Leukemia (1999) 13, 1770–1775  1999 Stockton Press All rights reserved 0887-6924/99 $15.00 http://www.stockton-press.co.uk/leu Detection of cytogenetic aberrations both in CD90 (Thy-1)-positive and (Thy-1)- negative (CD34) subfractions of patients with acute and chronic myeloid C Brendel1, B Mohr1, C Schimmelpfennig1,JMu¨ller1, M Bornha¨user1, M Schmidt1, M Ritter1, G Ehninger1 and A Neubauer2

1Universita¨tsklinikum Carl Gustav Carus, Medizinische Klinik I, Dresden; and 2Universita¨tsklinikum der Philipps-Universita¨t Marburg, Abt. fu¨r Ha¨matologie, Onkologie und Immunologie, Zentrum fu¨r Innere Medizin, Marburg, Germany

Acute myeloid (AML) and chronic myeloid leukemia early stem cells from malignant progenitor cells via distinct (CML) are thought to arise from malignant hematopoietic pro- surface marker molecules is a challenging perspective, since genitor cells representing early and undifferentiated stem cell various improved cell selection methods have been recently clones. In CML there is evidence for a progenitor cell subset 4 free of leukemic clones, depending on the course of the dis- applied. ease. Additionally, it has been suggested that in AML, the early Acute myeloid leukemia is also considered to be an early stem cell compartment (CD34+/90+) does not harbor the malig- stem cell disease with fulminant lethal course if untreated. nant clone. We analyzed white blood cells from leukemia Since only a minority of patients achieve long-term complete patients for the presence of aberrant cells in stem cell subfrac- remission, bone marrow or stem cell transplantation is still the tions. Sixteen patients with CML, six patients with AML, two patients with acute lymphatic leukemia (ALL) and one with treatment of choice if appropriate donor cells are available. chronic myelomonocytic leukemia (CMMOL), all with known Specific chromosomal abnormalities can be determined in cytogenetic abnormalities, were evaluated according to their some FAB subtypes. Furthermore, various gross chromosomal CD90 (Thy-1)-positive or -negative phenotype. Subsets were aberrations can be discovered in 60% de novo AML of differ- sorted on to slides and further characterized by FISH and/or ent subtypes; in secondary AML multiple aberrations occur standard cytogenetic testing. The bcr-abl translocation or even more frequently.5 Therefore, structural gross chromosomal abnormalities could be detected in equally high amounts of 92.2% and 89.2% in both stem cell subsets. abnormalities can often serve as distinct markers for the malig- We conclude, that in progressed AML and CML cells charac- nant cell clone. Interestingly, CD34-positive/CD90-positive or terized by specific genetic aberrations implicated in the malig- CD34/CD90-negative hematopoietic progenitor cells from nant state can be found in the CD34+/CD90+ and CD90− popu- patients with AML seem to exhibit different long-term prolifer- lation, thus making CD90 an inappropriate marker to ative abilities and different cytogenetic characteristics with distinguish benign from malignant cells in these leukemias. regard to the leukemic clone.6 These recent data offer new Keywords: hematopoietic progenitor cells; CD90; CD34; leukemia; FISH; cell sorting aspects for obtaining purged stem cell grafts for autologous transplantation in both diseases. Early hematopoietic progenitor cells are characterized through their undifferentiated proliferative potential and their Introduction expression of CD34 surface antigen. HLA-DR, CD38, c-, IL6-receptor and CD90 are also associated with hematopoietic Chronic myeloid leukemia is characterized through the progenitor cells, but not exclusively. The CD90 surface anti- expansion of an early stem cell clone mostly of myeloid origin gen is also expressed on B and T lymphocytes and harboring the Philadelphia (Ph) chromosome. At the molecu- neural cells.7,8 Cells coexpressing the CD34 and CD90 anti- lar level this typical cytogenetic finding represents the translo- gens are thought to be a stem cell subset that contains a high cation t(9;22)(q34;q11) with consecutive gene rearrangement number of long-term culture initiating cells.9,10 We sought to resulting in the chimeric bcr-abl fusion 210 kDa oncoprotein. investigate the occurrence of malignant clones in different The course of CML is usually fairly slow in the first phase of stem cell subsets of patients with CML and AML harboring the disease but will ultimately progress into an accelerated distinct chromosomal abnormalities. Cells sorted with a phase and an overt leukemic state, the blast crisis. Bone mar- + + fluorescence-activated cell sorter into CD34 /CD90 and row or peripheral stem cell transplantation from an HLA- + − CD34 /CD90 fractions were further analyzed cytogenetically identical sibling or from a matched unrelated donor is the only with FISH and/or standard cytogenetic techniques. We were chance for cure in these patients if they do not respond to unable to detect any cytogenetic difference in CD34-positive treatment with IFN-␣. If such a donor is not available autolog- progenitor cells of either CD90-positive or CD90-negative ous transplantation has been performed in some patients with phenotype. limited success because of residual leukemic cells in the graft. Different attempts have been made to obtain Ph chromosome- negative autologous graft cells.1,2 It has been shown pre- Materials and methods viously that stem cell subpopulations from patients with CML with either CD34-positive/DR-positive or DR-negative pheno- Patients type harbor bcr-abl rearrangements in a different manner, depending on the course of the disease.3 Differentiation of Blood samples were obtained from patients treated at the Universita¨tsklinikum Carl Gustav Carus in Dresden after infor- med consent. We examined blood from 13 patients with CML, Correspondence: A Neubauer, Universita¨tsklinikum der Philipps- two with primary diagnosis (PD), two in early chronic phase Universita¨t Marburg, Zentrum fu¨r Innere Medizin, Abt. Ha¨matologie, Onkologie und Immunologie, Baldinger Straβe, 35043 Marburg, (EAP), five in late chronic phase (LCP), five in accelerated Germany; Fax: ++49-6421-2866358 phase (AC) and three in blast crisis (BC). Additionally, blood Received 1 March 1999; accepted 24 June 1999 samples were taken from four patients with secondary AML, Cytogenetic aberrations in CD90 stem cell subfractions in AML and CML C Brendel et al 1771 one patient with de novo AML, one patient with first relapse regard to their CD90 phenotype as shown in Figure 1. In AML, two patients with ALL, one patient with CMMoL and selected cases purities of the sorted cells proved to be more two healthy donors (Table 1). Cells were either taken from the than 90% (range 90–98%). Cells were sorted on to slides and leukapheresis product or 20 ml peripheral blood were drawn into culture medium using a fluorescence-activated (FACS into heparinized collection tubes and submitted to Lympho- Vantage; Becton Dickinson) cell sorter. PBS (phosphate- prep (Nycomed, Oslo, Norway) gradient centrifugation buffered saline without Ca++ or Mg++) was used as sheath and according to standard protocols in order to separate a mono- sample fluid and cells were air dried in the PBS drop after nuclear cell fraction (MNC). sorting on to the slide. These stem cell subfractions were further submitted to FISH analysis and/or cytogenetic evaluation, respectively. Cell selection and sorting

Depending on the amount of CD34-positive cells, MNC were Interphase-fluorescence in situ hybridization (I-FISH) either stained directly or further enriched for CD34 engaging the Isolex (Baxter, Unterschleissheim, Germany) or MACS Slides with sorted cells were incubated at room temperature (Miltenyi Biotec, Bergisch Gladbach, Germany) technique. (RT) in 3:1 methanol:acetic acid for 15 min and air-dried. The MNCs or selected cell fractions were stained with fluoro- After incubation in hypotonic KCl (0.56%) for 3 min, chrome-conjugated monoclonal , CD34-FITC 2 × saline sodium citrate (pH 7.0) for 2 min and PBS for 3 min (HPCA-2; Becton Dickinson, Heidelberg, Germany), CD90-PE and pretreatment with pepsin solution, additional washing (in

(F15.42; Coulter Immunotech, Hamburg, Germany) or CD90- PBS and PBS-MgCl2), fixation (1% formaldehyde in PBS- PE (5E10; PharMingen, Hamburg, Germany) and isotype con- MgCl2) for 10 min and dehydration procedures (70%, 85% trols (679 1Mc7; Coulter Immunotech). Both CD90 antibodies and 100% ethanol), the hybridization was done according to were found to have almost equal staining performances on the manufacturer’s recommendations. The following DNA the sorting instrument. Staining was done for 30 min at 4°C probes were used: the LSI bcr/abl translocation probe (VYSIS), as described by the manufacturer followed by a washing step the CEP (Chromosome Enumeration DNA FISH Probes) 7 with phosphate-buffered saline (PBS). After setting adjustments spectrum orange probe (locus D7Z1; VYSIS), the CEP 8 spec- with isotype standard controls, CD34+ cells were sorted with

Table 1 Amount of CD34 subpopulations in different leukemia samples and normal blood donors

No. Diagnosis CD34+/CD90− CD34+/CD90+ CD34 (%) (%) enriched

1 CML ECP 92.4 3.3 yes 2 CML AC 0.6 0.0 no 3 CML AC 19.7 5.0 no 4 CML AC 1.4 0.1 no 5 CML BC 12.0 0.8 no 6 CML BC 9.0 0.0 no 7 CML LCP 0.2 0.0 no 8 CML ECP 4.7 0.05 no 9 CML LCP 4.2 0.0 no 10 CML AC 65.0 0.0 yes 11 CML LCP 20.0 5.0 yes 12 CML BC 60.0 1.5 no 13 CML LCP 59.0 27.0 yes 14 CML LCP 17.1 1.0 yes 15 CML PD 21.8 3.3 yes 16 CML PD 17.1 1.3 yes 17 CML AC 1.0 0.01 no

18 pre B-ALL 43.9 37.5 no 19 T-ALL 0.6 0.0 no

20 sec. AML 72.8 0.12 yes 21 AML relapse 17.9 1.3 no 22 sec. AML 61.5 0.17 no 23 sec. AML 38.0 1.3 no 24 sec. AML 70.0 0.0 no 25 de novo AML 69.0 0.5 no 26 CMMoL 93.0 0.0 yes

27 NB 73.0 0.1 yes 28 NB 1.5 0.0 yes Figure 1 The dot plot demonstrates the sorting strategy of CD34- ECP, early chronic phase; LCP, late chronic phase; AC, acceler- FITC (a) and CD90-PE (b) labeled cells from a patient with CML whose ated phase; BC, blast crisis; PD, primary diagnosis prior treatment; blood cells were magnetically preenriched for CD34. Quadrants and NB, normal blood donor. sort gates were set according to isotype control staining. Cytogenetic aberrations in CD90 stem cell subfractions in AML and CML C Brendel et al 1772 trum orange probe (locus D8Z2; VYSIS), the CEP 10 spectrum Table 2 Mean/median amount of CD34 subpopulations in differ- green probe (specific to bands 10p11.1-q11.1; VYSIS) and the ent leukemia samples

CEP 11 spectrum green probe (locus D11Z1; VYSIS). The CEP + − + + 11 probe was used as reference probe to reduce the cut-off Diagnosis n CD34 /CD90 (%) CD34 /CD90 (%) value for monosomy 7. For post-hybridization wash we used the formamide wash mean median mean median procedure. The cells were counterstained with DAPI II (125 ng/ml; VYSIS, Stuttgart, Germany). In order to visualize CML 10 11.3 4.5 0.75 0.03 ALL 2 22.3 22.3 18.8 18.8 the bcr (SpectrumGreen) and abl (SpectrumOrange) loci a tri- AML, CMMoL 5 51.3 61.5 0.65 0.5 ple band pass filter set was used. It excites and emits DAPI, total 17 24.3 12.0 2.8 0.1 rhodamine and fluorescein-type fluorophores. The same filter set was used for investigating the CEP results. n, number of samples. Internal threshold for trisomy was 3% cells showing three All samples are not enriched for CD34. signals and for monosomy it was 13% cells showing one sig- nal (done with reference probe). The cut-off value for BCR/ABL positivity in I-FISH was 10.5%.11 cells seemed to be low within the CD34+/CD90− population (33%), but in this case only three cells were evaluable by FISH analysis so that this result might rather resemble a method bias Cytogenetics then a true exception from general findings. Table 3 shows the percentages of detected aberrations within these subpopu- The FACS-sorted cells were incubated at standard culture con- lations from CML and AML patients and from one patient with ditions for 50 to 94 h in RPMI 1640 glutamax-I (Life Techno- pre B-ALL. logies, Karlsruhe, Germany) supplemented with 30% fetal calf If metaphase cytogenetic analysis was employed, aber- serum (Life Technologies), 2.6 ␮l/ml thymidine (ICN Bio- rations were detected in almost all metaphase nuclei, ie 100% chemicals, Eschwege, Germany), 10 ␮l/ml gentamycin in both progenitor cell subsets. Figure 2 depicts a metaphase (10 mg/ml; Life Technologies), 100 U/ml recombinant human karyotype of a sorted stem cell harboring the translocation (rH) granulocyte colony-stimulating factor (PromoCell, Heid- t(1;3). elberg, Germany), 100 U/ml rH granulocyte–macrophage The amount of cytogenetic changes detected with I-FISH colony-stimulating factor (PromoCell), 100 U/ml rH interleu- analysis was very high: mean 92.2% in the CD90+ fraction kin-3 (PromoCell), 1 U/ml rH erythropoietin (PromoCell) and and mean 89.2% in the CD90− fraction. The characteristic 20 ng/ml stem cell factor (PromoCell). Thirty to 53 h to the appearances of sorted bcr-abl-positive CD34 cells that have end of culture time 10 ␮l/ml KaryoMAX colcemid solution been analyzed with FISH technique are shown in Figure 3. (10 ␮g/ml; Life Technologies) was added. Chromosome prep- Fusion of the (green) bcr and red (abl) probe results in a bright aration was done according to classical standard procedures yellow signal indicating the neighborhood of both loci, ie with hypotonic incubation, methanol–acetic acid fixation and translocation t(9;22). The blue background color visualizes the Giemsa staining. DAPI stained interphase nucleus, ie the shape of the cell nucleus after sorting stress, in all cells fluorescence signals were easily detectable and discriminable. Results and discussion The detection of equally high amounts of cytogenetic aber- rations within both CD90-positive and CD90-negative We chose the CD90/Thy1 antigen and the CD34 antigen as subfractions in patients with AML is in contrast to findings candidate surface markers to define a phenotype of early stem from other groups who describe CD34+/CD90+ cells as a dis- cell subsets with distinct proliferative abilities.9,10 The amount tinct population with SCID repopulating ability and few cyto- of CD34/CD90 double positive cells was very low and often genetic changes.6 However, it should be considered that in undetectable in patients with myeloid malignancies even after different cases of AML (FAB classification, primary or second- pre-enrichment of CD34 cells with magnetic columns. Twelve ary leukemia, first diagnosis or relapse) different stem cell sub- out of 17 patients with CML and five out of six patients with sets might be affected, so that contrary results about malig- AML exhibited detectable numbers of CD34+/CD90+ cells nancy within the CD34/CD90 subset can be explained with (mean of 0.75% and 0.65%, respectively). Only one patient the limited numbers of patient samples in distinct subgroups, suffering from a Ph+ pre B-ALL showed 37.5% of CD34/CD90 and may as well be due to disease definition according to double positive cells. In two healthy volunteers the number morphological rather than biological features. of CD34+/CD90+ was also very low. The percentages and Although in early phase CML absence of Ph chromosome- mean/median values of CD90-negative and CD90-positive positive cells within the subset of CD34+/DR− cells has been progenitor cells in the samples examined are depicted in reported,3 we could not find any cytogenetic differences Tables 1 and 2. The finding of rare Thy-1 expression in acute between CD90-positive and CD90-negative fractions of CD34 myeloid leukemia is in keeping with the results from other cells in CML of different disease progression stages, in keeping groups.6,12 with others.13,14 Interestingly, the two CML patients with pri- Successful sorting was accomplished in 17/26 patients for mary diagnosis seem to have a slight difference in the percent- one or both CD90+ and CD90− fractions. Sort gates were set age of bcr-abl-positive cells within the CD90-positive and on the brightest CD90+ cells after isotype standardization as CD90-negative fractions. It could be assumed that the malig- illustrated in Figure 1. For all kinds of malignancies examined nant cell contamination is lower within that CD90-positive there was no difference in the frequencies of distinct cyto- fraction in these patients which is in concordance with Ver- genetic abnormalities in both CD90+ and CD90− stem cell faille’s observations in early phase CML,3 but that difference subfractions harboring these aberrations in an equal manner. is statistically not significant. Many authors describe an overall In one patient with CML (No. 4) the percentage of aberrant occurrence of Ph chromosome in different cells and progeni- Cytogenetic aberrations in CD90 stem cell subfractions in AML and CML C Brendel et al 1773 Table 3 Cytogenetic abnormalities found in the CD34 subpopulation of sorted samples

No. Diagnosis Cytogenetic Cell no. (×103) Material Percentage of cytogenetic aberrations (number of phases) aberrations CD90+ I-FISH CD90− I-FISH CD90+ CD90− metaphases metaphases

1 CML bcr-abl 1 Lphc 75 (3/4) 86 (87/101) 100 (3/3) 100 (8/8) 2 CML bcr-abl 1–1.5 PBMC ND 97 (193/200) NE 100 (2/2) 3 CML bcr-abl 3–10 Lph 94 (247/263) 97 (290/300) NE 100 (1/1) 4 CML bcr-abl 0.5 PBMC 70 (16/23) 33 (1/3) ND ND 5 CML bcr-abl 2–10 Lphc 94 (61/65) 90 (90/100) 100 (4/4) 100 (4/4) 11 CML bcr-abl 0.4–2 Lph NE NE 0 (0/1) NE 12 CML bcr-abl 1 PBMCc NE 96 (25/26) NE 100 (1/1) 13 CML bcr-abl 30 PBMC 96 (72/75) 96 (48/50) NE 100 (5/5) 14 CML bcr-abl 0.8–10 PBMC 91 (181/200) 92 (183/200) ND ND 15 CML bcr-abl 5–8 BM 65 (24/36) 95 (189/200) ND ND 16 CML bcr-abl 2–9 BM 68 (78/115) 86 (172/200) ND ND 17 CML bcr-abl 0.1–10 BM NE 100 (100/100) ND ND

18 B-ALL bcr-abl 10–20 PBMCc 96 (96/100) 95 (95/100) NE NE

20 sec. AML complexa 1 PBMC ND ND ND 100 (1/1) 21 AML relapse del(7) 1 PBMC 97 (281/289) 96 (131/137) NE NE 22 sec. AML t(1;3) 1–10 Lphc ND ND 100 (2/2) 100 (18/18) 23 sec. AML complexb 1 PBMC 72 (13/18) ND ND ND

PBMC, peripheral blood mononuclear cells; Lph, leukapheresis; BM, bone marrow; ND, not determined; NE, not evaluable; I-FISH, interphase-FISH. aAberrations: 42,XY,−5,−7,−9,+der(9),t(5;9),12p−,16q−,−17,+der(17),t(5;17),−18,−20. bAberrations: 49,XY,+8,+10,del(20),+21. cFrozen material. I-FISH and standard metaphase analysis were performed simultaneously and revealed comparable high amounts of cytogenetic aberrations within the stem cell subfractions.

Figure 2 A metaphase analysis of a sorted CD34 cell that harbors the translocation t(1;3) is depicted. tor cell types from CML patients, occasionally even within the sidered to be a preleukemic state predominantly affecting cells subset.15,16 of myeloid origin. About 30% of MDS patients develop an In contrast to these findings, recent data show that in MDS overt leukemia while CML patients ultimately progress into a no malignant cell clones are detectable within the blast phase. Although in MDS all three hematopoietic lineages CD34+/CD90+ pluripotent stem cell fraction.17 MDS is con- show dysplasic features there is now evidence for predomi- Cytogenetic aberrations in CD90 stem cell subfractions in AML and CML C Brendel et al 1774 References 1 Carella AM, Cunningham I, Lerma E, Dejana A, Benvenuto F, Pod- esta` M, Celesti L, Chimirri F, Abate M, Vassalo F, Figari O, Parodi C, Sessarego M, Valbonesi M, Carlier P, Prencipe E, Gatti AM, Van Den Berg D, Hoffman R, Frassoni F. Mobilization and trans- plantation of Philadelphia-negative peripheral-blood progenitor cells early in chronic myelogenous leukemia. J Clin Oncol 1997; 15: 1575–1582. 2 Brugger W, Henschler R, Heimfeld S, Berenson R, Mertelsmann R, Kanz L. Positivly selected autologous blood CD34+ cells and unseparated peripheral blood progenitor cells mediate identical hematopoietic engraftment after high dose VP16, ifosfamide, car- boplatin, and epirubicin. Blood 1994; 84: 1421–1426. 3 Verfaille CM, Bhatia R, Miller W, Mortari F, Roy V, Burger S, McCullough J, Stieglbauer K, Dewald G, Heimfeld S, Miller JS, MvGlave PB. BCR/ABL-negative primitive progenitors suitable for transplantation can be selected from the marrow of most early- chronic phase but not accelerated-phase chronic myeloid leuke- mia patients. Blood 1996; 87: 4770–4779. 4 McNiece IK, Briddell R, Stoney G, Kern B, Zilm K, Recktenwald Figure 3 Ph chromosome-positive interphases of sorted D, Miltenyi S. Large-scale isolation of CD34+ cells using the CD34+/CD90− cells in chronic myeloid leukemia are shown. Amgen cell selection device results in high levels of purity and recovery. J Hematother 1997; 6: 5–11. 5 Grimwade D, Walker H, Oliver F, Wheathley K, Harrison C, Harri- son G, Rees J, Hann I, Stevens R, Burnett A, Goldstone A. The importance of diagnostic cytogenetics on outcome in AML: analy- nance of malignant clones within the myelomonocytic pro- sis of 1,612 patients entered into the MRC AML 10 trial. Blood genitor subpopulation.17 It is an interesting observation that 1998; 92: 2322–2333. the CD90-positive stem cell fraction seems to be completely 6 Blair A, Hogge DE, Ailles LE, Lansdorp PM, Sutherland HJ. Lack of expression of Thy-1 (CD90) on acute myeloid leukemia cells free of genetically altered cells since the CD90 antigen defines with long-term proliferative ability in vitro and in vivo. Blood a heterogenous population with enrichment of LTC-IC cells, 1997; 89: 3104–3112. with committed progenitors still present in this cell popu- 7 Ritter MA, Sauvage CA, Delio D. Human Thy-1 antigen: cell sur- lation. face expression on early T and B lymphocytes. 1983; These controversial results may reflect the differences in 49: 555–564. biology of AML, CML and MDS and provide new insights into 8 Seeger RC, Danon YL, Rayner SA, Hoover H. Detection of a Thy- 1 determinant on human neuroblastoma, glioma, sarcoma, and the diversity of stem cell biology. teratoma cells with monoclonal antibody. J Immunol 1982; 128: We conclude that the CD90 antigen is not an appropriate 983–989. marker to distinguish benign from malignant blast cells in 9 Humeau L, Bardin F, Maroc C, Alario T, Galindo R, Mannoni P, either CML or AML. The very early hematopoietic progenitor Chabannon C. Phenotypic, molecular, and functional characteriz- + + cells are still poorly understood regarding phenotype, regu- ation of human peripheral blood CD34 /THY1 cells. Blood lation of self-renewal, maturation and differentiation. Recent 1996; 87: 949–955. 10 Murray L, Chen B, Galy A, Chen S, Tushinski R, Uchida N, Negrin findings indicate different proliferation kinetics even between R, Tricot G, Jagannath S, Vesole D, Barlogie D, Hoffmann R, normal progenitor cells from patients with CML compared to Tsukamoto A. Enrichment of human hematopoietic progenitor cells from healthy donors.18 This is probably due to differ- stem cell activity in the CD34+Thy-1+Lin-subpopulation from ences in patterns.19 mobilized peripheral blood. Blood 1995; 85: 368–378. The dogma of CD34 antigen positivity of pluripotent stem 11 Bornha¨user M, Mohr B, Ehrenlechner U, Neubauer A, Ehninger G. cells with SCID or sheep in utero repopulating capacity Fluorescence in situ hybridization for the bcr/abl rearrangement is dependent on the percentage of nonlymphocytic cells in peri- has also been a current focus of controversial discussion. pheral blood stem cell harvests. J Hematother 1998; 7: 425–430. Recent studies suggest the existence of a subset of cells with 12 Inaba T, Shimazaki C, Sumikuma T, Takahashi R, Hirai H, Kikuta long-term culture initiating potential with a CD34- T, Sudo Y, Yamagata N, Ashihara E, Goto H, Fujita N, Yoshimura negative/lineage-negative phenotype.20–22 Nevertheless, pur- M, Nakagawa M. Flow cytometric analysis of Thy-1 expression in ified CD34 cells reconstitute hematopoiesis rapidly and CD34-positive acute leukemia. Int J Hematol 1997; 66: 315–323. reliably.2,10 Detection of additional surface markers23–25 or 13 Maguer-Satta V, Petzer AL, Eaves AC, Eaves CJ. BCR-ABL 22 expression in different subpopulations of functionally charac- distinct biological precursor cell features is required to terized Ph+ CD34 cells from patients with chronic myeloid leuke- define progenitor cell subpopulations with distinct properties mia. Blood 1996; 88: 1796–1804. so that cell sorting strategies can be applied to patients with 14 Petzer AL, Eaves CJ, Landsdorp PM, Ponchio L, Barnett MJ, Eaves leukemic diseases with therapeutic intent. AC. Characterization of primitive subpopulations of normal and leukemic cells present in the blood of patients with newly diag- nosed as well as established chronic myeloid leukemia. Blood 1996; 88: 2162–2171. Acknowledgements 15 Fialkow PJ, Jacobson RJ, Papayannopoulou T. Chronic myelocytic leukemia: clonal origin in a stem cell common to the granulocyte, erythrocyte, platelet and monocyte/makrophage. Am J Med 1977; This work was kindly supported by the Deutsche Forschungs- 63: 125–130. gemeinschaft (Ne310/6-3). We additionally thank Ms Leontine 16 Martin PJ, Najfeld V, Hansen JA, Penfold GK, Jacobson RJ, Fialkow Ko¨hler, Ms Jeanette Mundt, Ms Ulrike Ehrenlechner and Ms PJ. Involvement of the B-lymphoid system in chronic myelogenous leukemia. Nature 1980; 287: 49–50. Jana Scha¨fer for their expert technical assistance. We also 17 Saitoh K, Miura I, Takahashi N, Miura AB. Fluorescence in situ wish to thank Dr med Thomas Wu¨ndisch and Dr med Hein- hybridization of progenitor cells obtained by fluorescence- rich Gu¨nther for providing blood samples and clinical advice. activated cell sorting for the detection of cells affected by Cytogenetic aberrations in CD90 stem cell subfractions in AML and CML C Brendel et al 1775 chromosome abnormality trisomy 8 in patients with myelodysplas- adis G, Grupp SA, Sieff CA, Mulligan RC, Johnson RP. Dye efflux tic syndromes. Blood 1998; 92: 2886–2892. studies suggest that hematopoietic stem cells expressing low or 18 Fogli M, Amabile M, Martinelli G, Fortuna A, Rondelli D, Ratta undetectable levels of CD34 antigen exist in multiple species. M, Curti A, Tura S, Lemoli RM. Selective expansion of normal Nature Med 1998; 3: 1337–1345. hematopoietic progenitors from chronic myelogenous leukemia 23 Yin AH, Miraglia S, Zanjani ED, Almeida-Porada G, Ogawa M, marrow. Br J Haematol 1998; 101: 119–129. Leary AG, Olweus J, Kearney J, Buck DW. AC133, a novel marker 19 Peters R, Leyvraz S, Perey L. Apoptotic regulation in primitive for human hematopoietic stem and progenitor cells. Blood 1997; hematopoietic precursors. Blood 1998; 92: 2041–2052. 90: 5002–5012. 20 Zanjani ED, Almeida-Porada G, Livingston AG, Flake AW, Ogawa 24 Miraglia S, Godfrey W, Yin AH, Atkins K, Warnke R, Holden JT, M. Human bone marrow CD34-negative cells engraft in vivo and Bray RA, Waller EK, Buck DW. A novel five-transmembrane hem- undergo multilineage expression that includes giving rise to atopoietic stem cell antigen: isolation, characterization, and mol- CD34-positive cells. Exp Hematol 1998; 26: 353–350. ecular cloning. Blood 1997; 90: 5013–5021. 21 Bhatia M, Bonnet D, Murdoch B, Gan OI, Dick JE. A newly dis- 25 De Waele M, Renmanns W, Damiaens S, Flament J, Schots R, Van covered class of human hematopoietic cells with SCID-repopulat- Riet I. Different expression of adhesion molecules on myeloid and ing activity. Nature Med 1998; 4: 1038–1054. B-lymphoid CD34+ progenitors in normal bone marrow. Eur 22 Goodell MA, Rosenzweig M, Kim H, Marks DF, DeMaria M, Par- J Hematol 1997; 59: 277–286.