A 3D In Vitro Cortical Tissue Model Based on Dense Collagen to Study the Effects of Gamma Radiation on Neuronal Function

Abstract

Studies on gamma radiation-induced injury have long been focused on hematopoietic, gastrointestinal, and cardiovascular systems, yet little is known about the effects of gamma radiation on the function of human cortical tissue. The challenge in studying radiation-induced cortical injury is, in part, due to a lack of human tissue models and physiologically relevant readouts. Here, a physiologically relevant 3D collagen-based cortical tissue model (CTM) is developed for studying the functional response of human iPSC-derived neurons and astrocytes to a sub-lethal radiation exposure (5 Gy). Cytotoxicity, DNA damage, morphology, and extracellular electrophysiology are quantified. It is reported that 5 Gy exposure significantly increases cytotoxicity, DNA damage, and astrocyte reactivity while significantly decreasing neurite length and neuronal network activity. Additionally, it is found that clinically deployed radioprotectant amifostine ameliorates the DNA damage, cytotoxicity, and astrocyte reactivity. The CTM provides a critical experimental platform to understand cell-level mechanisms by which gamma radiation (GR) affects human cortical tissue and to screen prospective radioprotectant compounds.

Keywords: cortical tissue models; gamma rays; hiPSC; ionizing radiation; natural polymers.

Figure 1

CTM preparation and experimental workflow. CTMs were generated by combining cells with rat tail type I collagen and basement membrane proteins. Upon gelation, highly hydrated collagen‐based hydrogels underwent plastic compression to remove excess water and increase the protein content, resulting in a dense collagen‐based CTM. After 26 days in culture, CTMs were irradiated with a Co⁶⁰ GR source at 2.5 Gy min⁻¹ for 2 min to simulate a 5 Gy dose to cortical tissue. Cultures were treated before radiation exposure with the commercially available radioprotectant (RP) amifostine by addition to culture media. CTMs were morphologically and mechanically characterized. The effect of GR on CTM function was quantified via cytotoxicity, DNA damage, cellular morphology, and electrophysiological behavior.

Figure 2

CTM morphological and mechanical characterizations. A) Macroscopic image of SH‐SY5Y seeded CTM cultured to DIV 64. B) SEM analysis demonstrating suprastructural (quarter‐staggered) organization of collagen fibrils. Scale bar represents 2 µm. C) Representative stress/strain curve of collagen‐based constructs undergoing unconfined compression with compressive modulus in the range of native soft tissues (D). E) Immunocytochemistry staining for DAPI and β III tubulin in SH‐SY5Y seeded CTM cultured to DIV 64. Scale bar represents 50 µm.

Figure 3

Biological response of SH‐SY5Y ‐seeded CTM to GR. A) Representative maximum intensity projections of confocal laser scanning microscopy (CLSM) analysis of CTMs seeded with SH‐SY5Y cells stained for DAPI (blue), γH2AX (green), and 53BP1 (red) at 30 min post‐treatment with 5 Gy GR. Scale bar represents 100 µm. B) Quantification of DNA damage markers γH2AX and 53BP1 mean intensity per nuclei at 30 min post GR treatment. Significant effect of radiation treatment on DNA damage marker expression in CTMs for both γH2AX (p = 0.00021) and 53BP1 (p = 0.018) nuclear intensities. C) LDH release from CTMs at 24 h after treatment (p = 0.00093).

Figure 4

Biological response of hiPSC neuron and astrocyte‐seeded CTMs to GR. A) Representative maximum intensity projections of CLSM analysis of CTMs seeded with hiPSC‐derived neurons and astrocytes stained for DAPI (blue), γH2AX (green), and 53BP1 (red) at 30 min post‐treatment with 5 Gy GR. Scale bar represents 50 µm. B) Distribution of cells in a 3D volume of CTM. C) Quantification of DNA damage markers γH2AX and 53BP1 mean intensity per nuclei at 30 min post GR treatment. Significant effect of radiation treatment on DNA damage marker expression in CTMs for both γH2AX (p = 2.22 × 10⁻¹⁶) and 53BP1 (p = 2.22 × 10⁻¹⁶).

Figure 5

Morphological changes of hiPSC neuron and astrocyte‐seeded CTMs to GR. A) Representative images of maximum intensity projections of CLSM analysis of CTMs stained for NeuN and GFAP. B) Significant effect of GR treatment in neurite length quantification (p = 4.4 × 10⁻¹⁴) in CTMs relative to untreated samples at 24 h after treatment. Scale bars represent 50 µm.

Figure 6

Electrophysiological activity changes of hiPSC neuron and astrocyte‐seeded CTMs to GR. A) Representative voltage versus time traces for EAPs recorded from 2D co‐cultures. Vertical and horizontal axes represent amplitude (µV) and time (ms), respectively. B) Time‐progression and stability of 2D co‐cultures in terms of AEY and MFR (left and right axis, respectively). C) Representative raster plots before (left) and 24 h (right) following GR exposure. Synchronized network bursts highlighted by pink rectangles. D) Difference‐over‐sum (DoS) normalized changes in MFR (left) and NBR (right) in 2D co‐cultures at 1‐ and 24‐h following GR‐exposure. E) Representative raw trace (left) of voltage versus time exhibiting MUA activity recorded from 3D CTMs. Pharmacological responsiveness (right) of 3D CTMs to glutamate (10 μm) and sodium channel antagonist lidocaine (100 μm).

Figure 7

Biological response of hiPSC‐derived neuron and astrocyte‐seeded CTMs to radioprotectant amifostine upon GR. Quantification of DNA damage markers γH2AX (A) and 53BP1 (B) mean intensity per nuclei at 30 min post GR treatment with and without radioprotectant. Significant effect of radiation treatment on DNA damage marker expression in CTMs for both γH2AX (p = 2.22 × 10⁻¹⁶) and 53BP1 (p = 2.22 × 10⁻¹⁶) and significant effect of amifostine on DNA damage marker expression in CTMs for both γH2AX (p = 5.055 × 10⁻¹⁴) and 53BP1 (p = 7.51 × 10⁻¹⁸). C) Representative images of maximum intensity projections of CLSM analysis of CTMs stained for neuronal marker NeuN in gamma‐irradiated or untreated CTMs treated with the radioprotectant amifostine or vehicle. Scale bar represents 50 µm. Quantification of neurite length in 2D (D) and CTMs (E). Significant effect of amifostine in neurite length quantification (p = 8.191 × 10⁻⁷) in CTMs relative to vehicle control upon RG exposure.