ID Number		1723
Study Type		In Vitro
Model		900 MHz (GSM) exposure to cells and analysis of cell cycle genes (ERK, MAPK)
Details		Rat1 (fibroblasts) and HeLa (human cervical cancer) cells were exposed to 900 MHz (GSM) for 5 to 30 minutes and assayed for expression of several MAPK genes including Erk-1, Erk-2, and p38 and JNK. The exposure system consisted of a panel antenna placed within the incubator and power density (no SAR) mapped throughout the incubator in terms of mW/cm2. Results are reported in terms of ~0.1, ~0.2, and ~0.3 mW/cm2. The authors report short term exposure (10-20 minutes) resulted in a transient increased phosphorylation (activation) of the MAPK elements Erk-1 and Erk-2 that returned to baseline after 30 minutes (literature reports such transient increases can be caused by a wide range of things and is not necessarily sufficient to promote cell division). The authors also report a slight decrease in phosphorylation (activation) of MAPK elements (p38 and JNK) which each represent different MAPK paths responsive to stress, inflammatory cytokines, and certain growth factors. The authors finally report evidence to suggest the first step in Erk-1 and Erk-2 phosphorylation (activation) is mediated by NADH oxidase, producing ROS that in turn stimulate metalloproteinases that activate the EGF receptor, and further activate the MAPKKK, MAPKK, MAPK (Erk) cascade.
Findings		Effects
Status		Completed With Publication
Principal Investigator		Weizmann Institute of Science, Israel
Funding Agency		Private/Instit.
Country		ISRAEL
References		Friedman, J et al. Biochem J, (2007) 405:559-568
Comments		The Erk (MAPK) pathway is often driven by growth factors, mitogens, and other positive stimuli, and downstream activates MAPKKK elements (Raf, Mos TPL2), which in turn downstream activates MAPKK elements (Mek-1, Mek-2), which in turn downstream activates MAPK elements (Erk-1 and Erk-2) and generally results in growth, differentiation, metabolism, and development. This pathway may also be instrumental in the biochemical pathways linked to learning and development in the brain. In contrast, other MAPK pathways (p38 and JNK, respectively) can be activated by various types of stress (heat, cadmium), inflammatory cytokines, and certain growth factors. These stress response pathways (through p38 and JNK, respectively) often end in inflammation and apoptosis. In terms of the stress response, p38 and JNK act to hyper-phosphorylate HSF-1 causing it to trimerize and thus allowing it to bind to stress inducible promoters that regulate the expression of heat shock genes (e.g., hsp70). In contrast, activated P-Erk-1 and P-Erk-2, as well as other elements such as GSK3beta, are negative regulators of HSF-1 (Kim et al J Cell Biochem (1997) 67:43-54), phosphorylating HSF-1 on different sites (ser 307, allowing secondary phosphorylation on ser 303 by GSK3beta) and preventing trimerization (and thus activation), and would oppose hsp-70 induction. Friedmann et al report short-term (10-20 minute) RF exposure increases P-Erk-1 and P-Erk-2 (i.e., activation) causes increased Erk-1 and Erk-2 phosphorylation (activation) as well as a slight decrease in p38 and JNK phosphorylation (deactivation). The Friedmann et al paper suggests that the early (10-20 minute) response to RF is an activation of the Erk MAPK pathway, with deactivation after ~30 minutes and longer. Exposures such as those used in Leszczynski 2002 and Caraglia 2005 using higher SARs [3.6 and 2.5 W/kg, respectively] and longer exposure times [1 hour or more] resulted in activation (phosphorylation) of p38 and JNK stress response pathways (including the heat shock response and apoptosis). Several studies in the literature show that many different factors (Yoon, S. and Seger, R. 2006, Growth Factors 24, 2144) can rapidly induce Erk-1 and Erk-2 activation (phosphorylation), and that several additional kinases (e.g., PI-3-kinase, Akt, Elk-1) are involved in the upstream and downstream processes. The complex response to different levels of heat is not well characterized, especially at the lower levels (38-40 degrees). What is known is that traditional heat shock (42 degrees) can result in short term activation (phosphorylation) of Erk-1 and Erk-2 in neuroblastoma cells (Bijur et al, J Neurochem (2000) 75:2401-2408) and in the brain of animals (Maroni et al, Brain Res Mol Brain Res (2003) 119:90-99), similar to the results reported by Friedmann et al. In addition, when the stressor cadmium is used in high concentration, p38 and hsp27 are activated (phosphorylated) driving the activation of HSF-1 and downstream induction of heat shock proteins, similar to the results of Caraglia (2005) and Leszczynski (2002). At lower concentrations, cadmium causes short-term activation (phosphorylation) of Erk-1 and Erk-2, which is eventually followed by a heat shock response (Hung et al 1998, J Biol Chem 273:31924-31931), similar to the findings of Friedmann et al (2007). Thus, the findings could well be the result if different levels and time courses of heating. From what I can see in the study by Butiglione (2007)  we do not yet have the paper  they exposed neuroblastoma cells to 900 MHz at 1 W/kg for both short term (5, 15, 30 minutes) and long term (6, 24 hours) and reported a max increase in Egr-1 at 30 minutes and a decrease to baseline at 24 hours. Since Erk kinases directly regulate Egr-1, this would also support the time course observed by Friedmann et al for Erk kinase activation. Many things stimulate NADH oxidase to generate ROS and downstream activate the EGF receptor (as suggested by Friedman et al to be the mechanism of Erk-1 and Erk-2 activation). Vascular cell adhesion molecule VCAM-1 (Deem et al 2007, J Immunol. 2007 Mar 15;178(6):3865-73), mitogens and growth factors (Thannickal et al FASEB J 2000, 14:1741-1748), ubiquinone (Vaillant et al 1996, J Bioenrg Bioemebr 28:531-540), GTP & GDP nucleotides / energy source (Morre et al 1993, 292:647-653), and increasing cluture temperature (De Figueroa et al 2001, Microbiol Res 155 :257-262 ; Nagata et al Jpn J Genet 1991, 66 :255-261). One study Mays initiated (at my suggestion, as I remember) in the early Motorola program examined the effects on incremental temperature to cells in culture (VanderWaal, Higashikubo, Roti Roti et al 2001). The findings were that slight temperature increases resuted in cell line-specific changes in proliferation. C3H 10T1/2 and LN71 (human glioma) cells cultured at 38 degrees resulted in accelerated growth (proliferation), while 39 degrees slowed proliferation (as compared to 37 degrees). At 40 degrees, the growth cycle shut down and cells stopped and accumulated at the G1/S boundary (i.e., proliferation stopped). C-fos and c-myc mRNA levels were increased at all temperatures, but MAPK genes were never examined. Interestingly, U87MG (human glioma) cells did not respond to 38 degrees with increased proliferation  they remained flat until ~40 degrees or so and also shut down at the G1/S boundary. Another strong indicator that the results are thermal in nature is that cells and animals (presumably) have developed very complex and efficient regulatory mechanisms to control and deal with changes in temperature  if not we would not last long. To this end, if the effects observed by Friedmann were novel RF energy interactions (other than heat), than similar exposures would be expected to reveal themselves in vitro as changes in proliferation, apoptosis, or response to growth factors. Such effects would further be expected to reveal themselves in animal studies as cancers and developmental effects. The overall weight of evidence does not support such effects. This all suggests to me that the studies by Friedmann (2007), Buttiglione (2007), Caraglia (2005) and Leszczynski (2002) are looking at the effects of very small temperature increases that cells can initially respond to as a positive growth stimulus, but too much or prolonged heat triggering a stress response. With regard to easy criticisms of the Friedmann et al paper, it does have several inconsistencies. It references Stagg et al (2001, 5 W/kg Iridium exposure in rats) and Chauhan, McNamee et al (2006) as reporting increases in c-fos and c-jun with RF exposure, and they did not - just increases following restraint stress in animals. Friedmann et al also incorrectly referenced Hook et al (2004) as indicating changes in apoptosis from RF, which the authors did not report. Friedmann et al also incorrectly referenced Capri et al (2004) as indicating RF exposure to primary human monocytes resulted in changes in cell cycle. The Capri et al paper reported a slight decrease in proliferation and no change in apoptosis in cells exposed to RF after several days. Regarding the exposure system, it is not well defined, and consisted of an antenna placed on the central shelf of an incubator. Exposure was reported in terms of power density mapped in mW/cm2 within the incubator - no SAR. They assumed no reflections because they put absorber on the incubator walls, but I cannot see where they say they also lined the shelves (might be glass, but often metal). They also did not detail where the culture dishes were placed relative to the antenna, just that the temperature changes in the culture media were less than 0.05 degrees. Such small temperature changes are surprising, because there is usually not a lot of culture media in a dish (to keep a high surface-to-volume ratio for gas exchange) and normal temperature fluctuation in incubators is often on the order a degree or more.