intensity measured with a fluorescent plate reader (FlexStation 3, Molecular Devices) comparing WT and swip-10. 400 animals were added in 100 μl to each well. Five technical replicates averaged for each biological replicate. Five biological replicates used for analysis. Statistical analyses were performed using one-sided Student’s t test. *P ≤ 0.05. ***P ≤ 0.001.

Fig. 3.

swip-10 mutation impairs mitochondrial activity. (A) Basal Oxygen consumption rate (OCR) recording from Oroboros Oxygraph 2 K. 400 animals used for each experiment. Five biological replicates used for analysis. (B) Quantification of ATP using an ATP determination kit. Data shown as average fold change of swip-10 levels relative to N2. A one-sided Student’s t test was used to determine significance comparing raw values (swip-10/N2). (C–F) OCR measured by Seahorse Respirometer, five biological replicates used for analysis. (C) Representative trace recording from Seahorse experiments. The X-axis displays each measurement period where probe is inserted into well and when the different drug compounds are added by the instrument. (D) Basal recordings from animals before the addition of any drug. (E) Maximal respiration induced by addition of FCCP (10 μM). (F) Nonmitochondrial respiration measured after the addition of the mitochondrial inhibitor sodium azide. (A–F) All experiments noted here were performed using whole, intact worms. (A, C–F) were analyzed using one-sided Student’s t test. *P ≤ 0.05.

Fig. 4.

Elevated oxidative stress in swip-10 mutants. (A and B) Representative images for DCFDA staining obtained by confocal microscopy. Images captured with a 20× lens using an excitation laser of 488 nm. Scale bar is 50 microns. (C) Quantification of fluorescence intensity of animals treated with 50 μM DCFDA. Intensity average normalized to total size of animal (mm²). Normalized intensity/area measurements were taken from 10 animals and averaged. Analysis was performed from the average of 5 independent measurements. Statistical analyses were performed using one sided Student’s t test. (D–F) Targeted glutathione measurements using HPLC. (D) Total levels of reduced glutathione (GSH). (E) Total levels of oxidized glutathione (GSSG). (D and E) Statistical analyses were performed using Student’s t test. F) Redox potential (Eh) measured as the ratio of GSSG/GSH. Wilcoxon rank sum test performed for analysis of Eh ratio. W = 20. (D–F) Each independent measurement was performed using approximately 500 animals. Eight biological replicates tested. *P ≤ 0.05

Fig. 5.

Role for Cu(I) in supporting DA neurons via homeostasis of oxidative stress and mitochondrial function. (A–C) Animals either untreated or grown on agar plates containing 10 µM CuCl₂ or 5 µM elesclomol (ES). (D–F) N2 animals grown on agar plates containing 10 µM BCS until ready for experimentation. (A) Basal OCR determined by the Oroboros Oxygraph respirometer as an average of steady state recordings over 10 min. 400 animals used for recording. Ordinary one-way ANOVA used for analysis. (B) Measurements of ROS levels of animals treated with DCFDA. Confocal microscopy images obtained were used to quantify relative fluorescence intensity normalized to size of each animal. Student’s t test used for statistical analysis. (C) Relative gene expression levels quantified via quantitative PCR. All genes normalized to the housekeeping gene actin. Two technical replicates used for each measurement. At least seven biological replicates performed for all experiments. The one-sample t test was used for statistical analysis. The dashed red line indicates the normalized value of 1 for fold change relative to wildtype (N2) animals. (D) Basal OCR recordings (Oroboros) as described previously. Ordinary one-way AVOVA used for statistical analysis. (E) Measurements of ROS levels of animals treated with DCFDA. Confocal microscopy images obtained were used to quantify relative fluorescence intensity normalized to size of each animal. Student’s t test was used for statistical analysis. (F and G) Quantification of morphology characteristics of DA neurons. Total population percentage of animals displaying one or more characteristic of degeneration. (G) Animals either untreated or grown on agar plates containing 10 µM CuCl2 or 5 µM elesclomol (ES). (F) Statistical analysis was performed by one-sided Student’s t test. (G) Statistical analysis was performed by ordinary one-way ANOVA. *P ≤ 0.05. **P ≤ 0.01. ****P ≤ 0.0001.

Fig. 6.

Glial expression of swip-10 dictates global Cu(I) homeostasis, oxidative stress levels, and mitochondrial function. (A) Glial rescue of attenuated Cu(I) levels throughout the entirety of the body in swip-10 mutants. 25 animals assessed per condition. Statistical analyses were performed using one-sided Student’s t test. (B) Basal OCR determined by the Oroboros Oxygraph respirometer as an average of steady state recordings over 10 min. 400 animals used for recording. Ordinary one-way ANOVA used for analysis. (C) Relative gene expression levels quantified via quantitative PCR. Red bars indicate changes in swip-10 mutants, blue bars are the “glial rescue” strain. Two technical replicates used for each measurement. A minimum of six biological replicates were tested for all experiments. Data were analyzed using the one-sample t test. The dashed red line indicates the normalized value of 1 for fold change relative to WT (N2) animals. (D) Measurements of ROS levels of animals treated with DCFDA. Confocal microscopy images obtained were used to quantify relative fluorescence intensity normalized to size of each animal. Student’s t test was used for statistical analysis. Glial rescue strain contains approximately 80 to 90% transgenic animals expressing swip-10 only under the pan-glial promoter (ptr-10). *P ≤ 0.05. **P ≤ 0.01. ****P ≤ 0.0001.

Fig. 7.

Metabolomic assessment of swip-10 mutants. Synchronized animals were grown up to the L4 stage prior to metabolite extraction and subsequent chromatography. Raw mass/charge(m/z) values were used for bioinformatic analyses to assess pathways altered between N2 and swip-10 animals. (A and B) PLS-DA and heatmap from HILIC positive column. (C and D) PLS-DA and heatmap from C18-negative column. (A–D) Statistical analyses revealed nonoverlapping metabolomic signatures between genotypes. (B and D) Rows represent individual biological replicate samples for each genotype. Columns represent altered metabolite changes. 500 healthy, intact animals were pooled for each sample for analysis.

Fig. 8.

Metabolic and mitochondrial networks altered by loss of swip-10. Mummichog 2.0 was used for pathway assessment of altered metabolites. Enrichment scores indicate abundance of known metabolites in a defined pathway. (A) Metabolites measured Hilic positive column (B) Metabolites measured from C18 negative column.