Regulation of Cl(-) channel conductance by Ins(3,4,5,6)P(4) provides receptor-dependent control over salt and fluid secretion, cell volume homeostasis, and electrical excitability of neurones and smooth muscle. Ignorance of how Ins(3,4,5,6)P(4) is synthesized has long hindered our understanding of this signaling pathway. We now show Ins(3,4,5,6)P(4) synthesis by Ins(1,3,4,5,6)P(5) 1-phosphatase activity by an enzyme previously characterized as an Ins(3,4,5,6)P(4) 1-kinase. Rationalization of these phenomena with a ligand binding model unveils Ins(1,3,4)P(3) as not simply an alternative kinase substrate, but also an activator of Ins(1,3,4,5,6)P(5) 1-phosphatase. Stable overexpression of the enzyme in epithelial monolayers verifies its physiological role in elevating Ins(3,4,5,6)P(4) levels and inhibiting secretion. It is exceptional for a single enzyme to catalyze two opposing signaling reactions (1-kinase/1-phosphatase) under physiological conditions. Reciprocal coordination of these opposing reactions offers an alternative to general doctrine that intracellular signals are regulated by integrating multiple, distinct phosphatases and kinases.

Combinatorial expression of the various isoforms of diphosphoinositol synthases and phosphohydrolases determines the rates of phosphorylation/dephosphorylation cycles that have been functionally linked to vesicle trafficking, stress responses, DNA repair, and apoptosis. We now describe two new 19-kDa diphosphoinositol polyphosphate phosphohydrolases (DIPPs), named types 3␣ and 3␤, which possess the canonical Nudix-type catalytic motif flanked on either side by short Gly-rich sequences. The two enzymes differ only in that Pro-89 in the ␣ form is replaced by Arg-89 in the ␤ form, making the latter ϳ2-fold more active in vitro. ). Catalytic activity in vivo was established by individual overexpression of the human (h) DIPP3 isoforms in HEK293 cells, which reduced cellular levels of diphosphoinositol polyphosphates by 40 -50%. The hDIPP3 mRNA is preferentially expressed in testis, accompanied by relatively weak expression in the brain, contrasting with hDIPP1 and hDIPP2 which are widely expressed. The hDIPP3 genes (NUDT10 encodes hDIPP3␣; NUDT11 encodes hDIPP3␤) are only 152 kbp apart at p11.22 on chromosome X and probably arose by duplication. Transcription of both genes is inactivated on one of the X chromosomes of human females to maintain appropriate gene dosage. The hDIPP3 pair add tissue-specific diversity to the molecular mechanisms regulating diphosphoinositol polyphosphate turnover.Hydrolases containing the GX 5 EX 7 REUXEEXGU 1 "Nudixtype" motif comprise a protein superfamily whose members have been proposed to act as "surveillance enzymes" (2) that function both to eliminate potentially toxic metabolites from the cell as well as to regulate concentrations and availability of substrates, cofactors, and signaling molecules (3). Almost all of the substrates for these hydrolases are nucleotide diphosphates. The unique exception is provided by a subgroup of phosphohydrolases (DIPPs) 2 that preferentially attack diphosphoinositol polyphosphates (4, 5).The diphosphoinositol polyphosphates provide the most extreme example of the degree to which phosphate groups can be clustered in high density around the inositol ring. These compounds are formed when specific monoester phosphate groups on Ins(1,3,4,5,6)P 5 and InsP 6 are converted to diphosphates, by enzymes known as diphosphoinositol polyphosphate synthases (DINS). For example, InsP 6 is phosphorylated to PP-InsP 5 and [PP] 2 -InsP 4 , which contain either one or two diphosphate groups, respectively (6, 7). The DIPPs rapidly degrade the diphosphates back to their monoester precursors. This phosphate release relieves some of the severity of the electrostatic and steric constraints imposed upon these polyphosphorylated molecules, and so a substantial and physiologically purposeful free-energy change ensues. Indeed, evidence has accumulated that this DINS-and DIPP-catalyzed substrate cycling contributes to the control of vesicle trafficking. Initially, this hypothesis arose from observations we made several years ago (8 -10) showing that the diphosphoinositol p...

We previously described paralogous human genes [NUDT10 and NUDT11 [where NUDT is (nucleoside diphosphate attached moiety 'X')-type motif, also known as the 'nudix'-type motif]] encoding type 3 diphosphoinositol polyphosphate phosphohydrolases (DIPP3) [Hidaka, Caffrey, Hua, Zhang, Falck, Nickel, Carrel, Barnes and Shears (2002) J. Biol. Chem. 277, 32730-32738]. Normally, gene duplication is redundant, and lacks biological significance. Is this true for the DIPP3 genes? We address this question by characterizing highly-conserved murine Nudt10 and Nudt11 homologues of the human genes. Thus these genes must have been duplicated prior to the divergence of primates and sciurognath rodents, approx. 115 million years ago, greatly exceeding the 4 million year half-life for inactivation of redundant paralogues; our data therefore indicate that the DIPP3 duplication is unusual in being physiologically significant. One possible functional consequence is gene neofunctionalization, but we exclude that, since Nudt10 and Nudt11 encode identical proteins. Another possibility is gene subfunctionalization, which we studied by conducting the first quantitative expression analysis of these genes. We demonstrated high Nudt10 expression in liver, kidney and testis; Nudt11 expression is primarily restricted to the brain. This differential, but complementary, expression pattern indicates that subfunctionalization is the evolutionary consequence of DIPP3 gene duplication. Our kinetic data argue that diphosphoinositol polyphosphates are more physiologically relevant substrates for DIPP3 than are either diadenosine hexaphosphate or 5-phosphoribosyl 1-pyrophosphate. Thus the significance of the Nudt10/Nudt11 duplication is specific hydrolysis of diphosphoinositol polyphosphates in a tissue-dependent manner.

No abstract