Cancer ResearchResearch PaperOpen Access

Mitochondrial Protein SFXN1 Drives Bladder Cancer Spread by Blocking Cellular Cleanup

SFXN1 suppresses PINK1-dependent mitophagy in bladder cancer, triggering toxic ROS buildup and EMT-driven metastasis.

Tuesday, July 7, 2026 1 view
Published in Oncogene
A fluorescence microscopy image of bladder cancer cells with red-stained mitochondria and green autophagy markers visible as bright puncta against a dark background, on a lab microscope stage

Summary

Researchers discovered that SFXN1, a mitochondrial protein, fuels bladder cancer metastasis not through its known role in metabolism but by blocking mitophagy — the cell's process for clearing damaged mitochondria. When SFXN1 is overexpressed, it interacts with two mitochondrial proteins (PARL and MPP-β) to accelerate degradation of PINK1, a key mitophagy initiator. This blocks mitochondrial cleanup, causes toxic reactive oxygen species (ROS) to accumulate, and activates TGF-β signaling that drives cancer cells to invade and spread. Knocking down SFXN1 in bladder cancer cells and mouse models significantly suppressed tumor growth and lymph node metastasis, identifying SFXN1 as a promising new therapeutic target.

Detailed Summary

Bladder cancer (BLCA) is the tenth most common cancer globally, responsible for roughly 549,000 new cases and 200,000 deaths per year. While most cases are non-muscle invasive and manageable, the 25–30% that progress to muscle-invasive disease carry a poor prognosis, and metastatic spread remains the primary driver of mortality. This study from Nanjing Drum Tower Hospital set out to characterize the oncogenic role of SFXN1 (sideroflexin 1), a mitochondrial inner membrane protein previously known mainly as a serine transporter involved in one-carbon metabolism.

Using immunohistochemistry on 155 paraffin-embedded BLCA tumor tissues (including 21 matched normal tissue pairs and 13 fresh tissue pairs), the researchers found that SFXN1 was significantly overexpressed in tumor versus normal bladder tissue. UALCAN database analysis of TCGA data confirmed this upregulation, and high SFXN1 expression was positively correlated with advanced tumor stage and poor overall survival. These findings established a clinical rationale for deeper mechanistic investigation.

Functional experiments in T24 and J82 bladder cancer cell lines showed that SFXN1 knockdown (via siRNA) markedly reduced cell migration and invasion in transwell and wound-healing assays, and suppressed proliferation in MTT assays. Conversely, SFXN1 overexpression enhanced these metastatic behaviors. Critically, serine deprivation and formate rescue experiments demonstrated that SFXN1's pro-metastatic effects were independent of its classical serine transport function, pointing to a previously unknown mechanism.

Mechanistic investigation revealed that SFXN1 acts as a bridging factor between two inner mitochondrial membrane (IMM) proteases — PARL (presenilin-associated rhomboid-like protein) and MPP-β (mitochondrial processing peptidase-β) — to accelerate degradation of PINK1, the master initiator of mitophagy. Co-immunoprecipitation confirmed direct interactions among SFXN1, PARL, and MPP-β. In cells with high SFXN1, PINK1 protein levels were reduced, mitophagy markers (LC3B–mitochondria co-localization, mitochondria–lysosome co-localization) were suppressed, and transmission electron microscopy revealed accumulation of damaged mitochondria. Conversely, SFXN1 knockdown restored mitophagy flux. Treatment with CCCP (a mitophagy inducer) reversed the pro-metastatic effects of SFXN1 overexpression, while Mdivi-1 (a mitophagy inhibitor) negated the anti-metastatic benefits of SFXN1 knockdown.

With mitophagy blocked, dysfunctional mitochondria accumulated and mitochondrial ROS (mtROS) levels rose substantially, as measured by MitoSOX flow cytometry. Elevated mtROS activated TGF-β signaling, which in turn drove epithelial-to-mesenchymal transition (EMT), increasing expression of mesenchymal markers (vimentin, N-cadherin) and decreasing epithelial markers (E-cadherin). Scavenging mtROS with mitoTEMPO abrogated EMT activation and metastasis in SFXN1-overexpressing cells. In vivo, shRNA-mediated SFXN1 knockdown in T24 cells significantly reduced subcutaneous tumor growth and lymph node metastasis in nude mouse models over 40 days, validated by bioluminescence imaging and IHC. Together, these findings define a SFXN1 → PARL/MPP-β → PINK1 degradation → mitophagy arrest → mtROS accumulation → TGF-β/EMT → metastasis axis, identifying SFXN1 as a novel and druggable target in bladder cancer.

Key Findings

  • SFXN1 was significantly overexpressed in 155 BLCA tumor tissues vs. matched normal bladder tissue, with high expression correlating with advanced tumor stage and poor overall survival (TCGA/UALCAN analysis)
  • SFXN1 knockdown substantially reduced T24 and J82 cell migration and invasion in transwell assays; SFXN1 overexpression enhanced these behaviors, effects confirmed independent of serine transport via serine-deprivation and formate rescue experiments
  • SFXN1 interacted directly with PARL and MPP-β on the inner mitochondrial membrane (confirmed by Co-IP), accelerating PINK1 protein degradation and suppressing PINK1-dependent mitophagy flux
  • SFXN1 knockdown restored LC3B–mitochondria and mitochondria–lysosome co-localization (immunofluorescence), while SFXN1 overexpression reduced these mitophagy markers and caused accumulation of damaged mitochondria on TEM
  • Mitophagy arrest from high SFXN1 caused measurable mtROS accumulation (MitoSOX flow cytometry); scavenging mtROS with mitoTEMPO (25 μM) abrogated TGF-β/EMT activation and rescued the metastatic phenotype
  • CCCP-induced mitophagy reversed pro-metastatic effects of SFXN1 overexpression; Mdivi-1 (25 μM) negated anti-metastatic benefits of SFXN1 knockdown, confirming mitophagy as the key mediator
  • In vivo, shSFXN1 T24 cells showed significantly reduced subcutaneous tumor weight and lymph node metastasis volume vs. shNC controls in nude mouse models (n=8 subcutaneous, n=10 lymph node; 40-day experiment)

Methodology

The study combined clinical tissue analysis (155 paraffin-embedded BLCA samples, 21 matched normal pairs, 13 fresh tissue pairs from Nanjing Drum Tower Hospital) with in vitro experiments in T24 and J82 bladder cancer cell lines using siRNA knockdown and plasmid overexpression, and in vivo nude mouse subcutaneous and popliteal lymph node metastasis models (n=8 and n=10 per group). Mechanistic assays included Co-IP, immunofluorescence co-localization, TEM, flow cytometry (MitoSOX, JC-1, MitoTracker), RNA-seq, and GSEA of TCGA BLCA cohort data. Statistical analyses used Student's t-test with significance threshold P<0.05, and all in vitro experiments were performed in at least triplicate.

Study Limitations

The study is predominantly preclinical, relying on two bladder cancer cell lines and nude mouse models, which may not fully recapitulate human tumor heterogeneity or immune microenvironment interactions. While 155 clinical tissue samples support the prognostic association, survival analyses were based on a single-center cohort and TCGA database mining without prospective validation. The authors do not discuss potential off-target effects of SFXN1 manipulation on other mitochondrial functions, nor do they address potential conflicts of interest explicitly in the text.

Enjoyed this summary?

Get the latest longevity research delivered to your inbox every week.

Enter your email to subscribe: