Genetic analysis of a synaptic calcium channel in Drosophila: intragenic modifiers of a temperature-sensitive paralytic mutant of cacophony.

Our previous genetic analysis of synaptic mechanisms in Drosophila identified a temperature-sensitive paralytic mutant of the voltage-gated calcium channel alpha1 subunit gene, cacophony (cac). Electrophysiological studies in this mutant, designated cac(TS2), indicated cac encodes a primary calcium channel alpha1 subunit functioning in neurotransmitter release. To further examine the functions and interactions of cac-encoded calcium channels, a genetic screen was performed to isolate new mutations that modify the cac(TS2) paralytic phenotype. The screen recovered 10 mutations that enhance or suppress cac(TS2), including second-site mutations in cac (intragenic modifiers) as well as mutations mapping to other genes (extragenic modifiers). Here we report molecular characterization of three intragenic modifiers and examine the consequences of these mutations for temperature-sensitive behavior, synaptic function, and processing of cac pre-mRNAs. These mutations may further define the structural basis of calcium channel alpha1 subunit function in neurotransmitter release.

Fast synaptic fatigue in shibire mutants reveals a rapid requirement for dynamin in synaptic vesicle membrane trafficking.

The GTPase dynamin is involved in endocytosis in many cell types, as first revealed by temperature-sensitive paralytic mutations in the Drosophila dynamin gene, shibire (shi), which disrupt synaptic vesicle endocytosis and deplete synaptic terminals of vesicles. Here we report that shi synapses exhibit a fast synaptic fatigue phenotype within 20 ms of repetitive stimulation, which cannot be explained by vesicle depletion, as we confirmed by electron microscopy. These results suggest that, in addition to its well-characterized role in synaptic vesicle recycling, dynamin may be required for short-term maintenance of the readily releasable pool of synaptic vesicles.

A temperature-sensitive paralytic mutant defines a primary synaptic calcium channel in Drosophila.

Neurotransmission at chemical synapses involves regulated exocytosis of neurotransmitter from the presynaptic terminal. Neurotransmitter release is thought to be triggered by calcium influx through specific classes of voltage-gated calcium channels. Here we report genetic and functional analysis implicating a specific calcium channel gene product in neurotransmitter release. We have isolated a temperature-sensitive paralytic allele of the Drosophila calcium channel alpha1 subunit gene, cacophony (cac). This mutant, referred to as cac(TS2), allows functional analysis of synaptic transmission after acute perturbation of a specific alpha1 subunit. Electrophysiological analysis at neuromuscular synapses revealed that neurotransmitter release in cac(TS2) is markedly reduced at elevated temperatures, indicating that cac encodes a primary alpha1 subunit functioning in synaptic transmission. These observations further define the molecular basis of voltage-gated calcium entry at synapses and provide a new starting point for further genetic analysis of synaptic mechanisms.

Genetic modifiers of the Drosophila NSF mutant, comatose, include a temperature-sensitive paralytic allele of the calcium channel alpha1-subunit gene, cacophony.

The N-ethylmaleimide-sensitive fusion protein (NSF) has been implicated in vesicle trafficking in perhaps all eukaryotic cells. The Drosophila comatose (comt) gene encodes an NSF homolog, dNSF1. Our previous work with temperature-sensitive (TS) paralytic alleles of comt has revealed a function for dNSF1 at synapses, where it appears to prime synaptic vesicles for neurotransmitter release. To further examine the molecular basis of dNSF1 function and to broaden our analysis of synaptic transmission to other gene products, we have performed a genetic screen for mutations that interact with comt. Here we report the isolation and analysis of four mutations that modify TS paralysis in comt, including two intragenic modifiers (one enhancer and one suppressor) and two extragenic modifiers (both enhancers). The intragenic mutations will contribute to structure-function analysis of dNSF1 and the extragenic mutations identify gene products with related functions in synaptic transmission. Both extragenic enhancers result in TS behavioral phenotypes when separated from comt, and both map to loci not previously identified in screens for TS mutants. One of these mutations is a TS paralytic allele of the calcium channel alpha1-subunit gene, cacophony (cac). Analysis of synaptic function in these mutants alone and in combination will further define the in vivo functions and interactions of specific gene products in synaptic transmission.

The Drosophila NSF protein, dNSF1, plays a similar role at neuromuscular and some central synapses.

The N-ethylmaleimide sensitive fusion protein (NSF) was originally identified as a cytosolic factor required for constitutive vesicular transport and later implicated in synaptic vesicle trafficking as well. Our previous work at neuromuscular synapses in the temperature-sensitive NSF mutant, comatose (comt), has shown that the comt gene product, dNSF1, functions after synaptic vesicle docking in the priming of vesicles for fast calcium-triggered fusion. Here we investigate whether dNSF1 performs a similar function at central synapses associated with the well-characterized giant fiber neural pathway. These include a synapse within the giant fiber pathway, made by the peripherally synapsing interneuron (PSI), as well as synapses providing input to the giant fiber pathway. The latency (delay) between stimulation and a resulting muscle action potential was used to assess the function of each class of synapses. Repetitive stimulation of the giant fiber pathway in comt produced wild-type responses at both 20 and 36 degrees C, exhibiting a characteristic and constant latency between stimulation and the muscle response. In contrast, stimulation of presynaptic inputs to the giant fiber (referred to as the "long latency pathway") revealed a striking difference between wild type and comt at 36 degrees C. Repetitive stimulation of the long latency pathway led to a progressive, activity-dependent increase in the response latency in comt, but not in wild type. Thus the giant fiber pathway, including the PSI synapse, appears to function normally in comt, whereas the presynaptic inputs to the giant fiber pathway are disrupted. Several aspects of the progressive latency increase observed in the long latency pathway can be understood in the context of the activity-dependent reduction in neurotransmitter release we observed previously at neuromuscular synapses. These results suggest that repetitive stimulation causes a progressive reduction in neurotransmitter release by presynaptic inputs to the giant fiber neuron, resulting in an increased latency preceding a giant fiber action potential. Thus synapses presynaptic to the giant fiber appear to utilize dNSF1 in a manner similar to the neuromuscular synapse, whereas the PSI chemical synapse may differ with respect to the expression or activity of dNSF1.

Synaptic physiology and ultrastructure in comatose mutants define an in vivo role for NSF in neurotransmitter release.

N-Ethylmaleimide-sensitive fusion protein (NSF) is a cytosolic protein thought to play a key role in vesicular transport in all eukaryotic cells. Although NSF was proposed to function in the trafficking of synaptic vesicles responsible for neurotransmitter release, only recently have in vivo experiments begun to reveal a specific function for NSF in this process. Our previous work showed that mutations in a Drosophila NSF gene, dNSF1, are responsible for the temperature-sensitive paralytic phenotype in comatose (comt) mutants. In this study, we perform electrophysiological and ultrastructural analyses in three different comt alleles to investigate the function of dNSF1 at native synapses in vivo. Electrophysiological analysis of postsynaptic potentials and currents at adult neuromuscular synapses revealed that in the absence of repetitive stimulation, comt synapses exhibit wild-type neurotransmitter release at restrictive (paralytic) temperatures. In contrast, repetitive stimulation at restrictive temperatures revealed a progressive, activity-dependent reduction in neurotransmitter release in comt but not in wild type. These results indicate that dNSF1 does not participate directly in the fusion of vesicles with the target membrane but rather functions in maintaining the pool of readily releasable vesicles competent for fast calcium-triggered fusion. To define dNSF1 function further, we used transmission electron microscopy to examine the distribution of vesicles within synaptic terminals, and observed a marked accumulation of docked vesicles at restrictive temperatures in comt. Together, the results reported here define a role for dNSF1 in the priming of docked synaptic vesicles for calcium-triggered fusion.

Distinct roles for N-ethylmaleimide-sensitive fusion protein (NSF) suggested by the identification of a second Drosophila NSF homolog.

The N-ethylmaleimide-sensitive fusion protein (NSF) is a cytoplasmic protein implicated in the fusion of intracellular transport vesicles with their target membranes. NSF is thought to function in the fusion of essentially all types of vesicles, including endoplasmic reticulum, Golgi, and endocytic vesicles, as well as secretory vesicles undergoing regulated fusion (for review see Rothman, J.E. (1994) Nature 372, 55-63). However, little experimental evidence exists to address the possibility that organisms might have multiple NSF proteins serving distinct functions in the same or different cells. We previously cloned a neurally expressed Drosophila homolog, dNSF-1 (Ordway, R.W., Pallanck, L., and Ganetzky, B. (1994) Proc. Natl. Acad. Sci. U.S.A. 91, 5715-5719), and have subsequently identified mutations in this gene that confer an apparent failure of synaptic transmission at elevated temperature (Pallanck, L., Ordway, R.W., and Ganetzky, B. (1995) Nature, 376, 25; Siddiqi, O., and Benzer, S. (1976) Proc. Natl. Acad. Sci. U.S.A. 73, 3253-3257). Here we report that 1) Drosophila contains a second NSF homolog, termed dNSF-2, that exhibits 84% amino acid identity to dNSF-1, 2) dNSF-1 and dNSF-2 display overlapping but different temporal expression, and 3) multiple transcripts are derived from the dNSF-2 gene. These findings raise the possibility that different NSF gene products serve distinct or overlapping functions with the organism.

Stretch activation of a toad smooth muscle K+ channel may be mediated by fatty acids.

1. using standard single channel patch clamp techniques we studied the stretch sensitivity of a 20 pS K(+)-selective channel which is activated by fatty acids and found in freshly dissociated smooth muscle cells from the stomach of the toad Bufo marinus. 2. A pulse of suction applied to the back of the patch pipette in order to stretch the membrane resulted in activation of this K+ channel. A train of suction pulses resulted in a gradually increased level of channel activity during each successive pulse, as well as an increase in baseline activity between pulses. This pattern contrasts markedly with many other stretch-activated channels whose activation is limited to the duration of the suction pulse. 3. Application of fatty acids augmented the response to stretch. In contrast, application of 10 microM defatted albumin, which removes fatty acids from membranes, rapidly and reversibly decreased the response to stretch. 4. These results are consistent with the hypothesis that fatty acids which are generated by mechanical stimuli, perhaps by mechanically activated phospholipases, are the intermediaries in activation of certain mechanically sensitive ion channels.

Direct effects of fatty acids and other charged lipids on ion channel activity in smooth muscle cells.

A variety of fatty acids increase the activity of certain types of K+ channels. This effect is not dependent on the three enzymatic pathways that convert arachidonic acid to various bioactive oxygenated metabolites. One type of K+ channel in toad stomach smooth muscle cell membranes in activated by fatty acids and other single chain lipids which possess both a negatively charged head group and a sufficiently hydrophobic acyl chain. Neutral lipids have no effect on K+ channel activity, while positively charged lipids with a sufficiently hydrophobic acyl chain suppress channel activity. Acyl Coenzyme A's, which do not flip across the bilayer, act only from the cytosolic surface of the membrane, suggesting that the binding site for channel activation is also located there. This fatty acid-activated channel is also activated by membrane stretch. Moreover, this mechanical response is either mediated or modulated by fatty acids. Thus, fatty acids and other charged single chain lipids may comprise another class of first or second messenger molecules that target ion channels.

Neurally expressed Drosophila genes encoding homologs of the NSF and SNAP secretory proteins.

Several lines of investigation have now converged to indicate that the neurotransmitter release apparatus is formed by assembly of cytosolic proteins with proteins of the synaptic vesicle and presynaptic terminal membranes. We are undertaking a genetic approach in Drosophila melanogaster to investigate the functions of two types of cytosolic proteins thought to function in this complex: N-ethylmaleimide-sensitive fusion protein (NSF) and the soluble NSF attachment proteins (SNAPs). We have identified Drosophila homologs of the vertebrate and yeast NSF and SNAP genes. Both Drosophila genes encode polypeptides that closely resemble their vertebrate counterparts and are expressed in the nervous system; neither appears to be in a family of closely related Drosophila genes. These results indicate that the Drosophila NSF and SNAP genes are excellent candidates for mutational analysis of neurotransmitter release.

Structural requirements for charged lipid molecules to directly increase or suppress K+ channel activity in smooth muscle cells. Effects of fatty acids, lysophosphatidate, acyl coenzyme A and sphingosine.

We determined the structural features necessary for fatty acids to exert their action on K+ channels of gastric smooth muscle cells. Examination of the effects of a variety of synthetic and naturally occurring lipid compounds on K+ channel activity in cell-attached and excised membrane patches revealed that negatively charged analogs of medium to long chain fatty acids (but not short chain analogs) as well as certain other negatively charged lipids activate the channels. In contrast, positively charged, medium to long chain analogs suppress activity, and neutral analogs are without effect. The key requirements for effective compounds seem to be a sufficiently hydrophobic domain and the presence of a charged group. Furthermore, those negatively charged compounds unable to "flip" across the bilayer are effective only when applied at the cytosolic surface of the membrane, suggesting that the site of fatty acid action is also located there. Finally, because some of the effective compounds, for example, the fatty acids themselves, lysophosphatidate, acyl Coenzyme A, and sphingosine, are naturally occurring substances and can be liberated by agonist-activated or metabolic enzymes, they may act as second messengers targeting ion channels.

Both membrane stretch and fatty acids directly activate large conductance Ca(2+)-activated K+ channels in vascular smooth muscle cells.

Large conductance Ca(2+)-activated K+ channels in rabbit pulmonary artery smooth muscle cells are activated by membrane stretch and by arachidonic acid and other fatty acids. Activation by stretch appears to occur by a direct effect of stretch on the channel itself or a closely associated component. In excised inside-out patches stretch activation was seen under conditions which precluded possible mechanisms involving cytosolic factors, release of Ca2+ from intracellular stores, or stretch induced transmembrane flux of Ca2+ or other ions potentially capable of activating the channel. Fatty acids also directly activate this channel. Like stretch activation, fatty acid activation occurs in excised inside-out patches in the absence of cytosolic constituents. Moreover, the channel is activated by fatty acids which, unlike arachidonic acid, are not substrates for the cyclo-oxygenase or lypoxygenase pathways, indicating that oxygenated metabolites do not mediate the response. Thus, four distinct types of stimuli (cytosolic Ca2+, membrane potential, membrane stretch, and fatty acids) can directly affect the activity of this channel.

Direct regulation of ion channels by fatty acids.

A variety of fatty acids regulate the activity of specific ion channels by mechanisms not involving the enzymatic pathways that convert arachidonic acid to oxygenated metabolites. Furthermore, these actions of fatty acids occur in patches of membrane excised from the cell and are not mediated by cellular signal transduction pathways that require soluble factors such as nucleotides and calcium. Thus, fatty acids themselves appear to regulate the action of channels directly, much as they regulate the action of several purified enzymes, and might constitute a new class of first or second messengers acting on ion channels.

Hyperpolarization-activated cationic channels in smooth muscle cells are stretch sensitive.

The properties of hyperpolarization-activated channels were studied in single smooth muscle cells from the stomach of the toad, Bufo marinus, using the patch-clamp technique. In cell-attached patches, inward channel currents were activated by hyperpolarizing pulses from a holding potential of -20 mV to potentials more negative than -60 mV. The activity of the channels increased and their latency of activation decreased as the hyperpolarization was increased. The slope conductance of the channels with standard high sodium concentration pipette solution was 64.2 +/- 9.1 pS (SD, n = 17). Stretching the patch, by suction applied to the back of the patch pipette, also increased the activity and shortened the latency of activation. We designate these channels as HA-SACs (hyperpolarization- and stretch-activated channels). HA-SACs were observed in 83% (175/210) of the patches studied. HA-SAC currents were carried by sodium and potassium ions, but their amplitude was increased by replacing extracellular sodium with potassium. Extracellular magnesium and calcium ions significantly reduced the single-channel conductance of HA-SACs. These permeation characteristics and the single-channel conductance of HA-SACs were indistinguishable from those of stretch-activated channels (SACs) previously described in these cells. The following observations are consistent with HA-SACs being a subset of SACs. First, SACs were at times found in cell-attached patches which lacked HA-SACs. Second, the number of channels in a cell-attached patch simultaneously activated by stretch (usually 5-10 and often more) exceeded by far the number simultaneously activated by hyperpolarization (usually one or two).(ABSTRACT TRUNCATED AT 250 WORDS)

Arachidonic acid and other fatty acids directly activate potassium channels in smooth muscle cells.

Arachidonic acid, as well as fatty acids that are not substrates for cyclooxygenase and lipoxygenase enzymes, activated a specific type of potassium channel in freshly dissociated smooth muscle cells. Activation occurred in excised membrane patches in the absence of calcium and all nucleotides. Therefore signal transduction pathways that require such soluble factors, including the NADPH-dependent cytochrome P450 pathway, do not mediate the response. Thus, fatty acids directly activate potassium channels and so may constitute a class of signal molecules that regulate ion channels.

Endogenous adenosine inhibits catecholamine contractile responses in normoxic hearts.

The importance of endogenous myocardial adenosine in attenuating catecholamine-elicited contractile responses was investigated in perfused oxygenated rat hearts. Perfusion of the isolated hearts with adenosine deaminase potentiated the isoproterenol-induced increases of three contractile variables (left ventricular pressure development and rates of both left ventricular pressure development and relaxation). The peak (maximal, within 30 s) and maintained (after 1 min) increases of the contractile variables caused by 10(-8) M isoproterenol were enhanced by 15-22 and 31-43%, respectively. Adenosine deaminase appeared in epicardial surface transudates of similarly perfused hearts, indicating that the enzyme had entered the myocardial interstitial space. Isoproterenol alone elevated the release of adenosine into coronary effluents of isoproterenol-stimulated hearts, and adenosine deaminase prevented the release of the nucleoside. The higher the level of adenosine in the effluent, the greater the reduction of the peak contractile variables. Phenylisopropyladenosine at 10(-8) M prevented the adenosine deaminase potentiation of 10(-9) M isoproterenol-induced contractile responses. The adenosine analogue at 10(-6) M blocked completely the isoproterenol-produced increases in the contractile variables. These results suggest that endogenous adenosine prevents full mechanical responsiveness to beta-adrenoceptor stimulation in the oxygenated myocardium. In addition, the findings support the notion that adenosine serves as an important negative feedback modulator in the oxygenated heart.