Homogeneous Oligomers of Pro-apoptotic BAX Reveal Structural Determinants of Mitochondrial Membrane Permeabilization.

BAX is a pro-apoptotic protein that transforms from a cytosolic monomer into a toxic oligomer that permeabilizes the mitochondrial outer membrane. How BAX monomers assemble into a higher-order conformation, and the structural determinants essential to membrane permeabilization, remain a mechanistic mystery. A key hurdle has been the inability to generate a homogeneous BAX oligomer (BAXO) for analysis. Here, we report the production and characterization of a full-length BAXO that recapitulates physiologic BAX activation. Multidisciplinary studies revealed striking conformational consequences of oligomerization and insight into the macromolecular structure of oligomeric BAX. Importantly, BAXO enabled the assignment of specific roles to particular residues and α helices that mediate individual steps of the BAX activation pathway, including unexpected functionalities of BAX α6 and α9 in driving membrane disruption. Our results provide the first glimpse of a full-length and functional BAXO, revealing structural requirements for the elusive execution phase of mitochondrial apoptosis.

Polymodal Mechanism for TWIK-Related K+ Channel Inhibition by Local Anesthetic.

BACKGROUND: Local anesthetics cause reversible block of pain and robustly inhibit TWIK-related K channel (TREK-1) currents. Before local anesthesia onset, injection of local anesthetics can cause unwanted transient pain. TREK-1 is an anesthetic-sensitive potassium channel that when inhibited produces pain. A disordered C-terminal loop of TREK-1 is thought to contribute to anesthetic sensitivity, but the molecular basis for TREK-1 inhibition by local anesthetics is unknown. Phospholipase D2 (PLD2) is an enzyme that produces phosphatidic acid (PA) required for TREK-1 activation and also binds to the channel's C terminus. METHODS: Here, we use biophysical and cellular techniques to characterize direct and indirect lipid-mediated mechanism for TREK-1 inhibition (respectively). We characterized direct binding of local anesthetic to TREK-1 by reconstituting the purified channel into artificial membranes and measuring ion flux. We characterized indirect PA-mediated inhibition of TREK-1 by monitoring lipid production in live whole cells using a fluorescent PLD2 product release assay and ion channel current using live whole-cell patch-clamp electrophysiology. We monitored anesthetic-induced nanoscale translocation of PLD2 to TREK-1 channels with super-resolution direct stochastic reconstruction microscopy (dSTORM). RESULTS: We find local anesthetics tetracaine, lidocaine, and bupivacaine directly bind to and inhibit PLD2 enzymatic activity. The lack of PLD2 activity indirectly inhibited TREK-1 currents. Select local anesthetics also partially blocked the open pore of TREK-1 through direct binding. The amount of pore block was variable with tetracaine greater than bupivacaine and lidocaine exhibiting a minor effect. Local anesthetics also disrupt lipid rafts, a mechanism that would normally activate PLD2 were it not for their direct inhibition of enzyme catalysis. CONCLUSIONS: We propose a mechanism of TREK-1 inhibition comprised of (1) primarily indirect PLD2-dependent inhibition of lipid catalysis and (2) limited direct inhibition for select local anesthetics through partial open pore block. The inhibition through PLD2 explains how the C terminus can regulate the channel despite being devoid of structure and putative binding sites for local anesthetics.

An ion selectivity filter in the extracellular domain of Cys-loop receptors reveals determinants for ion conductance.

Neurotransmitter binding to Cys-loop receptors promotes a prodigious transmembrane flux of several million ions/s, but to date, structural determinants of ion flux have been identified flanking the membrane-spanning region. Using x-ray crystallography, sequence analysis, and single-channel recording, we identified a novel determinant of ion conductance near the point of entry of permeant ions. Co-crystallization of acetylcholine-binding protein with sulfate anions revealed coordination of SO4(2-) with a ring of lysines at a position equivalent to 24 A above the lipid membrane in homologous Cys-loop receptors. Analysis of multiple sequence alignments revealed that residues equivalent to the ring of lysines are negatively charged in cation-selective receptors but are positively charged in anion-selective receptors. Charge reversal of side chains at homologous positions in the nicotinic receptor from the motor end plate decreases unitary conductance up to 80%. Selectivity filters stemming from transmembrane alpha-helices have similar pore diameters and compositions of amino acids. These findings establish that when the channel opens under a physiological electrochemical gradient, permeant ions are initially stabilized within the extracellular vestibule of Cys-loop receptors, and this stabilization is a major determinant of ion conductance.

Galanthamine and non-competitive inhibitor binding to ACh-binding protein: evidence for a binding site on non-alpha-subunit interfaces of heteromeric neuronal nicotinic receptors.

Rapid neurotransmission is mediated through a superfamily of Cys-loop receptors that includes the nicotinic acetylcholine (nAChR), gamma-aminobutyric acid (GABA(A)), serotonin (5-HT(3)) and glycine receptors. A class of ligands, including galanthamine, local anesthetics and certain toxins, interact with nAChRs non-competitively. Suggested modes of action include blockade of the ion channel, modulation from undefined extracellular sites, stabilization of desensitized states, and association with annular or boundary lipid. Alignment of mammalian Cys-loop receptors shows aromatic residues, found in the acetylcholine or ligand-binding pocket of nAChRs, are conserved in all subunit interfaces of neuronal nAChRs, including those that are not formed by alpha subunits on the principal side of the transmitter binding site. The amino-terminal domain containing the ligand recognition site is homologous to the soluble acetylcholine-binding protein (AChBP) from mollusks, an established structural and functional surrogate. We assess ligand specificity and employ X-ray crystallography with AChBP to demonstrate ligand interactions at subunit interfaces lacking vicinal cysteines (i.e. the non-alpha subunit interfaces in nAChRs). Non-competitive nicotinic ligands bind AChBP with high affinity (K(d) 0.015-6 microM). We mutated the vicinal cysteine residues in loop C of AChBP to mimic the non-alpha subunit interfaces of neuronal nAChRs and other Cys loop receptors. Classical nicotinic agonists show a 10-40-fold reduction in binding affinity, whereas binding of ligands known to be non-competitive are not affected. X-ray structures of cocaine and galanthamine bound to AChBP (1.8 A and 2.9 A resolution, respectively) reveal interactions deep within the subunit interface and the absence of a contact surface with the tip of loop C. Hence, in addition to channel blocking, non-competitive interactions with heteromeric neuronal nAChR appear to occur at the non-alpha subunit interface, a site presumed to be similar to that of modulating benzodiazepines on GABA(A) receptors.

Structural comparison of three crystalline complexes of a peptidic toxin with a synaptic acetylcholine recognition protein.

Many peptidic toxins from animal venoms target neuronal or peripheral synaptic receptors with high affinities and specificities. Hence, these toxins are not only potent natural weapons but also precise molecular tools for pharmacological studies of their receptors. Although they belong to various structural and/or functional subfamilies, they often share similar molecular features, such as a highly reticulated scaffold presenting specific binding determinants.

Solution NMR of acetylcholine binding protein reveals agonist-mediated conformational change of the C-loop.

Previous X-ray crystallography, molecular dynamics simulation, fluorescence spectroscopy, and deuterium-hydrogen exchange of acetylcholine binding protein (AChBP) suggest that after binding of the agonist, the C-loop at the periphery of the binding site draws inward to cap the site and envelop the agonist. In this study, we use high-resolution solution NMR to monitor changes in the chemical environment of the C-loop without and with acetylcholine (ACh) bound. Substitution of [15N]cysteine for the native cysteines 123, 136, 187, and 188 provided intrinsic monitors of the chemical environments of the Cys- and C-loops, respectively. Two-dimensional transverse relaxation-optimized spectroscopy 15N-1H HSQC spectroscopy of apo-AChBP revealed seven well resolved cross-peaks for the group of cysteines. The spectrum of AChBP with Ser substituted for Cys 187 and 188 shows only two main cross-peaks, corresponding to Cys 123 and 136 from the Cys-loop, enabling resonance assignments. After binding of ACh, the five cross-peaks associated with cysteines from the C-loop condense into two predominant cross-peaks not observed in the spectrum from the apo protein, indicating a restricted range of conformations and change in chemical environment of the C-loop. The results show that isotopic cysteine can be incorporated into specified positions of AChBP expressed from a eukaryotic source, that the C-loop assumes multiple conformations without ACh, but that its conformation becomes restricted with ACh bound. The collective findings suggest a structural mechanism for agonist recognition in AChBP and related Cys-loop receptors.

Structural dynamics of the alpha-neurotoxin-acetylcholine-binding protein complex: hydrodynamic and fluorescence anisotropy decay analyses.

The three-fingered alpha-neurotoxins have played a pivotal role in elucidating the structure and function of the muscle-type and neuronal alpha7 nicotinic acetylcholine receptors (nAChRs). To advance our understanding of the alpha-neurotoxin-nAChR interaction, we examined the flexibility of alpha-neurotoxin bound to the acetylcholine-binding protein (AChBP), which shares structural similarity and sequence identities with the extracellular domain of nAChRs. Because the crystal structure of five alpha-cobratoxin molecules bound to AChBP shows the toxins projecting radially like propeller "blades" from the perimeter of the donut-shaped AChBP, the toxin molecules should increase the frictional resistance and thereby alter the hydrodynamic properties of the complex. alpha-Bungarotoxin binding had little effect on the frictional coefficients of AChBP measured by analytical ultracentrifugation, suggesting that the bound toxins are flexible. To support this conclusion, we measured the anisotropy decay of four site-specifically labeled alpha-cobratoxins (conjugated at positions Lys(23), Lys(35), Lys(49), and Lys(69)) bound to AChBP and free in solution and compared their anisotropy decay properties with fluorescently labeled cysteine mutants of AChBP. The results indicated that the core of the toxin molecule is relatively flexible when bound to AChBP. When hydrodynamic and anisotropy decay analyses are taken together, they establish that only one face of the second loop of the alpha-neurotoxin is immobilized significantly by its binding. The results indicate that bound alpha-neurotoxin is not rigidly oriented on the surface of AChBP but rather exhibits segmental motion by virtue of flexibility in its fingerlike structure.