A mitochondria-targeted peptide for optic nerve injury
In a mouse optic nerve crush model, the research peptide HDAP2 was associated with less mitochondrial loss in injured axons and higher retinal ganglion cell survival. Early preclinical neuroprotection, not a human result.
When an optic nerve is injured, the damage is not just “a cut wire.” Retinal ganglion cells, the neurons that send visual information from the eye to the brain, start a slow die-off after the initial trauma. One reason is mundane but brutal: the axons lose mitochondrial support, and the cells run out of energy to keep themselves alive.
A preclinical study in Neuroscience reports that a research peptide called HDAP2 (described by the authors as a mitochondria-targeted aromatic peptide) was associated with less mitochondrial loss inside injured axons and higher retinal ganglion cell survival in a mouse optic nerve crush model. It is the kind of result that reads like a small step, but it is aimed at a very specific bottleneck: what happens in the first days after axons are damaged.
The paper is here: MacNeil MA, et al. (2026) (DOI: https://doi.org/10.1016/j.neuroscience.2026.01.045).
The basic idea
Most neuroprotection headlines talk about “saving neurons” as if neurons fail all at once. In reality, neurons often fall apart in stages. After an axonal injury, mitochondria can disappear from the damaged segment and transport can break down. Once that happens, the cell body can survive for a while, but it is living on borrowed time.
HDAP2 is positioned as an attempt to intervene in that vulnerable window by stabilizing mitochondrial membranes and reducing the downstream cascade that follows energy failure.
What the study actually did (and what it measured)
The model here is straightforward: optic nerve crush in mice, followed by systemic treatment with HDAP2 or saline.
Instead of relying on a single readout, the authors focused on two linked outcomes:
First, they measured retinal ganglion cell survival in the retina after injury. Second, they examined the injured optic nerve itself using electron microscopy, looking at mitochondrial density and ultrastructure inside axons.
A claim that matters for translation, if it holds up, is delivery. The authors report that systemically administered HDAP2 crossed the blood-retinal barrier and localized to retinal layers that are rich in mitochondria.
What changed in the treated animals
Two results are the heart of the story.
One is downstream: the HDAP2-treated mice had higher retinal ganglion cell densities than the saline group across regions of the retina. The other is upstream: in the crushed optic nerve, the authors report substantially less mitochondrial loss in axons with HDAP2, describing mitochondrial density as closer to uninjured controls.
A nuance worth keeping: they report mitochondrial morphology looked broadly similar across groups, which they interpret as HDAP2 preventing mitochondrial loss rather than “repairing” mitochondria that were already structurally abnormal.
Why this is interesting (and why it is still early)
If you are trying to build peptide-based therapies for nervous system injury, this is a clean, concept-driven pitch: hit the mitochondria early, then see if neuron survival looks better later. The study also tries to answer a practical question that many preclinical neuroprotection projects dodge: can you get a systemically delivered molecule to the right retinal compartments in the first place?
But the limitations are just as important as the signal:
This is one acute injury model in mice, not a chronic human disease. Optic nerve crush is a useful stress test, but it is not glaucoma, and it is not the slow, multi-year neurodegeneration that most people actually experience.
We also do not yet have the human basics that decide whether a molecule becomes medicine: pharmacokinetics, tolerability, and the kind of safety profile that would be required for a therapy intended for nervous system use.
If this line of work grows up, the next confidence-building steps will look boring: replication across labs, a better map of the timing window, and evidence that benefits persist when the injury is less “clean” than a crush.