Abstract

Background

   Ionizing radiation (IR) can be extremely harmful for human cells since
   an improper DNA-damage response (DDR) to IR can contribute to
   carcinogenesis initiation. Perturbations in DDR pathway can originate
   from alteration in the functionality of the microRNA-mediated gene
   regulation, being microRNAs (miRNAs) small noncoding RNA that act as
   post-transcriptional regulators of gene expression. In this study we
   gained insight into the role of miRNAs in the regulation of DDR to IR
   under microgravity, a condition of weightlessness experienced by
   astronauts during space missions, which could have a synergistic action
   on cells, increasing the risk of radiation exposure.

Methodology/Principal Findings

   We analyzed miRNA expression profile of human peripheral blood
   lymphocytes (PBL) incubated for 4 and 24 h in normal gravity (1 g) and
   in modeled microgravity (MMG) during the repair time after irradiation
   with 0.2 and 2Gy of γ-rays. Our results show that MMG alters miRNA
   expression signature of irradiated PBL by decreasing the number of
   radio-responsive miRNAs. Moreover, let-7i*, miR-7, miR-7-1*, miR-27a,
   miR-144, miR-200a, miR-598, miR-650 are deregulated by the combined
   action of radiation and MMG. Integrated analyses of miRNA and mRNA
   expression profiles, carried out on PBL of the same donors, identified
   significant miRNA-mRNA anti-correlations of DDR pathway. Gene Ontology
   analysis reports that the biological category of “Response to DNA
   damage” is enriched when PBL are incubated in 1 g but not in MMG.
   Moreover, some anti-correlated genes of p53-pathway show a different
   expression level between 1 g and MMG. Functional validation assays
   using luciferase reporter constructs confirmed miRNA-mRNA interactions
   derived from target prediction analyses.

Conclusions/Significance

   On the whole, by integrating the transcriptome and microRNome, we
   provide evidence that modeled microgravity can affects the DNA-damage
   response to IR in human PBL.

Introduction

   Eukaryotic cells have evolved efficient DNA-damage response to
   genotoxic agents in order to eliminate any detrimental effect of DNA
   lesions. Ionizing radiation (IR) in out-of Earth represents an
   environmental mutagen to which humans are daily exposed on Earth.
   Crewmembers of space mission are even more exposed to IR because the
   cosmic radiation field is rather different from that experienced on
   Earth, and fragmentation with spacecraft shielding modifies the
   radiation quality spectrum, hence modifying the biological
   effectiveness of IR in a manner that is still undetermined. The cell
   response to the space environment, which is characterized by a
   condition of weightlessness (i.e. microgravity, 10^−4–10^−6 g),
   includes immune cell function suppression [35][1], [36][2], skeletal
   muscle atrophy [37][3]–[38][5], cardiovascular disorders [39][6], loss
   of bone [40][7], [41][8], changes of gene expression [42][9], [43][10],
   increase in chromosomal aberrations and apoptosis [44][11], [45][12].
   Studies carried out with systems simulating on Earth some aspects of
   microgravity, such clinostats and Rotating Wall Vessel bioreactors,
   reported similar results, indicating that the experiments performed
   with modeled microgravity can be used as surrogate of space conditions
   [46][13]–[47][20]. Despite the abundance of data about the biological
   effects of space and simulated microgravity, it is still unclear
   whether microgravity can affects the DNA-damage response (DDR) to IR.
   DDR is a complex pathway addressed to maintain genome integrity through
   the activation of proteins involved in sensing, signaling, and
   transducing the DNA damage signal to effector proteins of cell cycle
   progression/arrest, DNA repair and apoptosis [48][21]. While several
   studies reported additive/synergistic interactions of radiation and
   microgravity in different biological systems [49][22]–[50][26], other
   studies did not report such interactions [51][27]–[52][29]. In
   particular, the repair of radiation-induced DNA damage seems to be
   unaffected by microgravity in bacteria and human fibroblasts [53][30],
   [54][31] and in yeast [55][32]. On the contrary, in our previous work
   we detected a significant delay in the rejoining of DNA double-strand
   breaks induced by IR in human peripheral blood lymphocytes incubated in
   microgravity conditions [56][33]. Recent findings show the tendency of
   radioadaptation to DNA damage when space flown cells recovered on Earth
   are exposed to subsequent irradiation [57][34]. By considering the
   complexity of DDR and the controversial impact of reduced gravity on
   radio-sensitivity, we expected that the analysis of microRNA (miRNA)
   profiles could contribute to increase our knowledge on the features of
   space environment.

   MiRNAs are endogenous small noncoding RNAs (18–24 nt), acting as
   post-transcriptional modulators of gene expression, by pairing to
   target mRNAs and leading to decreased translational efficiency and/or
   decreased mRNA levels [58][35]. A single miRNA can influence the
   expression of up to thousand genes, thus the function of a miRNA is
   ultimately defined by the genes it targets. Besides a physiological
   role of miRNAs in a variety of important biological processes including
   differentiation, apoptosis [59][36], fat metabolism [60][37], viral
   infection [61][38] and pathological processes, such as tumorigenesis
   [62][39]–[63][43], the miRNA-mediated gene regulation operates also in
   response to cellular stress. Ionizing radiation induces changes in
   miRNA expression both in vitro and in vivo, according to cell type,
   radiation dose and post-irradiation time [64][44]–[65][49]. Several
   studies suggested that miRNA expression is regulated in the DDR at the
   transcriptional level, in a p53-dependent manner [66][50], and through
   modulation of miRNAs processing and maturation steps [67][51]. Whilst
   miRNA-mediated DDR has been studied after ionizing radiation, UV
   radiation and hypoxic stress [68][52], [69][53], the response to IR
   combined with microgravity has not been studied yet, and should give
   important information about the risk of the exposure to space
   environment.

   In our previous studies carried out with peripheral blood lymphocytes
   (PBL) incubated in modeled microgravity (MMG) during the repair time
   after IR, we reported significant decrease of cell survival, delay of
   double strand break repair, increase of mutant frequency and apoptosis
   [70][33], [71][54]. In the present study we analyzed miRNA expression
   profile of human PBL irradiated in vitro with 0.2 and 2Gy of γ-rays and
   incubated for a short (4 h) and medium-long (24 h) period in MMG and in
   parallel ground conditions (1 g). The results obtained from this study
   show differences in miRNA expression profile as a function of the dose
   and the time after irradiation in both gravity conditions.
   Interestingly, under MMG many miRNAs were not responsive to radiation
   compared with 1 g-condition. Analysis of mRNA expression profiles and
   further miRNA-mRNA anti-correlation analyses allowed the identification
   of DDR genes differently modulated in the two gravity conditions.

Methods

Ethics Statement

   Human peripheral blood lymphocytes (PBL) were obtained from freshly
   collected “buffy coats” of healthy donors at the Blood Centre of the
   City Hospital of Padova (Italy). This study obtained ethics approval
   from the Transfusion Medicine (TM) ethics committee of Blood Centre of
   the City Hospital of Padova. The informed consent from donors was not
   required by the TM/ethics committee because PBL samples were analyzed
   anonymously.

Cells, irradiation and microgravity simulation

   PBL were isolated by separation on Biocoll (Biochrom KG, Seromed)
   density gradient from freshly collected buffy coats from 12 healthy
   donors. After overnight incubation, PBL, consisting of peripheral
   mononuclear cells depleted of monocytes, were irradiated with γ-rays
   (0.2 and 2Gy) at the Department of Oncological and Surgical Sciences of
   Padova's University with a ^137Cs source (dose-rate: 2.8Gy/min). PBL
   from six donors (named D, E, F, I, L, M) were irradiated with 0.2Gy
   whereas PBL from other six donors (named A, B, C, G, H, P) were
   irradiated with 2Gy, for a total of 12 independent experiments. For
   each experiment, irradiated and non-irradiated PBL of the same donor
   were incubated in 1 g and MMG conditions for 4 and 24 h. MMG was
   simulated by the Rotating Wall Vessel (RWV) bioreactor (Synthecon,
   Cellon), as previously described [72][14], [73][20]. PBL incubated in 1
   g, irradiated and non, were kept in 75 cm^2 flasks at the same density
   ([74]Figure 1). In all experiments were used unstimulated quiescent
   (G[0]) PBL.

Figure 1. Experimental procedure of irradiation and microgravity simulation.

   [75]Figure 1
   [76]Open in a new tab

Total RNA isolation

   At the end of incubation time (4 and 24 h) in 1 g and MMG, total RNA
   was isolated from 10^7 irradiated and non-irradiated PBL, by using
   Trizol® Reagent (Invitrogen, CA), according to the manufacturer's
   protocol. Total RNA quantification was performed using the ND-1000
   spectrophotometer (Nanodrop, Wilmington, DE); RNA integrity and the
   content of miRNAs were assessed by capillary electrophoresis using the
   Agilent Bioanalyzer 2100, with the RNA 6000 Nano and the small RNA Nano
   chips, respectively (Agilent Technologies, Palo Alto, CA). Only total
   RNA samples with RNA Integrity Number (RIN) values ≥6 and with miRNA
   <20% were used for microarray analysis.

MiRNA and gene expression profiling

   MiRNA expression profiles were carried out in irradiated (0.2, 2Gy) vs.
   non-irradiated PBL, incubated for 4 and 24 h in 1 g and MMG. Analyses
   were performed by using the “Human miRNA Microarray kit (V2)” (Agilent
   Technologies), that allows the detection of 723 known human (miRBase
   v.10.1) and 76 human viral miRNAs. Total RNA (200 ng) was labeled with
   pCp Cy3, according to the Agilent protocol and unincorporated dyes were
   removed with MicroBioSpin6 columns (BioRad) [77][55]. Probes were
   hybridized at 55°C for 22 hours using the Agilent's Hybridization Oven
   that is suited for bubble-mixing and microarray hybridization
   processes. Then, the slides were washed by Agilent Gene expression wash
   buffer 1 and 2 and scanned using an Agilent microarray scanner (model
   G2565CA) at 100% and 5% sensitivity settings. Agilent Feature
   Extraction software version 10.5.1.1 was used for image analysis.

   Gene expression profiling was carried out in 2Gy-irradiated vs.
   non-irradiated PBL, incubated for 24 h in 1 g or MMG, previously
   analyzed for miRNA profiling. We used the “Whole Human Genome Oligo
   Microarray” (Agilent), consisting of ∼41.000 (60-mer) oligonucleotide
   probes, which span conserved exons across the transcripts of the
   targeted full-length genes. 800 ng of total RNA were labeled with
   “Agilent One-Color Microarray-Based Gene Expression protocol” according
   to the manufacturer's instructions. 1.65 µg of labeled cRNA were used
   to prepare the hybridization samples and the hybridization was carried
   out at 65°C for 17 hours in a hybridization oven rotator (Agilent). The
   arrays were washed by Agilent Gene expression wash buffers and
   Stabilization and Drying Solution as suggest by the supplier. Slides
   were scanned on an Agilent microarray scanner (model G2565CA) and
   Agilent Feature Extraction software version 10.5.1.1 was used for image
   analysis. Raw data are available on the Gene Expression Omnibus (GEO)
   website ([78]http://www.ncbi.nlm.nih.gov/geo/) using accession number
   [79]GSE20120 for miRNA expression profiling (72 experiments) and
   accession number [80]GSE20173 for mRNA expression profiling (20
   experiments).

Statistical analysis of miRNA and gene expression data

   Inter-array normalization of expression levels was performed with
   cyclic Lowess for miRNA experiments and with quantile for gene
   expression profiling [81][56] to correct possible experimental
   distortions. Normalization function was applied to expression data of
   all experiments and then values of spot replicates within arrays were
   averaged. Furthermore, Feature Extraction Software provides spot
   quality measures in order to evaluate the goodness and the reliability
   of hybridization. In particular flag “glsFound” (set to 1 if the spot
   has an intensity value significantly different from the local
   background, 0 otherwise) was used to filter out unreliable probes: flag
   equal to 0 will be noted as “not available (NA)”. So, in order to make
   more robust and unbiased statistical analysis, probes with a high
   proportion of “NA” values were removed from the dataset. We decided to
   use the 40% of NA as threshold in the filtering process obtaining a
   total of 270 available human miRNAs. Principal component analysis,
   cluster analysis and profile similarity searching were performed with
   tMev that is part of the TM4 Microarray Software Suite [82][57]. The
   identification of differentially expressed genes and miRNAs was
   performed with one and two class Significance Analysis of Microarray
   (SAM) program [83][58] with default settings. The expression level of
   each miRNA and mRNA was calculated as the log 2
   (irradiated/non-irradiated) PBL of the same donor.

Identification of miRNA target genes and anti-correlation analysis of miRNA
and mRNA expression data

   To predict miRNA targets we have performed a computational analyses
   using PITA algorithm based on thermodynamic stability of the RNA-RNA
   duplex, considering free energy minimization [84][59]. PITA algorithm
   was applied over up-to-date version 38 of RefSeq transcript sequences
   and it used miRNAs sequences downloaded from mirBase version 14. To
   identify the most likely targets, we have integrated mRNA and miRNA
   expression data, obtained on the same biological samples, using MAGIA
   web tool [85][60]. We used a non-parametric index (Spearman correlation
   coefficient), the most indicated statistical coefficient for a small
   number of measures, to estimate the degree of anti-correlation (e.g.
   up-regulated miRNA and corresponding down-regulated mRNA target)
   between any putative pairs of miRNA and mRNA [86][61], [87][62] and we
   have selected as functional only those anti-correlated less than
   −0.775. To identify biological processes most involved in the
   biological phenomena under study we have performed a Gene Ontology (GO)
   analysis, using DAVID tool [88][63], on significant anti-correlated
   target genes identifying biological pathways significantly enriched
   (P<0.05). Pathway enrichment analysis was carried out by KEGG web tool
   [89][63] whereas miRNA-mRNA anti-correlations were visualized by
   Cytoscape software package [90][64].

Validation of miRNA and mRNA expression levels with qRT-PCR

   The data of miRNA expression analysis were validated by using the
   TaqMan® MicroRNA Assay kit (Applied Biosystems, Foster City, CA), that
   incorporate a target-specific stem-loop reverse transcription primer to
   provide specificity for the mature miRNA target. In brief, each RT
   reaction (15 µl) contained 10 ng of total purified RNA, 5× stem-loop RT
   primer, 1× RT buffer, 0.25 mM each of dNTPs, 50 U MultiScribe™ reverse
   transcriptase and 3.8 U RNase inhibitor. The reactions were incubated
   in a Mastercycler EP gradient S (Eppendorf) in 0.2 ml PCR tubes for 30
   min at 16°C, 30 min at 42°C, followed by 5 min at 85°C, and then held
   at 4°C. The resulting cDNA was quantitatively amplified in 40 cycles on
   an ABI 7500 Real-Time PCR System, using TaqMan Universal PCR Master Mix
   and Taqman MicroRNA Assays for miR-34a, miR-424*, miR-181a-2*, miR-144,
   miR-598, miR-27a, and for U48 small nuclear (RNU48) as endogenous
   control.

   For mRNA detection, 1 µg of total RNA was retrotranscribed with
   ImProm-II Reverse Transcription System (Promega). qRT-PCR was performed
   with the GoTaq qPCR Master Mix (Promega) and gene-specific primers for
   ATM, BAX, FANCF, STAT5A, TNFRSF10B genes and for GADPH as reference.
   qRT-PCR reactions were always performed in quadruplicates, in PBL
   samples from 4–6 donors. The relative expression levels of miRNAs and
   mRNAs between samples were calculated using the comparative delta CT
   (threshold cycle number) method (2-^ΔΔCT) implemented in the 7500 Real
   Time PCR System software [91][65].

Luciferase reporter assays

   Luciferase reporter vectors containing the 3′-UTR of miR-27a, miR-144
   and miR-424* target genes ATM, FANCF, STAT5A, TNFRSF10B, BAX, were
   generated following PCR amplification from human cDNA and cloned into
   the pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega,
   Madison, WI), immediately downstream from the stop codon of the
   luciferase gene. The sequence of each insert was confirmed by
   sequencing. MiR-27a-sensor, miR-144-sensor and miR-424*-sensor, were
   obtained by annealing, purifying and cloning short oligonucleotides
   containing the perfect miRNA binding site into the SacI and XbaI sites
   of the pmirGLO vector. A549 cells were plated in 24-well plates
   (14×10^5 cells/well) and 24 h later co-transfected with 50 ng of the
   pmirGLO dual-luciferase constructs, containing the indicated 3′UTRs of
   target genes, and with 32 nM pre-miR™ miRNA Precursor
   Molecules-Negative Control or pre-miR™ miRNA Precursor hsa-miR-27a
   (PM10939), hsa-miR-424*(PM12641), and hsa-miR-144 (PM11051) (all from
   Ambion, Austin, TX), using Lipofectamine2000 (Invitrogen Life
   Technologies). Lysates were collected 24 h after transfection and
   Firefly and Renilla Luciferase activities were consecutively measured
   by using Dual-Luciferase Reporter Assay (Promega) according to
   manufacturer's instructions. Relative luciferase activity was
   calculated by normalizing the ratio of Firefly/Renilla luciferase to
   that of negative control-transfected cells. Transfections were
   performed in triplicate and repeated 3–4 times.

Results

Radio-responsive miRNAs in static condition (1 g)

   Human PBL were irradiated in vitro with γ-rays (0.2Gy; 2Gy) and
   incubated in normal gravity (1 g) and in MMG during the
   post-irradiation time. Twelve total donors were analyzed, six for each
   dose, by performing independent experiments, in which irradiated and
   non-irradiated PBL of the same donor were incubated in 1 g and in MMG
   for 4 and 24 h. MicroRNA expression profiling was performed on total
   RNA extracted at the end of incubation times ([92]Figure 1), by
   comparing the expression profile of irradiated vs. non-irradiated PBL
   of the same donor. Data obtained from PBL incubated in 1 g allowed to
   identify 26 (0.2Gy) and 20 (2Gy) radio-responsive miRNAs at 4 h after
   irradiation; miRNAs differentially expressed at 24 h after irradiation
   were 17 (0.2Gy) and 52 (2Gy), ([93]Figure 2A and [94]Table 1). Raw data
   and means of miRNA expression values are available on [95]Table S1.
   Besides a fraction of radio-responsive miRNAs common between 0.2 and
   2Gy (28% and 19% at 4 and 24 h after IR, respectively), most of miRNA
   species was activated in a dose-related manner ([96]Figure 2B). Our
   results showed that, early after irradiation, both doses induced
   consistent changes in miRNA expression, whereas late after irradiation,
   the effect of the higher dose was predominant. Furthermore, most of
   radio-responsive miRNAs showed a time-related expression pattern, with
   a substantial down-regulation at 4 h and up-regulation at 24 h after IR
   ([97]Figure 2C).

Figure 2. Differentially expressed miRNAs in irradiated PBL incubated in 1 g
condition.

   [98]Figure 2
   [99]Open in a new tab

   (A) Number of radio-responsive miRNAs at 4 and 24 h after irradiation
   with 0.2 and 2Gy of γ-rays. (B) Percentage of dose-responsive miRNAs at
   the same time points. The expression level of each radio-responsive
   miRNA is the mean of expression values from six different donors/dose
   determined by the log2 (irradiated/non-irradiated) PBL. (C) Dendrogram
   showing radio-responsive miRNAs common to 0.2 and 2Gy of γ-rays, whose
   expression changed between 4 and 24 h after irradiation. Six different
   donors were analyzed for each dose, as indicated by the capital
   letters. The range of expression value is from -3.0 (green,
   down-regulation) to 3.0 (red, up-regulation). Grey boxes correspond to
   not available (N/A) fluorescent signal from the microarray platform.

Table 1. Differentially expressed miRNAs in γ-irradiated versus
non-irradiated human PBL.

   4 h after irradiation 24 h after irradiation
   miRNA name 0.2Gy 2Gy miRNA name 0.2Gy 2Gy
   hsa-miR-21* −0.84 −0.72 hsa-miR-1225-5p 0.44 0.60
   hsa-miR-34b* −0.89 −0.85 hsa-miR-135a* 0.48 0.67
   hsa-miR-210 0.53 0.46 hsa-miR-152 −0.43 −0.41
   hsa-miR-630 −1.75 −1.41 hsa-miR-181a-2* −0.57 −0.99
   hsa-miR-886-3p −0.59 −0.91 hsa-miR-188-5p [100][46] 0.60 0.73
   hsa-miR-199b-5p 0.58 0.39 hsa-miR-34a [101][44], [102][48], [103][52],
   [104][70], [105][71] 1.53 1.76
   hsa-miR-582-5p 0.48 0.39 hsa-miR-34b* [106][68] 1.02 1.66
   hsa-miR-378 −0.71 −0.46 hsa-miR-424* 1.00 2.08
   hsa-miR-513b −0.96 −0.65 hsa-miR-638 [107][49], [108][70] 0.57 0.59
   hsa-miR-923 −0.99 −0.53 hsa-miR-663 [109][46], [110][49], [111][68],
   [112][70] 0.46 0.99
   hsa-let-7e [113][45], [114][47], [115][49] −0.31 hsa-miR-765 0.63 0.71
   hsa-miR-16 [116][52] 0.30 hsa-miR-1226* 0.76
   hsa-miR-23a* 0.17 hsa-miR-150* 0.38
   hsa-miR-34a −0.64 hsa-miR-202 0.43
   hsa-miR-145 0.75 hsa-miR-601 0.78
   hsa-miR-181b 0.30 hsa-miR-760 0.42
   hsa-miR-196b 0.55 hsa-miR-886-3p −0.60
   hsa-miR-202 0.49 hsa-miR-100 [117][72] −0.50
   hsa-miR-221 [118][52], [119][69] 0.51 hsa-miR-101* −0.37
   hsa-miR-339-3p −0.34 hsa-miR-10a −0.57
   hsa-miR-345 −0.62 hsa-miR-141 −0.38
   hsa-miR-425* 0.43 hsa-miR-151-3p −0.32
   hsa-miR-450a 0.52 hsa-miR-16-2* −0.66
   hsa-miR-494 −0.76 hsa-miR-17 [120][69], [121][73] −0.22
   hsa-miR-629* 0.71 hsa-miR-181a [122][44], [123][69] −0.31
   hsa-miR-801 −0.85 hsa-miR-18b [124][52] −0.35
   hsa-miR-223 0.47 hsa-miR-196a −0.91
   hsa-miR-301a 0.45 hsa-miR-196b [125][70] −0.51
   hsa-miR-513a-5p −1.19 hsa-miR-19b −0.25
   hsa-miR-940 0.53 hsa-miR-200b [126][70] −0.24
   hsa-miR-768-5p −0.38 hsa-miR-210 [127][44], [128][53] −0.31
   hsa-miR-146a −0.41 hsa-miR-221* −0.36
   hsa-miR-575 −0.47 hsa-miR-29b-1* −0.47
   hsa-miR-378* −0.47 hsa-miR-30d [129][69] −0.18
   hsa-miR-188-5p −0.59 hsa-miR-30e* −0.24
   hsa-miR-126* [130][69] −0.60 hsa-miR-330-3p −0.34
   hsa-miR-335 −0.35
   hsa-miR-345 [131][44], [132][53], [133][71] 1.16
   hsa-miR-363 0.99
   hsa-miR-371-5p 0.55
   hsa-miR-421 0.50
   hsa-miR-483-5p 1.33
   hsa-miR-494 0.87
   hsa-miR-505* −0.40
   hsa-miR-513a-5p 1.06
   hsa-miR-513b 1.19
   hsa-miR-513c 1.22
   hsa-miR-551b −0.40
   hsa-miR-574-5p 0.97
   hsa-miR-630 [134][68], [135][73] 0.96
   hsa-miR-769-5p −0.34
   hsa-miR-801 0.66
   hsa-miR-873 −0.64
   hsa-miR-877* 0.72
   hsa-miR-923 0.89
   hsa-miR-940 0.49
   hsa-miR-95 −0.44
   hsa-miR-99a −0.64
   [136]Open in a new tab

   Irradiated and non-irradiated PBL of the same donors were incubated in
   static gravity (1 g) for 4 and 24 h, and miRNA expression profile was
   analyzed at the end of each incubation time. The expression value of
   each radio-responsive miRNA is the mean of expression levels calculated
   as the log2 (irradiated/non-irradiated) PBL from six donors/dose (see
   [137]Table S1). References of miRNAs differentially expressed in other