Radiation-Induced Genomic Alterations

Experiments in a controlled setting exploring the impact of radiation exposure on laboratory animals or cell lines can help understand the extent of radiation damage possible in humans. A genome-wide study on the impact of radiation on germline changes in mice noted increased de novo CNVs and indels in offspring due to exposure to 3 Gy (9). This study also identified a significantly higher number of clustered mutations in the progeny of irradiated male mice. A signature of radiation exposure is the association of these mutations with clustered damage sites. These clustered DNA lesions occur within one or two helical turns of DNA, i.e., within a short distance. DNA lesions are mostly seen as an aftermath of radiation exposure. Complex double-stranded and single-stranded DNA breaks and clustered mutations in specific genomic areas are the most damaging as the repair process of these types of changes are challenging. Single-strand DNA breaks in the clustered damage sites can lead to inefficient base excision repair pathways resulting in persistent lesions during subsequent replication cycles (10, 11). Complex clusters of radiation-induced DNA lesions initiate further mutagenesis and genomic instability, which could eventually lead to cell death (12). This property of radiation is used in medicine to target and kill tumor cells. The increased number of DNA breaks induced by γ radiation (<2.0 Gy) and the increased replication rate in tumor cells act in synergy to eliminate cancerous cells (13). Cells with a high-proliferation rate are prone to radiation-induced DNA damage clusters (14), suggesting that different tissues might respond to different dose, dose rates and radiation quality.

In vitro studies have also demonstrated the ability of radiation to induce CNVs (5, 15). Experiments conducted on human fibroblast cells detected genome-wide distributions of CNVs associated with exposure (<=3 Gy), with several hotspots corresponding to aphidicolin (APH) and hydroxyurea (HU) induced replication stress (5). This study also reported a higher percentage of duplications than other stressors and that CNVs were detected more than seven days postirradiation. The aftermath of radiation exposure on the germline of male mice identified an eightfold increase in CNVs in offspring, indicating a transgenerational impact (9). Several other studies using different radiation sources, such as laser-driven electron accelerators or X rays, identified increasing CNVs in chromosome hotspots (15, 16). It was also observed that radiation-induced tumors in Trp53 mutant mice developed CNV changes in chromosome hotspots associated with specific tumor types (7). For example, radiation-induced mammary tumors exhibited amplification of the MET locus on chromosome 6. Consistent with this observation, radiation-induced gliomas in Trp53/PTEN mutant mice also developed recurrent amplification of the locus containing the RTK and MET gene in chromosome 6 (17), changes that were associated with cancer stem cell maintenance in both of these cancers (18, 19).

High doses of γ radiation on hematopoietic cells induced gene fusions associated with leukemogenesis, including AML1-ETO, BCR-ABL, DEK-CAN, and DEK-ABL, indicated that DNA breakage and mis-repair is a significant risk factor for leukemia (20). Low and therapeutic dose irradiation on hematopoietic stem and progenitor cells (HSPC) isolated from human umbilical cord blood also led to increased formation of preleukemic fusion genes associated with radiation-induced DNA damage (21). A significant increase in BCR-ABL fusion was reported in this study compared to controls, suggesting that the risk of developing leukemia exists even at low doses.

Exposure to high-intensity radiation is also known to cause excessive skin damage. A study investigating the impact of γ radiation (53 Gy, dose rate 0.6 Gy/minute) on minipig skins found elevated apoptosis and weakened stem cell replication after irradiation (22). A higher percentage of γH2AX phosphorylation and increased frequency of radiation-induced 53PB1 foci indicated radiation-induced DNA damage. Extensive radiation-induced fibrosis (RIFs) persisted for weeks in a minor fraction of cells (<1%) and reflect the complex DNA damage, including single-strand and double-strand DNA breaks, and base lesions that are refractory to DNA repair mechanisms. A similar observation of persistent RIFs after low-intensity in vivo irradiation of human skin was observed 24 h after 4 Gy exposure with significant inter-individual variation in RIF levels that could be associated with differences in DNA repair mechanisms (23).

Recent advancements that aid in studying three-dimensional (3D) DNA organization have helped the research community to understand the effect of radiation stress on genome-wide 3D organizational structure. CCCTC-binding factor (CTCF) and cohesion, involved in chromatin organization, are early responders to DNA damage induced by radiation (24, 25). In this context, incipient role of G protein-coupled receptors (GPCRs) signaling system in DNA damage response has opened a new direction to controlling the damage due to radiation and subsequently developing target-specific therapeutics (26). A previous study showed that protein G2A (member of GPCR family) responds to DNA damage when induced in lymphocytes (27). Similarly, the combined role of GPCR relaxin family peptide 3 receptor (RXFP3) and GPCR interacting protein 2 (GIT2) in response to oxidative stress and DNA damage have been identified by van Gastel et al. and Chadwick et al. (28, 29). These studies also concluded that RXFP3-GIT2 system plays a significant role in regulating cellular degradation associated with DNA damage. Nevertheless, further studies are needed to identify the involvement of GPCRs in radiation-induced DNA damage. There are currently no reports that explore the consequences of γ radiation on chromatin organization, but one study revealed the ability of 5 Gy of X-ray radiation to induce changes in the genome-wide 3D organization in human fibroblasts and lymphoblastoid cells (30). A strengthening of topologically associating domain (TAD) formation was observed after X-ray exposure, suggesting noticeable 3D genome changes after exposure. TADs, formed through the interaction of CTCF, play a crucial role in bringing genes and regulatory elements in close contact with each other, leading to preserving genome integrity after irradiation. This impact lasted for days and was more pronounced in lymphoblastoid cells than in fibroblasts (30), pointing to differences in radiosensitivity across cell types. This study also linked the structural change to the ataxia-telangiectasia mutated (ATM) DNA repair pathway. These reports noted that radiation-induced DNA damage was extensive and included point mutations, DNA breaks, and changes in the genome-wide 3D organization, highlighting the importance of functional DNA repair mechanisms to mitigate against the lasting effects of such radiation-induced genomic insults.

Radiation-Induced Transcriptomic Changes

Messenger RNA (mRNA)-expression. Genome-wide expression analysis of the peripheral blood (PB) in different mammalian systems has shown radiation-induced alterations at different exposure intensities. Experiments in partial-body irradiated at 0.5 Gy, 2 Gy and 10 Gy mice identified PB expression signatures that differentiated radiation damage with 79–100% accuracy (31). Additional studies on global gene expression profiles in the blood of male C57BL/6 mice identified a 74-gene signature associated with moderate to low doses (at 0.5, 2, 5 and 8 Gy). More than one-third of these genes were regulated by TP53, suggesting that it plays a vital role in radiation responses (32). The most marked response was in genes associated with natural killer (NK) cell functions, reflecting a relative loss of NK cells from the population. T- and B-cell mediated immunity genes were also significantly influenced by 48 h after γ-ray irradiation (0, 0.5, 2, 5, 8 Gy). Downregulation of NK-cytotoxicity-associated genes was observed, including NKG7, GNLY, and GZMA (33). Several of these genes were further validated in independent studies, including human PB, indicating that these genes could be further explored as reliable radiation biomarkers (34, 35) (Table 1). Other reports that analyzed the global expression profile of irradiated mice have identified trends in gene expression. An increasing number of genes were differentially expressed within days after irradiation with 2, 3, and 5 Gy, and gene expression increased with increasing radiation doses (21). Specifically, the expression of SLC25A51 and CCNA2 showed an initial suppression followed by recovery of the transcript levels by seven days after exposure to mice with doses less than 10 Gy (21). Another study focused on differential expression of genes in mice exposed to 4 and 8 Gy found many upregulated genes including CDKN1A, MDM2, BBC3, and CCNG1, and downregulated genes including TCF4 and MYC. The same study also identified genes, DDB2, PCNA, GADD45A, SESN1, RRM2B, KCNN4, IFI30, and PTPRO, which are downregulated in mice but upregulated in human (36). The difference in gene expression were also observed in wild-type C57BL/6 (8 Gy) and DNA double-strand break repair-deficient Atm^–/– (4 Gy) and Prkdcscid (3 Gy) mutants of C57BL/6, concluding that regulators including TP53 and NFjB are activated by radiation only in wild type mice (37). The study further identified that mutant strains show inflammatory responses after radiation exposure. The source of radiation can be a key factor for the transcriptomic changes, as discussed by Broustas et al. (38) based on the experiments where 7,285 and 5,045 genes were differentially expressed in the blood of irradiated mice that received 0.25 or 1 Gy of neutron or 1 or 4 Gy X-ray irradiation, respectively (38). Similarly, effects of age on transcriptomics profile have been observed after irradiating (4 Gy X rays) young (2 months) and old (21 months) male mice for 24 h. The microarray-based gene expression analysis suggested that young mice were more active against radiation by upregulating pathways related to apoptosis and phagocytosis. The pathways associated with fibroblast growth factor signaling were underrepresented in old mice, while hematologic malignancies related pathways were enriched (39).

TABLE 1

List of Potential mRNA Biomarkers in Radiation-Induced Injury in Different Animal Models

TABLE 1

Continued

TABLE 1

Continued

In ex vivo irradiated (0.56 Gy, 2.23 Gy and 4.45 Gy, acute dose rate of 1.03 Gy/min, low-dose-rate = 3.1 mGy/ min) human whole blood, 454 genes were differentially expressed 24 h postirradiation to all doses, while 598 genes were differentially expressed after acute exposure. These genes were mainly enriched in functions related to immune response, B-cell mediated immunity, cell-to-cell signaling, and natural killer cell activation (40). Similarly, in the whole-thorax irradiated (single dose of 10 Gy) NHP, 1,187 mRNA transcripts were significantly dysregulated at day 30 postirradiation. The differentially regulated genes mostly belonged to functions related to immune responses (41). The transcriptomics effect of radiation doses (2.5 Gy and 0.1 Gy of low-LET protons) was also observed in a human 3-D tissue model EPI-200 at 4, 16 and 24 h postirradiation. The study found that high dose reduced terminal differentiation and structural integrity, while low dose was mainly associated with recovery and tissue repair (42). A systematic review included 27 previous radiation-based studies and identified 27 potential genes that have significant correlation with radiation dose. Top five discriminatory genes, TNFSF4, FDXR, MYC, ZMAT3, and GADD45A were identified when compared between doses <2 Gy and ≥2 Gy (43). The roles of several other long ncRNAs in radiation responses have been discussed in previous studies (44–46); however not covered in the present review.

Experiments in NHPs, such as baboons, can model radiation-induced changes comparable to humans. Baboons that developed the clinically relevant disease, hematologic acute radiation syndrome (HARS, classified based on severity ranging from H1-3), had a persistent change in peripheral gene expression within 1 or 2 days postirradiation to 2.5 or 5 Gy of γ radiation (57). These three genes (WNT3, POU2AF1, and ZZZ3) showed a persistent change in gene expression over time. Several genes associated with immune-related functions that impact T-cell migration (CCR7, CD117) and T-cell responses (VSIG4), antimicrobial function (RNASE3), promotion of Interleukin-12 (IL-12) response signal transducer and activator of transcription 4 protein (STAT4), stimulation of B- and T-cells (SH2D1A), and initiation of apoptosis in pathogen-infected cells (GZMH) and cytolysis (PRF1, NCR3, and KLRF1) were associated with the radiation-induced response (58). Another study in the same mammalian system identified radiation-induced overexpression of genes involved in cell cycle regulation (CDCA7L), modification of the T-cell and B-cell immune responses (GBP2, GLUL, HERC5, and PPP3CC), erythropoiesis (GBP2), and cell migration of cancer and hematopoietic cells (HERC5, HMHA1) (59) when irradiated with 2.5 or 5 Gy. A recent study showed that IL-3 signaling, ephrin receptor signaling, ErbB signaling, nitric oxide signaling in the cardiovascular system, Wnt/β-catenin signaling, and inflammasome pathways were associated with positive survival outcomes in NHPs after acute exposure to 6.5 Gy of γ radiation (95).

Experiments to study the effect of radiation (0.1 Gy of γrays) on gene expression in a human myeloid tumor cell line (ML-1) identified genes that responded in a dose-dependent manner (2, 72). Among the set of genes identified in these studies (CDKN1A, GADD45, MDM2, ATF3, and BAX), CDKN1A, GADD45A, and MDM2 were also overexpressed after low-dose-gamma irradiation. Of the transcriptionally active genes identified after irradiation, induction of CDKN1A and GADD45A was proportional to exposure intensity. Another study that explored the transcriptional changes in human peripheral blood lymphocytes (PBL) also identified CDNK1A and GADD45A as radiation-induced genes after 2 Gy X-ray irradiation (73). This study also identified a linear dose-dependent response in the expression of DDB2, CDKN1A, and XPC. Human PB signatures are of interest because radiation-exposure biomarkers can distinguish between the irradiated and non-irradiated samples with 100% accuracy (54).

Exposure to high-dose gamma radiation (5, 10 and 20 Gy) in human lymphoblastoid TK6 cells elevated mRNA levels of GPX, GADD45, P21, and PCNA, moderately repressed XRCC1 expression, and strongly down-regulated the expression of KU80 (77). A similar observation was noted when non-immortalized T cells were exposed to low- to moderate-doses of gamma radiation (between 0.15–12 Gy), with transcriptional induction of several genes, including CDKN1A, GADD45A, TNFSF4, KIF20A, PSRC1, and CDCA3 (85). Another in silico network analysis identified eight ARS-associated genes (BRD4, NFKBIA, CDKN1A, TFPI, MMP9, CBR1, ZAP70, IDH3B) and confirmed through literature mining (69, 70). In these experiments, commonly studied genes such as CDKN1A and GADD45 in mice, NHPs, and human cell lines showed promise to be effectively utilized as radiation biomarkers.

MicroRNA (miRNA) expression. Serum microRNA (miRNA) signatures indicate the long-term impact of total-body irradiation [(TBI) 6.5 and 8 Gy] in mice when measured within 24 h postirradiation (96). These miRNA signatures distinguished irradiated from unirradiated animals, which correlated with the impact of radiation on hematopoietic stem cells (HSCs). Several studies in mice have explored the miRNA signatures associated with radiation exposure. An analysis identified long-term elevation of miR-21 in the brain when hippocampal cells and brain tissue from 0.5 Gy X-ray irradiated mice at different times for 1 year (97), and upregulation of miR-145 and miR-663 in AHH-1 cells and HPBLs, when 4 Gy gamma-ray irradiated for 4 h or 24 h (98). Radiation-induced miRNA profiles were also found to be time- and dose-dependent. Serum miRNAs (miR-27a-3p, miR-187-3p, miR-30a-3p, and miR-30c-5p) were detectable within 24 h postirradiation in mice. Of these miRNAs, the expression of miR-30a-3p and miR-30c-5p were able to differentiate between radiation doses (6.5 Gy vs. 8 Gy) after a week postirradiation (96). The microRNAs were also correlated with injuries to human bone marrow cells based on serum profiling of mice exposed to sublethal (6.5 Gy) and lethal (8 Gy) doses of radiation (95). In a different study, 600 miRNAs from serum of irradiated (1–12 Gy) mice were compared in a dose dependent manner at time points of 24 and 48 h (99). Similarly, serum miRNA signatures containing miR-130a-3p, miR-150-5p, miR-142-5p, miR-706, and miR-342-3p can be used to identify low-dose (2 Gy) radiation cohorts irradiated within 24 h (96). The dose reconstruction algorithms that were developed based on serum miR-150-5p depletion normalized with miR-23a-3p in mouse models can approximate the absorbed dose at various time points during the recovery phase (100). Leukemia specimens from patients treated with fractionated doses showed depletion of miR-150-5p in blood (100). Changes in plasma miRNAs such as miR-34a-5p, miR-100-5p, and miR-150-5p were associated with pro-inflammatory NF-jB-mediated functions as studied in heart and lung damaged C3H mouse after irradiation. In the same study, expression of miRNAs, miR-34b-3p, miR-96-5p, and miR-802-5p, were significantly altered in C57Bl/6 mice after 13.99 Gy exposure to the heart and lung (101). In a separate study using rats, miRNAs, miR-144-5p, miR-144-3p, miR-142-5p and miR-19a-3p, were differentially regulated in the blood after lung-specific injuries induced by a single 15 Gy dose (dose rate of 1.43 Gy/min) (103). miRNA miR-150 found in blood and lung also showed significant downregulation after thoracic irradiation in female WAG/RijCmcr rats (102). Similarly, thoracic irradiation of NHPs causing injuries in lung and heart have shown differences in the expression profile of blood miRNAs, miR-199a-3p and miR-25-3p, after a whole-thorax and lung irradiation (WTLI) at 9.8 or 10.7 Gy (103). In C3H mice after whole-thorax 13.92 Gy irradiation, differentially expressed miRNAs miR-34a-5p, miR-100-5p, and miR-150-5p were identified associated with survival (101). These studies have improved our ability to consider circulatory miRNAs as organ-specific markers upon radiation damage.

Similarly, microarray analysis identified seven miRNA signatures altered by irradiation at 5.8, 6.5 or 7.2 Gy (dose rate of 0.6 Gy/min) in NHPs (104). Conserved serum miRNA signatures have the potential to serve as predictive biomarkers for radiation injury in humans, NHPs, and mice. A combination of three miRNAs (miR-133b, miR-215, and miR-375) was found to identify radiated versus unexposed NHPs, accurately. Two microRNAs, miR-199a-3p and miR-25-3p, from blood of whole-thorax lung irradiated NHPs at 10.7 Gy were differentially expressed and associated with survival (103). Similarity in another study, microRNAs associated with neutropenia were identified in NHPs after TBI (2–6.5 Gy) and whole-thorax lung irradiation (9.8 or 10.7 Gy) (105). Radiation-induced mortality could also be predicted by a five-miRNA (miR-133b, miR-215, miR-375, miR-126, and miR-30a) composite signature in macaques (104). The same study identified a set of two miRNAs (miR-30a and miR-126), which together can be used as a survival indicator (106).

Global miRNA expression changes measured in the PB of baboons within the days after irradiation at 2.5 or 5 Gy (dose rate of 0.08 Gy/min for 5 Gy TBI and 5 Gy 50% partial-body irradiation, and 0.32 Gy/min for other) identified miR-425-5p expression that distinguished different HARS groups (107). Six miRNA species (miR-133, miR-124, miR-29c, miR-378, miR-574-3p, and rno-miR-7) were identified as promising candidates that can distinguish HARS groups with and without pancytopenia (59). Many miRNA species were already known to be linked with radiosensitivity (e.g., miR-22, miR-29c, miR-195, miR-212) or chemotherapy resistance (e.g., miR-331-5p). In particular, miR-212 involved in radiosensitivity and immune modulation was upregulated 48 to 77-fold over an extended period (60). Another study that compared radiation-induced global gene expression profiles in baboons when total-body irradiated at 2.5 or 5 Gy (dose rate of 0.08 Gy/min for 5 Gy TBI and 5 Gy 50% partial-body irradiation, and 0.32 Gy/min for other), identified a significant downregulation of miR-342-3p, which allowed an almost complete separation of HARS categories (108). These miRNAs can be further explored as reliable radiation-biomarkers for accurate screening and evaluation of radiation exposure in humans.

A systematic review and meta-analysis study showed that seven miRNAs that include miR-150, miR-29a, miR-29b, miR-30c, miR-200b, miR-320a, and miR-30a showed significant correlation with dose of radiation across different species that mainly include human, mouse, rat and NHPs (109, 110). These studies support the hypothesis that miRNAs from body fluids can be used as biomarkers for detecting organ-specific injuries induced by high dose radiation exposures. Studies in humanized mice engrafted with human CD34⁺ HSCs showed that the expression profile of a serum miRNA signature containing miR-27a-3p, miR-187-3p, miR-30a-3p, and miR-30c-5p was altered in response to 6.5 and 8 Gy irradiation. This indicates that radiation-induced miRNA signatures may be conserved between mice and humans and can serve as viable biomarkers of radiation injury in humans, however, further studies are needed to validate their utility as biomarkers (96).

An extensive list of important mRNAs, miRNAs, and their tested effects in different animal models is shown in Tables 1 and 2, respectively. The summary of miRNA and gene-based biomarkers studied in different animal models is illustrated in Fig. 2A. Note that the overlap among the gene or miRNA biomarkers across the species is little to none warranting the need to further investigate and validate their utility as effective radiation biomarkers.

FIG. 2

Panel A: Summary of genes and miRNAs biomarkers studied in different animal models. The genes or miRNAs studied in different animal models such as baboons, human, mouse, and NHP were compared. Panel B: Statistics of radiation countermeasures, their classification, and experimental studies in different animal models. The drugs studied in animal models e.g., NHP, human, mouse/rat, and minipig were compared to identify unique and common ones in each model.

TABLE 2

List of Potential miRNA Biomarkers of Radiation-Induced Injuries in Different Animal Models

Continued

Radiation effects on mammalian microbiota. Microorganisms are essential to host cell maintenance in health and disease; therefore, the human gut microbiota has attracted increasing attention in recent years. The gastrointestinal (GI) tract hosts the largest and most diverse microbiome in the human body, where the microbiome plays a vital role in several metabolic processes that are essential for human health. ARS occurs after TBI, and death is frequently attributed to poor hematopoietic recovery and/or death of epithelial cells lining the GI tract. The GI tract is the fastest-renewing adult tissue, making it highly sensitive to radiation (124). It also harbors a diverse microbial community comprising 10 to 100 trillion microorganisms, raising the possibility that the gut microbiota is highly prone to radiation-induced damage (125–129). Although some studies have shown associations between gut microbiota and radiation-induced damage (129–131), possible mechanisms are poorly understood.

Most cancer patients that undergo pelvic irradiation experience side effects such as fatigue and diarrhea; however, the impact of radiation on the microorganisms colonizing the GI tract's mucosal surfaces is unexplored. Severe diarrhea may also complicate radiotherapy as there is no current clinical or experimental information on the role of gut microbiota in this pathogenesis. A study focusing on the fecal microbiota of patients receiving pelvic radiotherapy and suffering from acute post-radiotherapy diarrhea revealed changes in microbial diversity. In healthy volunteers and patients without diarrhea, microbial diversity was stable throughout the study. However, patients exhibiting diarrhea showed a progressive modification in their microbial diversity. Patients who developed diarrhea harbored bacterial phyla, Actinobacteria, and some Firmicutes, which were either not detected or least abundant in patients who did not develop diarrhea. On the other hand, patients who developed diarrhea had a reduced abundance of some Clostridia species, which are members of the phylum Firmicutes (132).

Radiation-induced changes in the bacterial compositions of the large and small intestines at the genus level are one of the hallmarks of radiation injury. For instance, a single dose of 8 Gy increased several genera, including Alistipes, Lactobacillus, and Akkermansia in the large intestine and Corynebacterium and Turicibacter in the small intestine. Compared with the corresponding unexposed control group, the abundance of the genera Prevotella was lower in the irradiated large intestine, while levels of Alistipes were lower in the irradiated small intestine (131).

The gut microbiota may contribute to radiation-induced pathogenesis, and it differs between patients with and without radiation enteropathy. Reports also indicated that microbial diversity decreased over time in patients with rising radiation enteropathy. A consistent association between bacterial diversity and late radiation enteropathy was observed, and higher counts of Clostridium IV, Roseburia, and Phascolarctobacterium were significantly associated with radiation enteropathy (132).

A study that analyzed “elite survivors” in a population of mice that recovered from high-dose total-body irradiation (8.0 to 9.2 Gy) had overrepresented members of the bacterial taxa, Lachnospiraceae and Enterococcaceae. These microorganisms were associated with postirradiation restoration of hematopoiesis and GI repair. These bacteria were also found to be more abundant in leukemia patients undergoing radiotherapy, who also displayed milder GI dysfunction (133). A different study in the radiation enteritis (RE) mice model, considering a single dose of 18 Gy X rays at a rate of 5 Gy/min, identified an abundance of bacteria representing 12 genera from Alloprevotella, Alistipes, Akkermansia, Bacteroides, Dubosiella, Eggerthellaceae, Enterococcus, Escherichia-Shigella, Lactobacillus, Lachnospiraceae, Muribaculaceae, and Rikenellaceae after irradiation (134). These studies highlight the importance of GI microbiota in radiation-induced pathogenesis and potentially pave the way for novel treatment protocols involving alterations of human gut microbiota.

In a recent study, the bacterial 16S rRNA amplicon sequencing and untargeted metabolomics were accomplished to investigate the effects of BIO 300, a promising radiation countermeasure under advanced development, on the gut microbiome and metabolome of CD2F1 male mice exposed to ⁶⁰Co-gamma radiation (9.2 Gy, 0.6 Gy/min, LD_70/30 dose). Radiation changed the ratio of Firmicutes/Bacteroidetes and also decreased the relative abundance of Lactobacillus in BIO 300-treated as well as control mice after irradiation (135). The ratio returned to pre-irradiated levels in BIO 300-treated animals by day 14 postirradiation. Concurrently, there was corrective shifts in metabolic pathways that were altered after radiation exposure. In brief, these results demonstrated that exposure resulted in a relative depletion of commensals like Lactobacillus leading to an inflammatory metabolic phenotype in untreated control mice while the BIO 300-treated mice demonstrated alleviation of this condition by restoring normal gut microbiota.

Alterations to the gut microbiome of NHPs exposed to high doses of radiation detected several bacterial species. After irradiation, significant increases in the relative abundance of Treponema and Helicobacter genera were observed in rhesus macaques when exposed to 6.8 Gy of radiation (136). In another study, relative abundances of Prevotella, Lactobacillus, Clostridium XIVa, Oscillibacter, and Treponema were found to be highly correlated with radiation intensity. Prevotella, Oscillibacter, and Treponema were closely associated with the overall survival, while Streptococcus was associated with death in macaques (137). The Firmicutes/Bacteriodetes ratio, a factor related to the disruption of metabolic homeostasis, declined from 1.2 to below one after 7.4 Gy irradiaiton. Actinobacillus, Bacteroides, Prevotella, and Veillonella genera were significantly increased by more than twofold, and Acinetobacter and Aerococcus genera were diminished by greater than tenfold after irradiation (138).

Other Omic Profiles Altered by Radiation

Altered lipid profiles. Concerns over a potential exposure of radiation in large populations have emphasized the need for rapid and reliable biodosimetry methods to determine the absorbed dose and required triage. Indirect consequences of radiation exposure include the generation of reactive oxygen species (ROS) through water hydrolysis, enhanced NADPH oxidase activity, compromised mitochondrial function leading to damaged cellular lipids. Since lipidomics is a powerful technique for large-scale identification and quantification of lipids, rapid identification of altered lipid molecules are promising biomarkers for biodosimetry.

Experiments conducted in mice and NHP models to characterize lipidomic profiles that respond to radiation exposure have helped identify important radiation biomarkers. A study that analyzed global lipid profiling from mouse serum identified significant molecular alterations after γ irradiation. A low abundance of oxygenated, polyunsaturated fatty acids (PUFAs) was observed after 8 Gy gamma-ray irradiation (dose rate of 1.67 Gy/min). Exposure to γradiation induced a significant increase in the serum levels of phosphatidylcholines (PCs) and arachidonic acid, while levels of diacyl PCs carrying PUFAs were decreased (139).

Radiation exposure in NHPs induced significant perturbations in lipid metabolism, affecting all major lipid species, including free fatty acids, glycerolipids, glycerophospholipids, and esterified sterols. A significant increase in the levels of PUFA-containing lipids and polyunsaturated triglycerides in the serum of NHPs exposed to 10 Gy was identified. These triglycerides contained primarily arachidonic acid and docosahexaenoic acid acyl moieties (140). Temporal changes in the serum lipidome from hours to days postirradiation in NHPs have also been reported. Marked lipidomic changes occurred within 24 h postirradiation with 6.5 Gy, along with increased levels of cytokine, free fatty acids, monoacylglycerides, and C-reactive protein. Simultaneous decreases in di- and tri-acylglycerides, sphingomyelins, lysophosphatidylcholines, and esterified sterols were also observed in this study (141). Decreased sphingomyelins and increased levels of lysophosphatidylcholines may be important markers for biodosimetry within days following irradiation. The biphasic and dynamic alterations linked to radiation exposure in the serum lipidome emphasized the importance of determining the long-term temporal response of these promising radiation biomarkers (141).

Global lipidomic analysis using ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS) performed on intestinal tissue specimens from acute RE after 18 Gy X irradiation at a dose rate of 5 Gy/min revealed distinct lipid metabolite fingerprints. Several lipids were significantly altered by the occurrence of peroxidation in the acute or chronic radiation enteritis group compared with the controls, which included glucosylceramide, phosphatidylethanolamine, lysophosphatidylcholine, lysophosphoglycerol, lysophosphatidylinositol, phosphatidylcholinephosphatidylglycerol, phosphatidylinositol, and sphingosine (134). Metabolism of sphingolipids is a very complex process that involves the catabolism of several lipids by enzymes such as glucosylceramide synthase and sphingosine kinase 1 (142). Though current technologies can perform lipid profiling for various exposure conditions, data interpretation becomes very difficult because of their complex biological roles; hence, lipid profiling is not widely used for biomarker identification in radiation studies.

Altered metabolic profiles. The rapid identification of radiation-induced metabolomic markers in biological samples, such as urine, blood, tissue, and saliva, have also been investigated for developing countermeasures to a radiological or nuclear public health emergency (143). Radiation metabolomics has primarily focused on the mass spectrometry (MS) analysis of samples from irradiated animals to monitor an altered biological response. In this manner, metabolic studies are valuable for understanding the systems-level biological impact of radiation exposure. Moreover, radiosensitive biodosimetry methods were developed to accurately measure exposure levels and understand the impact of radiation on tissues and organs. Metabolomic studies helped to understand changes in metabolite levels in normal and irradiated organs in mice, NHPs, and humans. For example, Gao et al. measured 31 low molecular weight metabolites such as lactamide, 1,2,4-benzenetriol, taurine, and piperine to assess the metabolic changes in rat lungs when exposed to either a single 10 or 20 Gy dose of radiation (144). Similarly, age-dependent correlation was observed for N(1)-acetylspermidine and 2′-deoxyuridine, but a poor correlation for elevated xanthine and N(1)-acetylspermidine in older irradiated mice when exposed to 3 Gy γ radiation (145). These findings suggested that aging might be associated with higher levels of oxidative stress and a decline in DNA damage-repair efficiency while also implying a specific role for polyamine metabolism. Other studies have identified a time- and dose-dependent response in the GI metabolites to 4–8 Gy in mice (146) or in the urinary metabolites, β-thymidine and N-hexanoylglycine, in mice when exposed to 0, 3 and 8 Gy (2.57 Gy/min) for 24 h (147). Altered metabolic profiles of tissues, including bone marrow, ileum, liver, muscle, and lung, were also observed within 12 h postirradiation (6 Gy, dose rate of 0.92 Gy/min), which are associated with DNA methylation, energy metabolism, and amino acid metabolism (148). Li and colleagues analyzed T cells from irradiated mice that received 0.1, 0.5, or 3 Gy TBI at a dose rate of 1.7 Gy/min, and concluded that radiation effects were correlated with a decrease in key metabolic pathways such as glycolysis and energy metabolism (149).

A sex-specific study confirmed that thirteen dose-responsive metabolic biomarkers that include L-carnitine, xanthine, L-acetylcarnitine, and xanthosine were increased in male NHPs when exposed to 2, 4, 6, 7 and 10 Gy for 7 days (150). A follow-up study demonstrated that several derivatives of carnitine and acylcarnitines were also significantly altered in gamma-ray irradiated NHPs (151). Numerous LC-MS-based studies have been used to measure radiation-induced metabolic markers pertaining to energy metabolism, DNA damage, organ injuries, and inflammation (152–154). Metabolic profiles have also been evaluated for their use in the therapeutic intervention of after exposure to radiation. There is a recent report for conducting a study in irradiated NHP to evaluate metabolic changes in plasma and plasma-derived exosomes. NHP were gamma irradiated, blood samples were collected at various time points after 5.8 Gy or 6.5 Gy TBI. Exosomes were isolated and analyzed for untargeted metabolomic and lipidomic profiling. Plasma profiling demonstrated markers of dyslipidemia, inflammation and oxidative stress after irradiation. Based on the differences in metabolite composition between plasma and exosomes, it was suggested that exosomal profiling may augment the identification of low abundance biomarkers that would otherwise be obscured in plasma (155). In another study using 7.2 Gy TBI NHP serum, authors reported temporal fluctuations of metabolites within 96 h postirradiation and higher fold changes of altered metabolites at 7.2 Gy compared to 6.5 Gy at 24 h postirradiation. These findings highlight the importance of biofluid collection timepoint for successful interpretation of serum metabolic profiles (156, 157). The profound elevation of long-chain acylcarnitines in irradiated NHPs across multiple tissue types underscored the role of this class of metabolites as a generic indicator of radiation-induced tissue injury (158).

Metabolomics studies have also been applied to biodosimetry-based analyses (159, 160). For example, in human cell line studies, depleted metabolites such as glutathione, adenosine monophosphate (AMP), and nicotinamide adenine dinucleotide (NAD) were linked to oxidative stress and DNA repair pathways 1 h after 0.5–8.0 Gy dose of radiation (161).

Low-dose irradiated human keratinocytes also exhibited a disruption in energy metabolism when exposed to 0.03, 0.1 and 2 Gy of X rays for 3, 24 and 48 h (162). An integrated study involving transcriptomics and metabolomics of irradiated human bronchial epithelial (HBE) cells measured 326 differentially expressed genes and 147 altered metabolites 24 h after 4 Gy exposure. These omics results suggested that after irradiation cellular metabolism may be regulated by p53 (163). There is also a report for metabolomic and lipidomic profiles in multiple tissues (liver, kidney, jejunum, heart, lung, and spleen) of NHP that received 7.2 Gy of γ TBI. There were robust metabolic changes in the kidney and liver and modest changes in other tissues. Overall, metabolomics has identified numerous potential biomarkers that may be useful for analyzing, diagnosing, or treating ARS or other symptoms of exposure to radiation.

Altered proteomic profiles. Like metabolomics profiling, proteomics analyses can also be exploited to measure the detrimental effects of radiation exposure. Radiation might alter regulatory networks or cause proteomic modifications such as decarboxylation, disulfide bonds, or aggregation (164) and subsequently affect the steady-state levels of specific proteins. Several studies have analyzed different tissues to measure changes in protein abundance resulting from radiation exposure. For example, 19 differentially expressed proteins associated with biological functions such as the DNA damage response, stress response, and cytoskeleton system were observed in the intestines of irradiated mice (exposure to 9 Gy for 24 and 72 h) when analyzed using two-dimensional gel electrophoresis (165). A single dose study on the liver proteome of mice (from 0.02 to 1 Gy) showed a remarkable downregulation of glycolysis and pyruvate dehydrogenase availability. A subsequent long-term outcome was an increase in liver inflammation (166). Another study focused on the spleen proteome of tocotrienol (GT3) treated mice 24 h prior to 7 Gy TBI, which showed a difference in the expression of several proteins, including the upregulation of Wnt signaling pathways (167). This study also identified significant alterations in the levels of metabolic enzymes such as aldehyde dehydrogenase and propionyl coenzyme A, guanylate cyclase, and glycine amidinotransferase, which suggested possible changes in the carbon and amino acid metabolism due to radiation. Other differentially expressed proteins suggested that cell signaling proteins attenuated radiation-induced injuries.

Previous studies characterized the radiation-induced proteome changes in different body fluids such as urine and blood plasma (168) from NHPs exposed to 6.7 Gy and 7.4 Gy (168, 169). Some of the differentially expressed proteins based on dose and time points in both the samples from the same animal included ferritin (F6ZV45), angiotensinogen preproprotein (G7MFR4), putative uncharacterized proteins (G7NQN4 and G7MJ28), angiotensinconverting enzyme isoform 1 (H9FJ99), and other uncharacterized proteins (F6TLR3, F7DHQ1, and F7GRY2). Further, the resulting urine proteome profile identified many proteins associated with cell adhesion, disease progression, and key metabolic pathways (169). Proteomic changes in the serum of high-dose irradiated NHPs after administrating BIO 300, a synthetic genistein nanosuspension, identified upregulated proteins, e.g., tubulin α chain (B3KT06), cDNA FLJ57036 (B4E3P1), CP protein (A5PL27), glutathione peroxidase (V9HWN8), carbonic anhydrase I (V9HWE3), glutathione S-transferase pi 1 (V9HWE9), which helped understand the metabolic changes associated with drug effects (170).

Several previous studies also focused on understanding the proteome changes in human tissues such as endothelial cells (<2.5 Gy γ irradiation) (171, 172), peripheral lymphocytes (1, 2 and 4 Gy γ irradiation) (173), skin fibroblast cell lines (10 Gy γ irradiation) (174). Proteomics-based studies have also suggested that tumor-derived factors are upregulated by radiation and may serve as potential therapeutic targets during radiation treatment (175). The changes in the proteomic profiles due to radiation exposure is associated with biological phenotype such as inflammation, toxicity, immune response and retinoic acid signaling in human (176, 177). Some studies have noted that protein expression changes were associated with radiotherapy and tumor resistance in different organs such as the breast, lungs, and brain (175, 178–181). A study that treated breast cancer cell lines with a single 10 Gy dose of γ radiation showed the top differentially regulated proteins such as C-type mannose receptor 2, arginino-succinate synthase 1, gelsolin, peroxiredoxin 5, and cathepsin D (182), while the lung cells in the same study showed cofilin-1, HSPB1, annexin A4, and vimentin as the top upregulated proteins (183). Overall, these reports have demonstrated the potential of proteins as reliable biomarkers for high-dose exposure.

Radiation Countermeasures: Radioprotectors and Radiomitigators

As discussed above, exposure to high or low doses of radiation can affect human health in many ways, primarily resulting in ARS, cutaneous radiation syndrome (CRS), neurovascular syndrome, or even death. These health effects are radiation-induced damage such as genomic alterations or direct DNA damage, destruction of blood-forming stem cells, and damaging the immune system and other metabolic functions.

Current drug discovery efforts are focused on developing radioprotectors and radiomitigators for their use during pre- and postirradiation, respectively. To date, several agents have been tested as potential radioprotectors and/or radiomitigators to counter ARS. The manifestation of ARS is related to disorders of the GI, neurovascular, hematopoietic, and cutaneous systems (184).

Radiomitigators

Under this category, the thrombomodulin (Thbd)-activated protein C (aPC) pathway has been shown to stimulate blood cell production and help attenuate radiation injuries in mice (dose rate of 1.37 Gy/min and 0.52 Gy/min for two separate mouse models) (185). In general, aPC has an association with several cellular activities such as anti-coagulation, anti-inflammation in blood, and other cytoprotective properties such as endothelial barrier protection, protection against vascular leakage, inflammation, apoptosis, and inflammasome activation (186). The drug TP508 (rusalatide acetate, Chrysalin) also mitigates the effects of radiation by activating radioresistant stem cells in the intestines and colonic regions of mice when treated 24 h after a lethal 9 Gy dose of radiation (187). In another study, TP508 was found to stimulate cellular events linked with the repair of bone, skin, and muscle tissues in rats (188). Among the FDA-approved radiomitigators, Neupogen and Neulasta are the most efficient for treating radiation damage and neutropenia (189, 190). These drugs help improve granulocyte colony-stimulating factor (G-CSF) pharmacokinetics and subsequently empower the immune response against radiation injuries (191). Another FDA-approved drug, Leukine (Sargramostim), has also shown potential radiomitigative effects to treat neutropenia in irradiated patients (192–195). Apart from G-CSF, NPLATE (Romiplostim) is the first FDA-approved thrombopoietin receptor agonist used as radiomitigator to treat low-platelet counts due to acute radiation exposure as studied by Micewicz et al. (198). In a recent study, Singh and colleagues analyzed metabolic and lipidomic profiles in the serum of NHPs (7.2 Gy of γ radiation) after treatment of a candidate drug, Ex-Rad (ON01210), which showed significant alterations in biochemical pathways towards the recovery of radiation-injured organs (199). Several other studies and reviews were focused on repurposing of drugs as radiomitigators (198–202).

Radioprotectors

Unlike radiomitigators, radioprotectors are administered before radiation exposure and investigated for their counter effects against ARS in different models, including mice and NHPs. BIO 300 is one example of a promising candidate that causes alterations in the metabolic profile, which further helped identify potential biomarkers associated with lung and other radiation-induced tissue injuries in the mice model when exposed to a single dose of 11.0 or 12.5 Gy of whole-thorax lung irradiation (203–205). In NHPs, serum-based global lipidomics and metabolomic changes were observed after treatment with BIO 300, which were again used to understand the regulation of different pathways associated with the radiation-induced injuries. Specifically, significant alterations in the levels of important metabolites such as tyrosine, glycerophosphoserine, glycerophosphocholine, and phenylalanine were observed (206). A similar study with proteomics concluded that BIO 300 causes elevation of actin by minimizing its nitration in NHP model (170).

Similarly, amifostine (WR-2721), as a radio-protecting candidate, showed significant positive impacts against ARS in humans. Metabolomics and lipidomics analyses of control and amifostine-treated mice, when exposed to 9.6 Gy of whole-body irradiation, revealed that radiation triggered the dysregulation of 1,614 metabolites in bone marrow, jejunum, and lung samples. Further analysis concluded that bone marrow exhibited a heightened response to the protective outcomes of amifostine, while jejunum and lung had a modest response (207). In another study, reduced levels of metabolites such as hypoxanthine, glutamic acid, and L-valine due to exposure to radiation was mitigated, while, some elevated metabolites like PS (18:0/ 20:4) and L-arginine were corrected after treatment by amifostine in mice, which were exposed to 9.6 Gy of γ radiation (208). Similarly, MS and NMR studies of serum of mice (14 Gy of γ radiation) and NHPs (5.8 Gy or 7.2 Gy γ radiation, dose rate of 0.6 Gy/min) showed dysregulation of twenty-three pathways after γ irradiation. These altered metabolic pathways mainly included lipid biosynthesis, glycolysis, and nucleotide metabolism (209). This study further observed reversed metabolic signatures of ARS progression through pretreatment of Amifostine, showing the potential of the drug as a radioprotector; however, side effects such as nausea and vomiting are still a major concern. Amifostine has been tried to treat ARS but showed toxic side effects, however later approved by FDA for xerostomia (210–212).

Antioxidants, especially the group of vitamin E compounds, have been widely studied as radioprotectors (213, 214). For example, α-tocopherol enhanced the survival of mice when administered 24 h before γ-irradiation (dose rate of 0.6 Gy/min) (214). Similarly, GT3 has also been investigated as radioprotective agent. This agent exerted metabolic changes in NHPs, which were linked with increasing levels of antioxidants, suggesting a protective benefit to countermeasure radiation-induced injuries (215). Global metabolomic changes were analyzed in another study using irradiated NHP treated with GT3. Analysis of serum samples identified several altered metabolites after irradiation, including compounds involved in fatty acid beta-oxidation, purine catabolism, and amino acid metabolism. A machine-learning algorithm, Random Forest, separated control, irradiated GT3-treated, and irradiated vehicle-treated NHPs at 12 h and 24 h. Primary metabolites validated included carnitine/acylcarnitines, amino acids, creatine, and xanthine. Overall, GT3 administration reduced high fluctuations in serum metabolite levels, suggesting an overall beneficial effect on animals exposed to radiation (157). Recently conducted studies with serum and jejunum samples of GT3-treated mice that received 11 Gy TBI demonstrated the restoration of radiation-induced proteomic changes by GT3 (216, 217). Likewise, an injection of selenium-containing compounds, such as sodium selenite or selenomethionine, resulted in the improved survival of mice after irradiation.

Agents that Serve Both as Radiomitigators or Radioprotectors

Some of the drugs that have been investigated as radioprotectors and radiomitigators to modulate ARS effectively include glucans, 5-androstenediol (5-AED), and meloxicam (218, 219). In animal trials, β-glucan has shown potential both as a radioprotector and a radiomitigator when mice were exposed to a 6, 7 and 8 Gy dose γ radiation and a 4 Gy dose (dose rate of 0.15 Gy/min) in two different studies (220, 221). β-glucan combined with other radioprotectors, such as cystamine or WR-2721, showed a positive impact in protecting from radiation damage (222). The administration of β-glucan with WR-2721 and selenium also resulted in favorable outcomes (223). Several other studies revealed similar successes in the use of β-glucan either pre- or postirradiation to treat radiation toxicity and improve animal survivability (224–227). A pre- or postirradiation treatment with 5-AED was also effective in modulating the immune response to counter the impact of radiation. 5-AED stimulated myelopoiesis and the recovery from infections in mice (exposed to 3 Gy of total-body γ radiation) and NHPs (4 Gy γ TBI) (218, 228). Increased G-CSF production in 5-AED treated mice 24–48 h prior to 7.5 Gy TBI (dose rate of 0.6 Gy/min), synergistically improved the activation of monocytes, granulocytes, and NK cells, along with an increase in the number of innate immune cells (229–231). 5-AED is an open Investigational New Drug (IND) used as a radioprotector or radiomitigator for radiation-related injuries (232).

The mechanism of action and efficacy of a variety of such compounds and biomolecules were comprehensively discussed in previous reviews (201, 233, 234). Various highlighted compounds, especially the repurposed ones, that can be used to modulate ARS pre- and postirradiation, which include immunomodulators, are listed in  Supplementary Table S1 (89_rare-199-01-02_s01.docx) ( https://doi.org/10.1667/RADE-2100187.1.S1). A short summary of radiation countermeasures studied in different animal models is provided in Fig. 2B. Different radiation drugs were categorized into four main classes i.e., immunomodulator, prostaglandin, antioxidant and other biologicals. Interestingly, our literature search showed 12 compounds that have been studied as radioprotectors and radiomitigators (Fig. 2B and  Supplementary Table S1 (89_rare-199-01-02_s01.docx)).

In the recent past, countermeasures have been used for studying metabolomics in NHP models with different doses of TBI. The metabolomic profile in serum samples of NHPs treated with Ex-Rad, also known as ON01210, a chlorobenzyl sulfone derivative (organosulfur compound), after irradiation (7.2 Gy, 0.6 Gy/min, LD_70/60) has been investigated. Two different dose administration schedules (24 and 36 h postirradiation as well as 48 and 60 h postirradiation) were used to investigate the global profiling approach. Alterations in the biochemical pathways relating to inflammation and oxidative stress after irradiation were alleviated in animals that received Ex-Rad treatments (201).