Single Cell Irradiation Nuclear Microscopy
Using a Radioactive Source
P. Rossi*#, B.L. Doyle*, J.C. Banks*, A. Battistella+, G. Gennaro$,
F.D. McDaniel*¶, M. Mellon£, E. Vittone&, G. Vizkelethy*, N.D. Wing*
*
Sandia National Laboratories, Albuquerque, NM, USA
Department of Physics and INFN, via Marzolo 8, 35131 Padova, Italy
+
Laboratori Nazionali Legnaro, INFN, Legnaro(PD), Italy
$
Azienda Ospedaliera and INFN, Padova, Italy
¶
Ion Beam Modification and Analysis Laboratory, University of North Texas, Denton, TX
£
Quantar Technology inc., Santa Clara, CA
&
Department of Experimental Physics of the University and INFN, 10125 Turin, Italy
#
Abstract. Irradiation of a single biological cell, instead of a whole tissue, with ions in a known number and position, is a powerful means to study very low dose biological effectiveness. Present methods employ accelerated ion beams which are 1) either
collimated with micro-apertures and affected by a halo of 3-5µm at best, or 2) focused to a sub-micron spot, whose resolution is
degraded when extracted into air. We have studied the efficacy of a new micro-radiobiological method, originally developed for
materials research. This new approach uses an IPEM, Ion Photon Emission Microscope, which employs a specially shaped Po210 alpha particle source for in-air irradiation. Alpha particles strike the cells, which are previously grown directly on a 10-20
µm thick scintillating plastic blade and placed in the focal plane of a conventional optical microscope. Photons produced at the
single ion impact point are projected at high magnification onto a single-photon position sensitive detector, which provides the
position of each ion that hits the cells. Adequacy of this setup for Single Cell Radio-Biology will be discussed.
tions stressing the cells and blurring dim endpoints
could lead to safer, but more demanding measuring
methods. An example is that cells rest horizontally on a
Petri-dish floor and are covered by a thick culture medium, which requires that the beam must have a vertical
up direction. Eventually, in Bystander Effect studies,
one considers endpoints of an untouched cell, when few
irradiated cells surround it. In this case, it seems better
to avoid a random irradiation of the culture medium to
have clearer experimental conditions.
For Single Cell irradiation, we mean in-air irradiation of a living-cell culture, where cells are spatially
separated and every particle hits the culture in a known
position. Facilities for this kind of irradiation can be
divided in two groups: those in which the single ion is
explicitly aimed to strike a predetermined position and
those in which the ion strikes randomly, but its position
is measured and recorded.
INTRODUCTION
Few facilities in the world use accelerated ion
beams to irradiate single biological cells for studies of
dose effects. Whenever a low-dose radiation risk has to
be assessed, as in the much-studied case of radon exposure, low fluence broad beams and statistical methods
proved totally inadequate. Single cell irradiation methods were developed by two pivotal groups at Gray
Laboratory in London [1] and at RARAF of Columbia
University in New York [2].
Actually, there are a number of radiobiological
measurements that require diversified techniques. In
the following, we review some of them. Studies of rare
endpoints (e.g. “what” happens to cells after being hit)
require hitting several 100,000 cells, which led to systems for high speed automated cell pattern recognition,
aiming, and hitting. If biological intracellular signaling
models have to be tested, a micrometer spatial accuracy
is required. Sometimes, very different and penetrating
high-energy ion species are needed to study the socalled “Bystander Effect,” when a whole tissue has to
be targeted. Finally concern about experimental condi-
Aiming at Cells
A facility of the first type requires a micro-beam,
which can be provided by either collimating or focusing
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
364
optically inspected after the irradiation run, and (2) the
lack of highly controlled experimental conditions, when
this is required, as in the Bystander Effect studies. This
method, which originally gave interesting results, is
now very seldom used due to the difficulty in providing
good statistics. The second drawback alone does not
prevent employing the “random” irradiation method to
study several biological endpoints, but it does seriously
complicate experiments of the Bystander Effect.
We are now proposing an alternate way to perform
single cell random irradiations with on-line impact
point measurements. This new approach will provide
high statistics, is inexpensive, and offers reliable operation. This new approach is an extension of an idea of
Doyle et al. at Sandia National Laboratories, who proposed that nuclear microscopy can be performed without microbeams [3,4]. An early suggestion to apply
this microscopy to radiobiology can be found in ref. [5].
We have developed an IPEM, Ion Photon Emission Microscope, to be employed with a specially
shaped alpha particle source for in-air irradiation. See
Fig. 1. Although, it was originally planned for Radiation Effects Microscopy of ICs, it could also become a
radiobiology tool. Alpha particles randomly strike the
cells, which are previously grown directly on a 10-20
µm thick scintillating plastic blade, and placed in the
focal plane of a conventional optical microscope. Photons produced in the blade at the single ion impact point
are projected at high magnification onto a single photon
position sensitive detector. These positions are later
correlated with the measured position of the cells under
exposure.
a broad beam. Collimated micro-beams are the easiest
approach, since all that have to be taken into account
are a small hole for the ions and a 100% efficient trigger to detect the single ion, which then switches on a
mechanical or electrostatic beam-blanking shutter.
Fused silica capillaries of about 1 µm bore are used at
Gray Lab [1]. A pair of laser-drilled apertures is used
at RARAF [2]. Examples of such triggers are thin plastic phosphor foils giving light upon the particle crossing
(Gray-Lab), or miniaturized pulse ion counters (RARAF).
Collimation methods, while providing a huge
amount of data and significant achievements, have two
main drawbacks: (1) the presence of a prominent halo,
due to ions hitting the collimator walls, which extends
the actual beam size to 3-5 µm at best; and (2) the need
for a relatively slow mechanical movement to position
the single cell in front of the collimator. The second
drawback forbids a very high irradiation speed and reduces statistics. Although really fast stages are now
available, one wonders whether such a mechanical impulse could stress the cells or even change their position.
A proposed solution has been to employ focused
micro-beams with an electrostatic or magnetic scanning
system, which are more costly and technically demanding. Moreover, a micrometer spot size beam can exist
only in high vacuum and degrades immediately whenever extracted in air. A very thin window coupled to an
ultra-precise positioning of the Petri dish, which should
lean upon the window, would seem to solve the problem.
According to widespread opinion, the cell irradiation facilities should be devoted to this sole task and be
under control of a “biological” group. Simplicity and
low cost of both installation and operation are therefore
important. Because these systems must be so dedicated, one cannot easily exploit accelerators, beam lines
and microbeams already built for other purposes. This
is particularly true when large production or simply
suitable biological conditions are required, e.g. a vertical-up beam. For the time being, no large production of
real biological interest has taken place in focused beam
installations and it seems that collimation will remain
the leading approach to single cell irradiations for some
time.
OM40
60 X
X
PSD
0.5 mCi Po210
400 TEM grid
BC400
Random Irradiation
PinDiode
Diode
Pin
Diode
A second approach for single cell irradiation consists of randomly hitting the culture and measuring each
impact point relative to the position of the cells. Traditionally, nuclear emulsions stacks or sensitive plastics
like CR39 are employed. This method has two drawbacks: (1) the practical impossibility to handle a large
quantity of data since the “nuclear” images must be
IBICC
FIGURE 1. Setup for an alpha particle IPEM in vacuum.
365
7
IN-VACUUM IPEM SETUP
AND PERFORMANCE
-1
alpha particle source (activity = 2·10 s , diameter =
12.5 mm, manufacturer Isotope Products).
Because of geometric restrictions, we initially had
to place the source on the other side of the OM40 hole
with respect to the sample (at 40 mm), and evacuate the
chamber (see Fig. 1). We obtained a detected alpha
particle rate of only 0.05 Hz using this initial trial system, and had to run over the weekend to get the image
shown in Fig. 2. While these results are clearly not
optimal, they do represent the first demonstration of inair nuclear microscopy performed using a radioactive
source.
This alpha-source-IPEM is currently being evaluated for resolution, efficiency and its adequacy for radiobiology. We are proceeding using two approaches:
1) exploiting an existing “accelerator” IPEM setup,
which has to be operated under vacuum for an early
assessment of performance; and 2) building and testing
a prototype alpha particle IPEM, which will involve inair alpha particle source irradiation.
The existing setup includes a scattering chamber, a
specially shaped microscope (JEOL OM-40), with a
hole allowing the beam to pass through during viewing,
a commercially available 210Po alpha source, and a
Quantar Position Sensitive Detector (PSD) for light
measurement [6]. Among alpha particle emitters, 210Po
has many advantages. The alpha particle energy is 5.3
MeV and the source has minimal gamma and beta
emission. It has a half-life of 138 d, which is long
enough to allow a practical use without continued replacements and short enough to have a high specific
activity, leading to a spectroscopic-grade energy resolution.
To maintain single photon sensitivity, the PSD has
a low signal threshold, which results in a quantum noise
of few hundreds counts per second (cps). For IPEM
experiments, and particularly for single cell hit experiments, it is necessary to eliminate this noise. To accomplish this, we demand a coincidence with a particle
detector, and therefore a silicon “pin-diode” detector is
located downstream from the sample.
The spatial resolution has been analyzed in ref. [7]:
The FWHM is about 12 µm and there is a prominent
halo (Tail Percentage, TP, for |x| > FWHM of about
20%). This poor result is due to scattered and stray light
rays in the blade and the environment proximate to the
blade. Enhancements in resolution are currently being
explored. These improvements include a better preparation of the environment, which has to be as opaque as
possible, and adhering the blade to the sample.
A comparison between hits detected by the IPEM
and the pin-diode gave a measured overall IPEM efficiency of about 0.2. For this test, we employed a provisional PSD detector, with a S2 photo-cathode (quantum
efficiency (q.e. = 0.05). We foresee alpha detection
efficiency close to 100% for the new system, because
the optical microscope on this system has a much larger
Numerical Aperture and Transmission Coefficient. In
addition, a new Quantar detector (Mepsicron) is being
used on the new setup, which has a bi-alkali photocathode with q.e. = 0.25.
The alpha particle exposure rate has been extremely
low up until now, due to the geometry of the JEOL
OM40 microscope and the low activity of the 210Po
FIGURE 2. Alpha particle IPEM on a 400 TEM-grid. The
full-energy IBIC signal from the pin-diode is put in coincidence with the PSD (x, y) signals
ALPHA PARTICLE TABLE-TOP IPEM
Of course these exposure rates are much too low to
allow a practical use of the alpha-source-IPEM. Our
solution to increase this rate has been to employ a usual
table-top microscope (see Fig. 3) and place a needle
source between the lens and sample. See Fig. 4.
The distance from source to sample would be between 5 and 10 mm. The solid angle fraction (SAF),
for a viewing field of 0.5 mm radius and a distance of 6
mm is SAF = Ω/(4π) = 1.7·10-3. If Nhit is the number of
cells per second we want to hit and CAF the Cell Area
Fraction, we can calculate the total activity of the
source that is needed to be A = Nhit/(CAF*SAF). For
CAF=0.01, which corresponds to a cell diameter of 10
µm and average cell distance of 100 µm, and for Nhit
between 10 and 1000, which is the highest speed focused beam facilities that we have, we get an activity
between 6·105 and 6·107 Bq, or 0.015-1.5 mCi.
366
From Isotope Products, we obtained a custom 2
mCi 210Po source, electrodeposited on a thin silver foil
of 100 mm2 surface area. A new source with 50 times
more activity will be delivered soon. We will then be
able to cut the 2 mCi radioactive foil into pieces of 1.3
mm on a side, with an activity of 106 s-1 each. The reduced size will allow the source, when attached to a
needle, to be brought much closer to the sample with a
substantial increase of the solid angle fraction. This
distance could be as small as 6-8 mm, allowing in-air
irradiation. Also, the needle can easily be mounted
close to the lens of a standard table-top optical microscope. See Figures 3 and 4.
FIGURE 4. A high activity (1.3 mm)2 source is placed between lens and sample, to be covered by a phosphor blade.
The use of a thick culture medium appears to be
possible with our set-up, provided we modify the current system to utilize an upward pointing beam and
have a “rejection” detector placed before the phosphor
blade (e.g. underneath it), which is thin enough to
transmit the alpha particles at minimal energy loss. An
example of such an upstream ion detector is a miniaturized pulsed ion counter with thin windows and filled
with P10 gas, developed by the RARAF group [2],
which is also transparent to light (Fig.5). We note here
that this detector cannot be employed “before” the target for the “aiming at cells” approach because the thin
windows (2.5 µm optical clear mica) would slightly
deviate the beam, whose position would be lost. This is
not an issue for us, because with IPEM we measure the
strike point of each ion, in close proximity (a few microns) upstream of the cells.
FIGURE 3. Prototype of alpha particle table-top IPEM, with
the Mepsicron PSD on top of the Microscope.
Combining the improved light collection efficiency
of this microscope, the q.e. = 0.25 of the Mepsicron
PSD, and the use of a 10 µm thick BC400 blade, we
can foresee an alpha particle detection efficiency approaching 100%. The anticipated exposure rate for
such a system using the existing 2 mCi Po source
should be more than 10 hit cells per second. This rate
would increase nearly two orders of magnitude with the
new 100 mCi alpha source.
We employ a “pin-diode” downstream from the
sample for PSD noise rejection (of some hundreds of
Hz), and this requires a thin sample. A final pin diode
is what most focused and unfocused beam installations
use for single cell exposure experiments, but as noted in
the introduction, this should be avoided, because cells
prefer a thick medium.
Thin (10 µm)
phosphor blade
window
Pulsed Ion
Counter
Alpha source tip
PSD COINCID
Thick culture medium
Optical microscope + PSD
FIGURE 5. Schematic of the “Up Side Down” Irradiation
geometry of thick cell culture samples.
367
CONCLUSION
REFERENCES
A new irradiation device, based on a table-top optical microscope with photon position detection and a
high activity alpha particle source, has been invented
and built. Although it was originally planned for Radiation Effects Microscopy of ICs, it could also find
considerable application as a new radiobiology tool.
Performance evaluation is still taking place and we
have presented data here only for an alpha particleIPEM prototype. This non-optimized system shows a
reduced resolution of 10s of microns and a prominent
halo. This poor resolution could preclude most of the
radiological applications, if remedies are not found.
However, improvements look quite possible and are
based on 1) better preparation of the sample, 2) reducing lateral scattering of photons in the phosphor, and 3)
eliminating stray light photons. If all of these issues are
resolved, the alpha-IPEM device would become appealing for radiobiology applications for four reasons:
(1) It will be extremely inexpensive and have a
very low equipment footprint (a “table-top” device),
(2) It will be possible to strike cells in a relatively
large field in air, and at high speed without the need of
a fast mechanical displacement system,
(3) Irradiation can be vertically oriented and in an
advantageous bottom to top geometry, with a suitable
noise rejection detector, and
(4) Alpha particles remain the most-often used and
interesting ion in this kind of research.
1. Folkard, M., Prise, K.M., Vojnovic, B., Newman, H.C.,
2.
3.
4.
5.
Roper, M. J., and Michael, B., Int. J. Radiat. Biol. 69,
729 (1996).
Randers-Pherson, G., Geard, C. G., Johnson, G., Elliston,
C. D., and Brenner, D. J., The Columbia University Single-Ion Microbeam, Radiation Research 156, 210-214
(2001).
Doyle, B. L., Vizkelethy, G., Walsh, D. S., Senftinger,
B., and Mellon, M., Nucl. Instr. and Meth. B 158, 6
(1999).
Doyle, B. L., Walsh, D. S., Renfrow, S. N., Vizkelethy,
G., Schenkel, T., and Hamza, A. V., Nucl. Instr. And
Meth. B 181, 199-210 (2001).
Rossi, P., Vittone, E., Doyle, B.L., Cherubini, R., Battistella, A., Gennaro, G., Giacomelli, L., Giudice, A. Lo,
Manfredotti, C., and Zafiropoulos, D., ALBA: AlphaLight Based Analysis. A new approach to Single Cell Irradiation, Radiation Research, 158 (#3), 378-379 (2002).
6. Quantar Technology Inc., Santa Clara, CA.
7.
ACKNOWLEDGMENTS
Sandia is a multi-program laboratory operated by
Sandia Corporation, a Lockheed Martin Company, for
the United States Department of Energy under Contract
DE-AC04-94AL85000. Work supported in part by the
Italian “Istituto Nazionale di Fisica Nucleare” (INFN).
Work at UNT supported in part by the National Science
Foundation, Office of Naval Research, Texas Advanced
Technology Program, and the Robert A. Welch
Foundation.
368
Rossi, P., Doyle, B. L., Banks, J. C., Battistella, A., Gennaro, G., McDaniel, F.D., Mellon, M., Vittone, E., Vizkelethy, G., and Wing, N.D., Ion Photon Emission
Microscopy, to be published in Nucl. Instr. And Meth. B,
proceedings of the ICNMTA8 (Sep 2002), Takasaki, Japan.