Paul Finnon

Gene expression following ionising radiation: identification of biomarkers for dose estimation and prediction of individual response.

Purpose: To establish a panel of highly radiation responsive genes suitable for biological dosimetry and to explore inter-individual variation in response to ionising radiation exposure. Materials and methods: Analysis of gene expression in response to radiation was carried out using three independent techniques (Microarray, Multiplex Quantitative Real-Time Polymerase Chain Reaction (MQRT- PCR) and nCounter® Analysis System) in human dividing lymphocytes in culture and peripheral blood leukocytes exposed ex vivo from the same donors. Results: Variations in transcriptional response to exposure to ionising radiation analysed by microarray allowed the identification of genes which can be measured accurately using MQRT PCR and another technique allowing direct count of mRNA copies. We have identified genes which are consistently up-regulated following exposure to 2 or 4 Gy of X-rays at different time points, for all individuals in blood and cultured lymphocytes. Down-regulated genes including cyclins, centromeric and mitotic checkpoint genes, particularly those associated with chromosome instability and cancer could be detected in dividing lymphocytes only. Conclusions: The data provide evidence that there are a number of genes which seem suitable for biological dosimetry using peripheral blood, including sestrin 1 (SESN1), growth arrest and DNA damage inducible 45 alpha (GADD45A), cyclin-dependent kinase inhibitor 1A (CDKN1A), cyclin G1 (CCNG1), ferredoxin reductase (FDXR), p53 up-regulated mediator of apoptosis (BBC3) and Mdm2 p53 binding protein homolog (MDM2). These biomarkers could potentially be used for triage after large-scale radiological incidents and for monitoring radiation exposure during radiotherapy.

Aberrant CDKN1A transcriptional response associates with abnormal sensitivity to radiation treatment

Normal tissue reactions to radiation therapy vary in severity among patients and cannot be accurately predicted, limiting treatment doses. The existence of heritable radiosensitivity syndromes suggests that normal tissue reaction severity is determined, at least in part, by genetic factors and these may be revealed by differences in gene expression. To test this hypothesis, peripheral blood lymphocyte cultures from 22 breast cancer patients with either minimal (11) or very severe acute skin reactions (11) have been used to analyse gene expression. Basal and post-irradiation expression of four radiation-responsive genes (CDKN1A, GADD45A, CCNB1, and BBC3) was determined by quantitative real-time PCR in T-cell cultures established from the two patient groups before radiotherapy. Relative expression levels of BBC3, CCNB1, and GADD45A 2 h following 2 Gy X-rays did not discriminate between groups. However, post-irradiation expression response was significantly reduced for CDKN1A (P<0.002) in severe reactors compared to normal. Prediction of reaction severity of ∼91% of individuals sampled was achieved using this end point. Analysis of TP53 Arg72Pro and CDKN1A Ser31Arg single nucleotide polymorphisms did not show any significant association with reaction sensitivity. Although these results require confirmation and extension, this study demonstrates the possibility of predicting the severity of acute skin radiation toxicity in simple tests.

High and low dose responses of transcriptional biomarkers in ex vivo X-irradiated human blood

Abstract Purpose: Modifications of gene expression following ionizing radiation (IR) exposure of cells in vitro and in vivo are well documented. However, little is known about the dose-responses of transcriptionally responsive genes, especially at low doses. In this study, we investigated these dose-responses and assessed inter-individual variability. Materials and methods: High dose (0.5–4 Gy) and low dose (5–100 mGy) gene expression responses at 2 h and 24 h using 13 biomarkers transcriptionally regulated through the DNA damage response by the tumor suppressor p53 were investigated. Inter-individual variation was also examined. Results: High dose-response curves were best constructed using a polynomial fit while the low dose-response curves used a linear fit with linear R2 values of 0.841–0.985. Individual variation was evident in the high and low dose ranges. The FDXR, DDB2 high dose gene combination produced a mean dose estimate of 0.7 Gy for 1 Gy irradiated ‘unknown’ samples (95% CIs of 0.3–1.1 Gy) and 1.4 Gy for 2 Gy exposure (95% CIs of 0.6–2.1 Gy). The FDXR, DDB2, CCNG1 low dose gene combination estimated 98 mGy (95% CIs of 27–169 mGy) for 100 mGy exposure. Conclusions: These findings identify genes that fulfill some of the requirements of a good exposure biomarker even at low doses, such as sensitivity, reproducibility and simple proportionality with dose.

Correlation of in vitro lymphocyte radiosensitivity and gene expression with late normal tissue reactions following curative radiotherapy for breast cancer

BACKGROUND AND PURPOSE Identification of mechanisms of late normal tissue responses to curative radiotherapy that discriminate individuals with marked or mild responses would aid response prediction. This study aimed to identify differences in gene expression, apoptosis, residual DNA double strand breaks and chromosomal damage after in vitro irradiation of lymphocytes in a series of patients with marked (31 cases) or mild (28 controls) late adverse reaction to adjuvant breast radiotherapy. MATERIALS AND METHODS Gene expression arrays, residual γH2AX, apoptosis, G2 chromosomal radiosensitivity and G0 micronucleus assay were used to compare case and control lymphocyte radiation responses. RESULTS Five hundred and thirty genes were up-regulated and 819 down-regulated by ionising radiation. Irradiated samples were identified with an overall cross-validated error rate of 3.4%. Prediction analyses to classify cases and controls using unirradiated (0Gy), irradiated (4Gy) or radiation response (4-0Gy) expression profiles correctly identified samples with, respectively, 25%, 22% or 18.5% error rates. Significant inter-sample variation was observed for all cellular endpoints but cases and controls could not be distinguished. CONCLUSIONS Variation in lymphocyte radiosensitivity does not necessarily correlate with normal tissue response to radiotherapy. Gene expression analysis can predict of radiation exposure and may in the future help prediction of normal tissue radiosensitivity.

A minimally invasive assay for individual assessment of the ATM/CHEK2/p53 pathway activity

Ionizing radiation induces DNA Double-Strand Breaks (DSBs) which activate the ATM/CHEK2/p53 pathway leading to cell cycle arrest and apoptosis through transcription of genes including CDKN1A (p21) and BBC3 (PUMA). This pathway prevents genomic instability and tumorigenesis as demonstrated in heritable syndromes [e.g. Ataxia Telangiectasia (AT); Li-Fraumeni syndrome (LFS)]. Here, a simple assay based on gene expression in peripheral blood to measure accurately ATM/CHEK2/p53 pathway activity is described. The expression of p21, Puma and Sesn2 was determined in blood from mice with different gene copy numbers of Atm, Trp53 (p53), Chek2 or Arf and in human blood and mitogen stimulated T-lymphocyte (MSTL) cultures from AT, AT carriers, LFS patients, and controls, both before and after ex vivo ionizing irradiation. Mouse Atm/Chek2/p53 activity was highly dependent on the copy number of each gene except Arf. In human MSTL, an AT case, AT carriers and LFS patients showed responses distinct from healthy donors. The relationship between gene copy number and transcriptional induction upon radiation was linear for p21 and Puma and correlated well with cancer incidence in p53 variant mice. This reliable blood test provides an assay to determine ATM/CHEK2/p53 pathway activity and demonstrates the feasibility of assessing the activity of this essential cancer protection pathway in simple assays. These findings may have implications for the individualized prediction of cancer susceptibility.

TPD52 and NFKB1 gene expression levels correlate with G2 chromosomal radiosensitivity in lymphocytes of women with and at risk of hereditary breast cancer.

Purpose: To evaluate a transcriptomic approach to identify healthy women at increased risk of breast cancer due to G2-radiosensitivity and look at transcripts that are differentially expressed between individuals. Materials and methods: We perform the first study to assess the association of G2 radiosensitivity with basal gene expression in cultured T-lymphocytes from 11 women with breast cancer and 12 healthy female relatives using Affymetrix GeneChips. Results: Transcripts associated with radiosensitivity and breast cancer risk were predominantly involved in innate immunity and inflammation, such as interleukins and chemokines. Genes differentially expressed in radiosensitive individuals were more similarly expressed in close family members than in un-related individuals, suggesting heritability of the trait. The expression of tumour protein D52 (TPD52), a gene implicated in cell proliferation, apoptosis, and vesicle trafficking was the most strongly correlated with G2 score while nuclear factor (kappa) – B (NFKB1) was highly inversely correlated with G2 score. NFKB1 is known to be activated by irradiation and its inhibition has been previously shown to increase radiosensitivity. Conclusions: Gene expression analysis of lymphocytes may provide a quantitative measure of radiation response potential and is a promising marker of breast cancer susceptibility.

Quantifying murine bone marrow and blood radiation dose response following 18F-FDG PET with DNA damage biomarkers

The purpose of this study was to quantify the poorly understood radiation doses to murine bone marrow and blood from whole-body fluorine 18 ((18)F)-fluorodeoxyglucose (FDG) positron emission tomography (PET), by using specific biomarkers and comparing with whole body external low dose exposures. Groups of 3-5 mice were randomly assigned to 10 groups, each receiving either a different activity of (18)F-FDG: 0-37MBq or whole body irradiated with corresponding doses of 0-300mGy X-rays. Blood samples were collected at 24h and at 43h for reticulocyte micronucleus assays and QPCR analysis of gene expression in peripheral blood leukocytes. Blood and bone marrow dose estimates were calculated from injected activities of (18)F-FDG and were based on a recommended ICRP model. Doses to the bone marrow corresponding to 33.43mGy and above for internal (18)F-FDG exposure and to 25mGy and above for external X-ray exposure, showed significant increases in radiation-induced MN-RET formation relative to controls (P<0.05). Regression analysis showed that both types of exposure produced a linear response with linear regression analysis giving R(2) of 0.992 and 0.999 for respectively internal and external exposure. No significant difference between the two data sets was found with a P-value of 0.493. In vivo gene expression dose-responses at 24h for Bbc3 and Cdkn1 were similar for (18)F-FDG and X-ray exposures, with significant modifications occurring for doses over 300mGy for Bbc3 and at the lower dose of 150mGy for Cdkn1a. Both leucocyte gene expression and quantification of MN-RET are highly sensitive biomarkers for reliable estimation of the low doses delivered in vivo to, respectively, blood and bone marrow, following (18)F-FDG PET.

Sfpi1/PU.1 mutations in mouse radiation-induced acute myeloid leukaemias affect mRNA and protein abundance and associate with disrupted transcription

Radiation-induced acute myeloid leukaemias (AMLs) in mice are characterised by deletions and point mutations in the Sfpi1/PU.1 transcription factor. Six AML cell lines were used to examine the impact of three previously described R235 point mutations. AML cells carry myeloid and stem cell markers and the R235 mutations differentially affect mRNA and protein abundance. Expression of Sfpi1/PU.1 target genes was deregulated in a broadly similar fashion irrespective of R235 mutation including Flt3, which is frequently subject to activating mutations in human myeloid leukaemias. While R235 mutations differentially affect protein abundance they resulted in similar disruption of Sfpi1/PU.1 functions.

Seeking genetic signature of radiosensitivity - a novel method for data analysis in case of small sample sizes

BackgroundThe identification of polymorphisms and/or genes responsible for an organisms radiosensitivity increases the knowledge about the cell cycle and the mechanism of the phenomena themselves, possibly providing the researchers with a better understanding of the process of carcinogenesis.AimThe aim of the study was to develop a data analysis strategy capable of discovering the genetic background of radiosensitivity in the case of small sample size studies.ResultsAmong many indirect measures of radiosensitivity known, the level of radiation-induced chromosomal aberrations was used in the study. Mathematical modelling allowed the transformation of the yield-time curve of radiation-induced chromosomal aberrations into the exponential curve with limited number of parameters, while Gaussian mixture models applied to the distributions of these parameters provided the criteria for mouse strain classification. A detailed comparative analysis of genotypes between the obtained subpopulations of mice followed by functional validation provided a set of candidate polymorphisms that might be related to radiosensitivity. Among 1857 candidate relevant SNPs, that cluster in 28 genes, eight SNPs were detected nonsynonymous (nsSNP) on protein function. Two of them, rs48840878 (gene Msh3) and rs5144199 (gene Cc2d2a), were predicted as having increased probability of a deleterious effect. Additionally, rs48840878 is capable of disordering phosphorylation with 14 PKs. In silico analysis of candidate relevant SNP similarity score distribution among 60 CGD mouse strains allowed for the identification of SEA/GnJ and ZALENDE/EiJ mouse strains (95.26% and 86.53% genetic consistency respectively) as the most similar to radiosensitive subpopulatioConclusionsA complete step-by-step strategy for seeking the genetic signature of radiosensitivity in the case of small sample size studies conducted on mouse models was proposed. It is shown that the strategy, which is a combination of mathematical modelling, statistical analysis and data mining methodology, allows for the discovery of candidate polymorphisms which might be responsible for radiosensitivity phenomena.

Lymphocyte telomere length correlates with in vitro radiosensitivity in breast cancer cases but is not predictive of acute normal tissue reactions to radiotherapy.

Purpose: To examine the hypothesis that lymphocyte telomere length may be predictive of both breast cancer susceptibility and severity of acute reactions to radiotherapy. Materials and methods: Peripheral blood lymphocyte cultures from breast cancer patients (with normal or severe skin reactions to radiotherapy) and normal individuals were assessed for in vitro radiosensitivity as measured by apoptosis, cell cycle delay and cytotoxicity. Telomere lengths were determined by a flow cytometric fluorescence in situ hybridization assay (FLOW-FISH). Results: Female breast cancer cases (n = 24) had reduced lymphocyte telomere lengths by comparison with healthy controls (n = 20, p < 0.04). However, the average age of healthy controls was less (45.4) than cases (53). When the control group was modified to give a better age match (51.5, n = 13) the reduced telomere length in cases was not significantly different from controls. Lymphocytes from breast cancer cases also showed reduced cell cycle delay (p < 0.001) and increased apoptosis (p < 0.01) following irradiation in vitro at 3 and 5 Gy respectively, compared to healthy controls. Statistical significance was maintained with the improved age matching of groups. Comparison of lymphocytes from breast cancer patients with normal (n = 11) and severe (n = 13) skin reactions to radiotherapy failed to identify differences in telomere length or cellular radiosensitivity in this limited sample. Conclusions: This study adds to the evidence suggesting a correlation between altered cellular radiosensitivity and breast cancer. However, in the cases investigated, telomere length does not appear to be predictive of acute skin reactions to radiotherapy.