JAK2 V617F tyrosine kinase mutation in cell lines derived from ...

Leukemia (2006) 20, 471–476 & 2006 Nature Publishing Group All rights reserved 0887-6924/06 $30.00 www.nature.com/leu

ORIGINAL ARTICLE JAK2 V617F tyrosine kinase mutation in cell lines derived from myeloproliferative disorders H Quentmeier, RAF MacLeod, M Zaborski and HG Drexler DSMZ-German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany

A mutation in the JH2 pseudokinase domain of the Janus kinase 2 gene (JAK2 V617F) has been described in chronic myeloproliferative disorders (MPD). We screened 79 acute myeloid leukemia (AML) cell lines and found five positive for JAK2 V617F (HEL, MB-02, MUTZ-8, SET-2, UKE-1), 4/5 with histories of MPD/MDS. While SET-2 expressed both mutant (mu) and wild-type (wt) JAK2, remaining positives carried homo-/hemizygous JAK2 mutations. Microsatellite analysis confirmed losses of heterozygosity (LOH) affecting the JAK2 region on chromosome 9p in MB-02, MUTZ-8 and UKE-1, but also in HEL, the only JAK2mu cell line lacking any reported MPD/MDS history. All five JAK2mu cell lines displayed cytogenetic hallmarks of MDS, namely losses of 5q or 7q, remarkably in 4/5 cases affecting both chromosomes. Our combined FISH and microsatellite analysis uncovered a novel mechanism to supplement mitotic recombination previously proposed to explain JAK2 LOH, namely chromosome deletion with/without selective JAK2mu amplification. Confirming the importance of the mutated JAK2 protein for growth and prevention of apoptosis, JAK2mu cell lines displayed higher sensitivities to JAK2 inhibition than JAK2wt cell lines. In summary, JAK2 V617F cell lines, derived from patients with history of MPD/MDS, represent novel research tools for elucidating the pathobiology of this JAK2 mutation. Leukemia (2006) 20, 471–476. doi:10.1038/sj.leu.2404081; published online 12 January 2006 Keywords: HEL; JAK2; MB-02; MUTZ-8; SET-2; UKE-1

interferes with wild-type (wt) JAK2 function remains to be clarified. Continuous cell lines are useful tools for the analysis of basic aspects of normal and malignant cell biology. Therefore, we set out to screen a panel of leukemia cell lines for expression of the JAK2 V617F mutation. Five of 79 cell lines tested JAK2 V617F positive. Cell lines MB-02, MUTZ-8, SET-2 and UKE-1 were established from AML patients with histories of MPD or MDS (42 cell lines tested), while cell line HEL was the only testing positive for JAK2 V617F of 37 leukemia cell lines without a reported history of MPD or MDS. Sequencing and site-specific restriction analysis showed that cell line SET-2 expressed both wt and the mutant JAK2, while the other cell lines expressed mutated JAK2 only. Classic cytogenetic analysis, fluorescence in situ hybridization (FISH) and microsatellite mapping were performed to investigate whether loss of heterozygosity at chromosome 9p (9pLOH) was the result of mitotic recombination as previously reported.1,2,4 JAK2mu cell lines were more sensitive to JAK inhibition than JAK2wt cell lines. The JAK2 V617F positive cell lines may help to elucidate whether and how JAK2mu participates in the rise of these diseases.

Materials and methods

Human cell lines Introduction An activating point mutation in the JH2 domain of Janus kinase 2 (JAK2) was recently described in chronic myeloproliferative disorders (MPD). The JAK2 V617F mutation occurs frequently in polycythemia vera (PV) (65–97% of reported cases), essential thrombocythemia (ET) (23–57%) and idiopathic myelofibrosis (IM) (35–95%).1–6 In isolated cases, it occurs in Philadelphianegative, but not in Philadelphia-positive chronic myeloid leukemia (CML).6 At low percentages, the JAK2 V617F mutation has also been found in ‘atypical MPD-like’ chronic myelomonocytic leukemia (CMML), systemic mastocytosis (SM) and chronic neutrophilic leukemia (CNL), and also in myelodysplastic syndromes (MDS) and acute megakaryocytic leukemia (AML M7), but not in other forms of acute myeloid leukemia (AML M1–M6) or in acute lymphoblastic leukemias (ALL).6–8 JAK2 is involved in cellular growth factor signalling, and deregulation of JAK2 by chromosomal aberrations may contribute to leukemogenesis.9–12 The mechanism whereby JAK2 V617F mutation effects protein activation, which in turn Correspondence: Dr H Quentmeier, DSMZ, Mascheroder Weg 1b, 38124 Braunschweig, Germany. E-mail: [email protected] Received 4 November 2005; accepted 15 November 2005; published online 12 January 2006

The continuous cell lines were either taken from the stock of the cell bank (DSMZ – German Collection of Microorganisms and Cell Cultures)13 or were generously provided by the original investigators. The cell lines were cultivated according to the protocols described previously.13,14 The following cell lines (in alphabetical order) were tested for JAK2 V617F mutation: 380 (derived from patient with B-cell precursor ALL, BCP ALL), AML193 (AML FAB M5), AP-217 (CML megakaryocytic blast crisis, meg BC), AS-E2 (AML M6), CCRF-CEM (T-ALL), CMK (AML M7), CML-T1 (Philadelphia-negative CML), CMY (AML M7), ELF-153 (IM-AML M7), EOL-1 (hypereosinophilic syndrome-eosinophilic leukemia), F-36P (MDS-AML M6), GDM-1 (MPDAML M4), HEL (AML M6), HL-60 (AML M2), HNT-34 (MDSAML M4), HU-3 (AML M7), JK-1 (CML-erythroid BC) JURKAT (T-ALL), JURL-MK1 (CML meg BC), K-562 (CML BC), KARPAS45 (T-ALL), KASUMI-1 (AML M2), KE-37 (T-ALL), KG-1 (AML), KMOE-2 (AML M6), KU-812 (CML myeloid BC), LAMA-84 (CML meg BC), M-07e (AML M7), MARIMO (ET-AML M2), MB-02 (IM-AML M7), MC-3 (CML meg BC), MEG-A2 (CML meg BC), MEG-J (CML meg BC), MEGAL (AML M7), MHHCALL-2 (BCP ALL), MHH-CALL-4 (BCP ALL), MKPL-1 (AML M7), ML-2 (AML M4), MN-60 (B-ALL), MOLM-7 (CML meg BC), MOLM-13 (MDS-AML 5a), MOLM-16 (AML M7), MOLM-17 (MDS-AML 5a), MOLM-20 (CNL), MOLT-4 (T-ALL), MONOMAC-6 (myeloid metaplasia-AML M5), MTT-95 (AML M7), MUTZ-3 (AML M4), MUTZ-8 (MDS-AML M4), MUTZ-11 (MDS-AML M4), MV4;11 (AML M5), NALM-1 (CML BCP BC),

JAK2 V617F mutation in cell lines H Quentmeier et al

472 NALM-6 (BCP ALL), NB-4 (AML M3), NS-MEG (CML meg BC), OCI-AML2 (AML M4), OCI-AML3 (AML M4), OCI-AML5 (AML M4), OCI-M2 (MDS-AML M6), PF-382 (T-ALL), REH (BCP ALL), RPMI-8402 (T-ALL), RS4;11 (BCP ALL), SET-2 (ETleukemic transformation), SIG-M5 (AML M5a), SKH-1 (CML meg BC), SKM-1 (MDS-AML M5), SUP-B15 (BCP ALL), TALL-1 (T-ALL), TANOUE (B-ALL), TF-1 (AML M6), THP-1 (AML M5), TOM-1 (BCP ALL), TS9;22 (CML meg BC), U-937 (histiocytic lymphoma), UKE-1 (ET-AML), UT-7 (AML M7), YS9;22 (CML meg BC), and YT (T-ALL).

Mutational analysis for detection of JAK2 V617F mutation JAK2 mutations were identified using the amplification-refractory mutation system (ARMS).5 This assay uses four primers specifically to detect the mutant (279 bp) and wt (229 bp) versions of the gene plus a positive control band (463 bp). Polymerase chain reaction (PCR) primers were: forward outer (FO) 50 -TCCTCAGAACGTTGATGGCAG-30 , reverse outer (RO) 50 -ATTGCTTTCCTTTTTCACAAGAT-30 , forward wt specific (FWT) 50 -GCATTTGGTTTTAAATTATGGAGTATaTG-30 , reverse mutant specific (RMU) 50 -GTTTTACTTACTCTCGTCTCCA CAaAA-30 . Unless stated otherwise, all primers used in this study were purchased from MWG Biotech (Ebersberg, Germany). Genomic PCR amplification was performed with a DNA thermal cycler (Perkin-Elmer Cetus, Heidelberg, Germany) under the following conditions: 30 s at 941C for denaturation, 30 s at 581C for annealing and 2 min at 721C for extension. In the first of 36 cycles, HotStart Taq polymerase (Qiagen, Hilden, Germany) was activated for 15 min at 951C. To verify the JAK2 1849G4T mutation, the 463 bp products from PCR with outer primers FO and RO were sequenced (MWG Biotech). Additionally to confirm the JAK2 wt or mutant status and to distinguish between homozygous or heterozygous mutations, JAK2 463 bp PCR products were treated with BsaXI (New England Biolabs, Frankfurt, Germany). The BsaXI site in the wt version of JAK2 is removed by the G4T transition.1

Cytogenetic analysis Karyotypic analysis and FISH were performed as described previously.15 Tilepath bacterial artificial chromosome (BAC) clones were sourced from BAC-PAC Resources (Children’s Hospital, Oakland, CA, USA), and chromosome painting probes from Cambio (Cambridge, UK). Probe preparation and labelling were as described previously.15 Imaging and analysis were performed using an Axioscope 2 fluorescence microscope system (Zeiss, Goettingen, Germany) and Smart Capture 2 software (Applied Imaging, Newcastle, UK).

Microsatellite analysis To verify whether JAK2 V617F mutations were paralleled by LOH at chromosome 9p, we tested six microsatellite markers for heterozygosity.4 The size of the PCR products was verified by capillary gel electrophoresis (CEQ 8000, Beckmann Coulter, Krefeld, Germany) as previously described.16 Asterisks indicate positive labelling for the fluorescent dye D4. Labelled oligonucleotides were obtained from Proligo (Paris, France).

D9S288

Leukemia

sense 50 -GTTTCTTAGCAACCTCAACAGGG-30 antisense 50 -*AATCATCCAGAAAGGCCA-30

D9S1810 D9S1681 D9S1852 D9S925 D9S161

sense 50 -GTTTCTTCTGACAGCAGAGCATCC-30 antisense 5´-*CAAGCAAAACTTTTTATTGTGA-30 sense 50 -GTTTCTTCAGATTCAGCCATGTTC-30 antisense 50 -*AGGCAGTTGCACAGATAG-30 sense 50 -GTTTCTTGAATCACAACATACACCCAC-30 antisense 50 -*GAAACATTCTTTTACAAGTAACATT-30 sense 50 -*TGTGAGCCAAGGCCTTATAG-30 antisense 50 -GTCTGGGTTCTCCAAAGAAA-30 sense 50 -*TGCTGCATAACAAATTACCAC-30 antisense 50 -CATGCCTAGACTCCTGATCC-30

Genomic PCR amplification was performed for 35 cycles with Takara Taq polymerase (Cambrex, Verviers, France) under the following conditions: 30s at 941C for denaturation, 30s at 531C for annealing and 2 min at 721C for extension.

[3H]Thymidine uptake and detection of apoptotic cells

Assays of [3H]thymidine incorporation were executed as follows: 2.5 104 cells (in 100 ml) were seeded in triplicate in 96-well flat-bottom microtiter plates. JAK inhibitor I (Merck Biosciences, Bad Soden, Germany) was added as 2 concentrated solution in a 100 ml volume. For the last 3 h of the incubation period, 1 mCi [3H]thymidine (Amersham Pharmacia Biotech, Freiburg, Germany) was added to each well. Cells were harvested as described.17 For cell cycle analysis, cells were fixed with 70% ethanol (201C, 20 min on ice), washed with phosphate-buffered saline, and stained with propidium iodide (PI) (20 mg/ml). DNA content of the cells was determined by flow cytometry on the FACScan (Becton Dickinson, Heidelberg, Germany). Apoptotic cells were detected and quantitated with the annexin-V/PI method, applying a commercially available kit (R&D Systems, Wiesbaden, Germany). Binding of fluorescein isothiocyanate-labeled annexin-V and PI staining of the cells was determined by flow cytometry.

Results and discussion

Cell lines with JAK2 V617F mutations It has recently been shown that the majority of patients with PV, and substantial numbers of patients with IM and ET, carry a single nucleotide mutation in the JAK2 gene.1–6 The JAK2 V617F mutation was also found in patients with other MPD and the megakaryoblastic variant of AML, although at lower percentages.6–8 Since the JAK2 point mutation may contribute to the etiology of these diseases, elucidation of its cellular function(s) may reveal potential therapeutic targets. In numerous cases, the effects of a mutationally altered protein on key cellular functions, including proliferation, differentiation or apoptosis, have been revealed using immortalized cell lines that carry the mutation in question. To find such a model system for JAK2 V617F in hematopoietic neoplasia, we tested a panel of 42 cell lines derived from patients with MPD, MDS and AML (M7) and compared these with 37 cell lines representing other types of AML (without reportedly antecedant MPD or MDS) and ALL. Cell lines were tested for JAK2 V617F expression applying the PCR-based ARMS system, a method that specifically amplifies the normal and mutant sequences plus a positive control band in a single reaction.5 According to ARMS and direct sequencing of genomic PCR products, 5/79 cell lines carried the JAK2 V617F mutation (Figure 1a and b): MB-02 derived from a patient with a history of IM and myeloid metaplasia that developed into AML M7;14 MUTZ-8 established from a patient with AML M4 25 years after the onset of MDS;18 SET-2 from a patient at the stage


473

Figure 1 JAK2 V617F mutations in continuous cell lines. (a) ARMS assay with JAK2wt (229 bp), JAK2mu (279 bp) plus JAK2 PCR control (463 bp). Note that in a 1:1 mu/wt DNA mix, the 229 bp JAK2wt PCR product is under-represented. (b) Sequencing of JAK2 463 bp fragments. Sequencing was performed with reverse primer RO; therefore, JAK2 1849G4T mutation is seen as C4A transition. Cell lines HEL, MB-02 and UKE-1 show the same results as cell line MUTZ-8. (c) BsaXI digestion of JAK2 463 bp fragments. BsaXI site in JAK2 is destroyed by the point mutation. Note that JAK2wt signal in cell line SET-2 are weaker than JAK2mu signals in all three assays.

of leukemic transformation of ET;19 and UKE-1 from a patient with ET, which transformed into acute leukemia.20 The AML M6-derived cell line HEL was the only JAK2 V617F positive cell line tested lacking a known patient-history of MPD or MDS.14 However, like other JAK2mu cell lines found in this study, also this cell line showed cytogenetic rearrangements, which are considered hallmarks of AML/MDS:21 these included losses of long-arm material from both chromosomes 5 or 7 (5q/7q) affecting cell lines HEL, MB-02, MUTZ-8, and SET-2 (Figure 2a– d), while cell line UKE1 displayed monosomy for chromosome 7 (data not shown). In the case of cell line HEL, formation of 5q/ 7q may be related to treatment received for preceding Hodgkin lymphoma.14 Coincident 5q and 7q occur rather rarely in AML, o10% according to Mitelman’s database (http:// cgap.nci,nih.gov/Chromosomes/Mitelman), and their presence in 4/5 JAK2mu cell lines is noteworthy, raising the intriguing possibility of their nonrandom association. Sequencing of JAK2 PCR products suggested that cell line SET-2 expressed the wt and the mutant allele of JAK2, the four other cell lines expressed the mutant allele only (Figure 1b). These results were confirmed by analyses using the JAK2 wt specific restriction enzyme BsaXI (Figure 1c). JAK2 V617F mutations occur often in MPD patients with PV, ET and IM, but not in patients with Philadelphia-positive CML.1–6 Accordingly, 3/5 cell lines from patients with a history of ET and IM, and 0/17 CML cell lines expressed the JAK2 point mutation.

Figure 2 Cytogenetic analysis. Images (a–d) depict FISH with wholechromosome library probes for chromosomes 5 (red) and 7 (green) and show consistent losses of chromosomes 5 and 7 long-arm material in MB-02, MUTZ-8, HEL and SET-2 cells, while UKE-1 cells showed monosomy for chromosome 7 (data not shown). Note that all cell lines are near-diploid apart from HEL which is near triploid. FISH images (e–h) depict results of cohybridizing a BAC clone (RP11-39k24) covering the JAK2 locus (green) and a library probe for chromosome 9 (red). Arrows indicate JAK2 signals. Note quasi-normal diploid hybridization patterns for MB-02 and MUTZ-8 (similar data for UKE1 not shown), while HEL and SET-2 display selective JAK2 chromosomal amplification and concomitant deletion of the residual homolog, either wholly (HEL) or in part (SET-2). Note triple arrows (g) indicating multiple JAK2 signals due to genomic amplification in HEL cells.

The JAK2 V617F mutation has been described as an infrequent event in ‘atypical MPD’, MDS and in megakaryocytic leukemia.6–8 We tested 20 cell lines from patients with a history of CNL (n ¼ 1), hypereosinophilic syndrome (n ¼ 1), myeloid metaplasia (n ¼ 1), MDS (n ¼ 8) and AML (M7) (n ¼ 9): of these, Leukemia


474 Ad (i): none of the four homozygously JAK2mu cell lines display declining heterozygosity at more distant microsatellite markers D9S925 and D9S161, 13 and 22 Mbp centromeric of JAK2 (Table 1), respectively, contrasting with previous reports invoking mitotic crossing over where heterozygosity rises from 21 to 66% over the same distance.4 Ad (ii): FISH analysis of JAK2 V617F cell lines with a panel of BAC clones covering almost the entire short-arm of chromosome 9 revealed wide discrepancies in both copy numbers and chromosomal configurations. While MB-02, MUTZ-8 and UKE1 cells yielded apparently normal diploid signals (Figure 2e and f; Table 2), HEL and SET-2 cells displayed patterns consistent with both genomic amplification and deletion (Figure 2g and h). In the case of SET-2 cells, JAK2 amplification occurred via formation of a series of small marker chromosomes, including an isochromosome which together increased the JAK2 copy number three-fold over diploid levels (Figure 2h). No normal chromosome 9 homolog was present, replaced by a singleton del(9)(p11p23–24) from which material centromeric of clone 175e13 (8.4 Mbp) was deleted (Figure 2h; Table 2). Thus, LOH centromeric of JAK2 was directly attributable to a combination of partial 9p deletion and selective JAK2 amplification in SET-2 cells. In triploid HEL cells, similar processes effected B10-fold (diploid-corrected) JAK-2 amplification via an estimated 8-fold tandem duplication, affecting each of a pair of closely related marker chromosomes (Figure 2g; Table 2). Again no normal homolog was present explaining LOH in this cell line also. Comparison with the distinct DAMI subclone of HEL, isolated at early passage,22 revealed identical 9p configurations, including JAK2 amplification, consistent with primordial JAK2 rearrangement (data not shown).

MUTZ-8 was the sole example which expressed the JAK2 mutation. The JAK2 mutation had not been reported in AML (M1–M6)5,6 or ALL.6,8 We found no JAK2 V617F in ALL cell lines (0/19), and only in 1/18 AML (M1–M6) cell lines. Cell line HEL alone tested positive for JAK2 V617F of 18 AML (M2–M6) cell lines without a known background of MPD/MDS. However, as pointed out, also this cell line carries chromosomal stigmata of MDS similar to the remaining JAK2 V617F positives described in this study. In conclusion, the percentages of cell lines carrying the JAK2 V617F mutation reflect the situation in the primary diseases: 3/5 (60%) cell lines from patients with a history of ET or IM, 1/20 (5%) cell lines from patients with ‘atypical MPD’, MDS and AML M7 (without history of IM), 0/17 CML cell lines, 0/16 ALL cell lines, and 1/18 AML (M1–M6) cell lines expressed the JAK2mu.

Chromosome 9p LOH About 40% of JAK2 V617F positive MPD patients express the homozygous mutation.2,4,5 We observed that also 4/5 JAK2mu cell lines (80%) show exclusive expression of the mutant variant of the JAK2 gene (Figure 1). Microsatellite analysis showed that these cell lines display chromosome 9p21/p22 LOH (Table 1). LOH might result from losses of chromosome 9, either wholly (monosomy), or in part (9p), or by mitotic recombination. The current view is that mitotic recombination effects partial chromosome 9 LOH in JAK2mu MPD.1,2,4 However, the combined results of microsatellite (i) and FISH (ii) analyses provide compelling evidence for alternative mechanisms underlying LOH.

Table 1

Chromosome 9 LOH according to microsatellite analysis

Cell line

JAK2

D9S288 (p24.2) 3,941792 [0.81]

D9S1810 (p24.2) 4,817678 [0.81]

D9S1681 (p24.1) 5,276071 [0.31]

D9S1852 (p24.1) 6,225983 [0.5]

D9S925 (p22.1) 18,279058 [0.81]

D9S161 (p21.2) 27,622327 [0.63]

HEL MB-02 MUTZ-8 SET-2 SIG-M5 TF-1 UKE-1

mu mu mu mu/wt wt wt mu

1 1 1 1 2 2 1

1 1 1 1 2 2 1

1 1 1 2 1 2 1

1 1 1 1 2 2 1

1 1 1 1 2 2 1

1 1 1 0 2 2 1

The five JAK2mu cell lines and two JAK2wt control cell lines chosen at random were analyzed. Shown are the numbers of alleles for six microsatellites on chromosome 9p. Location of microsatellites is indicated in brackets. JAK2 is located at 9p24.1 (5,011988), between D9S1810 and D9S1681. D9S925 and D9S161 are, respectively, located 13 and 22 Mbp centromeric of JAK2. Database estimates of heterozygosity (HET) for the microsatellites are given in square brackets. The low HET value may explain D9S1681 homozygosity in the JAK2 wild-type cell line SIG-M5.

Table 2

Copy number and configuration of Chromosome 9p in JAK2 mutant cell lines

Cell line

HEL MB-02 MUTZ-8 SET-2 UKE-1

Clones (distance) 43n6 (0)

31f19 (0.6)

+ +/+ +/+ +/+/ ++ +/+

+/+ +/+ +/+ +/+/++ +/+

125k10 (4.9)

39k24 (5.1)

12d24 (5.4)

+++/+++ +++/+++ +++/+++ +/+ +/+ +/+ +/+ +/+ +/+ +/+/+/++ +/+/+/++ +/+/+/++ +/+

+/+

+/+

207c16 (5.7)

307l3 (6.5)

175e13 (8.4)

264o11 (10.6)

149i2 (21.9)

CTD-3002e24 (37.0)

+++/+++ +/+ +/+ +/+/+/++

+++/+++ +/+ +/+ +/+/+/++

+++/+++ +/+ +/+ +/+/+/++

+++/+++ +/+ +/+ ++

+/+ +/+ +/+ +

+/+ +/+ +/+

+/+

+/+

+/+

+/+

+/+

+/+

Table shows quantity of FISH signals along 9p in cell lines indicated by: no signal; +, singleton; ++, doublet; and +++, X8 signals present on individual chromosomal elements separated by ‘/’. Clones derive from RP11 library unless otherwise indicated. Distances in Mbp from 9p telomere are from UCSC genome browser May 2004 assembly. Leukemia


475 228,017:201143,321 200000

175000

Dye Signal

150000

125000

100000

75000

50000

25000

0 230

240

250

260

270

280

290

300

310

320

222,084:164515,308 175000

150000

Dye Signal

125000

100000

75000

50000

25000

0 230

240

250

260

270

280

290

300

310

320

Figure 4 JAK inhibition. (a) 3H-labeled thymidine incorporation after 2 days with JAK inhibitor I: JAK2mu cell lines are more sensitive to JAK inhibition than JAK2wt cell lines (IC50 of JAK2mu cell lines: HEL 0.5 mM, SET-2 0.1 mM, UKE-1 0.15 mM; IC50 of JAK2wt cell lines: K-562 10 mM, SKM-1 10 mM). (b) Annexin-V staining after 1 day with JAK inhibitor I: JAK inhibitor I induces apoptosis in JAK2mu cell lines HEL, SET-2 and UKE-1, but not in JAK2wt cell lines K-562 and SKM-1.

271,423:200227,054 200000

and deletion of the homologous chromosome, either in part, sparing JAK2wt (SET-2), or in entirety (HEL). While the data on remaining cell lines (MB-02, MUTZ-8, UKE-1) are less unequivocal, the absence of any differential LOH along 9p therein provides no positive support for mitotic crossing over, again leaving open the possibility of deletion/duplication.

175000

Dye Signal

150000

125000

100000

75000

50000

Inhibition of JAK2

25000

0 230

240

250

260

270

280

290

300

310

320

Figure 3 Microsatellite analysis. JAK2mu cell line MB-02 shows D9S1681 LOH, JAK2wt cell line TF-1 and JAK2mu/wt cell line SET-2 show heterozygous expression of D9S1681. Note that one allele is underrepresented in cell line SET-2 (arrow).

Microsatellite analysis confirmed that in SET-2 cells, one of the JAK2 loci was amplified: the signal intensity of one allele of the JAK2 adjacent microsatellite D9S1681 was much higher than that of the other (Figure 3). JAK2 sequencing and BsaXI digestion experiments confirmed these results and suggested furthermore that it was the mutant variant of JAK2 that was amplified as implied by the cytogenetic findings: in both assays, signal intensities for JAK2mu were distinctly higher than for JAK2wt (Figure 1b and c). Thus, results of FISH analysis in cell lines HEL and SET-2 suggest that homozygous expression of JAK2mu may not only be the result of mitotic recombination, but also the consequence of loss and amplification of chromosome 9p21. Taken together, the cytogenetic and LOH data are not generally explicable by mitotic recombination alone, and instead favor a combination of selective JAK2mu amplification

JAK2 protein was expressed in JAK2mu and JAK2wt cell lines likewise (data not shown). It had been reported that JAK inhibitor I downregulated the phosphorylation of JAK2 in the JAK2mu cell line HEL.2 We show that at least 20 lower concentrations of the inhibitor were necessary for growth inhibition in JAK2mu than in JAK2wt cell lines (Figure 4a). Likewise, apoptosis was induced at lower doses in JAK2mu than in JAK2wt cell lines (Figure 4b). These data suggest that the JAK2 V617F mutated kinase has an important effect on cell growth and prevention of apoptosis.

Conclusion To devise targeted, ‘intelligent’ therapies, the pathobiology of MPD requires elucidation. Although clinically related, MPD are biologically diverse, demanding a wide range of well characterized in vitro models. Hitherto only one model of JAK2 V617F mutation, HEL cells derived from erythroleukemia, has been available. We describe a further four cell lines, MB-02, MUTZ-8, SET-2, UKE-1 – all established from MPD/MDS patients to serve as appropriate models of JAK2 V617F mutation. We have in addition shown that selective JAK2mu amplification combined with whole or partial deletion of the residual homolog may Leukemia


476 underly some instances of 9p LOH, accompanying formation of JAK2 V617F.

11

References 1 Baxter EJ, Scott LM, Campbell PJ. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet 2005; 365: 1054–1061. 2 Levine RL, Wadleigh M, Cools J, Ebert BL, Wernig G, Huntly BJP et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thromobocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell 2005; 7: 387–397. 3 James C, Ugo V, LeCouedic JP, Staerk J, Delhommeau F, Lacout C et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature 2005; 434: 1144–1148. 4 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. 5 Jones AV, Kreil S, Zoi K, Waghorn K, Curtis C, Zhang L et al. Widespread occurrence of the JAK2 V617F mutation in chronic myeloproliferative disorders. Blood 2005; 106: 2162–2168. 6 Jelinek J, Oki Y, Gharibyan V, Bueso-Ramos C, Prchal JT, Verstovsek S et al. JAK2 mutation 1849G4T is rare in acute leukemias but can be found in CMML, Philadelphia-chromosome negative CML and megakaryocytic leukemia. Blood 2005; 106: 3370–3373. 7 Steensma DP, Dewald GW, Lasho TL, Powell HL, McClure RF, Levine RL et al. The JAK2 V617F activating tyrosine kinase mutation is an infrequent event in both ‘atypical’ myeloproliferative disorders and the myelodysplastic syndrome. Blood 2005; 106: 1207–1209. 8 Levine RL, Loriaux M, Huntly BJP, Loh ML, Beran M, Stoffregen E et al. The JAK2V617F activating mutation occurs in chronic myelomonocytic leukemia and acute myeloid leukemia, but not in acute lymphoblastic leukemia or chronic lymphocytic leukemia. Blood 2005; 106: 3377–3379. 9 Parganas E, Wang D, Stravopodis D, Topham DJ, Marine JC, Teglund S et al. Jak2 is essential for signaling through a variety of cytokine receptors. Cell 1998; 93: 385–395. 10 Peeters P, Raynaud SD, Cools J, Wlodarska I, Grosgeorge J, Philip P et al. Fusion of TEL, the ETS-variant gene 6 (ETV6), to the receptor-associated kinase JAK2 as a result of t(9;12) in a lymphoid

Leukemia

12

13 14 15 16

17

18 19

20

21 22

and t(9;15;12) in a myeloid leukemia. Blood 1997; 90: 2535–2540. Nunez CA, Zipf TF, Roberts WM, Medeiros LJ, Hayes K, BuesoRamos CE. Molecular monitoring of cerebrospinal fluid can predict clinical relapse in acute lymphoblastic leukemia with eosinophilia. Arch Pathol Lab Med 2003; 127: 601–605. Reiter A, Walz C, Watmore A, Schoch C, Blau I, Schlegelberger B et al. The t(8;9)(p22;p24) is a recurrent abnormality in chronic and acute leukemia that fuses PCM1 to JAK2. Cancer Res 2005; 65: 2662–2667. Drexler HG, Dirks W, MacLeod RAF, Quentmeier H, Steube KG, Uphoff CC. DSMZ Catalogue of Human and Animal Cell Lines, 8th edn. DSMZ: Braunschweig, Germany, 2001. Drexler HG. The Leukemia-Lymphoma Cell Line Factsbook. Academic Press: San Diego, CA, 2000. MacLeod RAF, Drexler HG. Cytogenetic analysis of cell lines. Methods Mol Biol 2005; 290: 51–70. Quentmeier H, Martelli MP, Dirks WG, Bolli N, Liso A, MacLeod RAF et al. Cell line OCI/AML3 bears exon-12 NPM gene mutationA and cytoplasmic expression of nucleophosmin. Leukemia 2005; 19: 1760–1767. Quentmeier H, Zaborski M, Drexler HG. Effects of thrombopoietin, interleukin-3 and the kinase inhibitor K-252a on growth and polyploidization of the megakaryocytic cell line M-07e. Leukemia 1998; 12: 1603–1611. Hu ZB, MacLeod RAF, Meyer C, Quentmeier H, Drexler HG. New acute myeloid leukemia-derived cell line: MUTZ-8 with 5q. Leukemia 2002; 16: 1556–1561. Uozumi K, Otsuka M, Ohno N, Moriyama T, Suzuki S, Shimotakahara S et al. Establishment and characterization of a new human megakaryoblastic cell line (SET-2) that spontaneously matures to megakaryocytes and produces platelet-like particles. Leukemia 2000; 14: 142–152. Fiedler W, Henke RP, Ergu¨n S, Schumacher U, Gehling UM, Vohwinkel G et al. Derivation of a new hematopoietic cell line with endothelial features from a patient with transformed myeloproliferative syndrome. Cancer 2000; 88: 344–351. Charrin C. del(5q) in myeloid malignancies. Atlas Genet Cytogenet Oncol Haematol 1998, URL: http://www.infobiogen.fr/services/ chromcancer/Anomalies/ del5qID1092.html. MacLeod RAF, Dirks WG, Reid YA, Hay RJ, Drexler HG. Identity of original and late passage Dami megakaryocytes with HEL erythroleukemia cells shown by combined cytogenetics and DNA fingerprinting. Leukemia 1997; 11: 2032–2038.