Disruption of C/EBP Function in Acute Myeloid Leukemia
http://www.100md.com
《新英格兰医药杂志》
In acute myeloid leukemia (AML), the most common type of acute leukemia in adults, there is clonal expansion and arrested maturation of myeloid precursor cells in the bone marrow, frequently resulting in granulocytopenia, thrombocytopenia, and anemia. The genetic abnormalities underlying these events fall into distinct classes. One class arises from mutations that affect lineage-specific transcription factors involved in hematopoietic-cell differentiation.1,2 These mutations, which disrupt cellular differentiation, are crucial events in the pathogenesis of a subgroup of cases of AML. Another class of genetic events comprises mutations that affect intracellular signaling molecules in a way that enhances cell proliferation and survival without much effect on differentiation. Examples of this class are the gain-of-function mutations in genes encoding tyrosine kinases in hematopoietic cells. There is experimental and epidemiologic evidence that these two types of mutations have a joint role in the pathogenesis of AML.
Among the transcription factors with a role in hematopoiesis is the CCAAT enhancer binding protein (C/EBP), also known as CEBPA, a key molecule in the mediation of lineage specification and the differentiation of multipotent myeloid progenitors into mature neutrophils through the activation of myeloid-specific genes (see diagram). The differentiation-inducing and antiproliferative functions of C/EBP require the integrity of the N-terminal transcriptional-activation domains of the molecule, as well as the basic-region–leucine zipper motif in the C-terminal region that mediates protein–protein interactions, DNA binding, and dimerization of C/EBP (see diagram).
Role of Transcription Factors in Normal Myeloid Development and AML.
The differentiation of hematopoietic progenitor cells into particular lineages depends on the coordinated activity of specific transcription factors. The role of three crucial transcription factors, PU1, GATA1, and C/EBP, in the commitment and maturation of myeloid progenitors is shown in Panel A. Down-regulation of PU1 and up-regulation of GATA1 are required for the differentiation of common myeloid progenitors into megakaryocyte–erythroid progenitors. Increases in the relative level of expression of GATA1 also promote the differentiation of megakaryocyte–erythroid progenitors into erythrocytes. Up-regulation of C/EBP initiates the transition from common myeloid progenitors to granulocyte–macrophage progenitors. Further downstream, C/EBP induces granulocytic development and blocks monocytic development from bipotential granulocyte–macrophage progenitors, presumably through the inhibition of PU1 function. Mutational inactivation of PU1, GATA1, and C/EBP has been observed in patients with AML. Characterized functional domains of the C/EBP protein and the two most common types of C/EBP mutations are shown in Panel B. Transactivation domains 1 and 2 (TAD1 and TAD2) mediate interactions with the transcriptional machinery. The basic-region–leucine zipper domain (bZIP) mediates DNA binding as well as homodimerization or heterodimerization with other C/EBP proteins. N-terminal nonsense mutations abolish expression of the full-length 42-kD C/EBP protein and up-regulate the formation of a 30-kD isoform with dominant negative properties, resulting in a substantial reduction in C/EBP function. C-terminal missense mutations result in C/EBP proteins with decreased DNA-binding or dimerization activity.
The observation that C/EBP-deficient mice lack mature granulocytes raises the possibility that mutations in the gene encoding C/EBP (CEBPA) could contribute to the block in differentiation of myeloid progenitor cells in AML. This hypothesis is supported by the fact that two types of heterozygous CEBPA mutations have been identified in sporadic AML. First, nonsense mutations can affect the N-terminal region of the molecule. They prevent expression of the full-length 42-kD C/EBP protein, thereby eliminating upstream initiation sites of the transcription factor and up-regulating the formation of a truncated 30-kD isoform with dominant negative properties. Second, in-frame mutations in the basic-region–leucine zipper domain result in C/EBP proteins with decreased DNA-binding or dimerization activity. In addition, point mutations affecting specific amino acids in the basic region are associated with a loss of C/EBP function.
In this issue of the Journal, Smith and coworkers (pages 2403–2407) describe a family with three members affected by AML associated with an identical N-terminal CEBPA mutation that was predicted to result in a loss of C/EBP function. Several points are particularly relevant to this work.
First, until now, only one genetic abnormality, involving heterozygous mutations in the gene encoding runt-related transcription factor 1 (RUNX1), has been reported in familial AML outside the setting of a syndrome such as trisomy 21.3 Heterozygous germ-line mutations in RUNX1 have been associated with the syndrome called familial platelet disorder with predisposition to AML, and several lines of evidence argue that the susceptibility to leukemia in affected members of families with this syndrome results from haploinsufficiency of RUNX1 (i.e., the RUNX1 protein produced by the remaining normal gene is not sufficient for normal function of the protein).
The heterozygous CEBPA mutation in the family studied by Smith et al. is predicted to result in the 30-kD isoform of C/EBP that, at least in vitro, can inhibit the remaining wild-type protein in a dominant negative fashion. This possibility would indicate that a substantial reduction in C/EBP function, rather than haploinsufficiency, confers a predisposition to AML in these cases.
Second, it has been found that the inactivating CEBPA mutation is associated with AML characterized by a long latency period (10 to 30 years), suggesting that additional genetic lesions are contributing factors — a likelihood that would support the concept that different mutations act jointly in the pathogenesis of AML. Despite extensive analysis of other AML-related genes, however, the investigators failed to identify any mutations in them. For this reason, the mechanism underlying the accumulation of immature myeloid precursors in the presence of a disabling mutation in CEBPA appears to be relatively straightforward, but the causal relationship between the mutation and the development of overt leukemia is unclear. One possibility is that carriers of a CEBPA mutation have a large population of poorly differentiated myeloid cells associated with an increased risk of a "second genetic hit" that would lead to AML.
Third, and perhaps most exciting, the observations of Smith et al. confirm and extend the clinical observation that in sporadic AML, mutant CEBPA predicts a favorable outcome in younger patients whose cytogenetic characteristics are associated with an intermediate level of risk.4 The case of Patient II-3 in the family studied by Smith et al., who has remained in complete remission for almost four decades despite having received therapy that would be considered inadequate by current standards, highlights the question of whether patients with AML who carry a CEBPA mutation are being overtreated when they receive the multiple courses of intensive chemotherapy or high-dose therapy followed by hematopoietic stem-cell transplantation that is given to most patients with AML. In the absence of additional data on function or findings from animal models, any hypothesis regarding the molecular basis of the apparent chemoresponsiveness of AML in patients with a CEBPA mutation must remain highly speculative. The same holds true for the core-binding-factor leukemias, characterized by the presence of t(8;21)(q22;q22) or inv(16)(p13q22) mutations, in which the mechanism of sensitivity to high-dose cytarabine or other agents remains elusive. We speculate that leukemias involving CEBPA mutations have a good prognosis because they are relatively "simple" diseases affecting patients with few genetic abnormalities; most of these patients have a normal karyotype, and it appears that CEBPA mutations, particularly N-terminal nonsense mutations, only rarely occur in combination with other genetic abnormalities.
Studies of C/EBP and other transcription factors that regulate hematopoiesis have led to a greater understanding of myeloid leukemogenesis and to the development of a molecularly based classification. However, the most challenging task for future research will be to develop therapies that target these specific genetic pathways.
Source Information
From the Department of Internal Medicine III, University Hospital Ulm, Ulm, Germany.
References
Kelly LM, Gilliland DG. Genetics of myeloid leukemias. Annu Rev Genomics Hum Genet 2002;3:179-198.
Tenen DG. Disruption of differentiation in human cancer: AML shows the way. Nat Rev Cancer 2003;3:89-101.
Song WJ, Sullivan MG, Legare RD, et al. Haploinsufficiency of CBFA2 causes familial thrombocytopenia with propensity to develop acute myelogenous leukaemia. Nat Genet 1999;23:166-175.
Fr?hling S, Schlenk RF, Stolze I, et al. CEBPA mutations in younger adults with acute myeloid leukemia and normal cytogenetics: prognostic relevance and analysis of cooperating mutations. J Clin Oncol 2004;22:624-633.(Stefan Fr?hling, M.D., an)
Among the transcription factors with a role in hematopoiesis is the CCAAT enhancer binding protein (C/EBP), also known as CEBPA, a key molecule in the mediation of lineage specification and the differentiation of multipotent myeloid progenitors into mature neutrophils through the activation of myeloid-specific genes (see diagram). The differentiation-inducing and antiproliferative functions of C/EBP require the integrity of the N-terminal transcriptional-activation domains of the molecule, as well as the basic-region–leucine zipper motif in the C-terminal region that mediates protein–protein interactions, DNA binding, and dimerization of C/EBP (see diagram).
Role of Transcription Factors in Normal Myeloid Development and AML.
The differentiation of hematopoietic progenitor cells into particular lineages depends on the coordinated activity of specific transcription factors. The role of three crucial transcription factors, PU1, GATA1, and C/EBP, in the commitment and maturation of myeloid progenitors is shown in Panel A. Down-regulation of PU1 and up-regulation of GATA1 are required for the differentiation of common myeloid progenitors into megakaryocyte–erythroid progenitors. Increases in the relative level of expression of GATA1 also promote the differentiation of megakaryocyte–erythroid progenitors into erythrocytes. Up-regulation of C/EBP initiates the transition from common myeloid progenitors to granulocyte–macrophage progenitors. Further downstream, C/EBP induces granulocytic development and blocks monocytic development from bipotential granulocyte–macrophage progenitors, presumably through the inhibition of PU1 function. Mutational inactivation of PU1, GATA1, and C/EBP has been observed in patients with AML. Characterized functional domains of the C/EBP protein and the two most common types of C/EBP mutations are shown in Panel B. Transactivation domains 1 and 2 (TAD1 and TAD2) mediate interactions with the transcriptional machinery. The basic-region–leucine zipper domain (bZIP) mediates DNA binding as well as homodimerization or heterodimerization with other C/EBP proteins. N-terminal nonsense mutations abolish expression of the full-length 42-kD C/EBP protein and up-regulate the formation of a 30-kD isoform with dominant negative properties, resulting in a substantial reduction in C/EBP function. C-terminal missense mutations result in C/EBP proteins with decreased DNA-binding or dimerization activity.
The observation that C/EBP-deficient mice lack mature granulocytes raises the possibility that mutations in the gene encoding C/EBP (CEBPA) could contribute to the block in differentiation of myeloid progenitor cells in AML. This hypothesis is supported by the fact that two types of heterozygous CEBPA mutations have been identified in sporadic AML. First, nonsense mutations can affect the N-terminal region of the molecule. They prevent expression of the full-length 42-kD C/EBP protein, thereby eliminating upstream initiation sites of the transcription factor and up-regulating the formation of a truncated 30-kD isoform with dominant negative properties. Second, in-frame mutations in the basic-region–leucine zipper domain result in C/EBP proteins with decreased DNA-binding or dimerization activity. In addition, point mutations affecting specific amino acids in the basic region are associated with a loss of C/EBP function.
In this issue of the Journal, Smith and coworkers (pages 2403–2407) describe a family with three members affected by AML associated with an identical N-terminal CEBPA mutation that was predicted to result in a loss of C/EBP function. Several points are particularly relevant to this work.
First, until now, only one genetic abnormality, involving heterozygous mutations in the gene encoding runt-related transcription factor 1 (RUNX1), has been reported in familial AML outside the setting of a syndrome such as trisomy 21.3 Heterozygous germ-line mutations in RUNX1 have been associated with the syndrome called familial platelet disorder with predisposition to AML, and several lines of evidence argue that the susceptibility to leukemia in affected members of families with this syndrome results from haploinsufficiency of RUNX1 (i.e., the RUNX1 protein produced by the remaining normal gene is not sufficient for normal function of the protein).
The heterozygous CEBPA mutation in the family studied by Smith et al. is predicted to result in the 30-kD isoform of C/EBP that, at least in vitro, can inhibit the remaining wild-type protein in a dominant negative fashion. This possibility would indicate that a substantial reduction in C/EBP function, rather than haploinsufficiency, confers a predisposition to AML in these cases.
Second, it has been found that the inactivating CEBPA mutation is associated with AML characterized by a long latency period (10 to 30 years), suggesting that additional genetic lesions are contributing factors — a likelihood that would support the concept that different mutations act jointly in the pathogenesis of AML. Despite extensive analysis of other AML-related genes, however, the investigators failed to identify any mutations in them. For this reason, the mechanism underlying the accumulation of immature myeloid precursors in the presence of a disabling mutation in CEBPA appears to be relatively straightforward, but the causal relationship between the mutation and the development of overt leukemia is unclear. One possibility is that carriers of a CEBPA mutation have a large population of poorly differentiated myeloid cells associated with an increased risk of a "second genetic hit" that would lead to AML.
Third, and perhaps most exciting, the observations of Smith et al. confirm and extend the clinical observation that in sporadic AML, mutant CEBPA predicts a favorable outcome in younger patients whose cytogenetic characteristics are associated with an intermediate level of risk.4 The case of Patient II-3 in the family studied by Smith et al., who has remained in complete remission for almost four decades despite having received therapy that would be considered inadequate by current standards, highlights the question of whether patients with AML who carry a CEBPA mutation are being overtreated when they receive the multiple courses of intensive chemotherapy or high-dose therapy followed by hematopoietic stem-cell transplantation that is given to most patients with AML. In the absence of additional data on function or findings from animal models, any hypothesis regarding the molecular basis of the apparent chemoresponsiveness of AML in patients with a CEBPA mutation must remain highly speculative. The same holds true for the core-binding-factor leukemias, characterized by the presence of t(8;21)(q22;q22) or inv(16)(p13q22) mutations, in which the mechanism of sensitivity to high-dose cytarabine or other agents remains elusive. We speculate that leukemias involving CEBPA mutations have a good prognosis because they are relatively "simple" diseases affecting patients with few genetic abnormalities; most of these patients have a normal karyotype, and it appears that CEBPA mutations, particularly N-terminal nonsense mutations, only rarely occur in combination with other genetic abnormalities.
Studies of C/EBP and other transcription factors that regulate hematopoiesis have led to a greater understanding of myeloid leukemogenesis and to the development of a molecularly based classification. However, the most challenging task for future research will be to develop therapies that target these specific genetic pathways.
Source Information
From the Department of Internal Medicine III, University Hospital Ulm, Ulm, Germany.
References
Kelly LM, Gilliland DG. Genetics of myeloid leukemias. Annu Rev Genomics Hum Genet 2002;3:179-198.
Tenen DG. Disruption of differentiation in human cancer: AML shows the way. Nat Rev Cancer 2003;3:89-101.
Song WJ, Sullivan MG, Legare RD, et al. Haploinsufficiency of CBFA2 causes familial thrombocytopenia with propensity to develop acute myelogenous leukaemia. Nat Genet 1999;23:166-175.
Fr?hling S, Schlenk RF, Stolze I, et al. CEBPA mutations in younger adults with acute myeloid leukemia and normal cytogenetics: prognostic relevance and analysis of cooperating mutations. J Clin Oncol 2004;22:624-633.(Stefan Fr?hling, M.D., an)