PINK1 is required for timely cell-type specific mitochondrial clearance during Drosophila midgut metamorphosis

Author’s Accepted Manuscript PINK1 is required for timely cell-type specific mitochondrial clearance during Drosophila midgut metamorphosis Yan Liu, Jingjing Lin, Minjie Zhang, Kai Chen, Shengxi Yang, Qun Wang, Hongqin Yang, Shusen Xie, Yongjian Zhou, Xi Zhang, Fei Chen, Yufeng Yang

PII: DOI: Reference:

www.elsevier.com/locate/developmentalbiology

S0012-1606(16)30145-2 http://dx.doi.org/10.1016/j.ydbio.2016.08.028 YDBIO7240

To appear in: Developmental Biology Received date: 16 March 2016 Revised date: 23 August 2016 Accepted date: 25 August 2016 Cite this article as: Yan Liu, Jingjing Lin, Minjie Zhang, Kai Chen, Shengxi Yang, Qun Wang, Hongqin Yang, Shusen Xie, Yongjian Zhou, Xi Zhang, Fei Chen and Yufeng Yang, PINK1 is required for timely cell-type specific mitochondrial clearance during Drosophila midgut metamorphosis, Developmental Biology, http://dx.doi.org/10.1016/j.ydbio.2016.08.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

PINK1 is required for timely cell-type specific mitochondrial clearance during Drosophila midgut metamorphosis Yan Liua1, Jingjing Lina1, Minjie Zhanga, Kai Chena, Shengxi Yanga, Qun Wanga, Hongqin Yangb, Shusen Xieb, Yongjian Zhouc, Xi Zhanga, Fei Chena*, Yufeng Yanga,b* a

Institute of Life Sciences, Fuzhou University, Fuzhou, Fujian Province, China, 350108.

b

Key Laboratory of Optoelectronic Science and Technology for Medicine, Ministry of Education, Fujian Normal University, Fuzhou, China 350007

c

Department of Gastric Surgery, Union Hospital of Fujian Medical University, Fujian, China

350001 [email protected] [email protected] * Corresponding authors.

Abstract Mitophagy is the selective degradation of mitochondria by autophagy, which is an important mitochondrial quality and quantity control process. During Drosophila metamorphosis, the degradation of midgut involves a large change in length and organization, which is mediated by autophagy. Here we noticed a cell-type specific mitochondrial clearance process that occurs in enterocytes (ECs), while most mitochondria remain in intestinal stem cells (ISCs) during metamorphosis. Although PINK1/PARKIN represent the canonical pathway for the elimination of impaired mitochondria in varied pathological conditions, their roles in developmental processes or normal physiological conditions have been less studied. To examine the potential contribution of PINK1 in developmental processes, we monitored the dynamic expression pattern of PINK1 in the midgut development by taking advantage of a newly CRISPR/Cas9 generated knock-in fly strain expressing PINK1-mCherry fusion protein that presumably recapitulates the endogenous expression pattern of PINK1. We disclosed a spatiotemporal correlation between the expression pattern of PINK1 and the mitochondrial clearance or persistence in ECs or ISCs respectively. By mosaic genetic analysis, we then demonstrated that PINK1 and PARKIN function epistatically to mediate the specific timely removal of mitochondria, and are involved in global autophagy in ECs during Drosophila midgut metamorphosis, with kinase-dead PINK1 exerting dominant negative effects. Taken together, our studies concluded that the PINK1/PARKIN is crucial for timely cell-type specific mitophagy under physiological conditions and demonstrated again that Drosophila midgut metamorphosis might serve as an elegant in vivo model to study autophagy.

Introduction In Drosophila, metamorphosis transforms a larva into a completely new body structure through pupae stage. During the course of metamorphosis, certain larvae tissues such as larvae 1

These authors contribute equally to this work.

midgut are significantly reshaped and reconstituted (Jiang et al., 1997). The Drosophila midgut contains multipotent ISCs scattered along its basement membrane. ISCs generate two different cell types, ECs and enteroendocrine (ee) cells (Guo and Ohlstein, 2015; Ohlstein and Spradling, 2006). During metamorphosis, the larvae intestine undergoes a significant reduction in length at the onset of puparium formation. Timely histolysis of larvae midgut appears especially in ECs and a new adult midgut epithelium forms from midgut progenitor cells (Bender et al., 1997; Santhanam et al., 2014). In Drosophila melanogaster the steroid hormone ecdysone regulates molting and metamorphosis (Truman and Riddiford, 2002), and also regulates programmed cell death (PCD) during metamorphosis (Baehrecke, 2000; Jiang, 1997; Kumar and Cakouros, 2004). PCD in larvae midgut begins upon puparium formation in response to the late larvae pulse of ecdysone (Baehrecke, 2000), and autophagy is essential for the developmental PCD (Denton et al., 2009). The most prominent role of mitochondria is to produce ATP through respiration, and to regulate cellular metabolism (Voet et al., 2006). Mitochondria also play a central role in many other physiological tasks, such as apoptosis (Green, 1998), signal transduction, storage of calcium ions and cellular proliferation (Hajnoczky et al., 2006). Mitochondria need to be eliminated when it has aged or been damaged by reactive oxygen species (ROS). A selective autophagy for mitochondria is termed mitophagy. The primary functions of mitophagy are for turnover and clearance of dysfunctional or superfluous mitochondria in order to prevent further cell injuring. The number of mitochondria in a cell can vary widely by organisms, tissues and cell types. Besides the role of quality control, mitophagy also regulates organelle number in response to developmental or physiological cues (Kanki and Klionsky, 2008). Mitophagy mediates removal of superfluous mitochondria from developing erythrocytes (Ney, 2011), and also contributes to maternal inheritance of mitochondrial DNA through the clearance of sperm-derived mitochondria (Rawi et al., 2011; Sato and Sato, 2011). However, the role of mitophagy needs further exploration in the degradation of the Drosophila intestine during metamorphosis. A canonical pathway described for mammalian mitophagy is mediated by PINK1 and PARKIN. Mutations in the serine/threonine-protein kinase PINK1 and the cytosolic E3 ligase PARKIN can cause Parkinson’s disease (PD). In Drosophila, PINK1 and PARKIN function in a convergent pathway to maintain mitochondrial integrity and cell survival, with PINK1 functioning upstream of PARKIN (Clark et al., 2006; Park et al., 2006; Yang et al., 2006). According to the current model, PINK1 accumulates on the outer membrane of depolarized mitochondria, and this process depends on its kinase activity. PINK1 recruits PARKIN to mitochondria, then PARKIN ubiquitinates some outer membrane protein(s) and then triggers mitochondrial elimination (Vives-Bauza et al., 2010). Reported as well, PINK1 functions with the fission/fusion machinery to regulate mitochondrial dynamics (Yang et al., 2008). PINK1/PARKIN pathway can interact with the mitochondrial fusion protein Mfn2 and mitochondrial adaptor protein Miro, and the degradation of either Miro or Mitofusin is thought to quarantine damaged mitochondria before removal (Chen and Dorn, 2013; Wang et al., 2011). Previous studies suggested that PARKIN activity is regulated by PINK1-mediated phosphorylation during mitophagy (Iguchi et al., 2013; Kane et al., 2014; Kondapalli et al., 2012). However, there have been evidences suggesting that PARKIN may differentially regulate distinct types of autophagy in a PINK1-independent way (Chen et al., 2010), and vice versa (Lazarou et al., 2015). Although PINK1/PARKIN represents the canonical pathway for the elimination of depolarized

mitochondria in varied pathological conditions, few data demonstrated whether this classical pathway affects mitophagy under physiological conditions. Here we showed that PINK1 and PARKIN function together to mediate the timely removal of mitochondria and macroautophagy in ECs during Drosophila midgut metamorphosis, with PINK1 acting upstream of PARKIN and kinase-dead PINK1 displaying dominant negative impacts.

Results Concomitant autophagy and mitophagy in the process of programmed cell size reduction during midgut metamorphosis Drosophila midgut exhibits extensive morphological changes during metamorphosis. We examined this transformation from early-third-instar larvae stage to white pre-puparium (WPP) stage. Noticeably, polyploid ECs (with large nuclei) underwent a dramatic reduction in size from early-third instar larvae stage to WPP stage (Fig1A). We used mCherry-Atg8a as an autophagosome marker to observe the global autophagy. The emergence of mCherry-Atg8a puncta corresponds to the formation of autophagosome, as a characteristic of ongoing autophagy. We found that Atg8a scattered in ECs, and had little co-localization with mitochondria, which had dot or tubular morphology during early-third-instar larvae stage (Fig1B, 1F, 1G). Upon late-third-instar larvae stage, Atg8a started to aggregate and extensively overlap with mitochondria. Besides, there seemed to be an overshoot in the mitochondrial abundance at this stage (Fig1C, 1F, 1G). At WPP stage, mitochondria in ECs were significantly eliminated, which implicated the execution of mitophagy. Meanwhile, the number of Atg8a puncta reached the maximum while Atg8a overlapping with mitochondria decreased (Fig1D, 1F, 1G). Finally, after puparium formation (A-WPP), ECs were ultimately removed by cytolysis while ISCs rich in dot or tubular mitochondria but little Atg8a survived metamorphosis (Fig1E). Therefore, we observed that midgut cells underwent a large reduction in size accompanied by global autophagy induction and mitochondrial clearance during Drosophila larvae development, which is consistent with previous results (Denton et al., 2009). Programmed cell size reduction in midgut transformation requires Atg1, Atg12 and Atg5 Evolutionarily conserved autophagy-related (ATG) genes mediate canonical autophagy (Boya et al., 2013). To determine whether these ATG genes are required for midgut cells shrinking and mitochondrial clearance, we used a series of RNAi lines for ATG genes for genetic mosaic analysis. First, we examined whether the cell size reduction of ECs depended on autophagy. We knockdowned Atg1 or Atg12 in GFP-marked clone cells. Indeed, unlike the significant shrinkage of control cells, cells expressing Atg1 RNAi or Atg12 RNAi were obviously larger (Fig2A-D). In parallel, the degradation of mitochondria in the clone cells with knockdown of Atg1 or Atg12 was blocked compared to their normal neighbor cells (Fig2E-G). The Atg5-Atg12 conjugate, a key regulator of the autophagic process, functions in autophagosome formation (Mai et al., 2012). Consistently, the clone cells with knockdown of both Atg5 and Atg12 had much more mitochondria than control cells (Fig2H). In conclusion, these results suggested that the cellular shrinkage and mitophagy in ECs are mediated through the autophagic machinery during larvae-to-pupae transition. PINK1-mCherry can be an appropriate alternative to endogenous PINK1 protein

PINK1 involves in mitochondrial quality control by identifying damaged mitochondria, and PINK1 selectively recruits PARKIN to specific mitochondria for degradation (Narendra et al., 2012). It has been reported that PINK1 and PARKIN regulate the mitochondrial integrity in Drosophila (Greene et al., 2003). Hence, whether the PINK1/PARKIN pathway mediates mitochondrial clearance in midgut appealed to us. Due to the lack of effective antibodies against Drosophila PINK1 (dPINK1), we tried to use CRISPR-Cas9 mediated genome editing to generate a PINK1 knock-in allele, in which endogenous PINK1 protein could be tagged with linker-mCherry-FLAG at the C-terminus. To this end, we constructed a donor DNA plasmid vector containing the linker-mCherry-FLAG sequence immediately upstream of Pink1 stop codon, flanked by two 1kb homologous arms close to the selected PAM sites (Fig3A). The circular donor DNA vector was co-injected with the U6-sgRNA plasmid into Act5c-cas9 Drosophila embryos subsequently. Positive F1 KI flies were identified by PCR and verified by sequencing (Fig3B, 3C), with a heritable rate of 8.7% (data not shown). Western blotting experiments with an anti-FLAG antibody confirmed the expression of PINK1-linker-mCherry-FLAG fusion protein (abbreviated as PINK1-mcherry thereafter, about 100kD for the full-length fusion protein) (Fig3D). We then examined whether PINK1-mcherry could be a true functional alternative to endogenous PINK1. Pink1-deficient flies display severe physiological defects, such as male sterility and abnormal wing posture (Yang et al., 2006; Zhang et al., 2013). On the contrary, the male PINK1-mCherry flies were fully fertile and the wings of the male PINK1-mCherry flies were completely normal (data not shown), as suggested that adding a tag at the C-terminus of PINK1 didn't interfere with its protein functions. In parallel, we found that while nearly 100% of 15-day-old males of a Pink1-deficient allele (Pink1[B9] null) exhibited abnormal wing posture, all Pink1 [B9] / PINK1-mCherry flies had normal wings, similar to Pink1[B9]/+ flies when maintained at 29℃ (Fig3E), demonstrating that PINK1-mCherry was able to fully complement Pink1 deficiency. Thus, we concluded that the PINK1-mCherry could be functionally equivalent to endogenous PINK1. And due to its KI nature, PINK1-mCherry might recapitulate the endogenous expression pattern of PINK1. The specific expression pattern of PINK1 is MMP-related during midgut metamorphosis To study the involvement of PINK1 in mitochondrial clearance during metamorphosis, we then observed the expression pattern of PINK1 in midgut using PINK1-mCherry fly strain. PINK1 was lowly expressed in ECs in early-third-instar larvae stage, which was consistent with western blot data (Fig4A, 4B). Unlike the scattered distribution of PINK1-mCherry in early-third-instar larvae stage, the intensity of PINK1-mCherry began to elevate in late-third-instar larvae and PINK1-mCherry started to aggregate and partially overlap with mitochondria in ECs (Fig4C, 4F). In line with western blotting data, the gradually increasing expression of PINK1 might coincide with the initiation of mitophagy at the late-third-instar larvae stage (Fig4C, 4A). With midgut cells shrinking during the WPP stage, PINK1-mCherry co-localized more with mitochondria in ECs (Fig4D, 4F). After puparium formation (A-WPP), the mitochondria in ECs were completely removed before the disappearance of ECs. However, in contrast, diploid ISCs rich in mitochondria persisted during metamorphosis (Fig4E), consistent with previous observations (Fig1E). We speculated the recruitment of PINK1 to mitochondria might be due to the loss of mitochondrial membrane potential (MMP) at late-third-instar larvae stage. Therefore, we assessed MMP from early-third-instar larvae stage to late-third-instar larvae stage using

5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolylcarbocyanine iodide (JC-1) dye. Upon late-third-instar larvae stage, a decrease in the red (590nm)/green (530nm) fluorescence intensity ratio indicated depolarization of MMP (Fig4G-I), as coincided with the emerging co-localization between PINK1 and mitochondria (Fig4C, 4F). PINK1 signal correlates with the cell-type specific mitochondrial clearance in fly midguts We then turned to ISCs that could be nicely labeled by escargot-Gal4 [esg-Gal4]-driven GFP. The expression of PINK1 in ISCs remained low from early-third-instar larvae stage to A-WPP stage (Fig5A-D). We then tried to figure out whether mitophagy occurred in ISCs by using a ubiquitously expressing mitochondria targeted EYFP reporter (sqh-mito::EYFP). In parallel, ISCs can be identified by their typical smaller nuclei or by using esg-Gal4-driven RFP (Surprisingly, there were no RFP expressing cells with smaller nucleus found upon WPP stage. We speculated that the expression of escargot could be so low that esg+ cells were not faithfully labeled. It was also puzzling that RFP positive puncta was found in ECs with larger nucleus. It may be due to uncharacterized distribution/turnover pattern of mRFP. However, the exact reasons remained unknown. Fig5G). Indeed, ISCs had most mitochondria remained from early-third-instar larvae stage to late-third-instar larvae stage (Fig5E-F). At the WPP stage, mitochondria were retained in ISCs while removed in ECs (identified by their larger nuclei) (Fig5G). At the A-WPP stage, only mitochondria-bearing ISCs were present, while ECs were removed completely, as was consistent with our previous results (Fig5H, Fig1E, Fig4E). These results indicated that ISCs with barely detectable PINK1 protein could avoid clearance of their mitochondria and be exempted from programmed cell death during metamorphosis. Taken together, we have disclosed a cell-type specific mitochondrial clearance, which spatiotemporally correlates with the protein level of PINK1 in this specific developmental process. PINK1 is required in the timely clearance of mitochondria in dying midguts To determine whether PINK1 is essential for the cell-type specific mitochondrial clearance in midgut metamorphosis, we employed mosaic analysis. Firstly, clone cells with PINK1 null mutation generated by both negatively- and positively-labeling strategies had significant retention of EYFP-labeled mitochondria when compared with neighboring control cells (Fig6A, Fig6B), concluding that PINK1 was required for the timely clearance of mitochondria in ECs. Consistently, PINK1 RNAi clone cells had significantly more mitochondria than their neighboring cells (Fig6C). On the other hand, clone cells with dPINK1 or human PINK1 (hPINK1) overexpression exhibited pronounced mitochondrial clearance (Fig6D and data not shown). To determine the function of kinase activity of PINK1, we overexpressed human kinase-dead PINK1 (hPINK1-KD) in RFP-marked clone cells. Surprisingly, the clones with hPINK1-KD overexpression showed blocked mitophagy with more residual mitochondria and had larger cell size than neighboring cells (Fig6E), implying a dominant negative effects imposed by kinase-dead hPINK1 proteins. However, the clone cells overexpressing other hPINK1 pathogenic alleles with mutations in the kinase domain (G426D or L464P), displayed similar extent of mitochondria clearance as neighboring cells (Fig6F, 6G), implicating residual kinase activity of these two alleles. Although PINK1 deficiency or hPINK1-KD overexpression could cause suppression of mitochondria clearance at WPP stage, the suppression turned out to be only temporary since mitochondria were eventually cleared in Pink1 deficient or hPINK1-KD clone cells when observed with a longer

interval after clone induction (Fig6H and data not shown). In addition, complete mitochondrial clearance in ECs could be accomplished even in the Pink1 null mutant flies (Fig6I), although a considerable delay was observed in the elimination of ECs in Pink1 null flies compared to the control flies (Fig6J, 6K, 6h APF vs 4h APF). Although these findings suggested that there might be redundant or alternative fail-safe mechanisms to complete mitophagy in the absence of PINK1, PINK1 appears both sufficient and necessary for the timely clearance of mitochondria in Drosophila midgut metamorphosis and its kinase activity is indispensable. PINK1 regulates the autophagy and cell size reduction occurring in midgut metamorphosis The contribution of PINK1 in mitochondrial clearance prompted us to examine whether PINK1 also involves in the global autophagy. Indeed, knockdown of PINK1 markedly attenuated the formation of mCherry-Atg8a puncta when compared with control cells (Fig7A). In addition, the GFP-marked clone cells overexpressing hPINK1 contained more mCherry-atg8a puncta (Fig7B). Knockdown of PINK1 significantly blocked the shrinking of midgut cells (Fig7D), while size of cells with hPINK1 overexpression showed no significant difference compared with neighboring cells (Fig7E). Also, consistent with the dominant negative impact mentioned above, the clones with ectopical expression of hPINK1-KD showed a decrease in mCherry-Atg8a puncta (Fig7C). Taken together, PINK1 might involve in mitophagy, global autophagy and programmed size reduction. Moreover, the kinase activity of PINK1 is essential for its function. Forced expression of PINK1 in ISCs elicits mitophagy but does not lead to cellular depletion during metamorphosis Despite that PINK1 was hardly detected in larval ISCs, we wondered whether mitophagy in ISCs could still be triggered by PINK1 overexpression. To address this possibility and whether the activity of PINK1 is coordinated in ISCs, we investigated the impact of PINK1 overexpression on mitophagy in ISCs by using esg-Gal4-driven PINK1 expression. Consistent with our previous results, ISCs possessed abundant mitochondria from early-third-instar larvae stage to A-WPP stage in control flies (Fig8A-D, left panel). With forced ISCs-specific PINK1 overexpression, GFP-positive mitochondria could hardly be observed in ISCs from late-third-instar larvae stage to A-WPP stage (Fig8B-8D, right panel), while a large number of mitochondria were still present at early-third-instar larvae stage (Fig8A, right panel). It is particularly noteworthy that ISCs survived these stages despite the remarkable loss of mitochondrial content. Intriguingly enough and due to the same possibilities we considered for Fig5G, upon late-third-instar larvae stage, mito::GFP signal that was presumably produced originally in ISCs driven by esg-Gal4 could be detected in ECs in the PINK1 overexpression genetic background (Fig8B& 8C, right panels). Further investigations might disclose the underlying mechanisms. Thus, our results indicated that overexpression of PINK1 is sufficient to induce mitochondrial clearance in ISCs but not result in cellular elimination during the transitional period. Interestingly, with PINK1 overexpression, GFP labeled mitochondria still (re)-appeared in ISCs at adult stage, similar to control flies (Fig8E), as suggested that mitochondria in ISCs might not be completely removed during metamorphosis and could be reconstituted in ISCs before eclosion under this manipulation. Altogether, we concluded that forced expression of PINK1 causes mitophagy in larval but not adult ISCs, and cannot lead to ISCs elimination in the pupae midguts.

PARKIN is required for programmed cell size reduction and timely clearance of mitochondria To examine whether PINK1/PARKIN pathway is conserved in midgut metamorphosis, we then carried on the research of PARKIN by genetic manipulation in this system. We found that knockdown of parkin prevented the timely clearance of mitochondria in midgut (Fig9A). Moreover, the clone cells overexpressing dPARKIN or human PARKIN (hPARKIN), which is functional equivalent of dPARKIN, had less residual mitochondria compared to control cells (Fig9B and data not shown). Furthermore, the mCherry-Atg8a puncta in parkin knockdown clone cells significantly reduced (Fig9C), while clones with hPARKIN overexpression possessed equivalent puncta with the neighboring cells (Fig9D). In addition, the clone cells with parkin knockdown were larger in size than the control cells; on the other hand, the cellular size of clone cells with hPARKIN overexpression did not alter significantly (Fig9E, 9F). Together with mCherry-Atg8a results, overexpression of hPARKIN appeared insufficient to enhance macroautophagy in midgut removal. This insufficiency is probably due to rather drastic global autophagy occurring during this process (Fig1C), which could not be easily intensified further by more PARKIN in this particular mosaic experimental setup (Fig9D, 9F). Again, mitochondria in the parkin knockdown clone cells were removed eventually after puparium formation (data not shown), similar to what happened in the case of PINK1 deficiency. In conclusion, these data suggested that PARKIN mediates the timely mitochondrial clearance and the autophagy-dependent shrinkage of midgut cells. PINK1 functions upstream of PARKIN in a convergent pathway to regulate mitophagy in Drosophila midgut metamorphosis To further determine whether PINK1 and PARKIN function in one common pathway in midgut metamorphosis, we used MARCM strategy and found that the RFP-labeling MARCM clone cells with PINK1 null mutation and hPARKIN overexpression showed no significant difference from neighboring control cells, suggesting that hPARKIN overexpression could prevent the suppression of mitochondrial clearance caused by PINK1 null mutation (Fig10A). Not unexpectedly, the clones with both PINK1 RNAi and hPARKIN overexpression exhibited significant removal of mitochondria (Fig10B), suggesting that PINK1 RNAi resulted in only partial but not complete knock-down of PINK1 such that hPARKIN overexpression dictated the phenotype (Fig9B). Furthermore, the mCherry-Atg8a puncta in clone cells with both PINK1 knockdown and hPARKIN overexpression obviously increased compared with the clones with PINK1 knockdown alone (Fig10C, 7A). Furthermore, the mCherry-Atg8a puncta in the clones with overexpression of hPINK1 and hPARKIN together significantly increased (Fig10D). In conclusion, these data demonstrated that PINK1 acts upstream of PARKIN in a convergent pathway to mediate the timely cell-type specific mitochondrial clearance and autophagy in Drosophila midgut metamorphosis. Discussion PINK1 and PARKIN are implicated in a common genetic and cellular pathway involving removal of damaged mitochondria upon depolarization. However, it remains unclear whether PINK1/PARKIN pathway plays the same role in normal developmental process. Here we identified the role of PINK1/PARKIN pathway upon mitophagy in physiological development. We

reported that PINK1/PARKIN functions in a common mitochondrial quality control pathway to regulate mitophagy during larva-pupae metamorphosis in Drosophila midgut, with PINK1 upstream of PARKIN. The Drosophila midgut system has a number of obvious advantages to study mitophagy in vivo. The Drosophila midgut has typical cellular makeup with ISCs adjoining the basement membrane and ECs form the majority of the intestinal epithelial monolayer, interspersed with ee cells (Ohlstein and Spradling, 2006). We focused on ECs that are polyploidy with large nuclei and easy to be monitored. Besides, mitochondria can be conveniently observed in great details in ECs, which allow us to study mitophagy at the single-cell level. More refined analysis can be done at the cellular organelle level if super high-resolution imaging technologies are employed. Our results showed that different types of intestine cells have distinct fates during metamorphosis. With their mitochondria relatively intact, the ISCs are preserved, presumably to form a new adult midgut epithelium upon puparium formation. In contrast, with acute mitophagy the mitochondria of ECs are removed completely, the process is accompanied by global autophagy within a short time window from late-third instar larvae stage to WPP stage, and all ECs are completely eliminated at the end. This mitochondrial quantitative control in distinct cell types coincides with the expression pattern of PINK1. The stereotypic developmental programs provide us a delicate system to conduct research on mitophagy and macroautophagy. In fact, the rapid removal of larvae midgut is a critical developmental process directed by molting hormone ecdysone during Drosophila metamorphosis (White et al., 1997). Larvae midgut degradation occurs with classic hallmarks of autophagy (Lee et al., 2002), which is reported to be essential for midgut PCD (Denton et al., 2009). It has been reported that in Drosophila ecdysone triggers autophagy only when juvenile hormone concentration is low (Riddiford, 1993). Interestingly, recent work in mouse liver identified the fed-state sensing nuclear receptor farnesoid X receptor (FXR) and the fasting transcriptional activator cAMP response element-binding protein (CREB) axis as a crucial physiological switch coordinately regulating the hepatic autophagy gene network (Lee et al., 2014). We then speculate that ecdysone, a Drosophila FXR homologue, may also function as the master physiological regulator upon midgut autophagy during metamorphosis. Mitophagy mediates clearance of damaged mitochondria, and is also involved in normal development process. The removal of mitochondria is vitally important for erythrocyte maturation, and Nix is required specifically for elimination mitochondria of reticulocyte by incorporating mitochondria with autophagosomes (Schweers et al., 2007). It was revealed that the BNIP3L/NIX-involving mitochondrial dynamic regulatory and degradation pathways can function to maintain mitochondrial quantity in the lens epithelium and to eliminate mitochondria in maturing lens fiber cells (Chauss et al., 2014). Furthermore, the clearance of mitochondria represents a developmental process that contributes to the survival of mature T cells (Pua et al., 2009). Considering that mitochondria need to exit their arena at the right time, the precise control of gene expression and protein turnover must be carried out. However, only damaged or superfluous mitochondria can be ‘labeled’ and then removed. It might be intriguing to figure out how the mitochondria of ECs are targeted and then recruit PINK1. Our results suggested that the decrease of mitochondrial membrane potential was followed by the inchoate aggregation of PINK1 in early-third-instar larvae stage. Previous studies have shown that mitochondrial depolarization can be induced by apoptosis, and ecdysone can trigger apoptosis in Drosophila (Ress et al., 2000). Given the possible connection between autophagy and apoptosis (Schwarten et al., 2014), we tend to think that ecdysone may act as a key physiological regulator to mediate cell

apoptosis and mitochondrial permeability transition that triggers mitophagy during midgut PCD. However, whether mitophagy is subject to direct transcriptional regulation by ecdysone needs further investigations. And how is PINK1 regulated when ecdysone is negligible, such as in the adult stages? Furthermore, our data showed that the specific expression of PINK1 seems to be crucial for the cell-type specific degradation in the midgut. It is very likely PINK1 could be coordinated by the same control of ecdysone as well. Then does the control occur at the transcriptional level or post-translational level? How can ISCs act against the control and manage to keep PINK1 level nearly undetectable? Through overexpressing PINK1 in ISCs and monitoring the mitochondria quantity, our results showed that forced expression of PINK1 is able to trigger mitophagy in ISCs during larvae-to-pupae stage despite the mitophagy activity in normal ISCs is low. This finding indirectly demonstrated our guess that the level of PINK1 could be actively regulated by ISCs that might act against the control of ecdysone during transition period. Moreover, overexpression of PINK1 did not result in ISCs depletion during larvae-to-pupae stage, which also implied that mitophagy may be separated from global macrophagy. Furthermore, mitochondria of ISCs appear to be replenished presumably to meet the energy demands of tissue remodeling before eclosion, indicating that mitophagy may have different regulatory mechanisms for ISCs in distinct developmental stages and the processes of mitochondrial biogenesis and mitophagy have to be well coordinated. Recent study reported that an Atg5-independent autophagic process mediates mitochondrial clearance, leading to a metabolic switch that promotes iPSC reprogramming (Ma et al., 2015), as bridges stem cell induction with mitochondrial quality control. Furthermore, daughter cells that receive fewer old mitochondria maintain stem cell traits after asymmetric cell division (Katajisto et al., 2015). The aged mitochondria can cause cumulative damage to exhaust stem cells and compromise tissue function eventually, therefore stem cells may get rid of old or inferior mitochondria during division to slow the accumulation of such damage. Combined with our results, it may imply that the reason why mitochondria of ISCs could survive and there is barely detectable PINK1 is that ISCs possess or preserve more young and healthy mitochondria during midgut metamorphosis. More future work is required to elucidate the underlying mechanisms of the cell-type specific mitophagy during midgut metamorphosis. To date, our understanding of the connection between macroautophagy and mitophagy is limited. A priority for mitophagy research is to identify putative effectors that selectively target mitochondria to degradation. As reported, mitophagy-specific effectors such as ATG32 account for the selectivity of mitophagy in yeast (Kanki et al., 2009; Okamoto et al., 2009). The homologs of ATG32 have not been identified in higher eukaryotes (Kanki et al., 2010), while Nix and PARKIN could function as mitophagy-specific proteins in mammals (Schweers et al., 2007; Youle and Narendra, 2011). Nix contains a conserved LC3-binding motif known as LIR, it may function in a manner similar to the yeast ATG32/ATG11 pair (Novak and Dikic, 2011). Recently, NDP52 and optineurin was identified as essential receptors for PINK1- and PARKIN-mediated mitophagy (Lazarou et al., 2015). Therefore, there could be unidentified mitophagy-specific effectors for mitochondrial clearance during Drosophila midgut metamorphosis. Atg5 is the essential component for the canonical autophagy (Komatsu et al., 2005; Kuma et al., 2004). In our results, Atg5 and its autophagic partner Atg12 are essential for mitophagy during midgut shrinkage. In addition, as shown by mCherry-Atg8 results, Atg5 also affects the subsequent global autophagy process. As a selective process, the very presence of mitophagic adapters is undoubted. In yeast,

Atg32 acts as a receptor with Atg11 for recruiting mitochondria to the vacuole when mitophagy is induced (Kanki et al., 2009). The results of our data showed that the other organelles such as nucleus are still present while mitochondria are almost removed completely (Fig1E, Fig6G, 6H). Therefore, unique pathways that mediate mitophagy may differ from those mediating macroautophagy. Hence, we speculate that Atg5 may interact with unidentified mitophagic receptor/adapter to specifically target mitochondria for degradation. Notably, we noticed that those clone cells with overexpression hPINK1 KD showed significant retention of mitochondria compared with neighboring cells, which indicated that the kinase-dead PINK1 mutant proteins are able to display a dominant negative effect. In fact, mutations in PINK1 were believed to result in an autosomal recessive form of PD (Valente et al., 2004; Valente et al., 2001), as is partially supported by our results of hPINK1 G426D or L464P experiments and previous reports (Song et al., 2013). Therefore the dominant negative impact we observed at the cellular level is unexpected and might argue for a PINK1 mutation in the heterozygous state might increase the risk for PD (Abou-Sleiman et al., 2006). The kinase dead form of hPINK1 might act as a scaffold competing for binding substrates. However, the mitophagy in ‘kinase-dead’ clone cells was just delayed rather than prevented, all clone cells were removed finally; this feature applied to PINK1 or parkin knockdown case as well. Therefore, we tend to believe that PINK1/PARKIN pathway may just act as only one mechanism for removal of mitochondria in midgut metamorphosis, and there must be other complementary or fail-safe pathways to ensure mitophagy in the process. Previous reports suggested that PINK1 and PARKIN function in the same pathway in regulating mitochondrial function, with PINK1 functioning upstream of PARKIN in Drosophila and mammalian (Clark et al., 2006; Whitworth and Pallanck, 2009). In our results, PINK1 and PARKIN can mediate the cell-type specific mitophagy in one common pathway during Drosophila midgut metamorphosis. Moreover, besides the well-known cooperative PINK1/ PARKIN pathway, either PINK1 or PARKIN can function independently. PINK1 recruits autophagy receptors to mitochondria to activate mitophagy directly, independently of PARKIN (Lazarou et al., 2015). On the other hand, mitochondrial recruitment of PARKIN and activation of mitophagy is PINK1-dispensable in cardiac myocytes (Kubli et al., 2015). Taken together, we believe that the pathways other than the common one may act as the backup ‘scavenger’ when PINK1 or PARKIN is blocked. There may be other mediators functioning in PINK1/PARKIN pathway or there could even be feedforward or feedback loops in the mitochondrial quality and quantity control. We are now continuing with this line of research.

Methods and Materials Drosophila Stocks and nomenclature Drosophila stocks were maintained at 21-23 ℃ , unless mentioned. The following temperature-sensitive fly stocks were used: hsFLP; actin << Gal4, UAS-myrRFP and hsFLP; actin << Gal4, UAS-GFP, these FLP stocks were gifts from Tsinghua University Drosophila Resource Center. hsFLP; actin << Gal4, UAS-myrRFP; sqh-mito::EYFP and hsFLP, FRT19A, tubGal80; actin << CD2 << GAL4, UAS-myrRFP; sqh-mito::EYFP and hsFLP, FRT19A, Ubi-mRFP; sqh-mito::EYFP were produced by standard crosses. hsFLP; pmCherry-Atg8a; act << Gal4, UAS-nlsGFP was a gift from Eric Baehrecke (Chang et al., 2013). “<<” stands for tandem FRTs. FRA19A, Pink1 [B9]/ FM7 was produced by standard crosses. The y[1], M(Act5c-Cas9, [w+]) was a gift from Fillip Port and Simon Bullock. Transgenic strain UAS-hparkin and UAS-hPINK1 were generated previously (Yang et al., 2006). UAS-hPINK1-3KD (or as hPINK1-KD) was obtained by mutating key residues (K219, D362, and D384) in kinase domain of hPINK1 to alanine residues as described previously (Beilina et al., 2005), and generated by standard microinjection protocol in our lab using w1118 stock. UAS-hPINK1(G426D) and pUAS-hPINK1(L464P) fly strains were gifts from J. Chuang (Song et al., 2013). The other fly strains were obtained from Bloomington Drosophila Stock Center and Vienna Drosophila Research Center. Further information on genes and symbols can be found in Flybase (http://flybase.bio.indiana.edu). The terms ‘intestinal stem cell (ISC)’ and ‘enterocyte (EC)’ are proposed to denote Drosophila midgut stem cells and their daughters, respectively (in this paper). Embryo injections Preblastoderm embryos were de-chorinated and injected by standard protocol with a FemtoJet apparatus (Eppendorf). Injected embryos were kept for 2 days at 18℃ and the hatched larvae were collected and grown at 25℃. y[1], M(Act5C-Cas9 )ZH-2A, w[1118] embryos were injected with both sgRNA (100ng/μL) and pBS-donor plasmid (500ng/μL). All injection mixtures were prepared in mili-Q water. Screening All the germline transmissions were screened by PCR. Adults developed from injected embryos were out-crossed to balancer flies. Each single first generation offspring were backcrossed to the balancer stock. The genomic DNA of individual F1 progeny were isolated and screened by PCR after the second generation progeny emerged. Genomic knock-in plasmid construction For sgRNA, pU6-BbsI-chiRNA (Addgene plasmid 45946) plasmid was used for expression of chiRNA in the Drosophila. Targeting chiRNAs are easily cloned by annealed oligos into the pU6-BbsI-chiRNA plasmid via the BbsI restriction sites. Target sites were designed to direct Cas9-mediated cleavage specifically. CRISPR optimal target finder (http://tools.flycrispr.molbio.wisc.edu/targetFinder/) was used to choose the target site for PINK1, which does not have highly homologous sites elsewhere in the genome. For donor DNA, candidate DNA sequence fragments were amplified by overlap PCR and connected to pBlueScript II vector sequence fragment, which was also amplified by PCR using Infusion kit (TaKaRa). To avoid the cleavage of donor plasmid, the PAM site was replaced with synonymous codon. Sequences of the plasmids can be acquired upon request. Quantification of cell size The cell size of all control and mutant midgut was quantified as previously described (Chang

et al., 2013). Assessment of mitochondrial membrane potential Mitochondrial membrane potential was assessed in Drosophila midgut with the probe JC-1 (Invitrogen). JC-1 accumulates within the intact mitochondria to form multimer J-aggregates that result in a shift of fluorescence from green (530nm) to red (590nm). The potential-sensitive color shift is due to concentration-dependent formation of red fluorescent J-aggregates. Intact guts from early-3-instar larvae and late-3-instar larvae were dissected, loaded with 3μg/ml of JC-1 for 30 minutes at 37℃. The guts were washed with phosphate-buffered saline, and mitochondrial JC-1 was monitored by confocal microscopy (a Leica SP5 confocal microscope). Mitochondrial depolarization is indicated by a decrease in the red/green fluorescence intensity ratio or an increase in the green/red ratio. Immunoblotting Midguts were dissected at various times in PBS and lysed in 100μl SDS lysis buffer (20 mM Tris/HCl at pH 7.6, 150 mM NaCl, 5mM EDTA, 10% glycerol, 1% SDS and 1mM PMSF) for 30 min on ice. Guts debris was pelleted at 16000g and 4℃ for 10 min, then eluted in 5×SDS loading buffer and boiled for 5 min at 95℃. Samples were separated on 7% SDS-PAGE and transferred onto a PVDF membrane (BIO-RAD). Membranes were blocked in 5% BSA in TBS-T, incubated overnight with mouse anti-DDK(Flag) (1:1000,Sangon Biotech) at 4℃, followed by incubation with HRP-conjugated secondary antibodies for 2h at RT. Proteins were visualized using the Immobilon Western chemiluminescent HRP substrate (Beyotime) on ChemiDocTM XRS+ (BIO-RAD). Three independent biological experiments were performed. Induction of cell clones and imaging To induce corresponding expression in clones of cells, virgin females of hsFLP; mCherry-Atg8a; actin << CD2 << GAL4 (<< is FRT site), UAS-nlsGFP or hsFLP; actin << CD2 << GAL4, UAS-myrRFP; sqh:mitoEYFP or hsFLP, FRT19A, tubGal80; actin << CD2 << GAL4, UAS-myrRFP; sqh:mitoEYFP were collected and crossed to indicated RNAi or FRT-recombined null mutant alleles or transgenic lines under standard crosses. To induce FLP expression, larvae were heat-shocked at 37℃ for 15 min, 30min or 1h, respectively. After heat-shock, these eggs were allowed to continue to grow at 29℃. Pupa were dissected 2 hours later after white pupa, and stained with Hoechst33342 (Beyotime) by standard protocol, and scanned by a Leica SP5 confocal microscope. These clones were identified by the expression of nlsGFP or myrRFP and compared with neighbor cells for mCherry-Atg8a or mito::EYFP signal. For GFP or mCherry-Atg8 imaging, we briefly fixed samples with 4% formaldehyde in PBS and stained DNA with Hoechst32242 as described previously (Chang et al., 2013). For hsFLP, FRT19A, Ubi-mRFP; sqh-mito::EYFP mediated mosaic analysis, clone cells were identified with those lacking RFP or with very little residual RFP signal. Other protocols were the same as above. Imaging processing and analysis All images were prepared using Adobe Photoshop and subjected to identical post-acquisition brightness/contrast effects. Quantification was performed in individual frames after thresholding using the ImageJ software (NIH). Co-localization percentage was calculated using the JACoP plugin in single Z-stack sections of de-convolved images (Martinez-Lopez et al., 2013). Co-localization analysis between mCherry-Atg8a or PINK1-mCherry and mito::EYFP was expressed as Manders' coefficients, which are defined as the proportion of intensity in the green channel (mito::EYFP reporter) that overlapped with that in the red channel (PINK1mCherry

reporter) or the proportion of intensity in the red channel (mCherry-Atg8a reporter) that overlapped with that in green channel (mito::EYFP reporter) (Casari et al., 2014). Quantification of relative mitochondria content was by calculating the mean ratio of EYFP-mitochondria area to total cell area (n=5 intestines with 2-5 cells measured per genotype), and was shown as mean ± s.d. Quantification of relative effective Atg8a autophagic signal was by calculating the mean ratio of mCherry-Atg8a puncta area to total cell area (n=5 intestines with 2-5 cells measured per genotype), and was shown as mean ± s.d. Statistics Statistical analyses were carried out with Prism GraphPad software. Indicated P values were obtained using a two-tailed t-test, and all quantitative data are shown as mean ± standard deviation. No statistical method was used to predetermine sample size. No samples were excluded. The experiments were not randomized. The investigators were not blinded to allocation during the experiments and outcome assessment.

Figure 1 Concomitant autophagy and mitophagy in the process of EC programmed cell size reduction during metamorphosis. (A) Representative differential interference contrast (DIC) microscopy images of ECs (large blue nuclei) from wild-type (Canton-S) fruit flies at the early-third-instar larvae (Early 3rd) stage, late-third-instar larvae (Late 3rd) stage and during puparium formation (white prepupal, WPP) stages. Intestines were stained for DNA (Hoechst33258, blue). (B-E) Representative confocal images of autophagy detected by formation of mCherry-Atg8a punctate spots in midguts dissected at the Early 3rd larvae (B), Late 3rd (C), WPP (D) and after puparium formation (after white prepupal, A-WPP) stages (E) Mitochondria were labeled with ubiquitously expressed mito::EYFP (sqh-mito::EYFP). (F) The mean ratio quantification of EYFP-mitochondria area to total cell area from B-C (n=5 intestines per stage with 2-5 cells measured per intestine) is shown as mean ± s.d. (G) Overlapping of EYFP-mitochondria and mCherry-Atg8a puncta for B-C was expressed as Manders' coefficients (Dunn et al., 2011; Manders et al., 1993), which are defined as the proportion of intensity in the green channel (EYFP reporter) that colocalized with that in the red channel (mCherry reporter). Quantifications are shown as mean ± s.d. *P<0.05; **P<0.01; ***P<0.001. Representative images are shown. Scale bars, 50μm. Genotypes: (a) Canton-S; (b-e) yw; pmCherry-Atg8a/ +; sqh-mito::EYFP/ +.

Figure 2 Role of canonical autophagy machinery in programmed midgut cell size reduction and mitophagy during metamorphosis. (A) Control clone cells labeled with GFP monitored by confocal and DIC microscopy. The dotted lines indicate cell boundary. Note that clone cells had similar size with the neighboring cells. (B) Midguts overexpressing Atg1 RNAi specifically in GFP-marked clones of cells were dissected during puparium formation and analyzed by confocal microscopy. The dotted lines indicate cell boundary. Note that clone cells were larger than the neighboring control cells. (C) Midguts overexpressing Atg12 RNAi specifically in GFP-marked clones of cells were dissected during puparium formation and analyzed by confocal microscopy. The dotted lines indicate cell boundary. Note that clone cells were larger than the neighboring control cells. (D) Quantification (μm2) from B and C, n=5 fruit fly intestines per genotype with 2-5 cells measured per intestine. Quantification is shown as mean ± s.d *P<0.05. (E-H) Midguts overexpressing Atg1 RNAi (F), Atg12 RNAi (G) and Atg12 RNAi + Atg5 RNAi (H) or RFP only (E) specifically in RFP-marked clones of cells were dissected during puparium formation and analyzed by confocal and DIC. Mitochondria were labeled with ubiquitously expressed mito::EYFP. Intestines are stained for DNA (Hoechst, blue). The mean ratio quantification of EYFP-mitochondria area to total cell area from E-H (n=5 intestines per genotype with 2-5 cells measured per intestine) is shown as mean ± s.d. ns, not significant;

**P<0.01; ***P<0.001. Representative images are shown. Scale bars, 20μm. Genotypes: (a) hsFLP; actin << Gal4, UAS-GFP/ +; (b) hsFLP; UAS-Atg1 RNAi/ +; actin << Gal4, UAS-GFP/ +; (c) hsFLP; actin << Gal4, UAS-GFP/ UAS-Atg12 RNAi; (e) hsFLP; act << Gal4, UAS-RFP/ +; sqh-mito::EYFP/ +; (f) hsFLP; act << Gal4, UAS-RFP/ UAS-Atg1 RNAi; sqh-mito::EYFP/ +; (g) hsFLP; act << Gal4, UAS-RFP/ +; sqh-mito::EYFP/ UAS-Atg12 RNAi; (h) hsFLP; act << Gal4, UAS-RFP/ UAS-Atg5 RNAi; sqh-mito::EYFP/ UAS-Atg12 RNAi.

Figure 3 Genomic knock-in editing to establish PINK1-mCherry fly strain. (A) Schematic overview of strategy to generate the PINK1-linker-mCherry-Flag knock-in allele. (B-C) Successful HR of PINK1-linker-mCherry-Flag was confirmed by genomic PCR amplicon and Sanger sequencing. The asterisks indicate the target bands of PINK1-mCherry. Marker: DL2000. (D) Western blot analysis of the endogenous expression of w1118 control and PINK1-mCherry flies with an anti-FLAG antibody, with the expected full-length PINK1-mCherry around 100kd. (E) PINK1-mcherry is functionally equivalent to endogenous PINK1. The males of w1118 (w) or PINK1-mCherry were crossed to Pink1 [B9] females and the percentages of offspring with abnormal wing posture phenotype were scored at 15-day after eclosion for each indicated genotype. Numbers indicated the percentages of abnormal wings for each indicated genotype. The flies were aged at 29℃. Genotypes: (b-c) PINK1-mCherry/ +; (d) w1118; PINK1-mCherry; (e) Pink1 [B9]/ +; Pink1 [B9]/ Y; Pink1 [B9]/ PINK1-mCherry.

Figure 4 Monitoring dynamic PINK1 signal during midgut metamorphosis. (A) The endogenous expression level of PINK1-linker-mCherry-FLAG determined by immunoblotting of Drosophila intestine blotted with an anti-FLAG antibody at different stages, alpha-tubulin as the internal input control. (B-E) Patterns of PINK1-mcherry signal in midgut during metamorphosis at the early 3rd (B), Late 3rd (C), WPP (D) and A-WPP stages (E). Note that basally located ISCs (the inset, cells with smaller nuclei) contained more mitochondria than neighboring ECs (with larger nuclei) at A-WPP stage. Representative confocal images are shown. (F) Co-localization between mCherry-PINK1 puncta and EYFP-mitochondria for C-D was expressed as Manders' coefficients, which are defined as the proportion of intensity in the red channel (mCherry reporter) that co-localized with that in the green channel (EYFP reporter). Quantifications are shown as mean ± s.d. *P<0.05. (G-H) Mitochondrial membrane potential (MMP) were determined by JC-1 staining at the early third instar larval (G) and late third instar larval (H). (I) The mean ratio of JC-1 intensity in green channel to red channel for B and C was shown as mean ± s.d. ***P<0.001. Note that MMP depolarization is indicated by a decrease in the red (590nm)/green (530nm) fluorescence intensity ratio or an increase in the green/red ratio. The emerging mitochondrial enrichment of PINK1 signal appears to correlate with the depolarization of MMP in ECs during midgut metamorphosis. Intestines were stained for DNA (Hoechst, blue). Representative confocal images are shown. Scale bars, 20μm. Genotypes: (a-e) PINK1-mCherry; (g-h) Canton-S.

Figure 5 Mitochondrial retention coincides with low expression of PINK1 in ISCs during midgut metamorphosis. (A-D) Intestines of flies with esg-Gal4, UAS-GFP (ISC/EB marker; green) and mCherry-PINK1 transgenes were dissected at the early 3rd (A), Late 3rd (B), WPP (C) and A-WPP (D) stages and monitored. Note that mCherry signal was nearly undetectable in GFP-marked ISCs, indicating that PINK1 was lowly expressed in ISCs. (E-H) Intestines of flies with esg-Gal4, UAS-RFP, sqh-mito::EYFP were dissected at the early 3rd (E), Late 3rd (F), WPP (G) and A-WPP (H) stages and monitored. Note that ECs (dotted line), probably with residual un-degraded RFP protein, possessed much fewer mitochondria compared to its neighboring ISCs upon WPP stage. Meanwhile, there is no expression of [esg-Gal4]-driven RFP in ISCs at A-WPP stage due to unknown reasons. Intestines were stained for DNA (Hoechst, blue). Representative confocal images are shown. Scale bars, 20μm. Genotypes: (a-d) PINK1-mCherry/ +; esg-Gal4, UAS-GFP/ +; (e-h) esg-Gal4, UAS-RFP/ +; tub-Gal80ts/ sqh-mito::EYFP.

Figure 6 PINK1 is involved in mitophagy in the Drosophila midgut. (A-K) Mitochondria were labeled with ubiquitously expressed mito::EYFP. (A-G) The mean ratio quantification of EYFP-mitochondria area to total cell area (n=5 intestines with 2-5 cells measured per genotype) was shown as mean ± s.d. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ns, not significant. (A) Midguts that contained PINK1 null mutation clone cells (lacking RFP or with little residual RFP) were dissected during puparium formation and monitored. Note that clone cells had more mitochondria than neighboring control cells. (B) Midguts with RFP-marked MARCM cell clones harboring PINK1 null mutation were dissected during puparium formation and monitored. Note that clone cells had more mitochondria than neighboring control cells. (C) Midguts expressing PINK1 RNAi specifically in RFP-marked clones of cells were dissected during puparium formation and monitored. Note that clone cells had more mitochondria than neighboring cells. (D) Midguts overexpressing PINK1 specifically in RFP-marked clones of cells were dissected during puparium formation and monitored. Note that clone cells had fewer mitochondria than neighboring cells. (E) Midguts overexpressing kinase-dead human PINK1 specifically in RFP-marked clones of cells were dissected during puparium formation and monitored. Note that clone cells possessed more mitochondria than neighboring cells. (F) Midgut overexpressing PINK1 with a mutation in the kinase domain (G426D) specifically in RFP-marked clones of cells were dissected during puparium formation and monitored. Note that there were no significant difference of mitochondria quantity between clone cells and neighboring control cells. (G) Midgut overexpressing PINK1 with a mutation in the kinase domain (L464P) specifically in RFP-marked clones of cells were dissected during puparium formation and monitored. Note that there were no significant difference of mitochondria quantity between clone cells and neighboring control cells. (H) Midguts expressing PINK1 RNAi specifically in RFP-marked clones of cells were dissected 4h APF. Note that mitochondria in RFP-marked clone cells had been removed almost completely (I) Representative confocal images of intestines from PINK1 null mutant fruit flies at the 4h after puparium formation (APF). Note that at this stage mitochondria in ECs (with larger nuclei) had been removed almost completely in Pink1 [B9] males, but the nuclei of ECs remained. (J) Representative images of intestines from PINK1 null mutant fruit flies at the 6h APF. Note that ECs have been nearly completely removed while mitochondria-bearing ISCs remained at this stage. (K) Representative images of intestines from control fruit flies at the 4h APF. Note that the phenotype was similar to PINK1

null

mutant

fruit

flies

at

6h

APF.

Intestines

were stained for

DNA

(Hoechst,

blue).

Representative confocal images are shown. Scale bars, 20μm. Genotypes: (a) hsFLP, FRT19A, Ubi-mRFP/ FRA19A, Pink1 [B9]; sqh-mito::EYFP/ +; (b) hsFLP, FRT19A, tubGal80/ FRT19A, Pink1 [B9]; act << Gal4, UAS-RFP/ +; sqh-mito::EYFP/ +; (c) hsFLP; act << Gal4,UAS-RFP/ +; sqh-mito::EYFP/ UAS-dPINK1 RNAi; (d) hsFLP; act << Gal4, UAS-RFP/ +; sqh-mito::EYFP/ UAS-hPINK1; (e) hsFLP; act << Gal4, UAS-RFP/ UAS-hPINK1-KD; sqh-mito::EYFP/ +; (f) hsFLP; act << Gal4, UAS-RFP/ UAS-hPINK1-GD; sqh-mito::EYFP/ +; (g) hsFLP; act << Gal4, UAS-RFP/ UAS-hPINK1-LP; sqh-mito::EYFP/ +; (h) hsFLP; act << Gal4,UAS-RFP/ +; sqh-mito::EYFP/ UAS-dPINK1 RNAi; (i-j) Pink1 [B9]/ Y; sqh-mito::EYFP/ +; (k) +/ Y; sqh-mito::EYFP/ +.

Figure 7 PINK1 is involved in cell size reduction in the Drosophila midgut. (A-C) Midguts overexpressing PINK1 RNAi (A), hPINK1 (B) and hPINK1 KD mutant(C) specifically in GFP-marked clones of cells were dissected during puparium formation with ubiquitously expressed mCherry-Atg8a. The mean ratio quantification of mCherry-Atg8a puncta area to total cell area (n=5 intestines with 2-5 cells measured per genotype) was shown as mean ± s.d. *P<0.05; **P<0.01; ***P<0.001. (D-E) Intestines overexpressing PINK1 RNAi fruit flies (D) and hPINK1 (E) specifically in GFP-marked clones of cells were dissected during puparium formation and analyzed by confocal and DIC microscopy. The dotted lines indicate cell boundary. Cell size quantification (μm2) (n=5

intestines with 2-5 cells measured per genotype) was shown as mean ± s.d. ns, not significant; **P<0.01. Note that PINK1 RNAi clone cells were larger than neighboring control cells. Intestines were stained for DNA (Hoechst, blue). Representative images are shown. Scale bars, 20μm.

Genotypes: (a) hsFLP; pmCherry-Atg8a/ +; act <<

Gal4, UAS-GFP/ UAS-dPINK1 RNAi; (b) hsFLP; pmCherry-Atg8a/ +; act << Gal4, UAS-GFP/ UAS-hPINK1; (c) hsFLP; pmCherry-Atg8a/ UAS-hPINK1-KD; act << Gal4, UAS-GFP/ +; (d) hsFLP; act << Gal4, UAS-GFP/ UAS-dPINK1 RNAi; (e) hsFLP; act << Gal4, UAS-GFP/ UAS-hPINK1.

Figure 8 Forced PINK1 overexpression is sufficient to induce mitophagy in larval ISCs but not adult ISCs, and cannot lead to global ISCs elimination during metamorphosis. (A-E) Intestines of flies with esg-Gal4, UAS-mito::GFP (left panel) and esg-Gal4, UAS-mito::GFP; UAS-dPINK1 (right panel) were dissected at the early third instar larval (A), late third instar larval (B), puparium formation (C), after puparium formation (D), adult (E) and monitored. Intestines were stained for DNA (Hoechst, blue). Representative confocal images are shown. Scale bars, 20μm. Genotypes: (a-e, left panels) esg-Gal4, UAS-mito::GFP/ +; +/+. (a-e, right panels) esg-Gal4, UAS-mito::GFP/ +; UAS-dPINK1/ +.

Figure 9 PARKIN is involved in mitophagy and programmed cell size reduction of ECs. (A-B) Midguts overexpressing parkin RNAi (A) and hPARKIN (B) specifically in RFP-marked clones of cells were dissected during puparium formation. Mitochondria were labeled with ubiquitously expressed mito::EYFP. The mean ratio quantification of EYFP-mitochondria area to total cell area (n=5 intestines with 2-5 cells measured per genotype) was shown as mean ± s.d. ****P<0.0001. (C-D) Midguts overexpressing parkin RNAi (C) and hPARKIN (D) specifically in GFP-marked clones of cells were dissected during puparium formation, with ubiquitously expressed mCherry-Atg8a. The mean ratio quantification of mCherry-Atg8a puncta area to total cell area (n=5 intestines with 2-5 cells measured per genotype) was shown as mean ± s.d. ns, not significant; ***P<0.001. (E-F) Midguts overexpressing parkin RNAi (E) and hPARKIN (F) specifically in GFP-marked clones of cells were dissected and analyzed by confocal and DIC microscopy during puparium formation. The dotted lines indicate cell boundary. Note that parkin RNAi cells were larger than neighboring control cells. Cell size quantification (n=5 intestines with 2-5 cells measured per genotype) was shown as mean ± s.d. ns, not significant; **P<0.01. Intestines were stained for DNA (Hoechst, blue). Representative images are shown. Scale bars, 20μm. Genotypes: (a) hsFLP; act << Gal4, UAS-RFP/ +; sqh-mito::EYFP/ UAS-dParkin RNAi; (b) hsFLP; act << Gal4, UAS-RFP/ UAS-hParkin; sqh-mito::EYFP/ +; (c) hsFLP; pmCherry-Atg8a/ +; act << Gal4, UAS-GFP/ UAS-dParkin RNAi; (d) hsFLP; pmCherry-Atg8a/ UAS-hParkin; act << Gal4, UAS-GFP/ +; (e) hsFLP; act << Gal4, UAS-GFP/ UAS-dParkin RNAi; (f) hsFLP; UAS-hParkin/ +; act << Gal4, UAS-GFP/ +.

Figure 10 PINK1 functions upstream of PARKIN in a convergent pathway to regulate mitophagy. (A-B) Mitochondria were labeled with ubiquitously expressed mito::EYFP. The mean ratio quantification of EYFP-mitochondria area to total cell area (n=5 intestines with 2-5 cells measured per genotype) was shown as mean ± s.d. ns, not significant; **P<0.01 (A) Midguts with RFP-marked MARCM cell clones containing PINK1 null mutation and hparkin overexpression were dissected during puparium formation and monitored. (B) Midguts overexpressing PINK1 RNAi and hparkin specifically in RFP-marked clones of cells were dissected during puparium formation and monitored. Note that clone cells had less mitochondrion than neighboring normal cells. (C-D) mCherry-Atg8a was ubiquitously expressed. The mean ratio quantification of mCherry-Atg8a puncta area to total cell area (n=5 intestines with 2-5 cells measured per genotype) was shown as mean ± s.d. ***P<0.001, **P<0.01. (C) Midguts overexpressing PINK1 RNAi and hparkin specifically in GFP-marked clones of cells were dissected during puparium formation and monitored. (D) Midguts overexpressing hPINK1 and hparkin specifically

in GFP-marked clones of cells were dissected during puparium formation and monitored. Intestines were stained for DNA (Hoechst, blue). Representative confocal images are shown. Scale bars, 20μm. Genotypes: (a) hsFLP, FRT19A, tubGal80/ FRT19A, Pink1 [B9]; act << Gal4, UAS-RFP/ UAS-hParkin; sqh-mito::EYFP/ +; (b) hsFLP; act << Gal4, UAS-RFP/ UAS-hParkin; sqh-mito::EYFP/ UAS-dPINK1 RNAi; (c) hsFLP; pmCherry-Atg8a/ UAS-hParkin; act << Gal4, UAS-GFP/ UAS-dPINK1 RNAi; (d) hsFLP; pmCherry-Atg8a/ UAS-hParkin; act << Gal4, UAS-GFP/ UAS-hPINK1.

Author Contributions Y.Y. conceived the project. Y.Y. and F.C. supervised the project. Y.L., J.L., F.C., Q.W., S.X., H.Y., Y.Z. performed fly genetics, immunohistochemistry and imaging experiments. M.Z., S.Y., X.Z. performed gDNAs, donor DNA construction, microinjection and KI characterization. J.L., K.C., X.Z. performed p-element constructions and microinjections. J.L., M.Z. performed western blotting. J.L., F.C. and Y.Y. wrote the manuscript, with contributions from other authors. Author Information Reprints and permissions information is available at https://www.elsevier.com/permissions. The authors declare no competing financial interests. Readers are welcome to comment on the online version of the paper. Correspondence and requests for materials should be addressed to Yufeng Yang ([email protected]).

Acknowledgements We thank Bingwei Lu, Yu Cai for critical comments on the manuscript. We thank Eric Baehrecke for helpful discussions and sharing key fly strains. We thank Fillip Port and Simon Bullock for providing act5C-cas9 fly strain. We thank J. Chuang for sharing UAS-hPINK1 (G426D) and pUAS-hPINK1 (L464P) fly strains. We thank Core Facility for Fruit Flies at the Institute of Biochemistry and Cell Biology/ Shanghai Institutes for Biological Science/ CAS and Tsinghua Drosophila Resource Center for providing fly stocks. This work was supported in part by the National Natural Science Foundation of China (No. 31071296), National Key Basic Research Program of China (973 project, No.2015CB352006), the National Natural Science Foundation of China (No. 31070877), Industry-Academia-Research project of Fujian, China. (No. 2010Y4006) and the National Natural Science Foundation of China (No.61335011).

References Abou-Sleiman, P.M., Muqit, M.M., McDonald, N.Q., Yang, Y.X., Gandhi, S., Healy, D.G., Harvey, K., Harvey, R.J., Deas, E., Bhatia, K., Quinn, N., Lees, A., Latchman, D.S., Wood, N.W., 2006. A heterozygous effect for PINK1 mutations in Parkinson's disease? Ann Neurol 60, 414-419. Baehrecke, E.H., 2000. Steroid regulation of programmed cell death during Drosophila development. Cell Death Differ 7, 1057-1062. Beilina, A., Van Der Brug, M., Ahmad, R., Kesavapany, S., Miller, D.W., Petsko, G.A., Cookson, M.R., 2005. Mutations in PTEN-induced putative kinase 1 associated with recessive parkinsonism have differential effects on protein stability. Proc Natl Acad Sci U S A 102, 5703-5708. Bender, M., Imam, F.B., Talbot, W.S., Ganetzky, B., Hogness., D.S., 1997. Drosophila Ecdysone Receptor Mutations Reveal Functional Differences among Receptor Isoforms. Cell 91, 777-788. Boya, P., Reggiori, F., Codogno, P., 2013. Emerging regulation and functions of autophagy. Nat Cell Biol 15, 713-720. Casari, A., Schiavone, M., Facchinello, N., Vettori, A., Meyer, D., Tiso, N., Moro, E., Argenton, F., 2014. A Smad3 transgenic reporter reveals TGF-beta control of zebrafish spinal cord development. Dev Biol 396, 81-93. Chang, T.K., Shravage, B.V., Hayes, S.D., Powers, C.M., Simin, R.T., Wade Harper, J., Baehrecke, E.H., 2013. Uba1 functions in Atg7- and Atg3-independent autophagy. Nat Cell Biol 15, 1067-1078. Chauss, D., Basu, S., Rajakaruna, S., Ma, Z., Gau, V., Anastas, S., Brennan, L.A., Hejtmancik, J.F., Menko, A.S., Kantorow, M., 2014. Differentiation State-Specific Mitochondrial Dynamic Regulatory Networks Are Revealed by Global Transcriptional Analysis of the Developing Chicken Lens. G3: Genes|Genomes|Genetics 4, 1515-1527. Chen, D., Gao, F., Li, B., Wang, H., Xu, Y., Zhu, C., Wang, G., 2010. Parkin mono-ubiquitinates Bcl-2 and regulates autophagy. J Biol Chem 285, 38214-38223. Chen, Y., Dorn, G., 2013. PINK1-phosphorylated mitofusin 2 is a Parkin receptor for culling damaged mitochondria. Science 340, 471-475. Clark, I.E., Dodson, M.W., Jiang, C., Cao, J.H., Huh, J.R., Seol, J.H., Yoo, S.J., Hay, B.A., Guo, M., 2006. Drosophila pink1 is required for mitochondrial function and interacts genetically with parkin. Nature 441, 1162-1166. Denton, D., Shravage, B., Simin, R., Mills, K., Berry, D.L., Baehrecke, E.H., Kumar, S., 2009. Autophagy, not apoptosis, is essential for midgut cell death in Drosophila. Curr Biol 19, 1741-1746. Dunn, K.W., Kamocka, M.M., McDonald, J.H., 2011. A practical guide to evaluating colocalization in biological microscopy. Am J Physiol Cell Physiol 300, C723-742. Green, D.R., 1998. Apoptotic pathways: the roads to ruin. Cell 94, 695-698. Greene, J.C., Whitworth, A.J., Kuo, I., Andrews, L.A., Feany, M.B., Pallanck, L.J., 2003. Mitochondrial pathology and apoptotic muscle degeneration in Drosophila parkin mutants. Proceedings of the National Academy of Sciences 100, 4078-4083. Guo, Z., Ohlstein, B., 2015. Stem cell regulation. Bidirectional Notch signaling regulates Drosophila intestinal stem cell multipotency. Science 350. Hajnoczky, G., Csordas, G., Das, S., Garcia-Perez, C., Saotome, M., Sinha Roy, S., Yi, M., 2006. Mitochondrial calcium signalling and cell death: approaches for assessing the role of mitochondrial Ca2+ uptake in apoptosis. Cell Calcium 40, 553-560.

Iguchi, M., Kujuro, Y., Okatsu, K., Koyano, F., Kosako, H., Kimura, M., Suzuki, N., Uchiyama, S., Tanaka, K., Matsuda, N., 2013. Parkin-catalyzed ubiquitin-ester transfer is triggered by PINK1-dependent phosphorylation. J Biol Chem 288, 22019-22032. Jiang, C., Baehrecke, E.H., Thummel, C.S., 1997. Steroid regulated programmed cell death during Drosophila metamorphosis. Development 124, 4673-4683. Jiang, C., Baehrecke, E. H., and Thummel, C. S, 1997. Steroid regulated programmed cell death during Drosophila metamorphosis. Kane, L.A., Lazarou, M., Fogel, A.I., Li, Y., Yamano, K., Sarraf, S.A., Banerjee, S., Youle, R.J., 2014. PINK1 phosphorylates ubiquitin to activate Parkin E3 ubiquitin ligase activity. The Journal of Cell Biology 205, 143-153. Kanki, T., Klionsky, D.J., 2008. Mitophagy in yeast occurs through a selective mechanism. J Biol Chem 283, 32386-32393. Kanki, T., Wang, K., Cao, Y., Baba, M., Klionsky, D.J., 2009. Atg32 is a mitochondrial protein that confers selectivity during mitophagy. Dev Cell 17, 98-109. Kanki, T., Wang, K., Klionsky, D.J., 2010. A genomic screen for yeast mutants defective in mitophagy. Autophagy 6, 278-280. Katajisto, P., Dohla, J., Chaffer, C.L., Pentinmikko, N., Marjanovic, N., Iqbal, S., Zoncu, R., Chen, W., Weinberg, R.A., Sabatini, D.M., 2015. Stem cells. Asymmetric apportioning of aged mitochondria between daughter cells is required for stemness. Science 348, 340-343. Komatsu, M., Waguri, S., Ueno, T., Iwata, J., Murata, S., Tanida, I., Ezaki, J., Mizushima, N., Ohsumi, Y., Uchiyama, Y., Kominami, E., Tanaka, K., Chiba, T., 2005. Impairment of starvation-induced and constitutive autophagy in Atg7-deficient mice. J Cell Biol 169, 425-434. Kondapalli, C., Kazlauskaite, A., Zhang, N., Woodroof, H.I., Campbell, D.G., Gourlay, R., Burchell, L., Walden, H., Macartney, T.J., Deak, M., Knebel, A., Alessi, D.R., Muqit, M.M.K., 2012. PINK1 is activated by mitochondrial membrane potential depolarization and stimulates Parkin E3 ligase activity by phosphorylating Serine 65. Open Biology 2, 120080-120080. Kubli, D.A., Cortez, M.Q., Moyzis, A.G., Najor, R.H., Lee, Y., Gustafsson, A.B., 2015. PINK1 Is Dispensable for Mitochondrial Recruitment of Parkin and Activation of Mitophagy in Cardiac Myocytes. PLoS One 10, e0130707. Kuma, A., Hatano, M., Matsui, M., Yamamoto, A., Nakaya, H., Yoshimori, T., Ohsumi, Y., Tokuhisa, T., Mizushima, N., 2004. The role of autophagy during the early neonatal starvation period. Nature 432, 1032-1036. Kumar, S., Cakouros, D., 2004. Transcriptional control of the core cell-death machinery. Trends Biochem Sci 29, 193-199. Lazarou, M., Sliter, D.A., Kane, L.A., Sarraf, S.A., Wang, C., Burman, J.L., Sideris, D.P., Fogel, A.I., Youle, R.J., 2015. The ubiquitin kinase PINK1 recruits autophagy receptors to induce mitophagy. Nature 524, 309-314. Lee, C.Y., Cooksey, B.A., Baehrecke, E.H., 2002. Steroid regulation of midgut cell death during Drosophila development. Dev Biol 250, 101-111. Lee, J.M., Wagner, M., Xiao, R., Kim, K.H., Feng, D., Lazar, M.A., Moore, D.D., 2014. Nutrient-sensing nuclear receptors coordinate autophagy. Nature 516, 112-115. Ma, T., Li, J., Xu, Y., Yu, C., Xu, T., Wang, H., Liu, K., Cao, N., Nie, B.M., Zhu, S.Y., Xu, S., Li, K., Wei, W.G., Wu, Y., Guan, K.L., Ding, S., 2015. Atg5-independent autophagy regulates mitochondrial clearance and is essential for iPSC reprogramming. Nat Cell Biol 17, 1379-1387.

Mai, S., Muster, B., Bereiter-Hahn, J., Jendrach, M., 2012. Autophagy proteins LC3B, ATG5 and ATG12 participate in quality control after mitochondrial damage and influence lifespan. Autophagy 8, 47-62. Manders, E.M.M., Verbeek, F.J., Aten, J.A., 1993. Measurement of co‐localization of objects in dual‐colour confocal images. Journal of Microscopy 169, 375–382. Martinez-Lopez, N., Athonvarangkul, D., Mishall, P., Sahu, S., Singh, R., 2013. Autophagy proteins regulate ERK phosphorylation. Nat Commun 4, 2799. Narendra, D., Walker, J.E., Youle, R., 2012. Mitochondrial quality control mediated by PINK1 and Parkin: links to parkinsonism. Cold Spring Harb Perspect Biol 4. Ney, P.A., 2011. Normal and disordered reticulocyte maturation. Curr Opin Hematol 18, 152-157. Novak, I., Dikic, I., 2011. Autophagy receptors in developmental clearance of mitochondria. Autophagy 7, 301-303. Ohlstein, B., Spradling, A., 2006. The adult Drosophila posterior midgut is maintained by pluripotent stem cells. Nature 439, 470-474. Okamoto, K., Kondo-Okamoto, N., Ohsumi, Y., 2009. Mitochondria-anchored receptor Atg32 mediates degradation of mitochondria via selective autophagy. Dev Cell 17, 87-97. Park, J., Lee, S.B., Lee, S., Kim, Y., Song, S., Kim, S., Bae, E., Kim, J., Shong, M., Kim, J.M., Chung, J., 2006. Mitochondrial dysfunction in Drosophila PINK1 mutants is complemented by parkin. Nature 441, 1157-1161. Pua, H.H., Guo, J., Komatsu, M., He, Y.W., 2009. Autophagy is essential for mitochondrial clearance in mature T lymphocytes. J Immunol 182, 4046-4055. Rawi, S., Louvet-Vallee, S., Djeddi, A., Sachse, M., Culetto, E., Hajjar, C., Boyd, L., Legouis, R., Galy, V., 2011. Postfertilization autophagy of sperm organelles prevents paternal mitochondrial DNA transmission. Science 334, 1144-1147. Ress, C., Holtmann, M., Maas, U., Sofsky, J., Dorn, A., 2000. 20-Hydroxyecdysone-induced differentiation and apoptosis in the Drosophila cell line, l(2)mbn. Tissue Cell 32, 464-477. Riddiford, L.M., 1993. Hormone receptors and the regulation of insect metamorphosis. Receptor 3, 203-209. Santhanam, A., Peng, W.H., Yu, Y.T., Sang, T.K., Chen, G.C., Meng, T.C., 2014. Ecdysone-induced receptor tyrosine phosphatase PTP52F regulates Drosophila midgut histolysis by enhancement of autophagy and apoptosis. Mol Cell Biol 34, 1594-1606. Sato, M., Sato, K., 2011. Degradation of paternal mitochondria by fertilization-triggered autophagy in C. elegans embryos. Science 334, 1141-1144. Schwarten, M., Mohrlüder, J., Ma, P., Stoldt, M., Thielmann, Y., Stangler, T., Hersch, N., Hoffmann, B., Merkel, R., Willbold, D., 2014. Nix directly binds to GABARAP: A possible crosstalk between apoptosis and autophagy. Autophagy 5, 690-698. Schweers, R.L., Zhang, J., Randall, M.S., Loyd, M.R., Li, W., Dorsey, F.C., Kundu, M., Opferman, J.T., Cleveland, J.L., Miller, J.L., Ney, P.A., 2007. NIX is required for programmed mitochondrial clearance during reticulocyte maturation. Proc Natl Acad Sci U S A 104, 19500-19505. Song, S., Jang, S., Park, J., Bang, S., Choi, S., Kwon, K.Y., Zhuang, X., Kim, E., Chung, J., 2013. Characterization of PINK1 (PTEN-induced Putative Kinase 1) Mutations Associated with Parkinson Disease in Mammalian Cells and Drosophila. Journal of Biological Chemistry 288, 5660-5672. Truman, J.W., Riddiford, L.M., 2002. Endocrine insights into the evolution of metamorphosis in insects. Annu Rev Entomol 47, 467-500. Valente, E.M., Abou-Sleiman, P.M., Caputo, V., Muqit, M.M., Harvey, K., Gispert, S., Ali, Z., Del Turco,

D., Bentivoglio, A.R., Healy, D.G., Albanese, A., Nussbaum, R., Gonzalez-Maldonado, R., Deller, T., Salvi, S., Cortelli, P., Gilks, W.P., Latchman, D.S., Harvey, R.J., Dallapiccola, B., Auburger, G., Wood, N.W., 2004. Hereditary early-onset Parkinson's disease caused by mutations in PINK1. Science 304, 1158-1160. Valente, E.M., Bentivoglio, A.R., Dixon, P.H., Ferraris, A., Ialongo, T., Frontali, M., Albanese, A., Wood, N.W., 2001. Localization of a novel locus for autosomal recessive early-onset parkinsonism, PARK6, on human chromosome 1p35-p36. Am J Hum Genet 68, 895-900. Vives-Bauza, C., Zhou, C., Huang, Y., Cui, M., de Vries, R.L., Kim, J., May, J., Tocilescu, M.A., Liu, W., Ko, H.S., Magrane, J., Moore, D.J., Dawson, V.L., Grailhe, R., Dawson, T.M., Li, C., Tieu, K., Przedborski, S., 2010. PINK1-dependent recruitment of Parkin to mitochondria in mitophagy. Proc Natl Acad Sci U S A 107, 378-383. Voet, D., V., J.G., P., C.W., 2006. BAMBED Changes Publisher: Farewell, ASBMB; welcome, John Wiley. Biochem Mol Biol Educ 34, 401. Wang, X., Winter, D., Ashrafi, G., Schlehe, J., Wong, Y.L., Selkoe, D., Rice, S., Steen, J., LaVoie, M.J., Schwarz, T.L., 2011. PINK1 and Parkin target Miro for phosphorylation and degradation to arrest mitochondrial motility. Cell 147, 893-906. White, K.P., Hurban, P., Watanabe, T., Hogness, D.S., 1997. Coordination of Drosophila metamorphosis by two ecdysone-induced nuclear receptors. Science 276, 114-117. Whitworth, A.J., Pallanck, L.J., 2009. The PINK1/Parkin pathway: a mitochondrial quality control system? J Bioenerg Biomembr 41, 499-503. Yang, Y., Gehrke, S., Imai, Y., Huang, Z., Ouyang, Y., Wang, J.W., Yang, L., Beal, M.F., Vogel, H., Lu, B., 2006. Mitochondrial pathology and muscle and dopaminergic neuron degeneration caused by inactivation of Drosophila Pink1 is rescued by Parkin. Proceedings of the National Academy of Sciences 103, 10793-10798. Yang, Y., Ouyang, Y., Yang, L., Beal, F., McQuibban, A., Vogel, H., and Lu, B., 2008. Pink1 regulates mitochondrial dynamics through interaction with the fission/fusion machinery. Proceedings of the National Academy of Sciences 105, 17585-17585. Youle, R.J., Narendra, D.P., 2011. Mechanisms of mitophagy. Nat Rev Mol Cell Biol 12, 9-14. Zhang, L., Karsten, P., Hamm, S., Pogson, J.H., Muller-Rischart, A.K., Exner, N., Haass, C., Whitworth, A.J., Winklhofer, K.F., Schulz, J.B., Voigt, A., 2013. TRAP1 rescues PINK1 loss-of-function phenotypes. Hum Mol Genet 22, 2829-2841.

Highlights 1. An orchestrated cell-type specific mitophagy process in Drosophila midgut metamorphosis; 2. A pink1 genomic knock-in allele was generated to monitor the dynamic expression pattern of PINK1; 3. The spatiotemporal expression pattern of PINK1 correlates with the cell-type specific mitochondrial clearance or persistence; 4. PINK1 and PARKIN function epistatically to mediate timely specific mitophagy during Drosophila midgut metamorphosis.