Transcriptome and proteome analysis of tyrosine kinase inhibitor treated canine mast cell tumour cells identifies potentially kit signaling-dependent genes

Cell proliferation, metabolism and death after masitinib treatment

Treatment of C2 cells with 100 nM masitinib induced a complete growth arrest with stable cell counts during 72 hours of treatment. In contrast, cell numbers in untreated C2 cells constantly increased during the experiment with an almost 3-fold rise of cell counts after 72 hours.

A WST-1 assay was performed to assess mitochondrial function before exposure, and after 24, 48 and 72 hours of masitinib exposure (Figure 1). There was an approximately 60% reduction in WST-1 conversion of masitinib treated cells at all indicated time points when compared to the initial activity of the untreated cells. A starting point analysis was chosen to reduce the influence of variable cell concentrations of treated and untreated cell cultures and the different metabolic activity during the different growth curve phases of untreated cells.

LDH-leakage assay identified a significant increase of 11.4% after 72 hours but no changes after 24 and 48 hours, indicating only a reduced level of cell death and cytotoxicity at the masitinib concentration used (Figure 2).

Changes in the transcriptome after masitinib treatment

Treatment of C2 cells with masitinib (100nM) induced a massive change in their global gene expression pattern. A total of 2,116, 3,087 and 3,502 genes had significant changes in their expression levels of > 1.5-fold after 12, 24 and 72 h, respectively (Figure 3, Additional file 1). Approximately 59% of these genes had decreased expression levels while the rest had increased expression. Approximately one third of these genes code for nuclear proteins while 18–26% of the gene products are expressed in the cytoplasm and in cell organelles (Figure 4).

Most nuclear factors were involved in mitosis and DNA replication (Figure 5), which were mostly down-regulated after masitinib treatment. In addition, genes associated with stress response, glycolysis and the citrate cycle were significantly down-regulated (Figure 5).

An up-regulation of mRNA expression levels was mostly observed for genes associated with Golgi apparatus, endoplasmic reticulum and lysosomes and genes associated with apoptosis and proteolysis (Figure 6). Of note, a set of pro-apoptotic genes were significantly enriched in both, up-regulated and down-regulated, groups of genes.

Pathway analysis identified a significant down-regulation of gene expression levels associated with p53, steroid receptor and GTPase-associated signal transduction pathways. In contrast, there was a time dependent increase in the number of up-regulated genes associated with signal transduction pathways during masitinib treatment. After 12 hours of masitinib treatment there was a significant up-regulation of genes associated with three signal transduction pathways, i.e. T-cell receptor, insulin receptor and steroid hormone receptor. At 24 hours genes associated with five additional pathways were up-regulated, i.e. thyroid receptor, vitamin D receptor, Ras cascade, IL10 receptor and IGE receptor. Finally, at 72 hours up-regulation of genes associated with 15 signal transduction pathways were recorded, the aforementioned pathways and the signalling cascades of the B-cell receptor, MAP kinase, EGF receptor, focal adhesion, CXCR4 receptor, EPO signalling and PTEN signalling (Figure 7).

A correlation analysis of expression levels with the different timepoints identified 89 genes with a time dependent, continuous up-regulation in gene expression levels during masitinib treatment, including the cyclin-dependent kinase inhibitor 1A (p21, Cip1), parathyroid hormone (PTH) and platelet/endothelial cell adhesion molecule 1 (PECAM1) ( Additional file 2). DAVID analysis identified a significant enrichment of the functional annotations apoptosis, ATM-signalling pathway, RAS protein signal transduction, aging, B-cell proliferation and unfolded protein response in this group of genes. A correlation analysis identified 55 genes that had a time dependent, continuous decrease in expression levels during masitinib treatment, including EIF2 and EIF5. Enriched functional annotations in this gene subset were butyrate and pyruvate metabolism, mitochondrial functions, cell migration, apoptosis and mitosis ( Additional file 2).

Changes in the proteome after masitinib treatment

2D-DIGE and MALDI-TOF analyses of the proteome after 24 and 72 hours of masitinib treatment identified 24 proteins with significant differences in protein expression levels when compared to the proteome before masitinib treatment (Figure 8A and 8B). Three proteins, TAR-DNA binding protein (TARDBP, 1.52-fold, p = 0.012), eukaryotic translation initiation factor 3 (EIF3, 1.30-fold, p = 0.014) and the actin related protein 2 (ACTR 2, 1.09-fold, p = 0.0054) were down-regulated after 24 hours of masitinib treatment (Table 1). Only two proteins, annexin A1 (ANXA1, 1.66-fold, p = 0.0087) and the gelsolin-like capping protein (CAPG, 1.66-fold, p = 0.0039) were up-regulated after 24 hours of masitinib treatment.

Comparison with the set of genes identified in the transcriptome analysis identified 15 gene products to be present in the list of mRNA and proteins with significant changes in expression levels. mRNA expressions from 6 of the 8 down-regulated proteins after masitinib treatment were also down-regulated. Furthermore, mRNA from 9 of the 15 proteins up-regulated in C2 treated cells was also present in the transcriptome analysis. However, only five of the transcripts were up-regulated whereas four were down-regulated, in contrast to the situation at the protein level.

Cell line

C2 cells were kindly provided by Patrice Dubeuil (Institut national de la santé et de la recherche médicale (INSERM), Marseille, France). Cells were cultured in RPMI-1640 with stable glutamine medium (Biowest, Nuaillé, France) supplemented with 10% FCS (Biowest, Nuaillé, France), sodium pyruvate (Biochrome, Berlin, Germany), MEM non-essential Amino Acids (Biowest, Nuaillé, France) and penicilline/streptomycine (Gibco, Darmstadt, Germany) and kept at 5% CO2 and 37°C. Cells were passaged every 6 to 7 days and rethawed from an original stock every 10 to 12 weeks. In all experiments, cells from passages 3 through 10 were used. On time point 0 cell culture flasks were incubated with 100 nM masitinib Mesylate (AB Science, Paris, France) and kept at 5% CO2 and 37°C. Three replicates of RNA and protein were obtained at time point 0 before and after 12, 24, 48 and 72 hours of masitinib incubation. The supernatant of each flask was centrifuged at 500 g for 5 minutes and washed twice with un-supplemented RPMI-1640 medium (Biowest, Nuaillé, France) with centrifugation steps at 1,000 g for 1 minute. Finally the cell pellet of each flask was equally divided and used for protein or mRNA isolation, respectively. Cell pellets were resolved in 250 μl protein lysis buffer (GE Healthcare, Freiburg, Germany) or 500 μl RA1 lysis buffer (NucleoSpin RNA; Macherey & Nagel, Düren, Germany) as previously described. Samples were stored at −80°C until further use. Cell numbers were determined at each time point using Trypan Blue exclusion (Biochrome, Berlin, Germany) and viable cells were counted in a Neubauer microscope counting chamber. The experiments with the C2 cell line were performed according to the European, German and local ethical guidelines of the Freie Universitaet Berlin. Animals, humans and their tissues were not otherwise involved in the study presented here.

WST-1 assay

For WST and LDH assays, cells were seeded at a density of 1.8 × 10⁴ cells / ml in 96-well plates (Greiner, Frickenhausen, Germany). Mitochondrial activity was quantified using the Cell Proliferation Reagent WST-1 (Roche Diagnostics GmbH, Mannheim, Germany). Following the indicated treatments of cells, 10 μl WST-1 reagent was added per well (1:10 final dilution). After a 1 hour incubation at 37°C, the absorbance at 450/630 nm was measured by using an ELISA reader (Bio-Rad Laboratories GmbH, München, Germany).

Lactate dehydrogenase release

Lactate dehydrogenase (LDH) activity was determined by using the CytoTox-ONETM Homogeneous Membrane Integrity Assay (Promega GmbH, Mannheim, Germany), a fluorimetric method for measuring the release of LDH from cells with damaged membranes. All reagents were prepared according to the manufacturer’s instructions. CytoTox-ONETM reagent (100 μl) was added following treatments to each well and incubated for a further 10 minutes. Next, 50 μl stop solution was added, the plate was shaken for 10 seconds, and the fluorescent signal was recorded at the 560/590nm excitation/emission wavelength pair by using Fluostar Optima (BMG Labtech GmbH, Offenburg, Germany). Sample triplicates were treated with 5 μl lysis solution to perform a 100% cell lysis control in order to determine the maximum amount of LDH.

Protein and mRNA isolation

Proteins were extracted in 250 μl protein lysis buffer containing 7 M urea, 2 M thiourea, and 4% CHAPS (3-(3-Cholamidopropyl)dimethylammonio-1-propanesulfonate). Lysates were sonicated twice for 2 minutes and then centrifuged at 2,200 g for 2 minutes. The supernatant was collected and stored at −80°C until analysis. Protein concentrations were determined with the 2-D Quant Kit (GE Healthcare, Freiburg, Germany). For mRNA isolation pellets were transferred into 500 μl of RA1 lysis buffer (NucleoSpin RNA; Macherey & Nagel, Düren, Germany) containing 5 μl β-mercaptoethanol and homogenized by pipetting. mRNA was extracted and purified using a commercial system (NucleoSpin RNA; Macherey & Nagel, Düren, Germany) [22, 23].

RNA quality was controlled using the BioAnalyzer (Agilent Technologies, USA) and only high quality RNA (RIN > 9) was used for microarray analyses.

Microarray data analysis

Affymetrix GeneChip hybridization (Canine Genome 2.0 Array) was performed with 2 μg total RNA according to the manufacturer’s recommendations. Three chips for each timepoint of treatment and pretreated cells were stained and washed with the GeneChip Fluidics Station 450 and visualized on an Affymetrix GeneChip Scanner 3000. Microarray data were deposited at the Gene Expression Omnibus data repository under the number GSE32657.

Affymetrix CEL files were imported into Partek Genomic Suite Software (Version 6.4, Partek Inc., St. Louis, USA) and processed by the implemented gcRMA workflow (median polish probe set summarization, RMA background correction, quantile normalization) [24]. Differences in gene expression between samples at the different time points of masitinib treatment were analysed by ANOVA and false discovery rate was controlled by using the q-value method [25]. Differentially expressed genes were selected by applying a filter of q < 0.001 and a fold-change of >1.5 in both directions. Un-named genes were excluded from the list. Hierarchical clustering of the samples and genes was conducted using Pearson correlation and complete linkage. Change in expression levels were correlated with the different timpoints and a partial correlation >0.95 was accepted as linear correlated.

To supplement the gene annotations of differentially expressed genes with functional information, BLAST search and Affymetrix-provided human to canine microarray comparisons were used to map canine genes to their human equivalents as shown before [26, 27]. Using the human equivalents as templates, the DAVID database was queried for gene ontology information [28]. To study enriched functional gene families and functional annotation, all down-regulated and all up-regulated genes were submitted separately to DAVID [29]. In the case of redundant probes with a fold-change in the same direction only the probe set with the highest fold-change was included in further analyses. Selection criteria for DAVID included a medium stringency, ≥ 4 probes within a cluster and an enrichment factor > 1.3. In case of multiple appearances of similar gene families or functional annotation terms, the cluster with the higher enrichment factor was selected.

2D-DIGE and MALDI-TOF

Two-dimensional difference gel electrophoresis (2D-DIGE) was used to quantify and compare the proteome in triplicates of C2 cell pellets before and after 24 and 72 hours of masitinib treatment. Protein extracts were labelled with CyDyes (GE Healthcare, Freiburg, Germany) as previously described [30]. The internal standard was composed of equal amounts of all protein lysates used. 50 μg of protein of the respective samples were labelled with 400 pmol of the respective dye. Two cell pellet probes and the internal standard were then combined and an equal volume of 2× sample buffer (7 M urea, 2 M thiourea, 4% CHAPS, 2% Pharmalyte IPG Buffer, 2% DTT, 0.04% bromophenol blue) was added. Rehydration buffer (7 M urea, 2 M thiourea, 4% CHAPS, 2% pharmalyte IPG buffer, 40 mM DTT) was used to yield a final volume of 450 μl. The Cy-labelled samples were applied to immobilised non-linear pH gradient (IPG) strips, pH 3–7 (GE Healthcare, Freiburg, Germany), and strips were allowed to rehydrate in the dark at room temperature overnight. Isoelectric focusing (IEF) was performed using an Ettan IPGphor 3 Isoelectric Focusing Unit (Ettan IPGphor Manifold; GE Healthcare, Freiburg, Germany) for a total of 50 kVh at 20°C, 75 μA/strip.

Two steps of equilibration followed IEF: 15 minutes with equilibration buffer (6 M urea, 30% glycerol, 2% SDS, and 50 mM Tris, 0.02% bromophenol blue, pH 8.8) containing 100 mg DTT, followed by 15 minutes with equilibration buffer containing 250 mg iodoacetamide. Strips were transferred on top of 24 cm width, 12.5% SDS-PAGE gels and sealed with 0.5% low-melting-point agarose. The second-dimension molecular weight separation was carried out using an Ettan DALTsix Electrophoresis Unit (GE Healthcare, Freiburg, Germany). Running parameters used were 60 mA for 1 hour, 240 mA for 1 hour and 300 mA for 5 hours [31].

CyDye-stained protein spots were visualised with a Typhoon 9400 fluorescence scanner (GE Healthcare, Freiburg, Germany) at the respective wavelengths for the three CyDyes. Spot detection, matching and quantification of spot intensity were performed using the DeCyder 2D Software, Version 7.0 (GE Healthcare, Freiburg, Germany). Differences in expression between the different durations of masitinib treatment were analysed using an unpaired student’s t test with p-values < 0.05 considered significant. No multiple testing or FDR adjustment was done. Gels with 350 μg of protein were silver-stained and spots were picked for subsequent MS analysis [32].

Protein identification by MALDI-TOF-MS

For protein identification by matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF-MS) an Ultraflex-II TOF/TOF instrument (Bruker Daltonics, Bremen, Germany) equipped with a Smart beam™ laser was used. Peptides were obtained by trypsin in-gel digestion as previously described [32]. Protein digests were measured in the reflector mode using α-cyano-4-hydroxycinnamic acid (CHCA) as matrix. For the database search, listed contamination peaks from keratin and autoproteolytic products were excluded for peptide mass fingerprint database search with the Mascot server (http://www.matrixscience.com) in the NCBInr database. The search was restricted to mammalian sequences and one missed tryptic cleavage was considered. A mass accuracy of 50−100 ppm was used for the searches.