Analysis of B-CLL Transcriptomic and Proteomic Profiles: Differences between Molecular Subgroups.

INTRODUCTION: B cell chronic lymphocytic leukemia (B-CLL) is a lymphoproliferative disorder with a variable clinical course. Patients (pts) with unmutated (unmut) IgV_H gene show a shorter progression free survival and overall survival than the patients with IgV_H mutated (mut). To understand the differences between molecular subgroups of B-CLL we have studied transcriptomic and proteomic profiles on samples from 40 B-CLL pts (Binet stage A).

MATERIAL AND METHODS: 100 μg of total PBMC proteins were used for IEF followed by 2D electrophoresis. Image analysis of scanned gels was used to identify statistically differentially expressed proteins. Image acquisition and analysis were performed using the Ludesi software (http://www.ludesi.com). Selected spots were subjected to automatic digestion and the proteins were identified by MALDI-TOF (Voyager DE-Pro, Applied Biosystems) peptide mass fingerprint using the Protein Prospector software. To confirm the initial results, sequencing of selected peptide ions was carried out by collision-induced dissociation (CID) with a nESI-QTRAP mass spectrometer from Applied Biosystems. Eight proteins were validated by Western Blot.

Total RNA from B cells was used to analyze the expression profile by hybridization with Whole Human Genome U133 Plus 2.0 Array from Affymetrix. Normalization, differential gene expression and functional annotations were analyzed using the GEPAS suite (http://www.gepas.org). qPCR using TaqMan primers/probes was used for validation of the microarray data

RESULTS: When we compared IgV_H mut vs unmut transcriptomic and proteomic profiles, we found more than 600 differentially expressed genes and 12 proteins ( fdr <0.05 adjusted for multiple test contrast and p<0.05, Mann-Whitney’s test, for gene and proteins, respectively). In tables 1 and 2 we show some of the most differentially expressed gene/proteins in each group of pts. Validation of results from microarrays and proteomic data using qPCR and Western blot are in progress. We obtained positive correlation between transcriptomic and proteomic profiles, (corr=0.21, p=0.04, Pearson’s correation test) suggesting that common features are found using both approximations.

CONCLUSION: We found a number of interesting gene/proteins that could be able to differentiate molecular subgroups of B-CLL pts. The study of these proteins and genes may lead to better understand the different clinical behaviour of IgV^H mut and unmut B-CLL forms, but validation with a larger group of pts is still necessary.

Table 1: Genes differentially expressed