Transcriptomic profiling reveals gene expression in human peripheral blood after exposure to low-dose ionizing radiation.

Although the health effects of exposure to low-dose ionizing radiation have been the focus of many studies, the affected biological functions and underlying regulatory mechanisms are not well-understood. In particular, the influence of radiation exposure at doses of less than 200 mGy on the regulation of genes and pathways remains unclear. To investigate the molecular alterations induced by varying doses of low-dose radiation (LDR), transcriptomic analysis was conducted based on ribonucleic acid (RNA) sequencing following exposure to 50 and 150 mGy doses. Human peripheral blood was collected, and the samples were divided into three groups, including two treatments and one control (no radiation). A total of 876 (318 upregulated and 558 downregulated) and 486 (202 upregulated and 284 downregulated) differentially expressed genes (DEGs) were identified after exposure to 50 mGy and 150 mGy, respectively. Most upregulated genes in both the 50 mGy and 150 mGy groups were associated with 'antigen processing and presentation,' which appeared to be the major targets affected by LDR exposure. Several interacting genes, including HLA-DQA1, HLA-DQA2, HLA-DQB2, HLA-DRB1, and HLA-DRB5 were mapped to 'antigen processing and presentation,' 'immune system-related diseases' and the 'cytokine-mediated signaling pathway,' suggesting that these genes might drive the downstream transmission of these signal transduction pathways. Our results suggest that exposure to LDR may elicit changes in key genes and associated pathways, probably helping further explore the biological processes and molecular mechanism responsible for low-dose occupational or environmental exposures in humans.

INTRODUCTION

The United Nations Scientific Committee on the Effects of Atomic Radiation defines low-dose radiation (LDR) that may cause deterministic effects (tissue effects) as those of 200 mGy or less and low dose rates as 0.1 mGy.min−1 (averaged over one hour or less) for radiation such as external X-rays and gamma rays [1]. Humans are continuously exposed to low doses of natural and anthropogenic ionizing radiation every day. The increasing use of ionizing radiation in medical procedures, industrial applications, and scientific research continues to increase the potential for exposure. In America, England and Australia, the average yearly medical exposure dose increased dramatically between 1982 and 2006 [2]. In America, for example, the average yearly medical exposure dose increased from 0 to 5 mGy in 1982 to 30 mGy in 2006; and the average yearly medical exposure dose doubled in England and tripled in Australia between 1982 and 2006 [2]. In China, with the increasing use of diagnostic and therapeutic procedures in the fields of radiology and nuclear medicine, the annual per capita dose has doubled over the past two decades [3]. Exploring the biological effects of exposure to low-dose/low-dose-rate ionizing radiation is therefore necessary for adequately protecting both individuals and the environment [4-5]. A better understanding of the biological processes affected by of LDR could lead to advances in patient care and reduce the detrimental impact of radiation therapy on patients [6].

Since the 1970s, the linear no-threshold (LNT) model has been used to estimate the health risks of LDR by extrapolating the risks determined in high-dose studies [7-8]. However, advances in LDR biology and cell molecular techniques have demonstrated that the LNT model does not appropriately reflect the biological or health effects in the low-dose range [9-12]. In fact, LDR can positively affect biological processes such as immunity, DNA repair, and cellular stress resistance [11, 13–15], and the dose–effect relationship could be hormesis, which is a biphasic dose response relationship in which there are beneficial effects at low doses and harmful effects at high doses [16-17]. There have been many large-scale epidemic studies in populations exposed to low doses of radiation, including atomic bomb survivors [18-20], those who have worked with high-background radiation [21-23] and those who have received medical diagnosis and disease treatment [24-26], but the association between LDR and carcinogenesis is still controversial. For example, among the atomic bomb survivors or radiation workers, LDR exposure has a positive association with leukemia, but no radiation-associated risks for either Hodgkin lymphoma or multiple myeloma were found [18, 22–23]. Low-dose exposure from medical X-ray examination and treatment is a risk factor of disease, such as atherosclerosis, leukemia and brain tumors [24-25]. However, other studies have shown that mortality among both medical professionals and nuclear workers who are exposed to LDR is similar to that among non-exposed workers [27-31]. Moreover, even though epidemiological studies of exposed human cohorts have found that some diseases appear under LDR exposure, such as visual disturbances, leukemia and cataract [22, 32–33], there are still uncertainties regarding the determining mechanisms associated with low-dose occupational or environmental exposures in humans. The issue remains controversial, and the exact health risks of exposure to LDR are unclear [34].

Advanced genomic tools have provided new insights related to the hazard assessment of radiation exposure; however, most radiation-inducible gene expression profiling has been performed using hybridization-based microarrays [35-37]. Microarray technology presents some limitations, including indirect quantification from hybridization-signal intensities [38], background and cross-hybridization problems [39] and reproducibility issues [40]. In addition, previously published work has mainly focused on differentially expressed genes (DEG) and their potential to be used as biomarkers or for individualized biodosimetry [36, 41–44]. There are very few existing studies on the regulation of genes and related pathways, especially the molecular alterations caused by exposure to a dose of less than 200 mGy. The effects of radiation on gene expression have biological consequences [45-47]. Recently published data show that LDR can elicit changes in metabolic and immune pathways, potentially increasing the risk of immune dysfunctions and metabolic disorders [14, 37]. Further investigation is needed to elucidate the possible transcriptional regulation mechanisms in response to LDR.

Here, we present the whole-transcriptome gene expression profiling of LDR-induced human peripheral blood through RNA sequencing and pathway enrichment analysis. RNA-Seq presents several advantages over microarrays, including the following: (i) RNA-Seq is not dependent on prior knowledge about the target sequence, (ii) it exhibits a large dynamic range and high sensitivity because of its digital nature, and (iii) the quantification of the transcriptome is more accurate because the quantification of each transcript is directly based on the digital counts of the transcript [48]. This study provides novel insights into specific gene expression and pathway-level differences that are associated with different radiation doses based on technological improvements and combined approaches. The overarching goal of this study is to use global transcriptomic and pathway analysis to identify candidate genes and pathways for understanding the biological processes and molecular mechanisms that occur in human peripheral blood exposed to LDR.

MATERIALS AND METHODS

Sample collection and radiation

In this study, blood samples from nine healthy adult donors (five males and four females; aged 20–48 years) were collected in PAXgene Blood RNA tubes for transcriptomic sequencing. The eligibility of the donors was evaluated using questionnaires and regular medical examinations. Peripheral blood samples (5 ml) were collected from each subject. The subjects had no history of chronic disease, substance abuse, smoking, or toxic chemical exposure. In addition, they had not been exposed to radiation nor did they have a history of viral infections during the months preceding the study. The present study was approved by the Ethics Committee. Written informed consent was obtained from all human subjects prior to enrollment in the present study.

Irradiation with 137Cs γ-rays was performed in the Beijing Radiation Center (Beijing, China). The exposure setup was calibrated by physical measurement using a tissue-equivalent ionizing chamber. The radiation dose rate was calculated using the source radioactivity and the distance between the source and samples: dose rate of 58.4 cGy/h corresponds to a source-sample distance of 70 cm. The homogeneous irradiation field was 30 × 30 cm; the samples were placed within a 5 cm-radius circle, and the uncertainty of the calibration was 1%. A control group (non-radiation) and two different radiation groups were used (50 and 150 mGy). Nine blood samples were divided into three dose groups, and each group included three biological replicates. Following irradiation, blood samples were incubated for 6 hours at 37°C in RPMI-1640 medium of equal volume (Thermo Fisher Scientific, Inc., Waltham, MA, USA), supplemented with 10% fetal bovine serum (FBS; HyClone; GE Healthcare Life Sciences, Logan, UT, USA), see Supplementary Fig. S1.

RNA-seq and analysis

Red blood cell (RBC) lysis was performed before RNA extraction. Total RNA was extracted using Trizol reagent (Invitrogen, Gaithersburg, MD) following the manufacturer’s instructions and then treated with DNase I (Fermentas, Hanover, MD). We checked the purity of the samples using a Nanodrop 2000 (Thermo Fisher Scientific, Waltham, MA), the concentration was assessed in a Qubit 2.0 fluorometer (Life Technologies, Carlsbad, CA), and RNA integrity was verified using an Agilent 2100 BioAnalyzer (Agilent Technologies, Santa Clara, CA). In this study, the RNA integrity number (RIN) ranged between 7.9 and 9.2, which is greater than 7 and could be acceptable for further analysis.

Library construction and sequencing were performed in a BGISEQ-500 system at the Beijing Genomic Institution (www.genomics.org.cn, BGI, Shenzhen, China). Clean reads were mapped to human reference gene and genome hg19 available in the NCBI database (https://www.ncbi.nlm.nih.gov/assembly) by using Bowtie2 and HISAT [49-50], respectively. The original sequence data have been submitted to the NCBI Sequence Read Archive (Accession number: PRJNA622427). For gene expression analysis, the matched reads were calculated and then normalized to RPKM values using RSEM software [51]. The differential expression analysis between treatments and controls was performed using Cuffdiff program, and the significant DEGs for further data analysis were filtered with false discovery rate (FDR) corrected P value ≤0.001 and fold change ≥1.5.

Bioinformatics analysis

DEGs were used as the input for a series of bioinformatics analyses performed with the WEB-based GEneSeTAnaLysis Toolkit (WebGestalt) [52-53]. WebGestalt is an open online analytical platform that integrates Gene Ontology (GO) [54], KEGG [55], and protein interaction networks for a variety of functional enrichment analyses (http://www.webgestalt.org). The GO, pathway annotation and enrichment analyses were based on the GO database (http://www.geneontology.org/) and the KEGG pathway database (http://www.genome.jp/kegg/).

Quantitative real-time PCR and data analysis

In the above mentioned selected DEGs, 10 newly discovered radiation-related genes were selected to be validated. Validations of the mRNA levels of DEGs were performed using quantitative real-time PCR (qRT-PCR). Peripheral blood samples (30 ml) were collected from another three healthy adult subjects (two males and one female: aged 28–30 years). The health status of the sample and sample irradiation were same as the above materials for RNA-seq. Then, blood samples were incubated at 37°C for 6 hours in RPMI-1640 medium of equal volume (Thermo Fisher Scientific, Inc., Waltham, MA, USA), supplemented with 10% FBS (HyClone; GE Healthcare Life Sciences, Logan, UT, USA). After RBC lysis, total RNA for qPCR was obtained using Trizol, treated with DNase (Fermentas, Hanover, MD), as described above (also see Supplementary Fig. S1). First-stranded cDNA was synthesized by RevertAid First Strand cDNA synthesis Kit (Thermo Scientific, Waltham, MA, USA). QRT-PCR was performed on an ABI7500 Real-Time PCR System using SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) and gene primer sequences are provided in Supplementary Table S3. The relative expression levels of each gene were normalized according to β-actin gene expression and analyzed using the 2-ΔΔCt method. Each group, composed of three parallel samples, was analyzed after irradiation. All experiments were performed three times.

Each data represents the means ± standard deviation (SD) of three independent biological experiments. Data management and statistical analysis was performed using MedCalc version 15.2.2 (MedCalc Software bvba, Ostend, Belgium). Comparisons were analyzed using Student’s t test with statistical significance at P < 0.05.

RESULTS

Identification of differentially express genes (DEGs)

High-throughput sequencing was used to identify mRNA expression in the radiation-treated and control groups. The clean reads obtained from the control, 50 and 150 mGy groups totaled 24.11, 23.87 and 24.11 Mb, respectively, and all Q30 values were higher than 90%. The comparison of differential gene expression between the radiation and control groups revealed 318 upregulated DEGs and 558 downregulated DEGs in the 50 mGy group, and the number of DEGs decreased to 202 upregulated and 284 downregulated genes in the 150 mGy group. In addition, 489 DEGs were found in the 150 vs 50 mGy groups, among which 263 DEGs were upregulated and 225 DEGs were downregulated. The complete DEG list can be found in Table 1. The Venn diagram analysis (Supplementary Fig. S2) showed that only 112 upregulated and 136 downregulated DEGs were altered by both gamma doses, whereas large numbers of DEGs were uniquely regulated by the different gamma doses, especially at the 50 mGy dose.

Table 1

Total number of DEGs

Comparison (mGy)	Up-regulated genes	Down-regulated genes
50 vs 0	318	558
150 vs 0	202	284
150 vs 50	263	225

GO and KEGG pathway enrichment analysis of DEGs

To understand the distribution of the DEGs at a macro level, the set of DEGs identified under different doses of radiation were mapped in accordance with GO terms and KEGG pathways. The goal of the GO consortium is to produce a dynamic, controlled vocabulary that can be applied to all eukaryotes even as knowledge of the roles of genes in cells accumulates, and our understanding of these roles changes [54]. The GO enrichment analysis revealed a wide range of cellular components, molecular functions and biological processes, including 54 important functional groups (Fig. 1). KEGG enrichment revealed that 758 DEGs and 1335 DEGs were enriched in 291 pathways in the 50 mGy and 246 pathways in the 150 mGy group, respectively, within six main categories. For both the 50 and 150 mGy groups, ‘human diseases,’ ‘organismal systems’ and ‘metabolism’ were the top three categories with the highest proportions of different DEGs. With a P-value <0.05 as the threshold, the top 20 significant enriched pathways of the DEGs were determined and are shown in Fig. 2. Among these pathways, ‘signal transduction,’ ‘immune system,’ ‘cancer: overview,’ ‘transport and catabolism’ and ‘signaling molecules and interaction’ showed significant enrichment. In addition, several pathways related to metabolism and human diseases were found, including ‘global and overview maps,’ ‘lipid metabolism’ and ‘amino acid metabolism’ as well as ‘infectious diseases.’ The results indicated that the greatest numbers of genes were active in environmental information processing, organismal systems, human diseases, cellular processes and metabolism.

Fig. 1

Gene ontology (GO) plot of genes with different expression patterns, including comparisons of the 50 vs 0 mGy, 150 vs 0 mGy and 150 vs 50 mGy groups.

Fig. 2

Pathway classifications of DEGs identified in the 50 vs 0 mGy, 150 vs 0 mGy and 150 vs 50 mGy comparisons.

Among the most enriched pathways, 11 pathways were identified under both the 50 and 150 mGy doses, and nine pathways were unique to the 50 and 150 mGy groups (Fig. 3). We found that the most downregulated genes at 150 mGy were related to the ‘rap1 signaling pathway,’ ‘calcium signaling pathway,’ ‘ras signaling pathway,’ ‘cAMP signaling pathway,’ and ‘cGMP-PKG signaling pathway,’ while at 50 mGy, the most downregulated genes were enriched in numerous diseases, including herpes simplex infection, Huntington’s disease, Alzheimer’s disease, Parkinson’s disease and non-alcoholic fatty liver disease and in cytokine-cytokine receptor interaction. At both radiation doses, ‘antigen processing and presentation’ was found to be the most enriched pathway among the upregulated genes.

Fig. 3

The most enriched pathways of DEGs, including upregulated and downregulated genes under 50 and 150 mGy.

Gene interaction network

The genes and the associated pathways were grouped into categories, which included ‘antigen processing and presentation,’ ‘immune system process,’ ‘immune system disease,’ ‘cytokine-mediated signaling pathway’ and ‘gas transport’ under both doses, the ‘response to stress’ at 50 mGy and the ‘calcium signaling pathway’ at 150 mGy. The interaction network was constructed based on co-expression data and experimental validation. Fig. 4 shows that the radiation-sensitive genes (HLA-DQA1, HLA-DQA2, HLA-DQB2, HLA-DRB1, and HLA-DRB5) located on chromosome 6 were not only mapped to ‘antigen processing and presentation,’ ‘immune system-related diseases’ and ‘cytokine-mediated signaling pathway,’ but may also directly interact with each other, mediating the effects of varying doses of radiation.

Fig. 4

Gene interaction networks specific to varying doses of 137Csγ radiation exposure. (A) Gene interactions associated with antigen processing and presentation; (B) gene interactions associated with immune system diseases; (C) gene interactions associated with immune system processes; (D) gene interactions associated with the cytokine-mediated signaling pathway; (E) gene interactions associated with gas transport; (F) gene interactions associated with the response to stress; (G) gene interactions associated with the calcium signaling pathway.

Key candidate DEGs and pathways involved in the radiation-responsive mechanisms

At both doses, differential expression analysis showed that there were more downregulated DEGs than upregulated DEGs (Table 1), and these downregulated DEGs were enriched in immune system- and human disease-related processes and signaling molecules and interaction pathways (Fig. 3). Interestingly, signaling molecules and interaction pathways such as the‘rap1 signaling pathway,’ ‘calcium signaling pathway,’ ‘ras signaling pathway,’ ‘cAMP signaling pathway’ and ‘cGMP-PKG signaling pathway,’ were specific to the 150 mGy group, which is an indication that molecular signaling may be activated when irradiation reaches a certain dose, which was not observed in the initial stage at 50 mGy. The upregulated DEGs were generally enriched in antigen processing and presentation at both doses (Fig. 3), demonstrating the importance of the immune system after LDR. The top 10 most enriched pathways of the DEGs included antigen processing and presentation, asthma, rheumatoid arthritis, type I diabetes mellitus, malaria, the cytokine-mediated signaling pathway and gas transport (see Supplementary Tables S1 and S2). At both doses, many DEGs such as HLA-DQA1, HLA-DQA2, HLA-DQB2, HLA-DRB1, HLA-DRB5, KIR2DL1, KIR2DL3, TNFSF10, TNFSF13B and TNFRSF4 were shared among the top 10 most enriched pathways (Supplementary Tables S1 and S2) and even directly interacted with each other (Fig. 4), which indicated that these genes could play important roles in the molecular mechanisms responding to LDR.

Verification of the key genes’ expression profiles by qRT-PCR

To verify the expression profiles of the key candidate DEGs (Supplementary Table S3) in the RNA sequencing experiment, we performed qRT-PCR assays using samples of human peripheral blood obtained from three healthy donors. These DEGs were selected for their potential role involving in the radiation-responsive mechanisms, because they were shared among the top 10 most enriched pathways (Supplementary Tables S1 and S2) and directly interacted with each other (Fig. 4). Samples were analyzed at 6 hours following ex vivo irradiation with 0, 50 and 150 mGy of γ-rays. As shown in Fig.5, the qRT-PCR results showed that the expression patterns of the selected genes were similar to those of RNA sequencing experiment (Fig. 5A and B), indicating their signal perception and transcription reactions after receiving radiation. Thus, the specific functions of these genes in human peripheral blood at different dose and different time should be further investigated in the future studies.

Fig. 5

The expression profiles of 10 genes in human peripheral blood by the quantitative RT-PCR. (A) The expression patterns of DEGs under irradiation and no radiation treatment by RNA-seq; (B) the expression patterns of DEGs following irradiation (50 mGy and 150 mGy) and no radiation by qPCR; Data are presented as means ± SD of three independent experiments. The asterisks indicate significant differences compared with the control treatment, *P < 0.05; **P < 0.01; ***P < 0.005, t-test.

DISCUSSION

According to ICRP Publication 90, the lowest experimentally observed doses causing persistent damage at the anatomical and structural level are in the range of 100–300 mGy of acute low-LET radiation [56]. Even evidence of malformations found at considerably lower doses, i.e. within a dose range of 50–250 mGy, cannot be ignored [56]. As mentioned in the ICRP103 report, the occupational exposure from existing and planned exposure sources should meet the requirements of individual dose limit [57]. However, the temporary constraint value of emergency exposure may reach 100 mGy and higher, such as nuclear accident or emergency disposal of nuclear facilities. Moreover, the irradiation of patients in medical irradiation should meet the requirements of medical irradiation guidance level for radiation diagnosis and nuclear medicine diagnosis, but there may be cases where the cumulative dose exceeds 150 mGy in a short time, such as multiple abdominal multi-slice spiral CT diagnosis [57]. The radiation in the range of 50–150 mGy is the focus of effect research, which is closely related to some scenes in emergency and medical irradiation [57]. Therefore, this study focused on the DEGs of transcriptome in peripheral blood of healthy people irradiated by 50 and 150 mGy dose, providing targets for the study of LDR effect mechanism and theoretical reference for the study of radiation protection.

The dose–response relationship is usually nonlinear and difficult to establish under LDR, potentially due to the dynamics between the compensatory mechanisms defending against radiation-induced damage in an organism [56]. In the present study, different dose-related gene expression patterns were observed at the transcriptional level. These changes suggest that radiation, even at low levels, could modulate cellular processes [37], and that gene expression, even at low levels, could be used as a sensitive endpoint for evaluating the effect of radiation under environmentally realistic low-dose exposure. In this study, we found a decreasing number of DEGs that were affected by radiation at doses from 50 mGy to 150 mGy (Table 1), showing that the effect of radiation on gene expression may be dose dependent at the transcriptome level. Similar observations of different dose–response patterns after LDR exposure have been reported at the transcriptional level [56-58]. The high radiation dose resulted in a lower number of DEGs, indicating a possible phase shift in the response to radiation doses above 150 mGy. The genes that were responsive to 50 mGy radiation were mostly associated with the immune system and diseases (e.g. Parkinson’s disease, rheumatoid arthritis, type I diabetes mellitus and asthma), whereas a radiation dose of 150 mGy induced gene expression changes related not only to the immune system and human diseases but also to environmental information processing (e.g. calcium signaling pathway) (Figs 3 and 4; Supplementary Tables S1 and S2). Significant changes in pathways observed at different doses indicated that they may be involved in the response to radiation.

GO terms represent gene product properties, and KEGG enrichment analysis can supply information for understanding gene functions and signaling pathways in metabolism. On the basis of GO analysis, we found that LDR significantly changed cellular components, biological processes, and molecular functions in human peripheral blood (Fig. 1). The relevant genes were included TNFSF10, TNFSF13B and TNFRSF4 (Supplementary Table S1), which belong to the tumor necrosis factor (TNF) superfamily (TNFSF), the TNF cell surface receptors, and the TNF receptor superfamily (TNFRSF), respectively. TNFSF–TNFRSF interactions activate signaling pathways for cell survival, death, and differentiation that control immune function and disease [59-66]. As members of the TNF superfamily, TNFSF10 induces apoptosis in transformed and tumor cells, and TNFSF10 could stimulate the activation of MAPK8/JNK, caspase 8 and caspase 3 by binding to its receptors (TNF-related apoptosis-inducing ligand, TRAIL) [67]. TRAIL has been reported to mediate the calcification of aortic valve interstitial cells through the apoptosis mechanism [68]. Previous reports have shown that TNFRSF4 can be upregulated by activated T cells [62], playing an important role in supporting/sustaining the immune response [69]. In this study, TNFRSF4 was downregulated after LDR by both RNA-seq and qRT-PCR experiments (Fig. 5; Supplementary Table S1), suggesting its impact on the immune process. Based on the pathway enrichment results, DEGs related to human disease were more enriched at 50 mGy (53 genes) than at 150 mGy (31 genes) (Fig. 3; Supplementary Table S2), and most of these genes have been reported to play important roles in human diseases. For instance, the HLA-DQA1 and HLA-DQB2 genes, which regulate type I diabetes and rheumatoid arthritis [70-74], were highly expressed under both the 50 and 150 mGy doses in our study (Fig. 5). Pathways involved in immunity were altered after radiation exposure at both the 50 and 150 mGy doses (Supplementary Table S1), which indicated that radiation exposure, even at low dose levels, may have harmful effects disrupting the immune system. Previous studies have shown that the mature lymphocytes and bone marrow stem cells of atomic bomb survivors were severely damaged, which significantly weakened their immune system, thus reducing their ability to resist microbial invasion [75-76]. Thus, many of these people died from infections. In our study, we found many more upregulated genes involved in antigen processing and presentation, as well as human diseases such as type I diabetes mellitus, malaria, rheumatoid arthritis and systemic lupus erythematosus (Fig. 3; Supplementary Table S2). Researchers also found fewer CD4 T-cells and increased levels of inflammatory proteins in A-bomb survivor blood samples [75-76]. LDR has been shown to impact type I diabetes in a mouse model receiving whole-body radiation [77]. Additionally, the Radiation Effects Research Foundation reported that mortality attributed to heart diseases increased with the radiation dose in a Life Span Study cohort [78]. Metabolic risk factors, including diabetes mellitus and high blood pressure, may increase the risk of coronary heart disease [79], and radiation exposure might enhance the risk of coronary heart disease via metabolic pathway disruption [37]. This is supported by the enriched KEGG pathways, which showed that type I diabetes mellitus was significantly enriched after 50 mGy radiation, and the number of enhanced genes related to type I diabetes mellitus was greater than the number of suppressed genes (Fig. 3, Supplementary Table S2).

The KEGG analysis indicated that one of the most important pathways responding to radiation exposure was antigen processing and presentation (Fig. 3; Supplementary Table S2), which has been reported to be related to immune response, transport pathogenesis, secretion, and phagocytosis. As shown in Fig. 4A, in the gene interaction network, the expression of several genes was associated with the antigen processing and presentation pathway, including the HLA-DQA1, HLA-DQA2, HLA-DQB2, HLA-DRB1, HLA-DRB5, KIR2DL1 and KIR2DL3 genes. As protein-coding genes, HLA-DQA1 and HLA-DQB2 belong to the HLA class II alpha chain paralogues and are anchored in the membrane. These genes are expressed in antigen presenting cells and play a central role in the immune system by presenting peptides derived from extracellular proteins [80-81]. Overexpression of HLA-DQA1 and HLA-DQB2 was demonstrated to be associated with diseases including type I diabetes and rheumatoid arthritis [70-74]. Our research also indicated that HLA-DQA1 and HLA-DQB2 were upregulated in type I diabetes and rheumatoid arthritis. The KIR2DL3 gene has been revealed to be associated with protection against rheumatoid arthritis (RA) [82-85]. Previous studies suggested that RA subtypes with different KIRs could result in clinical heterogeneity of the disease. For example, the presence of the KIR2DL3 gene in the absence of the KIR2DL2 gene is associated with seronegative RA, while the presence of the KIR2DL2 gene is associated with seropositive anti-CCP [83]. Interestingly, 17 relevant genes were found to be uniquely enriched in the calcium signaling pathway among the KEGG pathways at 150 mGy (Fig. 3; Supplementary Table S2), and this pathway has been reported to be important in radiation-induced bystander effects [86-89]. As shown by the gene interaction network (Fig. 4G), ATP2B2 (ATPase plasma membrane calcium transporter 2) exhibited co-expression with many other genes in this pathway, which was experimentally validated, indicating its importance in the calcium signaling pathway, especially in bystander effects. ATP2B2 encodes a human Ca2+-pumping ATPase [90] with critical roles in intracellular calcium homeostasis [91]. It has been reported to be expressed predominately in the brain and mammary gland, acting as an emerging player in autism and breast cancer [63, 92–96]. ATP2B2 exhibits the fastest activation and is one of the primary ATP pumps found in the brain and sensory epithelium. Investigations of this gene have highlighted its pivotal role in the normal function of the auditory system as well as the vestibular system. The differentially expressed HLA-DQA1, HLA-DQB2, KIR2DL3 and ATP2B2 genes, which were enriched in the immune system processes and human diseases, probably play important roles in the response to LDR.

In this study, we used RNA-Seq to screen DEGs and signaling pathways involved in LDR impacts and verify the expression patterns of some key candidate genes in the RNA-seq experiment by qRT-PCR. Although our study provides a more complete list of candidate genes or pathways that play important roles in response to LDR and indicates the coregulation of several new signaling pathways, the precise molecular mechanisms of these interacting pathways remain unclear. Furthermore, different tissues respond to radiation differently, the conclusions in this study are only referred from human peripheral blood responsible for LDR. Further efforts are needed to assess more novel DEGs and study how these signaling pathways guide humans to respond to LDR.

Click here for additional data file.

Click here for additional data file.

Click here for additional data file.

Click here for additional data file.

Click here for additional data file.