Energy Deposition leads to Increase, Mutations

The biological rationale for linking direct deposition of energy by ionizing radiation to mutation induction is strong. The structural and functional relationships in this KER contribute sufficiently to the overall biological plausibility.

There are numerous studies that demonstrate, using various model systems, an increase in mutation frequency in response to radiation exposure (Russell et al., 1957; Winegar et al., 1994; Gossen et al., 1995; Suzuki & Hei 1996; Albertini et al., 1997; Dubrova et al., 1998; Kraemer et al., 2000; Dubrova, Plumb, et al., 2000; Canova et al., 2002; Dubrova et al., 2002; Dubrova & Plumb, 2002; Masumura et al., 2002; Somers et al., 2004; Burr et al., 2007; Ali et al., 2012; Bolsunovsky et al., 2016; Mcmahon et al., 2016; Matuo et al., 2018; Nagashima et al., 2018; Wu et al., 1999; Hei et al., 1997; Nagasawa and Little, 1999; Barnhart and Cox, 1979; Thacker at al., 1982; Zhu et al., 1982; Metting et al., 1992; Schwartz et al., 1991; Chen et al., 1984; Albertini et al., 1997). The process of mutation induction by radiation is initiated when cells are exposed to ionizing radiation. These high-energy waves or particles interact with the genetic material in the nucleus, damaging the DNA and triggering a cascade of signalling events and activities aimed at repairing the damage. It has been shown that various dose rates of radiation exposure can lead to distinct types of damage. High dose-rate radiation has been observed to generate a higher number of DNA strand breaks, resulting in a variety of mutations, including small base changes and deletions. Moreover, the likelihood of insufficient repair is elevated, contributing to an overall increase in mutation frequency. In contrast, low dose-rate radiation has been found to have a significantly lower  mutation frequency, particularly in deletions and rearrangements (Brooks et al., 2016; Sankaranarayanan & Nikjoo, 2015). Of note, radiation is not likely to impact only one gene; more often than not, the random nature of energy deposition by radiation results in mutations to many genes and genomic sites clustered in the same area (Sankaranarayanan & Nikjoo, 2015; Adewoye et al., 2015). Many of the radiation-induced mutations have been documented as deletions (Gossen et al., 1995; Behjati et al., 2016), often of differing sizes in a number of different genes (Sankaranarayanan & Nikjoo, 2015). The mechanism for radiation-induced mutations is thought to be similar to the process for spontaneously-occurring mutations, as the structure of radiation-induced mutations examined at expanded simple tandem repeat (ESTR) loci was not found to differ from the structure of spontaneous mutations (Dubrova, 2005). Moreover, exposure to radiation may produce specific mutational signatures. Two ionizing radiation-specific mutational signatures were found when 12 radiation-induced secondary tumours across 4 different tumour types underwent whole-genome sequencing and bioinformatics processing. In particular, these radiation-exposed tumours were significantly enriched in small deletions and balanced inversions. These results were validated when the same mutational signatures were observed in radiation-exposed but not radiation-naïve prostate tumours from a previously-published dataset (Behjati et al., 2016). Similarly, another study examining mutations present in radiation-induced tumours of Nf1 heterozygous and wild-type mice revealed three distinctive mutational signatures. Interestingly, these signatures were found in all of the tumours regardless of its histology or of the animal’s genotype. Moreover, these signatures were still present after removal of the 33 most mutated samples from the analysis, after analysis of only the non-synonymous substitutions, and after analysis of only the synonymous substitutions (though the third mutational signature could not be extracted in this last analysis group) (Sherborne et al. 2015). There were also common cellular pathways that were found to be frequently mutated in the tumours of these mice. In sarcomas from mice of both genetic backgrounds (Nf1 heterozygous and wild-type), the top two pathways harbouring mutations were those influencing cellular assembly and organization, and those involved in cellular function and maintenance. Additionally, Ras pathways were commonly mutated in tumours from both genetic backgrounds. Specific to wild-type sarcomas, mutations were also found in cell cycle and cell signalling pathways (Sherborne et al., 2015). Supporting the finding that different genetic backgrounds in mice do not affect mutational signatures in tumours (Sherborne et al., 2015), there also does not appear to be strain-specific differences in ESTR mutational frequencies in response to radiation. One study examined five different strains of male mice that were irradiated and mated to unirradiated females at least 4 weeks post-irradiation. Although there was a difference in doubling doses between strains, the ESTR mutations themselves were not significantly different. Furthermore, there were no significant differences found between strains in terms of germline mutation induction (Dubrova, 2005).

Germline mutations have been further interrogated in studies examining the effects of radiation exposure on germ cells. There is evidence from mouse studies suggesting that the germ cells of radiation-exposed males have elevated ESTR mutations and that the offspring of these irradiated males inherit more ESTR mutations as a result of the germline mutations (Dubrova et al., 1998; Dubrova, Bersimbaev, et al., 2000; Dubrova & Plumb, 2002; Somers et al., 2004; Barber et al., 2009; Ali et al., 2012; T.E. Wilson et al., 2015). This was reviewed by Somers et al. (2006). Interestingly, in utero irradiation of embryos at day 12 resulted in increased ESTR mutations across several tissue types in males and females; however, only the offspring of the irradiated males showed an elevated ESTR mutation rate (Barber et al., 2009). On a genome-wide scale, the offspring of irradiated males were found to have significantly more clustered single nucleotide variants (SNVs) and insertion/deletion events compared to offspring from unirradiated fathers (Adewoye et al., 2015).

Human studies have also shown correlations in radiation exposure and increased germline mutations. This relationship was assessed in families exposed accidently to high doses of ionizing radiation after the Chernobyl accident in Ukraine, and in families living in close proximity to the Semipalatinsk nuclear test site in Kazakhstan. In both cases, germline mutations were evaluated using eight hypervariable minisatellite probes. In the Chernobyl study, the paternal mutation rate in the exposed group was significantly increased by 1.6-fold relative to an unexposed control group; there was, however, no significant difference in the maternal germline mutation rates between the exposed group and the unexposed control group (Dubrova et al., 2002C). In the Semipalatinsk study, analysis of families living in the affected region over three generations found that germline mutations in the first and second generation were significantly increased relative to unexposed families living in a low-radiation area. Overall, the germline mutation rate in the families exposed to radiation from this test site was doubled (Dubrova, Bersimbaev, et al., 2000).

Overall, there is strong supporting evidence that direct deposition of energy increases the frequency of mutations. The evidence presented below is summarized in table 2, here (click link). In general, exposure to ionizing radiation has been documented to elevate mutation frequencies in a number of different studies spanning different models and cell types (Russell et al., 1957; Winegar et al., 1994; Gossen et al., 1995; Suzuki & Hei, 1996; Albertini et al., 1997; Kraemer et al., 2000; Canova et al., 2002; Dubrova & Plumb, 2002; Masumura et al., 2002; Bolsunovsky et al., 2016; Mcmahon et al., 2016; Matuo et al., 2018; Nagashima et al., 2018; Wu et al., 1999; Hei et al., 1997; Nagasawa and Little, 1999; Barnhart and Cox, 1979; Thacker at al., 1982; Zhu et al., 1982; Metting et al., 1992; Schwartz et al., 1991; Chen et al., 1984; Albertini et al., 1997) . Furthermore, several reviews outline evidence of the relationship specifically between radon gas exposure and mutation frequency (Jostes, 1996; Robertson et al., 2013; ICRP, 2005). At low doses (<1 Gy) the induction of mutations in cells has been observed for high-LET radiation such as alpha particles (Wu et al., 1999; Hei et al., 1997; Nagasawa and Little, 1999; Barnhart and Cox, 1979; Thacker at al., 1982; Zhu et al., 1982; Metting et al., 1992; Schwartz et al., 1991; Chen et al., 1984; Albertini et al., 1997).

Figure 1: Plot of example studies (y-axis) against equivalent dose (Sv) used to determine the empircal link between direct deposition of energy and increased cell mutation rates. The z-axis denotes the equivalent dose rate used in each study. The y-axis is ordered from low LET to high LET from top to bottom.

Figure 2: Plot of example studies (y-axis) against time scales used to determine the empircal link between direct deposition of energy and and increased cell mutation rates. The z-axis denotes the equivalent dose rate used in each study. The y-axis is ordered from low LET to high LET from top to bottom.

Dose and Incidence Concordance

It is clear that increasing doses of ionizing radiation is concordant with increased incidence of mutations (see table under Quantitative Understanding of the Linkage). Extensive evidence from in vitro studies using human cells (Suzuki & Hei 1996; Canova et al., 2002), animal cells (Schmidt and Kiefer, 1998; Canova et al., 2002; Mcmahon et al., 2016; Nagashima et al., 2018), hybrid cell lines (Kraemer et al., 2000), yeast cells (Matuo et al., 2018), and bacteria (Bolsunovsky et al., 2016) demonstrates this concordance. In vivo studies using mice have also found a dose-dependent increase in mutations across several different types of radiation (Russell et al., 1957; Dubrova & Plumb 2002).

This relationship between radiation exposure and mutation incidence is impacted by several different factors. Higher LET radiation, such as high LET carbon ions and neutrons, were found to induce more mutations in comparison to radiation of a lower LET, including low LET carbon ions, gamma-rays and X-rays (Dubrova & Plumb, 2002; Matuo et al., 2018). Similarly, more mutations were present in the gametes of mice exposed to acute X-rays compared to those exposed to chronic gamma-rays (Russell et al., 1957). The tissue being irradiated may also have a role in determining mutant frequency, as whole body irradiation of mice led to a significant increase in mutations (mostly deletions) of the spleen, liver, lung and kidneys (Gossen et al., 1995; Masumura et al., 2002), but not the testis (Masumura et al., 2002). Furthermore, the specific kind of mutation may be dependent on the type of radiation. In one study, irradiation of the liver with carbon ions resulted in a significant increase in deletion mutations, while irradiation with X-rays or gamma-rays resulted in a significant increase in point mutations (Masumura et al., 2002).

Temporal Concordance

Temporal concordance is well established. As described above, energy deposition happens immediately upon radiation exposure, with an increased incidence of mutations documented days or weeks after irradiation (Russell et al., 1957; Winegar et al., 1994; Gossen et al., 1995; Albertini et al., 1997; Canova et al., 2002; Dubrova & Plumb, 2002; Masumura et al., 2002; Matuo et al., 2018; Nagashima et al., 2018).

Essentiality

Not identified.

Uncertainties and inconsistencies in this KER are as follows:

1. In a review paper describing the role ionizing radiation plays in elevating mutation frequency in the germline and therefore genetic risk, Sankaranarayanan & Nikjoo (2015) stated that most radiation-induced mutations tended to be deletions. In contrast, an examination of ESTR loci mutations in offspring and their irradiated fathers found that the ESTR mutations tended to be gains more often than losses (Dubrova ,2005). This may, however, highlight a characteristic specific to ESTR mutations rather than mutations in general.
2. In a study examining the long-term of effects of in utero radiation exposure, males irradiated at embryonic day 12 showed significant increases in both somatic and germline ESTR mutations as adults, and produced offspring with significantly elevated ESTR mutations in their sperm (Barber et al., 2009). In contrast, male mice exposed to radiation during their neonatal days (6 - 8 days old) or pubertal stage (18 - 25 days) did not have increased mutations in adult spermatozoa, as mutant frequencies that were present in spermatogenesis stages immediately after radiation returned to normal levels later in the spermatogenesis process (Xu et al., 2008).
3. Factors such as dose, dose-rate, tissue type and radiation quality can influence mutation rate induction (Hooker et al., 2004; Rydberg et al., 2005; Day et al., 2007; Okudaira et al., 2010; Brooks et al., 2016). 
4. Difference in measurements of mutational frequency can affect the interpretation of the data.

There is evidence of a positive response-response relationship between the radiation dose and the frequency of mutations (Russell et al., 1957; Suzuki & Hei, 1996; Albertini et al., 1997; Kraemer et al., 2000; Canova et al., 2002; Dubrova & Plumb, 2002; J.W. Wilson et al., 2015; Bolsunovsky et al., 2016; Mcmahon et al., 2016; Nagashima et al., 2018) . Most studies found that the response-response relationship was linear (Russell et al., 1957; Albertini et al., 1997; Canova et al., 2002; Dubrova et al., 2002; Nagashima et al., 2018). There were however, two exceptions. In a study using normal human bronchial epithelial cells irradiated with 1 - 6 Gy of gamma-rays, the relationship between the number of induced HPRT mutants and the radiation dose was described as non-linear (Suzuki & Hei, 1996) Similarly, in a study examining HPRT mutations in isolated peripheral blood T-lymphocytes irradiated with low LET gamma-rays, the slope of the line from 0 - 2 Gy differed from the slope at the 2 - 4 Gy interval; thus this was described as two different linear relationships or an overall linear-quadratic relationship (Albertini et al., 1997). In a study with V79 Chinese hamster cells, a curvilinear response was also seen as a result of x-ray response while a linear response was seen for Am-241 alpha-particle exposure (Schmidt and Keifer, 1998).

The time scale relationship between radiation exposure and the frequency of mutations is not well defined. Most studies look for manifestation of mutations days or weeks after irradiation, making it particularly difficult to pinpoint exactly when the mutations first occur. Analysis of various organs from mice after in vivo radiation found that mutations were present at 2 days (Winegar et al., 1994; Masumura et al., 2002) and 3 days (Gossen et al., 1995)(Gossen, 1995) post-exposure. Mutations were still present at 7 days and 14 days (Winegar et al., 1994), and 10 days and 21 days (Gossen, 1995) following irradiation. One study documented a doubling in the number of mutations from 7 to 14 days (Winegar et al., 1994) while the other reported a two-fold decrease from 3 to 21 days  (Gossen et al., 1995).

An attempt to better define this time scale relationship was made in a study using Salmonella typhimurium bacteria. This study was designed to determine how mutation frequency was affected by constant cesium-137 gamma-ray radiation exposure at defined dose rates of 67.8 uGy/hour, 3.2 uGy/hour, and 0.6 uGy/hour; these mutation frequencies were compared to a control group exposed to background radiation levels (0.09 uGy/hour). Mutation frequencies were evaluated after 24, 48, 72 and 96 hours of constant exposure. At 24 hours, the 67.8 uGy/hour, 3.2 uGy/hour and 0.6 uGy/hour mutant frequencies were significantly higher than background exposure controls. Interestingly, however, these levels were decreased at 48 hours and continued to decline gradually towards control frequencies over time. This decline was proposed to be due to an elimination of the highly mutated cells, leaving behind an increasing number of cells that had adapted to the radiation and were thus more equipped for survival (Bolsunovsky et al., 2016). Other studies are required to build a more complete understanding of this timeline.

Not identified.