Marianne Sowa Resat, Gregory A. Kimmel, John H. Miller, Joe C.
McDonald, Mark K. Murphy, Daniel J. Strom, Brian D. Thrall, Noelle F.
Metting and Steven D. Colson
Environmental Molecular Sciences Laboratory, Pacific Northwest National
Laboratory, P.O. Box 999, MS K8-88, Richland WA 99352 USA
Currently, the standards for human exposure to ionizing radiation are determined by extrapolating data obtained at high doses to the low dose regime using a linear no threshold model. This extrapolation of the radiation risks presented from low doses implicitly assumes that cells in an irradiated population respond individually, not collectively.1) When alpha particles enter tissue they traverses only a few cells before stopping. Therefore, unless the fluence is extremely high, only a small portion of the exposed population receives a direct dose. For such high-LET radiation, it has been shown that the number of cells responding to the radiation exceeds the number of cells actually hit.2) The bystander effect refers to the observation of a biological response in the absence of direct irradiation. Although bystander effects have been demonstrated for high-LET radiation, they have not been shown for low doses of low-LET radiation. In an effort to directly address this latter case, we have developed a novel single cell irradiation device. This device has been designed such that high energy electrons deposit energy in a pre-selected subset of cells for which the un-irradiated neighbors can be easily identified. By targeting individual cells with a highly spatially resolved dose, the biological responses of a single irradiated cell or a bystander can be studied.
The device consists of a pulsed electron beam capable of operating at energies from 20 to 80 keV. Rather than making use of a tightly focused beam where it is difficult to achieve sufficiently low electron fluxes, we have instead chosen to use a broad source (~ 5 mm) combined with a spatial collimator. The electron gun is housed in a standard vacuum chamber pumped by a turbo molecular pump (base pressure 1 x 10-8 Torr). The chamber is equipped with a Faraday cup for monitoring beam current and an optical shutter to ensure no electron dark current between pulses.
The spatial resolution of the device is achieved in two stages by passing the electron beam first through a pre-collimator and then through a high aspect ratio hole (~15:1) in the final collimator/interface platen. The pre-collimator is fabricated from an Al disk with a 0.005" hole for transmitting electrons. Ten µm of gold is plated on the exit side of the pre-collimator. The pre-collimator serves to reduce the number of electrons hitting the final collimator and decreases the chances for x-ray production at the cell interface.
After exiting the pre-collimator, the electrons hit the collimator platen, which establishes the final spatial resolution of the electron beam. The platen has been designed to minimize the production of x-rays while optimizing the spatial resolution of the delivered dose It is fabricated from a 0.15" thick aluminum disk. A 0.05" hole is drilled leaving an aluminum membrane 25 µm thick, sufficient to stop 50 kV electrons. This membrane is then coated with 10 µm of gold. To provide the final beam collimation, a 2 µm hole is laser driller thru the Al/Au membrane. Both the collimator and pre-collimator can be constructed with one hole or a series of holes depending on the biological experiment of interest.
The vacuum interface is obtained by coating the platen with a 200 nm polyimide window. The irradiation device is interfaced with a biological irradiation chamber mounted on an X-Y scanning stage of a standard optical microscope. Cells are plated on a 1.5 µm mylar membrane that is place directly on the electron gun interface. Thin, low density films of polyimide and mylar are highly transparent to energetic electrons. The device can deliver doses of a few to 100's of electrons per cell. At very low fluences, the number of electrons actually passing thru the hole will be governed by Poison statistics. The feasibility of the design has been investigated by a Monte Carlo simulation for irradiation of a cellular monolayer.3) The results indicate that the majority of the calculated beam spreading would be contained within a volume typical of mammalian cell lines.
The dosimetry of the device has been measured using several different methods. The electron beam is characterized by measuring the current on a Faraday both before and after collimation. Very low currents are measured in pulse counting mode using a channeltron. Imagining of the beam is done on a ruggedized phosphor screen.
Dosimetric measurements are performed using GAFCHROMICÒ HD-810 dosimetry film. These films are suitable for beam profiling and dose mapping over a wide range of absorbed dose. The films are composed of materials with low atomic number so there is minimal alteration of the radiation field. Film calibrations were made at various doses using a 60Co source. Films were analyzed using a CCD camera/optical microscope arrangement to measure optical density differences and indicated good spatial localization of the dose. Using a Canberra Ge spectrometer, measurements have also been made to check for the production of X-rays under various conditions. We find that the contribution from X-rays are a minimal component of the total dose delivered to the cell.
REFERENCES
1) R. Cox, Int J Radiat Biol 73 (1998) 373-376.
2) H. Nagasawa, J. B. Little, Cancer Res 52 (1992) 6394-6396.
3) J. H. Miller et al., Radiat Environ Biophys 39, (2000) 173-177.