The Annexin A1 mimetic peptide Ac 2-26 preserved mitochondrial function and neuronal survival in human retinal organoids and organoid-derived retinal ganglion cells exposed to high glucose and inflammatory stress. The protection was associated with FPR2-dependent suppression of the p38/NOX4…
The Annexin A1 mimetic peptide Ac 2-26 preserved mitochondrial function and neuronal survival in human retinal organoids and organoid-derived retinal ganglion cells exposed to hyperglycemic and inflammatory stress, according to a laboratory study reported in August 2026. The protective effects were associated with FPR2 -dependent suppression of the p38/NOX4 signaling axis, a mechanism the investigators propose as a candidate target for early intervention in diabetic retinopathy .
Diabetic retinopathy is a neurodegenerative disease characterized by hyperglycemia-induced inflammation, mitochondrial dysfunction, and progressive neuronal cell death leading to irreversible vision loss. Retinal ganglion cells are particularly vulnerable and undergo apoptosis before microvascular pathology becomes clinically evident, meaning that by the time vascular lesions appear on examination, a portion of the neuronal population may already be lost. Therapeutic strategies targeting these early neurodegenerative events remain limited, and most existing treatments act on the later vascular phase of the disease.
The study positions Ac 2-26 as a pro-resolving neuroprotective agent that targets convergent inflammatory and mitochondrial pathways. Because the experiments used human retinal tissue models rather than rodent retina, the findings provide a human-relevant readout of a mechanism previously studied mainly in animal systems. The work also extends peptide-based intervention into diabetic retinopathy, a field dominated by anti-VEGF biologics, laser photocoagulation, and vitrectomy.
In retinal organoids, exposure to high glucose produced the degenerative changes typical of the diabetic retina, and Ac 2-26 significantly attenuated high glucose-induced retinal degeneration. The peptide suppressed both the intrinsic and extrinsic apoptotic pathways, the two principal routes by which retinal neurons are programmed to die. It also reduced Müller cell gliosis , the reactive response of the retina's principal glial cells to neuronal injury, and selectively modulated inflammatory mediators associated with diabetic retinopathy progression.
The organoid-derived retinal ganglion cell experiments addressed the mitochondrial dimension. Ac 2-26 reduced mitochondrial oxidative stress and restored mitochondrial bioenergetics under hyperglycemic and inflammatory conditions. Those metabolic improvements were accompanied by prevention of neuronal apoptosis, which links the peptide's bioenergetic effects to cell survival in the neuronal population most vulnerable in diabetic retinopathy.
The molecular correlate of these effects was FPR2-dependent suppression of the p38/NOX4 signaling axis. The abstract reports an association between that pathway and the protective phenotype; causation was not explicitly proven. What is notable is the specificity: the study identifies a defined receptor-mediated signaling cascade rather than a generalized antioxidant mechanism, giving follow-up work a concrete molecular target to test.
The study was a laboratory investigation using human retinal organoid models and organoid-derived retinal ganglion cell models exposed to high glucose and inflammatory conditions, with Ac 2-26 as the intervention. The sample size and duration of exposure were not reported in the study abstract, and no quantitative data on sample sizes, doses, or effect magnitudes are provided in the abstract. The evidence is therefore directional: it establishes the presence of protective effects and a candidate mechanism, but not their magnitude or kinetics.
The endpoints measured were broad, reflecting an attempt to capture the disease as a tissue-level process rather than a single biochemical readout:
Human retinal organoids are three-dimensional cultures differentiated from induced pluripotent stem cells that recapitulate key features of retinal development and physiology. They are among the closest in vitro approximations of human retinal tissue available, and they allow the study of neuroglial interactions and neuronal injury in a human genetic background. Organoid-derived retinal ganglion cells take this further by isolating the cell type most vulnerable in diabetic retinopathy for focused analysis.
What the design cannot demonstrate is clinical efficacy. Organoids lack a vasculature, so the model cannot capture the microvascular pathology, ischemia, and blood-retinal barrier dysfunction that define the clinical disease. The findings establish that Ac 2-26 acts directly on retinal neurons, but they say nothing about its effects on endothelial cells, pericytes, or the diabetic circulation. The design demonstrates association between FPR2/p38/NOX4 suppression and neuroprotection; it does not prove causation.
Annexin A1 is a glucocorticoid-regulated protein with established roles in the resolution of inflammation. Ac 2-26 is the N-terminally acetylated peptide fragment spanning residues 2 through 26 of the protein, and it acts as a peptide agonist at formyl peptide receptor 2 FPR2 , a G protein-coupled receptor expressed on immune cells and, on the evidence of this study, relevant to retinal neurons.
FPR2 belongs to the pro-resolving machinery of the innate immune system. Where conventional anti-inflammatory approaches block individual pro-inflammatory signals, pro-resolving mediators actively terminate inflammation and promote tissue repair. A peptide that engages FPR2 does more than neutralize a single cytokine; it recruits an endogenous resolution program that acts across multiple inflammatory pathways. That is consistent with the study's finding that Ac 2-26 selectively modulated inflammatory mediators rather than suppressing them indiscriminately.
The p38 MAPK pathway is a stress-responsive kinase cascade activated by inflammatory cytokines, oxidative stress, and metabolic disturbance. NOX4 is a NADPH oxidase that generates reactive oxygen species, and in the retina it has been linked to hyperglycemia-induced mitochondrial dysfunction. The p38/NOX4 connection matters because it links inflammatory signaling directly to oxidant production. By suppressing this axis, Ac 2-26 appears to interrupt the circuit between inflammation and mitochondrial damage, which would explain why both mitochondrial oxidative stress and bioenergetic failure were attenuated.
That circuit also helps account for the breadth of the observed effects. The intrinsic apoptotic pathway is triggered by mitochondrial stress and the release of pro-apoptotic factors from the organelle; the extrinsic pathway is triggered by death receptor engagement at the cell surface. Müller cell gliosis is the retinal signature of metabolic injury, and its reduction suggests the peptide's effects extend beyond neurons to the glial cells that support them. Together, these observations frame the FPR2/p38/NOX4 axis as a promising therapeutic target for early intervention in diabetic retinopathy.
For researchers, the study identifies a specific axis to validate with loss-of-function experiments. FPR2 knockdown or pharmacological antagonism should abolish Ac 2-26's protective effects if the receptor is truly required, and NOX4 inhibition should phenocopy the peptide's mitochondrial protection. The open question of whether other Annexin A1 mimetic peptides show similar or enhanced efficacy invites structure-activity comparisons across the annexin peptide family, and the organoid platform provides a human-relevant system for performing them.
For clinicians, the implications concern timing. If retinal ganglion cells die before microvascular pathology is clinically evident, then neuroprotective intervention must be delivered early, during a phase of diabetes currently managed with glycemic control and surveillance. A peptide that prevents neuronal apoptosis and preserves mitochondrial bioenergetics represents a fundamentally different approach from anti-VEGF injections, which treat established vascular leakage and neovascularization. The study does not show that Ac 2-26 can be delivered to the human retina, but it supports the rationale for trying.
For the peptide supply chain, Ac 2-26 is a 25-residue N-terminally acetylated peptide, a molecule of a size and composition well suited to solid-phase synthesis. The acetylation at the N-terminus is a standard modification that can improve stability against aminopeptidases. The open questions are delivery and duration: intravitreal injection is the established route for retinal therapeutics but burdens patients with repeated procedures, while systemic delivery must contend with peptide clearance and the blood-retinal barrier. Formulation strategies that extend ocular residence time would be a prerequisite for any clinical program. The broader relevance to peptide research is that the same receptor-mediated logic, a pro-resolving peptide acting through FPR2 to suppress a defined stress kinase and oxidase axis, could inform similar peptides for other neurodegenerative and inflammatory diseases.
The evidence base has clear boundaries. The findings derive from in vitro organoid models, not from human clinical subjects, and they were not confirmed in vivo. The sample size, duration, doses, and effect magnitudes were not reported in the study abstract, which limits the ability to assess the strength of the effects. And the central mechanistic claim, FPR2-dependent suppression of p38/NOX4 signaling, is an association reported in the abstract rather than a demonstrated causal chain.
The open questions are correspondingly concrete. The optimal dosing, route, and timing of Ac 2-26 administration are unknown. Whether the protective effects translate to in vivo diabetic retinopathy models and, ultimately, human patients has not been tested. Whether Ac 2-26 also affects retinal microvascular pathology or other retinal cell types is unanswered, because the organoid models lack a functional vasculature. The long-term safety and off-target effects of FPR2 activation by Ac 2-26 have not been characterized, and FPR2 is expressed on multiple immune cell populations where sustained activation could have unintended consequences.
What would settle these questions is also clear. In vivo studies in diabetic rodent models could test whether Ac 2-26 preserves retinal ganglion cell survival, retinal function, and the retinal vasculature, and they could establish dosing and pharmacokinetics. Genetic experiments in FPR2-deficient animals could convert the reported association into a causal demonstration. Human clinical evaluation would ultimately require a neuroprotective endpoint measured before vascular pathology appears, which would in turn require a shift in how diabetic retinopathy trials are designed. For now, the study's contribution is to identify a mechanism worth pursuing: a pro-resolving peptide acting through FPR2 to suppress p38/NOX4 signaling and preserve the mitochondria that retinal neurons depend on.
Related reading: GLP-1 Drugs May Cut Protein Intake, Raising Sarcopenia Risk, Illumina Beats Earnings as Director Meister Sells $140M in Shares, Semaglutide hsCRP cut of 37.8% in SELECT suggests anti-inflammatory role, FDA's Compounded GLP-1 Warning Letters and the Embody Telehealth Model.