PD-1 Marks Senescent T Cells That Drive Rheumatoid Arthritis via Mitochondrial Breakdown
A newly identified PD-1/DRP1/mitophagy axis in CD4+ T cells fuels the inflammation and joint destruction characteristic of rheumatoid arthritis.
Summary
Researchers at Dalian Medical University discovered that CD4+PD-1+ T cells accumulate abnormally in rheumatoid arthritis patients and behave as pathogenic senescent cells. These cells show reduced expression of DRP1, a protein critical for mitochondrial fission and cleanup. Without adequate DRP1, defective mitochondria pile up, generating excess reactive oxygen species that trigger a senescence-associated secretory phenotype — a cocktail of pro-inflammatory cytokines. PD-1 signaling itself drives the DRP1 reduction by suppressing HIF-1α. In mouse arthritis models, adoptive transfer of these cells worsened disease, while restoring mitochondrial function with MitoQ or a DRP1 inhibitor reduced inflammation. The findings reveal a new therapeutic target pathway in rheumatoid arthritis.
Detailed Summary
Rheumatoid arthritis (RA) is an autoimmune disease driven in part by dysregulated CD4+ T cells that acquire premature senescence features. While senescent T cells are known to promote inflammation through their senescence-associated secretory phenotype (SASP), the precise subset responsible and the underlying mechanism have remained unclear. This study from Dalian Medical University identifies CD4+PD-1+ T cells as the critical pathogenic senescent subset and maps out a mechanistic axis — PD-1 → DRP1 suppression → impaired mitophagy → mitochondrial ROS accumulation → SASP — that drives RA progression.
The researchers enrolled RA patients meeting ACR 1987 criteria and age- and sex-matched healthy controls. Flow cytometry confirmed a significant expansion of CD4+PD-1+ T cells in RA peripheral blood. These cells displayed classical senescence hallmarks compared with CD4+PD-1− T cells from the same patients: elevated SA-β-galactosidase activity, reduced Ki67 proliferation marker, increased expression of p16, p21, and p53, loss of CD28 co-stimulatory molecule, and upregulation of SASP cytokines including IL-1β, IL-6, IL-8, TNF-α, and MMP-3. The PD-1+ subset also showed impaired proliferative capacity by CFSE dilution assay and elevated cytotoxic markers (perforin, granzyme B), painting a picture of cells that have stopped dividing but remain metabolically destructive.
Mitochondrial phenotyping revealed that RA CD4+PD-1+ T cells harbored dysfunctional mitochondria: reduced mitochondrial membrane potential by JC-10 assay, altered morphology (rounder, less elongated aspect ratio by Mitotracker Green/ImageJ), and substantially elevated mitochondrial ROS (MtROS) measured by MitoSOX Red flow cytometry. Critically, these cells showed significantly lower DRP1 protein and mRNA expression and impaired mitophagy flux, quantified using the mt-mKeima adenoviral reporter system — a ratiometric pH-sensitive probe that distinguishes mitochondria in acidic lysosomes (active mitophagy, 550 nm excitation) from healthy ones (440 nm excitation). PINK1 and Parkin, the canonical mitophagy mediators, were also reduced.
To establish causality, the team performed siRNA knockdown of DRP1 and Parkin in Jurkat T cells, which recapitulated the MtROS accumulation and SASP induction seen in primary RA cells. Conversely, DRP1 overexpression (pcDNA3.1-DRP1) or treatment with the mitochondria-targeted antioxidant MitoQ (200 nM) significantly reduced MtROS and SASP markers. The DRP1 inhibitor mdivi-1 (1 μM), paradoxically, also reduced SASP in some contexts, suggesting complex dynamics. PD-1 ligation experiments using plate-bound PD-L1 or PD-L2 demonstrated that PD-1 signaling directly suppressed DRP1 transcription via inhibition of HIF-1α, a known transcriptional activator of DRP1.
In collagen-induced arthritis (CIA) mouse models, adoptive transfer of 1×10⁶ CD4+PD-1+ T cells from CIA spleens significantly accelerated disease progression compared with CD4+PD-1− T cell transfer, as measured by paw thickness, clinical arthritis scores, and histological synovitis/cartilage damage (H&E and safranin O staining). Pre-treatment of CD4+PD-1+ T cells with MitoQ before transfer attenuated the disease-accelerating effect, validating MtROS as a functional driver. Coculture experiments showed that CD4+PD-1+ T cells drove greater plasmablast differentiation from B cells, induced higher IL-1β and IL-6 expression in RA fibroblast-like synoviocytes, and caused greater chondrocyte apoptosis than CD4+PD-1− T cells — collectively explaining their tissue-destructive capacity. These findings establish a therapeutically actionable PD-1–DRP1–mitophagy–SASP axis in RA.
Key Findings
- CD4+PD-1+ T cells were significantly expanded in RA patients vs. healthy controls, displaying elevated SA-β-galactosidase activity, increased p16/p21/p53 expression, and loss of CD28
- RA CD4+PD-1+ T cells showed markedly elevated mitochondrial ROS (MitoSOX Red signal) and reduced mitochondrial membrane potential (JC-10 assay) compared with autologous CD4+PD-1− T cells
- Mitophagy flux was significantly impaired in CD4+PD-1+ T cells vs. CD4+PD-1− T cells by mt-mKeima reporter assay, with reduced PINK1 and Parkin protein levels
- DRP1 mRNA and protein expression were significantly lower in RA CD4+PD-1+ T cells; siRNA-mediated DRP1 knockdown in Jurkat cells reproduced MtROS accumulation and SASP induction
- Adoptive transfer of 1×10⁶ CIA-derived CD4+PD-1+ T cells significantly worsened arthritis clinical scores and paw thickness vs. CD4+PD-1− T cell transfer in CIA mice
- MitoQ pre-treatment (200 nM) of CD4+PD-1+ T cells before adoptive transfer substantially attenuated disease acceleration, confirming MtROS as a functional mediator
- PD-1 ligation with plate-bound PD-L1/PD-L2 transcriptionally suppressed DRP1 expression via HIF-1α inhibition, establishing the upstream signaling mechanism
Methodology
This was a translational mechanistic study combining human samples (RA patients meeting ACR 1987 criteria vs. age- and sex-matched healthy controls, with ethics approval No. 2023-253) and a CIA mouse model (male DBA/1J mice, 6–8 weeks). CD4+PD-1+ and CD4+PD-1− T cells were isolated by immunomagnetic bead separation from PBMCs. Key readouts included flow cytometry for senescence markers, MitoSOX Red for MtROS, JC-10 for mitochondrial membrane potential, mt-mKeima adenoviral reporter for mitophagy flux, Western blot for protein quantification, and qPCR for gene expression; statistical methods and exact n values per patient cohort are detailed in supplemental tables.
Study Limitations
The human cohort size is not fully detailed in the available text, limiting statistical power assessment; the study relies on peripheral blood samples and may not fully reflect the joint microenvironment where pathology occurs. The CIA mouse model, while standard, does not perfectly replicate human RA immunopathology, and causality in humans remains correlative. The authors do not report funding conflicts explicitly in the provided text, though institutional affiliations are academic.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe:
