Mitochondrial DNA depletion by ethidium bromide decreases neuronal mitochondrial creatine kinase: Implications for striatal energy metabolism

Abstract

Mitochondrial DNA (mtDNA), the discrete genome which encodes subunits of the mitochondrial respiratory chain, is present at highly variable copy numbers across cell types. Though severe mtDNA depletion dramatically reduces mitochondrial function, the impact of tissue-specific mtDNA reduction remains debated. Previously, our lab identified reduced mtDNA quantity in the putamen of Parkinson's Disease (PD) patients who had developed L-DOPA Induced Dyskinesia (LID), compared to PD patients who had not developed LID and healthy subjects. Here, we present the consequences of mtDNA depletion by ethidium bromide (EtBr) treatment on the bioenergetic function of primary cultured neurons, astrocytes and neuron-enriched cocultures from rat striatum. We report that EtBr inhibition of mtDNA replication and transcription consistently reduces mitochondrial oxygen consumption, and that neurons are significantly more sensitive to EtBr than astrocytes. EtBr also increases glycolytic activity in astrocytes, whereas in neurons it reduces the expression of mitochondrial creatine kinase mRNA and levels of phosphocreatine. Further, we show that mitochondrial creatine kinase mRNA is similarly downregulated in dyskinetic PD patients, compared to both non-dyskinetic PD patients and healthy subjects. Our data support a hypothesis that reduced striatal mtDNA contributes to energetic dysregulation in the dyskinetic striatum by destabilizing the energy buffering system of the phosphocreatine/creatine shuttle.

Fig 1. The mitochondrial genome (mtDNA).

rRNA-encoding and protein-encoding genes are shown; tRNA-encoding genes are withheld for clarity. Teal arrows depict the direction of mtDNA replication from the heavy strand or light strand origins of replication (O_H, O_L); red arrows depict the direction of polycistronic transcription from the heavy strand or light strand promoters (HSP1/2, LSP). *—ND6 is the only protein-encoding gene on the light chain. Adapted from [29].

Fig 2. Exposure to EtBr dose- and time-dependently decreases mtDNA in rat striatal co-cultures and purified neuronal cultures, but not in astrocytes.

(A-C) Log2-fold change in nDNA-normalized mtDNA quantity in response to EtBr treatment, in (A) neuron-enriched cocultures (NECo), (B) pure neuronal cultures (Neu), and (C) astrocytes (Ast). N for each group is included in figure legends. *** = p < 0.001, ** = p < 0.01, relative to vehicle-treated controls following Dunn’s post-hoc test.

Fig 3. Exposure to EtBr decreases mtRNA in rat striatal co-cultures, purified neuronal cultures, and astrocytes.

(A-C) Log2-fold decrease in mtRNA expression after EtBr treatment assayed with RNASeq, in (A) NECos, (B), pure neuronal cultures, and (C) astrocytes. Each bar is a comparison of three control samples to three EtBr-treated samples, each pooled from two independently dissected culture experiments. MultiRankSeq analysis with false discovery rate–adjusted p-values taken from the DESeq comparison. mtRNAs were the most significantly changed RNA transcripts in neurons with a p-value of 0. (D-F) Log2-fold decrease in mtRNA expression after EtBr treatment assayed with qPCR. Error bars reflect delta-method propagated +/-SEM, with level of significance determined following Dunn’s post-hoc test. 2D[5] = 5ng/ml EtBr for 2 days, 4D[5] = 5ng/ml EtBr for 4 days, 2D[50] = 50ng/ml EtBr for 2 days, 4D[50] = 50ng/ml EtBr for 4 days. N for each group is included in figure legends. *** = p < 0.001, ** = p < 0.01, relative to controls. Transcripts are arranged according to their distance from HSP2 (see Fig 1A). The ND6 gene, the only mRNA on the light chain, is shown last.

Fig 4. EtBr treatment reduces oxygen consumption and increases mitochondrial spare capacity in astrocytes.

(A) Diagram of the procedure and measurements of the Seahorse assay. The sequential addition of mitochondrial toxins permits the measurement of different respiratory states. After recording basal respiration, oligomycin (OGM) is added to block complex V and to eliminate ATP production-linked oxygen consumption. Addition of FCCP allows the free flux of protons through the mitochondrial inner membrane and maximum oxygen consumption. Rotenone and antimycin A (ROT+AMA) inhibit complex I and III, and prevent proton pumping. Non-mitochondrial residual oxygen consumption is subtracted from all measurements. (B) Oxygen consumption rate (OCR) of NECos, neurons, and glia, in the presence and absence of EtBr. (C) Basal, leak, max, and spare capacity OCR of NECos, neurons, and astrocytes, in the presence and absence of EtBr. N for each group is included in figure legends. Error bars reflect +/-SEM. *** = p < 0.001; ** = p < 0.01; * = p < 0.05, after Dunn’s post-hoc test.

Fig 5. Impaired mitochondrial respiration increases astrocytic, but not neuronal, rate of glycolysis.

(A) Basal extracellular acidification rate (ECAR) of NECos, neurons, and astrocytes, in the presence and absence of EtBr. (B) Relative quantity of electron donors (i.e. NAD(P)H, FADH₂) across culture conditions, measured by MTS absorbance and normalized to cell number. (C) Relative ADP plotted against relative ATP. Arrows show the direction of change from control to EtBr treatment for each culture condition. RLU, relative luminescence units. Error bars reflect +/-SEM. N for each group is included in figure legends. (A) *** = p < 0.001; * = p < 0.05, following Dunn’s post-hoc test. (B), (C) *** = p < 0.001, ** = p < 0.01 compared to respective controls.

Fig 6. Representative NECo ¹H NMR spectra.

¹H NMR spectra at 600MHz from control NECo cultures from 0.8–4.5 ppm and from 5.0–8.8 ppm. 5.0–8.8 ppm section is magnified 10x relative to 0.8–4.5 ppm section. Leu–leucine; Ile–isoleucine; Val–valine; Lac–lactate; Thr–threonine; Ala–alanine; Ace–acetate; NAA–N-acetyl aspartate; Glx–glutamine/glutamate; GABA- gamma-amino butyrate; Glu–glutamate; Pyr–pyruvate; Suc–succinate; Gln–glutamine; Asp–aspartate; Cr–creatine; PCr–phosphocreatine; PCho–phosphocholine; GPC–glycerophosphocholine; Tau–taurine; Gly–glycine; Ser–serine; Ino–myoinositol; Glc–glucose; Fum–fumarate; Phe–phenylalanine; Tyr–tyrosine; AXP–combined adenine nucleotides; For–formate. Acetate and formate peaks are contaminants from sample preparation. Insets–Metabolites quantified for this study. Right, PCr and Cr peaks, 10x magnified. Left, ADP+ATP peak, 50x magnified.

Fig 7. Reduced mitochondrial function decreases neuronal PCr/Cr ratio, and decreases glial tCr.

(A) Phosphocreatine (PCr), creatine (Cr), and PCr/Cr ratio in NECos, neurons, and astrocytes, with and without EtBr. (B) Absolute PCr, Cr, and PCr/Cr ratio in neuronal cultures supplemented with 5 mM Cr 24h before harvest, with and without EtBr. (C) ATP+ADP in NECos, neurons, and astrocytes, with and without EtBr. N for each group is included in figure legends. Error bars reflect +/-SEM. *** = p< 0.001; ** = p < 0.01; * = p < 0.05, compared to respective controls.

Fig 8. Decreased mtDNA quantity corresponds to decreased expression of mtCK.

(A) Log2-fold change of mtCK and B-CK expression in rat NECos, neurons, and astrocytes, normalized to reference genes. (B) Log2-fold change of mtCK and B-CK expression in matched human dyskinetic and non-dyskinetic PD patients, normalized to reference genes and relative to control subjects. Inset: Relative mtDNA quantity from the same cohort as in (B), normalized to nDNA [43]. (A) *** = p < 0.001; ** = p < 0.01, relative to respective controls. (B, inset). ** = p < 0.01, * = p < 0.05, following Tukey’s HSD post-hoc test.