henhaobio
Sign in

bioRxivCell BiologyPreprint

Local ROS and inefficient autophagosome formation limit Optineurin-mediated mitophagy in neurons

Jordan R. GreeniD, Anne E. OjoiD, Molly Katherine Gooden, Lillian D. PapelliD, Titilola D. KalejaiyeiD, Chantell S. EvansiD

PDFDOI

Abstract

Mitophagy is a quality control mechanism that maintains neuronal health by removing damaged mitochondria. In neurons, mitophagy can be driven by Optineurin (OPTN), a selective autophagy receptor recruited in response to oxidative stress. However, mechanistic studies of mitophagy have relied mainly on non-neuronal cell lines and acute damage paradigms. It remains unknown whether these findings translate to neurons under physiological stress conditions. Here, we investigated the temporal dynamics of OPTN-dependent mitophagy under mild oxidative stress using live-cell imaging in primary rat hippocampal neurons. We found that autophagosomes failed to readily engulf OPTN-positive (OPTN+) mitochondria, revealing a previously unidentified rate-limiting step in neuronal OPTN-mediated mitophagy. Additionally, OPTN+ mitochondria colocalized with acidified lysosomes over timescales significantly longer than previously reported. These OPTN+ mitochondria displayed locally elevated reactive oxygen species (ROS). Scavenging mitochondrial ROS increased lysosomal colocalization with damaged mitochondria and accelerated mitophagy progression. Together, these findings indicate that redox signaling contributes to the slow timescale of OPTN-mediated mitophagy in neurons.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (3 Oct 2026) · CC BY 4.0

In rat hippocampal neurons under mild oxidative stress, Optineurin-marked damaged mitochondria are engulfed slowly and reach lysosomes over days; scavenging mitochondrial reactive oxygen species or nutrient deprivation speeds this.

Mechanistic mitophagy work has mostly used non-neuronal lines such as HeLa treated with high-dose mitochondrial toxicants, where damaged mitochondria are engulfed within hours, and neuronal mitophagy kinetics lack consensus. This study times each step of Optineurin (OPTN)-mediated mitophagy in primary neurons under mild antioxidant-free stress and places the bottleneck at autophagosome engulfment, not the lysosomal acidification step the authors' earlier work had proposed.

Key results

  1. ShownAt 6 h, most OPTN-positive mitochondrial fragments had not yet recruited LC3B, so autophagosome initiation and engulfment, rather than lysosomal fusion, is the stalled step; bulk autophagosomes formed and acidified normally, but those at OPTN+ mitochondria acidified less.

    ~0.6 of OPTN+ events were LC3B-negative and ~0.15 fully engulfed in both control and AOF (n = 47-114 events, 3 biological replicates); in the DMSO/BafA1 experiment most events stayed LC3B-negative in both conditions and LC3B+/CatD+ events were rare (0.06); AOF did not change total LC3B puncta (n = 71 neurons, 7 replicates); the acidified mCherry+ LC3B fraction was significantly lower at OPTN+ events than elsewhere. · Autophagosomes fail to form in OPTN-mediated…; Autophagosome initiation and formation are…; Figure legends

  2. ShownDelivery of OPTN+ mitochondria to lysosomes takes days in neurons, whereas in HeLa cells most OPTN+ events colocalize with lysosomes within 2 h even under mild stress; a SNAP-tag pulse-chase separating old from new mitochondria showed lysosomal colocalization rising between 24 and 72 h after AOF.

    At 6 h the vast majority of neuronal OPTN+ events were LysoTracker-negative, versus ~0.6 LysoTracker-positive in HeLa at 0.5 or 4 µM AA/OA (n = 30-32 cells). Pulse-chase validation: 96.5% dual-labeled with co-added ligands, 87.5% first-pulse-only when sequential; LysoT+ fraction of old OPTN+ events rose from 24 to 72 h under AOF (p < 0.05; n = 24-60 events). · Autophagosome initiation and formation limit the…; “Aged” mitochondria remain in hippocampal…; Figure legends

  3. ShownOPTN+ mitochondria are depolarized and carry locally elevated mitochondrial ROS relative to the rest of the network in both control and AOF neurons, while ATP at these sites did not differ from the soma.

    Mito-Grx1-roGFP2 405/488 ratio significantly higher at OPTN+ events than in the network (n = 19-36 events, 3 biological replicates; p < 0.01 to p < 0.0001); TMRE significantly lower at events (p < 0.0001; n = 14-17 events); QUEEN-37C 400/494 ratio not different between soma and events (n = 4 events in 48 neurons). · AOF treatment induces low levels of ROS and…; Local ROS is elevated at OPTN+ mitophagy events; Figure legends

  4. SuggestedScavenging ROS for 24 h after AOF increased the fraction of OPTN+ mitochondria inside acidic lysosomes without changing how many neurons had OPTN+ events, implicating the local redox state in the delay.

    NAC (500 µM) lowered MitoSOX (p < 0.01) and raised LysoT+ OPTN+ events (p < 0.05; n = 44-55 events, 70 neurons, 4 biological replicates); EUK-134 (1 µM) lowered MitoSOX (p < 0.0001) with a similar rise in LysoT+ events (p = 0.0507; n = 47-64 events). · Scavenging mitochondrial ROS accelerates…; Mitochondrial damage and treatments; Figure legends

  5. ShownBoosting autophagy by 24 h partial nutrient deprivation (equal parts medium and Hanks' Balanced Salt Solution, HBSS) increased lysosomal delivery of OPTN+ mitochondria without altering mitophagy induction or the local ROS elevation.

    HBSS lowered pS6 and raised LC3B-II:LC3B-I; total LC3B puncta (p < 0.0001) and acidified fraction (p < 0.01) rose (n = 49-50 neurons); mitochondrial volume and OPTN+ event counts unchanged; CatD+ fraction of OPTN+ events increased significantly (n = 17-27 events, 5 biological replicates). A 4 h deprivation lowered pS6 without changing LC3B-II:LC3B-I. · Partial nutrient deprivation induces autophagy…; Figure legends

Methods, data and reproducibility

Primary rat hippocampal neurons were stressed for 6 h in antioxidant-free (AOF) medium and each mitophagy step (OPTN recruitment, LC3B engulfment, lysosomal delivery) was scored by multi-channel live confocal imaging, plus a SNAP-tag pulse-chase following old mitochondria for 24 or 72 h. Redox and ATP sensors, ROS scavengers, HBSS starvation, immunoblots and HeLa comparisons probed mechanism.

  • Primary hippocampal neurons from embryonic day 18 Sprague Dawley rats, transfected with tagged OPTN, mitochondrial, LC3B and lysosomal reporters for live imaging · rat · n = e.g. 45-52 neurons across 3 biological replicates (Fig. 2D)
  • Neuronal lysates for fluorescent immunoblotting (Parkin, OPTN, LC3B, CytC, MTMR2, MTMR5, Rubicon, pS6, S6) · rat · n = 3 biological replicates per blot
  • HeLa cells transfected with Mito-DsRed2 and Halo-OPTN, treated 2 h with 0.5 or 4 µM Antimycin A/Oligomycin A · human (HeLa cell line) · n = 30-32 cells from 3 biological replicates
Methods & tools
Leica STELLARIS 8 confocal microscope, Leica Application Suite X (Lightning deconvolution), FIJI, Imaris, Prism, SNAP-tag and HaloTag ligands (SNAP-Cell 430, JF549cp, JF646/JFX646), mCherry-EGFP-LC3B tandem reporter, Mito-EGFP-mCherry mitophagy reporter, QUEEN-37C ATP sensor, Mito-Grx1-roGFP2 redox sensor, CellROX, MitoSOX, TMRE and LysoTracker dyes, Image Studio (LI-COR) and Image Lab (Bio-Rad)
Reproducibility gaps
Data available upon request; no code or analysis scripts are provided, and OPTN+ events were identified manually in FIJI. Plasmid sources and reagent catalogue numbers are listed (QUEEN-37C is Addgene #129318).
Limitations (6)
  • Bulk immunoblotting may miss small protein changes because mitophagy accounts for a small share of turnover, so the stable protein levels under cycloheximide and the unchanged MTMR2, MTMR5 and Rubicon could hide small changes. · stated by the authors
  • OPTN knockdown did not change the general Mito-EGFP-mCherry mitophagy signal, but compensation by parallel autophagy receptors or mitophagy pathways could contribute to this result. · stated by the authors
  • The antioxidant-free paradigm cannot be applied to HeLa cells, so the neuron-versus-HeLa comparison used low-dose (0.5 µM) Antimycin A/Oligomycin A as a mild-stress stand-in; CellROX showed lower ROS than with 4 µM, but equivalence to neuronal AOF stress was not measured directly. · stated by the authors
  • How local ROS impairs autophagosome formation, and the exact timing of engulfment, remain unresolved. · stated by the authors
  • The ROS-scavenging effect is significant for NAC (p < 0.05) but not for EUK-134 (p = 0.0507), and was scored only by LysoTracker colocalization; LC3B engulfment was not imaged under scavenging, so an effect on that step is inferred. · noted in this summary
  • Event counts per condition are modest (e.g. 17-27 events in the HBSS experiments; 4 events for the ATP comparison), all from manually scored somata of cultured embryonic rat hippocampal neurons, with no in vivo or other neuron-type validation. · noted in this summary

Read this if you study neuronal mitochondrial quality control or autophagy and want step-resolved live-imaging readouts of mitophagy, including a SNAP-tag pulse-chase for tracking old mitochondria over days.

Citations
1

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.