Biomedical subjects
Publications and source records attributed to J H Hurley.
Phosphatidylinositol 3-phosphate directs the endosomal localization of regulatory proteins by binding to FYVE and PX domains. New structures of these domains complexed with the phosphoinositide headgroup show how interactions with phosphate and hydroxyl groups differentiate this lipid from all others.
Inositol polyphosphate 5-phosphatases are central to intracellular processes ranging from membrane trafficking to Ca(2+) signaling, and defects in this activity result in the human disease Lowe syndrome. The 1.8 resolution structure of the inositol polyphosphate 5-phosphatase domain of SPsynaptojanin bound to Ca(2+) and inositol (1,4)-bisphosphate reveals a fold and an active site His and Asp pair resembling those of several Mg(2+)-dependent nucleases. Additional loops mediate specific inositol polyphosphate contacts. The 4-phosphate of inositol (1,4)-bisphosphate is misoriented by 4.6 compared to the reactive geometry observed in the apurinic/apyrimidinic endonuclease 1, explaining the dephosphorylation site selectivity of the 5-phosphatases. Based on the structure, a series of mutants are described that exhibit altered substrate specificity providing general determinants for substrate recognition.
The reversible localization of signaling proteins to both the plasma and the internal membranes of cells is critical for the selective activation of downstream functions and depends on interactions with both proteins and membrane lipids. New structural and biochemical analyses of C1, C2, PH, FYVE, FERM and other domains have led to an unprecedented amount of information on the molecular interactions of these signaling proteins with regulatory lipids. A wave of studies using GFP-tagged membrane binding domains as reporters has led to new quantitative insights into the kinetics of these signaling mechanisms.
GAF domains are ubiquitous motifs present in cyclic GMP (cGMP)-regulated cyclic nucleotide phosphodiesterases, certain adenylyl cyclases, the bacterial transcription factor FhlA, and hundreds of other signaling and sensory proteins from all three kingdoms of life. The crystal structure of the Saccharomyces cerevisiae YKG9 protein was determined at 1.9 A resolution. The structure revealed a fold that resembles the PAS domain, another ubiquitous signaling and sensory transducer. YKG9 does not bind cGMP, but the isolated first GAF domain of phosphodiesterase 5 binds with K:(d) = 650 nM. The cGMP binding site of the phosphodiesterase GAF domain was identified by homology modeling and site-directed mutagenesis, and consists of conserved Arg, Asn, Lys and Asp residues. The structural and binding studies taken together show that the cGMP binding GAF domains form a new class of cyclic nucleotide receptors distinct from the regulatory domains of cyclic nucleotide-regulated protein kinases and ion channels.
VHS domains are found at the N-termini of select proteins involved in intracellular membrane trafficking. We have determined the crystal structure of the VHS domain of the human Tom1 (target of myb 1) protein to 1.5 A resolution. The domain consists of eight helices arranged in a superhelix. The surface of the domain has two main features: (1) a basic patch on one side due to several conserved positively charged residues on helix 3 and (2) a negatively charged ridge on the opposite side, formed by residues on helix 2. We compare our structure to the recently obtained structure of tandem VHS-FYVE domains from Hrs [Mao, Y., Nickitenko, A., Duan, X., Lloyd, T. E., Wu, M. N., Bellen, H., and Quiocho, F. A. (2000) Cell 100, 447-456]. Key features of the interaction surface between the FYVE and VHS domains of Hrs, involving helices 2 and 4 of the VHS domain, are conserved in the VHS domain of Tom1, even though Tom1 does not have a FYVE domain. We also compare the structures of the VHS domains of Tom1 and Hrs to the recently obtained structure of the ENTH domain of epsin-1 [Hyman, J., Chen, H., Di Fiore, P. P., De Camilli, P., and Brünger, A. T. (2000) J. Cell Biol. 149, 537-546]. Comparison of the two VHS domains and the ENTH domain reveals a conserved surface, composed of helices 2 and 4, that is utilized for protein-protein interactions. In addition, VHS domain-containing proteins are often localized to membranes. We suggest that the conserved positively charged surface of helix 3 in VHS and ENTH domains plays a role in membrane binding.
N- and P/Q-type Ca(2+) channels regulate a number of critical physiological processes including synaptic transmission and hormone secretion. These Ca(2+) channels are multisubunit proteins, consisting of a pore-forming alpha(1), and accessory beta and alpha(2)delta subunits each encoded by multiple genes and splice variants. beta subunits alter current amplitude and kinetics. The beta(2a) subunit is associated with slowed inactivation, an effect that requires the palmitoylation of two N-terminal cysteine residues in beta(2a). In the current manuscript, we studied steady state inactivation properties of native N- and P/Q-type Ca(2+) channels and recombinant N-type Ca(2+) channels. When bovine alpha(1B) and beta(2a) and human alpha(2)delta were coexpressed in tsA 201 cells, we observed significant variations in inactivation; some cells exhibited virtually no inactivation as the holding potential was altered whereas others exhibited significant inactivation. A similar variability in inactivation was observed in native channels from bovine chromaffin cells. In individual chromaffin cells, the amount of inactivation exhibited by N-type channels was correlated with the inactivation of P/Q-type channels, suggesting a shared mechanism. Our results with recombinant channels with known beta subunit composition indicated that inactivation could be dynamically regulated, possibly by alterations in beta subunit palmitoylation. Tunicamycin, which inhibits palmitoylation, increased steady-state inactivation of Ca(2+) channels in chromaffin cells. Cerulenin, another drug that inhibits palmitoylation, also increased inactivation. Tunicamycin produced a similar effect on recombinant N-type Ca(2+) channels containing beta(2a) but not beta(2b) or beta(2a) subunits mutated to be palmitoylation deficient. Our results suggest that Ca(2+) channels containing beta(2a) subunits may be regulated by dynamic palmitoylation.
Chromaffin cells express N-type calcium channels identified on the basis of their high sensitivity to block by omega-conotoxin GVIA (omega-CgTx GVIA). In contrast to neuronal N-type calcium currents that inactivate during long depolarizations and that require negative holding potentials to remove inactivation, many chromaffin cells exhibit N-type calcium channel currents that show little inactivation during maintained depolarizations and that exhibit no decrease in channel availability at depolarized holding potentials. N-type calcium channels are thought to be produced by combination of the pore-forming alpha(1B) subunit and accessory beta and alpha(2)/delta subunits. To examine the molecular composition of the non-inactivating N-type calcium channel, we cloned the alpha(1B) and accessory beta (beta(1b), beta(1c,) beta(2a), beta(2b), and beta(3a)) subunits found in bovine chromaffin cells. Expression of the subunits in either Xenopus oocytes or human embryonic kidney 293 cells produced high-threshold calcium currents that were blocked by omega-CgTx GVIA. Coexpression of bovine alpha(1B) with beta(1b), beta(1c), beta(2b), or beta(3a) produced currents that were holding potential dependent. In contrast, coexpression of bovine alpha(1B) with beta(2a) produced holding potential-independent calcium currents that closely mimicked native non-inactivating currents, suggesting that non-inactivating N-type channels consist of bovine alpha(1B), alpha(2)/delta, and beta(2a).
Structures are now available for the majority of the enzyme families involved in the phosphorylation, dephosphorylation and hydrolysis of signaling phospholipids. Lipid kinase and phosphatase structures recapitulate catalytic motifs involved in protein phosphorylation and dephosphorylation, whereas cytosolic phospholipase A(2) manifests novel catalytic geometry. Structures have been determined for most known intracellular phospholipid 'receptor' domains, both those that bind membrane-embedded phospholipids and those that bind lipid monomers.
Phosphatidylinositol-4,5-bisphosphate plays a pivotal role in the regulation of cell proliferation and survival, cytoskeletal reorganization, and membrane trafficking. However, little is known about the temporal and spatial regulation of its synthesis. Higher eukaryotic cells have the potential to use two distinct pathways for the generation of phosphatidylinositol-4,5-bisphosphate. These pathways require two classes of phosphatidylinositol phosphate kinases, termed type I and type II PIP kinases. While highly related by sequence, these kinases localize to different subcellular compartments, phosphorylate distinct substrates, and are functionally nonredundant. Here, we show that a 20- to 25-amino acid loop spanning the catalytic site, termed the activation loop, determines both enzymatic specificity and subcellular targeting of PIP kinases. Therefore, the activation loop controls signaling specificity and PIP kinase function at multiple levels.
The steroidogenic acute regulatory protein (StAR) regulates acute steroidogenesis in the adrenal cortex and gonads by promoting the translocation of cholesterol to the mitochondrial inner membrane where the first step in steriod biosynthesis is catalyzed. StAR-related lipid transfer (START) domains occur in proteins involved in lipid transport and metabolism, signal transduction, and transcriptional regulation. The 2.2 A resolution crystal structure of the START domain of human MLN64 reported here reveals an alpha/beta fold built around a U-shaped incomplete beta-barrel. The interior of the protein encompasses a 26 x 12 x 11 A hydrophobic tunnel that is large enough to bind a single cholesterol molecule. The StAR and MLN64 START domains bind 1 mole of 14C cholesterol per mole of protein in vitro. Based on the START domain structure and cholesterol binding stoichiometry, it is proposed that StAR acts by shuttling cholesterol molecules one at a time through the intermembrane space of the mitochondrion.
Protein kinase C homology-1 and -2, FYVE, and pleckstrin homology domains are ubiquitous in eukaryotic signal transduction and membrane-trafficking proteins. These domains regulate subcellular localization and protein function by binding to lipid ligands embedded in cell membranes. Structural and biochemical analysis of these domains has shown that their molecular mechanisms of membrane binding depend on a combination of specific and nonspecific interactions with membrane lipids. In vivo studies of green fluorescent protein fusions have highlighted the key roles of these domains in regulating protein localization to plasma and internal membranes in cells.
Type II beta phosphatidylinositol phosphate kinase is a representative phosphatidylinositol phosphate kinase that is active against membrane-bound substrates. The structure of the enzyme contains a flattened basic face that spans the crystallographic dimer interface and is adjacent to the active site. Analytical ultracentrifugation shows that phosphatidylinositol phosphate kinase is a dimer in solution. Modeling suggested that the flattened face binds to acidic phospholipids by electrostatic interactions. The enzyme binds to acidic vesicles containing phosphatidylserine, phosphatidic acid, or phosphoinositides mixed with phosphatidylcholine, but not to neutral phosphatidylcholine vesicles. Binding to acidic vesicles is abolished in the presence of 1.0 M NaCl, consistent with an essential electrostatic contribution to the free energy of binding. The +14 charge on the flattened face of the dimer was reduced to +2 in the triple mutant Lys72Glu/Lys76Glu/Lys78Glu. The mutation has no effect on dimerization, but reduces the apparent KA for 25% phosphatidylserine/75% phosphatidylcholine mixed vesicles by 16-fold. The reduction in the level of binding can be ascribed to a loss of electrostatic interactions based on the finite difference solution to the Poisson-Boltzmann equation. The mutant reduces catalytic activity toward phosphatidylinositol 5-phosphate by approximately 50-fold. The wild-type enzyme binds half-maximally to phosphatidylinositol 4,5-bisphosphate-containing vesicles at a mole fraction of 0.3% in a phosphatidylcholine background, as compared to a 22% mole fraction in phosphatidylserine. The binding to phosphatidylinositol 4,5-bisphosphate-containing membranes is less sensitive to salt and to the triple mutation than binding to phosphatidylserine-containing membranes, suggesting that at least part of phosphatidylinositol 4,5-bisphosphate's interaction with the enzyme is independent of the flattened face. It is concluded that the flattened face of type II beta phosphatidylinositol phosphate kinase binds to membranes through nonspecific interactions, and that this interaction is essential for efficient catalysis.
Phosphatidylinositol 3-phosphate regulates membrane trafficking and signaling pathways by interacting with the FYVE domains of target proteins. The 1.15 A structure of the Vps27p FYVE domain reveals two antiparallel beta sheets and an alpha helix stabilized by two Zn2+-binding clusters. The core secondary structures are similar to a rabphilin-3A Zn2+-binding domain and to the C1 and LIM domains. Phosphatidylinositol 3-phosphate binds to a pocket formed by the (R/K)(R/K)HHCR motif. A lattice contact shows how anionic ligands can interact with the phosphatidylinositol 3-phosphate-binding site. The tip of the FYVE domain has basic and hydrophobic surfaces positioned so that nonspecific interactions with the phospholipid bilayer can abet specific binding to phosphatidylinositol 3-phosphate.
Explore the source record for details and available documents.
The most prominent structural feature of the G protein-coupled receptor superfamily is their seven hydrophobic domains, which are postulated to form membrane-spanning alpha helices. Some members of the G protein-coupled receptor family, specifically several serotonin (5-HT) receptors, possess eight hydrophobic domains. The importance of this extra hydrophobic domain, located at the N terminus of the receptor, is unknown. This question was addressed by deleting the extra hydrophobic region from the 5-HT2C receptor and comparing its function and topology with those of the wild-type receptor. Immunofluorescence microscopy was used to determine the location of the N terminus of the epitope-tagged wild-type and mutant receptors. The N terminus of both receptors was extracellular, suggesting that the extra hydrophobic domain does not change the topology of this receptor and is unlikely to be a membrane-spanning alpha helix. Radioligand-binding studies in transfected cells and expression studies in Xenopus oocytes demonstrated that seven hydrophobic domains were sufficient for normal function in these assays. Interestingly, the mutant receptor, now containing seven hydrophobic domains, is expressed at higher levels in transfected cells than the wild-type receptor containing eight hydrophobic domains, suggesting that the extra hydrophobic domain does impact the activity of this receptor by regulating its expression.
Phosphoinositide kinases play central roles in signal transduction by phosphorylating the inositol ring at specific positions. The structure of one such enzyme, type IIbeta phosphatidylinositol phosphate kinase, reveals a protein kinase ATP-binding core and demonstrates that all phosphoinositide kinases belong to one superfamily. The enzyme is a disc-shaped homodimer with a 33 x 48 A basic flat face that suggests an electrostatic mechanism for plasma membrane targeting. Conserved basic residues form a putative phosphatidylinositol phosphate specificity site. The substrate-binding site is open on one side, consistent with dual specificity for phosphatidylinositol 3- and 5-phosphates. A modeled complex with membrane-bound substrate and ATP shows how a phosphoinositide kinase can phosphorylate its substrate in situ at the membrane interface.
Guanylyl cyclases (GCs) and adenylyl cyclases (ACs) have fundamental roles in a wide range of cellular processes. Whereas GCs use GTP as a substrate to form cGMP, ACs catalyze the analogous conversion of ATP to cAMP. Previously, a model based on the structure of adenylate cyclase was used to predict the structure of the nucleotide-binding pocket of a membrane guanylyl cyclase, RetGC-1. Based on this model, we replaced specific amino acids in the guanine-binding pocket of GC with their counterparts from AC. A change of two amino acids, E925K together with C995D, is sufficient to completely alter the nucleotide specificity from GTP to ATP. These experiments strongly validate the AC-derived RetGC-1 structural model and functionally confirm the role of these residues in nucleotide discrimination.
The hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2) by phosphoinositide-specific phospholipase C (PLC) is absolutely dependent on Ca2+. The PH domain truncated catalytic core of rat phospholipase C delta1 (PLC-delta1) has Ca2+ binding sites in its catalytic and C2 domains, and potential Ca2+ binding sites in two EF-hands. A catalytically inactive PLC-delta1 catalytic core bound with low affinity to PIP2-containing vesicles in the presence of Ca2+. A mutant PLC-delta1 has been engineered which lacks the C2 domain Ca2+ binding site and the surrounding loops known as the jaws. Isothermal calorimetric titration showed four Ca2+ ions bind to the wild-type PLC-delta1 catalytic core in solution but only one binds to the C2 domain jaws deletion mutant. The activity and Ca2+ dependence of wild-type and mutant phospholipase Cs were determined using substrate incorporated in detergent micelles and in large unilamellar vesicles. The activities of wild-type and mutant were identical to each other in both assay systems. Wild-type and the C2 jaws deletion mutant of PLC have Hill coefficients of 1.12-1.16 with respect to [Ca2+]. We conclude that a single Ca2+ bound to the catalytic domain is entirely responsible for the Ca2+ dependence of the basal activity of PLC-delta1.