Molecular mimicry in the chronic myeloproliferative disorders: reciprocity between quantitative JAK2 V617F and Mpl expression

Polycythemia vera (PV), idiopathic myelofibrosis (IMF), and essential thrombocytosis (ET) share their origin in a multipotent hematopoietic progenitor cell¹ ; cytogenetic abnormalities of chromosomes 1, 8, 9, 13, and 20² ; growth factor–independent in vitro hematopoietic colony formation³ ; and molecular abnormalities, including increased granulocyte PRV-1 mRNA expression,⁴  impaired megakaryocyte and platelet thrombopoietin receptor (Mpl) expression,⁵  and the acquisition of JAK2 V617F.^6-9  Because JAK2 has important biologic associations with regard to Mpl expression as well as signal transduction,¹⁰  we examined the relationship between JAK2 V617F and platelet Mpl expression in PV, IMF, and ET.

The study was approved by our institutional review board and written informed consent was obtained from all patients, in accordance with the Declaration of Helsinki. Diagnoses of PV and ET were based on the Polycythemia Vera Study Group criteria¹¹ ; the diagnosis of IMF was based on the Italian Consensus Conference.¹²  Genomic DNA was prepared from density gradient–purified granulocytes using the Puregene Cell kit (Gentra Systems, Minneapolis, MN). RNA was prepared from platelets isolated by differential centrifugation using the RNAeasy kit (Qiagen, Valencia, CA). Complementary DNA was prepared from platelet RNA using the TaqMan Reverse Transcription Reagent kit (Applied Biosystems, Foster City, CA) and random hexamers.

Platelet lysate (20 μg) was applied to 10% precast gradient gels (Bio-Rad, Hercules, CA), transferred, and immunoblotted with a rabbit polyclonal antiserum to the C-terminal of the Mpl protein and anti–glycoprotein IIB antiserum as described previously.¹³  Quantitative densitometry was performed using ImageJ software (National Institutes of Health, Bethesda, MD) by a blinded observer; Mpl normalized to IIB was expressed as the percent of 2 averaged controls on each blot. Sixteen secondary erythrocytosis patients defined the normal range, in which the mean was 98% and the range was 52% to 149%. The intra-assay correlation coefficient for replicates was 0.99, and the interassay coefficients of variation for the normal and low controls were 5% and 22%, respectively.

Custom TaqMan SNP Genotyping Assays (Applied Biosystems) for JAK V617F with templates from either genomic DNA or cDNA consisting of real-time polymerase chain reaction (PCR) primers flanking the mutant region plus 2 TaqMan MGB real-time PCR probes specific for the normal or mutant sequence (JAK2 genomic assay: forward primer, AAGCTTTCTCACAAGCATTTGGTTT and reverse primer, AGAAAGGCATTAGAAAGCCTGTAGTT; JAK2 cDNA assay: forward primer, AAGCTTTCTCACAAGCATTTGGTTT and reverse primer, CCAAATTTTACAAACTCCTGAACCAGAA; JAK2 genomic assay: probes, VIC-TCTCCACAGACACATAC and FAM-TCCACAGAAACATAC; JAK2 cDNA assay: probes, VIC-CTCCACAGACACATAC and FAM-TCCACAGAAACATAC) were performed on an Applied Biosystems Prism 7700 sequence detection system using the default PCR conditions. DNA mixtures corresponding to V617F allele percentages of 5%, 10%, 25%, 50%, 75%, and 90% were assay reference standards. Intra-assay replicates did not vary more than 5%, and interassay replicates did not vary more than 12%.

Statistical calculations were performed using SigmaStat (STSTAT Software, Point Richmond, CA).

Our study population consisted of a clinically well-defined group of PV, IMF, and ET patients followed for up to 30 years. In agreement with previous studies, we detected JAK2 V617F in neutrophil genomic DNA in 92% of PV (n = 92), 45% of ET (n = 84), and 42% of IMF (n = 19) patients.^6,7,14,15  There were no significant differences in age, disease duration, sex, or degree of splenomegaly between the JAK2 V617F–positive and –negative patients.

The percent of JAK2 alleles that are JAK2 V617F is related both to the extent of dominance of the JAK2 V617F–positive clone¹⁶  and also to the clone's JAK2 V617F genotype, which can be heterozygous or homozygous due to a mitotic recombination event.¹⁷  Therefore, we measured the percentage of JAK2 V617F alleles in PV, IMF, and ET neutrophils using a quantitative, allele-specific assay sensitive to the presence of 5% JAK2 V617F. We found that the median neutrophil JAK2 V617F allele percentage was greater in PV (67%; range, 35%-100%) than in ET (47%; range, 10%-63%; P < .001), and that allele percentages more than 63% were restricted to PV and IMF, a finding similar to some¹⁸  but not all¹⁹  studies. When stratified by disease duration, 100% neutrophil JAK2 V617F allele percentages were present in only 15% of PV patients less than 3 years from diagnosis compared with 40% of PV patients 3 to 10 years and 40% of PV patients between 10 and 26 years since diagnosis, although these differences were not significant (Figure 1A).

In PV patients, there was no difference between the median JAK2 V617F percentage in platelet cDNA and neutrophil genomic DNA (65% versus 61%, respectively) (Figure 1B). However, in ET, the median JAK2 V617F percentage was significantly higher in platelet cDNA than in neutrophil genomic DNA (42% versus 22%, respectively; P = .001) (Figure 1B). In 3 of 11 ET patients, JAK2 V617F was not detected in neutrophil genomic DNA but was easily detected in platelet cDNA. The differences in neutrophil JAK2 V617F allele percentage in ET compared with PV and the observation that JAK2 V617F can be detected in platelets only in ET suggest that factors such as the level of stem-cell commitment upon acquisition of JAK2 V617F or the degree of retained polyclonal hematopoiesis are related to differences in clinical phenotypes in JAK2 V617F–positive PV and ET.

JAK2 is the cognate tyrosine kinase of Mpl, and impaired megakaryocyte and platelet Mpl expression have been documented in many laboratories in the myeloproliferative disorders (MPDs).^20-22  Therefore, we examined the concordance of impaired platelet Mpl expression with disease phenotype and JAK2 V617F status in 152 MPD patients. Platelet Mpl expression was significantly different among the MPDs, with a median of 55% in ET (range, 1-246; reduced in 43%), 16% in PV (range, 1-227; reduced in 89%; P < .001 with respect to ET), and 5% in IMF patients (range, 1-22; reduced in 100%; P = .076 with respect to PV). In PV, median platelet Mpl expression was lower in patients with only the JAK2 V617F allele compared with patients with both the wild-type and mutant alleles (P = .001) (Figure 2). The median platelet Mpl expression of 98% in JAK2 V617F–negative ET patients was higher than the median of 28% in JAK2 V617F–positive ET patients (P = .001). Linear regression analysis of neutrophil JAK2 V617F allele percentage and platelet Mpl expression in the JAK2 V617F–positive cases demonstrated a significant inverse relationship (r =–0.36, P < .001) (Figure 2). However, while a negative correlation between neutrophil JAK2 V617F allele percentage and platelet Mpl expression existed, the JAK2 V617F allele percentage did not completely account for the variability of platelet Mpl expression. Of importance, the same association of impaired Mpl expression with an MPD phenotype occurred in JAK2 V617F–negative patients. This suggests that the same phenotype may develop as a consequence of different molecular mechanisms.

A previous study did not find a strong association between JAK2 V617F and impaired Mpl expression²³ ; however, the study was limited by a small number of Mpl measurements (n = 25), all of which were in ET patients, and neither neutrophil JAK2 V617F nor platelet Mpl expression was measured quantitatively. Impaired Mpl expression is in part related to defective Mpl processing, which may lead to retention in the Golgi or aberrant localization in the cytoplasm.¹³  Constitutively activated JAK2 may also reduce Mpl protein expression by causing receptor down-regulation. Indeed, the inverse correlation between neutrophil JAK2 V617F allele percentage and platelet Mpl reduction suggests that these are related abnormalities. Identification of other lesions in Mpl and the JAK signaling pathways in the JAK2 V617F–negative MPDs will further define the significance of aberrant Mpl protein expression to the molecular pathogenesis of the MPDs.

A.R.M. and D.M.W. designed and performed research and wrote the paper; O.R. designed and performed the research; J.L.S. designed the research and wrote the paper.

The authors declare no competing financial interests.

We thank Drs Sophie M. Lanzkron and Michael B. Streiff for patient recruitment.

This work was supported by the Myeloproliferative Disorders Foundation and the National Institutes of Health (RO1-HL082995).