USP16 regulates mitochondrial function by deubiquitinating TOM40, a core component of the translocase in the mitochondrial outer membrane

Abstract

Polycomb Repressive Complex 1 (PRC1) and its antagonistic H2A deubiquitinase USP16 have been regarded as nuclear proteins acting on chromatin. However, our recent studies revealed that a fraction of PRC1 and USP16 also localizes to mitochondria, where PRC1 plays a critical role in maintaining mitochondrial function. The role of USP16 in this context, however, has not yet been investigated. This study shows that USP16 interacts with and stabilizes TOM40, the core subunit of the TOM complex. USP16 deubiquitinates K48-linked ubiquitin chains from lysine 175, 184, and 309 of TOM40. USP16 knockout disrupts mitochondrial homeostasis, resulting in impaired membrane potential, elevated reactive oxygen species, reduced ATP synthesis, and altered oxygen consumption rates. Expression of FLAG-tagged USP16 restores TOM40 protein levels and rescues mitochondrial function. TOM40 re-expression in USP16 KO cells partially restores mitochondrial function, confirming its essential role. In summary, this study reveals that USP16 regulates mitochondrial function through TOM40.

Introduction

Mitochondria are essential organelles that regulate cellular energy metabolism, reactive oxygen species (ROS) homeostasis, and apoptotic signaling pathways [1–3]. These functions are vital for maintaining cellular health and responding to metabolic and environmental challenges. Most mitochondrial proteins are encoded in the nucleus but imported into the mitochondria. Thus, the functional integrity of mitochondria relies on the precise regulation of protein import, folding, and turnover, all of which are orchestrated by a range of molecular machines [4–6]. Of these, the translocase of the mitochondrial outer membrane (TOM complex) plays a pivotal role in mediating the import of nucleus-encoded proteins [7].

Post-translational modifications (PTMs), such as phosphorylation, acetylation, and ubiquitination, are key regulatory mechanisms that govern protein stability, activity, and subcellular localization [8–10]. Ubiquitination, in particular, plays a dual role by marking proteins for degradation via the ubiquitin-proteasome system (UPS) or for modulating their non-degradative functions [11, 12]. Ubiquitin itself contains seven lysine residues, K6, K11, K27, K29, K33, K48, K63; each can serve as an attachment point for polyubiquitin chain formation, enabling distinct signaling pathways and biological functions [13]. Deubiquitinases (DUBs), on the other hand, remove ubiquitin moieties from target proteins and are critical regulators of these processes [14]. While much research has focused on DUBs in nuclear and cytoplasmic pathways [15–17], their roles in mitochondria are only beginning to be understood [18–21].

Ubiquitin-specific protease 16 (USP16) is a well-characterized DUB that has been extensively studied in the context of chromatin biology [22, 23]. USP16 is known to deubiquitinate histone H2A at Lys119, influencing chromatin dynamics, gene expression, and cell cycle progression [23, 24]. Beyond histones, USP16 has been implicated in regulating other substrates involved in diverse processes, such as ribosomal biogenesis, immune responses, and cancer progression [25–29]. It has also been reported to interact with mitochondrial matrix protein DBT under overexpression conditions and with the mitochondria outer membrane interacting protein Dynamin-related protein 1 (Drp1) [30, 31], raising an intriguing possibility about its involvement in regulating mitochondrial function. In a recent study, we reported that a subset of the histone H2A ubiquitin ligase Polycomb Repressive Complex 1 (PRC1) and one of the H2A deubiquitinase USP16 also localize to mitochondria, where PRC1 plays a critical role in maintaining mitochondrial function [32]. However, the function of USP16 remains unclear.

In this study, we reveal a critical function of USP16 in mitochondria. We further validate our conclusion that a fraction of USP16 is localized to mitochondria by transmission electron microscopy. Importantly, we showed that it is required for maintaining mitochondrial function by stabilizing TOM40 through deubiquitination. Using an integrated approach combining proteomics, immunofluorescence imaging, and biochemical assays, we identified TOM40 as a direct substrate of USP16. TOM40 is the central pore-forming subunit of the TOM complex indispensable for maintaining mitochondrial homeostasis [33, 34]. Loss of USP16 results in the destabilization of TOM40, leading to impaired mitochondrial membrane potential, increased ROS production, and decreased ATP synthesis. These mitochondrial dysfunctions highlight the essential role of USP16 in regulating mitochondrial homeostasis. Importantly, the restoration of TOM40 expression in USP16-deficient cells partially rescues these defects, underscoring the mechanistic link between USP16 and TOM40 stability. These findings provide novel insights into the non-nuclear functions of USP16 and its role in regulating mitochondrial functions through the stabilization of a key mitochondrial protein.

Materials and methods

Reagents and antibodies

Reagents used in this article include: Puromycin dihydrochloride (#B7587, ApexBio), Polybrene (#H8761, Solarbio), MG132 (#HY-13259, Med Chem Express), Cycloheximide (#HY-12320, Med Chem Express), Triton X-100 (#T8200, Solarbio), Proteinase K Solution (#P1120, Solarbio), Sucrose (#S8271, Solarbio), Bovine Serum Albumin (#ST023, Beyotime), Anti-HA beads (#C0323, BioMag), Protein A + G Agarose (#P2055, Beyotime), Anti-FLAG® M2 Affinity Gel (#A2220, Millipore), FLAG Peptide (#T510060, Sangon Biotech), Anti-Diglycine Lysine Antibody Conjugated Agarose Beads (#PTM-1104, PTM BIO), Pierce™ GST Protein Interaction Pull-Down Kit (#21516, Thermo Fisher Scientific).

Antibodies used in this article include: USP16 (#sc-390683, Santa Cruz), VDAC1 (#CY5416, Abways), HSP60 (#15282-1-AP, Proteintech), ubH2A (#8240, Cell Signaling Technology), Histone H3 (#68345-1-Ig, Proteintech), TOM40 (#PTM-5693, PTM BIO), PGAM5 (#28445-1-AP, Proteintech), CISY (#16131-1-AP, Proteintech), FLAG-Tag (#20543-1-AP, Proteintech), V5-Tag (#13202, Cell Signaling Technology), HA-Tag (#51064-2-AP, Proteintech), GST-Tag (#2622, Cell Signaling Technology), His-Tag (#2365, Cell Signaling Technology), β-Tubulin (#ABL1030, Abbkine).

Plasmids and transfection

The lip2000 Efficient DNA Transfection Reagent (#ML0201, Kermey) was used to transfect plasmid DNA into cells according to the supplier's instructions. Firstly, we added 4 μg of plasmid to 250 μL of serum-free DMEM medium (#10-013-CVRC, Corning) and mixed well. Meanwhile, we added 10 μL of transfection reagent to 250 μL of serum-free DMEM medium and mixed them thoroughly. Then, the two solutions were combined and allowed to stand at room temperature for 20 min, and then the mixture was added to 2 × 10⁵ cells (taking one well of a six-well cell culture plate as an example). The cells were then placed in a 37 °C, 5% CO₂ incubator for further cultivation.

The transfection of plasmids based on the lentiCRISPR v2 backbone was performed by infecting cells with packaged lentivirus in the presence of polybrene. Firstly, pMD2.G, psPAX2, and the target plasmid were transfected into HEK293T cells at a ratio of 1:3:4. After 72 h, the packaged lentivirus in the medium was collected and used to infect HeLa cells. The infected HeLa cells were then placed in a 37 °C, 5% CO₂ incubator for further cultivation.

Small interfering RNA (siRNA) and transfection

The cell transfection using siRNA was carried out with RNA TransMate (#E607402, Sangon Biotech). First, siRNA (siRNA sequences are listed in Table S2) and RNA TransMate were separately diluted in serum-free DMEM medium and then mixed and left to stand at room temperature for 10 min. The mixture was inoculated and added to HeLa cells, which were then placed in a 37 °C, 5% CO₂ incubator for cultivation. After 48-72 h, experimental studies were conducted.

Construction of USP16 KO cell line

CRISPR-Cas9 gene knockout technology [35] was used to construct a USP16 KO HeLa cell line. First, we designed sgRNAs through the website “http://crispor.gi.ucsc.edu/” and synthesized a pair of sequence-complementary DNA Oligos according to the sgRNA template sequence (Table S3). Then, we constructed sgRNA expression vectors (lentiCRISPR v2-USP16 KO sgRNA1, lentiCRISPR v2-USP16 KO sgRNA2, and lentiCRISPR v2-USP16 KO sgRNA3, as shown in Table S1). By lentiviral infection, lentiCRISPR v2-USP16 KO sgRNA1 and lentiCRISPR v2-USP16 KO sgRNA2 were co-transfected into HeLa cells, and lentiCRISPR v2-USP16 KO sgRNA3 single plasmid was transfected into HeLa cells. Puromycin dihydrochloride (puro) was used to obtain two resistant colonies. By immunoblotting, the USP16 KO cells were identified. Furthermore, primers were designed on both sides of the sgRNA editing site (Table S3). Changes in the position of the gel bands were observed by PCR amplification and agarose gel electrophoresis. The target gel bands were recovered and sent to TsingkeBiotechnology Co., Ltd. for Sanger sequencing. The sequencing results were analyzed using DNAMAN 8 software to confirm USP16 KO and to identify DNA sequence alterations in the KO clones.

Construction of FLAG-USP16 KI Cell Line

Prime editing (PE) [36] was used to directionally insert a FLAG tag sequence in frame with the N-terminus of the USP16 protein for in situ labeling of the target gene. First, we selected a sequence from the N-terminus of USP16 and designed sgRNAs using the website “http://crispor.gi.ucsc.edu/”. Then, we used the website “https://www.pridict.it/” to design epegRNA primers for the editing site, including Spacer, RTT-PBS, and Scaffold primers (Table S3). We constructed the epegRNA plasmid (FLAG-USP16KI-epegRNA, Table S1). Next, we co-transfected the FLAG-USP16KI-epegRNA plasmid with the pCMV-PEMax-P2A-hMLH1dn-EGFP plasmid into HeLa cells. Positive monoclonal cells expressing EGFP fluorescence were selected using an inverted fluorescence microscope. Finally, we verified successful FLAG tag insertion into USP16 by immunofluorescence (IF) and immunoblotting.

RNA isolation and real-time PCR

Total RNA was extracted using the RNA extraction kit (#RC112-01, Vazyme) according to the manufacturer’s instructions. The HiScript II 1st Strand cDNA Synthesis Kit (#R212-01, Vazyme) was used to reverse transcribe the RNA into cDNA. Using primers listed in Table S3, PCR amplification was conducted and real-time fluorescence measurement using the ChamQ Universal SYBR qPCR Master Mix (#Q711-02, Vazyme) in a real-time quantitative PCR instrument (LightCycler 480 Ⅱ, Roche). The relative expression levels of the genes of interest were calculated using the 2-ΔΔCt method with β-actin as an internal reference.

Immunofluorescence

Cells were seeded at a density of 2 × 10⁴ cells per dish in laser confocal microscopy-specific culture dishes and cultured for 36 h. Subsequently, the cells were fixed with 4% paraformaldehyde (#P1110, Solarbio) and permeabilized with 0.1% Triton X-100 for 20 min. The cells were then blocked with a 3% Bovine Serum Albumin (BSA) solution at room temperature for 30 min. Primary antibodies were added and incubated at 4 °C for 12 h, followed by incubation with fluorescently labeled secondary antibodies for 1 h. For multiplex staining, the cells were re-blocked with 3% BSA solution and incubated with another primary antibody and its corresponding secondary antibody (ensuring that the two primary antibodies are from different species). After the final secondary antibody incubation, Antifade Mounting Medium with DAPI (#P0131, Beyotime) was added, and images were captured using a laser confocal microscope (LSM 980, ZEISS).

Immunoblotting

Protein samples for immunoblotting were prepared using an enhanced RIPA lysis buffer (#AR0102, Boster) and boiled in 1 × loading buffer. The protein samples were loaded and separated by SDS-PAGE, then transferred onto a PVDF membrane and incubated with primary antibodies. After incubation, the membrane was washed with 1 × TBST. Subsequently, the membrane was incubated with secondary antibodies for 1 h, followed by further washing with 1 × TBST. The membrane was then developed using a highly sensitive multifunctional imaging system (ImageQuant 800, Amersham). Finally, the immunoblot images were analyzed for grayscale intensity using ImageJ software.

Co-immunoprecipitation

Cells were pelleted, and IP cell lysis buffer (#P0013, Beyotime) was added. The lysed cells were centrifuged at 12,000 rpm for 10 min at 4 °C, and the supernatant was collected. A portion of the supernatant was taken as the input protein sample, while the remaining supernatant was mixed with the primary antibody and incubated overnight at 4 °C for immunoprecipitation. The protein-antibody complex solution was then thoroughly mixed with Protein A + G Agarose at 4 °C for 3 h. After mixing, the Agarose was washed five times with TBS buffer, and 5 × SDS-PAGE protein loading buffer was added. The mixture was heated in a 100 °C boiling water bath for 5 min to denature the proteins, followed by centrifugation at 5800 rpm for 1 min at 4 °C to remove the solid. The resulting supernatant was used for subsequent immunoblot analysis.

GST pull-down assay

Purified GST‑USP16 or GST was incubated with His‑TOM40 at 4 °C overnight, then captured with glutathione‑Sepharose beads. After washing, bound proteins were eluted and analyzed by western blotting using anti‑GST and anti‑His antibodies.

Immunoelectron microscopy

For immunoelectron microscopy, cells were first treated with a specialized immunoelectron microscopy fixative (#G1124, Servicebio), dehydrated in a graded ethanol series and embedded in LR White resin (#14381-UC, Head Biotechnology). Ultrathin sections of 70–80 nm were mounted on nickel grids and blocked with 1% BSA at room temperature. The sections were incubated with a primary antibody mixture (FLAG antibody) diluted in 1% BSA at 4 °C for 12 h, followed by three washes with TBS buffer. Subsequently, the sections were labeled with 12 nm colloidal gold-conjugated goat anti-mouse secondary antibody for 2 h and washed three times with TBS buffer. The nickel grids were then stained in the dark with 2% uranyl acetate in saturated ethanol for 15 min, washed three times with 70% ethanol, rinsed three times with ddH₂O, and dried in a 37 °C oven for 10 min. Finally, the samples were observed, and images were captured using a transmission electron microscope (HT7800, HITACHI). Black 12 nm gold particles indicated positive signals.

Subcellular fractionation

Based on previous publications [37, 38], we successfully constructed 2 HeLa cell lines. One expressed 3 × HA-EGFP-tagged mitochondrial out member protein OMP25. The second cell line expressed 3 × HA-EGFP-tagged nuclear outer membrane protein Nesp4. The cell lines were verified by immunoblotting and by inverted fluorescence microscopy (see Supplementary Fig. S1K for the methodology). To isolate mitochondria, cold RSB hypotonic buffer (10 mM NaCl, 1.5 mM MgCl₂, and 10 mM Tris-HCl) was added to 3 × HA-EGFP-OMP25 expressing cells and allowed the cells to swell for 10 min. The cell suspension was then ruptured with a Dounce homogenizer, releasing the mitochondria. Anti-HA magnetic beads were then mixed with the mitochondrial suspension and incubated at 4 °C for 4 h. The mixture was placed on a magnetic rack for separation, yielding purified mitochondria. The same protocol was used to isolate nuclei from cells expressing 3 × HA-EGFP-Nesp4.

To obtain the cytoplasm, the cells were ruptured as described above. The homogenate was then transferred to a centrifuge tube and centrifuged at 1300 × g for 5 min at 4 °C. The supernatant was collected and further centrifuged at 10,000 × g for 15 min at 4 °C. The resulting supernatant was subject to another round of centrifugation at 14,000 × g for another 15 min at 4 °C. The final supernatant obtained was the cytoplasm.

Purification of mitochondria by sucrose step density gradient centrifugation

The cells were ruptured with a Dounce homogenizer as described in the previous section. The suspension was then centrifuged at 1300 × g for 5 min at 4 °C. The resulting supernatant was transferred to a high-speed centrifuge and centrifuged at 10,000 × g for 15 min at 4 °C to obtain a crude mitochondrial fraction. A sucrose density gradient solution was prepared in Ultraclear centrifuge tubes. The crude mitochondrial fraction was layered above the sucrose solution and spun in an ultracentrifuge at 60,000 × g for 20 min at 4 °C. After centrifugation, the mitochondrial fraction from the middle of the sucrose density gradient was collected by aspiration and spun at 7000 × g for 15 min at 4 °C, yielding purified mitochondria.

Protease K treatment assay

In this experiment, purified mitochondria were used and subjected to digestion with 100 ng/mL proteinase K on ice for 30 min, either in the absence or presence of 1% Triton X-100. Following digestion, the protein levels of USP16, VDAC1, and HSP60 were detected by immunoblotting.

Mass spectrometry

We utilized 4D Label-free ubiquitination proteomics qualitative technology from PTM Bio Inc. to perform mass spectrometric analysis of the ubiquitinated proteome. First, we isolated proteins from HeLa cells and digested them with trypsin. Then, based on the PTM method using anti-diglycine lysine antibody-conjugated agarose beads, we enriched modified peptides. The peptides were separated using an ultra-high performance liquid chromatography (UHPLC) system (NanoElute, Bruker Daltonics) and analyzed by mass spectrometry (timsTOF Pro, Bruker Daltonics). Using the raw data files obtained from mass spectrometry, we searched the ubiquitination modification proteomics database to identify proteins with ubiquitination sites.

Similarly, through PTM Bio Inc., we employed 4D Label-free proteomics quantitative technology for mass spectrometric analysis of the proteome. First, proteins from different groups of HeLa cells (Control and USP16 KO) were digested with trypsin. The resulting peptides were separated using a UHPLC (NanoElute, Bruker Daltonics) and analyzed by mass spectrometry (timsTOF Pro, Bruker Daltonics). Based on the raw data files obtained from mass spectrometry, we searched the proteomics database to identify proteins downregulated following USP16 KO.

For the Affinity Purification coupled with Mass Spectrometry (AP-MS) experiment to identify USP16-interacting proteins, mass spectrometric analysis was performed through BGI Genomics Co., Ltd. We first transfected HEK293T cells with either the pcDNA3.1 or pcDNA3.1-FLAG-USP16 plasmids. At 72 h post- transfection, the cell pellets were collected and treated with the DSP crosslinking reagent. The crosslinked cells were then lysed, and an Anti-FLAG® M2 Affinity Gel was used to perform immunoprecipitation. The bound proteins were eluted using FLAG Peptide through competitive elution. The eluted proteins were subsequently treated with DTT and subjected to SDS-PAGE gel electrophoresis, followed by staining with Coomassie Brilliant Blue. Target gel bands were excised and subjected to in-gel trypsin digestion. The resulting peptide fragments were separated using a UHPLC (UltiMate 3000, Thermo Fisher Scientific), and then ionized using nanoESI. The ionized peptides were transferred to a tandem mass spectrometer (Q-Exactive HF, Thermo Fisher Scientific) for data-dependent acquisition mode detection. Finally, the results were analyzed based on database search outcomes.

Mitochondrial superoxide measurement

The Mitochondrial Superoxide Assay Kit with MitoSOX Red (#S0061M, Beyotime) was used to detect changes in mitochondrial superoxide levels in cells according to the kit instructions. MitoSOX Red staining working solution was added to cells and incubated in a 37 °C, 5% CO₂ incubator for 30 min. After incubation, the cells were washed twice with PBS buffer. Then, they were observed and photographed under an inverted fluorescence microscope (Axio Observer, ZEISS), and the fluorescence intensity of each cell was analyzed using ZEN software.

Mitochondrial membrane potential assay

The Enhanced mitochondrial membrane potential assay kit with JC-1 (#C2003S, Beyotime) was used to detect mitochondrial membrane potential. According to the kit’s instructions, the JC-1 staining working solution was prepared, and 10µM of the CCCP positive solution was used to treat cells as a positive control group. Then, JC-1 staining working solution was added to cells from different treatment groups and the positive control group, and incubated at 37 °C for 20 min. After incubation, cells were washed twice with JC-1 staining buffer. Subsequently, cells were observed and photographed under an inverted fluorescence microscope, and the fluorescence intensity of each cell was analyzed using ZEN software. When the mitochondrial membrane potential is high, JC-1 aggregates in the mitochondrial matrix, forming JC-1 aggregates, that emit red fluorescence. When the mitochondrial membrane potential is low, JC-1 cannot aggregate in the mitochondrial matrix, and JC-1 exists as monomers, that emits green fluorescence. We measured changes in the mitochondrial membrane potential by calculating the relative ratio of green fluorescence to red fluorescence.

Intracellular ATP level detection

The ATP Assay Kit (#S0026, Beyotime) was used to measure ATP levels in cells from different groups. First, cells were pelleted and lysed using the lysis buffer provided in the kit. The lysed cell suspension was then centrifuged at 12,000 rpm for 10 min at 4 °C, and the supernatant was collected. Next, the protein concentration was quantified using the BCA Protein Concentration Assay Kit (#P0012, Beyotime). Samples with equal protein concentrations were then taken and mixed with the ATP assay working solution. The Relative Light Unit (RLU) values of cells from different groups were measured using the luminometer on the multifunctional microplate reader (INFINITE M PLEX, Tecan).

Mitochondrial permeability transition pore assay

The Mitochondrial Permeability Transition Pore Assay Kit or MPTP Assay Kit (#C2009S, Beyotime) was used to detect MPTP in cells from different groups. Appropriate volumes of Calcein AM staining solution, fluorescence quenching solution (CoCl₂), or Ionomycin positive control solution were added to cells from different groups, followed by incubation at 37 °C for 40 min. After incubation, the culture medium was replaced, and the cells were further incubated at 37 °C for 30 min to ensure complete hydrolysis of Calcein AM by intracellular esterase, resulting in the formation of green fluorescent Calcein. The culture medium was then removed, and the cells were washed three times with PBS buffer. Finally, the detection buffer was added, allowing observation and photography under a fluorescence microscope, and the fluorescence intensity of each cell was analyzed using ZEN software. After Calcein AM enters living cells, it is catalyzed by esterase to generate Calcein, which is non - membrane - permeable and emits strong green fluorescence. After treatment with CoCl₂, it forms non-fluorescent Calcein-2Co. Under normal conditions, the MPTP is closed, preventing CoCl₂ from entering the mitochondria. As a result, only the mitochondria exhibit green fluorescence. In the positive control group treated with Ionomycin, Ca²⁺ enters the cells, leading to the opening of the MPTP. Subsequently, Co²⁺ enters the mitochondria and forms non - fluorescent Calcein - 2Co. Therefore, we can observe the degree of mitochondrial MPTP opening in different groups of cells by measuring the green fluorescence intensity of mitochondrial Calcein.

Oxygen consumption rate (OCR) measurements

Cells were seeded into XF24 plates at a density of 10,000 cells per well and incubated overnight in a 37 °C, 5% CO₂ incubator. Simultaneously, the sensor plate containing XF calibration solution was hydrated overnight in a 37 °C, CO₂-free incubator. On the second day, the medium in the XF24 plate was replaced with XF assay medium (#103680-100, Agilent) containing 10 mM glucose, 1 mM pyruvate, and 2 mM glutamine, and the cells were incubated at 37 °C for 1 h. The fully hydrated sensor plate was then placed onto the cell plate. Following the instructions of the Seahorse XF Cell Mitochondrial Stress Test Kit (#103015-100, Agilent), 1.5 µM Oligomycin, 1 µM FCCP, and 0.5 µM Rotenone and Antimycin A were sequentially added. The oxygen consumption rate (OCR) of the cells from different groups was measured using the Seahorse XF24 analyzer (Agilent).

Evaluation of mitochondrial respiratory chain (MRC) activity

According to the manufacturer’s instructions, the activities of MRC complexes I, III, and V were detected using the MRC Activity Detection Kits (#BC0515, #BC3245, and #BC1445, Solarbio). The enzyme activities were assessed using a microplate reader at wavelengths of 340 nm (MRC complex I), 550 nm (MRC complex III), and 660 nm (MRC complex V).

Quantitative assessment of energy metabolism

We utilized the services of Metware Biotechnology Co., Ltd. to conduct qualitative and quantitative analysis of various energy metabolism-related substances in different cell groups [Control (parental HeLa cells), Vector-transfected HeLa cells, HeLa USP16 KO + Vector, and HeLa USP16 KO + TOM40] using LC-MS/MS analysis. Proteins were isolated from cells and digested with trypsin. The resulting peptides were then separated and analyzed using UHPLC (ACQUITY H-ClassD, Waters) and tandem mass spectrometry (QTRAP® 6500 + , SCIEX). A database was constructed based on standards, and quantitative analysis of the mass spectrometry data was performed using Multiple Reaction Monitoring on a triple quadrupole mass spectrometer. Finally, we processed the mass spectrometry data using Analyst 1.6.3 software, allowing for multi-angle analysis based on data characteristics to accurately identify differential metabolites between the different groups.

Statistical analysis

All statistical analyses were performed using GraphPad Prism 8.0 software. Quantitative data obtained from experiments with biological replicates are presented as mean ± SD, and “n” represents the number of biological reproducibility experiments. Statistical analysis was performed using Student’s t test, and p-value < 0.05 were considered statistically significant. */^# P < 0.05, **/^## P < 0.01, ***/^### P < 0.001, ns = not significant.

Ethics approval

This study was only performed in vitro cell experiments. No human participants and live vertebrate animals were involved. Accordingly, ethical approval and informed consent are not required. All experimental operations complied with institutional laboratory research guidelines and relevant specifications.

Results

USP16 is localized in mitochondria

USP16 is primarily localized in the cytoplasm and nucleus [39]; however, interactions of USP16 with mitochondria matrix protein DBT and outer membrane interacting protein Drp1 have been reported [30, 31]. To investigate a possible role of USP16 in mitochondria, we used an immunofluorescent assay. A green fluorescence-labeled USP16 partially overlapped with the red fluorescence-labeled mitochondria, indicating that USP16 localizes in or interacts with mitochondria (Fig. 1A).

Fig. 1. Localization of USP16 in mitochondria.

A Immunofluorescent staining with antibody to USP16 and red fluorescent mitochondrial tracker showing co-localization of a fraction USP16 with mitochondria in HeLa cells. scale bar = 20 μm, Enlarged: scale bar = 2 μm. B Immunogold electron microscopy refined the localization of USP16 on the mitochondrial membrane and in the matrix. Red arrows point to FLAG colloidal gold particles in the mitochondrial membrane, and blue arrows point to FLAG colloidal gold particles in the mitochondrial matrix; scale bar = 200 nm. C Images of HeLa cells stably expressing 3 × HA-EGFP-OMP25. Top panels: EGFP fluorescent and phase-contrast images. Scale bar = 100 μm. Middle panels: EGFP-OMP25 co-localized with mitochondria that were stained with a mitochondria red fluorescent tracker; scale bar = 10 μm. Bottom panels, EGFP-OMP25 is co-localized with mitochondria purified via immunoprecipitation (IP) with the anti-HA antibody. Red fluorescence signifies mitotracker. Scale bar = 50 μm. D Images of HeLa cells stably expressing 3 × HA-EGFP-Nesp4. Top panels: EGFP fluorescent and phase-contrast images. Scale bar = 100 μm. Middle panels, Nucleus were stained with Hoechst dye (blue). EGFP-Nesp4 co-localized with the nucleus. scale bar = 5 μm. Bottom panels, Nesp4 co-localized with nuclei purified via IP from the cell lines expressing 3 × HA-EGFP-Nesp4. Green fluorescence denotes Nesp4, and blue fluorescence signifies the nucleus. Scale bar = 50 μm. E Immunoblots confirming the subcellular localization of USP16 in total cell lysates, purified mitochondria, purified nucleus, and cytoplasm (after removal of nuclei and mitochondria). β-tubulin serves as a cytoplasm marker, VDAC1 serves as a mitochondrial marker, and H3 serves as a nuclear marker. F Proteinase K treatment demonstrated that USP16 was localized to both the mitochondrial outer membrane and matrix. VDAC1 serves as a mitochondrial outer membrane marker, and HSP60 serves as a mitochondrial matrix marker.

For precise localization, we generated USP16 KO as well as FLAG-USP16 knock-in (KI) HeLa cell lines using CRISPR-Cas9 and Prime editing technologies (Supplementary Fig. S1A–J). We also expressed FLAG-USP16 in HeLa cells by transient transfection of a FLAG-USP16 overexpression (OE) plasmid. After staining with an antibody conjugated to colloidal gold particles, the cells were examined with a transmission electron microscope. The dark gold particles with a size of 12 nm represent positive signals. We observed no immunogold labeling in USP16 KO cells. In contrast, colloidal gold marked mitochondrial membrane and matrix in FLAG-USP16 KI cells and a significantly increased labeling was observed in FLAG-USP16 OE cells (Fig. 1B).

To extend these positive findings, we purified mitochondria and nucleus via affinity purification from two HeLa cell lines that expressed 3 × HA-EGFP tagged mitochondria Outer Membrane Protein 25 (OMP25) or similarly tagged nuclear outer membrane protein Nesprin 4 (Nesp4), respectively (Supplementary Fig. S1K and Fig. 1C, D). We also purified cytoplasm (after removal of nuclei and mitochondria) using differential centrifugation. Immunoblots confirmed the presence of USP16 in the nucleus, cytoplasm, and mitochondria (Fig. 1E). Finally, a proteinase K protection assay was performed to refine the localization of USP16 in mitochondria. Upon treatment with proteinase K, the levels of USP16 were somewhat reduced. In contrast, the outer mitochondrial membrane protein VDAC1 was digested, while the mitochondrial matrix protein HSP60 remained intact. When 1% Triton X-100 was added, all three proteins were digested. Together, these results indicate USP16 is localized to both the mitochondrial outer membrane and matrix (Fig. 1F).

USP16 knockout impairs mitochondrial function

After confirming the mitochondrial localization of USP16, we investigated its role in regulating mitochondrial biology. First, we examined the USP16 knockout (KO) HeLa cell line (described in Supplementary Fig. S1A–F) for effects on multiple mitochondrial functions. This was then followed by the transient reintroduction of FLAG-USP16 into these KO cells to confirm the finding.

Immunoblots confirmed that USP16 was detected in the parental HeLa cells but absent in USP16 KO cells and that USP16 levels were restored effectively following the transient transfection of FLAG-USP16 (Fig. 2A, compare lanes 3, 5 with 2, compare lanes 4, 6 with 3, 5, quantified in 2B). Further immunofluorescence experiments demonstrated that the reintroduction efficiency of USP16 is close to 100% (see Supplementary Fig. S2). MitoSOX Red staining, as analyzed via confocal microscopy and flow cytometry, revealed increased mitochondrial superoxide production in USP16 KO cells. Its levels were significantly reduced when USP16 was reintroduced. These results indicate that the exogenous USP16 was able to rescue mitochondrial oxidative stress caused by USP16 KO (Fig. 2C, D). We then examined additional mitochondrial functions. Using the JC-1 probe, we performed a mitochondrial membrane potential (MMP) assay and observed a significant reduction of MMP in USP16 KO cells. The addition of CCCP, which causes a decrease in MMP, was used as a control (3rd column of panel E). This defect was partly rescued upon transient transfection with the FLAG-USP16 OE plasmid (Fig. 2E, F). We also found that there was a decreased of ATP production in USP16 KO cells. The level was partially restored by introducing FLAG-USP16 (Fig. 2G). Moreover, USP16 KO significantly increased mitochondrial permeability transition pore (MPTP) opening. Lonomycin, which increases MPTP opening, was used as a control. This defect was also partly rescued by FLAG-USP16 (Fig. 2H, I). Finally, to investigate the changes in cellular metabolic activities, we used the Seahorse XF Cell Mitochondrial Stress Test Kit to measure the oxygen consumption rate (OCR) in USP16 KO and FLAG-USP16-rescued cells. OCR was assessed following sequential addition of mitochondrial electron transport chain (ETC)-targeting drugs (oligomycin, FCCP, rotenone/antimycin A). Parameters such as basal respiration, ATP-linked respiration, proton leak, and maximal respiration were all significantly reduced in USP16 KO human cells, consistent with what we observed in USP16 KO mouse ESC cells [32]. Introduction of FLAG-USP16 restored all these OCR parameters to near normal levels (Fig. 2J–N). Collectively, these findings show that USP16 KO disrupts mitochondrial functions and introducing FLAG-USP16 can partially or largely mitigate these defects, demonstrating the critical role of USP16 in maintaining mitochondrial homeostasis.

Fig. 2. Negative effects of USP16 knockout on mitochondrial functions were rescued by exogenous expression of FLAG-USP16.

A Immunoblot results of USP16 protein levels in parental HeLa cells, USP16 KO cells and KO cells rescued by FLAG-USP16 via transient transfection. B Quantification of results in A (n = 3). C Representative images of mitochondrial superoxide changes, as detected with the MitoSOX Red fluorescent probe, in USP16 KO cells and KO cells rescued with exogenous USP16; scale bar = 20 μm. D Bar graph quantifying red fluorescence intensity from MitoSOX Red staining in (C) (n = 5). E Representative images of mitochondrial membrane potentials using JC-1 in USP16 KO cells and USP16-rescued KO cells; scale bar = 200 μm. F Bar graph showing green fluorescence (JC-1 monomers) and red fluorescence (JC-1 aggregates) intensity of MMP changes in KO cells and rescued KO cells, as shown in (E) (n = 3). G Bar graph showing ATP levels in different groups of cells. (n = 3). H Calcein/CoCl₂ staining is performed on USP16 KO cells and two rescued KO cell lines to observe the intensity of the green fluorescence of calcein, thereby determining the degree of mitochondrial MPTP opening; scale bar = 200 μm. I Bar graph quantifying green fluorescence intensities from Calcein as shown in (H) (n = 3). J. Measurement of oxygen consumption rate (OCR) in different groups of cells. (n = 3). K–N Basal OCR: Oxygen consumption required for cellular ATP production and proton leak; ATP-linked OCR: Decrease in OCR upon addition of oligomycin, indicating oxygen consumption related to ATP synthesis; Proton Leak: Residual oxygen consumption after subtracting ATP-linked OCR from OCR; Maximal OCR: Maximum oxygen consumption achieved after FCCP addition (n = 3). Statistical analyses were performed using Student’s t test. *P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant.

Identification of the substrates of USP16 in mitochondria

Since USP16 KO significantly affects mitochondrial function, we proceeded to investigate the underlying mechanism by identifying its mitochondrial targets. We reasoned that USP16 KO cells may have reduced levels of substrates that are critical to mitochondrial functions due to the ubiquitin-proteasome pathway. For this purpose, we purified mitochondria from USP16 KO HeLa cells and the control parental cells and performed ubiquitination proteomics and quantitative proteomics analyses (Fig. 3A). The ubiquitination proteomics identified 1436 ubiquitin-modified proteins within mitochondria (Supplementary Fig. S3A). Quantitative proteomics identified 3496 proteins that significantly changed abundance, with 1129 upregulated and 973 downregulated, in USP16 KO cells (Supplementary Figs. S3B and 3B). To identify proteins that may directly interact with USP16, we expressed FLAG-USP16 by transient transfection in HeLa cells, treated the cells with cross-linking agent DSP, and performed affinity-purification under denaturing conditions (Fig. 3C). The eluent was then separated via SDS-PAGE and analyzed by mass spectrometry (Fig. 3A). Total 712 proteins were identified (Fig. 3D). Venn diagram analyses of ubiquitinated proteins (Supplementary Fig. S3A), proteins downregulated after USP16 KO (Fig. 3B), potential USP16 interacting proteins (Fig. 3D) and mitochondrial protein databases (including the Uniprot mitochondria localization database) identified three mitochondria proteins: TOM40, CISY, and PGAM5, as potential USP16 targets (Fig. 3E). When analyzed by immunoblotting, we found that USP16 KO had the most dramatic effects on TOM40 protein levels, whereas its effects on CISY and PGAM5 protein levels were much less pronounced (Fig. 3F–H). We thus focused on TOM40 for further studies.

Fig. 3. Identification of USP16 substrate in mitochondria.

A Schematic representation of the experimental workflow used to identify USP16 substrates in mitochondria. B Volcano plots showing protein quantification in Control (parental HeLa cells) and USP16 KO HeLa cells. Blue dots represent downregulated proteins, red dots represent upregulated proteins, and gray dots represent proteins without significant changes. C Coomassie Brilliant Blue staining showing immunoprecipitation samples from vector and FLAG-USP16 transfected HeLa cells treated with DSP. D Venn diagram showing results from AP-MS results (C), identifying 712 potential USP16-interacting proteins. E Venn diagram comparison of results from ubiquitination proteomics, proteomics, and AP-MS, combined with mitochondrial protein databases and UniProt, to identify mitochondrial-localized proteins. Three potential USP16 targets in mitochondria were revealed: TOM40, CISY, and PGAM5. F–H Immunoblots of TOM40, CISY, and PGAM5 levels in control and USP16 KO cells. I Immunofluorescent image showing co-localization of endogenous USP16 and TOM40. Green fluorescence represents USP16, and red fluorescence represents TOM40; scale bar = 20 μm, Enlarged: scale bar = 2 μm. J TOM40 interacts with USP16 under overexpression conditions. Plasmids as indicated above were co-transfected into USP16 KO HeLa cells, followed by immunoprecipitation using FLAG antibody agarose beads. K FLAG-USP16 interacts with endogenous TOM40. FLAG-USP16 expression vector was transfected into USP16 KO HeLa cells, followed by immunoprecipitation using FLAG antibody agarose beads. L Endogenous USP16 and TOM40 are co-IPed. USP16 was immunoprecipitated with the USP16 antibody, which also co-IPed TOM40 from HeLa cells.

We confirmed that USP16 co-localized with TOM40 using immunofluorescent staining (Fig. 3I). When FLAG-USP16 and TOM40-V5 were co-expressed, Flag-USP16 co-immunoprecipitated (co-IP) with TOM40-V5 (Fig. 3J). To eliminate interference from endogenous USP16 and ensure the specificity of the FLAG-USP16 signal, we transfected Flag-USP16 into USP16-KO cells and performed co-IP experiments. Flag-USP16 was found to co-IP the endogenous TOM40 (Fig. 3K). Further experiments confirmed that endogenous USP16 co-IP with endogenous TOM40 (Fig. 3L). Conversely, endogenous TOM40 also co-IP with endogenous USP16 (Supplementary Fig. S4A). In addition, the GST pull‑down assay further verified the direct interaction between these two proteins (Supplementary Fig. S4B). Together, these data identified TOM40 as a target of USP16 in mitochondria.

USP16 regulates TOM40 levels through deubiquitination

To investigate how USP16 regulates TOM40 protein levels, we measured TOM40 RNA and protein levels in control and two USP16 KO cell lines using RT-qPCR and immunoblot assays. USP16 KO did not affect TOM40 mRNA levels, but significantly reduced its protein levels (Fig. 4A–C). Reintroducing USP16 into USP16 KO cells restored TOM40 protein levels, suggesting that USP16 stabilizes TOM40 (Fig. 4D, E). Proteasome inhibitor MG132 treatment mitigated the reduction in TOM40 levels caused by USP16 KO, suggesting that TOM40 levels were regulated through the ubiquitination-proteasome pathway (Fig. 4F, G). Thus, TOM40 was primarily stabilized through deubiquitination by USP16. To substantiate this interpretation, we measured the half-life of TOM40 in control and two USP16 KO cell lines. USP16 KO significantly reduced TOM40 stability and shortened its half-life (Fig. 4H, I). These findings demonstrate that USP16 plays an important role in maintaining TOM40 levels via deubiquitination, thereby escaping the ubiquitination-proteasome pathway.

Fig. 4. USP16 regulates TOM40 level through deubiquitination.

A RT-qPCR analyses of TOM40 mRNA levels in control and USP16 KO cells (n = 3). B Immunoblots of TOM40 protein levels in control and two USP16 KO cell lines. C Quantification of results in (B) (n = 3). D Immunoblots of TOM40 protein levels in control, two USP16 KO cell lines, and USP16 KO cells rescued with FLAG-USP16. E Quantification of results in (D) (n = 3). F, G USP16 stabilizes TOM40. Treatment with MG132 (20 μM, 2 h) prevented TOM40 protein decrease induced by USP16 KO. Representative Immunoblots were shown in panel F and quantification was shown in (G) (n = 3). H, I Cycloheximide (CHX, 10 μg/mL) chase assay showing changes in TOM40 protein levels in control and USP16 KO cells over 8 h. Immunoblots were shown in panel H and quantification was shown in (I) (n = 3). J TOM40 is ubiquitinated in vivo and its ubiquitination level is regulated by USP16. Expression plasmids as indicated were transfected into HeLa cells. Cells were treated with MG132 (20 μM) for 2 h. Lysates were immunoprecipitated with anti-V5 antibody. Input and eluents were shown. K Identification of ubiquitin linkage types on TOM40 using HeLa cells. L Identification of ubiquitination sites on TOM40 using HeLa cells. M, N TOM40 was ubiquitinated in K175/184/309. Immunoblots of lysates of HeLa cells transfected with expression plasmids indicated above (n = 3). The level of each of the mutant forms of TOM40 increased in response to exogenous USP16. However, the K175/184/309 to R did not change with USP16 expression, confirming the sites of ubiquitination. Statistical analyses were performed using Student’s t test. *P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant.

To investigate whether USP16 deubiquitinates TOM40 directly, we co-transfected HA-ubiquitin (HA-ub) and TOM40-V5 with either wild-type USP16 or a catalytically inactive mutant form, namely USP16-C205S, into HeLa cells. We detected TOM40 ubiquitination, which was significantly increased by USP16 knockdown (Fig. 4J, compare lanes 2 to 5). Importantly, TOM40 ubiquitination was completely abolished by exogenous wild-type USP16, whereas the presence of USP16-C205S did not significantly alter the ubiquitination levels of TOM40 compared to the control (Fig. 4J, lanes 2-4). To determine the ubiquitin linkage, we co-transfected expression plasmids of FLAG-USP16 and TOM40-V5, as well as WT HA-ub or ubiquitin chain-restricted mutants in which only a single lysine residue was retained (K6, K11, K27, K29, K33, K48, K63) while all other lysines were mutated to arginine. A lysine-free mutant (7KR) was included as a negative control. USP16 reduced TOM40 ubiquitination in cells expressing the K11, K27, K33, K48, or K63-only ubiquitin mutants. Notably, USP16 displays preferential activity toward K48‑linked ubiquitin chains, which show the most dramatic reduction upon USP16 expression (Fig. 4K), indicating that USP16 primarily influences K48-linked ubiquitination of TOM40.

Human TOM40 contains 14 lysine sites that might be ubiquitinated. Using the phosphosite analysis (www.phosphosite.org/homeAction.action), three major ubiquitination sites, K175, K184, and K309, were identified (Supplementary Fig. S5). Mutation on these lysines individually or simultaneously (K175R, K184R, K309R, or K175/184/309R) significantly reduced TOM40 ubiquitination, with the most substantial reduction observed when all three lysines were mutated (Fig. 4L). Consistent with the role of these three amino acids in mediating USP16‑regulated ubiquitination, mutation of TOM40 at K175/K184/K309 abolishes USP16‑mediated regulation of protein stability (Fig. 4M, N). These results confirm that USP16 is a bona fide deubiquitinating enzyme for TOM40 and that USP16 regulates TOM40 levels primarily by reducing ubiquitination of TOM40 at K175, K184, and K309.

TOM40 levels are critical to maintain mitochondrial functions

Previous studies have shown that a decrease in TOM40 leads to mitochondrial dysfunction [40–42]. To confirm these observations, we knocked down (KD) TOM40 in HeLa cells using two siRNAs. RT-qPCR and immunoblot assays confirmed that the levels of TOM40 mRNA and protein were significantly reduced (Fig. 5A–C). Transfection of an expression plasmid of TOM40-V5 restored the levels of TOM40 in the KD cells (Fig. 5D, E). Mitochondrial superoxide levels, as evaluated with the MitoSOX Red fluorescent probe, were significantly increased in TOM40 KD cells, and the increase was reversed by the exogenous TOM40 (Fig. 5F, G). Mitochondrial membrane potential, as assessed with the JC-1 probe, was altered in TOM40 KD cells and was partially restored by TOM40 expression plasmid (Fig. 5H, I). Intracellular ATP levels were significantly reduced in TOM40 KD cells and were partially rescued upon TOM40 expression plasmid transfection (Fig. 5J). MPTP opening, as assessed using Calcein fluorescence, was significantly reduced in TOM40 KD cells and was partially restored by TOM40 plasmid transfection (Fig. 5K, L). Oxygen consumption rate (OCR) measurements showed disruptions in mitochondrial respiration in TOM40 KD cells, as revealed by sequential addition of ETC inhibitors (Fig. 5M). Finally, detailed analyses of OCR parameters, including basal OCR, ATP-linked OCR, proton leak, and maximal OCR, all exhibited significant impairment in TOM40 KD cells, while a partial restoration was observed in TOM40-rescued cells (Fig. 5N–Q). These findings demonstrated that certain levels of TOM40 are required for maintaining mitochondrial function.

Fig. 5. The levels of TOM40 are important for mitochondrial function.

A RT-qPCR analyses of TOM40 mRNA levels in control and siRNA treated HeLa cells (n = 3). B, C Immunoblots of control and siRNA treated HeLa cells (n = 3). D, E Immunoblots of TOM40 protein levels in control, TOM40 KD, and TOM40 KD HeLa cells rescued with TOM40 expression plasmid (n = 3). F Measurements of mitochondrial superoxide levels using the MitoSOX Red fluorescent probe in control, TOM40 KD, and TOM40-rescued KD cell lines. scale bar = 20 μm. G Bar graphs showing of fluorescence intensities generated by MitoSOX Red in the groups shown in (F) (n = 5). H Measurements of mitochondrial membrane potential (JC-1) in control, TOM40 KD, and two rescued TOM40 KD cell lines, scale bar = 200 μm. I Bar graphs showing green fluorescence intensities generated by JC-1 monomers and red fluorescence intensities generated by JC-1 aggregates in the groups shown in (H) (n = 3). J Bar graph analyses of intracellular ATP levels in different groups of cells (n = 3). K Measurements of mitochondrial permeability transition pore (MPTP) opening in control, TOM40 KD, and two rescued TOM40 KD cell lines, scale bar = 200 μm. L Bar graphs showing green fluorescence intensities of Calcein in the groups shown in (K) (n = 3). M Measurement of oxygen consumption rate (OCR) in control, TOM40 KD, and two rescued TOM40 KD cell lines (n = 3). N–Q Basal OCR: Oxygen consumption required to meet cellular ATP demands and proton leakage; ATP-linked OCR: Reflecting oxygen consumption related to ATP synthesis; Proton Leak: Residual oxygen consumption after subtracting ATP-linked respiration from basal respiration. Maximal OCR: Maximum oxygen consumption achieved after FCCP addition (n = 3). Statistical analyses were performed using Student’s t test. *P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant.

TOM40 partially rescues the mitochondrial dysfunction caused by USP16 KO

To demonstrate that TOM40 indeed has a significant role in mediating the USP16-regulated mitochondria function, we restored the levels of TOM40 in USP16 KO cells by transfection of TOM40-expressing plasmid (Fig. 6A, B). We observed that, upon TOM 40 re-expression, all the dysfunctions detected in USP16 KO cells were largely or partially restored, including mitochondrial superoxide levels (Fig. 6C, D), mitochondrial membrane potential (Fig. 6E, F), intracellular ATP levels and MPTP opening (Fig. 6G–I), and finally the oxygen consumption rates (OCR) (Fig. 6J), and key parameters such as basal OCR, ATP-linked OCR, proton leak, and maximal OCR (Fig. 6K–N). Furthermore, overexpression of TOM40 in USP16 KO cells improved respiratory chain complex I, III and V activities (Fig. 6O–Q). Overall, these findings indicate a pivotal role of TOM40 in USP16-regulated mitochondrial functions.

Fig. 6. Role of TOM40 in mitochondrial dysfunction caused by USP16 KO.

A, B Immunoblot analyses of USP16 KO cells transfected with plasmids as indicated as the top of the panel with TOM40 and USP16 antibodies. Tubulin was used as controls (n = 3). C Images showing the measurements of mitochondrial superoxide levels using the MitoSOX Red fluorescent probe in cells as treatment above, scale bar = 20 μm. D Bar graph showing of red fluorescence intensity generated by MitoSOX Red in the groups shown in (C) (n = 5). E Measurements of mitochondrial membrane potential (JC-1) in cells as treated above, scale bar = 200 μm. F Bar graph showing green fluorescence intensity generated by JC-1 monomers and red fluorescence intensity generated by JC-1 aggregates in the groups shown in (E) (n = 3). G Measurements and bar graph of intracellular ATP levels in cells as treated above (n = 3). H Measurements of mitochondrial permeability transition pore (MPTP) opening in cells as treated above, scale bar = 200 μm. I Bar graph of green fluorescence intensity of Calcein in the groups shown in (H) (n = 3). J Measurements of oxygen consumption rate (OCR) in different groups by sequentially adding mitochondrial electron transport chain (ETC) inhibitors (Oligomycin, FCCP, Rotenone/Antimycin A). K–N Basal OCR: Oxygen consumption required to meet cellular ATP demands and proton leakage; ATP-linked OCR: Reflecting oxygen consumption related to ATP synthesis; Proton Leak: Residual oxygen consumption after subtracting ATP-linked respiration from basal respiration; Maximal OCR: Maximum oxygen consumption achieved after FCCP addition (n = 3). O–Q Overexpression of TOM40 after USP16 KO and detection of changes in mitochondrial respiratory chain complex I/III/V activity (n = 3). Statistical analyses were performed using Student’s t test. *P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant.

TOM40 partially rescues metabolism changes induced by USP16 KO

The aforementioned studies indicated that TOM40 is a target protein of USP16 and that re-expression of TOM40 in USP16 knockout cells partially rescued mitochondrial energy metabolism dysfunction caused by impaired oxidative phosphorylation, mitochondrial respiratory chain, and ATP reduction. To further examine the role of TOM40 in USP16-regulated mitochondrial functions, we used targeted metabolomics to evaluate the impact of restoring TOM40 on the energy metabolism in USP16 KO cells. CV (Coefficient of Variation) distribution analysis demonstrated stable data from all experimental groups (Supplementary Fig. S6A). PCA (Coefficient of Variation) scores and metabolic clustering analysis (Supplementary Fig. S6B, C) revealed significant differences in metabolites between sample groups (USP16 KO + Vector group and Vector group; USP16 KO + TOM40 group and USP16 KO + Vector group), with no significant difference between the Control group and the Vector group. Using Student’s t-test (P < 0.05, VIP > 1) and a > 2-fold change as criteria, we identified 71 metabolites through energy metabolism analysis. No significant changes in metabolites were observed between the Vector group and the Control group (Fig. 7A). In the USP16 KO group compared to the Vector group, 15 metabolites were downregulated, and 4 were upregulated (Fig. 7B). Overexpression of TOM40 in USP16 KO cells, compared to the USP16 KO group, resulted in 9 upregulated metabolites and 1 downregulated metabolite (Fig. 7C). Venn diagram showed that TOM40 re-expression in USP16 KO cells restored the levels of 6 downregulated metabolites (D-Glucose-6-phosphate, D-Mannose-6-phosphate, L-2-Hydroxyglutaric-acid-disodium, Guanosine-diphosphate, Argininosuccinic-acid, D-Fructose-6-phosphate) and 1 upregulated metabolite (Glycerol-3-phosphate) (Fig. 7D–F). Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis on these 7 metabolites revealed that they primarily affect energy metabolic pathways and the biosynthesis of secondary metabolites (Fig. 7G). These findings confirm that TOM40 re-expression partially rescues the metabolic disturbances caused by USP16 knockout, affirming its role as a critical downstream target of USP16 in maintaining metabolic homeostasis.

Fig. 7. Metabolite changes in USP16 KO cells and KO cells rescued by TOM40 re-expression.

A–C Volcano plots of differential metabolites in two groups of samples. A: Vector vs Control (parental HeLa cells); B: USP16 KO + Vector vs Vector; C: USP16 KO + TOM40 vs USP16 KO + Vector. Each point in the plot represents a metabolite, where green points indicate downregulated metabolites, red points indicate upregulated metabolites, and gray points indicate metabolites detected without significant changes. D, E Venn diagram showing differential metabolites identified in USP16 KO cells rescued with TOM40. F Boxplot showing differential levels of metabolites produced after reintroducing TOM40 in USP16 KO cells. G KEGG pathway analysis of metabolic changes associated with differential metabolites produced after reintroducing TOM40 in USP16 KO cells.

Discussion

In this study, we establish USP16 as a key regulator of mitochondrial function through its interaction with TOM40, a central component of the translocase of the outer mitochondrial membrane (TOM). Our findings reveal a previously unrecognized role of USP16 in maintaining mitochondrial homeostasis by stabilizing TOM40 via deubiquitination. The localization of USP16 to both the mitochondrial outer membrane and matrix, as shown by immunofluorescence and immunoelectron microscopy, underscores its direct involvement in the biological processes of the mitochondria.

USP16 knockout (KO) cells displayed significant mitochondrial dysfunction, including increased oxidative stress, reduced mitochondrial membrane potential (MMP), decreased ATP production, and impaired oxygen consumption rate (OCR). These observations indicate a critical role for USP16 in maintaining mitochondrial integrity and energy metabolism. Proteomics and interaction assays identified TOM40 as a primary substrate of USP16, with specific deubiquitination of K48-linked ubiquitin chains at lysines K175, K184, and K309 of TOM40. The stabilization of TOM40 by USP16 highlights a post-translational mechanism through which USP16 preserves mitochondrial functionality. Reintroducing TOM40 into USP16 KO cells largely or partially restored multiple mitochondrial functions, including improvements in MMP, ATP production, OCR, and oxidative stress levels. Additionally, targeted metabolomics revealed that TOM40 restoration corrected metabolic disturbances caused by USP16 deficiency. These rescue experiments underscore the importance of TOM40 as a downstream effector of USP16. These findings not only demonstrate the critical role of TOM40 in mediating the effects of USP16 but also provide insight into the metabolic consequences of USP16 dysfunction.

Our results align with previous studies highlighting the essential role of the TOM complex in protein translocation and mitochondrial homeostasis [33, 41, 43]. Using a wide range of approaches, this study has elucidated a direct regulatory relationship between USP16 and TOM40. Although the current mechanistic analyses were performed primarily in HeLa and 293T cells, the involvement of USP16 in mitochondria regulation was also supported in other physiologically relevant cell types. The observed deubiquitination of TOM40 by USP16 shows that USP16 was present and functioned in the mitochondria, consistent with our recent observation that a fraction of PRC1 components, Ring 1 and Ring 2, are located in and affect mitochondrial functions [32].

In all, our study identifies USP16 as a novel regulator of mitochondrial functions and highlights its critical role in maintaining mitochondrial integrity through the stabilization of TOM40. Several questions remain. For example, the molecular determinants governing USP16 mitochondrial targeting (e.g., cryptic targeting sequences or protein-mediated import) and the broader implications of its activity on other mitochondrial proteins warrant further investigation. Additionally, while our study highlights the importance of TOM40 in mitochondrial function, other USP16 targets such as CISY and PGAM5 may also contribute to its effects, as suggested by the proteomics data. Future studies will need to characterize these additional substrates and define their roles in mitochondrial dynamics and bioenergetics. Moreover, the in vivo significance and clinical implications of the USP16-TOM40 axis will need to be further investigated using animal models and clinical specimens.

Author contributions

HW contributed to Conceptualization, Methodology, Supervision, Writing–Review and Editing; GC contributed to Conceptualization, Methodology, and Writing–Review and Editing; ZJ conducted Investigation, Data Curation, Formal Analysis, Writing–Original Draft, Editing and was responsible for manuscript submission; ZD performed Investigation, Formal Analysis and Editing; LTC contributed to Supervision, Writing–Review and Editing; YL, KL, XJ, XZ and SZ contributed to Investigation, Data Curation, and Writing–Review and Editing.

Funding

This work was supported by the Second-level Scientific Research Startup Fund of Henan University (Grant No. S23009Y).

Data availability

The mass spectrometry raw data of the ubiquitinated proteome, proteome, and Affinity Purification coupled with Mass Spectrometry from this study have been deposited in PRIDE (PRoteomics IDEntifications Database). The shared PXD accession number for these three projects is PXD065823, and the link is: https://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD065823. The raw mass spectrometry data of Energy Metabolism from this study have been deposited in MetaboLights. The project accession number is MTBLS12680, and the access link is: https://www.ebi.ac.uk/metabolights/reviewerf552e5c0-d678-4597-ba91-50693e6b3341. The original Western blot images are provided as supplementary material (Supplementary Fig. S7). The authors declare that all other data and materials supporting the findings of this study are available in this article and its supplementary files.

Competing interests

The authors declare no competing interests.

Consent for publication

All authors agree to the publication of the article.

Supplementary information

The online version contains supplementary material available at 10.1038/s41419-026-08948-y.