Home | Help | Feedback | Subscriptions | Archive | Search | Table of Contents

This Article PDF (Full Text) Alert me when this article is cited Services Email this article Similar articles in this journal Alert me to new content in the JGP Download to citation manager Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Search for Related Content Social Bookmarking                      What's this?

Speaker Abstracts
1. SR Ca²⁺-ATPase Atomic Structure and Molecular Dynamics Simulations. YUJI SUGITA and CHIKASHI TOYOSHIMA, Institute of Molecular and Cellular Biosciences, The University of Tokyo, Tokyo, Japan (Sponsor: David Gadsby)

Sarcoplasmic reticulum (SR) Ca²⁺ATPase (SERCA1a) is an integral membrane protein of 110 kD that establishes the concentration gradient of Ca²⁺ across the SR membrane by transporting two Ca²⁺ per ATP hydrolyzed. Recent X-ray structures of SR Ca²⁺-ATPase provide atomic models for two high-affinity Ca²⁺-binding sites in the transmembrane region, which consists of 10 helices. We have studied how the Ca²⁺-binding sites are stabilized in the transmembrane region by using all-atom molecular dynamics (MD) simulations with explicit solvent and lipids. During the simulations, three cytoplasmic domains as well as the transmembrane domain were very stable. However, the Ca²⁺-coordination and the H-bonds formed by the protonation of Glu58 and Glu908 are stable in an MD simulation, whereas the H-bonds are disrupted and the Ca²⁺-coordination geometry is severely altered in another simulation in which these residues are treated as unprotonated. The results clearly indicate that the H-bonds formed by protonation of Glu58 and Glu908 provide extra stability for the Ca²⁺-binding sites of Ca²⁺-ATPase (Sugita et al. 2005. J. Am. Chem. Soc. In press).

2. Structure and Mechanism of Sarcoplasmic Reticulum Ca²⁺-ATPase. JESPER V. MØLLER,¹ CLAUS OLESEN,¹ POUL NISSEN,² ANNE-MARIE L. JENSEN,² RIKKE C. NIELSEN,² and THOMAS L.-M. SØRENSEN,^{3 1}Department of Biophysics and ²Department of Molecular Biology, University of Aarhus, DK-8000 Aarhus C, Denmark; ³Diamond Light Source Ltd., Rutherford Appleton Laboratory, Oxfordshire, UK

Recent progress in the X-ray analysis of sarcoplasmic reticulum Ca²⁺-ATPase crystals with bound nucleotides and phosphate analogs (AlF₄^–, MgF₄^2–) by our and Toyoshima's groups has led to the structural description, at atomic resolution, of intermediates and transition states related to the Ca²⁺ transport cycle (summarized in Olesen et al. 2004. Science. 306:2251–2255). After reaction with ADP and AlF₄ (to mimic the transition state leading to the Ca2E1~P intermediate following reaction with ATP), the cytoplasmic N- and P-domains of Ca²⁺-ATPase are glued together by AlF₄, forming a transitional complex by an associative SN2 mechanism. This is accompanied by changes in the conformation of the M1/M2 transmembrane segments and a dampened ATPase dynamics that lead to closure of the cytosolic gate for bound Ca²⁺. After reaction of Ca²⁺-ATPase with AlF₄ in the E2 state, another complex is formed, which represents the transition state corresponding to E2P dephosphorylation. In this state, the Ca²⁺ liganding residues are occupied by occluded protons, originating from luminal Ca²⁺/H⁺ exchange. In the E2P dephosphorylation transition state, both the A-domain and N-domain are rotated to enable the Thr-181 and Glu-183 in the TGES motif, together with Mg²⁺, to dephosphorylate Asp-351 by a mechanism that structurally resembles the reaction by which the same Asp-351 is normally phosphorylated from ATP. When dephosphorylation has been accomplished, the TGES motif is disengaged from the phosphorylation site, and the ATPase returns to a conformational state in which it can react with Ca²⁺. Our data also comprise the structure of an E2 state with bound nucleotide, and a K⁺ (Rb⁺) site, relevant for understanding the modulatory effect of these ligands on ATPase activity. Overall, our data provide a structural explanation for ATPase function in terms of an orderly sequence of events that leads to effective coupling between the phosphorylation/dephosphorylation and Ca²⁺/H⁺ exchange events. [Supported by the Danish Medical Science Research Council, The Novo Nordisk Foundation, The Lundbeck Foundation, and The Aarhus University Research Foundation.]

3. Purification of Na,K-ATPase expressed in Pichia pastoris. Prospects for structural work? STEVEN. J.D. KARLISH, Department of Biological Chemistry, Weizmann Institute of Science, Rehovot, Israel (Sponsor: David Gadsby)

Na⁺,K⁺-ATPase (porcine /his₁₀-ß) expressed in Pichia Pastoris, has been dissolved in n-dodecyl-ß-maltoside (DDM) and purified in a functional state by metal chelate bead chromatography (Cohen et al. 2005. J. Biol. Chem. In press; see abstract for this symposium). The ß subunit is expressed as two lightly glycosylated polypeptides and is easily deglycosylated by endoglycosidase-H at 0°C. Added lipid is required to maintain Na,K-ATPase activity, and evidence has been obtained for specific interactions of the protein with the acid phospholipid, dioleoyl phosphatidylserine (DOPS), and probably also the neutral phospholipid, dioleoyl phosphatidylcholine (DOPC), and cholesterol. Recombinant Na⁺,K⁺-ATPase and pig kidney Na⁺,K⁺-ATPase, dissolved in DDM, appear to be mainly stable monomers (/ß) as judged by size exclusion HPLC and sedimentation velocity. Their Na⁺,K⁺-ATPase activities at 37°C are similar but are lower than that of membrane-bound renal Na⁺,K⁺-ATPase. Both DDM-soluble recombinant and renal Na⁺,K⁺-ATPase are stabilized in an E₁ conformation, perhaps explaining the lower activities. Human 1 and 2 isoforms have also been expressed with porcine ß1 and 1/his₁₀-ß1 and 2/his₁₀-ß1 complexes purified, and, in addition, FXYD proteins have been expressed, and /ß/FXYD complexes are being purified (Lifshitz, Garty, and Karlish, abstract for this symposium). The purified complexes of Na,K-ATPase could become important tools for structure–function and biophysical studies, and for studying interactions with lipids and other proteins. Because the recombinant Na,K-ATPase can be produced pure, active, stable, mono-disperse, and deglycosylated, in quantities up to 1 mg, structural work may become feasible. 3D crystallization trials have recently been initiated (with S. Iwata, Imperial College, London).

4. Binding of Na⁺ or ATP Control Opening or Closing of the Cytoplasmic Gate to the Na⁺ Sites in Na,K-ATPase. PETER L. JORGENSEN, Institute of Molecular Biology and Physiology, University of Copenhagen, 2100 Copenhagen OE, Denmark

The initial steps in active Na⁺ extrusion are entry of Na⁺ through a cytoplasmic gate, followed by binding and occlusion of three Na⁺ ions in intramembrane sites to prevent the exchange of ions with those on either side of the membrane, but the details of the structural changes accompanying occlusion have so far not been resolved. High resolution structure analyses of Ca-ATPase of SR (PDB code 1T5S, 1VFP) show that ATP binding closes the gate for access to the intramembrane cation sites I and II primarily by moving the M1 helix. The A-domain tilts by 30° to pull up the M1-M2 helices and to strain the loop to M3. In a homology model of the -subunit of Na,K-ATPase, the gate for Na⁺ entrance on the cytoplasmic side between M2, M4, and M6 can be visualized by Grasp analysis of electrostatic surface potential. Opening and closure of this gate can be monitored by chymotryptic cleavage at Leu-266 in the loop to M3. In a Tris medium, addition of Na⁺ (K_1/2 1.6 mM) increases the rate constant for chymotryptic cleavage by three- to fourfold, and this effect of Na⁺ is reversed by Mg²⁺ and nucleotides. Binding of both Mg²⁺ and the nonhydrolyzable ATP analogs is required to close the gate for entry of Na⁺. These structural changes are interpreted to reflect opening of the cytoplasmic gate by Na⁺, and closure of the gate upon binding of MgAMP-PCP, accompanied by tilting of the A-domain and changes of the strain on the loop to M3 that can be monitored by ease of chymotryptic cleavage at Leu266.

5. Heavy Metal binding Sites in P_IB-type ATPases. JOSÉ M. ARGÜELLO, Department of Chemistry and Biochemistry, Worcester Polytechnic Institute, Worcester, MA (Sponsor: Jack H. Kaplan)

P_IB-type ATPases transport diverse heavy metals (Cu⁺, Cu²⁺, Zn²⁺, Co²⁺, etc.) across membranes. The most recognized members of this subfamily are the two human Cu-ATPases. Their mutations lead to Wilson or Menkes diseases. However, present in archaea, bacteria, and eukaryotes, P_IB-ATPases are widely distributed. Previous experimental evidence has shown that individual proteins can transport various metals and that diverse subgroups might present alternative specificities, suggesting a confusing and unpredictable metal selectivity for P_IB-ATPases. This can be clarified by identifying and characterizing their various metal binding/transport sites. Many P_IB-ATPases, in addition to the transmembrane metal binding site (T-MBS), have NH₂-terminal or COOH-terminal cytoplasmic metal binding domains (MBD). These are apparently regulatory and do not participate in determining metal specificity. By analogy with P_II-ATPases, we assumed that P_IB-ATPases T-MBS would be constituted by amino acid side chains in transmembrane segments (TMs) flanking the large ATP binding cytoplasmic loop. Bioinformatics analysis of TMs from more than 200 available P_IB-ATPase sequences revealed conserved sequences only in TMs H6, H7, and H8 (equivalent to H4, H5, and H6 of P_II-ATPases). These signature sequences allowed the identification of at least five subgroups of P_IB-ATPases. Cloning, heterologous expression, and functional characterization of proteins from four of these subgroups revealed unique metal specificities: IB-1 Cu⁺/Ag⁺; IB-2 Zn²⁺/Cd²⁺/Pb²⁺; IB-3 Cu²⁺; IB-5 Pb²⁺. Site-directed mutagenesis experiments testing the participation of conserved amino acids in metal coordination during transport indicated that the T-MBS of Cu⁺-ATPases (group IB-1) is constituted by two Cys in H6, a Tyr and an Asn in H7, and a Met and a Ser in H8. The essential participation of these residues in heavy metal transport indicates a singular metal coordination in the transmembrane region of Cu-ATPases, distinct from that observed in metalloproteins where the metals play a structural or a catalytic role. [Supported by NSF grant MCM-0235165.]

6. "Next Stop: Binding Sites"—Ion Transport and Binding Sites in P-type ATPases. HANS-JÜRGEN APELL, Department of Biology, University of Konstanz, 78464 Konstanz, Germany

Recent biophysical and electrophysiological studies of the three major members of the family of P-type ATPases, i.e., the Na,K-ATPase, the SR Ca-ATPase and the H,K-ATPase, support the concept that a common mechanism of ion transport exists for this family of ion pumps. The basic principle is a central ion-binding moiety deep within the membrane domain of the protein which is connected to the aqueous phases on both sides of the membrane by access channels. A short-circuit pathway is prevented by strict control of two gates which, under physiological conditions, allow only alternating opening to the channels on either side of the binding sites. Although the access channels are (mostly) so narrow that ions migrate through them in an electrogenic manner, the ions move very rapidly, probably controlled by electro-diffusion. Recent kinetic investigations support this concept: only the ion movements in the half channels are electrogenic, while no, or no significant, charge movements could be detected during the phosphorylation/dephosphorylation reactions and the conformation transitions. Subsequent to the approach to the central moiety, binding of the ions in their respective sites is a more complex process which is apparently correlated with conformational relaxations that modify the protein structure adjacent to the bound ion, and thus create or at least optimize the binding sites to allow defined, sequential binding. In addition, complete occupancy of the sites communicates a trigger signal for the subsequent enzymatic action. Such a concept of active ion transport separates rather strictly (passive) ion movements, which are affected by the membrane potential, from the enzymatic activity of the protein, which has to account fully for the energy transduction between chemical energy (stored in ATP) and a "conformational excitation" of the protein that eventually drives the vectorial ion transport. [Supported by DFG grant Ap45/4.]

7. Na,K-ATPase Ion Translocation Pathway. PABLO ARTIGAS, NICOLÁS REYES and DAVID C. GADSBY, The Rockefeller University, New York, NY

Palytoxin (PTX) opens an ion channel in the Na/K-ATPase, an effect antagonized by ouabain. We studied PTX-induced channels in outside-out patches, excised from Xenopus oocytes heterologously expressing Xenopus pumps (1ß3) mutated to be ouabain resistant, with 125-mM Na solutions containing 100 µM ouabain to inhibit endogenous pumps. Mutation of N131 (corresponding to D129 in rat 1) modified single PTX-induced pump-channel conductance (_PTX) according to introduced charge: _PTX 1 pS for positive, _PTX 4 pS for neutral, and _PTX 6 pS for negative residues. Effects on macroscopic PTX-induced conductance (G_PTX) of covalent modification of mutant N131C with charged sulfhydryl reagents resembled those of mutagenesis on _PTX: cationic MTSET⁺ reduced, whereas anionic MTSES^– increased, G_PTX. This suggests that residue 131 in transmembrane (TM) segment 2, lies near the external entrance to the channel and electrostatically influences its conductance, by controlling the effective local concentration of the conducted ions. MTSET⁺ and MTSES^– produced corresponding electrostatic effects on G_PTX in pumps with single cysteines introduced at several other positions near the extracellular ends of TMs 1, 2, 4, 5, and 6 (including G805C in TM6). However, for the mutant T806C (just intracellular to G805), G_PTX was reduced both by MTSET⁺ (90%) and by MTSES^– (20%), suggesting that the channel narrows abruptly at that point, precluding accommodation of MTSES^– without sterically hindering ion conduction. Consistent with these results, homology models based on the SERCA X-ray crystal structures locate T806 at the deepest point of a wide cavity surrounded by TMs 1, 2, 4, 5, and 6. Because many of the residues we mutated in this cavity are reported determinants of ouabain affinity, and were water accessible in the unmodified pump (without PTX), we studied the influence of the presence of the steroid on their modification by MTSET⁺. The implications of these results for ouabain binding will be discussed. [NIH-HL36783.]

8. Mechanism of SR Ca²⁺-ATPase by Functional Analysis of Site Mutants. JENS PETER ANDERSEN,¹ JOHANNES D. CLAUSEN,¹ DAVID B. MCINTOSH,² ANJA P. EINHOLM,¹ ANNE NYHOLM ANTHONISEN,¹ and BENTE VILSEN,^{1 1}Institute of Physiology and Biophysics, Department of Physiology, University of Aarhus, Aarhus, Denmark; ²Institute of Infectious Disease and Molecular Medicine, University of Cape Town, Cape Town, South Africa (Sponsor: David C. Gadsby)

The molecular mechanism of the sarcoplasmic reticulum Ca²⁺-ATPase can be addressed by combining information from X-ray crystallography with analysis of functional changes in site-specific mutants. We have established a panel of assays for the partial reactions in the Ca²⁺ transport cycle of expressed wild-type and mutant Ca²⁺-ATPase, including rapid kinetic measurements of the rates of phosphorylation and dephosphorylation, of conformational changes, and of Ca²⁺ binding and dissociation, as well as measurement of the binding affinities for nucleotides, vanadate, fluorides, and other inhibitors. These assays have allowed us to analyze functionally the structural features seen in the atomic models derived from Ca²⁺-ATPase crystals, such as the movement of the conserved A-domain TGES motif into the catalytic site in E₂ conformations, interaction of the conserved P1-helix residue E340 with transmembrane segment M3 and cytoplasmic loop L6-7, interaction of the bent transmembrane segment M1 with the Ca²⁺-binding glutamate, E309, of M4, and interaction of conserved N- and P-domain residues with nucleotide and Mg²⁺. The long known K⁺-induced activation of E₂P dephosphorylation by K⁺ acting from the cytoplasmic side was disrupted by mutation E732A in domain P, thereby leading to identification of the regulatory K⁺ site (Sorensen et al. 2004. J. Biol. Chem. 279:46355–46358). To further understand the regulation of E₂P dephosphorylation, we have analyzed a series of mutants with alterations to the K⁺ site and nearby residues of the A-M3 linker that approaches the bound K⁺ ion in E₂ conformation. These results will be discussed.

9. Cation Pump Subunit Interactions. KAZUYA TANIGUCHI, Biological Chemistry, Division of Chemistry, Graduate School of Science, Hokkaido University, Sapporo, Japan (Sponsor: Jack Kaplan)

Na/K-ATPase and other P-type ATPase enzymes utilize the free energy of ATP for ion transport across membranes. Na/K-ATPase and gastric H/K-ATPase retain a catalytic subunit and a glycosylated ß subunit, and kidney Na/K-ATPase also contains a subunit. The coupling mechanism of ATP hydrolysis and cation transport in P-type-ATPase can essentially be explained by the Post-Albers mechanism, the sequential appearance of dephosphoenzyme with a high affinity for ATP (E1ATP), an ADP-sensitive phosphoenzyme (E1P), a K-sensitive (and ADP insensitive) phosphoenzyme (E2P), and a dephosphoenzyme with a reduced affinity for ATP (E2), which becomes E1ATP. While protomeric Na/K-ATPase (ß) is thought to be sufficient for Na/K-ATPase activity, some controversy has arisen concerning whether the functional unit of the enzyme (or the transporter) is a protomer or a much higher molecular weight oligomer, (ß)n, which would be related to the mechanism of transport, either sequential or simultaneous. Our group, in collaboration with Y. Hayashi (Kyorin University) has successfully demonstrated the simultaneous presence of EP:EATP both in pig kidney Na/K-ATPase and gastric H/K-ATPase, and has been studying enzymatic properties of the oligomer. These data have added new dimensions to the oligomeric properties of the enzyme and are consistent with numerous previous studies by Askari's group (Medical College of Ohio) and W. Schoner's group (Justus-Liebig University). We also succeeded in confirming the presence of tetrameric H/K-ATPase using total internal reflection fluorescence microscopy. Y. Hayashi also was successful in isolating the active teraprotomer of Na/K-ATPase in C12E8 solubilized condition. The presence of subunit interactions in Na/K-ATPase has recently been reported by R.W. Mercer (Washington University), J. Kaplan (Oregon Health Science University), and K. Sweadner (Massachusetts General Hospital). Although the monomeric crystal structure of SR-Ca-ATPase, reported by C. Toyoshima's group (The University of Tokyo) showed an ATP binding site, we showed that each single amino acid mutation in the ATP binding pocket induced different effects on the high and low affinity ATP effects, suggesting a conformational difference in 1 mol of ATP binding/catalytic subunit. Studies of the molecular events occurring in each subunit, and of subunit interactions, during ATP hydrolysis would be the key to a better understanding of the mechanism of energy transduction in P-type ATPases.

10. Characterization of Subunit Interactions and their Role in Na,K-ATPase Delivery to the Plasma Membrane. YIQING CHI, MELISSA LAUGHERY, EDWARD B. MARYON, REBECCA CLIFFORD, and JACK H. KAPLAN, Department of Biochemistry and Molecular Genetics, University of Illinois at Chicago, Chicago, IL (Sponsor: Jack H. Kaplan)

The most widespread form of the Na,K-ATPase is the 1ß1 heterodimer. Using the baculovirus-infected insect cell system we have shown that the -subunit expressed alone is retained in the endoplasmic reticulum, while the ß-subunit is able to traffic alone to the plasma membrane. In addition, substitution of the three N-linked glycosylation sites in ß1 has little effect on the assembly, activity, or plasma membrane delivery of the Na pump. We have investigated whether this is seen in mammalian cells using polarized monolayers of MDCK cells. In addition, we have investigated similar substitutions in the ß2 isoform, which has up to eight potential N-linked glycosylation sites, in both the insect cell and MDCK cell system. Inappropriate expression of the ß2 isoform in kidney had been associated with apical mis-targeting of Na pump subunits and the development of cysts in polycystic kidney disease. It has been reported that expression of ß2 in MDCK cells causes apical delivery of Na pump subunits. We have developed MDCK cell lines that simultaneously express the ß1 and ß2 isoforms. By using tetracycline-regulated expression and selective siRNA knockdown of one of the isoforms, we have investigated the effects of altered levels and ratios of each ß-subunit isoform on Na,K-ATPase targeting and delivery. The effects of these manipulations in ß-subunit expression will be discussed as well as the roles of the ß-subunit in Na pump delivery. [Supported by NIH grants HL30315 and GM39500.]

11. Na,K-ATPase in Epithelial Cell Polarity and Signaling. AYYAPPAN K. RAJASEKARAN, Department of Pathology and Laboratory Medicine, David Geffen School of Medicine, University of California, Los Angeles, CA

The Na,K-ATPase is a highly studied molecule best appreciated for its role in intracellular electrolyte homeostasis. Recent studies from our laboratory have established a role for Na,K-ATPase in the formation of tight junctions and establishment of polarity in epithelial cells. In this talk, I will present new evidence that Na,K-ATPase associates with key proteins localized to the apical junctional complex, which contains tight and adherens junctions, in epithelial cells. The significance of these findings in relation to establishment and maintenance of epithelial polarity and its implication in cancer will be discussed.

12. Na,K-ATPase Subunit Mutations in Neurological Disease. KATHLEEN J. SWEADNER and JOHN T. PENNISTON, Laboratory of Membrane Biology, Massachusetts General Hospital and Harvard Medical School, Boston, MA (Sponsor: David Gadsby)

The Na,K-ATPase consumes half of the ATP supply of the brain, and so it would be expected that defects in Na,K-ATPase would result in neurological problems. What is surprising is the very specific nature of the neurological disorders in humans carrying mutations in 3 and 2 Na,K-ATPase subunits. In the nervous system, 3 is found only in neurons. It restores ion gradients drained by electrical activity and makes a small direct hyperpolarizing contribution to membrane potential. Six different mutations of 3 have been found in eight unrelated families with a unique, dominantly inherited disease phenotype, Rapid Onset Dystonia Parkinsonism (RDP). Individuals carrying the gene are symptom free until a metabolically stressful precipitating event, and then they develop symptoms over a period of hours to days: involuntary muscle contractions that result in abnormal movements and postures. We have generated structural models of 3 based on SERCA1a crystal structures and found that all of the mutations lie in places that should be critical for activity. Transfectants showed reduced expression of most of the mutated 3 subunits, suggesting impaired folding or stability. A scheme will be presented that may explain the peculiar characteristics of the disease. In the nervous system, 2 is found in some neurons, but is more abundant in astrocytes. As many as 15 mutations in 2, in 16 families, segregate with familial hemiplegic migraine (FHM2). The disease, like some other migraines and epilepsies, may be due to impairment of astrocyte K⁺ clearance. Interestingly, the same disease is produced by mutation of a calcium channel, Ca_v2.1, that is expressed exclusively in neurons, and a scheme will be presented that shows a plausible link through the development of cortical spreading depression. Homology maps of FHM2 mutation sites predict, as with the RDP mutations, that most appear to be at structurally critical sites. [Supported by NIH HL036271.]

13. Alterations in the 2 Isoform of Na,K-ATPase Associated with Familial Hemiplegic Migraine Type 2. RHODA BLOSTEIN,¹ LAURA SEGALL,¹ ROSEMARIE SCANZANO,¹ ALESSANDRA MEZZETTI,² ENRICO PURISIMA,² and J. JAY GARGUS,^{3 1}Department of Biochemistry and Department of Medicine, McGill University, Montreal, QC, Canada; ²Biotechnology Research Institute, National Research Council of Canada, Montreal, QC, Canada; ³Department of Physiology and Department of Biophysics and Pediatrics, Section of Human Genetics, University of California, Irvine, CA

At least nine missense mutations in the 2 subunit of Na,K-ATPase have been identified in familial hemiplegic migraine with aura (FHM2). Whereas two alleles (L764P and W887R) showed loss of function (De Fusco et al. 2003. Nature Gen. 33:192–196), we observe that at least three others, namely T345A (Kaunisto et al. 2004. Neurogenetics 5:141–146) as well as R689Q and M731T (Vanmolkot et al. 2003. Ann. Neurol. 54: 360–366), are functional but display altered Na,K-ATPase kinetics. Kinetic analyses (apparent cation and ATP affinities, catalytic turnover, and steady-state E₁/E₂ conformational poise) reveal changes effected by T345A, R689Q, and M731T, and suggest that the disease phenotype is the consequence of lowered molecular activity of the 2 pump isoform. The lower activity is due to either decreased K⁺ affinity (T345A) or catalytic turnover (R689Q and M731T), thus causing a delay in extracellular K⁺ clearance and/or altered localized Ca²⁺ handling/signaling secondary to reduced activity in colocalized Na⁺/Ca²⁺ exchange. Information about the mechanistic bases for the kinetic alterations has been obtained from consideration of the structural changes effected by the residue replacements based upon homologous replacements in the known crystal structure of the sarcoplasmic reticulum Ca-ATPase. [Supported by the CIHR grant MT-3876.]

14. Behavioral Abnormalities in Na,K-ATPase Alpha Subunit Haploinsufficient Mice. AMY E. MOSELEY,¹ MICHAEL T. WILLIAMS,² TORRI L. SCHAEFER,² CHARLES V. VORHEES,³ and JERRY B LINGREL,^{1 1}Department of Molecular Genetics, University of Cincinnati, Cincinnati, OH; ²Department of Child Neurology and ³Department of Developmental Biology, Cincinnati Children's Research Foundation, Cincinnati, OH

Three isoforms of the catalytic subunit of the Na,K-ATPase are expressed in adult brain. The 1 isoform is expressed in most cell types, while the 2 isoform is mainly expressed in astrocytes and the 3 isoform is expressed in neurons. To explore the role of the Na,K-ATPase in behavior, we performed a battery of tests on adult male mice made heterozygous for the 1, 2, and 3 isoforms. We first tested for anxiety-related behavior. Using the zero maze test, the 2^+/– mice spent less time in the open compared with wild type (ANOVA, F_3,79 = 4.7, P < 0.05), whereas the 1^+/– and 3^+/– mice were the same as wild type. All three heterozygous mice performed the same as wild type in the novel object recognition test, spending a similar amount of time with the novel object. General locomotor activity was also assessed. While the 1^+/– and 3^+/– activity was similar to that of wild-type mice over a period of 1 h, 2^+/– mice exhibited reduced locomoter activity throughout this time (ANOVA, F_3,84 = 5.5, P < 0.05). The Morris Water maze was used to test for spatial learning deficits. The 3^+/– mice displayed longer latency to the platform compared with wild type, using the cued platform version of the Morris Water maze. Using the submerged platform version of the Morris Water maze, the 2^+/– and 3^+/– mice, but not the 1^+/– mice, exhibited longer latency to the platform compared with wild-type animals. While the 1^+/– mice displayed behavior similar to wild type, both the 2^+/– and 3^+/– mice showed memory/learning impairment compared with wild-type mice. The 2^+/– mice also showed reduced locomotor activity compared with wild-type, 1^+/–, or 3^+/– mice. In summary, having examined all three Na,K-ATPase isoform–deficient mice concurrently, we conclude that the Na,K-ATPase isoforms can differentially modulate behavior. [Supported by NIH grants HL28573, HL66062, DA06733, and DA14269.]

15. Function and Regulation of Human Copper-transporting ATPases, the Menkes Disease and Wilson Disease Proteins. SVETLANA LUTSENKO, NATALIE BARNES, and RUSLAN TSIVKOVSKII, Department of Biochemistry and Molecular Biology, Oregon Health and Science University, Portland, OR

Human copper-transporting ATPases (Cu-ATPases) play an essential role in cell metabolism. These proteins deliver copper to the secretory pathway that participates in the biosynthesis of secreted copper-dependent enzymes. In addition, Cu-ATPases transport excess copper out of the cells, thereby maintaining the intracellular copper concentration. Mutations in the genes encoding the copper-transporting ATPases ATP7A and ATP7B lead to the severe metabolic disorders Menkes disease and Wilson disease, respectively. Though these Cu-ATPases have distinct tissue-specific distribution, in several tissues, including brain, kidney, and placenta, both ATP7A and ATP7B are present; however, the specific roles of Cu-ATPases in these tissues remain poorly understood. Using high-resolution fluorescent imaging, we demonstrate that in murine cerebellum, ATP7A and ATP7B have distinct cell-specific distribution (Bergmann glia and Purkinje neurons, respectively) and are regulated differently during development. Heterologous expression of ATP7A and ATP7B in Sf9 cells revealed that ATP7A and ATP7B have comparable affinities for either ATP or copper, but they differ substantially in the time courses of their partial reactions, such as catalytic phosphorylation and dephosphorylation. Lastly, we provide experimental evidence that the lack of functional ATP7B in the cerebellum of knock-out mice lacking the Wilson disease gene is compensated by ATP7A, illustrating a tight link between copper homeostasis of Purkinje neurons and of Bergmann glia. [Supported by NIH grant PO1-GM067166.]

16. Na,K-ATPase Regulation by FXYD Proteins. KAETHI GEERING, Department of Pharmacology and Toxicology, University of Lausanne, CH-1005 Lausanne, Switzerland (Sponsor: David Gadsby)

Recent studies have provided evidence that four out of seven members of the FXYD family, FXYD1 (phospholemman), FXYD2 ( subunit), FXYD4 (CHIF), and FXYD7, are auxilary subunits of Na,K-ATPase and regulate its activity in a tissue- and isoform-specific way (for review see Crambert and Geering. 2003. Sci. STKE. 2003. 166:RE1). We are interested (1) to identify interaction sites in the transmembrane domain (TM) of the Na,K-ATPase subunit and of FXYD proteins that mediate the efficient association of the two proteins and/or the functional effect of FXYD proteins, and (2) to elucidate structural and functional properties of FXYD proteins that have so far not been studied. Mutational analysis combined with protein modeling revealed that distinct amino acids in TM9 of the Na,K-ATPase subunit are involved in either the efficient association or the functional effect of FXYD proteins. Moreover, tryptophan scanning permitted us to identify distinct domains in the TM helix of FXYD7 that are important for the efficient interaction with Na,K-ATPase. Characterization of FXYD3 (Mat-8), a protein that is mainly expressed in stomach and colon and that is up-regulated in certain tumors, revealed that, like other FXYD proteins, FXYD3 also associates with Na,K-ATPase and regulates its transport properties. However, FXYD3 exhibits some unusual characteristics. In contrast to other FXYD proteins, which are type I proteins, FXYD3 may have two TM domains due to lack of cleavage of a signal peptide. Moreover, when expressed in Xenopus oocytes, FXYD3 can associate not only with Na,K-ATPase but also with H,K-ATPase. However, in situ (stomach), FXYD3 is associated only with Na,K-ATPase since its expression is restricted to mucous cells in which H,K-ATPase is absent. Finally, we identified a transcript variant of FXYD3 that is expressed in undifferentiated but not in differentiated Caco2 cells. [Supported by the Swiss National Fund grant 31-64793.01.]

17. Structural and Functional Interactions of FXYD Regulatory Protein with Shark Na,K-ATPase. FLEMMING CORNELIUS and YASSER A. MAHMMOUD, Department of Biophysics, University of Aarhus, Denmark

In recent studies, we have characterized an FXYD regulatory protein, phospholemman-like protein from shark, PLMS or FXYD10, which is specifically associated with shark Na,K-ATPase (Mahmmoud et al. 2000. J. Biol. Chem. 274: 35969–35977; Mahmmoud et al. 2003. J. Biol. Chem. 278: 37427–37438). As for phospholemman, the FXYD10 protein contains a COOH-terminal protein kinase multi-phosphorylation domain permitting its interaction with the Na,K-ATPase to be dynamically regulated.

Structural interaction of FXYD10 with the Na,K-ATPase -subunit is investigated by intermolecular cross-linking using homobifunctional thiol cross-linking agents. Cross-linking is identified to take place between the COOH-terminal Cys-74 of FXYD10 and Cys-254 in the A-domain of , a position optimal for functional regulation. Thus, interaction of FXYD10 with the A-domain inhibits Na,K-ATPase activity by restricting the free rotation of the A-domain toward the N/P-domains, thereby stabilizing the E1P conformation. Such structural interactions are in accord with the kinetic observations of an increased rate of phosphorylation following truncation of FXYD10 to relieve its interaction with the A-domain.

From controlled proteolysis of purified shark Na,K-ATPase, it is proposed that the FXYD10 functional regulation of Na,K-ATPase may be influenced by interactions with the NH₂ terminus, which control the steady-state E1/E2 conformational poise, and that PKC phosphorylation of the NH₂ terminus may regulate this interaction. Thus NH₂-terminal truncation of the shark Na,K-ATPase a-subunit abolished FXYD10/ association and the FXYD10 functional interactions were abrogated. [Supported by The Danish Medical Research Council.]

18. Regulation of the Na,K-ATPase by FXYD proteins. HAIM GARTY and STEVEN J.D. KARLISH. Department of Biological Chemistry, The Weizmann Institute of Science, Rehovot 76100, Israel

The FXYD proteins are a family of single-span transmembrane proteins with unique tissue distributions and cellular regulations. Five members of this group have been shown to interact with the Na,K-ATPase and alter its kinetic properties. Recently, we have demonstrated that a sixth, and the most distinct, family member, FXYD5 (RIC), also interacts with the Na,K-ATPase (Lubarski et al., abstract at this symposium). CHIF and are two FXYD proteins that are preferentially expressed in the kidney and have nonoverlapping distributions along the nephron. They have opposite effects on the apparent affinity of the Na,K-ATPase for cytoplasmic Na⁺ and thereby provide a convenient means to adjust pumping rates to the unique requirements of different nephron segments. The role of CHIF in aldosterone-dependent electrolyte homeostasis is further established by the phenotypic analysis of CHIF knockout mice. Other studies have characterized structural and functional interactions between these FXYD proteins and the 1ß1 subunits of the pump in transfected HeLa cells and in partly purified renal Na,K-ATPase. Using functional assays, coimmunoprecipitations, and covalent cross-linking we have identified domains and residues participating in the structural and functional interaction of these proteins with the ß pump complex. These studies have highlighted a central role for a number of specific residues in structural interactions, and demonstrated a similar general disposition of the two FXYD proteins with respect to and ß (Lindzen et al. and Fuzesi et al., abstracts at this symposium). Additional residues have been shown to account for the different functional effects of CHIF and . A model of interaction between the transmembrane segments and cytoplasmic sequences of the subunit and a homology model of the 1 subunit has been proposed (Fuzesi et al. 2005. J. Biol. Chem. In press).

19. Molecular Mechanism of Cardiac Calcium Pump Regulation by Phospholamban, Revealed by Site-directed Spectroscopy. DAVID D. THOMAS, Department of Biochemistry, Molecular Biology, and Biophysics, University of Minnesota, Minneapolis, MN

We have used an array of site-directed labeling and spectroscopic techniques to probe directly the interactions, structural changes, and dynamics that are crucial to the action of the Ca-ATPase (SERCA) and its regulation by phospholamban (PLB) in the sarcoplasmic reticulum of the heart. Here are some recent highlights:

(1) Spin-labeling and EPR of single-Cys mutants of PLB show that PLB undergoes a large-scale structural change upon SERCA binding, in which the cytoplasmic domain of PLB is lifted high above the membrane surface (Kirby, T., C.B. Karim, and D.D. Thomas. 2004. Biochemistry. 42:5842–5852).

(2) Fluorescence resonance energy transfer (FRET) from SERCA to PLB in reconstituted membranes shows that PLB binds extremely tightly to SERCA with a K_d that is much less than the concentrations of SERCA and PLB in the cardiac SR membrane. Activation of SERCA by saturating (10 µM) Ca²⁺ causes a structural change in the SERCA-PLB complex but does not cause significant dissociation of PLB from SERCA under physiological conditions (Mueller, B., C.B. Karim, Negrashov, H. Kutchai, and D.D. Thomas. 2004. Biochemistry. 43:8754–8765).

(3) Time-resolved phosphorescence anisotropy (TPA) of SERCA in membranes shows a large-scale change in the structure and dynamics of the P domain of SERCA, with the largest changes observed upon ATP binding, and smaller but opposite changes observed upon Ca binding. ATP causes the P domain to tilt and undergo an order-to-disorder transition (Mueller, B., M. Zhao, I.V. Negrashov, R. Bennett, and D.D. Thomas. 2004. Biochemistry. 43:12846–12854). PLB inhibits the ATP-induced dynamic disorder within the P domain.

(4) NMR and EPR of labeled PLB were used to determine the average structure and topology of the PLB monomer in lipid micelles (Zamoon, J., A. Mascioni, D.D. Thomas, and G. Veglia. 2003. Biophys. J. 85:2589–2598) and to determine the principal sites on PLB that interact with SERCA (Zamoon, J., F. Nitu, C.B. Karim C., D.D. Thomas, and G. Veglia. 2005. Proc. Natl. Acad. Sci. USA. 102:4747–4752).

(5) Solid-phase peptide synthesis was used to label PLB with TOAC, a novel amino acid in which a spin label is rigidly coupled to the carbon and thus reports directly the dynamics of the peptide backbone. EPR in membranes shows that the transmembrane domain of PLB is a highly ordered helix, but the cytoplasmic domain of PLB is in dynamic equilibrium between helical and dynamically disordered conformations (Karim, C.B., T.L. Kirby, Z. Zhang, Y. Nesmelov, and D.D. Thomas. 2004. Proc. Natl. Acad. Sci. USA. 101:14437–14442). Phosphorylation of PLB at Ser 19 increases the population of the disordered conformation (Paterlini, M.G. and D.D. Thomas. 2005. Biophys. J. 88:3243–3251). SERCA binding restricts PLB dynamics. Phosphorylation of SERCA-bound PLB does not dissociate PLB from SERCA, but does change the structure of the bound complex.

These results support a model in which PLB binds tightly to SERCA under all physiological conditions. SERCA is inhibited at low Ca²⁺ by transmembrane interactions that prevent an ATP-induced order-to-disorder transition within the P domain. Phosphorylation of PLB induces a disorder-to-order transition within the PLB cytoplasmic domain. It is the dynamically disordered (extended) conformation of the PLB cytoplasmic domain that is poised to bind to the SERCA cytoplasmic domain and actively relieves SERCA inhibition. [Supported by NIH grants GM27906 and GM64742.]

20. Elucidation of the Ouabain-Binding Site in Na-K-ATPase by Chimeric Approaches. JAN JOEP H.H.M. DE PONT, LI YAN QIU, ELMAR KRIEGER, GIJS SCHAFTENAAR, HERMAN G.P. SWARTS, PETER H.G.M. WILLEMS, and JAN B. KOENDERINK, Department of Biochemistry, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, and Centre for Molecular and Biomolecular Informatics, Radboud University Nijmegen, Nijmegen, The Netherlands (Sponsor: Carel van Os)

Although the global structure of gastric H,K-ATPase is similar to that of Na,K-ATPase, it does not bind ouabain. Previously, we showed that a chimera of these enzymes, in which only the M3-M4 and M5-M6 hairpins originated from Na,K-ATPase, bound ouabain with a high affinity (Koenderink et al. 2000. Proc. Natl. Acad. Sci. USA. 97:11209–11214). We also demonstrated that only three amino acids (Phe⁷⁸³, Thr⁷⁹⁷, and Asp⁸⁰⁴) from the M5-M6 hairpin of Na,K-ATPase were sufficient to confer high-affinity ouabain binding to a chimera that contained only the M3-M4 hairpin of Na,K-ATPase (Qiu et al. 2003. J. Biol. Chem. 278:47240–47247). To further pinpoint the ouabain-binding site, we used a loss-of-function strategy and found that M3 was not important, but that four amino acids present in the extracellular half of M4 were crucial for ouabain binding. In a final gain-of-function study, we showed that a gastric H,K-ATPase that contained these seven (=three plus four) amino acids of Na,K-ATPase bound ouabain with a similar affinity to that of the native enzyme. Similar studies were performed with the nongastric H,K-ATPase that has an intrinsic low affinity for ouabain. In the extracellular half of M4, M5, and M6, there are 14 amino acids that differ from those in Na,K-ATPase. Upon introduction of these Na,K-ATPase amino acids into nongastric H,K-ATPase, a high-affinity ouabain-binding site was obtained. By similar approaches to those noted above, we demonstrated that introduction of only five of these amino acids was sufficient to obtain a high-affinity ouabain-binding site in nongastric H,K-ATPase. Based on the E₂P crystal structure of Ca²⁺-ATPase (Toyoshima et al. 2004. Nature. 432:361–368), we constructed a homology model for the ouabain-binding site of Na,K-ATPase in which most of the amino acids we found, as well as several amino acids postulated earlier, play crucial roles.

21. GPCR Signals and Intracellular Traffic of Na⁺,K⁺-ATPase. ALEJANDRO M. BERTORELLO, Department of Medicine, Membrane Signaling Networks, Karolinska Institutet, Karolinska University Hospital-Solna, 171 76 Stockholm, Sweden (Sponsor: David Gadsby)

The establishment of cell polarity in transporting epithelia requires the localization of ion transport proteins (e.g., Na⁺,K⁺-ATPase) in specific domains of the cell. In response to G protein–coupled receptor signals (such as dopamine, parathyroid hormone), a decrease in Na⁺,K⁺-ATPase activity from renal epithelial cells is mediated by endocytosis of active units, whereas stimulation of its activity in renal (angiotensin) and lung epithelial cells (isoproterenol, dopamine) is the result of increased recruitment from intracellular organelles (endosomes) to the plasma membrane. It is thus important for the development and maintenance of such polarized structures that during regulation by receptor signals the Na⁺,K⁺-ATPase units be sorted adequately. Therefore, it is envisioned that movement of Na⁺,K⁺-ATPase molecules from, or into, the plasma membrane during their regulation by receptor signals would require a highly developed organization and synchronization of spatial and temporal interactions between many signaling networks. An example of such signal compartmentalization is provided by the renal and lung epithelia, where dopamine regulates the same target, the Na⁺,K⁺-ATPase, by operating different signaling networks that result in opposite consequences for activity of the target.

22. Significance of the Conserved Cardiac Glycoside Binding Site of the 2 Isoform of the Na,K-ATPase. JERRY B. LINGREL,¹ IVA DOSTANIC,¹ JOHN LORENZ,² JAMES W. VAN HUYSSE,³ and JONATHAN NEUMANN,^{1 1}Department of Molecular Genetics, Biochemistry, and Microbiology and ²Department of Molecular and Cellular Physiology, University of Cincinnati College of Medicine, Cincinnati, OH 45267; ³University of Ottawa Heart Institute, Hypertension Unit, Ottawa, Ontario, Canada

The Na,K-ATPase contains a highly conserved cardiac glycoside binding site, which occurs in organisms as diverse as Drosophila, frogs, chickens, mice, and humans. One exception is the 1 isoform of rodents, which is relatively insensitive to this class of compounds. It is reasonable to hypothesize that this conserved binding site plays a biological role. To test this, we developed animals where the 2 isoform of the Na,K-ATPase of mice is made resistant to ouabain. These animals survive and have normal development, survival rate, and basal cardiovascular hemodynamics. However, when animals are treated with ACTH, which induces hypertension in wild-type animals, no increase in blood pressure occurs in animals with the ouabain-resistant 2 isoform. These findings indicate that the cardiac glycoside binding site plays a physiological role, at least in ACTH-induced hypertension, and further suggests that there is an endogenous ligand interacting with this site. The most logical candidate is the endogenous cardiac glycosides, which have been identified by several laboratories. Our studies indicate that these compounds increase in both the wild-type and targeted mice. The conclusion from our studies is that the cardiac glycoside binding site of the Na,K-ATPase plays an intrinsic physiological role.

23. Gene Targeting Studies of Ca²⁺-transporting ATPases. GARY E. SHULL,¹ GBOLAHAN W. OKUNADE,¹ MARIAN MILLER,² and VIKRAM PRASAD,^{1 1}Department of Molecular Genetics, Biochemistry, and Microbiology and Department of Environmental Health,² University of Cincinnati College of Medicine, Cincinnati, OH 45267 (Sponsor: Jack Kaplan)

Ca²⁺ gradients required for Ca²⁺-signaling and homeostasis are maintained by P-type plasma membrane and intracellular Ca²⁺-ATPases. Gene-targeting technology is being used to systematically analyze the biological functions of specific Ca²⁺-ATPase isoforms. The phenotypes of mice carrying null mutations in plasma membrane Ca²⁺-ATPases PMCA1, PMCA2, and PMCA4 (Kozel et al. 1998. J. Biol. Chem. 273:18693–18696; Okunade et al. 2004. J. Biol. Chem. 279:33742–33750; Reinhardt et al. 2004. J. Biol. Chem. 279:42369–42373; Schuh et al. 2004. J. Biol. Chem. 279:28220–28229) indicate that PMCA1 serves essential housekeeping functions, whereas PMCA2 and PMCA4 serve more specialized physiological functions. PMCA1 null mutants die during the early stages of embryogenesis, and loss of a single copy of the PMCA1 gene can exacerbate an apoptosis phenotype observed in vascular smooth muscle of PMCA4 null mice. Although PMCA4 is widely expressed and is the most abundant isoform in many tissues, null mutants appear to be healthy. However, male PMCA4 null mutants are infertile due to a failure of hyperactivated sperm motility resulting from the absence of PMCA4 in the principal piece of the sperm tail, where it serves as the major Ca²⁺ extrusion mechanism controlling sperm Ca²⁺ concentrations. Loss of PMCA2 in sensory hair cells of the inner ear causes profound deafness and balance defects, and loss of PMCA2 in lactating mammary glands causes a severe deficit in the Ca²⁺ concentrations in milk. Homozygous null mutations in sarco(endo)plasmic reticulum Ca²⁺-ATPase isoform 2 (SERCA2) leads to embryolethality, whereas loss of a single copy of the gene causes impaired cardiac contractility and squamous cell tumors involving keratinized epithelial cells (Prasad et al. 2004. Biochem. Biophys. Res. Comm. 322:1192–1203). The latter finding provides the first direct demonstration that a perturbation of Ca²⁺ homeostasis or signaling can be a primary initiating event in cancer. [Supported by NIH grant HL61974.]

Poster Abstracts
24. Purification of Na,K-ATPase Expressed in Pichia Pastoris: Specific Interactions with Lipids. EYTAN COHEN,¹ RIVKA GOLDSHLEGER,¹ DANIEL M. TAL,¹ CHRISTINE EBEL,² MARC LE MAIRE,³ and STEVEN J.D. KARLISH,^{1 1}Department of Biological Chemistry, Weizmann Institute of Science, Rehovot, Israel; ²Institut de Biologie Structurale J.P. Ebel, UMR 5075 CEA-CNRS-UJF, F-38027 Grenoble, Cedex 01, France; ³Unité de Recherche Associée CNRS 2096, Commissariat à l'Energie Atomique Saclay, 91191 Gif-sur-Yvette cedex, France (Sponsor: David Gadsby)

Na⁺,K⁺-ATPase (porcine /his₁₀-ß) has been expressed in Pichia Pastoris, solubilized in n-dodecyl-ß-maltoside and purified by metal chelate bead chromatography combined with size exclusion HPLC. The recombinant protein is inactive if the purification is done without added phospholipids. The neutral phospholipid dioleoyl phosphatidylcholine (DOPC) preserves Na⁺,K⁺-ATPase activity of protein prepared in a Na⁺-containing medium, but activity is lost in a K⁺-containing medium. By contrast, the acid phospholipid dioleoyl phosphatidylserine (DOPS) preserves activity in either Na⁺- or K⁺-containing media. The presence of cholesterol inhibits Na,K-ATPase activity measured at 0°C, while at 37°C, cholesterol stabilizes the protein against thermal inactivation without affecting Na,K-ATPase activity. The stabilizing effect of cholesterol is detectable in the presence of DOPC and not in the presence of DOPS. In optimal conditions, the protein is stable for about 2 wk at 0°C. Both recombinant Na⁺,K⁺-ATPase and native pig kidney Na⁺,K⁺-ATPase, dissolved in n-dodecyl-ß-maltoside, appear to be mainly stable monomers (/ß), as judged by size exclusion chromatography and sedimentation velocity. Na⁺,K⁺-ATPase activities at 37°C of the size exclusion HPLC-purified recombinant and renal Na⁺,K⁺-ATPase are comparable but are lower than that of membrane-bound renal Na⁺,K⁺-ATPase. The ß subunit is expressed in Pichia Pastoris as two lightly glycosylated polypeptides and is quantitatively deglycosylated by endoglycosidase-H at 0°C, to a single polypeptide. Deglycosylation inactivates Na⁺,K⁺-ATPase prepared with dioleoyl phosphatidylcholine, whereas dioleoyl phosphatidylserine protects after deglycosylation, and Na⁺,K⁺-ATPase activity is preserved. This work demonstrates an essential role of phospholipid interactions with Na⁺,K⁺-ATPase, including a direct interaction of dioleoyl phosphatidylserine, probably another interaction of either the neutral or acid phospholipids, as well as cholesterol. A role for the ß subunit in stabilizing conformations of Na⁺,K⁺-ATPase (or H⁺,K⁺-ATPase) with occluded K⁺ ions can also be inferred. Purified recombinant Na⁺,K⁺-ATPase could become an important experimental tool for various purposes including, hopefully, structural work.

25. Thermal Stability of a Thermophilic P-type ATPase. F. LUIS GONZÁLEZ FLECHA,¹ DIEGO I. CATTONI,¹ ATIN K. MANDAL,² DIPTI SHARMA,³ GERMANO S. IANNACCHIONE,³ and JOSÉ M. ARGÜELLO,^{2 1}Instituto de Química y Fisicoquímica Biológicas, Universidad de Buenos Aires-CONICET, Argentina; ²Department of Chemistry and Biochemistry and ³Department of Physics, Worcester Polytechnic Institute, Worcester, MA (Sponsor: Jack H. Kaplan)

Protein stability is the result of a delicate balance between stabilizing and destabilizing interactions. While thermal denaturation of globular proteins is a well-characterized process, little is known about the thermal stability of membrane proteins. In addition, the lack of information on the stability of thermophilic membrane proteins is remarkable. The aim of this work was to initiate the characterization of the thermal denaturation process of CopA, a thermophilic P_IB-type Cu-ATPase from Archaeoglobus fulgidus. CopA was heterologously expressed in Escherichia coli, solubilized in dodecylmaltoside (DDM), and affinity purified. The resulting enzyme retained thermophilic characteristics with maximum activity at 75°C and an E^a = 103 kJ/mol. DSC analysis of CopA showed a thermal transition at 81°C, a value significantly higher than that determined for mesophilic P-type ATPases. The presence of ATP-Mg further stabilized the protein, shifting its T_m to 105°C. As expected, CopA denaturation was found to be much slower than that of mesophilic P-type ATPases in similar conditions. The enzyme preparation incubated at 75°C showed an irreversible exponential decrease in enzyme activity and intrinsic fluorescence intensity. This inactivation was not associated with either fragmentation or formation of SDS-stable aggregates of the protein. Moreover, the first-order rate of thermal inactivation suggests a two-state process involving only fully active and inactive molecules. CopA reconstitution in mixed micelles of asolectin and DDM before the inactivation further increased the enzyme stability. These results indicate that thermophilic membrane proteins are more stable than their mesophilic counterparts, that they retain their stability even when heterologously expressed, and therefore, that their stability appears to depend largely on intramolecular interactions. [Supported by NSF grants MCM-0235165 and OISE-0436435 to J.M. Argüello, and DMR-0092786 to G.S. Iannacchione, and ANPCyT grant PICT-11138 to F.L.G. Flecha.]

26. The Modulatory ATP Binding Site of the Calcium Pump. ANNE-MARIE LUND JENSEN,¹ THOMAS LYKKE-MØLLER SØRENSEN,¹ JESPER VUUST MØLLER,³ and POUL NISSEN,^{1 1}Department of Molecular Biology and ²Department of Biophysics, University of Aarhus, Aarhus, Denmark

In skeletal muscles, the dominant P-type ATPase is sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA). SERCA is responsible for the reuptake of cytosolic Ca²⁺ (released during muscle contraction) into the sarcoplasmic reticulum. The active transport by P-type ATPases is fueled by ATP and involves the formation of a covalent aspartyl-phosphoanhydride intermediate. How ATP is involved as the key substrate in formation of the E1~P state has now become well characterized at the biochemical and structural level (Sørensen et al. 2004. Science. 304:1672–1675). However, ATP also exhibits a general, stimulatory effect on the functional transitions relating to the E2 states (Stahl et al. 1984. Biochemistry. 23:5389–5392), which indicates the existence of a noncatalytic, modulatory site.

We have determined the crystal structure of SERCA in the E2 state with the inhibitor thapsigargin and in presence and absence of the ATP analogue AMPPCP at 3.1 Å and 2.8 Å resolution, respectively. The E2:TG:AMPPCP structure shows important aspects of the modulatory binding site that indeed are different from the catalytic site. Although the N-domain interacts with the adenosine moiety in essentially the same way as in the Ca2E1-AMPPCP state, it is the Glu439 residue of the N-domain, rather than the phosphorylation site centered on Asp351 of the P-domain, that interacts with the phosphate groups of ATP. This is consistent with iron-cleavage data and biochemical studies of an E439A mutant (Patchornik et al. 2002. Biochemistry. 41:11740–11749; Inesi et al. 2004. J. Biol. Chem. 279:31629–31637).

We anticipate that the modulatory ATP binding site is fully occupied at typical physiological conditions with millimolar ATP concentrations available. We therefore propose a new route of reactivation of the functional cycle of SERCA, going directly from an E2-ATP state to the Ca2E1-ATP state. [Supported by the Danish Medical Science Research Council, the Novo Nordisk Foundation, and the Lundbeck Foundation.]

27. Crystallization of a Mammalian Membrane Protein Overexpressed in Saccharomyces cerevisiae. MARIE JIDENKO,¹ RIKKE C. NIELSEN,² THOMAS LYKKE-MØLLER SØRENSEN,² JESPER V. MØLLER,³ MARC LE MAIRE,¹ POUL NISSEN,² and CHRISTINE JAXEL,^{1 1}Unité de Recherche Associée 2096 of the Centre National de la Recherche Scientifique and Service de Biophysique des Fonctions Membranaires, Département de Biologie Joliot Curie, CEA Saclay, 91191 Gif sur Yvette Cedex, and Laboratoire de Recherche Associé 17V and Institut Fédératif de Recherches 46, Université Paris Sud, Paris, France; ²Department of Molecular Biology and ³Department of Biophysics, Institute of Physiology and Biophysics, University of Aarhus, DK-8000 Aarhus C, Denmark (Sponsor: Philippe Champeil)

The Ca²⁺-ATPase SERCA1a from rabbit has been overexpressed in Saccharomyces cerevisiae. This membrane protein was purified by avidin agarose affinity chromatography based on natural biotinylation in the expression host, followed by HPLC gel filtration. Both the functional and structural properties of the overexpressed protein validate the method. Thus, calcium-dependent ATPase activity and calcium transport are intact after reconstitution in proteoliposomes. Moreover, the recombinant protein crystallizes in a form that is isomorphous to the native SERCA1a protein from rabbit and the diffraction properties are similar. Even if other protein purification methods based on biotin–avidin interaction were used, this is the first example of a successful crystallization of a mammalian membrane protein derived from a heterologous expression system, and it opens the door for the study of mutant forms of SERCA1a. This procedure is likely to be successful also in the case of other eukaryotic membrane proteins, which are generally difficult to purify and crystallize.

28. Atomic Structure of a Covalently Phosphorylated Intermediate of SERCA1 Ca²⁺-ATPase: Normal Mode Fits of Electron Densities. EDWARD BEAUMONT,¹ BERTRAND FOURNIER,² DAVID STOKES,³ KONRAD HINSEN,² and JEAN-JACQUES LACAPERE,^{1 1}U683 INSERM 16, F-75870 Paris Cédex 18, ²Laboratoire Léon Brillouin (CEA-CNRS), F-91191 Gif sur Yvette Cedex, Paris, France; ³Skirkball Institute of Biomolecular Medicine, New York University, New York, NY

Three-dimensional structure of a stabilized phosphoenzyme intermediate of the Ca²⁺-ATPase has been recently obtained using cryoelectron microscopy of two-dimensional crystals (Stokes et al. 2005. J. Biol. Chem. 280:18063–18072). The electron microscopy (EM) images show tubular crystals formed in the presence of decavanadate where the protein dimers are arranged in rows winding around the tubes. The structure was solved at 8-Å resolution. In a first step, a single protein has been extracted and characterized. We used a recently published method (Hinsen et al. 2005. Biophys. J. 88:818–827) for the flexible docking of high-resolution structures into the EM density for this new conformation of the same protein. In brief, the structure fitting consists of an initial step of orientation of the atomic structure relative to the EM density, and an iterative deformation of the structure using a set of low energy normal modes was used to improve the fit of the EM density. We previously showed that normal modes can describe the massive domain changes that the Ca-ATPase undergoes in its catalytic cycle (Reuter et al. 2003. Biophys. J. 85:2186–2197). The atomic structures of different reaction intermediates of the Ca-ATPase (Toyoshima et al. 2002. Nature. 418:605–611; Sorensen et al. 2004. Science. 304:1672–1675; Toyoshima et al. 2004. Nature. 430:529–535) were used as starting points leading to different fitted models that were compared. The motions associated with the conformational transition between initial and fitted-final models were further analyzed. In a second step, part of the crystalline tube was used and several Ca-ATPases were fitted into it. Final fitted models obtained after flexible docking were used to characterize the protein–protein interactions within the crystal and the atomic location of the decavanadate complexes.

29. Structural Changes in Sodium Potassium ATPase Induced by Ion Binding and Membrane Cholesterol Depletion. CLAUS HELIX NIELSEN,¹ SALIM ABDALI,¹ JENS AUGUST LUNDBÆK,² and FLEMMING CORNELIUS,^{3 1}Quantum Protein Center, Technical University of Denmark, Lyngby, Denmark; ²Department of Physiology and Biophysics, Weill Medical College, Cornell University, New York, NY; ³Department of Biophysics, University of Aarhus, Aarhus, Denmark

We have used Raman spectroscopy (30 mW at = 532 nm, 5 min acquisition time) as a tool to investigate conformational changes of the unphosphorylated sodium-potassium-adenosine-triphosphatase (Na⁺,K⁺-ATPase) enzyme from shark rectal gland that occur upon sodium ion binding or membrane cholesterol depletion. Using the well-established effect of buffer composition on the E2–E1 conformational equilibrium, our results show that the protein embedded in native membrane fragments becomes more helical when going from the E2 to the E1 conformation (i.e., upon binding sodium) as evidenced by the changes in the amide I and III bands. Upon cholesterol depletion with methylated-ß-cyclodextrin (20 mM for 30 min at 20°C), the E1_CholDepl and E2_CholDepl spectra are similar and both resemble the E1 spectrum more than the E2 spectrum in the amide I and III bands. For the four conformers (E1, E2, E1_CholDepl, E2_CholDepl) the region between 1020 and 1120 cm^–1 is characterized by at least five overlapping peaks. All conformers share a peak at 1060 cm^–1, but E1, E1_CholDepl, and E2_CholDepl all have higher intensities at higher wavenumber bands compared with the E2 spectrum in that region, consistent with the changes observed for the amide I and III bands. In all conformers, the I₈₂₅/I₈₅₃ tyrosine doublet in the spectra is not significantly changed, indicating no differences in the hydrogen-bonding environment, which poses constraints on models for the E2E1 transition both in normal and cholesterol-depleted membranes. Taken together, our results show that (part of) the conformational changes in the Na⁺,K⁺-ATPase upon sodium binding can be mimicked by depletion of membrane cholesterol. This suggests that membrane protein function can be regulated by altering membrane physical properties. [Supported by the Danish National Research Foundation.]

30. Dephosphorylation of the Calcium pump Coupled to Counterion Occlusion. CLAUS OLESEN,¹ THOMAS LYKKE-MØLLER SØRENSEN,² RIKKE C. NIELSEN,² ANNE-MARIE L. JENSEN,² JESPER VUUST MØLLER,¹ and POUL NISSEN,^{2 1}Department of Physiology and Biophysics and ²Centre for Structural Biology, University of Aarhus, Aarhus, Denmark

To understand the dephosphorylation mechanism and to reveal the intramolecular coupling between cation transport and ATP hydrolysis, we crystallized sarcoplasmic reticulum Ca²⁺-ATPase (SERCA1a) in complex with aluminium fluoride. This represents the transition state of hydrolysis of the counterion-bound (protonated) phosphoenzyme (Olesen et al. 2004. Science 306:2251–2255). The planar aluminium fluoride group is located between the conserved Asp351 side chain and a water molecule, thus representing the transition state of hydrolysis of the phosphoenzyme. The water molecule is positioned, and activated for a nucleophilic attack, by Ser181 and Glu183 of the conserved TGES motif of the A-domain. This arrangement overlaps with the position of ADP:AlF₄^– in phosphoryl transfer in the E1~P structure (Sorensen et al. 2004. Science 304:1672–1675). The domain movements associated with the formation of the dephosphorylation site depend on the release of ADP after ATP phosphorylation, and the dephosphorylation reaction cannot proceed before the bound Ca²⁺ ions have been exchanged for protons that become occluded. The helix bundle constituting the proper arrangement of the dephosphorylation site for catalytic activity is stabilized by an integral K⁺ site (Sorensen et al. 2004. J. Biol. Chem. 279:46355–46358), explaining the stimulatory effect of monovalent cations on dephosphorylation. The new structure provides a rationale for the vectorial transport of Ca²⁺ and couples it with the counterion exchange needed for dephosphorylation.

31. Purification, Kinetic Characterization, and Initial Crystallization of an Archaeal P-type ATPase. BJØRN PANELLA PEDERSEN,¹ THOMAS LYKKE-MØLLER SØRENSEN,² and POUL NISSEN,^{1 1}Department of Molecular Biology, University of Aarhus, Aarhus, Denmark; ²Diamond Light Source Ltd., Rutherford Appleton Laboratory, Oxfordshire, UK

The sarcoplasmic reticulum Ca²⁺-ATPase from skeletal muscle (SERCA 1a) is so far the only P-type ATPase cation pump for which atomic structures are available. We therefore aim at structure-based investigations of other P-type ATPases, and we describe the successful cloning and heterologous expression of a number of bacterial and archaeal P-type ATPases.

A large group of P-type ATPases comprising both bacterial and eukaryotic proteins is specific to soft cations such as Cu²⁺ and Zn²⁺ and these cation pumps are essential for many detoxification systems and for maintaining the intracellular metal-ion homeostasis in cells. The best known example being the Wilson and Menkes proteins from humans involved in intracellular copper trafficking.

A P-type ATPase from the archaeal species Thermoplasma acidophilum (gene code TA1143) shows sequence homology to the Cu/Ag transporting ATPases and is found to be highly expressed in Escherichia coli membranes. Solubilization in DDM followed by affinity chromatography and size-exclusion chromatography yields up to 3–4 mg pure, monodisperse protein per liter E. coli culture.

Using a para-nitrophenyl phosphate assay, the protein expressed is found to be an active ATPase with a V_max of 200 nmol/(mg*min) and a K_m of 5.5 mM. This activity can be competitively inhibited by the addition of ATP or an ATP homologue and it is dependent on Mg²⁺ ions as expected for a P-type ATPase. Identification of the cation specificity is currently in progress.

Crystallization screening using the vapor diffusion technique has produced several hits. The use of the detergent C₁₂E₈ proved to be critical for the formation of protein crystals.

32. The Average Conformation at Micromolar [Ca²⁺] of Ca²⁺-ATPase with Bound Nucleotide Differs from that Adopted with the Transition State Analogue ADP.AlFx or with AMPPCP Under Crystallization Conditions at Millimolar [Ca²⁺]. MARTIN PICARD,¹ CHIKASHI TOYOSHIMA,² and PHILIPPE CHAMPEIL,^{1 1}Unité de Recherche Associée 2096 (CNRS) and Service de Biophysique des Fonctions Membranaires (DBJC, CEA), 91191 Gif-sur-Yvette Cedex, France; ²Institute of Molecular and Cellular Biosciences, University of Tokyo, Tokyo, Japan

Crystalline forms of detergent-solubilized sarcoplasmic reticulum Ca²⁺-ATPase, obtained in the presence of either a substrate analogue, AMPPCP, or a transition state complex, ADP.fluoroaluminate, were recently described to share the same general architecture, despite the fact that, when studied in a test tube, these forms show different functional properties. Here, we show that the differences in the properties of the E1.AMPPCP and the E1.ADP.AlFx membranous (or solubilized) forms are much less pronounced when these properties are examined in the presence of 10 mM Ca²⁺ (the concentration prevailing in the crystallization media) than when they are examined in the presence of the few µM Ca²⁺ known to be sufficient to saturate the transport sites. This concerns various properties, including ATPase susceptibility to proteolytic cleavage by proteinase K, ATPase reactivity toward SH-directed Ellman's reagent, ATPase intrinsic fluorescence properties (here described for the E1.ADP.AlFx complex for the first time), and also the rates of ⁴⁵Ca²⁺–⁴⁰Ca²⁺ exchange at site "II." These results solve the above paradox at least partially and suggest that the presence of a previously unrecognized Ca²⁺ ion in the Ca²⁺-ATPase.AMPPCP crystals should be reinvestigated. On the other hand, they emphasize the fact that the average conformation of the E1.AMPPCP complex under usual conditions in the test tube differs from that found in the crystalline form. The extended conformation of the nucleotide revealed in the E1.AMPPCP crystalline form might be indicative only of the requirements for further processing of the complex, toward the transition state leading to phosphorylation and Ca²⁺ occlusion. [Supported by HFSPO grant RGP 0060/2001-M.]

33. Structure of Na.K-ATPase as Analyzed by Cryoelectron Microscopy. PASI PURHONEN,¹ HANS HEBERT,¹ KAREN THOMSEN,² and ARVID B. MAUNSBACH,^{2 1}Department of Biosciences, Karolinska Institutet and School of Technology and Health, Royal Institute of Technology, S-14157 Huddinge, Sweden; ²The Water and Salt Research Center, Department of Cell Biology, Institute of Anatomy, University of Aarhus, DK-8000 Aarhus, Denmark (Sponsor: David Gadsby)

We have analyzed the molecular structure of Na,K-ATPase by electron crystallography from frozen-hydrated two-dimensional crystals induced in purified membranes from the outer medulla of pig kidney. Data were collected from 141 small Na,K-ATPase 2-D crystals in E2 conformation, and used for determination of a 3-D structure at <10 Å. The Na,K-ATPase protomer contains all three subunits (, ß, and ) present in the kidney. We have compared our Na,K-ATPase model from cryoelectron microscopy to the atomic structure of the related Ca-ATPase in different conformations obtained by x-ray crystallography in the laboratories of Toyoshima (Toyoshima et al. 2000. Nature. 405:647–655; 2002. 418:605–611; 2004. 430:529–535; 2004. 432:361–368) and Nissen (Nissen et al. 2004. Science. 304:1672–1675; 2004. 306:2251–2255). The best fit between these two ATPases is observed in the transmembrane region, where the Na,K-ATPase structure shows two groups of distinct densities, being composed of the 10 helices from the catalytic subunit. Extra density areas in the transmembrane region suggest the positions for the single helices arising from the Na,K-ATPase ß and subunits, which are absent in Ca-ATPase. Thus, comparisons with Ca-ATPase suggest that the subunit of Na,K-ATPase is located in close association with the M2 and M9 transmembrane helices. The overall structure of the subunit of Na,K-ATPase with respect to the nucleotide binding (N), phosphorylation (P), and actuator (A) domains is similar to the X-ray structure of Ca-ATPase in E2 conformation. The angular difference for the N-domain position of Na,K-ATPase in the observed E2 and in the modeled E1 conformation is 40°, similar to the difference observed between the E1 and E2 forms of Ca-ATPase. The present observations and comparisons suggest that large conformational changes occur also between the E1 and E2 forms of Na,K-ATPase.

34. Implications of SERCA Structural Models for ATP Binding Events in Na,K-ATPase ("NaKA"). JOHN S. WILLIS and MARK A. MILANICK, Department of Cellular Biology, University of Georgia, Athens, GA 30601; Department of Medical Pharmacology and Physiology, School of Medicine, and Dalton Cardiovascular Research Center, University of Missouri, Columbia, MO 65211

Based on SERCA structural models, ATP is thought to bind to the N-domain in E1 when N and P domains are separated and there is open access to the nucleotide binding site (PDB 1SU4). Subsequently, the N and P domains converge and the -phosphate approaches D351. The close apposition of N and P domains encloses the ATP binding site at this stage and continues to surround it throughout the remainder of the pump cycle. In NaKA, this scenario raises the questions of when ADP can escape and ATP can bind. In NaKA 1 mM ATP binds the enzyme in E2 and promotes the conversion from E2 to E1. In the SERCA model for E2 (1IWO), the ATP binding site is obscured. Similarly, the occurrence of ADP-ATP exchange in NaKA indicates that the binding site is available at E1PADP, even though the SERCA model for this stage is tightly closed (1WPE, F487-D351, 11 Å). However, the subsequent SERCA structure (E2P, 1WPG), although still loosely closed, shows a gap between N and P large enough to allow the passage of nucleotides to the interior. In 1WPG, however, the F487-D351 distance is not only longer than an ATP (27 Å vs. 14 Å), but the TGES region from the A domain also intrudes between N and P. Thus, while the nucleoside head of ATP could slip into the N pocket, the -phosphate could not approach D351. With conversion to E1, however, ATP would be well positioned to phosphorylate D351. This scheme offers several explanations for low affinity in E2 (constrained geometry, limited time) and the possibility that the fully open configuration of SERCA (1EUL) might not exist during the NaKA pump cycle at physiological [ATP]. It also offers the prediction that ATP is entrapped during the interval between E2 and E1P.

35. Protonation of Ca²⁺-ATPase Residues upon Ca²⁺ Release. JULIA ANDERSSON,¹ KARIN HAUSER,² and ANDREAS BARTH,^{1 1}Department of Biochemistry and Biophysics, Stockholm University, Stockholm, Sweden; ²Institute for Biophysics, Johann Wolfgang Goethe-University, Frankfurt am Main, Germany

We have studied protonation of the sarcoplasmic reticulum Ca²⁺-ATPase during Ca²⁺ release and E2P formation (Ca₂E1E2P) using rapid scan Fourier transform infrared spectroscopy. The reaction has been investigated from pH 6.0 to 9.0. Infrared spectra show four signals in the spectral region of protonated carboxyl groups at pH 6.0–7.5 and only two at pH 8.0–9.0. The results show that at least two of the protonated carboxyl groups of E2P have a pK of 7.7. We have concluded that these carboxyl groups participate in H⁺ countertransport. To identify these carboxyl groups and to assign the IR bands, multiconformation continuum electrostatic calculations (MCCE) have been performed to calculate the residues' ionization, at various pH, for the calcium-free and the calcium-occluded structure, respectively. The combination of infrared measurements and MCCE calculations clearly indicates that Asp⁸⁰⁰ is involved in the proton countertransport whereas Glu³⁰⁹ is not. The second carboxyl group involved in the countertransport might be Glu⁹⁰⁸. Our results also indicate a pH-dependent conformational change in a ß-sheet or turn structure of E2P.

Additionally, we have tentatively assigned a band to the C=O bond of the phosphorylated Asp 351. Based on infrared data, we concluded that the bond strength is essentially unchanged but is slightly reduced in E2P compared with Ca₂E1P. This reduction is larger when Mg²⁺ is bound to the aspartyl phosphate.

36. Role of the Subunit Second Extracellular Loop in the Accessibility of K⁺ Ions to their Binding Site in Na,K-ATPase. OIHANA CAPENDEGUY and JEAN-DANIEL HORISBERGER, Department of Pharmacology and Toxicology, University of Lausanne, Lausanne, Switzerland (Sponsor: J.-D. Horisberger)

By homology with the structure of the SERCA and supported by experimental studies, the fourth, fifth, and sixth transmembrane helices of the subunit of Na,K-ATPase have been proposed to participate in Na⁺ and K⁺ binding sites. To improve the understanding of control of access of K⁺ ions to their binding sites from the extracellular side, we have focused on the second extracellular loop linking transmembrane segments 3 and 4. To this end we have mutated 12 residues of this loop in the rat 1 subunit, from E314 to G326, into cysteine and studied these mutants by electrophysiological techniques. We measured the functional expression and the accessibility of the cysteines to a membrane-impermeant cationic thiol reagent (MTSET); and, in selected mutants, we studied the voltage-dependent activation by extracellular K⁺, and the effect of MTSET binding on activation by K⁺ and on the presteady-state ouabain-sensitive current. Several mutations resulted in a large increase of the apparent affinity for extracellular K⁺ both in the presence and in the absence of extracellular Na⁺. Four mutants (E314C, Y315C, W317C, and I322C) were strongly modified by MTSET, three of these were inhibited and one E314C was stimulated. The effects of the E314 mutation (reduced Vmax, increased apparent affinity for K⁺, decreased maximal translocated charges) were all corrected by MTSET treatment. For W317C and I322C, MTSET induced a right shift of the midpoint potential, indicating an increase of the extracellular affinity for Na⁺. These results combined with those of a preceding study (Capendeguy and Horisberger. 2005. J. Physiol. In press) on the third extracellular loop show that these two loops have a complementary function, controlling the accessibility of K⁺ ions to their binding site from the extracellular side by a kind of "gates game" and modulating the E1/E2 equilibrium. [Supported by the Swiss National Fund grant 31-65441.01 to J.-D. Horisberger.]

37. The Cadmium Transport Site of CadA, the Cd²⁺-ATPase from Listeria monocytogenes. CHEN-CHOU WU, ANNE MARTEL, AURELIE GARDARIN, ELISABETH MINTZ, FLORENT GUILLAIN, and PATRICE CATTY, Laboratoire de Biophysique Moléculaire et Cellulaire, UMR 5090 CEA-CNRS-Université Joseph Fourier, CEA/DRDC/BMC, 38054 Grenoble Cedex 9, France

The Zn²⁺/Cd²⁺/Pb²⁺-ATPases constitute a bacterial subfamily of P1-type ATPases that behave as detoxification pumps, among which is found CadA, the Cd²⁺-ATPase from Listeria monocytogenes (Bal et al. 2001. FEBS Lett. 506:249–252; Bal et al. 2003. Biochem. J. 369:681–685).

Whereas it is the major determinant of the resistance to Cd²⁺ in L. monocytogenes, CadA expressed in the yeast Saccharomyces cerevisiae did just the opposite to what was expected, as it strikingly decreased the Cd²⁺ tolerance of these cells. Yeast cells expressing the nonfunctional CadA mutant Asp³⁹⁸Ala could grow on selective medium containing up to 100 µM Cd²⁺, whereas those expressing the functional protein could not grow in the presence of 1 µM Cd²⁺. The CadA-GFP fusion protein was localized in the endoplasmic reticulum membrane, suggesting that yeast hypersensitivity was due to Cd²⁺ accumulation in the reticulum lumen (Wu et al. 2004. Biochem. Biophys. Res. Comm. 324:1034–1040).

This phenotype was used as a powerful tool to select among 35 mutations in the transmembrane region of CadA, those that could affect the activity of the protein. Functional studies of the selected mutants produced either in yeast or in Sf9 cells revealed that not only the two cysteines of the canonical CPC motif, but also amino acids located in the third, fourth, and eighth transmembrane helices, were important for CadA activity. This leads us to propose a metal transport site different from those recently proposed for the Cu⁺-ATPases, Ccc2p from S. cerevisiae (Lowe et al. 2004. J. Biol. Chem. 279:25986–25994) and CopA from Archeoglobus fulgidus (Mandal et al. 2004. J. Biol Chem. 279:54802–54807).

38. Testing the Accuracy of Homology Modeling: Similar Inhibition Patterns by Cytosolically Acting Organic Cations on Na,K- and PM Ca-ATPase. CRAIG GATTO,¹ JEFF B. HELMS,¹ SHENG-YOU HUANG,² XIAOQIN ZOU,² KRISTA L. ARNETT,³ and MARK A. MILANICK,^{3 1}Department of Biological Sciences, Illinois State University, Normal, IL; ²Dalton Cardiovascular Research Center and Department of Biochemistry and ³Department of Medical Pharmacology and Physiology, University of Missouri, Columbia, MO (Sponsor: Jack H. Kaplan)

We studied the effect of organic cations on the inhibition of the Na,K pump (sheep kidney) and the PM Ca pump (cow erythrocytes). Homology models suggested that the monovalent cations, guanidine and tetramethyl guanidine (TMG), would inhibit both pumps at cytoplasmic accessible binding sites. On the Na pump, inhibition of ATPase activity by guanidine, TMG, and diamine was competitive with Na, and their inhibition of pNPPase, but not of ATPase, was competitive with K, consistent with these organic cations binding predominantly from the inside. The concentration dependence for guanidine and TMG inhibition was consistent with cooperative binding of two inhibitor molecules. These results are in contrast with a known extracellular site inhibitor, tetrapropylammonium (TPA), which does compete with K for ATPase but not with Na (Gatto et al. 2005. Am. J. Physiol. Cell Physiol. In press). As also predicted by homology modeling, guanidine and TMG were competitive with calcium for PM Ca-ATPase activity. Again, the concentration dependence of inhibition was consistent with cooperative binding of two inhibitor molecules. Interestingly, the modeling of PMCA predicted that diamine inhibition should differ from guanidine. Subsequent experiments confirmed that diamine inhibited PMCa, but it was not competitive with calcium. Moreover, diamine inhibition was not cooperative and could be accounted for with single molecule binding. In addition, we observed that the extracellular site inhibitor TPA poorly inhibited the PM Ca pump (<10% inhibition at 100 mM TPA). The homology models of both the Na,K and PM Ca pumps provide likely explanations for TPA exclusion from cytoplasmic binding. Even though the models were based on the SERCA structure with occluded calcium (a conformation not expected to bind cytosolic calcium), we were able to predict the distinct cytosolic inhibitor binding characteristics of the Na,K pump and PM Ca pump. Further, these models held up to experimental testing. [Supported by NIH-DK37512 to M.A. Milanick, NIH-GM 061583 and AHA-030161N to C. Gatto, AHA-0315236Z to J.B. Helms, and AHA-0265293Z and NIH-DK61529 to X. Zou.]

39. The Third Na⁺ Binding Site of Na,K-ATPase. CIMING LI, OIHANA CAPENDEGUY, KÄTHI GEERING, and JEAN-DANIEL HORISBERGER, Department of Pharmacology and Toxicology, University of Lausanne, CH-1005 Lausanne, Switzerland (Sponsor: J.-D. Horisberger)

Na,K-ATPase exports three intracellular sodium ions in exchange for two extracellular potassium ions. In the high resolution structure of the related sarcoplasmic and endoplasmic reticulum Ca²⁺-ATPase, two cation binding sites have been identified (Toyoshima et al. 2000. Nature. 405:647–655). The two corresponding sites in Na,K-ATPase are thought to be alternatively occupied by sodium and potassium ions, while the location of a third, sodium-specific site has been proposed on the basis of modeling and valence analysis (Ogawa and Toyoshima. 2002. Proc. Natl. Acad. Sci. USA. 99:15977–15982), but has not been demonstrated experimentally. Mutants of residues in the fifth, sixth, and ninth transmembrane segments (TMS) of the 1 subunit of the rat Na, K-ATPase were expressed in Xenopus oocytes and studied by two-electrode voltage clamp to evaluate the effects of these mutations on the affinity for intra- and extracellular Na⁺ and extracellular K⁺. The voltage-dependent translocation of Na⁺ to the extracellular medium was also studied by measurement of ouabain-sensitive presteady-state currents upon fast voltage perturbations. Mutation of E961 in TMS-9 did not alter K⁺ affinity, but reduced both intracellular and extracellular sodium binding affinity, and altered the voltage-dependent kinetics of Na⁺ translocation. Similarly, mutations of G813 and T814 from TMS-6 and of Y778 from TMS-5 altered the voltage-dependent sodium translocation. These results enabled us to define the location of a third sodium-specific binding site in a space between the fifth, sixth, eighth, and ninth transmembrane domains of the subunit of the Na, K-ATPase at about the same level as the two previously defined cation sites. [Supported by the Swiss National Fund grant 31-64793.01 to K. Geering and 31-65441.01 to J.-D. Horisberger.]

40. The Third Na⁺ Binding Pocket Is Located Near the Amino Acid Residues Thr-774 and Gln-923 in Rat Na,K-ATPase. TOSHIAKI IMAGAWA, TETSUYA YAMAMOTO, SHUNJI KAYA, KAZUYASU SAKAGUCHI, and KAZUYA TANIGUCHI, Biological Chemistry, Division of Chemistry, Graduate School of Science, Hokkaido University, Sapporo, Japan

Na,K-ATPase, a member of the P-type ATPase family, is an integral membrane protein found in all mammalian cells and forms an acid-stable phosphoenzyme (EP) during the hydrolysis of ATP. Na,K-ATPase is an electrogenic pump that transports three Na⁺ ions from the cytosol to the extracellular space and two K⁺ ions in the opposite direction. The location of the two Na⁺ binding sites were estimated from homology modeling based on the three-dimensional structure o