Clinical Trial in a Dish for Space Radiation Countermeasure Discovery

Abstract

NASA aims to return humans to the moon within the next five years and to land humans on Mars in a few decades. Space radiation exposure represents a major challenge to astronauts’ health during long-duration missions, as it is linked to increased risks of cancer, cardiovascular dysfunctions, central nervous system (CNS) impairment, and other negative outcomes. Characterization of radiation health effects and developing corresponding countermeasures are high priorities for the preparation of long duration space travel. Due to limitations of animal and cell models, the development of novel physiological relevant radiation models is needed to better predict these individual risks and bridge gaps between preclinical testing and clinical trials in drug development. “Clinical Trial in a Dish” (CTiD) is now possible with the use of human induced pluripotent stem cells (hiPSCs), offering a great tool for drug safety or efficacy testing using patient-specific cell models. Here we review the development and applications of CTiD for space radiation biology and countermeasure studies, focusing on progress made in the past decade.

1. Introduction

Since the first human journey to space in 1961, mission durations have progressively increased, so have cumulative times in space for individual astronauts, with the longest record standing at 438 days. Currently, long-duration missions are limited to the space stations at Low-Earth Orbit (LEO), where astronauts are exposed to ionizing radiation (IR) originating from three sources: galactic cosmic rays (GCRs), solar particle events (SPEs), and electrons and protons trapped in the Van Allen Belts (TPs) (Benton and Benton, 2001). High-energy, high charge (HZE) ions are key components of GCR and understanding their health effects is an essential goal of space radiobiology. HZE ions and their fragmentation products can easily penetrate spacecraft shielding and cause molecular, cellular and tissue damage in astronauts. Space travel beyond the protection of Earth’s magnetic field (e.g., Moon, Mars and deep space) will result in increased exposures to HZE ions significantly increasing effective doses to astronauts (Rapp, 2006). HZE ions create dense ionization tracks along their trajectories in living tissue, known as high-linear energy transfer (LET).

Health concerns about HZE ion exposures arise from epidemiologic studies of exposures in medical and occupational settings to another source of high-LET radiation, alpha particles, which were found to be far more effective for cancer induction than low-LET radiations (Mortezaee et al., 2019). Rodent irradiation studies also confirmed the enhanced carcinogenic efficiency of charged particles, which highly depended on genetic background of rodent strain (Bielefeldt-Ohmann et al., 2012). However, the health risks of space radiation remain poorly understood due to the lacking of a sufficiently large exposed cohort for an adequately powered epidemiological study. Current understanding of space radiation-related health risks is mainly based on effects of anticipated absorbed dose ranges using ground-based radiation analogues and limited flight research. Two previous studies compared hundreds of astronauts with space flight experience to control populations on the ground, and neither study found cancer risks increased by space flight (Hamm et al., 1998; Reynolds et al., 2019). In 2019, NASA performed a study to systematically compare a male astronaut during and after a year-long mission to his twin brother as a ground control, by integrating multiple data types. Most measurements of biological functions of the flown twin were not significantly affected or they returned to a prefight state compared to the ground control. However, some changes persisted even 6 months after the flight, including gene expression changes, increased levels of DNA damage, shortened telomeres, and attenuated cognitive function (Garrett-Bakelman et al., 2019). Space radiation exposure affects multiple organs and physiological systems, and its health risks are divided into four areas by NASA (Chancellor et al., 2014): degenerative tissue effects, carcinogenesis, acute and late CNS effects, and radiation syndromes due to SPEs.

Protection from space radiation-related health issues using biological strategies is still at the preliminary stage. Reagents that can potentially ameliorate radiation-induced DNA damage and free radicals have drawn much interest in the form of radioprotective drugs, especially antioxidants (Brink and Jr, 2012; Smith et al., 2017). However, most radioprotective drugs so far were developed for acute high-dose radiation exposure from radiotherapy, which may not offer similar protection against chronic exposure to low dose/dose rate of high-LET radiation. Drugs that can protect astronauts from injuries caused by both rare high-dose radiation from SPEs and constant low-dose GCRs are needed.

To understand adverse effects of space radiation on human health, it is important to gain basic knowledge about the radiosensitivity of each tissue type and individual susceptibility to radiation, particularly in developing radiation countermeasures for extended space travel. Existing animal models cannot fully recapitulate the human physiology or mutational changes induced by radiation, and the development of radioprotective drugs also requires access to various cell types/tissues from a large population carrying different genetic backgrounds. This was not feasible until the advent of human induced pluripotent stem cells (hiPSCs) and their subsequent applications in preclinical drug development known as “clinical trial in a dish (CTiD)”. This review covers three main areas: an introduction about hiPSCs and CTiD; radiation studies done in the past decade using human primary cells; and recent advances in radiation biology and radioprotective countermeasures development using hiPSCs and CTiD (Figure 1).

Figure 1.

Schematic overview of the application of Clinical Trial in a Dish (CTiD) in space radiation countermeasure discovery. Created with BioRender.com.

2. hiPSCs as ideal cell source for CTiD

2.1. Definition of CTiD

The average cost of developing a new drug passing Food and Drug Administration (FDA) approval was $2.87 billion in 2016, according to a report by Tufts Center for the Study of Drug Development (DiMasi et al., 2016). Only 20% of the cost was spent on the drug discovery and preclinical tests, with the remaining 80% being spent on clinical trials and applications for Investigative New Drug (IND) status and FDA approvals. The low successful rate (11.83%) of clinical trials was a major reason for the extremely high cost (DiMasi et al., 2016). Therefore, there is an urgent need for the pharmaceutical industry to develop new strategies to shorten the gap from preclinical testing to clinical trial, which involves improving the efficiency of screening candidate drugs and accuracy in predicting outcomes of clinical trials. CTiD involve the testing of the incidence and degree of a target population’s responses to candidate therapies using human cell-based in vitro systems. CTiD has attracted tremendous interest in the past few years because it provides a powerful new tool to test candidate drugs systematically on patient-specific cells before moving toward clinical trials. This platform is valuable not only for drug development for a specific type of disease, but also for assessing individual patient’s susceptibility to the harm in question (e.g., radiation exposure) and designing of personalized medical treatments (countermeasures) if needed.

2.2. Role of hiPSCs in CTiD

Major requirements for CTiD include the access to high numbers of high-quality patient-specific cell types and ability to accurately simulate the physiological in vivo condition using these cells. With the emergence and advancement of hiPSCs, the potential and feasibility of CTiD significantly improved. Dr. Shinya Yamanaka made the seminal discovery of hiPSCs in 2006 by introducing the Yamanaka factors (Oct3/4, Sox2, Klf4, and c-Myc) into terminally differentiated somatic cells (Takahashi and Yamanaka, 2006). hiPSCs can be generated from donor’s cells of various origins and have an unlimited renewal capacity and differentiation potential into nearly all cell types (Paik et al., 2020). hiPSCs have significantly advanced the biomedical research in many aspects since their discovery due to several unprecedented advantages. First, hiPSC-derived cells can provide an unlimited source of autologous cells for cellular therapies of degenerative diseases (Yamanaka, 2020). Second, hiPSCs carrying patient-specific genetic information offer an excellent opportunity to model genetic disorders and investigate their etiologies (Rowe and Daley, 2019). Moreover, hiPSCs can be engineered using genome editing techniques, such as zinc finger nucleases (ZFN), transcription activator–like effector nuclease (TALEN), and the clustered regularly interspaced short palindromic repeat (CRISPR) (Kim and Kim, 2014). These novel tools allow us to generate human genetic disease models in a precise and predictable manner. Already, genome editing has been successfully applied to correct certain genetic mutations and to model various diseases such as cardiomyopathies caused by single genetic mutation, including dilated cardiomyopathy, Barth syndrome, long-QT syndrome, and Duchenne muscular dystrophy (Karakikes et al., 2015; Long et al., 2014; G. Wang et al., 2014; Y. Wang et al., 2014). Third, hiPSCs provide unlimited number and all types of patient-specific cells to be used for screening of candidate drugs in a high-throughput manner. Lastly, hiPSCs offer a unique tool to generate physiologically relevant tissues in vitro to predict individual disease risks and drug responses by providing crucial information on their genetic backgrounds, greatly facilitating the drug development process (Sayed et al., 2016).

2.3. High-throughput drug screening using hiPSCs

More than 80% of tested drugs fail during clinical trial stage due to toxicity and/or lack of efficacy (Kola and Landis, 2004). Drug toxicities accounted for 35% and 28% of drug failure in phase I and phase II clinical trials, respectively, implicating mainly the cardiovascular system, liver, gastrointestinal system, and CNS. Many high-profile drugs have been withdrawn from the market because of cardiotoxicity, such as the nonsteroidal anti-inflammatory drug rofecoxib, the gastrointestinal prokinetic drug cisapride, and the broad-spectrum antibacterial drug grepafloxacin (Sallam et al., 2015). More efficient systems are needed to better predict patients’ drug responses.

Immortalized cell lines and animal models have been widely used for drug testing, but have had limited success due to limited relevance to human physiology and pathophysiology. Thus, scalable human alternatives are in great demand, as they would be useful for predictive screening of a big pool of early-stage drugs. The restricted availability and pronounced batch variations of primary cardiomyocytes (CMs) and hepatocytes preclude their application in systematic evaluation of cardiotoxicity and hepatoxicity of early-stage drug candidates. Unlike conventional approaches, hiPSCs offer a non-invasive and unlimited supply of patient-derived cells/tissue. With the advancement of robust differentiation methods, especially the application of defined induction medium, it is now feasible to obtain CMs/hepatocytes from variable hiPSC lines with minimal batch effects. This enables the high-throughput drug screening of human cells, bypassing the impracticality of harvesting primary cells.

Recent advances in the scalable production of CMs and hepatocytes from hiPSCs made it possible to develop automated assays to predict drug-induced arrhythmias and hepatotoxicity. For example, the comprehensive in vitro Proarrhythmia Assay (CiPA) initiative aims to replace ICH S7B guidance for nonclinical evaluation of compounds for potential QT prolongation, mainly by using multiple electrode arrays (MEAs) or voltage-sensing dyes to detect drug-induced repolarization defects in hiPSC-CMs (Gintant et al., 2017). However, the application of hiPSC-derived hepatocytes in drug discovery and preclinical toxicology is limited by their immature phenotypes compared to primary hepatocytes (Sachinidis et al., 2019). This problem is being tackled by the rapid advancement of differentiation and culture methods. By solving the inherent difficulty in obtaining primary neurons, the recent advances in scalable generation of hiPSC-derived neural cells types is a remarkable transformative development for neuroscience drug discovery. With the continuingly progress in developing well-validated and standardized protocol for various cell types, hiPSCs will become even more widely used by biopharmaceutical scientists and industry for the preclinical drug development.

2.4. Precision Medicine

Patients show various reactions to the same medication due to each individual’s unique genetic background. Patients with similar symptoms are usually treated with the same medication, despite possible different disease pathologies. Patient-specific hiPSCs carry the same genetic information with their donors, thus providing an ideal patient-specific drug screening platform to evaluate the effect of a personalized countermeasure for each individual (Collins and Varmus, 2015). This precision medicine approach allows us to investigate disease pathology in a patient-specific manner and design/screen most optimal drugs for each individual. In addition to individualized treatment, hiPSCs can also be used for high-throughput drug toxicity and efficacy assays with a large cohort of patients (Inoue et al., 2014), to categorize patients as drug responders and non-responders. Only drug responders would be selected for further clinical trials, which will dramatically reduce the cost and improve efficiency of clinical trials. Furthermore, drug response-related genes and single nucleotide polymorphisms (SNPs) can be identified using expression quantitative trait locus (eQTL) analysis (Rong et al., 2021), which can be applied as biomarkers to predict individual risks and responses.

3. Space radiation effects: lessons learned from conventional cell models

Extrapolating radiation effects from animal models to humans and understanding interindividual heterogeneity in radiation responses among humans requires in vitro and ex vivo studies in human cells and tissues. Several human primary cell types that are relatively easy to obtain or expand in vitro, including blood, endothelial cells (ECs) and fibroblasts, are widely used for current radiobiology studies. Despite lacking genetic diversity and physiologically relevant environment in most situations, in vitro models made from primary cells provide an accessible, robust, and cost-effective platform to investigate human-specific responses to space radiation and screen candidate countermeasures at initial stages of drug development. Most studies published so far have utilized low-LET radiation (e.g., x-ray and gamma rays), which compromises their translational value for understanding the biological effects. Although low- and high-LET radiation affect normal tissues to different extents (Niemantsverdriet et al., 2012), they can create some common effects on human cells, such as DNA damage, mitochondria and oxidative stress. Both low- and high-LET radiation could lead to mitochondrial dysfunctions, including excessive ROS and reduced ATP production in various cell types at different extents (Averbeck and Rodriguez-Lafrasse, 2021). By analyzing datasets deposited in NASA’s GeneLab database, investigators observed common activated molecular pathways between human umbilical vein ECs (HUVECs) exposed to simulated space radiation and HUVECs flown to the International Space Station (ISS) (Beheshti et al., 2019). Considering the large quantities of low-LET studies, they may still give critical insights into biological effects of space radiation and are discussed in this review.

3.1. Radiation effects on human endothelial cells

The endothelium within vascular networks is one of the major cell compartments affected by the whole-body radiation exposure and is considered as a promising target for potential space radiation countermeasures. EC responses to radiation exposure cover a wide range of molecular and phenotypic changes. Among them, inflammation is commonly observed in irradiated ECs. Gamma rays can upregulate ICAM-1 expression on various EC subtypes, including human dermal microvascular ECs (HDMECs), HUVECs and transformed human microvascular ECs (HMEC-1). However, another adhesion molecule, E-selectin, was found to be upregulated on HDMECs but not on other two types of ECs, suggesting a heterogeneity of radiation responses across different types of ECs (Prabhakarpandian et al., 2001). In a follow-up study done by the same group, PKCδ inhibition was found to protect ECs from radiation-induced inflammation (Soroush et al., 2018). Inflammatory responses of ECs induced by IR were confirmed across multiple studies (Dong et al., 2015; Eckert et al., 2021; Haubner et al., 2013; Jang et al., 2020; Wang et al., 2019; Wei et al., 2017). Importantly, multiple countermeasures that could mitigate radiation-induced EC inflammation have been tested using in vitro cell models and found to be protective, including NFkB inhibitor PS1145 (Dong et al., 2015), Pravastatin (Jang et al., 2020) and Captopril (Wei et al., 2017). Cell barrier, which is correlated to EC inflammation responses, was also reported to be significantly disrupted after radiation (Gabryś et al., 2007; Kabacik and Raj, 2017; Sharma et al., 2013; Tang et al., 2021), whereas the knocking-down of USP11 had been shown to improve EC junctions after x-ray irradiation (Tang et al., 2021).

Radiation exposure also leads to mitochondrial dysfunction and senescence of ECs, which can be mitigated or reversed by Rosiglitazone (Baselet et al., 2020) and Metformin (Park et al., 2022), respectively. In terms of DNA damage caused by IR, Puerarin and Metformin were reported as effective protective drugs for ECs (Liu et al., 2022; Park et al., 2022). In addition, radiation was also found to cause viability loss of ECs in a radiation dose-dependent manner (Kacem et al., 2021).

Next-generation sequencing and high-resolution mass spectrometry (HRMS) enables investigation of detailed gene/protein expression changes upon radiation and comprehensive evaluation of radiation effects on human cell models, which may improve our understanding of radiation injury and lead to new therapeutic approaches. Several studies have been done in the last decade to reveal the transcriptomic changes of ECs exposed to different types of IR (Benadjaoud et al., 2021; Bouten et al., 2021; Jaillet et al., 2017; Wagner-Ecker et al., 2010). Proteomic analysis was also done to discover molecular changes in ECs after gamma ray irradiation (Pluder et al., 2011). A limiting factor of these studies, however, is the radiation quality difference between the radiation used and space radiation. In 2015, human microvascular ECs (HMEC-1) were sent to the ISS for 160 days during the Soyuz TMA-18 M 44S manned mission. Transcriptomic analysis of space-flown ECs showed that space radiation and other space stressors activated pathways for hypoxia, inflammation, DNA repair and apoptosis. Inhibition of autophagic flux and an aged-like phenotype were also observed (Barravecchia et al., 2022). Common phenotypic changes were observed in ECs upon radiation from ground radiation facilities and space radiation. But a detailed comparison of EC responses to different sources of radiation (low/high-LET, ground analogues and real space radiation) is still needed.

Radiation-induced changes in ECs involve not only the transcriptional, translational, post-translational and cellular levels, but also the EC microenvironment through the secretion of reactive oxygen species (ROS), cytokines, chemokines and growth factors, that will further affect phenotypes of adjacent cell types (pericytes, smooth muscle cells, macrophages, fibroblasts) and tissues (Ritchie et al., 2015).

3.2. Radiation effects on human fibroblasts

Human fibroblasts are also commonly used as in vitro models to study biological effects of radiation, due to their relatively ready availability, proliferation capability, and known plasticity and sensitivity to external stressors such as radiation. Like ECs, fibroblasts are widely spread across the whole body and susceptible to IR-induced damage. Several studies have been published in recent years on the effects of low and/or high-LET radiation on human fibroblasts. In a series of studies on the effects of particle radiation on immortalized human fibroblasts, the transition from ATM to ATR signaling at DNA break sites was extended for longer periods of time by high-LET, compared to low-LET radiation (Saha et al., 2013). In a follow-up study, the investigators also compared effects of low-energy, high-LET ions and HZE ions on human fibroblast 82–6 cells, and they found that low-energy ions generated massive but localized DNA damage, leading to delayed DSB repair, and induced distinct cellular responses compared to HZE particles (Saha et al., 2014). A more recent study revealed that human fibroblasts may exert bystander effects via NO production post Fe-ion irradiation (Hada et al., 2019).

Bystander effects of irradiated fibroblasts have also been reported by several other studies. X-ray irradiated fibroblasts could secret miR-21 to mediate the bystander effects and activate DNA damage responses in non-radiated cells (Xu et al., 2014). Interestingly, human fibroblasts pretreated with low-LET protons were more resistant to high-LET radiation-induced chromosomal damage thereafter. Moreover, pretreated cells were able to protect adjacent cells from radiation-induced DNA damage through paracrine effects (Buonanno et al., 2015). In another study, human cancer associated fibroblasts (CAFs) were treated with x-rays and showed a senescence phenotype, which could promote cancer cell growth via the JAK/STAT signaling pathway. The same study found that FOXO4-DRI, a FOXO4-p53–interfering peptide, could kill CAFs and sensitize tumor cells to radiation treatment (Meng et al., 2021). Similar to the finding of this study, human skin fibroblasts exposed to x-rays could produce IL-6 that enhanced tumor cell migration (Suzuki et al., 2022). Proteomic studies might help to reveal the underlying mechanism of bystander effects of irradiated fibroblasts. Recently, proteomic analysis was performed on fibroblasts after exposure to 20 Gy of photon radiation and revealed elevated interferon signaling (Freyter et al., 2022). This further indicates that the inflammatory and secretory phenotypes after radiation exposure might underlie the observed bystander effects. The protein compositions of CAF-EVs (extracellular vesicles) after proton radiation exposure were analyzed using LC-MS/MS proteomics, and no significant difference was observed between control and radiated CAFs (Berzaghi et al., 2021). Further studies are needed to fully explain the observed bystander effects.

DNA damage is commonly observed in irradiated human fibroblasts. In 2014, AG1522 normal human foreskin fibroblast cells were sent to the ISS for 14 days, and a detailed characterization of radiation-induced DNA damage was accomplished by capturing 3D images of γ-H2AX foci by a laser scanning confocal microscopy (Lu et al., 2017). Different doses of IR with different energies could lead to unique DNA repair responses based on IR type and dosage (Sridharan et al., 2020). Even with comparable LET and energy, different types of radiation (boron and neon ions, and gamma-rays) could also lead to different types of DNA damage in fibroblasts (Jezkova et al., 2018). Compared to proton and helium ion radiation, human fetal lung fibroblasts irradiated by carbon ions showed slower DNA double strand break (DSB) repair (Oizumi et al., 2020). To better predict chromosome aberrations resulting from a range of irradiation conditions, Hada et al. established a computational model based on responses of human fibroblasts to different types of IR generated at the NASA Space Radiation Laboratory (Slaba et al., 2020). Microgravity was found to have synergistic effects on radiation-induced chromosome aberrations in human fibroblasts (Hada et al., 2018). Cell cycle-related genes were upregulated after simultaneous exposure to microgravity and radiation (Ikeda et al., 2019). The synergistic effect between microgravity and radiation needs to be further investigated. Interestingly, human fibroblasts pretreated with low-LET protons were more resistant to chromosomal damage induced by high-LET (Buonanno et al., 2015). Similar findings were also reported in another study later (Suzuki et al., 2020).

3.3. Radiation effects on other human cell types

Human neural stem cells (hNSCs) have also been used for radiation studies due to their proliferation capacity. X-rays and carbon ions were shown to induce growth inhibition and dose-dependent apoptosis of hNSCs. Carbon ions exerted more severe effects on hNSCs compared to x-ray (Isono et al., 2015). Another study comparing x-rays and carbon ions found that both could reduce hNSCs’ viability by 62% 7 days after of irradiation, with no apoptosis observed after 48 hours (Morini et al., 2018). In both studies, hNSCs showed a high radiosensitivity to both x-rays and carbon ions. Radiation-induced neurotoxicity has been reviewed elsewhere (Schielke et al., 2020).

Peripheral blood cells are sensitive to early radiation injury and can be easily biopsied from donors. Their gene expression changes induced by radiation persist several days (Turtoi et al., 2009). The most marked response of blood cells after gamma radiation exposure is the upregulation of genes associated with natural killer (NK) cell immune functions, as well as genes related to T- and B-cell mediated immunity (Paul et al., 2013). Keynton et al. performed a time- and dose-course study of gamma radiation on whole blood and identified appropriate sensitive and specific transcript biomarkers that were detectable in blood samples using microarray (Rouchka et al., 2019). In addition to the aforementioned cell types, immortalized human podocytes (Azzam et al., 2021) and epithelial cells (Daza et al., 2009) have also been used as cell models to investigate biological effects of radiation exposure on humans. All these in vitro radiation models using primary or immortalized human cell lines with various tissue origins offer valuable platforms to study shared and specific responses of different tissues upon exposure to various types of radiation.

4. Delineating radiation effects and developing countermeasures with CTiD

Although human primary and immortalized cell lines are suitable for various robust assays of radiation effects in vitro and ex vivo, their applications are strictly limited to the initial stage of the drug development process. The reasons for this include the limited cell types, mostly lack of genetic diversity, and physiological complexities of the whole human body. There is an unmet need for more complex and more relevant in vitro human systems for space radiation countermeasure discovery. hiPSCs and their application in CTiD provide a promising strategy for preclinical drug development, which may dramatically reduce time and cost.

hiPSC-derivatives offer an opportunity to study radiation responses of many tissue types that otherwise can hardly be obtained from human sources (cardiomyocytes, neurons etc.). Radiation is known to cause damage in all components of the heart, including the myocardium, pericardium, valves, coronaries, and conduction system (Chang et al., 2017; Shimizu et al., 2010). To investigate radiation effects on human cardiomyocytes, hiPSCs were differentiated into cardiomyocytes (hiPSC-CMs) and exposed to 5 and 10 Gy of x-rays. Irradiated hiPSC-CMs were characterized by RNA-sequencing after 48 hours. Radiation decreased the beating rate of hiPSC-CMs and higher doses of x-ray were more likely to change their electrophysiological spatial distribution (Becker et al., 2018b, 2018a). In another study, radiation effects of x-ray on hiPSC-CMs were assessed by multi-electrode array (MEA). High-dose x-rays irradiation (≥20 Gy) immediately and reversibly modified the electrical conduction, which activated cellular compensatory mechanisms and led to the immediate antiarrhythmic outcome of cardiac radioablation for refractory ventricular arrhythmias (Kim et al., 2021). To investigate the responses of hiPSC-CMs to actual space radiation, our team flew hiPSC-CMs onboard the ISS in collaboration with NASA during the SpaceX CRS-9 commercial resupply service mission (Wnorowski et al., 2019). With the combined effects of space radiation and microgravity, hiPSC-CMs onboard ISS for 5.5 weeks displayed altered calcium handling and dysregulated genes involved in mitochondrial metabolism compared to ground control CMs. This was the first study using hiPSCs to model space radiation effects on human CM structure and function.

To study radiation effects on the neural system, human embryonic stem cell (hESC)-derived NSCs were exposed to gamma rays, which subsequently showed impaired neuronal differentiation and altered gene expression even with a radiation dose below 100 mGy (Katsura et al., 2016). hiPSC-derived chondrocytes (hiPSC-DCHs) were used to investigate the gamma radiation effects on the human skeletal system, with a focus on cartilage. DNA DSBs were observed in hiPSC-DCHs in a radiation dose-dependent manner. The DNA damage was repaired efficiently in hiPSC-DCHs by activating homologous recombination and non-homologous end joining repair mechanisms. hiPSC-DCHs also displayed low ROS and cPARP levels and large degree of senescence upon radiation exposure (Stelcer et al., 2018). The same research group performed another study recently to investigate the effect of x-rays on hiPSC-DCHs and observed similar findings to gamma rays (Stelcer et al., 2021).

Space radiation-induced skin tissue damage is a significant risk for space crews (Radstake et al., 2022). Epidermal cells can be strongly affected by radiation exposures due to their distribution on the exposed body surface. To elucidate genome stability in epidermal cells following irradiation, skin keratinocytes were differentiated from hiPSCs and exposed to gamma rays. Compared to undifferentiated hiPSCs, keratinocytes showed decreased DNA damage responses, lower cellular sensitivity (expression of senescence genes) and less apoptosis upon IR (Miyake et al., 2019), indicating a reduced radiation sensitivity with differentiation.

Human pluripotent stem cells (hPSCs) can also be utilized to elucidate radiation effects on early embryonic and organ development. A study performed previously by our group showed that hESCs irradiated by x-ray suffered significant cell death and apoptosis, while remaining in a pluripotent cell state after irradiation (Wilson et al., 2010). Another study observed impaired differentiation capacity of hiPSCs toward CMs after exposure to α-particles irradiation (Baljinnyam et al., 2017). The different observations between two studies might be explained by the different radiation doses and types used.

A major advantage of using hiPSCs to study human-specific radiation effects is that various cell/tissue types can be obtained from genetically diverse patients or a cohort for a specific disease, with results that can be applied to test individual drug responses for radiation countermeasure development. Most studies using hiPSC models so far have focused on modeling human exposure to radiation instead of drug screening/discovery. More effort therefore needs to be devoted to translate current hiPSC radiation models to high-throughput drug screening, as well as individual susceptibility studies using genetically diverse hiPSC lines, in order to pinpoint genetic contributions to radiation-induced damage and develop new countermeasures.

5. Moving toward 3D tissue engineering

Translation of preclinical studies using CTiD into clinical applications is highly reliant on the successful development of hiPSC-derived models with cell-autonomous and radiation-related phenotypes. The establishment of physiologically relevant models requires the ability to faithfully replicate cell-cell interactions and 3D environments in human tissues/organs. Recent advances in co-culture systems, 3D tissue engineering, microphysiological systems (MPS), and organoids create new opportunities to further expand our knowledge of radiation biological effects and find new therapeutic strategies to counter these effects. Recent advancements of radiation studies using tissue engineering and their potential applications in space radiation countermeasure discovery are summarized in this section.

To investigate radiation effects on early human brain development, 3D neural spheres were generated from hiPSCs and exposed to 1 Gy of x-rays. The stemness of neural progenitor cells was impaired and genes involved in neurodevelopment and growth abnormalities were dysregulated, suggesting a negative impact of radiation on the early developing human brain (Klatt et al., 2019). In another recent study, cardiac spheres were generated from hESC-derived CMs using a similar technology to mimic heart tissue. These 3D cardiac spheres allowed CMs to reach maximum maturity after 100 days as evidenced by increased α-actinin lengths and typical multinucleation and branching. Exposing them to x-ray radiation (0.1–2 Gy) led to a significant increase of beating rates and a more arrhythmic sequence of cellular depolarization and repolarization in cardiac spheres. Proteomic analysis of cardiac spheres after exposure to 2 Gy of x-ray identified a dysregulated adrenergic signaling pathway cascade which might explain the dysregulated beating rate (Smit et al., 2021). This study showed that radiation-induced cardiac risks can be estimated adequately in a physiological engineered tissue model derived from hPSCs. Recently, Humanetics Corporation (Humanetics^®) developed a candidate drug, BIO 300 to mitigate cardiotoxicities of space-relevant radiation exposure. hiPSC-derived engineered heart tissues (EHTs) were employed as an in vitro heart model to test the efficacy of this drug.

Although there is a limited number of radiation studies published using engineered human tissues derived from hiPSCs, more studies have been published using 3D tissue models made from human primary or immortalized cells. HUVECs and normal human lung fibroblasts (NHFLs) were used to generate a 3D vessel-on-a-chip model to investigate the effect of x-ray radiation on human vasculature. This model could recapitulate the self-organizing EC sprouts formation and angiogenesis within a microfluidic device. ECs cultured in a 3D microfluidic chip were more resistant to radiation-induced damage, including adherent junction breakage, apoptosis and DNA damage (Guo et al., 2019). In another vessel-on-chip system, a vessel-like lumen structure was established in the channel of a microfluidic chip using HDMECs. Exposure to gamma radiation caused EC barrier disruption and reduced angiogenic potential in this chip model. This system also allowed characterization of cell behavior of tip and stalk cells after irradiation, which is difficult to monitor in 2D EC culture (Na et al., 2021). A 3D microvasculature tissue model made from HUVEC was also used to study IR effects, revealing a synergistic effect of low- and high-LET radiation on the early and late stage of angiogenesis (Wuu et al., 2020). Notably, a subset of these miRNAs was found to regulate radiation-induced vascular damage and their inhibition by antagomirs could successfully prevent this vascular damage (Malkani et al., 2020). In summary, these 3D vessel-on-chip and vascular tissue models made from primary ECs can be valuable tools to investigate radiation‑induced vascular injuries in human.

Besides blood vessel models, gastrointestinal models have been developed from human primary cells and used to explore radiation effects on the human gut. In a previous study, a human gut-on-a-chip system was established from human intestinal epithelial cells and ECs in a microfluidic chip setting. Gamma ray irradiation increased ROS generation, cytotoxicity, apoptosis, DNA fragmentation, and villus blunting within this gut-on-a-chip model. Intestinal barrier integrity was also compromised by radiation exposure. Importantly, the radiation-induced damage could be inhibited by a prophylactic radiation countermeasure, dimethyloxaloylglycine (DMOG) (Jalili-Firoozinezhad et al., 2018). In two more recent studies, 3D human intestinal organoids were generated using cells isolated from human ileum crypt, which were exposed to x-ray or gamma ray radiation. It was found that, SIRT1 inhibition with nicotinamide (NAM) could significantly improve intestinal organoid survival post-irradiation (Fu et al., 2021). A protective effect of valproic acid (VPA) against radiation-induced damage via NOTCH signaling was also reported in intestinal organoids (Park et al., 2021). These recent studies suggest that human 3D gut models can serve as a platform to reveal radiation-induced gastrointestinal syndrome, and test/screen candidate radio-protective countermeasures (Figure 2).

Figure 2.

Radiation countermeasures that have been tested with human in vitro cell models and shown protective effects. Created with BioRender.com.

6. Conclusions

With rapid growth of investments and plans for space exploration, the number of astronauts and space tourists will rise substantially in the near future. Despite the known detrimental effects of space radiation, effective protective countermeasures are still absent and needed. Unlike clinical trials on the ground, it is extremely difficult to recruit sufficient space travelers for clinical trials of space radiation countermeasures. As an alternative, CTiD derived from hiPSCs offers a unique opportunity to discover space radiation countermeasures based on human systems. hiPSCs can be obtained from different individuals and represent diverse genetic backgrounds of patients, making them useful for predicting individual drug responses within a large population.

Monolayer culture of human cells in simple 2D settings are ideal for the establishing robust and cost-effective assays of radiation responses, allowing them to be easily translated into high-throughput screening at the initial stage of drug discovery. A major drawback, however, is that simple 2D models differ in many aspects from in vivo physiological settings, making them less accurate in predicting candidate drug efficacy and toxicity in humans. By contrast, more sophisticated 3D models generated with advanced tissue engineering techniques can better resemble the physiological environment in vivo, although their fabrication and culture are more difficult and resulting in lower throughput (Figure 1). There is optimism that continuingly progress in improved tissue engineering techniques, more sensitive and higher throughput methods of analysis, and better hiPSC differentiation protocols will allow greater adaptation of 3D models in high-throughput drug discovery over time. Both 2D and 3D models can be applied in CTiD depending on the complexity and composition of the targeted tissue needed for drug testing. In summary, CTiD is ideally suited for the purposes of space radiation drug development because of its unique features, and its uses in this field will continuingly progress with more and more efforts being devoted to refine current 2D and 3D cell/tissue models (Cho et al., 2021; Kim et al., 2022; Thomas et al., 2021).

The application of CTiD in radiation drug discovery is still in its infancy. There is no published study yet using CTiD for high-throughput drug screening against space mission relevant radiation, to the best of our knowledge, except a most relevant report of using hiPSC-derived EHTs in the discovery of a potential radioprotective drug called BIO 300 developed by a pharmaceutical company. By contrast, a large number of studies have been done to understand human-specific radiation injuries and test a specific drug candidate using various human primary cell types or hiPSC-derivatives. These studies conducted in the past decade are summarized in this review, providing evidence of the great value of human in vitro cell models for the space radiation drug discovery.

Microgravity is another major deleterious factor in space travel that occurs simultaneously with radiation exposure. The synergetic effects of microgravity and radiation must be considered in space radiation countermeasure development. More efforts are needed to simulate the simultaneous exposure to both factors using random positioning or clinostat within radiation environment. By adapting the culture containers, current 2D/3D cells models could be also maintained and used for CTiD within the simulated microgravity and radiation environment. An ultimate goal would be the development of potent countermeasures for deleterious effects induced by the exposure to both factors simultaneously.