JAK2V617F-negative ET patients do not display constitutively active JAK/STAT signaling

Abstract

Objective: Presence of the JAK2V617F mutation in only 40% to 60% of patients with essential thrombocythemia (ET) underscores the heterogeneity of this myeloproliferative disorder (MPD). Several distinct mutations, either in JAK2 (exon 12) or in c-Mpl (W515L) have been described in subsets of other MPDs, polycythemia vera, and idiopathic myelofibrosis. Analogous to JAK2,V617F these mutations cause constitutive JAK2 and signal transducer and activation of transcription (STAT) activation. It has therefore been proposed that constitutive activation of the JAK/STAT pathway underlies the molecular etiology of all MPDs. We investigated the alternative hypothesis that distinct alterations, separate from the JAK/STAT signal transduction pathway, underlie a subset of JAK2V617F-negative ET.

Methods: cDNA microarrays and quantitative reverse transcriptase polymerase chain reactions were used to compare gene expression in 40 ET patients with and without the JAK2V617F mutation.

Results: Unsupervised clustering of gene-expression patterns in ET patients revealed two distinct subclasses of patients. These subclasses differed in presence or absence of the JAK2V617F mutation. Patients lacking the JAK2V617F mutation displayed significantly lower expression of the JAK/STAT target genes Pim-1 and suppressor of cytokine signaling-2. In addition, JAK2V617F-negative patients showed lower levels of STAT3 phosphorylation.

Conclusions: These data demonstrate that a large proportion of JAK2V617F-negative ET patients do not display constitutive JAK/STAT signaling. Hence, we propose that alterations in different signal transduction pathways can lead to the clinical phenotype of ET. Elucidation of novel ET-inducing changes will facilitate both a molecular classification of ET and development of rationally designed therapies.

Figure 1. Experimental Design

RNA was isolated from purified granulocytes of individual ET patients. Presence of the JAK2^V617F mutation was determined for each patient by qRT-PCR. A control pool was generated consisting of RNA from isolated granulocytes of 50 healthy volunteers. Individual patient RNA, labeled with one fluorochrome, together with this control pool, labeled with a second flurochrome, was hybridized to cDNA arrays. Data analysis reveals genes whose expression is increased or decreased in ET patients relative to healthy controls. A two-sample t-test used to identify genes, which are differentially expressed between patients bearing the JAK2^V617F mutation and those which do not (see Fig. 2).

Fig. 2. Gene Expression Profiling in Patients with ET

A cohort of 16 ET patients was analyzed for gene expression by cDNA microarray, as detailed in Fig. 1. A two-sample t-test was used to identify genes, which are differentially expressed between patients bearing the JAK2^V617F mutation (shown in red) and those negative for JAK2^V617F (shown in green). To control for multiple testing the obtained p-values are adjusted for the false discovery rate (fdr) [22, 23]. Differential expression was defined by p<0.05 (fdr corrected). Agglomerative hierarchical clustering was used to generate the dendrogram [37]. Columns represent individual patients and rows represent specific cDNAs identified by accession number. Gene expression is depicted according to the color scale shown below the figure. “1” indicates no difference in expression between patient and healthy controls, whereas “10” and “100” or “1/10” and “1/100” indicate the fold over- or underexpression respectively. White squares depict missing data points. Accession numbers of genes, whose transcription is regulated by the JAK7STAT pathway are underlined.

Figure 3. Validation of gene expression in JAK2^V617F and JAK2^wt ET patients by Quantitative RT-PCR

RNA was isolated from purified granulocytes of JAK2^V617F-positive or JAK2^V617F-negative ET patients and healthy controls as indicated and subjected to quantitative RT-PCR analysis for (A) Pim-1 and (B) SOCS-2 expression. A standard curve with known copy numbers of Pim-1 (A) or SOCS-2 (B) respectively and 18S rRNA was included on each plate. Sample copy numbers of target genes and 18S rRNA were determined from the standard curve and are expressed as relative ratios (molecules target gene per 10⁶ 18S molecules). The median is depicted by a vertical line; **, p< 0.001, *** p< 0.0001

Figure 4. Protein Phosphorylation in ET Patients with and without the JAK2^V617F mutation

Total cellular protein was isolated from purified granulocytes of 8 ET patients, not previously used for microarray analysis. 30 μg of protein were subjected to Western Blot and hybridized with an antibody against (A) phospho STAT3 (top) or total STAT3 (bottom) (B) phospho AKT (top) or total AKT (bottom) or (C) phospho ERK1/2 p44/p42 MAP kinases (top) or total ERK1/2 p44/p42 MAP kinases (bottom).