Structural interactions of a voltage sensor toxin with lipid membranes.

Protein toxins from tarantula venom alter the activity of diverse ion channel proteins, including voltage, stretch, and ligand-activated cation channels. Although tarantula toxins have been shown to partition into membranes, and the membrane is thought to play an important role in their activity, the structural interactions between these toxins and lipid membranes are poorly understood. Here, we use solid-state NMR and neutron diffraction to investigate the interactions between a voltage sensor toxin (VSTx1) and lipid membranes, with the goal of localizing the toxin in the membrane and determining its influence on membrane structure. Our results demonstrate that VSTx1 localizes to the headgroup region of lipid membranes and produces a thinning of the bilayer. The toxin orients such that many basic residues are in the aqueous phase, all three Trp residues adopt interfacial positions, and several hydrophobic residues are within the membrane interior. One remarkable feature of this preferred orientation is that the surface of the toxin that mediates binding to voltage sensors is ideally positioned within the lipid bilayer to favor complex formation between the toxin and the voltage sensor.

Structure and hydration of membranes embedded with voltage-sensing domains.

Despite the growing number of atomic-resolution membrane protein structures, direct structural information about proteins in their native membrane environment is scarce. This problem is particularly relevant in the case of the highly charged S1-S4 voltage-sensing domains responsible for nerve impulses, where interactions with the lipid bilayer are critical for the function of voltage-activated ion channels. Here we use neutron diffraction, solid-state nuclear magnetic resonance (NMR) spectroscopy and molecular dynamics simulations to investigate the structure and hydration of bilayer membranes containing S1-S4 voltage-sensing domains. Our results show that voltage sensors adopt transmembrane orientations and cause a modest reshaping of the surrounding lipid bilayer, and that water molecules intimately interact with the protein within the membrane. These structural findings indicate that voltage sensors have evolved to interact with the lipid membrane while keeping energetic and structural perturbations to a minimum, and that water penetrates the membrane, to hydrate charged residues and shape the transmembrane electric field.

Expression and purification of CB2 for NMR studies in micellar solution.

We demonstrate feasibility of biophysical characterization of the peripheral cannabinoid receptor CB2 produced by heterologous expression in E. coli membranes. Recombinant receptor was purified by affinity chromatography, and NMR diffusion experiments performed on CB2 solubilized in dodecylphosphocholine (DPC) micelles. Circular dichroism spectroscopy indicated high alpha-helical content (49 %) of CB2.

Engineering vanilloid-sensitivity into the rat TRPV2 channel.

The TRPV1 channel is a detector of noxious stimuli, including heat, acidosis, vanilloid compounds and lipids. The gating mechanisms of the related TRPV2 channel are poorly understood because selective high affinity ligands are not available, and the threshold for heat activation is extremely high (>50°C). Cryo-EM structures of TRPV1 and TRPV2 reveal that they adopt similar structures, and identify a putative vanilloid binding pocket near the internal side of TRPV1. Here we use biochemical and electrophysiological approaches to investigate the resiniferatoxin(RTx) binding site in TRPV1 and to explore the functional relationships between TRPV1 and TRPV2. Collectively, our results support the interaction of vanilloids with the proposed RTx binding pocket, and demonstrate an allosteric influence of a tarantula toxin on vanilloid binding. Moreover, we show that sensitivity to RTx can be engineered into TRPV2, demonstrating that the gating and permeation properties of this channel are similar to TRPV1.

Structural insights into the mechanism of activation of the TRPV1 channel by a membrane-bound tarantula toxin.

Venom toxins are invaluable tools for exploring the structure and mechanisms of ion channels. Here, we solve the structure of double-knot toxin (DkTx), a tarantula toxin that activates the heat-activated TRPV1 channel. We also provide improved structures of TRPV1 with and without the toxin bound, and investigate the interactions of DkTx with the channel and membranes. We find that DkTx binds to the outer edge of the external pore of TRPV1 in a counterclockwise configuration, using a limited protein-protein interface and inserting hydrophobic residues into the bilayer. We also show that DkTx partitions naturally into membranes, with the two lobes exhibiting opposing energetics for membrane partitioning and channel activation. Finally, we find that the toxin disrupts a cluster of hydrophobic residues behind the selectivity filter that are critical for channel activation. Collectively, our findings reveal a novel mode of toxin-channel recognition that has important implications for the mechanism of thermosensation.

Tarantula toxins use common surfaces for interacting with Kv and ASIC ion channels.

Tarantula toxins that bind to voltage-sensing domains of voltage-activated ion channels are thought to partition into the membrane and bind to the channel within the bilayer. While no structures of a voltage-sensor toxin bound to a channel have been solved, a structural homolog, psalmotoxin (PcTx1), was recently crystalized in complex with the extracellular domain of an acid sensing ion channel (ASIC). In the present study we use spectroscopic, biophysical and computational approaches to compare membrane interaction properties and channel binding surfaces of PcTx1 with the voltage-sensor toxin guangxitoxin (GxTx-1E). Our results show that both types of tarantula toxins interact with membranes, but that voltage-sensor toxins partition deeper into the bilayer. In addition, our results suggest that tarantula toxins have evolved a similar concave surface for clamping onto α-helices that is effective in aqueous or lipidic physical environments.

Identification of active site residues in mevalonate diphosphate decarboxylase: implications for a family of phosphotransferases.

A combination of sequence homology analyses of mevalonate diphosphate decarboxylase (MDD) proteins and structural information for MDD leads to the hypothesis that Asp 302 and Lys 18 are active site residues in MDD. These residues were mutated to replace acidic/basic side chains and the mutant proteins were isolated and characterized. Binding and competitive displacement studies using trinitrophenyl-ATP, a fluorescent analog of substrate ATP, indicate that these mutant enzymes (D302A, D302N, K18M) retain the ability to stoichiometrically bind nucleotide triphosphates at the active site. These observations suggest the structural integrity of the mutant MDD proteins. The functional importance of mutated residues was evaluated by kinetic analysis. The 10(3) and 10(5)-fold decreases in k(cat) observed for the Asp 302 mutants (D302N and D302A, respectively) support assignment of a crucial catalytic role to Asp 302. A 30-fold decrease in activity and a 16-fold inflation of the K(m) for ATP is documented for the K18M mutant, indicating that Lys 18 influences the active site but is not crucial for reaction chemistry. Demonstration of the influence of conserved aspartate 302 appears to represent the first documentation of the functional importance of a residue in the MDD catalytic site and affords insight into phosphotransferase reactions catalyzed by a variety of enzymes in the galactokinase, homoserine kinase, mevalonate kinase, phosphom-evalonate kinase (GHMP kinase) family.

Zn-, Cd-, and Pb-transcription factor IIIA: properties, DNA binding, and comparison with TFIIIA-finger 3 metal complexes.

Properties of the metal ion binding sites of Zn-transcription factor IIIA (TFIIIA) were investigated to understand the potential of this type of zinc finger to undergo reactions that remove Zn(2+) from the protein. Zn-TFIIIA was purified from E. coli containing the cloned sequence for Xenopus laevis oocyte TFIIIA and its stoichiometry of bound Zn(2+) was shown to depend on the details of the isolation process. The average dissociation constant of Zn(2+) in Zn-TFIIIIA was 10(-7). The dissociation constant for Zn-F3, the third finger from the N-terminus of TFIIIA, was 1.0 x 10(-8). The reactivity of Zn-TFIIIA with a series of metal binding ligands, including 2-carboxy-2'-hydroxy-5'-sulfoformazylbenzene (zincon), 4-(2-pyridylazo)-resorcinol (PAR), and 3-ethoxy-2-oxo-butyraldehyde-bis-(N(4)-dimethylthiosemicarbazone) (H(2)KTSM(2)) revealed similar kinetics. The reactivity of PAR with Zn-TFIIIA declined substantially when the protein was bound to the internal control region (ICR) of the 5S ribosomal DNA. Both Cd(2+) and Pb(2+) disrupt TFIIIA binding to its cognate DNA sequence. The Pb(2+) dissociation constant of Pb-F3 was measured as 2.5 x 10(-8). According to NMR spectroscopy, F3 does not fold into a regular conformation in the presence of Pb(2+).

Structural interactions between lipids, water and S1-S4 voltage-sensing domains.

Membrane proteins serve crucial signaling and transport functions, yet relatively little is known about their structures in membrane environments or how lipids interact with these proteins. For voltage-activated ion channels, X-ray structures suggest that the mobile voltage-sensing S4 helix would be exposed to the membrane, and functional studies reveal that lipid modification can profoundly alter channel activity. Here, we use solid-state NMR to investigate structural interactions of lipids and water with S1-S4 voltage-sensing domains and to explore whether lipids influence the structure of the protein. Our results demonstrate that S1-S4 domains exhibit extensive interactions with lipids and that these domains are heavily hydrated when embedded in a membrane. We also find evidence for preferential interactions of anionic lipids with S1-S4 domains and that these interactions have lifetimes on the timescale of ≤ 10(-3)s. Arg residues within S1-S4 domains are well hydrated and are positioned in close proximity to lipids, exhibiting local interactions with both lipid headgroups and acyl chains. Comparative studies with a positively charged lipid lacking a phosphodiester group reveal that this lipid modification has only modest effects on the structure and hydration of S1-S4 domains. Taken together, our results demonstrate that Arg residues in S1-S4 voltage-sensing domains reside in close proximity to the hydrophobic interior of the membrane yet are well hydrated, a requirement for carrying charge and driving protein motions in response to changes in membrane voltage.

Bacterial expression of functional, biotinylated peripheral cannabinoid receptor CB2.

A biotin-protein ligase recognition site (BRS) was inserted into a polypeptide comprised of the maltose-binding protein, the peripheral cannabinoid receptor (CB2), thioredoxin A, and a polyhistidine tag at the carboxy terminus. Expression levels of the recombinant receptor in Escherichia coli BL21(DE3) cells were approximately 1mg per liter of bacterial culture. The biotinylated CB2-fusion fully retained its ligand-binding capacity. Introduction of the BRS at the C-terminus of the CB2 fusion protein (construct CB2-109) resulted in its complete in vivo biotinylation; the biotinylated protein was streptavidin-binding competent. Positioning of the BRS near the N-terminus of CB2 (CB2-112) resulted in a very low level of biotinylation in vivo. However, the detergent solubilized and purified CB2-112 fusion protein were successfully biotinylated in vitro by action of a BirA biotin-protein ligase. The biotinylated CB2-112 fusion protein was cleaved by the tobacco etch virus protease at specifically inserted sites, and deposited onto monomeric avidin agarose beads. Biotinylation of the recombinant CB2 receptor enabled not only purification but also immobilization of the GPCR on a solid support in homogeneous orientation which is beneficial for subsequent structural characterization.

Investigation of the functional contributions of invariant serine residues in yeast mevalonate diphosphate decarboxylase.

Alignment of more than 20 deduced sequences for mevalonate diphosphate decarboxylase (MDD) indicates that serines 34, 36, 120,121, 153, and 155 are invariant residues that map within a proposed interdomain active site cleft. To test possible active site roles for these invariant serines, each has been mutated to alanine. S34A exhibits limited solubility and impaired binding of the fluorescent ATP analogue, trinitrophenyl-ATP (TNP-ATP), suggesting that Ser-34 substitution destabilizes proper enzyme folding. All other serine mutants retain structural integrity, as indicated by their ability to bind TNP-ATP at levels comparable to wild-type enzyme. S153A exhibits a 18-fold inflation in K(d) for Mg-ATP, as indicated by competitive displacement of TNP-ATP; the enzyme also is characterized by a 35-fold inflation in K(m) for Mg-ATP. S155A exhibits a 26-fold inflation in K(m) for Mg-ATP, but competitive displacement of TNP-ATP indicates only a 2-fold inflation in K(d) for this substrate. S155A exhibits both a 16-fold inflation in K(m) for mevalonate diphosphate and a 14-fold inflation in K(i(slope)) for the substrate analogue, diphosphoglycolylproline. These observations suggest roles for Ser-153 and Ser-155 in substrate binding. Catalytic consequences of mutating invariant serines 36, 120, 153, and 155 are modest (<8-fold diminution in k(cat)). In contrast, S121A, which exhibits only modest changes in K(d) for Mg-ATP and K(m) for mevalonate diphosphate, is characterized by a >42,000-fold diminution in k(cat), indicating the critical involvement of Ser-121 in reaction catalysis. The selective involvement of the latter of two tandem serine residues (Ser-120, Ser-121) in a conserved sequence motif suggests mechanistic similarities within the GHMP kinase superfamily of proteins.

Properties of the reaction of chromate with metallothionein.

The reactivity of chromate or Cr(VI) with rabbit liver metallothionein (MT) was explored in this study. Zn(7)-MT reacts very slowly with Cr(VI) in a process characterized by a second-order rate constant of 3.9 x 10(-)(4) M(-)(1) s(-)(1). During the reaction, Zn(2+) was released from the protein. In contrast, apo-MT reduces chromate quicker and in this reaction is much more effective as a reducing agent, when compared to Cys or GSH. The kinetics are consistent with a reaction pathway involving an initial binding step followed by the reduction of Cr(VI). In the process, MT sulfhydryl groups were oxidized at the same rate that Cr(VI) disappeared. A Cr(V) intermediate was detected by EPR spectroscopy immediately upon mixing apo-MT with Cr(VI). The Cr(V) signal decayed during the reaction but was quite stable and could be observed for hours once the supply of thiols was depleted. The g values for the Cr(V) species were 2.014 and 1.987. The kinetics of the reaction of Cr(VI) and the concentration of the intermediate Cr(V) signal were independent of the oxygen concentration and were unaffected by the presence of superoxide dismutase, catalase, or DMSO. In the presence of oxygen, oxy radicals were generated according to ESR spin-trapping experiments with 5,5'-dimethyl-1-pyrroline-N-oxide. Superoxide dismutase decreased and catalase or DMSO largely inhibited the formation of the spin-trapped adduct. Cr(III), the presumed final species of the Cr(VI) reduction, formed a stable complex with apo-MT in the absence of oxygen with an average stoichiometry of two Cr ions bound per protein molecule. Upon addition of O(2), the complex slowly dissociated.

Interaction of Cd2+ with Zn finger 3 of transcription factor IIIA: structures and binding to cognate DNA.

Finger 3 of transcription factor IIIA of Xenopus laevis was synthesized and constituted with Zn(2+) or Cd(2+). The C-block element of the internal control region of the promoter of the 5S rRNA gene binds to the Zn-F3 and Cd-F3 with dissociation constants of 2.6 x 10(-5) and 1.5 x 10(-4) M, respectively. According to NMR spectroscopy, Zn-F3, as well as Cd-F3, exists as a conformational equilibrium that is not susceptible to structural analysis by NMR methods. To restrict the observed conformational flexibility, a mutant F3 (mF3), which differs from F3 in the number and type of amino acids between the cysteine and the histidine ligands, was synthesized. The affinity of Zn-mF3 for the C-block DNA was greatly reduced relative to Zn-F3. Nevertheless, the metal ion dissociation constants of the Zn- and Cd-mF3 complexes remain similar to those of the native structures at 4.5 x 10(-9) and 3.2 x 10(-8) M, respectively. Zn-mF3 is more thermally stable than Cd-mF3, but both adopt similar conformations according to two-dimensional (1)H NMR spectroscopy. Each peptide displays a betabetaalpha fold for its backbone that is typical of this class of zinc finger domains. The(113)Cd ion in (113)Cd-mF3 is coupled to the protons of two cysteine and two histidine residues and characterized by a chemical shift of 567 ppm.

Cobalt-substituted zinc finger 3 of transcription factor IIIA: interactions with cognate DNA detected by (31)P ENDOR spectroscopy.

We show the first ENDOR study of the coordination environment of high-spin Co(II) in a biological system with a study of DNA binding to the Co-substituted Cys2/His2 single Zn-finger domain, Finger 3 (F3), from the prototypical zinc finger protein, transcription factor IIIA (TFIIIA) from Xenopus laevis. High covalency to cysteine and histidine is implied by ENDOR-derived 1H couplings to protons of cysteinyl ligands and 14N couplings to histidyl nitrogens, results which support the expectation that Zn(II) and Co(II) bind to F3 in a very similar manner. No changes in either 1H or 14N ENDOR were detected upon binding Co(II)-F3 to C-block DNA. Of particular importance to the use of Co(II) substitution for Zn(II), the ENDOR method shows that Co(II)-F3 undergoes sequence-specific binding to the cognate DNA for Zn(II)-F3, the internal control region (ICR) of the 5S rRNA (C-block). 31P ENDOR measurements yield a Co-31P distance of rCo-P = 8.1(3) A to the nearest backbone phosphodiester of the C-block. Interestingly, a 31P ENDOR doublet observed for Co(II)-F3 in phosphate buffer indicates that inorganic phosphate (Pi) binds at a comparable distance from Co as does the nearest phosphate of DNA, presumably at the same site.