Page 88 - Haematologica3

N. Landberg et al.
(ADCC), thus providing a putative new therapeutic oppor-
tunity for targeting imatinib-resistant CML stem cells.
Methods
Patient samples and CD34 enrichment
Bone marrow (BM) and peripheral blood (PB) from TKI-naive chronic phase CML patients (n=34; Online Supplementary Table S1) were obtained after written informed consent and in accordance with the Declaration of Helsinki. Ten of these patients were included in the NordCML006 study (clinicaltrials.gov identifier: 00852566) and 15 in the ongoing BFORE study (clinicaltrials.gov identifier: 02130557).15,16 Mononuclear cells were isolated using lymphoprep (GE Healthcare Bio-Sciences AB, Sweden) and CD34 enrichment was performed using magnetic beads (Miltenyl Biotec, Germany) according to manufacturer’s instructions. The study was conducted with the approval of a regional ethics committee in Lund (Dnr 2011/289).
Flow cytometric analyses and FACS sorting of primary cells
Analyses of cell surface protein expression and fluorescence- activated cell sorting (FACS) was performed on a LSR Fortessa or a FACS Aria II (BD Bioscience, USA). The antibodies and viability dyes used are listed in Online Supplementary Table S2. Isotype con- trols were used at corresponding concentrations. Two or more CML samples were analyzed for each cell surface marker. Prior to RNA extraction, carried out according to the manufacturer’s instructions (Thermo Fisher Inc, USA), viable, single CD34+CD38low cells (5% lowest CD38 expressing cells of the CD34+ cells) from CML and NBM were sorted into a PicoPure RNA Isolation Kit Extraction Buffer (Thermo Fisher Scientific Inc).
RNA sequencing
To analyze gene expression, complementary (c)DNA was amplified using the SMARTer Ultra Low Input RNA Kit for Sequencing (Takara Bio Europe, France). Sequencing libraries were prepared from the amplified cDNA using the Nextera Library DNA Preparation Kit (Illumina, USA). Paired 2x151 base pair (bp) RNA sequencing was performed on a NextSeq 500 (Illumina). The reads were aligned to human reference genome hg19 using TopHat 2.0.7.17 Gene expression values were calculated as frag- ments per kilobase of transcript per million reads (fpkm) using Cufflinks 2.2.0.18 A total of ten diagnostic CML samples and four NBM controls were analyzed. RNA sequencing data have been deposited at the European Genome-phenome Archive (EGA) under the accession code EGAS00001002421. Qlucore Omics Explorer (v 3.1 Qlucore AB, Sweden) was used to identify differ- entially expressed genes.
Cell cycle status and cell culture with imatinib
To distinguish cells in G0/G1 phase from cells in S/G2/M phase in primary CML patient samples, deep red anthraquinone 5 (DRAQ5; BioStatus, UK) was added after staining for CD34, CD38, IL1RAP and CD36. Cells were incubated at room tempera- ture for 20 minutes and subsequently analyzed using a LSR Fortessa (BD Bioscence). To determine sensitivity to imatinib treat- ment, 2000 CML cells per well were FACS sorted into 96-well plates according to CD34+CD38lowIL1RAP+CD36+ and CD34+CD38lowIL1RAP+CD36- phenotypes (Online Supplementary Figure S1) and challenged with imatinib at 5μM or dimethyl sulfox- ide (DMSO) at a corresponding concentration for 72 hours. Viable cells were evaluated using CountBright Absolute Counting Beads (Thermo Fisher Inc) and 4',6-diamidino-2-phenylindole (DAPI) on a LSR Fortessa (BD Bioscience) after three days in culture.
ADCC assay
For ADCC assays we used polyclonal antibodies targeting CD36 produced in rabbit (Innovagen, Sweden) and unspecific polyclonal rabbit immunoglobulin G (IgG) isotypes (Abcam) as control. Target cells subjected to ADCC were KU812 cells, CD34+- enriched CML samples and CD34+-enriched NBM samples. Target cells were labeled with PKH26 (Sigma-Aldrich, USA), plated at 10,000 cells/well and incubated for 30 minutes at room tempera- ture with antibodies of concentrations between 0.001-10μg/ml. One hundred thousand natural killer (NK) cells harvested from healthy donors and isolated using magnetic beads (Milteny Biotec) were added to each well, and the ADCC effect was analyzed the following morning using DAPI as a viability marker. Specific ADCC-induced cell death was calculated according to the follow- ing formula:
(Percentage DAPI+ cellsantibody – Percentage DAPI+ cellsno antibody) / (0.01 x Percentage DAPI– cellsno antibody).
Statistical Analyses
Prism 6 (GraphPad Software, USA) was used for statistical analyses. When possible, the Mann-Whitney U-test was used to determine statistically significant differences between groups, in other settings the Student’s t-test was used. The Spearman’s rank test was used to determine correlations.
Results
RNA sequencing of CD34+CD38low CML cells identifies a distinct gene expression profile
In order to identify cell surface markers that are upregu- lated on primitive CML cells, we performed RNA sequenc- ing of sorted CD34+CD38+ progenitor cells and more prim- itive CD34+CD38low CML cells from ten newly diagnosed patients in chronic phase (Figure 1A). Corresponding healthy BM cells, sorted using the same strategy, were used as controls (n=4). The CML CD34+CD38low population had a distinct gene expression profile when compared to normal CD34+CD38low and CML CD34+CD38+ cells, visualized using unsupervised principal component analysis with a variance threshold of 0.27 retaining 1005 genes (Figure 1B). Using a previously curated list of 1418 genes encoding cell surface proteins, a two-group comparison revealed a statis- tically significant leukemia-specific upregulation of 32 genes (fold change >3, Q-value < 0.05) in the primitive CML cell compartment (Figure 1C,D). Correspondingly, 24 cell surface-associated genes were found to be significantly downregulated in the same cells (fold change <0.33, Q- value < 0.05, Online Supplementary Figure S2).
Validation of cell surface protein expression of upregulated genes
To assess if the upregulated genes identified by RNA sequencing corresponded to an increased protein expres- sion at the cell surface, we performed flow cytometric analyses of 16 putative targets using commercially available antibodies (Table 1 and Online Supplementary Table S2). We confirmed the previously reported leukemia-specific upreg- ulation of the following four markers: IL1RAP, IL2RA (CD25), DPP4 (CD26) and NCAM1 (CD56) within the CD34+CD38low compartment (Table1).11-13,19 Seven additional proteins, including CD36, LEPR (CD295), TFRC (CD71), ITGB3 (CD61), CD7, FCGR2A (CD32), and GP6 were found to be expressed on primary CD34+CD38low CML cells by flow cytometry (Table 1). Most of these cell surface pro-
448
haematologica | 2018; 103(3)