2.1. Cell lines
Drug-sensitive CCRF-CEM and multidrug-resistant P-glycoprotein-overexpressing CEM/ADR5000 leukemic cells were kindly given by Prof. Axel Sauerbrey (Department of Pediatrics, University of Jena, Germany). Cells were cultured in RPMI1640 medium supplemented with 10 % fetal bovine serum (FBS) and 1 % penicillin (1,000 U/mL)/streptomycin (100 µg/mL) (P/S) (Life Technologies, Darmstadt, Germany). Doxorubicin (5,000 ng/mL) was supplied to retain overexpression of P-gp (MDR1, ABCB1) in resistant CEM/ADR5000 leukemic cells [26]. Human brain glioblastoma cell line U87.MG and the U87.MGΔEGFR that is transfected with a plasmid carrying an EGFR gene with a deletion of exons 2–7 were obtained from Dr. W. K. Cavenee (Ludwig Institute for Cancer Research, San Diego, CA, United States). The cell lines were cultured with 800 ng/mL geneticin [22, 23].
The panel of 60 human tumor cell lines of the Development of Therapeutics Program of the National Cancer Institute (NCI, USA) included leukemia, melanoma, non-small cell lung cancer, colon cancer, renal cancer, ovarian cancer, breast cancer and prostate carcinoma cells and tumor cells of the central nervous system [24].
2.2. Cell viability assay
The cytotoxic activities of triptolide (Sigma Aldrich, Taufkirchen, Germany; Fig 1A) were evaluated by the resazurin assay [25]. This assay is based on reduction of the indicator dye, resazurin, to the highly fluorescent resorufin by viable cells. Aliquots of 5,000 cells/100 µL of U87.MG and U87.MGΔEGFR were placed in 96-well plates and incubated for one day before treatment. However, for leukemic cells, 10,000 cells/100 µL cells were seeded into 96-well plates and immediately treated. Twenty microliters of resazurin 0.01% w/v solution were added to each well after 72 h at 37 °C incubation, and the plates were incubated at 37°C for 4 h. Fluorescence was detected by an Infinite M2000 Proplate reader (Tecan, Crailsheim, Germany) with an excitation wavelength of 544 nm and an emission wavelength of 590 nm. Each experiment was carried out at least three times with six replicates each. The viability was analyzed based on a comparison with untreated cells. Fifty percent inhibition (IC50) values imply the drug concentrations needed to inhibit 50% of cell proliferation and were calculated from a calibration curve by linear regression using Microsoft Excel [26, 27].
2.3. Compare and cluster analyses of microarray-based mRNA transcriptome-wide expression data
The mRNA microarray hybridization of the NCI cell lines has been published at the NCI Web site (http://dtp.nci.nih.gov) [28, 29]. COMPARE analyses were used to obtain rank-ordered lists of genes expressed in the NCI cell lines. The detailed method as a tool to determine candidate genes for drug resistance and sensitivity has been previously published [30-33]. In order to identify COMPARE rankings, a scale index of correlation coefficients (R-values) was generated from Iog10IC50 values of test compounds and microarray-based mRNA expression values. Greater mRNA expression correlated with enhanced drug resistance in the standard COMPARE, whereas greater mRNA expression in cell lines indicated drug sensitivity in reverse COMPARE analyses. Pearson´s correlation test was used to calculate significance values and rank correlation coefficients as relative measure for the linear dependency of two variables.
For hierarchical cluster analyses, objects were classified by calculation of distances according to the closeness of between individual distances. All objects were displayed into cluster trees (dendrograms). Merging of objects with similar features leads to cluster formation, where the length of the branch implies the degree of relation. Distances of subordinate cluster branches to superior cluster branches serve as criteria for the closeness of clusters. Therefore, objects with tightly related features were clustered closely together, if separation of objects in the dendrogram increased with progressive dissimilarity. Hierarchical clustering and heat map analyses were carried out using clustered image map (CIM) miner software by the one matrix CIM (https://discover.nci.nih.gov/cimminer/oneMatrix.d) [34].
2.4. Transcription factor gene promoter binding motif analysis
The top 40 genes, which directly or inversely correlated with log10IC50 values of the NCI cell lines in COMPARE analysis, were submitted to binding motif analysis. Promoter sequences 25 kb upstream of exon 1 of the corresponding genes were retrieved from UCSC Genome Browser Gene Sorter (http://genome.ucsc.edu). Promoter sequences were checked using the SeqPos tool implemented in the Galaxy Cistrome software [35].
2.5. Molecular docking
The interaction energy of triptolide with NF-ĸB pathway proteins was predicted using molecular docking: I-ĸB kinase β, I-ĸB kinase β-NEMO (NF-ĸB essential modulator) complex, NF-ĸB, and NF-ĸB-DNA complex. The protocol for molecular docking was reported by us [36]. Protein structures using X-ray crystallography were obtained from PDB database (http://www.rcsb.org/). I-ĸB kinase β (PDB ID:3RZF), I-ĸB kinase β-NEMO complex (PDB ID:3BRT), NF-ĸB (p52/RelB heterodimer, PDB ID:3DO7), and NF-ĸB-DNA complex (p50/p65 heterodimer bound to DNA, PDB ID: IVKX) were used in our study.
A grid box was defined for docking spaces in each protein according to its pharmacophores. Docking parameters were set to 250 runs and 2,500,000 energy evaluations for each docking. Dockings were performed three times independently. Lamarckian Genetic Algorithm was chosen for docking calculations. For the visualization of docking results, AutodockTools-1.5.7rcl was used. The surface representation image showing the binding pocket of proteins was made with Visual Molecular Dynamics (VMD) software developed with NIH support by the Theoretical and Computational Biophysics group at the Beckman Institute, University of Illinois at Urbana-Champaign (http://www.ks.uiuc.edu/Research/vmd/).
2.6. Ingenuity Pathway Analysis
Deregulated genes identified by COMPARE analysis as factors determining cellular responsiveness to triptolide were subjected to pathway analysis, in order to determine the biological function of these genes. Forty genes were imported into the Ingenuity Pathway Analysis (IPA) software (Qiagen Bioinformatics, Redwood City, CA, U.S.A) in Excel format [37].
Core analyses were carried out with general settings: “human” as species and “cell line” as type of biological sample. Canonical pathways, diseases and functions, and upstream regulators were determined by Fisher’s exact test at a significance value of p < 0.05. IPA core analyses identifies key regulators and networks in human cell lines [37, 38].
2.7. NF-ĸB reporter cell assay
HEK293 cells stably expressing HEK-Blue-Null1 vector and SEAP on a NF-ĸB promoter were obtained from Invivogen (San Diego, CA, USA). The cells were maintained according to manufacturer´s protocol and treated with various concentrations of triptolide (0 µM, 1.6 µM, 3.1 µM, 6.3 µM, 12.5 µM, 25 µM and 50 µM) for 1 h and triptolide was not removed. Afterwards, NF-ĸB activity was induced with 100 ng/mL of TNF-α for 24 h. The activation was evaluated by detecting SEAP spectrophotometrically at 630 nm with addition of Quanti Blue (Invivogen). The procedure has been published by us [39-41].
2.8. Electrophoretic mobility shift assay (EMSA)
Electrophoretic mobility shift assays were carried out as previously demonstrated [42]. And the sequence of the biotinylated probes (sequence from a regulatory region in ABO intron 1 [GenBank KC841429]) used for testing are shown in Table 1. Gel shifts were performed using LightShift™ Chemiluminescent EMSA Kit (Thermo-Fisher, Waltham, MA, USA) and nuclear extracts were prepared from K562 cells (ATCC CCL -243™) [43]. Triptolide was dissolved in DMSO to a 20 mM stock solution which was then further diluted. Nuclear extracts were pre-incubated with three different dilutions of triptolide (5 nM, 10 nM, and 20 nM) for 5 min and three concentrations (2.5 nM, 5 nM, and 100 nM) for 10 min to test if triptolide would inhibit GATA1 protein binding to the probes. The final dilution of DMSO was 0.1%. As a vehicle control, pre-incubation was performed with 0.1% DMSO only and as a negative binding control a probe with a disrupted GATA1 site was used. Supershift assay was performed with polyclonal anti-GATA-1 IgG (1 µg/µL; Active Motif, Waterloo, Belgium).
Table 1. EMSA probe designations and sequences (wildtype and mutated GATA1-binding motif highlighted in bold).
Name
|
Sequence 5’ to 3’
|
ABOi1 GATA F
|
AGAGTCTTCGCAATGCCTGGGAAAGGGAGAGATAAGGCTCACTAGCCA
|
ABOi1 GATA R
|
TGGCTAGTGAGCCTTATCTCTCCCTTTCCCAGGCATTGCGAAGACTCT
|
ABOi1 GAGA F
|
AGAGTCTTCGCAATGCCTGGGAAAGGGAGAGAGAAGGCTCACTAGCCA
|
ABOi1 GAGA R
|
TGGCTAGTGAGCCTTCTCTCTCCCTTTCCCAGGCATTGCGAAGACTCT
|