Gene Expression Profiling Distinguishes Essential Thrombocythemia from Polycythemia Vera Patients and Identifies a Common Expressed Set of Genes in Relation to JAK2V617F Status

Introduction. The existence of the JAK2V617F mutation in a high proportion of Myeloproliferative Disorders (MPD) BCR-ABL-negative has provided important insight into the pathogenesis of these diseases. However, much of the molecular abnormalities associated to BCR-ABL-negative MPD remain unknown, specially in those which do not display JAK2V617F. In a previous study, we performed gene expression analysis by using the microarray technique in 20 essential thrombocythemia (ET) patients (44K whole human genome oligo microarrays, Agilent Technologies) and the results were confirmed in 40 ET patients by using TaqMan® Low Density Arrays Arrays (LDA, Applied Biosystems). In our previous experience the results showed different gene expression patterns in ET and a supervised clustering of the data identified genes differentially expressed between JAK2V617F-negative and JAK2V617F-positive ET patients, and a characteristic gene expression profile for JAK2V617F-negative patients (Puigdecanet et al.,2008).

Aim. Our aim was to confirm the ET gene expression profile in an extended group of patients and to explore the differences and similarities in polycythemia vera (PV) and reactive thrombocytosis (RT) patients by real-time quantitative RT-PCR (RQ-PCR) technique using the LDA platform. In addition, we wanted to analyze the relationship between gene expression data and JAK2V617F status.

Patients and Methods. The following patients were included in the study: 58 ET (23 JAK2V617F-negative, 34 JAK2V617F in heterozygosity and one JAK2V617F in homozygosity) and 41 PV (7 JAK2V617F-negative, 25 JAK2V617F in heterozygosity and 9 JAK2V617F in homozygosity) patients, diagnosed according to the WHO criteria (2001) and who had never received cytoreductive treatment, and six patients with RT. Based on the previous results, we designed a new LDA platform containing 96 assays in duplicate, which included the most expressed genes in ET in relation to healthy controls and the most differentially expressed genes between JAK2V617F-negative and JAK2V617F-positive ET patients. The RQ-PCR analysis was performed in RNA from peripheral blood granulocytes and the relative gene expression quantification was achieved using GAPDH as the endogenous control and a pool of 10 healthy individuals as the calibrator.

Results. ETvsPV: The majority of the genes studied presented significant higher expression in ET than in PV patients. Interestingly, FOSB was one of the most differentially expressed gene (FC= 8.3), and CISH and C13orf18 did not distinguish between the two groups.

ET: We confirmed the differentially expression of the majority of the genes previously detected between JAK2V617F-negative and JAK2V617F-positive ET patients and we extended the set of genes. Among them, we highlight the differential expression of CISH, FOSB and C13orf18 genes (p<0.01).

PV: Supervised analysis showed that CD44, BATF and CISH clearly distinguish JAK2V617F-negative PV patients from the JAK2V617F-positive.

ET and PVvsRT: Some differentially expressed genes between ET and RT patients were detected, but the most significant gene was TNF (p<0.001), which presented a higher expression in RT (FC=5.9). The same difference was observed between PV and RT.

Conclusions. We have detected a different gene expression pattern in ET and in PV patients. However, we also identified a set of genes which expression was related to JAK2V617F status, both in ET and PV patients. These findings would be interesting to identify other signal transduction pathways besides JAK-STAT involved in the pathogenesis of ET and PV.