Abstract We investigated the effects of global system for mobile communication (GSM) microwave exposure on the permeability of the blood-brain barrier using a ...
Acta Neuropathol (1997) 94 : 465–470
© Springer-Verlag 1997
R E G U L A R PA P E R
K. Fritze · C. Sommer · B. Schmitz · G. Mies · K.-A. Hossmann · M. Kiessling · C. Wiessner
Effect of global system for mobile communication (GSM) microwave exposure on blood-brain barrier permeability in rat
Received: 18 February 1997 / Revised, accepted: 12 May 1997
Abstract We investigated the effects of global system for mobile communication (GSM) microwave exposure on the permeability of the blood-brain barrier using a calibrated microwave exposure system in the 900 MHz band. Rats were restrained in a carousel of circularly arranged plastic tubes and sham-exposed or microwave irradiated for a duration of 4 h at specific brain absorption rates (SAR) ranging from 0.3 to 7.5 W/kg. The extravasation of proteins was assessed either at the end of exposure or 7 days later in three to five coronal brain slices by immunohistochemical staining of serum albumin. As a positive control two rats were subjected to cold injury. In the brains of freely moving control rats (n = 20) only one spot of extravasated serum albumin could be detected in one animal. In the sham-exposed control group (n = 20) three animals exhibited a total of 4 extravasations. In animals irradiated for 4 h at SAR of 0.3, 1.5 and 7.5 W/kg (n = 20 in each group) five out of the ten animals of each group killed at the end of the exposure showed 7, 6 and 14 extravasations, respectively. In the ten animals of each group killed 7 days after exposure, the total number of extravasations was 2, 0 and 1, respectively. The increase in serum albumin extravasations after microwave exposure reached significance only in the group exposed to the highest SAR of 7.5 W/kg but not at the lower intensities. Histological injury was not observed in any of the examined brains. Compared to other pathological conditions
K. Fritze · B. Schmitz · G. Mies · K.-A. Hossmann · C. Wiessner1 Department of Experimental Neurology, Max-Planck-Institute for Neurological Research, Gleueler Strasse 50, D-50931 Cologne, Germany Tel.: 49-221-4726-210; Fax: 49-221-4726-325 C. Sommer · M. Kiessling Institute for Neuropathology, Ruprecht-Karls-University, Heidelberg, Germany Present address: 1 Novartis Pharma Inc., Nervous System Research, Basel, Switzerland
with increased blood-brain barrier permeability such as cold injury, the here observed serum albumin extravasations are very modest and, moreover, reversible. Microwave exposure in the frequency and intensity range of mobile telephony is unlikely to produce pathologically significant changes of the blood-brain barrier permeability. Key words Microwave irradiation · Mobile telephony · Blood-brain barrier · Vasogenic edema · Rat
Introduction The increasing use of cellular telephones and the close proximity of the antenna of such devices to the head have raised concerns about the biological interactions between microwave irradiation and the central nervous system. It is well established that microwave exposure resulting in whole-body specific absorption rates (SAR) of 1–4 W/kg produces pathological reactions in mammals [39]. Such levels may also be reached in circumscribed areas of the brain using commercially available GSM (global system for mobile communication) telephone devices. However, whole-body irradiation is accompanied by an increase in body temperature which may be responsible for the observed pathological reactions. In fact, as long as the body temperature is kept constant, microwave irradiation at frequency bands up to the GHz range seems to produce little if any biologically adverse effects [22]. This has been confirmed by a recent experimental study from our laboratory in which 4-h microwave exposure of rat brain at 900 MHz with SAR of up to 7.5 W/kg resulted only in minor and pathologically irrelevant alterations of the expression patterns of various transcription factors or stress proteins [11]. On the other hand, several studies reported permeability changes of the blood-brain barrier (BBB) following microwave exposure [1, 2, 8, 10, 20, 26, 31, 32] but others studies produced clearly negative results [13, 23, 28, 29, 40–44]. A major methodological problem of microwave irradiation studies is the design of the exposure setup because
466
the magnitude of the SAR in the brain depends crucially on the geometric position of the antenna in relation to the head of the animals. In the present study we used a novel exposure system which simulates the absorption characteristics of cellular phones operating in the 900 MHz frequency band in the heads of rats maintained in a fixed position in relation to the antenna [5]. This system enables the induction of well-defined field strengths and limits the variations of the brain-averaged SAR due to movement and different animal size to less than 16% [5]. Using this system we analyzed the immediate and late effects of 4-h microwave exposure on the permeability of the BBB of rat.
Material and methods Microwave exposure The exposure setup has been described in detail previously [11]. In short, unanesthetized male Wistar rats (250–300 g) were restrained in plastic tubes mounted circularly in a Plexiglas carousel. The heads of the rats were positioned towards a central antenna and exposed to microwave fields in the GSM standard with a carrier frequency of 900 MHz, a repetition frequency of 217 Hz and average brain SAR of 0.3 and 1.5 W/kg. An average SAR of 7.5 W/kg was produced by continuous wave 900 MHz irradiation. Microwave exposure was carried out for 4 h in groups of 20 animals. Ten animals from each group were killed immediately after exposure, and the other 10 animals 7 days later. As negative controls two additional groups were used: a freely moving control group (n = 20) and a sham-exposed control group (n = 20) which was restrained for 4 h in the exposure system. These animals were also killed either immediately after sham exposure (n = 10) or 7 days later (n = 10). As positive controls, focal BBB disturbances were produced in two rats by cold injury. For this purpose a metal rod precooled with liquid nitrogen was applied to the parietal skull for 3 and 6 min, and brains were removed 3 and 6 h later, respectively. This type of injury causes a cortical lesion and a breakdown of the BBB with massive extravasation of serum albumin into the brain parenchyma [7, 17].
was developed with diaminobenzidine solution (0.5 mg/ml, 1% hydrogen peroxide) for 5 min at RT. The reaction was stopped by washing with PBS and the sections were dehydrated and embedded with Eukitt (Kindler, Freiburg, Germany). The specificity of the staining was confirmed by control experiments in which either the primary antibody was omitted or replaced by rabbit serum (1 : 200). In both cases immunoreactivity was completely abolished. Brains of perfusion-fixed animals were embedded in paraffin, and 1-µm paraffin sections were prepared at three coronal levels, i.e., 9 and 1.2 mm posterior as well as 3 mm anterior to bregma, according to the stereotactic atlas of the rat brain [27]. Immunohistochemical staining of extravasated serum albumin was carried out as described above except that primary rabbit anti-rat antibodies were diluted 1 : 4000 in PBS/5% BSA. Histological evaluation of neuronal damage was performed on adjacent paraffin sections stained with cresyl-violet and hematoxylin-eosin. Statistical evaluations Differences between groups were tested for statistical difference by one-way ANOVA followed by Fisher’s protected least-square difference test for multiple comparisons. The level of statistical significance was set at P < 0.05.
Results Cold injury The validity of the immunohistochemical technique for the detection of extravasated serum albumin was tested in rats subjected to cortical cold injury. Albumin extravasation could be clearly detected within and in the periphery of the lesion, as well as along the fiber tracts of the subjacent white matter (Fig. 1). In addition, some weak immunostaining was observed in the choroid plexus and in the wall of some vessels located in more distant regions of the same or the opposite hemisphere. This basal staining pattern was similar to that of non-irradiated or freely moving control rats.
Immunohistochemistry and histology Animals were anesthetized with 2.0% halothane and killed either by decapitation (immediate post-exposure groups) or by transcardiac perfusion with 4% (w/v) paraformaldehyde (7-day survival groups). Brains of decapitated animals were immediately removed and frozen in isopentane at –70° C. Ten-micrometer cryostat sections of brain were taken at five coronal levels, i.e., 9 mm, 4 mm and 1.2 mm posterior as well as 3 mm and 4.4 mm anterior to bregma, according to the stereotactic atlas of the rat brain [27]. Sections were mounted on poly-L-lysine-coated slides and stored at –70° C until further use. Permeability changes of BBB were traced by immunohistochemical staining of extravasated serum albumin using the peroxidase-antiperoxidase (PAP) method [35]. For this purpose the nonfixed cryostat sections were incubated in phosphate-buffered saline (PBS) for 10 min to remove traces of blood serum from the tissue surface. PBS incubation was followed by 30-min fixation in methanol. Endogenous peroxidases were blocked by 30-min incubation in methanol containing 0.1% hydrogen peroxide. Nonspecific binding of the secondary antibody was blocked with swine serum [1 : 100 in PBS/5% bovine serum albumin (BSA); Dako, Hamburg, Germany) for 1 h at room temperature (RT), followed by incubation with rabbit anti-rat albumin antibodies (1 : 1000 in PBS/5% BSA, Bioscience, Oberschleißheim, Germany) for 16 h at RT. After incubation with swine anti-rabbit immunglobulins (1:50 in PBS/5% BSA, Dako) for 30 min at RT and rabbit PAP (1:200 in PBS/5% BSA, Dako) for 30 min at RT, the peroxidase reaction
Non-irradiated animals The number of areas showing serum albumin extravasation was counted by microscopical scanning of all cryostat or paraffin sections taken from the different coronal levels of each brain. In the freely moving control animals (n = 20) only one single spot with albumin extravasation was found (Table 1). The diameter of this leakage area was approximately 50 µm (Fig. 1 e). In the sham-exposed control group (n = 20) three circumscribed areas of albumin extravasation were found in 2 of the 10 animals decapitated immediately after the 4-h exposure, and another one in 1 of the 10 animals which survived 7 days (Table 1). Under control conditions the mean incidence of immunohistochemically detectable albumin extravasations was, therefore, only 3 spots in 20 animals. Microwave-irradiated animals The results are summarized in Table 1. Small areas with albumin extravasations were found in between 5 and 7 of
467
Fig. 1 a–g Immunohistochemical staining of serum albumin in cryostat sections of rat brain after 4-h microwave irradiation, as compared to cold injury (positive control). a Massive extravasation of serum albumin at 3 h after cold injury; b no macroscopical extravasations in freely moving rat (negative control); c rat exposed for 4 h to 900 MHz microwave irradiation at a specific absorption rate of 7.5 W/kg irradiation. d–g Higher magnifications show circumscribed albumin extravasations in a freely moving control rat (e), and at the end of 4-h exposure to 1.5 W/kg (f) and 7.5 W/kg (d, g), respectively. Micrograph shown in d is a magnification of the two immunopositive spots visible in c (arrow). Note paucity of microwave-induced alterations and the similarity of extravasates in freely moving control and irradiated animals. The total number of immunopositive spots is given in Table 1. Bars a–c = 0.5 cm; d = 50 µm; e–g = 100 µm
the 20 animals of each of the three groups irradiated with SAR of 0.3, 1.5 and 7.5 W/kg, respectively. The number of extravasations was highest immediately after microwave exposure, when 5 out of 10 animals of each group exhibited one or more spots of albumin immunoreactivity. However, compared to the non-irradiated groups, the number of extravasations only reached statistical significance in the group with the highest SAR, 7.5 W/kg, where a total of 14 spots was observed (Table 1). At 7 days after exposure the number of extravasations was much smaller and even in the group with the highest SAR amounted to only 1 spot in 10 animals. The size and appearance of albumin extravasations was very similar irrespective of the intensity of microwave exposure or the time of investigation. The diameter varied between 50 and 100 µm (Fig. 1 d–g), and there was no
468
Fig. 2 a–d Immunohistochemical staining of serum albumin in paraffin sections of rat exposed 7 days earlier to microwave irradiation. a Free moving rat (negative control); b 7 days after 4-h sham exposure; c 7 days after exposure to 0.3 W/kg; d 7 days after exposure to 7.5 W/kg microwave irradiation. Note similarity of perivascular extravasates of serum albumin in sham-exposed and irradiated animals. The total number of immunopositive spots is given in Table 1. Bars a = 1.0 mm; b–d = 0.1 mm
clear topological preference for any of the investigated coronal levels (Table 1). It should be noted, however, that the few extravasations observed at 7 days after sham or microwave exposure were all clearly associated with larger vessels (Fig. 2 b–d). Histological analysis of cresyl-violet or hematoxylineosin stained sections did not reveal any evidence for neu-
Table 1 Number and distribution of serum albumin extravasation after 900 MHz microwave irradiation of the brain. Sham or microwave exposure in restrained rats for 4 h. Immunohistochemical staining of albumin extravasation was carried out in cryostat sections of animals killed immediately after exposure and in paraffin
sections of animals fixed by perfusion at 7 days after exposure. Levels 1–3 correspond to coronal plains 4.4, 3.0 mm anterior, and 1.2, 4.0 and 9.0 mm posterior to bregma, respectively (SAR specific absorption rates, n.d. not determined)
Animals with extravasations
Number of extravasations Total
Level 1
Level 2
Level 3
Level 4
Level 5
Freely moving controls cryostat sections (n = 10) paraffin sections (n = 10)
1 0
1 0
1 n.d.
0 0
0 0
0 n.d.
0 0
Sham-exposed controls immediate (n = 10) 7 day (n = 10)
2 1
3 1
1 n.d.
0 1
0 0
0 n.d
2 0
SAR 0.3 W/kg immediate (n = 10) 7 day (n = 10)
5 2
7 2
2 n.d.
0 2
1 0
2 n.d.
2 0
SAR 1.5 W/kg immediate (n = 10) 7 day (n = 10)
5 0
6 0
1 n.d.
1 0
0 0
2 n.d.
2 0
SAR 7.5 W/kg immediate (n = 10) 7 day (n = 10)
5 1
14a 1
5 n.d.
3 1
2 0
2 n.d.
2 0
a Indicates
significant increase compared to sham-exposed controls (P < 0.05)
469
ronal damage, even not in the areas of serum albumin extravasation. This is clearly different from cold injury where substantial neuronal damage was present not only within the cortical lesion but also in the surrounding edematous tissue.
Discussion In the intact brain the BBB is impermeable for circulating serum proteins [4]. This is in line with our findings in freely moving control rats which exhibited almost no immunohistochemically detectable extravasations of serum albumin in brain parenchyma. Conversely, the induction of an established vasogenic type of brain edema by cold injury [18] resulted in intense albumin staining which confirms that the present immunohistochemical method is sensitive enough to detect pathologically relevant permeability changes of the BBB. Our data demonstrate that the immobilization of rats during 4-h sham exposure induces minor but detectable increases in the number of serum albumin extravasations, and that this number is slightly increased when immobilized rats are exposed to microwaves. At the highest SAR of 7.5 W/kg this increase becomes significant in the immediate post-exposure group when compared with either of the two control groups but even at this high intensity, which is far above the safety limits of 2 W/kg [3], only 5 of 10 animals exhibited such disturbances. Moreover, the barrier lesions could only be detected at the end of the microwave exposure and were not longer visible when animals were investigated 7 days later. Compared to the changes induced by cold injury or other pathologically relevant disturbances of barrier permeability, these alterations are negligible and in view of their reversibility probably without any pathological significance. This conclusion is confirmed by the histological examination which did not reveal any cellular injury. In a parallel investigation using the same experimental material, the induction of stress proteins and the expression of marker molecules for cellular activation of neurons, astrocytes and microglial cells was studied by in situ hybridization and immunohistochemistry [11]. In this study some minor changes in the expression patterns of immediate-early genes (c-fos and c-jun) and the heat shock gene hsp70 were observed at the end of the exposure but not 1 or 7 days later. We, therefore, concluded that 4-h microwave exposure in the 900 MHz band with SAR of up to 7.5 W/kg does not cause persisting injury of the central nervous system. This conclusion is at variance with the results of several previous studies which demonstrated alterations of BBB permeability following microwave irradiation alone or in combination with static magnetic fields, as used for magnetic resonance imaging (see above). Some authors suggested that barrier breakdown is mainly due to subtle thermal effects [12, 21, 24, 37, 38] but others have stressed that the disturbance may occur under athermal conditions [23, 41–44]. Salford et al. [32, 33] used either continuous or pulse-modulated 915 MHz electromagnetic
radiation at SAR of up to 5 W/kg and observed multifocal albumin, but not fibrinogen extravasations. The incidence of albumin extravasations rose significantly only at SAR of greater than 2.5 W/kg which is in line with our observation of significant changes of SAR at 7.5 but not 1.5 W/kg. Since these high intensities are close to the threshold of thermal induction, the observed barrier alterations may, therefore, be caused by spurious temperature effects. As far as the mechanisms of barrier disturbances following microwave exposure are concerned, most authors suggest pinocytotic transport across the vascular endothelial cells [1, 2, 25, 34]. However, the higher permeability change for the low molecular weight albumin as compared to the larger fibrinogen molecule, as described by Salford et al. [32, 33], argues against such a mechanism. An alternative explanation could be an opening of tight junctions or an increase of ornithine decarboxylase activity, which has been shown to respond to low frequency (ELF) magnetic field exposure [6] and which correlates with BBB disturbances [19]. However, our data are also at variance with this interpretation because the most pronounced increase in extravasation was observed at SAR 7.5 W/kg which – in contrast to the lower SAR – was not modulated by low frequency pulses. We, therefore, suggest that the observed barrier disturbances are due to unspecific side effects, such as stress-induced hypertensive episodes brought about by the immobilization of the exposed animals. In fact, the non-irradiated restrained control group also exhibited a few extravasations, and it is conceivable that the additional perception of temperature changes during irradiation enhances this effect. This would be in line with previous reports on reversible openings of the BBB following hypertensive episodes brought about by such different interventions as systemic application of vasoconstrictor agents, compression of the thoracic aorta or rapid intracarotid infusion of saline or blood [9, 14–16, 30, 36]. Obviously, the here observed changes were much less pronounced than described in these studies, which explains the absence of behavioral abnormalities (this investigation) or the paucity of molecular [11] indicators of stress. In conclusion, the present study demonstrates that microwave irradiation at a frequency and intensity corresponding to mobile telephony produces no or only negligible permeability changes which are not associated with lasting morphological alterations. Our data do not support the notion that mobile telephony is a health risk to the brain. Acknowledgements This investigation was supported by Motorola, Inc. The technical assistance of Mrs. S. Krause and Mr. St. Hennes, and the secretarial help of Mrs. M. Hahmann and Mrs. D. Schewetzky are gratefully acknowleged.
References 1. Albert EN (1977) Light and electron microscopic observations on the blood-brain barrier after microwave irradiation. In: Hazzard DG (ed) Symposium on biological effects and measurements of radiofrequency microwaves. HEW Publication, Rockville, pp 294-304
470 2. Albert EN, Kerns JM (1981) Reversible microwave effects on the blood-brain barrier. Brain Res 230 : 153–164 3. ANSI/IEEE (1992) IEEE Standard for the safety levels with respect to human exposure to radio frequency fields, 3 kHz to 300 GHz. The Institute of Electrical and Electronics Engineers, New York, p 10017 4. Brightman MW, Klatzo I, Olsson Y, Reese TS (1970) The blood-brain barrier to proteins under normal and pathological conditions. J Neurol Sci 10 : 215–239 5. Burkhardt M, Spinelli Y, Kuster N (1997) Exposure setup to test CNS effects of wireless communications systems. Health Phys (in press) 6. Byns CV, Kartun K, Pieper S, Adeg WR (1988) Increased ornithin decarboxylase activity in cultured cells exposed to low energy, modulated microwave fields and phorbol ester tumour promotous. Cancer Res 48 : 4222–4226 7. Chan PH, Longar S, Fishman RA (1983) Phospholipid degradation and edema development in cold-injured rat brain. Brain Res 277:329–337 8. Chang BK, Huang AT, Joines WT, Kramer RS (1982) The effect of microwave radiation (1.0 GHz) on the blood-brain barrier in dogs. Radio Sci 17 : 165–168 9. Dinsdale HB, Robertson DM, Haas DA, Davis PE (1976) Acute hypertension, blood-brain barrier damage, and the role of adrenal steroids. In: Cervôs-Navarro J, Betz E, Matakas F, Wüllenweber R (eds) The cerebral vessel wall. Raven Press, New York, pp 253-260 10. Frey AH, Feld SR, Frey B (1975) Neural function and behavior: defining the relationship. Ann NY Acad Sci 247 : 433–439 11. Fritze K, Wiessner C, Kuster N, Sommer C, Gass P, Hermann D, Kiessling M, Hossmann K-A (1997) Effect of GSM microwave exposure on the genomic response of the rat brain. Neuroscience (in press) 12. Goldman H, Lin JC, Murphy S, Lin MF (1984) Cerebrovascular permeability to 86Rb in the rat after exposure to pulsed microwaves. Bioelectromagnetics 5 : 323–330 13. Gruenau SP, Oscar KJ, Folker MT, Rapoport SI (1982) Absence of microwave effect on blood-brain barrier permeability to [14C]sucrose in the conscious rat. Exp Neurol 75 : 299–307 14. Hardebo JE, Nilsson B (1981) Opening the blood-brain barrier by acute elevation of intracarotid pressure. Acta Physiol Scand 111 : 43–49 15. Johansson B (1976) Brain barrier pathology in acute arterial hypertension. In: Levi G, Battistin L, Lajtha A (eds) Transport phenomena in the nervous system. Plenum Press, New York, pp 517-525 16. Johansson B (1974) Blood-brain barrier dysfunction in acute arterial hypertension after papaverine-induced vasodilatation. Acta Neurol Scand 50 : 572–580 17. Klatzo I (1987) Pathophysiological aspects of brain edema. Acta Neuropathol (Berl) 72 : 236–239 18. Klatzo I (1967) Neuropathological aspects of brain edema. J Neuropathol Exp Neurol 26 : 1–14 19. Koenig H, Goldstone AD, Lu CY, Trout JJ (1989) Polyamines and Ca2+ mediate hyperosmolal opening of the blood-brain barrier: in vitro studies in isolated rat cerebral capillaries. J Neurochem 52 : 1135–1142 20. Lange DG, Sedmak J (1991) Japanese encephalitis virus (JEV): potentiation of lethality in mice by microwave radiation. Bioelectromagnetics 12 : 335–348 21. Lin JC, Lin MF (1982) Microwave hyperthermia-induced blood-brain barrier alterations. Radiat Res 89 : 77–87 22. McKinlay AFE (1997) Possible health effects related to the use of radiotelephones. Recommendations of a European Commission Expert Group. Radiol Protect Bull 187 : 9–16 23. Merritt JH, Chamnes AF, Allen SJ (1978) Studies on bloodbrain barrier permeability after microwave radiation. Radiat Environ Biophys 15 : 367–377 24. Moriyama E, Salcman M, Broadwell RD (1991) Blood-brain barrier alteration after microwave-induced hyperthermia is purely a thermal effect. I. Temperature and power measurements. Surg Neurol 35 : 177–182
25. Neubauer C, Phelan AM, Kues H, Lange DG (1990) Microwave irradiation of rats at 2.45 GHz activates pinocytic-like uptake of tracer by capillary endothelial cells of cerebral cortex. Bioelectromagnetics 11 : 261–268 26. Oscar KJ, Hawkins TD (1977) Microwave alteration of the blood-brain barrier system of rats. Brain Res 126 : 281–293 27. Pellegrino LJ, Pellegrino AS, Cushman AJ (1979) Stereotactic atlas of the rat brain. Plenum Press, New York 28. Preston E, Préfontaine G (1980) Cerebrovascular permeability to sucrose in the rat exposed to 2450-MHz microwaves. J Appl Physiol 49 : 218–223 29. Preston E, Vavasour EJ, Assenheim HM (1979) Permeability of the blood-brain barrier to mannitol in the rat following 2450 MHz microwave irradiation. Brain Res 174 : 109–117 30. Rapoport SI (1976) Blood-brain barrier in physiology and medicine. Raven Press, New York 31. Salford L, Brun A, Eberhardt J, Malmgren L, Persson B (1992) Electromagnetic field-induced permeability of the blood-brain barrier shown by immunohistochemical methods. In: Norden B, Ramel C (eds) Resonance phenomena in biology. Oxford University Press, Oxford, pp 87-91 32. Salford LG, Brun A, Eberhardt JL, Persson BRR (1993) Permeability of the blood-brain barrier induced by 915 MHz electromagnetic radiation, continuous wave and modulated at 8, 16, 50, and 1200 Hz. Bioelectrochem Bioenerg 30 : 293–301 33. Salford LG, Brun A, Sturesson K, Eberhardt JL, Persson BRR (1994) Permeability of the blood-brain barrier induced by 915 MHz electromagnetic radiation, continuous wave and modulated at 8, 16, 50, and 200 Hz. Microsc Res Tech 27 : 535–542 34. Shivers RR, Kavaliers M, Teskey GC, Prato FS, Pelletier R-M (1987) Magnetic resonance imaging temporarily alters bloodbrain barrier permeability in the rat. Neurosci Lett 76 : 25–31 35. Sternberger LA, Hardy PHJ, Cuculis JJ, Meyer HG (1970) The unlabeled antibody enzyme method of immunohistochemistry. Preparation and properties of soluble antigen-antibody complex (horseradish peroxidase-antihorseradish peroxidase) and its use in identification of spirochetes. J Histochem Cytochem 18 : 315–333 36. Strandgaard S (1978) Autoregulation of cerebral circulation in hypertension. Acta Neurol Scand 66 : 1–82 37. Sutton CH, Carrol FB (1979) Effects of microwave-induced hyperthermia on the blood-brain barrier of the rat. Radio Sci 14 : 329–334 38. Sutton CH, Nunnally RL, Carroll FB (1973) Protection of the microwave-irradiated brain with body-core hypothermia. Cryobiology 10 : 513–514 39. UNEP/WHO/IRPA (1993) Electromagnetic Fields (300 Hz – 300 GHz), Environmental Health Criteria 137. United Nations Environment Programme/World Health Organization/International Radiation Protection Association, Geneva 40. Ward TR, Ali JS (1985) Blood-brain barrier permeation in the rat during exposure to lower-power 1.7 GHz microwave radiation. Bioelectromagnetics 2 : 131–143 41. Ward TR, Elder JA, Long MD, Svendsgaard D (1982) Measurement of blood-brain barrier permeation in rats during exposure to 2450 MHz microwaves. Bioelectromagnetics 3 : 371– 383 42. Williams WM, Hoss W, Formaniak M, Michaelson SM (1984) Effect of 2450 MHz microwave energy on the blood-brain barrier to hydrophilic molecules. A. Effect on permeability to sodium fluorescein. Brain Res Rev 7 : 165–170 43. Williams WM, Cerro M del, Michaelson SM (1984) Effect of 2450 MHz microwave energy on the blood-brain barrier to hydrophilic molecules. B. Effect on permeability to HRP. Brain Res Rev 7 : 171–181 44. Williams WM, Platner J, Michaelson SM (1984) Effect of 2450 MHz microwave energy on the blood-brain to hydrophilic molecules. C. Effect on permeability to [14C]sucrose. Brain Res Rev 7 : 183–190
