Gene Expression Profile of Protein Kinases Reveals a Distinctive Signature of Chronic Lymphocytic Leukemia (CLL) and Points to a Role of Second Generation Protein Kinase Inhibitors.

Background. Chronic lymphocytic leukemia (CLL) is characterized by the accumulation of relatively mature B cells and by a very variable clinical course. This clinical heterogeneity is sustained by different biologic parameters, such as the mutational status of the immunoglobulin variable genes (IgV_H), CD38 and ZAP-70 expression. In order to investigate the potential role of protein kinase (PK) inhibitors in CLL, we evaluated the gene expression profile of 1324 probesets annotated as PK using the HGU133 Plus2.0 Affymetrix arrays.

Methods. We evaluated 44 CLL and 137 acute lymphocytic leukemia (ALL) patients. Two additional sets of CLL (49 cases) were utilized to validate the results obtained. Probesets identified as PK genes were used for all the analyses, namely unsupervised clustering, Analysis of Variance (ANOVA) and t-test analysis. ANOVA was performed using a p-value of ≤0.001: further selection was performed retaining only those probesets whose mean expression level was ≥300 in at least one group and showed a fold change difference of ≥1.5 across all groups. Finally, to specifically identify genes differentially expressed between different subclasses of CLL, a t-test was applied: probesets were required to have a p-value ≤0.05 and a fold change>1.5.

Results. Unsupervised analysis, performed on CLL samples and different ALL subgroups, highlighted in CLL a unique and very homogeneous pattern characterized by the overexpression of a large set of PK; these results were further confirmed by ANOVA. Moreover, we identified 16 PK genes that were highly expressed in all 3 CLL sets analyzed. These genes codify for proteins with tyrosine kinase activity (SYK, LYN, BLK, LCK, JAK1, CSK and FGR), serin-threonin kinase activity (PIM2, PFTK1, TLK1, MAP4K1, PDPK1, PRKCB1 and STK10) or both (GRK6 and WEE1). Some of the selected genes are members of important protein kinase families, involved in cellular signaling, such as Src kinases (SFK), MAPK and JAK kinase family. PK expression was also analyzed in different CLL subclasses, subdivided according to different prognostic factors; in particular, we compared IgV_H mutated vs unmutated patients, CD38+ vs CD38- cases and, finally, ZAP-70+ vs ZAP-70- patients in the 3 experimental CLL sets. Comparison between IgV_H mutated vs unmutated cases highlighted a differential expression of ZAP-70 in all the 3 sets analyzed. Contrariwise, no PK was associated with the other prognostic parameters. Thus, these analyses did not show a specific signature associated with the abovementioned biologic features, suggesting that PK overexpression is specific of the disease itself rather than of CLL subclasses.

Conclusions. Our results show that CLL is characterized by a very peculiar PK signature and identify some potential molecular targets. Moreover, our findings indicate that a common mechanism of PK-mediated deregulation is operational in CLL cells, independently of other prognostic factors. Based on these pre-clinical data, we propose that second generation PK inhibitors may have a role in the management of all CLL patients.