Brain Research, 591 (1992) 351-355
351
© 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
BRES 25369
Octopamine selectively modifies the slow component of sensory adaptation in an insect mechanoreceptor B.G. Zhang, P.H. T o r k k e l i a n d A.S. F r e n c h Department of Physiology, Universityof Alberta, Edmonton, Alta. (Canada) (Accepted 23 June 1992)
Key words: Mechanoreceptor; Neuromodulation; Biogenic amine; Sensory adaptation; Cockroach
The effects of octopamine were studied on the dynamic behavior of the sensory neuron in the cockroach femoral tactile spine. The neuron is a rapidly adapting mechanoreceptor in which adaptation occurs by elevation of the threshold for action potential encoding. The threshold follows increases or decreases of membrane potential, with a delay that involves two separate exponential components. Previous evidence has associated the slow component with sodium pumping and the fast component with sodium channel inactivation. Octopamine reversibly raised the resting threshold and increased but slowed the slow component. These data indicate that octopamine has specific effects on membrane-ionic processes in insect sensory neurons.
Octopamine is a biogenic amine which has a neuromodulatory effect on a range of nervous system functions. It is a principal component of the known amines in the cockroach brain, which also include tyramine, dopamine, 5-HT and noradrenaline 2°. The most clearly established action of octopamine is to elevate cAMP levels by activating adenylate cyclase 8''~. At least two types of receptors for octopamine have been characterized in insect tissues I'~' and one of these has been found to be a G-protein coupled activator of adenylate cyclase 9. Several neuromodulatory effects of octopamine have been observed. In Limulus, octopamine was released synaptically onto photoreceptors and increased their rate of dark adaptation, probably by lowering intracellular calcium levels t2. In other systems octopamine appeared to act on ionic conductances in membranes. It produced active bursting and plateau potentials in locust interneurons and motoneurons ~7'1s, increased phasic firing in a locust mechanoreceptor m6, reduced ex¢itability in a lobster mechanoreceptor ~3 and increased synaptic efficiency in the lobster stomatogastric ganglion1°. However, the detailed ionic mechanisms underlying these effects are still ut~known. The cockroach femoral tactile spine is a mechanore-
ceptor that contains a single sensory neuron. The soma of the neuron is inside the spine lumen, close to the sensory dendrite 7. The neuron adapts rapidly to a sustained mechanical stimulus, so that the response to a step deformation is a burst of action potentials, decaying to silence after about 1 s or less. This rapid adaptation occurs during the encoding of the receptor current into action potentials and can be demonstrated without mechanical stimulation by both extracellular 5 and intracellular 2 electric current stimulation. Rapid adaptation in the tactile spine neuron is caused by elevation of the threshold for action potential production 5. The threshold is labile, increasing with depolarization and decreasing with hyperpolarization. Both increases and decreases follow a similar, double-exponential, time course, which is probably due to two distinct components. The two components can be distinguished by their different sensitivities to membrane potential 5, by systems analysis of action potential encoding 21, or by the selective actions of certain chemical agents 4'6'~5. The two components make approximately equal contributions to the total changes in threshold level and have time constants of approximately 100 ms and approximately 1000 ms, although both time constants vary with membrane potential 5.
Correspondence: A.S. French, Department of Physiology, University of Alberta, Edmonton, Alta., Canada T6G 2H7. Fax: (1) (403) 49-8915.
352
Stilulatin9current Tactilespine
Current monitor
Threshold Sttaulatin9current 30
iariabletime~ ~ Threshold nA -30 nA ~-[~measurement
Fig. 1. Measuring the decay in threshold following a step decrease in current stimulus. Upper paneh a glass microelectrode was lowered through the cut end of the spine to approach the sensory neuron. Current passing through the electrode produced action potentials th',tt were detected by differential extracellular electrodes in the lemur. Current was measured by a virtual ground monitor. Lower panel: a step decrease in current caused the threshold to decay to a lower keel. The decay was measured as a function of time after the step.
The fast component has been associated with sodium channels. Mild oxidizing agents which selectively reduce sodium channel inactivation in several systems can selectively eliminate the fast component 2'4. Also, phentolamine, which has local anesthetic properties via sodium channel blockade, can reversibly reduce the fast component is. Less is known about the slow component of threshold change. Ouabain caused a selective reduction in this component", suggesting that sodium pumping is involved. However, intraceilular measurements after bursts of action potentials gave only small afterhyperpolarizations with time constants of several seconds 2. Although normal adaptation involves a depolarization-induced increase in threshold, it is easier to measure the dynamic properties of the threshold during decreases in threshold, because no action potentials occur during the decrease. This approach was used here to observe the effects of octopamine on the two components of adaptation. Adult cockroaches, Periplanet. americana, of either sex were taken from a laboratory colony. The preparation and the procedures for stimulating the sensory neuron and measuring the threshold have been described before 5.6 and are illustrated in Fig. 1. The metathoracic leg was mounted in a plexiglass holder with the cut end of the tibia immersed in saline connected to an Ag/AgCi electrode. The spine was amputated above the sensory neuron and a glass microelec-
trode (1 M NaCI, 5-15 M n ) was advanced through the cut end into the spine lumen. The microelectrode was driven by a constant voltage-to-current convertor. Its tip was advanced to the position of maximum sensitivity to a negative current pulse, which is believed to be close to the site of normal action potential initiation. In this configuration, negative voltages applied to the electrode cause current to flow into the tip of the electrode. A small portion of this current flows outwards through the membrane adjacent to the electrode tip, depolarizing the membrane. Such depolarizing currents are shown as positive currents here. Fig. 1 also shows the protocol for measuring the change in threshold current following a step decrease in membrane polarization. A steady depolarizing current of 30 nA was applied for a period of 5 s to allow the threshold to settle to a stable, elevated, level. Then the current was stepped to - 3 0 nA. After a variable period, the threshold for producing an action potential was measured by computer-controlled successive approximation, using a current pulse of 5 ms and allowing a further period of 5 ms after the pulse for action potential conduction to the extracellular recording electrodes. The successive approximation procedure repeated the stimulus protocol seven times and halved the change in test pulse amplitude on each cycle to give a final resolution of 0.5 nA. The sign of each amplitude change was determined by the occurrence or absence of an action potential. Following a step decrease in stimulating current, the decrease in rheobasic threshold current could be fitted by a sum of two exponential decays, as reported previously 5. Fig, 2 shows typical thre~hold measurements following steps and the best fitting double exponential decay: i(t) = A -
B(I - e - , / v ) _
C(I - e - t / v )
(Eqn. l)
where l(t) is the threshold current at time t after the step, A is the static threshold current, B and C are the amplitudes of the two components of threshold decay and U and V are their respective time constants. Fitting was carried out by a non-linear, minimum square error method t4, that always converged rapidly. The fitting program produced values of the five parameters in Eqn. 1. Before octopamine application, the data (open squares) were fitted by: A = 36.0 nA, B - 17.8 nA, U - 8 1 ms, C - 17.1 nA and V - 3 6 7 ms. These are similar to values obtained before from normal preparations 5. After establishing a stable preparation, the bath solution surrounding the cut end of the spine was repiaced by 1 mM D,L-OCtopamine (Sigma) in cock-
353 roach saline, although the concentration inside the spine lumen was probably significantly lower. Octopamine levels are much lower in normal insect tissues, but the tactile spine neuron and other insect neurons are surrounded by a relatively impermeable gliai barrier 3. Bath octopamine levels of up to 100 mM were necessary to affect the behavior of locust flight interneurons t8 and levels up to 1 mM were applied to locust wing stretch receptors ~6. Mcasmc~nent of a complete threshold decay curve required a[:,proximately 8 rain, so data were taken at 15-rain intervals. The solid circles in Fig. 2 show the threshold 15 rain after octopamine application. The threshold was elevated ( A - 110.2 nA), both of the decay components were increased (B--42.5 hA, C = 36.3 hA), and both of the time constants were increased (U = 127 ms, V = 1149 ms). Replacement of the octopamine by normal saline, with several washes, reversed the effect (open circles). The experiment was performed on 17 different preparations and an increase in threshold was seen in 16 of these. However, the time required for the octopamine effect to appear varied from 15 to 90 rain and the time required for recovery after washing varied from 15 to 75 rain. It was difficult to keep the preparation stable for the long times required by the experiments and especially difficult to prevent small move-
4O
ISO1 ,
,-,
.~ 100
.
~
,r4
0
0~,
o
"T
5o0
[]
•
o
--
loo0
1500
Time after step (ms) Fig. 2. Threshold decay curves at the start of an experiment (open squares), 15 rain after octopamine application (filled circles), and |5 rain after washing with cockroach saline (open circles). The octopamine-induced increase in threshold current was reversed by washing. The solid lines show the fitted relationships of Eqn. 1. Inset: mean values and S.E.M. of the threshold current at 1500 ms for the five different experiments.
3000
1oo
o
Parameters"
200
*
A
B
o
C
o
U
V
Fig. 3. Bar graphs of the mean values and standard errors for the five parameters of Eqn. 1 before and after (shaded) the application of octopamine. Asterisks indicate significantly different values in a paired t-test.
ments of the electrode during solution changes. From the 17 experiments we chose five for statistical analysis. In each of these the removal of octopamine produced a complete return to the baseline condition as shown in Fig. 2, so that the position of the electrode and the condition of the neuron were presumed to be constant throughout the entire experiment. The mean and standard error values of threshold current at 1500 ms after the step are shown as inset bar graphs in Fig. 2 with the same symbols as the main graph and the octopamine data shaded. The mean value increased from 9.8 nA to 40.5 nA after octopamine application and returned to 9.2 nA after washing. The data after oetopamine treatment were significantly different to those before and after washing (paired t-test, * P < 0.05, ** P < 0.005). In the five experiments, the octopamine effect was fully developed after times of 15, 15, 15, 30 and 45 rain. Return to normal after washing took: 15, 15, 90, 30 and 45 rain, respectively. For each of the five experiments, Eqn. 1 was fitted to the threshold decay curves before and after octopamine treatment. The mean and standard error values for the five parameters in the equation are shown in Fig. 3, together with the statistical significance of differences from paired t-tests. Two major effects are apparent. The static threshold, A, increased significantly, so did both the amplitude, C, and time constant, V, of the slower component. The time constant of the fast component, U, also increased, but was so variable after octopamine treatment that the change was not statistically significant. The data indicate that octopamine raises the threshold of the tactile spine neuron and increases but slows the slow component of adaptation. The fast component
354 of adaptation, which has been associated with sodium channel properties, may also be slowed, but this was not statistically significant and its amplitude was not affected. These actions all appear to be completely reversible. At the present time, the ionic mechanisms underlying the slow component of adaptation are not well established. The only agent which has previously been shown to modify the slow component is ouabain, which is normally associated with inhibition of electrogenie sodium pumps ~'. A range of other chemicals which affect potassium channels, chloride channels, calcium channels and calcium-activated potassium channels have been applied to the tactile spine preparation without producing any modulation of the adaptation process 5. it is possible that both of the significant effects of octopamine result from modification of a single process. For example, incrca,3ing the pump's sensitivity (o intracellular sodium could raise the threshold and the amplitude of the slow component of adaptation. The elevation of the tactile spine neuron's static threshold by octopamin¢ agrees with the reduction in sensitivity seen in the lobster mechanoreceptor system ~3 but disagrees with the effects seen in central insect neurons ~z~s and in locust wing receptors ~'. However, the present results show that octopamine has at least two acdons on the tactile spine neuron, raising the static threshold but slowing its rate of change with membrane potential. This suggests that the apparent effects could vary with the type of stimulus applied, as well as the phasic properties of the receptor being studied. The lobster mechanorcceptors are relatively slowly adapting and were stimulated with step mov¢. ments and currents of about 500 ms duration. Octopamine would be expected to produce a lower rate of firing and even slower adaptation. The locust stretch receptors are more rapidly adapting and were stimulated with 4 Hz sinusoids to which they normally produced only ~,ne spike per cycle. The extra spikes produccd by octopaminc might be expected if the primary effect of octopamine in this situation was to reduce adaptation. It is difficult to compare octopamine actions on neurons containing disparate groups of ion channels, pumps and synaptic inputs. It is also difficult to know if the actions of octopamine on the tactile spine neuron arc physiologically important, There is no evidence for efferent neurons innervating the tactile spine lumen, although circulat;ng octopamine in the hemolymph might reach the neuron. However, the present results establish a useful model system on which to base future investigations of octopamine and other important new romodulators. It is now possible to obtain reliable
intracellular recordings from the tactile spine neuron 2 and we are currently developing more direct methods for measuring the currents involved in rapid sensory adaptation. The relatively strong actions of octopamine on the tactile spine neuron should allow us to explore the ionic mechanisms underlying these actions in the future. Support for this work was provided by the Medical Research Council of Canada and the Alberta Heritage Foundation for Medical Research. 1 Arakawa, S., Gocayne, J.D., McCombie, W.R., Urquhart, D.A., Hall, L.M., Fraser, C.M. and Venter J.C., Cloning, localization and permanent expression of a Drosophila octopamine receptor, Neuron, 4 (1990) 343-354. 2 Basarsky, T.A. and French, A.S., Intracellular measurements from a rapidly adapting sensory neuron, J. Neurophysiol., 65 (1991) 49-56. 3 Bernard, J., Guiliet, J.C., Collier, J.P., Evidence for a barrier between blood and sensory terminal in an insect mechanoreceptot, Comp. Biochem. Physiol., A67 (1980) 573-579. 4 French, A.S., Removal of rapid sensory adaptation from an insect mechanoreceptor neuron by oxidizing agents which affect sodium channel inactivation, J. Comp. Physiol. Ai61 (1987) 275-282. 5 French, A.S., Two components of rapid sensory adaptation in a cockroach mechanoreceptor neuron, J. Neurophysiol., 62 (1989) 768-777. 6 French, A.S., Ouabain selectively affects the slow component of sensory adaptation in an insect mechanoreceptor, Brain Res., 504 (1989) 112-114. 7 French, A.S. and Sanders, EJ., The mechanosensory apparatus of the femoral tactile spine of the cockroach, Periplaneta ameri. cana, Cell T/ss. Res., 219 (1981) 53-68.
8 Groome, J.R. and Watson, W.FI., Second messenger systems underlying amine and peptide actions on cardiac muscle in the horseshoe crab Limuh~s polyphemus, J. F:2rp. Biol., 145 (1989) 41q-437, 9 Guillen, A., Ilaro, A. and Municio, A.M., A possible new class of oc:topamine receptors coupled to adenylat¢ cyclase in the brain of tile dipterous Ceratitis cal~itata. Pharmacological characterization and regulation of "~H-octopamine binding, Lift, Sci,, 45 (1989) 655 -662. 10 Johnson, B.R. and Harris-Warrick, R,M,, Aminergic modulation of graded synaptic transmission in the lobster stomatogastric ganglion, J. Neurosci., 10 (1990) 2066-2076.
!! Nathanson, A. and Greengard, P., Octopamine sensitive adenylate cyclase. Evidence for the biological role of octopamine in nervous tissue, Scienct; 180 (1973) 398-430. 12 O'Day, P.M. and Lisman, J.E., Octopamine enhances dark-adap. tatioll ill Lhmdles ventral photoreceptors, J. Neurosci., 5 (1985) 1490-1496. 13 Pasztor,V.M. and Bush, B.M.H., Primaryafferent responsesof a crustacean mechanoreceptor are modulated by proctolin, octopamine and serotonin, J. Nearobiol., 20 (1989)234-254. 14 Press, W,H., Flannery~ B.P., Teukolsky, S.A. and Vetterling, W.T.. in Numerical Recipes in C The Art of Scientific Computing, Cambridge University Press, Cambridge, UK, 1988, pp. 540-547. 15 Ramil'ez,J..M. and French, A.S., Phentolamineselectivelyaffects the fast sodium compo~ient of sensory adaptation in an insect mechanoreceptor, J. Neurobioi., 21 (1990)893-899. 15 Ram,rez, J.-M. and Orchard, l., Octopaminergic modulation of the lorewingstretch receptor in the locust, Locusta migratoria, J. Exp Biol., 149 (1990) 255-279. 17 Rarairez, J.-M. and Pearson, K.G., Octopamine induces bursting -* anc~plateau potentials in insect neurones, Brain Res., 549 (1991) 332-337. 18 Ramirez, J.-M. and Pearson, K.G., Octopaminergic modulations
355 of interneurons in the flight system of the locust, J. Neurophysiol., 66 (1991) 1522-1537. 19 Roeder, T. and Gewecke, M., Octopamine receptors in locust nervous tissue, Biochem. Pharmacol., 39 (1990) 1793-1797. 20 Shaft, N., Midgley, J.M., Watson, D.G., Smail, G.A., Strang, R. and MacFarlane, R.G., Analysis of biogenic amines in the brain
of the American Cockroach (Periplaneta americana) by gas chromatography-negative ion chemical ionisation mass spectrometry, J. Chromatogr., 490 (1989) 9-19. 21 Stockbridge, L.L., Torkkeli, P.H. and French, A.S., IntraceUular non-linear frequency response measurements in the cockroach tactile spine neuron, Biol. Cybern., 65 (1991) 181-187.
351
© 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
BRES 25369
Octopamine selectively modifies the slow component of sensory adaptation in an insect mechanoreceptor B.G. Zhang, P.H. T o r k k e l i a n d A.S. F r e n c h Department of Physiology, Universityof Alberta, Edmonton, Alta. (Canada) (Accepted 23 June 1992)
Key words: Mechanoreceptor; Neuromodulation; Biogenic amine; Sensory adaptation; Cockroach
The effects of octopamine were studied on the dynamic behavior of the sensory neuron in the cockroach femoral tactile spine. The neuron is a rapidly adapting mechanoreceptor in which adaptation occurs by elevation of the threshold for action potential encoding. The threshold follows increases or decreases of membrane potential, with a delay that involves two separate exponential components. Previous evidence has associated the slow component with sodium pumping and the fast component with sodium channel inactivation. Octopamine reversibly raised the resting threshold and increased but slowed the slow component. These data indicate that octopamine has specific effects on membrane-ionic processes in insect sensory neurons.
Octopamine is a biogenic amine which has a neuromodulatory effect on a range of nervous system functions. It is a principal component of the known amines in the cockroach brain, which also include tyramine, dopamine, 5-HT and noradrenaline 2°. The most clearly established action of octopamine is to elevate cAMP levels by activating adenylate cyclase 8''~. At least two types of receptors for octopamine have been characterized in insect tissues I'~' and one of these has been found to be a G-protein coupled activator of adenylate cyclase 9. Several neuromodulatory effects of octopamine have been observed. In Limulus, octopamine was released synaptically onto photoreceptors and increased their rate of dark adaptation, probably by lowering intracellular calcium levels t2. In other systems octopamine appeared to act on ionic conductances in membranes. It produced active bursting and plateau potentials in locust interneurons and motoneurons ~7'1s, increased phasic firing in a locust mechanoreceptor m6, reduced ex¢itability in a lobster mechanoreceptor ~3 and increased synaptic efficiency in the lobster stomatogastric ganglion1°. However, the detailed ionic mechanisms underlying these effects are still ut~known. The cockroach femoral tactile spine is a mechanore-
ceptor that contains a single sensory neuron. The soma of the neuron is inside the spine lumen, close to the sensory dendrite 7. The neuron adapts rapidly to a sustained mechanical stimulus, so that the response to a step deformation is a burst of action potentials, decaying to silence after about 1 s or less. This rapid adaptation occurs during the encoding of the receptor current into action potentials and can be demonstrated without mechanical stimulation by both extracellular 5 and intracellular 2 electric current stimulation. Rapid adaptation in the tactile spine neuron is caused by elevation of the threshold for action potential production 5. The threshold is labile, increasing with depolarization and decreasing with hyperpolarization. Both increases and decreases follow a similar, double-exponential, time course, which is probably due to two distinct components. The two components can be distinguished by their different sensitivities to membrane potential 5, by systems analysis of action potential encoding 21, or by the selective actions of certain chemical agents 4'6'~5. The two components make approximately equal contributions to the total changes in threshold level and have time constants of approximately 100 ms and approximately 1000 ms, although both time constants vary with membrane potential 5.
Correspondence: A.S. French, Department of Physiology, University of Alberta, Edmonton, Alta., Canada T6G 2H7. Fax: (1) (403) 49-8915.
352
Stilulatin9current Tactilespine
Current monitor
Threshold Sttaulatin9current 30
iariabletime~ ~ Threshold nA -30 nA ~-[~measurement
Fig. 1. Measuring the decay in threshold following a step decrease in current stimulus. Upper paneh a glass microelectrode was lowered through the cut end of the spine to approach the sensory neuron. Current passing through the electrode produced action potentials th',tt were detected by differential extracellular electrodes in the lemur. Current was measured by a virtual ground monitor. Lower panel: a step decrease in current caused the threshold to decay to a lower keel. The decay was measured as a function of time after the step.
The fast component has been associated with sodium channels. Mild oxidizing agents which selectively reduce sodium channel inactivation in several systems can selectively eliminate the fast component 2'4. Also, phentolamine, which has local anesthetic properties via sodium channel blockade, can reversibly reduce the fast component is. Less is known about the slow component of threshold change. Ouabain caused a selective reduction in this component", suggesting that sodium pumping is involved. However, intraceilular measurements after bursts of action potentials gave only small afterhyperpolarizations with time constants of several seconds 2. Although normal adaptation involves a depolarization-induced increase in threshold, it is easier to measure the dynamic properties of the threshold during decreases in threshold, because no action potentials occur during the decrease. This approach was used here to observe the effects of octopamine on the two components of adaptation. Adult cockroaches, Periplanet. americana, of either sex were taken from a laboratory colony. The preparation and the procedures for stimulating the sensory neuron and measuring the threshold have been described before 5.6 and are illustrated in Fig. 1. The metathoracic leg was mounted in a plexiglass holder with the cut end of the tibia immersed in saline connected to an Ag/AgCi electrode. The spine was amputated above the sensory neuron and a glass microelec-
trode (1 M NaCI, 5-15 M n ) was advanced through the cut end into the spine lumen. The microelectrode was driven by a constant voltage-to-current convertor. Its tip was advanced to the position of maximum sensitivity to a negative current pulse, which is believed to be close to the site of normal action potential initiation. In this configuration, negative voltages applied to the electrode cause current to flow into the tip of the electrode. A small portion of this current flows outwards through the membrane adjacent to the electrode tip, depolarizing the membrane. Such depolarizing currents are shown as positive currents here. Fig. 1 also shows the protocol for measuring the change in threshold current following a step decrease in membrane polarization. A steady depolarizing current of 30 nA was applied for a period of 5 s to allow the threshold to settle to a stable, elevated, level. Then the current was stepped to - 3 0 nA. After a variable period, the threshold for producing an action potential was measured by computer-controlled successive approximation, using a current pulse of 5 ms and allowing a further period of 5 ms after the pulse for action potential conduction to the extracellular recording electrodes. The successive approximation procedure repeated the stimulus protocol seven times and halved the change in test pulse amplitude on each cycle to give a final resolution of 0.5 nA. The sign of each amplitude change was determined by the occurrence or absence of an action potential. Following a step decrease in stimulating current, the decrease in rheobasic threshold current could be fitted by a sum of two exponential decays, as reported previously 5. Fig, 2 shows typical thre~hold measurements following steps and the best fitting double exponential decay: i(t) = A -
B(I - e - , / v ) _
C(I - e - t / v )
(Eqn. l)
where l(t) is the threshold current at time t after the step, A is the static threshold current, B and C are the amplitudes of the two components of threshold decay and U and V are their respective time constants. Fitting was carried out by a non-linear, minimum square error method t4, that always converged rapidly. The fitting program produced values of the five parameters in Eqn. 1. Before octopamine application, the data (open squares) were fitted by: A = 36.0 nA, B - 17.8 nA, U - 8 1 ms, C - 17.1 nA and V - 3 6 7 ms. These are similar to values obtained before from normal preparations 5. After establishing a stable preparation, the bath solution surrounding the cut end of the spine was repiaced by 1 mM D,L-OCtopamine (Sigma) in cock-
353 roach saline, although the concentration inside the spine lumen was probably significantly lower. Octopamine levels are much lower in normal insect tissues, but the tactile spine neuron and other insect neurons are surrounded by a relatively impermeable gliai barrier 3. Bath octopamine levels of up to 100 mM were necessary to affect the behavior of locust flight interneurons t8 and levels up to 1 mM were applied to locust wing stretch receptors ~6. Mcasmc~nent of a complete threshold decay curve required a[:,proximately 8 rain, so data were taken at 15-rain intervals. The solid circles in Fig. 2 show the threshold 15 rain after octopamine application. The threshold was elevated ( A - 110.2 nA), both of the decay components were increased (B--42.5 hA, C = 36.3 hA), and both of the time constants were increased (U = 127 ms, V = 1149 ms). Replacement of the octopamine by normal saline, with several washes, reversed the effect (open circles). The experiment was performed on 17 different preparations and an increase in threshold was seen in 16 of these. However, the time required for the octopamine effect to appear varied from 15 to 90 rain and the time required for recovery after washing varied from 15 to 75 rain. It was difficult to keep the preparation stable for the long times required by the experiments and especially difficult to prevent small move-
4O
ISO1 ,
,-,
.~ 100
.
~
,r4
0
0~,
o
"T
5o0
[]
•
o
--
loo0
1500
Time after step (ms) Fig. 2. Threshold decay curves at the start of an experiment (open squares), 15 rain after octopamine application (filled circles), and |5 rain after washing with cockroach saline (open circles). The octopamine-induced increase in threshold current was reversed by washing. The solid lines show the fitted relationships of Eqn. 1. Inset: mean values and S.E.M. of the threshold current at 1500 ms for the five different experiments.
3000
1oo
o
Parameters"
200
*
A
B
o
C
o
U
V
Fig. 3. Bar graphs of the mean values and standard errors for the five parameters of Eqn. 1 before and after (shaded) the application of octopamine. Asterisks indicate significantly different values in a paired t-test.
ments of the electrode during solution changes. From the 17 experiments we chose five for statistical analysis. In each of these the removal of octopamine produced a complete return to the baseline condition as shown in Fig. 2, so that the position of the electrode and the condition of the neuron were presumed to be constant throughout the entire experiment. The mean and standard error values of threshold current at 1500 ms after the step are shown as inset bar graphs in Fig. 2 with the same symbols as the main graph and the octopamine data shaded. The mean value increased from 9.8 nA to 40.5 nA after octopamine application and returned to 9.2 nA after washing. The data after oetopamine treatment were significantly different to those before and after washing (paired t-test, * P < 0.05, ** P < 0.005). In the five experiments, the octopamine effect was fully developed after times of 15, 15, 15, 30 and 45 rain. Return to normal after washing took: 15, 15, 90, 30 and 45 rain, respectively. For each of the five experiments, Eqn. 1 was fitted to the threshold decay curves before and after octopamine treatment. The mean and standard error values for the five parameters in the equation are shown in Fig. 3, together with the statistical significance of differences from paired t-tests. Two major effects are apparent. The static threshold, A, increased significantly, so did both the amplitude, C, and time constant, V, of the slower component. The time constant of the fast component, U, also increased, but was so variable after octopamine treatment that the change was not statistically significant. The data indicate that octopamine raises the threshold of the tactile spine neuron and increases but slows the slow component of adaptation. The fast component
354 of adaptation, which has been associated with sodium channel properties, may also be slowed, but this was not statistically significant and its amplitude was not affected. These actions all appear to be completely reversible. At the present time, the ionic mechanisms underlying the slow component of adaptation are not well established. The only agent which has previously been shown to modify the slow component is ouabain, which is normally associated with inhibition of electrogenie sodium pumps ~'. A range of other chemicals which affect potassium channels, chloride channels, calcium channels and calcium-activated potassium channels have been applied to the tactile spine preparation without producing any modulation of the adaptation process 5. it is possible that both of the significant effects of octopamine result from modification of a single process. For example, incrca,3ing the pump's sensitivity (o intracellular sodium could raise the threshold and the amplitude of the slow component of adaptation. The elevation of the tactile spine neuron's static threshold by octopamin¢ agrees with the reduction in sensitivity seen in the lobster mechanoreceptor system ~3 but disagrees with the effects seen in central insect neurons ~z~s and in locust wing receptors ~'. However, the present results show that octopamine has at least two acdons on the tactile spine neuron, raising the static threshold but slowing its rate of change with membrane potential. This suggests that the apparent effects could vary with the type of stimulus applied, as well as the phasic properties of the receptor being studied. The lobster mechanorcceptors are relatively slowly adapting and were stimulated with step mov¢. ments and currents of about 500 ms duration. Octopamine would be expected to produce a lower rate of firing and even slower adaptation. The locust stretch receptors are more rapidly adapting and were stimulated with 4 Hz sinusoids to which they normally produced only ~,ne spike per cycle. The extra spikes produccd by octopaminc might be expected if the primary effect of octopamine in this situation was to reduce adaptation. It is difficult to compare octopamine actions on neurons containing disparate groups of ion channels, pumps and synaptic inputs. It is also difficult to know if the actions of octopamine on the tactile spine neuron arc physiologically important, There is no evidence for efferent neurons innervating the tactile spine lumen, although circulat;ng octopamine in the hemolymph might reach the neuron. However, the present results establish a useful model system on which to base future investigations of octopamine and other important new romodulators. It is now possible to obtain reliable
intracellular recordings from the tactile spine neuron 2 and we are currently developing more direct methods for measuring the currents involved in rapid sensory adaptation. The relatively strong actions of octopamine on the tactile spine neuron should allow us to explore the ionic mechanisms underlying these actions in the future. Support for this work was provided by the Medical Research Council of Canada and the Alberta Heritage Foundation for Medical Research. 1 Arakawa, S., Gocayne, J.D., McCombie, W.R., Urquhart, D.A., Hall, L.M., Fraser, C.M. and Venter J.C., Cloning, localization and permanent expression of a Drosophila octopamine receptor, Neuron, 4 (1990) 343-354. 2 Basarsky, T.A. and French, A.S., Intracellular measurements from a rapidly adapting sensory neuron, J. Neurophysiol., 65 (1991) 49-56. 3 Bernard, J., Guiliet, J.C., Collier, J.P., Evidence for a barrier between blood and sensory terminal in an insect mechanoreceptot, Comp. Biochem. Physiol., A67 (1980) 573-579. 4 French, A.S., Removal of rapid sensory adaptation from an insect mechanoreceptor neuron by oxidizing agents which affect sodium channel inactivation, J. Comp. Physiol. Ai61 (1987) 275-282. 5 French, A.S., Two components of rapid sensory adaptation in a cockroach mechanoreceptor neuron, J. Neurophysiol., 62 (1989) 768-777. 6 French, A.S., Ouabain selectively affects the slow component of sensory adaptation in an insect mechanoreceptor, Brain Res., 504 (1989) 112-114. 7 French, A.S. and Sanders, EJ., The mechanosensory apparatus of the femoral tactile spine of the cockroach, Periplaneta ameri. cana, Cell T/ss. Res., 219 (1981) 53-68.
8 Groome, J.R. and Watson, W.FI., Second messenger systems underlying amine and peptide actions on cardiac muscle in the horseshoe crab Limuh~s polyphemus, J. F:2rp. Biol., 145 (1989) 41q-437, 9 Guillen, A., Ilaro, A. and Municio, A.M., A possible new class of oc:topamine receptors coupled to adenylat¢ cyclase in the brain of tile dipterous Ceratitis cal~itata. Pharmacological characterization and regulation of "~H-octopamine binding, Lift, Sci,, 45 (1989) 655 -662. 10 Johnson, B.R. and Harris-Warrick, R,M,, Aminergic modulation of graded synaptic transmission in the lobster stomatogastric ganglion, J. Neurosci., 10 (1990) 2066-2076.
!! Nathanson, A. and Greengard, P., Octopamine sensitive adenylate cyclase. Evidence for the biological role of octopamine in nervous tissue, Scienct; 180 (1973) 398-430. 12 O'Day, P.M. and Lisman, J.E., Octopamine enhances dark-adap. tatioll ill Lhmdles ventral photoreceptors, J. Neurosci., 5 (1985) 1490-1496. 13 Pasztor,V.M. and Bush, B.M.H., Primaryafferent responsesof a crustacean mechanoreceptor are modulated by proctolin, octopamine and serotonin, J. Nearobiol., 20 (1989)234-254. 14 Press, W,H., Flannery~ B.P., Teukolsky, S.A. and Vetterling, W.T.. in Numerical Recipes in C The Art of Scientific Computing, Cambridge University Press, Cambridge, UK, 1988, pp. 540-547. 15 Ramil'ez,J..M. and French, A.S., Phentolamineselectivelyaffects the fast sodium compo~ient of sensory adaptation in an insect mechanoreceptor, J. Neurobioi., 21 (1990)893-899. 15 Ram,rez, J.-M. and Orchard, l., Octopaminergic modulation of the lorewingstretch receptor in the locust, Locusta migratoria, J. Exp Biol., 149 (1990) 255-279. 17 Rarairez, J.-M. and Pearson, K.G., Octopamine induces bursting -* anc~plateau potentials in insect neurones, Brain Res., 549 (1991) 332-337. 18 Ramirez, J.-M. and Pearson, K.G., Octopaminergic modulations
355 of interneurons in the flight system of the locust, J. Neurophysiol., 66 (1991) 1522-1537. 19 Roeder, T. and Gewecke, M., Octopamine receptors in locust nervous tissue, Biochem. Pharmacol., 39 (1990) 1793-1797. 20 Shaft, N., Midgley, J.M., Watson, D.G., Smail, G.A., Strang, R. and MacFarlane, R.G., Analysis of biogenic amines in the brain
of the American Cockroach (Periplaneta americana) by gas chromatography-negative ion chemical ionisation mass spectrometry, J. Chromatogr., 490 (1989) 9-19. 21 Stockbridge, L.L., Torkkeli, P.H. and French, A.S., IntraceUular non-linear frequency response measurements in the cockroach tactile spine neuron, Biol. Cybern., 65 (1991) 181-187.
