Distinct binding properties distinguish LQ-type calmodulin-binding domains in cyclic nucleotide-gated channels

Cyclic nucleotide-gated (CNG) channels operate as transduction channels in photoreceptors and olfactory receptor neurons. Direct binding of cGMP or cAMP opens these channels which conduct a mixture of monovalent cations and Ca(2+). Upon activation, CNG channels generate intracellular Ca(2+) signals that play pivotal roles in the transduction cascades of the visual and olfactory systems. Channel activity is controlled by negative feedback mechanisms that involve Ca(2+)-calmodulin, for which all CNG channels possess binding sites. Here we compare the binding properties of the two LQ-type calmodulin binding sites, both of which are thought to be involved in channel regulation. They reside on the isoforms CNGB1 and CNGA4. The CNGB1 subunit is present in rod photoreceptors and olfactory receptor neurons. The CNGA4 subunit is only expressed in olfactory receptor neurons, and there are conflicting results as to its role in calmodulin-mediated feedback inhibition. We examined the interaction of Ca(2+)-calmodulin with two recombinant proteins that encompass either of the two LQ sites. Comparing binding properties, we found that the LQ site of CNGB1 binds Ca(2+)-calmodulin at 10-fold lower Ca(2+) levels than the LQ site of CNGA4. Our data provide biochemical evidence against a contribution of CNGA4 to feedback inhibition. In accordance with previous work on photoreceptor CNG channels, our results indicate that feedback control is the exclusive role of the B-subunits in photoreceptors and olfactory receptor neurons.     

Ca2+ stabilizes the membrane potential of moth olfactory receptor neurons at rest and is essential for their fast repolarization

The role of Ca(2+) in insect olfactory transduction was studied in the moth Spodoptera littoralis. Single sensillum recordings were made to investigate in vivo the role of sensillar Ca(2+) on the electrophysiological properties of sex pheromone responsive olfactory receptor neurons (ORNs). Lowering the sensillar Ca(2+) concentration to 2 x 10(-8) M increased ORN spontaneous firing activity and induced long bursts of action potentials (APs) superimposed on spontaneous negative deflections of the transepithelial potential. We inferred that Ca(2+) stabilizes the membrane potential of ORNs, keeping the spontaneous firing activity at a low and regular level. Neither the amplitude and kinetics of the rising phase of sensillar potentials (SPs) recorded in response to pheromone stimuli nor the AP generation during stimulation depended on the extracellular Ca(2+) concentration. Thus, extracellular Ca(2+) is not absolutely necessary for ORN response. Partial inhibition of responses with a calmodulin antagonist, W-7, also indicates that intracellular Ca(2+) contributes to the ORN response and suggests that Ca(2+) release from internal stores is involved. In 2 x 10(-8) M Ca(2+), the repolarization of the SP was delayed when compared with higher Ca(2+) concentrations. Therefore, in contrast to depolarization, ORN repolarization depends on extracellular Ca(2+). Ca(2+)-gated K(+) channels identified from cultured ORNs with whole-cell recordings are good candidates to mediate ORN repolarization.     

Cellular and molecular constituents of olfactory sensation in vertebrates

Since the discovery of odorant-activated adenylate cyclase in the olfactory receptor cilia, research into the olfactory perception of vertebrates has rapidly expanded. Recent studies have shown how the odor discrimination starts at the receptor level: each of 700-1000 types of the olfactory neurons in the neural olfactory epithelium contains a single type of odor receptor protein. Although the receptors have relatively low specific affinities for odorants, excitation of different types of receptors forms an excitation pattern specific to each odorant in the glomerular layer of the olfactory bulb. It was demonstrated that adenosine 3',5'-cyclic monophosphate (cAMP) is very likely the sole second messenger for olfactory transduction. It was also demonstrated that the affinity of the cyclic nucleotide-gated channel for cAMP regulated by Ca(2+)/calmodulin is solely responsible for the adaptation of the cell. However, many other regulatory components were found in the transduction cascade. Regulated by Ca(2+) and/or the protein-phosphorylation, many of them may serve for the adaptation of the cell, probably on a longer time scale. It may be important to consider the resensitization as a part of this adaptation, as well as to collect kinetic data of each reaction to gain further insight into the olfactory mechanism.     

Calcium signalling in squid olfactory receptor neurons

Isolated squid olfactory receptor neurons respond to dopamine and betaine with hyperpolarizing conductances. We used Ca(2+) imaging techniques to determine if changes in intracellular Ca(2+) were involved in transducing the hyperpolarizing odor responses. We found that dopamine activated release of Ca(2+) from intracellular stores while betaine did not change internal Ca(2+) concentrations. Application of 10 mM caffeine also released Ca(2+) from intracellular stores, suggesting the presence of ryanodine-like receptors. Depletion of intracellular stores with 100 microM thapsigargin revealed the presence of a Ca(2+) store depletion-activated Ca(2+) influx. The influx of Ca(2+) through the store-operated channel was reversibly blocked by 10 mM Cd(2+). Taken together, these data suggest a novel odor transduction system in squid olfactory receptor neurons involving Ca(2+) release from intracellular stores. Copyright Copyright 1999 S. Karger AG, Basel     

Umami changes intracellular Ca2+ levels using intracellular and extracellular sources in mouse taste receptor cells

Recently, candidates for umami receptors have been identified in taste cells, but the precise transduction mechanisms of the downstream receptor remain unknown. To investigate how intracellular Ca(2+) increases in the umami transduction pathway, we measured changes in intracellular Ca(2+) levels in response to umami stimuli monosodium glutamate (MSG), IMP, and MSG + IMP in mouse taste receptor cells (TRCs) by Ca(2+) imaging. Even when extracellular Ca(2+) was absent, 1/3 of umami-responsive TRCs exhibited increased intracellular Ca(2+) levels. When intracellular Ca(2+) was depleted, half of the TRCs retained their response to umami. These results suggest that umami-responsive TRCs increase their intracellular Ca(2+) levels through two pathways: by releasing Ca(2+) from intracellular stores and by an influx of Ca(2+) from extracellular sources. We conclude that the Ca(2+) influx from extracellular source might play an important role in the synergistic effect between MSG and IMP.     

Kinetics of Ca2+, NADH, and oxidized flavoproteins in the frog olfactory lining under the effect of odorants

The kinetics of fluorescence of Ca(2+) - chlortertacyclin-cell membrane complex as well as of NADH and oxidized flavoproteins in receptor cells of the frog olfactory lining under the effect of odorants has been studied. Changes in the fluorescence of the olfactory lining upon stimulation by cineole and vanillin occurred more rapidly than under the effect of camphor and amyl alcohol. Differences in the kinetics of reactions of NADH and the Ca(2+)-CTC-CM complex to different odorants are apparently due to heterogeneity of molecular mechanisms associated with the involvement of different intracellular signal systems in the transduction of these odorants in the olfactory lining. In contrast to them, ammonia and beta3-mercaptoethanol penetrate into olfactory cells and inhibit the mitochondrial respiratory chain without the participation of second messengers. At the same time, the motor activity of olfactory cilia is depressed.     

A Model of the Intracellular Response of an Olfactory Neuron in Caenorhabditis elegans to Odor Stimulation

We developed a mathematical model of a hypothetical neuronal signal transduction pathway to better understand olfactory perception in Caenorhabditis elegans. This worm has only three pairs of olfactory receptor neurons. Intracellular Ca(2+) decreases in one pair of olfactory neurons in C. elegans, the AWC neurons, following application of an attractive odor and there is a transient increase in intracellular Ca(2+) following removal of odor. The magnitude of this increase is positively correlated with the duration of odor stimulation. Additionally, this Ca(2+) transient is induced by a cGMP second messenger system. We identified likely candidates for the signal transduction molecules functioning in this system based on available gene expression and physiological data from AWCs. Our model incorporated a G-protein-coupled odor receptor, a G-protein, a guanylate cyclase as the G-protein effector, and a single phosphodiesterase. Additionally, a cyclic-nucleotide-gated ion channel and a voltage-gated ion channel that mediated calcium influx were incorporated into the model. We posited that, upon odor stimulation, guanylate cyclase was suppressed by the G-protein and that, upon cessation of the stimulus, the G-protein-induced suppression ceased and cGMP synthesis was restored. A key element of our model was a Ca(2+)-dependent negative feedback loop that ensured that the calcium increases were transient. Two guanylate cyclase-activating proteins acted on the effector guanylate cyclase to negatively regulate cGMP signaling and the resulting calcium influx. Our model was able to closely replicate in silico three important features of the calcium dynamics of AWCs. Specifically, in our simulations, [Ca(2+)] increased rapidly and reached its peak within 10 s after the odor stimulus was removed, peak [Ca(2+)] increased with longer odor exposure, and [Ca(2+)] decreased during a second stimulus that closely followed an initial stimulus. However, application of random background signal ('noise') showed that certain components of the pathway were particularly sensitive to this noise.     

Characterization of p26olf, a novel calcium-binding protein in the frog olfactory epithelium

We have previously shown that p26olf is a novel S100-like Ca(2+)-binding protein in the frog olfactory epithelium. In this paper, we characterized the Ca(2+) binding property of p26olf, examined the precise localization in the frog olfactory epithelium, and searched for the possible target proteins of p26olf. By flow dialysis experiments using (45)Ca, p26olf was suggested to bind approximately 4 Ca(2+). Circular dichroism measurement showed that binding of Ca(2+) to p26olf induces an increase in the apparent content of both alpha-helix and beta-sheet with an apparent K(d) value of 2.4 micrometer. Electron microscopic observation disclosed p26olf immunoreactivity in the cilia, dendritic knob, and dendrite of olfactory receptor cells. Blot overlay analysis and affinity purification of p26olf-binding proteins showed that p26olf binds to a frog beta-adrenergic receptor kinase-like protein in a Ca(2+)-dependent manner. These results suggested that p26olf has some roles in the olfactory transduction or adaptation.     

Oscillatory current responses of olfactory receptor neurons to odorants and computer simulation based on a cyclic AMP transduction model

Neural oscillatory activities triggered by odorant stimulation have been often reported at various levels of olfactory nervous systems in vertebrates. To elucidate the origin of neural oscillations, we studied first the oscillatory properties of current responses of isolated olfactory receptor neurons (ORNs) of the rainbow trout to amino acid odorants, using a whole-cell voltage-clamp technique and found that the damped current oscillations were intrinsic in both ciliated and microvillous ORNs and occurred when ORNs were stimulated by odorants at high intensities. Continuous wavelet analysis using the Gabor function revealed that the dominant frequency of oscillations was 1.89 +/- 0.50 Hz (mean +/- SD, n = 92). There was no significant difference in oscillation frequency between the two types of ORNs and between different perfusion conditions with standard and Na(+)-free (choline) Ringer's solutions, but there was a slight difference in oscillation frequency between different holding potential conditions of negative and positive potentials. We then performed a computer simulation of the current responses with a cAMP olfactory transduction model. The model was based on the assumption that the current responses of ORNs were linearly related to the sum of concentrations of active cyclic-nucleotide-gated channels and Ca(2+)-activated Cl(-) channels, and was expressed by 12 differential equations with 44 different parameters. The simulation revealed that the oscillations of current responses of ORNs were mainly due to the oscillatory properties of intracellular cAMP and Ca(2+) concentrations. The necessary reaction component for the oscillations in the transduction model was direct inhibition of adenylate cyclase activity by Ca(2+). High Ca(2+) efflux by the Na(+)-Ca(2+) exchanger and cAMP-phosphodiesterase activity were most influential on the oscillations. The simulation completely represented the characteristics of current responses of ORNs: odorant-intensity-dependent response, intensity-dependent latency and adaptation. Thus, the simulation is generally applicable to current and voltage responses of ORNs equipped with cAMP olfactory transduction pathway in other vertebrate species. The simulation programs for Macintosh (cAMP 9.2.7 and 9.2.8 for MacOS 8.1 or later) and cAMP JAVA applet versions based on cAMP 9.2.8 have been published on the world wide web (http://bio2.sci.hokudai.ac.jp/bio/chinou1/noriyo_home.html).     

A novel calcium-regulated membrane guanylate cyclase transduction system in the olfactory neuroepithelium

This report defines the identity of a calcium-regulated membrane guanylate cyclase transduction system in the cilia of olfactory sensory neurons, which is the site of odorant transduction. The membrane fraction of the neuroepithelial layer of the rat exhibited Ca(2+)-dependent guanylate cyclase activity, which was eliminated by the addition of EGTA. This indicated that the cyclase did not represent a rod outer segment guanylate cyclase (ROS-GC), which is inhibited by free Ca(2+). This interpretation was supported by studies with the Ca(2+) binding proteins, GCAPs (guanylate cyclase activating proteins), which stimulate photoreceptor ROS-GC in the absence of Ca(2+). They did not stimulate the olfactory neuroepithelial membrane guanylate cyclase. The olfactory neuroepithelium contained a Ca(2+) binding protein, neurocalcin, which stimulated the cyclase in a Ca(2+)-dependent fashion. The cyclase was cloned from the neuroepithelium and was found to be identical in structure to that of the previously cloned cyclase termed GC-D. The cyclase was expressed in a heterologous cell system, and was reconstituted with its Ca(2+)-dependent activity in the presence of recombinant neurocalcin. The reconstituted cyclase mimicked the native enzyme. Immunocytochemical studies showed that the guanylate cyclase coexists with neurocalcin in the apical region of the cilia. Deletion analysis showed that the neurocalcin-regulated domain resides at the C-terminal region of the cyclase. The findings establish the biochemical, molecular, and functional identity of a novel Ca(2+)-dependent membrane guanylate cyclase transduction system in the cilia of the olfactory epithelium, suggesting a mechanism of the olfactory neuroepithelial guanylate cyclase regulation fundamentally distinct from the phototransduction-linked ROS-GC.     

Glutamate-mediated extrasynaptic inhibition: direct coupling of NMDA receptors to Ca(2+)-activated K+ channels

NMDA receptors (NMDARs) typically contribute to excitatory synaptic transmission in the CNS. While Ca(2+) influx through NMDARs plays a critical role in synaptic plasticity, direct actions of NMDAR-mediated Ca(2+) influx on neuronal excitability have not been well established. Here we show that Ca(2+) influx through NMDARs is directly coupled to activation of BK-type Ca(2+)-activated K+ channels in outside-out membrane patches from rat olfactory bulb granule cells. Repetitive stimulation of glutamatergic synapses in olfactory bulb slices evokes a slow inhibitory postsynaptic current (IPSC) in granule cells that requires both NMDARs and BK channels. The slow IPSC is enhanced by glutamate uptake blockers, suggesting that extrasynaptic NMDARs underlie the response. These findings reveal a novel inhibitory action of extrasynaptic NMDARs in the brain.     

Inward currents and increases in cytosolic Ca2+ concentration induced by cyclic ADP-ribose in turtle olfactory receptor cells

In olfactory receptor cells, it is well established that cyclic AMP (cAMP) and inositol-1,4,5-trisphosphate (IP(3)) act as second messengers during odor responses. In previous studies, we have shown that cAMP-increasing odorants induce odor responses even after complete desensitization of the cAMP-mediated pathway. These results suggest that at least one cAMP-independent pathway contributes to the generation of odor responses. In an attempt to identify a novel second messenger, we investigated the possible role of cyclic ADP-ribose (cADPR) in olfactory transduction. Turtle olfactory receptor cells were isolated using an enzyme-free procedure and loaded with fura-2/AM. The cells responded to dialysis with cADPR with an inward current and an increase of the intracellular Ca(2+) concentration, [Ca(2+)](i). Flooding of cells with 100 microM cADPR from the pipette also induced an inward current without changes in [Ca(2+)](i) in Na(+)-containing and Ca(2+)-free Ringer solution. In an Na(+)-free and Ca(2+)-containing Ringer solution, cADPR induced only a small inward current with a concomitant increase in [Ca(2+)](i). Inward currents and increases in [Ca(2+)](i) induced by cADPR were completely inhibited by removal of both Na(+) and Ca(2+) from the outer solution. The experiments suggest that cADPR activates a cation channel at the plasma membrane, allowing inflow of Na(+) and Ca(2+) ions. The magnitudes of the inward current responses to cAMP-increasing odorants were greatly reduced by prior dialyses of a high concentration of cADPR or 8-bromo-cyclic ADP-ribose (8-Br-cADPR), an antagonist. It is possible that the cADPR-dependent pathway contributes to the generation of olfactory responses.     

The electrochemical basis of odor transduction in vertebrate olfactory cilia

Most vertebrate olfactory receptor neurons share a common G-protein-coupled pathway for transducing the binding of odorant into depolarization. The depolarization involves 2 currents: an influx of cations (including Ca2+) through cyclic nucleotide-gated channels and a secondary efflux of Cl- through Ca2+-gated Cl- channels. The relation between stimulus strength and receptor current shows positive cooperativity that is attributed to the channel properties. This cooperativity amplifies the responses to sufficiently strong stimuli but reduces sensitivity and dynamic range. The odor response is transient, and prolonged or repeated stimulation causes adaptation and desensitization. At least 10 mechanisms may contribute to termination of the response; several of these result from an increase in intraciliary Ca2+. It is not known to what extent regulation of ionic concentrations in the cilium depends on the dendrite and soma. Although many of the major mechanisms have been identified, odor transduction is not well understood at a quantitative level.     

A diacylglycerol-gated cation channel in vomeronasal neuron dendrites is impaired in TRPC2 mutant mice: mechanism of pheromone transduction

Vomeronasal sensory neurons play a crucial role in detecting pheromones, but the chemoelectrical transduction mechanism remains unclear and controversial. A major barrier to the resolution of this question has been the lack of an activation mechanism of a key transduction component, the TRPC2 channel. We have identified a Ca(2+)-permeable cation channel in vomeronasal neuron dendrites that is gated by the lipid messenger diacylglycerol (DAG), independently of Ca(2+) or protein kinase C. We demonstrate that ablation of the TRPC2 gene causes a severe deficit in the DAG-gated channel, indicating that TRPC2 encodes a principal subunit of this channel and that the primary electrical response to pheromones depends on DAG but not Ins(1,4,5)P(3), Ca(2+) stores, or arachidonic acid. Thus, a previously unanticipated mechanism involving direct channel opening by DAG underlies the transduction of sensory cues in the accessory olfactory system.     

Analysis of cellular calcium fluxes in cardiac muscle to understand calcium homeostasis in the heart

Central to controlling intracellular calcium concentration ([Ca(2+)](i)) are a number of Ca(2+) transporters and channels with the L-type Ca(2+) channel, Na(+)-Ca(2+) exchanger and sarcoplasmic reticulum Ca(2+)-ATPase (SERCA) being of particular note in the heart. This review concentrates on the regulation of [Ca(2+)](i) in cardiac muscle and the homeostatic mechanisms employed to ensure that the heart can operate under steady-state conditions on a beat by beat basis. To this end we discuss the relative importance of various sources and sinks of Ca(2+) responsible for initiating contraction and relaxation in cardiac myocytes and how these can be manipulated to regulate the Ca(2+) content of the major Ca(2+) store, the sarcoplasmic reticulum (SR). We will present a simple feedback system detailing how such control can be achieved and highlight how small perturbations to the steady-state operation of the feedback loop can be both beneficial physiologically and underlie changes in systolic Ca(2+) in ageing and heart disease. In addition to manipulating the amplitude of the normal systolic Ca(2+) transient, the tight regulation of SR Ca(2+) content is also required to prevent the abnormal, spontaneous or diastolic release of Ca(2+) from the SR. Such diastolic events are a major factor contributing to the genesis of cardiac arrhythmias in disease situations and in recently identified familial mutations in the SR Ca(2+) release channel (ryanodine receptor, RyR). How such diastolic release arises and potential mechanisms for controlling this will be discussed.     

Opposite effects of PSD-95 and MPP3 PDZ proteins on serotonin 5-hydroxytryptamine2C receptor desensitization and membrane stability

PSD-95/Disc large/Zonula occludens 1 (PDZ) domain-containing proteins (PDZ proteins) play an important role in the targeting and the trafficking of transmembrane proteins. Our previous studies identified a set of PDZ proteins that interact with the C terminus of the serotonin 5-hydroxytryptamine (5-HT)(2C) receptor. Here, we show that the prototypic scaffolding protein postsynaptic density-95 (PSD-95) and another membrane-associated guanylate kinase, MAGUK p55 subfamily member 3 (MPP3), oppositely regulate desensitization of the receptor response in both heterologous cells and mice cortical neurons in primary culture. PSD-95 increased desensitization of the 5-HT(2C) receptor-mediated Ca(2+) response, whereas MPP3 prevented desensitization of the Ca(2+) response. The effects of the PDZ proteins on the desensitization of the Ca(2+) response were correlated with a differential regulation of cell surface expression of the receptor. Additional experiments were performed to assess how PDZ proteins globally modulate desensitization of the 5-HT(2C) receptor response in neurons, by using a peptidyl mimetic of the 5-HT(2C) receptor C terminus fused to the human immunodeficiency virus type-1 Tat protein transduction domain, which disrupts interaction between the 5-HT(2C) receptor and PDZ proteins. Transduction of this peptide inhibitor into cultured cortical neurons increased the desensitization of the 5-HT(2C) receptor-mediated Ca(2+) response. This indicates that, overall, interaction of 5-HT(2C) receptors with PDZ proteins inhibits receptor desensitization in cortical neurons.     

Intracellular Ca2+ regulates the sensitivity of cyclic nucleotide-gated channels in olfactory receptor neurons.

In olfactory receptor neurons, odorants stimulate production of cAMP, which directly activates cyclic nucleotide-gated (CNG) channels. Olfactory adaptation is thought to result from a rise in intracellular Ca2+. To determine whether inhibition of CNG channels plays a role in adaptation, we have investigated the action of Ca2+ on these channels in inside-out "macro" patches from the dendrite and cilia of catfish olfactory neurons. Internal Ca2+, with a K1/2 of 3 microM, profoundly inhibits CNG channels by shifting the dose-response relationship to higher cAMP levels without altering the maximal response. The inhibition does not appear to result from a direct interaction of Ca2+ with the CNG channel. Thus, the inhibition washes out after excision of the inside-out patch, and Ca2+ does not inhibit the cloned catfish CNG channel expressed in Xenopus oocytes. Hence we propose that a regulatory Ca(2+)-binding protein, distinct from the CNG channel, controls the gain of signal transduction and contributes to olfactory adaptation by decreasing the sensitivity of the CNG channel to cAMP.     

Noise analysis of ion channels in non-space-clamped cables: estimates of channel parameters in olfactory cilia.

Ion channels in the cilia of olfactory neurons are part of the transduction machinery of olfaction. Odorant stimuli have been shown to induce a biphasic current response, consisting of a cAMP-activated current and a Ca(2+)-activated Cl- current. We have developed a noise analysis method to study ion channels in leaky cables, such as the olfactory cilium, under non-space-clamp conditions. We performed steady-state noise analysis on ligand-induced currents in excised cilia, voltage-clamped at input and internally perfused with cAMP or Ca2+. The cAMP-activated channels analyzed by this method gave results similar to those of single-channel recordings (gamma = 8.3 pS). Single-channel currents have not yet been recorded for the Ca(2+)-activated Cl- channels. Using our noise analysis method, we estimate a unit conductance, gamma = 0.8 pS, for these channels. The density of channels was found to be approximately 70 channels/micron2 for both channel species.     

Mitochondrial calcium homeostasis as potential target for mitochondrial medicine

Mitochondria are crucial in different intracellular pathways of signal transduction. Mitochondria are capable of decoding a variety of extracellular stimuli into markedly different intracellular actions, ranging from energy production to cell death. The fine modulation of mitochondrial calcium (Ca(2+)) homeostasis plays a fundamental role in many of the processes involving this organelle. When mitochondrial Ca(2+) homeostasis is compromised, different pathological conditions can occur, depending on the cell type involved. Recent data have shed light on the molecular identity of the main proteins involved in the handling of mitochondrial Ca(2+) traffic, opening fascinating and ambitious new avenues for mitochondria-based pharmacological strategies.     

Presynaptic cyclic nucleotide-gated ion channels modulate neurotransmission in the mammalian olfactory bulb

Cyclic nucleotide-gated channels (CNGCs) on the dendritic cilia of olfactory receptor neurons (ORNs) are critical for sensory transduction in the olfactory system. Do CNGCs also play a role in the axons and/or nerve terminals of ORNs? We find that the cyclic nucleotides cAMP and cGMP can both facilitate and depress synaptic transmission between olfactory nerve fibers and their targets in olfactory bulb glomeruli. Cyclic nucleotides increase intracellular Ca(2+) in ORN terminals and enhance spontaneous transmitter release; at higher concentrations, cyclic nucleotides depress evoked transmission by altering olfactory nerve excitability. Cyclic nucleotides have no effect on transmission or nerve excitability, however, in mice lacking olfactory CNGCs. Taken together, our results identify a novel role for presynaptic CNGCs in modulating neurotransmission.