Energy sharing in the DRG: satellite glia deliver mitochondria to sensory neurons to protect against diabetic neuropathy
Aims: To test whether satellite glial cells (SGCs) in dorsal root ganglia (DRG) actively supply mitochondria to sensory neurons, and whether failure of this neuro–glial “mitochondrial support” contributes to diabetic peripheral neuropathy (DPN) and chemotherapy-induced neuropathy.
Methods: The authors combined SGC–neuron co-cultures with live imaging and pharmacological perturbations of tunnelling nanotubes (TNTs) and uptake pathways and used ultrastructural validation (SEM/TEM) in mouse and human DRG. They interrogated a candidate TNT regulator (myosin-10, MYO10) using transcriptomics (single-nucleus RNA-seq) and in situ hybridization and tested in vivo relevance across neuropathy models (paclitaxel CIPN, db/db type 2 diabetic neuropathy, and STZ type 1 diabetes). Therapeutic proof-of-concept used intra-DRG delivery of (i) healthy SGCs and (ii) purified SGC-derived mitochondria, including cross-species human-to-mouse transfer.
Results: SGCs transferred mitochondria to DRG neurons in an activity-dependent manner via TNT-like structures and blocking transfer in naïve mice was sufficient to trigger nerve degeneration and pain behaviours. MYO10 emerged as a key factor for this transfer. Interestingly, human DRG from people with diabetes showed reduced MYO10 expression and impaired SGC-to-neuron mitochondrial transfer. In vivo, adoptive transfer of healthy human SGCs into db/db mouse DRG reduced mechanical hypersensitivity, but the benefit was lost when MYO10 was knocked down in donor SGCs. Direct injection of healthy SGC-derived mitochondria also reduced neuropathic pain (with longer benefit from mitochondria from non-diabetic vs diabetic donors) and increased intraepidermal nerve fibre (IENF) density in db/db mice.
Conclusions: SGC-to-neuron mitochondrial transfer is a previously underappreciated homeostatic mechanism in the DRG that becomes compromised in diabetes, linking glial metabolic support to small-fibre degeneration and neuropathic pain.
Comments: This paper is conceptually important because it reframes DPN as, in part, a failure of glial bioenergetic caregiving rather than a purely neuron-intrinsic mitochondrial problem. The human DRG data of reduced MYO10 and reduced transfer in diabetes provides a much-needed bridge from mechanism to disease relevance. The therapeutic experiments are provocative: both SGC transfer and mitochondrial delivery improved pain readouts and, notably, IENF density, an endpoint closely aligned with small-fibre pathology in DPN. Key caveats remain. The interventional route (intra-DRG injection) is invasive, and the analgesic window appears short (hours to days), raising translational questions about durability, dosing, scalability, and safety (including immunogenicity for allogeneic cells/organelles). Mechanistically, TNTs are not the only route implicated (endocytosis and connexin-dependent communication also contribute), so “boosting MYO10” may be necessary but not sufficient. For DPN, the most actionable near-term opportunity would be to define why diabetic SGCs downshift MYO10/transfer (hyperglycaemia, lipotoxicity, inflammation, altered neuronal activity?), and whether less invasive approaches can restore this pathway (e.g., enhancing SGC mitochondrial health, TNT competency, or neuron-side mitochondrial uptake). If replicated across cohorts and neuropathy subtypes, neuro-glial mitochondrial transfer could become both a biomarker axis in human DRG and an unconventional therapeutic entry point for neuropathic pain.
Ali Jaafar
Reference. Xu J, Li Y, Novak C, Lee M, Yan Z, Bang S, McGinnis A, Chandra S, Zhang V, He W, Lechler T, Rodriguez Salazar MP, Eroglu C, Becker ML, Velmeshev D, Cheney RE, Ji RR. Mitochondrial transfer from glia to neurons protects against peripheral neuropathy. Nature. 2026 Jan 7. doi: 10.1038/s41586-025-09896-x. Epub ahead of print. PMID: 41501451.
