Use of cytogenetic indicators in radiobiology

Abstract

The study of ionising radiation has systematically relied on cytogenetic indicators to evaluate the biological effects and has led to theoretical approaches to explain observations associated with radiation exposure. In many of the early studies on radiobiology, the induction of chromosomal aberrations was the method of choice to evaluate dose–response relationships. But progressively, this and other cytogenetic biomarkers were used to obtain mechanistic insight on the biological effects induced by radiation. This paper attempts to give a view on the use of cytogenetic indicators in the study of various radiation-related phenomena, including radiation dosimetry, mechanisms involved in the various cellular responses to radiation, such as bystander effects, chromosomal instability and adaptive response, as well as DNA repair pathways. One future direction may involve the use of cytogenetic indicators to evaluate various molecular determinants in individuals' susceptibility to radiation, using other techniques such as fluorescence in situ hybridisation (FISH) and linking them to specific gene functions and single nucleotide polymorphisms.

INTRODUCTION

Biomarkers are used in preventing chemical and radiation-induced health effects, by providing an indication of risk associated with exposure. The complexity of bio-monitoring is underlined by the various methods of assessment and their association with exposure as depicted in Figure 1⁽¹⁾, in which external dose (assessed by environmental monitoring) is linked stepwise to late developing disease, in essence an irreversible effect (assessed by health surveillance). Cytogenetic biomarkers can be classified in this scheme as early biological effects, which also include gene mutations and unscheduled DNA synthesis⁽²⁾.

Figure 1.

Scheme of the levels and methodologies used in the biological monitoring of genotoxicity associated with exposure to ionising radiation and chemicals.

It has been shown in prospective follow-up studies^(3,4) that subjects with elevated levels of structural chromosomal aberrations (CAs) may be at an elevated risk for cancer, whereas no such association was observed with other biomarkers. This places the CAs assay as a reliable indicator of risk. On the other hand, cytogenetic biomarkers as a whole can be helpful in evaluating the in vitro and in vivo mechanisms underlying DNA damage from ionising radiation, namely the involvement of DNA repair. This paper gives a broad view of the use of cytogenetic markers in the study of diverse radiation-related effects (Table 1).

Table 1.

Usefulness of cytogenetic biomarkers in radiobiology.

Application	Biomarkers	Purpose
Radiation dosimetry
Recent exposure Retrospective exposure	DIC; MN Translocations	Evaluation of an exposure to ionising radiation and determination of the absorbed dose
High vs. low-LET radiation exposure	CAs—DIC and ‘Crea’; multi-aberrant cells, Translocations (increased complexity), MN and multi- micronucleated cells, SCEs	Evaluation of the type of exposure by the genotoxicity pattern of the biomarker
Mechanistic studies
Bystander effect Chromosomal instability Adaptive response	CAs; Translocations; MN; SCEs	Study of key radiation-induced phenomena by using cytogenetic biomarkers under different experimental protocols
DNA repair studies
NHEJ and HR repair PARPs Excision repair	CAs; Translocations; MN	Study of key DNA repair pathways by using cytogenetic biomarkers in DNA repair defective cells or in pharmacologically inhibited cells (e.g. DNA repair inhibitors)
Susceptibility studies	MN; CAs; Translocations	Functional characterisation of genetic polymorphisms of genes involved in the individual susceptibility to ionising radiation (e.g. cell cycle control, DNA repair, and apoptosis)

CAs, chromosomal aberrations; Crea, complex rearrangements; DIC, dicentric chromosomes; HR, homologous recombination; LET, linear energy transfer; MN, micronuclei; NHEJ, non-homologous end-joining; PARPs, Poly(ADP-ribose) polymerases; and SCEs: sister-chromatid exchanges.

DNA DAMAGE AND BIOLOGICAL RADIATION DOSIMETRY

Ionising radiation can induce a wide range of DNA lesions, including damage to bases, DNA–DNA and DNA–protein cross-links, DNA single (SSB) and double strand breaks (DSBs). It is generally agreed, however, that the formation of DSBs is the critical radiation-induced damage that leads to chromosomal aberrations such as dicentrics, reciprocal translocations and rings, which involve interaction of DSBs with each other.

One of the critical factors related to the study of biological effects of ionising radiation is the estimation of dose. The majority of such studies use human lymphocytes, which besides their availability are known to be very sensitive to ionising radiation. Biological monitoring of humans exposed to ionising radiation has relied heavily on cytogenetic indicators such as unstable chromosomal aberrations, especially dicentric chromosomes (DIC), which are suitable indicators of a recent exposure to ionising radiation⁽⁵⁾. Stable chromosomal aberrations, namely reciprocal translocations (fluorescence in situ hybridisation (FISH)/chromosome painting), also are an area of growing interest in radiation dosimetry because they can adequately be used to detect a retrospective exposure to ionising radiation. Other cytogenetic biomarkers such as micronuclei (MN), and to a lesser extent, sister-chromatid exchanges (SCEs) in peripheral lymphocytes, have also been used in dosimetry studies as well as in predicting cancer risk.

The use of cytogenetic biomarkers, mentioned above, has been helpful in confirming the exposure to ionising radiation and/or in evaluating the extent of DNA damage in different scenarios, namely after an environmental, occupational, accidental or medical exposure (Table 2). Of particular interest is the follow-up study of cancer patients treated with ionising radiation. Recently, we studied for a period of 24 months, a group of 19 thyroid cancer patients (papillary and follicular carcinoma) after ¹³¹I therapy⁽⁶⁾. Lymphocytes from these patients revealed an increase of the frequencies of the cytogenetic parameters studied (CAs, DIC, MN) demonstrating the existence of a mild but persistent DNA damage pattern. In such biological dosimetry studies, the dose of ionising radiation received can be derived by extrapolation from human data on the cytogenetic biomarker studied, compared with dose–response relationships obtained in vitro.

Table 2.

Examples of the use of cytogenetic biomarkers in different radiation exposure scenarios.

Exposure
Atomic bomb
Long-term atomic bomb survivors from Hiroshima and Nagasaki
Environmental
Domestic radon
Nuclear power plant vicinity
Fall-out from nuclear tests
Occupational
Nuclear power plant workers
Uranium miners
Medical workers
High altitude aircraft crew
Astronauts
Accidental
Chernobyl
Goiania
Medical
External radiotherapy 60Co
131I (e.g. thyroid carcinoma; hyperthyroidism)
Diagnostic X rays
Long-term thorotrast survivors

Dicentric chromosomes and centric rings (centric r) have been extensively validated as highly sensitive biomarkers for recent radiation exposure. The shape of the dose–effect relationship for the induction of such aberrations is linear–quadratic in general, but linear for low doses of ionising radiation. For dicentrics, linearity can be demonstrated down to 20 mGy, but for lower doses, statistical variations mask the effect of radiation⁽⁷⁾. Repair processes can reduce the quadratic component of the curve as the dose rate decreases.

The formation of DSBs can result in acentric DNA fragments that eventually are lost by the cell during division. These acentric fragments can be incorporated in micronuclei in cytokinesis-blocked binucleated cells. Dose–effect relationships of micronuclei are similar, albeit lower than those for CAs, since not all acentric fragments give rise to micronuclei. In human lymphocytes, the persistence of cytogenetic damage over time depends on various factors, including the type of biomarker and the severity of the outcome to the cell, which can induce mitosis-linked cell death and/or apoptosis, or renewal of the lymphocyte population. A more complete review on the usefulness and limits of cytogenetic markers in biological dosimetry is given in the accompanying article by Léonard et al.

HIGH-LET vs. LOW-LET RADIATION

Mechanistic knowledge of DNA and cell damage by high-linear-energy-transfer (LET) radiation, such as heavy ions or alpha-particles, is less extensive compared to that of low-LET radiation. Although alpha-particles are only capable of traversing a fraction of a cell volume, they are very effective in producing a high density of localised lesions⁽⁸⁾. DNA lesions produced by alpha-particles are characterised by clustering, inducing DSBs, which are difficult to repair. Complex chromosome rearrangements, sometimes referred to as ‘Crea’⁽⁹⁾, are thus induced, leading to enhanced cell death. This leads to a greater biological effectiveness per unit dose for high-LET radiation. Comparison of the relative biological effectiveness (RBE) for different types of radiation can provide information on the underlying mechanisms of damage induced by radiation, on assessment of risk associated with accidental, occupational and environmental exposure (e.g. to radon), and can also allow different therapy applications for human disease. One instance of high-LET radiation application is the use of the boron neutron capture (BNC) reaction in the treatment of malignant glioma and melanoma patients, the boron neutron capture therapy (BNCT). This reaction depends on the capture of thermal neutrons by the minor stable isotope of boron, ¹⁰B in boronated compounds, e.g. p-borono-l-phenylalanine (BPA), with the release of alpha and lithium particles. The propagation of the alpha and lithium particles in biological tissues is characterised by a short-range and a high-LET with a remarkable destructive power⁽¹⁰⁾. The exposure of human melanoma cells to BNC reaction revealed a characteristic pattern with the presence of multi-micronucleated cells (MNs)⁽¹¹⁾. In fact, for increasing doses of alpha particles the frequency of micronucleated cells containing one MN per cell decreased, whereas the frequencies of micronucleated melanoma cells with two, and especially three or more MNs per cell notoriously increased. These results are consistent with the formation of multiple damaged sites on the DNA molecule. Similar observations occur with CAs, with an increase of complex rearrangements involving two or more chromosomes as well as in multi-aberrant cells. Long-term biological consequences of exposure to high-LET radiation may be linked to misrepaired lesions or complex insertions after repair, which may occur even at low doses. The pattern of genotoxicity may be used to evaluate the quality of radiation involved, e.g. high- vs. low-LET radiation.

MECHANISTIC STUDIES: BYSTANDER EFFECT, CHROMOSOMAL INSTABILITY AND ADAPTIVE RESPONSE

Cytogenetic biomarkers may also be useful for the knowledge of the mechanisms underlying some intriguing phenomena induced by ionising radiation: the bystander effect, the chromosomal instability and the adaptive response. One interesting issue related to radiobiology is the observation that some effects arising from radiation exposure may be non-targeted, that is, effects arising from non-nuclear or even non-cellular exposure. Nagasawa and Little⁽¹²⁾ showed that in CHO cells irradiated with very low doses of alpha particles, >30% of cells had induced SCEs, even though <1% were traversed by the particles. Subsequent observations using this and other cytogenetic biomarkers (e.g. CAs and MNs) have highlighted a new field in radiation research, the bystander effect, whereby cells not actually irradiated displayed cytogenetic alterations. The mechanisms leading to this phenomenon are not known, but it may be due to some soluble transmissible factors produced by irradiated cells (e.g. cytokines and growth factors), by the generation of reactive oxygen species (ROS) or alternatively by direct cell–cell communications via gap-junctions⁽¹³⁾.

Chromosomal instability induced by ionising radiation is a phenomenon that can occur in the progeny of irradiated cells, sometimes several generations after the initial exposure, and can actually lead to delayed cell death and delayed mutations. The induction of transmissible chromosomal instability has been demonstrated in a number of cell systems, especially using in vitro models, although it has also been observed in a few in vivo experiments⁽¹⁴⁾. The chromosomal aberrations are the standard biomarkers chosen, but some reports have fully described the existence of delayed translocations⁽¹⁵⁾ or delayed micronuclei⁽¹⁶⁾, in irradiated cells. Like the bystander effect, the induction of chromosomal instability is frequently observed after densely ionising radiation exposure. Chromosomal instability is usually regarded as a condition of genomic instability, although some authors demonstrated that it can be mediated by a bystander effect⁽¹³⁾. No clear dose–response has been fully observed for the induction of instability by high-LET radiation, although some evidence shows that it can arise in cells exposed to very low particle fluences⁽¹⁷⁾. In fact, chromosomal instability may even arise in descendants of unirradiated surviving cells after alpha particle irradiation^(12,18). The implications for carcinogenesis are unknown, though it has been known for some time that genomic instability is a common feature of pre-malignant neoplastic and fully malignant cells⁽¹⁹⁾. Cells from different individuals are not equally susceptible to radiation-induced genomic instability, and there is possibly a link to altered DNA repair processes due to irradiation.

The adaptive response is another well-documented effect induced by ionising radiation by which cells/organisms previously treated with low doses of a genotoxicant become less sensitive to a subsequent higher dose of the same or another genotoxicant. The adaptive response has been assessed using cytogenetic biomarkers (CAs, MNs and SCEs), and probably reflects an enhanced process of DNA repair via signal transduction⁽²⁰⁾, although different mechanisms could be involved^(21,22). The persistence of the adaptive response is also a matter of debate. Recently, we evaluated a group of thyroidectomized patients with non-familial thyroid cancer treated with ¹³¹I for the existence of a cytogenetic adaptive response in their circulating lymphocytes after an in vitro challenge towards a testing dose of mitomycin C (MMC). At one month after ¹³¹I therapy, the induction of MN by MMC was markedly lower than before therapy, increasing thereafter, at six months, perhaps reflecting a transient adaptation due to radiation⁽²³⁾. These results strengthen the need for long-term cytogenetic evaluation of the adaptive response.

DNA REPAIR

A further issue related to cytogenetic biomarkers and radiation is the study of DNA repair processes. A wealth of information has been obtained by the study of the human chromosome breakage syndromes, e.g. Fanconi anaemia, Bloom syndrome, Werner syndrome, Nijmegen breakage syndrome and Ataxia telangiectasia. Patients with these diseases exhibit elevated frequencies of spontaneous CAs or SCEs and also hypersensitivity to DNA-damaging agents, including radiation. In addition, it is known for some time that certain cell lines exhibit a marked radiosensitivity, not only evaluated by clonogenic survival but also by using cytogenetic end-points (CAs and MNs). It is presently clear that some of these cell lines (e.g. xrs-5; xrs-6; M059J) are in fact defective in components of the non-homologous end-joining (NHEJ) pathway for the repair of DSBs. This pathway is first mediated by the heterotrimeric enzyme DNA-dependent protein kinase (DNA-PK). NHEJ is activated by DSBs, being a highly conserved process in which two broken DNA ends rejoin without the requirement of extensive homology⁽²⁴⁾. Further information has been accumulated by manipulation of DNA-PK activity by chemical inhibitors. The fungal metabolite wortmannin (WM) is a potent and irreversible inhibitor of the catalytical sub-unit of DNA-PK and has thus being used as a radiosensitiser essentially through cell survival assays. Recently, we compared the effect of WM in the induction of micronuclei by different doses of low-(⁶⁰Co gamma radiation) and high-LET radiation (BNC reaction) in V79 Chinese hamster cells⁽²⁵⁾. The results obtained showed a pronounced effect of WM for low-LET radiation exposure (3- to 4-fold) whereas the sensitising effect for the BNC reaction was about 2-fold. These results, however, do not preclude a role for DNA-PK in high-LET-induced damage repair.

The repair of DSBs may also be mediated by homologous recombination (HR). In contrast to NHEJ, HR requires extensive homology since the damaged chromosome retrieves information from the undamaged one, generally resulting in accurate repair. HR is mediated by the RAD52 group of proteins, which includes RAD50, RAD51, RAD52, RAD54, among others. In this case, various gene products of this pathway were identified in radiosensitising studies, and cellular phenotypes associated with mutations in these genes frequently display increased radiosensitivity with increased CAs⁽²⁶⁾.

A further example of the use of cytogenetic biomarkers in DNA repair studies is the evaluation of the involvement of the poly(ADP-ribose) polymerase (PARP-1) in the repair of radiation-induced lesions. PARP-1 is considered to be a constitutive factor of the DNA damage surveillance network present in eukaryotic cells acting through a DNA break sensor function. The use of PARP-1 inhibitors, especially 3-aminobenzamide, in cytogenetic approaches, has been important to demonstrate the importance of the ADP-rybosilation for the integrity of the chromosomes⁽²⁷⁾. These results have been generally confirmed using knock-out models. In fact, the exposure of PARP-1-deficient cells/animals to gamma radiation, increases recombination observed as an increase in sister-chromatid exchanges and chromosome aberrations^(27–29).

SUSCEPTIBILITY STUDIES

It is clear from the previous sections that the complexity of biological processes involved in radiobiology depends on numerous gene products, the combination of which determines the final outcome of exposure. The molecular and cellular determinants involve induction of DNA repair coupled with temporary growth arrest, with further control mechanisms associated with genotoxic stress response, including the p53 gene product⁽³⁰⁾. These control mechanisms prolong cell cycle arrest at different cell cycle checkpoints or eliminate damaged cells by apoptosis or irreversible growth arrest. The various DNA repair systems involved include, besides the aforementioned HR and NHEJ, also base excision repair (BER). In case of successful repair, the growth arrest is lifted and cells resume proliferation. However cell proliferation without efficient repair can also occur, in which case cell death and mitotic catastrophe arises. The number of genes involved in all these processes is huge, more than 100 genes having been identified for DNA repair alone⁽³¹⁾. Clearly, individual susceptibility to radiation could involve many if not all of the phenomena discussed. The discovery of DNA repair gene polymorphisms (for example in XRCC1, Ape1, XRCC3, Ku70, Ku80, DNAPKcs, Ligase IV, XRCC4, BRCA1, BRCA2) raises the issue of their importance in radiation susceptibility. Until now, results on the association of these polymorphisms with radiosensitivity have been inconsistent, mainly due to the complexity of the processes involved and also because the number of possible allelic variants found associated with radiosensitivity could be fairly high. Further studies on gene function and polymorphisms and their relation with cytogenetic indicators, in particular the FISH-based techniques (whole chromosome painting, translocation analysis) may shed some light on this issue.

We are deeply indebted to Professor G. Gerber for critical reading of the manuscript. This work was supported by the European Commission (EC) Contract No ICA2-CT-2000-10056 and from Fundação da Ciência e Tecnologia (FCT).