Letters to the Editor
1316 mutated allele. In the case of a mutation, the PCR generates an additional 203 bp PCR product.1 The second strategy consists of nested PCR, and subsequent restriction enzyme digestion with BsaXI (New England Biolabs, Hitchin, UK). In case of mutated DNA, the PCR product remains undigested, whereas unmutated DNA results in two fragments of 170 and 203 bp in size.4 None of the MDS/AML cases showed evidence of the JAK2 V617F mutation, regardless of the presence or absence of fibrosis. Interestingly, the mutation was detected in three of four cases (75%) diagnosed as MDS/MPD. In contrast to the MDS/AML with fibrosis, these four patients showed some clinical features of chronic myeloproliferative disease, including splenomegaly, leukocytosis and/or thrombocytosis, as well as, morphological features of CMPD in the trephine biopsy. One of the positive cases was classified as atypical chronic myelogenous leukemia negative for BCR/ABL fusion transcripts. In contrast to the results from Ohyashiki et al., our results inidicate that the JAK2 V617F mutation is exceedingly rare in bona fide MDS or de novo AML, regardless of the presence of fibrosis. Furthermore, the identification of the JAK2 mutation in MDS/MPD cases indicates that these cases are more closely related to classical CMPD. This interpretation is supported by other reports demonstrating the occurrence of the JAK2 mutation in MDS/MPD, such as chronic myelomonocytic leukemia, although the latter entity shows a lower incidence of JAK2 mutations than CMPD.5,6,8,9 We believe that the analysis of the JAK2 V617F mutation, in difficult-to-classify cases, will help to clarify the borderline between MDS, on one hand, and atypical CMPD and MDS/MPD on the other hand. The better understanding of these entities, potentially may lead in the near future to identify new therapeutic options for these patients.
M Kremer1, T Horn1, T Dechow2, A Tzankov3, L Quintanilla-Martı´nez4 and F Fend1 1 Institute of Pathology, Technical University Munich, Munich, Germany; 2 Department of Hematology and Oncology, Technical University Munich, Munich, Germany;
3
Institute of Pathology, University of Innsbruck, Innsbruck, Austria and 4 Institute of Pathology, GSF Research Center for Environment and Health, Oberschleissheim, Germany E-mail:
[email protected] References 1 Baxter EJ, Scott LM, Campbell PJ, East C, Fourouclas N, Swanton S et al. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet 2005; 365: 1054–1061. 2 James C, Ugo V, Casadevall N, Constantinescu S, Vainchenker W. A JAK2 mutation in myeloproliferative disorders: pathogenesis and therapeutic and scientific prospects. Trends Mol Med 2005; 11: 546–554. 3 Kralovics R, Passamonti F, Buser AS, Teo SS, Tiedt R, Passweg JR et al. A gain-of-function mutation of JAK2 in myeloproliferative disorders. N Engl J Med 2005; 352: 1779–1790. 4 Levine RL, Wadleigh M, Cools J, Ebert BL, Wernig G, Huntly BJ et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell 2005; 7: 387–397. 5 Steensma DP, Dewald GW, Lasho TL, Powell HL, McClure RF, Levine RL et al. The JAK2 V617F activating tyrosine kinase mutation is an infrequent event in both ‘atypical’ myeloproliferative disorders and myelodysplastic syndromes. Blood 2005; 106: 1207–1209. 6 Jelinek J, Oki Y, Gharibyan V, Bueso-Ramos C, Prchal JT, Verstovsek S et al. JAK2 mutation 1849G 4T is rare in acute leukemias but can be found in CMML, Philadelphia-chromosome negative CML and megakaryocytic leukemia. Blood 2005; 106: 3370–3373. 7 Ohyashiki K, Aota Y, Akahane D, Gotoh A, Miyazawa K, Kimura Y et al. The JAK2 V617F tyrosine kinase mutation in myelodysplastic syndromes (MDS) developing myelofibrosis indicates the myeloproliferative nature in a subset of MDS patients. Leukemia 2005; 19: 2359–2360. 8 Levine RL, Loriaux M, Huntly BJ, Loh M, Beran M, Stoffregen E et al. The JAK2V617F activating mutation occurs in chronic myelomonocytic leukemia and acute myeloid leukemia, but not in acute lymphoblastic leukemia or chronic lymphocytic leukemia. Blood 2005; 106: 3377–3379. 9 Johan M, Goodeve A, Bowen D, Frew M, Reilly J. JAK2 V617F Mutation is uncommon in chronic myelomonocytic leukemia. Br J Haematol 2005; 130: 968.
Bcl2 phosphorylation and active PKC a are associated with poor survival in AML
Leukemia (2006) 20, 1316–1319. doi:10.1038/sj.leu.2404248; published online 27 April 2006
Bcl2 was identified as the cellular oncogene product associated with the t(14,18) translocation commonly seen in B-cell lymphomas and is the founding member of a family of proteins that regulate apoptosis.1 Bcl2 protects cells from a wide range of stress challenges and thus high Bcl2 levels would be predicted to favor chemoresistance.1,2 However, Bcl2 expression levels alone do not always correlate with poor prognosis (e.g. some acute myeloid leukemias (AML) patient subsets, pediatric acute lymphoblastic leukemia patients).2 As Bcl2 expression alone cannot explain its full and potent antiapoptotic activity, other regulatory mechanisms must be involved. The phosphorylation status of Bcl2 influences its function.3–6 Growth agonist-induced phosphorylation of Bcl2 at serine 70 is required for Bcl2’s full and potent antiapoptotic function.4–6 Leukemia
PKC a and extracellular signal-related kinase (ERK) have been identified as Bcl2 kinases that promote survival.5,6 In hematopoietic cell lines, overexpression of exogenous kinase or treatment of cells with kinase agonist (i.e. bryostatin for PKC a and aurintricarboxylic acid for ERK) results in enhanced Bcl2 phosphorylation and increased resistance to apoptosis on induction of stress challenges (i.e. IL-3 withdrawal or treatment of cells with chemotherapeutic agents).5–7 Consequently, the phosphorylation status of Bcl2 in AML blast cells or any other human malignancy is unknown. Furthermore, the consequence of phosphorylated Bcl2 on the survival of AML blast cells has yet to be determined. In the present study, we investigate Bcl2 phosphorylation status using metabolic radiolabeling and activation status of PKC a and ERK in blast cells from AML patients and examine their effect on patient survival. We find that Bcl2 is phosphorylated in nearly half the AML blast cells tested. Furthermore, Bcl2 is always phosphorylated in AML blast cells with activated PKC a and ERK but never in cells that lack
Letters to the Editor
1317 Table 1 Patient #
Clinical data for patients Blast (%)
pBcl2
pERK
pPKC a
1
54
+
+
2 3 4 5 6 7 8 9 10 11 12 13 14 15
88 90 93 53 95 91 93 71 90 81 95 (PB) 93 (BM) 96 94 78
+ + + + + +
+ + ND + ND +
+ + + ND + + ND + +
FAB
Cytogenetics
Status
Secondary AML from MDS RAEBT Unknown M1 M4 M4 Unknown M1 M4 M2 M2 M1 M4 Unknown M1 M4
Diploid
At diagnosis
7 [19] 7[9], del(7)(q11.2)[10] Diploid Diploid t(2;7),t(9;22),t(15;17),del(20)[5] t(9;11)[14], del(7),t(9;11)[4], +8,t(9;11)[1] del(9)(q22q33)[19] +8[4], diploid [16] t(1;16)[18], del(9)[1], diploid [1] inv(16)[15], inv(16),+22[1], diploid [4] +X,+8,del(11),+13,+19,+20,+21[20] Diploid +8 [9], diploid [3] Diploid
Resistant Relapsed refractory At diagnosis At diagnosis Relapsed refractory Relapsed refractory Relapse Relapse Relapsed refractory At diagnosis Relapse Relapse Relapsed refractory Relapse
BM, bone marrow; PB, peripheral blood; FAB, French–American–British; MDS, myelodysplastic syndrome; RAEBT, refractory anemia with excess of blasts in transformation; ND, not done.
both activated kinases. Finally, AML patients with blast cells expressing phosphorylated Bcl2 exhibit shorter overall survival (particularly when PKC a was active) compared to patients with blast cells expressing unphosphorylated Bcl2. This finding is consistent with our previous study that determined that expression levels of proteins (i.e. Bax and PKC a) that affect the functional status of Bcl2 modify the prognostic impact of Bcl2.8 Commercially available phospho-Bcl2 antibodies have significant crossreactivity with nonphosphorylated Bcl2 protein and thus are not specific enough for analysis of Bcl2 phosphorylation status (data not shown). In studies on AML cell lines, we used metabolic radio-labeling to examine Bcl2 phosphorylation.5 In the present study, we utilized this assay to determine Bcl2 phosphorylation status in primary blast cells derived from bone marrow of 15 AML patients. Bone marrow or peripheral blood was obtained for in vitro studies from patients with newly diagnosed or recurrent AML with high (450%) blast count. Informed consent was obtained following institutional guidelines. Clinical data for AML patients are presented in Table 1. Mononuclear cells were separated by Ficoll–Hypaque density-gradient centrifugation. Samples were processed within 30 min after they were obtained. Cells (20–40 106) were labeled with [32P]orthophosphoric acid for 2 h and Bcl2 was immunoprecipitated using an antibody that was previously described.5 OCI-AML3 cells (kind gift of Dr Mark Minden, Ontario Cancer Institute, Toronto, ON, Canada) were used as a positive control for Bcl2 phosphorylation and were included in each experimental set. In one case (patient 12), we were able to perform experiments on both bone marrow and peripheral blood from a single patient. As shown in Table 1, Bcl2 is phosphorylated in approximately half of AML blast cells studied (i.e. seven of 15). A representative set is shown in Figure 1a. Bcl2 is phosphorylated in samples 10, 11, and 12 but not sample 8. These findings represent the first identification of phosphorylated Bcl2 in primary AML blast cells. With Bcl2 phosphorylation confirmed as a physiologic phenomenon, we next investigated which Bcl2 kinases might be involved. PKC a and ERK expression and phosphorylation were examined by Western analysis in 13 of the 15 AML patient samples (samples 9 and 12 are not included due to limited number of isolated cells). Activation of ERK and PKC a varied
Figure 1 Bcl-2 phosphorylation, extracellular signal-related kinase (ERK) and PKC a phosphorylation status in primary acute myeloid leukemias (AML) samples. (a) Representative group of samples (number 8, 10, 11, and 12) from the same experimental set show the result of metabolic 32P-labeling of immunoprecipitated Bcl-2. For samples 8, 10, 11, and 12, bone marrow (BM) was isolated. AML blast cells were extracted and incubated with the isotope for 2 h. Peripheral blood (PB) from patient 12 was similarly treated. Cells then were lysed and Bcl2 immunoprecipitated, and phosphorylation determined as previously described.5 (b) Blast cells were lysed and protein extracts were separated by Sodium dodecyl sulphate–polyacrylamide gel electrophoresis; Western blot analysis was performed with antibodies against phospho-ERK (Cell Signaling, Beverly, MA, USA), ERK (Cell Signaling), Bcl2 (Dako, Carpenteria, CA, USA), phospho-PKC a (Cell Signaling), PKC a (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and GAPDH (Chemicon International, Temecula, CA, USA). A representative group of samples is shown.
(Table 1; Figure 1b). As presented in Table 1, four of 13 samples were devoid of active kinase, two of 13 samples displayed active ERK alone, four of 13 samples displayed active PKC a alone, and three of 13 samples displayed both active PKC a and ERK. Every phospho-Bcl2 positive sample exhibited phosphorylated ERK and/or PKC a (Table 1). Protein kinase profiles from representative samples 8, 10, and 11 are shown in Figure 1b. Bcl2 was not phosphorylated in sample 8. Interestingly, neither Leukemia
Letters to the Editor
1318 Table 2
Activation status of Bcl2 kinase and Bcl2 phosphorylation status in AML blast cells # samples with no active kinase
# samples with one active kinase (either PKC a or ERK)
# samples with both active kinase
4 0
3 3
0 3
# samples with pBcl2 negative # samples with pBcl2 positive
AML, acute myeloid leukemias; ERK, extracellular signal-related kinase.
Table 3
Frequency counts for the number of deaths, proportions and median survival estimates for all variables and pairwise combinations
Variable
# deaths/N (%)
Median survival (days)
Univariate P-value
Hazard ratio
pBCL2 Negative Positive
6/8 (75%) 6/7 (86%)
197.5 79.0
F 0.0628
1.00 3.783
F 0.93–15.4
Perk Negative Positive
6/8 (75%) 4/5 (80%)
158.0 142.0
F 0.1723
1.00 5.46
F 0.48–62.5
pPKC a Negative Positive
4/6 (67%) 6/7 (86%)
213.5 84.0
F 0.0561
1.00 4.363
F 0.96–19.8
Combination of pBCL2 and pERK Negative pBCL2 and negative pERK Positive pBCL2 and positive pERK Negative pBCL2 and positive pERK Positive pBCL2 and negative pERK
4/6 3/4 1/1 2/2
(67%) (75%) (100%) (100%)
202.0 110.5 209.0 73.5
F 0.0926 0.2256 0.0912
F 9.440 9.037 5.229
F 0.68–129.23 0.26–317.88 0.77–35.64
Combination of pBCL2 and pPKC a Negative pBCL2 and negative pPKC a Positive pBCL2 and positive pPKC a Negative pBCL2 and positive pPKC a Positive pBCL2 and negative pPKC a
3/5 4/5 2/2 1/1
(60%) (80%) (100%) (100%)
218.0 79.0 101.0 142.0
F 0.0497 0.0549 0.1550
F 7.995 9.691 13.017
F 1.00–63.76 0.95–98.48 0.38–447.45
ERK nor PKC a were phosphorylated in this sample (Table 1; Figure 1b). Sample 10 represents an example where Bcl2 is phosphorylated but only one kinase is active (i.e. PKC a). Sample 11 represents the set of samples where both ERK and PKC a were active (Table 1; Figure 1a). While some samples that lacked phosphorylated Bcl2 exhibited active PKC a (i.e. samples 6 and 7) or active ERK (i.e. sample 4), Bcl2 was phosphorylated in all cases where both kinases are active (i.e. samples 5, 11, and 15; Table 1). The relationship between kinase activation and Bcl-2 was found to be statistically significant (P ¼ 0.046) using Pagano and Halvorsen’s variation of Fisher’s Exact Test.9 As shown in Table 2, all samples with no active Bcl2 kinase displayed unphosphorylated Bcl2. Conversely, Bcl2 was phosphorylated in all samples where both Bcl2 kinases were activated (Table 2). This finding suggests a close relationship exists between activation of PKC a and ERK and Bcl2 phosphorylation status in AML blast cells. For sample 12 both bone marrow and peripheral blood were analyzed (Figure 1a). Interestingly, Bcl2 phosphorylation levels were greater in bone marrow derived blast cells compared to blasts in peripheral blood. While this observation is for a single patient, it is tempting to speculate that perhaps Bcl2 phosphorylation status in blast cells might be affected by interactions with cells in the bone marrow stroma. Experiments to address this question are ongoing. We next evaluated the effect of phospho-Bcl-2, phospho-PKC a, phospho-ERK and their combination on median survival. Frequency counts for the number of deaths, proportions and median survival estimates, hazard ratios and 95% confidence Leukemia
95% CI for HR
intervals (CI) are reported for all potential risk factors and pairwise combinations of risk factors in Table 3. The time to death or censoring was computed in days since treatment for each patient. Survival was censored at the date of the last follow-up if death was not observed. Survival probabilities were estimated nonparametrically using Kaplan–Meier’s product limit method. Cox proportional hazards regression models were used to model survival as a function of each potential risk factor after adjusting for the patient being new or having undergone treatment previously. All statistical analyses were performed using SAS 9.1 (SAS Institute Inc., Cary, NC, USA). Acute myeloid leukemias patients whose blasts express phosphorylated Bcl-2 may have shorter overall survival (N ¼ 7; 79 days) compared to patients with no phosphorylated Bcl-2 (N ¼ 8; 197.5 days), although this difference is not statistically significant (hazard ratio ¼ 3.783; 95% CI, 0.93–15.4; P ¼ 0.0628). Survival of AML patients with active PKC a was shorter compared to patients with no phosphorylated PKC (hazard ratio ¼ 4.363; 95% CI, 0.96–19.8; P ¼ 0.0561). Extracellular signal-related kinase activation was not prognostic (P ¼ 0.1723). Patients expressing both, active PKC a and phosphorylated Bcl2 have shorter overall survival compared to patients with no active PKC a and unphosphorylated Bcl2 (hazard ratio ¼ 7.995; 95% CI, 1.00– 63.76; P ¼ 0.0497). This finding supports a role for PKC a in prosurvival Bcl2 phosphorylation.5,8 Previously, we determined that PKC a modulates the ability of Bcl2 to serve as a prognostic factor in AML.8 However, in that study PKC a phosphorylation of Bcl2 was implied as no direct analysis of Bcl2 phosphorylation was performed. The current data supports the ability of PKC a to
Letters to the Editor
1319 modulate Bcl2 phosphorylation and suggest that PKC a-mediated Bcl2 phosphorylation may be relevant to chemoresistance in AML.
S Kurinna1, M Konopleva2, SL Palla3, W Chen2, S Kornblau2, R Contractor2, X Deng4, WS May4, M Andreeff2 and PP Ruvolo1,2 1 Division of Cell Signaling, Institute of Molecular Medicine, University of Texas Health Science Center, Houston, TX, USA; 2 Department of Blood and Marrow Transplantation, MD Anderson Cancer Center, Houston, TX, USA; 3 Department of Biostatistics and Applied Mathematics, MD Anderson Cancer Center, Houston, TX, USA and 4 Shands Cancer Center, University of Florida, Gainesville, FL, USA E-mail:
[email protected]
References 1 Yang E, Korsmeyer SJ. Molecular thanatopsis: a discourse on the Bcl2 family and cell death. Blood 1996; 88: 386–401.
2 Deng X, Kornblau SM, Ruvolo PP, May WS. Regulation of Bcl2 phosphorylation and potential significance for leukemic cell chemoresistance. J Natl Cancer Inst 2000; Monograph 28: 30–37. 3 Haldar S, Jena N, Croce CM. Inactivation of Bcl2 by phosphorylation. Proc Natl Acad Sci USA 1995; 92: 4507–4511. 4 Ito T, Deng X, Carr BK, May WS. Bcl2 phosphorylation required for anti-apoptosis function. J Biol Chem 1997; 272: 11671–11673. 5 Ruvolo PP, Deng X, Carr BK, May WS. A functional role for mitochondrial PKC a in Bcl2 phosphorylation and suppression of apoptosis. J Biol Chem 1998; 273: 25436–25442. 6 Deng X, Ruvolo P, Carr B, May WS. Survival function of ERK1/2 as IL-3-activated staurosporine-resistant Bcl2 kinases. Proc Natl Acad Sci USA 2000; 97: 1578–1583. 7 Konopleva M, Tsao T, Ruvolo P, Stiouf I, Estrov Z, Leysath CE et al. Novel triterpenoid CDDO-Me is a potent inducer of apoptosis and differentiation in acute myelogenous leukemia. Blood 2002; 99: 326–335. 8 Kornblau SM, Vu HT, Ruvolo P, Estrov Z, O’Brien S, Cortes J et al. Bax and PKCa modulate the prognostic impact of Bcl2 expression in acute myelogenous leukemia. Clinical Cancer Res 2000; 6: 1401–1409. 9 Pagano M, Halvorsen K. An algorithm for finding the exact significance levels of r x c contingency tables. J Am Stat Assoc 1981; 76: 731–934.
The JAK2 V617F mutation identifies a subgroup of MDS patients with isolated deletion 5q and a proliferative bone marrow
Leukemia (2006) 20, 1319–1321. doi:10.1038/sj.leu.2404215; published online 13 April 2006
The detection of JAK2 V617F somatic mutation has greatly enhanced our understanding of the pathogenesis of the bcr/ablnegative chronic myeloproliferative disorders (MPD). An increased sensitivity to erythropoietin and growth factor independence is reported in the presence of the mutation.1–3 Four independent groups recently report JAK2 V617F mutation in the majority of patients with polycythaemia vera and up to 50% of essential thrombocythaemia and idiopathic myelofibrosis.1–4 An increased incidence of thrombosis, haemorrhage and fibrotic transformation in the presence of the mutant allele is reported.3 Larger studies are however required to elicit the true prognostic significance of the mutation. The finding of JAK2 V617F mutation outside of the classical MPDs is uncommon with reports of low incidence in chronic myelomonocytic leukaemia, atypical chronic myeloid leukaemia, hypereosinophilic syndrome and chronic neutrophilic leukaemia.5 The myelodysplastic syndromes (MDS) are a heterogeneous group of clonal haematopoietic stem cell disorders characterised by peripheral blood cytopenias, ineffective erythropoiesis and increased apoptosis. The 5q syndrome is a subgroup of MDS characterised by an interstitial deletion of the long arm of chromosome 5(q31–q33) with macrocytic anaemia, normal or elevated platelet count, hypolobated megakaryocytes and associated with a favourable prognosis. However, cases presenting with 5q deletion and marked elevation of the platelet count in association with a hypercellular bone marrow display characteristics more suggestive of an overlap syndrome (MDS/ MPD). We analysed 97 patients from six European centres known to have a diagnosis of MDS and deletion, 5q abnormality for the presence of JAK2 V617F mutation. Isolated deletion of 5q was
detected in 81/97 cases, whereas additional cytogenetic abnormalities were noted in 16/97. The diagnosis of MDS was based on the World Health Organization Classification. A summary of patient characteristics is outlined in Table 1. Ethical approval was obtained before study commencement. Samples for analysis were obtained from archived bone marrow aspirate slides, archived cytogenetic samples or peripheral blood. The mutant JAK2 allele was detected using an allele-specificpolymerase chain reaction (AS-PCR).4,6,7 The presence of the mutation and ratio of mutant to wild-type JAK2 allele was confirmed in 2/6 mutant cases using pyrosequencing. Polymerase chain reaction products were generated using AS-PCR primer sequences. Sequences were read from a reverse sequencing primer 50 -TCTCGTCTCCACAGA-30 . Pyrosequencing reactions were run on a Biotage PSQ HS 96 pyrosequencer. In a patient with JAK2 mutation, cells from bone aspirate were subjected to progenitor cell culture (CFU-GM- and BFU-Ederived colonies) or subfractionated into CD34 þ ve cells, followed by molecular analysis. Peripheral blood from the same case was also subfractionated into CD14, CD15, CD3 and CD19 þ cells using antigen-specific microbeads followed by selection using the AutoMacs cell separator Milteny. In the case of progenitor cultures, PCR was performed directly on the cells without prior DNA purification. DNA was extracted from subfractionated cells using Charge Switch reagents (Invitrogen, Inchinnan Business Park, Fountain Drive, Paisley, UK). All DNA extraction and cell separation kits were used according to the manufacturer’s instructions. Samples were processed in accordance with standard cytogenetic methods. CD34 þ cells were probed for 5q using the Vysis LSI EGR1 (5q31), spectrum orange D5S23 and spectrum green D5S721 probe, according to the manufacturer’s instructions (Abbot Laboratories). 100K Affymetrix SNP analysis of DNA extracted from CD34-positive cells was performed using Leukemia
