Inhibition of the I(h) current in isolated peripheral nerve: a novel mode of peripheral antinociception?

Although the alpha2-adrenergic agonist clonidine has been shown to promote peripheral antinociception, its mechanism of action has not yet been clearly elucidated. By the use of the sucrose-gap method, we have shown that in C fibers of the rabbit vagus nerve, clonidine at micromolar concentrations enhances activity-dependent hyperpolarizations generated by the Na+-K+ pump during and after repetitive stimulation. Similar results were obtained with 10 microM of ZD 7288, a specific blocker of the hyperpolarization-activated cation current (I(h)) and with 2 mM of Ba2+ that blocks the inwardly rectifying potassium current (I(KIR)). Furthermore, clonidine had no added effect on the ZD 7288-induced response, whereas it produced a marked enhancement of Ba2+induced response. From these results, it can be concluded that clonidine enhances activity-dependent hyperpolarization by inhibiting the current I(h). We propose that clonidine, by increasing the threshold for initiating the action potential, induces a slowing or block of conduction and that this mechanism is the origin of the clonidine-induced antinociception. Finally, this study suggests a novel role for inwardly rectifying hyperpolarization-activated conductances in peripherally mediated antinociception.

Mechanism of antinociceptive action of clonidine in nonmyelinated nerve fibres.

Despite a large body of clinical evidence in favour of a local anesthetic effect of clonidine, the underlying mechanism has not yet been elucidated. In this study we have used the sucrose-gap method to measure the effects of clonidine on the electrophysiological properties of nonmyelinated nerve fibers in the rabbit vagus nerve. The results showed that clonidine enhanced the hyperpolarizing and reduced the depolarizing afterpotential that follow compound action potentials during electrical activity. We showed that summation of these afterpotentials shifts the membrane potential toward more negative values, thus creating a region of low safety conduction, where the local circuit currents might fail to depolarize the axonal membrane to the threshold value needed to open voltage-dependent Na(+) channels. Yohimbine did not reverse the inhibitory effects of clonidine on impulse propagation, indicating that the observed effects of clonidine relies on mechanisms not mediated by alpha(2)-adrenoceptors.

Axonal and glial currents activated during the post-tetanic hyperpolarization in non-myelinated nerve.

Changes in membrane potential and potassium concentration in the extracellular space ([K+]e) of rabbit vagus nerve were measured simultaneously during electrical activity and during the period of recovery using a modified sucrose-gap method and potassium-sensitive microelectrodes. After stimulation for 15 s at 15 Hz the main activity-induced increase in [K+]e reached 16.9 mM. This increase in [K+]e was paralleled by a depolarization of the preparation. The period of activity was followed by a post-tetanic hyperpolarization (PTH) lasting tens of seconds, generated by the axonal electrogenic Na+-K+ pump and to a lesser extent by the pump of the surrounding Schwann cells. The amplitude of the PTH dramatically increased in experiments in which inward currents were blocked by removal of Cl­ or after application of Cs+ or Ba2+, indicating that under normal conditions the current generated by the Na+-K+ pump is strongly short-circuited. A pharmacological and kinetic study showed that these currents are: (1) the hyperpolarization-activated current I h, and (2) the inwardly rectifying I KIR current. The results show that the latter originates from Schwann cells. Our data indicate that in non-myelinated nerves there is a subtle association of inward ionic channels which (1) helps the cell to maintain an optimal membrane potential after a period of activity, and (2) contributes to the removal of excess K+ from the extracellular space.

Lactate is released and taken up by isolated rabbit vagus nerve during aerobic metabolism.

To determine if lactate is produced during aerobic metabolism in peripheral nerve, we incubated pieces of rabbit vagus nerve in oxygenated solution containing D-[U-14C]glucose while stimulating electrically. After 30 min, nearly all the radioactivity in metabolites in the nerve was in lactate, glucose 6-phosphate, glutamate, and aspartate. Much lactate was released to the bath: 8.2 pmol (microg dry wt)(-1) from the exogenous glucose and 14.2 pmol (microg dry wt)(-1) from endogenous substrates. Lactate release was not increased when bath PO2 was decreased, indicating that it did not come from anoxic tissue. When the bath contained [U-14C]lactate at a total concentration of 2.13 mM and 1 mM glucose, 14C was incorporated in CO2 and glutamate. The initial rate of formation of CO2 from bath lactate was more rapid than its formation from bath glucose. The results are most readily explained by the hypothesis that has been proposed for brain tissue in which glial cells supply lactate to neurons.

Glia-axon interactions and the regulation of the extracellular K+ in the peripheral nerve.

Changes in membrane potential of both axons and Schwann cells were measured simultaneously during electrical activity and during the period of recovery in the rabbit vagus nerve by the use of the sucrose-gap apparatus. During low-frequency stimulation (0.5-1 Hz) the preparation developed a ouabain-sensitive hyperpolarization. This hyperpolarization increased when the inwardly rectifying K+ channels in Schwann cells were blocked with Ba2+, indicating that the hyperpolarization was generated by the electrogenic glial Na(+)-K+ pump. During trains at higher frequencies (15 Hz), the preparation depolarized, but after cessation of the stimulation it developed a posttetanic hyperpolarization (PTH). The PTH was also ouabain-sensitive and was strongly enhanced by Cs+ which is known to block the hyperpolarization-activated inward current (Ih) in axons but not in glial cells. These results show that the PTH reflects mainly the axonal electrogenic pump. Our results indicate that during activity the K+ released from the firing axons is removed from the extracellular space by Schwann cells and that after cessation of the stimulation the K+ surplus returns from Schwann cells back to axons. Both the glial and axonal K+ uptake is mediated by successive activation of the glial and axonal Na(+)-K+ pump. The nature of the signalling mechanisms that control the pumping rates of the respective pumps remain unknown.

Uptake of potassium by nonmyelinating Schwann cells induced by axonal activity.

1. The electrophysiological properties of the rabbit vagus nerve (membrane potential, compound action potentials, and afterpotentials) and potassium accumulation were measured simultaneously during low-frequency stimulation (LFS) (0.5 and 1 Hz) by using a modified sucrose-gap apparatus and potassium-sensitive microelectrodes (KSM). 2. During LFS at 0.5 and 1 Hz, the concentration of K+ in the extracellular space ([K+]c) increased in approximately 30 s to a maximal level that was 0.6 and 1.5 mM, respectively, above the resting concentration. Concomitantly the preparation developed an ouabain-sensitive hyperpolarization. 3. The compound action potential (CAP) was followed by a fast hyperpolarizing afterpotential (fHAP), a depolarizing afterpotential (DAP), and a slow hyperpolarizing afterpotential (sHAP). During LFS the characteristics of all these afterpotentials were profoundly modified. In parallel to the increase in [K+]e, the fHAP was decreased and the amplitude of the DAP was dramatically enhanced. Furthermore, the sHAP which had a duration of < 1 s when it followed a single CAP, turned into a ouabain-sensitive hyperpolarization (indicating that it was generated by the electrogenic Na(+)-K+ pump) that lasted several minutes. 4. The application of external Ba2+ produced a hyperpolarizing sag on the sHAP following a single isolated CAP. During LFS, Ba2+ enhanced the build-up of the DAP, raised the maximal level of [K+]e, and increased the activity-induced ouabain-sensitive hyperpolarization. 5. The increase by Ba2+ of the activity-induced hyperpolarization shifted the spikes from both myelinated and nonmyelinated fibers toward a more negative potential but did not increase their amplitude, indicating that this Ba(2+)-induced hyperpolarization originated from an extra-axonal source, presumably the Schwann cells. 6. It is proposed that the electrogenic activity of the Na(+)-K+ pump was enhanced in Schwann cells situated near active axons. This hyperpolarization was, however, not recorded in normal conditions because it was fully short-circuited by a K+ influx through Ba(2+)-sensitive channels. 7. Our results lead to the hypothesis that the Na(+)-K+ pump of the nonmyelinating Schwann cells is important in the mechanisms maintaining the homeostasis of K+ in the axonal microenvironment. They show that the Na(+)-K+ pump contributes to the K+ buffering not only by actively pumping K+ but also by generating a hyperpolarization that drives a passive K+ influx through Ba(2+)-sensitive K+ channels.

Long-range intercellular signalling in glial cells of the peripheral nerve.

Schwann cells are considered to be electrically silent satellite cells surrounding axons, although they exhibit ionic channels, some of which are similar to those employed by axons for the generation and transmission of nerve impulses. Here, we show that Schwann cells generate, in response to a short and gentle electrical stimulus, a long-lasting depolarizing potential, slowly propagating along the Schwann cell synsitium. This electrical signal, which in situ might be generated by the Schwann cells in response to the axonal electrical activity, constitutes in the peripheral glia a novel form of long-range intercellular signalling, which may be involved in the regulation and modulation of the axonal excitability.

Hyperpolarizing afterpotentials in C fibers and local anesthetic effects of clonidine and lidocaine.

Effects of clonidine and lidocaine on the hyperpolarizing after-potential (HAP) and frequency-dependent block in C fibers were examined on desheathed rabbit vagus nerves, using the sucrose gap technique. A single action potential (AP) was followed by a fast and a slow HAP. Clonidine, at concentrations from 0.05 to 50 mumol/l, decreased the fast HAP, while the AP amplitude was unchanged. At a 500 mumol/l concentration of clonidine, the fast HAP amplitude was similar to control, the slow HAP was increased, and the AP amplitude decreased. Lidocaine at 500 mumol/l delayed and broadened the HAP, making a distinction between fast and slow HAP impossible, and decreased and delayed the AP amplitude. In the presence of lidocaine (500 mumol/l), clonidine at concentrations from 0.05 to 500 mumol/l decreased the HAP amplitude, without modifying the lidocaine-induced shape of the HAP. The modifications of the HAP, however, do not contribute to the local anesthetic effects of clonidine, as the addition of clonidine (0.5 and 500 mumol/l) to Locke or lidocaine (500 mumol/l) solution does not enhance the frequency-dependent block (3 and 10 Hz) observed with either Locke or lidocaine solution alone.

Clonidine enhances the effects of lidocaine on C-fiber action potential.

We examined local anesthetic effects of clonidine and its interaction with lidocaine with regard to tonic inhibition of the C-fiber action potential (AP) on the isolated, desheathed rabbit vagus nerve by the sucrose gap method. Clonidine and lidocaine at 500 microM concentrations caused a comparable degree of C-fiber inhibition, corresponding to an AP area under the curve of 75.8% +/- 9.4% (mean +/- SE) and 82.2% +/- 5.9% of control, respectively. Concentrations of clonidine less than 500 microM did not inhibit C-fiber AP. Clonidine, added in concentrations of 500 nM, 500 microM, and 5 mM to a 500 microM lidocaine perfusion, caused a significant decrease in fiber blockade of 18%, 20%, and 54%, respectively, as compared with clonidine added to Locke perfusion (P less than 0.05). The sodium channel blocker tetrodotoxin (3 microM) decreased the AP area to 9.3% +/- 1.3% of control. The remaining tetrodotoxin-resistant AP was almost completely blocked by clonidine (500 microM) and lidocaine (500 microM), indicating a higher susceptibility of tetrodotoxin-resistant fibers to the two drugs than the C-fiber population as a whole. The enhancing effect of a low dose of clonidine (500 nM) on lidocaine-induced (500 microM) inhibition of C-fiber AP might explain the clinical observation that clonidine, at approximately 1000-fold lower concentrations than lidocaine, prolongs the action of lidocaine in peripheral nerve block.

After potentials in nonmyelinated nerve fibers.

1. The sucrose-gap technique was employed to examine the different types of after potentials that follow, in desheathed rabbit vagus nerves, a single action potential (AP) elicited by a short (0.4 ms) supramaximal depolarizing pulse. 2. A fast and a slow hyperpolarizing after potential (fHAP and sHAP) as well as a depolarizing after potential (DAP) followed a single spike. Both the fHAP and the sHAP showed a dependence on the K+ electrochemical gradient, indicating that they are due to an outwardly oriented current of K+ ions. 3. The fHAP was sensitive to low concentrations of tetraethylammonium (TEA; 1 mM) and 4-aminopyridine (4-AP; 10 microM) and to millimolar concentrations of Ba2+. We conclude that the fHAP reflects the tail of the delayed rectifier K+ current. 4. The sHAP contained a Ca(2+)-sensitive component that showed a requirement for voltage-dependent Ca2+ entry during the AP. This component was completely blocked by low concentration of TEA (1 mM) and by Cd2+ (1 mM), but unaffected by 4-AP. These observations suggest that it reflects a current flowing through Ca(2+)-activated K+ channels. The remaining, apparently Ca(2+)-insensitive, component was insensitive to 4-AP and could be blocked by TEA only at concentrations greater than 50 mM. 5. The DAP usually appeared when the external concentration of K+ was increased to above approximately 8 mM, but sometimes it was clearly visible even at lower [K+]o. The DAP was TEA insensitive and entirely Ca2+ dependent. This latter property is inconsistent with the widely accepted hypothesis according to which the DAP reflect the accumulation of K+ in the extracellular space during the AP. 6. The origins of both the Ca(2+)-insensitive component of the sHAP and the DAP are not clear. However, in view of the fact that the sucrose-gap technique records not only the membrane potential of the nerve fibers but also of the surrounding glia, there is the possibility that these after potentials reflect changes in the electrical properties of the satellite Schwann cells.

Calcium efflux and intracellular exchangeable calcium in mammalian nonmyelinated nerve fibers.

Calcium efflux was measured in desheathed rabbit vagus nerves loaded with 45Ca2+. The effects of extracellular calcium, sodium, phosphate, potassium and lanthanum ions on the calcium efflux were investigated and the distribution of intracellular calcium determined by kinetic analysis of 45Ca2+ efflux profiles. The 45Ca2+ desaturation curve can be adequately described by three exponential terms. The rate constant of the first component (0.2 min-1) corresponds to an efflux from an extracellular compartment. The two slow components had rate constants of 0.03 and 0.08 min-1 and represent the efflux from two intracellular pools. The amounts of exchangeable calcium in these two pools, after a loading period of 150 min, were 0.170 and 0.102 mmol/kg wet weight, respectively. The total calcium efflux in physiological conditions amounted to about 24 fmol cm-2 sec-1. The magnitude of the two intracellular compartments as well as the total calcium efflux were markedly affected by extracellular phosphate, sodium and lanthanum, whereas the corresponding rate constants remained almost unchanged. Phosphate reversed the effect of sodium withdrawal on the calcium efflux: in the absence of phosphate, sodium withdrawal increased the calcium efflux to 224%, but in the presence of phosphate, sodium withdrawal decreased calcium efflux to 44%. Phosphate also affected the increase in calcium efflux produced by inhibitors of mitochondrial calcium uptake, suggesting that two different mitochondrial pools contribute to the control and regulation of intracellular calcium and of the transmembrane calcium transport.

Continuous measurement of calcium influx in mammalian nonmyelinated nerve fibers: effects of Nao, Cao, and electrical activity.

A new technique for continuous monitoring of the cellular calcium was developed and used for studying the effects of external and internal Na (Nao and Nai), external Ca (Cao), Ca ionophore A23187, and electrical activity on membrane-bound and intracellular Ca in mammalian nonmyelinated nerve fibers. Increasing Cao increased both the membrane-bound and the intracellular Ca. Lowering Nao increased the membrane-bound fraction of Ca indicating that lack of Nao enhanced the capacity of the plasma membrane to bind Ca, and produced an increase of the internal Ca pool. Increasing Nai by treatment with ouabain enhanced the Ca inflow in both, the presence and absence of Nao, presumably by stimulating the Cao/Nai exchange. The Ca ionophore A23187 produced a large and irreversible increase in the intracellular Ca without affecting the membrane-bound fraction. On the other hand, electrical activity, which is known to produce a large increase of the total Ca in squid axon, had no measurable effect on the total calcium content in our preparation. It is concluded that in mammalian nerve fibers a Ca load by exposition to Na-free solution or to A23187 produces an accumulation of Ca into the intracellular Ca stores, whereas during electrical activity the membrane-associated extrusion mechanisms are able to maintain the intracellular Ca2+ below the threshold for intracellular sequestration. Furthermore, the results indicate that the intracellular sequestration mechanisms are dependent on the internal concentration of Na.

Involvement of intracellular calcium in the phosphate efflux from mammalian nonmyelinated nerve fibers.

Phosphate efflux was measured as the fractional rate of loss of radioactivity from desheathed rabbit vagus nerves after loading with radiophosphate . The effects of strategies designed to increase intracellular calcium were investigated. At the same time, the exchangeable calcium content was measured using 45Ca. Application of calcium ionophore A23187 increased phosphate efflux in the presence of external calcium in parallel with an increase in calcium content. In the absence of external calcium, there was only a late, small increase in phosphate efflux. For nerves already treated with the calcium ionophore, the phosphate efflux was sensitive to small changes in external calcium, in the range 0.2 to 2 mM calcium, whereas similar increases in calcium in absence of ionophore gave much smaller increases in phosphate efflux. Removal of external sodium (choline substitution) produced an initial increase in phosphate efflux followed by a fall. The initial increase in phosphate efflux was much larger in the presence of calcium, than in its absence. The difference was again paralleled by an increase in calcium content of the preparation, thought to be due to inhibition of Na/Ca exchange by removal of external sodium. Measurements of ATP content and ATP, ADP, phosphate and creatine phosphate ratios did not indicate significant metabolic changes when the calcium content was increased. Stimulation of phosphate efflux by an increase in intracellular calcium may be due to stimulation of phospholipid metabolism. Alternatively, it is suggested that stimulation of phosphate efflux is associated with the stimulation of calcium efflux, possibly by cotransport of calcium and phosphate.

Effects of calcium and lanthanum on phosphate efflux from nonmyelinated nerve fibers.

Phosphate efflux was measured as the fractional rate of loss of radioactivity from rabbit vagus loaded with radiophosphate. The effects of changes in extracellular calcium and of lanthanum have been investigated. In Locke solution with normal, 0.9 mM, calcium and without phosphate, the fractional rate of loss was 1.62 X 10(-3) min-1 at 120 min after the beginning of the washing period and fell slowly (9% hr-1) during washing from 2 to 6 hr. Addition of calcium to the Locke solution produced a transient increase followed by a reversible maintained increase in phosphate efflux. The latter was 40 and 75% above efflux in normal calcium for 20 and 50 mM calcium, respectively. Removal of calcium, with or without addition of EGTA, produced only a transient increase in phosphate efflux, with no subsequent maintained change. Addition of low concentrations of lanthanum produced a reversible inhibition of phosphate efflux. Half-maximal inhibition was at 3.5 micro M lanthanum and appeared to be due to binding of lanthanum to more than one, probably two, sites. Measurements of inhibition by lanthanum at different calcium concentrations did not indicate any competition between calcium and lanthanum. It is suggested that a least a part of phosphate efflux depends on internal calcium and that lanthanum acts by preventing release of phosphate from the phosphate transport mechanism.

The correction factors for sucrose gap measurements and their practical applications.

The distribution of extracellular and intracellular potential in the sucrose gap apparatus, previously established for a single fiber using the cable equations for a core conductor model (Jirounek and Straub, Biophys. J., 11:1, 1971), is obtained for a multifiber preparation. The exact equation is derived relating the true membrane potential change to the measured potential differences across the sucrose gap, the junction potentials between sucrose and physiological solution, the membrane potential in the sucrose region, and the electrical parameters of the preparation in each region of the sucrose gap. The extracellular potential distribution has been measured using a modified sucrose gap apparatus for the frog sciatic nerve and the rabbit vagus nerve. The results indicate a hyperpolarization of the preparations in the sucrose region, of 60--75 mV. The hyperpolarization is independent of the presence of junction potentials. The calculation of the correction terms in the equation relating the actual to the measured potential change is illustrated for the case of complete depolarization by KC1 on one side of the sucrose gap. The correction terms in the equation are given for various experimental conditions, and a number of nomographic charts are presented, by means of which the correction factors can be rapidly evaluated.

Efflux of inorganic phosphate from mammalian non-myelinated nerve fibres.

1. Uptake and release of radiophosphate were measured in desheathed rabbit vagus nerve. 2. During incubation in Locke containing 0.2 mM-[32P]phosphate a slow labelling of water soluble compounds of the nerve was found; the labelling of the non water soluble compounds was much smaller. During washing with inactive Locke, the label was almost entirely released from the water soluble compounds; the radioactivity of these compounds was therefore used as the basis for the calculation of the efflux rate constants. 3. The efflux of radiophosphate increased with increasing phosphate concentration of the washing fluid. 4. A similar effect of external phosphate on the efflux of radiophosphate was seen when the phosphate concentration was suddenly changed. The rate constants were in 0.2 mM-phosphate 1.29 X 10(-3) min-1, in 0.2 mM 1.95 X 10(-3) min-1, and in 2 mM 3.21 X 10(-3) min-1 at 37 degrees C. After changing the external solution the efflux reached a new level with a time constant of about 9 min. 5. Addition of arsenate also increased the efflux of radiophosphate; on a molar basis the effect of arsenate was slightly smaller than the effect of phosphate. 6. Addition of malate or malonate did not affect the efflux of radiophosphate. 7. When the Na of the Locke was replaced by Tris, the efflux of radiophosphate was lowered by 76%, the new level was reached with a time constant of 7.7 min. In Tris-Locke changes in external phosphate did not affect the phosphate efflux. 8. A lowering of the phosphate efflux by 52% was found in Li-Locke; the efflux was then no longer affected by the external phosphate concentration. 9. The effect of external phosphate on the efflux and in the radiophosphate was also measured at intermediate Na concentrations. At different Na concentrations the ratio between the Na dependent effluxes in 2 and 0.2 mM-phosphate was approximately equal to the ratio between the Na dependent influxes in 2 and 0.2 mM-phosphate. 10. The efflux of 22Na had a rate constant of 0.050 min-1 in Locke and 0.046 min-1 in Tris-Locke. 11. It is concluded that part of the efflux of phosphate is mediated by a Na-dependent transport system; the system appears to be able to exchange phosphate between the inside and the outside and to mediate net movements of phosphate in both directions.