ABSTRACT: BACKGROUND: BCR-ABL1+ chronic myeloid leukemia (CML) is characterized by abnormal production of leukemic stem (LSC) and progenitor cells and their spread from the bone marrow into the blood resulting in extramedullary myeloproliferation. So far, little is known about specific markers and functions of LSC in CML. METHODS: We examined the phenotype and function of CD34+/CD38─/Lin─ CML LSC by a multi-parameter screen approach employing antibody-phenotyping, mRNA expression profiling, and functional studies, including LSC repopulation experiments in irradiated NOD-SCID-IL-2Rgamma-/- (NSG) mice, followed by marker-validation using diverse control-cohorts and follow-up samples of CML patients treated with imatinib. RESULTS: Of all LSC markers examined, dipeptidylpeptidase IV (DPPIV=CD26) was identified as specific and functionally relevant surface marker-enzyme on CD34+/CD38─ CML LSC. CD26 was not detected on normal CD34+/CD38─ stem cells or LSC in other hematopoietic malignancies. The percentage of CD26+ CML LSC decreased to undetectable levels during successful treatment with imatinib in all patients (p<0.001). Whereas the sorted CD26─ stem cells obtained from CML patients engrafted irradiated NSG mice with multilineage BCR-ABL1-negative hematopoiesis, CD26+ LSC engrafted NSG mice with BCR-ABL1+ cells. Functionally, CD26 was identified as target-enzyme disrupting the SDF-1alpha-CXCR4-axis by cleaving SDF-1alpha a chemotaxin for CXCR4+ stem cells. Whereas CD26 was found to inhibit SDF-1alpha-induced migration, CD26-targeting gliptins reverted this effect and blocked the mobilization of CML LSC in a stroma co-culture assay. CONCLUSIONS: CD26 is a robust biomarker of LSC and a useful tool for their quantification and isolation in patients with BCR/ABL1+ CML. Moreover, CD26 expression may explain the extramedullary spread of LSC in CML. To define specific mRNA expression patterns and to identify specific LSC markers in CML LSC, gene array analyses were performed. RNA was isolated from sorted CD34+/CD45+/CD38─ CML LSC, CD34+/CD45+/CD38+ CML progenitor cells, CML MNC, sorted CD34+/CD38─ cord blood (CB) SC, CB-derived CD34+/CD38+ progenitor cells, and CB MNC. Total RNA was extracted from sorted cells using RNeasy Micro-Kit (Qiagen) and used (100 ng total RNA) for Gene Chip analyses. Preparation of terminal-labeled cRNA, hybridization to genome-wide human PrimeView GeneChips (Affymetrix, Santa Clara, CA, USA) and scanning of arrays were carried out according to the manufacturer's protocols (https://www.affymetrix.com). Robust Multichip Average (RMA) signal extraction and normalization were performed according to http://www.bioconductor.org/ as described.18 Differences in mRNA expression levels (from multiple paired samples) were calculated as mRNA ratio of i) CML LSC versus CB SC, ii) CML LSC versus CD34+/CD38+ CML progenitors, and normal cord blood SC versus cord blood progenitors. To calculate differential gene expression between individual sample groups where appropriate, we performed a statistical comparison using the LIMMA package as described previously. Briefly, LIMMA estimates the fold change between predefined sample groups by fitting a linear model and using an empirical Bayes method to moderate the standard errors of the estimated log-fold changes for each probe set.
