RESEARCH INTERESTS
Nuclear Inositol Signaling
A recurring theme in intracellular signaling is the spatial restriction of pathways to selective intracellular compartments. Over the past decade, our research has examined nuclear inositol signaling pathways and more recently our studies in the budding yeast, Saccharomyces cerevisiae, have identified nuclear processes regulated by IP messengers. The budding yeast genome contains a single phosphoinositide-specific phospholipase C gene, PLC1 that closely resembles a nuclear PLCd from mammalian cells. Examination of IP metabolism in a variety of yeast strains reveal that activation of Plc1 results in the production of IP3, which is then sequentially phosphorylated to IP6 (York et al, Science, 1999). Biochemical studies indicate that two kinases are needed to perform these phosphorylation steps, and that these activities are present in nuclei. Through collaboration with Dr. Susan Wente's laboratory, we discovered that the two enzymes essential for these kinase activities are also critical for mRNA export from the nucleus. Through biochemical and genetic studies we have shown that the production of IP6 is required for efficient mRNA export. Furthermore, the sub-cellular localization of the IP5 2-kinase in this cascade, designated Ipk1, at the nuclear pore complex suggests that production of IP6 is in part a nuclear event. In conjunction with the Wente lab, we are further characterizing the mechanism and targets of IP6 regulated mRNA export (Ives et al., JBC, 2000). In cloning the yeast IP3 kinase, designated IPK2, we found it to be identical to ARG82, a previously characterized regulator of gene expression through the ArgR-Mcm1 transcription complex. Ipk2 is a dual-specificity 6-/3-kinase that sequentially converts IP3 to IP5, is localized within the nucleus and is required to assemble protein complexes on DNA-promoter elements. Both Plc1 activity and Ipk2-mediated IP4/IP5 production are required for ArgR-Mcm1 transcriptional activation. Our results indicate Ipk2 influences transcriptional responses through a two-step mechanism (Odom). First, Ipk2 protein but not IP synthesis is needed to enable formation of ArgR-Mcm1 complexes on DNA promoter elements. Second, production of IP4 and possibly IP5 through both phospholipase C and Ipk2 kinase activity is required to properly execute transcriptional control. This is consistent with a model in which ArgR-Mcm1 complexes are silently poised on promoter elements, awaiting activation of phospholipase C signaling pathways, which enables direct regulation of transcription. Furthermore, we determined that Ipk1 is not required for complex formation or transcription control, demonstrating that two independent signals arise from this pathway, IP4/IP5 as a regulator of transcription and IP6 as a regulator of mRNA export. Our work solidifies the notion of nuclear IP messengers, provides a direct mechanism connecting IP signaling to transcriptional control, demonstrates that Plc1 activation results in production of more than one messenger through a IP kinase pathway and illustrates that IP3 is a precursor to other IP messenger molecules.
Phosphatase Regulation of Inositol Lipid Messengers
While inositol lipids, such as phosphatidylinositol 4,5-bisphosphate (PIP2), are classically viewed as precursors to signaling molecules, recent evidence demonstrates that they also function as messengers in their own right. Our research has focused on the regulation of processes influenced by lipid messengers through studies of inositol lipid phosphatases. Interest in understanding the role of inositol lipid phosphatases has escalated with the discoveries that defects in these enzymes are observed in human diseases such as Lowe syndrome, cancer and myotubular myopathy. Our biochemical and genetic studies of four yeast inositol lipid 5-phosphatases, designated Inp51, Inp52, Inp53 and Inp54, have demonstrated that they play an essential role in survival, membrane trafficking and actin cytoskeleton (Stolz et al, Genetics, 1998; Stolz et al., JBC, 1998; Raucher et al., Cell, 2000). Yeast Inp52 and Inp53 share a highly similar domain structure with the mammalian synaptojanin proteins, which were first discovered as regulators of synaptic vesicle recycling. These proteins have a central 5-phosphatase catalytic domain that is flanked by an amino-terminal domain SAC1-like domain, and a variable proline-rich carboxyl-terminal region. While it was initially presumed that the inositol lipid 5-phosphatase activity of synaptojanin was important for membrane trafficking our recent work has challenged this hypothesis. We found that SAC1-like domains encode a novel inositol lipid phosphatase whose primary cellular substrate may be PI(4)P (Guo et al, JBC, 1999). This discovery provides a biochemical mechanism explaining genetic studies of SAC1, which linked it to regulating actin cytoskeleton, secretion from the Golgi and microsomal ATP transport. Analysis of cellular phosphoinositide levels in sac1 mutants demonstrate that PI(4)P increases over 8-fold, while the levels of PIP2 do not increase. Based on these data, we suggest that PI(4)P has a role independent of PIP2 in regulating secretion and actin cytoskeleton. Significantly, our work has identified the first class of inositol regulatory enzymes that harbor two distinct catalytic active sites on a single polypeptide chain and open a new area of research aimed at determining the concerted and independent roles of each activity.
Lithium Pharmacology
Bipolar or manic depressive disease has been effectively treated with lithium for over forty years. Despite the enduring pharmacologic impact of this drug, the molecular basis for its therapeutic effect has remained elusive, thus hindering the search for more effective drugs that have fewer harmful side effects. Insight into lithium's mode of action has come from the characterization of two enzymes in the IP signaling pathway, inositol monophosphatase and inositol polyphosphate 1-phosphatase, each potently inhibited by therapeutic doses of lithium. We have determined the three-dimensional structure of 1-phosphatase, which has led to the identification of a family of structurally conserved lithium-inhibited phosphatases (York et al., Biochemistry, 1994; York et al, PNAS, 1995). Members of this protein family function in diverse cellular pathways and it is our hypothesis that they represent the toxic and therapeutic targets of lithium. To further understand lithium's mechanism of action, we have determined the lithium-binding site of 1-phosphatase. Using this information, a lithium-binding sequence motif has been defined that was used to identify novel lithium-inhibited phosphatases from human genomic databases. We have characterized of one of these proteins, designated BPNT1, and have found that it possesses lithium-inhibited bisphosphate nucleotidase activity (Spiegelberg et al, JBC, 1999). BPNT1 is most highly expressed in regions of the distal nephron involved in sodium and water balance, which is intriguing given that a major side effect of chronic lithium use in humans is nephrogenic diabetes insipidus. Our recent studies have defined a mechanism by which lithium-induced inhibition of nucleotidase activity interferes with cAMP mediated water reabsorption pathways and we have discovered a pharmacological intervention to overcome this blockade even in the presence of lithium. By characterization of the cellular function of individual enzymes in the family we seek to elucidate lithium's mechanism of action, which ultimately will enable the development of improved therapies for treatment of bipolar disease.