GAK and PRKCD are positive regulators of PRKN-independent mitophagy

Background

Autophagy is an intracellular degradative pathway, essential for maintaining cellular homeostasis. Mitophagy is a form of selective autophagy, by which the cell captures mitochondria within a double membrane autophagosome and delivers this to the lysosome for degradation. This is essential for maintaining a healthy pool of mitochondria, and defects in this pathway have most prominently been linked to neurodegenerative diseases. Mitophagy regulation has infamously been characterised by PINK1 phosphorylation of the mitochondrial protein Parkin (PRKN). However, in vivo observations of basal mitophagy in mice and fly models are often characterised as PINK1- and PRKN-independent ^[1,2]. One form of PRKN-independent mitophagy is mediated by hypoxia and HIF1α ^[3]. To mimic this pathway, without hypoxia induction, HIF1α can be stabilised via binding to small molecules, such as the iron chelator deferiprone (DFP) ^[4,5].

 

Key Findings

GAK and PRKCD are identified as novel regulators of mitophagy

DFP was used in this study to induce mitophagy in the U2OS cell line, which lowly expresses PRKN and is therefore deficient in PRKN-dependent mitophagy ^[6]. It was established that DFP is a potent inducer of PRKN-independent mitophagy in this cell line. This was indisputably evidenced by an image-based reporter system of mitophagy that was validated by correlative light and electron microscopy (CLEM), decreases in enzyme activity of the mitochondrial matrix protein citrate synthase and overall depletion of mitochondrial proteins shown by proteomic analysis.

The authors next aimed to identify novel regulators of PRKN-independent mitophagy. Many regulatory proteins have been identified in autophagy and mitophagy, however, there is a glaring lack of research in lipid mediators and lipid-binding proteins. The authors, therefore, screened 197 proteins with lipid-interacting domains, known to be expressed in U2OS cells. This image-based siRNA screen identified eleven novel lipid-binding proteins that were implicated in the regulation of DFP-mediated mitophagy. Among these were the kinases cyclin G-associated kinase (GAK) and Protein Kinase C Delta (PRKCD). As PINK1 plays an important role in PRKN-dependent mitophagy by phosphorylating PRKN, it was contemplated that GAK and PRKCD kinases may substitute for PINK1 in a PRKN-independent pathway. Notably, GAK is also a risk factor for Parkinson’s Disease, and PRKCD is known to localise to mitochondria.

Pursuit of these novel regulators found that GAK and PRKCD are integral mediators of DFP-induced mitophagy, with PRKCD, but not GAK, localising to the mitochondria in unstimulated and DFP-stimulated cells. Using kinase inhibitors of these proteins, their kinase activity was found to be necessary for this process. Importantly, GAK and PRKCD kinase activity was dispensable for PRKN-dependent mitophagy and starvation-induced autophagy in cells overexpressing PRKN. This signifies that GAK and PRKCD are specific mediators of PRKN-independent mitophagy.

 

Investigating the mechanism of GAK and PRKCD in mitophagy regulation

To delineate the mechanism of GAK and PRKCD regulation, their potential effects on HIF1α stabilisation was explored. Inhibition of GAK and PRKCD kinase activity did not affect expression of HIF1a responsive genes BNIP3 and BNIP3L (NIX), nor did it affect their phosphorylation, which is known to promote autophagy protein, ATG8, recruitment during mitophagy. However, it was found that PRKCD kinase activity enhanced recruitment of key autophagy regulators ATG13 and ULK1 to the early autophagosome during DFP-induced mitophagy. This was not apparent with GAK kinase inhibition, but its activity was important for mitochondrial morphology and also affected lysosome size and number. This may suggest that GAK is important for mitochondrial delivery to autophagosomes and subsequently lysosomes for degradation. Interestingly, mass spectrometry revealed that inhibition of GAK kinase activity activated several glycolytic enzymes, suggesting there may be some metabolic alterations mediated by GAK during mitophagy.

 

GAK and PRKCD regulate mitophagy in vivo

Finally, it was determined that these kinases are important for basal mitophagy in vivo. Basal mitophagy was shown to be regulated by kinase activity of a GAK orthologue in C. elegans and PRKCD paralogue in zebrafish. Dimethyloxaloylglycine (DMOG), another HIF1α stabilising small molecule, also induced PRKN-independent mitophagy in zebrafish that was dependent on PRKCD kinase activity.

 

 

What I like about the paper…

This study has uncovered novel regulators of PRKN-independent mitophagy and proposes that different machinery is employed for PRKN-dependent and -independent mitophagy. With a plethora of research on the conventional PRKN-dependent mitophagy, it is refreshing to delve into work shifting this focus. It is important to investigate this peripheral pathway as the authors highlight that in vivo, basal mitophagy is often PRKN-independent, therefore, it may have important implications for health. GAK and PRKCD regulation of mitophagy was also found to be an evolutionarily conserved mechanism, with relevance in C. elegans and zebrafish. As GAK is a known risk factor for Parkinson’s disease, describing this pathway and its mechanism with further clarity could have important implications for understanding the role of mitophagy in neurodegenerative diseases.

 

Open questions to the authors:

• Why do you think DMOG, and not DFP, activated mitophagy in vivo? DFP acts as an iron chelator that inhibits proline hydroxylases (iron is a critical cofactor for their normal function), whereas, DMOG directly inhibits proline hydroxylases to prevent the degradation of HIF1a. As the mechanism of DFP and DMOG is subtly different, could this suggest that PRKN-independent mitophagy differs in vivo or more specifically in zebrafish?
• As DFP was used as a substitute for hypoxia-induced mitophagy, would you predict similar findings with a hypoxia model? Furthermore, as defective mitophagy has been implicated in cancer, where hypoxia conditions are well-characterised, do you think GAK or PRKCD could have a role in this context?
• It was very interesting that glycolytic proteins were altered by GAKi. As glycolytic switching and mitophagy have been linked to cell differentiation, most prominently in immune cells and neurons, would you hypothesize a role for GAK in this process?

 

References:

1. McWilliams, T. G. et al. Basal Mitophagy Occurs Independently of PINK1 in Mouse Tissues of High Metabolic Demand. Cell Metab. 27, 439-449.e5 (2018).
2. Lee, J. J. et al. Basal mitophagy is widespread in Drosophila but minimally affected by loss of Pink1 or parkin. Cell Biol. 217, 1613–1622 (2018).
3. Bruick, R. K. Expression of the gene encoding the proapoptotic Nip3 protein is induced by hypoxia. Natl. Acad. Sci. U. S. A. 97, 9082–7 (2000).
4. Allen, G. F. G., Toth, R., James, J. & Ganley, I. G. Loss of iron triggers PINK1/Parkin- independent mitophagy. EMBO Rep. 14, 1127–35 (2013).
5. Zhao, J.-F., Rodger, C. E., Allen, G. F. G., Weidlich, S. & Ganley, I. G. HIF1α-dependent mitophagy facilitates cardiomyoblast differentiation. Cell Stress. 4, 99–113 (2020).
6. Tang, M. Y. et al. Structure-guided mutagenesis reveals a hierarchical mechanism of Parkin activation. Commun. 8, (2017).
7. Asquith, C. R. M. et al. SGC-GAK-1: A Chemical Probe for Cyclin G Associated Kinase (GAK). J. Med. Chem. 62, 2830–2836 (2019).

 

Tags: autophagy, mitophagy

Posted on: 19 January 2021

doi: https://doi.org/10.1242/prelights.27046

Read preprint (2 votes)