PPTC7 maintains mitochondrial protein content by suppressing receptor-mediated mitophagy

Niemi, Natalie M.; Serrano, Lia R.; Muehlbauer, Laura K.; Balnis, Catherine E.; Wei, Lianjie; Smith, Andrew J.; Kozul, Keri-Lyn; Forny, Merima; Connor, Olivia M.; Rashan, Edrees H.; Shishkova, Evgenia; Schueler, Kathryn L.; Keller, Mark P.; Attie, Alan D.; Friedman, Jonathan R.; Pagan, Julia K.; Coon, Joshua J.; Pagliarini, David J.

doi:10.1038/s41467-023-42069-w

Download PDF

Article
Open access
Published: 13 October 2023

PPTC7 maintains mitochondrial protein content by suppressing receptor-mediated mitophagy

Nature Communications volume 14, Article number: 6431 (2023) Cite this article

4307 Accesses
2 Citations
43 Altmetric
Metrics details

Subjects

Abstract

PPTC7 is a resident mitochondrial phosphatase essential for maintaining proper mitochondrial content and function. Newborn mice lacking Pptc7 exhibit aberrant mitochondrial protein phosphorylation, suffer from a range of metabolic defects, and fail to survive beyond one day after birth. Using an inducible knockout model, we reveal that loss of Pptc7 in adult mice causes marked reduction in mitochondrial mass and metabolic capacity with elevated hepatic triglyceride accumulation. Pptc7 knockout animals exhibit increased expression of the mitophagy receptors BNIP3 and NIX, and Pptc7^-/- mouse embryonic fibroblasts (MEFs) display a major increase in mitophagy that is reversed upon deletion of these receptors. Our phosphoproteomics analyses reveal a common set of elevated phosphosites between perinatal tissues, adult liver, and MEFs, including multiple sites on BNIP3 and NIX, and our molecular studies demonstrate that PPTC7 can directly interact with and dephosphorylate these proteins. These data suggest that Pptc7 deletion causes mitochondrial dysfunction via dysregulation of several metabolic pathways and that PPTC7 may directly regulate mitophagy receptor function or stability. Overall, our work reveals a significant role for PPTC7 in the mitophagic response and furthers the growing notion that management of mitochondrial protein phosphorylation is essential for ensuring proper organelle content and function.

Small extracellular vesicles from young plasma reverse age-related functional declines by improving mitochondrial energy metabolism

Article Open access 16 April 2024

ROS-dependent S-palmitoylation activates cleaved and intact gasdermin D

Article 10 April 2024

Messenger RNA transport on lysosomal vesicles maintains axonal mitochondrial homeostasis and prevents axonal degeneration

Article Open access 10 April 2024

Introduction

Mitochondria engage in numerous cellular processes in eukaryotic cells, including ATP synthesis via oxidative phosphorylation, cofactor biosynthesis, and the commitment to cell death. Over time, mitochondria can become damaged, requiring quality control mechanisms to maintain a healthy mitochondrial population¹. One such mechanism is mitophagy, which engages the autophagic machinery to clear damaged or superfluous organelles². A growing body of work suggests that protein phosphorylation plays a key role in regulating mitophagy³, and in various other mitochondrial processes such as protein import^4,5, core metabolism^6,7, heme biosynthesis⁸, and programmed cell death⁹. Indeed, multiple candidate protein phosphatases reside at or within mitochondria^10,11,12,13 where they would be positioned to manage phosphorylation on mitochondrial proteins. Despite their prevalence, many of these phosphatases remain poorly studied.

Our recent work demonstrates that loss of one phosphatase, PPTC7, leads to the hyperphosphorylation of select mitochondrial proteins and significant loss of overall mitochondrial protein content¹⁴. Pptc7 knockout (KO) animals present with hypoketotic hypoglycemia and fail to survive a single day after birth. Pptc7 knockout tissues show widespread decreases in mitochondrial protein levels with little to no effects on matched mRNA levels, suggesting altered post-transcriptional control of mitochondrial content.

To further explore this phenomenon, we generated two independently derived Pptc7 knockout models: a conditional model that allows inducible Pptc7 knockout in adult mice, and isolated fibroblasts from our global Pptc7 (i.e., perinatal lethal) knockout mouse model. We find that knockout of Pptc7 decreases mitochondrial protein content in all tested systems. These molecular changes were accompanied by metabolic defects, including diminished oxygen consumption in cells exposed to multiple nutrients, as well as hepatic lipid accumulation in vivo. Interestingly, the mitophagy receptors BNIP3 and NIX were among the limited number of proteins that increased in expression across both mouse models and isolated fibroblasts. Our recent MITOMICS study revealed that PPTC7 loss also drives elevation of these receptors in human cells, and this response was unique across over 200 knockout cell lines¹⁵. CRISPR-mediated knockout of BNIP3 and NIX (gene name Bnip3l) rescued the expression of most mitochondrial proteins in Pptc7 KO cells, demonstrating that these receptors are necessary for the observed mitophagy. However, mitochondrial dysfunction persisted in Pptc7 knockout cells after Bnip3 and Bnip3l knockout, indicating that this dysfunction is connected at least partially to the underlying aberrant protein phosphorylation. Our phosphoproteomic analyses suggest a varied set of substrates for the phosphatase PPTC7, including the import protein TIMM50, multiple core metabolic proteins, and the BNIP3 and NIX mitophagy receptors themselves. Collectively, these data demonstrate that PPTC7 is critical for managing mitochondrial protein phosphorylation as well as regulating BNIP3- and NIX-associated mitophagy.

Results

PPTC7 maintains mitochondrial protein levels in adult mice

We demonstrated previously that knockout (KO) of the mitochondrial protein phosphatase Pptc7 causes decreased mitochondrial content, metabolic dysfunction, and fully penetrant lethality within one day of birth¹⁴. To overcome the limitations associated with these severe phenotypes, we generated a floxed Pptc7 mouse with the potential for conditional knockout (Fig. 1a). We bred floxed Pptc7 mice to the well-established UBC-Cre-ER^T2 mouse¹⁶ to allow widespread, inducible recombination in response to tamoxifen (Fig. 1b). We validated tamoxifen-induced genomic excision of Pptc7 exon 3 across seven tissues (Supplementary Fig. 1A), which we confirmed at the protein level in liver (Supplementary Fig. 1B). Furthermore, age-matched female Cre-positive floxed animals did not show significant levels of recombination in the absence of tamoxifen (Supplementary Fig. 1A), indicating that “leaky” Cre-induced knockout of Pptc7 is not a significant concern through the timeframe of our study.

**Fig. 1: Acute knockout of Pptc7 compromises hepatic mitochondrial content in adult mice.**

Upon confirming tamoxifen-inducible KO of Pptc7 in adult mice, we isolated liver tissue from control (i.e., UBC-Cre^+/-;Pptc7^+/+) and knockout (i.e., UBC-Cre^+/-;Pptc7^flox/flox) mice of both sexes two weeks post-tamoxifen administration and analyzed these tissues via 16-plex tandem mass tag (TMT) quantitative proteomics (Fig. 1c). In this experiment, we identified 6749 proteins, 1367 of which were altered in Pptc7^-/- liver relative to control tissue. Approximately equal numbers of proteins significantly increased (646) or decreased (721) in abundance (p < 0.05, Student’s t test). Notably, when stratified for mitochondrial localization, only 14 proteins in MitoCarta 3.0¹³ exhibited a log₂ fold change of >0.2 in Pptc7 KO liver, while 512 mitochondrial proteins exceeded a log₂ fold change of −0.2 (Supplementary Fig. 1C, Source data 1). These results demonstrate that acute loss of Pptc7 leads to decreased mitochondrial protein content in adult mouse liver comparable to what we observed in the perinatal global Pptc7 KO model¹⁴ (Fig. 1d, e). Importantly, PPTC7 was the most significantly downregulated mitochondrial protein in our proteomics analysis (Fig. 1d), providing confirmation of Pptc7 knockout in UBC-Cre^+/-;Pptc7^flox/flox liver tissue. We additionally confirmed decreased mitochondrial content through measurements of citrate synthase (CS) activity (Supplementary Fig 1D), CS protein expression (Supplementary Fig 1E), and mitochondrial DNA (mtDNA) levels (Supplementary Fig 1F). Despite the trend of decreased mitochondrial markers, a few proteins, including BNIP3 and its paralog, NIX, were significantly elevated in Pptc7 KO liver (Fig. 1d). An analysis of the proteomic changes between our two KO models revealed significant correlation (Fig. 1e), suggesting PPTC7 plays a similar role in liver tissue across developmental states. Collectively, these data demonstrate that acute loss of Pptc7 in adult mouse liver globally decreases mitochondrial content, suggesting that PPTC7 expression is required across at least two physiological states to maintain mitochondrial protein homeostasis.

To investigate the physiological response to acute Pptc7 loss, we performed gene function analysis on our proteomics dataset. This analysis revealed a significant enrichment in proteins involved in fatty acid oxidation (FAO, Fig. 1f), suggesting Pptc7 KO animals may have disrupted lipid metabolism. As global Pptc7 KO animals exhibit signs of diminished FAO such as hypoketosis¹⁴, we tested whether adult mice with acute loss of Pptc7 showed similar defects. We found that Pptc7 KO male mice did not have significantly altered ketone levels in response to an overnight fast (Fig. 1g) but that both sexes of Pptc7^-/- mice had significantly elevated liver triacylglycerols (TAGs) (Fig. 1h), consistent with hepatic steatosis. Interestingly, male Pptc7 KO mice also showed significantly increased expression of multiple perilipin proteins, which associate with lipid droplets and are elevated during steatosis^17,18 (Fig. 1i). We also examined hepatic and circulating cholesterol levels, as well as serum TAGs, which did not significantly change in Pptc7 KO animals relative to control littermates (Supplementary Fig. 1G–I). Collectively, these data suggest that acute loss of Pptc7 perturbs hepatic lipid homeostasis but in a manner distinct from that observed in the perinatal KO.

Loss of Pptc7 broadly decreases mitochondrial metabolism in isolated cells

To complement our studies in mice, we established a cell-based system to interrogate mechanisms driving the metabolic and mitochondrial abnormalities in Pptc7 KO models. We bred Pptc7^+/- mice from our global KO model to generate mouse embryonic fibroblasts (MEFs) from three wild-type and three knockout E14.5 embryos, which were validated at the gene (Fig. 2a) and protein (Fig. 2b) levels. A proteomic analysis of these cells revealed significant mitochondrial protein loss (Fig. 2c) consistent with our observations from the global Pptc7 KO heart and liver tissues and inducible Pptc7 KO liver tissue (although the relative reduction of mitochondrial proteins in vivo is larger than that seen in the MEFs) (Fig. 2d). These data demonstrate that loss of Pptc7 decreases mitochondrial protein content cell autonomously and in a third distinct cell or tissue type.

**Fig. 2: *Pptc7* knockout causes cell-autonomous decreases in mitochondrial protein content leading to broad metabolic defects.**

We performed correlation analysis of all overlapping proteins identified in MEFs, perinatal heart tissue, and perinatal liver tissue, and found significant positive correlation between Pptc7 KO models (Supplementary Fig. 2A, B). These data also reveal that proteins involved in metabolic pathways, such as oxidative phosphorylation, the TCA cycle, and fatty acid oxidation, were more strongly diminished in Pptc7 KO cells and tissues (Supplementary Fig. 2C–E) than other non-metabolic pathways (e.g., the mitochondrial ribosome, Supplementary Fig. 2F–H). To test whether Pptc7 KO cells have compromised metabolism, we performed Seahorse assays and found that primary Pptc7 KO fibroblasts have mildly decreased oxygen consumption rates (OCR) in basal conditions but have substantially lower spare respiratory capacity than wild-type fibroblasts (Fig. 2e). We also tested OCR in permeabilized wild-type and Pptc7 KO fibroblasts provided with pyruvate/malate or succinate/rotenone and found that KO cells given either substrate treatment showed decreased basal OCR and spare respiratory capacity (Fig. 2f, g). As our data suggested that acute loss of Pptc7 compromises fatty acid oxidation in mouse liver (Fig. 1), we also performed Seahorse analysis of MEFs exposed to palmitate-BSA in the presence or absence of the CPT1 inhibitor etomoxir. These data show that palmitate-BSA induces a significant increase in OCR in wild-type but not Pptc7 KO MEFs (Fig. 2h), consistent with deficient fatty acid oxidation in the latter. As an orthogonal approach, we performed a BioLog assay to test the capacity of Pptc7 KO cells to catabolize various nutrients. These experiments showed that Pptc7 KO cells have a compromised ability to metabolize substrates that feed into the TCA cycle (e.g., pyruvate, α-ketoglutarate, succinate, fumarate, or malate, Fig. 2i), glutamine catabolism (e.g., glutamate, glutamine, and Glutamax, Fig. 2j), and fatty acid catabolism (e.g., octanoyl-carnitine, Fig. 2i). Collectively, these data demonstrate that PPTC7 expression is required to maintain mitochondrial protein levels in isolated cells and that loss of this phosphatase disrupts multiple metabolic pathways.

Loss of mitochondrial content is mediated by the mitophagy receptors Bnip3 and Nix

The consistent loss of mitochondrial protein levels across each Pptc7 KO tissue and cell type (Fig. 2d) suggests a common molecular driver for this phenotype. The mitophagy receptor BNIP3 is significantly upregulated across all cell types and tissues assayed, with some systems, including the MEFs, showing additional upregulation of NIX (gene name Bnip3l), a BNIP3 paralog¹⁴ (Figs. 1c and 2d). We hypothesized that BNIP3 and NIX could promote excessive mitophagy, contributing to the decrease in mitochondrial protein levels seen in Pptc7 KO systems. To test this, we used CRISPR/Cas9 to disrupt Bnip3 and Bnip3l in the Pptc7 knockout background, which we validated by protein expression (Fig. 3a). We transduced these cells with the mt-Keima mitophagy reporter, a pH-dependent fluorescent protein that exhibits distinct emission spectra in mitochondria at physiological pH and in the acidic lysosomal compartment after mitophagy¹⁹. Pptc7 KO cells expressing mt-Keima showed elevated numbers of acidic mitochondria via both microscopy (Fig. 3b, c) and flow cytometry (Fig. 3d and Supplementary Fig. 3A), indicating excess mitophagic flux. Importantly, this increase was abolished following CRISPR/Cas9-based disruption of Bnip3 and Bnip3l, as seen in both microscopy (Fig. 3b, c) and FACS-based approaches (Fig. 3d and Supplementary Fig. 3A). Conversely, mitophagy was elevated when BNIP3 or NIX were overexpressed in wild-type MEFs (Supplementary Fig. 3B, C). Importantly, analysis of proteomics data in both Pptc7 KO liver tissue and MEFs found only BNIP3 and NIX to be significantly upregulated among identified proteins within the receptor-mediated or ubiquitin-meditated mitophagy pathways (Supplementary Fig. 3D, E). Most proteins within the autophagic pathway do not significantly change in expression in response to Pptc7 KO (Supplementary Fig. 3D, E). However, activated LC3-II accumulated to a greater extent in Pptc7 KO MEFs treated with bafilomycin A relative to wild-type cells (Supplementary Fig. 3F), suggesting higher autophagic flux. Collectively, these data suggest that BNIP3 and NIX drive the elevated mitophagy flux seen in Pptc7 knockout cells. Consistently, proteomic analysis of each of the Pptc7/Bnip3/Bnip3l knockout clones (herein referred to as TKO clones), as well as double Bnip3/Pptc7 and Bnip3l/Pptc7 KO clones revealed substantial rescue of mitochondrial proteins relative to Pptc7 KO cells (Fig. 3d).

**Fig. 3: *Pptc7* knockout causes excessive BNIP3- and NIX-mediated mitophagy.**

We next tested whether the knockout of Bnip3 and Bnip3l rescued the metabolic defects seen in Pptc7 KO cells. Basal oxygen consumption was partially rescued in both TKO clones, but spare respiratory capacity was not rescued in either TKO clone relative to Pptc7 KO cells (Fig. 3e). Furthermore, the TKO cells exhibited diminished OCR and ECAR relative to wild-type MEFs (Fig. 3f), similar to the Pptc7 KO cells. These data demonstrate that while the dual knockout of BNIP3 and NIX reversed mitochondrial protein loss in the Pptc7 KO cells, it did not rectify the underlying metabolic defects.

Due to the persistent metabolic defects seen in Pptc7/Bnip3/Bnip3l TKO cells, we considered the possibility that stabilization of BNIP3 and NIX may be an indirect, common downstream consequence of general mitochondrial dysfunction. To assess this, we analyzed the data from our recent MITOMICS study, in which we performed multiomic analyses on over 200 HAP1 cell lines carrying monogenic deletions of genes encoding mitochondria-localized proteins. Remarkably, Pptc7 was the only gene whose KO resulted in significant increase of both BNIP3 and NIX expression in these cells (Fig. 3g, h). These data suggest that upregulation of BNIP3 and NIX is not a general response to mitochondrial stress (at least in HAP1 cells), but rather that these mitophagy receptors are selectively elevated in response to the loss of Pptc7.

Phosphoproteomic analyses reveal BNIP3 and NIX as candidate PPTC7 substrates

Our analyses above suggest that the metabolic aberrations caused by PPTC7 loss are rooted in dysregulated protein phosphorylation that then leads to BNIP3/NIX-mediated mitophagy. Our new model systems offer an opportunity to identify common PPTC7 substrates across cells and tissues. To identify these putative substrates, we performed phosphoproteomic analyses on liver tissue from both sexes of mice from our inducible Pptc7 KO model (n = 16 mice; 4 per sex and genotype, Fig. 4a) and matched wild-type and Pptc7 KO MEFs (n = 3 independent cell lines for wildtype and KO, respectively, Fig. 4b). As expected from the disruption of a protein phosphatase, the majority of significantly altered mitochondrial phosphoisoforms were elevated in Pptc7 KO samples across both systems (Fig. 4a, b), whereas similar trends were not seen in the non-mitochondrial phosphoproteome (Supplementary Fig. 4A, B). The results, especially when combined with our earlier data¹⁴, suggest a complex regulatory picture. Collectively, hundreds of mitochondrial phosphoproteins are elevated across all systems, including marked changes on multiple metabolic proteins. However, only four phosphoisoforms are significantly elevated across all systems (Fig. 4c). Of note, there is substantially higher overlap of quantified phosphoisoforms in perinatal and adult model liver tissue from Pptc7 KO mice than between other samples (Fig. 4d), suggesting tissue specificity for PPTC7 substrates. These data are consistent with a recent study that found significant functional specialization of mitochondrial phosphoproteomes across tissues in mice²⁰. Validation of these potential substrates and their contribution to the Pptc7 KO metabolic phenotype across tissues will require further mechanistic investigations.

**Fig. 4: Phosphoproteomic analysis of *Pptc7* KO systems reveals candidate substrates, including BNIP3 and NIX.**

Notably, we identified multiple phosphorylation sites on the BNIP3 and NIX receptors themselves. These phosphorylation sites do not exhibit significant increases in the Pptc7 KO cells when the data are subject to the standard normalization to total protein levels. However, this normalization could mask meaningful hits in instances whereby the phosphosite in question promotes protein stability and is found at high stoichiometry. In this case, loss of a phosphatase could result in the accumulation of stabilized, phosphorylated substrates that is not evident by a change in apparent phosphorylation occupancy. Given this, and the substantial upregulation of BNIP3 and NIX in our systems, we examined our phosphoproteomics data without total protein normalization and found multiple elevated phosphopeptides for both BNIP3 and NIX across all Pptc7 KO systems (Fig. 4e). These phosphorylation events span eight unique residues in each of the BNIP3 and NIX proteins (Fig. 4f), suggesting complex phosphorylation-based regulation. Two of these observed sites, S60 and T66, were identified across each of our experimental systems (Fig. 4g) and were recently linked to BNIP3 stability²¹, adding further credence to the possibility that these phosphoisoforms may largely be representative of the total cellular pool of these proteins. Many aspects of this regulation remain to be elucidated, including the activity of kinase(s) responsible for BNIP3 and NIX phosphorylation, the effect of phosphorylation on protein half-life, and the absolute phosphorylation stoichiometry. Nonetheless, the consistent upregulation of these phosphopeptides in Pptc7 KO systems suggests that PPTC7 influences this phosphorylation status, possibly through direct interaction.

PPTC7 can directly interact with and dephosphorylate BNIP3 and NIX

We previously performed a large-scale affinity enrichment-mass spectrometry analysis using mitochondrial baits in two human cell lines²². In reanalyzing these data, we discovered that BNIP3 and NIX were the strongest interaction partners identified for PPTC7 in both 293T and HepG2 cells (Fig. 5a). Notably, of the 77 baits tested, PPTC7 is the only bait to capture endogenous BNIP3 or NIX beyond significance thresholds, suggesting that this interaction is both specific and reproducible across experiments and cell types. To confirm these results, we performed immunoprecipitations of overexpressed PPTC7-FLAG in U2OS cells. Endogenous BNIP3 and NIX are kept at low basal expression levels, resulting in difficulty capturing PPTC7-FLAG via immunoprecipitation in basal conditions (Fig. 5b). However, upon upregulation of BNIP3 and NIX through treatment with deferiprone (DFP), an iron chelator well known to induce BNIP3 and NIX²³, PPTC7 robustly immunoprecipitated both mitophagy receptors (Fig. 5b), further supporting a strong interaction between these proteins.

**Fig. 5: PPTC7 interacts with and can dephosphorylate BNIP3 and NIX.**

While these immunoprecipitations demonstrate that PPTC7 can interact with BNIP3 and NIX, it remained unclear whether these interactions were direct. To address this, we performed yeast two-hybrid analyses to analyze interactions between human PPTC7, BNIP3, and NIX. These data demonstrated strong interactions between PPTC7 and each mitophagy receptor in a reciprocal fashion (Fig. 5c), suggesting direct interaction. To confirm these interactions, we generated recombinant PPTC7 to query whether this phosphatase could directly dephosphorylate BNIP3 and NIX. We reasoned that such an experiment would test whether PPTC7 can directly and functionally interact with BNIP3 and NIX, as well as test whether PPTC7 has selectivity toward the phosphosites identified in our analyses. We isolated crude mitochondria from Pptc7 KO MEFs, which have elevated phosphorylation on multiple residues on BNIP3 and NIX (Fig. 4). We incubated these isolated mitochondria with either wild-type or catalytically inactive recombinant PPTC7 and subsequently performed phosphoproteomic analyses. These data demonstrate that wild-type, but not catalytically inactive, PPTC7 dephosphorylated BNIP3 (Fig. 5d) and NIX (Fig. 5e) at most identified phosphosites. Collectively, these data demonstrate that PPTC7 can directly interact with BNIP3 and NIX to promote their dephosphorylation. Whether these interactions are maintained in vivo and under the conditions in which dephosphorylation may occur within intact cells should be an area of active investigation in the future.

Discussion

Our data demonstrate that Pptc7 knockout leads to BNIP3 (and, in some systems, NIX) upregulation, excessive mitophagy, and decreased steady state levels of mitochondrial proteins. BNIP3 and NIX are phosphorylated in Pptc7 knockout systems, consistent with a model in which phosphorylation promotes BNIP3- and NIX-mediated mitophagic signaling. Notably, in S. cerevisiae, the single mitophagy receptor Atg32 is phosphorylated by casein kinase 2 (CK2) to promote its interaction with Atg11 and subsequently induce mitophagy^24,25. Similar mechanisms activate mammalian mitophagy receptors, including FUNDC1, which is phosphorylated by CK2 and dephosphorylated through the phosphatase PGAM5²⁶, and the Atg32 homolog BCL2L13, whose phosphorylation allows recruitment of the mitophagy machinery²⁷. Our data suggest that similar activation mechanisms may exist for BNIP3 and NIX, albeit ones that are more complex and that are potentially mediated through diverse signaling inputs, as evidenced by the numerous phosphorylation events identified on each receptor. Importantly, phosphorylation of BNIP3 and NIX has been previously linked to the function and stability of these proteins. Similar to FUNDC1, phosphorylation near the N-terminal LC3 interaction region (LIR) of BNIP3 and NIX stabilizes their interaction with LC3 to promote mitophagy^28,29,30. Phosphorylation has also been identified within the C-terminal regions of BNIP3 and NIX, influencing the ability of BNIP3 to promote cell death³¹ and the subcellular localization of NIX³². Finally, a recent study found that phosphorylation of BNIP3 at S60 and T66—two residues identified in our analysis—stabilizes this receptor by blocking its ubiquitination-mediated turnover²¹. Three independent studies recently identified the mitochondrial E3 ligase FBXL4 as responsible for the basal turnover of BNIP3 and NIX^33,34,35. Knockout of Fbxl4 increases BNIP3 and NIX protein levels and promotes excessive mitophagy, leading to diminished mitochondrial protein levels and perinatal lethality in mice³⁶—phenotypes that strikingly mirror those observed in our Pptc7 knockout models¹⁴.

Accumulating evidence suggests that cells devote significant resources to the proper regulation of BNIP3 and NIX activity. TMEM11, a transmembrane protein linked to the MICOS complex, was recently found to physically interact with BNIP3 and NIX and mediate their spatial localization³⁷. Importantly, knockout of TMEM11 triggers hyperactive mitophagy, likely due to an increased number of LC3-positive mitophagosome formation sites at the outer mitochondrial membrane³⁷. These data suggest that the physical localization of BNIP3 and NIX is likely important for the proper regulation of mitophagy. Beyond this, BNIP3 protein levels seem to be actively limited in basal cell culture conditions, as BNIP3 is one of the shortest-lived mitochondrial proteins with a half-life of 103 min³⁸. BNIP3 accumulates during proteasomal inhibition^21,28, suggesting efficient ubiquitin-mediated turnover. Consistently, Fbxl4 knockout systems show hyperactive basal mitophagy, further suggesting that efficient ubiquitination of BNIP3 and NIX is critical for maintaining mitochondrial content and proper organellar function^33,34,35. Our data, as well as data generated by He et al., suggest that phosphorylation of BNIP3 (and likely NIX) may promote protein stability by blocking ubiquitination-mediated turnover. However, He et al. identified PP1/2 A as the phosphatase responsible for dephosphorylating BNIP3²¹. This leads to an important question: is PPTC7 directly responsible for the dephosphorylation of BNIP3 and NIX, or does it mediate these effects indirectly (for instance, through PP1/2A activity)?

Certain observations support each of these models. Our data suggest that PPTC7 interacts with BNIP3 and NIX as demonstrated by unbiased affinity enrichment-mass spectrometry, immunoprecipitation, yeast two-hybrid experiments, and through the direct dephosphorylation of these mitophagy receptors on mitochondria isolated from Pptc7 knockout cells (Fig. 5). However, these data demonstrate only that PPTC7 can interact with BNIP3 and NIX in experiments in which they are proximal to one another. In cells, the matrix localization of PPTC7³⁹ should theoretically disallow the direct regulation of BNIP3 and NIX, which reside on the outer mitochondrial membrane^13,40. However, a handful of observations suggest that the location and function of PPTC7 may be more complex than currently appreciated. First, our western blot in Fig. 2b reveals two sets of PPTC7 bands, one that migrates at the expected molecular weight (~33 kDa), and one that migrates substantially higher than this (~40 kDa). These data suggest that PPTC7 may reside in distinct populations in cells, although the molecular details underlying such differences in migration patterns remain to be elucidated. Second, the yeast ortholog of Pptc7, PTC7, is alternatively spliced to encode isoforms that localize either to the mitochondrial matrix or to the ER or nuclear envelope^41,42. These data suggest that Ptc7p and PPTC7 may share evolutionarily conserved dual functions in distinct organellar or cellular locations. Finally, the potential dual localization of PPTC7 is further supported by the BioPlex interactome studies^43,44,45, which identify an interaction between PPTC7 and BNIP3. Notably, the BioPlex datasets also identify multiple interaction partners for PPTC7 in the mitochondrial matrix (e.g., DBT, RTN4IP1, TRMT2B, and TRMT61B) as well as on the outer mitochondrial membrane (e.g., RHOT1, RHOT2, HSDL1, MAOB, and TRABD), which could be consistent with dual localization of this phosphatase. We propose that these data, along with the data presented in this manuscript, support a role for PPTC7 in the direct dephosphorylation of BNIP3 and NIX. Whether these interactions are direct in cells and under what physiological contexts these interactions take place should be active areas of investigation in the future.

Despite the importance of BNIP3 and NIX in mediating excessive mitophagy in Pptc7 knockout systems, the lack of full metabolic rescue in our TKO cells suggests that this phosphatase influences mitochondrial functions in ways distinct from its functional interaction with these mitophagy receptors. One possibility is that dysregulated phosphorylation of mitochondrial proteins, as seen in Pptc7 KO systems, affects their mitophagic selectivity, as has recently been suggested in S. cerevisiae^46,47. This could happen independently of BNIP3 and NIX, which are not conserved in budding yeast. Despite the lack of BNIP3 and NIX in yeast, Ptc7p promotes mitochondrial function^7,48, suggesting that PPTC7 may have ancestral roles in enabling mitochondrial metabolism distinct from its role in mammalian mitophagy. One such function may be the regulation of mitochondrial protein import via TIMM50. Phosphorylation at threonine 33 of TIMM50 is one of the four consistently upregulated phosphorylation events in Pptc7 KO systems (Fig. 4c), and phosphorylation in a similar region of Tim50p (at serine 104) was identified in ptc7Δ yeast^7,14. We previously demonstrated that phosphomimetic mutation of S104 in Tim50p decreases mitochondrial protein import and that a non-phosphorylatable S104A Tim50p mutant partially rescues import defects in ptc7Δ yeast¹⁴. These data suggest that the dysregulated phosphorylation on TIMM50, or other candidate substrates identified in our analyses, likely contributes to the metabolic dysfunction seen in Pptc7 KO systems. The TKO cells generated for this study will serve as a critical tool for disentangling these metabolic phenotypes in the absence of diminished mitochondrial protein content.

Collectively, our data demonstrate that loss of the mitochondrial phosphatase PPTC7 decreases mitochondrial protein content in isolated cells, in multiple tissue types, and across developmental contexts. This loss in mitochondrial protein content is associated with elevated phosphorylation on numerous mitochondrial proteins, including the mitophagy receptors BNIP3 and NIX. The decreases in mitochondrial protein content seen in Pptc7 KO cells are largely rescued by genetic ablation of BNIP3 and NIX without rectifying the underlying metabolic issues. These data highlight the critical role of PPTC7 in maintaining mitochondrial homeostasis and further substantiate the importance of properly regulated protein phosphorylation within this organelle.

Methods

Generation of a conditional floxed Pptc7 mouse model

Pptc7 floxed animals were generated at the Biotechnology Center at the University of Wisconsin-Madison. The Easi-CRISPR method⁴⁹ was used to generate two loxP sites flanking exon 3 of Pptc7 (i.e., a floxed allele). Animals were generated with a CRISPR approach as previously described (Niemi et al.¹⁴), with a key modification the addition of a long, single stranded oligonucleotide (i.e., a megamer) to facilitate homologous recombination. Briefly, one-cell fertilized C57BL/6J embryos were microinjected with CRISPR reagents (ctRNA.int12.2 (20 ng/μL) + ctRNA.int23.3 (20 ng/μL) + Cas9 IDT (50 ng/μL) + megamer (50 ng/μL)). Sequences of ctRNA.int12.2, ctRNA.int23.3 and the sequence of the resulting floxed allele is provided in Supplementary Table 1. Microinjected embryos were surgically transferred into pseudeopregnant recipients. Pups were sequenced at weaning for each LoxP cassette. Founders were backcrossed to C57BL/6J mates and F1 pups analyzed by sequencing to evaluate germline transmission.

Generation of an inducible global Pptc7 knockout mouse

Pptc7 floxed animals were bred with UBC-Cre-ERT2 transgenic mice to generate animals a global, inducible Pptc7 knockout model. The UBC-Cre-ERT2 strain was acquired from Jackson laboratories (B6.Cg-Ndor1^{Tg(UBC-cre/ERT2)1Ejb}/1J, Strain #007001) and was bred to Pptc7^flox/flox animals. Pptc7 floxed animals were bred to homozygosity (i.e., Pptc7^flox/flox) while maintaining one copy of the UBC-Cre/ERT2 transgene (referred to as “Cre” hereafter) as experimental animals. Control animals include UBC-Cre/ERT2;Pptc7^+/+ animals and/or Pptc7^flox/flox animals carrying no copy of Cre transgene. Controls used in each experiment are clarified within the figure legends of the manuscript. Both male and female mice were used in our analyses. To induce knockout, mice were fed tamoxifen-containing chow for seven consecutive days. Tamoxifen chow was purchased from Envigo (Teklad custom diet TD.130859), which is formulated for 400 mg of tamoxifen citrate per kilogram of diet. The diet was modified to promote more consistent feeding and genomic knockout as follows: tamoxifen chow pellets were ground to a powder with a morter and pestle, as was standard chow (Formulab Diet 5008). The two diets were mixed 3:1 (e.g., 15 g of tamoxifen chow to 5 g of standard chow) and placed into a weigh boat. Water was added to the powder to create a “mash”; a fresh mash was given to each cage daily for the duration of the treatment. After tamoxifen administration was complete, mice were transitioned back to a standard chow diet (Formulab Diet 5008) until the end of the experiment. Mice were housed on a 12-h light:dark cycle, and group housed by strain and sex under temperature- and humidity-controlled conditions and received ad libitum access to water and food. All animal work complied with ethical regulations for animal testing and research, and was done in accordance with IACUC approval by the College of Agricultural and Life Sciences (CALS) Animal Care and Use Committee at the University of Wisconin- Madison (protocol/animal welfare assurance #A3368-01) or at Washington University School of Medicine (animal welfare assurance D1600245).

Genotyping analysis

Each model used in our studies was validated by genotyping analysis. Genotyping primers for the flox allele and Cre used for genotyping can be found in Supplementary Table 2. The flox primers produce a 667 bp amplicon for the floxed allele and a 599 bp amplicon for the wild type allele. Upon Cre-mediated excision, these primers produce a 167 bp amplicon reflecting a knockout allele. The Cre-specific primers produce a 480 bp amplicon for the Cre allele. Approximately 100 ng of genomic DNA isolated from tail tips served as a template for all genotyping reactions. Pptc7 knockout MEFs were generated from our previously published CRISPR knockout model and genotyped as previously described¹⁴.

Generation and validation of mouse embryonic fibroblasts

Mouse embryonic fibroblasts were generated from a cross of Pptc7^+/- heterozygous mice. A timed mating was performed and E0.5 was determined through the presence of a vaginal plug. The pregnant female was sacrificed at E14.5 by CO₂ asphyxiation. 11 embryos were removed, decapitated, and internal organs removed. Remaining tissue was finely minced, suspended in 0.25% trypsin-EDTA, and incubated at 37 °C for 30 min in a CO₂-controlled tissue culture incubator. After trypsinization, tissues were pipetted to a single cell suspension, filtered through a nylon mesh cell strainer (Thermo-Fisher #08-771-1), and plated into individual 60 mm² dishes. Cells were cultured in DMEM (high glucose, no pyruvate, Thermo-Fisher catalog #11965-092) supplemented with 10% heat inactivated fetal bovine serum (FBS) and 1× penicillin/streptomycin. Cells were grown in a temperature-controlled CO₂ incubator at 37 °C and 5% CO₂. For the first 3 passages, media was supplemented with gentamicin (10 μg/mL) to reduce risks of contamination. Each cell line was genotyped using leftover tissue from each embryo from which genomic DNA was isolated. gDNA isolation and genotyping was done as previously described¹⁴. The sex of the cells was not determined as male/female embryos are morphologically indistinguishable at E14.5. Immortalized MEFs were generated using plasmid encoding SV40 Large T antigen as follows: MEFs were split into 6 well plates to ~80% confluence and transfected with 2 µg SV40 1: pBSSVD2005 (a gift from David Ron; Addgene plasmid #21826) using FuGENE HD according to manufacturer’s protocol. MEFs were split 1:10 for 5 consecutive passages before considered immortalized. Immortalized MEFs were cultured identically to primary MEFs (see above) and were used to a maximum of 15 passages for all experiments. Cellular experiments in this paper were performed with both primary (e.g., Seahorse assays) and immortalized (e.g., phosphoproteomics, proteomics, and BioLog assays) MEF clones.

Proteomic and phosphoproteomic analysis

TMT-based proteomic and phosphorproteomic experiments were performed on mouse liver tissue and mouse embryonic fibroblasts, Supplementary Data 1 and 2, respectively. Label-free proteomics analysis was performed on Pptc7/Bnip3/Bnip3l TKO samples, Supplementary Data 3. Details for the generation of these datasets are as follows:

TMT phosphoproteomics

Tissue lysis and protein digestion

Cells or mouse liver tissue were resuspended in 1 mL 6 M guanidine hydrochloride, 100 mM Tris, pH 8 and probe-sonicated using a Misonix XL-2000 Series Ultrasonic Liquid Processor. The sonication regime, repeated twice, was the following: the probe was set to output 11 watts for 10 s with an off-time of 30 s. The protein concentrations of the homogenized tissue samples were determined using the BCA Protein Assay Kit (Thermo Pierce). Sample volumes corresponding to 400 μg of protein were set aside for further processing. Methanol was then added to each sample so that the final volume consisted of 90% methanol. The samples were then centrifuged at 9000 × g for 5 min, and the supernatant was decanted. The protein pellet was then resuspended in 200 μL of lysis buffer containing 8 M urea, 10 mM tris (2-carboxyethyl) phosphine, 40 mM chloroacetamide, and 100 mM Tris, pH 8. 8 μg of Lysyl Endopeptidase (FUJIFILM Wako Chemicals) was added and incubated for 4 h at room temperature. The samples were then diluted with 100 mM tris, pH 8 to achieve a final urea concentration of less than 2 M. 8 μg of trypsin (Promega) was added, and the samples incubated for 10 h at room temperature. Trypsin enzyme activity was then quenched by addition of 120 μL of 10% trifluoroacetic acid (TFA), and the samples were centrifuged at 9000 × g for 5 min. Peptides were desalted with Strata-X reversed phase solid phase extraction cartridges (Phenomenex).

TMT labeling

Desalted peptides were resuspended in 100 μL 200 mM triethylammonium bicarbonate buffer. Each TMTpro 16-Plex channel had been pre-aliquoted in 500 μg quantities was resuspended in 20 μL of acetonitrile (ACN). Each sample was added to distinct TMT channels and vortexed for 4 h at room temperature. The TMT labeling reaction was quenched with 0.64 μL of 50% hydroxylamine and 100 μg of labeled peptides per sample were combined. The combined sample was dried and desalted with Strata-X reversed phase solid phase extraction cartridges.

Phosphopeptide enrichment

Desalted TMT-labeled peptides were resuspended in 1 mL 8% can in 6% TFA. 104 μL of magnetic titanium dioxide beads (ReSyn Biosciences) that were washed with 80% ACN 6% TFA three times. The resuspended sample was added to the washed beads and vortexed for 1 h at room temperature. The bead-bound phosphorylated peptides were then sequentially washed three times with 1 mL of 80% ACN/6% TFA, once with 80% ACN, once with 80% ACN in 0.5 M glycolic acid, then three times with 80% ACN. The phosphorylated peptides were eluted from the beads with two washes of 300 μL 50% ACN in 1% ammonium hydroxide. Dried phosphorylated peptides were resuspended in 0.1% TFA and desalted using Strata-X reversed phase solid phase extraction cartridges.

Offline fractionation

The TMT-labeled phosphopeptides were loaded onto a 4.6 mm inner diameter 150 mm outer diameter BEH C18 column (Waters Corp) for offline fractionation into 16 fractions. The peptides were separated using a high pH reversed-phase gradient (10 mM ammonium formate Mobile Phase A, 10 mM ammonium formate in 80% MeOH Mobile Phase B) on an Agilent 1260 Infinity Binary LC equipped with an automatic fraction collector. The effective separation gradient started at 20% Mobile Phase B and increased to 60% Mobile Phase B in 6 min, flowing at a rate of 800 μL/min. The fractions were then concatenated into 8 fractions, combining fractions 1 and 9, 2 and 10, 3 and 11, 4 and 12, 5 and 13, 6 and 14, 7 and 15, and 8 and 16. The combined fractions were transferred to 2 mL Starstedt micro tubes for rapid evaporation under vacuum with the SpeedVac Vacuum Concentrator Kit (Thermo).

LC-MS acquisition

Each fraction was resuspended in 0.2% formic acid. 1 μg of peptide was loaded onto a 75 μm inner diameter × 360 μm outer diameter column (New Objective) packed in-house 1 to 30 cm with 1.7 micron BEH C18 particles (Waters Corp). The peptides were chromatographically separated over a 120-min reverse phase gradient (0.2% formic acid mobile phase A, 0.2% formic acid/80% ACN mobile phase B) using a Dionex UltiMate 3000 nano-HPLC (Thermo) at a column temperature of 50 °C and a flow rate of 335 nL/min. Eluting peptides were sprayed into an Orbitrap Eclipse using a Nanospray Flex ionization source (Thermo) at 2 kV. MS1 scans were recorded in the Orbitrap using a resolving power of 60,000, requiring an AGC target of 1e6 or a maximum injection time of 50 ms. MS1 scans included precursor ions ranging from 300-1350 m/z. The most intense ions were selected for HCD fragmentation (normalized collision energy 35%) and subsequent MS2 analysis until the 1 s duty cycle lapsed. Dynamic exclusion was set to 40 s and the MS2 isolation width was set to 1.5 Th. Only precursors of charge states 2–6 were selected. MS2 spectra were analyzed in the Orbitrap using a resolving power of 60,000 requiring an AGC target of 1e5 or a maximum inject time of 118 ms. MS2 scans included fragment ions ranging from 100 to 2000 m/z.

Data processing

Thermo RAW files were processed with MaxQuant (version 1.5.2.8)2, implementing the Andromeda3 algorithm to perform database searching of MS2 spectra against a database of canonical proteins and isoforms (Uniprot, Mus musculus, 2018). TMT labels were set as fixed modifications on N termini and lysine residues. Carbamidomethylation of cysteine was set as a fixed modification. Variable modifications included acetylation of the N terminus, phosphorylation of serine, threonine, and tyrosine, and oxidation of methionine. Peptide and protein matches were filtered to 1% FDR. Phosphorylation sites were considered localized if they delivered MaxQuant localization scores >0.75. The TMT reporter ion intensities of the phosphorylated peptides were median-normalized to account for variability introduced from sample-handling. To this end, the median of the reporter ion intensities for each sample was calculated. To obtain the correction factor for each phosphorylated peptide intensity, the average of the medians was divided by each of the respective sample medians, rendering a correction factor for each channel. This channel-specific correction factor was then multiplied to each of the reporter ion intensities within that channel.

Label-free proteomics

Protein extraction and digestion

Mouse embryonic fibroblast cells were resuspended in 1 mL 6 M guanidine hydrochloride, 100 mM Tris, pH 8 and sonicated using a Qsonica Q700 temperature-controlled sonicator. The sonication regime cycled 10 times applying an ultrasonic wave of amplitude of 35 for 20 s with a 10 s off time. The protein concentrations of the lysate were then determined using the Pierce BCA Protein Assay Kit. Methanol was then added to volumes corresponding to 200 μg for each sample, so that the final volume consisted of 90% methanol for protein extraction. The samples were then centrifuged at 9,000 g for 5 min and the supernatant was decanted. The protein pellet was then resuspended in 100 μL of lysis buffer containing 8 M urea, 10 mM Tris (2-carboxyethyl) phosphine, 40 mM chloroacetamide, and 100 mM Tris, pH 8. 4 μg of Lysyl Endopeptidase was added and incubated for 4 h at room temperature. The samples were then diluted with 100 mM tris, pH 8 to achieve a 2 M urea concentration. 4 μg of trypsin was added and the samples incubated for 10 h at room temperature. Trypsin enzyme activity was then quenched by adding 60 μL of 10% TFA and the samples were centrifuged at 9000 × g for 5 min. Peptides were desalted with Strata-X reversed phase solid phase extraction cartridges.

LC-MS acquisition method

Each sample was resuspended in 0.2% formic acid. 1.5 μg of peptide was loaded onto a 75 um inner diameter × 360 um outer diameter column (CoAnn Technologies) packed in-house to 30 cm with 1.7 micron BEH C18 particles (Waters Corp). The peptides were chromatographically separated over a 90-min reverse phase gradient (0.2% formic acid mobile phase A, 0.2% formic acid/80% ACN mobile phase B) using a Vanquish Neo UHPLC (Thermo) at a column temperature of 50 °C and a flow rate of 320 nL/min. Eluting peptides were sprayed into an Orbitrap Eclipse using a Nanospray Flex ionization source (Thermo) at 2.2 kV. MS1 scans were recorded in the Orbitrap using a resolving power of 240,000, requiring an AGC target of 8e5 or a maximum injection time of 50 ms. MS1 scans included precursor ions ranging from 350-2000 m/z. The most intense ions were selected for HCD fragmentation (normalized collision energy of 25%) and subsequent MS2 analysis until the 1 s duty cycle lapsed. Dynamic exclusion was set to 10 s and the quadrupole isolation width was set to 0.5 Th. Only precursor ions with charge states of 2–5 were selected. MS2 spectra were analyzed in the ion trap on the “turbo” resolution setting, requiring an AGC target of 2e4 or a maximum inject time of 14 ms. MS2 scans included fragment ions ranging from 150 to 1350 m/z.

Data processing

Thermo RAW files were processed with MaxQuant (version 1.5.2.8)2, implementing the Andromeda3 algorithm to perform database searching of MS2 spectra against a database of canonical proteins and isoforms (Uniprot, Mus musculus). Carbamidomethylation was set as a fixed modification on cysteines. Variable modifications included N terminus acetylation and methionine oxidation. Peptide and protein matches were filtered to 1% FDR. Fast LFQ was toggled off and a LFQ minimum ratio count was set to 1. Match between runs (MBR) was toggled on. All other parameters were set to default MaxQuant settings.

Phosphoproteomics on phosphatase-treated isolated mitochondria

Pptc7^-/- mouse embryonic fibroblasts were grown as detailed in “Generation and validation of mouse embryonic fibroblasts”. Crude mitochondria were isolated as previously described⁵⁰ with some modifications. Briefly, cells were scraped into dPBS, placed on ice, and pelleted at 600 × g at 4 °C for 10 min. Cells were resuspended in ice-cold IB_c (0.1 M Tris-MOPS, pH 7.4, 0.1 M EGTA). Cells were transferred to a prechilled 7 mL dounce homogenizer on ice, and cells were homogenized using a tight pestle with 30–35 strokes. Homogenized cells were spun at 600 × g at 4 °C for 10 min to pellet nuclei and unbroken cells. The supernatant was transferred to a new, prechilled tube and spun at 7000 × g for 10 min at 4 °C. The pellet contained crude mitochondria, which were resuspended in IB_c and quantified using the Pierce 660 nm protein assay. ~500 μg mitochondria per condition were diluted into IB_c containing 10 mM MnCl₂ and ~1 μg of PPTC7 or PPTC7 D78A (or vehicle only for untreated conditions). Phosphatase assays were incubated at room temperature for 60 min before being snap frozen and stored at −80 °C until processing for phosphoproteomics.

Sample preparation

Crude mitochondrial protein was precipitated in 90% LC-MS grade MeOH and centrifuged at 16,500 × g at 4 °C for 10 min. Pellets were solubilized in buffer containing 8 M Urea and 100 mM Tris by sonication. Samples were centrifuged again (16,500 × g at 4 °C for 10 min), and supernatant protein concentration was quantified using BCA assay. Total protein was normalized to 400 µg per sample prior to reduction and alkylation with 10 mM TCEP and 40 mM 2-CAA, respectively. Urea concentration was reduced to <2 M with 100 mM Tris prior to tryptic digestion (1:25) overnight shaking at 37 °C. Digested peptides were desalted using StrataX polymeric reverse phase resin (Phenomenex), and fully dried before further processing. 5% of total peptides were retained as a proteomics sample. The remaining peptides were subjected to Sequential enrichment using Metal Oxide Affinity Chromatography (SMOAC) using TiO₂ and Fe³⁺ enrichment columns, and dried to completion. Proteomic, TiO₂, and Fe³⁺ samples were resuspended in 0.2% formic acid (FA).

LC-MS/MS analysis

1 µg of peptides for proteomics analysis, and 4 µL of phosphoenriched samples were injected to a Thermo Orbitrap Exploris 240 following LC separation on an EASY-spray HPLC column (Thermo, ES902) using an UltiMate 3000 UHPLC system. Proteomics samples were separated using buffer A (0.2% formic acid) and buffer B (0.2% FA in 80% acetonitrile) on a 120 min gradient (3–60 min, 4–20% B, 60–100 min, 20–45% B, then washed and re-equilibrated). Phosphoproteomic samples were separated on a 120 min gradient (5–100 min, 8–40% B, then washed and re-equilibrated). Eluted peptides entered the mass spectrometer following positive mode ESI. For proteomics samples, MS1 scans were performed in the orbitrap (120,000 resolution, normalized AGC target 300%, RF Lens 80%, maximum ion injection time of 50 ms, with a 1.0e5 intensity threshold). MS2 analysis of HCD-generated (75% collision energies) fragmented ions were performed in the orbitrap (30,000, normalized AGC target 75%, with automatic maximum injection time). For phosphoproteomic samples, MS1 scans were performed in the orbitrap (120,000 resolution, normalized AGC target 300%, RF Lens 70%, maximum ion injections time 100 ms, with a 5.0e3 intensity threshold). MS2 analysis of HCD-generated (25% collision energies) fragmented ions were performed in the orbitrap (30,000, normalized AGC target 25%, with automatic maximum injection time). Monoisotopic precursor selection and dynamic exclusion (proteomics 60 s, phosphoproteomics 30 s) were enabled.

MS data analysis

Thermo raw files were searched with Proteome Discoverer 2.5 using Sequest HT with the uniprot Mus musculus fasta file (v2023-03-01). Searches were performed with a precursor mass tolerance of 10 ppm. Carbamidomethylation (+57.021 Da (C)) was included as a static modification, and N-terminal Acetyl (+42.011 Da), Met-loss (−131.040 Da (M)), and Met-loss+Acetyl (−89.030 Da (M)) were included as dynamic modifications in all searches. Phosphorylation (+79.966 Da (S, T, Y)) was included as a dynamic modification for all phosphoenrichment samples. The IMP-ptmRS node was used for scoring post-translational modifications, with a cut off set to >0.75. Search results were filtered to a peptide FDR of 1%. Phosphopeptide intensities were normalized to total reporter ion intensity. Imputation was performed by sampling values from the lowest 5% of the normal distribution within groups. Associated p-values were calculated using Student’s T-test assuming equal variance.

Quantification of ketones, triacylglycerols, and cholesterol

Ketone bodies were quantified as previously described¹⁴. Briefly, serum from mice fasted overnight (i.e., between 16 and 18 h) was collected and stored at −80 °C until the assay was performed. Serum was thawed on ice, diluted 1:10, and ketone bodies were quantified using the Wako Autokit Total Ketone Bodies Assay per the manufacturer’s directions, with the modification that measurements were made kinetically over a 30-min period rather than an endpoint measurement. A standard curve, made from Wako’s Calibrator 300, was simultaneously run and total ketone bodies in each sample were calculated according to this standard. Quantification of total triglycerides and total cholesterol in plasma were determined using the Triglyceride Infinity Kit and the Cholesterol Infinity Kit (both from Thermo Scientific). Briefly, in a 96 well plate format, 100 μL of each reagent was aliquoted and 1 μL of plasma sample or standard was added and mixed. Reactions were incubated for 30 min at room temperature. Total triglycerides were determined by readings at an absorbance of 540 nm, while total cholesterol was read at an absorbance of 490 nm. Sample concentrations were calculated from a standard curve correcting for blanks and a secondary wavelength at 660 nm. The same kits were used to quantify total triglycerides and total cholesterol in liver tissue. For sample preparation, 30-50 mg of fresh weight liver tissue was homogenized in 1.5 mL of chloroform: methanol (2:1 v/v). Samples were centrifuged at 14,000 × g for 10 min at 4 °C. An aliquot was evaporated in a 1.5 mL microcentrifuge tube and 100 μL of reagent was added to the dried sample followed by 30-min room temperature incubation. The developed reagent was then transferred to a 96 well plate with standards to determine the concentration. Final results were calculated by normalizing for the initial fresh weight of tissue.

Metabolic assays

For the Seahorse assays in Fig. 2, primary (i.e., not immortalized) Pptc7 knockout MEFs and matched wild-type MEFs were plated into Seahorse XFe96 plates at 12,500 cells/well. For the Seahorse assay in Fig. 3, immortalized wild-type, Pptc7 knockout, and Pptc7/Bnip3/Bnip3l triple knockout cells were plated into XFe96 plates at 15,000 cells/well. Cells were allowed to adhere overnight in standard DMEM media (25 mM glucose, 2 mM glutamine, 10% heat inactivated FBS, 1× penicillin/streptomycin). One hour before the Seahorse experiment, DMEM was aspirated, cells were washed with PBS, and were swapped into Seahorse XF Base Medium (i.e., unbuffered DMEM containing 25 mM glucose, 2 mM glutamine with no FBS) and allowed to equilibrate at 37 °C in a non-CO₂ incubator before initiating the experiment. While cells were equilibrating, a pre-hydrated Seahorse cartridge was loaded with oligomycin, FCCP, and rotenone/antimycin A prepared from a XF Cell Mito Stress Test Kit, which were reconstituted per the manufacturer’s directions. Compounds were injected into 180 μL assay medium in subsequent injections for final concentrations of 1.0 μM oligomycin, 1.0 μM FCCP, and 0.5 μM rotenone/antimycin A. For pyruvate/malate and succinate/rotenone experiments, primary Pptc7 knockout and wild-type MEFs were permeabilized using XF PMP per manufacturer’s instructions. MEFs were plated in a 96 well plate as described above. Plating DMEM was aspirated, and cells were washed with 1× MAS (220 mM mannitol, 70 mM sucrose, 10 mM KH₂PO₄, 5 mM MgCl₂, 2 mM HEPES, 1 mM EGTA, and 0.2% w/v fatty acid free BSA). To initiate the experiment, cells were swapped into 1x MAS containing 1 nM XF PMP (to permeabilize the cells) and either 1 mM malate and 10 mM pyruvate or 10 mM succinate and 2 μM rotenone. Permeabilized cells were subjected to the XF Cell Mito Stress Test Kit as described above. For the fatty acid oxidation Seahorse experiment, palmitate-BSA and BSA were purchased from Agilent to test the responsiveness of wild type and Pptc7 knockout MEFs to fatty acids. Briefly, 15,000 immortalized mouse embryonic fibroblasts were plated into XFe96 plates in DMEM and allowed to adhere overnight. Twenty-four hours before starting the assay, growth media was replaced with substate-limited medium (i.e., DMEM, 0.5 mM glucose, 1 mM glutamine, 0.5 mM carnitine, and 0.5% FBS). 45 min before starting the assay, cells were washed with PBS and media was replaced with FAO assay medium (KHB – 111 mM NaCl, 4.7 mM KCl, 1.25 mM CaCl₂, 2 mM MgSO₄, 1.2 mM NaH₂PO₄) supplemented with 2.5 mM glucose, 0.5 mM carnitine, and 5 mM HEPES, adjusted to pH 7.4. 15 min prior to starting the assay, 4 μM etomoxir or vehicle (DMSO) was added to select wells. Palmitate-BSA or BSA alone were injected to wells after three measurements of basal respiration. All data were collected on an XFe96 Seahorse Flux analyzer and analyzed on its associated Seahorse XF software. To perform the BioLog experiment, a BioLog MitoPlate S-1 Mitochondrial Substrate Metabolism assay plate was activated by adding “assay mix” (BioLog MAS, Redox Dye MC, saponin, and water) to each well and incubating for 1 h at 37 °C to dissolve substrates. Immortalized Pptc7 knockout MEFs and matched wild type MEFs were trypsinized, counted to a final cell number of 20,000 cells per well, and resuspended in 3 mL of 1× BioLog MAS (Biolog catalog #72303). The redox dye color formation (OD590) within the plate was tracked on an OmniLog instrument for 4 h. Data were collected on OmniLog software and analyzed by calculating the slope of redox dye formation as reported over this 4-h timeframe.

Citrate synthase assays

Citrate synthase assays were performed as previously described^7,14. Briefly, liver tissue was homogenized in Lysis Buffer A (LBA; 50 mM Tris-HCl, pH 7.4, 40 mM NaCl, 1 mM EDTA, and 0.5% (v/v) Triton X-100) supplemented with 1× protease inhibitor cocktail (0.5 μg/mL pepstatin A, chymostatin, antipain, leupeptin, and aprotinin) and 1× phosphatase inhibitor cocktail (0.5 mM imidazole, 0.25 mM sodium fluoride, 0.3 mM sodium molybdate, 0.25 mM sodium orthovanadate, and 1 mM sodium tartrate). Lysates were clarified, BCA normalized, and 5 μg of whole cell lysate was used for each reaction. 200 μL reactions were assembled into a single well of a 96 well plate containing the following (final concentrations): 100 mM Tris, pH 7.4, 300 μM acetyl CoA, 100 μM DTNB (5,5’-dithio-bis-[2-nitrobenzoic acid]). 10 µL of 10 mM oxaloacetic acid (OAA) was added per well (final [c] = 500 μM) to initiate enzyme assay. Absorbance at 412 nm (A412) was measured every minute for 30 min on a Cytation 3 plate reader (BioTek).

Western blotting and antibodies

Western blotting was performed as previously described¹⁴. Antibodies used in this study include: anti-PPTC7 (Novus Biologicals, catalog #NBP1-90654, 1:1000 dilution, 48 h incubation), anti-Citrate synthase (Cell Signaling Technologies, catalog #14309, 1:1000 dilution, overnight incubation), anti-beta actin (Abcam, catalog # ab170325, 1:1000 dilution, overnight incubation), anti-BNIP3 (rodent specific antibody, Cell Signaling Technologies, catalog #3769, 1:1000 dilution, 48 h incubation), anti-NIX (Cell Signaling Technologies, catalog #12396, 1:1000 dilution, overnight incubation), and anti-LC3B (Cell Signaling Technologies, catalog #3868, 1:1000 dilution, overnight incubation). For the immunoprecipitations, the following antibodies were used: anti-BNIP3 (Abcam, clone EPR4034, catalog #ab109362, WB 1:1000), anti-NIX (Cell Signaling Technology, clone D4R4B, catalog #12396, WB 1:1000); anti-FLAG (Sigma-Aldrich, catalog #SAB4301135, WB 1:1000), and anti-vinculin (SCBT, clone G-11; sc-55465, WB 1:1000).

mtDNA analysis

Genomic DNA (i.e., gDNA) was isolated from mouse liver tissue using the DNeasy Blood & Tissue Kit (Qiagen) according to manufacturer’s instructions. DNA was quantified and normalized to 100 ng input per reaction. Primers to analyze nuclear and mitochondrial DNA markers can be found in Supplementary Table 3. Quantitative real time PCR was performed on a QuantStudio 6 Real-Time PCR system with data was collected on QuantStudio Real-Time PCR v1.2 software. Relative quantitation was performed using the ΔΔCt method.

Generation of Pptc7/Bnip3/Nix triple knockout (TKO) cells

A single clone of Pptc7 knockout MEFs were used to generate Pptc7/Bnip3/Bnip3l triple knockout (TKO) cells. The TKO MEFs were created by the Genome Engineering & Stem Cell Center (GESC@MGI) at Washington University in St. Louis. Briefly, two gRNAs were designed to target each 5’ and 3’ of exon of interest, introduce double strain break and delete target exon. Synthetic gRNAs were purchased from IDT (sequences are provided in Supplementary Table 4), complexed with Cas9 recombinant protein and transfected into MEFs. The transfected cells were then single cell sorted into 96-well plates. Single cell clones were identified using deletion PCR and inside PCR to analyze the target site region for complete deletion of the respective gene (primers sequence and PCR products size are reported in Supplementary Table 4).

Mitophagy and autophagy assays

The mitochondrial keima (mt-Keima) assay was described previously¹⁹. Live-cell images were acquired using a Leica DMi8 SP8 Inverted Confocal microscope, equipped with a 63x Plan Apochromatic objective and environmental chamber (5% CO₂, 37 °C). Images were quantified with Image J/FIJI software. Single cells were segregated by generating regions of interest (ROI). Selected ROIs were cropped and split into separated channels for 561 nm and 458 nm, followed by threshold processing. The fluorescence intensity of mt-Keima at 561 nm (lysosomal signal) and mt-Keima at 458 nm (mitochondrial signal) was measured for each ROI and the ratio (561 nm/458 nm) was calculated. Three biological replicates were performed, with >50 cells analyzed per condition for each replicate. Individual ratios are represented as gray dots and the mean ratios from each biological replicate. For the flow cytometry experiments, cells were harvested by trypsinization and were resuspended in FluoroBrite media with 0.8% heat-inactivated FBS immediately prior to flow cytometry. Cells were analyzed on a LSR-Fortessa (BD Biosciences; Franklin Lakes, NJ, USA) flow cytometer using BDFACSDiva software (version 9.0). FSC (488 nm), SSC (excitation 488 nm, emission 488), QDot 605 (excitation 405 nm, emission 610), and PE-TexasRed (excitation 585 nm, emission 610) laser filters were used. For all experiments, wild-type MEFs (i.e., not expressing mtKeima) were used as a negative control to calibrate the laser intensities for Qdot 605 and PE-TexasRed, which were kept constant throughout experiments. Cells were gated to select for live cells, single cells, and mtKeima positive cells sequentially. Flow cytometry data were processed with FlowJo (version 10.2.2), and the high mitophagy gate was drawn during analysis. To determine autophagic flux in wild type and Pptc7 knockout fibroblasts, cells were treated with the 50 nM bafilomycin A for an acute timecourse. Cells were then washed in PBS, lysed in RIPA buffer, and processed for western blotting for LC3B. 30 μg total lysate was loaded onto a 4-20% gradient gel, blotted to nitrocellulose membrane, and probed for LC3B (Cell Signaling Technologies, 1:1000). Bands were visualized on a Li-COR Odyssey and analyzed using the associated ImageStudio software.

Yeast two hybrid analysis

Yeast two-hybrid analysis was performed as previously described³⁷. Briefly, the Matchmaker Gold Yeast Two-Hybrid System (Takara) was used. Human BNIP3 and human NIX were cloned into pGBKT7 and pGADT7 as previously described³⁷; for this study, full length human PPTC7 was cloned into pGBKT7 and pGADT7 using standard restriction-based cloning using the BamHI and XhoI restriction sites for pGADT7 and BamHI and NotI restriction sites for pGBKT7. Plasmids were verified by Sanger sequencing. Y2H Gold and Y187 yeast strains were transformed with bait (pGBKT7 plasmids and derivatives) and prey (pGADT7 plasmids and derivatives), respectively, by lithium acetate transformation. Haploid bait- and prey-expressing strains were mated on YPD plates (1% yeast extract, 2% peptone, and 2% glucose) for 24 h and diploids were subsequently selected on synthetic dextrose (SD; 0.7% yeast nitrogen base, 2% glucose, and amino acids) -leu-trp plates. Cells were grown to exponential phase in SD-leu-trp media, normalized to 0.5 OD 600 per mL, and cells were spotted on SD-leu-trp (permissive) and SD-leu-trp-his (selection) plates. Plates were then incubated at 30 °C prior to analysis.

Co-immunoprecipitation assays

U2OS cells were transfected with PPTC7-FLAG using Lipofectamine 2000 (Thermo Fisher). Where indicated, cells were treated with 1 mM diferoprone (DFP). Cells were lysed on ice for 30 min in Tris-Triton lysis buffer (50 mM Tris-Cl pH 7.5, 150 mM NaCl, 10% glycerol, 1 mM EDTA, 1 mM EGTA, 5 mM MgCl2, 1 mM β-glycerophosphate, and 1% Triton) containing protease inhibitor cocktail (Rowe Scientific; CP2778) and PhosSTOP EASYpack Phosphatase Inhibitor Cocktail (Roche; 4906837001). Cell lysates were collected by centrifugation at 18,800 × g for 10 min at 4 °C. To affinity-precipitate exogenously expressed FLAG-tagged PPTC7, cell lysates were incubated in a rotating incubator for 1 h at 4 °C with anti-FLAG affinity gel (Sigma; A2220). The immunoprecipitates were washed with Tris-Triton lysis buffer 5 times prior to elution with Bolt™ LDS Sample Buffer, electrophoresis on Bolt 4–12% gradient gels, and Western blotting.

Data and statistical analysis and software

Statistical analysis was carried out using either Microsoft Excel or Prism using the statistical tests listed in the associated figure legends. Volcano plots and linear regression plots were generated using Microsoft Excel. Bar graphs, box plots, and dot plot distributions (e.g., Fig. 2d) were generated using Prism 9 (GraphPad). All graphs were imported into Adobe Illustrator for final adjustments. Seahorse figures were generated using Wave software, Excel, or Prism. FACS plots were generated using FlowJo. The GO analysis graph plotted in Fig. 1f was generated at: http://bioinformatics.sdstate.edu/go/. BioRender was used for some graphics.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The raw data for the proteomics and phosphoproteomics datasets in this study have been deposited to the MassIVE database under the accession number MSV000091194. Source data are provided with this paper.

References

Song, J., Herrmann, J. M. & Becker, T. Quality control of the mitochondrial proteome. Nat. Rev. Mol. Cell Biol. 22, 54–70 (2021).
CAS PubMed Google Scholar
Pickles, S., Vigié, P. & Youle, R. J. Mitophagy and quality control mechanisms in mitochondrial maintenance. Curr. Biol. 28, R170–R185 (2018).
CAS PubMed PubMed Central Google Scholar
Wang, L., Qi, H., Tang, Y. & Shen, H.-M. Post-translational modifications of key machinery in the control of mitophagy. Trends Biochem. Sci. 45, 58–75 (2020).
PubMed Google Scholar
Kravic, B. et al. In mammalian skeletal muscle, phosphorylation of TOMM22 by protein kinase CSNK2/CK2 controls mitophagy. Autophagy 14, 311–335 (2018).
CAS PubMed PubMed Central Google Scholar
Walter, C. et al. Global kinome profiling reveals DYRK1A as critical activator of the human mitochondrial import machinery. Nat. Commun. 12, 4284 (2021).
ADS CAS PubMed PubMed Central Google Scholar
Grimsrud, P. A. et al. A quantitative map of the liver mitochondrial phosphoproteome reveals posttranslational control of ketogenesis. Cell Metab 16, 672–683 (2012).
CAS PubMed PubMed Central Google Scholar
Guo, X. et al. Ptc7p dephosphorylates select mitochondrial proteins to enhance metabolic function. Cell Rep 18, 307–313 (2017).
CAS PubMed PubMed Central Google Scholar
Chung, J. et al. Erythropoietin signaling regulates heme biosynthesis. eLife 6, e24767 (2017).
Niemi, N. M. & MacKeigan, J. P. Mitochondrial phosphorylation in apoptosis: flipping the death switch. Antioxid. Redox Signal. 19, 572–582 (2013).
CAS PubMed PubMed Central Google Scholar
Morgenstern, M. et al. Quantitative high-confidence human mitochondrial proteome and its dynamics in cellular context. Cell Metab 33, 2464.e18–2483.e18 (2021).
Google Scholar
Niemi, N. M. & Pagliarini, D. J. The extensive and functionally uncharacterized mitochondrial phosphoproteome. J. Biol. Chem. 297, 100880 (2021).
CAS PubMed PubMed Central Google Scholar
Pagliarini, D. J. et al. A mitochondrial protein compendium elucidates complex i disease biology. Cell 134, 112–123 (2008).
CAS PubMed PubMed Central Google Scholar
Rath, S. et al. MitoCarta3.0: an updated mitochondrial proteome now with sub-organelle localization and pathway annotations. Nucleic Acids Res. 49, D1541–D1547 (2021).
CAS PubMed Google Scholar
Niemi, N. M. et al. Pptc7 is an essential phosphatase for promoting mammalian mitochondrial metabolism and biogenesis. Nat. Commun. 10, 3197 (2019).
ADS CAS PubMed PubMed Central Google Scholar
Rensvold, J. W. et al. Defining mitochondrial protein functions through deep multiomic profiling. Nature 606, 1–7 (2022).
Google Scholar
Ruzankina, Y. et al. Deletion of the developmentally essential gene ATR in adult mice leads to age-related phenotypes and stem cell loss. Cell Stem Cell 1, 113–126 (2007).
CAS PubMed PubMed Central Google Scholar
Carr, R. M. & Ahima, R. S. Pathophysiology of lipid droplet proteins in liver diseases. Exp. Cell Res. 340, 187–192 (2016).
CAS PubMed Google Scholar
Okumura, T. Role of lipid droplet proteins in liver steatosis. J. Physiol. Biochem. 67, 629–636 (2011).
CAS PubMed Google Scholar
Sun, N. et al. A fluorescence-based imaging method to measure in vitro and in vivo mitophagy using mt-Keima. Nat. Protoc. 12, 1576–1587 (2017).
CAS PubMed Google Scholar
Mann, M. et al. Mitochondrial phosphoproteomes are functionally specialized across tissues. Preprint at https://doi.org/10.1101/2022.03.23.485457 (2022).
He, Y.-L. et al. BNIP3 phosphorylation by JNK1/2 promotes mitophagy via enhancing its stability under hypoxia. Cell Death Dis 13, 1–13 (2022).
MathSciNet Google Scholar
Floyd, B. J. et al. Mitochondrial protein interaction mapping identifies regulators of respiratory chain function. Mol. Cell 63, 621–632 (2016).
CAS PubMed PubMed Central Google Scholar
Allen, G. F. G., Toth, R., James, J. & Ganley, I. G. Loss of iron triggers PINK1/Parkin-independent mitophagy. EMBO Rep. 14, 1127–1135 (2013).
CAS PubMed PubMed Central Google Scholar
Aoki, Y. et al. Phosphorylation of Serine 114 on Atg32 mediates mitophagy. Mol. Biol. Cell 22, 3206–3217 (2011).
CAS PubMed PubMed Central Google Scholar
Kanki, T. et al. Casein kinase 2 is essential for mitophagy. EMBO Rep 14, 788–794 (2013).
CAS PubMed PubMed Central Google Scholar
Chen, G. et al. A regulatory signaling loop comprising the PGAM5 phosphatase and CK2 controls receptor-mediated mitophagy. Mol. Cell 54, 362–377 (2014).
CAS PubMed Google Scholar
Murakawa, T. et al. Bcl-2-like protein 13 is a mammalian Atg32 homologue that mediates mitophagy and mitochondrial fragmentation. Nat. Commun. 6, 7527 (2015).
ADS PubMed Google Scholar
Poole, L. P., Bock-Hughes, A., Berardi, D. E. & Macleod, K. F. ULK1 promotes mitophagy via phosphorylation and stabilization of BNIP3. Sci. Rep. 11, 1–15 (2021).
Google Scholar
Rogov, V. V. et al. Phosphorylation of the mitochondrial autophagy receptor Nix enhances its interaction with LC3 proteins. Sci. Rep. 7, 1131 (2017).
ADS PubMed PubMed Central Google Scholar
Zhu, Y. et al. Modulation of serines 17 and 24 in the LC3-interacting region of Bnip3 determines pro-survival mitophagy versus apoptosis. J. Biol. Chem. 288, 1099–1113 (2013).
CAS PubMed Google Scholar
Liu, K. E. & Frazier, W. A. Phosphorylation of the BNIP3 C-terminus inhibits mitochondrial damage and cell death without blocking autophagy. PLoS ONE 10, e0129667 (2015).
PubMed PubMed Central Google Scholar
da Silva Rosa, S. C. et al. BNIP3L/Nix-induced mitochondrial fission, mitophagy, and impaired myocyte glucose uptake are abrogated by PRKA/PKA phosphorylation. Autophagy 17, 2257–2272 (2021).
PubMed Google Scholar
Nguyen-Dien, G. T. et al. FBXL4 suppresses mitophagy by restricting the accumulation of NIX and BNIP3 mitophagy receptors. EMBO J. 12, e112767 (2023).
Google Scholar
Elcocks, H. et al. FBXL4 ubiquitin ligase deficiency promotes mitophagy by elevating NIX levels. EMBO J. 42, e112799 (2023).
CAS PubMed PubMed Central Google Scholar
Cao, Y. et al. A mitochondrial SCF-FBXL4 ubiquitin E3 ligase complex degrades BNIP3 and NIX to restrain mitophagy and prevent mitochondrial disease. EMBO J. 42, e113033 (2023).
CAS PubMed Google Scholar
Alsina, D. et al. FBXL4 deficiency increases mitochondrial removal by autophagy. EMBO Mol. Med. https://doi.org/10.15252/emmm.201911659 (2020).
Gok, M. O. et al. The outer mitochondrial membrane protein TMEM11 demarcates spatially restricted BNIP3/BNIP3L-mediated mitophagy. J. Cell Biol. 222, e202204021 (2023).
CAS PubMed PubMed Central Google Scholar
Schäfer, J. A., Bozkurt, S., Michaelis, J. B., Klann, K. & Münch, C. Global mitochondrial protein import proteomics reveal distinct regulation by translation and translocation machinery. Mol. Cell 82, 435.e7–446.e7 (2022).
Google Scholar
Rhee, H.-W. et al. Proteomic mapping of mitochondria in living cells via spatially restricted enzymatic tagging. Science 339, 1328–1331 (2013).
ADS CAS PubMed PubMed Central Google Scholar
Hung, V. et al. Proteomic mapping of cytosol-facing outer mitochondrial and ER membranes in living human cells by proximity biotinylation. eLife 6, e24463 (2017).
Juneau, K., Nislow, C. & Davis, R. W. Alternative splicing of PTC7 in Saccharomyces cerevisiae determines protein localization. Genetics 183, 185–194 (2009).
CAS PubMed PubMed Central Google Scholar
Williams, C. C., Jan, C. H. & Weissman, J. S. Targeting and plasticity of mitochondrial proteins revealed by proximity-specific ribosome profiling. Science 346, 748–751 (2014).
ADS CAS PubMed PubMed Central Google Scholar
Huttlin, E. L. et al. The BioPlex network: a systematic exploration of the human interactome. Cell 162, 425–440 (2015).
CAS PubMed PubMed Central Google Scholar
Huttlin, E. L. et al. Architecture of the human interactome defines protein communities and disease networks. Nature 545, 505–509 (2017).
ADS CAS PubMed PubMed Central Google Scholar
Huttlin, E. L. et al. Dual proteome-scale networks reveal cell-specific remodeling of the human interactome. Cell 184, 3022.e28–3040.e28 (2021).
Google Scholar
Kolitsida, P. et al. Phosphorylation of mitochondrial matrix proteins regulates their selective mitophagic degradation. Proc. Natl Acad. Sci. USA 116, 20517–20527 (2019).
ADS CAS PubMed PubMed Central Google Scholar
Kolitsida, P. et al. The pyruvate dehydrogenase complex regulates mitophagic trafficking and protein phosphorylation. Life Sci. Alliance 6, e202302149 (2023).
PubMed PubMed Central Google Scholar
Martín-Montalvo, A. et al. The phosphatase Ptc7 induces coenzyme Q biosynthesis by activating the hydroxylase Coq7 in yeast. J. Biol. Chem. 288, 28126–28137 (2013).
PubMed PubMed Central Google Scholar
Quadros, R. M. et al. Easi-CRISPR: a robust method for one-step generation of mice carrying conditional and insertion alleles using long ssDNA donors and CRISPR ribonucleoproteins. Genome Biol. 18, 92 (2017).
PubMed PubMed Central Google Scholar
Frezza, C., Cipolat, S. & Scorrano, L. Organelle isolation: functional mitochondria from mouse liver, muscle and cultured filroblasts. Nat. Protoc. 2, 287–295 (2007).
CAS PubMed Google Scholar

Download references

Acknowledgements

We would like to thank members of the Pagliarini lab and the Niemi lab for discussions and critical evaluation of this work. We would like to thank Sangeeta Adak and the Diabetes Models Phenotyping Core (P30DK020579) for assistance in the Biochemical analysis of tissue and serum. This work was supported by R01DK098672 (to D.J.P.), P41GM108538 (National Center for Quantitative Biology of Complex Systems, to D.J.P. and J.J.C.), R35GM151130 (to N.M.N.), the Diabetes Research Center at Washington University (P30DK020579 Pilot and Feasibility funds to N.M.N.), T32G002760 (Genomic Sciences Training Program training grant support to L.R.S.), R35GM137894 (to J.R.F.), and funds from the BJC Investigators Program (to D.J.P.). L.W. was supported by the MilliporeSigma Predoctoral Fellowship in Honor of Dr. Gerty T. Cori at Washington University. Support for this research was provided by the University of Wisconsin–Madison, Department of Biochemistry and Office of the Vice Chancellor for Research and Graduate Education with funding from the Wisconsin Alumni Research Foundation (to M.P.K.), and by NIH grants, R01DK101573, R01DK102948, and RC2DK125961 (to A.D.A.). J.K.P. was supported by Australian National Health and Medical Research Council grants (APP1183915 and APP2019993), a Brain Foundation Research grant (2020), and an Australian Research Council Future Fellowship (FT180100172). J.K.P and K.-L.K. were supported by a Mito Foundation Incubator Grant (2022) and Mito Foundation Ph.D. scholarship, respectively. We would like to thank Kathy Krentz and C. Dustin Rubenstein from the Animal Models and Genome Editing Cores at the University of Wisconsin-Madison for generating of the Pptc7 floxed model. The generation of the Pptc7 floxed model was supported by the University of Wisconsin Carbone Cancer Center Support Grant (P30CA014520). We thank the Genome Engineering and iPSC Center (GESC@MGI) at the Washington University in St. Louis for creating the Pptc7/Bnip3/Bnip3l triple knockout (TKO) MEFs lines. We would like to acknowledge the Washington University Diabetes Research Center (P30DK020579) for providing access to a Seahorse instrument used to generate data in this manuscript.

Author information

Authors and Affiliations

Morgridge Institute for Research, Madison, WI, 53715, USA
Natalie M. Niemi, Joshua J. Coon & David J. Pagliarini
Department of Biochemistry and Molecular Biophysics, Washington University School of Medicine, St. Louis, MO, 63110, USA
Natalie M. Niemi, Lianjie Wei, Merima Forny & David J. Pagliarini
Department of Biomolecular Chemistry, University of Wisconsin-Madison, Madison, WI, 53706, USA
Lia R. Serrano, Catherine E. Balnis, Evgenia Shishkova & Joshua J. Coon
Department of Chemistry, University of Wisconsin-Madison, Madison, WI, 53706, USA
Laura K. Muehlbauer & Joshua J. Coon
Department of Cell Biology and Physiology, Washington University School of Medicine, St. Louis, MO, 63110, USA
Andrew J. Smith & David J. Pagliarini
School of Biomedical Sciences, Faculty of Medicine, University of Queensland, Brisbane, QLD, 4072, Australia
Keri-Lyn Kozul & Julia K. Pagan
Department of Cell Biology, University of Texas Southwestern Medical Center, Dallas, TX, 75390, USA
Olivia M. Connor & Jonathan R. Friedman
Department of Biochemistry, University of Wisconsin-Madison, Madison, WI, 53706, USA
Edrees H. Rashan, Kathryn L. Schueler, Mark P. Keller, Alan D. Attie & David J. Pagliarini
National Center for Quantitative Biology of Complex Systems, Madison, WI, 53706, USA
Evgenia Shishkova & Joshua J. Coon
The University of Queensland, Institute for Molecular Bioscience, Brisbane, QLD, 4072, Australia
Julia K. Pagan
The University of Queensland Diamantina Institute, Faculty of Medicine, The University of Queensland, Brisbane, QLD, 4102, Australia
Julia K. Pagan
Department of Genetics, Washington University School of Medicine, St. Louis, MO, 63110, USA
David J. Pagliarini

Authors

Natalie M. Niemi
View author publications
You can also search for this author in PubMed Google Scholar
Lia R. Serrano
View author publications
You can also search for this author in PubMed Google Scholar
Laura K. Muehlbauer
View author publications
You can also search for this author in PubMed Google Scholar
Catherine E. Balnis
View author publications
You can also search for this author in PubMed Google Scholar
Lianjie Wei
View author publications
You can also search for this author in PubMed Google Scholar
Andrew J. Smith
View author publications
You can also search for this author in PubMed Google Scholar
Keri-Lyn Kozul
View author publications
You can also search for this author in PubMed Google Scholar
Merima Forny
View author publications
You can also search for this author in PubMed Google Scholar
Olivia M. Connor
View author publications
You can also search for this author in PubMed Google Scholar
Edrees H. Rashan
View author publications
You can also search for this author in PubMed Google Scholar
Evgenia Shishkova
View author publications
You can also search for this author in PubMed Google Scholar
Kathryn L. Schueler
View author publications
You can also search for this author in PubMed Google Scholar
Mark P. Keller
View author publications
You can also search for this author in PubMed Google Scholar
Alan D. Attie
View author publications
You can also search for this author in PubMed Google Scholar
Jonathan R. Friedman
View author publications
You can also search for this author in PubMed Google Scholar
Julia K. Pagan
View author publications
You can also search for this author in PubMed Google Scholar
Joshua J. Coon
View author publications
You can also search for this author in PubMed Google Scholar
David J. Pagliarini
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

N.M.N. and D.J.P. wrote the manuscript. N.M.N. and D.J.P. conceived the overall project and its design. N.M.N., L.R.S., L.K.M., C.E.B., A.J.S., E.S., D.J.P., and J.J.C. generated and prepared samples, performed proteomics or phosphoproteomics, or assisted with data analysis and interpretation of proteomics and/or phosphoproteomics data. L.W., K.-L.K., and J.K.P. generated reagents, performed experiments, and analyzed data regarding the mitophagy experiments. N.M.N., M.F., and E.H.R. performed experiments and analyzed data associated with cellular metabolic experiments. N.M.N., K.L.S., M.P.K., A.D.A., and D.J.P. assisted in generating and providing husbandry to the animals in this study, as well as advising on experimental design and data interpretation. N.M.N., A.J.S., K.-L.K., O.M.C., J.R.F., J.K.P., and D.J.P. designed and performed experiments or analyzed data regarding the BNIP3/NIX-PPTC7 interactions. All authors reviewed and edited the manuscript.

Corresponding authors

Correspondence to Natalie M. Niemi or David J. Pagliarini.

Ethics declarations

Competing interests

The authors declare the following competing interests: J.J.C. is a consultant for Thermo Fisher Scientific, 908 Devices, and Seer. The remaining authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Deepali Rathore, Shusheng Wang and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Niemi, N.M., Serrano, L.R., Muehlbauer, L.K. et al. PPTC7 maintains mitochondrial protein content by suppressing receptor-mediated mitophagy. Nat Commun 14, 6431 (2023). https://doi.org/10.1038/s41467-023-42069-w

Download citation

Received: 10 March 2023
Accepted: 28 September 2023
Published: 13 October 2023
DOI: https://doi.org/10.1038/s41467-023-42069-w

This article is cited by

Mitophagy mediated by BNIP3 and NIX protects against ferroptosis by downregulating mitochondrial reactive oxygen species
- Shun-ichi Yamashita
- Yuki Sugiura
- Tomotake Kanki
Cell Death & Differentiation (2024)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Subjects

Abstract

Similar content being viewed by others

Introduction

Results

PPTC7 maintains mitochondrial protein levels in adult mice

Loss of Pptc7 broadly decreases mitochondrial metabolism in isolated cells

Loss of mitochondrial content is mediated by the mitophagy receptors Bnip3 and Nix

Phosphoproteomic analyses reveal BNIP3 and NIX as candidate PPTC7 substrates

PPTC7 can directly interact with and dephosphorylate BNIP3 and NIX

Discussion

Methods

Generation of a conditional floxed Pptc7 mouse model

Generation of an inducible global Pptc7 knockout mouse

Genotyping analysis

Generation and validation of mouse embryonic fibroblasts

Proteomic and phosphoproteomic analysis

TMT phosphoproteomics

Tissue lysis and protein digestion

TMT labeling

Phosphopeptide enrichment

Offline fractionation

LC-MS acquisition

Data processing

Label-free proteomics

Protein extraction and digestion

LC-MS acquisition method

Data processing

Phosphoproteomics on phosphatase-treated isolated mitochondria

Sample preparation

LC-MS/MS analysis

MS data analysis

Quantification of ketones, triacylglycerols, and cholesterol

Metabolic assays

Citrate synthase assays

Western blotting and antibodies

mtDNA analysis

Generation of Pptc7/Bnip3/Nix triple knockout (TKO) cells

Mitophagy and autophagy assays

Yeast two hybrid analysis

Co-immunoprecipitation assays

Data and statistical analysis and software

Reporting summary

Data availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding authors

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Supplementary information

Source data

Rights and permissions

About this article

Cite this article

Share this article

This article is cited by

Comments

Search

Quick links