2.1.

Cell Culturing

Normal skin fibroblasts GM 1717 (p 9) from a 40-yr-old male adult and cells from an XP patient of complementation group A (XPA), XP12BE (GM05509A; p 12; 17-yr-old female) as well as XP cells with low repair (25%; GM05293; p 15; 6-yr-old male) were obtained from the Human Genetic Mutant Cell Repository (Camden, New Jersey). Normal cells from a 40-yr-old male (CRL 1221; p 16-22), repair deficient XPA cells (CRL 1223; p 15; 10-yr-old female), Cloudman S91 mouse melanoma cells of clone m-3 (CCL 53.1; p 40), and human white melanoma cells from a 53-yr-old man (CRL 1585; p 5) were purchased from the American Type culture collection (Rockville, Maryland). Foreskin-derived normal human fibroblasts 3229 derived from a 1-day-old boy were a gift from Dr. Robert Zimmerman (Boston, Massachusetts) and used in passage 13 or 117. Normal cells were plated 1:2, XP cells 1:3, and melanoma cells 1:10. For the determination of ultraweak photon emission, cells were cultured in tissue culture plastic flasks (surface, 75 cm², Gibco Basel, Switzerland) in 15 ml Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Basel, Switzerland) supplemented with 10% fetal calf serum and 100 units (U) ml⁻¹ penicillin-streptomycin, as previously described,^{10 22} and grown for 3 to 7 days, depending on the cell type, until confluency. Cells were counted in a hemocytometer from Neubauer (Flow Laboratories, Baar, Switzerland) and cells were counted in triplicates (±10%). Cells were frozen in liquid nitrogen using the cryopreservation apparatus from Biotech Research Laboratories (Rockville, Maryland). Controlled gradual temperature reduction during cryopreservation was critical for the maintenance of cell life and viability. This apparatus preserves cells at the rate of 1°C/min when placed in a −70 °C freezer. After 5 h or overnight, frozen samples were transferred to the liquid nitrogen container. A special transport storage system for liquid nitrogen (Cryo Diffusion, Lery, France) was used to transfer the cells to Catania (Italy) to perform ultraweak photon emission measurements. For this analytical procedure, cells were prepared as follows: all cells were gently thawed, centrifuged at 1200 rpm, and the medium was discarded. The cells were washed two times in 5 mL phosphate buffered saline (PBS), centrifuged at 1200 rpm, and resuspended in 250 μL PBS. For ultraweak photon emission measurements, 100 μL of this ultimate cell suspension was used.

2.2.

Delayed Luminescence Measurements in Mammalian Cells after UVA Laser Induction

The delayed luminescence or light-induced ultraweak photon emission has been measured using²⁶ the device ARETUSA, developed at the Southern National Laboratory in Catania. This measuring system consists of a cooled single-photon-count photomultiplier (PMT), a highly efficient optical system able to collect the emitted light from the sample and to transmit it to the PMT, a high-intensity pulsed nitrogen laser with pulses at 337 nm, an electronic device able to deactivate the PMT during the laser pulses and an electronic setup able to coordinate the measurement procedure and to collect, analyze, and elaborate the signals from the PMT. A block diagram of such a system is shown in Fig. 1.

Figure 1

Block diagram of the highly sensitive UV-A-laser-induced ultraweak photon emission detection system.

We used the Hamamatsu R1878 PMT. It is a multialkali photocathode having a spectral response from 300 to 850 nm with a maximum response at 420 nm. The effective area is small (only 4 mm in diameter) and therefore a high optical efficiency of the setup is required. However, under these conditions, the background noise is lowered to 0.01 compared to the 2 in. area of PMTs, which we described previously.^{9 17} To further reduce the noise, the PMT was cooled to −20 °C, using an original forced circulation cooling system (ethanol as liquid), which is placed in direct contact with the lateral surface of the PMT, as shown in Fig. 1. We placed 100 μL of cell suspension directly, without using any cuvette, as depicted in the same Fig. 1, on the upper quartz termination of a liquid light guide NT 53-691 (Edmund Optics, New Jersey), which shows a transmission coefficient greater than 50% in the range of 300 to 650 nm.

As the light source, we used a pulsed nitrogen laser (Laser Photonics LN203C, Lake Mary, Florida) characterized by a wavelength of 337.1 nm, a pulse width of 5 ns, and an energy of about 100 μJ/pulse (±3% standard deviation). The laser was used in a single pulse mode. In short, the sample was illuminated by a single pulse and immediately afterward the DL was measured. Every measurement required about 2 s (the operation frequency of the laser was 0.5 Hz, which is roughly two orders of magnitude less than its maximum operation frequency). To receive a uniform illumination of the measured samples (a hemisphere having a diameter of approximately 5 mm) we used a lighting system based on a trifurcated quartz fiber. The three terminations surround the sample at a constant distance with an angle of 120 deg to each other (see Fig. 1). Under these conditions almost all of the light emitted by the laser source (as controlled by an optical power meter) reached the sample with a very high power per unitary volume (about 6×10¹¹ W/m ³) .

One of the main problems in ultraweak photon emission measurements is the necessity to protect the PMT during the illumination time. This action was performed until now by electromechanical shutters whose operational times are^{9 27} of the order of 150 ms. In the system presented here, an electronic gate was developed that was controlled by a digital signal. This device is able to establish on all dynodes the same tension as a normal voltage divider circuit, when the digital signal has a lower level, while it inverts the sign of the voltage between the cathode and the first dynode and between the third and the second dynode if the digital signal has its higher level. Using this protocol, it enables us to start the photon counting only a few microseconds after the laser pulses have reached the sample.

The measurement process is performed and controlled by a microcomputer boosted through an Ortec multichannel scaler (MCS) plug-in card characterized by dwell times ranging from 100 ns to 1300 s, a memory length of 65.536 channels, and input counting rates up to 150 MHz. The signals comings from the PMT are first processed by an octal discriminator (Lecroy 6408B) and then the transistor-transistor logic output (TTL) is sent to the MCS for acquisition.

The synchronization of the process is managed by MCS. Its start-out TTL signal is split in three identical signals, and each of them is addressed to a gate and delay generator (Ortec GG 8000) to temporarily protect the PMT during the sample irradiation. After a delay time of about 6 μs, the inhibition stops and the acquisition can begin.

3.1.

Delayed Luminescence of S91 Melanoma Cells after UVA-Laser Irradiation

Table 1 shows a comparision between our classical design and our new setup. As depicted in this table, the improvement factor for the detection of ultraweak photon emission in XPA fibroblasts using the classical design^{9 10} compared to the new UV-A laser induction procedure is roughly 10⁴. Note that our classical device uses a mechanically based closing shutter system in the time range higher than 100 ms and cell quantities up to 10 ml. Our new detection procedure can measure ultraweak photon emission in cell quantities as low as 100 μL, which is a factor of 100 lower compared to the previous procedure (10 mL). Therefore, the improvement is increased by a factor of up to 10⁶.

Table 1

Figure 2(a) shows two fully independent experiments for emission of delayed luminescence following UV-A laser induction at 337 nm in Cloudman S91 mouse melanoma cells at a cell density of 4 million cells/mL. It is evident that our new technique enables us to determine this light-induced ultraweak photon emission in a very short time interval of less than 10 μs, leading to the observation of a light burst of more than 10⁷ photons.

Figure 2

Delayed luminescence of Cloudman S91 mouse melanoma cells of 4×10⁶ mouse melanoma cells after preillumination with 337-nm UV-A laser light. Error bars shown by arrows represent standard deviations of three fully independent experiments. Each experiment represents an average of 100 determinations in the same cells. Standard deviations are less than 5% up to 1 ms. Background values obtained with medium are not subtracted. (a) (▪) Experiment 1 and (○) experiment 2. (b) Experimental data (○) compared to hyberbolic decay (line). The first value after 8.5 μs represents 11.4×10⁶ counts; the last value after 62.1 ms corresponds to 931 counts. To outline the oscillations, the ratio R between the data trend and the hyperbolic fitting is calculated (□).

As shown in Fig. 2 for S91 cells at a cell density of 4 million/mL these two experiments are almost identical from 8.5 μs to 62 ms. Figure 2(b) confirms a further remarkable feature of DL, as calculated by the ratio R between the data trend and the hyperbolic fitting and previously described:^{15 28} the fact that cells display sinusoidal oscillations around the decay function. Note, however, that this phenomenological observation has not received any full theoretical explanation until now.

Because several laboratories described in cells following induction with white light ultraweak photon emission curves as a hyperbolic decay function,^{11 12 15 17 22} we tested additionally this observation with 1×10⁶ cell/mL in comparison to the control of PBS, as depicted in Fig. 3. While the cells show a high linear regression coefficient of 0.999 [see Fig. 3(b)], almost ideal to 1, the control value of PBS was with 0.989 slightly lower, as shown in Fig. 3(a).

Figure 3

Ultraweak photon emission dynamics of 1×10⁶ mouse melanoma cells (b) and PBS (a) after irradiation with 337-nm UV-A laser light. Points represent the average of 100 determinations. Standard deviations are less than 5%. (a) Experimental data (•) compared to hyberbolic decay (line) in PBS. First value after 8.5 μs represents 2.9×10⁵ counts. The calculated regression coefficient is 0.989. (b) Experimental data (○) compared to hyberbolic decay (line) in S91 cells. First value after 8.5 μs represents 3.8×10⁶ counts. The calculated regression coefficient is 0.999.

The DL in PBS shown in Fig. 3(a) is due to a contribution from the quartz window on the inner light guide, which could unfortunately not be excluded. Therefore, when a drop of PBS, was placed on the top of the light guide, a significantly lower light signal, compared to cell suspension, was measured, as shown in Fig. 3(a). Nevertheless, because no DL was shown from the liquid itself using the classical design,²⁷ this DL of PBS is clearly due to some contribution of the quartz window of the light guide because the drop acts as a hemispherical lens deviating the laser light on the top of the fiber.

A linear increase with higher cell densities was found in this mouse melanoma cell system for the first measured value after 8.5 μs in the density range from 1 to 8 million cells/mL, as described in Fig. 4. The control value of PBS in the range of 10⁵ photons was very low.

Figure 4

Initial rates of ultraweak photon emission of Cloudman S91 mouse melanoma after preillumination with UV-A laser light in nanoseconds at suspension densities ranging from 0 to 8×10⁶ mouse melanoma cells. Mean values with standard deviations of three independent experiments are given. One experimental point included in the graph represents the average of 100 determination with standard deviation less than 5%.

3.2.

UVA-Laser-Induced Ultraweak Photon Emission of Normal, Cancer-Prone, and Melanoma Cells as a Biophysical Marker of Cell Stage

Figure 5 depicts DL following UV-A laser induction in low passages (13) and high passages (117) of foreskin derived human skin fibroblasts 3229. Similar results were obtained with normal skin fibroblasts CRL 1221 (data not shown). As shown in Table 2, there is a significant tendency that normal cells can be induced more significantly by the UV-A laser system than cells originating from melanoma. To exclude the influence of melanin on ultraweak photon emission as shown before in S91 melanoma cells,²² white melanoma cells from a 53-yr-old donor were additionally used. A similar observation of reduced DL for melanoma cells was found when cancer-prone XP cells were used, as shown in the same Table 2.

Figure 5

Initial rates of DL from normal young skin fibroblasts 3229. Comparison of low passage number 13 with high p 117 after preillumination with UV-A laser light in nanoseconds at suspension densities ranging from 1 to 8×10⁶ cells. Mean values n with standard deviation of three independent experiments are given. One experimental point included in the graph represents the average of 100 determinations with standard deviation less than 8%.

Table 2