Letters to the Editor
1303
Clonal duplication of a germline PTPN11 mutation due to acquired uniparental disomy in acute lymphoblastic leukemia blasts from a patient with Noonan syndrome
Leukemia (2007) 21, 1303–1305. doi:10.1038/sj.leu.2404651; published online 15 March 2007
Noonan syndrome (NS; MIM 163950) is a common autosomal dominant disorder characterized by short stature, distinct facial features and congenital cardiac defects.1 In approximately 60% of NS cases, a heterozygous gain-of-function mutation can be identified in one of three genes: mutations are found in PTPN11, SOS1 or KRAS in 50, 10 and o5% of cases, respectively.1 The PTPN11 gene codes for SHP-2, a non-receptor tyrosine phosphatase that regulates multiple responses including proliferation, differentiation and migration. SHP-2 relays growth signals from stimulated growth factor receptors to other signaling molecules, including Ras. Ras is activated by the guanosine nucleotide exchange factor SOS1 and functions as a molecular switch, cycling between an inactive GDP-bound and an active GTP-bound state to control fundamental cellular pathways.2 Two of the three known NS genes, PTPN11 and KRAS, are well-known proto-oncogenes and specific somatic mutations of these genes are detected in cancer cells.1,2 This favors the notion that NS predisposes to malignancy. Indeed, infants with NS that harbor specific germline mutations in PTPN11 (e.g., T73I) or in KRAS (T58I) are predisposed to develop myeloproliferative disease (MPD) resembling juvenile myelomonocytic leukemia (JMML).1 Myeloproliferation is generally self-limiting, implicating that NS-associated mutants disrupt hematopoiesis only during early stages of development. When the proliferating clone acquires additional genetic abnormalities, the disorder may become aggressive and lethal. It is not well established whether older individuals with NS are at increased cancer risk, although there are several reports linking NS to leukemia and other malignancies.3,4 With an estimated incidence of NS of one per 2000 life births per year, these cases of malignancy may, however, be due to chance. For two reasons, an assumed increased cancer risk in NS may not be obvious: (1) the features of NS may be subtle and overlooked in cancer patients and (2) NS-associated mutants of PTPN11 or RAS are functionally mild in contrast to the cancer-associated mutants, thus requiring several genetic alterations cooperating during transformation. We present two childhood cases of NS with acute lymphoblastic leukemia (ALL). Both patients harbored germline mutations of PTPN11. In one patient’s leukemia cells we demonstrate uniparental disomy (UPD) at the PTPN11 gene locus giving rise to loss of the wild-type and duplication of the mutant PTPN11 allele. These data provide genetic evidence that NS-associated germline mutations contribute to malignancy and that loss of the wild-type PTPN11 is an important ‘second hit’. The first NS patient was diagnosed with hypertrophic cardiomyopathy shortly after birth. At the age of 28 months, she presented with failure to thrive, fever, malaise and progressive hepatosplenomegaly. The blood count revealed pancytopenia with hemoglobin level 7.1 g/dl, platelet count 25 000 mm3 and a leukocyte count of 3000 mm3 with 1% blasts. A bone marrow aspirate showed monomorphic infiltration with lymphoblasts. Flow cytometry was consistent with pre-B cell ALL. Genetic analysis uncovered a hyperdiploid karyo-
type without structural aberrations. She was treated according to the German ALL-BFM 2000 protocol (standard risk), responded well, and remains in remission 9 months after initial diagnosis. The second NS patient presented in early childhood with poor growth and NS-like facial features. At the age of 8 years she developed fever and organomegaly. Blood count demonstrated leukocytosis with 14 800 mm3, hemoglobin 5.3 g/dl and platelets 14 000 mm3. Bone marrow aspirates showed 99% blasts. Immunophenotyping revealed early pre B-cell ALL with aberrant CD13 and CD33 expression. Genetic analyses demonstrated complex chromosomal alterations including a TEL-AML1 fusion. She was treated according to the Nordic Society of Pediatric Hematology and Oncology (NOPHO) intermediate risk-protocol. Three months after the end of treatment the patient relapsed. Subsequently, allogeneic hematopoietic stem cell transplantation (HSCT) was performed from an unrelated donor. One year after HSCT a second relapse was diagnosed and palliative treatment was initiated. Mutation analysis of DNA extracted from buccal cells from the first patient revealed the presence of a heterozygous known NS-associated mutation, c.1510A4G, in PTPN11 predicting an M504V substitution in the protein (Figure 1). In this patient’s leukemic blasts, we identified the same mutation and an absence of the wild-type allele (Figure 1) consistent with loss of heterozygosity. To determine whether this finding was due to loss of one copy of the gene or due to uniparental disomy, we analyzed DNA from blasts using array-CGH. In agreement with the results obtained from karyotype analysis, array-CGH showed hyperdiploidy with additional copies of chromosomes 4, 6, 14, 17, 18 and 21. Notably, array-CGH showed no evidence of a deletion in chromosome 12 band q24.13, the gene locus of PTPN11 (Supplementary Figure 1). Therefore, allele loss in this specimen presumably results from mitotic recombination and consecutive UPD (Figure 1). We next analyzed bone marrow cells obtained during remission for mutations and for genomic rearrangements and found that, with the exception of the patient’s heterozygous germline PTPN11 mutation, all other genetic abnormalities were absent (Figure 1 and Supplementary Figure 1). Loss of wild-type PTPN11 allele has been previously described in a patient with isolated JMML and a somatic PTPN11A72V mutation.5 Additionally, clonal duplication of a mutant PTPN11N308D allele has recently been reported in a case of NS with therapy-related acute myeloblostastic leukemia.4 Based on these data, we hypothesize that wild-type SHP-2 has properties of a tumor suppressor and the capacity to reduce the transforming potential of oncogenically activated SHP-2 as has previously been shown for wild-type KRAS.6 Loss of the wild-type and duplication of the mutated allele due to UPD has recently been reported to occur in other oncogenes such as KRAS6 and JAK2.7 Likewise, Fitzgibbon et al. have recently identified concurrent homozygous mutations at four distinct loci (WT1, FLT3, CEBPA, and RUNX1) in myeloid leukemia specimens, indicating that mutation precedes mitotic recombination which acts as a ‘second hit’ responsible for removal of the remaining wild-type allele.8 Moreover, UPD at the NF1 tumor suppressor gene on chromosome 17 has been described as a frequent mechanism of LOH in neurofibromatosis type 1 (NF1)-associated leukeLeukemia
Letters to the Editor
1304
a
Leukemic blasts
Remission bone marrow
GGGATGG
GGGATGG
GGGATGG
G
G
G
Buccal cells
b
c
Chr. 12
WT
M504V PTPN11
M504V
M504V
mutation
M504V
M504V
WT M504V
UPD
PTPN11
mutation
Figure 1 Uniparental disomy leads to duplication of mutant PTPN11 in leukemic blasts from a patient with Noonan syndrome and ALL. (a) DHPLC chromatograms suggest the presence of a mutation in DNA extracted from buccal cells (left) and remission bone marrow cells (right) of the patient. Note the less abnormal chromatogram seen for leukemia cells (middle) due to loss of the wild-type allele with decreased heteroduplex formation (b and c). (b) In an independent experiment, sequencing of DNA from buccal and remission bone marrow cells shows a heterozygous germline mutation (M504V) consistent with the clinical diagnosis of Noonan syndrome. PTPN11 mutation analysis of DNA extracted from leukemia cells reveals loss of the wild-type allele. (c) As demonstrated by array-based comparative genome hybridization (Supplementary Figure 1), this allele loss is presumably due to uniparental disomy. Loss of the wild-type PTPN11 is restricted to malignant cells and not detectable in bone marrow cells in remission.
mias.9,10 Notably, NF1 shares many clinical features with NS and both syndromes are associated with JMML and aberrant Ras signaling.1 In the second patient we identified a known heterozygous NS-associated PTPN11E139D mutation. This patient’s leukemia cells retained the normal allele, indicating that ALL in NS is due to heterogeneous mechanisms. In conclusion, we present genetic evidence that NS-associated germline mutations in PTPN11 may cooperate with other genetic events, such as UPD at the PTPN11 gene locus, to contribute to ALL evolution. Our study therefore suggests that, although rare, the association of NS and ALL is not a result of chance alone. Somatic mutations of PTPN11 have been described in cases of sporadic ALL.11 In similar cases the presence of NS should be carefully excluded.
Acknowledgements We thank Marcel Tauscher and Cornelia Klein for their excellent technical assistance. We are grateful to Dr Mwe Mwe Chao for critical comments.
A Karow1, D Steinemann2, G Go¨hring2, H Hasle3, J Greiner4, A Harila-Saari5, C Flotho1, M Zenker6, B Schlegelberger2, CM Niemeyer1 and CP Kratz1 1 Department of Pediatrics, University of Freiburg, Freiburg, Germany; 2 Institute of Cell and Molecular Pathology, Hannover Medical School, Hannover, Germany; 3 Department of Pediatrics, Skejby Hospital, Aarhus University, Aarhus, Denmark; 4 Department of Pediatric Hematology and Oncology, Children’s Hospital of Eastern Switzerland, St Gallen, Switzerland; Leukemia
5
Department of Pediatrics, University of Oulu, Oulu, Finland and 6 Institute of Human Genetics, University of Erlangen-Nuremberg, Erlangen, Germany. E-mail:
[email protected]
References 1 Kratz CP, Niemeyer CM, Zenker M. An unexpected new role of mutant Ras: perturbation of human embryonic development. J Mol Med 2007; 85: 223–231. 2 Downward J. Signal transduction. Prelude to an anniversary for the RAS oncogene. Science 2006; 314: 433–434. 3 Roti G, La Starza R, Ballanti S, Crescenzi B, Romoli S, Foa R et al. Acute lymphoblastic leukaemia in Noonan syndrome. Br J Haematol 2006; 133: 448–450. 4 Chantrain CF, Jijon P, De Raedt T, Vermylen C, Poirel HA, Legius E et al. Therapy-related acute myeloid leukemia in a child with Noonan syndrome and clonal duplication of the germline PTPN11 mutation. Pediatr Blood Cancer 2007; 48: 101–104. 5 Loh ML, Vattikuti S, Schubbert S, Reynolds MG, Carlson E, Lieuw KH et al. Mutations in PTPN11 implicate the SHP-2 phosphatase in leukemogenesis. Blood 2004; 103: 2325–2331. 6 Zhang Z, Wang Y, Vikis HG, Johnson L, Liu G, Li J et al. Wildtype Kras2 can inhibit lung carcinogenesis in mice. Nat Genet 2001; 29: 25–33. 7 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. 8 Fitzgibbon J, Smith LL, Raghavan M, Smith ML, Debernardi S, Skoulakis S et al. Association between acquired uniparental disomy and homozygous gene mutation in acute myeloid leukemias. Cancer Res 2005; 65: 9152–9154. 9 Stephens K, Weaver M, Leppig KA, Maruyama K, Emanuel PD, Le Beau MM et al. Interstitial uniparental isodisomy at clustered breakpoint intervals is a frequent mechanism of NF1 inactivation in myeloid malignancies. Blood 2006; 108: 1684–1689.
Letters to the Editor
1305 10 Flotho C, Steinemann D, Mullighan CG, Neale G, Mayer K, Kratz C et al. Genome-wide single nucleotide polymorphism analysis in juvenile myelomonocytic leukemia identifies uniparental disomy surrounding the NF1 locus in cases associated with neurofibromatosis but not in cases with mutant RAS or PTPN11. Oncogene 2007, (in press).
11 Tartaglia M, Martinelli S, Cazzaniga G, Cordeddu V, Iavarone I, Spinelli M et al. Genetic evidence for lineage-related and differentiation stage-related contribution of somatic PTPN11 mutations to leukemogenesis in childhood acute leukemia. Blood 2004; 104: 307–313.
Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)
CD7/CD56-positive acute myeloid leukemias are characterized by constitutive phosphorylation of the NF-kB subunit p65 at Ser536
b
at rk
NK L
L/
Ju
6
AM
CD
34
cyt AM
a
p65 P536p65
at rk Ju
L
L/ AM
1
AM
CD
34
2
NK
actin
p65 CD34 AML AML/NK
nucl
c % Annexin V
CD7/CD56 myeloid/natural killer acute leukemia represents a novel rare distinct hematologic neoplasia.1 According to its immunologic markers, it is classified as an acute myeloid leukemia (AML) although its clinical presentation is that typical of lymphomas. In particular, CD7/CD56 AML is characterized by an important extramedullary involvement with mediastinal masses and/or peripheral lymphoadenopathies in spite of a generally modest peripheral leucocytosis. Analysis of leukemic blasts and immunohystochemistry of adenopathies are positive for the expression of both myeloid and NK markers (CD7, CD33, CD34, CD56 and frequently HLA-DR). All rare patients relapsed or were refractory to therapies, suggesting that this disease is particularly resistant to the conventional AML chemotherapy. Premature relapse after allogenic bone marrow transplantation have also been described.1 NF-kB is a transcription factor, which is generally composed of the p65/p50 heterodimer. In resting cells, NF-kB is retained into the cytoplasm by the inhibitory protein IkB-a. Upon stimulation, IkB-a becomes degraded and free NF-kB translocates into the nucleus, where it regulates the expression of several genes involved in survival and proliferation. NF-kB has been described as an essential transcription factor in tumorigenesis. In particular, it is believed to be necessary for myeloid leukemogenesis.2 Recently, it has been demonstrated that NF-kB could be activated via a non-canonical pathway.3–5 At least five kinases are responsible for the phosphorylation of p65 at Ser536.4 This phosphorylated form of p65 is no longer subject to IkB-a inhibition and does not bind to p50. Moreover, phospho-S536-p65 can translocate to the nucleus independently to conventional stimuli, where it may regulate the expression of genes which appear to be different with respect to those regulated by IkB-a-mediated NF-kB activation.4 In the lymphoblastic lymphoma cell line Jurkat and in primary T-cell lymphocytes, p65 is constitutively phosphorylated at Ser536.5 In this work we have isolated leukemic blasts obtained from two CD7/CD56-positive AML and seven conventional AML blasts. Both of the CD7/CD56 AML patients’ samples were collected at the time of the diagnosis. Both patients present the clinical characteristic, described by Suzuki,1 were treated with conventional AML protocols (one received allogenic bone
marrow transplantation), but both of them die for resistance to therapies. To quantify the level of activation of the transcription factor NF-kB in CD7/CD56 AML and in conventional AML, bone marrow blasts of two CD7/CD56 AML and seven conventional AML sample have been lysed, and 20 mg of nuclear lysate has been assayed for NF-kB DNA-binding activity as described elsewhere.6 Data have been compared with DNA-binding activity of NF-kB measured in a pool of three normal CD34 nuclear extracts. This value has been referred as 1 (Figure 1a). In conventional AML samples, NF-kB is highly activated with a mean level of activation of six-fold. CD7/CD56 AML samples share a lower level of NF-kB activation (mean 2). To assess whether the lower NF-kB activation was associated with a
Folds of NF-kB DNA binding
Leukemia (2007) 21, 1305–1306. doi:10.1038/sj.leu.2404581; published online 22 March 2007
40
15 10 - + CD34
- + AML
- + AML/NK
MG-132
Figure 1 Peculiar pattern of NF-kB activation in CD7/CD56 AML. (a) NF-kB DNA-binding activity has been measured as described in Materials and methods. Values obtained in normal CD34 cells have been referred as ‘1’. (b) Cytosolic and nuclear extracts of the indicated samples have been obtained as described, and separated on a 10 % SDS–PAGE. P536p65 means phosphorylated p65 at Ser536. (c) The indicated primary cells have been incubated with 1 mM MG-132 for 10 h and then apoptosis have been measured by quantification of Annexin V-positive cells. Leukemia
