Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Published Ahead of Print on August 6, 2015, as doi:10.3324/haematol.2015.129510. Copyright 2015 Ferrata Storti Foundation. DNMT3A mutations occur early or late in patients with myeloproliferative neoplasms and mutation order influences phenotype by Jyoti Nangalia, Francesca L. Nice, David C. Wedge, Anna L. Godfrey, Jacob Grinfeld, Clare Thakker, Charlie E. Massie, Joanna Baxter, David Sewell, Yvonne Silber, Peter J. Campbell, and Anthony R. Green Haematologica 2015 [Epub ahead of print] Citation: Nangalia J, Nice FL, Wedge DC, Godfrey AL, Grinfeld J, Thakker C, Massie CE, Baxter J, Sewell D, Silber Y, Campbell PJ, Green AR. DNMT3A mutations occur early or late in patients with myeloproliferative neoplasms and mutation order influences phenotype. Haematologica. 2015; 100:xxx doi:10.3324/haematol.2015.129510 Publisher's Disclaimer. E-publishing ahead of print is increasingly important for the rapid dissemination of science. Haematologica is, therefore, E-publishing PDF files of an early version of manuscripts that have completed a regular peer review and have been accepted for publication. E-publishing of this PDF file has been approved by the authors. After having E-published Ahead of Print, manuscripts will then undergo technical and English editing, typesetting, proof correction and be presented for the authors' final approval; the final version of the manuscript will then appear in print on a regular issue of the journal. All legal disclaimers that apply to the journal also pertain to this production process. DNMT3A mutations occur early or late in patients with myeloproliferative neoplasms and mutation order influences phenotype Jyoti Nangalia1,2,3, Francesca L. Nice1, David C. Wedge3, Anna L Godfrey1,2, Jacob Grinfeld1,2,3, Clare Thakker1, Charlie E. Massie1, Joanna Baxter2,4, David Sewell2,4, Yvonne Silber1, Peter J. Campbell1,2,3, and Anthony R. Green1,2 1Department of Haematology, Cambridge Institute for Medical Research and Wellcome Trust/MRC Stem Cell Institute, University of Cambridge, Cambridge, UK 2Department of Haematology, Addenbrooke’s Hospital, Cambridge, UK 3Wellcome Trust Sanger Institute, Hinxton, Cambridge, UK 4Cambridge Blood and Stem Cell Bank, University of Cambridge, Cambridge, UK Correspondence Professor Anthony R. Green, Cambridge Institute for Medical Research, Hills Road, Cambridge, CB2 0XY, UK. E-‐mail: [email protected] Running title: DNMT3A mutations in MPNs and influence of mutation order 1 Letter to the editor JAK2 CALR Somatic mutations in , and MPL are found in the majority of myeloproliferative neoplasms (MPN) but many patients also harbor somatic mutations in epigenetic regulators of DNA methylation TET2 DNMT3A ( , TET2, ASXL1 IDH1/2 and and EZH2 ) or chromatin structure ( JAK2V617F and TET2 /progenitor cell biology and clinical presentation 2. TET2, affecting 7-10% of patients DNMT3Amut) mutations ( JAK2V617F and EZH2 ). In MPN patients, mutations in occur either prior to or following the acquisition of order of mutation acquisition for in MPNs after ASXL1 3–5. JAK2V617F 1 and recently, the has been shown to influence hematopoietic stem DNMT3A is the next most frequently mutated gene DNMT3A However in contrast to other mutations, have only been reported early in myeloid disease: prior to acquisition of or in a separate clone in MPNs1,6; prior to NPM1 or FLT3 de novo mutations in acute myeloid leukemia (AML) 7; and prior to AML transformation of MPNs or myelodysplasia 8. Here, we have used clonal analysis of hematopoietic colonies from DNMT3Amut MPN patients to investigate timing of mutation acquisition, subclonal evolution, and the influence of mutation order. Thirteen DNMT3Amut MPN patients were identified in whom viable material was available for clonal analysis (Table S1): 9 (WES), 1 mutated JAK2V617F- JAK2exon12- mutated patient identified mutated and 2 patient by MPL- identified targeted gene by mutated patients identified by exome-sequencing whole-genome screening (TGS). sequencing MPN (WGS), diagnoses and conformed 1 CALR- to both British Committee for Standards in Haematology and World Health Organization 2008 classifications. All samples were obtained following written informed consent and ethical approval. Peripheral blood mononuclear cells were cultured to obtain BFU-Es 2 as described previously , and 2991 individual colonies (average of 230 colonies/patient) were genotyped by Sanger sequencing for mutations in DNMT3A, JAK2, CALR DNMT3Amut and MPL . occurred prior to acquisition of JAK2V617F as evidenced by single-mutant colonies harboring DNMT3Amut and JAK2V617F . Single-mutant colonies in four patients (‘ DNMT3Amut DNMT3A -first’ patients, Fig.1A) only and double-mutant colonies with represented a substantial proportion of BFU-Es (mean 58%, range 27%-86%), consistent with mutant allele fractions from granulocyte WES, and indicating the presence of significant ‘pre- These results accord with observations JAK2’ multi-lineage clonal hematopoiesis in these patients. of clonal hematopoiesis in ET patients despite low allele 2 burdens of JAK2V617F 9, and with reports DNMT3Amut- of associated DNMT3Amut individuals 10,11. In three patients, and JAK2V617F clonal hematopoiesis in normal colonies were mutually exclusive (‘biclonal’ patients, Fig.1B). In such patients, the two mutations have arisen either in separate cells downstream of a shared ancestral clone or in clonally unrelated cells. DNMT3Amut and JAK2V617F clones did not share additional mutations (identified by WES) in any of the 3 patients (Fig.2 patients #25, #81, #27). Moreover, in one of the two female patients, DNMT3Amut X-chromosomes (Fig.S1). These data demonstrate that and JAK2V617F DNMT3Amut clones harbored different active and JAK2V617F clones in the same patient may represent clonally-unrelated expansions. In three patients, DNMT3Amut patients, Fig.1C) but these occurred patients after acquisition were more of mutated JAK2 or MPL JAK2/MPL difficult to identify. In patient ( #650, the order mutation acquisition was initially unclear as the majority of colonies were double-mutant for and JAK2exon12, confirmed wildtype that and the DNMT3A no antecedent JAK2exon12 single-mutant mutation arose colonies first, and were that JAK2exon12 (Fig.1C). heterozygous- became undetectable following acquisition of from an earlier timepoint showed heterozygous- detected DNMT3Amut, JAK2exon12 DNMT3Amut order, the two homozygous- different mitotic antecedent colonies we with because colonies grown colonies with wildtype DNMT3A (Fig.S2A). JAK2exon12 in in only the later timepoint (Fig.S2B). Consistent with this mutation JAK2exon12 recombination JAK2/MPL of DNMT3A However, Furthermore, WGS of a diagnostic sample and a later timepoint demonstrated mutated both timepoints but -first subclones present in this patient (Fig.1C, h1 and h2) carried breakpoints (Fig.S2C). In two further JAK2/MPL- first patients, single-mutant colonies were detected but were present only at low levels (~3% of total colonies, Fig.1C). Overall, considering MPN patients with sequential acquisition of mutations within mutant the same clone, DNMT3Amut single-mutant JAK2/MPL subclones were significantly smaller than single- and double-mutant subclones (p=0.03 for both comparisons, t-test, Fig.S3). To exclude a confounding effect of mutations in other genes known to be recurrently mutated in myeloid malignancies, WES or WGS data were interrogated for all 13 patients. Four patients had additional mutations in TET2, CBL or SH2B3 . Delineation of full phylogenetic hierarchies using Sanger sequencing or Fluidigm SNP Genotyping (see supplementary methods) of individual colonies from these patients did not identify any preferential association of these mutations with either single or double-mutant subclones (Fig.2). Our data therefore suggest that JAK2/MPL single-mutant subclones may have a 3 competitive disadvantage compared with DNMT3Amut subclones, in which case JAK2/MPL -first patients may be enriched among those in whom mutation order could not be determined by colony assay (Fig.1D). To investigate competition between JAK2/MPL- mutated and DNMT3Amut subclones further, colonies were grown from paired samples obtained at different timepoints (median separation 35 months; range 6-179 months) in 10 patients (3 biclonal, 3 unclear). In 6 patients who harbored single-mutant JAK2/MPL DNMT3Amut -first, 3 colonies DNMT3 (2 -first, 1 order patients receiving hydroxycarbamide, 1 patient receiving pipobroman, 2 patients receiving interferon-alpha (IFN), and 1 patient not receiving cytoreduction), the proportions did not significantly change between timepoints (Fig.3A, B; blue shading). Similarly, in 7 patients with double-mutant colonies (4 patients receiving hydroxycarbamide, 1 patient receiving pipobroman and 2 patients not receiving cytoreduction) there was no significant change in the proportions over time (Fig.3A, B; purple shading). In contrast, in all 6 patients with single-mutant JAK2 MPL / colonies (1 patient receiving hydroxycarbamide, 2 patients receiving IFN and 3 patients not receiving cytoreduction), the proportions of single-mutant colonies fell significantly over time (Fig.3A, B; red shading, p=0.027, Wilcoxon signed rank test,). Whilst the observed reduction of JAK2/MPL colonies could have been influenced by IFN treatment in 2 patients (#81 and #25), 2 patients in whom this pattern was also observed had not received cytoreduction (#27 and #650), disadvantage. supporting the notion that single-mutant We found no associations between JAK2/MPL subclonal subclones have a competitive changes over time and treatment responses (measured in accordance with European Leukemia Net guidelines) in patients, possibly because individual patients harboured multiple and differing combinations of the various clones. We next assessed whether order of acquisition of DNMT3Amut influenced phenotype. In the 8 patients in whom mutation order was established, 4 had ET, and of 4 patients in whom JAK2V617F occurred on a DNMT3A JAK2V617F- mutated MPN DNMT3Amut- first patients all -nonmutated background (1 JAK2V617F –first and 3 biclonal patients), 3 had PV and 1 had MF. To expand this cohort, an additional 918 patients were screened by TGS, and a further 33 patients with identified. Copy-number corrected variant allele fractions were DNMT3Amut used to and determine JAK2V617F the were order of 2 mutation acquisition as recently described , and mutation order was unambiguously assigned in 8 further patients: 2 DNMT3A MF, 1 PMF and 1 ET. -first patients had ET, and 6 Combining both cohorts, all 6 JAK2 -first patients comprised 3 PV, 1 post PV- DNMT3A -first patients presented with ET. By 4 contrast, of 10 patients in whom JAK2V617F arose on a wildtype DNMT3A background, 7 presented with PV and only 1 with ET (p=0.003, Chi squared test). There were no significant differences in other clinical features (Table S2) in this cohort. Our results therefore indicate that mutation order influences clinical presentation, not only in MPN patients , JAK2V617F described2 but also in patients with TET2 and JAK2V617F with DNMT3A prior to mutation of DNMT3A or TET2 mutations, . JAK2 previously DNMT3A or JAK2V617F . By contrast, acquisition of TET2 is associated with PV. ‘ -first’ patients were older at presentation in our previous study 2, but no difference in age was identified between and as mutations Mutations in either are associated with an ET phenotype when acquired prior to JAK2V617F TET2 and DNMT3A -first -first patients. This may be due to the smaller number of patients in the current study or may reflect a real difference in the age at which TET2 In summary, we demonstrate that in MPNs, JAK2/MPL single-mutant DNMT3Amut subclones have and DNMT3A DNMT3Amut a mutations are acquired. can follow competitive JAK2 and disadvantage subclones. This concept is consistent with observations that MPL mutations, and that in vivo DNMT3A compared and TET2 with mutations confer an advantage to hematopoietic stem/progenitor cells 2,12–14, whereas this is not the case for JAK2V617F in DNMT3Amut some mouse models 15 . Furthermore, we show that mutation order of JAK2V617F and is associated with differences in MPN phenotype. This emphasizes the importance of the pattern of acquisition of phenotype of JAK2V617F- JAK2V617F with respect to mutations in epigenetic modifiers in influencing the mutated MPNs. Acknowledgements: We thank the Cambridge Blood and Stem cell Biobank (Cambridge University) and Haemato-Oncology Anthony Bench, Diagnostics Service (Addenbrooke’s hospital) for sample collection and storage, and the Cancer Genome Project (Wellcome Trust Sanger Institute, Hinxton) for next generation sequencing. Work in the Green lab is supported by Leukemia and Lymphoma Research, the Wellcome Trust, the Medical Research Council, the Kay Kendall Leukaemia Fund, the Cambridge NIHR Biomedical Research Center, the Cambridge Experimental Cancer Medicine Centre and the Leukemia and Lymphoma Society of America. ALG and JN were supported by Kay Kendall Leukaemia Fund clinical fellowships. PJC is a Wellcome Trust senior clinical fellow. 5 Authorship contributions: JN designed experiments, analysed data and wrote the paper. JN, ALG, FLN, CT and YS performed colony work. Microsatellite screening was by JN and ALG. X chromosome inactivation assessment was by JN. FLN performed Fluidigm SNP Genotyping and colony work. JN and CEM analysed WES and WGS data. EJB and DS prepared patient samples. DCW performed analysis of mutation order. JN and JG screened TGS data and assessed clinical correlates. ARG directed the research. All authors reviewed the manuscript. Disclosure of conflicts of interest: The authors declare no competing financial interests. References 1. Lundberg P, Karow A, Nienhold R, et al. Clonal evolution and clinical correlates of somatic mutations in myeloproliferative neoplasms. Blood. 2014;123(14):2220–2228. 2. Ortmann CA, Kent DG, Nangalia J, et al. Effect of mutation order on myeloproliferative neoplasms. N Engl J Med. 2015;372(7):601–612. 3. Stegelmann F, Bullinger L, Schlenk RF, et al. DNMT3A mutations in myeloproliferative neoplasms. Leukemia. 2011;25(7):1217–1219. 4. Abdel-Wahab O, Pardanani A, Rampal R, Lasho TL, Levine RL, Tefferi A. DNMT3A mutational analysis in primary myelofibrosis, chronic myelomonocytic leukemia and advanced phases of myeloproliferative neoplasms. Leukemia. 2011;25(7):1219–1220. 5. Nangalia J, Massie CE, Baxter EJ, et al. Somatic CALR mutations in myeloproliferative neoplasms with nonmutated JAK2. N Engl J Med. 2013;369(25):2391–2405. 6. Rao N, Butcher CM, Lewis ID, et al. Clonal and lineage analysis of somatic DNMT3A and JAK2 mutations in a chronic phase polycythemia vera patient. Br J Haematol. 2012;156(2):268–270. 7. Shlush LI, Zandi S, Mitchell A, et al. Identification of pre-leukaemic haematopoietic stem cells in acute leukaemia. Nature. 2014;506(7488):328–333. 8. Fried I, Bodner C, Pichler MM, et al. Frequency, onset and clinical impact of somatic DNMT3A mutations in therapy-related and secondary acute myeloid leukemia. Haematologica. 2012;97(2):246–250. 9. Kralovics R, Teo S-S, Li S, et al. Acquisition of the V617F mutation of JAK2 is a late genetic event in a subset of patients with myeloproliferative disorders. Blood. 2006;108(4):1377–1380. 10. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371(26):2488–2498. 11. Genovese G, Kähler AK, Handsaker RE, et al. Clonal hematopoiesis and blood-cancer risk inferred from blood DNA sequence. N Engl J Med. 2014;371(26):2477–2487. 12. Challen GA, Sun D, Jeong M, et al. Dnmt3a is essential for hematopoietic stem cell differentiation. Nat Genet. 2012;44(1):23–31. 6 13. Ko M, Bandukwala HS, An J, et al. Ten-Eleven-Translocation 2 (TET2) negatively regulates homeostasis and differentiation of hematopoietic stem cells in mice. Proc Natl Acad Sci U S A. 2011;108(35):14566–14571. 14. Moran-Crusio K, Reavie L, Shih A, et al. Tet2 loss leads to increased hematopoietic stem cell selfrenewal and myeloid transformation. Cancer Cell. 2011;20(1):11–24. 15. Li J, Kent DG, Chen E, Green AR. Mouse models of myeloproliferative neoplasms: JAK of all grades. Dis Model Mech. 2011;4(3):311–317. 7 Figure Legends Figure 1. Timing of acquisition of DNMT3A mutations Single cell derived hematopoietic erythroid colonies (BFU-Es) were grown in vitro from peripheral blood derived mononuclear cells and individually genotyped for mutations using Sanger sequencing Plots in (A), (B), (C) and (D) show colony genotyping results for mutations in or CALR for each patient and the order of mutation acquisition . DNMT3A JAK2/MPL/CALR JAK2, MPL Within each plot, each dot represents a single colony and its quadrant placement shows the corresponding genotype of axis) and and . DNMT3A (vertical (horizontal axis). Solid red arrows within quadrants show the confirmed path of clonal evolution. Wt, wildtype; het, heterozygous mutation; hom, homozygous mutation; PV, polycythemia vera; ET, myelofibrosis (A) essential DNMT3A- PMF, primary myelofibrosis; PPV-MF, . (B) Biclonal patients: 3 patients in whom separate clones. (C) JAK2/MPL- to the acquisition of mutated DNMT3A and DNMT3A first patients: 4 patients in whom mutated JAK2V617F acquisition of acquisition of thrombocythemia; DNMT3Amut first patients: 3 patients in whom mutated DNMT3A JAK2 post-PV occurred prior to the and or JAK2V617F MPL were in occurred prior . (D) Order unknown: 3 patients in whom the order of mutation JAK2/MPL/CALR could not be delineated as only wildtype colonies and/or double mutant colonies were detected. Figure 2. Clonal structures of DNMT3A-mutated MPNs Individual BFU-Es from 6 patients were genotyped using Sanger sequencing or Fluidigm SNP Genotyping for their respective somatic variants identified previously by whole exome sequencing. Genotyping results from individual colonies were then used to construct phylogenetic trees. Circles represent the subclones; wildtype (white); mutated (brown). The earliest detectable clone is represented at the top of each structure, with subsequent subclones shown below. Somatic mutations acquired in each subclone are indicated beside respective circles, and represent those that are acquired in addition to mutations present in earlier clones. Numbers of colonies identified for each subclone are shown inside circles. Mutations in Additional mutations in genes known to be JAK2, MPL, CALR recurrently and mutated DNMT3A in are highlighted in red. myeloid malignancies are 8 highlighted in green. ET, essential thrombocythemia; PMF, primary myelofibrosis; PV, polycythemia vera; het, heterozygous mutation; hom, homozygous mutation. Figure 3 Evolution of subclones in DNMT3A-mutated MPNs (A) Colonies grown months; 6-179 from paired samples months) in 10 patients. obtained at different timepoints Vertical axis shows the (median separation percentage of columns are shaded to represent the proportions of the different genotypes (red, DNMT3A total colonies JAK2/MPL 35 and -only; blue, -only; purple, double mutant). Numbers of colonies genotyped per patient are shown above columns and the timings of sample acquisition (months from diagnosis) are shown below. (B) Changes in subclonal proportions over time for the 10 patients in (A) for a total of 19 subclones. red, only; blue, DNMT3A -only; purple, double mutant; T1 and T2 represent the earliest JAK2/MPLand latest timepoints sampled for the patients. The median interval between T1 and T2 did not significantly differ between the different subclones (1-way analysis of variance). * <0.05 Wilcoxon ranked sum test 9 Figure 1 ET #98 ET #97 het hom JAK2 V617F wt DNMT3A R882H het wt wt het hom JAK2 V617F wt het hom JAK2 V617F het hom JAK2 V617F wt het hom JAK2 V617F PMF #27 PV #81 DNMT3A S770L wt het DNMT3A Y660F wt het PV #25 DNMT3A c.2597+1G>A wt het B Biclonal wt het hom JAK2 V617F wt het hom JAK2 V617F C JAK2/MPL-first DNMT3A Y908* wt het wt het hom JAK2 V617F ET #50 h2 h1 wt het hom JAK2 F537_K539delinsL DNMT3A F731V wt het hom PV #650 PV #152 DNMT3A I705T wt het wt het hom MPL W515L D Order unknown ET #78 PMF #42 wt het hom MPL W515L PPV-MF #28 DNMT3A R882H wt het DNMT3A D279FS*1 wt het hom wt ET #03 DNMT3A R882H wt het DNMT3A R882H wt het ET #052 DNMT3A R882H wt het DNMT3A c.1123-1_1144del23 wt het A DNMT3A-first wt het hom CALR R376fs*55 wt het hom JAK2 V617F Figure 2 PV #25 Wild type Wild type 58 JAK2 V617F TET2 N281fs*0 2 POLK C769R L1044L PLD1 15 DNMT3A PV #81 Y660F JAK2 V617F hom JAK2 24 DNMT3A S770L V617F het DNMT3A SLC6A6 L675V 2 V617F JAK2 V617F hom GRID1 ET #052 Wild type 2 DNMT3A R882H APLP2 V359V 5 12 GABRB3 R283W 28 V617F A487T 25 JAK2 C10orf71 T295T CD3E G170S R1178C ENSG0218819 LRRC6B F514S NBAS E1326D R55H OR2S2 RPGRIP1l R321H ZSCAN5C S568S SCNSA T413T N972N 12 PMF #42 4 DNMT3A splice L818P RICTOR JAK2 V617F RFTN2 R89* I65V TMEM87A 10 15 8 1 21 44 splice 1 ET #03 Wild type Wild type 62 4 MRPL49 Q28H PMF #27 TET2 Q744fs*10 JAK2 KIF4B Y2125* MYO9A G641R COL9A2 L383L PRKAG3 Mutation in JAK2, DNMT3A 5 GCAT V409M XIRP1 R1275C SLC4A9 I514N ENSG218819 RBM44 G1028D RENBP L144L ZNF143 S286R MUL1 G70R CALR R376fs*55 het TET2 p.K1785* het p.D279fs*1 DNMT3a hom splice SH2B3 hom N727S 79 CALR R376fs*55 C401S 64 42 CBL SCN11A M1706T Additional mutations associated with myeloid malignancies Other mutations hom Figure 3 A ET #052 JAK2-mt 100 ET #98 ET #03 JAK2-mt JAK2-mt ET #78 MPL-mt 197 93 79 76 PV #81 JAK2-mt 70 141 133 91 59 76 153 72 45 59 73 139 151 99 105 54 78 0 172 179 PV #25 PMF #27 ET #50 JAK2-mt JAK2-mt MPL-mt PV #152 JAK2-mt 91 134 112 185 138 0 31 91 132 32 152 20 174 76 82 PV #650 JAK2-mt 6 178 120 Colonies (n) Colonies (% of total) 80 60 40 20 0 Months from Diagnosis JAK2/MPL only DNMT3A only B Colonies (% of total) Wildtype 80 60 40 20 0 T1 JAK2/MPL only (n=6) T2 T1 DNMT3A only (n=6) T2 13 152 183 Double mutant * 100 45 T2 Double mutant (n=7) T1 Supplementary Appendix Construction of phylogenetic hierarchies by genotyping of individual haematopoietic colonies Haematopoietic colonies (BFU-‐E) were grown and individually genotyped using Sanger sequencing or Fluidigm SNP Genotyping for all the mutations that had been previously identified by next-‐generation sequencing in each patient. Colony genotyping for patients #81, #03, #42 and #25 was by PCR and Sanger sequencing. Colony genotyping for #052 and #27 was by Fluidigm SNP Genotyping. Fluidigm genotyping primers were designed according to the manufacturer’s recommendations, and obtained from Fluidigm (Fluidigm Corp, CA, USA). The Fluidigm SNPType Genotyping Reagent Kit (Fluidigm Corp, CA, USA) was used according to the manufacturer’s instructions. Colony DNA was first amplified using Specific Target Amplification as per the manufacturer’s instructions, to enrich for the DNA sequences for subsequent PCR. Genotyping reactions were then performed on a nanofluidic 192.24 Dynamic Array Integrated Fluid Circuit (IFC; Fluidigm Corp, CA, USA), which automatically assembles PCR reactions. End-‐point fluorescent images were acquired on a Biomark system and analysed using the Biomark SNP Genotyping Analysis software version 3.1.2. Colonies with ambiguous genotype results underwent repeat PCR and Sanger sequencing. Primer sequences for Sanger sequencing were as follows: Gene V617F JAK2 exon12 JAK2 R882 DNMT3A Y660F DNMT3A splice(#81) DNMT3A splice(#052) DNMT3A S770L DNMT3A Y908* DNMT3A D279fs*1 DNMT3A DNMT3A F731V I705T DNMT3A N281fs*0 TET2 Q744fs*10 TET2 K1785* TET2 CALR MPL POLK PLD1 MRPL49 APLP2 TMEM87A RFTN2 SH2B3 LANCL3 Forward primer TTTCCTTAGTCTTTCTTTGAAGCAGC CTCCTCTTTGGAGCAATTCA AATACTCCTTCAGCGGAGCGAAGA GTCCCCGACGTACATGATCT CAGGACGTTTGTGGAAAACA TGCCCTCATTTACCTTCTGG ATGAAGCAGCAGTCCAAGG CGCCTCTGTGGTTTTTGTTT CACCTCGTACTCTGGCTCGT ATGAAGCAGCAGTCCAAGG CAAGGAGGAAGCCTATGTG CACATGGTGAACTCCTGGAA GAGCAGATTCCCAAACTGAAA TGGTGAACATCATTCACCTTCT CCTGCAGGCAGCAGAGAAAC TGACCGCTCTGCATCTAGT TGTTCTAGTCTCCCAAGCAAGTC CATCCCCAGGAAGTCACTGT CGCCCAGCCTTTTATTTATG GACGTCACTGCCTCTGTCCT AATGCCAGACTATTAGATTTCAGAAC TAAAACCCCCATGTGTCCTT ACCACCTTTGCTGCTACCAC GCTCACGGCTTGTCGTCTAT Reverse Primer TAGTTTACACTGACACCTAGCTGTGATCC CATCTAACACAAGGTTGGCATA AAGATTCGGCAGAACTAAGCAGGC GACTTGGGCCTACAGCTGA CTCCATAAAGCAGGGCAAA CTCAGAGTCTGGCCTTGAGC TTGAGTTCTACCGCCTCCTG AGTCATCCGCCACCTCTTC CAGGAATGAATGCTGTGGAA GGCTTTCTCTTCCGACCTCT CTTCCTGTCTGCCTCTGTCC TCAGCATCATCAGCATCACA CTTTGGGGGTGAGGAAAAGT TGGTGAACATCATTCACCTTCT ACAGAGACATTATTTGGCGCG TACAGGCCTTCGGCTCCA AAGCAAACATCCACATGCAC CATTCGAGCTGAGGAGGAAC GGGATCTGGGATCCTGGTAG GCGAGGCAGGACTTACTCAT TTGAAGCCATCTGAGGCTAA CGGGGCTATTCATCCTGTTA CCCACCTTGGTTAAGGGAAT CAGTGCACCAGCTCATTCTC Microsatellite marker analysis Microsatellite analysis was used to determine the extent of chromosome 9p loss-of-heterozygosity. A panel of microsatellite markers along chromosome 9p were selected from the GeneLoc database. The PCR reaction contained 1.25µl 10x ReddyMix PCR buffer IV, 0.75µl 25mM magnesium chloride solution, 0.125µl Thermoprime plus polymerase (Thermo Scientific, MA, USA), 0.1µl 100mM dNTPs, 0.06µl 100µM forward primer, 0.06µl 100µM reverse primer, and 0.5-2µl DNA, in a total reaction volume of 12.5µl. PCR conditions were 11 minutes at 95 oC, followed by 38 cycles of 30 seconds at 95oC, 30 seconds at 57oC and 1 minute at 72oC, followed by a final extension at 72oC for 10 minutes. The PCR reaction was diluted between 1:5 and 1:40 in sterile water. Fragment detection was performed on a 3730xl Genomic analyser and results were analysed using the Peak Scanner software version 1.0 (Applied Biosytems, CA, USA). Heterozygosity for these markers was initially analysed using constitutional DNA obtained from a buccal swab. Informative markers were then taken forwards for testing on tumour DNA from individual homozygous JAK2exon12mutated colonies to ascertain the extent of loss-of-heterozygosity in heterozygous microsatellite markers. Primer sequences were as follows: Microsatellite marker D9S288 D9S1852 D9S235 D9S925 D9S162 D9S161 D9S43 D9S1817 D9S1791 D9S2148 D9S176 Primer name D9S288_F D9S288_R D9S1852_F D9S1852_R D9S235_F D9S235_R D9S925_F D9S925_R D9S162_F D9S162_R D9S161_F D9S161_R D9S43_F D9S43_R D9S1817_F D9S1817_R D9S1791_F D9S1791_R D9S2148_F D9S2148_R D9S176_F D9S176_R Sequence (5’-‐3’) GTTTCTTAGCAACCTCAACAGGG 6-‐FAM-‐AATCATCCAGAAAGGCCA GTTTCTTGAATCACAACATACACCCAC 6-‐FAM-‐GAAACATTCTTTTACAAGTAACATT 6-‐FAM-‐CTGTATGGAGAGAGAATACG GTTTCTTGGTCTCTCCGGTATACTCA 6-‐FAM-‐TGTGAGCCAAGGCCTTATAG GTCTGGGTTCTCCAAAGAAA 6-‐FAM-‐GCAATGACCAGTTAAGGTTC AATTCCCACAACAAATCTCC 6-‐FAM-‐TGCTGCATAACAAATTACCAC GTTTCTTCATGCCTAGACTCCTGATCC 6-‐FAM-‐TTCTGATATCAAAACCTGGC AAGGATATTGTCCTGAGGA 6-‐FAM-‐AGCTGTAGTGAGCCCTGAT CGTTAGGAGCCTTGAGACTT 6-‐FAM-‐GTAATCTTGGGCAACCTATGTATG TCAAAATAAGTCTGGGACAAAACC 6-‐FAM-‐TCAATCAACATCTGTCTATTCATC ACATCTGGCACTCTGGAGAG 6-‐FAM-‐AGCTGGCTGTTGGAGAAA TGACCAATGGCAGGGTAT Table S1: Clinical features of MPN patients with DNMT3A mutations MPN Patient At Diagnosis Sex Age Hb g/dl WCC Plts Spleen Karyotype x109/l x109/l FISH normal Karyotype normal At Follow-‐up Therapy during study Past therapy Disease (yrs) DNMT3A mutation JAK2/MPL/CALR status Death Thrombosis HU HU 7 c.1123-‐ 1_1144del23$ JAK V617F N N P HU 14 p.R882H JAK V617F Y N ET #052 F 76 14.5 7.7 804 N ET #98 M 84 15.0 10.7 1285 N ET #8697 F 66 12.0 36.3^ 1034 N 13q del HU HU 5 p.R882H JAK V617F N N ET #03 M 54 14.3 16.4 953 N FISH normal HU IFN 11 p.R882H JAK V617F N Y PV #152 M 72 18.6 18.8 750 N Del Y HU HU 15 p.I705T JAK V617F Y Y nil nil 19 p.Y908* JAK2 exon 12$$$ N N PV #650 F 53 18.5 5.1 308 N Karyotype normal ET #50 M 75 11.9 7.1 488 N NA nil nil 5 p.F731V MPL W515L N N ET #78 M 66 14.8 9.1 983 N NA HU HU 6 p.R882H MPL W515L N N PPV -‐MF #28 F 45 10.4 10.4 229 Y NA nil nil 8 p.R882H JAK V617F N Y PV #25 F 45 16.2 17.6 654 Y NA IFN IFN 4 p.Y660F JAK V617F N Y PV #81 M 33 17.7 11.2 853 Y IFN IFN 16 G>A, c.2597+1$$ JAK V617F N N PMF #27 F 57 10.7 5.2 293 N nil Thal 12 p.S770L JAK V617F N N PMF #42 F 62 12.6 15 238 Y Rux Imatinib 1 p.D279fs*1 CALR R376fs*55 N N Karyotype normal FISH normal NA MPN, myeloproliferative neoplasm; Hb, hemoglobin; WCC, white cell count; Plts, platelet count; PV, polycythemia vera; PMF, primary myelofibrosis; ET, essential thrombocythemia; PPV-‐MF, post PV myelofibrosis; HU, hydroxyurea; P, pipobroman; IFN, interferon alpha; Rux, Ruxolitinib; NA, not available; Y, yes; N, no, NA, not available; $ cDNA annotation of DNMT3A deletion affecting intron-‐exon boundary; $$ cDNA annotation of DNMT3A mutation affecting an essential splice site; $$$ JAK2 exon 12 mutation p.F537_539_delinsL; * stop codon. No transformation events to myelodysplasia, acute myeloid leukemia or polycythemia vera occured during the follow-‐up period.^ Significant leucocytosis due to concurrent chronic lymphocytic leukemia. Table S2 Total PV (n) ET (n) MF (n) Median age (years) Mean WBC (x109/l) Thrombosis (n) Progression (n) Death (n) Median follow-up (days) JAK2V617F-first* 10 6 1 3** 54.6 9.6 2 0 1 4170 (863-5866) DNMT3A-first 6 0 6 0 70.5 14.4 1 0 1 3149 (777-5036) P-value 0.002 0.40 0.57 NP NP NP 0.52 Clinical data for JAK2V617F and DNMT3Amut patients with an established order of mutation acquisition (8 patients in whom order was determined from initial colony analysis and 8 further patients in whom mutation order was determined using data from targeted next-generation sequencing). Data show number of patients or events (n) in the different mutation hierarchy groups. *includes three biclonal patients where JAK2V617F was also acquired on a DNMT3A-wildtype background. **includes one patient with post-PV MF. P-values show results of Chi-squared analysis for categorical variables and t-test for continuous variables. NP, not performed. Supplementary Figure Legends Figure S1. X chromosome inactivation status assessment by analysis of expression patterns of heterozygous SNPs in LANCL-‐3 and MPP1 in the two female patients (#25 and #27). Hematopoietic colonies were pooled by mutation status for JAK2 and DNMT3A. Pooled colonies underwent cDNA extraction followed by SNP genotyping. Sequencing traces are of cDNA derived from colonies sharing the same genotype of JAK2V617F and DNMT3A. #27 (upper panel) has the same X chromosome inactivation status in all subclones and is not informative. #25 (lower panel) has differing expression patterns in JAK2V617F-‐mutated and DNMT3A-‐mutated colonies confirming their independent origins. Figure S2 (a) Colony genotyping from an early sample from #650 showing colonies with heterozygous mutated JAK2 and wildtype DNMT3A (b) Whole genome sequencing reads from #650. Forward and reverse DNA reads are shown in blue and yellow, respectively, and mismatched bases (compared with the reference genome) are shown in red. Left panels shows a region in JAK2 exon 12 and right panels shows a region in DNMT3A. Lower panels show an early bone marrow sample from diagnosis and upper panels show a peripheral blood sample from 14 years later. JAK2 exon 12 mutation (p. F537_K539delinsL) is present in both early and late timepoints, however, DNMT3A mutation (p. Y990*) is not present in the earlier time point. (c) Microsatellite mapping of 9p loss-‐of-‐heterozygosity (LOH) in individual hematopoietic colonies (BFU-‐Es) to determine the breakpoint region leading to acquired uniparental disomy in colonies with homozygous mutations in JAK2 exon 12 from patient #650. The panel of microsatellite markers is shown along the bottom in relation to their position from the telomere and centromere on chromosome 9. Non-‐ informative markers are shaded grey, heterozygous markers are shaded in black and markers showing LOH are shaded in red. Colony numbers genotyped in the two homozygous clones h1 and h2 are shown on the right. Differing lengths of LOH in h1 and h2 confirm distinct breakpoints in the two homozygous clones. Figure S3 Differences in the proportions of JAK2/MPL/CALR-‐only, DNMT3A-‐only and double-‐mutant subclones in patients where mutations have been acquired sequentially (ie. excluding biclonal patients). * p<0.05 Students t-‐test performed on pairwise comparisons. Figure S1 Wildtype subclone JAK2 V617F subclone DNMT3A mutated subclone #27 MPP1 #25 LANCL-3 DNMT3a Y908* wt het hom Figure S2 wt a het hom JAK2 Exon 12 JAK2 DNMT3A p.Y908* (ii) (iii) (iv) At diagnosis 14 years post diagnosis (i) b JAK2 h2 n=5 h1 n=11 D9S176 Centromere D9S43 D9S1817 D9S1791 D9S2148 33.9 D95161 D9S235 D95925 D95162 18.3 6.2 D9S288 D9S1810 D9S1852 c Telomere T cell LOH Het Not informative Mb from Telomere Figure S3 Colonies (% of total) * 60 * 40 20 0 JAK2/MPL DNMT3A Double-mutant (n=3) (n=4) (n=10)