Page 111 - 2019_07 resto del Mondo-web

attributed to resistance to chemotherapy in the small sub- population of leukemia stem cells (LSC) with the capacity to regrow the tumor from a single cell.1,10 Thus, these LSC present an obvious target for preventing relapsed disease. However, there are no definitive markers to identify an LSC within the bulk ALL, and therefore our understand- ing of the molecular drivers of LSC fate in this disease is limited.
One promising method for studying CSC is the side population (SP) assay. This assay exploits the stem cell’s inherent ability to efflux the fluorescent dye Hoechst 33342 through an increased expression of certain ATP- binding cassette (ABC) transporters. This method was originally developed by Goodell et al., who discovered that normal bone marrow cells incubated with Hoechst 33342 could be analyzed by dual-wavelength flow cytometry to reveal a small subpopulation of cells that did not accumulate the dye.11 These cells were found to be highly enriched for hematopoietic stem and progenitor cells.11 The method has subsequently been applied to identify putative stem and progenitor cells in many other normal tissues12-15 as well as human cancers,16-18 including T-ALL.19
In this study, we characterize the SP cells within tumors generated in a robust model of zebrafish ALL that is mor- phologically and genetically similar to the human disease.20-22 Tumors are generated by co-injecting single cell zebrafish embryos with plasmids containing the zebrafish lymphocyte-specific rag2 promoter driving expression of a biofluorescent protein and the murine oncogene Myc.20 Most leukemias spread quickly beyond the thymus to invade the entire body. These tumors are transplantable by intraperitoneal injection into syngeneic and/or irradiated recipients. Through limiting-dilution transplantations, the frequency at which cells from a spe- cific tumor engraft can be calculated using Extreme Limiting Dilution Analysis (ELDA) software.23 These cells are often classified as LSC but are more precisely desig- nated leukemia-propagating cells (LPC). This approach allowed us to correlate the LPC frequency in relation to the frequency of SP cells, and then use the sorted side population as a proxy for LSC to identify transcriptional changes by RNA-sequencing.
Methods
Animals
All zebrafish experiments were performed as approved by the University of Chicago Institutional Animal Care and Use Committee using clonal golden 2 (CG2) syngeneic fish, which were a gift from Dr. Sergei Revskoy.24
Generation of mosaic transgenic fish that develop acute lymphoblastic leukemia
Rag2-GFP25 and rag2-Myc26 sequences were inserted into a pDEST-Tol2-pA2 backbone (gift from Dr. Christian Mosimann) to add inverted terminal repeats of minimal recognized tol2 sequence flanking the promoter/gene sequence. Co-injection of these vectors with M3 Tol2 transposase mRNA into single-cell embryos increases the efficiency of genomic integration27 and produces GFP+ ALL tumors as described previously.28,29 Fish are screened from 21-42 days post-fertilization (dpf) for green thymi to identify tumors for further experiments.
Cell transplantation and limiting-dilution analysis
Limiting-dilution transplantation of leukemia cells was per- formed as described previously.30 Briefly, fish were euthanized in a high dose (~0.1%) of Tricaine MS-222 (Western Chemical Inc.) and tumors harvested in 10% fetal bovine serum/90% 0.9X phosphate-buffered saline and heparin salt (300 U/mL) (Sigma), filtered through a 40 μm filter, and diluted into limiting cell doses along with carrier peripheral blood cells from a wild-type fish. Five microliters of the different cell dilutions were injected into the peritoneal cavity of anesthetized CG2 recipient fish. Engraftment was screened for weekly, starting at 2 weeks, and continued for 1 month after the final positive engraftment or until all members of the transplant group were positive or dead. Engraftment rates to calculate the LPC frequency were analyzed using ELDA software.31
Side population assay and fluorescent activated cell sorting
SP staining and analyses were performed as described previ- ously.28 Briefly, tumors were harvested as described for cell trans- plantations, then incubated with 15 μg/mL Hoechst 33342 (Life Technologies) at 28°C for 2 h protected from light. The inhibitor sample also included 250 μM verapamil (Sigma). It should be noted that in our hands, the Hoechst dye was not stable at 4°C for more than 1-2 months, so fresh stock solution was made monthly for optimal staining. Cells were then washed and ana- lyzed by flow cytometry on an LSR Fortessa X-20 instrument with a UV laser and Hoechst blue and Hoechst red bandpass fil- ters (450/50 and 670/30, respectively). For fluorescent activated cell sorting, a BD AriaFusion instrument was used and cells were sorted with a 100 μm nozzle tip into 10% fetal bovine serum/90% 0.9X phosphate-buffered saline.
haematologica | 2019; 104(7)
SP enriches for LPC and Wnt expression in zebrafish ALL
Quantitative polymerase chain reaction
RNA from each tumor was reverse transcribed in a 20 μL vol- ume with random primers using the Maxima First Strand cDNA Synthesis Kit for reverse transcriptase-quantitative polymerase chain reaction (qPCR) (Fermentas) as described in the manufac- turer’s instructions. qPCR was performed in 10 μL reactions that included 1-4 μL diluted cDNA, primers and SsoAdvanced SYBR Green Supermix supplemented with ROX (Biorad). Data were collected from triplicate reactions according to recommended settings (Biorad) on a 7900HT instrument with 384-well format (Applied Biosystems), and analyzed by the delta-delta CT method.32 Expression of ef1a was used as a reference gene.
RNA-sequencing and analysis
The RNA-sequencing and analysis are described in the Online Supplementary Methods.
EdU proliferation assay
Proliferation assays were performed with a Click-iT EdU Alexa Fluor 647 Flow Cytometry Assay Kit (Invitrogen) accord- ing to the manufacturer’s instructions with incubation adjusted to2hat28°C.
Statistical analysis
Linear correlations were assessed by R2 and the Pearson corre- lation coefficient, and non-parametric correlations were assessed by the Spearman rank correlation coefficient, with two-tailed P- values used in both cases. LPC frequencies, confidence intervals, and pairwise test P-values were all determined by ELDA soft- ware.31 The statistical significance of data from qPCR experi- ments was determined by multiple t-tests.
1389