International Journal of Radiation Biology

ISSN: 0955-3002 (Print) 1362-3095 (Online) Journal homepage: https://www.tandfonline.com/loi/irab20

Gene expression following ionising radiation:

Identification of biomarkers for dose estimation
and prediction of individual response

Sylwia Kabacik, Alan Mackay, Narinder Tamber, Grainne Manning, Paul

Finnon, Francois Paillier, Alan Ashworth, Simon Bouffler & Christophe Badie

To cite this article: Sylwia Kabacik, Alan Mackay, Narinder Tamber, Grainne Manning, Paul
Finnon, Francois Paillier, Alan Ashworth, Simon Bouffler & Christophe Badie (2011) Gene
expression following ionising radiation: Identification of biomarkers for dose estimation and
prediction of individual response, International Journal of Radiation Biology, 87:2, 115-129, DOI:
10.3109/09553002.2010.519424

To link to this article: https://doi.org/10.3109/09553002.2010.519424

Published online: 10 Nov 2010.

Submit your article to this journal

Article views: 598

View related articles

Citing articles: 84 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=irab20
Int. J. Radiat. Biol., Vol. 87, No. 2, February 2011, pp. 115–129

Gene expression following ionising radiation: Identification

of biomarkers for dose estimation and prediction of individual
response

SYLWIA KABACIK1, ALAN MACKAY2, NARINDER TAMBER2, GRAINNE MANNING1,

PAUL FINNON1, FRANCOIS PAILLIER1, ALAN ASHWORTH2, SIMON BOUFFLER1, &
CHRISTOPHE BADIE1
1
Biological Effects Department, Centre for Radiation, Chemical and Environmental Hazards, Health Protection Agency,
Chilton, Didcot, Oxfordshire, and 2Breakthrough Breast Cancer Research Centre, the Institute of Cancer Research,
London, UK

(Received 18 June 2010; Revised 3 August 2010; Accepted 25 August 2010)

Abstract
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 nCounter1
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.

Keywords: ionising radiation, biodosimetry, individual response, gene expression, biomarkers, peripheral blood lymphocytes

exposed individuals followed by more accurate

Introduction
Ionising radiation (IR) is a ubiquitous environmental
hazard and a well-known carcinogen. In general,
exposures are low but accidents and incidents have
led to significant exposures (Blakely et al. 2005).
The response to any radiation accident or incident
involving actual or potential IR exposure requires
accurate and rapid assessment of the doses received
by individuals. This is required for initial ‘triage’ of
assessment of doses in those exposed to determine
the most appropriate approach to treatment. As in
cases of accidental exposure, physical dosimetry is
rarely available, and biological dosimetry, based on
the scoring of chromosomal damage – specifically
dicentric chromosomes – in blood samples, plays a
key role (International Atomic Energy Agency
[IAEA] 2001). The dicentric assay is accurate and
sensitive; nonetheless, it is a labour-intensive assay

Correspondence: Dr Christophe Badie, Cancer Genetics and Cytogenetics, Radiation Effects Department, Health Protection Agency, Centre for Radiation,
Chemical and Environmental Hazards, Chilton, Didcot, Oxon, OX11 0RQ, UK. Tel: þ44 (0)1235 822698. Fax: þ44 (0)1235 833891.
E-mail: christophe.badie@hpa.org.uk
ISSN 0955-3002 print/ISSN 1362-3095 online Ó 2011 Informa UK, Ltd.
DOI: 10.3109/09553002.2010.519424
116 S. Kabacik et al.
(42 days from sampling for first results), and it has a
limited sample capacity available for mass screening
following an accident or incident. A recent estimation
including the 15 European Union countries where
biological dosimetry is established, gives a total
capacity for dosimetric triage of about 1500 cases per
week analysed with a dicentric assay (Wojcik et al.
2010). Therefore, it is clear that any minimally
invasive methods that could provide individual dose
estimates more rapidly with greater sample through-
put would be of great value in incident management.
Recently, many attempts have been made to
improve dosimetry using different end-points. Dif-
ferent cytogenetic techniques have been proposed, for
example, using premature chromosome condensa-
tion (Lindholm et al. 2010) or analysis of micronuclei
(Willems et al. 2010). Other techniques have shown
some potential, and studies on protein biomarkers
such as histone phosphorylation (g-H2AX) (Garty
et al. 2010), global proteomics approaches (Ossetrova
and Blakely 2009) and changes in gene expression
following IR exposure (Amundson et al. 1999, Fan
et al. 2001) are promising. These studies are based on
the knowledge that exposure to IR induces complex
changes at the level of RNA transcripts (quantity)
as well as proteins (stability, post-transcriptional
changes and quantity).
Over recent years, a greater understanding has
been obtained of the transcriptional response in cells
to radiation exposure. Exposure of cells to ionising
radiation activates multiple signal transduction path-
ways, which result in complex alterations in gene
expression. Expression of specific genes can depend
on radiation dose (Grace et al. 2002, Amundson and
Fornace 2003, Stassen et al. 2003), dose-rate
(Amundson et al. 2003), radiation quality
(Woloschak and Chang-Liu 1995) and lapse between
stress and analysis (Franco et al. 2005), making
gene-expression profiling a potentially powerful and
informative approach for biological dosimetry
(Blakely et al. 2003, Chaudhry 2008). A linear
dose-response relationship has been obtained for
some genes over a dose range of 1–3 Gy (Grace et al.
2002), 0.5–4 Gy (Kang et al. 2003) and down to
doses as low as 2 cGy (Amundson and Fornace
2003). Such studies have utilised human lympho-
blastoid cells (Jen and Cheung 2003), non-cycling
human fibroblasts (Ding et al. 2005), confluent
human keratinocytes (Franco et al. 2005) and
peripheral blood lymphocytes (Kang et al. 2003).
The transcriptional responses to the genotoxic
stresses can be monitored on a global scale using
microarrays (Tusher et al. 2001, Gruel et al. 2006) or
more accurately and sensitively in a few specific
genes by quantitative real time polymerase chain
reaction (QRT-PCR) (Grace et al. 2003). Radiation
responsive genes could potentially be used as
biomarkers of human exposure to radiation after an
accident (Dressman et al. 2007, Chaudhry 2008),
and selected molecular biomarkers are likely to
provide a more robust approach for high-throughput
minimally invasive radiation biodosimetry (Paul and
Amundson 2008).
Gene expression responses in blood lymphocytes
can also be used to understand the genetic basis for
variation in human radiosensitivity (Correa and
Cheung 2004), to characterise tumour subtypes
(e.g., Payton et al. 2009), to assess inter-individual
susceptibility to DNA damaging agents for the
prediction of therapeutic response to drugs (Fry
et al. 2008), to predict clinical outcome in human
cancers (Carter et al. 2006) and to identify in vivo
radiation-induced biomarkers (Amundson et al.
2004). For example, we recently have shown (Badie
et al. 2008) that cyclin-dependent kinase inhibitor
1A (CDKN1A) transcriptional response associates
with abnormal sensitivity to radiation treatment.
Several studies have identified potential biomar-
kers of radiation exposure. But there has been
relatively little effort placed on the identification of
genes that respond consistently in cycling and non-
cycling cells from all donors.
In this study, we compared the gene expression
response to IR in dividing, mitogen stimulated
lymphocytes and blood resting leukocytes exposed
ex vivo from three healthy normal donors. The goals
of this study were to establish a panel of highly
radiation responsive genes suitable for radiation
dosimetry, to compare the gene expression profiles
of cycling cells (stimulated T-lymphocytes) and non
dividing gap 0 (Go) blood leukocytes from the same
donors and to explore inter-individual variation in
response to IR and to compare three methods for
quantifying gene expression.
Materials and methods
Blood collection
Peripheral blood samples were obtained anon-
ymously from three healthy female working indivi-
duals (age range 37–53 years old) with informed
consent and ethical approval from Oxford Research
Ethics Committee. Lymphocyte samples were pur-
ified by separation on Histopaque-1077 (Sigma-
Aldrich, Poole, UK) and frozen in liquid nitrogen
prior to use. These lymphocytes were used to
produce T-cell lines.
Cell growth
T-lymphocyte cultures were prepared using the
method described previously (O’Donovan et al.
1995, Finnon et al. 2008).

Irradiation of lymphocyte cultures and blood
Cultured lymphocytes were disaggregated and
seeded at a density of 4 6 105 cells/ml in GR10
medium. Cells or fresh whole blood were irradiated
at 2 or 4 Gy using a Siemens Stabilipan (Siemens
AG, Munich, Germany) therapy X-ray set (output
14 mA, 250 kV peak, 1.2 mm Cu HVL, 0.7 Gy
min71) at room temperature at the Medical Re-
search Council (MRC), Harwell. Fresh blood and
lymphocyte cultures were maintained at 378C before
and after irradiation. RNA Later (Applied Biosys-
tems/Ambion, Warrington WA3 7QH, UK) was
added 2 or 24 h after irradiation and cells were kept
at 7808C until RNAs were extracted.
RNA purification from lymphocytes and whole
blood samples
Total RNA from human blood leukocytes (white
blood cells) was prepared by using LeukoLOCKTM
Total RNA Isolation Kit (Applied Biosystems/
Ambion, Warrington, Cheshire, UK) (for MQRT-
PCR and microarrays). This kit eliminates plasma,
platelets and red blood cells including reticulocytes.
RiboPureTM-Blood Kit (Applied Biosystems/Am-
bion) was used to extract RNA for Nanostring. Total
RNA from human T-cell lines was prepared by using
RNAqueous1-4PCR (Applied Biosystems/Ambion)
kit. DNA contamination was removed by using
DNA-freeTM (Applied Biosystems/Ambion). RNA
quantity was assessed by Nanodrop ND1000, and
RNA quality was assessed on 1.3% agarose gel. For
multiplex PCR, reverse transcriptase reactions were
performed using High Capacity cDNA Reverse
transcription kit, (Applied Biosystems, Foster City,
CA, USA) according to manufacturer’s protocol
with 700 ng of total RNA.
cDNA microarray hybridisation procedure and
data extraction
Total RNA integrity was confirmed on an Agilent
2100 BioAnalyser (Agilent Technologies, South
Queensferry, West Lothian, UK) before linear T7
amplification using an aminoallyl MessageAmp Kit
(Applied Biosystems/Ambion). Amplified RNA
(1 mg) was coupled with either CY3 or CY5 and
hybridised to Breakthrough 20K cDNA microarrays
in replicate dye swap hybridisations, as previously
described (Mackay et al. 2009). The Breakthrough
20K microarray platform and all primary microarray
data have been submitted to Array Express database
(http://www.ebi.ac.uk/microarray-as/ae/, with acces-
sion number E-TABM-1083). Annotation of the
Breakthrough 20K cDNA microarray was based on
build 216 of the Unigene database.
Gene expression response to ionising radiation 117
Expression values from spots with hybridisation
artefacts or extremely low intensities were flagged in
Genepix 5.1 (Axon Instruments/Molecular Devices
Corporation Sunnyvcale, CA, USA) and then con-
verted to missing values. The raw intensity values
were then converted to log2 ratios of sample to
reference (M values) and log2 average spot intensity
(A values) for all subsequent pre-processing and
analysis.
Data from lymphocyte samples and blood samples
were pre-processed separately. Probes with missing
values in more than 80% of hybridisations within
each cell type were removed from subsequent
analysis. The remainder of missing values were
imputed using a k-nearest neighbour imputation.
The loess local regression function was used to
remove dye biases in log2 ratios across the slide.
After loess normalisation log2 ratios were subjected
to quantile normalisation using the affy package in R.
Final processed data sets represented 10736 probes
for the analysis of lymphocytes and 9698 probes for
the analysis of blood samples.
Differential gene expression was identified using
the significance analysis of microarrays (SAM) in the
same package. Paired 2-class SAM comparisons
were made between irradiated samples and respec-
tive controls in each tissue, at each time point and
each dose. Individual comparisons used an overall
false discovery rate of 5% as a cut-off and were
ranked according to their fold change. A complete
list of differentially expressed genes was made by
concatenating the results of these tests and ranking
the combined gene lists based upon the sum of their
individual ranks in each test.
Multiplex quantitative Real-Time-PCR
(MQ-RT- PCR)
Real-time PCR was performed using iQ5 thermo-
cyclers (Bio-Rad, Hercules, CA, USA). All reactions
were run in triplicate using PerfeCTa1 MultiPlex
qPCR SuperMix (Quanta Biosciences, Inc.
Gaithersburg, MD, USA), primer and probe sets
for target genes at 300 nM concentration each and
2.5 ml of cDNA in 30 ml reaction volume. 30 6-
Carboxyfluorescein (FAM), 6- Hexachlorofluores-
cein (HEX), Texas Red and CY5 (Eurogentec Ltd,
Fawley, Hampshire, UK) were used as fluorochrome
reporters for the hydrolysis probes analysed in
multiplexed reactions. A list of the primers and
probes which we designed is given in Table I.
Cycling parameters were 2 min at 958C, then 45
cycles of 10 s at 958C and 60 s at 608C. Data were
collected and analysed by iQ5 Detection System
software. Gene target Ct (cycle threshold) values
were normalised to a Hypoxanthine-Guanine phos-
phoribosyltransferase 1 (HPRT1) internal control.
118 S. Kabacik et al.

Table I. Oligonucleotide primers and probes used in this study for MQRT-PCR analysis. They were designed by using in house programs
specifically developed to identify multiplex PCR compatible oligonucleotides.

Database acc no.

Genes Genebank/dbSNP PCR primers, Fwd, Rev Probes

HPRT1 NM_000194.2 TCAGGCAGTATAATCCAAAGATGGT CGCAAGCTTGCTGGTGAAAAGGACCC

AGTCTGGCTTATATCCAACACTTCG
CCNG1 NM_004060.3, GGAGCTGCAGTCTCTGTCAAG AACTGCTACACCAGCTGAATGCCC
NM_199246.1
TGACATCTAGACTCCTGTTCCAA
PLK3 NM_004073.2 ATCAGCGCGAGAAGATCCTA CGAGACCTGCAGCACCGCC
GATGTTGTCAGCGTCCTCAA
CDC25C NM_022809, ATGCATCATCAGGACCACAA TGAGTTGCTGAGGTGTCGAAGCCA
NM_001790
GCCAGTGGCTGGAATGTTA
PRC1 NM_199414, AATCTGATGCTACTTCTGGAATC CAATTCAACCAACATCCAGTCCTGAGAA
NM_199413,
NM_003981
AGCCACAGCTGGTTGACTGA
SESN1 NM_014454 GCTGTCTTGTGCATTACTTGTG ACATGTCCCACAACTTTGGTGCTGG
CTGCGCAGCAGTCTACAG
FDXR NM_024417, GTACAACGGGCTTCCTGAGA CGGGCCACGTCCAGAGCCA
NM_004110
CTCAGGTGGGGTCAGTAGGA
ATF3 NM_001030287, AGGTTTGCCATCCAGAACAA CCTCTGCCACCGGATGTCCTCT
NM_001040619,
NM_001674
CTGACAGTGACTGATTCC
MDM2 NM_002392 CCATGATCTACAGGAACTTGGTAGTA CAATCAGCAGGAATCATCGGACTCAG
ACACCTGTTCTCACTCACAGATG
DDB2 NM_000107 GTCACTTCCAGCACCTCACA AGCCTGGCATCCTCGCTACAACC
ACGTCGATCGTCCTCAATTC
BBC3 NM_014417.3, CGGAGACAAGAGGAGCAG CCCTCACCCTGGAGGGTCCTGT
NM_001127240.1
NM_001127241.1
NM_001127242.1
GGAGTCCCATGATGAGATTG
CDKN1A NM_000389.3, GCAGACCAGCATGACAG TTTCTACCACTCCAAACGCCGGCT
NM_078467.1
TAGGGCTTCCTCTTGGA
GADD45A NM_001924.2 CTGCGAGAACGACATCAAC ATCCTGCGCGTCAGCAACCCG
AGCGTCGGTCTCCAAGAG
PHPT1 NM_001135861.1, TCGCTCTCATTCCTGATGTG CTTGTAGCCGCGCACGATCTCCTT
NM_014172.4
TCGTAGATGTCCGCATGGTA

Ct values were converted to transcript quantity using
standard curves obtained by serial dilution of PCR-
amplified DNA fragments of each gene. The linear
dynamic range of the standard curves covering six
orders of magnitude (serial dilution from 3.2 6 1074
to 8.2 6 10710) gave PCR efficiencies between 93%
and 105% for each gene with R2 4 0.998. Relative
gene expression levels after irradiation were similarly
obtained for comparison with pre-irradiation controls.
Nanostring method
The nCounter Analysis System uses a novel digital
technology (NanoString Technologies1, Inc., Seat-
tle, WA, USA) that is based on direct multiplexed
measurement of gene expression (Geiss et al. 2008).
The technology uses molecular barcodes and single
molecule imaging to detect and count hundreds of
unique transcripts in a single reaction. The nCounter
Reporter Probes allow for direct measurement of
individual RNA molecules and provides a digital
signal.
Briefly, target molecules are detected by hybridisa-
tion to Capture and Reporter Probes, each approxi-
mately 50 nucleotides in length and targetting a
contiguous 100-base target region. Target regions of
each mRNA are screened to eliminate direct and
inverted repeat elements and evaluated for cross-
hybridisation against the human RefSeq database.
Potential 50-base probes are then selected for
melting temperatures (Tm) between 78–838C. Cap-
ture and Reporter Probes are ligated to the synthetic

DNA backbones containing the barcode as described
(Geiss et al. 2008), and supplied in one tube
containing all Capture probes, and one tube contain-
ing all Reporter probes.
Hybridisations are set up in four pipetting steps as
follows: For each sample, 100 ng of purified RNA or
lysate in 5 ml total is added, 10 ml hybridisation buffer,
10 ml supplied Reporter Probe mixture, and lastly 5 ml
of supplied Capture Probe mixture in separate thin-
walled PCR strip tubes. The tubes are then covered
and incubated at 658C for 12–18 h in a thermocycler
with heated lid. After hybridisation, the samples are
processed in the PrepStation and counted in the
DigitalAnalyzer. The number of counts of each gene
in the CodeSet is analysed by Microsoft Excel
software. Nanostring assays were performed by
Nanostring Technologies1, Inc., Seattle.
Results
Three independent experiments have been carried
out on the blood leukocyte population as well as
cultured T-lymphocytes from the same three donors.
Blood and T-lymphocytes were irradiated using
three doses (0, 2 and 4 Gy of X-rays) and two time
points (2 and 24 h post-irradiation). Blood leuko-
cytes trapped on the filters (LeukoLOCKTM) and
T-lymphocytes were kept 7808C thus allowing RNA
from all samples to be extracted and analysed
simultaneously. Freezing the cells does not interfere
with the gene expression results (data not shown)
and allows retrospective analysis of large batches of
samples.
Altogether 108 samples have been analysed by
DNA array (216 arrays) and MQRT-PCR. Micro-
array analyses identified the genes significantly up
and down-regulated in response to irradiation in the
three experiments in blood and lymphocytes. The
number of significantly responsive genes for each
time and dose are presented in Table II. Overall,
combining the data, we detected 570 genes up-
regulated and 31 probes down-regulated in blood
samples as compared with 232 probes up-regulated
and 1357 probes down-regulated in lymphocyte
cultures using a false discovery rate of 55%.
Table II. Radiation responsive genes (up and down-regulated) in
blood and cultured lymphocytes in response to radiation. Genes
were identified in paired SAM analyses using a false discovery rate
of 5%.
Blood Lymphocytes
2h 24 h 2h 24 h
2 Gy 4 Gy 2 Gy 4 Gy 2 Gy 4 Gy 2 Gy
up 296 234 13 247 31 100 107
down 0 1 7 29 29 267 113
Gene expression response to ionising radiation 119
The list of the top 20 genes up and down-regulated
in cultured lymphocytes and blood in response to
irradiation is presented in Table III, and the
corresponding heatmaps of microarray data are
shown in Figure 1.
The number of genes seems to be relatively similar
at 2 and 24 h with more up-regulated genes in blood
whereas down-regulated genes are mostly observed
in cultured lymphocytes. Some genes are specifically
up-regulated in cultured lymphocytes Table IIIA or
Blood Table IIIC; however, among the genes
significantly up-regulated in blood and cultured
lymphocytes, nine genes are consistently found in
blood and lymphocytes in the top 20 list for at least
one dose at one time point in all three repeat
experiments, (Table III): sestrin 1 (SESN1), growth
arrest and DNA damage inducible 45 alpha (GAD-
D45A), cyclin G1 (CCNG1), RNA binding motif
protein 15 (RBM15), CDKN1A, TNF receptor
superfamily, member 6 (FAS), F-box and WD repeat
domain containing two (FBXW2), protein phospha-
tise, Mg2þ/Mn2þ dependent, 1D (PPM1D) and
biogenesis of lyosomal organelles complex-1, subunit
2 (BLOC1S2), with the first four being present at all
time point for all doses.
Other genes are specifically very good markers for
blood such as PCNA (proliferating-cell nuclear
antigen) which is involved in DNA replication, repair
of DNA damage, chromatin structure maintenance,
chromosome segregation and cell-cycle progression
(Stoimenov and Helleday 2009) or for dividing
lymphocytes such as ATF3 (Activating transcription
factor 3), a transcription factor that plays a regulatory
role in inflammation, cell division apoptosis and
DNA damage (Fan et al. 2002). Both are signifi-
cantly up-regulated at all doses and time points in all
donors as shown on the heatmaps presented in
Figure 1.
There are a large number of genes down-regulated
in irradiated dividing lymphocytes (Tables II and
IIIB) unlike blood leukocytes where there are very
few genes consistently and significantly down-regu-
lated (Table II and Table IIID), especially, at the
earlier time point (2 h). This is not surprising as
these down-regulated genes are mainly associated
with cell cycle and are only responsive in cycling
lymphocytes in culture (as budding uninhibited by
benzimidazoles 1 homolog [BUB1], cyclin A2
[CCNA2], kinesin family member 23 [KIF23],
protein regulator of cytokinesis 1 [PRC1] or cen-
tromere protein E, 312 kDa [CENPE]).
MQRT-PCR assays have been used to confirm the
results obtained by microarray analysis in individual
4 Gy donors for three up-regulated genes (SESN1, GAD-
D45A and CDKN1A) as well as one down-regulated
116
1109
gene (PRC1) (Figure 2). Overall, all the genes
identified by microarray were confirmed by the other
120 S. Kabacik et al.

Table III. Top 20 most up-regulated genes in cultured lymphocytes (A) and peripheral whole blood (C) and most down-regulated in
cultured lymphocytes (B) and peripheral whole blood (D) in response to irradiation. The numbers indicate the ranking of genes based on
fold change (1 being the most responsive gene at a particular dose and time-point). False discovery rate of less than 5%. Some of the genes
did not pass this level of significance for a specific dose or time point (NS: not significant).

Symbol Description Unigene entry 2 h 2 Gy 2 h 4 Gy 24 h 2 Gy 24 h 4 Gy

(A) Most significantly up-regulated genes in irradiated cultured lymphocytes.

GADD45A Growth arrest and DNA-damage-inducible, alpha Hs.80409 1 1 1 1
SESN1 Sestrin 1 Hs.591336 3 5 3 3
CDKN1A Cyclin-dependent kinase inhibitor 1A (p21, Cip1) Hs.370771 2 3 7 5
ATF3 Activating transcription factor 3 Hs.460 5 6 13 4
FAS Fas (TNF receptor superfamily, member 6) Hs.244139 6 8 4 11
FBXW2 F-box and WD repeat domain containing 2 Hs.494985 7 14 6 6
RBM15 RNA binding motif protein 15 Hs.435947 4 4 42 8
ANKRA2 Ankyrin repeat, family A (RFXANK-like), 2 Hs.239154 23 33 31 9
PPM1D Protein phosphatase 1D magnesium-dependent, Hs.591184 11 20 60 35
delta isoform
C12orf5 Chromosome 12 open reading frame 5 Hs.504545 12 32 48 43
RRM2B Ribonucleotide reductase M2 B (TP53 inducible) Hs.512592 22 54 45 16
C11orf24 Chromosome 11 open reading frame 24 Hs.303025 19 47 85 29
IFNG Interferon, gamma Hs.856 NS 9 2 2
FBXO22 F-box protein 22 Hs.591115 28 69 94 80
CCNG1 Cyclin G1 Hs.79101 10 23 10 NS
BLOC1S2 Biogenesis of lysosome-related organelles Hs.576605 NS 15 12 18
complex-1, subunit 2
PRDM1 PR domain containing 1, with ZNF domain Hs.436023 NS 19 21 15
TncRNA Trophoblast-derived non-coding RNA Hs.523789 14 39 NS 7
RPS27L Ribosomal protein S27-like Hs.108957 NS 44 5 20
LRRC58 Leucine rich repeat containing 58 Hs.518084 NS 36 23 14
(B) Most significantly down-regulated genes in irradiated cultured lymphocytes.
BUB1 BUB1 budding uninhibited by benzimidazoles Hs.469649 8 19 NS 242
1 homolog (yeast)
CCNA2 Cyclin A2 Hs.58974 22 22 NS 474
KIF23 Kinesin family member 23 Hs.270845 6 3 NS 686
NCAPG Non-SMC condensin I complex, subunit G Hs.567567 20 79 NS 720
PPP1R3D Protein phosphatase 1, regulatory (inhibitor) Hs.42215 26 34 NS 771
subunit 3D
PRC1 Protein regulator of cytokinesis 1 Hs.567385 17 26 NS 826
CENPE Centromere protein E, 312kDa Hs.75573 5 11 NS 913
LDHA Lactate dehydrogenase A Hs.2795 NS NS 1 1
HMMR Hyaluronan-mediated motility receptor Hs.72550 1 2 NS NS
(RHAMM)
DKFZp762E1312 Hypothetical protein DKFZp762E1312 Hs.532968 2 1 NS NS
DLG7 Discs, large homolog 7 (Drosophila) Hs.77695 3 5 NS NS
CBLN2 Cerebellin 2 precursor Hs.569851 NS NS 8 2
PTGES3 Prostaglandin E synthase 3 (cytosolic) Hs.50425 NS NS 6 5
INSIG1 Insulin induced gene 1 Hs.520819 NS NS 3 8
BNIP3 BCL2/adenovirus E1B 19kDa Hs.144873 NS NS 2 9
interacting protein 3
PIF1 PIF1 50 -to-30 DNA helicase homolog Hs.112160 4 7 NS NS
(S. cerevisiae)
PTMA Prothymosin, alpha (gene sequence 28) Hs.459927 NS NS 13 4
PSRC1 Proline/serine-rich coiled-coil 1 Hs.405925 9 8 NS NS
RRM2 Ribonucleotide reductase M2 polypeptide Hs.226390 NS NS 7 11
KIF14 Kinesin family member 14 Hs.3104 7 18 NS NS
(C) Most significantly up-regulated genes in irradiated whole blood.
PCNA Proliferating cell nuclear antigen Hs.147433 1 1 1 1
DUSP21 Dual specificity phosphatase 21 Hs.534478 5 6 11 7
SESN1 Sestrin 1 Hs.591336 6 5 22 11
GADD45A Growth arrest and DNA-damage-inducible, alpha Hs.80409 2 3 38 15
TMEM30A Transmembrane protein 30A Hs.108530 4 4 40 18
CCNG1 Cyclin G1 Hs.79101 3 2 62 26
RBM15 RNA binding motif protein 15 Hs.435947 108 13 13 10
RPS27L Ribosomal protein S27-like Hs.108957 7 NS 3 4

(continued)
 Gene expression response to ionising radiation 121

Table III. (Continued).

Symbol Description Unigene entry 2 h 2 Gy 2 h 4 Gy 24 h 2 Gy 24 h 4 Gy

CDKN1A Cyclin-dependent kinase inhibitor 1A (p21, Cip1) Hs.370771 39 NS 6 3

IER5 Immediate early response 5 Hs.15725 50 NS 12 14
FAS Fas (TNF receptor superfamily, member 6) Hs.244139 32 NS 23 24
HSP90AA1 Heat shock protein 90kDa alpha (cytosolic), Hs.525600 73 NS 17 12
class A member 1
FBXW2 F-box and WD repeat domain containing 2 Hs.494985 90 NS 49 30
ARL6IP1 ADP-ribosylation factor-like 6 interacting protein 1 Hs.634882 104 NS 64 105
PERP PERP, TP53 apoptosis effector Hs.520421 56 NS 105 119
UBC Ubiquitin C Hs.520348 86 NS 134 99
BLOC1S2 Biogenesis of lysosome-related organelles Hs.576605 64 NS 169 90
complex-1, subunit 2
ZFR Zinc finger RNA binding protein Hs.435231 40 NS 179 156
PPM1D Protein phosphatase 1D magnesium-dependent, Hs.591184 48 NS 185 153
delta isoform
STX11 Syntaxin 11 Hs.118958 215 NS 72 150
(D) Most significantly down-regulated genes in irradiated whole blood.
BANK1 B-cell scaffold protein with ankyrin repeats 1 Hs.480400 NS NS 1 2
IGJ Immunoglobulin J polypeptide Hs.651109 NS NS 2 4
IL2RB Interleukin 2 receptor, beta Hs.474787 NS NS 3 3
EDG1 Endothelial differentiation, sphingolipid Hs.154210 NS NS 5 13
G-protein-coupled receptor, 1
GLCCI1 Glucocorticoid induced transcript 1 Hs.131673 NS NS 7 24
CD8A CD8a molecule Hs.85258 NS NS 6 26
KLRF1 Killer cell lectin-like receptor subfamily F, member 1 Hs.183125 NS NS NS 1
TCL1A T-cell leukemia/lymphoma 1A Hs.2484 NS 1 NS NS
GZMK Granzyme K (granzyme 3; tryptase II) Hs.277937 NS NS 4 NS
SLC2A3 Solute carrier family 2, member 3 Hs.419240 NS NS NS 5
TRA@ T cell receptor alpha locus Hs.74647 NS NS NS 6
IGHG1 Immunoglobulin heavy constant gamma 1 (G1m marker) Hs.510635 NS NS NS 7
GFRA3 GDNF family receptor alpha 3 Hs.58042 NS NS NS 8
FGFBP2 Fibroblast growth factor binding protein 2 Hs.98785 NS NS NS 9
FOXRED1 FAD-dependent oxidoreductase domain containing 1 Hs.317190 NS NS NS 10
INSIG1 Insulin induced gene 1 Hs.520819 NS NS NS 11
SH3BP5 SH3-domain binding protein 5 (BTK-associated) Hs.654642 NS NS NS 12
PBEF1 Pre-B-cell colony enhancing factor 1 Hs.489615 NS NS NS 14
ZNF182 Zinc finger protein 182 Hs.189690 NS NS NS 15
FOLH1 Folate hydrolase (prostate-specific membrane antigen) 1 Hs.654487 NS NS NS 16

NS: Not Significant (false discovery rate higher than 5%).

assays (i.e., MQRT-PCR and the Nanostring tech-
nology); nevertheless, some promising candidate
biomarkers were not identified in the microarray data.
Ferredoxin reductase (FDXR) for example is not
present on the array and could not therefore be
detected. For p53 up-regulated mediator of apoptosis
(BBC3), also known as PUMA, and Mdm2 p53
binding protein homolog (MDM2), the primers and
probes used for quantitative PCR were designed to
cover most if not all the alternative transcripts. Some
alternative transcripts might be more radiation-
responsive than others and it is possible that the
oligonucleotides on the array were not detecting
the radiation-responsive ones, thus missing the up-
regulation in response to IR revealed by MQRT-PCR.
MQRT-PCR confirmed that these three up-
regulated genes which are significantly responsive
in all individuals, at all time points (2–24 h) at
different doses (2 and 4 Gy) in blood as well as in
stimulated lymphocytes from the same donors.
CDKN1A and GADD45A are known to be good
biomarkers of radiation exposure (Jen and Cheung
2005) but they are also involved in signal transduc-
tion pathways important for genome stability.
SESN1 is an anti-oxidant gene activated through
p53 transcriptional programme as are CDKN1A and
GADD45A (Budanov and Karin 2008).
PRC1 encodes a regulator of cytokinesis which
contributes to the correct formation of the spindle
during the metaphase (Li et al. 2004). It is responsive
to IR in cultured lymphocytes where cells are
blocked in their cell-cycle progression and modestly
up-regulated at 2 Gy/24 h (as detected by MQRT-
PCR) as surviving cells override the block and
progress through mitosis. PRC1 is a good marker
of radiation exposure in dividing lymphocytes but is
not radiation-responsive in blood. There are no
genes which are down-regulated in all samples in
122 S. Kabacik et al.
Figure 1. Heat maps of microarray data including some of the most radiation responsive genes are shown in cultured lymphocytes (left
panels) and blood (right panels). A colour bar showing the level of expression is shown in the bottom right of the Figure.
blood and lymphocytes. It should be noted that other
genes like FDXR, MDM2 (Figure 3) and damage-
specific DNA binding protein 2, 48 kDa (DDB2)
(data not shown) although not detected by micro-
array analysis (or not represented on the arrays) are
also good candidates for biodosimetry; they showed a
consistent up-regulation between 2 and 24 h after
exposure to 2 and 4 Gy. FDXR is particularly
promising as the fold of change is about 15-fold at
24 h and can be easily detectable even at lower doses
and later time points (data not shown).
We have compared the three techniques used in
this study, and the data for four radiation-responsive
genes (CDKN1A, GADD45A, PHPT1, CCNG1)
are presented in Figure 4. There is a good agreement
between the three methods in that the up-regulation
of expression in blood and lymphocytes is detected
by all three techniques. The nCounter Analysis
System and particularly the MQRT-PCR are more
accurate for detecting differences between doses and/
or time points and both of these methods appear to
have a much superior ‘dynamic range’ over which
they can detect changes. It is important to note that
the Nanostring method avoids possible bias due to
enzymatic amplification of RNA prior to analysis.
These methods enable highly sensitive detection and
quantification of the expression of specific gene. For
example, BBC3 and PLK3 (polo-like Kinase 3)
genes, are up-regulated and were both identified as
radiation responsive in all samples by MQRT-PCR
(Figure 5) but not by microarray. The data presented
for four genes, suggest that, the contribution of
genetic factors to gene expression response to IR
exposure is strong enough to affect the level of
response; the highest transcriptional response in
blood appear to be also the highest in cultured
lymphocytes. Donor two for example (middle
sample), is consistently showing a stronger up-
regulation in blood and T-lymphocytes for the four
genes presented; in other words, when comparing
 Gene expression response to ionising radiation 123

Figure 2. Gene expression changes in four key index genes. Log2 ratios of each gene are plotted for each donor after 2 and 24 h in response
to either 2 or 4 Gy of X-rays. CDKN1A (panel A), GADD45A (panel B), PRC1 (panel C) and SESN1 (panel D) gene expression changes
obtained by Microarray and MQRT-PCR in cultured lymphocytes (upper panels) and blood (lower panels) for the three donors (1, 2 and 3).
Microarray: Gene expression changes in four key index genes. Log2 ratios of each gene are plotted for each donor after 2 and 24 h in
response to either 2 Gy or 4 Gy X-ray irradiation. Data are presented for one of the three replicate experiments. MQRT-PCR: Gene
expression changes in the same four key index genes. Fold changes in expression compared to non-irradiated cells (and relative to HPRT
gene) of each gene are plotted for each donor after 2 and 24 h in response to either 2 or 4 Gy irradiation. The Figures represent the ratio of
level of expression after irradiation divided by the non-irradiated cell expression levels. The mean of three independent experiments with
triplicate reactions are presented. Values were normalised using HPRT as a standard. Data for donor 1 are presented in blue, in green for
donor 2 and in red for donor 3. Error bars indicate the standard error of the mean (SEM) for n ¼ 3 independent experiments.
124 S. Kabacik et al.

Figure 2. (Continued).

individuals, the up-regulation level in cultured cells previous findings demonstrating that genetic factors
is representative of the level of up-regulation in blood influence the differences in individual transcriptional
(although not true for all genes, e.g., CDKN1A and response as shown in heterozygous carriers of
SESN1 in Figure 1). This is in good agreement with Nijmegen breakage syndrome (Cheung and Ewens

Figure 3. Up-regulation in cultured lymphocytes and blood of two
genes MDM2 (panel A) and FDXR (panel B) not previously
identified by microarray. Fold changes in expression assessed by
MQRT-PCR compared to non-irradiated cells (and relative to
HPRT gene) of each gene are plotted for each donor after 2 and
24 h in response to either 2 or 4 Gy irradiation. The Figures
represent the ratio of level of expression after irradiation divided by
the un-irradiated cell expression levels. The mean of three
independent experiments with triplicate reactions are presented.
Values were normalised using HPRT as a standard. Error bars
indicate the standard error of the mean (SEM) for n ¼ 3
independent experiments.
2006) and Ataxia telangiectasia (Watts et al. 2002,
Smirnov and Cheung 2008).
Discussion
We have used whole genome microarray expression
profiling to identify genes which could potentially help
to estimate quantitatively radiation exposures across a
dose range relevant for medical decision making in a
radiologic emergency. MQRT- PCR was performed
to confirm the array results and increase the sensitivity
to evaluate inter-individual differences. We also tested
a recently developed high throughput digital technol-
ogy (NanoString nCounter) (Geiss et al. 2008) on
selected genes to assess its potential role in quantita-
tive radiation dosimetry.
Exposure to IR leads to complex cellular responses
that include changes in gene expression, and these
gene expression responses can differ between in-
dividuals. Nevertheless, in the present study, we
Gene expression response to ionising radiation 125
monitored the expression level of numerous genes
amongst which some may be suitable for biological
dosimetry.
The following genes have been identified either by
microarray analysis or by MQRT-PCR: SESN1,
GADD45A, CDKN1A, CCNG1, FDXR, BBC3
DDB2, PHPT1, MDM2, PCNA, PLK3 as well as
FAS, FAS and v-mic v-myc myelocytomatosis viral
oncogene homolog (c-MYC) (data not shown) as
being up-regulated in blood in the three donors
analysed. For most of these genes radiation re-
sponses are long lasting with CDKN1A, SESN1 and
GADD45A still being up-regulated at 48 h post-
exposure (2 Gy) (data not shown). It will be very
important to demonstrate that the up-regulation of
the identified biomarkers last in time (72 h or even
longer). These experiments will have to be carried
out on blood exposed in vivo as in our experience
RNA extracted from blood after more than 48 h
tends to show some degradation. Our data confirm
previous work (Meadows et al. 2008) which showed
that gene expression response to radiation is specific,
can be long lasting and is quantitatively accurate. It is
probable that these methods can be refined in the
near future by for example, sorting of lymphocytes
sub-populations which have a more pronounced
transcriptional response to radiation. For instance,
it has been shown that several biological responses in
cluster of differentiation (CD) 4þ cells could be more
sensitive to low doses of radiation than CD56þ and
CD8þ (Gruel et al. 2008) although these refinements
may prove to be time-consuming and not suitable for
mass screening. Other tissues sustain long term
changes in gene expression after in vivo exposure to
IR. Mouse liver maintains a high level of CDKN1A
following a sublethal dose of 6 Gy for at least 10
weeks (Pawlik et al. 2009).
In order to study the impact of IR on changes in
gene expression in cycling cells, we used a protocol
where leukocytes are stimulated to obtain extended-
term cultures of human T-lymphocytes exclusively
(O’Donovan et al. 1995); the gene expression
profiles are specific to dividing T-lymphocytes. On
the other hand, resting blood leukocytes are con-
stituted of different cell-types such as monocytes and
neutrophils. Each sub-population has a specific
expression pattern and some of the differences in
gene expression detected here are cell-type depen-
dant. Nevertheless, many of the genes which expres-
sion is significantly down-regulated by IR-exposure
in cycling T-lymphocytes are associated with the cell-
cycle, not to a specific cell type. Genes like BUB1,
CCNA2, PRC1, CENPE (Table IIB) and others,
like cell division cycle 25 homolog C (CDC25C),
cyclin B1 (CCNB1), mitotic arrest deficient-2
(MAD2L1) are down-regulated and there repression
was confirmed by MQRT-PCR. Although they are
126 S. Kabacik et al.
Figure 4. Comparison of data obtained from the three different techniques (MQRT-PCR, microarray and NanoString nCounter). The final
ratios, expressed as fold change for the three methods, were generated by comparing expression levels in irradiated samples with basal levels
in non-irradiated samples. The mean of three independent experiments (microarray) and of triplicate reactions (MQRT-PCR and
NanoString nCounter) are presented for donor 1.
poor candidates for biological dosimetry purposes as
they are not responsive in blood, the modification in
transcription level in response to IR of genes
associated with chromosome instability and aneu-
ploidy is extremely interesting. Genes like PRC1
which associates with mitotic spindles, BUB1 master
regulator of spindle assembly checkpoint (Baker
et al. 2009, Marchetti and Venkatachalam 2010)
and MAD2L1, haplo-insufficiency of which causes
premature anaphase and chromosome instability in
mammalian cells (Michel et al. 2001) are all down-
regulated in cycling lymphocytes in response to IR.
The assessment of expression level of genes asso-
ciated with the spindle assembly checkpoint and
proper chromosome segregation can have important
consequences for tumorigenesis and developmental
abnormalities studies. A signature of chromosomal
instability obtained from gene expression profiles has
been identified recently which predicts clinical out-
come in multiple human cancers (Carter et al. 2006).
Remarkably, 11 out of the 25 genes which form their
classifier are significantly down-regulated in our
irradiated cultured lymphocyte samples.
There is concern surrounding the increasing use of
IR in medical diagnostics and the associated risk of
radiation-induced cancer. The biomarkers identified
in this study can potentially be used to identify
individuals who have suffered accidental exposure.
The existence of heritable radiosensitivity syn-
dromes (e.g., Ataxia telangiectasia [AT]) 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. This is
also suggested by the results shown in Figure 5. Lack
of or low expression in response to IR exposure could
also help to identify individuals who are predisposed
to negative outcome of exposure. For example, we
have previously shown that post-irradiation expres-
sion response of CDKN1A in peripheral blood was
significantly reduced in patients with severe acute
skin reactions (Badie et al. 2008). A clear differential
transcriptional response was also demonstrated be-
tween healthy donors and individuals with a defined
genetic defect like AT (Badie et al. 2008).
To summarise, radiation-responsive genes show-
ing relatively little intra-individual variation in
irradiated blood are the best candidates for dose
estimation. The main purpose of this manuscript was
to establish a panel of highly radiation responsive
genes suitable for biological dosimetry and this has

Figure 5. Individual transcriptional responses to radiation exposure 24 h after 4 Gy in four radiation-responsive genes, PLK3, BBC3,
GADD45A and ATF3 assessed by MQRT-PCR. Error bars indicate the standard error of the mean (SEM) for n ¼ 3 independent
experiments.
been achieved. As the ultimate aim is to be able to
estimate the radiation dose, we started additional
work to study the low dose sensitivity of these genes
and to produce full dose-responses and time-courses
in blood exposed ex vivo and in vivo in mice.
Some others may be particularly good as suscept-
ibility biomarkers, for example the down-regulated
genes in cultured lymphocytes.
Conclusion
We have used three different techniques available to
study gene expression in response to IR exposure in
human blood and cultured lymphocytes from three
healthy donors. Changes in gene expression assessed
by DNA microarrays provided a list of radiation
responsive genes for which the variations in tran-
scriptional response can be validated and measured
more accurately and with greater sensitivity by
MQRT-PCR. The results obtained were validated
by a third independent technique scoring mRNA
copies directly.
Array analysis is very useful for identification of
responsive transcripts but the MQRT-PCR and
Nanostring methods have distinct advantages for
quantitative analysis sensitive for the detection of
Gene expression response to ionising radiation 127
patterns of gene expression in association with
radiation exposure. The ability of a single gene set
to predict radiation exposure throughout a window
of time is an important advance in the development
of gene expression for practical biological dosimetry
(Paul and Amundson 2008, Brengues et al. 2010).
Some of the genes identified in this study are
involved in signal transduction networks which
could well be also relevant to those of radiation
induced cancer. They provide research tools to
examine in vivo dosimetry, but also therapeutic
response and mechanisms of radiation tumorigen-
esis. Confirmation and extension of these results
in vivo with animal models or radiotherapy patient
samples may enable the development of simple
clinical tests to predict the likely level of radiation
toxicity and to individualise patient treatment as
well as provide robust tools for biological dosime-
try.
Acknowledgements
We thank Richard Doull (MRC Harwell) for
irradiations. This work was supported in part by
Cancer Research UK (C181/A6976) and Break-
through Breast Cancer.
128 S. Kabacik et al.
Declaration of interest: The authors report no
conflicts of interest. The authors alone are respon-
sible for the content and writing of the paper.
References
Amundson SA, Do KT, Fornace AJ Jr. 1999. Induction of stress
genes by low doses of gamma rays. Radiation Research
152:225–231.
Amundson SA, Fornace AJ Jr. 2003. Monitoring human radiation
exposure by gene expression profiling: Possibilities and pitfalls.
Health Physics 85:36–42.
Amundson SA, Lee RA, Koch-Paiz CA, Bittner ML, Meltzer P,
Trent JM, Fornace AJ Jr. 2003. Differential responses of stress
genes to low dose-rate gamma irradiation. Molecular Cancer
Research 6:445–452.
Amundson SA, Grace MB, McLeland CB, Epperly MW, Yeager
A, Zhan Q, Greenberger JS, Fornace AJ Jr. 2004. Human
in vivo radiation-induced biomarkers: Gene expression changes
in radiotherapy patients. Cancer Research 64:6368–6371.
Badie C, Dziwura S, Raffy C, Tsigani T, Alsbeih G, Moody J, Finnon
P, Levine E, Scott D, Bouffler S. 2008. Aberrant CDKN1A
transcriptional response associates with abnormal sensitivity to
radiation treatment. British Journal of Cancer 98:1845–1851.
Baker DJ, Jin F, Jeganathan KB, van Deursen JM. 2009. Whole
chromosome instability caused by Bub1 insufficiency drives
tumorigenesis through tumor suppressor gene loss of hetero-
zygosity. Cancer Cell 16:475–486.
Blakely WF, Miller AC, Grace MB, McLeland CB, Luo L,
Muderhwa JM, Miner VL, Prasanna PG. 2003. Radiation
biodosimetry: Applications for spaceflight. Advanced Space
Research 31:1487–1493.
Blakely WF, Salter CA, Prasanna PG, 2005. Early-response
biological dosimetry – recommended countermeasure en-
hancements for mass-casualty radiological incidents and
terrorism. Health Physics 9:494–504.
Brengues M, Paap B, Bittner M, Amundson S, Seligmann B, Korn
R, Lenigk R, Zenhausern F. 2010. Biodosimetry on small blood
volume using gene expression assay. Health Physics F8:179–185.
Budanov AV, Karin M. 2008. p53 target genes sestrin1 and
sestrin2 connect genotoxic stress and mTOR signaling. Cell
134:451–460.
Carter SL, Eklund AC, Kohane IS, Harris LN, Szallasi Z. 2006. A
signature of chromosomal instability inferred from gene
expression profiles predicts clinical outcome in multiple
human cancers. Nature Genetics 8:1043–1048.
Chaudhry MA. 2008. Biomarkers for human radiation exposure.
Journal of Biomedical Science Sep;15(5):557–563.
Cheung VG, Ewens WJ. 2006. Heterozygous carriers of Nijmegen
Breakage Syndrome have a distinct gene expression phenotype.
Genome Research 8:973–979.
Correa CR, Cheung VG. 2004. Genetic variation in radiation-
induced expression phenotypes. American Journal of Human
Genetics 5:885–890.
Ding LH, Shingyoji M, Chen F, Chatterjee A, Kasai KE, Chen
DJ. 2005. Gene expression changes in normal human skin
fibroblasts induced by HZE-particle radiation. Radiation
Research 4:523–526.
Dressman HK, Muramoto GG, Chao NJ, Meadows S, Marshall
D, Ginsburg GS, Nevins JR, Chute JP. 2007. Gene expression
signatures that predict radiation exposure in mice and humans.
PLoS Medicine 4:e106.
Fan F, Jin S, Amundson SA, Tong T, Fan W, Zhao H, Zhu X,
Mazzacurati L, Li X, Petrik KL, Fornace AJ Jr, Rajasekaran B,
Zhan Q. 2001. ATF3 induction following DNA damage is
regulated by distinct signaling pathways and over-expression of
ATF3 protein suppresses cells growth. Oncogene 21:7488–7496.
Finnon P, Robertson N, Dziwura S, Raffy C, Zhang W, Ainsbury
L, Kaprio J, Badie C, Bouffler S. 2008. Evidence for significant
heritability of apoptotic and cell cycle responses to ionising
radiation. Human Genetics J3:485–493.
Franco N, Lamartine J, Frouin V, Le Minter P, Petat C, Leplat JJ,
Libert F, Gidrol X, Martin MT. 2005. Low-dose exposure to
gamma rays induces specific gene regulations in normal human
keratinocytes. Radiation Research. 3:623–635.
Fry RC, Svensson JP, Valiathan C, Wang E, Hogan BJ,
Bhattacharya S, Bugni JM, Whittaker CA, Samson LD.
2008. Genomic predictors of interindividual differences in
response to DNA damaging agents. Genes Development
22:2621–2626.
Garty G, Chen Y, Salerno A, Turner H, Zhang J, Lyulko O,
Bertucci A, Xu Y, Wang H, Simaan N, Randers-Pehrson G,
Yao YL, Amundson SA, Brenner DJ. 2010. The RABIT: A
rapid automated biodosimetry tool for radiological triage.
Health Physics 8:209–217.
Geiss GK, Bumgarner RE, Birditt B, Dahl T, Dowidar N,
Dunaway DL, Fell HP, Ferree S, George RD, Grogan T,
James JJ, Maysuria M, Mitton JD, Oliveri P, Osborn JL, Peng
T, Ratcliffe AL, Webster PJ, Davidson EH, Hood L, Dimitrov
K. 2008. Direct multiplexed measurement of gene expression
with color-coded probe pairs. Nature Biotechnology 3:
317–325.
Grace MB, McLeland CB, Blakely WF. 2002. Real-time
quantitative RT-PCR assay of GADD45 gene expression
changes as a biomarker for radiation biodosimetry. Interna-
tional Journal of Radiation Biology 8:1011–1021.
Grace MB, McLeland CB, Gagliardi SJ, Smith JM, Jackson WE 3rd,
Blakely WF. 2003. Development and assessment of a quan-
titative reverse transcription-PCR assay for simultaneous
measurement of four amplicons. Clinical Chemistry. 49:
1467–1475.
Gruel G, Lucchesi C, Pawlik A, Frouin V, Alibert O,
Kortulewski T, Zarour A, Jacquelin B, Gidrol X, Tronik-
Le Roux D. 2006. Novel microarray-based method for
estimating exposure to ionizing radiation. Radiation Re-
search. 166:746–756.
Gruel G, Voisin P, Vaurijoux A, Roch-Lefevre S, Grégoire E,
Maltere P, Petat C, Gidrol X, Voisin P, Roy L. 2008. Broad
modulation of gene expression in CD4þ lymphocyte subpo-
pulations in response to low doses of ionizing radiation.
Radiation Research 170:335–344.
International Atomic Energy Agency (IAEA). 2001. Cytogenetic
analysis for radiation dose assessment: A manual. Technical reports
Series No. 405, Vienna: IAEA. Available from the website: http://
www-pub.iaea.org/MTCD/publications/. . ./TRS405_scr.pdf
Jen KY, Cheung VG. Transcriptional response of lymphoblastoid
cells to ionizing radiation. 2003. Genome Research 132092–
132100.
Jen KY, Cheung VG. 2005. Identification of novel p53 target
genes in ionizing radiation response. Cancer Research 65:
7666–7673.
Kang CM, Park KP, Song JE, Jeoung DI, Cho CK, Kim TH, Bae
S, Lee SJ, Lee YS. 2003. Possible biomarkers for ionizing
radiation exposure in human peripheral blood lymphocytes.
Radiation Research 159:312–319.
Li C, Lin M, Liu J. 2004. Identification of PRC1 as the p53 target
gene uncovers a novel function of p53 in the regulation of
cytokinesis. Oncogene 23:9336–9347.
Lindholm C, Stricklin D, Jaworska A, Koivistoinen A, Paile W,
Arvidsson E, Deperas-Standylo J, Wojcik A. 2010. Premature
chromosome condensation (PCC) assay for dose assessment in
mass casualty accidents. Radiation Research 173:71–78.
Mackay A, Tamber N, Fenwick K, Iravani M, Grigoriadis A,
Dexter T, Lord CJ, Reis-Filho JS, Ashworth A. 2009. A high-
resolution integrated analysis of genetic and expression profiles

of breast cancer cell lines. Breast Cancer Research Treatment
118:481–498.
Marchetti F, Venkatachalam S. 2010. The multiple roles of Bub1
in chromosome segregation during mitosis and meiosis. Cell
Cycle 9:58–63.
Meadows SK, Dressman HK, Muramoto GG, Himburg H,
Salter A, Wei Z, Ginsburg GS, Chao NJ, Nevins JR, Chute
JP. 2008. Gene expression signatures of radiation response
are specific, durable and accurate in mice and humans. PLoS
One 3:e1912.
Michel LS, Liberal V, Chatterjee A, Kirchwegger R, Pasche B, Gerald
W, Dobles M, Sorger PK, Murty VV, Benezra R. 2001. MAD2
haplo-insufficiency causes premature anaphase and chromosome
instability in mammalian cells. Nature 409:355–359.
O’Donovan MP, Freemantle MR, Hull G, Bell DA, Arlett CF,
Cole J. 1995. Extended-term cultures of human T-lympho-
cytes: A practical alternative to primary human lymphocytes for
use in genotoxicity testing. Mutagenesis 10:189–201.
Ossetrova NI, Blakely WF. 2009. Multiple blood-proteins
approach for early-response exposure assessment using an
in vivo murine radiation model. International Journal of
Radiation Biology 85:837–850.
Paul S, Amundson SA. 2008. Development of gene expression
signatures for practical radiation biodosimetry. International
Journal of Radiation Oncology Biology Physics 71:
1236–1244.
Pawlik A, Delmar P, Bosse S, Sainz L, Petat C, Pietu G, Thierry D,
Tronik-Le Roux D. 2009. Changes in transcriptome after in vivo
exposure to ionising radiation reveal a highly specialised liver
response. International Journal of Radiation Biology 85:656–671.
Payton JE, Grieselhuber NR, Chang LW, Murakami M, Geiss
GK, Link DC, Nagarajan R, Watson MA, Ley TJ. 2009. High
throughput digital quantification of mRNA abundance in
Gene expression response to ionising radiation 129
primary human acute myeloid leukemia samples. Journal of
Clinical Investigation 119:1714–1726.
Petrik KL, Fornace AJ Jr, Rajasekaran B, Zhan Q. 2002. ATF3
induction following DNA damage is regulated by distinct
signaling pathways and over-expression of ATF3 protein
suppresses cells growth. Oncogene 21:7488–7496.
Smirnov DA, Cheung VG. 2008. ATM gene mutations result in
both recessive and dominant expression phenotypes of genes and
microRNAs. American Journal of Human Genetics 83:243–253.
Stassen T, Port M, Nuyken I, Abend M. 2003. Radiation-induced
gene expression in MCF-7 cells. International Journal of
Radiation Biology 79:319–331.
Stoimenov I, Helleday T. 2009. PCNA on the crossroad of cancer.
Biochemical Society Transactions 37:605–613.
Tusher VG, Tibshirani R, Chu G. 2001. Significance analysis of
microarrays applied to the ionizing radiation response.
Proceedings of the National Academy of Sciences of the
USA 985116–985121.
Watts JA, Morley M, Burdick JT, Fiori JL, Ewens WJ, Spielman
RS, Cheung VG. 2002. Gene expression phenotype in
heterozygous carriers of Ataxia telangiectasia. American
Journal of Human Genetics 71:791–800.
Willems P, August L, Slabbert J, Romm H, Oestreicher U, Thierens
H, Vral A. 2010. Automated micronucleus (MN) scoring for
population triage in case of large scale radiation events.
International Journal of Radiation Biology 6:2–11.
Wojcik A, Lloyd D, Romm H, Roy L. 2010. Biological dosimetry
for triage of casualties in a large-scale radiological emergen-
cy:capacity of the EU member states. Radiation Protection
Dosimetry 138:397–401.
Woloschak GE, Chang-Liu CM. 1995. Modulation of expression of
genes encoding nuclear proteins following exposure to JANUS
neutrons or gamma-rays. Cancer Letters 97:169–175.