DNA Breaks, Chromosomal Aberrations And Cell Inactivation Induced By K
Ionization Events In DNA
M.A. Hervé du Penhoat1, A. Boissière1, A. Eschenbrenner1, F. Abel1, M.
Lamoureux1, M.F. Politis1, L. Sabatier2, E. Sage3, A. Touati1, A. Chetioui1
1
Groupe de Physique des Solides, Universités Paris 6 et Paris 7, UMR CNRS 75-88, 2 place Jussieu,
75251 Paris cedex 05, France
2
Laboratoire deRadiobiologie et Oncologie, CEA.DSV.DRR, BP 6, 92265 Fontenay aux Roses,
France
3
Laboratoire de Genotoxixité et Modulation de l'expression génique, Institut Curie-Recherche, 91405
Orsay cedex, France
Abstract. It is often thought that unrepaired or misrepaired complex DNA double-strand breaks are responsible for cell
inactivation and chromosomal aberrations. Here the role of inner-shell (core) ionizations in DNA atoms is studied. Ultrasoft
X-rays from LURE synchrotron radiation have been used to produce core events which mimic the ones induced by ionizing
radiations. By tuning the X-ray energy below and above the carbon K-threshold, it is possible to achieve a two-fold increase
of the number of core-ionizations in DNA for a same dose in cell. Cell survival, DNA double-strand breaks and chromosomal
aberrations have thus been studied, at three energies iso-attenuated in biological samples:~ 250, 350 and 800 eV. The relative
biological efficiencies of a given dose in cells to induce cell inactivation and chromosome aberrations appear governed by the
yields of core events in DNA. On the other hand, the lethal and DNA-breaking efficiencies of core events in DNA appear
strongly correlated.
correlated induction of CA4, presumably via DSB
misrejoining. However it is sometimes conjectured
that only a minority of DSBs, called reactive ones,
participate in misrejoining5. It has been suggested
that reactive DSBs might be particularly complex
DSBs or DSBs formed in clusters 6.
Analyses of the microscopic features of
radiation tracks have initially pointed out the
capacity of low-energy electrons to induce clustered
damages in DNA and therefore to produce strong
biological consequences. Their large capacity to
produce soft electrons was for instance thought to
play a key role in the large relative biological
efficiency (RBE) of ultrasoft X-rays, defined as
usual with respect to γ-rays. More precisely, for the
various electromagnetic radiations investigated,
RBEs were found correlated with the frequencies of
clusters of energy deposition of more than 100 eV in
2 nm-size DNA segments 7.
Here we want to investigate the biological role
of another type of energy clusters in particle tracks :
those from inner-shell ionizations (core-ionizations)
in DNA atoms.
INTRODUCTION
DNA double-strand breaks (DSBs) are a
particularly genotoxic form of DNA damage. Their
connection with cell inactivation and chromosomal
aberrations (CA) is well established. For radiations
with low linear energy transfer (LET), a correlation
between the cell fraction surviving to a given dose
and the corresponding mean number of DSBs per
cell was initially pointed out1. However, this
correlation was shown not to hold for large-LET
ions. Instead, for two cell lines and for a large range
of incident particles, the inactivation probability was
found roughly proportional to the number of DSBs
unrejoined 4 hrs after irradiation2. Also, for many
cell lines, the surviving fraction after a 2-Gy dose in
γ-rays was found strongly correlated with the
fraction of DSBs unrejoined 24 hrs after irradiation3.
This raised the picture of a category of more
complex DSBs as being responsible for cell
inactivation. DSBs are also involved in the formation
of CA, as shown by experiments in which an onset
of these lesions was found accompanied by a
CP680, Application of Accelerators in Research and Industry: 17th Int'l. Conference, edited by J. L. Duggan and I. L. Morgan
© 2003 American Institute of Physics 0-7354-0149-7/03/$20.00
52
events on DNA is thus about twice larger for X-ray
energies between 290 and 540 eV than for energies
just below 290 eV or just above 540 eV 11,12.
IMPLICATION OF DNA COREEVENTS IN THE BIOLOGICAL
EFFECTIVENESS OF IONIZING
RADIATIONS
U
Early experiments have demonstrated a lethal
and mutagenic effectiveness of core ionizations in
phosphorus (P) atoms of DNA 8,9. However, because
of the lower abundance and the smaller ionization
cross sections of P atoms in comparison with the
other ones of DNA, P core ionizations are expected
to have a small contribution to the overall radiation
effects. On the other hand, due to their much larger
occurence, inner-shell ionizations in carbon (C)-,
nitrogen (N)-, oxygen (O)- atoms of the DNA might
have a major biological role. This was first
suggested by the correlation observed 10, for heavy
ions, between cell inactivation cross sections and
cross sections for inner-shell ionization of one of the
C, N, O atoms of the DNA (fig.1).
Note that the hypothesis of a strong biological
role of core events in DNA provides a
straightforward explanation of the large RBE of
ultrasoft X-rays since most of these particles interact
with biological matter dominantly via inner-shell
ionization (see below).
100
mammal
Ar
F
Fe
Ar
10
Inactivation cross sections (µm2)
Xe
Ni
U+Pb
Xe
Kr
1
Ar
yeast
U
Ar
Kr Xe
bact
Pb
0,1
0,01
10
100
1000
LET (keV/µm)
10000
100
2
Inactivation cross sections (µm )
ULTRASOFT X-RAYS AS A
PROBE OF CORE-EVENTS
Ultrasoft X-Rays Used To Mimic The Core
Events Induced By Standard Ionizing Radiations
The effect of core ionizations cannot be
selectively investigated using usual charged particles
-fast electrons, ions- since those produce core events
with a very low probability (a few ‰). On the
contrary, for most ultrasoft X-rays, inner-shell
ionization is the dominant channel of interaction
with biological matter (table I). Moreover by
adjusting the X-ray energy above a atomic K
threshold it is possible to set the energy of the
photoelectron and to mimic -up to a certain extentthe core events created by electrons or ions. Finally
by selecting a given K-threshold among the C, N and
O ones, one preferentially ionizes a specific atomic
species and thereby can target specific cell
compartments. For instance, on account of the DNA
richness in carbon atoms, by using X-rays above the
C K-threshold (290 eV) and below the O K-threshold
(540 eV) it is possible to excite preferentially the
DNA. For a same dose in cell, the number of core
10
1
Fe
Ne
U
Kr
Ar
0,1
F
He
H
H
0,01
10
100
1000
10000
LET (keV/µm)
Figure 1. Cross sections for the inactivation of
mammalian cells by various ions (top) and cross
sections for the production of an efficient K vacancy
in one of the C, N, O DNA-atoms (bottom)
The effectiveness of ultrasoft X-rays to induce
cell inactivation, DSBs and CAs have thus been
compared at three energies : 250 eV, ~350 eV, ~800
eV.
53
the core events in DNA only. Indeed the calculated
yield of core events in DNA is much larger in the
second case. Two campaigns of experiments with
mammalian cells (V79) indeed demonstrated this
enhancement (fig 2)11,12. The measured value of
RBEinac (340 eV)/RBEinac (250 eV) = 2.0 ± 0.412
strikingly reproduces the 2.1 ratio calculated for the
corresponding yields of core events in DNA.
Because of the strong X-ray absorption inside
the biological medium, precise energies have been
chosen for each biological sample to ensure
isoattenuation inside it. For the above energies the
ratio between the mean nuclear dose and the entrance
dose in cells was around 15%.
Table I. Fractions and schemes of the various
photoionization events in partially hydrated DNA
(12 water molecules per nucleotide).
RBE
2
X-ray
380 eV
360 eV
230 eV
Outer-shell
ionization
70%
740 eV
26%
H, C, N,
O, Na, P
H, C, N,
O, Na, P
115 eV
Inner-shell
ionization
760 eV
260 eV
68%
0
O
90 eV
P (240 eV ; 115 eV)
13%
1
500 eV
61%
C
110 eV
6%
H, C, N,
O, Na, P
30%
P
Normalized RBE
250 eV
Typeenergy
of
ionization
220 eV
200
300
400
500
600
700
800
900
Energy (eV)
C (470 eV ; 260 eV) 14%
N (370 eV ; 360 eV) 9%
P (620 eV ; 115 eV)
3%
Figure 2. Inactivation RBEs measured for ultrasoft
X-rays at three isoattenuated energies. Here RBEs
are normalized to the one at 250 eV
Experimental Device
In this picture, the extracted probabilities that
core events in DNA lead to cell inactivation
(inactivation efficiency εinac) are respectively:
εinac (250 eV)= 1%
εinac (340 eV)= 0.9%
εinac (850 eV)= 2%
for L events produced in P atoms at 250 eV, K
events produced in C atoms at 340 eV and K events
produced in O atoms at 850 eV. The largest
efficiency at 850 eV is probably due to the largest
energies of the involved electrons (table I)
Experiments require the use of tunable
monochromatic X-rays which can only be obtained
at synchrotron radiation facilities. The present
studies have been performed in the LURE laboratory
at Orsay.
The experimental set-up is described elsewhere
11
. Due to the strong absorption of ultrasoft X-rays
by matter, a major necessity is to reduce as much as
possible the thickness of all materials placed ahead
the biological samples which have to be kept at
atmospheric pressure. In order to extract the photon
beam , a three-stage differential pumping connects a
very thin (150 nm) silicon nitride window to the
beam line (~ 10-9 Torr). The culture flasks are
moreover equipped with a very thin (~ 1 µm) 1 cm2
mylar foil through which cells are irradiated.
From the above efficiency values, the estimated
contribution of DNA core events to the overall ion
lethal effectiveness ranges from 50% to 100% 12.
Preliminary calculations indicate a similar core-event
contribution in the case of γ-rays.
DNA breaks
RESULTS AND DISCUSSION
DNA double-strand breaks (DSBs) induced by
ultrasoft X-rays have been measured for isolatedDNA samples (pBS plasmids). This technique allows
for the determination of DSBs selectively induced by
core events on the DNA 12.
The measured core-event efficiencies to induce
DSBs in situ are respectively 12:
Cell inactivation
For ultrasoft X-rays , a sharp increase of RBE
for inactivation (RBEinac, proportional to the number
of lethal events per dose unit in cell) was predicted
from below to above the C K-threshold when
assuming that the lethal effectiveness comes from
54
εDSB(250 eV) = 12%
εDSB(380 eV) = 10%
εDSB(760 eV) = 25%
Again the largest efficiency at 760 eV is
probably due to the largest energies of the involved
electrons. A strong correlation is observed between
the core-event efficiencies to inactivate cells and the
ones to induce DSBs in plasmid DNA. This might
indicate a scenario in which DSBs from core events
in DNA are the critical events responsible for cell
inactivation.
3. Dikomey E., Dahm-Daphi J., Brammer I.,
Martensen R. and Kaina B., Int J Radiat Bio 73, 269278 (1998)
4. Natarajan A.T., Obé G., Chromosoma 90, 120-127
(1984)
5. Chen A.M., Lucas, J.N., Simpson P.J., Griffin
C.S., Savage J.K.R., Brenner D.J., Hlatry L.R., Sachs
R.K., Radiat. Res. 148, S93-S101 (1997)
6. Sach R.K., Chen A.M., Simpson P.J., Hlatry L.R.,
Hahnfeld P., Savage J.K.R, Int. J. Radiat. Biol. 75,
657-672 (1999)
7. Goodhead D. T., Leenhouts H. P. , Paretzke H. G;,
Terrissol M., Nikjoo H; and Blaauboer R., Radiat.
Prot. Dos. 52, 217-223 (1994)
8. Kobayashi K., Hieda K., Maezawa H., Furusawa
Y., Suzuki M., Ito T., Int J Radiat Biol 59,643-650
(1991).
9. Saigusa S., Ejima Y., Kobayashi K., Sasaki M.S.,
Int J Radiat Biol 61, 785-790 (1992).
10. Chetioui A., Despiney I., Guiraud L., Adoui L.,
Sabatier L. and Dutrillaux B., Int J Radiat Biol. 65,
511-522 (1994).
11. Hervé du Penhoat M.A., Fayard B., Abel F.,
Touati A., Gobert F., Despiney-Bailly I., Ricoul M.,
Sabatier L., Stevens D.L., Hill M.A., Goodhead D.T.
and Chetioui A., Radiat Res 151, 649-658 (1999 ).
12. Fayard B., Touati A., Abel F., Hervé du Penhoat
M.A., Despiney-Bailly I., Gobert F., Ricoul M.,
L’Hoir A., Politis M.F., Hill M.A., Stevens D.L.,
Sabatier L., Sage E., Goodhead D.T. and Chetioui
A., Radiat Res 157, 128-140 (2002).
13. Touati A., Hervé du Penhoat M. A., Fayard B.,
Champion C., Abel F., Gogert F., Lamoureux M.,
Politis M. F., Martins L., Ricoul M., Sabatier L.,
Sage E. and Chetioui A., Rad. Prot. Dos. 99, 83-84
(2002)
14. Savage JRK., Mutat Res 404,139-147 (1998).
Chromosomal aberrations
Dicentric + ring chromosomal aberrations were
measured during two recent campaigns of
experiments at LURE 13. The measured ratios of
RBE to induce CAs (RBEaber) at 350 and 810 eV
versus 250 eV are respectively:
RBEaber(350)/RBEaber(250) = 2.5 ± 0.5
RBEaber(810)/RBEaber(250) = 1.2 ± 0.4
These ratios are very close to those for RBEinac.
This suggests that core ionizations on DNA might be
common initiating events for both end-points. Due to
the large energy locally deposited by these events,
they might induce complex unrejoined DSBs or
clusters of DSBs prone to misrejoining, as already
conjectured for ions 6. Note that an alternative
hypothesis, based on a single DSB and a
recombinational repair mechanism 14 was proposed
to explain the overall large RBEaber of ultrasoft Xrays. The connection presented here between CAs
and core events on DNA suggest new approaches for
investigating
the
still
unknown
involved
mechanisms.
ACKNOWLEDGEMENTS
This work was supported by CEA (LRC N°6), CNES
(contract N°793/99), and CNRS (contract PCV00033). Experiments were performed at the LURE
facilities. Work in L.Sabatier's laboratory was
supported
by
FIGH-CT-1999-00003
(RADINSTAB).
REFERENCES
1. Goodhead D. T., Radiat Res 91,45-76 , (1982)
2. Weber J. J. and Flentje M., Int J Radiat Biol 64,
167-178 (1993)
55