Correspondence
875
Figure 4 Regulation of IAP and caspase ubiquitination by integrins. (a) Nalm-6 cells were stimulated for 2 days with Fn (F2) or PLL (P2) in the absence of serum. IAP2, XIAP or caspase 7 were immunoprecipitated and analyzed by Western blot using anti-Ubiquitin and anti-IAP2, XIAP or caspase 7 antibodies. TCL ¼ total cell lysate. An immunoprecipitation with an irrelevant antibody is also shown for XIAP (IpC vs IpX). For ubiquitination experiments, Fn/PLL ratios were quantified by densitometry analyses and indicated at the bottom of each panel (F/P: x%). (b) Cells were first stimulated with Fn or PLL for 18 h before adding MG132 (1 mM) or not (Ctrl) for 8 h. The samples were analyzed by Western blot using anti-Ubiquitin and anti-XIAP or caspase 7 antibodies. These experiments are representative of three independent experiments.
AA was partly supported by CNRS, France. We thank Dr Hamid Band (Boston, MA) for helpful discussion.
AL Astier1,2 M Svoboda1,2 E Hinds1 R de Beaumont1 O Munoz1,2 AS Freedman1,2
1
Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA; and 2 Department of Medicine, Harvard Medical School, Boston, USA
References 1 Manabe A, Murti KG, Coustan-Smith E, Kumagai M, Behm FG, Raimondi SC et al. Adhesion-dependent survival of normal and leukemic human B lymphoblasts on bone marrow stromal cells. Blood 1994; 83: 758–766. 2 Mudry RE, Fortney JE, York T, Hall BM, Gibson LF. Stromal cells regulate survival of B-lineage leukemic cells during chemotherapy. Blood 2000; 96: 1926–1932.
3 de La Fuente MT, Casanova B, Moyano JV, Garcia-Gila M, Sanz L, Garcia-Marco J et al. Engagement of alpha4beta1 integrin by fibronectin induces in vitro resistance of B chronic lymphocytic leukemia cells to fludarabine. J Leukocyte Biol 2002; 71: 495–502. 4 Astier AL, Xu R, Svoboda M, Hinds E, Munoz O, de Beaumont R et al. Temporal gene expression profile of human precursor B leukemia cells induced by adhesion receptor: identification of pathways regulating B-cell survival. Blood 2003; 101: 1118–1127. 5 Kumagai M, Manabe A, Pui CH, Behm FG, Raimondi SC, Hancock ML. Stroma-supported culture in childhood B-lineage acute lymphoblastic leukemia cells predicts treatment outcome. J Clin Invest 1996; 97: 755–760. 6 Matsunaga T, Takemoto N, Sato T, Takimoto R, Tanaka I, Fujimi A et al. Interaction between leukemic-cell VLA-4 and stromal fibronectin is a decisive factor for minimal residual disease of acute myelogenous leukemia. Nat Med 2003; 9: 1158–1165. 7 Yang Y, Fang S, Jensen JP, Weissman AM, Ashwell JD. Ubiquitin protein ligase activity of IAPs and their degradation in proteasomes in response to apoptotic stimuli. Science 2000; 288: 874–877. 8 Shi Y. Mechanisms of caspase activation and inhibition during apoptosis. Mol Cell 2002; 9: 459–470.
Minimal residual disease cells in AML patients have an apoptosis-sensitive protein profile Leukemia (2004) 18, 875–877. doi:10.1038/sj.leu.2403299 Published online 5 February 2004 TO THE EDITOR
Relapse in acute myeloid leukemia is common and is thought to arise from minimal residual disease (MRD) cells surviving Correspondence: Dr GJ Schuurhuis, Department of Hematology, VU University Medical Center, BR240, PO Box 7057, 1007 MB Amsterdam, The Netherlands; Fax: þ 31 20 4442601; E-mail:
[email protected] Received 15 December 2003; accepted 22 December 2003; Published online 5 February 2004
chemotherapy.1 Survival of MRD cells has been attributed to drug resistance mechanisms. Defects in apoptosis pathways are probably contributing significantly to resistance to a variety of chemotherapeutic agents. Despite an emerging knowledge of the regulatory mechanisms of apoptosis, Bcl-2 still stands at the top of the list of parameters that contribute to chemotherapeutic resistance. In AML, it has been shown that Bcl-2 is of independent prognostic significance. High Bcl-2 predicts failure to achieve complete remission2 and shorter overall survival.3 In this study the expression profile of Bcl-2 and Bcl-2 related proteins in MRD cells was determined to examine whether chemotherapy-induced upregulation of an apoptosis-resistant protein profile or selection of subpopulations with a high Leukemia
Correspondence
876 anti-apoptosis protein profile occurs during chemotherapeutic treatment. In all, 61 AML patients were enrolled in this study after informed consent. Fresh bone marrow (BM) aspirates were obtained at diagnosis (n ¼ 61), after first cycle of induction chemotherapy (n ¼ 22), second cycle of induction (n ¼ 15) and after postinduction cycle (n ¼ 10) of chemotherapy, at follow-up (n ¼ 13), and at relapse (n ¼ 9). The median age was 54 years, 29 male and 32 female subjects distributed over RAEB-t (5 ), M0 (5 ), M1 (7 ), M2 (14 ), M3 (2 ), M4 (12 ), M5 (10 ), M6 (3 ), unknown (3 ) according to the FAB classification. In toal, 11 normal BM (nBM) from healthy donors were included. To enable the study of apoptosis-related protein expression in MRD cells, we relied upon an intracellular flow cytometry method.4,5 Briefly, this method consists of two parts; firstly, patient-specific leukemia-associated phenotypes (LAPs) are determined at diagnosis followed by an intracellular procedure to detect the Bcl-2-related proteins in these cells. The LAP can then be used to detect MRD cells in follow-up BM samples. Figure 1a shows an example of an AML patient with a crosslineage LAP. Gating on the LAP þ cells enables quantification of the apoptosis-related proteins of these cells (Figure 1c). At diagnosis, blasts from 61 AML patients showed a heterogeneous Bcl-2 expression (mean7s.d.: 16.1710.5), Bcl-xL (2.971.9), Mcl-1 (8.272.4) and Bax (4.274.3). The mean Bcl-2 expression was significantly lower in nBM CD34 þ cells (Bcl-2: 6.672.5, P ¼ 0.001). Also, Bcl-xL (2.270.8) and Mcl-1 (6.873.3) were lower and Bax was higher (5.774.0), although these differences were not significant. All this suggests that the majority of AML cases have an increased anti-apoptosis protein profile potentially offering the leukemic cells a survival advantage under therapy. This was tested with the same analysis procedure as at diagnosis, but was now performed on the MRD cells from patients in CR. Pgp activity analysis6 was performed in 17/51 MRD samples, showing deviant Pgp activity in MRD cells as compared to normal CD34 þ cells confirming the malignant origin of the MRD cells. Figure 1b shows the MRD cells in a BM sample after the second cycle of induction chemotherapy. Surprisingly, the Bcl-2 expression (Figure 1d) of the MRD cells had decreased from 48 to 13. Of interest, this value now is near the range of nBM (range: 3.6–10.6). Figures 1e–h show the Bcl-2 value of all measured samples. In particular samples with high Bcl-2 levels (Figure 1e) but also with intermediate Bcl-2 levels (Figure 1f) and even with Bcl-2 levels within the normal range (Figure 1g) show decreases in expression. Overall, it can be concluded that MRD cells have significantly lower Bcl-2 levels after chemotherapeutic treatment (cycle 1: P ¼ 0.01, cycle 2: P ¼ 0.0008, cycle 3: P ¼ 0.04) and at follow-up (FU1: P ¼ 0.01 and FU2: P ¼ 0.01) than at diagnosis. In contrast, Bcl-2 levels from refractory patients showed no change after first (P ¼ 0.64) or second cycle (P ¼ 0.79) (Figure 1h). Despite the fact that the MRD cells in CR were apoptosissensitive as compared to diagnosis, they were able to survive chemotherapy resulting in some cases in relapse. Bcl-2 expression at relapse had again increased to the level comparable to diagnosis (no difference with diagnosis; P ¼ 0.69) after the described decrease in an MRD situation. Two other anti-apoptosis proteins Bcl-xL and Mcl-1 and the proapoptotic protein Bax were also studied (Table 1). In summary, although less prominent compared to Bcl-2, in responding patients Bcl-xL and Mcl-1 significantly decreased from the second cycle onwards (Table 1a), while no change was observed in refractory patients (Table 1b). Relapse samples had Leukemia
Figure 1 Bcl-2 expression in AML and MRD by flow cytometry. In all cases gating was performed on CD45low/SSClow cells (a) AML blasts at diagnosis with a CD34 þ CD7 þ LAP. (b) BM sample after the second cycle of chemotherapy from the AML patient of (a) with MRD cells identified by the LAP (CD34 þ CD7 þ ). MRD cells are 0.25% of all BM cells. Such cells are not present in control BM samples. Gating on the LAP þ cells enables quantification of Bcl-2 (c) at diagnosis and (d) after second cycle of induction chemotherapy. (e–f) Summary of Bcl-2 levels of AML patients in CR after first, second, third cycle of chemotherapy at follow-up (FU1 and FU2) and at relapse. Data are devided into three graphs for clarity reasons; diagnosis samples with Bcl-2 levels (e) higher than the mean þ 6 s.d. of nBM CD34 þ cells, (f) higher than the mean þ 4 s.d. of nBM CD34 þ cells and (g) within the range found in nBM CD34 þ cells (h) Bcl-2 levels of patients not in CR (NCR) at time of BM aspiration after first and second cycle of chemotherapy and at relapse.
similar protein levels as at diagnosis. On the other hand, Bax showed an increased expression compared to diagnosis, reaching significance after the first cycle of chemotherapy in responding patients. This shows that the prominent results found for the anti-apoptosis proteins are not the results of aspecific treatment-related cellular changes. Overall, our data show that MRD blasts do not have a more apoptosis resistant profile. Instead, when a complete remission is achieved, in the majority of patients MRD cells are apoptosis sensitive compared to diagnosis and remain apoptosis sensitive throughout the remission period. Only when patients do not reach a CR or when patients relapse, the leukemic blasts retain their diagnosis apoptosis protein profile. These unexpected results clearly show that neither chemotherapy-induced upregulation of an apoptosis-resistant protein profile nor selection of subpopulations with a high anti-apoptosis protein profile occurs.
Correspondence
877 Table 1
Changes in apoptosis-regulating proteins in leukemic blasts
(a) Samples in CR Bcl-2 Bcl-xL Mcl-1 Bax
(b) No CR Bcl-2 Bcl-xL Mcl-1 Bax
Diagnosisa mean %
Cycle 1 (n ¼ 13) mean %
Cycle 2 (n ¼ 15) mean %
Cycle 3 (n ¼ 10) mean %
FU1 (n ¼ 9) mean %
FU2 (n ¼ 4) mean %
Relapse (n ¼ 9) mean %
100 100 100 100
56* 96 130 345*
45** 58* 76* 136
41* 84 86 73
54* 61 62 167
24* 80 79 183
95 117 110 95
Diagnosisa mean %
Cycle 1 (n ¼ 9) mean %
100 100 100 100
150 91 149 113
a
The apoptosis-related protein values at diagnosis are set at 100%. For each patient, protein changes in MRD cells were calculated relative to the corresponding diagnosis level. The mean of the groups are depicted. * and ** indicate significantly different from diagnosis: *: 0.054PX0.005 and **: 0.005XP40.0001. Leukemic blasts at all measurements were identified by a low CD45 expression in combination with a LAP marker expression.
It does however suggest that external factors likely regulate expression of apoptosis-related proteins. Either AML BM stromal cells or AML blasts themselves may account for this. Pertinent to this is the observation of Milojkovic et al (Milojkovic D, Buggins AG, Devereux S, Thomas NSB, Mufti GJ. Blood 2002; 100: 556a, 2184) that AML tumor supernatants can inhibit apoptosis of malignant hematopoietic cells. Combined with our observation this might indicate that AML cells at diagnosis, in follow-up situations of refractory patients and at relapse produce apoptosis-regulating factors that affect neighbouring cells, while in a CR BM due to the very limited number of AML cells, production is ineffective. Such cells might however be affected by the BM stroma: Konopleva et al7 found that a stromal cell line was able to increase Bcl-2 expression in blasts of 5/11 AML patients. It may then be hypothesized that intact BM stroma from AML patients may induce upregulation of anti-apoptosis profiles in intact stroma, while such may be impossible in damaged stroma, the latter in fact resembling the actual situation in remission BM after chemotherapy.
A van Stijn1 A Kok1 MA van Stalborch1 MA van der Pol1 N Feller1 AH Westra1 GJ Ossenkoppele1 GJ Schuurhuis1
1 Department of Hematology, VU University Medical Center, Amsterdam, The Netherlands
References 1 San Miguel JF, Vidriales MB, Lopez-Berges C, Diaz-Mediavilla J, Gutierrez N, Canizo C et al. Early immunophenotypical evaluation of minimal residual disease in acute myeloid leukemia identifies different patient risk groups and may contribute to postinduction treatment stratification. Blood 2001; 98: 1746–1751. 2 Campos L, Rouault JP, Sabido O, Oriol P, Roubi N, Vasselon C et al. High expression of bcl-2 protein in acute myeloid leukemia cells is associated with poor response to chemotherapy. Blood 1993; 81: 3091–3096. 3 Del Poeta G, Venditti A, Del Principe MI, Maurillo L, Buccisano F, Tamburini A et al. Amount of spontaneous apoptosis detected by Bax/Bcl-2 ratio predicts outcome in acute myeloid leukemia (AML). Blood 2003; 101: 2125–2131. 4 van Stijn A, Kok A, van der Pol MA, Feller N, Roemen GM, Westra AH et al. A flow cytometric method to detect apoptosis-related protein expression in minimal residual disease in acute myeloid leukemia. Leukemia 2003; 17: 780–786. 5 van Stijn A, Kok A, van der Pol MA, Feller N, Roemen GM, Westra AH et al. Multiparameter flow cytometric quantification of apoptosis-related protein expression. Leukemia 2003; 17: 787–788. 6 van der Pol MA, Pater JM, Feller N, Westra AH, van Stijn A, Ossenkoppele GJ et al. Functional characterization of minimal residual disease for P-glycoprotein and multidrug resistance protein activity in acute myeloid leukemia. Leukemia 2001; 15: 1554–1563. 7 Konopleva M, Konoplev S, Hu W, Zaritskey AY, Afanasiev BV, Andreeff M. Stromal cells prevent apoptosis of AML cells by upregulation of anti-apoptotic proteins. Leukemia 2002; 16: 1713–1724.
Leukemia
