(J-390) A 3D in vitro cortical tissue model based on dense collagen to study the effects of gamma radiation on neuronal function

Introduction:: Nuclear attacks and accidental exposures represent rare but critical modes of ionizing radiation exposure and may lead to acute to long-term injury or death. Understanding the biological mechanisms behind gamma radiation (GR)-induced injury is paramount for both medical and public health preparedness. Most IR-induced injury studies have addressed one or more of the following systems/tissues: bone marrow, gastrointestinal tract, lungs, skin, vasculature, and kidneys. However, IR, including GR, is also associated with radiation-induced cognitive decline (RICD); likely attributable to disruption of ‘healthy’ neuronal network function. Currently, there are limited experimental models which mimic the structural, physical, biological, and phenotypic characteristics of human cortical tissue to support short-to-long-term investigation of GR exposure on neuronal network function. Traditional in vivo systems have successfully illuminated certain impacts of IR on the brain, but face limitations due to intrinsic differences between the human brain and those of model organisms. More recently, In vitro CNS models have been used to study IR damage, however, these studies have primarily been carried out on 2D substrates, using tumorigenic cell lines or embryonic animal tissues and lack important aspects of human cortical tissue physiology including 3D structure, network connectivity, and glial cell types. Here, we describe a novel 3D, human cell-based cortical tissue model (CTM) capable of integrating hiPSC based neurons and astrocytes and a GR treatment workflow that enables traditional measures of cytotoxicity and DNA damage as well as high-content phenotypic assessments of neuronal network function.

Materials and Methods:: Dense collagen CTMs were fabricated by applying a known compressive load to polymerized type-I rat tail collagen pre-mixed with human ECM proteins and human SHSY-5Y cells or iPSC derived glutamatergic neurons and human iPSC derived spinal astrocytes (fig. 1A). To characterize morphological and mechanical properties, acellular CTMs were imaged using scanning electron microscopy. hiPSC cell seeded CTMs were cultured for 26 days before treatment with 5 Gy of GR delivered at a dose rate of 2.5 Gy/ min from a ⁶⁰Co source at the UMass Lowell Radiation Laboratory. DNA damage was quantified at 30 min post treatment by immunocytochemistry (ICC) staining for canonical DNA double stranded break markers gH2AX and 53BP1. Cellular morphology was assessed at 24 hrs following treatment by ICC for the neuronal marker NeuN and the astrocyte marker GFAP. All ICC quantification was analyzed using custom FIJI macros. Cytotoxicity was measured at 24 hrs following treatment using a standard lactate dehydrogenase assay (LDH). Extracellular electrophysiological recordings for both 2D and 3D samples were carried out using the Axion Maestro Pro multiwell multielectrode array (MEA) recording platform. To identify extracellular action potentials (EAPs) recordings were subject to an adaptive threshold set to ±5.5 standard deviations of the channel's adaptive RMS. Mean firing rates (MFR) and average electrode yield (AEY), calculated from 10-min recordings were compared to their baseline values through difference-over-sum (DoS) normalization.

Results, Conclusions, and Discussions:: Results: SEM analysis revealed superstructural (quarter-staggered) organization of collagen fibrils, as expected for this biofabrication method. Cell seeded-CTMs were cultured for 26 days prior to IR exposure and exhibited markers of neuronal and astrocyte maturation. We found that upon 5 Gy GR exposure CTMs showed significantly increased cytotoxicity, DNA damage, and astrocyte reactivity while neurite length and neuronal network activity as measured by MEAs significantly decreased. Additionally, we found that the clinically deployed radioprotectant amifostine ameliorated the DNA damage, neurite length reduction (fig 1B-E), and astrocyte reactivity.

Discussion: The described CTM and IR induced injury measurement workflows fill a critical technology gap left by current models and are necessary for examining the biological mechanisms which underly IR induced cortical tissue injury. We have demonstrated the potential utility of this platform for evaluating prospective radioprotectants, a valuable research tool for use in both clinical and military applications. We report the development of a dense type-I collagen-based CTM that incorporates human basement membrane proteins and supports the growth, differentiation, and function of SH-SY5 cells as well as human iPSC neurons and astrocytes. Although type-I collagen is not present in the brain, it is a widely used and well-characterized fibrillar protein that provides a biocompatible, tunable, structurally stable 3D environment, as biomanufacturing platform for CTM.

Conclusion: Here, we report on the development of a physiologically relevant cortical tissue model (CTM) that incorporates the fundamental neuroglial unit and approaches the structural and mechanical features of native cortical tissue. Our results indicate that our CTM recapitulates standard measures of IR-induced injury reported in vivo (e.g., DNA damage, cytotoxicity), and responds to the effect of radioprotectant countermeasures in agreement with clinical studies. Additionally, our data indicates that phenotypic activity measured using in vitro MEAs could support addressing fundamental questions regarding the GR effects on cortical networks and serving as a platform for radioprotectant drug identification and development. Importantly, the CTM’s stability and robustness makes it suitable for a variety of radiation sources and durations/repetitions of exposure, including proton beam and mixed exposures for modeling short-to-long term radiation therapy and/or space radiation exposure.


Acknowledgements (Optional): : UMass Lowell Radiation Laboratory, NASA NSTGRO 2023 fellowship

References (Optional): :