p13 protects against Parkinson's disease

Abstract

Mitochondrial dysfunction is thought to contribute to Parkinson's disease progression, and factors that can overcome mitochondrial defects could potentially be used to combat the disease and prevent neuronal death. In this issue, Inoue et al 1 report that reduction of p13, a mitochondrial protein that inhibits complex I assembly, rescues the cellular and behavioral defects of Parkinson's disease models. This work suggests that stabilizing the mitochondrial electron transport chain may be beneficial in the context of Parkinson's disease.

Parkinson's disease (PD) is among the most common neurodegenerative diseases and is characterized by specific movement and non‐motor symptoms 2. Dopaminergic neurons die, but the disease is also systemic and other areas of the brain are affected as well 2. There is no cure and only symptomatic treatment based on neurotransmitter replacement exists. However, the disease relentlessly progresses and the effectivity of these treatments trails over time. A combination of environmental and genetic factors promote the onset and progression of the disease 2. The task ahead is to discover strategies that neutralize these risks as to stop disease progression and, eventually, arrive at a cure.

The search for disease‐modifying therapies takes advantage of several animal models that recapitulate PD features 3. Researchers have further used toxins that cause parkinsonian symptoms in humans, and administration of rotenone, MPTP, or paraquat to animals recapitulates cardinal features of the disease. More recently, mutations in more than 20 genes have been identified in families with PD. While these genes do not all map onto one specific molecular pathway, what is exciting is that both the toxins and a subset of PD mutations affect mitochondrial function, including electron transport chain function, directly 3. In addition, mitophagy defects may contribute to the accumulation of defective mitochondria 4. Indeed, Pink1, Parkin, and DJ1 mutant animals, across species, often show defects in mitochondrial membrane potential, a decrease in complex I activity, an increase in oxidative stress, and/or altered mitochondrial morphology 5, 6. Similarly, rotenone and MPTP block complex I activity. Mitochondrial defects have now also been observed in samples from sporadic PD patients and in patient‐derived induced human neurons. Given that many familial and idiopathic patients as well as toxin‐induced parkinsonism show mitochondrial dysfunction, a major strategy to combat PD has focused on counteracting these mitochondrial defects.

In this issue, Inoue et al 1 show that lower expression of p13, a protein that resides in mitochondria, suppresses the mitochondrial defects seen in toxin‐induced and genetic PD models. p13 is a mitochondrial matrix protein known to be reduced in pancreatic cells upon oxidative stress 7. p13 binds to NDUFAB1 and ATPAF2 subunits of the electron transport chain complexes I and V, respectively 8, and Inoue et al show that p13 directly modulates the activity of complex I: lower p13 levels promote complex I assembly, providing an explanation for how p13 might functionally interact with the PD models.

Inoue et al tested whether p13 levels are altered in PD models and whether this protein could modulate mitochondrial function. They show that p13 levels are decreased in an in vitro toxin‐induced model of PD (human neuroblastoma cells treated with rotenone, CCCP, and MPTP) and in vivo in MPTP‐treated mice. Curiously, such a compensatory lowering of p13 levels was not observed in the genetic model the authors use, suggesting a potentially different underlying mechanism or a difference between acute and chronic models. Nonetheless, in a surprising set of results, they find that further lowering p13 levels is protective and rescues well‐established phenotypes of the toxin‐induced and genetic PD models, including mitochondrial depolarization, defects in complex I activity, and cell death. Lowering p13 levels using shRNA also prevents the disruption of complex I assembly by toxin treatment (Fig 1). The opposite, overexpression of p13, mimics the phenotypes of the toxin‐induced PD models and exacerbates the phenotypes in rotenone‐treated cells. Hence, p13 is a dosage‐sensitive modifier of mitochondrial activity.

Figure 1. p13 downregulation promotes the assembly of mitochondrial complex I and prevents mitochondrial dysfunction and cell death in Parkinson's disease models.

Parkinson's disease (PD) toxins, such as rotenone or MPTP, affect mitochondrial function: assembly of complex I (shown in orange) is decreased, mitochondrial membrane potential (Ψm) is reduced, and reactive oxygen species accumulate. This ultimately culminates in the degeneration of dopaminergic neurons in the substantia nigra. Reduction of p13 levels promotes the assembly of complex I, rescues the mitochondrial defects of PD models, and prevents neuronal death.

The timing of lowering p13 levels, that is, before or after toxin treatment, is key and dictates to what extent p13 needs to be lowered to achieve rescue. Indeed, simple toxin treatment without genetically manipulating p13 expression levels results in a compensatory drop in p13 levels, but this is not enough to rescue the phenotypes. However, if p13 levels are lowered by shRNA before toxin treatment to a very similar level as those achieved by simple toxin treatment alone, the authors do observe rescue. The authors propose that it is conceivable that the p13 reduction in the toxin PD models occurs as a response to the pathogenic insult, perhaps as an attempt to compensate for the cellular stress caused by mitochondrial dysfunction. Acute lowering of p13 might “hyperstabilize” complex 1, thus protecting the mitochondria before the toxins are added. Interestingly, a similar “stabilizing‐compensatory” effect has previously been observed. The levels of aconitase and fatty acid synthase (FASN) were also found to be lower in PD models (here, this was observed in genetic PD models). High aconitase levels facilitate the production of mitochondrial ROS, while high FASN lowers levels of the mitochondrial lipid cardiolipin that stabilizes complex I 6, 9. Further reducing the expression of aconitase or the activity of FASN in genetic models of PD rescued the mitochondrial defects 6, 9. A potentially interesting avenue to pursue could be to investigate which other proteins are downregulated in genetic animal PD models and assess whether lowering their levels or activity further is protective.

Inoue et al use toxins that kill dopaminergic neurons in the substantia nigra of treated mice, and as a result, these mice suffer from motor defects. MPTP is internalized specifically in dopaminergic neurons leading to decreased levels of p13 in the midbrain (which contains dopaminergic neurons), while other brain areas remain unaffected. Using CRISPR/Cas9, the authors created a novel knockout mouse lacking p13 expression. Excitingly, p13 knockout mice treated with MPTP do not present motor deficits or dopaminergic neuron loss. The authors go on to show that in p13 knockout mice, complex I assembles more readily in the presence of toxins, thus maintaining a higher level of mitochondrial function. This feat might be exploited also in the context of other mitochondrial disease models.

In the slipstream of failing clinical trials, it will be important to assess whether lowering p13 levels is protective in human neurons as well. Inoue et al took a first step by showing that p13 mRNA tends to be decreased in brains of PD patients. While exciting, the sample size is still small and further confirmation will be needed. In addition, it will be interesting to see whether the findings are recapitulated in patient material or in induced neurons from patient‐derived iPSC. Another point to consider is the need for stratification. The work by Inoue et al addresses mitochondrial dysfunction, and while there are clear connections to PD as outlined above, it is conceivable that other genes and mutations trigger PD through other cellular pathways 10. In this context, testing mitochondrial function in patients, or starting with specific familial cases where mitochondrial dysfunction is central, will be important in bringing this work to the next level.

As therapeutic strategies are suggested by the work by Inoue et al (as well as by work from several other groups), the necessity to develop early and reliable diagnosis, ideally in the prodromal phase of the disease, becomes even more important. Currently, PD is diagnosed several years after dopaminergic neuron degeneration has begun. The work by Inoue et al shows that lower p13 levels protect mitochondria from toxins. It will be interesting to determine whether the acute inhibition of p13 can also protect in the context of ongoing neurodegeneration, to stop or slow disease progression. Finally, it will be exciting to address whether lower p13 levels also restore non‐motor symptoms of PD.

See also: N Inoue et al (March 2018)