Letters to the Editor
or C/EBPa is functionally inactivated (for example, phosphorylation by activated FLT3.)1 These results would suggest that reduced levels of C/EBPa may be sufficient to impair myeloid differentiation, a hypothesis reinforced by our resultsFbut not observed in Cebpa þ / mouse models. In summary, our results underline the critical role of C/EBPa in human myelopoiesis, distinct to that observed in murine myelopoiesis, which correlates with its frequent deregulation in AML. We would like to emphasize that although knockdown and ectopic expression analyses in mouse systems have provided, and will certainly continue to provide, essential insights into normal hematopoiesis and leukemogenesis, the limitations of these models should not be overlooked.
Acknowledgements We thank Alan Friedman (Baltimore, MD) for providing critical constructs for these studies and for discussion. We are also indebted to Uwe Herwig, MD, and his team at the AlbertinenKrankenhaus, Hamburg, for supplying umbilical cord blood. Experiments using primary human cells were approved by the Ethics Committee of the A¨rtztekammer Hamburg. We acknowledge with special thanks the contribution of Ulla Bergholz, Arne Du¨sedau and Marion Zeigler in assisting with experiments. This study was supported by grants from the Deutsche Jose´ Carreras Leuka¨mie Stiftung (CS), the Deutsche Krebshilfe (JC and CS) and the Werner Otto Stiftung (GI). The HeinrichPette-Institut is supported by the Freie und Hansestadt Hamburg and the Bundesministerium fu¨r Gesundheit und soziale Sicherung.
B Niebuhr1,4, GB Iwanski1,4, M Schwieger1, S Roscher1, C Stocking1 and J Cammenga1,2,3 1 Division of Molecular Pathology, Heinrich-Pette-Institut fu¨r experimentelle Virologie und Immunologie, Hamburg, Germany; 2 Department of Molecular Medicine and Gene Therapy, Lund Strategic Center for Stem Cell Biology and Cell Therapy, Lund, Sweden and
3
Department of Hematology, Lund University Hospital, Lund, Sweden E-mail:
[email protected] 4 These authors contributed equally to this work.
983
References 1 Nerlov C. C/EBPa mutations in acute myeloid leukemia. Nat Rev Genet 2004; 4: 394–400. 2 Tenen D. Disruption of differentiation in human cancer: AML shows the way. Nat Rev Cancer 2003; 3: 89–101. 3 Friedman A. Transcriptional control of granulocyte and monocyte development. Oncogene 2007; 26: 6816–6828. 4 Vellenga E, Ostapovicz D, O’Rourke B, Griffin J. Effects of recombinant IL-3, GM-CSF, and G-CSF on proliferation of leukemic clonogenic cells in short-term and long-term cultures. Leukemia 1987; 1: 584–589. 5 Schwieger M, Lo¨hler J, Fischer M, Herwig U, Tenen D, Stocking C. A dominant-negative mutant of C/EBPalpha, associated with acute myeloid leukemias, inhibits differentiation of myeloid and erythroid progenitors of man but not mouse. Blood 2004; 103: 2744–2752. 6 Pabst T, Mueller B, Zhang P, Radomska H, Narravula S, Schnittger S et al. Dominant-negative mutations of CEBPA, encoding CCAAT/ enhancer binding protein-a (C/EBPa), in acute myeloid leukemia. Nat Genet 2001; 27: 263–270. 7 Kirstetter P, Schuster M, Bereshchenko O, Moore S, Dvinge H, Kurz E et al. Modeling of C/EBPalpha mutant acute myeloid leukemia reveals a common expression signature of committed myeloid leukemia-initiating cells. Cancer Cell 2008; 13: 299–310. 8 Liu H, Keefer J, Wang Q, Friedman A. Reciprocal effects of C/EBPa and PKCd on JunB expression and monocytic differentiation depend upon the C/EBPa basic region. Blood 2003; 101: 3885–3892. 9 Cammenga J, Mulloy J, Berguido F, MacGrogan D, Viale A, Nimer S. Induction of C/EBPa activity alters gene expression and differentiation of human CD34+ cells. Blood 2003; 101: 2206–2214. 10 Bjorregaard M, Jurlander J, Klausen P, Borregaard N, Cowland J. The in vivo profile of transcription factors during neutrophil differentiation in human bone marrow. Blood 2003; 101: 4322–4332.
Early in vivo changes of the transcriptome in Philadelphia chromosome-positive CD34 þ cells from patients with chronic myelogenous leukaemia following imatinib therapy
Leukemia (2009) 23, 983–985; doi:10.1038/leu.2008.337; published online 4 December 2008
The selective bcr-abl tyrosine kinase inhibitor imatinib has become the therapy of choice for patients with newly diagnosed chronic myelogenous leukaemia (CML) including those previously considered candidates for allogeneic haematopoietic stem cell transplantation. Major pathways downstream of bcr-abl are Grb2-Ras and p42/44 mitogen-activiated proteinkinase, phosphatidylinositol 3-kinase, nuclear factor-kB and the signal transducers and activators 1/5.1 Still, how bcr-abl downstream molecules are affected by bcr-abl inhibition through imatinib is not fully understood. To identify ‘early response genes’ during imatinib therapy, we looked for changes in the transcriptome of Philadelphia
chromosome-positive (Ph þ ) CD34 þ cells from patients with de novo CML following 7 days of treatment. Between November 2006 and July 2007, six patients with newly diagnosed Ph þ CML in chronic phase were included in this analysis (one female, five male; median age 61 (32–79) years; median follow-up 10.5 (6–13) months; median initial bcr-abl/G6PDH ratio 26.72 (5.52–128.54)%; median bcr-abl/ G6PDH ratio after median follow-up 0.1955 (0.003–0.77); median Hasford score 1093.40 (850–1488.5)). The study was approved by the ethics committee of the Medical Faculty of Heinrich-Heine University, Duesseldorf, and all patients gave written informed consent according to the declaration of Helsinki. Ethylenediaminetetraacetic acid blood (20 ml) was obtained before and 1 week after treatment with 400 mg imatinib daily. Samples were prepared within 2 h after collection. The Leukemia
Letters to the Editor
984 mononuclear cells were obtained by density centrifugation (Lymphoprep Axon Shield, PoC AS, Oslo, Norway). Samples yielded a median of 6 107 mononuclear cells per ml (range: 2 107–13 107) at the time of diagnosis and 6 107 mononuclear cells/ml (range: 1 107–12 107) 7 days after treatment with imatinib (Novartis, Basel, Switzerland). Peripheral blood CD34 þ cells were selected by using the ‘Human CD34 Selection Kit’ and the Robosep system (both from StemCell Technologies, Vancouver, British Columbia, Canada) according to the manufactur’s instructions. Following separation, the enriched CD34 þ cells had a median purity of 96% (range: 92–98%) as assessed by flow cytometry. Isolation of total RNA from a median 1 106 (range 3.5 105–1 107) of CD34 þ cells was performed with the RNeasy Mini Kit (Qiagen AG, Hilden, Germany). The amount of extracted RNA was quantified using the NanoDrop spectrophotometer (NanoDrop Technologies, Welmington, DE, USA). An aliquot of 120 ng RNA was used to generate biotin-labelled cRNA with MessageAmp II-Biotin Enhanced Kit (Ambion, Austin, TX, USA). Following fragmentation, labelled cRNA (5 mg) was hybridized to Affymetrix Gene Array Chips (Human Genome U133A 2.0, Affymetrix UK Ltd, High Wycombe, UK). The quality of extracted RNA and of cRNA was checked using the Agilent Bioanalyzer 2100 (Agilent Technologies, Waldbronn, Germany). Array data were analysed using dChip (http://www.hsph.harvard.edu/~cli/complab/dchip/). We performed smoothing spline normalization. Expression values were calculated applying the perfect match–mismatch difference model algorithm (dChip). We considered a gene differentially expressed if the transcript level as a multiple of the control was 41.5, with Po0.05 and a lower confidence bound 41.2. The use of lower confidence bound provides a 90% confidence that the actual change in expression as a multiple of the control is a value above the reported lower confidence bound. Hierarchical cluster analysis was performed using the correlation-based centroid-linkage algorithm (dChip). Following a median time of 0.9 months (range: 0.9–3) after starting imatinib, all six patients achieved a complete haematological remission. Thereafter, we found a median 3-log reduction (range 1.5–3) of the bcr-abl transcript after a median time of 10 months (range 6–13 months). The patients tolerated imatinib very well without any significant side effects. A supervised hierarchical cluster analysis of the CD34 þ cells obtained before and 7 days after starting imatinib provides a distinct set of genes permitting a clear discrimination between the two sets of samples. Looking at the entire 22 000 probe sets containing 14 500 genes spotted on the array, we found that the expression level of only 303 of these genes was significantly affected by treatment with imatinib (Figure 1). In particular, 183 genes were downregulated, whereas 120 genes were upregulated with a lower bound of at least 1.2-fold. These differentially expressed genes were related mainly to the following functional groups. Genes governing cell cycle and DNA replication generally had a reduced level of expression. On the other hand, genes involved in adhesive interactions showed an increased expression following the treatment with imatinib. Differentially expressed genes that we considered to be of relevance from a pathophysiological point of view are highlighted in Supplementary Tables 1 and 2 (Supplementary material) and in the following. Differential expression of eight genes was confirmed by quantitative real-time PCR (Supplementary Figure 1 and Supplementary primer sequences). FISH analysis yielded a median proportion of 87% (range 85–89%) CD34 þ cells with a typical bcr-abl-positive pattern. After 7 days of imatinib Leukemia
Figure 1 Cluster tree. Differentially expressed genes between the patient samples before imatinib (pre-Im) and after 7 days of therapy (post-Im).
treatment at 400 mg per day, we observed a similar median proportion of 85% (range: 84–87%). The reversal of bcr-abl-induced abnormal phosphorylation as a consequence of blocking the fusion protein by imatinib indicates signal transduction pathways relevant to the pathophysiology of CML. In an attempt to better understand and describe these pathways, we looked for molecular changes in the peripheral blood CD34 þ cells after 1 week of tyrosine kinase inhibitor treatment. Although there was no significant change within the first week with regard to CD34 þ cell and leukocyte count, all patients had reached a haematological remission, which later resulted in a complete cytogenetic and major molecular remission. Thus, in the first week of treatment, the molecular brakes are apparently effective in silencing the Ph þ CD34 þ progenitor cells. From the results of our gene expression analysis, we conclude that this reversal is probably explained by three major mechanisms. First, a significant downregulation of genes associated with cell cycle and proliferation, second a downregulation of genes associated with DNA replication and third with an upregulation of genes coding for proteins involved in cell migration and adhesion. Regarding genes associated with cell cycle, imatinib affected several key regulators of the different phases of the cell cycle. A significant downregulation of S-phase governing genes, such as cell division cycle 2 and 20 (CDC2, CDC20), Geminin and minichromosomal maintenance complex component 4, was observed.2 The largest difference with a 3.5-fold downregulation was observed for CDC2. For progression into the G2-phase,
Letters to the Editor
985 CDC2 and cell division cycle 45-like (CDC45L) are required, which we found to be downregulated in the imatinib-exposed Ph þ CD34 þ cells.3 Further along the cell cycle pathway towards mitosis, genes such as Cyclin B1 and B2 (CCNB1, CCNB2) are active.2 Both genes involved in the G2/M cell cycle arrest were downregulated, whereas other genes controlling this arrest, such as CDP-diacylglycerol synthase, WEE1 homologue or mitotic inhibitor kinase Mik1 were not differentially expressed. These findings suggest a slowing of the mitotic clock rather than a complete arrest. With regard to genes involved in DNA replication and repair, we found a substantial number of genes downregulated in the CD34 þ cells as a result of treatment with imatinib. The unwinding of DNA, a prerequisite for DNA replication, requires two tightly regulated complexes called topoisomerase 2 alpha and minichromosomal maintenance complex component helicase complex.4 Among the genes forming this complex, minichromosomal maintenance complex components 2, 3, 4, 6 and 7 had a significantly higher level of expression in the CD34 þ cells earlier compared with 7 days after imatinib treatment. A significant downregulation was also noted for the genes TIMELESS and TIPIN, which encode proteins supposedly interacting with the replication proteins along the replication fork.4 This reasoning is in line with our finding that one of these replication proteins, RPA3, was also downregulated in the CD34 þ cells following treatment with imatinib. Similarly, there was a decreased expression level of the replication factor activators RFC3 and RFC4 as well as for POLE2, the DNA polymerase e that is needed for the synthesis of the new DNA on the lagging strand.3,4 Finally, the gene encoding PCNA (proliferation cell nuclear antigen), a co-factor of polymerase d,5 the second enzyme for the DNA strand synthesis,4 was also downregulated. Coincident with the downregulation of DNA replication-associated genes, this group of genes, such as PCNA, RFC3, RFC4, RPA3 or POLE2, all involved in DNA repair,5 were also affected. They showed a significantly lower level in CD34 þ cells following treatment with imatinib in comparison with CD34 þ cells examined before treatment. The third group of genes affected early by imatinib relates to proteins involved in adhesion and motility. For instance, a significantly greater expression level was observed for the gene encoding the CD44 molecule that mediates binding of the CD34 þ cell to hyaluronic acid,6 a component of the bone marrow matrix. Upregulation was also observed for the gene encoding the cell surface receptor L-selectin (SELL), a lymphocyte adhesion molecule that is downregulated in bone marrow of newly diagnosed CML7 before any therapy. Other genes with a significantly higher expression level in the imatinib-exposed CD34 þ cells are IL-18 (interleukin 18) and RHOB (ras homologue gene family, member B).8 As a net result, upregulation of adhesion-associated genes may result in a reduced motility of CD34 þ cells and therefore decreased migration and trafficking in the peripheral blood. As yet, we do not have a clear picture of the sequence of molecular events that happen early during imatinib treatment. This would require repeated expression analysis in Ph þ CD34 þ cells at different time points following the start
of treatment. Although various aspects of cellular functions are closely related to each other, it is reasonable to assume that early during imatinib therapy, improved adhesion of Ph þ CD34 þ cells to matrix components of the bone marrow goes along with a decelerated cell cycle, including decreased DNA replication and mitotic activity. Subsequently, this would translate into reduced production of leukaemic stem, progenitor and ultimately mature cells.
Acknowledgements DB is supported by a research grant of the ‘Forschungskommission’ of the Medical Faculty of the Heinrich-Heine-University Duesseldorf. The work is supported by a grant of the ‘Leukaemie Liga, Duesseldorf’ and by Novartis pharma. Thanks to Sabrina Pechtel and Gernot Roeder for preparing the Affymetrix gene arrays. Contribution: DB: experiments and manuscript preparation; AC and IB: experiments and statistical analysis; RK: concept of study, manuscript preparation; NG and RH: manuscript preparation, concept of study; FN: experiments, manuscript preparation and concept of study.
D Bruennert, A Czibere, I Bruns, R Kronenwett, N Gattermann, R Haas and F Neumann Department of Haematology, Oncology and Clinical Immunology, Heinrich-Heine-University Duesseldorf, Duesseldorf, Germany E-mail:
[email protected]
References 1 Kawauchi K, Ogasawara T, Yasuyama M, Ohkawa S-I. Involvement of Akt kinase in the action of STI571 on chronic myelogenous leukemia cells. Blood Cells Mol Dis 2003; 31: 11–17. 2 Singhal S, Amin K, Kruklitis R, DeLong P, Friscia M, Litzky L et al. Alterations in cell cycle genes in early stage lung adenocarcinoma identified by expression profiling. Cancer Biol Ther 2003; 2: 291–298. 3 Bauerschmidt C, Pollok S, Kremmer E, Nasheuer H-P, Grosse F. Interactions of human Cdc45 with the Mcm2–7 complex, the GINS complex, and DNA polymerases delta and epsilon and during S phase. Genes Cells 2007; 12: 745–758. 4 Gotter AL, Suppa C, Emanuel BS. Mammalian TIMELESS and Tipin are evolutionarily conserved replication fork-associated factors. J Mol Biol 2007; 366: 36–52. 5 Venclovas C, Colvin ME, Thelen MP. Molecular modeling-based analysis of interactions in the RFC-dependent clamp-loading process. Protein Sci 2002; 11: 2403–2416. 6 Gotte M, Yip GW. Heparanase, hyaluronan, and CD44 in cancers: a breast carcinoma perspective. Cancer Res 2006; 66: 10233–10237. 7 Diaz-Blanco E, Bruns I, Neumann F, Fischer JC, Graef T, Rosskopf M et al. Molecular signature of CD34+ hematopoietic stem and progenitor cells of patients with CML in chronic phase. Leukemia 2007; 21: 494–504. 8 Morel JCM, Park CC, Woods JM, Koch AE. A novel role for interleukin-18 in adhesion molecule induction through NFkappa B and phosphatidylinositol (PI) 3-kinase-dependent signal transduction pathways. J Biol Chem 2001; 276: 37069–37075.
Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)
Leukemia
