1. Introduction
Microwave technology, non-ionizing electromagnetic radiation ranging from 300 MHz to 300 GHz, has been widely used in various fields, such as mobile communication, medicine, industrial synthesis and so on [
1,
2,
3,
4]. In addition, the anxiety and speculation about the potential health hazards caused by microwaves have been growing in recent years [
5,
6]. In the past decades, most of the researchers aimed to uncover the biological effects caused by microwaves with a single frequency at indicated power density [
7,
8]. Our group has previously reported that S band microwaves, ranging from 2 GHz to 4 GHz, could cause significant injuries on several organs and tissues, including the nervous system, cardiovascular system, reproductive system and immune system [
9,
10,
11,
12]. Moreover, the potential underlying mechanisms were explored [
7,
13,
14]. However, people are always exposed to complex microwave environment with different frequencies or power densities in daily life [
7,
8]. For example, the remote sensing satellites constantly equipped with synthetic aperture radar simultaneously emit a signal at frequencies of L-band (1–2 GHz), C-band (4–8 GHz) and X-band (8–12 GHz) [
1]. Therefore, it is important to explore the biological effects and underlying mechanisms caused by multi-frequency microwaves.
The immune system is the first line of defense against pathogens and foreign molecules such as bacteria, viruses and mutated or dead cells. It has been widely demonstrated that the immune system is one of the most sensitive tissues to both ionizing and non-ionizing radiation [
15,
16]. Unlike ionizing radiation, the biological effects of non-radiation on immune cells, tissues and organs are controversial. Limited data suggested that microwaves could affect immune functions, including phagocytosis, proliferation of immune cells and antibody production [
17,
18]. However, opposite results might be presented when exposed to microwaves with different parameters, such as average power density, exposure period and so on [
19,
20].
Non-ionizing radiation, including microwave radiation, could produce both thermal effects and non-thermal effects. We have previously established an animal model exposed to low frequency and low power density to evaluate the non-thermal effects. In this study, we exposed Wistar rats to multi-frequency microwaves of 1.5 GHz (L band) and 4.3 GHz (C band) at an average power density of 10 mW/cm2. We showed that multi-frequency microwaves slightly injured the thymus and spleen, but not the bone marrow, which resulted in the decrease in white blood cells (WBC) and lymphocytes. Multi-frequency microwave induced immune suppressions were also demonstrated by transcriptomic and proteomic analysis. Further studies should be conducted to clarify the underlying mechanisms according to the results of transcriptomics and proteomics.
3. Discussion
In recent years, the potential health hazards caused by microwaves have been attracting more and more attention. The World Health Organization (WHO) had classified electromagnetic field radiation as a new form of environmental pollution and “possibly carcinogenic to humans” [
2,
3]. Microwave exposure could induce biological responses in various tissues via both thermal and non-thermal mechanisms [
4]. Our group has been focusing on microwave-induced non-thermal effects. We previously reported that microwaves could reduce learning and memory ability and decrease the synaptic plasticity of the hippocampus, as well as impair cardiac function in animal models [
5,
6,
7,
8]. Now, we are greatly interested in investigating biological responses to multi-frequency microwaves [
5,
9,
10].
The immune system could rapidly respond to external and internal stimuli and play pivotal roles in protecting the body against viral infection, as well as cancer initiation and development. It has been reported that hematopoietic cells are more sensitive to microwaves than cells in many other tissues [
11]. Microwave ablation has been used as a potential adjuvant therapy for cancer therapy. Clinical studies have suggested that microwave ablation induced anti-tumor effects, both through directly destroying tumor tissues and via activating immune responses in patients. Moreover, thermal effects have been recognized as the major mechanism for immune activation during microwave ablation [
12,
13,
14,
15]. However, potential non-thermal effects of microwaves on immune cells were also investigated in animal models after exposure to low-level electromagnetic field [
16,
17]. Our previous in vitro study demonstrated that microwaves induced cellular apoptosis and impaired cytotoxicity in NK cells [
18]. In this study, we found that both single and multi-frequency microwaves decreased WBC and lymphocytes in peripheral blood, indicating that immune suppressions were induced by microwaves.
With the growing application of microwave technologies in our daily life, people are constantly under long-term environmental exposure to multi-frequency microwave radiation. For example, the satellite communication system and synthetic aperture radar simultaneously emit multi-frequency microwaves located in the L and C bands [
1,
19,
20]. Therefore, we evaluated the biological effects on the immune system in rats after exposure to multi-frequency microwaves of 1.5 GHz (L band) and 4.3 GHz (C band) with the average power density of 10 mW/cm
2. Our data suggested that both single and multi-frequency microwave exposure evoked immune suppressive responses, such as down-regulation of WBC and lymphocytes in peripheral blood. Otherwise, no obvious difference could be detected between single and multi-frequency exposure. However, we found that multi-frequency microwaves could decrease the percentage of T lymphocytes and increase the percentage of B lymphocytes much more strongly than single frequency microwaves, suggesting that T lymphocytes might be more sensitive to multi-frequency microwaves.
The cytokines and chemokines secreted from other tissues could stimulate or inhibit immune cells, while the immune cells in peripheral blood could enter into various tissues. The complex interactions between the immune system and other tissues increase the difficulties in studying the immune regulatory mechanisms. To clarify the mechanisms underlying microwave mediated immune dysfunctions, researchers evaluated the genotoxicity, cell proliferation and level of calcium in immune cells in vitro, as well as immunologically relevant parameters in vivo. However, controversial results have been reported due to differences in exposure parameters, immune cells and evaluation methods [
17,
21,
22,
23,
24,
25,
26]. Both transcriptomics and proteomics are efficient and high throughput tools for basic studies and have been widely used in investigating potential mechanisms. In this study, we showed that multi-frequency microwaves induced much more DEGs and DEPs than single frequency microwaves in the spleen, suggesting that multi-frequency microwaves produced much stronger biological effects on the spleen. Similar results were also observed in DEGs from peripheral blood at the transcriptional level. Unexpectedly, the number of DEPs in peripheral blood after single or multi-frequency microwave radiation were at the similar level. Moreover, the DEPs in peripheral blood were much greater than that in the spleen, which was contrary to the DEGs. We speculated that the DEPs in peripheral blood were not only derived from the expression of immune cells, but were also secreted from other tissues. In addition, proteins secreted from other tissues could in turn stimulate or inhibit protein expression in immune cells. Therefore, the DEPs in peripheral blood might be a potential indicator for the whole body, not only for the immune system. Furthermore, bioinformatics analysis suggested that several T lymphocytes’ development, differentiation and activation associated genes were down-regulated both in peripheral blood and in the spleen, while B lymphocytes’ activation related genes were up-regulated, indicating that T lymphocytes were much more sensitive to multi-frequency microwaves. However, the mechanisms should be further investigated.
In conclusion, both 1.5 GHz (L band) and 4.3 GHz (C band) microwaves could cause immune suppression at the average power density of 10 mW/cm2. However, multi-frequency microwaves of 1.5 GHz and 4.3 GHz produced much more impressive immune suppressive responses, especially in inhibiting T lymphocytes. The transcriptomic and proteomic analysis suggested that multi-frequency microwaves regulated numerous genes associated with immune activation and metabolism, both at the mRNA and protein level. Moreover, several T lymphocytes’ development, differentiation and activation associated genes were down-regulated, while B lymphocytes’ activation related genes were up-regulated. Our data provided useful information for exploring potential mechanisms underlying multi-frequency induced immune suppression.
4. Materials and Methods
4.1. Animals
One hundred six- to eight-week-old male Wistar rats (180 g ± 20 g) (Vital River Laboratory Animal Technology, Beijing, China) were randomly divided into four groups: sham-exposed group (Sham), 1.5 GHz microwave (L10) exposed group, 4.3 GHz microwave (C10) exposed group and multi-frequency (1.5 GHz and 4.3 GHz) microwave (LC10) exposed group. The animal experiments were approved by the Institutional Animal Care and Use Committee at Beijing Institute of Radiation Medicine (Beijing, China), and all experimental procedures were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.
4.2. Microwave Exposure
The microwave exposure system, based on a klystron amplifier model JD 2000 (Vacuum Electronics Research Institute, Beijing, China), had been previously reported [
5,
8]. Rats in L10, C10 and LC10 groups were exposed to microwaves of 1.5 GHz, 4.3 GHz and multi-frequency (1.5 GHz and 4.3 GHz), respectively, at the average power intensity of 10 mW/cm
2. Briefly, rats in microwave exposed groups were fixed in fan-shaped boxes, which were made of plexiglass and free of metal, and were exposed in the far field of a 12 dB-gain horn antenna. The power density was measured with a calibrated waveguide antenna over an area of 10 cm intervals horizontally on the plane with respect to the rat’s position. The distribution of the microwave density was described previously [
5]. Rats in the Sham group were processed in parallel with those in the microwave exposed groups, but without microwave exposure.
4.3. Phenotypes of Immune Cells
At 6 h, 7 d and 14 d after microwave exposure, the rats in Sham, L10, C10 and LC10 groups were anaesthetized by 1% pentobarbital sodium at 30 mg/kg. Peripheral blood was collected from IVC. The immune cells in peripheral blood, including WBC, lymphocytes and neutrophils, were counted by automatic blood cell counter (Sysmex XN-1000, Kobe City, Japan).
Moreover, the subtypes of lymphocytes were detected by flow cytometry (C6, BD, Franklin Lake, NJ, USA) after being labeled with the corresponding antibody panels, as followsd: Panel 1 for NK cells (CD45, CD3, CD161A), Panel 2 for T lymphocytes and B lymphocytes (CD45, CD3, CD45R), Panel 3 for the phenotype of T lymphocytes (CD4, CD8a) (eBioscience, San Diego, CA, USA).
4.4. Secretion of Cytokines
At 6 h and 7 d after microwave exposure, peripheral blood was collected as described above. The sera were separated by centrifugation and the expression of interleukin (IL)-1α, IL-4, IL-6, IL-10, IL-17A and IFN-γ was analyzed by Cytokine&Chemokine 22-Plex Rat ProcartaPlexTM Panel (Invitrogen, Carlsbad, CA, USA).
4.5. Histopathological Analysis
At 6 h, 7 d, 14 d and 28 d after microwave exposure, rats in each group were euthanized and the corresponding tissues were removed. For the thymus and the spleen, tissues were fixed in 10% buffered formalin solution, embedded in paraffin and cut at 3 μm thick in the coronal plane. For bone marrow, tissues were fixed with Helly solution immediately after isolation for 24 h, and thereafter fixed with 10% buffered formalin solution for 3 weeks, followed by decalcification for another 3 weeks. Then, the tissues were embedded in paraffin and cut at 3 μm thick in the coronal plane. The sections were stained with H&E, and the histopathological changes were observed by light microscopy (Leica, Wetzler, Germany).
4.6. Transmission Electron Microscopy (TEM)
The spleens of rats were collected at 7 d after microwave exposure, as described above. After being fixed in 2.5% glutaraldehyde and 1% osmium acid in sequence, the tissue blocks were processed with graded ethyl alcohols and embedded in EPON618. Seventy nm thin slices were laid on copper mesh and were then stained with heavy metals, uranyl acetate and lead citrate. The ultra-structures of the spleen in each group were observed and photographed by TEM (Hitachi, Tokyo, Japan).
4.7. Transcriptomics Analysis
The peripheral blood and spleen were collected at 7 d after exposure. Total RNA was isolated using the Trizol Reagent (Invitrogen Life Technologies, Carlsbad, CA, USA), and cDNA library was constructed by using NEBNext® Ultra™ Directional RNA Library Prep Kit for Illumina® (NEB, Ipswich, MA, USA). Then, the cDNA library was purified (AMPure XP system, Beckman Coulter, Brea, FL, USA) and quantified using the Agilent high sensitivity DNA assay on a Bioanalyzer 2100 system (Agilent, Palo Alto, CA, USA). Finally, the samples were then sequenced on a NovaSeq 6000 platform (Illumina, San Diego, CA, USA). The DEGs were analyzed by DESeq2 with screened conditions as follows: expression difference multiple |log2FoldChange| > 1, significant q-value ≤ 0.05.
Then, we used topGO to perform GO enrichment analysis to find the GO term with significantly enriched DEGs (p < 0.05), the p-value of which was calculated by hypergeometric distribution method. The enrichment analysis of KEGG pathway of DEGs also focused on the significant enrichment pathway with p < 0.05.
4.8. Proteomics Analysis
The peripheral blood and spleen were collected at 7 d after exposure. The proteins were extracted, and protein concentration was determined using the Bradford method. A 100 μg extracted protein from each sample was mixed, enzymatic degraded and desalinated. For spectral library generation, the proteomics analyses were performed using an U3000 UHPLC system (Thermo Fisher, Waltham, MA, USA) coupled with an Orbitrap fusion mass spectrometer (Thermo Fisher, Waltham, MA, USA) operating in the DDA mode. While the proteomics analysis of each sample was operated in DIA, the acquisition method consisted of one MS1 scan (350 to 1350
m/
z, resolution 120,000 maximum injection time 50 ms, AGC target 4E5) and 60 segments at varying isolation windows from 349
m/
z to 1500
m/
z. Stepped normalized collision energy was 35. Data analysis was carried out as described in Bruder et al., with minor modifications [
27]. Briefly, data extraction and extraction window were set to “dynamic” with correction factor 1 and identification was set to “normal distribution
p-value estimator” with
q-value cutoff of 0.01. The profiling strategy was set to “iRT profiling” with
q-value cutoff of 0.01. Ultimately, protein inference was set to “from search engine”, protein quantity was set to “Average precursor quantity” and smallest quantitative unit was set to “Precursor ion” (summed fragment ions).
GO and InterPro (IPR) analysis were conducted using the interproscan-5 program against the non-redundant protein database and the databases Clusters of Orthologous Groups (COG). The KEGG was used to analyze the protein family and pathway. The enrichment pipeline was used to perform the enrichment analysis of GO and KEGG.
4.9. Real-Time Reverse Transcription Polymerase Chain Reaction (RT-PCR)
The peripheral blood and spleen were collected at 7 d after exposure. The total RNA was isolated by the Trizol Reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) from peripheral blood and the spleen, and then the cDNA was synthesized by using revert aid first strand cDNA synthesis kit (Thermo Fisher Scientific, Waltham, MA, USA). The expression of indicated DEGs was analyzed by real-time RT-PCR using 2 × RealStar Green Fast Mixture with ROX II (Genstar, Beijing, China) on 7500 Fast Real-Time PCR System (Applied Biosystem Technologies, Foster City, CA, USA). The relative expression of DEGs was calculated by 2
−∆CT, using rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as control. The primers for each gene were shown in
Supplementary Table S5.
4.10. Mass Spectrometry Analysis
Protein samples for mass spectrometry analysis were prepared as described above. After protein extraction, digestion and desalination, the samples were purified by U3000 UHPLC system (Thermo Fisher Scientific, Waltham, MA, USA). The expressions of DEPs were detected by parallel reaction monitoring (PRM).
4.11. Statistical Analysis
Data were presented as mean ± s.e.m. and analyzed by GraphPad Prism software version 6.0 (GraphPad software, San Diego, CA, USA). Grouped data were analyzed by one-way ANOVA followed by Bonferroni post hoc tests. Longitudinal data were analyzed by a two-way repeated measure ANOVA followed by Bonferroni post hoc tests. Differences were considered significant at two-sided p < 0.05.