- Email: [email protected]
Intracellular recordings from two cell types in an in vitro preparation of the salamander olfactory epithelium
Neuroscience Letters, 35 (1983) 59-64
59
Elsevier Scientific Publishers Ireland Ltd.
I N T R A C E L L U L A R R E C O R D I N G S F R O M TWO CELL T Y P E S IN A N IN VITRO P R E P A R A T I O N OF T H E S A L A M A N D E R O L F A C T O R Y EPITHELIUM
LEONA M. MASUKAWA, JOHN S. KAUER and GORDON M. SHEPHERD
Sections of Neuroanatomy and Neurosurgery, Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06510 (U.S.A.) (Received September 17th, 1982; Revised version received October 28th, 1982; Accepted November 23rd, 1982)
Two electrophysiologically distinct cell types were found with intracellular recordings for the first time in an in vitro preparation of the olfactory epithelium of the salamander, Arnbystoma tigrinum. Intracellular recordings showed that Type I cells did not discharge action potentials but had high resting membrane potentials ( - 50 to - 104 mV) and relatively low input resistances. Type I1 cells had resting membrane potentials of - 24 to - 52 mV, high input resistances, and discharged upon penetration and to depolarizing current steps. The discharge pattern of Type II cells showed the following characteristics: (1) decreased spike latency and increased discharge frequency with increasing current step intensity; (2) relatively slowly adapting spike trains; and (3) varying spike amplitude during repetitive discharges. The superficial location in the epithelium of the Type I cells implies that they may be sustentacular cells with glial-like electrophysiological properties. The Type II cells are presumably olfactory receptor cells, based on the characteristics of their spike discharge to depolarizing current and their intermediate location within the olfactory epithelium.
The vertebrate olfactory epithelium is the site of transduction of odor molecules into neural signals. Unfortunately, information about the functional properties of the epithelial cells is very limited. It is known that the epithelium responds to odor stimuli by generating an electrical field potential called the electro-olfactogram [11]. Extracellular spike discharges, presumably recorded from receptor cell bodies or axons, are present in response to odor stimuli [2, 3, 5, 6, 8]. There are preliminary observations of intracellular responses of cells in the salamander to odor stimuli [1, 4], and there has been a study of lamprey receptor cells [13]. The intraceUular studies have been limited by the difficulties of recording from the small cells in the epithelium. An in vitro preparation would provide a significant improvement in the ability to record from epithelial cells and analyze their properties. In addition to better physical stability, it would provide the opportunity to carry out experiments under defined biochemical and pharmacological conditions. Building on recent experience 0304-3940/83/0000-0000/$ 03.00 © 1983 Elsevier Scientific Publishers Ireland Ltd.
60
in developing an in vitro preparation of the olfactory bulb [10], we have begun development of a similar preparation of the olfactory epithelium in the salamander. We report here that intracellular recordings can be obtained routinely, and that they reveal the existence of two different cell types, presumably receptor cells and supporting cells. This is the first study in which the membrane properties of these two cell types could be analyzed and compared. As in most other vertebrates, the olfactory epithelium in the salamander is composed of three main cell types: sustentacular (supporting) cells, receptor cells and basal cells [7]. The sustentacular cell bodies are located below the superficial mucus layer, and their processes extend to the basal lamina. Positioned in the middle portion of the epithelium are the mature receptor cell bodies, which send dendrites toward the surface and axons through the basal lamina. At the base of the epithelium are immature receptor cells, or basal cells. Salamanders (Ambystorna tigrinum) were anesthetized by cooling and the olfactory sac was exposed as described by Kauer [9]. Segments from the dorsal wall of the olfactory sac were dissected out and were pinned, mucus side up, in a recording chamber perfused with Ringer solution (in mM; NaC1, 104; KCI, 2.1; CaCI2, 3.6; MgCl2, 0.7; NaHCO3, 26; and glucose, 5.0) at r o o m temperature. The solution, bubbled with 95070 02 and 5O7o CO2, was maintained at pH 7.6. Recording micropipettes filled with 4 M potassium acetate and having impedances of 100-300 Mfi were pulled using the P-77 Brown-Flaming Micropipette Puller (Sutter Instrument Co., CA). For determining electrical properties of the cells, current was passed through the recording pipette using an active bridge circuit. When cells were penetrated (as indicated by an abrupt change in the DC potential), the depth was recorded with reference to the first contact of the pipette tip with the epithelial surface. Depth measurements were read f r o m the digital display of a micro-step apparatus driving the micropipette. An advantage of this preparation, when compared to other in vitro preparations of neuron tissue, is that physical stress on the tissue is minimal, and correlated with this was the fact that recordings could be made iramediately after the tissue was removed and placed in the chamber, without need for a period of 'rest'. Most preparations remained viable for 6 - 8 h, and yielded recordings from a number of cells. Two m a j o r groups of cells were found (see Table I). Type I cells were usually observed soon after entering the epithelium, but a few of these were also encountered deeper, down to 350 #m, a depth which is approximately equivalent to the thickness of the epithelium, from mucus surface to basal lamina. Some error was present in these depth measurements because of dimpling by the pipette, and because it was not feasible to flatten completely the tissue in the recording chamber (see below). Type I cells ranged in resting membrane potential from - 5 0 to - 104 mV upon penetration, and could be held for at least 15 min in most cases. The input resistances were low, ranging from too small to measure to 30 Mfi, and were characteristically less than 10 Mfl. The input resistances were measured by passing
61
TABLE I POSITION WITHIN OLFACTORY EPITHELIUM CHARACTERISTICS OF TWO CELL TYPES
AND
ELECTROPHYSIOLOGICAL
Resting membrane potentials are mean values for n cells. For input resistance measurements, the bridge imbalances were taken into account when present due to electrode rectification. Type II cells were hyperpolarized with steady current while input resistances were measured. Cell type
Depth
Resting membrane potential
Input resistances
Spiking present
n
1 II
0-350/zm 70-300 #m
- 8 6 mV - 36 mV
0-30 MI2 165-600 Mfl
+
19 13
a hyperpolarizing current step and recording the plateau of the voltage change produced. Action potentials were never observed in these cells when the membrane potential was depolarized by injected current, nor did they occur spontaneously. A second, distinct cell type, Type II, was located at a somewhat greater depth than Type I. Type II cells were seen between 70 and 300/~m. Although this overlaps with the distribution of Type I cells, this type was never found as superficially as the Type I cells, and there was the clear impression of a more deeply situated population. These cells often were observed to discharge spikes upon penetration by the recording electrode. The membrane potential ranged from - 24 to - 52 mV. Ceils of this type could usually be held for less than 5 min; however, in a few cases stable recordings lasted up to 1 h. The input resistance was measured while the membrane was hyperpolarized by steady background current to no more than 20 mV above the initial resting potential in order to bring the membrane potential nearer to that of the Type I cells. Resistances with this technique were substantially higher than Type I cells, ranging from 165 to 600 Mfl. When depolarizing current steps were superimposed on the holding current, spike activity was invariably seen. Fig. 1 illustrates impulse responses to current steps. Spike latency can be seen to decrease with increasing current intensity, while the frequency of the spikes increased. Initial action potentials were on the order o f 60-100 mV in amplitude, and were followed by transient afterhyperpolarizations. During repetitive firing at moderately strong depolarizations, as in Fig. 1B, the initial action potential was followed by smaller amplitude spikes. The spike discharge tended to be relatively slowly adapting, with the cell continuing to fire impulses in response to prolonged depolarizing pulses of moderate (Fig. 1B) to strong (Fig. 1C) intensity. Spike amplitudes decreased further during higher depolarizing current steps, as in Fig. 1C. In a few instances an action potential was observed to occur just after the termination of the current step (see Fig. 1D). With regard to the identification of the cell types, the relatively more superficial recording depths for Type I cells are in accord with the locations of the sustentacular
62 ceils in the epithelium. In a d d i t i o n , this cell type n ev er showed a c t i o n potentials. W e t h e r e f o r e tentatively i d e n t i f y the T y p e I cell as a sustentacular cell. It is o f interest to n o t e t h a t the high resting m e m b r a n e p o t e n t i a l s and very low i n p u t resistances o f T y p e I cells are w e l l - k n o w n characteristics o f glial cells [1 l]. Thus, these experiments p r o v i d e evidence consistent with G e t c h e l l ' s [4] suggestion that the sustentacular cells h a v e glial-like properties. T h e d ep t h d i s t r i b u t i o n o f T y p e II cells is consistent with the i n t e r p r e t a t i o n that these are recordings f r o m r e c e p t o r cells. T h e fact that this type was never r e c o r d e d within 7 0 / zm o f the surface suggests that the recordings were p r o b a b l y never f r o m the dendrites o f these cells, but rather f r o m their cell bodies. T h e m o s t n o t a b l e prope r t y o f this cell type was the ability to fire a c t i o n potentials in response to d e p o l a r i z i n g c u r r e n t injection, as expected for the o l f a c t o r y r e c e p t o r cell, which W
Fig. 1. Response of Type II cells to depolarizing current steps. A, B and C: discharge frequency increased with each current step. The action potential amplitude was seen to change during a repetitive discharge. Successive spikes within a train were always smaller in magnitude than the initial event. (The latency of the first action potential decreased as the current intensity was increased.) D: although action potentials usually occurred during a current step, this record shows a spike following the termination of the step. Spontaneous discharge was not seen in this cell. The top traces and bottom traces of frames A, B and C are voltage and current records, respectively. In D, the top trace is the current record and the bottom trace the voltage record. It was not always possible to maintain the bridge in balance due to the high resistances of the microelectrodes and additional electrode tip rectification during current passage. Calibrations: A, B and C, 20 mV, 0.066 nA and 50 msec; D, 20 mV, 0.066 nA and 20 msec.
63 gives rise to an unmyelinated axon that connects to the olfactory bulb (cf. ref. 7). Our results show that this cell type responds to depolarizing current with a repetitive impulse discharge which is relatively slowly adapting. Electrophysiological data similar to ours were obtained from identified olfactory receptor cells in lamprey [13], and such data tend to support our proposal that Type II cells are receptor cells in the salamander. This similarity also gives evidence that the receptor properties remain normally viable in the in vitro preparation. The finding of a slowly-adapting impulse discharge is further in accord with the prolonged intracellular spike firing discharges in response to odor stimulation that Getchell [4] recorded from presumed receptor cells, and also with the sustained discharges in response to prolonged odor pulses that have been recorded extracellularly from presumed receptor cells (see especially refs. 5 and 6). This implies that the depolarization produced by current passage mimicked the generator potential in the receptor cell in a qualitative way. Our data also indicate that the site of impulse generation is likely to be located at some distance from the site of electrode penetration and current passage in these cells. The penetrations were presumably in the cell bodies, from considerations of recording depths mentioned above, and from the small dimensions of the dendrites (approximately 1.0/zm) and axons (0.2/zm) compared with the cell bodies (9 × 18 /zm; see ref. 4). Compatible with a distant locus of impulse generation were the attenuated action potentials recorded during large current steps (see Fig. 1B, C), and the delayed action potentials that occasionally occurred after cessation of the current (Fig. 1D). Both of the observations can be explained by the invasion of action potentials generated at a distance (i.e. in the axon) from the cell body. The attenuation may also signal a turning on of a slow conductance in the cell membrane within a short period after the current step begins. The resulting decreased input resistance would act as a shunt for a passively-conducted spike. This is the first report of an in vitro preparation for intracellular analysis of the olfactory epithelium. The data strongly indicate recordings from two distinctly different cell populations which are consistent with supporting cells and olfactory receptor cells. The use of intracellular recordings from an in vitro preparation will allow controlled application of odors and molecules through the aqueous phase for examination of mechanisms of sensory transduction. In addition, it should also permit a thorough study of the physiological properties of receptor cells during all stages of development, turnover and regeneration. This work was supported by Research Grants NS-07609 and NS-10174 of the National Institute for Neurological and Communicative Disorders and Stroke.
64 1 Aoki, K. and Tagaki, S.F., lntracellular recordings of the olfactory cell activity, Proc. Jap. Acad., 44 (1968) 856-857. 2 Gesteland, R.C., Lettvin, J.Y. and Pitts, W.H., Chemical transmission ii~_the nose of the frog, J. Physiol. (Lond.), 181 (1965) 525-559. 3 Getchell, T.V., Analysis of unitary spikes recorded extracellularly from frog olfactory receptor cells and axons, J. Physiol. (Lond.), 234 (1973) 533-551. 4 Getchell, T.V., Analysis of intracellular recordings from salamander olfactory epithelium, Brain Res., 123 (1977) 275-286. 5 Getchell, T.V. and Shepherd, G.M., Responses of olfactory receptor cells to step pulses of odour at different concentrations in the salamander, J. Physiol. (Lond.), 282 (1978) 521-540. 6 Getchell, T.V. and Shepherd, G.M., Adaptive properties of olfactory receptors analyzed with odour pulses of varying durations. J. Physiol. (Lond.), 282 (1978) 541-560. 7 Graziadei, P.P.C. and Monti Graziadei, G.A., Olfactory epithelium of Necturus maculous and Ambystoma tigrinum, J. Neurocytol., 5 (1976) 11-32. 8 Holley, A., Duchamp, A., Rivial, M.F., Juge, A. and MacLeod, P., Qualitative and quantitative discrimination in the frog olfactory receptors: analysis from electrophysiological data, Ann. N.Y. Acad. Sci., 237 (1974) 102-114. 9 Kauer, J.S., Response patterns of amphibian olfactory bulb to odour stimulation, J. Physiol. (kond.), 243 (1974) 675-715. 10 Mori, K., Nowycky, M.C. and Shepherd, G.M., Electrophysiological analysis of mitral cells in the isolated turtle olfactory bulb, J. Physiol. (Lond.), 314 (1981) 281-294. / 1 0 t t o s o n , D., The electro-olfactogram. In L.M. Beidler (Ed.), Handbook of Sensory Physiology, Vol. 4, Chemical Senses, Part 1: Olfaction, Springer-Verlag, New York, 1971. 12 Somjen, G.G., Electrophysiology of neuroglia, Ann. Rev. Physiol., 37 (1975) 163- 170. 13 Suzuki, N., lntracellular responses of lamprey olfactory receptors to current and chemical stimulation. In Y. Katsuki, M. Sato, S. Takagi and Y. Omura (Eds.), Food Intake of Chemical Senses, Jap. Sci. Soc. Press, Japan, 1977.
59
Elsevier Scientific Publishers Ireland Ltd.
I N T R A C E L L U L A R R E C O R D I N G S F R O M TWO CELL T Y P E S IN A N IN VITRO P R E P A R A T I O N OF T H E S A L A M A N D E R O L F A C T O R Y EPITHELIUM
LEONA M. MASUKAWA, JOHN S. KAUER and GORDON M. SHEPHERD
Sections of Neuroanatomy and Neurosurgery, Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06510 (U.S.A.) (Received September 17th, 1982; Revised version received October 28th, 1982; Accepted November 23rd, 1982)
Two electrophysiologically distinct cell types were found with intracellular recordings for the first time in an in vitro preparation of the olfactory epithelium of the salamander, Arnbystoma tigrinum. Intracellular recordings showed that Type I cells did not discharge action potentials but had high resting membrane potentials ( - 50 to - 104 mV) and relatively low input resistances. Type I1 cells had resting membrane potentials of - 24 to - 52 mV, high input resistances, and discharged upon penetration and to depolarizing current steps. The discharge pattern of Type II cells showed the following characteristics: (1) decreased spike latency and increased discharge frequency with increasing current step intensity; (2) relatively slowly adapting spike trains; and (3) varying spike amplitude during repetitive discharges. The superficial location in the epithelium of the Type I cells implies that they may be sustentacular cells with glial-like electrophysiological properties. The Type II cells are presumably olfactory receptor cells, based on the characteristics of their spike discharge to depolarizing current and their intermediate location within the olfactory epithelium.
The vertebrate olfactory epithelium is the site of transduction of odor molecules into neural signals. Unfortunately, information about the functional properties of the epithelial cells is very limited. It is known that the epithelium responds to odor stimuli by generating an electrical field potential called the electro-olfactogram [11]. Extracellular spike discharges, presumably recorded from receptor cell bodies or axons, are present in response to odor stimuli [2, 3, 5, 6, 8]. There are preliminary observations of intracellular responses of cells in the salamander to odor stimuli [1, 4], and there has been a study of lamprey receptor cells [13]. The intraceUular studies have been limited by the difficulties of recording from the small cells in the epithelium. An in vitro preparation would provide a significant improvement in the ability to record from epithelial cells and analyze their properties. In addition to better physical stability, it would provide the opportunity to carry out experiments under defined biochemical and pharmacological conditions. Building on recent experience 0304-3940/83/0000-0000/$ 03.00 © 1983 Elsevier Scientific Publishers Ireland Ltd.
60
in developing an in vitro preparation of the olfactory bulb [10], we have begun development of a similar preparation of the olfactory epithelium in the salamander. We report here that intracellular recordings can be obtained routinely, and that they reveal the existence of two different cell types, presumably receptor cells and supporting cells. This is the first study in which the membrane properties of these two cell types could be analyzed and compared. As in most other vertebrates, the olfactory epithelium in the salamander is composed of three main cell types: sustentacular (supporting) cells, receptor cells and basal cells [7]. The sustentacular cell bodies are located below the superficial mucus layer, and their processes extend to the basal lamina. Positioned in the middle portion of the epithelium are the mature receptor cell bodies, which send dendrites toward the surface and axons through the basal lamina. At the base of the epithelium are immature receptor cells, or basal cells. Salamanders (Ambystorna tigrinum) were anesthetized by cooling and the olfactory sac was exposed as described by Kauer [9]. Segments from the dorsal wall of the olfactory sac were dissected out and were pinned, mucus side up, in a recording chamber perfused with Ringer solution (in mM; NaC1, 104; KCI, 2.1; CaCI2, 3.6; MgCl2, 0.7; NaHCO3, 26; and glucose, 5.0) at r o o m temperature. The solution, bubbled with 95070 02 and 5O7o CO2, was maintained at pH 7.6. Recording micropipettes filled with 4 M potassium acetate and having impedances of 100-300 Mfi were pulled using the P-77 Brown-Flaming Micropipette Puller (Sutter Instrument Co., CA). For determining electrical properties of the cells, current was passed through the recording pipette using an active bridge circuit. When cells were penetrated (as indicated by an abrupt change in the DC potential), the depth was recorded with reference to the first contact of the pipette tip with the epithelial surface. Depth measurements were read f r o m the digital display of a micro-step apparatus driving the micropipette. An advantage of this preparation, when compared to other in vitro preparations of neuron tissue, is that physical stress on the tissue is minimal, and correlated with this was the fact that recordings could be made iramediately after the tissue was removed and placed in the chamber, without need for a period of 'rest'. Most preparations remained viable for 6 - 8 h, and yielded recordings from a number of cells. Two m a j o r groups of cells were found (see Table I). Type I cells were usually observed soon after entering the epithelium, but a few of these were also encountered deeper, down to 350 #m, a depth which is approximately equivalent to the thickness of the epithelium, from mucus surface to basal lamina. Some error was present in these depth measurements because of dimpling by the pipette, and because it was not feasible to flatten completely the tissue in the recording chamber (see below). Type I cells ranged in resting membrane potential from - 5 0 to - 104 mV upon penetration, and could be held for at least 15 min in most cases. The input resistances were low, ranging from too small to measure to 30 Mfi, and were characteristically less than 10 Mfl. The input resistances were measured by passing
61
TABLE I POSITION WITHIN OLFACTORY EPITHELIUM CHARACTERISTICS OF TWO CELL TYPES
AND
ELECTROPHYSIOLOGICAL
Resting membrane potentials are mean values for n cells. For input resistance measurements, the bridge imbalances were taken into account when present due to electrode rectification. Type II cells were hyperpolarized with steady current while input resistances were measured. Cell type
Depth
Resting membrane potential
Input resistances
Spiking present
n
1 II
0-350/zm 70-300 #m
- 8 6 mV - 36 mV
0-30 MI2 165-600 Mfl
+
19 13
a hyperpolarizing current step and recording the plateau of the voltage change produced. Action potentials were never observed in these cells when the membrane potential was depolarized by injected current, nor did they occur spontaneously. A second, distinct cell type, Type II, was located at a somewhat greater depth than Type I. Type II cells were seen between 70 and 300/~m. Although this overlaps with the distribution of Type I cells, this type was never found as superficially as the Type I cells, and there was the clear impression of a more deeply situated population. These cells often were observed to discharge spikes upon penetration by the recording electrode. The membrane potential ranged from - 24 to - 52 mV. Ceils of this type could usually be held for less than 5 min; however, in a few cases stable recordings lasted up to 1 h. The input resistance was measured while the membrane was hyperpolarized by steady background current to no more than 20 mV above the initial resting potential in order to bring the membrane potential nearer to that of the Type I cells. Resistances with this technique were substantially higher than Type I cells, ranging from 165 to 600 Mfl. When depolarizing current steps were superimposed on the holding current, spike activity was invariably seen. Fig. 1 illustrates impulse responses to current steps. Spike latency can be seen to decrease with increasing current intensity, while the frequency of the spikes increased. Initial action potentials were on the order o f 60-100 mV in amplitude, and were followed by transient afterhyperpolarizations. During repetitive firing at moderately strong depolarizations, as in Fig. 1B, the initial action potential was followed by smaller amplitude spikes. The spike discharge tended to be relatively slowly adapting, with the cell continuing to fire impulses in response to prolonged depolarizing pulses of moderate (Fig. 1B) to strong (Fig. 1C) intensity. Spike amplitudes decreased further during higher depolarizing current steps, as in Fig. 1C. In a few instances an action potential was observed to occur just after the termination of the current step (see Fig. 1D). With regard to the identification of the cell types, the relatively more superficial recording depths for Type I cells are in accord with the locations of the sustentacular
62 ceils in the epithelium. In a d d i t i o n , this cell type n ev er showed a c t i o n potentials. W e t h e r e f o r e tentatively i d e n t i f y the T y p e I cell as a sustentacular cell. It is o f interest to n o t e t h a t the high resting m e m b r a n e p o t e n t i a l s and very low i n p u t resistances o f T y p e I cells are w e l l - k n o w n characteristics o f glial cells [1 l]. Thus, these experiments p r o v i d e evidence consistent with G e t c h e l l ' s [4] suggestion that the sustentacular cells h a v e glial-like properties. T h e d ep t h d i s t r i b u t i o n o f T y p e II cells is consistent with the i n t e r p r e t a t i o n that these are recordings f r o m r e c e p t o r cells. T h e fact that this type was never r e c o r d e d within 7 0 / zm o f the surface suggests that the recordings were p r o b a b l y never f r o m the dendrites o f these cells, but rather f r o m their cell bodies. T h e m o s t n o t a b l e prope r t y o f this cell type was the ability to fire a c t i o n potentials in response to d e p o l a r i z i n g c u r r e n t injection, as expected for the o l f a c t o r y r e c e p t o r cell, which W
Fig. 1. Response of Type II cells to depolarizing current steps. A, B and C: discharge frequency increased with each current step. The action potential amplitude was seen to change during a repetitive discharge. Successive spikes within a train were always smaller in magnitude than the initial event. (The latency of the first action potential decreased as the current intensity was increased.) D: although action potentials usually occurred during a current step, this record shows a spike following the termination of the step. Spontaneous discharge was not seen in this cell. The top traces and bottom traces of frames A, B and C are voltage and current records, respectively. In D, the top trace is the current record and the bottom trace the voltage record. It was not always possible to maintain the bridge in balance due to the high resistances of the microelectrodes and additional electrode tip rectification during current passage. Calibrations: A, B and C, 20 mV, 0.066 nA and 50 msec; D, 20 mV, 0.066 nA and 20 msec.
63 gives rise to an unmyelinated axon that connects to the olfactory bulb (cf. ref. 7). Our results show that this cell type responds to depolarizing current with a repetitive impulse discharge which is relatively slowly adapting. Electrophysiological data similar to ours were obtained from identified olfactory receptor cells in lamprey [13], and such data tend to support our proposal that Type II cells are receptor cells in the salamander. This similarity also gives evidence that the receptor properties remain normally viable in the in vitro preparation. The finding of a slowly-adapting impulse discharge is further in accord with the prolonged intracellular spike firing discharges in response to odor stimulation that Getchell [4] recorded from presumed receptor cells, and also with the sustained discharges in response to prolonged odor pulses that have been recorded extracellularly from presumed receptor cells (see especially refs. 5 and 6). This implies that the depolarization produced by current passage mimicked the generator potential in the receptor cell in a qualitative way. Our data also indicate that the site of impulse generation is likely to be located at some distance from the site of electrode penetration and current passage in these cells. The penetrations were presumably in the cell bodies, from considerations of recording depths mentioned above, and from the small dimensions of the dendrites (approximately 1.0/zm) and axons (0.2/zm) compared with the cell bodies (9 × 18 /zm; see ref. 4). Compatible with a distant locus of impulse generation were the attenuated action potentials recorded during large current steps (see Fig. 1B, C), and the delayed action potentials that occasionally occurred after cessation of the current (Fig. 1D). Both of the observations can be explained by the invasion of action potentials generated at a distance (i.e. in the axon) from the cell body. The attenuation may also signal a turning on of a slow conductance in the cell membrane within a short period after the current step begins. The resulting decreased input resistance would act as a shunt for a passively-conducted spike. This is the first report of an in vitro preparation for intracellular analysis of the olfactory epithelium. The data strongly indicate recordings from two distinctly different cell populations which are consistent with supporting cells and olfactory receptor cells. The use of intracellular recordings from an in vitro preparation will allow controlled application of odors and molecules through the aqueous phase for examination of mechanisms of sensory transduction. In addition, it should also permit a thorough study of the physiological properties of receptor cells during all stages of development, turnover and regeneration. This work was supported by Research Grants NS-07609 and NS-10174 of the National Institute for Neurological and Communicative Disorders and Stroke.
64 1 Aoki, K. and Tagaki, S.F., lntracellular recordings of the olfactory cell activity, Proc. Jap. Acad., 44 (1968) 856-857. 2 Gesteland, R.C., Lettvin, J.Y. and Pitts, W.H., Chemical transmission ii~_the nose of the frog, J. Physiol. (Lond.), 181 (1965) 525-559. 3 Getchell, T.V., Analysis of unitary spikes recorded extracellularly from frog olfactory receptor cells and axons, J. Physiol. (Lond.), 234 (1973) 533-551. 4 Getchell, T.V., Analysis of intracellular recordings from salamander olfactory epithelium, Brain Res., 123 (1977) 275-286. 5 Getchell, T.V. and Shepherd, G.M., Responses of olfactory receptor cells to step pulses of odour at different concentrations in the salamander, J. Physiol. (Lond.), 282 (1978) 521-540. 6 Getchell, T.V. and Shepherd, G.M., Adaptive properties of olfactory receptors analyzed with odour pulses of varying durations. J. Physiol. (Lond.), 282 (1978) 541-560. 7 Graziadei, P.P.C. and Monti Graziadei, G.A., Olfactory epithelium of Necturus maculous and Ambystoma tigrinum, J. Neurocytol., 5 (1976) 11-32. 8 Holley, A., Duchamp, A., Rivial, M.F., Juge, A. and MacLeod, P., Qualitative and quantitative discrimination in the frog olfactory receptors: analysis from electrophysiological data, Ann. N.Y. Acad. Sci., 237 (1974) 102-114. 9 Kauer, J.S., Response patterns of amphibian olfactory bulb to odour stimulation, J. Physiol. (kond.), 243 (1974) 675-715. 10 Mori, K., Nowycky, M.C. and Shepherd, G.M., Electrophysiological analysis of mitral cells in the isolated turtle olfactory bulb, J. Physiol. (Lond.), 314 (1981) 281-294. / 1 0 t t o s o n , D., The electro-olfactogram. In L.M. Beidler (Ed.), Handbook of Sensory Physiology, Vol. 4, Chemical Senses, Part 1: Olfaction, Springer-Verlag, New York, 1971. 12 Somjen, G.G., Electrophysiology of neuroglia, Ann. Rev. Physiol., 37 (1975) 163- 170. 13 Suzuki, N., lntracellular responses of lamprey olfactory receptors to current and chemical stimulation. In Y. Katsuki, M. Sato, S. Takagi and Y. Omura (Eds.), Food Intake of Chemical Senses, Jap. Sci. Soc. Press, Japan, 1977.