【作者】 曾群力；

【导师】 姜槐； 许正平；

【作者基本信息】 浙江大学 ， 劳动卫生与环境卫生学， 2007， 博士

【摘要】 随着无线通讯技术和电力事业的飞速发展，电磁辐射已成为环境中增长最快、影响最为普遍的因素之一，对其健康危害的认识和预防事关我国科技、经济和社会的可持续发展。有流行病学调查显示极低频电磁场(ELF EMF)暴露可引起白血病和乳腺癌等发病率增高；移动电话的射频电磁场(RF EMF)暴露可影响中枢神经系统功能，导致脑瘤等恶性病变。这些流行病学调查结果推动了电磁场对生物体的生物学效应及其机理的研究。体内、体外的实验研究提示低强度电磁场对神经系统、生殖系统和免疫系统等可产生一定影响，但同时也有大量的阴性报道存在，导致无法对电磁场的健康危险度进行正确评估。造成这种现象的原因在于电磁场与生物体作用的原初物理过程、引发的生物学反应以及产生生物学效应的机制不清，电磁场生物学效应的研究存在一定的盲目性。因此，揭示低强度电磁场生物学效应及作用机制成为目前迫切需要解决的问题。生物系统受电磁场辐照所产生的各种生理生化改变可能涉及到基因的表达调控。有研究发现，电磁场可改变原癌基因、凋亡相关基因、周期调控基因等的mRNA水平，如极低频电磁场可诱导原癌基因c-myc、c-jun和c-fos的转录，改变鼠胚胎干细胞凋亡相关基因bcl-2和bax、细胞周期调控相关基因GADD45的转录：一定强度射频电磁场可下调神经元特异性Nurrl基因的表达，上调bax、GADD45 mRNA的水平；低频电磁场间断辐照可上调p53缺陷型细胞中c-jun、p21和egr-1 mRNA的水平，但野生型细胞不受影响：1710 MHz射频电磁场可显著上调p53缺陷型细胞中Hsp70 mRNA的转录，同时使c-jun、c-myc和p21 mRNA瞬时低幅度增加。基因在生物体的功能最终由其编码的蛋白质在细胞水平上体现，因此从蛋白质的角度入手才能真正揭示生命活动的规律。电磁场对细胞蛋白质表达的作用研究不多，主要集中在对鸟苷酸脱羧酶ODC、热休克蛋白HSP27／70以及一些信号转导途径中信号分子PKA、PKC、TPK、MAPK等表达水平或磷酸化等翻译后修饰的影响上。然而在这些研究中，实验者通常是根据推测的电磁场作用的可能效应、作用靶点和机制，选择相关的单个或几个蛋白质进行检测。这种研究思路是以假说为前提的，可能产生主观偏差；同时，由于所选择的指标是零散的，无法得到系统性、整体性的结果，不能全面揭示电磁场的生物学效应，勾画出其反应通路。一般认为，电磁场作为一种低能量的环境因素，可能通过复杂的信号传递过程作用于细胞，改变多个蛋白质的表达水平和／或翻译后修饰，进而产生一系列后续效应。因此，从蛋白质组的角度研究电磁场的生物学效应是必要的。以双向电泳作为分离技术和质谱作为鉴定技术的蛋白质组学方法能同时分离细胞内成百上千种蛋白质，并比较不同生理或病理状态下蛋白质表达的变化，为揭示外界因素对生物体的影响和疾病发生机制等提供了一种全新的研究方法。2001年，本实验室和芬兰Leszczynski研究小组率先将该技术引入到电磁场生物学效应及机制的研究中。为探讨电磁场对肿瘤发生的可能促进效应和比较环境中最常见的两类电磁场作用的异同，本博士论文第一部分选择人乳腺癌细胞株MCF-7，采用双向电泳技术(2-DE)研究了50 Hz极低频磁场和1800 MHz射频电磁场对细胞蛋白质表达的影响，建立了该细胞的电磁场蛋白质差异表达图谱，进而利用质谱技术(MS)鉴定了部分电磁场反应蛋白质。根据第一部分的研究结果和一些文献的报道，我们认为有必要筛选确定电磁场作用敏感细胞，为此又设定了第二部分的研究内容。一般认为，电磁场对细胞的生物学效应受电磁场自身多因素的影响，如电磁场频率、强度、暴露时间和模式等。然而，相对于电磁场自身因素的影响而言，生物系统(细胞／组织等)的来源和辐照时的具体状态更能影响实验的最终结果。Leszczynski等发现SAR为2.4 W／kg的GSM 900射频场辐照EA．hv926细胞1小时可引起38个蛋白质的表达发生改变；而在相同条件下，EA．hy926v1细胞(EA．hy926的转化细胞株)中有另外45个蛋白点的表达发生变化，说明射频电磁场影响了两种细胞中不同蛋白质的表达。Sul等将4种不同组织来源的细胞暴露于2 mT正弦磁场中，每天辐照1、3和6小时，共14天，发现4种细胞对电磁场的反应性不同。生物系统的遗传特性决定了各生物系统对不同频率电磁场的反应敏感性不同，是导致目前许多研究结果不一致的原因之一。因此，我们认为只有以电磁场敏感细胞为研究对象，才能正确揭示电磁场的生物学效应和作用机制。在本博士论文的第二部分，我们利用传统双向电泳技术进行了电磁场相对敏感细胞的筛选，以为今后的深入研究奠定基础。第一部分：应用双向电泳技术研究电磁场对人乳腺癌细胞蛋白质表达的影响流行病学调查显示极低频电磁场暴露可引起乳腺癌发病率增高。在以往研究的基础上，我们选用50 Hz 0.4 mT正弦磁场对人乳腺癌细胞MCF-7进行辐照和假辐照处理24小时，提取总蛋白质进行双向电泳。银染图谱经PDQuest7.1软件分析显示，磁场辐照组中有6个蛋白质斑点的表达量发生显著改变，同时，在磁场辐照组中有19个蛋白点消失和19个蛋白点新出现。3个差异表达的蛋白质斑点经LC-ESI-IT串联质谱分析，鉴定为RNA结合蛋白调节亚基、蛋白酶体β亚基7型前体和翻译调控肿瘤蛋白。为系统研究射频电磁场对MCF-7细胞蛋白质表达的影响，选择不同时间(1、3、6、12和24小时)、不同强度(SAR为2或3.5 W／kg)、不同辐照模式(5 min-on／10 min-off或连续辐照)的217 Hz调制的全球移动通讯系统(GSM)1800 MHz射频电磁场辐照细胞，然后提取总蛋白质进行双向电泳。结果显示，在本实验条件下，1800 MHz射频电磁场对MCF-7细胞蛋白质表达谱有一定影响，但不明显，且依赖于电磁场暴露的强度、时间和模式。在上述基础上，选择作用较为明显的实验参数(SAR为3.5 W／kg，间断辐照3小时)对MCF-7细胞进行辐照，提取总蛋白质进行荧光差异双向电泳(DIGE)。采用“Decyder”软件进行分析，发现5个蛋白质点表达受电磁场作用上调。三个蛋白经MALDI-TOF／TOF质谱鉴定为CLIC1蛋白、翻译调控肿瘤蛋白和硫醇特异性抗氧化蛋白。另外两个蛋白未得到鉴定。第二部分：应用双向电泳技术筛选电磁场敏感细胞选用来源于不同物种或组织的细胞，包括中国仓鼠肺成纤维细胞CHL、小鼠胚胎成纤维细胞NIH3T3、大鼠肾上腺嗜铬细胞PC12、人眼晶状体上皮细胞SRA01／04、人羊膜上皮细胞FL、人早幼粒白血病细胞HL60和人皮肤成纤维细胞HSF分别暴露于0.4 mT的50 Hz磁场24小时或SAR为3.5 W／kg的1800 MHz射频电磁场间断辐照3小时后，立即提取全蛋白，进行双向电泳。结果显示，工频磁场辐照后，PC12和FL细胞中分别检测到差异表达蛋白点共14个和23个，分别占总检测蛋白点数2.2％和3.2％，而在其余细胞中仅检测到小于1.4％的蛋白质表达发生变化；射频电磁场辐照后，NIH3T3、FL和HL60细胞中分别检测到表达差异蛋白点共20个、23个和17个，分别占总检测蛋白点数2.4％、3.5％和2.0％，在其余细胞中仅检测到小于1.3％的蛋白质表达发生变化。根据检测到的差异点数量及其占总检测蛋白质点数的百分比，结合第一部分结果，初步认为在本实验条件下，MCF-7、PC12和FL细胞为工频磁场的相对敏感细胞，NIH3T3、FL和HL60细胞为射频电磁场的相对敏感细胞。结论：1．0.4 mT50 Hz磁场可诱导人乳腺癌细胞MCF-7蛋白质表达谱发生显著改变。已鉴定的三个差异蛋白和细胞骨架结构存在一定联系，提示细胞骨架很可能是电磁场作用的靶标。2．1800 Mnz射频电磁场处理并不能显著改变MCF-7细胞的蛋白质表达模式，提示MCF-7细胞对较高频率的射频电磁场反应性较弱。同时，该弱作用受电磁场辐照强度、作用时间和作用模式等参数的影响。3．细胞遗传和／或表观遗传(epigenetic)背景决定了其对电磁场的敏感性。在本实验条件下，MCF-7、PC12和FL细胞为工频磁场的相对敏感细胞，NIH3T3、FL和HL60细胞为射频电磁场的相对敏感细胞。不同细胞对电磁场的敏感性不同，同种细胞对不同频段的电磁场反应也可以不一样。4．蛋白质组学技术可以应用于电磁场生物学效应及机制研究。但由于环境低强度电磁场是一种弱作用因素，易受外界其它因素和细胞本身状态的影响；而蛋白质组学这种高通量技术本身是以牺牲敏感性为代价的，在应用于低强度电磁场这种弱效应研究的过程中，还存在一定的不足。因此，一方面需探索发展更灵敏、更高通量的技术；另一方面需通过国际合作，探索建立蛋白质组学技术在电磁场生物学效应研究中应用的技术标准和规范。从目前的情况看，由于蛋白质学技术本身存在的局限性，对所获得的结果还需谨慎对待，并应通过低通量的常规方法验证。5．通过对传统双向电泳技术与DIGE技术的比较，我们认为DIGE技术在电磁场应用中并不比传统双向电泳具有更大的优势。本博士论文的创新点：1．在国际上率先采用蛋白质组学技术进行了电磁场对人乳腺癌细胞蛋白质表达影响的研究及电磁场敏感细胞的筛选工作，在技术手段上有所创新。2．首次在国际上报道0.4 mT 50 Hz磁场可诱导人乳腺癌细胞MCF-7蛋白质表达谱发生显著改变，并鉴定了3个差异蛋白。3．首次从蛋白质组学的角度证明MCF-7细胞对1800 MHz射频电磁场的反应性较弱。4．在国际上首次利用蛋白质组学技术筛选了电磁场的敏感细胞，确定在本实验条件下，MCF-7细胞、PC12细胞和FL细胞为工频磁场的相对敏感细胞；NIH3T3细胞、FL细胞和HL60细胞为射频电磁场的相对敏感细胞。5．通过对两类电磁场的平行研究，证明不同细胞对电磁场的敏感性不同，而同一种细胞对不同电磁场的反应不同。更多还原

【Abstract】 Electromagnetic radiation has become a rapid-increasing and universal-existing environment risk factor due to the rapid development of mobile communication and electric power transmission. The health risk assessment of electromagnetic radiation will protect population from the possible hazards and prmote a sustaining development of technology, economics and society. Epidemiological studies have identified a potential positive association between exposure to extremely low-frequency (ELF) electromagnetic fields and leukaemia and breast cancer. Radiofrequency (RF) electromagnetic fields from mobile phones might affect the function of the central nervous system and induce malignant pathological changes such as brain cancer. The epidemiological findings drive the experimental studies on bioeffects and mechanisms of action of EMF. However, the data from in vivo and in vitro studies are inconsistent, and it is difficult to make a clear conclusion on the overall health impact induced by EMF. Moreover, lack of a coherent hypothesis for a mechanism by which EMF might interact with biological systems has limited such studies to a phenomenological rather than a mechanistic approach. There is thus pressing motivation to delineate the biological effects and the underlining mechanisms of low-energy EMF.Gene expression regulation is believed to play a role in the various physiological ang biochemical changes of biological system induced by EMF. Some studies have found EMF could alter the transcription of certain genes, including pro-oncogene, apopotosis-related gene, genes regulating the cell cycle and so on. For example, c-myc, c-jun, c-fos, p21. egr-1, bax and GADD45 were found to be ELF EMF-resposive genes, and Nurr1, bax, GADD45, hsp70, c-jun, c-myc and p21 were reported to be affected by RF EMF exposure. The cell behavior will be affected if the gene expression change results in variation of protein expression and/or modification. Therefore, exploring the effects of EMF on protein would be more direct to elucidate the biological effects of EMF at a cell or organism level. There are so far only a few studies looking at the protein expression or modification after exposing to EMF, and focusing on ornithine decarboxylase (ODC), heat shock proteins (HSP27/70), and some signalling proteins of certain signal transduction pathways. Those studies are hypothesis-driven, and might introduce bias while selecting end-points to be detected. Meanwhile, the selected end-points are scattered, and the experiments were not designed in a systematic way. Therefore, it is difficult to evaluate the EMF-induced bioeffects based on the obtained data. We reason it is necessary to analyze the effect of EMF on protein at a proteomic scale. Theoretically, a proteomics approach allows simultaneously monitoring hundreds or even thousands of proteins in a sample. In 2001 our laboratory and the Leszczynski group in Finland initiated the application of proteomics in EMF research.To evaluate the potential co-carcinogenic effect of EMF, and compare the effects of two different environmental prevailing EMF, the first part of this dissertation intend to reveal the differential protein expressions in human breast cancer cells (MCF-7) induced by 50 Hz ELF EMF or 1800 MHz RF EMF using two-dimensional electrophoresis (2-DE) technique, then identify differential expression proteins by mass spectrum analysis.Based on the results of the first part and other studies, we further reason it is necessary to scan EMF-sensitive cell types and set it up as the second part of the dissertation. It is realized that the bioeffects of EMF depend on many factors, such as exposure frequency, intensity, duration, and pattern. However, the origin and status of tested biological systems (cell, tissue) may be the most important factor to affect the results. Leszczynski and his colleagues, for example, found 38 protein spots with altered expression levels in the EA.hy926 cell line following to 900 MHz RF exposure for 1 h at SAR of 2.4 W/kg, whereas in the EA.hy926vl cell line (a subcloning of EA.hy926) 45 different protein spots showed altered (Nylund R, Leszczynski D. Mobile phone radiation causes changes in gene and protein expression in human endothelial cell lines and the response seems to be genome- and proteome-dependent. Proteomics, 2006, 6: 4769-4780). Sul exposed four cell lines of different origins to sinusoidal electromagnetic fields at 2 mT for 1, 3 or 6 hours per day. After 14 days, he found cell type-specific reaction to EMF (Sul AR, Park SN, Suh H. Effects of sinusoidal electromagnetic field on structure and function of different kinds of cell lines. Yonsei Med J, 2006, 47: 852-861). Thanks to these studies, the message is becoming clear: the diverse genetic backgroud of biological systems make them react differently to EMF, the key in EMF research is to identify EMF-sensitive cells and then explore the effects and mechanism of action of EMF. Therefore, the second part of my dissertation is to screen EMF-sensitive cell lines using 2-DE approach.Part I: The effects of EMF on protein profiles in human breast cancer cellsTo reveal the effect of ELF MF on protein expression, MCF-7 cells were exposed to 50 Hz, 0.4 mT ELF MF for 24 h. Immediately after the exposure and sham-exposure, proteins were extracted from the cells and subjected to be analyzed by 2-DE. The analysis of protein distribution in the gels was carried out with the aid of the PDQuest software, version 7.1. The results showed that 6 spots have been statistically significantly altered, and 19 additional spots were detected only in exposed group while 19 ones were detected only in control group. Three proteins were identified by LC-ESI-IT tandem MS as RNA binding protein regulatory subunit, proteasome subunit beta type 7 precursor, and tanslationally controlled tumor protein.To analyze the effect of RF EMF on protein expression, MCF-7 cells were exposed to 1800 MHz RF EMF modulated by 217 Hz (or sham-exposed) at different duration (1, 3, 6, 12 and 24 h), different intensities (SAR of 2 or 3.5 W/kg) and different patterns (5 min-on and 10 min-off exposure, or continuous exposure). After exposure or sham-exposure, total proteins were extracted and analyzed by 2-DE. The analysis of protein distribution in the gels was carried out as above. The results showed the protein expression changes induced by 1800 MHz RF EMF in MCF-7 cells were faint and depended on exposure intensity, duration and pattern.We further analyzed the protein expression change induced by RF EMF using fluorescence difference gel electrophoresis (DIGE). MCF-7 cells were intermittent exposed to 1800 MHz RF EMF at SAR of 3.5 W/kg for 3 h (under this exposure condition, we found 18 proteins expression were altered in 2-DE approach). The total proteins were extracted and separated by DIGE, and the three-color images were analyzed by the "Decyder" software. The results showed that 5 proteins were up-regulated by RF EMF. Three of these could be identified in MALDI TOF/TOF as CLIC 1 protein, translationally controlled tumor protein 1, and thiol-specific antioxidant protein.Part II: Screening EMF-sensitive cells types using 2-DEChinese hamster lung fibroblast cells line CHL, rat skin fibroblast cells line NIH3T3, rat pheochromocytoma cells line PC 12, human lens cells line SRA01/04, human amnion epithelial fibroblast cells line FL, human leukemic cell line HL60 and human skin fibroblast cell line HSF were exposed to 0.4 mT ELF MF for 24 h or 1800 MHz RF EMF at SAR of 3.5 W/kg for 3 h. The extracted proteins were separated using 2-DE respectively. Compare to sham-exposure group, ELF MF exposure induced 14 and 23 differentially expressed proteins in PC 12 and FL cells, representing 2.2 % and 3.2 % of the total detected protein spots, respectively. Only less than 1.4 % of the total protein spots were changed by ELF MF in other cell types. On the other hand, RF EMF exposure produced 20, 23 and 17 differentially expressed proteins in NIH3T3 cells, FL cells and HL60 cells representing 2.4 %, 3.5 % and 2.0 % of the total protein spots detected, respectively. Meanwhile, only less than 1.3 % of the total proteins were altered by ELF MF in other cells types. Combined the results of Part I, we concluded MCF-7, PC 12 and FL cells as ELF MF-sensitive cells; and NEH3T3, FL and HL60 cells as RF EMF-sensitive cells based on the absolute number of differentially expressed proteins and their ratios to the total protein spots detected.The main conclusions are:1. 0.4 mT 50 Hz ELF MF could significantly alter proteins expression in MCF-7 cells. The three identified proteins are related with the cellular cytoskeleton, implying cytoskeletal might be an interaction target to EMF.2. 1800 MHz RF EMF did not significantly alter protein expression in MCF-7 cells, suggesting this cell type react weakly to RF EMF. The protein expression changes induced by 1800 MHz RF EMF in MCF-7 depend on exposure intensity, duration and pattern.3. The genetic and/or epigenetic background of a biological system determines its response to EMF. Under the experimental conditions, we identified that MCF-7, PC12 and FL cells are ELF MF-sensitive cells, and NIH3T3, FL and HL60 cells are RF EMF-sensitive cells. Meanwhile different cell types have different response to EMF, and the same cell type reacts differently to different frequency of EMF.4. Proteomics approach is applicable in EMF research to investigate bioeffects and mechanism of action. However, due to the weak interaction of low intensity EMF with a biological system, and the low sensitivity of high through-put technology such as proteomics, there are defects in applying such a technique in elucidation EMF effects. It is urgert to develop more sensitive and more high through-put techniques, and to establish special technique criterion for using proteomics in EMF research. Due to the limitations of proteomics analysis, the produced candidates should be validated using non-HTST methods.5. After comparing the application of classical 2-DE and DIGE, we do not recommend DIGE as a better method in EMF research.Innovation points of this dissertation:1. We first employ proteomics approach in revealing the effects of EMF on protein expression in MCF-7 cells and in screening the EMF-responsive cell types. Thus, there is innovation in selecting experimental technique.2. We first report the differential proteins expression profile of MCF-7 cells induced by 0.4 mT 50 Hz ELF MF, and identified three ELF MF-responsive proteins3. We first report the MCF-7 cell line is not a sensitive cell line to 1800MHz RF EMF from a point of view of proteomics analysis.4. We first identify MCF-7, PC12 and FL cells are ELF MF-sensitive cells, and NIH3T3, FL and HL60 cells are RF EMF-sensitive cells under the current experimental conditions. 5. By conducting a parallel proteomics study on ELF MF and RF EMF, we conclude that different cells react differently to EMF, and the same cell type has different response to different frequency of EMF.更多还原