Electron and photon induced damage to biomolecular systems M. Folkard Gray Cancer Institute, PO Box 100, Mount Vernon Hospital, Northwood HA6 2JR, UK

Ionising radiations damage biomolecules (including DNA) by breaking bonds. Bond-breaks occur either: - Directly, by direct ionisation of the biomolecule - Indirectly, through the ionisation of water, and the formation of damaging reactive radicals Radiation damage of biomolecules

Ionizing radiation damages ALL biomolecules similarly We now know that the most radiation-sensitive biomolecule in living tissue is DNA Consequently, it is damage to DNA that leads to all observed macroscopic biological effects

repair mis-repair mutation viable cell not repaired cancer cell death Radiation damage of biomolecules

Physical sionisation, excitation Timescale of events: Early boil. hours - weekscell death, animal death Late boil. yearscarcinogenesis Radiation damage of biomolecules Chemical sfree radical damage s - hourschemical repair

Nevertheless, the effectiveness of an ionising radiation critically depends both on its type (i.e. photon, particle) and on its energy Therefore, these differences arise solely because radiations of different quality and type produce different patterns of ionisation Radiation damage of biomolecules For the same dose, both the quality and the number of ionisations produced by ALL ionising radiations is the same

Biological effectiveness: radiation type Energetic X-rays

1 Gy ~ 1000 tracks per cell ~ 100,000 ionisations per cell Biological effectiveness: radiation type

 -particles 1 Gy ~ tracks per cell ~ 100,000 ionisations per cell Biological effectiveness: radiation type

Millar et al. Biological effectiveness: radiation type C3H 10T1/2 cells transformants / 10 4 surviving cells 250 kVp X-rays 4 He 2 dose / Gy

surviving fraction dose / Gy V79 cells energetic X-rays 1.5 keV Al K X-rays Prise, Folkard & Michael, keV C K X-rays Goodhead and Nikjoo, 1989 Biological effectiveness: radiation quality

The primary factor that determines biological effectiveness is ionisation density -  -particles and low-energy X-rays are densely ionising - energetic X-rays are sparsely ionising Biological effectiveness In general, densely ionising radiations are more effective than sparsely ionising radiations

2  m 200 nm 20 nm 2 nm Biophysical Models of radiation damage - Develop a mathematical model of the cell and radiation track-structure

200 nm energetic X-rays Biophysical Models of radiation damage Breckow & Kellerer, 1990  e-e-

20 nm 1.5 keV Al K X-rays Biophysical Models of radiation damage Nikjoo, Goodhead, Charlton, Paretzke, keV X-ray e-e- e-e-

2 nm 0.28 keV C K X-ray Biophysical Models of radiation damage 0.28 eV X-ray Nikjoo, Goodhead, Charlton, Paretzke, 1989 e-e-

 - particle Biophysical Models of radiation damage  -particle  e-e- 2 nm

photon single-strand break DNA Damage

double-strand break e-e- photon DNA Damage

complex damage Locally multiply damaged sites (LMDS) DNA Damage

The track-structure models are very good at mapping the pattern of ionizations relative to the DNA helix The next key step is to map the pattern of breaks in the DNA helix For this, we need to know the amount of energy deposited through ionisation, and the amount of energy required to produce strand-breaks

1 MeV electrons Energy E / eV Frequency per eV liquid water DNA most probable E loss: 23 eV Re-drawn from; LaVerne and Pimblott, 1995 DNA Damage Theoretical spectrum of energy depositions by energetic electrons

100 keV electrons 300 eV electrons 2 nm Energy E / eV Freq. Events >E per target / Gy Re-drawn from; Nikjoo and Goodhead, 1991 Frequency of energy depositions >E in a 2 nm section of the DNA helix Most energy depositions ~few 10’s eV Few energy depositions >200 eV DNA Damage

Questions: How much energy is involved in the induction of single- and double-strand breaks by ionizing radiations? What is the minimum energy required to produce: 1) a single-strand break 2) a double-strand break

DNA Damage probability of break energy in DNA / eV SSB DSB Nikjoo et al calculated the probability of SSB and DSB, based on data for strand breaks from I 125 decays Minimum energy to produce SSB ~20 eV Minimum energy to produce DSB ~50 eV Re-drawn from; Nikjoo, Charlton, Goodhead, 1994

ionising synchrotrons characteristic X-ray sources vacuum tubes linacs gas discharge sources isotope sources Energetic photon sources typical cluster size 1 eV1 keV1 MeV1 GeV ultra-violetsoft X-rays X- and  -rays photon energy / eV

Measurement of DNA damage Use Plasmid DNA (circular double-stranded molecules of DNA, purified from bacteria) i.e. pBR322 (4363 base-pairs) Un-damaged DNA (supercoiled) linear double-strand break relaxed single-strand break

relaxed linear supercoiled Measurement of DNA damage These forms can be easily separated by gel-electrophoresis

energy / eV SEYA, LiF, MgF window TGM, polyimide window SEYA, aluminium window photons s -1 cm -1 Experiments using the Daresbury Synchrotron

window electrometer valve pump VUV grid sample sample ‘wobbler’ Experiments using the Daresbury Synchrotron ‘dry’ DNA irradiator

SSB induction in ‘dry’ DNA 150 eV photons % supercoiled DNA Photons / cm 2 0 1x x x

0 1x x x eV x x x x eV x x x eV x x eV 1 % supercoiled DNA Photons / cm 2 SSB induction in ‘dry’ DNA

150 eV photons DSB induction in ‘dry’ DNA % linear DNA 0 1x x x Photons / cm 2

x x x x eV 0 1x x x eV 8 0 2x eV 1x x10 14 % linear DNA x x eV Photons / cm 2 DSB induction in ‘dry’ DNA

SSB DSB Quantum Efficiency /  Photon Energy / eV ~20-fold Q.E. for SSB & DSB (dry plasmid) Prise, Folkard et. al, 1995, Int. J. Radiat. Biol. 76, SSB thresholdDSB threshold

Observations x x x x eV x x x x eV % % supercoiled % linear photons / cm 2 The 37% ‘loss of super- coiled’ level represents an average of one ssb per plasmid. At an equivalent dose, about 4% dsb produced Induction of dsb is linear with dose, and has non- zero initial slope Therefore dsbs are NOT due to the interaction of two (independent) ssbs

Free radical damage of DNA photon H 2 O H 2 O + + e - H + + OH

020 scale / mm VUV ‘DNA in solution’ VUV irradiator MgF DNA in 50  m gap

‘DNA in solution’ VUV irradiator

ionising synchrotrons gas discharge sources Energetic photon sources 1 eV1 keV1 MeV1 GeV ultra-violetsoft X-rays X- and  -rays photon energy / eV Useful region for ‘solution irradiator’

Wavelength / nm Output Peak at 147 nm ( = 8.5 eV) RF-excited Xenon Lamp VUV spectrum

source (Xenon lamp) VUV irradiator (lamp) concave grating monochromator

VUV irradiator (lamp)

DNA damage yields in solution: % supercoiled DNA Dose / Gy 50 SSB % linear DNA Dose / Gy DSB 7 eV photons

7eV % supercoiled DNA Dose / Gy 50 SSB Dose / Gy % linear DNA DSB 8.5 eV photons DNA damage yields in solution:

% supercoiled DNA Dose / Gy 50 SSB Dose / Gy % linear DNA DSB 8.5 eV photons DNA damage yields in solution:

% supercoiled DNA Dose / Gy SSB Dose / Gy % linear DNA DSB 8.5 eV photons mM Tris (OH radical scavenger) DNA damage yields in solution:

Dose / Gy 8.5 eV % linear DNA no scavenger scavenger Observations At all dose levels, the addition of a radical scavenger reduces the number of induced dsb The OH mediated damage is linear with dose This suggests that a single OH radical can produce a dsb

Are the strand-breaks due to (non-ionizing) UV damage? It is possible that ssb and dsb are caused by contaminating UV radiation UV-induced DNA damage consists mostly of the formation of pyrimidine dimers Addition of T4 endonuclease V converts pyrimidine dimers to strand-breaks

SSB DSB % supercoiled Dose / Gy Dose / Gy % linear no T4 with T4 +T4 endonuclease V DNA damage yields in solution: 8.5 eV photons

Mechanisms for ssb and dsb induction at low-energies Boudaiffa et al. have demonstrated that ssb and dsb can be induced in DNA by electrons with energies as low as 5 eV, through the process of ‘electron attachment’ Resonant Formation of DNA Strand Breaks by Low-Energy (3 to 20eV) Electrons. Science 287, (2000). B. Boudaiffa, P. Cloutier, D. Hunting, M.A. Huels et L. Sanche. “This finding presents a fundamental challenge to the traditional notion that genotoxic damage by secondary electrons can only occur at energies above the onset of ionization…”

Mechanisms for ssb and dsb induction at low-energies Incident electron energy / eV DNA breaks / incident electron (x10 -4 ) DSBs SSBs

Mechanisms for ssb and dsb induction at low-energies Below 15 eV, electrons can attach to molecules and form a ‘resonance’ e - + RH RH * transient molecular anion (TMA) RH * R + H _ electron autodetachment, or dissociation DSB induction occurs when fragmentation components react with the opposite strand This can induce an SSB

K.M. Prise G.C. Holding D. Cole C. Turner S. Gilchrist B Vojnovic B.D. Michael F.A. Smith B. Brocklehurst C.A. Mythen A.Hopkirk M. Macdonald I.H. Munro Acknowledgments GCI other

 The action spectra for ssb and dsb induced in dry DNA are similar, indicative of a common precursor. Conclusions  DNA in solution irradiated with 7 eV, or 8.5 eV photons gives a linear (or linear-quadratic) dsb induction, indicative of a single-event mechanism.  Addition of tris suggests that a single OH radical has a significant probability of inducing a dsb.

701.9x x *03.2x x * 11.0x x x x x x Co x x Co x x E/eVtris/mM ssb / Gy -1 bp -1 dsb/ Gy -1 bp -1 ssb/dsb synchrotron * DNA damage yields in solution:

% supercoiled DNA Dose / Gy % linear DNA Dose / Gy SSB no tris 1mM tris no tris 1mM tris Co 60  -rays (+ 1mM tris) DNA damage yields in solution:

yield ferric ions / photon energy / eV Water radical yields by Fricke dosimetry Watanabe, R., Usami, N., Takakura, K., Hieda, K. and Kobayashi, K., 1997, Radiation Research, 148, dsb/ Gy -1 bp x x x x x x10 -6 DSB

yield ferric ions / photon energy / eV Water radical yields by Fricke dosimetry Watanabe, R., Usami, N., Takakura, K., Hieda, K. and Kobayashi, K., 1997, Radiation Research, 148, ssb/ Gy -1 bp x x10 -5 SSB