Abstract

   The exhaustion of tumor-infiltrating CD8^+ T cells poses a substantial
   challenge in cancer immunotherapy, with mitochondrial health essential
   for sustaining T cell functionality. Mitophagy, a critical process for
   mitochondrial quality control, is severely impaired in exhausted CD8^+
   T cells, yet the underlying mechanisms remain unclear. We identified
   ubiquitin-specific protease 30 (USP30), a mitochondrial deubiquitinase
   that inhibits mitophagy, as a key factor up-regulated in exhausted
   CD8^+ T cells. Notably, prolonged antigen stimulation triggers the T
   cell receptor and nuclear factor of activated T cell 1 signaling, which
   drives the transcriptional up-regulation of USP30. Excitingly, our
   interventions targeting USP30 through genetic deletion or
   pharmacological inhibition effectively restored mitophagy, improved
   mitochondrial fitness, and rejuvenated CD8^+ T cell effector functions.
   These interventions reinvigorated antitumor responses and markedly
   suppressed tumor growth. Our findings establish USP30 as a critical
   regulator of mitophagy and a promising therapeutic target for reversing
   T cell exhaustion and enhancing the efficacy of cancer immunotherapy.
     __________________________________________________________________

   USP30 inhibition restores mitophagy and reverses CD8^+ T cell
   exhaustion to boost antitumor immunity.

INTRODUCTION

   CD8^+ T cells play a critical role in immune surveillance and cytotoxic
   response against cancer cells ([50]1). However, one of the major
   challenges in cancer immunotherapy is the functional decline of
   tumor-infiltrating CD8^+ T cells ([51]2). Hallmark features of this
   exhaustion process, such as impaired self-renewal, increased expression
   of inhibitory receptors, and diminished cytotoxic function, are driven
   by the suppressive mechanisms within the tumor microenvironment (TME)
   ([52]3–[53]5). Recent studies have highlighted mitochondrial
   dysfunction as a key driver of T cell exhaustion ([54]6–[55]8).
   Exhausted T cells exhibit notable mitochondrial abnormalities,
   including increased mitochondrial fragmentation, loss of mitochondrial
   membrane potential (MMP; depolarization), and the accumulation of
   dysfunctional mitochondria ([56]6, [57]9). These alterations are
   coupled with elevated reactive oxygen species production and impaired
   oxidative phosphorylation, leading to a decline in adenosine
   5′-triphosphate production. This mitochondrial deficit limits T cell
   proliferation, impairs effector function, and exacerbates exhaustion
   during chronic antigenic stimulation ([58]6, [59]10). Despite these
   observations, the precise mechanisms underlying the loss of
   mitochondrial integrity and quality in exhausted T cells remain poorly
   understood. Both animal studies and human clinical trials have
   demonstrated that tumor-infiltrating CD8^+ T cells with healthier
   mitochondria are less exhausted and exhibit improved functionality,
   which is associated with better treatment outcomes ([60]11). Therefore,
   strategies aimed at improving mitochondrial function in CD8^+ T cells
   have emerged as promising therapeutic approaches, with recent studies
   showing encouraging results ([61]10, [62]11).

   Mitophagy, the selective removal of damaged or dysfunctional
   mitochondria through autophagy, is a critical mitochondrial quality
   control mechanism ([63]12, [64]13). Dysregulation of mitophagy has been
   increasingly linked to a broad spectrum of pathological conditions and
   aging ([65]14, [66]15). In neurodegenerative diseases such as
   Parkinson’s and Alzheimer’s, impaired mitophagy leads to the
   accumulation of defective mitochondria, promoting oxidative stress and
   neuronal degeneration ([67]14, [68]16). Similarly, in metabolic and
   cardiovascular disorders, defective mitophagy contributes to
   bioenergetic failure and increased cellular stress ([69]17). Aging is
   also associated with a progressive decline in mitophagic activity,
   which exacerbates mitochondrial dysfunction, chronic inflammation, and
   stem cell exhaustion ([70]14, [71]15, [72]18). Although mitophagy is
   recognized as a critical process in various diseases and aging, its
   specific role in T cells has only recently begun to be elucidated.
   Growing evidence indicates that mitophagy is essential for maintaining
   mitochondrial integrity, promoting efficient energy production, and
   sustaining overall metabolic balance in T cells ([73]19, [74]20). Under
   conditions of mitochondrial stress, such as nutrient deprivation in the
   TME, mitophagy plays a crucial role in preventing the accumulation of
   damaged mitochondria that could, otherwise, compromise cellular
   function ([75]21). Recent studies suggest that impaired mitophagy may
   lead to the accumulation of dysfunctional, depolarized mitochondria,
   and contribute to CD8^+ T cell exhaustion ([76]8). These defects
   exacerbate metabolic stress and further impair the functionality of
   exhausted T cells ([77]8). However, the molecular mechanisms underlying
   the dysregulation of mitophagy during T cell exhaustion are not
   well-defined.

   One of the key pathways for mitophagy regulation involves phosphatase
   and tensin homologue (PTEN)–induced putative kinase 1 (PINK1) and the
   E3 ubiquitin ligase, Parkin ([78]22, [79]23). Under mitochondrial
   stress, PINK1 accumulates on the outer mitochondrial membrane (OMM) of
   damaged mitochondria, where it recruits and activates Parkin ([80]23,
   [81]24). This process leads to the ubiquitination of mitochondrial
   proteins, which promotes mitophagy and subsequent degradation of
   dysfunctional mitochondria ([82]22, [83]25–[84]27). Notably, Parkin
   deficiency has been associated with impaired mitochondrial clearance in
   CD8^+ T cells ([85]8, [86]20, [87]28). Although the PINK1/Parkin
   pathway is well characterized, the role of mitochondrial
   deubiquitination, notably by ubiquitin-specific protease 30 (USP30), is
   less explored. USP30 counteracts Parkin-mediated mitophagy by
   deubiquitylating OMM proteins ([88]29–[89]31). Emerging evidence
   suggests that USP30-mediated suppression of mitophagy contributes to
   the accumulation of damaged mitochondria ([90]31–[91]33). Deletion or
   pharmacological inhibition of USP30 could enhance mitophagy and improve
   mitochondrial function ([92]29, [93]34), suggesting a specific strategy
   to selectively promote mitophagy. However, the extent to which USP30
   contributes to impaired mitophagy in exhausted CD8^+ T cells, as well
   as the therapeutic potential of targeting USP30, remains to be fully
   determined. Further investigation is warranted to understand USP30’s
   impact on CD8^+ T cell mitophagy and to assess its therapeutic
   potential.

   In this study, we identify USP30 as a T cell receptor (TCR)/nuclear
   factor of activated T cell 1 (NFATc1)–induced negative regulator of
   mitophagy that promotes CD8^+ T cell exhaustion. Inhibiting USP30
   restores mitophagy and mitochondrial fitness, rejuvenates effector
   function, and enhances antitumor immunity. These findings highlight
   mitophagy enhancement via USP30 inhibition as a promising
   immunotherapeutic strategy to combat T cell exhaustion in cancer.

RESULTS

Mitophagy is suppressed during T cell exhaustion

   Mitochondrial dysfunction and impaired metabolism are hallmarks of
   terminally exhausted T cells ([94]9). To investigate how changes in
   mitochondrial function and cellular metabolism contribute to the
   development of T cell exhaustion, we used a well-established in vitro
   model of CD8^+ T cell exhaustion ([95]6). Activated T cells were
   expanded in the absence (“acute”) or presence (“chronic”) of persistent
   antigenic stimulation via anti-CD3–mediated TCR activation. Cells were
   passaged every 2 days with or without persistent stimulation, and both
   acutely and chronically stimulated T cells were briefly restimulated
   for 6 hours before collection ([96]Fig. 1A). Multiple lines of evidence
   indicate that inhibitory receptors, such as programmed cell death
   protein 1 (PD-1) and Tim3, positively correlate with CD8^+ T cell
   exhaustion and are highly expressed on terminal exhausted CD8^+ T cells
   ([97]8, [98]35, [99]36). Flow cytometry analysis demonstrated that
   CD8^+ T cells in the chronic stimulation model began exhibiting
   exhaustion by day 4, as marked by increased PD-1 and Tim3 expression.
   Prolonged chronic stimulation further led to an elevated proportion of
   PD-1^+Tim3^+ terminal exhausted CD8^+ T cells (fig. S1, A and B).

Fig. 1. Decline of mitophagy in exhausted CD8^+ T cells.

   [100]Fig. 1.
   [101]Open in a new tab

   (A) Experimental schematic of chronic and acute stimulation of CD8^+ T
   cells. (B) Pathway enrichment analysis of differentially expressed
   genes showing down-regulated signatures related to mitophagy in
   exhausted CD8^+ T cells. (C) Representative flow cytometry analysis of
   mitophagy in chronically and acutely stimulated CD8^+ T cells from the
   mt-Keima reporter mouse. (D) Quantification of high mitophagy
   populations on day 6 in acutely and chronically stimulated CD8^+ T
   cells. (E) Representative images of mt-Keima fluorescence in CD8^+ T
   cells under acute and chronic TCR stimulation. Scale bars, 20 μm. (F)
   Representative flow cytometry analysis of mitophagy and quantification
   of high mitophagy population (G) in PD-1^−Tim3^−, PD-1^+Tim3^−, and
   PD-1^+Tim3^+ subsets. (H) Representative flow cytometry analysis of
   mitochondrial mass and membrane potential [(MitoTracker Green (MG) and
   tetramethylrhodamine methyl ester (TMRM)] and quantification of
   MG^highTMRM^high population (I) and MG^highTMRM^low population (J) in
   PD-1^−Tim3^−, PD-1^+Tim3^−, and PD-1^+Tim3^+ subsets. (C to I) Numbers
   in flow cytometry plots represent the percentage of the parent
   population within each gate, with gating based on CD44^+CD8^+ live
   singlets. Data represent means ± SEM, and the points correspond to the
   number of samples. ***P < 0.001; ****P < 0.0001.

   Because chronic stimulation in vitro was sufficient to activate a
   transcriptional signature associated with T cell exhaustion, we next
   examined global transcriptional changes induced by acute versus chronic
   stimulation. Consist with the signature of exhausted CD8^+ T cells,
   genes related to the PD-1/programmed death-ligand 1 (PD-L1) signaling
   axis and TCR pathways were significantly enriched in chronically
   stimulated CD8^+ T cells ([102]Fig. 1B). Notably, genes associated with
   mitophagy and lysosome were significantly reduced in CD8^+ T cells
   subjected to chronic stimulation ([103]Fig. 1B). Mitophagy is a
   critical mechanism for maintaining mitochondrial quality control
   ([104]37). To directly assess dynamic changes of mitophagy during
   chronic CD8^+ T cell stimulation, we used the pH-sensitive
   mitochondria-targeted Keima (mt-Keima) fluorescent reporter mouse. The
   mt-Keima probe enables rapid and accurate determination of whether
   mt-Keima–tagged mitochondria are at the physiological pH (pH 8.0,
   green) or within autophagosomes that have fused with lysosomes for
   mitophagic degradation (pH 4.5, red) ([105]38). This differentiation
   can be effectively detected using flow cytometry. While more than 90%
   of native CD8^+ T cells displayed high levels of mitophagy, TCR
   stimulation led to a significant reduction in mitophagy over a 48-hour
   period ([106]Fig. 1C). Additionally, the mitophagy^high population
   progressively declined with prolonged in vitro stimulation ([107]Fig.
   1C). Notably, after day 4, chronically stimulated cells exhibited much
   lower mitophagy activity, with only ~1% of cells showing mitophagy on
   day 6, compared to more than 10% in acutely stimulated cells ([108]Fig.
   1, C and D), indicating that mitophagy is inhibited in exhausted CD8^+
   T cells. We further analyzed mitophagy using laser-scanning confocal
   microscopy and confirmed that chronic TCR stimulation resulted in
   decreased overall red mt-Keima fluorescence compared to those under
   acute stimulation, consistent with suppressed mitophagy ([109]Fig. 1E).
   To validate these findings using mt-Keima–independent approaches, we
   assessed autophagic flux by analyzing lipidation of
   microtubule-associated protein 1 light chain 3 (LC3; LC3-II/LC3-I
   ratio) ([110]39) in CD8^+ T cells following acute or chronic
   stimulation. Immunoblotting revealed a reduced LC3-I to LC3-II
   conversion under chronic stimulation, along with elevated levels of the
   mitophagy adaptor p62 ([111]40), both indicative of impaired autophagic
   activity (fig. S1C). We extended these findings in tumor-bearing
   mt-Keima mice by analyzing tumor-infiltrating CD8^+ T cells across
   different stages of exhaustion, as defined by the expression dynamics
   of PD-1 and Tim3 expression ([112]41–[113]43). Flow cytometry results
   demonstrated that the nonexhausted PD-1^−Tim3^− subset exhibited the
   highest proportion of the mitophagy^high population, while the
   terminally exhausted PD-1^+Tim3^+ CD8^+ T cells showed the lowest
   proportion ([114]Fig. 1, F and G). These findings suggest that CD8^+ T
   cells experience impaired mitophagy dynamic as the exhaustion
   progresses.

   A decline in mitophagic flux has been widely proposed to contribute to
   the accumulation of damaged mitochondria in T cells ([115]43). To
   investigate whether mitochondrial dysfunction is associated with
   impaired mitophagy in exhausted CD8^+ T cells, we assessed
   mitochondrial fitness using MitoTracker Green (MG), which labels
   mitochondria mass, and tetramethylrhodamine methyl ester (TMRM), a
   cationic dye used to measure MMP. Consistent with previous findings
   ([116]8), we observed depolarized mitochondria characterized by
   increased MG staining and reduced TMRM signal, while functional
   mitochondria displayed high levels of both MG and TMRM staining. In
   line with impaired mitophagy, the proportion of cells harboring
   functional mitochondria (MG^highTMRM^high) progressively declined,
   reaching its lowest level in terminally exhausted PD-1^+Tim3^+ CD8^+ T
   cells ([117]Fig. 1, H to J). Concurrently, this terminally exhausted
   subset exhibited a significant accumulation of depolarized mitochondria
   (MG^highTMRM^low) ([118]Fig. 1, H to J). Moreover, CD8^+ T cells with
   dysfunctional mitochondria showed elevated expression of inhibitory
   markers and a higher proportion of terminally exhausted CD8^+ T cells
   (fig. S1, D to G). These findings suggest that disruption of mitophagy
   may contribute to mitochondrial dysfunction in exhausted CD8^+ T cells.

Up-regulation of USP30 contributed to mitophagy inhibition in exhausted CD8^+
T cells

   To further elucidate how mitophagy becomes impaired during T cell
   exhaustion, we conducted a transcriptome analysis focusing on genes
   that regulate mitophagy, both positively and negatively. In contrast to
   our findings, which show that exhausted T cells exhibit lower
   mitophagy, reanalysis of single-cell RNA sequencing (RNA-seq) data from
   recently published human colorectal cancer (CRC) datasets (SCP259)
   ([119]44) revealed a general up-regulation of genes that promote
   mitophagy in exhausted CD8^+ T cell subsets, compared to naïve T cells
   ([120]Fig. 2A). Notable up-regulated genes include BECN1, PINK1, and
   PARK2 and mitophagy receptors such as OPTN, CALCOCO2, and SQSTM1
   ([121]Fig. 2A). The expression levels of USP8 and USP30, negative
   regulators of mitophagy ([122]31), were increased in the progenitor,
   effector, and exhausted CD8^+ T cell subsets, compared to those in
   naïve T cells ([123]Fig. 2A). In particular, USP30 exhibited the
   highest expression levels specifically in exhausted CD8^+ T cell
   subsets ([124]Fig. 2A). This suggests that T cell exhaustion may, at
   least, in part, suppress mitophagy through the up-regulation of USP30.
   Supporting this hypothesis, we observed a progressive increase in USP30
   transcript levels over a 6-day chronic stimulation period in CD8^+ T
   cells ([125]Fig. 2B). To validate this observation in a defined
   antigen-specific model, we used CD8^+ T cells from Pmel-1 transgenic
   mice, which specifically recognize GP100 and subjected them to chronic
   stimulation with the GP100 peptide (fig. S2A). Prolonged exposure to
   GP100 led to an increased frequency of PD-1^+Tim3^+ terminally
   exhausted Pmel-1 CD8^+ T cells (fig. S2B). In line with the above
   observation, chronically stimulated Pmel-1 cells exhibited higher USP30
   expression compared to acutely stimulated counterparts (fig. S2C).
   These results indicate that USP30 up-regulation is a conserved feature
   of CD8^+ T cell exhaustion across both polyclonal and antigen-specific
   settings and may contribute functionally to the suppression of
   mitophagy in exhausted CD8^+ T cells.

Fig. 2. TCR/NFATc1-mediated up-regulation of USP30 inhibits mitophagy in
exhausted CD8^+ T cells.

   [126]Fig. 2.
   [127]Open in a new tab

   (A) Heatmap of mitophagy-related gene expression in naive (CD44^−),
   progenitor (Ly108^+), effector (CX3CR1^+), and exhausted (Tim3^+) colon
   tumor-infiltrated CD8^+ T cells. Data from recently published
   single-cell analyses of human CRC datasets are available via the Single
   Cell Portal (SCP259). (B) Quantitative polymerase chain reaction (qPCR)
   analysis of USP30 expression in CD8^+ T cells subjected to chronic TCR
   stimulation for 0, 2, 4, and 6 days. Significance was determined using
   one-way analysis of variance (ANOVA). *P < 0.05; **P < 0.01. (C) qPCR
   analysis of USP30 expression in CD8^+ T cells stimulated with anti-CD3
   antibody at doses of 0, 0.25, 0.5, and 1 μg/ml. Significance was
   determined using one-way ANOVA. (D) Transposase-accessible chromatin
   sequencing (ATAC-seq) signal tracks for chronically stimulated CD8^+ T
   cells at day 0 (top) and day 10 (bottom); chr5:114,100,000-114,220,000
   exon of USP30 locus showed in the peak. Data from the public ATAC-seq
   datasets (GSE203593). (E) The motif analysis was performed on published
   ATAC-seq data (GSE203593) from activated CD8^+ T cells. The consensus
   motif enriched within the regulatory region of USP30 loci. (F) qPCR
   analysis of USP30 expression in CD8^+ T cells after CRISPR-mediated
   NFATc1 KO. Data represent means ± SEM, and the points correspond to the
   number of samples. *P < 0.05; ***P < 0.001.

USP30 expression is transcriptionally regulated by TCR-induced activation of
NFATc1

   We next investigated the mechanism of USP30 up-regulation in CD8^+ T
   cells. Given that TCR signaling is a well-established driver of T cell
   exhaustion, we hypothesized that TCR activation might directly regulate
   USP30 expression. As such, we treated mouse CD8^+ T cells with varying
   concentrations of anti-CD3 antibody to simulate different levels of TCR
   stimulation. We demonstrated a dose-dependent increase in USP30
   expression, with higher concentrations of anti-CD3 antibody resulting
   in greater USP30 up-regulation ([128]Fig. 2C). To investigate the
   transcriptional regulation of USP30 during exhaustion, we reanalyzed
   the transposase-accessible chromatin sequencing (ATAC-seq) datasets
   from chronic TCR stimulated CD8 T cells ([129]8, [130]45). We found
   that genomic regions surrounding the USP30 locus exhibited increased
   chromatin accessibility ([131]Fig. 2D). To identify transcription
   factors (TFs) that could be involved in regulating USP30 expression, we
   performed TF binding motif enrichment analysis around this locus. This
   analysis revealed multiple TFs with enriched motifs, including NFATc1,
   a key TF downstream of TCR signaling that plays a pivotal role in T
   cell activation and function ([132]Fig. 2E). To further confirm the
   role of NFATc1, we used a CRISPR-Cas9–mediated approach to knockout
   (KO) NFATc1 in CD8^+ T cells, followed by TCR activation. Notably, the
   loss of NFATc1 led to a significant reduction in USP30 expression
   compared to control cells ([133]Fig. 2F). Therefore, USP30 expression,
   at least partially, is transcriptionally regulated by NFATc1, in
   response to TCR signaling, and plays a role in the progression of CD8^+
   T cell exhaustion.

USP30 deficiency improves effector function in exhausted CD8^+ T cells

   To investigate the functional role of USP30 in CD8^+ T cell responses,
   we used a CRISPR-Cas9–based KO strategy to delete USP30 in
   antigen-specific Pmel-1 CD8^+ T cells. Splenocytes from Pmel-1
   transgenic mice were stimulated with GP100 peptide and interleukin-2
   (IL-2) for 24 hours and then electroporated with Cas9 ribonucleoprotein
   complexes containing either a nontargeting control or a USP30-specific
   guide RNA. We confirmed efficient USP30 deletion by reverse
   transcription–quantitative polymerase chain reaction (qPCR) (fig. S3A).
   Compared to control cells, USP30-deficient CD8^+ T cells exhibited
   reduced the expression of PD-1 and Tim3 following prolonged GP100
   stimulation (fig. S3, B and C). Notably, the deletion of USP30
   significantly elevated the maximal respiration rate (fig. S3D),
   indicative of functional mitochondria. Together, these findings
   demonstrate that USP30 deletion mitigates T cell exhaustion and
   preserves mitochondrial integrity, underscoring USP30 as a promising
   therapeutic target for enhancing antitumor CD8^+ T cell responses.

   The particular role of USP30 in CD8^+ T cell–mediated antitumor
   immunity has not been described. To determine whether USP30-mediated
   suppression of mitophagy affects CD8^+ T cell function, we generated
   mice with T cell–specific deletion of USP30 by crossing USP30^flox/flox
   mice with transgenic mice expressing Cre recombinase under the control
   of the Lck (lymphocyte protein tyrosine kinase) promoter ([134]46,
   [135]47) (USP30^CKO). We confirmed the ablation of USP30 expression in
   CD8^+ T cells (fig. S4A). These USP30-deficient CD8^+ T cells exhibited
   elevated autophagic flux, indicated by an increased LC3-II/I ratio and
   reduced expression of the mitophagy receptors p62 and OPTN (fig. S4B).
   The USP30^CKO mice were phenotypically normal and displayed unaltered T
   cell profiles (fig. S4C) at homeostasis. To investigate the role of
   USP30 in tumor immunity, we inoculated MC38 murine colon adenocarcinoma
   cells into both USP30^CKO and control USP30^WT (USP30^flox/flox;
   Lck^Cre−) mice. Notably, tumor growth was significantly reduced in
   USP30^CKO mice, suggesting that USP30 deletion in CD8^+ T cells
   enhanced antitumor efficiency ([136]Fig. 3A). Analysis of
   tumor-infiltrating lymphocytes (TILs) revealed that USP30 deletion
   significantly increased the MG^+TMRM^+ CD8^+ T cell population,
   indicating improved mitochondrial fitness ([137]Fig. 3, B to D).
   Furthermore, USP30^CKO mice exhibited a notable reduction in the
   proportion of terminally exhausted PD-1^+Tim3^+ CD8^+ T cells
   ([138]Fig. 3, E and F) accompanied by a significant increase in
   granzyme B expression ([139]Fig. 3G), reflecting enhanced cytotoxic
   function of CD8^+ T cells within the TME. In contrast, the analysis of
   CD4^+ and CD8^+ T cells from the draining lymph nodes showed that USP30
   deletion had no significant impact on T cells outside the TME (fig. S4,
   D and E). Collectively, these data indicate that USP30 deletion
   improves mitochondrial fitness and restores effector function in
   exhausted CD8^+ T cells. This suggests that targeting USP30-regulated
   mitophagy could represent a promising therapeutic strategy to
   reinvigorate exhausted CD8^+ T cells, offering potential benefits in
   the treatment of cancer.

Fig. 3. USP30 deletion enhances effector function in CD8^+ T cells.

   [140]Fig. 3.
   [141]Open in a new tab

   (A) USP30^WT and USP30^CKO mice were subcutaneously inoculated with
   MC38 tumor cells. Tumor growth kinetics in USP30^WT and USP30^CKO mice
   over 16 days post–MC38 tumor cell inoculation, with tumor sizes
   measured every other day; significance was calculated by two-way ANOVA.
   ****P < 0.0001. (B) Representative flow cytometry analysis of MMP
   (TMRM) and mitochondrial mass (MG), (C) quantification of the
   MG^+TMRM^+ population, and (D) MG^+TMRM^− population in
   tumor-infiltrating CD8^+CD44^+ T cells from USP30^WT and USP30^CKO
   mice. Significance was calculated by unpaired t test. *P < 0.05. (E)
   Representative flow cytometry plots showing PD-1 and Tim3 expression in
   tumor-infiltrating CD8^+CD44^+ T cells from USP30^WT and USP30^CKO
   mice. (F) Flow cytometry analysis of the percentage of
   PD-1^+Tim3^+CD8^+ tumor-infiltrating T cells in USP30^WT and USP30^CKO
   mice; significance was calculated by unpaired t test. ***P < 0.001. (G)
   Normalized flow cytometry analysis of granzyme B expression. GMI,
   geometric mean intensity. in tumor-infiltrating CD8^+ T cells from
   USP30^WT and USP30^CKO mice; significance was calculated by unpaired t
   test. **P < 0.01. Data represent means ± SEM, and the points correspond
   to the number of samples.

Pharmacological inhibition of USP30 restores effector function in exhausted
CD8^+ T cells

   We next explored whether pharmacological inhibition of USP30 could
   promote mitophagy and improve the function of exhausted CD8^+ T cells.
   We used ST-539, a USP30-specific inhibitor known to stimulate mitophagy
   ([142]32). We tested the ability of ST-539 to reverse mitophagy
   impairment in exhausted CD8^+ T cells in vitro in response to
   polyclonal T cell activation. ST-539 treatment led to over a threefold
   increase in CD8^+ T cell mitophagy ([143]Fig. 4, A and B) and markedly
   enhanced mitochondrial efficiency (fig. S5, A to C). Furthermore,
   ST-539 significantly reduced the proportion of terminally exhausted
   PD-1^+Tim3^+ CD8^+ T cells while also boosting the expression of the
   effector molecular granzyme B, a key mediator of CD8^+ T cell
   cytotoxicity ([144]Fig. 4, C to E). Comparable improvements in
   mitochondrial fitness (fig. S6, A and B), reduced exhaustion marker
   expression (fig. S6, C to E), and enhanced effector function (fig. S6F)
   were also observed in the Pmel-1 model following ST-539 treatment.
   These findings indicate that ST-539 effectively augments mitophagy and
   limits CD8^+ T cell exhaustion.

Fig. 4. Inhibition of USP30 restores effector functions in exhausted CD8^+ T
cells.

   [145]Fig. 4.
   [146]Open in a new tab

   (A) CD8^+ T cells were isolated from the mt-Keima mice and subjected to
   chronic TCR stimulation. Two days after initial stimulation, cells were
   treated with either vehicle or the USP30 inhibitor. On day 6
   poststimulation, mitophagy levels were assessed by flow cytometry; a
   representative plot is shown. (B) Quantification of the mitophagy^high
   population in CD8^+CD44^+ T cells treated with vehicle or USP30
   inhibitor. Significance was calculated by unpaired t test.
   ****P < 0.0001. (C) CD8^+ T cells from C57BL/6 mice were subjected to
   the same chronic TCR stimulation and USP30 inhibitor treatment. The
   expression of PD-1 and Tim3 on CD8^+CD44^+ T cells was assessed by flow
   cytometry. (D) Quantification of Tim3^+PD-1^+CD8^+ T cell populations.
   Significance was determined by unpaired t test. ***P < 0.001. (E)
   Normalized flow cytometry analysis of Granzyme B expression (GMI) in
   CD8^+ T cells treated with vehicle or USP30 inhibitor. Significance was
   determined by unpaired t test. *P < 0.05. (F) C57BL/6 mice were
   subcutaneously inoculated with MC38 tumor cells. Beginning on day 6
   postinoculation, mice received either vehicle or the USP30 inhibitor
   daily for 8 days. Tumor growth was monitored over a 20-day period.
   Tumor sizes were measured every other day; significance was calculated
   by two-way ANOVA. ****P < 0.0001. (G) Representative flow cytometry
   analysis of TMRM and MG, and (H) quantification of the MG^+TMRM^+
   population in tumor-infiltrating CD8^+CD44^+ T cells from mice treated
   with vehicle or USP30 inhibitor; significance was calculated by
   unpaired t test. **P < 0.01. (I) Normalized flow cytometry analysis of
   granzyme B expression (GMI) in tumor-infiltrating CD8^+ T cells from
   mice treated with vehicle or USP30 inhibitor; significance was
   calculated by unpaired t test. **P < 0.01. Data represent means ± SEM,
   and the points correspond to the number of samples.

   To further assess the antitumor efficacy of ST-539 in vivo, MC38 tumor
   cells were subcutaneously inoculated into C57BL/6 mice, followed by
   oral administration of ST-539 every other day. The tumor growth curve
   showed a significant reduction in the ST-539–treated group compared to
   that in controls, underscoring its potent antitumor activity ([147]Fig.
   4F). Notably, ST-539 demonstrated minimal direct cytotoxicity against
   MC38 tumor cells (fig. S7A) and had mild impact on T cell expansion
   (fig. S7B). We then analyzed TILs to evaluate mitochondrial fitness.
   Consistent with observations in USP30^CKO mice, ST-539 treatment
   resulted in a twofold increase in the MG^+TMRM^+ CD8^+ T cell
   population ([148]Fig. 4, G and H), suggesting enhanced mitochondrial
   fitness. ST-539 also significantly increased granzyme B levels in CD8^+
   TILs from MC38 tumors ([149]Fig. 4I), further indicating improved
   effector functionality. Subsequent investigations into tumor-associated
   CD4^+ T cells revealed that ST-539 treatment had no significant impact
   on their mitochondrial activity (fig. S7C). Similarly, the frequencies
   of regulatory T cells, macrophages, myeloid cells, and natural killer
   (NK) cells remained unchanged (fig. S7, D to G), suggesting that the
   effects of USP30 inhibition are specific to CD8^+ T cells.
   Collectively, these findings demonstrate that pharmacological
   inhibition of USP30 with ST-539 alleviates mitochondrial dysfunction
   and enhances antitumor immunity by protecting CD8^+ T cells from
   exhaustion.

USP30 inhibitor ST-539 improves mitochondrial health and mitigates exhaustion
in chronically stimulated human CD8^+ T cells

   To explore the translational value of USP30 inhibition, we reanalyzed
   published RNA-seq datasets from CD19 chimeric antigen receptor (CAR) T
   cells and found that CAR-specific stimulation significantly
   up-regulated USP30 compared to mock stimulation ([150]Fig. 5A)
   ([151]48). Similarly, in an in vitro model of human T cell exhaustion,
   bioinformatic analysis of a published dataset ([152]49) confirmed that
   USP30 expression is significantly elevated in dysfunctional CD8^+ T
   cells compared to that in acutely stimulated counterparts ([153]Fig.
   5B) ([154]49). To assess the functional consequences of USP30
   inhibition, we used a cord blood mononuclear cell (CBMC) model of CD8^+
   T cell exhaustion ([155]Fig. 5C) ([156]49). CBMCs were stimulated under
   acute and chronic conditions to induce distinct T cell states
   ([157]Fig. 5C). Flow cytometry analysis revealed that prolonged chronic
   stimulation significantly increased the proportion of PD-1^+Lag3^+
   exhausted CD8^+ T cells, along with elevated expression levels of PD-1
   and Lag3 ([158]Fig. 5, D to F), which confirm terminally dysfunctional
   state. Notably, USP30 inhibition enhanced mitochondrial efficiency in
   chronically stimulated human CD8^+ T cells ([159]Fig. 5, G and H) and
   boosted granzyme B expression ([160]Fig. 5I), reflecting improved
   cytotoxic potential. Consistently, Lag3 protein levels were
   significantly reduced following ST-539 treatment ([161]Fig. 5J).
   Together, these findings suggest that USP30 inhibition promotes
   mitochondrial health and effector function in chronically stimulated
   human CD8^+ T cells.

Fig. 5. Inhibition of USP30 mitigates exhaustion in human CD8^+ T cells.

   [162]Fig. 5.
   [163]Open in a new tab

   (A) The bulk sequencing dataset from CD19 CAR T cells received mock or
   CAR-specific stimulation ([164]48). The gene expression level of USP30
   was compared between these two groups. (B) The transcript level of
   USP30 was evaluated in dysfunctional and acute stimulated CD8 T cells
   derived from cord blood mononuclear cell (CBMC) model of T cell
   exhaustion ([165]49). (C) In vitro model of human CD8^+ T cell
   exhaustion using CBMCs. (D to F) Representative flow cytometry plots of
   PD-1 and Lag3 expression in CD8^+ T cells under acute or chronic
   antigen stimulation. (G and H) Representative flow cytometry analysis
   and quantification of mitochondrial mass (MG) (G) and MMP (TMRM) (H) in
   CD8^+ T cells from CBMCs treated with vehicle or USP30 inhibitor.
   Significance was determined by unpaired t test. *P < 0.05. (I and J)
   Normalized flow cytometry analysis of granzyme B (I) and Lag3 (J)
   expression (GMI) in CD8^+ T cells from CBMCs treated with vehicle or
   USP30 inhibitor. Significance was determined by unpaired t test.
   **P < 0.01; ****P < 0.0001.

DISCUSSION

   One of the key features of T cell exhaustion is mitochondrial
   dysfunction, which substantially impairs their effector function,
   contributing to a reduced immune response during chronic infections and
   cancer ([166]6, [167]8). While strategies such as enhancing
   mitochondrial biogenesis have been explored, these approaches fail to
   address the accumulation of damaged mitochondria in T cells ([168]50,
   [169]51). As such, considerable progress has been made in understanding
   the role of mitophagy as a mechanism of mitochondrial quality control
   in T cells ([170]52). Central to this process are PINK1 and Parkin,
   which mediate the ubiquitination and removal of damaged mitochondria
   ([171]53, [172]54). However, the role of mitochondrial deubiquitination
   in T cell mitophagy has remained largely unexplored ([173]8, [174]20,
   [175]55). Here, we demonstrate that mitophagy declines in CD8^+ T cells
   upon TCR activation and further exacerbated as T cells progress toward
   exhaustion ([176]Fig. 1, C and D). Concurrently, USP30, a mitochondrial
   deubiquitinase, is up-regulated in exhausted CD8^+ T cells and
   contributes to mitophagy suppression ([177]Fig. 2, A and B). We
   identified NFATc1, a TCR-activated TF, as a key regulator of USP30
   expression ([178]56, [179]57), establishing a direct link between TCR
   signaling and mitochondrial quality control. These findings reveal a
   previously unrecognized mechanism by which chronic antigen stimulation
   drives mitochondrial dysfunction and T cell exhaustion through
   transcriptional up-regulation of USP30.

   Although mitophagy has been implicated in T cell homeostasis, the
   specific regulatory mechanisms in the context of exhaustion remain
   incompletely understood ([180]8). The functional significance of
   mitophagy in maintaining T cell mitochondrial homeostasis has been
   evaluated through various loss-of-function genetic approaches targeting
   various mitophagy regulators, such as Parkin and PINK1 ([181]20,
   [182]58–[183]60). While recent studies suggested that USP30 may
   influence mitochondrial translation and in vitro cytotoxic function
   ([184]61, [185]62), its role in T cell exhaustion and in vivo antitumor
   immunity had not been defined. Furthermore, the USP30 whole-body KO
   mice ([186]62), which exhibit broad systemic deletion, may display
   phenotypes distinct from those observed in our T cell–specific KO
   model. We selectively delete USP30 in T cells and investigate its T
   cell–intrinsic role in vivo. Consistent with the previous study
   ([187]62), we confirmed the ablation of USP30 specifically in T
   lymphocyte cells did not affect baseline T cell development or
   homeostasis. However, we found that USP30 deletion markedly improved
   mitochondrial fitness and partially restored effector function in
   exhausted CD8^+ T cells in a colon cancer setting. Together, these
   findings suggest that USP30 may have tissue-specific functions and that
   its role in T cells may be various depending on in vitro activation and
   in vivo exhaustion.

   Our study provides several lines of evidence indicating that genetic
   deletion or pharmacological inhibition of USP30 improves mitochondrial
   fitness, which is essential for T cell function. While we focused on
   mitophagy, USP30 may also regulate other aspects of mitochondrial
   homeostasis ([188]31, [189]61, [190]63). Given that T cell dysfunction
   is a major limitation to immune checkpoint blockade, targeting
   mitochondrial dysfunction via USP30 inhibition may offer a
   complementary approach. Previous study indicates the possibility of
   activating mitophagy by modulating the sirtuin pathway using
   nicotinamide riboside (NR) or nicotinamide mononucleotide treatment
   ([191]8). NR treatment markedly reduces the accumulation of
   dysfunctional mitochondria and improves the antitumor effector function
   ([192]8). Consequently, dietary supplementation with NR elicited
   additive antitumor responses in combination with anti–PD-1 therapy
   ([193]8). One important consideration is the potential for excessive
   mitophagy upon USP30 inhibition, which could lead to bioenergetic
   insufficiency due to mitochondrial depletion. This highlights the need
   for optimizing dose and treatment timing based on tumor context and T
   cell metabolic demand. Fine-tuning the dose and treatment schedule
   could potentially modulate the impact of USP30 inhibition, maximizing
   its therapeutic benefit while minimizing off-target effects. This
   dose-dependent effect could also vary across tumor types, as different
   cancers may present distinct mitochondrial and metabolic challenges for
   infiltrating CD8^+ T cells. Future studies will need to explore the
   long-term impact of USP30 inhibition, including any possible unintended
   consequences of overstimulating mitophagy, such as excessive removal of
   mitochondria, which could impair overall T cell bioenergetics. In
   addition, a detailed analysis of the metabolic and transcriptional
   landscapes in human CD8^+ T cells following USP30 inhibition is crucial
   to understanding its broader immunological effects and evaluating its
   safety as a therapeutic approach. These studies are essential for
   advancing USP30 inhibition as a promising strategy to reverse T cell
   exhaustion and enhance antitumor immunity in clinical applications. In
   summary, our findings establish USP30 as a key regulator of mitophagy
   and mitochondrial fitness in CD8^+ T cells. Targeting USP30 represents
   a promising, mitochondria-directed strategy to restore T cell function
   in exhausted states and may enhance the efficacy of immunotherapies in
   cancer.

METHODS

Cell culture

   The MC38 cell line was obtained from the American Type Culture
   Collection and cultured in RPMI 1640 supplemented with 10% fetal bovine
   serum (FBS). Mouse CD8^+ T cells were cultured in T cell medium
   containing RPMI 1640, 10% FBS, 2 mM l-glutamine,
   penicillin-streptomycin, nonessential amino acids, 1 mM sodium
   pyruvate, 5 mM Hepes buffer, and β-mercaptoethanol. The cultures were
   incubated in temperature-stable and partial-pressure-stable conditions
   at 37°C and 5% CO[2].

CD8^+ T cell activation in vitro

   CD8^+ T cells were isolated from mice spleen cells using the EasySep
   Mouse CD8^+ T Cell Isolation Kit (catalog no. 19753). CD8^+ T cells
   were activated in T cell medium by adding the anti-CD3 antibody (3
   μg/ml, coated on the 24-well plate) (catalog no. 100359), anti-CD28
   antibody (1 μg/ml) (catalog no. 102116), and IL-2 (100 IU) for early 2
   days. On day 2, for acute TCR stimulation, activated T cells were
   cultured in T cell medium adding IL-2 (100 IU/ml). For chronic TCR
   stimulation, activated CD8^+ T cells were cultured in T cell medium in
   plates precoated with anti-CD3 antibody (3 μg/ml, 50 μl per well,
   coated at 4°C overnight). The medium was supplemented with recombinant
   IL-2 (100 IU/ml) throughout the culture period. After 2 days, cells
   were collected, washed, and transferred to freshly coated plates, where
   the same stimulation protocol was repeated (anti-CD3 coating and IL-2
   supplementation). This process was continued for multiple stimulation
   cycles, as indicated in [194]Fig. 1A. For all in vitro experiments, the
   concentration of T cells was kept under 1 million cells/ml. For Pmel-1
   experiments, splenocytes were isolated from Pmel-1 mice and activated
   with GP100 peptide (1 μg/ml). In the acute stimulation condition, the
   peptide was present only during the initial 24 hours. For chronic
   stimulation, Pmel-1 CD8^+ T cells were continuously maintained with
   GP100 peptides (200 ng/ml) continuously. In both conditions, cells were
   cultured in the presence of IL-2 (100 U/ml), with medium refreshed
   every other day.

Mice

   All experiments involving mice were conducted in accordance with the
   guidelines of the Institutional Animal Care and Use Committee at The
   Ohio State University. Both male and female mice were used in this
   study. We have previously described the mt-Keima mouse ([195]38). The
   mt-Keima mouse line was backcrossed for over 10 generations and
   maintained on a C57BL/6J background. USP30^fl/fl mice were obtained
   from the European Mouse Mutant Archive [EM:09734, C57BL/6 N-Atm1Brd
   Usp30tm2a [the European Conditional Mouse Mutagenesis Program (EUCOMM)
   Hmgu/WtsiH] and crossed with the transgenic strain expresses Cre under
   the control of the lymphocyte protein tyrosine kinase promoter
   (LCK-cre) ([196]47). Pmel-1 mice (strain no. 005023) were obtained from
   the Jackson Laboratory.

Tumor collection and mechanical disruption

   Single-cell suspensions of T cells were prepared by mechanically
   disrupting mouse spleens with the blunt end of a syringe plunger,
   followed by filtration through 70-μm strainers (Fisherbrand). For tumor
   single-cell suspensions, intact tumors were enzymatically treated by
   injecting a solution containing collagenase I (catalog no. C0130) and
   collagenase type IV (catalog no. C5138) in buffered RPMI medium. The
   tumors were incubated at 37°C for 60 min, mechanically disrupted with
   the blunt end of a syringe plunger and filtered through 70-μm strainers
   (Fisherbrand). Lysed tumor tissues were passed through strainers to
   obtain a cell mixture. Lymphocytes were subsequently isolated from the
   cell mixture using lymphocyte separation medium (Ficoll, catalog no.
   17-5442-02).

Flow cytometry

   For cell surface marker staining, collected cells were washed with
   phosphate-buffered saline (PBS) and incubated with antibodies in
   fluorescence-activated cell sorting buffer (1% FBS, 2 mM EDTA, and 2 mM
   NaN[3]) for 30 min on ice. The antibodies used include Tim-3-APC Fire
   750 (catalog no. 119721, clone RMT3-23), PD-1–phycoerythrin–cyanine 7
   (PE–Cy7; catalog no. 135225, clone 29F.1A12), CD3–Pacific Blue (catalog
   no. 100214, clone 17A2), CD4-BUV737 (catalog no. 367-0041-82, clone
   GK1.5), CD44-BV785 (catalog no. 103049, clone IM7), CD8-BV510 (catalog
   no. 100752, clone 53-6.7), CD45-BV605 (catalog no. 103149, clone
   30-F11), CD45–fluorescein isothiocyanate (FITC; catalog no. 103108,
   clone 30-F11), CD11b-BV510 (catalog no. 50402941, clone M1/70),
   F4/80-BV785 (catalog no. 123141, clone BM8), CD206-BV650 (catalog no.
   141723, clone C068C2), MHCII-BUV395 (catalog no. 569244, clone
   M5/114.15.2), CD11c-AF700 (catalog no. 117326, clone N418), CD301-PE
   (catalog no. 146803, clone URA-1), CD80-BV605 (catalog no. 104729,
   clone 16-10A1), CD3–Pacific Blue (catalog no. 100214, clone 17A2),
   Ly6c-APC-fire810 (catalog no. 128056, clone HK1.4), Ly6G-AF647 (catalog
   no. 135225, clone 1A8), and NK1.1-BV570 (catalog no. 7453, clone
   PK136). For intracellular marker staining, after surface marker
   staining, cells were fixed using a fixation buffer and permeabilized
   overnight with True-Nuclear Fix. On the second day, permeabilized cells
   were incubated with antibodies in Perm buffer (True-Nuclear, catalog
   no. 421403) in the dark at room temperature. The antibodies used
   include granzyme B–FITC (catalog no. 515403, clone GB11), FOXP3–Alexa
   Fluor 532 (catalog no. 126407, clone MF-14), T cell factor 1
   (TCF1)–Alexa Fluor 647 (catalog no. 6709, clone C63D9), the thymocyte
   selection-associated high mobility group box (TOX)–PE (catalog no.
   12-6502-82, clone TXRX10), and Arg^1-PE-Cy7 (catalog no. 105036, clone
   A1exF5). After staining, cells were analyzed using the Cytek Aurora
   flow cytometer. For mitochondrial staining, cells were washed with PBS
   and incubated with TMRM (100 nM; catalog no. T668) and MG (10 nM;
   catalog no. M7514) in PBS for 30 min in a 37°C incubator. Keima Green
   was excited with a wavelength of 488 nm and emitted at 615 nm, while
   Keima Red was excited with a wavelength of 561 nm and emitted at 615
   nm. For live/dead staining, after mitochondrial staining and before
   cell surface marker staining, cells were incubated with LIVE/DEAD
   Fixable Blue (catalog no. L23105) in PBS on ice in the dark for 15 min.

Human CBMCs

   Human CBMCs were seeded into 96-well plates at 200,000 cells per well
   in RPMI medium supplemented with 10% FBS, 2 mM l-glutamine,
   penicillin-streptomycin, nonessential amino acids, 1 mM sodium
   pyruvate, 5 mM Hepes buffer, and β-mercaptoethanol. Cells were cultured
   in medium containing recombinant human IL-2 (100 IU/ml; PeproTech) and
   recombinant human IL-15 (20 ng/ml; PeproTech) and stimulated with
   antihuman CD3 (5 μg/ml, clone OKT3) and antihuman CD28 (1 μg/ml)
   antibodies. For the acute stimulation condition, anti-CD3/CD28
   stimulation was applied only at the initiation of culture. For the
   chronic stimulation model, CBMCs were restimulated every 48 hours with
   anti-CD3/CD28 and fresh medium supplemented with IL-2 and IL-15
   ([197]49). Cells were maintained in culture for a total of 8 days. For
   USP30 inhibitor treatment, ST-539 (10 μM) was added on day 4
   poststimulation and maintained throughout the remaining culture period.
   The cytotoxic potential was evaluated by intracellular granzyme B
   staining, and exhaustion markers were measured by surface staining for
   PD-1, Tim-3, and Lag3. The antibodies used include CD8-BV421(catalog
   no. 50165670, clone RPA-T8), PD-1–PE–Cy7 (catalog no. 25-9969-42, clone
   MIH4), TIM-3-BUV805 (catalog no. 569859, clone F38-2E2), LAG3-APC/Fire
   750 (catalog no. 369214, clone 7H2C65), and granzyme B–AF700 (catalog
   no. 369214, clone GB11).

CRISPR-Cas9–mediated gene deletion of murine CD8^+ T cells

   The CRISPR-Cas9–mediated KO procedure was adapted from a previously
   published method ([198]64). Briefly, high-ranked guide sequences with
   the highest on-target and off-target scores were selected by CHOP-CHOP.
   Single guide RNA (sgRNA), Cas9 Nuclease V3, and Alt-R Cas9
   Electroporation Enhancer were purchased from Integrated DNA
   Technologies (IDT) Inc. IDT sgRNA was resuspended in nuclease-free 1×
   Tris-EDTA (TE) buffer at 100 pmol/μl. Cas9/ribonucleoprotein (RNP)
   complex was assembled by combining Alt-R S.p. Cas9 Nuclease V3 with an
   NFATc1-targeting sgRNA (sequence: AAGCTGGTCATTATCACCGC) or
   USP30-targeting sgRNA (sequence: GGCGTCCAAGACCTCGTCCTCGG,
   TCAAACAAGTGGGTGACTCGAGG). Isolated CD8^+ T cells were stimulated in
   24-well plates precoated with anti-CD3 (3 μg/ml; 17A2) and anti-CD28 (1
   μg/ml; 37.51) for 1 day. Two million activated CD8^+ T cells were mixed
   with the Cas9/RNP complex in the P4 nucleofection buffer, along with
   Alt-R Cas9 Electroporation Enhancer. Electroporation was performed
   using the CM137 program on the Lonza 4D Nucleofector.

Tumor xenograft and therapies

   MC38 cells were cultured in DMEM with 10% FBS supplemented and 1%
   penicillin-streptomycin. Before injection, MC38 cells were diluted in
   PBS for 1 × 10^7/ml. Tumor cell solution (100 μl) was injected
   subcutaneously into the mouse’s flank using insulin syringe. ST51000539
   (ST-539) was purchased from TimTec. For in vitro experiment, ST-539 in
   dimethyl sulfoxide was added to the medium for a working concentration
   of 10 μg/ml. For in vivo experiment, ST-539 was homogenized in 20%
   sulfobutylether-β-cyclodextrin (Captisol) water for a final
   concentration of 5 mg/ml. Drug mixture (100 μl) was injected into the
   stomach of mice using oral gavage needle every other day.

Confocal microscopy

   Fluorescent samples were analyzed using a Zeiss LSM 780 confocal
   microscope (Carl Zeiss MicroImaging). Fluorescence of mt-Keima was
   captured in two channels with sequential excitations at 458 nm (Keima
   Green) and 561 nm (Keima Red), and emissions were collected within the
   range of 570 to 695 nm. Laser power was carefully optimized to ensure
   clear signal detection and maintained consistently across all
   experimental conditions.

Seahorse assay

   The mitochondrial function of CD8^+ T cells was assessed using the
   Seahorse XF Cell Mito Stress Test Kit (Agilent) on an XF96
   extracellular flux analyzer. Approximately 5 × 10^5 cells were plated
   per well in an XF96 cell culture microplate. During the assay,
   mitochondrial stressors were sequentially injected at defined final
   concentrations, and the oxygen consumption rate was measured to
   evaluate key parameters of mitochondrial respiration.

Western blotting

   Samples were lysed in radioimmunoprecipitation assay buffer [50 mM
   tris-HCl, at pH 8.0, 150 mM NaCl, 1% (v/v) Nonidet P-40, 0.5% sodium
   deoxycholate, 0.1% SDS and protease inhibitor cocktail (Roche) on ice].
   Primary antibodies were used at the following concentrations:
   p62/SQSTM1 (Abnova, H00008878-M01; 1:1000); OPTN (Proteintech,
   10837-1-AP; 1:1000); LC3 (Cell Signaling Technology, 2775; 1:1000);
   glyceraldehyde-3-phosphate dehydrogenase (Cell Signaling Technology,
   51332; 1:1000); and ACTIN (Cell Signaling Technology, 3700S; 1:1000).
   The membranes were incubated with anti-rabbit (LI-COR, 926-32211;
   1:15,000) or anti-mouse (LI-COR, 926-68072; 1:15,000) immunoglobulin G
   secondary antibodies for 1 hour at room temperature. Images were
   captured using the Odyssey system (LI-COR).

RNA extraction and cDNA synthesis

   Total RNA was extracted from cells or tissues using TRIzol reagent
   (Direct-zol RNA MicroPrep, catalog no. R2062) according to the
   manufacturer’s instructions. The RNA concentration and purity were
   assessed using a NanoDrop spectrophotometer by measuring the absorbance
   at 260 nm and 280 nm. RNA with an A260/A280 ratio of ~2.0 was used for
   further experiments. Total RNA (1 μg) was treated with
   deoxyribonuclease I (Thermo Fisher Scientific) to remove genomic DNA
   contamination. First-strand cDNA was synthesized using a high-capacity
   cDNA reverse transcription kit (Applied Biosystems) with random
   hexamers as primers. The reaction was performed in a thermocycler at
   the following settings: 25°C for 2 min, 55°C for 10 min, and 95°C for
   1 min. The synthesized cDNA was diluted 1:10 with nuclease-free water
   for qPCR.

qPCR setup

   qPCR reactions mix was performed in a total volume of 20 μl, consisting
   of 10 μl of SYBR Green Master Mix (Applied Biosystems or equivalent), 1
   μl of forward primer (10 μM),1 μl of reverse primer (10 μM), 2 μl of
   cDNA template (diluted), and 6 μl of nuclease-free water. Forward
   primer of USP30: 5′-CTGTACAAAGATTGAAGCGAG-3′. Reverse primer of USP30:
   5′-ACTGTTTAACAAAGGTGCTC′. qPCR was performed on a StepOnePlus Real-Time
   PCR System.

Transcriptomics analysis

   CD8^+ T cells were isolated and cultured under acute or chronic
   TCR-mediated activation conditions, as described in [199]Fig. 1A.
   Following stimulation, a total of 3 × 10^6 cultured CD8^+ T cells were
   flash-frozen in an ethanol/dry ice slush and shipped to the core
   facilities at Novogen, where library construction and sequencing were
   performed following standard protocols. The raw sequencing reads were
   trimmed using Trim Galore v0.6.10 and aligned to the reference genome
   (mm10) with HISAT2 v2.2.1 in combination with Samtools v1.2.
   Differentially expressed gene analysis was performed using DESeq2
   v3.19. Gene set enrichment analysis (GSEA) was conducted with GSEA
   desktop software v4.3.3 in classic preranked mode. Data visualization
   included volcano plots generated with VolcaNoseR and heatmaps created
   using Morpheus (Broad Institute) and pheatmap v1.0.12.

Reanalysis of USP30 expression in public datasets

   To assess USP30 expression in CD8^+ T cells under different conditions,
   we reanalyzed publicly available transcriptomic datasets ([200]48,
   [201]49). Normalized read counts for USP30 were extracted from the
   processed data and compared across experimental groups using a
   one-tailed Student’s t test to evaluate directional changes in
   expression. Statistical significance was defined as P < 0.05.

Statistical analysis

   Statistical analyses were performed in GraphPad Prism using unpaired or
   paired Student’s t tests for comparing two groups and one-way analysis
   of variance (ANOVA) with Tukey’s post hoc test for multiple group
   comparisons. The tumor growth curve was analyzed using two-way ANOVA
   with Tukey’s post hoc test. Specific statistical tests used for
   individual experiments are detailed in the figure legends.

Acknowledgments