During the past 13 yrs, our laboratory, firstly, and others have highlighted the presence of multiple nuclear signaling pathways involving phospholipid hydrolysis. A remarkable fact is that the changes induced by external stimuli on nuclear lipid pathways are not a mere duplication of those occurring at the plasma membrane level. Indeed, in almost all of the model systems which have so far been investigated, the modifications seen in nuclear lipid metabolism are different from those taking place at the plasma membrane or in other cell domains. These observations have strengthened the concept that autonomous lipid signaling systems operate within the nucleus, and their regulation is independent of similar events which occur elsewhere in the cell. The enzymes that are central intermediaries in nuclear lipid signal transduction pathways are the phospholipases C (PI-PLC), because from their activation results the generation of second messengers. In addition, the kinases for inositol lipids and inositol phosphates as well as phosphatases have been detected in the nucleus. Of the second messengers which are generated at the nuclear level, diacylglycerol (DAG) is certainly the most intensively studied. In some instances the whole amount of DAG comes from inositol lipid hydrolysis, whilst in some others they come from phosphatidylcholine (PC) hydrolysis. However DAG, wherever they come from, seem to be related to the translocation of protein kinase C (PKC), mainly PKCα, to the nuclear compartment and to constitute the optimal substrate for nuclear DAG-kinase. The data showing the existence of PI-PLC activity in the nucleus are numerous. The isoforms of PI-PLC which have been reported to reside in the nucleus vary according not only to the cell lines or tissues investigated, but also depending on their metabolic state. Even though the β family of PI-PLC is the more abundantly represented family of isozymes in nuclei of several cell types, other PI-PLC such γ1,δ4 and δ1 can be found, the latter mainly involved in shuttling between cytoplasm and nucleus. Nuclear PI-PLCβ has been demonstrated to be directly involved in the control of cell cycle in that there is a direct effect of its signaling activity in G1 progression. In fact, nuclear PI-PLCβ1 induces expression and activation of cyclin D3, along with its kinase (cdk4), which complex in turn phosphorylates retinoblastoma protein (pRb) and as a consequence the transcription factor E2F-1 is activated, giving rise to an acceleration of the cell cycle. Moreover, nuclear PI-PLCβ1 plays a crucial role during erythroid differentiation, affecting the p45 subunit of NF-E2 transcription factor, specific for hematopoietic lineages and inhibiting the expression of β-globin. About the regulation of nuclear PI-PLC activity, the possibility that a GTP-binding protein could be responsible for the activation of nuclear PI-PLC-β1 is not consistent with data obtained till now. Unexpectedly, we have found that, after IGF-I stimulation of quiescent Swiss 3T3 cells, there was a transient serine hyperphosphorylation of nuclear PI-PLC-β1 which paralleled enzyme activation. When the cytoskeleton of Swiss 3T3 cells was depolymerized with colchicine, MAP kinase did not translocate to the nucleus and nuclear PI-PLC-β1 was neither hyperphosphorylated nor activated, suggesting a decisive role played by MAP kinase in controlling nuclear PI-PLC-β1 activity, even though additional experiments are required to definitively support such a mode of activation. The analysis of chromosomal localization offers other clues for the central role of nuclear PI-PLCβ1. We first showed that the rat gene coding for PI-PLCβ1 is localized on rat chromosome 3q35-36. Such a localization appeared extremely intriguing to us, as this region is frequently rearranged in rat tumors induced by chemical carcinogenesis. Then by screening a human fetal brain cDNA library with a rat PI-PLCβ1 probe, we have been able to identify a clone with a 89% homology to the rat sequence, which first exon completely differs from the rat one. By using this human cDNA in fluorescence in in situ hybridization on human metaphases, we have mapped human PI-PLCβ1 gene on chromosome 20p12, which is a region amplified and/or deleted in several human solid tumors. This suggests that the human PI-PLCβ1 gene might be directly involved in tumorigenesis.
