The Proteomic Landscape of Parkin-Deficient and Parkin-Overexpressing Rat Nucleus Accumbens: An Insight into the Role of Parkin in Methamphetamine Use Disorder

Abstract

In recent years, methamphetamine (METH) misuse in the US has been rapidly increasing, and there is no FDA-approved pharmacotherapy for METH use disorder (MUD). We previously determined that ubiquitin-protein ligase parkin is involved in the regulation of METH addictive behaviors in rat models of MUD. Parkin is not yet a "druggable" drug target; therefore, this study aimed to determine which biological processes, pathways, and proteins downstream of parkin are likely drug targets against MUD. Employing young adult Long Evans male rats with parkin deficit or excess in the nucleus accumbens (NAc), label-free proteomics, and molecular biology, we determined that the pathways downstream of parkin that are candidates for regulating METH addictive behaviors in young adult male rats are mitochondrial respiration, oxidative stress, AMPA receptor trafficking, GABAergic neurotransmission, and actin cytoskeleton dynamics.

Keywords: methamphetamine; mitochondria; nucleus accumbens; parkin; proteome.

Figure 1

Experimental Design. Nucleus accumbens tissue from wild-type (WT), parkin knockout (PKO), and parkin-overexpressing (PO) rats was assessed for proteomic landscape and the levels of several Krebs cycle and electron transport chain enzymes by SDS-PAGE/western blotting. Bioinformatic analyses were then used to identify biological processes impacted by parkin deficit or excess in the NAc. The tissues were also assessed for oxygen consumption rate and ATP levels.

Figure 2

Differentially Expressed Proteins in Parkin-Deficient and Parkin-Overexpressing Rat Nucleus Accumbens. (A,B) Principal component analysis (PCA) of proteomic data derived from parkin-deficient (PKO), parkin-overexpressing (PO), and wild-type (WT) rat nucleus accumbens (n = 3/group). Each mark represents an individual rat. Note that the PKO and PO groups are clearly separate from the WT group. (C,D) Volcano plots showing relationships between the magnitude of change in protein expression (log2 of fold change; x-axis) and statistical significance of this change (−log10(p); y-axis] in a comparison of PKO to WT and PO to WT. Colored points represent differentially expressed proteins (DEPs) at a cutoff p = 0.01. (E,F) Heatmaps depicting DEPs upregulated (red) and downregulated (blue) in rat nucleus accumbens by parkin knockout (PKO) or overexpression (PO) relative to wild-type (WT) nucleus accumbens (p < 0.01, Qlucore Explorer).

Figure 3

Parkin Deficit or Overexpression in Rat Nucleus Accumbens Changes Aerobic Respiration in Opposite Directions. (A,B) Gene set enrichment analysis (GSEA)-enrichment plots showing protein sets from GOBP_AEROBIC RESPIRATION signature significantly overrepresented (p < 0.05) in parkin knockout (PKO) or parkin overexpression (PO) rat nucleus accumbens, as compared to wild-type nucleus accumbens. (C,D) Hierarchical clustering heatmap of the leading-edge proteins from the GOBP_AEROBIC_RESPIRATION (p < 0.05) in PKO NAc and PO NAc. Three Krebs cycle enzymes, citrate synthase (CS), mitochondrial malonate dehydrogenase (MDH2), and dihydrolipoamide S-succinyltransferase (DLST), were decreased in PKO NAc and increased in PO NAc relative to wild-type NAc (red arrows).

Figure 4

Identification of Differentially Expressed Proteins in Parkin-Deficient and Parkin-Overexpressing Rat Nucleus Accumbens. (A,B) Venn diagrams showing numbers of overlapping DEPs between PKO and PO rats detected by Qlucore Explorer at p < 0.01 (A) or p < 0.05 (B). (A) At p < 0.01, there were 58 DEPs unique to the PKOs, 41 DEPs unique to the POs, and no shared DEPs. (B) At p < 0.05, there were 196 DEPs unique to the PKOs, 145 DEPs unique to the POs, and 32 shared DEPs. (C) Top 20 pathways altered by parkin loss in the nucleus accumbens (q < 0.05, ShinyGO). (D) Top 20 pathways altered by parkin overexpression in the nucleus accumbens (q < 0.05, ShinyGO). (E) Pathways to which DEPs shared by PKOs and POs belong (q < 0.05, ShinyGO). (F) GSEA analysis of the shared DEPs (p < 0.05).

Figure 5

Adeno-Associated, Virus-Mediated Overexpression of Parkin in Rat Nucleus Accumbens Is Variable and Negatively Correlates with Methamphetamine Self-Administration. (A) Assessment of parkin levels in wild-type, parkin knockout (PKO), and parkin-overexpressing nucleus accumbens (n = 7) by SDS-PAGE and western blotting, with GAPDH as a loading control. The top panel shows the results from samples analyzed by proteomics, whereas the bottom panel shows the results from additional samples. Variability in parkin overexpression should be noted. (B) In our previous study, parkin overexpression negatively correlated with METH intake in the drug self-administration paradigm [4]. Green dots represent parkin levels in wild-type nucleus accumbens, whereas the red dots represent parkin levels in parkin-overexpressing nucleus accumbens.

Figure 6

Parkin Deficit or Overexpression in Rat Nucleus Accumbens Changes Krebs Cycle Enzyme Levels in Opposite Directions. SDS-PAGE and western blotting showed statistically significant differences between parkin knockout (PKO) and parkin-overexpressing (PO) rats, as well as between these groups and wild-type (WT) controls. (A) dihydrolipoamide dehydrogenase (DLST), (B) mitochondrial malate dehydrogenase (MDH2), (C) 2-oxoglutarate dehydrogenase (OGDH), (D) dihydrolipoyl dehydrogenase (DLD), and (E) citrate synthase (CS). One-way ANOVA with Holm-Sidak’s post hoc test). * p < 0.05, ** p < 0.01, *** p < 0.001, n = 7/group. The data is expressed as mean ± SEM. (F) CS levels show a positive correlation with parkin levels (p < 0.05, Pearson correlation test). Abbreviations: GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Figure 7

Parkin Deficit and Overexpression in Rat Nucleus Accumbens Change Mitochondrial Function. (A) Oxygen consumption by complex IV (COX), normalized to (B) the levels of subunit IV (COXIV) of complex IV measured by SDS-PAGE and western blotting. (C) ATP levels generated in the nucleus accumbens of wild-type (WT), parkin knockout (PKO), and parkin-overexpressing (PO) rats. * p < 0.05, ** p < 0.01, ns, not significant (one-way ANOVA with Holm–Sidak post hoc test), n = 4/group. The data is expressed as mean ± SEM.

Figure 8

Top Upstream Transcriptional Regulators in Parkin-Deficient and Parkin-Overexpressing Rat Nucleus Accumbens. IPA-predicted upstream transcriptional regulators in the nucleus accumbens of parkin knockout (PKO) vs. wild-type (WT) and parkin-overexpressing (PO) vs. WT rats presented according to (A) Benjamini–Hochberg (B-H) p-value or (B) z-score, which predicts activation or inhibition state of a gene/pathway.