Induction of sister chromatid exchanges in fibroblasts from normal donors and from patients with xeroderma pigmentosum after combined treatment with ultraviolet radiation and modulated low frequency electric currents

ARTICLE

Although some experimental evidence over the past two decades points to potential cytogenetic effects caused by low frequency electromagnetic fields, this matter is highly controversial. In view of the fact that these fields not only play an important role in daily life but are also applied in therapy, it is necessary to examine their effects on genetic material.

Studies conducted by Knedlitschek and coworkers among others have shown that signal transduction processes mediated by second messengers such as cyclic AMP or calcium become modulated by interaction of electromagnetic fields with receptor structures or ion channels [1]. By activating protein kinases, this primary biochemical signal may even lead to cellular responses at the level of gene transcription, DNA- and protein synthesis [2-4]. In particular, electromagnetic fields have been shown to display different effects on proliferation and protein synthesis in lymphocytes from healthy donors and from patients with chronic myeloid leukaemia [5, 6] suggesting field actions depend upon genetic factors.

Induction of sister chromatid exchanges (SCE) is a key process associated with cytogenetic alterations, is thus a suitable indicator for studying the effects of low frequency electromagnetic fields. In a recent publication [7], SCE-induction in dermal fibroblasts was observed when the cells were treated with so-called interferential current (IFC), a waveform widely used in physiotherapy which is applied via skin electrodes. The maximum effect was found for cells from patients with xeroderma pigmentosum at a field strength of 1 V/m (50 Hz), corresponding to an amperage of 1 mA. Raising the current to 4 mA, however, did not result in higher SCE rates and even lowered the response at 50 Hz. This observation and results obtained by Liburdy [8] led the authors to the conclusion that SCE-induction by IFC is not an expression of cytogenetic damage as for example caused by UV [7].

IFC is generated by superposition of two alternating currents of equal amplitudes but with slightly different frequencies close to 4,000 Hz. Due to interference of the two currents, IFC is an alternating current with approximately this frequency but with its amplitude varying periodically between zero and a maximum value in proportion to the difference in frequency between the two currents. This "modulation" frequency of amplitude quantitatively and qualitatively influences cellular response [1]. IFC is largely free of side effects and is therefore widely used as a very effective treatment e.g. in physiotherapy. A recent study conducted by Wolf et al. has shown that IFC has a pronounced antipsoriatic effect comparable to anthralin [9]. In view of its good compatibility it might become an alternative to antipsoriatic drugs.

As the study by Fuhrmann et al. [7] has demonstrated, effects of IFC alone are rather small but it might be able to modulate effects of other agents. Here we present the results of a study on SCE-induction in human dermal fibroblasts where IFC was used in combination with ultraviolet radiation of 254 nm wavelength. Fibroblasts from donors without a hereditary skin disease which formed the control group, and fibroblasts from xeroderma pigmentosum (XP) patients were exposed in vitro. XP is a rare disease characterized by extreme photosensitivity and the development of multiple cutaneous malignancies at early onset. Patients with XP display a defective excision repair of UV-induced DNA photoproducts. The purpose of combining IFC with ultraviolet treatment was to find out a possible interaction between the two agents. In particular, we wanted to know whether the UV-induced cytogenetic damage could possibly be modified by IFC-treatment.

Materials and methods

Cell strains and culture conditions

Dermal fibroblasts were obtained from 10 normal donors (mean age 41.3 years) and 8 XP-patients (mean age 16.75 years). Cells were cultured in 25 cm² tissue culture flasks (Becton Dickinson) using RMPI 1640 medium supplemented with 10% fetal calf serum, penicillin and streptomycin (100 units/ml each). Incubation was at 37° C in air with 5% CO₂. Twenty-four hours prior to IFC treatment, 400,000 cells were seeded onto Falcon cell culture inserts of 25 mm diameter (Cyclopore membrane, 0.45 µm pore size). Cell density at the time of exposure was approximately 100,000 cells/cm².

Exposure to ultraviolet radiation

Fibroblasts from normal donors were exposed to UV light (254 nm) for 20 and 40 s, corresponding to doses of 2.5 and 5 mJ/cm², respectively. They were then treated with IFC as described below. Fibroblasts from XP-patients were exposed to UV for 8 and 20 s (doses 1 and 2.5 mJ/cm², respectively), either immediately before or after IFC treatment. The reduction in exposure time was necessary due to the increased responsiveness of XP-cells to ultraviolet radiation. In 5 XP-cases the influence of the order of UV- and IFC-treatment was analysed. The influence of order of treatment was not examined in normal donors.

Exposure to interferential current

Cells were exposed on the membrane filters which were placed between two rhodium-coated electrodes. All components were immersed in medium allowing IFC to flow through the filters. A full description of the system is given elsewhere [1]. IFC of 4,000 Hz was generated by means of an EDiT-2-device (Nemectron GmbH, Karlsruhe) and fed to the electrodes via an external attenuator and a digital ammeter. Exposure time was 20 min at 1 or 10 mA amperage. The electric field between the electrodes was 1 V/m at 1 mA, corresponding to a current density of 250 µA/cm² [1]. These ratings are far below the level for physiological damage to the cells. Experiments were carried out at a modulation frequency of 50 Hz.

Determination of sister chromatid exchanges

Immediately after exposure, cells were treated with trypsin, subsequently washed with PBS, seeded onto glass slides, and placed in square Petri dishes (100 x 100 mm). After the cells had settled, 20 ml of medium containing BrdU (final concentration 20 µM) were added, the dishes were wrapped in aluminium foil and incubated for another 65 hrs. Mitosis was arrested by adding 0.2 ml colcemid (0.25 µg/ml) per dish for 4 hrs. Medium was then replaced by 25 ml of 75 mM KCl (37° C) for 30 min to induce the hypotonic spreading of mitosis. The KCl solution was renewed once before fixation, which was started by slowly adding a mixture of acetic acid and methanol (1:3) at room temperature. As soon as 25 ml of fixative had been added, 25 ml of the mixture were removed. The procedure was repeated twice. Finally, the slides were treated with pure fixative for 20 min and then air-dried. Staining was performed according to the Hoechst-Giemsa method [10]. SCEs were observed at a 1,000-fold magnification and counted per metaphase. Every exchange point including those in the centromere was rated as SCE. Twenty metaphases were analysed in each specimen.

Statistical analysis

From all cell cultures, 20 mitoses were examined and a mean value of SCEs was calculated. The square root of this value was used for statistical analysis. Square root transformation was used in order to transform the non-symmetrical heteroscedastical (unequal variances) distribution in a symmetrical homoscedastical one [11-13].

The experimental design controlled the influences of person, three levels of UV, three levels of IFC and within person cell pathology. This means that cells of each person were exposed to 3 x 3 combinations of conditions of tests. Therefore the statistical analysis represents these factors of influence in analyses of variance (ANOVA). The analysis of variance was performed for both groups of donors using the general linear model (GLM) of SAS [12]. Three different ANOVAs were calculated: 1. The following factors were taken into account within the group of normal donors: a) person (10 donors within the normal group), b) UV-level (0, 2.5 and 5 mJ/cm²), c) IFC-amperage level (0, 1 and 10 mA). Interactions between UV-levels*IFC-levels, person*UV-levels and person*IFC-levels were calculated. 2. Within the collective of XP-patients the following factors were used as source of variation: a) person (8 donors within the XP group), b) UV-level (0, 1 and
2.5 mJ/cm²), c) IFC-amperage level (0, 1 and 10 mA). Interactions between UV-levels*IFC-levels, person*UV-levels and person*IFC-levels were analysed in the same way as in the normal group of donors. 3. The third ANOVA was performed considering the order of the factors (first current vs first UV), person, IFC and UV as above, interactions IFC*UV-levels, person*UV-levels, person*IFC-levels, UV-levels*order, IFC-levels*order and IFC*UV-levels*order. Results with p < 0.05 are considered significant.

Results

In Figures 1 and 2 the SCE-rates of normal and XP-fibroblasts are plotted against UV-dose with post-irradiation exposure to IFC at 0, 1 and 10 mA amperage as parameter. With the exception of the higher rate of SCE-induction by UV in XP-fibroblasts, the figures display similar features: 1) nearly additive contribution of UV and IFC to total SCE-rate between 0 and the intermediate UV-doses (2.5 and 1 mJ/cm², respectively); 2) "saturation" of SCE-rates beyond these doses which is most pronounced at 10 mA IFC-amperage; 3) some of the dose-effect curves are parallel up to the highest UV-dose, but also crossing of the lines is observed, which points to some non-linear interaction between the two agents; 4) when XP-fibroblasts are exposed to IFC prior to UV-irradiation (Fig. 3), the total SCE-rates for 1 and 10 mA amperage are always lower than for the reverse order of treatment (Fig. 2). This influence of treatment order is statistically significant (see below).

Table I shows means and means of square roots for the different exposure conditions. Table II summarizes the results of statistical analysis of all SCE-data. As can be seen, the influence of the person is significant in both groups (p{person} = 0.0001), i.e. each patient has its own individual SCE-basis rate. Interaction between person and field intensity is not significant in normal donors (p{person*IFC} = 0.5206) which means that patients of this group show a homogenous responsiveness to IFC-exposure, whereas XP-patients react heterogenously (p{person*IFC} = 0.017 which is significant). Interaction between person and UV-treatment shows a significant influence on the SCE-rate in both groups which means that the relationship between dose and effect differs from patient to patient. Treatment with UV-radiation increases the SCE-rate in both groups (p{UV} = 0.0001). The magnitude of the effect depends upon the dose. SCE-rate after UV-exposure is on a higher level in XP-patients which is a well-known phenomenon. Treatment with IFC only, shows a significant rise in SCE-rate in both groups. The effect is higher at 10 mA amperage than at 1 mA. In addition to all other influence factors (person, IFC, UV), the order of treatment was analysed in 5 XP-patients. The order of treatment has a statistically significant effect on SCE-rate (p = 0.0208): SCEs are higher when UV is given first.

Discussion

When comparing IFC and UV-radiation with respect to cytogenetic effects it has to be kept in mind that these agents differ widely in their biophysical properties. While UV-radiation of 254 nm wavelength is a destructive agent known to induce DNA-damage, the energies imparted to cellular structures by IFC-exposure are by many orders of magnitude too weak to modify or destroy any chemical bond. Moreover, as discussed, for example, by Liburdy [8], low frequency electric fields cannot effectively penetrate the cell membrane since the lipid bilayer acts as an electrical insulator. Therefore, the IFC cannot be expected to directly influence internal cell structures (e.g. the nucleus or DNA). Consequently, the mechanisms of SCE-induction by these two agents must be different.

SCEs are usually considered as indicating a cytogenetic effect due to DNA alterations caused by a damaging agent. While this explanation may hold for UV-radiation, it clearly fails to explain SCE-induction by IFC, i.e. in the absence of DNA-damage. This problem has been thoroughly discussed by Fuhrmann et al. [7] who studied SCE-formation in skin fibroblasts by IFC of different amperages and modulation frequencies. Their results can by no means be understood in terms of DNA-damage. Here we adopt the conclusions made by these authors, that SCE-induction by IFC may result from second messenger-mediated changes in gene expression. This interpretation is in line with the experimental proof that IFC causes activation of cellular signal transduction pathways [1].

In the absence of interactions between the two different mechanisms, SCE-formation after combined treatment should be statistically independent for either agent, leading to an additive contribution of UV and IFC to the total SCE-rate. This appears to be reflected by Figures 1 and 2 for the range from 0 to the intermediate UV-doses of 2.5 and 1 mJ/cm², respectively. However, above these doses, IFC-treatment tends to depress SCE-induction which is particularly evident for the XP-fibroblasts (Fig. 2). This means that at higher doses, UV-radiation and IFC interfere with respect to SCE-production, which, for example, could mean that processes supporting repair of UV-induced DNA-lesions are stimulated by IFC.

Figure 3, where UV-treatment followed IFC-exposure, seems to favour this explanation. For both UV-doses, the SCE-rates in the IFC-exposed XP-fibroblast are below the corresponding levels of Figure 2 to a statistically significant extent. When IFC is applied first, more time is left for the cells to activate and develop this repair function.

The way in which DNA-repair could be improved via IFC-mediated membrane stimuli, is, however, not clear. The fact that even XP-cells, which are defective in excision repair, profit from IFC pre-exposure points to other mechanisms of protection against SCE-formation by UV. However, even without precise knowledge of these details, the results obtained here are highly interesting, since they open the perspective of potentially mitigating adverse UV-effects. Although the effects of IFC observed here appear to be rather small in this context, it may be expected that other modulation frequencies along with an optimized time interval between current- and UV-exposure could improve IFC-efficiency.

CONCLUSION

Acknowledgements

This study was supported by the Manfred and Ursula Müller-Stiftung, Stifterverband für die Deutsche Wissenschaft, Essen, Germany. We thank Prof. Dr. E. G. Jung for providing the facilities at his clinic. The skillful and friendly technical assistance of Mrs C. Herbst is gratefully acknowledged. We also thank Mr T. Pelz for his good advice regarding the statistics.

REFERENCES

1. Knedlischek G, Noszvai-Nagy M, Meyer-Waarden H, Schimmelpfeng J, Weibezahn KF, Dertinger H. Cyclic AMP response in cells exposed to electric fields of different frequencies and intensities. Radiat Environ Biophys 1994; 33: 141-7.

2. Goodman R, Bassett CA, Henderson AS. Pulsing electromagnetic fields induce cellular transcription. Science 1983; 220: 1283-5.

3. Liboff AR, Williams T Jr, Strong DM, Wirtar R Jr. Time-varying magnetic fields: effect on DNA synthesis. Science 1984; 223: 818-9.

4. McLeod KJ, Lee RC, Ehrlich HP. Frequecy dependence of electric field modulation of fibroblast protein synthesis. Science 1987; 236: 1465-9.

5. Cadossi R, Bersani F, Cossarizza A, Zucchini P, Emilia G, Torelli G, Franceschi C. Lymphocytes and low-frequency electromagnetic fields. FASEB J 1992; 6: 2667-74.

6. Cossarizza A, Monti D, Sola P, Moschini G, Cadossi R, Bersani F, Franceschi C. DNA Repair after irradiation in lymphocytes exposed to low-frequency pulsed electromagnetic fields. Radiat Res 1989; 118: 161-8.

7. Fuhrmann E, Knedlitschek K, Wolf GK. Induction of sister chromatid exchanges by modulated, low frequency electric fields in fibroblasts from normal donors and from patients with xeroderma pigmentosum and dysplastic nevus syndrome. Eur J Dermatol 1997; 7: 85-8.

8. Liburdy RP. Calcium signalling in lymphocytes and ELF fields: evidence for an electric field metric and a site of interaction involving the calcium ion channel. FEBS Letter 1992; 301: 53-9.

9. Wolf B, Dertinger H, Wolf G. Die Wirkung von Interferenzstrom auf Psoriasis vulgaris. Akt Dermatol 1996; 22: 94-7.

10. Perry P, Wolff S. New Giemsa method for the differential staining of sister chromatids. Nature (London) 1974; 251: 156-8.

11. Tice RR, Lambert B, Morimoto K, Hollaender A. A review of the international symposium on sister chromatid exchanges: twenty-five years of experimental research. Envirom Mutag 1984; 6: 737-52.

12. Tucker JD, Auletta A, Cimino MC, Dearfield KL, Jacobsen-Kram D, Tice RR, Carrano AV. Sister chromatid exchange: second report of the Gene-Tox program. Mutation Res 1993; 297: 101-80.

13. Whorton EB Jr, Tice RR, Stetka DG. Statistical design, analysis, and inference issues in studies using sister chromatid exchange. Basic Life Sci 1984; 29 Pt A: 431-40.

14. SAS Institute Inc. SAS/STAT^® User's Guide, Version 6, Release 6.11 for OS/2, Fourth Edition, Volume 1 and 2, Cary, NC: SAS Institute Inc., 1996.

15. Butler MG, Yost J, Jenkins BB. Chromosome breakage and sister chromatid exchange analysis in computer operators. J Environ Sci Health 1987; A 22 (8): 729-41.