Effects on the brains of Wistar-Han rats exposed head-only to GSM-1800 or UMTS signals: preliminary results

OralInternational audienceOb jective. The Swedish group of Leif Salford in Lund reported the occurrence of brain damage (permeability of the blood-brain barrier and presence of dark neurons), 50 days after a single 2-hour exposure of rats to a mobile telephony GSM-900 signal (Salford L.G., Brun A. E., Eberhardt J.L., Malmgren L., Persson B.R.R. Environnemental Health Per- spectives, (2003), 111:881-883). Besides, new signals for mobile communication are being developed and their health effects have to be assessed. In its RF research agenda, WHO (www.who.emf ) emphasized as accuracy and reproducibility of published RF effects on the permeability of the blood- brain barrier and other neuropathologies (e.g., dura mater inflammation, dark neurons) Within the German BfS pro ject it was therefore planned to study the effects of GSM-1800 and UMTS signals on the blood-brain barrier and dark neurons in rats. Methods. Exposure and dosimetry: the characterization of the brain averaged SAR (BASAR) had been performed at 900 MHz using the loop antenna (Levˆeque P., Dale C., Veyret B., and Wiart J.. IEEE MTT (2004) 52: 2067-2075). The same approach was used as part of the pro ject at 1800 MHz at IRCOM, experimental measurements and numerical simulations (FDTD). Biological system and experiments: Male Wistar rats were housed under controlled tem- perature (22◦ C) and lighting conditions, and supplied with water and food ad libitum. After a one-week acclimation period, rats were randomly distributed in each experimental group and progressively trained to the rocket-type exposure setup during one week. Two exposure protocols were used: (a) 10 weeks old rats (300-325 g) were exposed once during 2 hours at 12 weeks or (b) 6 weeks old rats (200-225 g) were exposed at 8 weeks to a repetitive scheme of 2 hours per day, 5 days per week, during 4 weeks until they were 12 weeks old. Four SAR levels were used: 0.02 W/kg, 0.2 W/kg, 2.0 W/kg, and 10.0 W/kg. Sham-exposed rats (0 W/kg, restrained in a rocket), cage controls, and positive controls were included in the protocol. Injection of kainate was used as a positive control for BBB disruption (albumin leakage) and and 50 days after exposure. To ensure the blinding of the experiments, rats were coded before exposure and brains were then recoded before slicing and analysis. Rats were eu- thanized using isofurane inhalation (5% in air). The rat brains were fixed by intracardiac perfusion using a paraformaldehyde solution and frozen. Serial 10-μm-thick brain sections were prepared from 3 different brain regions (between bregma -0.80 and -1.20 mm; -4.00 and -4.80 mm; -8.00 mm and -8.80 mm). Biological parameters: The generescence were identified using (a) Cresyl-violet staining applied on the tissue sections (1-5 min) and rinsed with distilled water and (b) Fluoro-Jade staining for a more specific detection of dark neurons (Schmued L.C., Albertson C., and Slikker W. Jr. Brain Research, (1997), 751, 37-46). Briefly, tissue slices were treated with successive baths (1% NaOH and 80 % ethanol, 5 min; 70% ethanol, 2 min; 0.06% potassium permanganate, 10 min), rinsed and stained using a 0.001% Fluoro-jade solution (30 min, gentle agitation). Slices were then rinsed in distilled water and immersed in xylene. The permeation of the blood-brain-barrier was assessed by detecting endogenous albumin. Briefly, quenching of endogenous peroxi- dase was done with 0.3% H2 O2 in 0.3% horse serum (HS) in PBS. Non-specific binding sites were saturated with PBS-HS (5%) (10 min). Tissue sections were incubated during 1 h with an anti-human albumin antibody (Dakocytomation, France) and the first anti- body revealed using a indirect immunoperoxydase method (Vectastain ABC kit) (Vector SA, France). After various staining conditions, coverslips were mounted on slides before mi- croscopy observation, and analysis performed using the Aphelion image software (ADCIS SA, France). Statistics. For each exposure condition, groups of 16 rats were used allowing the detection of a significant variation (p ≤ 0.05) for an arbitrary average error of 35%(worse case arbitrary value considering variability usually reported for biological tests). Two successive series of 8 rats per exposure condition (n = 16) were performed. Statistical analysis was made using the Student t-test. Results. The proposed program is being performed over a 3-year period (beginning in 2004 and ending in 2007). Preliminary experimental results will be presented at the meetin