INT. J. RAD IAT. BIO L
1996,
V O L.
69,
NO .
6, 729 ± 738
Inactivation of V 79 cells by low-energy protons, deuterons and
helium -3 ions
M. FOLKARD*, K.M. PRISE, B. VOJNOVIC, H.C. NEWMAN, M.J. ROPER and
B.D. MICHAEL
(Received 19 December 1995; revision received 23 February 1996; accepted 5 M arch 1996)
Abstract. Previous work by ourselves and by oth ers has
dem onstrate d th at protons with a linear energy transfer
- 1
(LET) about 30 keV l m
are m ore effe ctive at killing cells
th an doubly charged particles of th e sam e LET. In th is work
we show th at by using deuterons, which have about tw ice the
range of protons with the sam e LET, it is possible to extend
m easurem ents of the RBE of singly charged particles to
- 1
higher LET (u p to 50 keV l m ). We report the design and
use of a new arrangement for irradiatin g V79 m am m alian
cells. C ell survival m easurem ents have been m ade using
protons in th e energy range 1 . 0 ± 3 . 7 M eV, deuterons in the
3
2+
energy range 0 . 9 ± 3 .4 M eV and H e ions in th e energy range
3 .4 ± 6 .9 M eV. This corresponds to volum e-averaged LET
- 1
(within the cell nucleus) betw een 10 and 28 keV l m
for
- 1
protons,
18 ± 50 keV l m
for
deuterons,
an d
59 ±
- 1
106 keV l m
for helium ions. Our results show no
difference in the effectiveness of protons an d deuterons
m atch ed for LET. H owever, for LET above about
- 1
30 keV l m
singly charged particles are m ore effective at
inactivating cells than doubly-charged particles of the sam e
LET and that this difference can be understood in term s of
th e radial dose distribution around th e primary ion track.
experim ental opportunity for investigating the
role of ionization den sity. Stu dies using lowenergy singly charged particles are also of
im portance to understanding the risk associated
with environm ental and occupational exposures
to fast neutron s where m uch of the dam age is due
to low -en ergy recoil proton s. Also, there is
increasing interest in th e use of protons in radiosurgery and radioth erapy (Raju 1995). We h ave
previously reported experim ents m easuring the
relative biological effectiven ess (R BE) for cell
survival (Folkard et al. 1989), and D N A dam age
(P rise et al. 1990) using protons with selected
energies <2 M eV. O ur ® ndings from these initial
stu dies were that for protons with LET
- 1
>17 keV l m , th e RB E for inactivating V79
m am m alian cells increases with increasing LET
an d th at protons with an LET of about
- 1
30 keV l m
were m ore effec tive than earlier
data had shown for doubly charged particles of
the sam e L ET. Sim ilar results regarding the
increased effectiveness of proton s have been
reported by Belli et al. (1993) w ho have studied
the inactivation of V79 cells and by G oodhead et al.
(1992) and Belli et al. (1992a) w ho have studied
1/ 2
both the inactivation of V79, H eLa and C3H 10T
cells and th e m utation of V79 cells by low -en ergy
protons and a -p articles. The observed differences
in th e lethality of protons and a -particles with the
sam e LET highlights th e im portan ce of trackstructure (as distinct from ionization density
alon g the track) in determ ining the biological
effect of a radiation. Protons and a -p articles
wh ich have the sam e ionization den sity along the
prim ary ion track (i.e. the sam e LET ) neverth eless
differ in respect of th e distribution of dose aroun d
the track.
Little reliable data so far exists for the
effectiveness of proton s w ith LET values greater
- 1
than about 30 keV l m . The data of Belli et al.
(1993) includes m easurem ents for th e R BE of
- 1
protons with LET values of 64 and 89 keV l m .
Th ese LET values are however, a re-evaluation of
1. Introduction
Stu dies related to th e fu ndam ental m echan ism s
of radiation action seek to un derstan d the spatial
qualities of ionizing radiation th at determ ine its
effect on tissue. It is now widely believed that it is
the extent to which ionization s are clustered over
nanom etre distances that determ ines radiation
effect (Frankenberg et al. 1986, Th acker et al.
1986, G oodhead 1994). T he relationship between
clustering of ionizations, th e subsequen t com plexity of the D NA lesion produced and the
eventual radiobiological effect, has so far largely
been the dom ain of track-structure m odelling
stu dies. Low-energy ligh t ions can h ave radiobiological properties associated with both
sparsely and densely ionizing radiation s, depen ding upon th eir energy, and therefore provide an
*Auth or for correspondence.
G ray Laboratory Can cer Reasearch Trust, PO Box 100,
M ount Vernon H ospital, N orth wood HA6 2JR, UK.
0955-3002 /96 $12.00
€
1996 Taylor & Francis Ltd
730
M . Folkard et al.
earlier work (Belli et al. 1989, 1992b) and the
auth ors indicate that these tw o data points are
now un reliable. A dif® culty of experim enting
- 1
with protons w hich have L ET > 30 keV l m
is
that the range of the proton in tissue is n ot
m uch greater th an the diam eter of a m am m alian
cell. Also, as the proton LET is increased, the
spread of LET within th e nucleus becom es large
and track-segm ent con dition s no longer apply.
Furth erm ore, it is dif® cult to estim ate reliably
the average LET, as sm all deviation s from the
idealized experim en tal arrangem en t can sign i® cantly affect th is quan tity. To overcom e this lim itation, we have perform ed experim en ts using
deuterons as well as protons. Deuterons with the
sam e LET as protons also h ave the sam e velocity
and track-structure and therefore (p resum ably)
sim ilar radiobiological properties. H owever, the
ran ge of th e deuterons is roughly tw ice th at of
protons of the sam e LET, and th e correspon ding
spread of LET w ithin th e nucleus is less. By using
deuterons, w e have been able to exten d our
m easurem ents of the RB E for inactivation by
singly charged particles to higher L ET. We h ave
furth er im proved th e reliability of m easurem en ts
m ade at higher LET by re-d esigning th e exp erim en tal arrangem ent to reduce, as far as possible,
the energy lost by particles reaching the cells. This
developm ent m inim ises the energy spread inciden t at the cell surface, an d also allows us to use
3
2+
H e ions with suf® cient energy, such that th eir
LET is com parable w ith that obtained using singly
ch arged particles.
We h ave also investigated the RBE of deuterons
with higher energies, so that their effect can be
com pared with protons at lower LET to con® rm
that the RBE values for LET-m atched protons and
deuterons are th e sam e. We h ave given th is aspect
of th e study particular atten tion following a
reported difference in th e RBE of proton and
deuterons with th e sam e LET, both for cell
inactivation (Belli et al. 1994) and m utation
(C h erubini et al. 1993).
In this paper, we describe the design, construction and dosim etry of our new arrangem ent
for irradiating V79 m amm alian cells using lowenergy protons in the energy range 1 .0 ± 3 .7 MeV,
deuterons in th e energy range 0 .9 ± 3 .4 M eV and
helium ions in the energy range 3 .4 ± 6 .9 MeV.
This corresponds to volume-averaged LET (with in
- 1
the cell nucleus) between 10 and 28 keV l m
for
- 1
protons, 18 ± 50 keV l m
for deuterons and 59 ±
- 1
106 keV l m
for helium ions. O ur results suggest
- 1
that at m oderate LET (about 40 ± 50 keV l m ),
singly-charged particles are m ore effective at
inactivating cells than doubly charged particles
with sim ilar L ET. We also ® nd no difference in th e
effectiven ess of protons and deuterons m atched for
LET.
2. M aterials and m ethods
2.1. Experimen tal arrangem en t
Energetic particles are produced using the G ray
L aboratory Van de G raaff accelerator. This is
nom inally rated to operate at an accelerating
voltage of 4M V, but in practice will operate
stably at any selected value between about 2 an d
4 . 2M V. An analyzing m agnet th at bends the beam
through 90 Êis used to select accelerated particles
of th e required type and energy so that nearm onoenergetic particles are delivered to the
sam ple, at the en d of a horizontal beam line
about 6m from th is m agnet. The usual com plem en t of electrostatic de¯ ectors and quadrupole m agnets are available to steer and shape
the beam .
An arran gem en t for irradiating cells has been
reported previously (Fo lkard et al. 1989), h ow ever
this under went partial m odi® cation (Folkard et al.
1995) an d th en furth er extensive m odi® cation for
the current study. The current m odi® cations h ave
been design ed to reduce as far as possible the
energy lost by th e particles reaching th e cell
sur face. T his is bene® cial for several reason s;
® rst, the reproducibility of th e energy at the
sam ple position w ith successive experim ents is
now m uch im proved. Second, the energy-spread
at the sam ple position due to straggling is reduced.
Finally, it extends the range of energies that can be
3
2+
stu died and allows us to use H e ions with which
very little energy loss can be tolerated if we are to
approach th e LET possible for singly-charged par4
2+
ticles. Although it is possible to accelerate H e
ions, the m axim um accelerating potential available does not allow us to ach ieve a condition
4
2+
wh ere w e could m atch the LET of H e to that
of deuteron s. N ote th at som e data have been
incorporated in this study from the partially
m odi® ed version, which gave sim ilar results to
the current version (where LET values could be
m atched).
The n ew arrangem ent is illustrated in ® gure 1a
an d b. U p to 12 sam ples are supported on a 29 cm
diam eter rotating alum inium platter that sweeps
each sam ple once th rough the radiation ® eld. Th e
speed of rotation (and hence the dose) can be
accurately an d independently preset for each
RBE of low-energy charged particles
731
(b )
Figure 1.
(a) O verall and (b) plan views of the sam ple irradiation apparatu s.
sam ple. The exit window of the beam line is m ade
from either 13 l m , or 25 l m thick polyim ide
(K apton, D u Pont) ® lm , supported by a vacuum tigh t ¯ ange over a 3 ´ 26 m m vertical slit. The
window is recessed 2 m m into the end-face of the
beam line and this distance represents the sm allest
window-to -sam ple gap that can be achieved. A slit
that de® nes th e shape of th e ® eld is m ounted just
before th e exit window (i.e. in the vacuum ). The
edges of the slit are m achined from 5m m diam eter
steel rod and de® ne a near-rectangular ® eld (th e
edges are aligned w ith radii of th e rotating platter
and are therefore not parallel) about 1 ´ 26 m m .
M oun ted to the slit are four (tw o per slit)
12 ´ 4 m m charge-collecting plates, sandw ich ed
between 20 l m th ick m ica for electrical isolation
an d each connected by insulated wire to an electrical vacuum feedthrough . Th ese serve to m onitor the dose to the sam ple and alth ough the four
plates operate independently, it has been found
suf® cient to sim ply sum and then m easure the
charges on a single electrom eter. A position
sensor on the w heel provides a sign al that
`enables’ this electrom eter, such that charge is
integrated only when the sam ple is crossing the
® eld. Additionally, a con tinuous analogu e an d
732
M . Folkard et al.
digital readout of the curren t from the m onitor is
displayed. These are used to check (and if necessary adjust) th e dose-rate before each sam ple is
irradiated, and to verify th at the dose-rate does
not change during the irradiation (which would
cause the dose to vary across the sam ple).
A 1 l m th ick gold scattering foil is m oun ted 1m
from th e exit window in the vacuum . The scattering area is reduced to 1 ´ 30 m m by a slit sim ilar to
that m ounted near the exit window. Th e window slit-m onitor assem bly can be positioned up to
30 m m horizontally off-axis, so th at only particles
scattered by th e gold foil can pass th rough th e exit
window. A quartz disk can be inserted into the path
of th e beam close to th e gold foil. Th e ¯ uorescence due to incident accelerated particles strikin g
the quartz can be viewed rem otely through a
vacuum window and is used at the beginning of
each session to ensure th at the beam is spread
evenly over th e scattering foil. A fu rth er check is
m ade by using an integrating electron ic cam era to
view the ¯ uorescence induced in a quartz slide
tem porarily ® xed over the exit window. T he im age
is processed and viewed using a 486-person al
com puter w hich en ables th e dose distribution
across the exit window to be readily evaluated. By
using a scattering foil, a un iform dose distribution
can be achieved over th e portion of the beam to
wh ich th e cells are exposed.
2.2. D ose and energy m easurem en t
The m eth ods for m easuring the dose and en ergy
at th e sam ple position have been described in
detail elsewh ere (Fo lkard et al. 1989). A parallelplate extrapolation cham ber is used to m easure
the dose at th e sam ple position and th us calibrate
the m onitor. Particles enter the ch am ber cavity
through the polarizing electrode, wh ich is m ade
from 3 l m thick alum inized M ylar. The other
electrode is a 13 m m diam eter copper plate
surrounded by a guard-rin g an d th e gap between
this and the polarizing electrode can be accurately
adjusted down to 0 .3 m m . The cham ber can be
m oun ted in place of th e rotatin g sam ple platter at
the appropriate distan ce from the window such
that it can be swept through the particle beam at
a precise angular velocity. The total electric charge
accum ulated by th e cham ber during on e sweep is
m easured using an electrom eter (K eithley, typ e
616) and by repeating th e process for a n um ber
of plate spacings, x, the charge, Q, per unit plate
spacing, dQ /dx, can be ascertained. Th e dose, D,
can be calculated using the following expression,
D =
(l / q )s
W
eA q
a
(l / q )a
k tp
dQ
dx
(1)
wh ere W / e is the average energy per ion pair for
singly charged protons and deuterons or doubly3
2+
- 1
charged H e
ions (35 J C
w as used th roughout), A is th e area of the collecting electrode
2
(136 m m ), q a is th e den sity of air at STP, k tp is a
tem peratu re and pressure correction factor, an d
(l / q )s / (l / q )a is th e ratio of stopping pow ers of the
sam ple and air. A value of 1 . 15 ± 1 .17 (depending
on th e particle and its energy) w as used, derived
from th e stopping power data for liquid water an d
dry air (p rotons and a -p articles), tabulated in
ICRU (1993).
To m easure the energy at the sam ple position , a
ruggedized silicon sur face-b arrier detector was
used (EG & G O rtec, 300 l m depletion depth) in
conjunction w ith conventional spectroscopy
electronics. The detector w as calibrated in
241
vacuum using an un sealed
Am isotope source.
Th e detector is con structed such that th e active
region is recessed 4 m m into the m ount. Sin ce the
vacuum window on th e Van de G raaff beam line is
recessed 2 m m , the m inim um air path that can be
ach ieved betw een th e w indow and th e detector is
6 m m . To en able energy (and dose) m easurem ents
to be m ade at th e sam ple position, th e window to
sam ple distance w as also set at 6 m m , although
sam ples can be irradiated closer than this, if
necessary.
2.3. Sam ple preparation
The particles of interest in this study have a short
ran ge, th erefore it is necessary to support th e cells
as a m onolayer. Chinese ham ster V79-379A cells
were m aintained in Eagles m inim al m edium
(E M EM ) containing 10% foetal calf serum an d
an tibiotics. A suspension of cells in expon en tial
7
- 1
phase w as concentrated to 10 m l
in H epesbuffered E M EM and 20 l l spread onto 13 m m
diam eter
polyvinylidine
di¯ uoride
® lters
(0 . 22 l m pore size, M illipore Corp.). Th e ® lters
rested on 1 . 0% (w/v) agar m ade up in cell culture
m edium . A fter a few m inutes, the m edium surrounding th e cells soaked through the ® lter to
leave an unattached m onolayer of cells on the
® lter in contact w ith enough m edium to keep
the cells viable, but not enough to cause appreciable radiation shielding. Th e prepared ® lters
have a m atte appearance w hen th is condition is
RBE of low-energy charged particles
reach ed. T he ® lter was then transferred to the
irradiation platter wh ere it was supported on
m oistened ® lter paper (W hatm an N o. 1) by surface ten sion. The platter w as cooled to about 10 ÊC
below the am bient tem peratu re to preven t the
sam ples drying out. Th e cooling was achieved by
circulating chilled antifreeze through a cavity
within the platter. Cells w ere typ ically on the
platter for < 10 m in in am bient atm osph ere. After
irradiation, th e cells w ere washed off the ® lters,
coun ted, diluted and plated out. Th e plated cells
were incubated for 6 days in an atm osph ere of 95%
air:5% CO 2 after w hich, they w ere stained and
colonies containing > 50 cells were scored. At
least three independent experim en ts w ere perform ed for each cell survival data point.
2.4. X -irradiation
The X -irradiations were perform ed using
240 kVp X -rays. The cells w ere exposed on m em brane ® lters as described in §2 . 3 at a dose-rate of
- 1
1 . 6 G y m in . A thick (5 m m ) Perspex lid was
placed over th e cells to provide build-up. The
irradiations took place with cells at 4 ÊC in atm ospheric air.
3. R esults and discussion
733
ach ieve the desired energy at the cell sur face
without the use of absorbers.
The shape of th e cells while supported by the
® lter is assum ed to be similar in appearance to that
described by D atta et al. (1976) for studies of trackend a -particles. They depict the cell as a `¯ attenedsph ere’, 10 l m th ick. This is clearly a sim pli® cation
of the tru e conditions, w here the cells have a range
of shapes and sizes. Lim ited m easurem en ts m ade
by us using a confocal m icroscope show that this is
a reasonable representation for calculation purposes (Folkard et al., unpublished data). These
observations h ave been used to constru ct a th reedim ensional com puter-m odel of th e cell and its
nucleus (Folkard et al. 1989), from w hich we can
calculate th e LET spectrum (and hence, the
volum e-averaged LE T) within th e cell nucleus for
all exp erim ental arrangem ents. E nergy losses
within the cell were estim ated using tabulated
data for proton and a -p article stopping powers
in liquid water (ICRU 1993). D euteron stopping
powers w ere assum ed to be th ose for protons at
3
2+
half th e energy. H e ions are assigned stopping
powers equivalent to those for a -particles at fourthirds th e energy. The calculated volum e-averaged
LET and the spread of LET within the cell nucleus
are shown in ® gure 2 as a function of th e incident
m ean energy. It is evident that we get th e expected
im provem ent (i.e. reduction) in th e spread of LET
wh en deuterons are used instead of protons of
equivalent LET. The m easurem ents perform ed
3.1. Energy measurem en ts and L ET evaluations
The particles crossing each cell will have a
spread of LET within th e n ucleus. This spread is
due both to the en ergy distribution of the inciden t particles and to th e en ergy lost by each
particle as it crosses th e cell. It is possible to
calculate the LET spectrum with in th e n ucleus
provided th e incident en ergy spectru m an d the
shape of th e cells and their nuclei are known .
From th e LET spectrum we can derive th e volum eaveraged LET, a single-value th at can be assign ed
to any given experim en tal con dition s. The
m easured m ean energy and ran ge of particles
inciden t at the cell surface (6 m m window-to sam ple distance) is sum m arised in table 1. In all
cases, the fu ll energy peak can be described by a
sim ple Gaussian curve and very few particles are
detected outside th is peak. The m easured fu ll
width at half m axim um (FW H M ) energy spread
is between 80 and 140 keV in all cases except for
the lowest energy proton data, w hich has a FW H M
of 180 keV. This is because it was not possible to
reduce the accelerator voltage suf® ciently to
Figure 2. C alculated volum e-averaged LET with in th e cell
nucleus as a fun ction of the m ean incident energy. Th e
dash ed lines above an d below each data set indicate
th e m axim um and m inim um LET present with in th e
cell nucleus.
734
M . Folkard et al.
using h igher energy particles reason ably resem ble
true track-segm ent experim ents, as the LET is
sim ilar through out th e cell nucleus. At low er
energies, the average LET with in the nucleus and
the spread of LET both increase. Above about
- 1
- 1
25 keV l m
for protons and 40 keV l m
for
deuterons, the average L ET is sen sitive to sm all
ch anges in both the incident en ergy and cell
shape. Variability in the cell (an d nucleus) th ickness will affect the value of the m axim um LET
present within th e nucleus, m uch m ore than the
m inim um value (which is affected only by the
thickness of the cytoplasm ). For exam ple, using
0 . 93 M eV deuterons (w hich is th e `w orst-case’
regarding sensitivity to cell shape), a ±10% uncertainty in cell nucleus thickness would m ean th at
the uncertainty in th e m axim um LET within
the nucleus is also about ±10% in th is instance.
Th e corresponding un certainty in the volum eaveraged L ET will be rough ly h alf of this value
(i.e. about ±5% ). For oth er particles and en ergies,
the uncertainty is less than th is.
3.2. M easurem ents of cell sur vival
Figure 3a ± c shows represen tative cell survival
curves after irradiation w ith protons, deuterons,
3
2+
H e and for com parison , 240 kVp X -rays. The
error-bars are ±1 standard error. All the data are
® tted using th e linear-q uadratic m odel such th at
the surviving fraction (SF ) is described by the
equation,
SF = exp - ( a D
+
2
b D ).
It is evident from both the proton and the
deuteron data th at as th e energy of the incident
particles is reduced (i.e. the average L ET
increased) the appearance of the survival curves
ch ange from low LET in character to h igh L ET.
Th at is, the curves becom e steeper an d shoulder is
reduced. For th e highest LET deuteron data, the
shoulder disappears com pletely. It can been seen
that for both th e highest proton an d deuteron L ET
survival curves there is no evidence of a `tail’ or
plateau at high doses, w hich m ight be seen if a
fraction of the cells w ere shielded. At these en ergies, th e range of the particles is n ot m uch greater
that th e width of the cell, th erefore even a sm all
am oun t of unwanted shielding, or `piling-u p’ of
cells would cause a plateau at relatively m odest
surviving fraction levels. T here is an indication of
shielding for the helium ion data an d in th is
instan ce, th e shielded data are not included in
the curve-® ts.
Figure 3. Surviving fraction of V79-379A cells afte r exposure
3
2+
to (a) protons, (b) deuterons, an d (c) He ions. Th e
survival afte r exposure to 240 kVp X-rays is also
depicted. The data are ® tte d by th e m ethod of leastsquares using the linear-quadratic m odel. Error bars
are ±1 stan dard error.
The RBE at th e 10% surviving fraction for all
proton, deuteron and h elium ion incident energies are plotted against volum e-averaged LET in
® gure 4, an d are tabulated alon g with other experim en tal param eters in table 1. E ach R BE is the
average of a m inim um of th ree experim ents an d
the error bars are derived from `worst-case’ ® ts to
the particle and X-ray data, when the respective
735
RBE of low-energy charged particles
3
2+
Figure 4. RBE at 10% surviving fraction for protons, deuterons an d He ions as a fu nction of volum e-averaged LE T within th e
cell nucleus. Th e error bars are derived from `worst-case’ ® ts to th e survival curves. Th e curves th rough the data are drawn
by eye.
errors for a and b are considered. T he m easured
- 1
RB E of protons at around 30 keV l m
is less th an
our previous m easurem ent (Folkard et al. 1989). In
this study however, the spread of LET w ithin the
cell nucleus has been m uch reduced (in particular,
there are few er contam inating low-energy protons,
wh ich h ave higher R BE) and we believe our curren t result better re¯ ects the RBE at th is L ET.
It can be seen that at all LET values used in th is
stu dy, the effectiveness of singly-charged particles
increases w ith increasing LET. It is evident
that the sam e RBE ± L ET relationship can be
used to describe th e effects of both protons and
deuterons. This result contrasts with a report
by Belli et al. (1994) and a prelim inary report
by Cherubini et al. (1993) who ® n d th at
Table 1.
X -rays
Protons
D euterons
He
2+
1
deuterons are less effective th an
protons with the sam e L ET for cell survival. Their
® ndings are unexpected as current physical
description s indicate that the track structures of
protons and deuterons are identical wh en th e LET
is the sam e. For singly ch arged particles, the tren d
of increasing RBE with L ET begin s to `¯ atten-out’
- 1
at the highest L ET (49 . 8 keV l m ) such that a
peak in effectiveness m ay exist around 55 ±
- 1
3
2+
65 keV l m . T he data for
He
ions an d
a -p articles indicates a m axim um R BE for doubly
- 1
charged particles about 90 ± 100 keV l m
in
agreem ent with other stu dies (i.e. Thacker et al.
1979).
Despite th e im provem ents to our irradiation
apparatus, w e have been unable to achieve a
In cident energy, th e range, the volum e-averaged LET, values for a and b from th e linear-quadratic ® ts to the data an d
RBE (initial slopes and at 10% surviving fraction) for th e particles used in th is stud y.
Radiation
3
<31 keV l m
Incident
energy (M eV)
CSDA range
in water ( l M )
240 kVp
3 . 66
1 . 83
1 . 07
3 . 40
2 . 14
1 . 40
0 . 93
6 . 90
4 . 18
3 . 39
210
65
27
115
55
28
16
74
33
24
LET
(K eV l m
10 .1
17 .8
27 .6
18 .5
26 .3
36 .1
49 .8
58 .9
88 .3
105 . 8
1
a
)
(G y
0 . 13
0 . 32
0 . 45
0 . 74
0 . 43
0 . 76
1 . 10
1 . 23
1 . 24
1 . 44
1 . 33
- 1
b
)
± 0 .022
± 0 .058
± 0 .035
± 0 .025
± 0 .050
± 0 .051
± 0 .014
± 0 .033
± 0 .025
± 0 .008
± 0 .019
(G y
- 2
)
0 . 048 ± 0 . 003
0 . 039 ± 0 . 011
0 . 028 ± 0 . 006
0 . 011 ± 0 . 004
0 . 055 ± 0 . 009
0 . 013 ± 0 . 008
0 .0
0 .0
0 .0
0 .0
0 .0
a /a
RBE
(X-ray)
1 . 00
2 . 49
3 . 42
5 . 63
3 . 28
5 . 77
8 . 35
9 . 32
9 . 41
10 . 90
10 . 11
RBE
(10% SF )
1 .00
1 .25
1 .40
1 .91
1 .56
1 .97
2 .74
3 .04
3 .07
3 .56
3 .31
736
M . Folkard et al.
condition where singly an d doubly charged
particles can be exactly m atched for L ET.
N evertheless, if lines drawn through the data are
extrapolated slightly, th en there is the suggestion
- 1
that at m oderate LET (30 ± 50 keV l m ), singlych arged particles are m ore effective th at doublych arged particles of the sam e LET. We do n ot
expect th at it w ill be possible to m easure reliably
the RBE for singly charged particles with L ET
- 1
m uch beyon d about 50 keV l m , because the
particle range will th en be too short. O ur highest
LET data point for singly ch arged particles was
obtained using deuterons with a m ean incident
energy of 0 . 93 M eV, which corresponds to a range
in th e cell of about 16 l m . To reduce the en ergy
still furth er increases the risk that particles w ill be
fully stopped within th e cell. An other dif® culty is
that the spread of LET with in th e nucleus is large at
high L ET (® gure 2) which m akes interpretation of
the data less straightforw ard. Clearly, th e m easured RBE correspon ds to an average effect of
the distribution of LET w ithin the nucleus and is
therefore not representative of a true tracksegm en t experim ent in this instance. O ne
m ethod of countering th is problem is to use
thinn er or attached cells (which ¯ atten) so th at
the particles have less cell th ickn ess to traverse.
Although th e cells rem ain rounded using the
m ethod reported here, th is has the advantage
that it probably ensures a greater un iform ity in
the cell-to-cell exposure com pared to an attached
cell system . Belli et al. (1989) irradiate V79 cells
Figure 5.
attached to 52 l m th ick M ylar (through the M ylar)
an d suggest th at `plateau’ in their cell survival data
could be caused by poorly attached or shielded
cells, and also by th e existence of a sub-p opulation
of rounded m itotic cells.
W here our data for proton s and deuterons overlap with those of other workers using th e sam e cell
line as ourselves, th ere is broad agreem ent with
their ® ndings. This is eviden t in ® gu re 5, wh ere we
have plotted our RBE ± LET data alon gside that of
Perris et al. (1986) and Belli et al. (1993, 1994). In
this instance the RBE is de® n ed in term s of the
ratio of th e initial slopes of the proton (o r deuteron) and X -ray sur vival curves (i.e. a / a x), in
accordance with the m ethod used in these
papers. O ur results agree well with the proton
data of Perris et al. (1986) and w ith th e proton
data (b ut n ot, as explained earler, the deuteron
- 1
data) of Belli et al. (1993, 1994) up to 30 keV l m .
Beyond th is value, th e data of Belli et al. show s
the R BE of protons decreasing w ith increasing
LET, although they state th at their high LET data
are unreliable, and therefore do not claim to
have identi® ed a m axim um in the RBE ± LET
relationship.
It has been pointed out in previous studies using
low-energy protons that the increased RB E of
protons and deuteron s com pared with helium
ions m ost likely re¯ ects differences in the particle
track-structure. Singly charged particles h ave a
lower velocity th an helium ions of the sam e LET,
therefore the energy spectrum of the secondary
RBE derived from the initial slope of the survival curves ( a / a X) as a fu nction of volum e-averaged LET. Also shown for
com parison are th e data of Belli et al. (1 993, 1994) and Perris et al. (1 986).
RBE of low-energy charged particles
electrons is also reduced with the con sequence
that the ionizations produced by th ese electrons
are m ore tightly clustered around the prim ary ion
track. Th e increased lethality of proton s and deuteron s com pared to helium ions is th erefore consisten t with m odels that place im portan ce on the
exten t to w hich ionizations are clustered at the
nanom eter level (G oodh ead 1994). Another point
to n ote is th at radiation protection m on itoring
m ethods th at rely on m icrodosim etric m easuring
techniques cannot distinguish between singly and
doubly charged particles with the sam e L ET. This is
particularly relevant to neutron ® elds wh ere a
signi® cant fraction of th e dose is due to low energy proton recoils. If, as these data suggest,
the energies of the secondary electrons are im portant in determ ining the biological effect of an
energetic particle, th en LET m ay n ot be the best
param eter for characterizing the particle track.
2
2
Th e quantity z* / b
(where z* is the effective
ch arge and b is the relative velocity) h as been
suggested as a m ore relevan t alternative (Katz
1970) as th e energy deposited by the secondary
electrons of particles m atched using th is param eter are sim ilar. It can be shown that our data
2
2
supports the notion that z* / b is a relevant param eter by noting that our data indicate peak effec- 1
tiveness of about 55 ± 60 and 95 keV l m for singly
and doubly charged ions respectively, and th at th e
2
2
corresponding value of z* / b for both of th ese is
about 1660. Th e conclusion th at can be drawn
from this is that th e RB E of a charged-p article is
not sim ply a function of the ionization density
alon g the track, but also depen ds on the radial
dose-d istribution aroun d th e track. H owever,
2
2
alth ough the use of z* / b
brings th e peaks into
approxim ate alignm ent, th e peak RBEs appear to
differ and are therefore not determ ined by the
2
2
value of z* / b .
4. C onclusions
In agreem ent with our previous work, and th at
of oth er workers, our data suggest that th e R BE for
cell survival of V79 cells exp osed to singly-charged
particles is greater than th at for h elium ions of th e
sam e LET at m oderate L ET values (i.e. about 40 ±
- 1
50 keV l m ). We have also shown th at deuterons
have a sim ilar RBE to protons with th e sam e LET,
and th at we can exploit the greater range of
deuterons to extend th e m easurem ents to higher
LET th an are possible using on ly protons. O ur data
show th at th e R BE for protons an d deuterons
737
increases with increasing LET up to th e highest
- 1
LET used (50 keV l m ), and suggest th at the peak
- 1
RB E m igh t be about 55 ± 65 keV l m . Fin ally, we
have shown that the differences between LETm atched singly an d doubly charged particles can
be understood in term s of th e differences in radial
dose distribution around the prim ary ion track.
Acknowledgem ents
This work is supported by th e Cancer Research
Cam paign and by grants from th e Radiation Protection R esearch Action Program m e of the European Com m unity. We should also like to thank
the staff of th e Gray L aboratory m echan ical an d
electrical w orkshops.
References
B ELLI , M ., C H ERUBINI , R., F INO TTO , S., M OSCH INI , G ., S APOR A , O .,
S IM O N E , G. and T ABOCCH INI, M . A., 1989, RBE ± LE T
relationsh ip for the survival of V79 cells irradiated
with low energy protons. International Journal of Radiation Biology, 55, 93 ± 104.
B ELLI , M., C E RA , F., C H ERU BIN I, R., I A N ZIN I, F., M O SCHIN I, G .,
S APO RA , O ., S IM ON E , G ., T A BO CCH INI, M . A. an d
T IVERO N , P., 1992b, RBE-L ET relationsh ip for survival
an d m utatio n induction of V79 cells irradiated with
low-energy protons: re-evaluation of the LE T values at
th e L NL facility. International Journal of Radiation
Biology, 61, 145 ± 146.
B ELLI , M ., C ERA , F., C HE RUBIN I , R., H A QU E , A. M . I., I A N ZIN I, F.,
M OSCHIN I, G., S AP ORA , O., S IM ON E , G ., T ABO CCH INI, M . A.
an d T IVERO N , P., 1993, In activation and m utatio n induction of V79 cells by low-energy protons: re-evaluation of
th e results at th e LNL facility. International Journal of
Radiation Biology, 63, 331 ± 337.
B ELLI , M ., C ERA , F., C HER UBIN I , R., G OO DH EAD , D . T., H AQ UE ,
A. M . I., I AN ZIN I , F., M OSCHIN I, G., N IKJO O , H., S APO RA , O .,
S IM O N E , G., S T EVEN S , D . L., T ABO CCH INI, M . A. an d
T IVERO N , P., 1994, Inavtivation induced by deuterons
of various LETs in V79 cells. Radiation Protection
Dosimetry, 52, 305 ± 310.
B ELLI , M., G OO DH EAD , D . T., I A NZINI, F., S IM O N E , G . an d
T A BO CCH INI, M . A., 1992a, Direct com parison of biological effe ctiveness of protons and a -particles of th e
sam e L ET. II. M utation induction at th e HPRT locus in
V79 cells. International Journal of Radiation Biology, 61,
625 ± 629.
C H ERU BIN I, R., C ERA , F., H A Q UE , A. M . I., T IVERO N , P., M OSCH INI ,
G ., S IM O N E , G ., B ELLI, M., I A N ZIN I, F., S A PORA , O . an d
T A BO CCH INI, M . A., 1993, M utatio n induction of low
energy protons in V79 cells. Abstracts of papers for th e
41st Ann ual M eeting of th e Radiation Research Society
an d the 13th An nual Meetin g of the North Am erican
H ypertherm ia Society, D allas, Texas, 135.
D AT TA , R., C O LE , A. and R OBIN SON , S., 1976, Use of track-end
241
alpha particles from
Am to stu dy radiosensitive sites
in CH O cells. Radiation Research, 65, 139 ± 151.
738
RBE of low-en ergy charged particles
F O LKA RD , M ., P RISE , K. M., V O JN OVIC , B., D AVIE S , S., R OPER , M . J.
an d M ICHAE L , B. D ., 1989, The irradiation of V79
M am m alian cells by protons with energies below
2 M eV. Part I: Experimental arrangem ent and
m easurem ent of cell survival. International Journal of
Radiation Biology, 56, 221 ± 237.
F O LKA RD , M ., P RISE , K. M ., V OJN O VIC , B., N EW M AN , H. C ., R OPER ,
M . J., H OLL IS , K. J. and M ICH AEL , B. D ., 1995, Conventional an d m icrobeam stu dies using low-energy
charged particles relevant to risk assessm ent and the
m echanism s of radiation action. Radiation Protection
Dosimetry, 61, 215 ± 218.
F R AN KENBERG , D ., G OO DHE AD , D . T., F RANK ENBE RG -S CHWAGE R , M .,
H AR BICH , R., B A NCE , D . A. and W ILKINSO N , R. E., 1986,
E ffe ctiveness of 1 .5 keV alum inium K and 0 . 3 keV
carbon K characteristic X-rays at inducing D NA
double-strand breaks in yeast cells. International Journal
of Radiation Biology, 50, 727 ± 741.
G OO DHE AD , D . T., B ELLI , M ., M ILL , A. J., B AN CE , D . A., A LL EN , L .
A., H ALL , S. C ., I AN ZIN I, F., S IM O N E , G., S TE VEN S , D. L.,
S TRETCH , A., T ABOCCH INI, M . A. an d W IL KINSON , R. E.,
1992, D irect com parison of biological effe ctiveness of
protons and a -particles of the sam e LET. I. Irradiation
m eth ods and inactivation of asyn chronous V79, HeLa
an d C3H 10T1/2 cells. International Journal of Radiation
Biology, 61, 611 ± 624.
G OO DHE AD , D. T., 1994, In itial events in th e cellular effects of
ionizing radiations: clustered dam age in D NA. International Journal of Radiation Biology, 65, 7 ± 17.
ICRU, 1993, Stopping Powers and Ranges for Protons and
a -Particles. Report 49 (Washington: In ternational
C om m ission on Radiation Units and Measurem ents).
2
2
K AT Z , R., 1970, RBE, L ET an d Z* / b , Health Physics, 18, 175.
P ERR IS , A., P IALO GLO U , A. A., K ATSA N OS , A. A. an d S ID ERIS , E. G .,
1986, Biological effectiveness of low energy protons. I.
Survival of Chinese ham ster cells. International Journal
of Radiation Biology, 50, 1093 ± 1101.
P RISE , K. M., F O LKARD , M ., D AVIES , S. and M ICH AEL , B. D., 1990,
The irradiation of V79 Mam m alian cells by protons
with energies below 2 M eV. Part II: M easurem ent of
oxygen enhancem ent ratios and D NA dam age. International Journal of Radiation Biology, 58, 261 ± 277.
R AJU , M . R., 1995, Proton radiobiology, radiosurgery an d
radiotherapy. International Journal of Radiation Biology,
67, 237 ± 259.
T HACKER , J., S TRE TCH , A. and S T EPHE N S , M . A., 1979, M utation
an d inactivation of cultured m am m alian cells exposed
to beam s of accelerated heavy ions. II. Chinese ham ster
V79 cells. International Journal of Radiation Biology, 36,
137 ± 148.
T HACKER , J., W ILKINSO N , R. E . an d G OO DH EAD , D . T., 1986, Th e
induction of chrom osom e aberrations by carbon ultrasoft X-rays in V79 ham ster cells. International Journal of
Radiation Biology, 49, 645 ± 656.