Regarding Article: Arendash GW et al. (2010) J Alzheimers Dis 19, 191-210

1 May 2010

We thank Dr. Kumlin and colleagues for their insightful comments regarding our paper [1]. Inasmuch as we were unaware of their earlier study involving EMF exposure to normal rats [2], we did not include it among the references in our paper and apologize for this oversight. Kumlin et al. [2] did indeed provide initial evidence that long-term EMF exposure (2 hrs/day, 5 days/week, for 5 weeks) can improve cognitive performance in rodents. It is important to note, however, that they provided EMF exposure to very young rats from 3-8 weeks of age. Thus, the authors were actually investigating effects of EMF exposure on immature rats whose brains were still developing—not adult animals. In utilizing the Morris water maze at a single test point, Kumlin et al. found a positive EMF effect only on the first two days of acquisition testing (not on the final two days of testing or overall), so their acquisitional effect was limited to the “rate” of learning. In the retention phase of testing, they found that rats given the higher of two EMF levels utilized (3.0 W/kg SAR) spent more time in the former platform area compared to non-exposed rats (15 sec vs. 10 sec average).

Kumlin and colleagues are incorrect in their assumption that our study utilized a continuous signal without modulation. In fact, our EMF exposure of 918 MHz frequency involved modulation with GMSK signal and was non-continuous with carrier bursts repeated every 4.6 ms, giving a pulse repetition rate of 217 Hz. The electrical field strength varied between 17 and 35 V/m. This resulted in calculated specific absorption rate (SAR) levels that varied between 0.25 W/kg and 1.05 W/kg. SAR was calculated from the below equation, with σ (0.88 s/m) and ρ (1030 kg/m-3) values attained from Nightingale et al. [3]:
SAR = σE2 σ = mean electrical conductivity of mouse brain tissue ρ ρ = mass density of mouse brain E = electrical field strength 

In addition, our animals received “near-field” EMF exposure, given that the antenna was one wavelength long and that far-field exposure begins at 2 antenna lengths away from the antenna. Since our antenna length was 12 inches and the distance to mouse cages was 10 inches, mice in our studies received near field EMF exposure, with far-field exposure beginning at 24 inches from the antenna.

We regret that a number of the aforementioned EMF parameters were not mentioned within the methodology of our paper. However, it should now be clear that the EMF parameters used in our mouse whole-body exposure studies closely mimic the EMF exposure provided to the human brain by a typical cell phone, which is why our study is directly relevant to human cell phone use.

In reference to Kumlin et al. observing cognitive benefits after 5 weeks of EMF exposure, while we reported cognitive benefits in our study beginning at 5 months into EMF exposure, this difference in onset of cognitive benefit is most likely due to: 1) their use of immature, adolescent rats wherein EMF effects could be more easily attained in comparison to our use of adult mice, and/or 2) the much higher (above cell phone level) SAR level employed in their study. Regarding the later, there really was no effect of their EMF exposure on Morris maze “overall” or “final” acquisition, with only a modest enhancement of retention in the probe trial observed at a very high SAR level. By contrast, our paper in “adult” mice involved both normal and Alzheimer’s mice (approximately 100 mice total) given cell phone-level SAR exposure in both protection- and treatment-based studies, tested at multiple time points, and in multiple cognitive tasks. Moreover, several of our cognitive tasks (i.e., the radial arm water maze and cognitive interference task) are far more challenging than the Morris water maze used by Kumlin et al. and involve working/short-term memory rather than the spatial long-term learning/memory evaluated in the Morris maze.

Regarding the comment that there was a surprisingly large increase in body temperature with long-term EMF exposure in mice of our study, we wish to make three points. First, the increase in body/brain temperature (of approximately 1ºC) was consistently observed only in the Alzheimer’s mice and only with long-term (not acute) EMF exposure—normal mice did not consistently show an increase in temperature during EMF exposure. Second, our mice were given longer and more consistent EMF exposure (daily for 8-9 months) than any prior study involving cell phone parameters. Third, Alzheimer’s transgenic mice and their brain Aβ burdens had never before been a subject of EMF studies. Thus, Kumlin et al. are questioning a result that had no prior precedent of what to expect from the literature. What is clear is that the 1ºC increase in body/brain temperature during our EMF exposure to AD mice is a minimal increase that is well below the 41ºC level that begins to result in mammalian brain damage. Moreover, fluctuations of 2ºC or higher in mammalian brains occur regularly, depending on behavioral and metabolic state. Thus, our observed 1ºC elevation in temperature during EMF exposure to AD mice would appear to be safe and cognitively beneficial. It is noteworthy that this increase in body/brain temperature of AD mice during EMF exposure periods occurred at SAR levels (0.25 – 1.05 W/kg) that are well within, and indeed typical of, SAR levels during cell phone use.

Kumlin et al. are mistaken in suggesting that the temperature data from our long-term and acute studies are not directly comparable since only the acute studies involved brain temperature measurement. In fact, the group comparisons for both types of studies are comparable in view of the high correlation (r=0.98) between body temperature and brain temperature that we also presented. Thus, an increase in body temperature would certainly mean an identical increase in brain temperature for the long-term studies, despite our not having measured brain temperature in the long-term studies. Kumlin et al. also believe that, because our temperature recording occurred during EMF exposure, that the temperature probes may have been subjected to interference from the EMF signal. If this was an issue, however, all measurements would be expected to have the same interference, not just those from AD mice and not just those from long-term studies.

Kumlin et al. further suggest that the EMF dose used in our studies is not known. However, as we have now provided fuller details of our EMF exposure methodology, the EMF dose utilized in our studies is now abundantly clear. It should also be evident that the EMF parameters utilized were indeed very similar to those emitted by typical cell phones during their use. We agree with Kumlin et al. that it is important to investigate the dose-response relation in future EMF studies. In that context, our ongoing studies are investigating the best set of EMF parameters for achieving cognitive benefits in the shortest amount of time (if previous cognitive impairment is present), and without incurring any undesirable side-effects.

Finally, we would like to indicate our excitement over the inception of a new field of cognitive neuroscience by recent studies such as ours and Kumlin et al.—namely, Cognitive Benefits of EMF Exposure. Explored in a conscientious manner, with well-designed studies and appropriate behavioral endpoints, this new emerging field could provide a surprising array of benefits against diseases of brain aging.

Gary W. Arendash
Dept. of Cell Biology, Microbiology, and Molecular Biology
University of South Florida
Tampa, FL

Chuanhai Cao
USF/Byrd Alzheimer’s Disease Research Institute
University of South Florida
Tampa, FL

References
[1] Arendash GW, Sanchez-Ramos J, Mori T, Mamcarz M, Lin X, Runfeldt M, Wang L, Zhang G, Sava V, Tan J, Cao C. Electromagnetic field treatment protects against and reverses cognitive impairment in Alzheimer's disease mice. J Alzheimers Dis. 2010 Jan;19(1):191-210.
[2] Kumlin T, Iivonen H, Miettinen P, Juvonen A, van Groen T, Puranen L, Pitkäaho R, Juutilainen J, Tanila H. Mobile phone radiation and the developing brain: behavioral and morphological effects in juvenile rats. Radiat Res. 2007 Oct;168(4):471-9.
[3] Nightingale NR, Goodridge VD, Sheppard RJ, Christie JL. The dielectric properties of the cerebellum, cerebrum and brain stem of mouse brain at radiowave and microwave frequencies. Phys Med Biol. 1983 Aug;28(8):897-903.