McNally, M. | bioRxiv Science

bioRxiv2025

Application of machine learning to discriminate photoreceptor cell species in xenotransplanted chimeric retinas

Photoreceptor transplantation is being studied to improve visual function in retinal diseases causing blindness, such as age-related macular degeneration, hereditary eye diseases, and traumatic retinopathy, among others. Preclinical studies often involve the delivery of exogenous human photoreceptor cells into the retinas of animal models. In such experiments, a key readout is the differential frequency of donor cell somatic integration versus artificial labeling secondary to material transfer of cytosolic or nuclear labels from donor to recipient cells. For this analysis, the ability to recognize photoreceptor nuclei as being of donor (human) versus animal is key, but purely immunohistology discrimination can be challenging because of antigenic species overlap or intercellular antigen transfer. To address this challenge, we sought to develop and validate a computational technique to discriminate between photoreceptor cells of different animal species based on machine learning of nuclear morphology. Here, we aimed to evaluate the feasibility of using computer-assisted detection of separate nuclei and employing random forest classification to automate the species differentiation, among DAPI-stained photoreceptors after xeno-transplantation of human photoreceptors into the retinas of mice and pigs. Our models were trained on single-species samples and validated with mixed-species samples. We then transplanted human embryonic stem cell-derived retinal organoid cells into rodent and pig retinal degeneration models. The random forest model accurately determined cell identity post-xenotransplantation, validated by histological assessment using an anti-human nuclear antibody. Our results support the potential efficacy of employing machine learning image analysis and classification techniques that may promote experimental rigor, minimize observer bias, and enable high throughput semi-automated workflows for transplantation outcomes analysis. The methodological framework reported here may enable a more nuanced and precise analysis of the behavior of transplanted photoreceptors for the purposes of human retinal regeneration.

neuroscience↗

bioRxiv2024

Ongoing loss of viable neurons for weeks after mild perinatal hypoxia-ischemia

Mild hypoxic-ischemic encephalopathy is common in neonates with no evidence-based therapies, and 30-40% of patients experience adverse outcomes. The nature and progression of mild injury is poorly understood. Thus, we studied the evolution of mild perinatal brain injury using longitudinal two-photon imaging of transgenic fluorescent proteins as a novel readout of neuronal viability and activity at cellular resolution. In vitro, perinatal murine organotypic hippocampal cultures underwent 15-20 minutes of oxygen-glucose deprivation. In vivo, mild hypoxia-ischemia was completed in post-natal day 10 mouse pups of both sexes with carotid ligation and 15 minutes of hypoxia. Consistent with a mild injury, minimal immediate neuronal death was seen and there was no volumetric evidence of injury by ex vivo MRI 2.5 weeks after injury. In both the hippocampus and neocortex, these mild injuries resulted in a significantly delayed and progressive neuronal loss in the second week after injury, measured by fluorophore quenching. Mild hypoxia-ischemia transiently suppressed cortical network activity followed by normal maturation. No post-injury seizures were seen. The participation in network activity of individual neurons destined to die was indistinguishable from those that survived for 4 days post-injury. In conclusion, our results showed that mild perinatal brain injury resulted in a prolonged increase of neuronal death. Neurons that died late were functioning normally for days after injury, suggesting a new pathophysiology of neuronal death. Critically, the neurons destined to die late demonstrated multiple biomarkers of viability long after mild injury, suggesting their later death may be modified with neuroprotective interventions. SIGNIFICANCE STATEMENTNeonatal encephalopathy due to peripartum hypoxia-ischemia (HI) is a major cause of neonatal mortality and morbidity worldwide. Of these infants, most are categorized as having mild HI. Infants with mild HI have significant long-term disabilities. There are currently no evidence-based therapies, largely because the progression and pathophysiology of mild injury is poorly understood. We have identified, for the first time, that mild perinatal HI results in a delayed and prolonged increase in neuronal death. The cortical and hippocampal neurons that die over a week after injury participate normally in neural network activity and exhibit robust viability for many days after injury, indicating a novel pathophysiology of neuronal death. Clinically, these data suggest an extended therapeutic window for mild perinatal HI.

neuroscience↗