Microarray
A number of 1×106KG1a cells (human) were irradiated at 8Gy at a radiation dose of 1.557Gy/min with 1hr LiCl pre-treatment. After 48hr, cells were resuspended in 500μl of RNALater(#7020, Ambion, ThermoFisher Scientific, USA) and incubated on ice for 3hr. The microarray was performed at Genotypic Technologies (Agilent) Pvt. Ltd, Bangalore, India. Briefly, RNA was isolated using RNeasy Mini Kit (Qiagen) and quantified in Nanodrop Spectrophotometer. Following reverse transcription of RNA into cDNA, Cy3-labeled cRNA was produced by in vitro transcription and hybridized to human whole genome array chip (Human Whole Gene Expression Microarray, 4644 K array, Agilent Technologies). The fluorescent images were scanned and acquired using feature extraction software (v10.5, Agilent technologies). The microarray data was deposited in GEO database (accession no.: GSE73486 http://www.ncbi.nlm.nih.gov/geo/). Similarly, the microarray data of K562 irradiated at 8Gy was obtained from the GEO database (accession no.: GSE7505 https://dtp.cancer.gov/discovery_development/nci-60/).
Microarray expression data analysis
The probes which were detected in all the replicates were considered for further data analysis. In GeneSpring GX11, the data were normalizedby percentile shift normalization method. After normalization, the data were log2 transformed. The average log2 ratio with p-value≤0.05was used to identify differentially regulated genes. To identify biological processes and molecular functions associated with differentially regulated genes, functional annotation tool in DAVID Bioinformatics Resources 6.7 was used (https://david.ncifcrf.gov/) considering p-value≤0.05 as significant. Further, the signaling pathways controlled by significantly altered genes were identified in Pathway Miner (http://www.biorag.org/pathway.html), which provided annotation from KEGG, Biocarta and GenMAPP, taking Fisher exact pvalue≤0.05 as significant.
Principal component analysis (PCA)
To identify the common clustered and correlated genes, the samples were clustered by implementing the PCA. One advantage of PCA is that the principal components are always orthogonal (uncorrelated), whereas other methods often produce redundant correlated clusters. The parameters set for PCA clustering were, Z value=8.0, error model=4, error variance=500, Bayesian degree of freedom=10, FDR threshold=0.05. The data were uploaded as log2 transformed values.
Network generation
The GeneMANIACytoscape plugin was used for predicting gene functions(11). The genes involved in cell adhesion classification were input in GeneMANIA to find out interactions among them. The resulting predicted network of functional relationships among query and predicted genes was then used as an annotated Cytoscape network for analysis. Further,ClueGO app(http://www.ici.upmc.fr/cluego/cluegoDownload.shtml) was implemented to improve biological interpretation and functionally organize GO and pathway term network.
Cell lines, antibodies and radiation
All cell lines used in this study were cultured in cell culture medium containing foetal bovine serum (FBS #10270, GIBCO, Invitrogen, USA) and antibiotics solution (#A002A-50ML, HIMEDIA, India) at 37°C (in a 5% CO2 atmosphere). K562 cells (human, NCCS Pune) were cultured in 10 percent FBS containing IMDM (#I7633, Sigma, USA) while KG1a cells were grown in 20 percent FBS containing IMDM (#I7633, Sigma, USA). Bhabhatron II teletherapy unit with Cobalt 60 radiation source (Panacea, India) at 80 cm signal to source distance and 35 x 35 cm2 field size was used for irradiation. LiCl (#A6286, 0100, Applichem, GmbH, Germany) was used to pre-treat cells at concentrations of 2-32mM. Throughout the experiment, it was present. Fibronectin (#F2006, Sigma, USA) in concentrations ranging from 1 to 10 g was coated on cultureware. CD54 (#555511), CD49D (#555502), CD49E (#555615), CD29 (#556048), CD18 (#556084), and CD41B (#555469) antibodies were purchased from BD USA and used as directed by the manufacturer. Unless otherwise stated, all other chemicals were purchased from Sigma in the United States.
Cell viability and adhesion
Cells were pretreated with 2mMLiCl for 1hr before radiation. After 48hr, the adhered cells and those in the supernatant were collected separately in 100μl of PBS (#21600-044, Gibco, USA) in a 96 well plate. This was followed by addition of 10µl of Prestoblue reagent (#A13261, Invitrogen, USA). This reagent is quickly reduced by metabolically active cells, providing a quantitative measure of viability and cytotoxicity. After 1hr of incubation, fluorescence was recorded at a wavelength of 590nm (560nm excitation).
To check interaction of cells with fibronectin, 24 wells plates were coated with different concentration of fibronectin. In a few wells, about 1µl of 1:100 diluted anti-CD49e antibodywas added and incubated for 30min at 37°C. Rinsed wells with PBS and plated 6 x 104 cells/well of the coated plates. Plates were incubated for 90min in a CO2 incubator. After incubation, rinsed wells with PBS four times gently. Added 100µl of PBS in each well and mixed 20µl of MTS reagent (#G5421, Promega, USA) in each well. Incubated at 37°C for 1-4hr in CO2 incubator. Recorded absorbance at 490nm. In a separate experiment, K562 adhesion to culture plate was estimated similarly upon blocking of surface receptors for CAMs belonging to Ig superfamily (CD54), and integrins (CD49D, CD49E, C11a, CD29, CD18 and CD41B). The adhered cells were collected and assayed for reduction of PrestoBlue reagent.
To observe the effect of combined treatment of CD18 antibody, LiCl and Imatinib on K562 cells, viability assay was performed with MTT reagent. A total of 1 x 105 K562 cells were pelleted in U-shaped wells of a 96 well plate in DMEM containing 10% FBS. Added CD18 antibody/ LiCl (2mM) / their combinations to the individual wells containing K562 cells. The Imatinib drug was added to each well at 1μM or 10μM concentration to assess the combined effect of treatments on K562 cell’s growth. The culture plate was centrifuged at 100rpm for 10min. Incubated cells for 3 days to induce spheroid formation. After 3 days, 10μl of MTT reagent was added in each well. Incubated the plate for 4hr. Added 50μl of MTT solvent (DMSO and Isopropanol in 1:1 ratio) to each well to dissolve the formazen crystals. Measured absorbance at 595nm.
Flow cytometry
Exponentially growing 1x106cells with or without radiation and LiCl pre-treatment were collected after 48hr of incubation. Cells were washed with PBS and fixed in 4% PFA for 20min at 4°C. Labeled cells with CD49e-PE (#555617, BD, USA), CD18 and CD54 antibody for 1hr at 4°C. The cells were then stained with 1:100 diluted FITC labeled secondary antibody (#sc-2080, Santa Cruz, USA).Again, washed the cells with PBS and analyzed in flow cytometer (BD FACS Caliber).
Immunoprecipitation of Serine 9 phosphorylated GSK3β
Briefly, after LiCl and radiation treatment as described earlier, K562 cells were washed with PBS. All subsequent steps were performed on ice. Added 1ml of RIPA lysis buffer/ 40mg (5 x 106 cells) of wet cell pellet. Further, added 10µl of protease inhibitor cocktail (#5871, Cell Signaling Technology). Incubated the mixture on ice for 15min. The lysate was centrifuged at 14000xg for 15min. The lysate was transferred to an eppendorf and incubated with 10µl (2µg) of GSK3β antibody (Ser-9)(A component ofPhospho-Akt pathway antibody sampler kit, #9916, Cell Signaling Technology, USA) for 1hr. Added 20µl of resuspended volume of Protein A/G Plus-agarose beads (#sc-2003, Santa Cruz Biotechnology Inc.). Incubated at 4˚C on a rocker platform for 1hr. Collected the immunoprecipitate by centrifugation at 2500rpm for 5min at 4˚C. Washed pellet four times with 1ml of PBS/RIPA buffer. Resuspended the pellet in 40µl of 1X electrophoresis sample buffer. Boiled sample for 2-3min and analyzed 20µl aliquot in SDS-PAGE.
siRNA transfection
A total of 1 x 106 K562cells/ml were suspended in OptiMEM (#31985, Gibco, Thermo Fisher Scientific, USA). The integrin α5 siRNA (#sc-29372, Santa Cruz, USA) was transfected into K562 cells at a concentration of 40, 80, 120, 160 and 200nM with 10μl lipofectamine 2000CD reagent (#52888, Invitrogen, USA) according to manufacturer’s recommendations. After 72hr of incubation, cells were treated with 2mM LiCl for 1-2hr and plated on fibronectin coated wells (10µg/ml). The adhesion of cells to fibronectin coated plates was measured with MTT reagent (#M5655, Sigma, USA).
Immunostaining
The LiCl and IR treated K562 cells were incubated for 48hr. Then cells were fixed in0.5% paraformaldehyde for 20min at 4°C. For labeling, CD18 antibody (1:1000 diluted) and PE labeled secondary antibody (1:1000 diluted) was incubated with cells. Subsequently added Prolong gold antifade reagent with DAPI (#8961 Cell Signaling Technology, USA). This was followed by mounting of cells on a slide with 20µl of 90% glycerol. Visualized fluorescence of cells under microscope (Olympus, India).
Transwell migration assay
K562 cells were starved for 24-48hr prior to assay by growing in serum free medium. A total of 1.6 x 105 cells were dispensed in 12 well (0.8-3μm) transwell insert in 1ml of serum free IMDM. In another experiment, cells were irradiated at 10 and 15Gy or blocked with CD18 antibody for 1hr before dispensing into transwell insert. Filled wells of 12well plate with 1ml of 10% IMDM medium and placed transwell insert in the well. Covered the plate and incubated for 48-72hr at 5% CO2. After incubation, placed the insert into a clean well containing 225µl of pre-warmed cell detachment solution (#ECM505, QCM Chemotaxis cell migration assay kit) for 30min at 37°C. Dislodged cells completely from the underside by gently tilting the insert back and forth several times during incubation. The insert was discarded thereafter. Also collected the cells migrated to the lower chamber. Mixed cells dislodged from insert and from lower chamber. Cells were resuspended in 100µl of PBS and 100µl of MTT was added. After 30min of incubation in dark at room temperature (RT), added MTT solvent and again incubated for 15-30min. Read absorbance at 560nm in a spectrophotometer.
Chromatin immunoprecipitation
The LiCl and IR treated cells were harvested after 48hr of incubation and their chromatin was immunoprecipitated with 5µl of NFκB antibody(#N8523-2ml Sigma, USA) as per manufacturer’s instructions using SimpleChIP® Enzymatic Chromatin IP Kit (#9002, Cell Signaling Technology, USA). This was followed by qPCR for IL7, IL8 and MMP1 genes using SYBR green (#K0222, ThermoFisher Scientific, USA).
Briefly, to crosslink proteins to DNA in the growing cells, 540µl of 37% formaldehyde was added to each sample.Mixed and incubated for 10min atRT. A total of 2ml of 10X glycine solution was added to each dish, mixed briefly, and incubated for 5min at RT. Then cells were centrifuged at 500g for 5min at 4℃ and washed twice with ice-cold PBS. Then cells were resuspended in 1ml ice-cold 1X Buffer A with dithiothreitol (DTT) and Protease Inhibitor Cocktail (PIC) per immunoprecipitation preparation (IP prep). This was followed by incubation on ice for 10min with intermittent mixing. The nuclei were pelleted by centrifugation at 2000g for 5min at 4℃. The nuclei pellet was resuspended in 1ml of ice-cold 1X Buffer B with DTT per IP prep. The step was repeated, and the pellet was resuspended in 100µl of 1X Buffer B with DTT per IP prep. Then 0.5µl of Micrococcal Nuclease was added per IP prep, mixed by inverting the sample several times and incubated for 20min at 37℃ with frequent mixing to digest DNA to a length of approximately 150-900bp. The reaction was stopped by adding 10µlof 0.5M EDTA per IP prep and placing tube on ice for 1-2min. The nuclei were pelleted by centrifugation at 16,000g for 1min at 4℃. Subsequently, the nuclear pellet was resuspended in 100µlof 1X ChIPBuffer with PIC per IP prep and incubated on ice for 10min. An aliquot of 500µlof sample was sonicated with several pulses to break the nuclear membrane. Incubated samples for 30s on ice between sonication pulses. Further, the lysate was clarified by centrifugation at 9400g for 10min at 4℃.
For CHIP, mixed 1X ChIP Buffer and 100μl (5-10µg of chromatin) of the digested, cross-linked chromatin preparation. An aliquot of the diluted chromatin (10µl) was transferred to a microfuge tube. This was designated as 2% Input Sample. For each immunoprecipitation, transferred 500µlof the diluted chromatin to a 1.5mlmicrocentrifuge tube and added immunoprecipitatingNFκB antibody. The IP samples were incubated for 4hr at 4℃ with rotation. Immediately, 30µl of ChIP-Grade Protein G agarose beads were added to each IP reaction and again incubated for 2hr at 4℃ with rotation. The Protein G agarose beads were pelleted by centrifugation at 3400g for 1min. The pelleted beads were then washed with low salt wash buffer, followed by washing with high salt wash buffer. For elution of chromatin from antibody/Protein G agarose beads and breaking of cross-links, 150µlof the 1X ChIP Elution Buffer was added to the 2% Input sample tube and incubated at RT. An equal amount of 1X ChIP elution buffer (150µl) was added to each IP sample. The chromatin was eluted from the antibody/Protein G agarose beads complex by gentle vortexing for 30min at 65℃. Pelleted beads in each IP by centrifugation at 3400g for 1min. Finally, eluted chromatin was transferred to a new tube. The cross-links were reversed by adding 6µl of 5M NaCl and 2µlProteinase K and incubating for 2hr at 65℃.
This was followed by the isolation of DNA. For this, 750µlof DNA binding buffer was added to each DNA sample and vortexed briefly. Then each sample was transferred to a DNA spin column placed in the collection tube. The spin column was centrifuged at 18,500g for 30s. Added 750µlof wash buffer to the spin column and centrifuged at 18,500g for 30s. This was followed by the addition of 50µl of elution buffer to each spin column placed into a 1.5mleppendorf. For DNA elution, the column was centrifuged at 18,500g for 30s. Subsequently, qPCR was performed for GAPDH, IL7, IL8 and MMP1 genes using SYBR green dye (#K0222, ThermoFisher Scientific, Waltham, MA) (see Supplementary Table 1 for PCR primers sequence).
RT-PCR
Approximately 1x107cells were suspended in 600µl of lysis buffer supplemented with 2mM DTT provided with Gene JET RNA purification kit (Thermo Scientific, USA). This was followed by addition of 360µl of ethanol (96-100%). The cell lysate was transferred to RNA purification column and washed with wash buffer 1 and 2 provided with the kit. RNA was eluted in 50-100µl of RNase free water. cDNA was synthesized from the eluted RNA using cDNA synthesis kit (Fermentas, USA). Mixed the following reagents, template DNA (0.1ng-5µg), random hexamer primer, 5X reaction buffer, ribolock RNase inhibitor, 10mM dNTP and Revert Aid M-MuLV Reverse Transcriptase followed by water. The reaction was performed in a thermal cycler for 5min at 25°C, 60min at 42°C and 5min at 70°C. An aliquot of 2µl of reverse transcribed product was used to set up qPCR using SYBR green (#K0222, ThermoFisher Scientific, USA) for detection of amplified gene in RT-PCR machine (#CFX96, Bio-Rad, USA).
Statistical analysis
The data were presented as mean±standard deviation. Student’s t-test was used to compare data between groups. Differences between the groups at 5% were considered significant (p≤0.05). For the calculation of log fold change in RT-PCR, the ΔCT and ΔΔCT values were calculated(12).
Where ΔCT=CT(geneof interest)-CT(housekeepinggene),
ΔΔCT= ΔCT(treated sample)-ΔCT(untreatedsample), and
CT(cycle threshold of the sample)= Indicates cycle number, where the fluorescence generated by the PCR product is discernible from the background noise.