Is WDR45 the missing link for ER stress-induced autophagy in beta-propeller associated neurodegeneration?

ABSTRACT

Beta-propeller protein-associated neurodegeneration (BPAN) is caused by mutations in the autophagy gene WDR45/WIPI4. In human, BPAN is associated with static encephalopathy in childhood and neurodegeneration in adulthood (SENDA). It has been proposed that WDR45 mutations cause neurodegeneration due to defective autophagy. Whether these mutations cause a global attenuation or a defect in a subset of autophagy functions is unknown. Based on a recent study showing that wdr45 knockout mice exhibit defective autophagy associated with an increased ER stress, we propose that ER-mediated autophagy, a selective activation of autophagy, is defective in mouse and cellular models of BPAN. We discuss the implication of these findings on the pathophysiological relevance of the relationship between ER stress and autophagy in BPAN as well as other neurodegenerative diseases exhibiting ER stress and defective autophagy.

KEYWORDS:

• Autophagy
• ER stress
• neurodegeneration
• beta propeller associated neurodegeneration
• BPAN

Maintenance of protein homeostasis, or proteostasis, necessitates the correct folding of membrane and secreted proteins within the endoplasmic reticulum (ER). The ER adapts to physiological demands for protein synthesis by expanding its protein folding ability. Several physiological and pathological conditions alter protein folding at the ER, which promotes the accumulation of misfolded proteins leading to ER stress [1]. Upon ER stress, cells activate an adaptive dynamic signaling network known as the unfolded protein response (UPR), which restores proteostasis. The UPR is coordinated by 3 ER transmembrane sensors, ATF6 (activating transcription factor 6), ERN1/IRE1 (endoplasmic reticulum [ER] to nucleus signaling 1) and EIF2AK3/PERK (eukaryotic translation initiation factor 2 alpha kinase 3), whose activation results in attenuation of global protein translation and upregulation of protein folding capacity [2]. ER stress also induces macroautophagy/autophagy, a pro-survival mechanism responsible for the degradation and recycling of cytoplasmic contents, including damaged organelles or misfolded proteins, by lysosomal proteases [3]. We have previously shown that ER stress-induced autophagy is an efficient survival mechanism that inhibits apoptosis in mouse and fly models of Parkinson disease and retinal degeneration [4,5]. Autophagy also inhibits ER expansion by reticulophagy, a selective autophagy pathway by which ER membranes are removed by sequestration within autophagosomes [6]. Although it is clear that perturbation of proteostasis and impaired autophagy are observed in several neurodegenerative diseases such as Alzheimer, Parkinson and Huntington diseases, the pathophysiological relevance of ER stress-induced autophagy or reticulophagy remains to be demonstrated in neurological diseases [7,8].

In their recent work published in Autophagy, Liao and col. investigated BPAN that is caused by mutations in the autophagy gene WDR45/WIPI4 [9]. WDR45 is one of the 4 homologs of yeast Atg18, which plays important roles in autophagy [10]. In human, WDR45 mutations lead to a SENDA syndrome, which has been established as a subtype of neurodegeneration with brain iron accumulation (NBIA) [11]. Lymphoblastic cell lines derived from BPAN affected subjects harboring WDR45 mutation show lower autophagic activity, and defective autophagy flux is observed in wdr45 knockout (KO) mouse brains [11,12]. These findings provide evidence that an autophagy defect is associated with BPAN in humans and mice. Thus, the current hypothesis regarding the pathological basis of WDR45 mutation is that BPAN neuronal tissues exhibit a reduced autophagy. However, it is unclear why autophagy defect and neural damage are milder in Nes-Wdr45 floxed mice and constitutive wdr45 KO mice [9,12] than the massive neuron loss reported in neural-specific depletion of autophagic genes Atg5 or Atg7 in mice [13,14]. Interestingly, Liao et al. observed that lower autophagy in the constitutive wdr45 KO mutant is associated with ER expansion, ER stress and cell death due to an unleashed UPR in wdr45 KO mice. Their results suggest that WDR45 regulates a specific autophagic program dedicated to the control of ER stress. This is supported by the fact that ER stress induces: 1) the expression of wdr45 [15]; 2) the recruitment of WDR45 at the ER and 3) the WDR45-dependent degradation of a subset of ER-resident proteins [9]. Their data, however, do not support a role of reticulophagy receptors in this process but instead suggest that WDR45 regulates a distinct mechanism that controls ER stress by sending damaged ER protein and aggregates for autophagic elimination [9].

Mechanistically, it is tempting to hypothesize that WDR45 is a UPR target, which promotes autophagy in response to the ER stress. This is supported by a study showing that WDR45 transcripts are upregulated in response to ER stress induced by thapsigargin in HeLa cells [15]. Although more work will be required to characterize WDR45 function at the ER and its mechanism of activation, these results suggest that WDR45 is not a mere regulator of autophagy but rather could be a sensor of ER stress that promotes autophagy activation in response to ER stress. In conclusion, the degeneration observed in BPAN patients may be due to a defective ER stress-induced protective autophagy pathway in which WDR45-dependent autophagy serves as a homeostatic rheostat to buffer proteostasis fluctuation of the ER and subsequent ER stress-induced cell death. Interestingly, another study investigating the role of WDR45 and WDR45B/WIPI3 (WD repeat domain 45B) showed that autophagy and neurological deficits of mice knocked out for Wdr45 or Wdr45b are less severe than in the double KO [16]. Their results suggest that WDR45 and WDR45B act cooperatively in autophagy and neural homeostasis. An important pending question is whether glial cells and an inflammatory response contribute to neurodegeneration in WDR45 or WDR45B mutant mice. Zhao and colleagues found that GFAP-positive astrocytes accumulate in the brain of wdr45b KO mice, which suggests that an inflammatory response could contribute to the neurological deficit in the wdr45b KO. However, autophagy impairment visualized by the accumulation of SQSTM1 is restricted to neurons, suggesting that the primary deficit caused by WDR45B loss occurs in mouse neurons.

Finally, Liao and colleagues observed that activating autophagy (rapamycin) or limiting ER overload (TUDCA) is beneficial for wdr45 KO mice [9]. Collectively these results indicate that wdr45 KO mice represent a promising preclinical model to further explore if therapeutic intervention that aims at restoring defective autophagy and limiting ER overload will be beneficial for BPAN patients.

Acknowledgments

We thank the association Autour du BPAN for funding to BM.

Disclosure statement

No potential conflict of interest was reported by the authors.