Modulation of N-type Ca2+ channels by intracellular pH in chick sensory neurons

Both physiological and pathological neuronal events, many of which elevate intracellular [Ca2+], can produce changes in intracellular pH of between 0.15 and 0.5 U, between pH 7.4 and 6.8. N-type Ca2+ channels, which are intimately involved in exocytosis and other excitable cell processes, are sensitive to intracellular pH changes. However, the pH range over which N-type Ca2+ channels are sensitive, and the sensitivity of N-type Ca2+ channels to small changes in intracellular pH, are unknown. We studied the influence of intracellular pH changes on N-type calcium channel currents in dorsal root ganglion neurons, acutely isolated from 14-day-old chick embryos. Intracellular pH was monitored in patch-clamp recordings with the fluorescent dye, BCECF, and manipulated in both the acidic and basic direction by extracellular application of NH4+ in the presence and absence of intracellular NH4+. Changes in intracellular pH between 6.6 and 7.5 produced a graded change in Ca2+ current magnitude with no apparent shift in activation potential. Intracellular acidification from pH 7.3 to 7.0 reversibly inhibited Ca2+ currents by 40%. Acidification from pH 7.3 to pH 6.6 reversibly inhibited Ca2+ currents by 65%. Alkalinization from pH 7.3 to 7.5 potentiated Ca2+ currents by approximately 40%. Channels were sensitive to pHi changes with high intracellular concentrations of the Ca2+ chelator, bis-(o-aminophenoxy)-N,N,N',N'-tetraacetic acid, which indicates that the effects of pHi did not involve a Ca2+-dependent mechanism. These data indicate that N-type Ca2+ channel currents are extremely sensitive to small changes in pHi in the range produced by both physiological and pathological events. Furthermore, these data suggest that modulation of N-type Ca2+ channels by pHi may play an important role in physiological processes that produce small changes in pHi and a protective role in pathological mechanisms that produce larger changes in pHi.     

Forskolin prolongs action potential duration and blocks potassium current in embryonic chick sensory neurons

To determine if alterations in internal cyclic adenosine monophosphate (cAMP) play a role in modulation of voltage-dependent channels in embryonic chick sensory neurons in vitro, forskolin (a direct activator of adenylate cyclase) was tested on the cells. Forskolin, in concentrations between 1 and 100 M, produced dose-dependent, reversible increases in action potential duration. This effect of forskolin was blocked by incubation of the neurons in 1 mM 2,5-dideoxyadenosine, an inhibitor of forskolin-induced activation of cyclase in other cells. This suggests that the increase in action potential duration is likely to be mediated by activation of adenylate cyclase. Cholera toxin, another cyclase activator, also increased action potential duration when applied to the sensory neurons in a concentration of 10 g/ml. Forskolin applied to voltage-clamped neurons decreased a voltage-dependent outward current, a result consistent with its effect on the action potential. These effects of forskolin are mimicked by capsaicin, but are in marked contrast to those previously reported for norepinephrine on the action potential and membrane currents (Dunlap and Fischbach 1981). Furthermore, forskolin does not block (or attenuate) the effects of norepinephrine, suggesting that increases in adenylate cyclase activity are most likely not involved in norepinephrine's action on the calcium channel.     

Cholinergic properties of embryonic chick sensory neurons

Experiments were carried out to determine the cholinergic properties of sensory neurons of the chick embryo by measuring the choline acetyltransferase activity (ChAT) and [3H]choline uptake. The choline acetyltransferase activity in the dorsal root ganglia of an 8-day-old chick embryo was 24.2 +/- 2.52, which increased to 45.4 +/- 9.69 pmol ACh/mg protein/min in the ganglia of 12-day-old embryos. Sensory neurons derived from dorsal root ganglia of 10-day-old embryos and maintained in a serum-free culture medium supplemented with insulin, transferrin and nerve growth factor (NGF) also contained significant amounts of ChAT (21.9 pmol ACh/mg protein/min). Omission of NGF resulted in neuronal death, and the enzyme activity could not be measured in these cultures. A specific inhibitor of ChAT, hydroxyethyl naphthylvinyl pyridine (NVP), when added to the assay mix produced a dose-dependent inhibition of ChAT from cultured neurons. Cultured sensory neurons incubated with [3H]choline followed by repeated washouts took up and retained [3H]choline. The uptake of [3H]choline was reduced by about 45% when NaCl, in the incubation medium, was replaced by LiCl. A specific inhibitor of choline uptake, hemicholinium-3, caused about 75% inhibition of [3H]choline uptake. It is implied that sensory neurons of the chick dorsal root ganglia express cholinergic properties during development.     

Two calcium-binding proteins mark many chick sensory neurons

The first immunohistochemical results with a new neuronal calcium-binding protein, calretinin, are presented. Calretinin is related to the 28,000 mol. wt calcium-binding protein, calbindin, and a survey of the chick brain by in situ hybridization has identified the brain nuclei that expressed the genes for the two proteins [Rogers J.H., J. Cell Biol. 105, 1343 (1987)]. Now, antisera have been raised against calretinin fusion proteins in order to visualize individual neurons. The antisera have been used in an immunohistochemical survey of calretinin and calbindin in the chick sensory nuclei and ganglia, where these two proteins are found to be particularly prevalent. In the central nervous system, they are seen in many secondary sensory neurons and local circuit neurons, the two proteins being almost always in separate cells. However, in ganglion cells of the spinal nerves, inner ear, and retina, they are often expressed together. Their distribution in the brain is generally different from that of a third calcium-binding protein, parvalbumin. These proteins may modulate many important calcium-dependent processes in neurons, and probably have multiple functions.     

Voltage-dependent GABA-induced modulation of calcium currents in chick sensory neurons

Externally applied gamma-aminobutyric acid (GABA) quickly and reversibly reduces by 60% voltage activated Ca2+ currents in chick dorsal root ganglion cells. This action is antagonized by depolarization, with characteristic time and voltage requirements. Intracellular perfusion with guanosine 5'-O-(3-thiotriphosphate) (GTP gamma S) or guanosine 5'-O-(2-thiodiphosphate) (GDP beta S) mimicks and blocks the GABA effect, respectively. A 3-state model describing the reactions involved is proposed.     

Trifluoperazine blocks GABA-gated chloride currents in cultured chick spinal cord neurons

1. The effects of trifluoperazine (TFP) on gamma-aminobutyric acid- (GABA) gated chloride currents were investigated using cultured chick embryonic spinal cord neurons (SCN) and gigaseal patch clamp recording techniques. 2. TFP showed a dose-dependent attenuation of GABA responses with half-maximal block near 9 microM. The GABA antagonism was noncompetitive and voltage dependent with greater block at hyperpolarized potentials. 3. At 10 microM, GABA induced little desensitization in response to prolonged applications in the absence of TFP. At 100 microM, GABA responses desensitized with a time constant near 27 s. Coapplication of TFP did not alter the lack of desensitization to 10 microM GABA but revealed a second, faster component (time constant approximately 0.7 s) to the attenuation at 100 microM GABA. 4. Steady-state fluctuation analysis of the macroscopic GABA-gated current gave a power spectrum that was described by a simple Lorentzian function. The corner frequency of fluctuations to GABA [34.6 +/- 7.3 (SE) Hz] remained unchanged during simultaneous application of GABA and TFP (33.0 +/- 8.5 Hz), indicating that TFP does not alter the average GABA channel open duration. 5. Single-channel recording from isolated outside-out membrane patches showed GABA-gated chloride events with a primary conductance of 26 pS and a minor component (representing less than 5% of all events) of 10 pS. With simultaneous application of GABA and TFP, event amplitudes remained unchanged but the frequency of opening was decreased. 6. The distribution of the main conductance channel open times were well described by the sum of two exponentials with time constants of 0.82 +/- 0.18 and 7.4 +/- 2.0 ms. These time constants were not significantly altered by 50 microM TFP. 7. TFP, at 100 microM, attenuated responses to (% of control): GABA (7.5 +/- 2.3), glycine (15.4 +/- 5.5), and glutamate (64.5 +/- 5.5). The proconvulsant tendency of TFP, in part, may be due to the greater block of responses to inhibitory than to excitatory transmitters.     

gamma-Aminobutyric acid-induced depression of calcium currents of chick sensory neurons

The effects of gamma-aminobutyric acid (GABA) on calcium currents were investigated in avian dorsal root ganglion (DRG) cells. GABA was applied to the vicinity of the cells by ejection pipettes using constant-pressure pulses. GABA concentrations between 5 and 100 mu M reduced and slowed the calcium current in a dose-dependent manner. A contribution of K and Cl outward currents to the reduction of the inward current was minimized by using identical caesium chloride concentrations on both sides of the membrane. The onset of the effect was rapid and 80% of the effect was observed within 1 s. The attenuation of the Ca slope conductance by GABA was found to be independent of the membrane potential between -50 and +50 mV.     

Characterization of two kinds of high-voltage-activated Ca-channel currents in chick sensory neurons

High-voltage-activated (HVA) Ca-channel currents in chick sensory neurons were characterized by dihydropyridine compounds (DHPs) and -conotoxin GVIA (CTX) using patch-clamp methods. In single-channel recordings, two HVA-currents were identified by their single-channel conductances, 13 pS and 25 pS in 110 mM BaCl₂. DHPs selectively affected the large-conductance channel. CTX (5 M), on the other hand, irreversibly eliminated only the small-conductance channel, while the large-conductance channel was either unaffected or only transiently blocked. In whole-cell recordings the macroscopic HVA-current was completely and irreversibly blocked by CTX but insensitive to DHPs in 60% of the cells. This current presumably was carried by the 13 pS channel. In the remaining cells, a part of the HVA-current (10%, SD=11%) was either unaffected or transiently blocked by CTX and was sensitive to DHPs. This current presumably was carried by the 25 pS channel. Inactivation of both macroscopic current component was incomplete during a 150 ms long depolarization. Our data suggest that the HVA-currents in chick sensory neurons are carried by two distinct Ca-channels that are differentially affected by CTX and DHPs.     

Divalent cation dependent inactivation of the high-voltage-activated Ca-channel current in chick sensory neurons

We have used the whole-cell clamp technique to investigate inactivation of the -conotoxin sensitive high-voltage-activated Ca-channel current (HVA current [2]) carried either by Ca, Ba or Sr (2.5 mM) in chick sensory neurons. At a low internal EGTA concentration (0.1 mM), Ca-channel currents clearly inactivated irrespective of the species of divalent cation carrying the current. During 150 ms pulses, current inactivated to 0.57, 0.67 and 0.75 of the peak current in Ca, Ba and Sr solution, respectively. Time constants of inactivation (26±10 ms and 280±50 ms, mean±S.D., in Ba) were largely independent of the membrane potential. Double-pulse experiments showed that the amount of inactivation left by a pre-pulse was proportional to the amplitude of the current evoked by the pre-pulse. No inactivation was induced by an outward current elicited by a strong depolarization to ü60 mV. With an internal EGTA concentration of 20 mM, the amount of inactivation was significantly smaller. In conclusion, the inactivation of the HVA Ca-channel currents during current flow depends mostly on the entry of divalent cations irrespective of their species.     

Cyclic AMP selectively reduces the N-type calcium current component of mouse sensory neurons in culture by enhancing inactivation

1. The single-electrode voltage-clamp technique was used to assess the effect of elevated intracellular cyclic AMP levels on the three calcium current components of mouse dorsal root ganglion (DRG) neurons in culture. 2. Neither forskolin, cholera toxin, nor 8-Br-cyclic AMP affected the isolated transient low-threshold (T) calcium current. 3. When calcium currents were evoked at clamp potentials (Vc) positive to -20 mV from holding potentials (Vh) near the resting membrane potential, the calcium current consisted primarily of the transient high-threshold (N) and the slowly inactivating high-threshold (L) calcium current components. Under these conditions forskolin, cholera toxin, and 8-Br-cyclic AMP reduced the peak calcium current but had little or no effect on the late (greater than or equal to 300 ms) calcium current. When calcium currents were evoked from very negative Vh, however, there was no effect of these compounds. 4. Forskolin had no effect on the voltage-dependence of the current-voltage relation, nor on the rate of recovery of the calcium current from inactivation. 5. In other experiments, current traces were fitted using a multiexponential curve-fitting program that determined the amplitudes and inactivation time constants (tau i) of the three calcium current components. Forskolin selectively reduced the magnitude of the (curve-fitted) N current, and reduced its tau i. 6. Forskolin also enhanced steady-state inactivation of the N current, producing a -7.5 mV shift in the steady-state inactivation curve. 7. Cholera toxin, forskolin, and 8-Br-cyclic AMP had similar effects on calcium currents in mouse DRG neurons in culture.(ABSTRACT TRUNCATED AT 250 WORDS)     

Spatial and temporal control of intracellular free Ca2+ in chick sensory neurons

Digital imaging of fura-2 fluorescence and the voltage-clamp technique were combined to study cytoplasmic free Ca²⁺ concentration, [Ca]_i, in neurons cultured from chick dorsal root ganglia. Depolarizing pulses raised [Ca]_i to a new steady-state level which was achieved earlier in neurites than in the soma. The rise in [Ca]_i during stimulated bursting or rhythmic activity was also faster in neurites. After stimulation [Ca]_i recovered monoexponentially in the soma and biexponentially in neurites. Application of 50 mM KCl produced membrane depolarization and a concomitant increase of [Ca]_i. During wash-out [Ca]_i often declined to an intermediate steady-state level at which it stayed for several minutes. Thereafter the resting level of [Ca]_i was quickly restored. [Ca]_i recovery was delayed after treating the cell with 2 M thapsigargin, an inhibitor of the Ca²⁺ pump of internal Ca²⁺ stores. Caffeine (10 mM) transiently increased [Ca]_i. A second caffeine application produced smaller [Ca]_i changes due to the prior depletion of Ca²⁺ stores, which could be replenished by brief exposure to KCl. Thapsigargin (2 M) transiently increased [Ca]_i both in the standard and Ca²⁺-free solution. [Ca]_i transients due to caffeine and thapsigargin started in the cell interior, in contrast to [Ca]_i changes evoked by membrane depolarization, which were noticed first at the cell edge. Caffeine and thapsigargin induced a transient inward current which persisted in the presence of 1 mM La³⁺ and in Ca²⁺-free solutions, but which was greatly diminished in Na⁺-free solutions. The effects of caffeine and thapsigargin were mutually exclusive both in the generation of [Ca]_i transients and in the inward current induction.     

Peripheral projections of the chick primary sensory neurons expressing gamma-aminobutyric acid immunoreactivity.

The expression of gamma-aminobutyric acid was studied in sensory neurons and peripheral target tissues of the chick dorsal root ganglia by combining immunocytochemistry and electron microscopy. In the chick embryos, the first immunoreaction was observed at embryonic day 12 in 1.4% of ganglion cell bodies. The intensity of immunostaining gradually increased during development and the percentage of immunostained neurons reached an average of 7.3% after hatching. These immunostained cell bodies could be identified as sensory neurons belonging either to some large neurons of the A1 subclass or to a few small neurons of the B1 subclass. The other neuronal cell bodies, corresponding to the A2 and B2 subclasses, as well as the satellite and glial cells were apparently devoid of any gamma-aminobutyric acid immunostaining. Among the peripheral tissues innervated by the primary sensory neurons, the nerve endings of Achilles' tendon and the paravertebral autonomic ganglia appeared devoid of immunoreactivity. In contrast, immunoreactivity was found within nerve endings located in some neuromuscular spindles of the skeletal muscles and within some Herbst's corpuscles in the subcutaneous tissue of the skin. Thus, the present results provide evidence that gamma-aminobutyric acid may be expressed by neuronal cell bodies belonging to two subclasses of primary sensory neurons and could be a putative neurotransmitter involved in the peripheral sensory innervation of, at least in part, skin and skeletal muscles.     

Calbindin D-28k-immunoreactive neurons in chick dorsal root ganglion: ontogenesis and cytological characteristics of the immunoreactive sensory neurons

The expression of calbindin (28,000 mol. wt Vitamin D-dependent calcium binding protein) was studied in sensory neurons of the chick dorsal root ganglion by combining immunocytochemical and ultrastructural features. In the chick embryo at E10, about 20% of the neuroblasts were immunostained with antibodies raised to calbindin. After hatching, two subpopulations of primary sensory neurons were labeled with calbindin-antibodies and could be identified: the large Al cell bodies (6%) mainly characterized by huge blocks of rough endoplasmic reticulum and the small B1 cell bodies (14%) which contain parallel cisternae of rough endoplasmic reticulum. The other neuronal cell types A2 and B2 were devoid of any immunostaining. Thus, calbindin is an early and reliable marker of two subclasses of primary sensory neurons in the chick dorsal root ganglion.     

Forskolin mimics and blocks a serotonin-sensitive decreased K+ conductance in tail sensory neurons of Aplysia

Facilitation of synaptic connections between sensory neurons and motor neurons mediating the tail-withdrawal reflex in Aplysia is produced by sensitizing stimuli. These effects can be mimicked by perfusing the pleural and pedal ganglia of a semi-intact preparation with serotonin (5-HT). In addition to the synaptic facilitation, 5-HT produces a depolarization associated with an increase in input resistance in the sensory neurons. The 5-HT-induced changes appear to be mediated by an elevation in cAMP levels. To examine further the role of cAMP in mediating the 5-HT response we utilized the potent cyclase activator forskolin. Forskolin mimics the 5-HT response and blocks the response to subsequent 5-HT applications indicating that both 5-HT and forskolin act through a common saturable mechanism. Voltage clamp and ion substitution experiments indicate that the 5-HT response is due, at least in part, to a decrease in a resting K+ conductance.     

Functional and pharmacological differences between two types of GABA receptor on embryonic chick sensory neurons

Embryonic chick sensory neurons grown in dissociated cell culture exhibit two functional responses to GABA: an increase in resting membrane permeability to chloride (Cl) ions (resulting in membrane depolarization) and a decrease in voltage-dependent calcium (Ca) channel current (resulting in a decreased action potential duration). These two functional effects differ in a number of ways. (1) The increase in resting membrane permeability desensitizes in the maintained presence of GABA, while the decrease in action potential duration does not. (2) Muscimol is a selective agonist for the increase in resting conductance, while baclofen is a selective agonist for the decrease in action potential duration. (3) Bicuculline inhibits the GABA- or muscimol-induced increase in Cl permeability, but it does not block the GABA- or baclofen-induced decrease in action potential duration. These functional and pharmacological differences between the two effects of GABA suggest that two separate receptors are involved.     

Modulation of Ca-channel current by an adenosine analog mediated by a GTP-binding protein in chick sensory neurons

Inhibitory modulation of the high-voltage-activated (HVA) Ca-channel current by 2-chloroadenosine (2CA) was studied in chick sensory neurons using the whole-cell clamp method. 2CA reduced the CTX-sensitive HVA-current (Aosaki and Kasai 1989) in a dose-dependent manner with aK _d of 0.8 M. The inhibition by 2CA was also voltage-dependent, being maximal at hyperpolarized potentials, and completely removed at potentials more positive than 30 mV. This voltage-dependence of 2CA action was also evident as a progressive increase in Ca-channel current magnitude during a depolarization which could be described by a single exponential function and which became faster at larger depolarizations. The concentration of 2CA affected the steady-state reduction in Ca-channel current, but did not alter the time-course of current increase during depolarization. The voltage-dependent effect of 2CA was mimicked by intracellular application of GTP-S, but not by phorbol ester, arachidonic acid or nitroprusside. These results are consistent with model in which 2CA activates a G-protein, which then unmasks an additional activation gate on the Ca-channel.     

Low-threshold heat receptor in chick sensory neurons is upregulated independently of nerve growth factor after nerve injury

In mammals, the cloned low-threshold heat receptor, vanilloid receptor subtype 1 (VR1), is involved in the genesis of thermal hyperalgesia after inflammation. However, there is evidence that VR1 is not involved in the thermal hyperalgesia that occurs after nerve injury. In search for other heat receptors which might be involved in this phenomenon, we previously demonstrated that chick dorsal root ganglion neurons, which are insensitive to capsaicin, respond to low-threshold heat. Here, we investigated whether expression of the low-threshold noxious heat receptor in chicks is regulated by nerve growth factor (NGF), as VR1 is in mammals. Heat (44 degrees C) responsiveness of isolated dorsal root ganglion neurons of chicks was investigated (i) under culture conditions for up to 4 days with and without NGF and (ii) after a tight ligation of the sciatic nerve for up to 6 days, using cobalt-uptake method. In every case, a significant upregulation in the proportion of heat-responsive neurons was observed. On the molecular level, there was an increase of chick VR1 mRNA level in dorsal root ganglion cells cultured for 3 days in medium lacking NGF. In rat dorsal root ganglion neurons cultured for 1-4 days without NGF, patch-clamp experiments revealed that after 1 day almost all neurons responding to heat also responded to capsaicin, whereas after 3-4 days, more than one-half of the heat-responsive neurons did not respond to capsaicin.These data suggest the existence of low-threshold heat receptors in chick dorsal root ganglion neurons, the expression of which is regulated independently of NGF.