Cell lines

Cell lines were maintained in Iscove’s modified Dulbecco’s medium (IMDM) supplemented with 20% preselected horse serum (HS) and, where indicated, mouse IL-3 conditioned medium corresponding to 100 U rIL-3 per ml [5, 7, 8]. Cells were kept at a density of between 6x10⁴ to 1x10⁶ cells/ml and were regularly checked to be free of mycoplasma contamination by PCR. The clonal cell lines used in this study were:

• FDCP-Mix is a well-characterized clonal hematopoietic progenitor cell line derived from long term hematopoietic cultures from male BDF1 mice [6, 9]. When injected into the tail vain of irradiated syngeneic BDF1 mice, FDCP-Mix cells contribute to the reconstitution of the hematopoietic system [6]. FDCP-Mix cells exhibit a predominant blast cell morphology when maintained in the presence of IL-3 and HS, whereas when exposed to fetal calf serum (FCS) differentiation into all myeloid lineages is observed [6].
• GMV-FDCP-Mix cell clones were established from FDCP-Mix cells after retroviral-mediated transduction with vectors coexpressing murine GM-CSF (granulo-macrophage stimulating factor) and the G418-resistance gene neo [8]. Transduced clones were selected in the presence of G418 (1 mg/ml) and IL-3 in soft agar, isolated after 9–14 days, and expanded in IMDM with 20% HS and IL-3 and 1 mg/ml G418 [8]. Removal of IL-3 results in synchronized myeloid differentiation of the cultures mediated by the endogenous expression of GM-CSF [8].
• FIA18 and FIA10 cells are two clonal variants of GMV-FDCP-Mix that were selected by cloning assays performed on GMV-FDCP-Mix cells cultivated in IL-3 free medium at a cell density of 4x10⁵ cells/ml for two weeks [5]. Cells of the plucked clones were expanded and their capacity for proliferation, clonability, morphology and differentiation capacity in the absence of IL-3 was extensively examined [5]. Several independent clones were tested for leukemogenicity by injection into the tail vein of syngeneic female BDF1 mice and subsequent analyses for presence of male specific sequences in hematopoietic tissues of injected mice [5]. Clones FIA18 and FIA10 were selected for further studies, as outlined in results.

Specific growth conditions (e.g. G418 and HS), clonability, morphology, and differentiation capacity of the clonal cell lines excludes the presence of contaminating cells.

Characterization of cells

For morphological characterisation of cells, suspension cultures were concentrated on slides by cytocentrifugation (Cytospin, Shandon). The cells were stained with May-Grünwald-Giemsa. To assay differentiation capacity, cells were washed and cultured in IMDM supplemented with 20% pretested FCS, 5 U/ml rIL-3, and 2 U/ml erythropoietin [5]. After 7 to 10 days of incubation, cytospin preparations were made and the cells stained as described above. For in vitro colony-forming cell determinations, cells were plated in soft agar in IMDM supplemented with 20% HS in the absence of IL-3. After 7 to 10 days of incubation, colonies of more than 50 cells were scored. Cultures were incubated in a fully humified atmosphere of 5% CO₂, 5% O₂ in N₂.

Animal experiments and histopathology

All animal experiments were conducted in accordance with approved protocols and the requirements of the German Animal Welfare Act. Animal experiments and protocols were approved with written permission by local animal committees in Hamburg and Kiel (Ministerium für Landwirtschaft, Umwelt und ländliche Räume des Landes Schleswig-Holstein, Germany, permit no. V 312–72241.121–3 36-3/07).

Mice were kept on a C57BL/6 background and housed under pathogen-free conditions at room temperature under a 12 h light-dark cycle with free access to food and water. In 5 independent experiments 113 female mice (8–20 weeks) with a body weight of 18–24 g and 23 male mice (8–20 weeks) with a body weight of 22–35 g were randomly distributed into designated groups. B6D2F1/J mice were bred in the animal facility of the HPI Leibniz Institute for Virology, Hamburg, Germany, animal facility of the Christian-Albrechts-University zu Kiel, Germany or purchased from Charles River Laboratories. In total, 136 mice were included in this study.

Cells cultured in IMDM supplemented with 20% HS and with or without IL3 or GM-CSF were washed twice in IMDM and resuspended in PBS. 5x10⁶ cells in 0.5 ml PBS were injected into the tail vein of syngeneic B6D2F1 mice that had received a sublethal dose of irradiation (4 Gy Cobalt-60 source, dose rate 1 Gy/h) or were left untreated [5]. Mice were inspected every 1–2 days for the following signs of morbidity: loss of balance, ataxic gait, tremor, uninduced as well as induced seizures as signs for CNS metastasis and weight loss, fatigue or infections as signs for leukemia. Hind limb paralysis occured in both groups additionally to either CNS or leukemic symptoms. Animals were euthanized by cervical dislocation upon showing any signs of disease.

Organs were inspected macroscopically at autopsy. Spleen sizes were determined. Femurs were dissected and bone marrow was flushed out into IMDM supplemented with 10% FCS. Cells were counted and spun onto slides at 1000 rpm using a microspin cytocentrifuge (Shandon, USA) and stained with May-Grünwald-Giemsa. Peripheral blood was taken up with a heparinised syringe, blood smears were prepared and stained with May-Grünwald Giemsa. Peripheral blood was subjected to hemolysis by incubation for 5 min in 1 ml hemolysis buffer (155 mM NH₄Cl, 10 mM KHCO₃ 0.08 mM EDTA-Titriplex pH 7.4) at RT. Subsequently, cells from bone marrow and peripheral blood were centrifuged by 400xg for 10 min, resuspended in 500 μl Tens lysis buffer and DNA was extracted from the cell pellets as described below.

For pathological analysis of brains simultaneously with determination of y-chromosome sequences, whole brains were dissected unfixed, sectioned sagittally with 2 mm slices, examined, samples taken for DNA extraction to be used for y-chromosome specific PCR and the adjacent slice(s) fixed overnight in 4% buffered formaldehyde. These slices were embedded in paraplast (Sherwood Medical Industries, St. Louis, Missouri, USA), sectioned at 3 μm, placed on microscope slides, stained with hematoxylin and eosin and examined for leukemic infiltration.

In one experiment with 23 male and 12 female BDF1 mice, animals were anaesthetized with pentobarbital, then injected with saline into the right ventricle and subsequently perfused with 4% PFA for fixation. For histologic inspection of brain, animals were decapitated, skull preparated and after opening the foramen magnum the brain was further fixed in by Calfix (Quaratett), a formaline containing solution, for 1 week at 4°C, followed by decalcification for 10 days by Calless (Quaratett) before paraffin embedding. Embedded brains were sectioned sagittally into 3 levels and sectioned at 3 μm, placed on microscope slides, stained with hematoxylin and eosin and examined for leukemic infiltration.

Determination of male-specific sequences

Genomic DNA was extracted from crushed tissue pieces by digestion with freshly added 800 μg/ml Proteinase K in 3 ml Tens lysis buffer (10 mM Tris-HCl pH 8.0, 25 mM EDTA pH 8.0, 100 mM NaCl, 0.5% SDS) per tissue sample overnight at 56°C with shaking [10]. After debris removal by centrifugation at 5000 rpm, genomic DNA in the supernatant was precipitated with an equal amount of Isopropanol and subsequent centrifugation at 13000 rpm for 20 min at 4°C and washing twice with 80% EtOH. After air drying for 10 min, the DNA was dissolved in TE buffer pH 8.0 by incubation at 37°C for 2 hrs. In the first set of FIA18 and FIA10 injected mice, DNA was extracted using the Phenol:Cloroform method (Isolation of High- Molecular-Weight DNA using organic solvents, Sambrook, J., Fritsch, E.F., Maniatis, T.: Molecular Cloning: A Laboratory Manual, 2^nd ed., Cold Spring Harbour Laboratory Press, Cold Spring Harbour, New York 1989).

The 260/280 absorbance ratio of the DNAs used for subsequent PCR was between 1.8–2.0. The DNA yield was quantified spectrophotometrically by absorbance at 260. PCR was used to detect male-specific sequences in genomic DNA of female mice injected with male FIA10 cells. As primers for y-specific PCR sense: cactattttcccagtggtctgtgaa and antisense: aaagtacacggaaggattggctaga were used [11]. Cycling parameters were as follows: a hot start at 94°C for 2 minutes, denaturing at 94°C for 1 minute, annealing at 67°C for 2 minutes, and elongation at 72°C for 3 minutes, repeated for 30 cycles and followed by elongation at 72°C for 10 minutes. Standard curves were generated by serially diluting male mouse genomic DNA from FIA10 cells into female mouse genomic DNA prepared from brain. The presence of y-PCR amplified sequences was confirmed by Southern Blotting with pY353/B [11]. Detection limit of the semiquantitative PCR was 0.1% of male sequences in a female background. Determination of y-specific sequences was quantified with real-time PCR (n = 12) using SYBR Green qPCR Select Master Mix (Applied Biosystems) with primers sense: gttttgggactggtgacaattg and antisense: gtcttgcctgtatgtgg [12]. Detection limit of the real-time PCR was 0,03% (1.6 pg) of male sequences in a female background (5000 pg).

Microarrays

Frozen vials from time points, when the cells injected into syngeneic mice showed the respective phenotype, i.e. brain leukemia for FIA10 and myeloid leukemia for FIA18, and of FDCP-Mix control cells were thawed for microarray analyses. After establishing log-phase growth, 1x10⁷ cells at a cell density of 5x10⁵/ml from 3 independent cultures of FIA10, FIA18 cells and FDCP-Mix control cells, respectively, were harvested. Morphology of the cultured cells showed the respective phenotype, blast cells with myeloid differentiation for FIA10 and FIA18, and blast cells for FDCP-Mix. RNA from the different cell populations was isolated using RNeasy Midi Kit (Qiagen). A_260/280 ratios were between 2.0–2.1 and 18S and 28S ribosomal RNA bands clearly visible for all RNAs isolated. 5 μg of total RNA was used for cDNA synthesis (Expression Analysis Technical Manual, Affymetrix, Santa Clara, CA, USA). cRNA was generated (BioArray High-Yield Transcript Labeling kit, ENZO, Farmingdale, NY, USA) and hybridized to Affymetrix GeneChip Mouse Genome 430 2.0 Array (15 μg cRNA, 16 h, 45°C) that contain ∼45,000 probe sets of over 39,000 murine transcripts. GeneChip arrays were stained, washed and scanned according to the manufacturer’s specifications.) Analyses were performed with dChip Software. Principle component analysis (PCA) was performed using the R software (http://www.r-project.org/). Scanned GeneChip.DAT files were analyzed by dChip Software (Built date Aug 21, 2008;(http://dchip.org). Relative expression was visualized via heatmap using the MeV 4.2 of the TM4 software suite (http://www.tm4.org/; [13]). Values were Log2 transformed and normalized within each gene. Furthermore, genes were classified into Gene Ontology [14] or PANTHER groups [15] using the DAVID functional annotation analysis tool (http://david.abcc.ncifcrf.gov/). The GO and PANTHER terms used were: GOTERM_BP_ALL: GO:0006350∼transcription OR PANTHER_MF_ALL: MF00036: Transcription factor (‘Transcription and TFs’); GOTERM_BP_ALL: GO:0006915∼apoptosis; GOTERM_BP_ALL: GO:0030154∼cell differentiation; GOTERM_BP_ALL: GO:0007049∼cell cycle; GOTERM_BP_ALL: GO:0007154∼cell communication; GOTERM_BP_ALL: GO:0007155∼cell adhesion; GOTERM_BP_ALL: GO:0007242∼intracellular signaling cascade; GOTERM_BP_ALL: GO:0019222∼regulation of metabolic process; GOTERM_CC_ALL: GO:0005856∼cytoskeleton; GOTERM_BP_ALL: GO:0008283∼cell proliferation; PANTHER_BP_ALL: BP00248:Mesoderm development. Genes of the groups were reassigned to their induction values and visualized as a heatmap with the MeV 4.2 software.

qRT-PCR / real-time RT-PCR

Total RNA was isolated using RNeasy Mini Kit (Qiagen) and then reverse transcribed using Reverse Transcriptase Kit (ThermoFisher). Relative expression levels were determined by real-time PCR on a 7900HT Fast Real-Time PCR System (Applied Biosystems) in 384-well PCR-plates (ABgene) using the TaqMan Gene Expression Assays-on-Demand system (Applied Bioystems, Foster City, USA) as described previously [16] with minor changes. Primers used in qRT-PCR analyses are summarized in S1 Table. The relative expression of mRNAs was quantified by the ΔCt method with efficiency values measured in a pilot experiment for each expression assay. Relative expressions were visualized via heatmaps using MeV 4.2 software.

Protein mass spectrometry

Cells were grown in SILAC medium for at least five cell divisions as described previously to ensure complete labeling of cells [17]. Cells were washed 6 times in serum-free medium, and incubated in serum-free medium. The cells were lysed in SDS buffer (4% SDS in 0.1 M Tris/HCl ph 7.6) before heating the samples at 95°C for 5 min. DNA was sheared by sonication and cell debris was removed by centrifugation at 16.000 g for 10 min. DC protein assay (BioRad, Hercules, CA) was used to determine the concentration of the solubilized proteins in supernatants after centrifugation. Equal amounts of proteins (40 μg each) were loaded on a gradient Bis-Tris gel (4–12%, Novex) and separated by SDS-PAGE before in-gel digestion [18]. Proteins in the gel were fixed and stained with coomassie solution (Instantblue, Expedeon) before cutting each lane into ten slices. Gel slices were destained, washed and dehydrated with ethanol and 50mM ammonium bicarbonate. Proteins were reduced and alkylated with 10 mM dithiotreitol (Sigma) and 55 mM iodoacetamide (Sigma), respectively. Proteolytic digestion was carried out with 12.5 ng/μl trypsin solution (Promega, Mannheim, Germany) overnight. Obtained peptides were subsequently extracted in five steps with increasing concentrations of acetonitrile. Peptide fractions were pooled, concentrated in a vacuum concentrator and desalted via the stop and go extraction technique [19].

For liquid chromatography and mass spectrometry (MS) analysis, the LC-MS-system consisting of a nano-flow HPLC system (Easy nLC 1000; Thermo Fisher Scientific) connected to a QExactive HF mass spectrometer (ThermoFisher Scientific) via a nano electrospray ionization source (ThermoFisher Scientific) was used. Peptides were separated according to their hydrophobicities on a 15 cm in-house packed RP-column (ID 75μm; C-18-beads, diameter 3μm, Dr. Maisch GmbH, Ammerbuch, Germany) with a binary solvent system (solvent A: 0.5% acetic acid; solvent B: 80% acetonitrile, 0.1% formic acid). Peptide elution was achieved by increasing the relative amount of B from 10% to 38% in a linear gradient within 50 min, followed by 5 min up to 60% and another 5 min to 95%. Re-equilibration was done within 2 min with 5% of solvent B.

Mass spectra were acquired with a data dependent Top15 method with a resolution of 60,000 at 200 m/z and AGC target of 3x10⁶ at a maximum injection time of 20 ms. The 15 most intense peaks were further fragmented with higher energy collisional dissociation (HCD) and MS² spectra were generated at 15,000 resolution, AGC target of 1x10⁵ with a maximum injection time of 25 ms.

Proteomic data analyses

80 raw files of controls and mutants in duplicates were analyzed using MaxQuant (v1.5.3.12 [20]) and the implemented Andromeda search engine [21]. Protein assignment was accomplished with correlation of fragment spectra with the Uniprot mouse database (2016). A list of common contaminants was used to exclude these from the analysis. Searches were performed with tryptic specifications and settings for mass tolerance for MS and MS/MS spectra as default. Carbamidomethylation at cysteine residues was set as fixed modification together with oxidation of methionine, acetylation at the N-terminus and phosphorylation of serine, threonine and tyrosine as variable modifications. The minimal peptide length was set to 7 amino acids by default and the false discovery rate on protein and peptide level was 1%. Matching between runs was enabled for the analysis. Prior to further processing of the data, contaminants, reverse entries and proteins that were only identified by a modification site were filtered out. Gene Ontology (GO) annotations for biological process, molecular function and cellular compartments as well as Kyoto Encyclopedia of Genes and Genomes (KEGG) annotation and Pfam annotation for protein families were performed with Perseus (v1.5.0.31). For all comparisons, median values of the two replicates of each condition were used.

Bioinformatics and statistical analysis

For Venn diagrams the program Venny was applied [22]. Venn diagrams were generated to display intersecting or non-intersecting groups of differentially expressed RNA or protein. The Gene Ontology (GO) enrichment tool [23, 24] and QIAGEN Ingenuity Pathway Analysis (QIAGEN IPA, https://www.qiagen.com) software (QIAGEN, Redwood City, CA, USA) were employed to group differentially expressed RNA and protein into functional categories [25]. Statistical analysis for RNA and protein expression data T-Test was used. The p-value for GO analyses was computed according to the mHG or HG model. The Diseases & Functions QIAGEN IPA Analysis identified the biological functions and/or diseases that were most significant from the data set. Molecules from the dataset that met the p-value cutoff of 0.05 and log2ratio cutoff of ≥ 1.0 or ≤ 1.0 and were associated with biological functions and/or diseases in the QIAGEN Knowledge Base were considered for the analysis. A right-tailed Fisher’s Exact Test was used to calculate a p-value determining the probability that each biological function and/or disease assigned to that data set is due to chance alone.

Mice injected with FIA10 cells into the tail vein develop CNS leukemia

We have previously reported a multipotent cell line, GMV-FDCP-Mix, in which self-renewing cell growth is maintained by IL-3, whereas autocrine stimulation by GM-CSF induces differentiation along the granulocyte-macrophage pathway and clonal extinction [8]. Neither these cells nor the parental FDCP-Mix cells are tumorigenic in vivo. FIA mutant cell clones that escape clonal extinction upon IL3 removal can be isolated from GMV-FDCP-Mix cells at a low incidence [5]. In contrast to GMV-FDCP-Mix cells, the FIA mutants studied to date induce myeloid leukemia when transplanted into sublethally irradiated mice [5] and unpublished.

In this study, we characterize mutant cell line FIA10, which induced strikingly different symptoms in syngenic mice after injection into the tail vein. In contrast to mice injected with mutant FIA18, which developed myeloid leukemia, all mice injected with FIA10 cells (115 mice in 5 independent experiments) experienced neurological failure 2 to 11 weeks after injection, characterized by loss of balance, ataxic gait, tremor and seizures, leading to death within 3 months.

Microscopic examination of brains of FIA10 transplanted mice showing neurological symptoms revealed focal leukemic infiltration of the brain parenchyma and hemorrhagias (Fig 1A–1C). Leukemic infiltration of the brain parenchyma consisted of immature blasts and myeloid differentiated cells (Fig 1A–1C). Spleens of FIA10 injected mice varied considerably in size, with a median of 360 mg. Peripheral blood and bone marrow cellularity were within normal range (Table 1) and no obvious alterations in other organs except the CNS were observed. In contrast, control mice injected with leukemic FIA18 cells developed a myeloid leukemia with enlarged spleen (460-1120mg), a marked increase in in cellularity in peripheral blood (4-15x10⁷/ml), and a depression of bone marrow cellularity (1-5x10⁶/ml) in line with previous results [5]. Differential cell counts performed on stained smears from FIA18 transplanted mice revealed a marked elevation of myeloblasts, immature and mature neutrophils in peripheral blood and an increase in myeloblasts, monocytes and macrophages in stained bone marrow cytospin preparations (Table 1). Mice injected with FDCP-Mix control cells (n = 17) showed no evidence of disease over a 3 months observation period as described previously [5]. To exclude that the low dose of irradiation of the animals prior to the administration of FIA10 cells allows the cells to transit the BBB, FIA10 cells were injected into unirradiated recipients (8 mice). Irrespective of prior irradiation or not, all mice succumbed to neurological disease within 3 months and leukemic cells were found in the brain parenchyma. The origin of leukemic cells growing in the brain of diseased mice injected with FIA10 cells was determined assessing the presence of Y-chromosome specific sequences. In all diseased female mice analysed, brain samples were clearly positive for y-specific sequences at variable levels ranging from 0.1% to 0.7%, while Y-chromosome specific sequences were not detected in peripheral blood or bone marrow.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. FDCP-Mix derived leukemic cell line FIA10 invades the brain after injection into the tail vein.

A-C: Five months old female BDF1 mice 15 days after injection of 5 x 10⁶ FIA10 cells into the tail vein were analysed. Shown are histological examples of three different mouse brains. Clinical signs of these diseased animals were a clear ataxia and inducible, generalized seizures. (A) Mononuclear, hematopoietic cells including several cells with immature blast morphology infiltrate the stratum moleculare of the CNS from the subarachnoid cavity and penetrating leptomenigeal vessels. (B) Infiltrating hematopoietic cells clearly differ from the neural stratum granulosum cells of the cerebellum by their eosinophilic cytoplasm and partially segmented, dark nuclei and seem to be connected to the perivascular space. Large Purkinje cells are also present separating the molecular from the granular layer of the cerebellar cortex. (C) Nodular, partially already organized leukemic infiltrate of the leptomeninx and the subarachnoid cavity of a cistern at the base of the brain. Staining: HE.

https://doi.org/10.1371/journal.pone.0295641.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Differential analysis of blood and bone marrow from sublethally irradiated mice injected with mutants FIA10 and FIA18.

https://doi.org/10.1371/journal.pone.0295641.t001

In vitro comparison of brain tropic FIA10 cell line with leukemic FIA18 cell line

In vitro cultures of FIA10 cells and FIA18 cells grown in self-renewing conditions consist of a mixture of suspension and adherent cells with a high percentage of immature blasts and some differentiated myeloid forms including monocytic cells, macrophages and granulocytes (Table 2). In the presence of differentiation-inducing factors, such as FCS and hematopoietic cytokines, FIA10 cells as well as FIA18 cells differentiate further into mature granulocytes, monocytes, macrophages and erythroid cells (Table 2) in line with other factor-independent GMV-FDCP-mix derived mutants [5]. In soft agar without additional cytokines, mutant cell clone FIA10 form undifferentiated and multicenter colonies (S1 Fig), further indicating an immature phenotype with myeloid differentiation. Taken together these results demonstrate that FIA10 and FIA18 cells have very similar myeloid differentiation potential and cloning capacity in vitro, despite the quite distinct disease pattern induced after injection into mice.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Characterisation of FIA10 and FA18 cells in vitro.

https://doi.org/10.1371/journal.pone.0295641.t002

CNS infiltrating FIA10 cells display altered RNA and protein expression compared with leukemic clone FIA18 that does not metastasize into the brain and the parental hematopoietic progenitor FDCP-Mix line

In search for potential mechanism(s) that selectively mediate leukemic metastasis into the brain, we compared genome-wide RNA and protein expression of FIA10 cells, which are capable of breaking down the BBB and specifically infiltrate the brain parenchyma, with FIA18 cells, which give rise to leukemia but do not metastasize to the brain, as well as with the parental FDCP-Mix hematopoietic progenitor cell line [5, 8]. As shown in Fig 2, RNA expression of FIA10 cells had overlapping as well as opposing expression levels with the leukemic FIA18 as well as with parental FDCP-Mix hematopoietic progenitor cell lines. Concerning differences in RNA expression of FIA10 compared with FIA18 cells, 188 genes were overexpressed and 116 underexpressed (S2 and S3 Tables). 10 genes upregulated and 10 genes downregulated in FIA10 cells compared with FIA18 cells in the gene expression arrays were chosen for independent validation by quantitative real time PCR with reverse transcription (RT-qPCR, S4 Table). At the protein level, 189 genes were upregulated (log2 ratio FIA10/FIA18 ≥ 1) and 177 genes were downregulated (log2 ratio FIA10/FIA18 ≤ 1) in FIA10 cells compared with FIA18 cells (S5 Table). Comparing differentially expressed genes of FIA10 cells with FIA18 and FDCP-Mix cell lines at the RNA and protein level, most genes are either up- or downregulated in FIA10 at the RNA and protein level, which is shown at the Venn diagrams for differentially expresssed RNAs or proteins (Fig 3A and 3B).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Hierarchical cluster analysis between FIA10, FIA18 and FDCP-Mix cells.

Abundance (log2 values) from low (white) to intermediate (blue) to high (red) expression from all probe sets. Analyses were done in triplicates. Values within the triplicates are similar with minor variances. mRNA expressions patterns of FIA10 compared with FIA18 and FDCP-Mix are considerably different with FIA10 being more close to FDCP-Mix but also sharing expression levels for a number of mRNAs with FIA18.

https://doi.org/10.1371/journal.pone.0295641.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Venn diagrams of differential RNA and protein expression in FIA10 and FIA18 in relation to FDCP-Mix cells.

RNAs and proteins were selected as up- or down-regulated if they had a log2fold change ≥ 1.0 or ≤ 1.0 with a p value ≤ 0.05. (A) Venn diagram of differential RNA expression. (B) Venn diagram of differential protein expression.

https://doi.org/10.1371/journal.pone.0295641.g003

Ingenuity Pathway Analysis of differentially expressed RNA and protein of the brain leukemic FIA10 cell line compared with the parental hematopoietic FDCP-Mix and myeloid leukemic FIA18 cell lines

Biofunctional analyses were performed on the sets of differentially expressed RNAs and proteins of FIA10 cells vs controls (S2 and S3 Tables). To interpret the biological significance of expression data, QIAGEN Ingenuity Pathway Analysis (QIAGEN IPA) was performed on differentially expressed RNA and protein of FIA10 cells compared with FIA18 and FDCP-Mix cells, respectively. In line with the observed brain metastasis, top alterations in RNA and protein expression between FIA10 cells and control cells were found in processes involved in the inflammatory response, inflammatory disease, organismal injury and abnormalities, cellular movement, cell-to-cell signaling, cell death and survival, immune cell trafficking, hematopoietic system development and function and tissue morphology (Table 3). In all but two categories considerable correlation between the mRNA and protein results were found, corroborating the significance of these results. Some discordance is expected as the spatial and temporal variations of protein and mRNAs influence the relationship between protein levels and their coding transcripts [26, 27].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Top diseases and biological functions based on differentially expressed RNA and protein of brain metastasing FIA10 cells.

https://doi.org/10.1371/journal.pone.0295641.t003

GO enrichment analysis of differentially expressed RNA and protein of the brain leukemic FIA10 cell line compared with the parental hematopoietic FDCP-Mix and myeloid leukemic FIA18 cell lines

Next, we analysed functional groups that were significantly differentially represented in FIA10 cells with a GO enrichment analysis. Tables 4, 5 shows enrichment in specific cellular component, molecular function and biological processes of selected upregulated (Table 4) and down-regulated (Table 5) protein and RNA expression of FIA10 cells compared with FIA18 cells. All differentially expressed genes are categorized in S2–S5 Figs and S6–S17 Tables. In the cellular component domain, the categories extracellular region (RNA and protein up- and down-regulated) and MHC protein complex (RNA and protein up-regulated), lytic vacuoles (RNA and protein upregulated) and nucleosomes (protein upregulated) were prevalent in FIA10. In the molecular function domain, the most significantly regulated categories included cytokine receptor activity (RNA up-regulated and protein down-regulated). In addition, down-regulation of cell-adhesion molecule binding, lipoprotein particle binding, lipopolysaccharide binding and integrin binding (all RNA) as well as upregulation of MHC protein complex binding (RNA and protein), monocarboxylic acid binding (protein), antigen binding (RNA and protein) and chemokine activity (RNA), cytokine binding and signaling receptor activity (RNA), and single-stranded DNA 5’-3’ exodeoxyribonuclease activity (RNA) were all significantly differentially regulated.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Gene ontology enrichment analysis based on of differentially expressed RNA and protein of brain metastatic FIA10 cells: Upregulated protein and RNA expression of brain metastatic FIA10 cells compared with leukemic but non-brain metastatic FIA18 cells.

https://doi.org/10.1371/journal.pone.0295641.t004

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Gene ontology enrichment analysis based on of differentially expressed RNA and protein of brain metastatic FIA10 cells: Downregulated protein and RNA expression of brain metastatic FIA10 cells compared with leukemic but non-brain metastatic FIA18 cells.

https://doi.org/10.1371/journal.pone.0295641.t005

In the biological process domain, the prevalent categories for upregulated RNA as well as protein of FIA10 compared with FIA18 cells were immune system process and its regulation, regulation of multicellular organismal process, antigen processing and presentation and defense response (Table 4). For upregulated protein of FIA10 vs FIA18 cells, chromatin silencing and negative epigenetic regulation of gene expression were listed, while upregulated FIA10 vs FIA18 RNA further included response to cytokine, inflammatory response and its regulation, positive regulation of cell-cell adhesion, (positive) regulation of cell migration, cell motility and its regulation, collagen catabolic and metabolic process, cell chemotaxis, negative regulation of cell death, regulation of neuroinflammatory response, positive regulation of microgial and glial migration, long-chain fatty acid metabolic process and cellular extravasation (Table 4, S2 and S3 Figs and S6–S11 Tables). For down-regulated FIA10 protein and RNA, the prevalent category in the biological process domain was leukocyte migration involved in the inflammatory response (Table 5). In addition, downregulated FIA10 vs FIA18 RNA in the biological process domain extended to further immune response categories including the inflammatory response, defence response, innate immune response, regulation of immune system process and acute inflammatory response. Other categories for down-regulated FIA10 vs FIA18 RNA encompassed regulation of cell migration, pyrimidine dimer repair by nucleotide-excision repair and regulation of triglyceride metabolic process (Table 5, S4 and S5 Figs and S12–S17 Tables).