**2.3. Oxygen self-referencing micro-biosensor**

Self-referencing polarographic (SERP) micro-sensor technology was developed at the Bio-Currents Shared Resource [11–13]. These sensors are made of borosilicate glass capillaries (1B150, WPI) which are pulled to outer tip diameters of ~3 μm using a Sutter P-97 (Sutter Instrument, CA). Then, a 25-μm diameter gold wire (Alfa Aesar, Ward Hill, MA) is electro‐ chemically etched in an aqueous solution of 1N HCl to reduce the tip diameter to ~1 μm. After being rinsed with water and isopropyl alcohol, the etched Au wire is inserted into the capillary so that the wire protrudes slightly from the pipette tip. The electrode tip is dipped into UV curing epoxy (429, Dymax, Torrington, CT). The exposed Au electrode is then etched again (same conditions as above) to form a recessed electrode with a cavity 2–3 μm deep. Finally, the electrode is coated by dipping it in a solution of 10% cellulose acetate (30 kDa) for 60 s and drying for 5–10 min. The electrochemical sensor itself is attached to a BRC amperometric head stage via a modified BNC connecter. An L-shaped Ag/AgCl reference electrode, connected via a 3M KCl/5% agar bridge (reference probe) placed in the bulk solution, completes the circuit (**Figure 2**). Selectivity for amperometric electrodes is usually defined by the conditioning, operating voltage, and excluding membranes.

**Figure 2.** Self-referencing micro-biosensor detection system.

### **2.4. System integration**

resolution. The first radiobiology experiment was to analyze metabolic oxygen consumption in individual living lung cell after sub-cellular irradiation and to explore the radiation response of such cells. The self-referencing oxygen electrochemical system was developed at the Bio-Currents Shared Resource at the Marine Biological Laboratory (MBL), Woods Hole, MA. It was integrated with the RARAF single-particle, single-cell microbeam to form a single-cell

Self-referencing polarographic (SERP) micro-sensor technology was developed at the Bio-Currents Shared Resource [11–13]. These sensors are made of borosilicate glass capillaries (1B150, WPI) which are pulled to outer tip diameters of ~3 μm using a Sutter P-97 (Sutter Instrument, CA). Then, a 25-μm diameter gold wire (Alfa Aesar, Ward Hill, MA) is electro‐ chemically etched in an aqueous solution of 1N HCl to reduce the tip diameter to ~1 μm. After being rinsed with water and isopropyl alcohol, the etched Au wire is inserted into the capillary so that the wire protrudes slightly from the pipette tip. The electrode tip is dipped into UV curing epoxy (429, Dymax, Torrington, CT). The exposed Au electrode is then etched again (same conditions as above) to form a recessed electrode with a cavity 2–3 μm deep. Finally, the electrode is coated by dipping it in a solution of 10% cellulose acetate (30 kDa) for 60 s and drying for 5–10 min. The electrochemical sensor itself is attached to a BRC amperometric head stage via a modified BNC connecter. An L-shaped Ag/AgCl reference electrode, connected via a 3M KCl/5% agar bridge (reference probe) placed in the bulk solution, completes the circuit (**Figure 2**). Selectivity for amperometric electrodes is usually defined by the conditioning,

irradiation response detection platform.

114 Radiation Effects in Materials

**2.3. Oxygen self-referencing micro-biosensor**

operating voltage, and excluding membranes.

**Figure 2.** Self-referencing micro-biosensor detection system.

For single-cell irradiation response measurements, sensor access during and after irradiation for precision location of damage within single cells with the imaging system is crucial. Microprobe measurements without irradiation are usually performed on an inverted micro‐ scope giving open access from above for probe placement. With microbeam irradiation, there is a constraint from the exit window below the sample, so microscopy and probing must be done from above the sample. Access requires that the tip of the probe (microelectrode) approaches a single cell, at an angle of between 20° and 30°, to within microns of the plasma membrane. Access for the reference electrode is also required. Both the measuring probe and the reference electrode have body diameters of 1.5 mm, while the measuring tip is drawn to a point. A Nikon 10× long working distance (4 mm) dry microscope objective is placed with the probe in a sample dish. Because the probe approach angle is so severely constricted, the angle setting technique appropriate for an open-access system is completely inadequate. An offset hinge manipulator was designed (**Figure 3**) and built which allows rapid repeatable reposi‐ tioning of the probe and simple angle adjustments. The hinge was constructed in a stacked configuration using high-precision flex pivots in such a way that angular settings between 10° and 60° can be set. In order to use the manipulator and the stacked hinge to satisfy our needs, a universal mounting car (Thorlabs, NJ) riding on an optical rail with an integrated robotic retraction mechanism is used. The probe manipulators are mounted accurately on the car for simple interchange. This robotic manipulator structure and the associated fully integrated control systems allow us to meet all the micromanipulation, and capillary probe placement needs in an efficient manner.

#### **2.5. Measurement**

#### *2.5.1. Cell preparation*

The human telomerase reverse transcriptase (hTERT) immortalized human small airway epithelial (SAE) cells were thawed from liquid nitrogen and cultured in fresh medium. Cells were diluted in fresh medium, and cultured cells were maintained at 37°C in a humidified 5%

**Figure 3.** Electrochemical micro-sensor mounted on microbeam end station by the offset hinge system.

CO2 incubator. The microbeam cell culture dishes were custom-made for cell growth and cell irradiation. They are made of Falcon 60 mm Petri dish, and a 0.25-inch diameter hole is drilled into the center of the dish bottom. The polypropylene film covered on the bottom of microbeam dish wells was treated with Cell-Tak (BD Biosciences) to enhance cell attachment. Also it allows the chosen radiation to get through to the cells while allowing them to be placed upright on the microbeam end stations with minimal distance between the dish bottom and beam exit window. Dishes were incubated at 37°C for 30 min and were then rinsed. Then, the cells were trypsinized and diluted to 1.5 × 104/ml (about 30 cells in a total volume of 2 μl medium). A sterile 18–22 mm square coverslip covered the well after cells in a droplet were plated using a micropipetter as close as possible to the center of the dish. The dishes were placed in an incubator until cells attached to the polypropylene. After cell attachment, the coverslips were removed and 5 ml more of medium was replenished to the dishes. Cells will typically flatten out within 1–3 h. The cells were stained by exposure to a 50 nM solution of the vital DNAbinding stain Hoechst 33342 for 30 min prior to radiation. This low stain concentration necessitates the use of an EMCCD camera (Princeton Instrument) to obtain a high-contrast image and allows rapid location of the cell nuclei to be hit, or not hit, as the experiment calls for during irradiation.

### *2.5.2. Radiation beam setup*

Beam size and beam location were measured with a nickel knife edge scan [7]. Beam location was registered using scanning of fluorescent beads. This was done by placing a microbeam dish containing fluorescent beads on the stage. Then, an isolated bead was moved to the approximate beam position. A spiral energy loss mapping scan was conducted with a solidstate charge particle detector. Once the center of the beam was identified, the bead was moved to that position (the center of beam mapping) and the beam location was registered with the imaging system. The low-magnification pictures of the bead were taken, and a center‐of‐

**Figure 4.** Measurement setup on microbeam end station.

gravity program routine was used to transfer the bead coordinates to the computer. Then, the cell dish was removed from incubator and was mounted on the microbeam stage. The chosen cell was moved to the registered beam position with an XYZ stage (MadCity Labs, Inc. WI), while monitored with a Nikon Eclipse E600 microscope and a reference picture was taken with an attached EMCCD camera (**Figure 4**). Then, the micro-electrochemical sensor and reference electrode were mounted on the XYZ stage, and the sensor/probe was carefully moved close to the cytoplasm (about 10–15 μm away) monitored with a Nikon 10× objective lens, and a background/control measurement was run for about 30 min with the sensor moved rapidly between two positions 15 μm apart at a frequency of 0.3 Hz. Data were collected at each pole position for approximately 1 s or 70% of the cycle time. The current signals were averaged at each position, and then, a differential current was obtained that can be converted into a directional measurement of flux using the Fick equation. Referencing the signals in this manner has the advantage that sources of interference caused by random drift and noise are effectively filtered from the signal and fluxes can be monitored in real time.

#### *2.5.3. Radiation and real-time oxygen consumption measurement*

During the single-cell microbeam irradiation, the current changes were monitored with biosensor in self-referencing mode. The number of helium ions was set at 20 or 30 in the microbeam irradiation protocol for this single-cell irradiation. The particle beam count rate was measured at about 200 per second. A pulser (Ortec Inc, TN) was used to simulate the real count rate and to control the beam shutter, because the 5.6 MeV helium ions cannot pass all the way through the cell and the medium (~1 mm thick) without being absorbed. In the cytoplasm irradiation experiments, an exclusion zone around each fluorescing nucleus is automatically generated to ensure that the cytoplasm target positions from one cell are not accidentally within the nucleus of an adjacent cell. Mutation induction caused by cytoplasmic irradiation has been reported using this technique. The image analysis system defines the long axis of each cell nucleus, after which the computer system delivers particles at two target positions along this axis, 6 μm away from each end of the cell nucleus. During the experiment, an in-house code with Matrox Genesis imaging library has been used to handle the images (subtract background, correct for illumination variation, locate cells, and record location of nearest frame).

#### **2.6. Results**

CO2 incubator. The microbeam cell culture dishes were custom-made for cell growth and cell irradiation. They are made of Falcon 60 mm Petri dish, and a 0.25-inch diameter hole is drilled into the center of the dish bottom. The polypropylene film covered on the bottom of microbeam dish wells was treated with Cell-Tak (BD Biosciences) to enhance cell attachment. Also it allows the chosen radiation to get through to the cells while allowing them to be placed upright on the microbeam end stations with minimal distance between the dish bottom and beam exit window. Dishes were incubated at 37°C for 30 min and were then rinsed. Then, the cells were trypsinized and diluted to 1.5 × 104/ml (about 30 cells in a total volume of 2 μl medium). A sterile 18–22 mm square coverslip covered the well after cells in a droplet were plated using a micropipetter as close as possible to the center of the dish. The dishes were placed in an incubator until cells attached to the polypropylene. After cell attachment, the coverslips were removed and 5 ml more of medium was replenished to the dishes. Cells will typically flatten out within 1–3 h. The cells were stained by exposure to a 50 nM solution of the vital DNAbinding stain Hoechst 33342 for 30 min prior to radiation. This low stain concentration necessitates the use of an EMCCD camera (Princeton Instrument) to obtain a high-contrast image and allows rapid location of the cell nuclei to be hit, or not hit, as the experiment calls

Beam size and beam location were measured with a nickel knife edge scan [7]. Beam location was registered using scanning of fluorescent beads. This was done by placing a microbeam dish containing fluorescent beads on the stage. Then, an isolated bead was moved to the approximate beam position. A spiral energy loss mapping scan was conducted with a solidstate charge particle detector. Once the center of the beam was identified, the bead was moved to that position (the center of beam mapping) and the beam location was registered with the imaging system. The low-magnification pictures of the bead were taken, and a center‐of‐

for during irradiation.

116 Radiation Effects in Materials

*2.5.2. Radiation beam setup*

**Figure 4.** Measurement setup on microbeam end station.

To detect physiologically driven molecular movements around the cell membrane, which is normally very small (fA current), the drift and background have to be taken care of with selfreferencing technique. This requires extracting small electrical signals, fA differences (AC) on top of large offset signals 10s of pA (DC). With cells exposed to 10 μM anti-mycin, the basal cellular O2 flux was tested and the changes in oxygen concentration dependent current (DC changes) were detected. Then, both cytoplasm and nucleus irradiations with 100% cell radiation were conducted. **Figure 5** shows the background and drifting (DC) during the radiation. A very obvious O2 current change (AC) was recorded within seconds after cyto‐ plasm irradiation. **Figure 6** shows the AC current change from two measurements with two

**Figure 5.** Self-referencing oxygen ion–selective probe result with alpha microbeam irradiation (DC).

different probes. The small changes in oxygen consumption could be extracted from the background oxygen concentration with the self-referencing method. A similar but relative small change was observed after nucleus irradiation. Without radiation, any of these large spikes cannot be seen on a long-time background measurement. Twelve measurements were conducted resulting in a success rate of ~30% as determined by the individual cell flux test because of the cell-to-cell variations and the uncertainty of the probe locations. The results indicate a role for mitochondrial damage following irradiation and enable further evaluations of the radiation-induced bystander effect. This effect is hypothetically due to the result of damage signals received by non-hit cells from hit cells. Establishing the mechanistic basis for such responses in the form of damage signaling from hit to non-hit cells and continued signaling has proven to be elusive. However, evidence for both oxygen- and nitrogen-based small molecules and mitochondrial dysfunction has been produced. The approach outlined in this study suggests biosensor mediators in the form of oxygen radicals, nitric oxide, and hydrogen peroxide can be directly measured at a single-cell level in both hit cells and bystander cells providing an incisive method of evaluating such evidence. This establishment of a noninvasive, self-referencing biosensor/probe system in conjunction with the RARAF microbeam provides an additional means for probing biological responsiveness at the level of individual cell, after precise sub-cellular targeting in hit cells and bystander cells.
