- Email: [email protected]
Ionic conductances in two types of sensory neurons in the leech, macrobdella decora
Camp.
Biochem.
Vol.97A,
Physiol.
No.
4, pp.
511-582,
0300-9629/90
1990
$3.00
+ 0.00
c 1990PergamonPressplc
Printedin Great Britain
IONIC CONDUCTANCES IN TWO TYPES OF SENSORY NEURONS IN THE LEECH, MACROBDELLA DECORA JBRGENJOHANSEN*and ANNA L. KLEINHAUst *Department of Zoology, Iowa State University, Ames, IA 50011, USA. Telephone: (515) 294-2358; TDepartment of Cell Biology and Anatomy, New York Medical College, Valhalla, NY 10595, USA (Received 18 April 1990) Abstract-l.
Ion conductances were investigated in two kinds of leech sensory neurons (Pu and N, cells) which differ in their membrane excitability and action potential. 2. In the P, cell body excitable membrane behavior is dominated by only two currents, a sodium current (INa) and a delayed rectifier (I,& 3. In contrast, in the N, cell I,, and I, is supplemented with the presence of a transient potassium current, I,. 4. A comparison between I,, and I, in the two cell types did not reveal any significant difference in activation and inactivation kinetics of either current between neurons. 5. Thus, the properties and presence of the A-current in the N, cell and not in the P, cell may account for the functional difference in excitability of the two kinds of neurons.
INTRODUCTION In the leech identified neurons of different biological functions have distinctive and characteristic action potentials (Nicholls and Baylor, 1968). However, very little is known about the distribution and properties of the conductances which are responsible for the differences in excitability and how these properties relate to the biological function of the neurons. In this study as a beginning to address such questions, we analyze and compare the currents present in ligated cell bodies in situ of two types of functionally distinct sensory neurons. Each leech ganglion contains 14 bilaterally symmetrically organized sensory neurons (Nicholls and Baylor, 1968). In each half of the ganglion there are 3 T-cells (touch neurons), 2 P-cells (pressure neurons), and 2 N-cells (nociceptive neurons). The three types of sensory neurons are characterized by their different sensory modalities, morphology, action potential shapes, and excitable properties. Although cells belonging to the three types of sensory neurons seem to form homogeneous groups there can be differences between the properties of the cells within each class as has been demonstrated for the lateral and medial N cells (Johansen et al., 1984a,b; Johansen and Kleinhaus, 1985a; 1986a; 1987). For this reason we focus our analysis on the ionic conductances in two specific identified sensory neurons, the lateral N cell (N,) and the medial P cell (PM). The P cell responds to light pressure on the skin whereas the N cell only responds to noxious stimuli (Nicholls and Baylor, 1968) and thus has a much higher threshold for mechano-sensory transduction. Furthermore, the two cell types differ in the shape of their action potential and in their excitability: the N-cell action potential in comparison to the P-cell action potential *Author to whom all correspondence
should be addressed.
is characterized by its pronounced undershoot and low rate of repetitive firing. Our results suggest these differences can be partly explained by the differential expression of select ion conductances rather than due to variations in the activation and inactivation kinetics of otherwise similar currents. Thus in the P, cell body excitable behavior underlying the action potential is dominated by only two currents, a delayed rectifier (IK) and an inward sodium current (INa). In contrast, in the N, cell body I, and INa, which have the same kinetic properties as in the P, cell, are supplemented with the presence of a rapidly activating A-current (IA). MATERIALS
AND METHODS
The experiments were performed on sensory neurons, N, and P, cells, from segmental ganglia of the leech Macrobdella decora. The leeches were obtained from a commercial supplier and kept in spring water at 15°C. Dissection and preparation
The procedures for current clamp recording and for perfusion of the experimental chamber were similar to those described in previous reports (Kleinhaus, 1976; 1980). The ganglionic chain was dissected in normal leech Ringer and pinned out in Sylgard (Dow Corning) coated dishes. The connective capsules were removed with specially sharpened forceps to expose the cells. The sensory cells could be unambiguously identified by visual inspection under the stereomicroscope due to their size and characteristic position in the ganglion. In order to obtain good spatial voltage clamp conditions, the soma of individual neurons was isolated from the axonal branches of the cell by ligation with fine surgical thread as described for leech neurons by Ready and Nicholls (1979). The cells were left attached to the ganglia and appeared to be healthy and regained normal resting potentials in less than 1 hr after the ligation. Ringer solutions
Normal leech Ringer contained: IZOmM NaCI, 2mM CaCI,, 4mM KCI, IOmM glucose, IOmM Tris-HCl, 577
578
J~RGENJOHANSEN and ANNAL. KLEINHAUS
pH 7.4. Sodium-free solutions were prepared by substitution of NaCl with isoosmolar concentrations of sucrose. Sodium and calcium free solutions were made by the additional substitution of CaCl, with 2mM MnCl,. The pharmacological potassium channel blockers tetraethylammonium chloride (TEA) and 4”aminopyridine (4-AP) were added directly to the experimental Ringer solutions. When not otherwise stated the experiments were carried out at room temperature (22-24°C). However, in some experiments the perfusing Ringer was cooled to 10°C and the temperature in the experimental chamber was continuously monitored by a thermistor. Two electrode voltage
clamp
A total of 23 sensory cells were investigated with a Dagan 8500 two electrode voltage clamp system (Dagan Corp., USA). The clamp was operated at a fixed bandwidth of 1kHz. Membrane current was measured bjt a virtual ground circuit connected to the bath via a AgCl-pellet electrode placed as close to the preparation as possible to minimize series resistance. Step changes in membrane potential were 90% complete within 200 psec but the settling time for the membrane current was usually on the order of l-3 msec due to the cell’s capacitance. The clamp was normally adjusted to a slightly underdamped condition. Elect;od& were pblled from I mm ultradot glass (Frederick Haer). were filled with 2.5 M KCl. and were bevelled to have &istances of IO-15 Ma. To reduce capacitative coupling between the current- and voltage electrodes, the voltage electrode was shielded with conductive silver paint to within 50pm from the tip. The conductive shield was insulated from the bath solutions by a cover of lacquer and connected to the driven shield of the voltage probe. Current and voltage traces were recorded on a computer system equipped with a Tecmar Labmaster A-D converter which digitized the traces at 32 kHz. Leakage and capacitative currents in response to de- and hyperpolarizing potential steps of equal magnitude were assumed to be linear and symmetrical and were electronically subtracted from the records. Leakage current for a hyperpolarizing step of 50mV was generally less than 5 nA when the cell was successfully penetrated with two electrodes. Nonlinear leakage currents within the voltage range investigated were not observed. Only experiments in which the drift of the voltage electrode was less than + 5 mV were considered. No compensation for series resistance was attempted. However, errors due to series resistance were not likely to be an
appreciable factor, since most of the currents studied were less than 50nA in magnitude. The time course for activation and inactivation of the currents was analysed in terms of the Hodgkin and Huxley (1952) kinetic scheme developed for the sodium and potassium conductances in the squid giant axon. Further details of the application of this analysis to leech neuronal conductances is given in Johansen and Kleinhaus (1986b).
RESULTS The object of this paper is to compare the properties and distribution of conductances in two types of functionally identified sensory neurons, NI and P,. All experiments were performed on acutely ligated cell bodies in situ and the currents analysed by two electrode voltage clamp techniques. In ~a~~o~~~~a the cells have a soma diameter of between 60-80 pm, In neither of the cells were we able to demonstrate a calcium-dependent potassium-current; a result supported by the results of Stewart et al. (1989) which also found this particular current absent in cultured leech sensory neurons of Hirudo. However, this was unexpected since studies have suggested that
components of the afterhyperpolarization of the individual spike as well as the prolonged hyperpolarization following repetitive firing may be a result of a calcium-dependent potassium current (Jansen and Nicholls, 1973; Yang and Kleinhaus, 1984; Johansen et al., 1985). Whether our inability to detect an I K(Ca)was due to that this current is truly absent in these cells or too small to be resolved by the techniques applied in the present study will need further ex~~mentation. Ion conductances in the N, ceN
The action potential in the N, cell has previously been demonstrated to be predominantly carried by sodium ions (Kleinhaus and Prichard, 1983; Johansen and Kleinhaus, 1986a). However, by pharmacological means and ion-substitutions we identified three other currents (Fig. 1) which are present in the cell body of this neuron: a delayed rectifier, a rapidly inactivating A-current, and a calcium current. I NP. Sodium currents were recorded in Ca-free, Mn-containing Ringer solutions to which 25mM TEA and 3 mM 4-AP were added. This eliminates all outward currents and prevents current flow through the Ca-channels, whether carried by divalent cations or sodium ions (Johansen and Kleinhaus, 1985b; 1986b). Since the activation and inactivation kinetics of INa are very fast all recordings of I,, were made at 10°C. Under these conditions the peak Na-conductance (Fig. 1C) is for voltage steps from a resting membrane potential of - 50 mV to 0 to 5 mV with a threshold for activation close to -4OmV (Fig. 2). The kinetics of I,, can be well fitted by conventional Hodgkin-Huxley type equations of the form I,, = g,,m4h. At 0 mV at 10°C I,, activates with a time constant of approximately 2 ms and inactivates with a time constant of about 8 ms. IK. A delayed rectifier current was isolated in Naand Ca-free, Mn-containing Ringer solutions to which 3 mM 4-AP was added. It has been previously shown that this effectively separates I, from I, in leech neurons (Johansen and Kleinhaus, 1986b). Figure IA shows a family of currents flowing in response to depolarizing steps of 100 ms duration from a holding potential of -45 mV. The current was voltage dependent, had a sigmoidal rising phase, and was activated at potentials more positive than -40 to - 30 mV (Fig. 2). The kinetics of the current could be approximated by a Hodgkin-Huxley model of the form I, = gxn2h where the kinetics of the inactivation parameter was so slow (> 500 ms) as to be ignored in this analysis. The time constants for activation were voltage dependent and in the range of potential steps from - 15 to 15 mV varied from 14.5 to 6 ms (see also Fig. 7). I,,. Figure lB shows a family of A-currents in the N, cell recorded in Na- and Ca-free, Mn-containing Ringer solution where the delayed rectifier was blocked by 25 mM TEA. The currents were elicited from a holding potential of -70 mV. The current activated much more rapidly than IK and inactivated completely following an exponential time course with a voltage-independent time constant of about 65 ms (range: 61-71 ms; N = 5). From a holding potential of -70 mV the current activated in response to depoIarizations more positive than - 50 mV (Fig. 2).
519
Ion conductances in leech sensory neurons
I1OnA
100
Vh -45mV
kmA
1
ma
V,,: -70mV
C
'NZI
100
ms
D
‘B,
Es? I/-l IOnA
5nA
5mV
v,,:-50
mV
40mo
Vh: -40mV
100
ms
Fig. 1.Ion conductances in the N, cell. (A) Family of delayed rectifier currents, I,, elicited from a holding potential (V,) of -45 mV in response to depolarizing voltage steps incremented by 10 mV. (B) Family of transient potassium currents, I,. Voltage steps are in 10 mV increments. (C) Peak sodium current, I,,, obtained in Ca-free, Mn-containing Ringer solution with 25 mM TEA and 3 mM 4-AP at 10°C. (D) Currents through the Ca-conductance isolated with 10 mM [Bab as the current carrier. Voltage steps are in 10 mV increments. The duration of the voltage clamp steps are indicated on this and the following figures in milliseconds (ms).
As other A-currents in leech (Johansen and Kleinhaus, 1986b; Stewart et al.. 1989) this current was characterized by almost complete steady state
IK
Fig. 2. Current-voltage
relationships of the currents in the N, cell.
inactivation at negative potentials; however, the current was at least 50% available at normal resting membrane potentials of -45 to - 50 mV (data not shown). I,,. In addition to I,.+,, I,, and I, the N, cell also possesses a Ca-current. The Ca-conductance was isolated as a Ba-current (Fig. 1D) in 1OmM [Ba], in Na-free Ringer solutions; however, at physiological extracellular Ca-concentrations (2 mM) the peak Cacurrent is negligible (c 5 nA) and does not contribute to the action potential in any significant way. Thus the excitable membrane behavior during an action appears to be dominated by INa, I,, and I, in the N, cell. Ion conductances in the P, cell In the Phi cell we identified a sodium current I,, (Fig. 3) and a delayed rectifier current I, (Fig. 4) with properties and current-voltage relationships (Fig. 5) similar to the corresponding currents in the N, cell. In order to compare the activation and inactivation kinetics of Ik and I,, in both P, and N, cells we also conducted a Hodgkin-Huxley analysis of these currents in the P, cell. As in the N, cells I,, and 1s were best described by equations of the form I,, = gN,m4h and I, = g,n*h, respectively. Furthermore, the activation and inactivation kinetics of I,, in the N, and the P, cells were indistinguishable (Fig. 6). The
JBRGENJOHANSEN and ANNA L. KLEINHAUS
-10 mV
-15 mV
K
OmV
-5mV
-r-I
-IV,: -50 mV
40
me
‘Na
-60-1 -60
Fig. 3. Sodium currents in the P, cell. The currents were obtained at 10°C from a holding potential of - 50 mV by voltage steps to - 15, - 10, -5, and 0 mV, respectively. activation kinetics of I, was also very similar in the two cell types (Fig. 7). The inactivation of IK was slower than 500ms in both cells and was not quantified any further in either case. The P, cell also possesss a calcium-current (Johansen and Kleinhaus, 1989); however, as in the N, cell this current is very small and does not contribute significantly to the action potential. Notably, we were not able to detect an A-current in this cell in contrast to the situation in the N, cell which has a well developed A-current (Fig. 1B). Our inability to detect this current in the P, cell was not a function of the state of the preparations in which we looked for this current. We did not record an A-current in the P, cell even when such a current could readily be obtained in the Retzius cell, N, cell, and anterior pagoda cell of the very same ganglion. The N, cell is generally slightly smaller than the P, cell so the difference in current expression is unlikely to be due to increased cell damage from penetrating a smaller cell with two electrodes. In addition, Stewart et al. (1989) analysing cultured Hirudo neurons did not find an A-current in P cells. The action potential in the P, cell is therefore likely to be primarily the result of the interactions of only
I -40
.
I -20
.
! 0
1 20
.
1 40 mV
Fig. 5. Current-voltage relationships of I,, and I, in the P,
cell.
two currents, namely the delayed rectifier and the sodium-current. DlSCUSSlON
In this paper we have analysed the ionic conductances in two classes of sensory cells in the leech, the N, and the P, cell, which represent two different kinds of sensory modalities. In the P, cell body excitable membrane behavior is dominated by only two currents, a delayed rectifier current and an inward sodium-current. However, in the N, cell body I, and INa is supplemented with the presence of a transient potassium current (IA) which activates close to the resting potential of the cell. We did not detect any significant differences in the kinetic parameters or current-voltage relationships of I,, and I, between the P, and N, cells; and the kinetics of the sodium current delayed rectifier current in both cell types follows the same Hodgkin-Huxley kinetic schemes of I,, = gNam4h and I, = g,n2h, respectively. Thus the only major difference in the conductances governing the action potential in these cells are the absence or
‘K
“ETCH-
-20
100 ma
Vh:‘45mV Fig. 4. Family P, cell.
of delayed
rectifier potassium
currents
Voltage steps are in 10 mV increments.
in the
I
-10
ii
lb
2-O
Fig. 6. Voltage dependence of activation (7,) and inactivation (TV) time constants of IN, from N, (a, n ) and PM (0, 0) cells. The time constants are from currents obtained at 10 C.
Ion conductances in leech sensory neurons
The results of this paper exemplifies how cells may express different conductances which allow them to fine tune their membrane excitability to suit their biological function. However, in order to expand this analysis we plan to model the excitable properties of the different sensory neurons on the basis of the kinetic properties of the particular conductances they possess and to further correlate this with their role in
16-9 14 12IOa-
behavior.
641 2 1
o!
-30
Fig.
581
Acknowledgements-This work was supported by a Iowa State Biotechnology Grant and a Iowa State University Grant to J.J. and by grants from NIH (NS-18054) and the Whitehall Foundation to A.L.K.
.
I
-20
7. Voltage
*
I
-10
.
I
0
.
t
10
.
1
20
.
I
30
dependence of activation time constants for 1, in N, and P, cells.
the presence of an A-current. Analysis in other neurons has shown that I, modulates spiking frequency by prolonging the interval between action potentials due to its activation at subthreshold potentials (Connor, 1980). I, is inactivating during the spike, but full availability for activation is restored by the spike afterhy~rpolarization. Thus, the presence of an A-current in the N, cell, which is a sensory neuron responsive only to noxious stimuli, may account in part for the higher threshold of stimulation necessary to excite this cell. The specific properties of an A-current will tend to counteract brief and irrelevant stimuli, and make the cell responsive only to high intensity stimuli where an all out response, such as an escape response, may be warranted. In contrast the P cells and also the T cells role is to respond to low intensity stimulation. This third type of sensory neuron, the T-cell, which like the P-cell fires volleys of action potentials upon stimulation also appear not to possess a transient potassium current (Valkanov and Boev, 1988; Stewart et al., 1989; Johansen and Kleinhaus, unpublished results). The presence of an A-current in the N, cell in comparison to its absence in the P, cell and in T-cells therefore can convincingly account for the N, cell’s much lower rate of repetitive firing of action potentials. The properties and kinetics of the delayed rectifier potassium current in the N, and PM cells and the transient potassium current in the N, cell described in this study are very similar to those reported for the corresponding currents in the Retzius cell (Johansen and Kleinhaus, 1986b) suggesting that these cells possess the same basic type of potassium currents. In addition, in the Retzius cell it has been shown that I, and I, can be differentialIy modulated by serotonin (Acosta-Urquidi et al., 1989). Surprisingly we were not able to demonstrate the presence of a IK(caj in Macrobdella sensory neurons. Whether this was due to that this current is not expressed in these cells or that it is too small to be identified for technical reasons needs further experimentation to be resolved. However, we have been able to clearly demonstrate a Ca-dependent K current of slow kinetics in the ALG neuron of the glossiphoniid leech Haementeria (Johansen et al., 1987).
REFERENCES Acosta-Urquidi J., Sahley C. L. and Kleinhaus A. L. (1989) Serotonin differentially modulates two K+ currents in the Retzius cell of the leech. J. Exp. Biol. 145, 403-417. Connor J. A. (1980) The fast K+ channel and repetitive firing. In: ~ol~use~n Nerve Cellf: From ~iophysies to Behuoior (Edited by Koester J. and Byrne J. H.), pp. 125-133, Cold Spring Harbor, New York. Hodgkin A. L. and Huxley A. F. (1952) A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. 117, 500-544. Jansen J. K. S. and Nicholls J. G. (1973) Conductance changes, an electrogenic pump and the h~rpola~zation of leech neurons following impulses. J. Physioi. 229, 635-655. Johansen J. and Kleinhaus A. L. (1985a) A monosynaptic connection between the medial nociceptive and the nut cell in leech ganglia. J. Comp. Physiol. 156, 65-69. Johansen J. and Kieinhaus A. L. (1985b) Properties of action potentials carried by divalent cations in identified leech neurons. f. Comp. Physiot. 157, 491-497. Johansen J. and Kleinhaus A. L. (1986a) Differential sensitivity to tetrodotoxin of nociceptive neurons in four species of leeches. J. Neurosci. 6, 3499-3504. Johansen J. and Kleinhaus A. L. (1986b) Transient and delayed potassium currents in the Retzius cell of the leech, ~acrobdelia
decora. J. Neurophys~oi. 36, 8 12622.
Johansen J. and Kleinhaus A. L. (1987) The effects of procaine, strychnine and penicillin on nociceptive neurons in leech segmental ganglia. Comp. B&hem. Physiol. WC, 405-409.
Johansen J. and Kleinhaus A. L. (1989) Voltage clamp analysis of ionic conductances in two types of sensory neurons in the leech. Sot. Neurosei. Abstr. 15, 1146. Johansen J., Hockfield S. and McKay R. D. G. (1984a) Distribution and morphology of nociceptive cells in the central nervous system of three species of leeches. J. Comp. Neurol. 226, 263-273. Johansen J., Yang J. and Kleinhaus A. L. (1984b) Actions of procaine on specific nociceptive cells in leech central nervous system. J. Neurosci. 4, 1253-1261. Johansen J., Taft W. C., Yang J., Kleinhaus A. L. and DeLorenzo R. J. (1985) Inhibition of Ca*+ conductance in identified lee& ne&ons by benzodiazepines. Proc. Natf. Acad. Sci. USA 82, 3935-3939. Johansen J., Yang J. and Kleinhaus A. L. (1987) Voltageclamp analysis of the ionic conductances in a leech neuron with a purely calcium-dependent action potential. J. Neuroph vsiol. 58, 1468-l 484. Kleinhaus -A: L. (19?6) Divalent cations and the action potential of the leech Retzius cell. PJzZgers Arch. 363, 97-104. Kleinhaus A. L. (1980) Segregation of leech neurones by the effect of sparteine on action potential duration. J. Physiol. 299, 309-321.
582
JBRGENJOHANSEN and ANNA L. KLEINHAIJS
Kleinhaus A. L. and Prichard J. W. (1983) Differential action of tetrodotoxin on identified leech neurons. Camp. Biochem. Physiol. 74C, 2 1l-2 18. Nicholls J. G. and Baylor D. A. (1968) Specific modalities and receptive fields of sensory neurons in the CNS of the leech. J. Neurophysiol. 31, 740-756. Ready D. F. and Nicholls J. G. (1979) Identified neurones from leech CNS make selective connections in culture. Nature 281, 67-69.
Stewart R. R., Nicholls J. G. and Adams W. B. (1989) Na+,
K+, and Ca2+ currents in identified leech neurones in culture. J. Exp. Biol. 141, l-20. Valkanov M. and Boev K. (1988) Ionic currents in the somatic membrane of identified T-mechanosensory neurons from segmental ganglia of the medicinal leech. Gen. Physiol. Biophys. 7, 643-649. Yang J. and Kleinhaus A. L. (1984) Effects of tetraethylammonium-chloride and divalent cations on the afterhyperpolarization following repetitive firing in the leech. Brain Res. 311, 380-384.
Biochem.
Vol.97A,
Physiol.
No.
4, pp.
511-582,
0300-9629/90
1990
$3.00
+ 0.00
c 1990PergamonPressplc
Printedin Great Britain
IONIC CONDUCTANCES IN TWO TYPES OF SENSORY NEURONS IN THE LEECH, MACROBDELLA DECORA JBRGENJOHANSEN*and ANNA L. KLEINHAUst *Department of Zoology, Iowa State University, Ames, IA 50011, USA. Telephone: (515) 294-2358; TDepartment of Cell Biology and Anatomy, New York Medical College, Valhalla, NY 10595, USA (Received 18 April 1990) Abstract-l.
Ion conductances were investigated in two kinds of leech sensory neurons (Pu and N, cells) which differ in their membrane excitability and action potential. 2. In the P, cell body excitable membrane behavior is dominated by only two currents, a sodium current (INa) and a delayed rectifier (I,& 3. In contrast, in the N, cell I,, and I, is supplemented with the presence of a transient potassium current, I,. 4. A comparison between I,, and I, in the two cell types did not reveal any significant difference in activation and inactivation kinetics of either current between neurons. 5. Thus, the properties and presence of the A-current in the N, cell and not in the P, cell may account for the functional difference in excitability of the two kinds of neurons.
INTRODUCTION In the leech identified neurons of different biological functions have distinctive and characteristic action potentials (Nicholls and Baylor, 1968). However, very little is known about the distribution and properties of the conductances which are responsible for the differences in excitability and how these properties relate to the biological function of the neurons. In this study as a beginning to address such questions, we analyze and compare the currents present in ligated cell bodies in situ of two types of functionally distinct sensory neurons. Each leech ganglion contains 14 bilaterally symmetrically organized sensory neurons (Nicholls and Baylor, 1968). In each half of the ganglion there are 3 T-cells (touch neurons), 2 P-cells (pressure neurons), and 2 N-cells (nociceptive neurons). The three types of sensory neurons are characterized by their different sensory modalities, morphology, action potential shapes, and excitable properties. Although cells belonging to the three types of sensory neurons seem to form homogeneous groups there can be differences between the properties of the cells within each class as has been demonstrated for the lateral and medial N cells (Johansen et al., 1984a,b; Johansen and Kleinhaus, 1985a; 1986a; 1987). For this reason we focus our analysis on the ionic conductances in two specific identified sensory neurons, the lateral N cell (N,) and the medial P cell (PM). The P cell responds to light pressure on the skin whereas the N cell only responds to noxious stimuli (Nicholls and Baylor, 1968) and thus has a much higher threshold for mechano-sensory transduction. Furthermore, the two cell types differ in the shape of their action potential and in their excitability: the N-cell action potential in comparison to the P-cell action potential *Author to whom all correspondence
should be addressed.
is characterized by its pronounced undershoot and low rate of repetitive firing. Our results suggest these differences can be partly explained by the differential expression of select ion conductances rather than due to variations in the activation and inactivation kinetics of otherwise similar currents. Thus in the P, cell body excitable behavior underlying the action potential is dominated by only two currents, a delayed rectifier (IK) and an inward sodium current (INa). In contrast, in the N, cell body I, and INa, which have the same kinetic properties as in the P, cell, are supplemented with the presence of a rapidly activating A-current (IA). MATERIALS
AND METHODS
The experiments were performed on sensory neurons, N, and P, cells, from segmental ganglia of the leech Macrobdella decora. The leeches were obtained from a commercial supplier and kept in spring water at 15°C. Dissection and preparation
The procedures for current clamp recording and for perfusion of the experimental chamber were similar to those described in previous reports (Kleinhaus, 1976; 1980). The ganglionic chain was dissected in normal leech Ringer and pinned out in Sylgard (Dow Corning) coated dishes. The connective capsules were removed with specially sharpened forceps to expose the cells. The sensory cells could be unambiguously identified by visual inspection under the stereomicroscope due to their size and characteristic position in the ganglion. In order to obtain good spatial voltage clamp conditions, the soma of individual neurons was isolated from the axonal branches of the cell by ligation with fine surgical thread as described for leech neurons by Ready and Nicholls (1979). The cells were left attached to the ganglia and appeared to be healthy and regained normal resting potentials in less than 1 hr after the ligation. Ringer solutions
Normal leech Ringer contained: IZOmM NaCI, 2mM CaCI,, 4mM KCI, IOmM glucose, IOmM Tris-HCl, 577
578
J~RGENJOHANSEN and ANNAL. KLEINHAUS
pH 7.4. Sodium-free solutions were prepared by substitution of NaCl with isoosmolar concentrations of sucrose. Sodium and calcium free solutions were made by the additional substitution of CaCl, with 2mM MnCl,. The pharmacological potassium channel blockers tetraethylammonium chloride (TEA) and 4”aminopyridine (4-AP) were added directly to the experimental Ringer solutions. When not otherwise stated the experiments were carried out at room temperature (22-24°C). However, in some experiments the perfusing Ringer was cooled to 10°C and the temperature in the experimental chamber was continuously monitored by a thermistor. Two electrode voltage
clamp
A total of 23 sensory cells were investigated with a Dagan 8500 two electrode voltage clamp system (Dagan Corp., USA). The clamp was operated at a fixed bandwidth of 1kHz. Membrane current was measured bjt a virtual ground circuit connected to the bath via a AgCl-pellet electrode placed as close to the preparation as possible to minimize series resistance. Step changes in membrane potential were 90% complete within 200 psec but the settling time for the membrane current was usually on the order of l-3 msec due to the cell’s capacitance. The clamp was normally adjusted to a slightly underdamped condition. Elect;od& were pblled from I mm ultradot glass (Frederick Haer). were filled with 2.5 M KCl. and were bevelled to have &istances of IO-15 Ma. To reduce capacitative coupling between the current- and voltage electrodes, the voltage electrode was shielded with conductive silver paint to within 50pm from the tip. The conductive shield was insulated from the bath solutions by a cover of lacquer and connected to the driven shield of the voltage probe. Current and voltage traces were recorded on a computer system equipped with a Tecmar Labmaster A-D converter which digitized the traces at 32 kHz. Leakage and capacitative currents in response to de- and hyperpolarizing potential steps of equal magnitude were assumed to be linear and symmetrical and were electronically subtracted from the records. Leakage current for a hyperpolarizing step of 50mV was generally less than 5 nA when the cell was successfully penetrated with two electrodes. Nonlinear leakage currents within the voltage range investigated were not observed. Only experiments in which the drift of the voltage electrode was less than + 5 mV were considered. No compensation for series resistance was attempted. However, errors due to series resistance were not likely to be an
appreciable factor, since most of the currents studied were less than 50nA in magnitude. The time course for activation and inactivation of the currents was analysed in terms of the Hodgkin and Huxley (1952) kinetic scheme developed for the sodium and potassium conductances in the squid giant axon. Further details of the application of this analysis to leech neuronal conductances is given in Johansen and Kleinhaus (1986b).
RESULTS The object of this paper is to compare the properties and distribution of conductances in two types of functionally identified sensory neurons, NI and P,. All experiments were performed on acutely ligated cell bodies in situ and the currents analysed by two electrode voltage clamp techniques. In ~a~~o~~~~a the cells have a soma diameter of between 60-80 pm, In neither of the cells were we able to demonstrate a calcium-dependent potassium-current; a result supported by the results of Stewart et al. (1989) which also found this particular current absent in cultured leech sensory neurons of Hirudo. However, this was unexpected since studies have suggested that
components of the afterhyperpolarization of the individual spike as well as the prolonged hyperpolarization following repetitive firing may be a result of a calcium-dependent potassium current (Jansen and Nicholls, 1973; Yang and Kleinhaus, 1984; Johansen et al., 1985). Whether our inability to detect an I K(Ca)was due to that this current is truly absent in these cells or too small to be resolved by the techniques applied in the present study will need further ex~~mentation. Ion conductances in the N, ceN
The action potential in the N, cell has previously been demonstrated to be predominantly carried by sodium ions (Kleinhaus and Prichard, 1983; Johansen and Kleinhaus, 1986a). However, by pharmacological means and ion-substitutions we identified three other currents (Fig. 1) which are present in the cell body of this neuron: a delayed rectifier, a rapidly inactivating A-current, and a calcium current. I NP. Sodium currents were recorded in Ca-free, Mn-containing Ringer solutions to which 25mM TEA and 3 mM 4-AP were added. This eliminates all outward currents and prevents current flow through the Ca-channels, whether carried by divalent cations or sodium ions (Johansen and Kleinhaus, 1985b; 1986b). Since the activation and inactivation kinetics of INa are very fast all recordings of I,, were made at 10°C. Under these conditions the peak Na-conductance (Fig. 1C) is for voltage steps from a resting membrane potential of - 50 mV to 0 to 5 mV with a threshold for activation close to -4OmV (Fig. 2). The kinetics of I,, can be well fitted by conventional Hodgkin-Huxley type equations of the form I,, = g,,m4h. At 0 mV at 10°C I,, activates with a time constant of approximately 2 ms and inactivates with a time constant of about 8 ms. IK. A delayed rectifier current was isolated in Naand Ca-free, Mn-containing Ringer solutions to which 3 mM 4-AP was added. It has been previously shown that this effectively separates I, from I, in leech neurons (Johansen and Kleinhaus, 1986b). Figure IA shows a family of currents flowing in response to depolarizing steps of 100 ms duration from a holding potential of -45 mV. The current was voltage dependent, had a sigmoidal rising phase, and was activated at potentials more positive than -40 to - 30 mV (Fig. 2). The kinetics of the current could be approximated by a Hodgkin-Huxley model of the form I, = gxn2h where the kinetics of the inactivation parameter was so slow (> 500 ms) as to be ignored in this analysis. The time constants for activation were voltage dependent and in the range of potential steps from - 15 to 15 mV varied from 14.5 to 6 ms (see also Fig. 7). I,,. Figure lB shows a family of A-currents in the N, cell recorded in Na- and Ca-free, Mn-containing Ringer solution where the delayed rectifier was blocked by 25 mM TEA. The currents were elicited from a holding potential of -70 mV. The current activated much more rapidly than IK and inactivated completely following an exponential time course with a voltage-independent time constant of about 65 ms (range: 61-71 ms; N = 5). From a holding potential of -70 mV the current activated in response to depoIarizations more positive than - 50 mV (Fig. 2).
519
Ion conductances in leech sensory neurons
I1OnA
100
Vh -45mV
kmA
1
ma
V,,: -70mV
C
'NZI
100
ms
D
‘B,
Es? I/-l IOnA
5nA
5mV
v,,:-50
mV
40mo
Vh: -40mV
100
ms
Fig. 1.Ion conductances in the N, cell. (A) Family of delayed rectifier currents, I,, elicited from a holding potential (V,) of -45 mV in response to depolarizing voltage steps incremented by 10 mV. (B) Family of transient potassium currents, I,. Voltage steps are in 10 mV increments. (C) Peak sodium current, I,,, obtained in Ca-free, Mn-containing Ringer solution with 25 mM TEA and 3 mM 4-AP at 10°C. (D) Currents through the Ca-conductance isolated with 10 mM [Bab as the current carrier. Voltage steps are in 10 mV increments. The duration of the voltage clamp steps are indicated on this and the following figures in milliseconds (ms).
As other A-currents in leech (Johansen and Kleinhaus, 1986b; Stewart et al.. 1989) this current was characterized by almost complete steady state
IK
Fig. 2. Current-voltage
relationships of the currents in the N, cell.
inactivation at negative potentials; however, the current was at least 50% available at normal resting membrane potentials of -45 to - 50 mV (data not shown). I,,. In addition to I,.+,, I,, and I, the N, cell also possesses a Ca-current. The Ca-conductance was isolated as a Ba-current (Fig. 1D) in 1OmM [Ba], in Na-free Ringer solutions; however, at physiological extracellular Ca-concentrations (2 mM) the peak Cacurrent is negligible (c 5 nA) and does not contribute to the action potential in any significant way. Thus the excitable membrane behavior during an action appears to be dominated by INa, I,, and I, in the N, cell. Ion conductances in the P, cell In the Phi cell we identified a sodium current I,, (Fig. 3) and a delayed rectifier current I, (Fig. 4) with properties and current-voltage relationships (Fig. 5) similar to the corresponding currents in the N, cell. In order to compare the activation and inactivation kinetics of Ik and I,, in both P, and N, cells we also conducted a Hodgkin-Huxley analysis of these currents in the P, cell. As in the N, cells I,, and 1s were best described by equations of the form I,, = gN,m4h and I, = g,n*h, respectively. Furthermore, the activation and inactivation kinetics of I,, in the N, and the P, cells were indistinguishable (Fig. 6). The
JBRGENJOHANSEN and ANNA L. KLEINHAUS
-10 mV
-15 mV
K
OmV
-5mV
-r-I
-IV,: -50 mV
40
me
‘Na
-60-1 -60
Fig. 3. Sodium currents in the P, cell. The currents were obtained at 10°C from a holding potential of - 50 mV by voltage steps to - 15, - 10, -5, and 0 mV, respectively. activation kinetics of I, was also very similar in the two cell types (Fig. 7). The inactivation of IK was slower than 500ms in both cells and was not quantified any further in either case. The P, cell also possesss a calcium-current (Johansen and Kleinhaus, 1989); however, as in the N, cell this current is very small and does not contribute significantly to the action potential. Notably, we were not able to detect an A-current in this cell in contrast to the situation in the N, cell which has a well developed A-current (Fig. 1B). Our inability to detect this current in the P, cell was not a function of the state of the preparations in which we looked for this current. We did not record an A-current in the P, cell even when such a current could readily be obtained in the Retzius cell, N, cell, and anterior pagoda cell of the very same ganglion. The N, cell is generally slightly smaller than the P, cell so the difference in current expression is unlikely to be due to increased cell damage from penetrating a smaller cell with two electrodes. In addition, Stewart et al. (1989) analysing cultured Hirudo neurons did not find an A-current in P cells. The action potential in the P, cell is therefore likely to be primarily the result of the interactions of only
I -40
.
I -20
.
! 0
1 20
.
1 40 mV
Fig. 5. Current-voltage relationships of I,, and I, in the P,
cell.
two currents, namely the delayed rectifier and the sodium-current. DlSCUSSlON
In this paper we have analysed the ionic conductances in two classes of sensory cells in the leech, the N, and the P, cell, which represent two different kinds of sensory modalities. In the P, cell body excitable membrane behavior is dominated by only two currents, a delayed rectifier current and an inward sodium-current. However, in the N, cell body I, and INa is supplemented with the presence of a transient potassium current (IA) which activates close to the resting potential of the cell. We did not detect any significant differences in the kinetic parameters or current-voltage relationships of I,, and I, between the P, and N, cells; and the kinetics of the sodium current delayed rectifier current in both cell types follows the same Hodgkin-Huxley kinetic schemes of I,, = gNam4h and I, = g,n2h, respectively. Thus the only major difference in the conductances governing the action potential in these cells are the absence or
‘K
“ETCH-
-20
100 ma
Vh:‘45mV Fig. 4. Family P, cell.
of delayed
rectifier potassium
currents
Voltage steps are in 10 mV increments.
in the
I
-10
ii
lb
2-O
Fig. 6. Voltage dependence of activation (7,) and inactivation (TV) time constants of IN, from N, (a, n ) and PM (0, 0) cells. The time constants are from currents obtained at 10 C.
Ion conductances in leech sensory neurons
The results of this paper exemplifies how cells may express different conductances which allow them to fine tune their membrane excitability to suit their biological function. However, in order to expand this analysis we plan to model the excitable properties of the different sensory neurons on the basis of the kinetic properties of the particular conductances they possess and to further correlate this with their role in
16-9 14 12IOa-
behavior.
641 2 1
o!
-30
Fig.
581
Acknowledgements-This work was supported by a Iowa State Biotechnology Grant and a Iowa State University Grant to J.J. and by grants from NIH (NS-18054) and the Whitehall Foundation to A.L.K.
.
I
-20
7. Voltage
*
I
-10
.
I
0
.
t
10
.
1
20
.
I
30
dependence of activation time constants for 1, in N, and P, cells.
the presence of an A-current. Analysis in other neurons has shown that I, modulates spiking frequency by prolonging the interval between action potentials due to its activation at subthreshold potentials (Connor, 1980). I, is inactivating during the spike, but full availability for activation is restored by the spike afterhy~rpolarization. Thus, the presence of an A-current in the N, cell, which is a sensory neuron responsive only to noxious stimuli, may account in part for the higher threshold of stimulation necessary to excite this cell. The specific properties of an A-current will tend to counteract brief and irrelevant stimuli, and make the cell responsive only to high intensity stimuli where an all out response, such as an escape response, may be warranted. In contrast the P cells and also the T cells role is to respond to low intensity stimulation. This third type of sensory neuron, the T-cell, which like the P-cell fires volleys of action potentials upon stimulation also appear not to possess a transient potassium current (Valkanov and Boev, 1988; Stewart et al., 1989; Johansen and Kleinhaus, unpublished results). The presence of an A-current in the N, cell in comparison to its absence in the P, cell and in T-cells therefore can convincingly account for the N, cell’s much lower rate of repetitive firing of action potentials. The properties and kinetics of the delayed rectifier potassium current in the N, and PM cells and the transient potassium current in the N, cell described in this study are very similar to those reported for the corresponding currents in the Retzius cell (Johansen and Kleinhaus, 1986b) suggesting that these cells possess the same basic type of potassium currents. In addition, in the Retzius cell it has been shown that I, and I, can be differentialIy modulated by serotonin (Acosta-Urquidi et al., 1989). Surprisingly we were not able to demonstrate the presence of a IK(caj in Macrobdella sensory neurons. Whether this was due to that this current is not expressed in these cells or that it is too small to be identified for technical reasons needs further experimentation to be resolved. However, we have been able to clearly demonstrate a Ca-dependent K current of slow kinetics in the ALG neuron of the glossiphoniid leech Haementeria (Johansen et al., 1987).
REFERENCES Acosta-Urquidi J., Sahley C. L. and Kleinhaus A. L. (1989) Serotonin differentially modulates two K+ currents in the Retzius cell of the leech. J. Exp. Biol. 145, 403-417. Connor J. A. (1980) The fast K+ channel and repetitive firing. In: ~ol~use~n Nerve Cellf: From ~iophysies to Behuoior (Edited by Koester J. and Byrne J. H.), pp. 125-133, Cold Spring Harbor, New York. Hodgkin A. L. and Huxley A. F. (1952) A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. 117, 500-544. Jansen J. K. S. and Nicholls J. G. (1973) Conductance changes, an electrogenic pump and the h~rpola~zation of leech neurons following impulses. J. Physioi. 229, 635-655. Johansen J. and Kleinhaus A. L. (1985a) A monosynaptic connection between the medial nociceptive and the nut cell in leech ganglia. J. Comp. Physiol. 156, 65-69. Johansen J. and Kieinhaus A. L. (1985b) Properties of action potentials carried by divalent cations in identified leech neurons. f. Comp. Physiot. 157, 491-497. Johansen J. and Kleinhaus A. L. (1986a) Differential sensitivity to tetrodotoxin of nociceptive neurons in four species of leeches. J. Neurosci. 6, 3499-3504. Johansen J. and Kleinhaus A. L. (1986b) Transient and delayed potassium currents in the Retzius cell of the leech, ~acrobdelia
decora. J. Neurophys~oi. 36, 8 12622.
Johansen J. and Kleinhaus A. L. (1987) The effects of procaine, strychnine and penicillin on nociceptive neurons in leech segmental ganglia. Comp. B&hem. Physiol. WC, 405-409.
Johansen J. and Kleinhaus A. L. (1989) Voltage clamp analysis of ionic conductances in two types of sensory neurons in the leech. Sot. Neurosei. Abstr. 15, 1146. Johansen J., Hockfield S. and McKay R. D. G. (1984a) Distribution and morphology of nociceptive cells in the central nervous system of three species of leeches. J. Comp. Neurol. 226, 263-273. Johansen J., Yang J. and Kleinhaus A. L. (1984b) Actions of procaine on specific nociceptive cells in leech central nervous system. J. Neurosci. 4, 1253-1261. Johansen J., Taft W. C., Yang J., Kleinhaus A. L. and DeLorenzo R. J. (1985) Inhibition of Ca*+ conductance in identified lee& ne&ons by benzodiazepines. Proc. Natf. Acad. Sci. USA 82, 3935-3939. Johansen J., Yang J. and Kleinhaus A. L. (1987) Voltageclamp analysis of the ionic conductances in a leech neuron with a purely calcium-dependent action potential. J. Neuroph vsiol. 58, 1468-l 484. Kleinhaus -A: L. (19?6) Divalent cations and the action potential of the leech Retzius cell. PJzZgers Arch. 363, 97-104. Kleinhaus A. L. (1980) Segregation of leech neurones by the effect of sparteine on action potential duration. J. Physiol. 299, 309-321.
582
JBRGENJOHANSEN and ANNA L. KLEINHAIJS
Kleinhaus A. L. and Prichard J. W. (1983) Differential action of tetrodotoxin on identified leech neurons. Camp. Biochem. Physiol. 74C, 2 1l-2 18. Nicholls J. G. and Baylor D. A. (1968) Specific modalities and receptive fields of sensory neurons in the CNS of the leech. J. Neurophysiol. 31, 740-756. Ready D. F. and Nicholls J. G. (1979) Identified neurones from leech CNS make selective connections in culture. Nature 281, 67-69.
Stewart R. R., Nicholls J. G. and Adams W. B. (1989) Na+,
K+, and Ca2+ currents in identified leech neurones in culture. J. Exp. Biol. 141, l-20. Valkanov M. and Boev K. (1988) Ionic currents in the somatic membrane of identified T-mechanosensory neurons from segmental ganglia of the medicinal leech. Gen. Physiol. Biophys. 7, 643-649. Yang J. and Kleinhaus A. L. (1984) Effects of tetraethylammonium-chloride and divalent cations on the afterhyperpolarization following repetitive firing in the leech. Brain Res. 311, 380-384.