To the Editor, Much attention has been focused on the attempt to undertake global analysis of protein expression and in particular to analyze the changes in protein expression associated with several processes such as differentiation and cell cycle progression. In general, cell differentiation— or proliferation—requires de novo protein synthesis and/or post-translational modification, in particular phosphorylation of proteins such as transcription factors. Consequently, there is a need to identify the proteins that are being modified, as well as to know when and where in the cell modifications occur. We focused our studies on the Chinese hamster ovary (CHO) mammalian tumor cell line, a good model to study the cell differentiation. It is known that cyclic 30,50monophosphate adenosine (cAMP) transforms CHO cells, in vitro: they lose characteristics associated with malignancy and adopt characteristics of normal fibroblasts (‘‘reverse transformation’’) [3, 7, 9]. Puck et al. [8–10] found that cAMP regulates the structure of the cell cytoskeleton, which in turn strongly influences many differentiation properties including the assumption of normal or cancerous behavior in mammalian cells. In a previous study, we analyzed how CHO-K1 cells display different nuclear protein contents when undergoing differentiation by cAMP [14]; our Mass Spectrometry (MS) investigation revealed a change in the composition of nuclear proteins associated to an inhibition of the protein expression with possible implications for the control of cell reverse transformation [14]. The mechanism of the transformation, however, is as yet obscure and need further exploration in all compartments of cell proteome which we have carried out with this study. Fundamental changes in properties of the cell membrane have indeed been reported but the molecular mechanisms generating these structural changes are largely unknown. The changes in the concentration of the different metabolites and the different protein composition associated with morphological changes observed are also largely unknown [8]. This study wants to explore more deeply the effect of cAMP differentiation on the proteome of CHO cells, considering that the changes in the levels of nuclear envelopeassociated proteins and chromatin-condensing proteins [14] constitutes only the first step toward developing an understanding of cellular mechanisms for controlling nuclear shape. Due to the complexity of eukaryotic proteomes, and the divergence of protein properties, it is beneficial to prepare standardized partial proteomes in order to maximize the coverage of the proteome and to increase the chance to visualize low abundance proteins and make them accessible for subsequent analysis [1]. The strategies used to prepare such a complex proteomic samples for MS analysis involve many steps. For protein extraction we chose to perform a subcellular fractionation to separate the whole cellular proteome in four fractions. In literature, there are several works in which this approach to proteome analysis is adopted [1, 6, 12, 13], but always to analyze the behavior of a restrict number of proteins, for example to investigate changes in subcellular localization of regulatory proteins impacted by experimental or disease parameters. R. Spera C. Nicolini Nanoworld Institute, Fondazione El.B.A. Nicolini, Largo Redaelli 7, Pradalunga, Bergamo, Italy e-mail: rspera@ibf.unige.it