4.1. Markers of DNA Damage and Genome Instability in FC that Are Consistent with the Expected Biological Effects of CIR Exposure
This section of the review summarizes available data supporting a model wherein the elevated cancer risk observed in FC can be explained, at least in part, by an underlying molecular mechanism involving CIR-induced DNA damage. Since direct experimentation with actual CIR is not feasible, evidence for potentially harmful occupational exposure to CIR among FC has come from investigations of biomarkers in response to exposure to sources of terrestrial radiation meant to model the qualities of CIR. Although these types of radiation cannot exactly recapitulate the properties of bona fide CIR, they have provided an important model for how CIR may be expected to affect biological systems.
Each day, cells experience >100,000 spontaneous DNA lesions which includes base damages, oxidative damage products such as 8-hydroxy-2-deoxyguanosine (8oxoG), single strand breaks (SSBs) or double strand breaks (DSBs) [
74]. These endogenously produced DNA lesions are usually resolved efficiently by cellular DNA repair machinery. The same types of DNA lesions can be induced by IR, including CIR, but they are more likely to occur as clusters of lesions comprised of more than one type of damage [
75,
76,
77]. When two or more isolated lesions occur within 10-20 bp (i.e., ~1-2 helical turns of the DNA), the damage is referred to as a clustered DNA lesion [
74,
78]. Clustered lesions may include complex DSBs and small DNA fragments arising from multiple closely spaced DSBs [
21,
79].
An extremely low dose rate relative to background exposure makes it difficult to detect CIR-induced damage. However, this type of radiation has properties that distinguish it from other types of radiation and may enable specialized strategies for detecting its effects on the genome. CIR includes high-LET radiation, which differs from low-LET radiation, such as X-rays, in terms of the ionization density along the radiation track [
80] and the relative biological effectiveness (relative measure of DNA damage done by a radiation per unit track of energy deposition in a biological tissue) [
81], which govern the types of DNA damage induced [
82]. Complex clusters of oxidative DNA lesions are produced to a greater extent by high-LET IR than by low-LET IR and, as a result, have more harmful consequences [
76,
83]. For instance, human cells exposed to 1.0 Gy
56Fe ions produce 20-50 DNA fragments of <1000base pairs, about 30 times higher than 1.0 Gy of γ rays [
79]. DNA damage induced by heavy ions is more efficient in inducing cellular senescence [
84], premature chromosomal condensation (biodosimetry to detect interphase chromosomal damage) [
85], micronuclei formation (biomarker of IR induced un-repaired DSB during cell cycle) [
86], and γ - H2AX (biomarker of DSBs) [
87] compared to high doses of low-LET IR. Interestingly, and consistent with the induction of a different spectrum of DNA lesions, mutational signatures induced by high-LET IR are distinct from those induced by low-LET IR [
88]. The studies discussed in this section involve much higher levels of DNA damage than would be expected upon CIR exposure in FC. According to the Linear No Threshold (LNT) model, it is assumed that risk associated at higher doses can be extrapolated to lower doses. However, alternative dose response relationships have been proposed, including some that depend upon the details of the of exposure including dose, dose-rate, type of radiation [
89].
The biological effects of CIR component particles on cells depend significantly on the dose rate. One study quantitated the level of γ- H2AX foci following exposure to low dose rate (0.015 Gy/min) vs. high dose rate (0.400 Gy/min) neutrons at several total doses (0.125 Gy to 2.0 Gy) in human peripheral blood mononuclear cells (PBMCs). In this study, averaging over all doses, 40% greater induction of γ- H2AX foci was observed after high neutron dose rate exposure compared to low dose rate exposure. The foci levels decreased 24 h after irradiation, and foci remained significantly higher than the background levels irrespective of the neutron dose rate [
90]. This indicates that the dose rate is a factor that may need to be considered for cancer risk estimations from neutron exposures, which are the most abundant component of CIR.
A very limited number of studies have directly measured biomarkers of DNA damage in FC. A study on 44 male airline pilots compared to 36 factory workers was conducted to determine whether indirect CIR estimates can correlate with 8oxoG levels in blood and urine samples [
30]. The study showed that levels of 8oxoG were higher in pilots, though it was not possible to establish a CIR dose-response relationship. It has also been hypothesized that interactions with non-ionizing radiofrequency (e.g., in-flight Wi-Fi) and low-frequency electromagnetic fields aboard aircraft could augment the deleterious effects of CIR by increasing cellular oxidative stress [
91,
92,
93], however, adverse health effects from non-IR exposure are inconsistent across the literature [
94]. A study using comet assays to compare civil FC flying long haul routes versus matched ground staff showed a non-significant trend toward higher levels of basal DNA damage among FC (measure oxidative damage, SSBs and DSBs) [
95]. When comparing high LET radiation exposure versus low LET radiation exposure, divergent patterns of DNA damage, gene expression, repair protein mobilization, cytokine activation, and cellular microenvironment remodeling are observed [
96]. Furthermore, some particle components of CIR, which have not been studied as thoroughly as photon radiation, also have distinctive properties concerning their biological effects [
97], as well as the types of DNA damage they produce, as detailed above. Sensitive assays capable of detecting these rare, distinctive types of DNA damage are needed to investigate the biological effects of CIR.
Several studies using cytogenetic analysis (chromosomal aberrations, formation of micronuclei, sister chromatid exchanges) have yielded mixed results with regard to evidence of CIR-induced damage in FC. Chromosomal aberrations, formed by inaccurately repaired or unrepaired DSBs [
98] in human lymphocytes have been considered a reliable indicator of low-dose radiation exposure [
3]. They can be observed several years after radiation exposures in PBMCs, such as in atomic bomb survivors [
99,
100]. Some FC studies have found significantly elevated levels of chromosomal aberrations versus the selected control group [
67,
68,
69] however others found no difference [
70,
71]. The different findings of these studies may be due to differences in study design including the selection of the control group, and whether a sufficient number of cells were scored to detect an expected small increase in these relatively rare cytological events. Significant higher levels of chromosomal aberrations and micronuclei were observed comparing the Concorde pilot group (estimated mean dose per year from 11 to 37mSv depending on radiation weighting factor of neutron) versus controls who were matched for age, health, and socioeconomic status [
67]. Similar statistically significant differences were also observed for civilian pilots and cabin crew of subsonic aircraft (n=192; 120 males and 72 females) compared to non-flying healthy volunteers (n=55; 24 males and 31 females) [
68]. On the other hand, a different study showed no difference between 83 airline pilots versus 50 comparison subjects (mean age 47 versus 46 years, respectively) [
71]. This underscores the point that the choice of control group is a critical factor. In addition, the parameters used for scoring chromosomal aberrations can play an important role in the outcome. For example, interlaboratory variation in recognizing, rejecting, classifying various type of chromosomal aberrations and setting a different threshold for scoring has been documented [
101]. The frequency of chromosomal aberrations is expected to be directly proportional to radiation exposure. Yet, while some FC studies have yielded findings that are consistent with this expectation [
70,
71] others have not [
69]. Significant higher levels of chromosomal translocations per cell was observed in long term male pilots (healthy, non-smoking, aged 40-60 years, recruited by single airline) versus non pilot controls (aged matched, no frequent flying history) [
69]. However, the number of translocations per cell showed no dose response relationship among the pilots [
69]. Similarly, the study done by Grajewski et al., 2018 [
42], observed no association in translocation frequency and estimated absorbed dose from all types of flying male pilots. By contrast, linear relationship between cytokinesis-blocked micronuclei and the average annual effective CIR dose of radiation received or the average annual flying hours in FC was observed in PBMCs [
72]. The DNA damage underlying cytogenetic abnormalities in FC could potentially be attributed to other in-flight exposures. Nevertheless, studies that have found elevated levels of chromosomal aberrations and particularly those finding a dose-response relationship with estimated CIR exposure are consistent with genome instability induced by CIR.
Animal and cell culture-based models have provided insights into the mechanisms that may be involved in the biological effects of CIR. These models remove the complexity inherent in population settings and enable more detailed investigations of the impact of CIR or its components in a cell- or tissue-specific manner that cannot be achieved using the biological samples usually available from human subjects. CIR exposures are anticipated to have tissue-specific effects on the human body. It is well-established that the biological effects of IR are tissue-dependent (
Figure 4) [
102,
103], partly due to tissue-specific sensitivity to the induction of apoptosis [
104]. Thus, there is a need for additional studies to understand the underlying mechanisms. Li et al. reported that exposure to
48Ti ions (one of the components of GCR) induces deletions in lung-derived epithelial cells in vitro when DSBs are induced in regions of the genome that have flanking short microhomologies that promote error-prone alternative end-joining pathways. In addition, in bronchial epithelial cells, they found an increase in chromosomal rearrangements, which are associated with increased lung cancer risk [
77]. Persistent epigenetic effects, such as alterations in DNA methylation, have also been reported in lung cells exposed to components of GCR [
105].
Despite important advances, our understanding of the impact of CIR exposure on human health remains limited and warrants further investigation. DNA repair mechanisms remove damage induced by CIR. Dysregulation of DNA repair pathways promote genomic instability and may underly the elevated cancer risk observed in FC. The next section delineates the major DNA repair pathways with emphasis on their roles in repair of DNA damage expected to be induced by CIR. The challenge of measuring CIR induced damage has hampered efforts to study it’s health effects, but given the central importance of DNA damage and repair for cancer [
106], CIR-induced genomic instability provides a biologically plausible explanation for elevated cancer risk in FC that needs to be further investigated.
4.2. DNA Repair Mechanisms Involved in the Cellular Response to Damage Induced by CIR Components
Genomic instability is a hallmark of cancer and drives radiation-induced carcinogenesis [
107,
108]. To maintain genome stability, cells employ numerous DNA repair mechanisms [
109]. Although it has not been feasible to study these mechanisms directly in the context of CIR, various laboratory-based model systems have been used [
110,
111,
112]. Each component of CIR is expected to induce a distinct spectrum of DNA lesions, each recognized and processed by a specific DNA repair mechanism. To understand the implications of this complexity, many groups have investigated the DNA damage response (DDR) in human PBMCs exposed to various doses of alpha particles, X-rays, and mixed beams of radiation [
113]. For example, expression of DDR genes such as ataxia telangiectasia mutated (
ATM) protein, tumor protein 53 (
P53), and DNA protein kinase catalytic subunit (
DNA PKcs) have been reported to be induced to a greater degree when cells are exposed to comparable doses of mixed beam versus pure high-LET alpha particles or pure low-LET X-rays [
113]. This suggests that mixed high- and low-LET radiation may present a greater challenge to genomic integrity. Experiments using neutron sources in vitro circumvent some of these challenges and have provided insights into the potential genotoxic effects of CIR. For instance, compared with mock-irradiated controls, a significant increase in micronuclei frequency, DNA damage levels, and ɣ-H2AX foci was observed after neutron irradiation from
241Am-
9Be source at doses as low as 9.0 mGy [
114]. In addition, consistent with induction of a DDR, increases in
GADD45A, CDKN1A, and
PARP1 transcript levels were reported at 4 hours post-irradiation with low doses of neutrons in human resting PBMCs [
114]. The molecular mechanisms of DNA repair have been extensively studied for higher doses of IR. However, there is a gap in our understanding of the biological effects and health risks associated with exposure to low-dose or low-dose-rate IR and complex mixtures of low- and high-LET radiation exposures, as seen in CIR.
In the remainder of this section, we summarize available literature on the possible role of CIR-induced DNA damage and DNA repair pathways in FC health outcomes. Since very little work has been done with actual CIR-induced DNA damage, most of this work relies on model systems and assumptions based on our understanding of the biological effects of IR. IR produces a variety of types of DNA damage. 1.0 Gy of gamma rays induces thousands of base lesions and SSBs and about 40 DSBs per human cell [
23]. Relative to low-LET photon radiation, the high-LET components of CIR may induce a higher proportion of DSBs [
115], which are mainly repaired by the
non-homologous end joining (NHEJ) and
homologous recombination (HR) pathways. Defects in NHEJ and HR activity have been associated with chromosomal aberrations, immune dysfunction, clinical radio-sensitivity, and elevated cancer risk [
116,
117]. The presence of one or more DNA lesions at or near the ends of DSBs requires additional factors for repair by the NHEJ pathway [
118], and the need for DNA end processing increases the possibility of slower and alternative error-prone DNA repair mechanisms [
119,
120]. A linear dose-response relationship indicates that a single ionization track gives rise to most DSBs for both high- and low-LET radiation [
23], but the dense ionization tracks associated with high-LET radiation are more likely to give rise to complex lesions that are repaired slowly compared to those induced by photons [
121]. In addition to directly induced DSBs, radiation can lead to DSBs by indirect means that remain incompletely understood. Genome instability can also occur in the progeny of irradiated cells and is termed delayed hyperrecombination [
122,
123]. Several-fold increases in delayed hyperrecombination that last for multiple weeks have been reported in response to low-LET X-rays or high-LET carbon ion radiation in mice [
122]. In further support of the notion that for a more prominent role in DSB repair of DNA damage induced by high-LET radiation, some studies have demonstrated HR-deficient cell lines having greater sensitivity to proton radiation than low-LET photons of the same energy [
124].
Microhomology-mediated end joining (MMEJ) (mutagenic DSB repair mechanism that uses microhomologies flanking break the site to guide repair) [
125] also contributes to the repair of radiation-induced DSB; the radio-sensitizing effects of MMEJ deficiency depend upon cell type and genetic context [
126,
127]. However, role of NHEJ, HR and MMEJ still remains unexplored in repair of CIR induced damage in FC cohort
.
SSBs, abasic sites, and oxidative DNA damage, the most abundant types of DNA lesions produced by IR, are primarily repaired by the
base excision repair (BER) machinery [
128]. Several DNA glycosylases including hNTH1(endonuclease III homolog), hOGG1 (8-oxoguanine DNA glycosylase), NEIL1 (Nei like DNA glycosylase 1), initiate repair of oxidative DNA damage induced by IR, and this processing can lead to their conversion to more dangerous SSB and DSB [
129]. The essential BER protein apurinic/apyrimidinic endonuclease 1 (APE1) plays a vital role in processing IR-induced oxidative DNA damage and clustered breaks [
130]. APE1 deficiency sensitizes cells to IR despite resulting in a smaller number of radiations induced DSBs, presumably arising when replication forks collide with SSB intermediates downstream of APE1. Failure to resolve radiation-induced clustered DNA lesions can result in the accumulation of genetically unstable cells [
131]. BER can be presumed to play an important role in CIR induced DNA damage but has not been studied directly in FC.
Bulky DNA adducts, IR induced inter-strand crosslinks, cyclopurines and variety of oxidative lesions are repaired by the
nucleotide excision repair (NER) pathway. Although NER-deficient cells are not hypersensitive to IR, genetic polymorphisms in some NER genes have been associated with radiation-related cancers [
132]. Some NER proteins are also involved in the repair of inter-strand crosslinks that IR can induce. IR also induces some types of DNA lesions that can be repaired by NER, such as cyclopurines [
133], and a variety of oxidative lesions [
134]. Little is known about the role of NER in the response to CIR-induced damage in FC. A study evaluating FC cumulative CIR dose exposure (assuming an exposure rate of 6 µSv/hr during flight) showed a significant correlation between CIR dose and NER activity in PBMCs measured by the unscheduled DNA synthesis assay [
73]. This could reflect a healthy worker effect [
18] but would also be consistent with a radio-adaptive response that has been proposed to underlie radiation hormesis [
135] and has been observed in other settings [
80,
136].
Additional DNA repair pathways including Fanconi Anemia repair, and others are expected to play a role in the response to CIR-induced DNA damage but require further study. The
Fanconi Anemia repair pathway eliminates inter-strand DNA crosslinks and protects cells from killing by IR. It is essential for maintaining genome integrity, and defective repair pathway have been associated with increased cancer risk and immune disorders [
137]. Based on the complexity of CIR induced DNA damage, multiple DNA repair pathways are expected to be involved in its repair. To address this complexity, there is a need to investigate the role of multiple DNA repair mechanisms for CIR-induced DNA damage in FC. Taken together, these results underscore the importance of the involvement of multiple repair pathways in response to CIR based on the complexity of induced DNA damage and also the need for additional mechanistic studies aimed at understanding the origins of differential sensitivity to high- versus low-LET IR in FC. Such studies are expected to assist in evaluating the potential health effects of in-flight exposure and the eventual design strategies to mitigate them.