Microtubular structure impairment after GSM-modulated RF radiation exposure

For the experiment we used the Chinese hamster fibroblast cell line (V79) because of its well-known properties and common use in cytotoxic studies (24). Cells were maintained in a strict line with Freshney’s instructions (25), and cultivation procedure was the same as described elsewhere in detail (26).

Electromagnetic field was created inside a 5402 Gigahertz Transversal Electromagnetic Mode Cell (GTEM, ETS-Lindgren, Cedar Park, TX, USA). 915 MHz field was generated with a tracking generator and an Anritsu 2721B spectrum analyser combined with an RF 3146 power amplifier module (RF Micro Devices, Greensboro, NC, USA) and a Polaris chipset RF 2722 signal modulator (RF Micro Devices). Quality and uniformity of the produced signal was measured by an antenna connected to the spectrum analyser. RF signal was modulated to the 900-MHz GSM band in line with mobile network systems common in Europe, Asia, Africa, and mainly South America. This signal is designed for wide-area cellular operations with maximum output powers permissible for mobile operations. Although 2G is being decommissioned around the world, the selected frequency is also used in the 3G and 4G networks.

Each exposure protocol was run on three independent cell samples. The field strength of the GSM-modulated electromagnetic field (EMF) was 10, 20, and 30 V/m. The cells were exposed in parallel for one, two, and three hours.

The strength and uniformity of the electric field created inside the GTEM cell was determined with an HI 468 measuring instrument (ETS Lindgren, Cedar Park, TX, USA) and an HI 6005 test probe (ETS Lindgren). The field in reference was controlled with the Anritsu 2721B spectrum analyser (Atsugi, Kanagawa, Japan). The strength of the field created inside the GTEM cell was proven to be homogenous, with only slight variations of ±1 V/m.

Sham-exposed (negative control) cells were treated in the same manner as the exposed ones but were not irradiated at any point, that is, the radiofrequency generator was switched off during simulation.

In addition to negative control, radiation exposed cells treated with 0.1 mmol/L colchicine (Sigma-Aldrich, St. Louis, MO, USA) were used as a positive control. Colchicine is an antimitotic agent able to attach to free tubulin subunits and suppress polymerisation and destruction of microtubules.

The average specific absorption rate (SAR) established on a single-cell level was 0.23, 0.8, and 1.6 W/kg at 10, 20, and 30 V/m, respectively. SAR was calculated by averaging the individual parameters of cell components in accordance with their volume fraction in a living cell (27, 28). Following the method described by Buttiglione et al. (29) dosimetry results in our study were obtained in a homogeneous cell culture arranged in the form of plastic chamber slides filled with 5 mL of culture medium. In order to calculate SAR values, we used the medium dielectric constant (ε_r) of 75, its effective electric conductivity (σ) of 1.8 S/m (numerical calculation), and aqueous sample density of 1 kg/L as input parameters (26, 30). To provide a steady temperature throughout irradiation the temperature of the nutrient medium (contained in plastic chamber slides and placed inside the GTEM cell) was continuously measured using a temperature sensor (NTA 880, Oregon Scientific, Tualatin, OR, USA). Temperature did not rise during irradiation and kept at 36.3±0.3 °C, which corresponds to physiological cell temperature.

Microtubular proteins in irradiated, negative, and positive control cell samples were determined with indirect immunocytochemical analysis (31). In brief, cells (in the total volume of 5 mL) were initially seeded on Permanox Lab-Tek Chamber Slides (Nunc, Roskilde, Denmark) at the concentration of 2.5x10⁴ cells/mL. After 24 hours, cell were irradiated for 1, 2, and 3 h. After irradiation, the samples were washed, permeabilised, and fixed in paraformaldehyde. Microtubular proteins were marked with a complex of primary IgG anti-β-tubulin antibody produced in mouse (Sigma-Aldrich, St. Louis, MO, USA) and secondary antibody representing a conjugate of anti-mouse IgG and fluorescein isothiocyanate (FITC, Sigma-Aldrich, St. Louis, MO, USA). Once mounted in a fluorescent medium, 1,000 cells per slide were analysed with a fluorescent light microscope (400x magnification, Olympus BX61, Tokyo, Japan). Microtubule damage was evaluated by determining structural differences in irradiated cells. Changes identified as grainy fluorescent clusters were compared with those observed in positive control cells. This grainy structure suggests that microtubule fibres are highly dissipated and therefore damaged.

In order to measure cell proliferation rate, 1x10⁴ cells/mL from each group were seeded on 24-well plates and counted for six post-exposure days under a light microscope at 400x magnification using improved Neubauer counting chamber (Sigma-Aldrich, St. Louis, MO, USA).

Figure 1 shows the percentage of damaged V79 cells after 1, 2, and 3 h of exposure to GSM-modulated 915 MHz radiation. Although the cells were exposed to different electric field strengths, they showed the same pattern of time-dependent microtubular impairment. Three-hour irradiation at all electric field strengths caused significant microtubular damage. At shorter irradiations, microtubules developed normally and did not differ from negative controls (Figure 2).

Figure 1

Figure 2

Three days after the three-hour exposure, cells counts were significantly lower than in negative control (P<0.05) (Figure 3A, 3B, and 3C). Two-hour exposure to 20 and 30 V/m fields (corresponding to 0.8 and 1.6 W/kg SAR, respectively), also resulted in significantly lower proliferation rate (Figures 3B and 3C) on day 3 post-exposure. However, on day 4 post-exposure, cells counts returned to normal in every exposed group. The decline in the cell count on the fifth post-exposure day is most likely a result of cell culture confluence and depletion of nutrient in the medium, considering that parallel decrease was observed in control cells (there was no difference in proliferation rate between exposed and negative control cells).

Figure

Our results clearly show that 915 MHz radiation impairs microtubular proteins of V79 cells in a time-dependent manner. This confirms the hypothesis that electromagnetic fields in GSM frequency range might interfere with the mechanisms driving the cytoskeleton network, as this process is based on the bipolarity of the basic protein building units (32). Microtubules, as part of the cytoskeleton, meet the basic requirements for the onset of excitation that results in vibrations theoretically predicted by Fröhlich (33, 34, 35) and for the generation of an endogenous oscillating electric field (32). In a healthy cell, this endogenous electromagnetic field is perfectly balanced, but external electromagnetic fields may disrupt it.

Earlier studies, such as the one by Ortner et al. (36) showed no effect of 2450 MHz microwave radiation during in vitro polymerisation of purified microtubular proteins. Still, small differences were seen in the dynamics of light scattering curves out of the standard error of the mean of measurement. A more recent study (37), on the other hand, showed that 24-hour intermittent exposure to 1800 MHz RF radiation affected gene expression in rat neurons associated with multiple cellular functions, including cytoskeleton. In another study (38), mobile-phone radiation was found to affect the expression of a cytoskeleton protein vimentin and of stress fibres forming F-actin. Similarly, in another study (39) one-hour 902.4 MHz irradiation with SAR of 0.6 W/kg affected fibroblast morphology through increased expression of genes coding for structural proteins. Furthermore, changes in actin-binding proteins were observed in human astrocytoma cells exposed to 835 MHz at power density of 40 mW/cm², while changes in cell proliferation were detected at power density of 8.1 mW/cm² (40). All these findings support the hypothesis that cytoskeleton responds to RF radiation.

Our proliferation rate findings complement previous research. We believe that the marked drop three days after 2- and 3-hour exposure was the consequence of damage done to microtubular structure. These changes seem to be reversible, however, as the count grew back to normal on day 4 post-exposure, indicating that the cells did not lose their ability to divide. This implies that the observed microtubule damage was not significant for the cell cycle. Comparable findings were reported by Ballardina et al. (41) for V79 cells exposed to 15-minute 2.45 GHz exposure. It induced changes in the mitotic apparatus, apoptosis, and a moderate drop in proliferation rate. The authors suggested that spindle alterations were not permanent and that the length of the exposure was insufficient to make them irreversible. It is interesting to note that a similar mechanism of microtubule structure impairment is seen at low-intensity, intermediate-frequency (100–300 kHz) radiation. These fields arrest cell proliferation by interfering with the formation of mitotic spindle and cause rapid disintegration of dividing cells. Their use in tumour treatment has been approved by the US Food and Drug Administration (42).

Our study shows that V79 cells are sufficiently sensitive to be affected by GSM RF radiation, as it caused time-dependent microtubule protein impairment and, consequently, lower proliferation rate. It has also established that radiation effects on microtubules did not depend significantly on electric field strength and corresponding SAR, which calls for further investigation. A more detailed analysis of microtubule structure is also warranted to cover the entire cell cycle and look into the effects on mitotic spindle. This will be in the focus of our future research.