Correspondence
1311 3 Legras S, Gu¨nthert U, Stauder R, Curt F, Oliferenko S, KluinNelemans H et al. A strong expression of CD44-6v correlates with shorter survival of patients with acute myeloid leukemia. Blood 1998; 91: 3401–3413. 4 Magyarosy E, Sebestyen A, Timar J. Expression of metastasis associated proteins, CD44v6 and NM23-H1, in pediatric acute lymphoblastic leukemia. Anticancer Res 2001; 21: 819–823. 5 Bendall LJ, Gottlieb DJ, Bradstock KF. The expression of CD44 variants in normal haemopoietic progenitors and acute myeloid leukaemia cells. Leukemia 2000; 14: 1239–1246.
6 Bendall L, James A, Zannettino A, Simmons P, Gottlieb D, Bradstock K. A novel CD44 antibody identifies an epitope that is aberrantly expressed on acute lymphoblastic leukaemia cells. Immunol Cell Biol 2003; 81: 311–319. 7 Nilsson S, Johnston H, Coverdale J. Spatial localization of transplanted hemopoietic stem cells: inferences for the localization of stem cell niches. Blood 2001; 97: 2293–2299. 8 Van Driel M, Gunthert U, van Kessel A, Joling P, Stauder R, Lokhorst H et al. CD44 variant isoforms are involved in plasma cell adhesion to bone marrow stromal cells. Leukemia 2002; 16: 135–143.
Expression of APO2.7, bcl-2 and bax apoptosis-associated proteins in CD34 bone marrow cell compartments from patients with myelodysplastic syndromes Leukemia (2004) 18, 1311–1313. doi:10.1038/sj.leu.2403386 Published online 29 April 2004
TO THE EDITOR Myelodysplastic syndromes (MDS) are clonal haematopoietic stem cell disorders in which excessive apoptosis has been involved. It has been postulated that this phenomenon in MDS patients could help to explain the ineffectiveness of haematopoiesis and may contribute to the development of peripheral cytopenias, more frequently observed in early MDS. This increased intramedullary apoptosis has been demonstrated through different methodological approaches. However, discrepant results have been reported and, therefore, their clinical interpretation remains controversial. Among other factors, the cellular heterogeneity of bone marrow (BM) samples in MDS together with the use of different ex vivo methods to identify apoptotic cells may explain, at least in part, such discrepancies. On the other hand, precise information on the cell type undergoing apoptosis is frequently lacking since results usually refer to the whole sample cellularity. Finally, most recent studies have focused on BM CD34 þ cells, while few reports analyse the more mature BM CD34 cells.1 Taking all these variables into consideration, in the present study, we have used four-colour stainings analysed by flow cytometry in order to evaluate specifically the expression of early apoptosis-related proteins in the mature BM compartments of mature nucleated red cells, myelomonocytic cells and lymphocytes in a series of 31 MDS patients, and compared them with corresponding cell compartments from the normal BM of 12 healthy donors. The markers analysed included the APO2.7/7A6 antigen – a 38 kDa mitochondrial membrane protein specifically expressed at the early stages of apoptosis, which has been reported as a valuable marker for the evaluation of apoptosis2 – together with the antiapoptotic bcl-2 and the proapoptotic bax proteins. To the best of our knowledge, this is the first report in which these apoptosis-related proteins are
Correspondence: Dr MB Vidriales, Department of Hematology, Hospital Universitario de Salamanca, Paseo de San Vicente 58-182, Salamanca 37007, Spain; Fax: þ 34 23 29 46 24; E-mail:
[email protected] Received 12 December 2003; accepted 18 March 2004; Published online 29 April 2004
specifically analysed in mature BM cell populations of both lowand high-risk MDS patients. BM samples from a total of 31 MDS patients were collected at diagnosis (FAB classification: RA: 33%; RARS: 19%; RAEB: 33%; RAEB-t: 11%; and chronic myelomonocytic leukaemia: 4%) and processed within the first 24 h. MDS patients were also subclassified from the prognostic point of view according to both the International Prognosis Scoring System (IPSS) and the Spanish Prognosis Scoring System (SpPSS). Of the 31 MDS patients, 21 (68%) were classified as ‘low-risk’ MDS (IPSSp1.5 and SpPSSp3) and 10 (32%) as ‘ high-risk’ MDS (IPSS41.5 and SpPSS43). Erythrocyte-lysed, freshly obtained BM samples were analysed using a four-colour immunofluorescence technique for the simultaneous staining of surface and cytoplasmic (cyt) antigens according to well-established methods previously described in detail.3 The following combinations of fluorochrome-conjugated (fluorescein isothiocyanate/phycoerythrin (PE)/PE-cyanine 5/allophycocyanine) monoclonal antibodies (MAb) were used: and cytbcl-2/CD32/CD34/CD45; cytbax/CD32/CD34/CD45; CD32/cytAPO2.7/CD34/CD45. Appropriate isotype-matched negative controls were stained in parallel, as described previously.3 BM cells were acquired in a FACScalibur flow cytometer (Becton Dickinson Biosciences-BDB, San Jose´, CA, USA), using the CellQuest software program (BDB) and the PAINT-A-GATE PRO software program (BDB) for subsequent analysis. In each BM sample, the following subpopulations of CD34 haematopoietic cells were identified on the basis of their unique CD45 expression and sideward light scatter (SSC) properties: (1) lymphocytes (CD45hi and SSClo), (2) myelomonocytic cells (CD45int/hi and SSCint/hi), and (3) nucleated red cells (CD45 and SSClo). For each of these cell subsets, quantitative expression of cytbcl-2, cytbax and cytAPO2.7 was evaluated by the mean relative fluorescence intensity (RFI) obtained.3 The results revealed significant differences in the reactivity for the three proteins among the different cell subsets analysed (Table 1). Accordingly, bax expression was significantly lower (P ¼ 0.003) among nucleated red blood cells (mean RFI7s.d. of 0.970.2) as compared to the other cellular compartments (lymphocytes 1.470.5 and myelomonocytic cells 1.470.4). Similarly, intracellular levels of the APO2.7 protein were significantly lower (P ¼ 0.004) in myelomonocytic cells (6.773.2) as compared to both nucleated red cells (12.076.8) and lymphocytes (13.175.7). In addition, reactivity for the Leukemia
Correspondence
1312 Table 1 RFI of expression of the cyt APO2.7, bcl-2 and bax apoptosis-related proteins in different subsets of BM haematopoietic cells from both low-risk MDS (n ¼ 21) and high-risk MDS (n ¼ 10) patients as compared to normal BM (n ¼ 12) cyt APO2.7
Cell types
Low-risk MDS
RFI
High-risk MDS
cyt BCL-2
Normal BM
Low-risk MDS
RFI
High-risk MDS
cyt BAX
Normal BM
Low-risk MDS
RFI
High-risk MDS
Normal BM
Lymphocytes
13.075.5 9.577 13.175.7 13.475.7 17.977.1 18.475.5 1.270.2 2.070.9 1.470.5 12.7 7.6 11.3 14.4 19.8 16.8 1.2 1.8 1.3 ~ ~~ 2.271.2 2.470.7 2.070.4 1.670.7 1.770.9 1.470.4 Myelomonocytic cells 10.274.3* 9.876.3 6.773.2 9.4 9.2 6.1 1.8 2.5 2.0 1.3 1.4 1.3 Nucleated red cells 25.9713** 12.9710.9 12.076.8 1.370.3*** 1.770.4 1.870.4~~ 1.570.4*** 1.570.7* 0.970.2~~~ 23.3 8.3 10.5 1.3 1.8 1.7 1.3 1.3 1.0 a
Results expressed as mean7s.d. and median of RFI (mean fluorescence intensity (MFI) of the analysed cells divided by the MFI of the corresponding unstained isotype-negative control cells). [1] *Po0.05, **Pp0.01, ***Pp0.001 as compared to normal BM. ~ P ¼ 0.004 upon comparing the cytAPO2.7 RFI of myelomonocytic cells with that of the other groups of cells; ~~Pp0.0001 upon comparing the ~~~ P ¼ 0.003 cytbax RFI of nucleated red cells with cytbcl-2 RFI of myelomonocytic cells and nucleated red cells with the lymphocyte groups; and that of the other groups of cells.
antiapoptotic bcl-2 protein was significantly lower (Pp0.0001) in nucleated red cells (1.870.4) and myelomonocytic BM cells (2.070.4) as compared to the lymphocyte compartment (18.475.5). Upon comparing the expression of these proteins in normal vs MDS BM cells, different reactivity patterns were observed for both the myelomonocytic and nucleated red cell compartments, but not for the lymphocytes (Table 1). In this sense, with respect to the corresponding cell compartments from normal BM, an increased expression of APO2.7 was found in low-risk MDS (mean RFI7s.d. of 10.274.3 and 25.9713, vs 6.773.2 and 12.076.8, for myelomonocytic (P ¼ 0.007) and nucleated red cells (P ¼ 0.02), respectively), but not in high-risk cases (mean RFI7s.d. of 9.876.3 and 12.9710.9 vs 6.773.2 and 12.076.8 for the myelomonocytic and nucleated cells, respectively). Interestingly, the increased reactivity for the APO2.7 antigen observed in nucleated red cells from low-risk MDS patients as compared to normal BM was also associated with increased levels of bax (1.570.4 vs 0.970.2, P ¼ 0.001) and decreased bcl-2 expression 1 (1.370.3 vs 1.870.4, P ¼ 0.001). Once again, no statistically significant differences were observed in the expression of bcl-2 in BM nucleated red cells from high-risk MDS patients vs normal BM. However, bax expression by nucleated red cells was increased over normal values (1.570.7 vs 0.970.2, P ¼ 0.02) in this group. As regards the expression of both bcl-2 and bax in BM lymphocytes and myelomonocytic cells, no statistically significant differences were found between MDS and normal individuals. Our results suggest that, compared with normal BM, expression of APO2.7, bcl-2 and bax in MDS may vary depending on the specific cellular compartment studied as well as on the type of MDS (low- vs high-risk cases). Accordingly, nucleated red cells from low-risk MDS patients would constitute the cell compartment most prone to apoptosis, since it shows increased levels of both APO2.7 and bax, in the context of an abnormally low bcl-2 expression. In the present study, myelomonocytic cells from low-risk MDS also displayed increased reactivity for APO2.7. However, in contrast to nucleated red cells, they showed normal expression of both bcl-2 and bax. In turn, in high-risk MDS patients, expression of these three apoptosisrelated proteins was not significantly different from that of normal BM, except for a significantly higher expression of bax observed in the nucleated red blood cell compartment. None of these proteins was abnormally expressed within the mature Leukemia
lymphoid cell compartment in either low- or high-risk MDS patients. Overall, these results are in line with previous observations indicating occurrence of apoptotic cell death in BM CD34 cells from early MDS patients.1 Although in most of these studies specific identification of the cells undergoing apoptosis was not carried out, several groups1 have reported that this process predominates in the red cell compartment. Similarly, van de Loosdrecht et al4 have recently found an inverse relationship between the degree of apoptosis of erythroblasts evaluated at the ultrastructural level and the stage of the disease in MDS; simultaneously, they show a significant increase in mitochondrial abnormalities, which could represent an additional alteration involved in ineffective erythropoiesis in MDS patients in the more advanced stages of the disease, and could help to explain the lower apoptotic rate observed in highrisk MDS patients. In contrast with our results and those previously mentioned, other groups have either reported the existence of more prominent apoptotic features in MDS patients with more advanced disease involving myelomonocytic, erythroid and megakaryocytic, but not lymphoid cells5 or failed to find any significant differences between RAEB-t and other categories of MDS.6 Deregulation of the expression of apoptosis-associated proteins of the bcl-2 family has been demonstrated in different malignancies, including myeloid disorders.1 In MDS, the patterns of expression of apoptosis-related proteins differs significantly between the FAB subgroups. Accordingly, it has been postulated that progression of MDS is accompanied by an increased expression of bcl-2 and a decreased ratio between the expression of proapoptotic and antiapoptotic proteins of the bcl2 family.1 However, these findings have not been fully confirmed and they are almost entirely restricted to BM CD34 þ cells. Expression of an apoptotic phenotype by nucleated red cells from low-risk MDS and, to a lesser extent, also from high-risk MDS cases, would support the existence of an ineffective erythropoiesis in these patients, in line with the clinical behaviour of the disease. As regards myelomonocytic cells, our results show a higher reactivity for APO2.7 in low-risk MDS as compared to normal BM. In contrast, myelomonocytic cells from low-risk MDS showed normal expression of both bcl-2 and bax. APO2.7/7A6 is a mitochondrial membrane protein expressed during the early stages of apoptosis in relation to the release of cytochrome c outside the mitochondria. Previous studies suggest that an increase in APO2.7 expression does not suggest apoptosis since
Correspondence
it could be prevented if bcl-2 or bcl-xL are overexpressed.7 Although bcl-xL expression was not evaluated in the present study, bcl-2 levels were normal, supporting the notion that in low-risk MDS an increased expression of APO2.7 by BM myelomonocytic cells could reflect a proapoptotic phenotype. This would be in line with the hypothesis that the cytochrome c-mediated apoptosis pathway could be involved in the excessive apoptosis of MDS patients,8 and at the same time indicates that APO2.7 could represent a useful marker to evaluate susceptibility of BM myelomonocytic cells to apoptosis. In summary, our results indicate that in MDS, and particularly in low-risk MDS, CD34 nucleated red cells and, to a lesser extent, myelomonocytic cells, but not mature lymphocytes, display a proapoptotic phenotype, which probably reflects a greater susceptibility to apoptosis in these BM cells, which is consistent with the clinical behaviour of the disease.
Acknowledgements L Sua´rez was a recipient of a grant (Ref. 63022) from the ‘Agencia Espan˜ola de Cooperacio´n Internacional’ from Madrid (Spain) during the first months of the study and from the ‘Secretarı´a de Estado de Educacio´n y Universidades’ (Ref. SB 2000-0089) as part of the Spanish program ‘Estancia de Profesores, Investigadores, Doctores y Tecno´logos Extranjeros en Espan˜a.’
L Sua´rez1,2 MB Vidriales1,2 G Sanz3 A Lo´pez4 MC Lo´pez-Berges1 M de Santiago4 L Palomera5 T Bernal6 ME Pe´rez de Equiza7 JF San Miguel1,2 A Orfao,2,4 for the PETHEMA Cooperative GROUP
1
Department of Hematology, University Hospital, Salamanca, Spain; 2 Centro de Investigacio´n del Ca´ncer, Salamanca, Spain; 3 Department of Hematology, Hospital La Fe, Valencia, Spain; 4 Department of Cytometry, University of Salamanca, Spain; 5 Department of Hematology, Lozano Blesa Hospital, Zaragoza, Spain; 6 Department of Hematology, Central de Asturias Hospital, Oviedo, Spain; and
7
Department of Hematology, Navarra Hospital, Pamplona, Spain
1313
References 1 Parker JE, Mufti GJ. The role of apoptosis in the pathogenesis of the myelodysplastic syndromes. Int J Hematol 2001; 73: 416–428. 2 Koester SK, Roth P, Mikulka WR, Schlossman SF, Zhang C, Bolton WE. Monitoring early cellular responses in apoptosis is aided by the mitochondrial membrane protein-specific monoclonal antibody APO2.7. Cytometry 1997; 29: 306–312. 3 Sua´rez L, Vidriales B, Garcı´a-Larana J, Lo´pez A, Martı´nez R, PETHEMA Cooperative Group et al. Multiparametric analysis of apoptotic and multi-drug resistance phenotypes according to the blast cell maturation stage in elderly patients with acute myeloid leukemia. Haematologica 2001; 86: 1287–1295. 4 van de Loosdrecht AA, Brada SJ, Blom NR, Hendriks DW, Smit JW, van den Berg E et al. Mitochondrial disruption and limited apoptosis of erythroblasts are associated with high risk myelodysplasia. An ultrastructural analysis. Leukemia Res 2001; 25: 385–393. 5 Raza A, Gezer S, Mundle S, Gao X-Z, Alvis S, Borok R et al. Apoptosis in bone marrow biopsy samples involving stromal and hematopoietic cells in 50 patients with myelodysplastic syndromes. Blood 1995; 1: 268–276. 6 Huh YO, Jilani I, Estey E, Giles F, Kantarjian H, Freireich E et al. More cell death in refractory anemia with excess blasts in transformation than in acute myeloid leukemia. Leukemia 2002; 16: 2249–2252. 7 Carthy CM, Granville DJ, Jiang H, Levy JG, Rudin CM, Thompson CB et al. Early release of mitochondrial cytocrome c and expression of mitochondrial epitope 7A6 with a porphyrin-derived photosensitizer: Bcl-2 and Bcl-xL overexpression do not prevent early mitochondrial events but still depress caspase activity. Lab Invest 1999; 79: 953–965. 8 Tehranchi R, Fadeel B, Forsblom AM, Christensson B, Samuelsson J, Zhivotovsky B et al. Granulocyte colony-stimulating factor inhibits spontaneous cytochrome c release and mitochondria-dependent apoptosis of myelodysplastic syndrome hematopoietic progenitors. Blood 2003; 101: 1080–1086.
Mitochondrial mutations in acute leukaemia Leukemia (2004) 18, 1313–1316. doi:10.1038/sj.leu.2403380 Published online 6 May 2004
TO THE EDITOR Mutations in the mitochondrial genome have been reported in various forms of carcinoma. For acute leukaemia, however, there is only one definitive report1 of their presence in one patient with acute lymphoblastic leukaemia (ALL), together with another
Correspondence: Professor AA Morley, Department of Haematology and Genetic Pathology, Flinders University and Flinders Medical Centre, Bedford Park, South Australia 5042, Australia; Fax: þ 61 8 82045450; E-mail:
[email protected] Received 16 November 2003; accepted 16 March 2004; Published online 6 May 2004
report2 on alteration of restriction fragment length polymorphisms in ALL cells, which could be interpreted as being due to acquired mutations. We therefore sought mitochondrial mutations in 22 patients with acute myeloid leukaemia (AML) (11 males, 11 females, ages 16–69) and 26 patients with ALL (14 males, 12 females, 23 children, three adults). In all, 15 of the ALL patients were specifically selected because they had relapsed. All patients were studied by sequencing the mitochondrial D loop from nucleotides 16111–190 and in 37 patients; the same region was also studied by denaturing gradient gel electrophoresis (DGGE). Mutations were found at diagnosis by sequencing in eight of the 22 patients (36%) with AML and 15 of the 26 patients (58%) with ALL. The difference in frequency between AML and ALL was not significant (P ¼ 0.09, Fisher’s exact test). Aberrant bands resulting from mutation were detected by DGGE with a sensitivity of 93% and a specificity of 98%. Leukemia
