Cell lines and treatments
Echinaforce® (batch nr. 040070, A. Vogel Bioforce AG, Roggwil, Switzerland) is a standardized preparation obtained by ethanol extraction of freshly harvested Echinacea purpurea herb and roots (95:5). Echinaforce® is marketed as a registered medicinal product and produced under conditions of Good Manufacturing Practice (GMP). Thus, a consistent quality for each produced batch is mandatory and equal to the requirements for an allopathic remedy. The composition of marker compounds like alkylamides (i.e. those compounds known to characterize this species of Echinacea) was described previously [3, 7, 26]. In contrast to pressed juice extracts, Echinaforce® extract does not contain polysaccharides which are known to stimulate the immune system nonspecifically [27-30]. The alcohol concentration of Echinaforce® tincture extract was 65% v/v and solvent controls have been included in all experimental in vitro experiments to rule out nonspecific effects. In addition, the preparation was free of detectable endotoxin as determined by means of a commercial assay kit with a lower limit of detection 0.1 unit/ml (Lonza Walkersville Inc., MD).
THP1 cells were grown in RPMI-1640 medium supplemented with glutamine, 10 % heat inactivated Fetal Bovine Serum, 50 IU/mL Penicillin, 50 µg/mL Streptomycin, 10 mM HEPES and 0.05 mM β-mercaptoethanol. Cells were treated with 1% Echinaforce® tincture versus ethanol solvent control. Each treatment condition consisted of six biological replicates.
Genome-wide gene expression analysis
Sample preparation and microarray processing
THP1 cells were treated for 48h with 1% Echinaforce® or ethanol solvent control. RNA was isolated using the RNeasy mini kit (Qiagen) according to manufacturer’s instructions. RNA concentration and purity was measured using the Nanodrop 1000 spectrophotometer (ThermoFischer, CA, USA). RNA integrity of each sample was checked using using the Experion Automated Electrophoresis System (Bio-Rad, MO, USA). Total RNA (500ng) was amplified using the Illumina TotalPrep RNA Amplification kit (Life Technologies, Carlsbad, CA, USA). Briefly, RNA was reverse transcribed using T7 oligo(dT) primers, after which biotinylated complementary or anti-sense RNA (cRNA) was synthesized through an in vitro transcription reaction. Then, 750 ng of amplified cRNA was hybridized to a HumanHT12 beadchip array (Illumina, San Diego, CA, USA) and further incubated for 18 hours at 58 °C in a hybridization oven under continuous rocking. After several consecutive washing steps, bead intensities were read on an Illumina iScan. Microarray data and raw gene expression intensities were preprocessed and analyzed using the beadarray R package [31]. Intensities were quantile normalized and log2 transformed. Raw and normalized array data were uploaded to the Gene Expression Omnibus (GEO) database and have accession number: GSE117904. Probes with a P-detection value higher than 0.05 in at least six samples were removed. Also, probes annotated as “bad” and “no match” as described before [32] were not kept for further analysis. Differentially gene expression was performed using the limma R package [33]. P-values were corrected for multiple testing using the method of Benjamini and Hochberg. Probes with a log2 fold change higher than 0.4 and an adjusted p-value less than 0.05 were defined as significant and kept for further analysis [34]. The probes were annotated with gene information using the illuminaHumanv4.db annotation dataset [35]. The gene IDs of the significant Illumina expression probes were uploaded into the IPA software to find enriched biological pathways, diseases and networks. Metascape protein-protein-interaction enrichment analysis of differentially expressed genes was performed with the following databases: BioGrid, InWeb_IM, OmniPath [36]. The Molecular Complex Detection (MCODE) algorithm in Metascape identified densely connected network components and applies pathway and process enrichment analysis to each MCODE component independently. The three best-scoring terms by p-value are retained as the functional description of the corresponding MCODE components (as shown in the tables).
Quantitative realtime PCR
To validate microarray data, THP1 cells were treated with 1% Echinaforce® or Solvent for the indicated time-points (3, 6, 12, 24 and 48h) in three independent experiments. The effect of JAK1 inhibition was determined by treating the cells with 1 µM JAK1 inhibitor Filgotinib (GLPG0634, Selleckchem) for 30 min before adding Echinaforce®. Total RNA was isolated using the RNeasy mini kit (Qiagen, Hilden, Germany) including a DNAse treatment step as suggested by the manufacturer. Then 750 ng RNA was reverse transcribed into cDNA using oligo dT (Invitrogen), M-MLV reverse transcriptase (Promega, Wisconsin USA), 2.5 mM dNTPs and RNaseOUT (Invitrogen). Samples were incubated on 42°C for 60 min and 75°C for 15 min. For the HERV genes, cDNA synthesis was performed using random primers (Invitrogen) and incubation of the samples at 37°C for 60 min and 75°C for 15 min. qPCR was performed using the GoTaq qPCR Master Mix (Promega, Wisconsin USA) on a StepOnePlus Real-Time PCR machine (Applied Biosystems). Following primers were used: MX1 forward primer 5’-GTTTCCGAAGTGGACATCGCA-3’, MX1 reverse primer 5’-CTGCACAGGTTGTTCTCAGC-3’ (NM_001144925), IFITM1 forward primer 5’-CCAAGGTCCACCGTGATTAAC-3’, IFITM1 reverse primer 5’-ACCAGTTCAAGAAGAGGGTGTT-3’ (NM_003641), STAT1 forward primer 5’- CCATCCTTTGGTACAACATGC-3’, STAT1 reverse primer 5’-TGCACATGGTGGAGTCAGG-3’ (NM_007315), IL8 forward primer 5’-GCTCTCTTGGCAGCCTTCCTGA-3’, IL8 reverse primer 5’-ACAATAATTTCTGTGTTGGCGC-3’ (NM_000584), CXCL10 forward primer 5’-GAAAGCAGTTAGCAAGGAAAGGT-3’, CXLC10 reverse primer 5’-GACATATACTCCATGTAGGGAAGTGA-3’ (NM_001565), ACTB forward primer 5’-CTGGAACGGTGAAGGTGACA-3’, and ACTB reverse primer 5’- AAGGGACTTCCTGTAACAATGCA-3’ (NM_001101). Primer sequences for HERVs were derived from [37]. Each sample was ran in triplicate and the median Ct-values between each replicate group was selected. Ct-values were normalized using ACTB housekeeping gene. The ddCt-values or log fold changes (logFC) were calculated using the solvent control as reference sample. A paired t-test t-test was used to determine the significance of the differences between Echinaforce® and solvent expression levels.
Kinase activity profiling
Sample preparation
THP1 cells were treated with 1% Echinaforce® or ethanol solvent control for 15 min. Cell lysates were prepared according to manufacturer’s instructions. In short, cells were washed twice with cold 1X PBS and lysed with lysis buffer (1:100 dilution of Halt Phosphatase Inhibitor Cocktail and Halt Protease Inhibitor Cocktail EDTA free in M-PER Mammalian Extraction Buffer (ThermoFisher Scientific™, Rockford, USA) at a ratio of 100 µl buffer per 1x106 cells. Lysates were then incubated on ice for 15 min and centrifuged for 15 minutes at 16000 x g at 4°C. Protein concentration was quantified using the Pierce BCA Protein Assay Kit (ThermoFisher Scientific™, Rockford, USA).
Serine/threonine kinases (STK) and tyrosine kinase (PTK) pamgene assay and data analysis
Kinase activity profiling was performed PamChip® preprocessing and kinase activity profiling was performed according to manufacturer’s instructions (PamGene International BV, ’s-Hertogenbosch, The Netherlands). The first part of the protocol consisted in the blocking of the arrays with 2% BSA followed by several washing steps. Then 0.5 µg for STK and 5 µg for PTK assays together with the correspondent reaction mixes (purchased from the Pamgene) were loaded onto the arrays and incubated in the microarray system PamStation ® 12 instrument (PamGene International, Den Bosch, The Netherlands). In this step, the ATP contained in the mix leads to the activation of the kinases in the lysate which will result in the phosphorylation of the peptides on the array. Peptide phosphorylation intensities are then detected with the primary STK antibody mix and FITC-labeled antibody for STK assay and with the FITC-labelled PTK antibody (PTK assay). Images are then taken by the CCD camera in the PamStation®12 and processed by the Bionavigator software. Peptide intensities data were log2 transformed and differences in phosphorylation between Echinaforce® treated and control cultures were determined by using an univariate student t-test analysis corrected for multiple testing using the Benjamini and Hochberg method [34].
To identify potentially activated or inhibited kinases we used the STK or PTK Upstream Kinase analysis PamApp from the Bionavigator Software. The analysis is based on “in silico predictions” for the upstream kinases of phosphorylation sites in the human proteome that are retrieved from the phosphoNET database [38]. In short, a prediction algorithm is derived from known interactions between kinases and phosphorylation sites. The prediction algorithm is then used to predict the strength of undocumented interactions. The Bionavigator application uses PhosphoNet database to map putative kinases upstream of the phospho-peptides (a kinase can have multiple possible phosphosites, and a single site can be phosphorylated by different kinases). For each set of peptides mapped to a specific kinase, a “difference statistics” is calculated (=normalized kinase statistics) using following formula: as the sample mean and variance of the intensity of peptide i in group j, respectively, whereas n is the number of peptides linked with a specific kinase. A positive kinase statistic means that the kinase is activated, while a negative statistic means the kinase is inactivated compared to the control group. The kinases are subsequently ranked based on a specificity and significance score which are calculated using permutation of the peptides and samples, respectively. Following formula is used: where m is the number of times out of M permutations that |τp| > |τ|, where τp is the value of the difference statistic obtained after permutation of the samples or peptides. The significance score represents the magnitude of the change represented by the normalized kinase statistic. The specificity score represents the specificity of the of normalized kinase statistic in terms of the set of peptides used for the corresponding kinase. The higher the score the less likely it is that the observed normalized kinase statistics could have been obtained using a random set of peptides from the data set. The sum of the significance and specificity score is used to rank the kinases [39].
Genome-wide DNA methylation analysis
Sample preparation
THP1 cells were cultured for 48h with 1% Echinaforce® or ethanol solvent control. Corresponding cellular genomic DNA was isolated using the DNeasy Blood & Tissue kit (Qiagen, Hilden, Germany) according to manufacturer’s instructions. DNA concentration and purity was measured using the Nanodrop 100 spectrophotomer and 1 µg of DNA was used for bisulfite conversion using the EZ DNA methylation Kit of Zymo Research according to manufacturer’s instructions. Successful bisulfite conversion was checked using a methylation-specific PCR in a region of the SALL3 gene (see [40] for primer sequences).
EPIC DNA methylation array
The Infinium HumanMethylationEPIC BeadChip array (Illumina, San Diego, CA, USA) was used to measure genome-wide DNA methylation. 4 µL of bisulfite-converted DNA from each sample was amplified, fragmented, precipitated, resuspended and subsequently hybridized onto the BeadChips. After overnight incubation of the BeadChips, unhybridized fragments were washed away, while hybridized fragments were extended using fluorescent nucleotide bases. Finally, the BeadChips were scanned using the Illumina iScan system to obtain raw methylation intensities of each probe.
EPIC DNA methylation data preprocessing and analysis
The R package RnBeads was used to preprocess the Illumina 450K methylation data [41]. CpG-probes were filtered before normalization based on following criteria: probes containing a SNP within 3 bp of the analyzed CpG site, bad quality probes based on an iterative greedycut algorithm with a detection p-value threshold of 0.01, and probes with missing values in at least one sample. After filtering these CpG-probes, methylation values were within-array normalized using the beta mixture quantile dilation (BMIQ) method [42]. Another filtering step was performed after normalization based on following criteria: probes measuring methylation not at CpG sites (CC, CAG, CAH, …) and probes on sex chromosomes.
The methylation beta-values were transformed to M-values (M = log2(β/(1-β))) prior to further analyses. The moderated t-test incorporated in the limma R package [33] was used to calculate the statistics and p-values of the methylation differences between Echinaforce®- and solvent-treated samples. Significant differentially methylated probes (DMPs) were selected based on a false discovery rate (FDR) < 0.1 and a difference in beta-value of at least 0.05. The DMPs were annotated with gene information using the IlluminaHumanMethylationEPICmanifest R package [43]. Further gene information was retrieved from the UCSC genome browser (human hg19). Enrichment of genomic regions was calculated using the Fisher’s exact test. Pathway analysis of the genes harboring a DMP was performed using the Ingenuity Pathway Analysis (IPA) software. Raw and normalized array data were uploaded to the Gene Expression Omnibus (GEO) database and have accession number: GSE117904.
Protein expression of Mx1, STAT1 and IFITM1 proteins using western blotting
Protein expression levels of MX1, STAT1 and IFTIM1 were determined in THP1 cells treated with 1% Echinaforce® or ethanol solvent control for 48 hours, as explained before. Then, cells were washed and incubated 15 minutes on ice in lysis buffer containing: 150mM NaCl, 1mM EGTA, 1mM EDTA, 1mM ß-glycerolphosphate, 1% Triton X-100 (w/v), 20mM Tris HCl, pH = 7.5 and proteinase inhibitor (cOmplete™, EDTA-free Protease Inhibitor Cocktail, Sigma-Aldrich, USA) plus PhosphataseArrest™ Phosphatase Inhibitor Cocktail (phosphataseArrest™, G-Biosciences, USA). Cells were subsequently centrifuged for 15 minutes at 200g at 4°C and supernatant containing the soluble proteins were stored at -20°C until use. Protein lysates (20 µg) were mixed with 5X sample buffer (5% SDS, 20 % glycerol, 0.2 % bromophenol-blue, 250 mM DTT, 65 mM Tris HCl) all purchased from Sigma Aldrich (Missouri, USA), heated for 5 minutes at 95°C and loaded in a 12% SDS-PAGE gel. Proteins contained in the homogenates were separated during 30 minutes at 60-70V and 1 hour at a constant voltage of 130V. Further, 10 ml of BenchMark™ Pre-Stained Protein Standard (Life Technologies, CA, USA) was also loaded next to the samples. After separation proteins ttransferred onto a Nitrocellulose Membrane (BioRad, CA, USA) during 2 hours at 45 V. Non-specific binding sites were blocked by incubating the membranes with blocking buffer (0.05 % Tween 20, 1x TBS, 5% BSA) for 1 hour at room temperature. The membrane was then incubated with the primary antibodies: MX1 (D3W7I) Rabbit mAb #37849, IFITM1 Antibody Rabbit pAb #13126 and the STAT1 (42H3) Rabbit mAb #9175 (all purchased from Cell Signaling Technology, Massachusetts, USA) or rabbit polyclonal Anti-GAPDH antibody (ab9485, Abcam, Cambridge, UK) overnight at 4°C. After membranes were washed, they were incubated with (1:10000) Donkey anti-Rabbit IgG (H+L) Secondary Antibody-HRP (Thermo Fisher Scientific, Massachusetts, USA) for one hour at room temperature. Chemiluminiscence detection was performed using the ECL detection kit (Pierce™ ECL Western Blotting Substrate (Thermo Fisher Scientific, Massachusetts, USA) in a ChemiDoc MP system (BioRad, CA, USA).
Assessment of IL-8 and CXCL10 levels
Cell culture supernatants were collected after 3, 6, 12, 24 and 48 hours and assayed for chemokines CXCL10 and IL8 by means of an enzyme-linked immunosorbent assay (ELISA) purchased from Invitrogen (CA, USA) following manufacturer’s instructions. The assays have a detection limit of 2pg/ml for CXCL10 and 5 pg/mL for IL-8.