Letters to the Editor 1046 grants CA30969, CA29139, CA98543, CA98413, and CA114766 for unfavorable early treatment response. Blood 2009; 114: awarded to the Children’s Oncology Group. 1053–1062. 4 Gutierrez A, Sanda T, Grebliunaite R, Carracedo A, Salmena L, LYu1, ML Slovak2,5, K Mannoor1, C Chen1, SP Hunger3, Ahn Y et al. High frequency of PTEN, PI3K, and AKT abnormalities in T-cell acute lymphoblastic leukemia. Blood 2009; 114: AJ Carroll4, RA Schultz2, LG Shaffer2, BC Ballif2 and Y Ning1 1 647–650. Department of Pathology, University of Maryland School 5 Duker AL, Ballif BC, Bawle EV, Person RE, Mahadevan S, of Medicine, Baltimore, MD, USA; 2 Alliman S et al. Paternally inherited microdeletion at 15q11.2 Division of Cytogenetics, Signature Genomic Laboratories, confirms a significant role for the SNORD116 C/D box snoRNA Spokane, WA, USA; 3 cluster in Prader-Willi syndrome. Eur J Hum Genet 2010; 18: Center for Cancer and Blood Disorders, University of 1196–1201. Colorado Denver School of Medicine, Aurora, CO, USA and 6 Raschke S, Balz V, Efferth T, Schulz WA, Florl AR. Homozygous 4 Department of Genetics, University of Alabama at deletions of CDKN2A caused by alternative mechanisms in various Birmingham, Birmingham, AL, USA human cancer cell lines. Genes Chromosomes Cancer 2005; 42: E-mail: [email protected] 58–67. 5Current address: Division of Cytogenetics, Quest Diagnostics 7 De Keersmaecker K, Marynen P, Cools J. Genetic insights in the Nichols Institute, Chantilly, VA, USA. pathogenesis of T-cell acute lymphoblastic leukemia. Haemato- logica 2005; 90: 1116–1127. 8 Colaizzo-Anas T, Aplan PD. Cloning and characterization of the SIL promoter. Biochim Biophys Acta 2003; 1625: 207–213. References 9 Van Vlierberghe P, van Grotel M, Tchinda J, Lee C, Beverloo HB, van der Spek PJ et al. The recurrent SET-NUP214 fusion as a new 1 Schultz KR, Pullen DJ, Sather HN, Shuster JJ, Devidas M, Borowitz HOXA activation mechanism in pediatric T-cell acute lympho- MJ et al. Risk- and response-based classification of childhood blastic leukemia. Blood 2008; 111: 4668–4680. B-precursor acute lymphoblastic leukemia: a combined analysis of 10 Kleppe M, Lahortiga I, El Chaar T, De Keersmaecker K, Mentens N, prognostic markers from the Pediatric Oncology Group (POG) and Graux C et al. Deletion of the protein tyrosine phosphatase gene Children’s Cancer Group (CCG). Blood 2007; 109: 926–935. PTPN2 in T-cell acute lymphoblastic leukemia. Nat Genet 2010; 2 Van Vlierberghe P, Homminga I, Zuurbier L, Gladdines-Buijs J, van 42: 530–535. Wering ER, Horstmann M et al. Cooperative genetic defects in 11 Mendell JT. miRiad roles for the miR-17-92 cluster in development TLX3 rearranged pediatric T-ALL. Leukemia 2008; 22: 762–770. and disease. Cell 2008; 133: 217–222. 3 Remke M, Pfister S, Kox C, Toedt G, Becker N, Benner A et al. 12 Mavrakis KJ, Wolfe AL, Oricchio E, Palomero T, de Keersmaecker High-resolution genomic profiling of childhood T-ALL reveals K, McJunkin K et al. Genome-wide RNA-mediated interference frequent copy-number alterations affecting the TGF-beta and PI3K- screen identifies miR-19 targets in Notch-induced T-cell acute AKT pathways and deletions at 6q15-16.1 as a genomic marker lymphoblastic leukaemia. Nat Cell Biol 2010; 12: 372–379.

DNA mismatch repair status affects cellular response to Ara-C and other anti-leukemic nucleoside analogs

Leukemia (2011) 25, 1046–1049; doi:10.1038/leu.2011.38; mechanisms leading to cell death remain unclear, although one published online 4 March 2011 hypothesis suggests that recognition of O6-MeG:T or Me6-TG:T mispairs in DNA by MutSa leads to repeated abortive repair, Nucleoside analogs form the backbone of therapeutic regimes ultimately resulting in the formation of pro-cytotoxic DNA for acute myeloid leukemia (AML) and are used to treat strand breaks.1 MMR-defective cells lack the capacity to numerous other hematological malignancies, including B-cell recognize mispairs and consequently demonstrate a ‘tolerant’ acute lymphoblastic leukemia and Hodgkin lymphoma. Chemo- phenotype and chemoresistance. resistance presents a major obstacle to the successful treatment The cytotoxicity of Ara-C is mediated in part by incorporation of both relapsed and therapy-related AML (t-AML), and altered into replicating DNA leading to inhibition of chain extension, DNA repair function is one possible mechanism underlying this stalling of replication forks, activation of the S-phase checkpoint phenotype. and cell cycle arrest. However, the presence of Ara-C residues at Microsatellite instability, indicative of defective DNA internucleotide linkage positions in leukemic cells treated mismatch repair (MMR) is a rare finding in de novo AML but in vitro provides evidence that Ara-C is not an absolute chain is common in relapsed AML and t-AML, in which it is reported terminator and that extension can occur following incorporation to occur in 30–90% of cases.1–3 DNA MMR is responsible for into DNA.4 Given this, we hypothesized that incorporation of maintaining the integrity of the genome by mediating repair of Ara-C might generate a substrate recognized by the MMR spontaneous DNA damage arising during DNA replication. The machinery and that, as in the case of 6-TG, cellular response to MutSa heterodimer (comprising hMSH2 and hMSH6) initiates Ara-C-induced toxicity is influenced by MMR status. repair of base/base mispairs. A second complex, MutSb MT-1, a B-lymphoblastoid cell line mutated in both copies of (comprising hMSH2 and hMSH3), mediates repair of short- MSH6, expressed almost negligible levels of hMSH6 protein and long insertion-deletion loops, but has no activity toward compared with its MMR-proficient parental cell line, TK6 mispairs. Subsequent recruitment of the MutLa heterodimer (assessed by western immunoblotting). Loss of MSH6 was (hMLH1 and hPMS2) facilitates strand excision and gap filling, accompanied by an approximate 50% decrease in hMSH2 restoring the DNA to its original structure. Paradoxically, protein but no difference in hMSH3 (Figure 1a). MT-1 was less functional DNA MMR facilitates cytotoxicity of certain sensitive to the cytotoxic effects of the methylating agent chemotherapeutic agents, including methylating agents and N-methyl-N-nitrosourea (MNU) (Figure 1b) and to 6-TG the base analog 6-thioguanine (6-TG). In both the cases, the (Figure 1c) compared with parental TK6 cells, consistent with

Leukemia Letters to the Editor 1047 the established tolerance of MT-1 to these agents and its known defect in DNA MMR. MT-1 and TK6 also displayed differential toxicity to Ara-C, consistent with a role for DNA MMR in affecting cellular response to this nucleoside analog. However, in contrast to the phenotype observed following treatment with 6-TG, DNA MMR-defective MT-1 cells were acutely sensitive to Ara-C relative to parental TK6 cells (Figure 1d). We next investigated whether MMR status affects cellular response to other nucleoside analogs used in the treatment of hematological malignancy. Clofarabine and fludarabine, like Ara-C, are good substrates for incorporation into DNA by DNA polymerases. Cellular response following treatment with each of these agents was consistent with that observed following treatment with Ara-C, in that MMR-defective MT-1 cells were more sensitive relative to MMR-competent parental TK6 cells (Figures 1e and f). There was no significant difference in sensitivity to UV radiation between TK6 and MT-1 (Figure 1g). Given that UV radiation is not thought to induce lesions, which are substrates for DNA MMR, this finding excludes a generic DNA repair defect in MT-1. These data suggest that the role of DNA MMR as a modulator of cellular response to nucleoside analogs are not limited to the pyrimidine analog Ara-C, but can be extended to other arabinosyl nucleoside analogs. Moreover, the mode of interaction is clearly different to that reported for 6-TG and methylating agents. To investigate the effect of defects in specific components of DNA MMR, we developed fully isogenic cell line pairs using short-hairpin RNA-mediated gene interference to reduce expression of MSH2, MSH3 or MSH6 in lymphoid cells (TK6) and myeloid leukemic cells (HL-60) (Figure 2a). Lentiviral particles containing short-hairpin RNA constructs targeting MSH2, MSH3 or MSH6 were used to transduce TK6 or HL-60, according to the manufacturers protocols (Sigma-Aldrich, Dorset, UK). Stably transfected cells were selected and maintained by transfer to cell culture medium supplemented with 2 mg/ml puromycin. Knockdown of MSH2 resulted in complete loss of hMSH2 protein in both HL-60 (HL-60 MSH2i) and TK6 (TK6 MSH2i), accompanied by complete loss of both hMSH3 and hMSH6. MSH3 knockdown resulted in near- complete loss (approximately 95%) of hMSH3 in both HL-60 (HL-60 MSH3i) and TK6 (TK6 MSH3i), but there was no observed effect on hMSH2 or hMSH6. Knockdown of MSH6 completely abolished hMSH6 protein (HL-60 MSH6i and TK6 MSH6i), and was accompanied by an approximate 50% decrease in hMSH2 but did not affect hMSH3 expression relative to respective parental cell lines. Cells depleted for either hMSH2 or hMSH6 demonstrated tolerance to the cytotoxic effects of MNU (Figure 2b) and 6-TG (Figure 2c) compared with respective parental cell lines, consistent with the phenotype expected in MutSa-defective cells when exposed to these agents. Depletion of hMSH3 had no effect on cytotoxicity induced by MNU or 6-TG (Figures 2b and c), confirming that the MutSb complex has no role in mediating cytotoxicity in response to these agents. Figure 1 MSH6 mutation modulates nucleoside analog-induced Like MSH6-mutated MT-1, cells depleted in hMSH2 or cytotoxicity in MMR-defective lymphoblastoid cells. (a)Western hMSH6 were sensitive to the cytotoxic effects of Ara-C, immunoblotting of MutS components (MSH2, MSH3 and MSH6) in MMR- clofarabine and fludarabine relative to their respective parental proficient TK6 and its MMR-deficient subclone MT-1. HeLa and LoVo cells (Figures 3a–c). In contrast, both TK6 and HL-60 cells were included as MMR proficient and deficient controls, respectively. deficient in hMSH3 were consistently resistant to the cytotoxic b-Actin was used as a loading control. Each blot is representative of three effects of Ara-C, clofarabine and fludarabine (Figures 3a–c). We independent experiments. (b–g) Cell survival in response to MNU (b), 6-TG (c), Ara-C (d), clofarabine (e), fludarabine (f) or UV radiation (g)was treated all isogenic cell lines with UV radiation and observed no compared in TK6 and MT-1 via clonogenic assay. Data are presented as differences in sensitivity in any of the cell lines relative to the number of viable cells from dosed cell suspensions as a percentage parental cells (Figure 3d). of the number of viable cells from untreated cell suspensions, and Taken together, these data demonstrate that DNA MMR status represents the mean and s.e. of three independent experiments. affects cellular response to Ara-C and other nucleoside analogs,

Leukemia Letters to the Editor 1048 and also suggests that the mode of interaction between DNA MMR components and nucleoside analogs is defect-specific, and different to that postulated for 6-TG.

Figure 2 Knockdown of MutS components by short hairpin RNA-mediated gene interference affects cellular response to MNU and 6-TG. (a) Western immunoblotting of MutS components (MSH2, MSH3 and MSH6) following stable short-hairpin RNA-mediated knockdown of MSH2, MSH3 or MSH6 in HL-60 and TK6. Non-target short-hairpin RNA was used to control for effects of the transduction Figure 3 Knockdown of MutS components by short-hairpin process on protein expression. b-Actin was used as a loading control. RNA-mediated gene interference affects cellular response to nucleo- (b, c) Cell survival in response to MNU (b) and 6-TG (c) was compared side analogs. Cell survival in response to Ara-C (a), clofarabine (b), in MMR-knockdown cell lines and their respective MMR-proficient fludarabine (c) or UV radiation (d) was compared in MMR-knockdown parental counterparts by clonogenic assay (TK6) or by growth cell lines and their respective MMR-proficient parental counterparts by inhibition assay (HL-60). Data are presented as described in the clonogenic assay (TK6) or by growth inhibition assay (HL-60). Data are legend to Figure 1. presented as described in the legend to Figure 1.

Leukemia Letters to the Editor 1049 A number of possible explanations could account for these specific DNA MMR defects, including MSH2 mutation and findings, and it should be noted that they are not necessarily loss of protein expression, have been reported in high-risk mutually exclusive. First, we can speculate that unlike 6-TG, patients. Given the increasing use of clofarabine in the treatment Ara-C and other nucleoside analogs induce substrates in DNA of elderly and other high-risk AML subgroups, our data suggest recognized by MutSb, as well as MutSa. Indeed, the observation that DNA MMR status may be a relevant prognostic factor. that hMSH3 loss confers resistance to the cytotoxic effects of In summary, we have shown that DNA MMR status affects nucleoside analogs suggests a role for MutSb. Second, it is also cellular response to the cytotoxic effects of nucleoside analogs, possible that some lesions induced by Ara-C, clofarabine and and these data suggest that the development of tailored fludarabine can be successfully repaired by DNA MMR. chemotherapeutic regimens to exploit differences in DNA repair Sensitization to nucleoside analog-induced killing in MutSa- status could be of clinical use. deficient cells is consistent with this model. A third possible explanation is that some DNA MMR components are also Conflict of interest involved in other repair pathways, which may be invoked following exposure to nucleoside analogs. Homologous recom- The authors declare no conflict of interest. bination repair is the primary repair mechanism for double- strand breaks occurring at stalled replication forks.5 Potential deregulation of homologous recombination repair, as a con- Acknowledgements sequence of DNA MMR loss, therefore represents a mechanism that could confer sensitivity to nucleoside analog-induced This work was supported by Leukemia and Lymphoma Research, cytotoxicity, as we have observed. UK (Grant no. 06002 to JMA). There is mounting evidence that DNA MMR components can SE Fordham1, EC Matheson1, K Scott2, JAE Irving1 and modulate cellular response to other chemotherapeutic nucleo- JM Allan1 side analogs, including 5-fluorouracil, although the mode of 1Northern Institute for Cancer Research, Newcastle University, interaction remains to be determined. 5-fluorouracil-mispaired Newcastle-upon-Tyne, UK and opposite guanine is recognized by MutSa, and studies have 2Department of Biology, University of York, York, UK reported resistance to 5-fluorouracil-induced cytotoxicity in cell E-mail: [email protected] lines deficient in components of DNA MMR.6 Similarly, there is some evidence that DNA MMR components might also have a References role in modulating in vitro response to gemcitabine, a nucleo- side analog of deoxycytidine also used to treat leukemia. 1 Allan JM, Travis LB. Mechanisms of therapy-related carcinogenesis. Notably, Takahashi et al.7 reported gemcitabine to be more Nat Rev Cancer 2005; 5: 943–955. toxic to hMLH1-deficient cells than their MMR-proficient 2 Worrillow LJ, Travis LB, Smith AG, Rollinson S, Smith AJ, Wild CP et al. An intron splice acceptor polymorphism in hMSH2 and risk of counterparts. Although these data suggest involvement of DNA leukemia after treatment with chemotherapeutic alkylating agents. MMR components in response to nucleoside analogs, they do Clin Cancer Res 2003; 9: 3012–3020. not confirm a role for the DNA MMR pathway per se,as 3 Zhu YM, Das-Gupta EP, Russell NH. Microsatellite instability and response could be mediated through other pathways in which p53 mutations are associated with abnormal expression of the MMR components participate, as we have suggested. MSH2 gene in adult acute leukemia. Blood 1999; 94: 733–740. Exposure to methylating agents and 6-TG contribute to t-AML 4 Ross DD, Chen SR, Cuddy DP. Effects of 1-beta-D-arabinofuranosyl- cytosine on DNA replication intermediates monitored by pH-step etiology through a mechanism involving the loss of DNA MMR alkaline elution. Cancer Res 1990; 50: 2658–2666. and the acquisition of genomic instability. A growing body of 5 Arnaudeau C, Lundin C, Helleday T. DNA double-strand breaks evidence has highlighted a possible association between t-AML associated with replication forks are predominantly repaired by and previous nucleoside analog therapy. For example, a homologous recombination involving an exchange mechanism in multicentre study reported that chronic lymphocytic leukemia mammalian cells. J Mol Biol 2001; 307: 1235–1245. patients treated with fludarabine were at increased risk of 6 Meyers M, Wagner MW, Mazurek A, Schmutte C, Fishel R, Boothman 8 DA. DNA mismatch repair-dependent response to fluoropyrimidine- developing t-AML. Although our data demonstrate a role for generated damage. JBiolChem2005; 280: 5516–5526. DNA MMR components in mediating cellular response to 7 Takahashi T, Min Z, Uchida I, Arita M, Watanabe Y, Koi M et al. fludarabine and other nucleoside analogs, it remains to be Hypersensitivity in DNA mismatch repair-deficient colon carcino- determined whether these agents facilitate the development of ma cells to DNA polymerase reaction inhibitors. Cancer Lett 2005; t-AML through loss of DNA MMR. 220: 85–93. Defective DNA MMR, as evidenced by high levels of 8 Morrison VA, Rai KR, Peterson BL, Kolitz JE, Elias L, Appelbaum FR et al. Therapy-related myeloid leukemias are observed in patients microsatellite instability, is often associated with AML sub- with chronic lymphocytic leukemia after treatment with fludarabine groups classified as high risk, such as elderly patients and those and chlorambucil: results of an intergroup study, cancer and with relapsed or therapy-related disease.2,3 Furthermore, leukemia group B 9011. J Clin Oncol 2002; 20: 3878–3884.

The t(4;9)(q11;q33) fuses CEP110 to KIT in a case of acute myeloid leukemia

Leukemia (2011) 25, 1049–1050; doi:10.1038/leu.2011.40; cell maintenance and mast cell development.1 KIT is normally published online 15 March 2011 activated by engagement with its ligand, stem cell factor, but a variety of acquired KIT point mutations have been described in KIT is a class III receptor tyrosine kinase (RTK) that has an acute myeloid leukemia (AML) and systemic mastocytosis that important role in hemopoiesis, particularly with regard to stem result in constitutive, ligand-independent activation.1 In contrast

Leukemia