- Email: [email protected]
Ouabain selectively affects the slow component of sensory adaptation in an insect mechanoreceptor
Brain Research, 504 (1989) 112-114
112
Elsevier
BRES 23817
Short Communications
Ouabain selectively affects the slow component of sensory adaptation in an insect mechanoreceptor Andrew S. French Department of Physiology, Universityof Alberta, Edmon:on, Alia. (Canada) (Accepted 1 August 1989)
Key words: Sensory transduction; Mechanoreceptor; Rapid adaptation; Sensory encoding; Threshold; Sodium pump; Ouabain
Sensory activity in the cockroach tactile spine neuron adapts rapidly to a step deflection. This rapid adaptation is caused by a rise in the threshold for action potential production, which has two components with different time constants and drug sensitivities. The basis of the slow component is unknown but it is insensitive to a wide range of ion channel blockers. In the present experiments the slow component was selectively reduced by ouabain, suggesting that it is due to the activation of an electrogenic sodium pump.
Rapid adaptation of sensory activity following a step change in stimulus is a widespread phenomenon in vertebrates and invertebrates. The cause of rapid adaption is not completely understood in any preparation and vertebrate sensory endings are generally too small to allow detailed electrophysiological investigation of the phenomenon. However, arthropods have peripherally located sensory cell bodies, permitting experiments on sensory proces,-,es which are often impossible in other phyla. The femoral tactile spine of the cockroach, Periplaneta americana, is a phasic mechanoreceptor containing a single sensory neuron9. The neuron adapts rapidly to a sustained stimulus, becoming silent about 1 s or less after a step change in spine position. This rapid adaptation occurs during the encoding of action potentials from the receptor current, since there is no evidence of adaption in the receptor current 3 but strong adaption in the discharge produced by direct electrical stimulation of the neuron 4. During rapid adaptation the threshold of the sensory neuron rises to exceed any constant stimulating current, preventing continuous discharge of action potentials at any level of stimulation. This behavior can be demonstrated by a preparation in which current is applied through the tip of an extraceilular glass microelectrode to the action potential initiating region at the axo-somatic junction4. Using this preparation it is possible to quantify the threshold behavior by strength-duration experiments6 which provide estimates of the rheobasic threshold current and the membrane time constant. The dynamic behavior of the rheobasic threshold
current (hereafter called threshold current) was studied recentlys. After step changes in stimulating current, the threshold current followed a double exponential trajectory, with time constants of approximately 50 ms and 500 ms. These two components of adaptation were clearly separable by several techniques and contributed approximately equally to the total change in threshold current. The faster component was selectively reduced by drugs which remove sodium channel inactivation, supporting earlier suggestions that sodium channel inactivation is involved in the rapid adaptation of the tactile spinel However, the slow component of adaptation was insensitive to a range of agents which affect sodium or potassium channelss. Several alternative mechanisms have been suggested for producing adaptation of spike discharge in neurons ~3 including the activation of electrogenic membrane pumps. The present experiments were therefore designed to test the hypothesis that an electrogenic sodium pump is involved in the adaptation of the tactile spine neuron. The preparation and experimental techniques used were identical to those described previouslys and are illustrated in Fig. 1. The tactile spine was amputated above the sensory neuron and a glass stimulating microelectrode (0.5 M NaCl, -- 5 M~9) was advanced through the spine lumen to the axo-somatic junction of the sensory neuron in the base of the spine. The microelectrode tip was moved to the point of maximum sensitivity, which is believed to be the normal site of action potential initiation. Action potentials were detected by extraceilular electrodes on
Correspondence: A.S. French, Department of Physiology, University of Alberta, Edmonton, AIta. T6G 2H7, Canada. 0006-8993/89/$03.50 © 19~9 Elsevier Science Publishers B.V. (Biomedical Division)
113 the sensory nerve in the femur. A steady depolarizing current of 30 nA was applied through the electrode for at least 2 s to allow threshold stabilization, and then the current was stepped to a lower value. The threshold was tested by a 20 ms depolarizing current pulse at various intervals after the step. An automated successive approximation technique was used to measure the threshold with a resolution of 0.2 hA. Ouabain (Sigma) was dissolved in cockroach saline' before being applied to the neuron via the cut end of the spine. The maximum effect of ouabain was reached about 15 min after its application, so a period of 30 min was allowed between applying ouabain and making the threshold measurements. Fig. 1 shows examples of changes in threshold current after steps from 30 nA to -30 nA. The absolute relationship between stimulating current and membrane potential is unknown but has been estimated to be 0.75 mV/nA for small currents 6 giving an approximate range of +20 mV for the membrane potential step. Before ouabain treatment the threshold current decayed with two components having time constants of 37.4 ms and 525 ms and amplitudes of 20.8 nA and 25.5 nA respectively (fitted line m normal). After treatment with 0.1 mM ouabain for 30 min the slow component was reduced to 13.5 nA while the fast component was almost unchanged at 22.9 nA (fitted line - - ouabain). Similar experiments to those of Fig. I were performed on 25 different preparations. The stimulating current was always initially 30 nA and it was then stepped to a range of values from 20 nA to -30 hA, giving total step decreases of 10 nA to 60 hA. For each step, the threshold was measured at a series of times after the step up to 1500
,~ 8 0 "
S ~ ~ M/,/~ / L.13e.A I__
Threshold
Normal C>
0
Ouabatn (O,l mH) 17, s 0 tO
[]
30
20-
E o o
10LL
0 0
1'8 2'8 3'8 4'0 5'8 60 Step decrease tn hotdtng current (hA)
Fig. 2. The amplitude of the fast component of threshold change as a function of step amplitude for normal and ouabain-treated preparations. Dat~.points are mean values from 25 experiments with bars indicating s.tandard deviations. Note the lack of effect of ouabain on the fast component.
ms, as shown in Fig. 1, and the data were fitted by a double exponential function. In each case the experiments were performed before and after the application of 0.1 mM ouabain. Fig. 2 shows average values of the fast component of threshold change for the normal and ouabain-treated preparations as functions of step amplitude. The relationship had an initially linear region with a slope of about 0.6 and then saturated with larger steps. This behavior has been described previously s. There were no detectable differences between the results from ouabaintreated and normal preparations. Fig. 3 shows average values of the slow component of threshold change vs step amplitude. Again, the normal data had an initial slope of about 0.6 but saturated at larger steps, as reported previously s. However, in the ouabain-treated preparations the amplitude of the slow
measured
Normal N
[]
[]
30Ea)
o
e-01 o ¢._
~-
0uabatn (0, l raM) []
0'94uabatnCO,1 raM) o
0
co
20-
~
10-
o
0
750
t560
Time after step (ms] Fig. l. Changes in threshold current following stepsfrom 30 nA to-30 nA stimulating current applied to the axe-somatic region of the tactile spine neuron. Solid lines fitted to the data are double exponential decay functions (parameters in the text). Normal data were obtained from the same preparation before treatment with ouabain. Inset
shows the tactile spine neuron embedded in the cuticular wall of the spine (shaded) with its nucleus (N), sensory ending (S) and axon, together with the stimulating microelectrode (M) and current step protocol. Action potentials in the axon were detected by extraccllular electrodes located in the femur adjacent to nerve 5.
O3
0
10 20 30 40 50 60 Step decrease tl~ holdtng current [nA)
Fig. 3. The amplitude of the slow component of threshold change as a function of step amplitude for the same preparations as shown in Fig 2. Bars again indicate standard deviations. In this case the ouabain treatment reduced the amplitude of the slow component to approximately half its normal value for all step amplitudes.
114 component was reduced to about half its normal value for each step amplitude. Ouabain is a cardiac glycoside which inhibits the sodium pump t ~.t2. Previously, it was shown to reduce the hyperpolarization which follows bursts of action potentials in the tactile spine neuron with a time constant of about 270 ms 5. In the present experiments it had a significant and selective effect on the slow component of the threshold change underlying adaptation. These results suggest that electrogenic sodium pumping causes the slow component of threshold change in the tactile spine neuron, which in turn accounts for about half of the threshold change underlying rapid sensory adaptation. One difficulty with this hypothesis is the relatively short time constant of the slow component (about 500 1 Chesler, M. and Fourtner, C.R., Mechanical properties of a slow muscle in the cockroach, J. Neurobiol., 12 (1981) 391-402. 2 Fohlmeister, J.E, Poppele, R.E. and Purple, R.L., Repetitive firing: quantitative analysis of encoder behavior of slowly adapting stretch receptor of crayfish and eccentric cell of Limulus, J. Gen. Physiol., 69 (1977) 849-877. 3 French, A.S., The receptor potential and adaptation in the cockroach tactile spine, J. Neurosci., 4 (1984) 2063-2068. 4 French, A.S., Action potential adaptation in the cockroach tactile spine, J. Comp. Physiol., A155 (1984) 803-812. 5 French, A.S., After-hyperpolarization and receptor potential attenuation following bursts of action potentials in an insect mechanoreceptor, Can. J. Physiol. Pharmacol., 63 (1985) 18-22. 6 French, A.S., Strength-duration properties of a rapidly adapting insect sensory neuron, J. Comp. Physiol., A159 (1986) 757-764. 7 French, A.S., Removal of rapid sensory adaptation from an insect mechanoreceptor neuron by oxidizing agents which affect sodium channel inactivation, J. Comp. Physiol., A161 (1987) 275-282.
ms) compared to the time constants reported for electrogenic pumps in other preparations of 1-10 s2"~°'14. It is also surprising that the stimulus used in the present experiments, which attempts to depolarize a small area of membrane without producing any action potentials, is sufficient to activate the pump. Definite answers to these problems are only likely to be provided by intracellular recordings from the neuron, which have not yet been reliably achieved. However, the present results implicate sodium pumping as a m a j o r component of sensory adaptation in the tactile spine. Rodney Gramlich provided expert technical assistance throughout this work, which was supported by the Canadian Medical Research Council and the Alberta Heritage Foundation for Medical Research. 8 French, A.S., Two components of rapid sensory adaptation in a cockroach mechanoreceptor neuron, J. Neurophysiol., in press. 9 French, A.S. and Sanders, E.J., The mechanosensory apparatus of the femoral tactile spine of the cockroach, Periplaneta americana, Cell Tiss. Res., 219 (1981) 53-68. 10 Gestrelius, S. and Grampp, W., Impulse firing in the slowly adapting stretch receptor neurone of lobster and its numerical simulation, Acta Physiol. Scand. 118 (1983) 253-261. 11 Glynn, I.M., The action of cardiac glycosides on sodium and potassium movements in human red cells, J. Physiol. (Lond.), 136 (1957) 148-173. 12 Hobbs, A., Comparative effects of external monovalent cations on sodium pump activity and ouabain inhibition rates in squid giant axon, J. Physiol. (Lond.). 331 (1982) 567-576. 13 Jack, J.J.B., Noble, D. and :i~;ien, R.W., Electric Current Flow in Excitable Cells, Oxford University Press, 1983, 518 pp.. 14 Sokolove, P.G. and Cooke, I.M., Inhibition of impulse activity in a sensory neuron by an electrogenic pump, J. Gen. Physiol., 57 (1971) 125-163.
112
Elsevier
BRES 23817
Short Communications
Ouabain selectively affects the slow component of sensory adaptation in an insect mechanoreceptor Andrew S. French Department of Physiology, Universityof Alberta, Edmon:on, Alia. (Canada) (Accepted 1 August 1989)
Key words: Sensory transduction; Mechanoreceptor; Rapid adaptation; Sensory encoding; Threshold; Sodium pump; Ouabain
Sensory activity in the cockroach tactile spine neuron adapts rapidly to a step deflection. This rapid adaptation is caused by a rise in the threshold for action potential production, which has two components with different time constants and drug sensitivities. The basis of the slow component is unknown but it is insensitive to a wide range of ion channel blockers. In the present experiments the slow component was selectively reduced by ouabain, suggesting that it is due to the activation of an electrogenic sodium pump.
Rapid adaptation of sensory activity following a step change in stimulus is a widespread phenomenon in vertebrates and invertebrates. The cause of rapid adaption is not completely understood in any preparation and vertebrate sensory endings are generally too small to allow detailed electrophysiological investigation of the phenomenon. However, arthropods have peripherally located sensory cell bodies, permitting experiments on sensory proces,-,es which are often impossible in other phyla. The femoral tactile spine of the cockroach, Periplaneta americana, is a phasic mechanoreceptor containing a single sensory neuron9. The neuron adapts rapidly to a sustained stimulus, becoming silent about 1 s or less after a step change in spine position. This rapid adaptation occurs during the encoding of action potentials from the receptor current, since there is no evidence of adaption in the receptor current 3 but strong adaption in the discharge produced by direct electrical stimulation of the neuron 4. During rapid adaptation the threshold of the sensory neuron rises to exceed any constant stimulating current, preventing continuous discharge of action potentials at any level of stimulation. This behavior can be demonstrated by a preparation in which current is applied through the tip of an extraceilular glass microelectrode to the action potential initiating region at the axo-somatic junction4. Using this preparation it is possible to quantify the threshold behavior by strength-duration experiments6 which provide estimates of the rheobasic threshold current and the membrane time constant. The dynamic behavior of the rheobasic threshold
current (hereafter called threshold current) was studied recentlys. After step changes in stimulating current, the threshold current followed a double exponential trajectory, with time constants of approximately 50 ms and 500 ms. These two components of adaptation were clearly separable by several techniques and contributed approximately equally to the total change in threshold current. The faster component was selectively reduced by drugs which remove sodium channel inactivation, supporting earlier suggestions that sodium channel inactivation is involved in the rapid adaptation of the tactile spinel However, the slow component of adaptation was insensitive to a range of agents which affect sodium or potassium channelss. Several alternative mechanisms have been suggested for producing adaptation of spike discharge in neurons ~3 including the activation of electrogenic membrane pumps. The present experiments were therefore designed to test the hypothesis that an electrogenic sodium pump is involved in the adaptation of the tactile spine neuron. The preparation and experimental techniques used were identical to those described previouslys and are illustrated in Fig. 1. The tactile spine was amputated above the sensory neuron and a glass stimulating microelectrode (0.5 M NaCl, -- 5 M~9) was advanced through the spine lumen to the axo-somatic junction of the sensory neuron in the base of the spine. The microelectrode tip was moved to the point of maximum sensitivity, which is believed to be the normal site of action potential initiation. Action potentials were detected by extraceilular electrodes on
Correspondence: A.S. French, Department of Physiology, University of Alberta, Edmonton, AIta. T6G 2H7, Canada. 0006-8993/89/$03.50 © 19~9 Elsevier Science Publishers B.V. (Biomedical Division)
113 the sensory nerve in the femur. A steady depolarizing current of 30 nA was applied through the electrode for at least 2 s to allow threshold stabilization, and then the current was stepped to a lower value. The threshold was tested by a 20 ms depolarizing current pulse at various intervals after the step. An automated successive approximation technique was used to measure the threshold with a resolution of 0.2 hA. Ouabain (Sigma) was dissolved in cockroach saline' before being applied to the neuron via the cut end of the spine. The maximum effect of ouabain was reached about 15 min after its application, so a period of 30 min was allowed between applying ouabain and making the threshold measurements. Fig. 1 shows examples of changes in threshold current after steps from 30 nA to -30 nA. The absolute relationship between stimulating current and membrane potential is unknown but has been estimated to be 0.75 mV/nA for small currents 6 giving an approximate range of +20 mV for the membrane potential step. Before ouabain treatment the threshold current decayed with two components having time constants of 37.4 ms and 525 ms and amplitudes of 20.8 nA and 25.5 nA respectively (fitted line m normal). After treatment with 0.1 mM ouabain for 30 min the slow component was reduced to 13.5 nA while the fast component was almost unchanged at 22.9 nA (fitted line - - ouabain). Similar experiments to those of Fig. I were performed on 25 different preparations. The stimulating current was always initially 30 nA and it was then stepped to a range of values from 20 nA to -30 hA, giving total step decreases of 10 nA to 60 hA. For each step, the threshold was measured at a series of times after the step up to 1500
,~ 8 0 "
S ~ ~ M/,/~ / L.13e.A I__
Threshold
Normal C>
0
Ouabatn (O,l mH) 17, s 0 tO
[]
30
20-
E o o
10LL
0 0
1'8 2'8 3'8 4'0 5'8 60 Step decrease tn hotdtng current (hA)
Fig. 2. The amplitude of the fast component of threshold change as a function of step amplitude for normal and ouabain-treated preparations. Dat~.points are mean values from 25 experiments with bars indicating s.tandard deviations. Note the lack of effect of ouabain on the fast component.
ms, as shown in Fig. 1, and the data were fitted by a double exponential function. In each case the experiments were performed before and after the application of 0.1 mM ouabain. Fig. 2 shows average values of the fast component of threshold change for the normal and ouabain-treated preparations as functions of step amplitude. The relationship had an initially linear region with a slope of about 0.6 and then saturated with larger steps. This behavior has been described previously s. There were no detectable differences between the results from ouabaintreated and normal preparations. Fig. 3 shows average values of the slow component of threshold change vs step amplitude. Again, the normal data had an initial slope of about 0.6 but saturated at larger steps, as reported previously s. However, in the ouabain-treated preparations the amplitude of the slow
measured
Normal N
[]
[]
30Ea)
o
e-01 o ¢._
~-
0uabatn (0, l raM) []
0'94uabatnCO,1 raM) o
0
co
20-
~
10-
o
0
750
t560
Time after step (ms] Fig. l. Changes in threshold current following stepsfrom 30 nA to-30 nA stimulating current applied to the axe-somatic region of the tactile spine neuron. Solid lines fitted to the data are double exponential decay functions (parameters in the text). Normal data were obtained from the same preparation before treatment with ouabain. Inset
shows the tactile spine neuron embedded in the cuticular wall of the spine (shaded) with its nucleus (N), sensory ending (S) and axon, together with the stimulating microelectrode (M) and current step protocol. Action potentials in the axon were detected by extraccllular electrodes located in the femur adjacent to nerve 5.
O3
0
10 20 30 40 50 60 Step decrease tl~ holdtng current [nA)
Fig. 3. The amplitude of the slow component of threshold change as a function of step amplitude for the same preparations as shown in Fig 2. Bars again indicate standard deviations. In this case the ouabain treatment reduced the amplitude of the slow component to approximately half its normal value for all step amplitudes.
114 component was reduced to about half its normal value for each step amplitude. Ouabain is a cardiac glycoside which inhibits the sodium pump t ~.t2. Previously, it was shown to reduce the hyperpolarization which follows bursts of action potentials in the tactile spine neuron with a time constant of about 270 ms 5. In the present experiments it had a significant and selective effect on the slow component of the threshold change underlying adaptation. These results suggest that electrogenic sodium pumping causes the slow component of threshold change in the tactile spine neuron, which in turn accounts for about half of the threshold change underlying rapid sensory adaptation. One difficulty with this hypothesis is the relatively short time constant of the slow component (about 500 1 Chesler, M. and Fourtner, C.R., Mechanical properties of a slow muscle in the cockroach, J. Neurobiol., 12 (1981) 391-402. 2 Fohlmeister, J.E, Poppele, R.E. and Purple, R.L., Repetitive firing: quantitative analysis of encoder behavior of slowly adapting stretch receptor of crayfish and eccentric cell of Limulus, J. Gen. Physiol., 69 (1977) 849-877. 3 French, A.S., The receptor potential and adaptation in the cockroach tactile spine, J. Neurosci., 4 (1984) 2063-2068. 4 French, A.S., Action potential adaptation in the cockroach tactile spine, J. Comp. Physiol., A155 (1984) 803-812. 5 French, A.S., After-hyperpolarization and receptor potential attenuation following bursts of action potentials in an insect mechanoreceptor, Can. J. Physiol. Pharmacol., 63 (1985) 18-22. 6 French, A.S., Strength-duration properties of a rapidly adapting insect sensory neuron, J. Comp. Physiol., A159 (1986) 757-764. 7 French, A.S., Removal of rapid sensory adaptation from an insect mechanoreceptor neuron by oxidizing agents which affect sodium channel inactivation, J. Comp. Physiol., A161 (1987) 275-282.
ms) compared to the time constants reported for electrogenic pumps in other preparations of 1-10 s2"~°'14. It is also surprising that the stimulus used in the present experiments, which attempts to depolarize a small area of membrane without producing any action potentials, is sufficient to activate the pump. Definite answers to these problems are only likely to be provided by intracellular recordings from the neuron, which have not yet been reliably achieved. However, the present results implicate sodium pumping as a m a j o r component of sensory adaptation in the tactile spine. Rodney Gramlich provided expert technical assistance throughout this work, which was supported by the Canadian Medical Research Council and the Alberta Heritage Foundation for Medical Research. 8 French, A.S., Two components of rapid sensory adaptation in a cockroach mechanoreceptor neuron, J. Neurophysiol., in press. 9 French, A.S. and Sanders, E.J., The mechanosensory apparatus of the femoral tactile spine of the cockroach, Periplaneta americana, Cell Tiss. Res., 219 (1981) 53-68. 10 Gestrelius, S. and Grampp, W., Impulse firing in the slowly adapting stretch receptor neurone of lobster and its numerical simulation, Acta Physiol. Scand. 118 (1983) 253-261. 11 Glynn, I.M., The action of cardiac glycosides on sodium and potassium movements in human red cells, J. Physiol. (Lond.), 136 (1957) 148-173. 12 Hobbs, A., Comparative effects of external monovalent cations on sodium pump activity and ouabain inhibition rates in squid giant axon, J. Physiol. (Lond.). 331 (1982) 567-576. 13 Jack, J.J.B., Noble, D. and :i~;ien, R.W., Electric Current Flow in Excitable Cells, Oxford University Press, 1983, 518 pp.. 14 Sokolove, P.G. and Cooke, I.M., Inhibition of impulse activity in a sensory neuron by an electrogenic pump, J. Gen. Physiol., 57 (1971) 125-163.