Membrane properties of interneurons in stratum oriens-alveus of the CA1 region of rat hippocampus in vitro

Neuroscience Vol. 36, No. 2, pp. 349-359, 1990

0306-4522/90 $3.00+ 0.00 Pergamon Press plc 1990 IBRO

Printed in Great Britain

MEMBRANE PROPERTIES OF INTERNEURONS IN STRATUM ORIENS-ALVEUS OF THE CA1 REGION OF RAT HIPPOCAMPUS IN VITRO J.-C. LACAILLEand S. WILLIAMS Centre de Recherche en Sciences Neurologiques et D6partement de Physiologie, Universit6 de Montr6al, Montr6al, Qu6bec, Canada H3C 3J7 Abstract--The membrane properties of interneurons situated near the border of stratum oriens and the

alveus of the CA1 region were examined with intracellular recording and staining in rat hippocampal slices in vitro. Cellular staining with Lucifer Yellow indicated that the somata of these interneurons were multipolar and their dendrites projected horizontally along the alveus and vertically toward stratum lacunosum-moleculare. Intrinsic properties (input resistance, action potential amplitude, time constant) and spike afterpotentials were typical of non-pyramidal cells. Action potential duration, however, was of relatively medium duration (1.15 ms) and slow afterhyperpolarizations followed depolarization-induced trains of action potentials. Spontaneous activity of interneurons was prominent and of either of two types: single action potentials or high frequency bursts of action potentials. Interneurons displayed marked, voltageand time-dependent inward rectification and anodal break excitation. Analysis of the slope of the charging function of hyperpolarizing transients, suggested that these interneurons were electrically compact (dendrite to soma conductance ratio, p ~ 2.7; and electrotonic length constant, L ,~ 1.1). Characteristically, interneurons sustained high frequency repetitive firing during long depolarizing pulses. The slope of the frequency-current relation was 442 Hz/nA for the first interspike interval and 117 Hz/nA for later intervals (no. 60), suggesting the presence of spike frequency adaptation. Physiologically, these interneurons resembled more closely basket cells of stratum pyramidale than stellate cells of stratum lacunosum-moleculare.

G A B A is the principal inhibitory neurotransmitter at supraspinal levels in the vertebrate CNS. z2~ In hippocampus and cortex, GABAergic interneurons play a crucial role in maintaining an appropriate level of cellular excitability by counterbalancing the intrinsic bursting properties of pyramidal cells and their excitatory inputs. 35'43'5°-52'54GABAergic interneurons achieve this mainly by hyperpolarizing pyramidal cells via two postsynaptic mechanisms. Firstly, through recurrent and feedforward inhibition produced via postsynaptic GABAA receptors and the opening of C1- channels--the early inhibitory postsynaptic potential (early IPSP). 3 6,9,1336,34,44Secondly, by way of feedforward inhibition generated via postsynaptic GABAB receptors and the opening of K + channels--the late IPSP. 1'3'1° 12,19.33,34,49 Despite the wealth of information concerning the postsynaptic actions of GABA, relatively little is Abbreviations: ACSF, artificial cerebrospinal fluid; AHP,

afterhyperpolarization; AP, action potential; B-cells, basket cells; EPSP, excitatory postsynaptic potential; fAHP, fast AHP; IPSP, inhibitory postsynaptic potential; ISI, interspike interval; L, electrotonic length constant; LY, Lucifer Yellow; O-A, stratum oriens-alveus region; RMP, resting membrane potential; sAHP, slow AHP; S-cells, stellate cells; V-cells, cells with vertical dendrites; V - L voltage-current; p, dendrite to soma conductance ratio; z0, membrane time constant; rt, first equalizing time constant.

known about the interneurons mediating these actions. Two distinct physiological types of interneurons have been identified in the CA1 region of the hippocampus: basket cells (B-cells) situated near stratum pyramidale 7'17'18'42and aspinous stellate cells (S-cells) in stratum lacunosum-moleculare) 7'18~25 Their morphology and membrane properties are different from pyramidal cells) 7'25'42Interneurons display a prominent afterhyperpolarization (AHP) following individual action potentials, and, supposedly, little or no spike frequency adaptation during depolarizing current pulses. 25'42 Other distinctive membrane properties are also associated with each type of interneuron: for B-cells, short duration action potentials (APs), many spontaneous synaptic potentials and numerous spontaneous APs; for S-cells, longer duration APs, little spontaneous synaptic activity, no or very few spontaneous APs, and a voltage-dependent mode of firing similar to thalamic neurons.14 These two different types of interneurons may mediate the two varieties of pyramidal cell IPSPs (early and late IPSPs, respectively). 2~'27 Another group of interrieurons has also been identified in stratum oriens near the alveus junction (O-A). 24 Some of these interneurons also appear to mediate feedforward and recurrent inhibition of pyramidal cells. 24 Their membrane properties, however, have not been extensively characterized 24 and it 349

350

J.-C. LACA1LLEand S. WILLIAMS

remains unclear if they represent another distinct type o f interneuron. The purpose o f the present studies was to measure the m e m b r a n e properties o f interneurons in O - A to c o m p a r e them with the intrinsic properties o f other k n o w n interneuron types. We have also examined their passive cable properties and their repetitive firing behaviour. A preliminary report o f this work has appeared. 28 EXPERIMENTAL PROCEDURES

Slices Experiments were performed on male Sprague-Dawley rats (n = 26, Charles River Laboratories) weighing 150 225g. Slices were obtained as previously described.24'25 The animals were decapitated, the brain rapidly dissected out and cooled in ice-cold, oxygenated artificial cerebrospinal fluid (ACSF; see below). The hippocampus was isolated and cut in transverse slices (450-650 ttm thickness) with a mechanical tissue chopper (McIlwain). Slices were transferred to a nylon net of a gas-fluid interface chamber where their undersides were bathed in ACSF and their upper surfaces exposed to warm, humidified 95% 02/5% CO2 atmosphere. ACSF consisted of (in raM): NaCI, 124; KC1, 5; NaH2PO4, 1.25; MgSO4, 2; CaC12, 2; NaHCO 3, 26; dextrose, 10. ACSF was saturated with 95% 05/5% CO2 and continuously perfused at 1 ml/min. Slices and ACSF were maintained at 35 + 0.5°C with a feedback-controlled heater. Intracellular recordings Micropipettes were pulled from fiber-filled capillary tubing of borosilicate glass with a Brown-Flaming micropipette puller (Sutter Instr. P-80). Micropipettes (40-75 MI)) were filled with 4 M potassium acetate and 0.01 M KCI. For intracellular recordings from O-A interneurons, the micropipette was positioned under visual guidance in the superior zone of stratum oriens at the junction with the alveus. The micropipette was advanced in the slice in small steps with a motorized hydraulic microdrive (David Kopf) until neurons were encountered. CA1 pyramidal cell recordings were obtained routinely by positioning the micropipettes in stratum pyramidale. Intracellular responses were recorded with intracellular recording amplifiers (Neuro Data IR-283) equipped with an active bridge circuit for current injection, displayed on a digital storage oscilloscope (Gould 1604) and stored in digitized format on a video cassette recorder (Neuro Corder DR 886) for later retrieval and analysis. Intracellular responses were evoked with a stimulator (WPI A300 series) connected to the bridge circuit. Because of the low density of interneurons, their recordings were more difficult to obtain. However, once interneurons were impaled, the stability of recordings was comparable to pyramidal cells. Recordings were judged acceptable if cell responses were stable without the aid of steady hyperpolarizing current injection. Intracellular recordings were accepted as interneuron recordings if a large AHP followed individual action potentials [in pyramidal cells a depolarizing afterpotential follows a brief fast AHP (fAHP)]fl 4'39'48 Intradendritic recordings of pyramidal cells were rarely encountered in O-A region. Occasional pyramidal cell dendritic penetrations displayed characteristic features, easily distinguishable from interneuron responses.24,26.55 Intr&sie properties To allow comparisons with other interneuron types, the following intracellular responses were measured. 7A7,24.25,42 Resting membrane potential (RMP) was taken on withdrawal from the cell. AP duration and the fAHP following individual APs were measured from averages of four or

eight traces with the cell firing at 1-3 Hz either spontaneously or during steady current injection. AP duration was measured at the base and at 50% amplitude of the AP. The fAHP amplitude was measured from the peak of the hyperpolarization and to both the baseline membrane potential and the base of the AP. The peak latency of the fAHP was taken as the difference in time between the base of the AP and the peak of the fAHP. The recovery time was measured as the time to decay from the peak of the fAHP to baseline membrane potential. Threshold was measured as the membrane potential at the base of the AP. The presence of a slower AHP (sAHP) following a train of APs was also evaluated. Care was taken to exclude the fAHP from these measures. APs were evoked with a 200 ms, 0.5 nA current pulse. The amplitude of the sAHP was taken as the difference between the peak of the hyperpolarization and baseline membrane potential. The sAHP peak latency was measured from the offset of the current pulse to the peak of the sAHP. The decay of the sAHP was measured from the peak to the return to baseline membrane potential. Voltage-current ( V - I ) relationship was examined from a family of responses to hyperpolarizing current pulses (200ms, -0.1 to - 1.0 nA) and cellular input resistance (RiO was measured as the slope of the linear regression. Repetitive firing behaviour was assessed during long depolarizing current pulses (700 ms) of increasing intensity (0.05-1.8 nA). Passive cable properties Membrane time constant and passive cable properties were estimated from responses to small hyperpolarizing current pulses (100 ms; -0.15 to -0.25 nA) according to the methods of Rail, 36Brown et al. s and Turner. 53 Responses (n = 4-8) were averaged with a digital oscilloscope and a waveform processor (Gould 1604). Responses were evoked within the voltage range corresponding to the linear portion of the V ~ relationship with the bridge carefully balanced. With the capacitance neutralization well adjusted, the intrinsic settling time of the electrodes was less than 250ps. Hyperpolarizing membrane responses (V) were transformed to charging function (V r - V). The charging function can be expressed as a sum of exponentials:s6

V f - V = Coe '~" + Ct e ,/T,+ . . .

(1)

where Vr is the final displacement in membrane potential, V is the time-dependent change in membrane potential, ro is the membrane time constant, ~1 the first equalizing time constant, and Co and C~ are coefficients. Higher order equalizing time constants were not extracted. The charging function was fitted with an exponential function using the least square methods and the parameters 30 and Co obtained from this function. Exponential peeling was performed according to Rall s6 to obtain the parameters t~ and Ca. Using Rall's model of a lumped soma with a dendritic cylinder, 8's6'sT's3the dendrite to soma conductance ratio (p) and the electrotonic length (L) were estimated. The dendrite to soma conductance ratio was obtained as described by Brown et al.: 8 p = [ro/Vf(Co/% + Ct/~l) ] - 1

(2)

The electrotonic length was estimated according to Rail: 36 lr ( P / ( P + 1)Y '2

(3)

Basic assumptions underlie Rail's model of a lumped soma and dendritic cylinder. Of primary concern are (i) that the membrane be passive and (ii) that the dendritic tree be represented as an equivalent uniform cylinder, s,36.37.53 To satisfy the first assumption our measures were taken within the linear portion of the V-1 relation. The second assumption is unlikely to be met since for these ceils the dendritic 3/2 power rule may not apply and since dendritic processes may not terminate at equal electrotonic distance from the

O-A interneuron membrane properties

351

soma. However, the model is still appropriate since in other hippocampal neurons (pyramidal and granule cells), despite the fact that these assumptions have not been met, the theoretically derived charging functions obtained from Rall's model closely approximate the experimentally observed somatic transients,s,53Finally, the equivalent cylinder analysis allows comparisons with other neuronal types also evaluated using Rail's model, s,53

IntraceUular injections of Lucifer Yellow Micropipettes were filled with 2-4% Lucifer Yellow (LY; Lucifer Yellow dilithium salt, Sigma) in l M LiC1.2s'47After cell penetration and a brief analysis of its cellular properties, LY was injected intracellularly (-0.5 to - 2 . 0 n A for 3 15 min). After injection the slice remained in the chamber for 10min. It was then fixed by immersion between two pieces of filter paper in 10% formaldehyde for 1 2 h. Slices were rinsed and kept in 0.1 M phosphate buffer (pH 7.4) at 4~C for 12-36 h. Slices were dehydrated in alcohols, cleared in methyl salicylate, mounted in DPX medium (BDH Chemicals), and examined with an epifluorescence microscope (Zeiss) equipped with appropriate filters. Cells were photographed and traced from the projected negatives to view cellular processes in their entirety.

B ALVEUS

.................

ORIENS

Statistics Statistical significance of difference between groups was assessed with Student's t-tests. Group measures are expressed as mean _+ S.D., unless otherwise noted. RESULTS

Interneuron recordings and morphology Intracellular recordings were obtained from 27 cells (in 26 slices) in the superior zone of stratum oriens near the alveus border. The distribution of these recordings within the CAI region is illustrated in Fig. I. Recordings were obtained throughout the C A l a - c extent. Six well-filled interneurons were recovered following electrophysiological characterization and intracellular injection of LY. An example of a LY-filled interneuron is shown in Fig. 1. Somata were multipolar and the mean soma size was 20 + 6.0 #m. Dendritic processes projected horizontally along the alveus but also vertically toward stratum pyramidale. The vertically oriented (ascending) dendrites were followed into stratum lacunosummoleculare in three interneurons (Fig. 1).

RADIATUM

50 p.m \,

Membrane properties

Fig. I. Interneurons recorded in O-A zone of CAI region. (A) Schematic representation of a transverse section of hippocampus showing the locations of the interneuron recordings. Filled circles represent individual interneurons, the filled square shows the location of the interneuron depicted in B, and open triangles indicate interneurons with spontaneous synaptic bursting activity (cf. Results). (B) Drawing of a physiologically-identified interneuron in O-A filled with LY showing horizontal dendrites in O-A and vertical (ascending) dendrites in stratum radiatum.

The intrinsic properties of interneurons in O - A are listed in Table 1. Average R M P ( - 6 3 mV), Rin ( 5 8 M ~ ) and AP amplitude (61 mV) suggest "'healthy" interneuron impalements. Representative examples of interneuron responses (with pyramidal cell responses displayed for comparison) are shown in Fig. 2. In this interneuron, the duration of the AP was 1.05 ms at the base and 0.49 ms at 50% amplitude. The pyramidal cell AP had a duration of 1.82 ms at the base and 0.86 ms at 50% amplitude. All interneurons examined displayed an afterhyperpolarization following individual APs (Fig. 2). In Fig. 2, the interneuron AP is followed by an A H P of - 14.0 mV lasting 39 ms. This interneuron A H P is referred to as a fAHP to distinguish it from the sAHP

evoked after bursts of APs (Fig. 2). In the interneuron shown in Fig, 2, the amplitude of the sAHP was - 4 . 9 mV, its peak latency 172 ms, and it decayed in 642 ms. In contrast, in pyramidal cells, a fAHP, a depolarizing afterpotential (DAP), and a medium duration A H P follow individual APs 39~48(Fig. 2). The amplitude and time course of interneuron fast and slow AHPs are given in Table 2 for all cells. In the majority of cells (21 of 27), APs were produced spontaneously (Fig. 3). A subgroup of interneurons (9 of 21), displayed spontaneous bursts of APs (Fig. 3). For the interneuron shown in Fig. 3B2, the frequency of firing during these bursts was 455 and 488 Hz. These were spontaneous

352

J.-C. LACAILLEand S. WILLIAMS Table 1. Intrinsic properties of interneurons in oriens-alveus Membrane potential (mV) Input resistance (Mr1) AP amplitude (mV) AP duration base (ms) AP duration 50% (ms) Threshold (mV) Membrane time constant (ms) Spontaneous firing rate (Hz)

Mean

S.D.

Range

n

- 62.7 58.0 61.3 1.15 0.55 -54.5 6.55 11.4

6.0 30.4 12.2 0.51 0.25 4.7 1.87 17.3

- 52, - 73 17,134 42, 87 0.40, 2.63 0.22, 1.28 -41, - 6 2 3.7, 10.2 0, 61.8

27 26 27 26 26 25 12 27

synaptic events since membrane hyperpolarization prevented the bursts responses and uncovered the underlying excitatory postsynaptic potential (EPSPs) (not shown). Response properties (listed in Table 1) were not statistically different between bursting and non-bursting interneurons (P > 0.05). Also, the distributions, within the CA1 region, of interneurons with single vs bursting type of spontaneous synaptic activity appeared similar (Fig. 1). In two interneurons, spontaneous, fast-rising depolarizations were observed (Fig. 3). These appeared similar to the dendritic spikes ( " d " spikes) reported in pyramidal cells. 39'4°'46The origin of these events is unknown, but tonic hyperpolarization of the membrane increased their amplitude and eventually prevented their occurrence (Fig. 3). Thus, they may be generated endogenously. In all interneurons tested (19 cells), the mode of firing during depolarizing current pulses was not altered by membrane hyperpolarization (i.e. cells did not change from sustained to burst type of firing25).

A

INTERNEURON O/A

B

Voltage-current relationship The V-I relationship was examined in 23 cells during 200 ms long hyperpolarizing current pulses (0 to - 1 nA). In the low range of current intensities (0 to - 0 . 3 nA), the membrane charged smoothly and exponentially to its maximum. However, at higher current intensities ( - 0 . 3 to - 1.0 nA), time-dependent inward membrane rectification occurred (Fig. 4). All but one cell exhibited a depolarizing "sag" in the response in this range of current intensity. Membrane rectification for maximum (Vmax) and steady-state (Vss) voltage measures was examined in voltage vs current graphs (Fig. 4). Vm,x measurements were made early ( 1 0 - 4 0 m s ) and Vss late (175-185ms) during the hyperpolarizing response, at the different current intensities. The difference in mean Rin ( + S . E . ) obtained from the slope of the linear regression for Vmax (57.1 ___5.9Mf]) and for V~ (46.7 + 4.8 Mf~) were statistically significant (P < 0.001). In most cells (15 of 19), V-I relationship

PYRAMIDAL CELL

C

INTERNEURON O/A

-58~ -58 : r _ ~ L _

.......

i _ A ~ I

30 mV 1.5nA 75 ms

75O~

.58 30ms

~ - - -.-ff¢~_...... -( 30rm

............

I .J---I

7.5 mV 1.5 nA

300rns

Fig. 2. AP and AHPs of interneurons in O-A and pyramidal cells. (A) Average of four APs of an interneuron at fast (A1) and slower (A2) time calibrations. AP duration was taken at the base of the AP (AI; broken line) and 50% amplitude. A large AHP (fAHP) followed the AP (A2, arrow; broken line = RMP). (B) Average of four APs of a CAI pyramidal cell also at fast (B1) and slow (B2) time calibrations. Pyramidal cell AP duration was longer (B1; broken line) and a fAHP (single arrow B2), a DAP (double arrow) and a mAHP (triple arrow) followed individual APs. (C) In a different interneuron, a depolarizing current pulse evoked a train of APs (C1), which is followed by a long duration AHP (sAHP; C2, arrow). In C2, APs are truncated.

O A interneuron membrane properties Table 2.

Hyperpolarizing afterpotentials of interneurons in oriens-alveus (mean + S.D.) Spike AHP (fAHP)

Amplitude from RMP (mV) Amplitude from AP base (mV) Peak latency (ms) Time to decay (ms) Membrane potential (mV) n

-6.9 ± 3.3 -13.2 4-4.6 4.9 _+2.9 36.3 _+25.6 - 62.9 _+6.9 26

was linear for Vm,x (Fig. 4). In the other four cells, inward rectification was present. In the majority of cells (15 of 19), V - I relationship for V~ displayed inward rectification (Fig. 4). In the remaining cells (different from the four cells with rectification in Vm,x), the relationship was linear. As shown for the interneuron in Fig. 4, most cells displayed anodal break excitation when hyperpolarizing current pulses were turned off. Passive cable properties

Cable properties were evaluated from the analysis of the slope of the charging function of hyperpolarizing responses evoked by small current injections (n = 8 cells). These small current injections ( - 0 . 1 5 to - 0 . 2 5 nA) fell within the linear portion of the V - I relationship. In some cells, this analysis was prevented by the presence of spontaneous EPSPs contaminating the passive responses. A representative A

353

SINGLE

B

Burst AHP (sAHP) -4.6 + 1.5 138 4- 92 838 4- 319 -62.2 4- 6.2 12

example of the procedure for obtaining the electrotonic parameters is shown in Fig. 5. In this interneuron, the exponential functions of the charging function (Vf - V ) and of the "peeled" faster component equation (1) yielded a membrane time constant (t0) of 9.7 ms, a first order equalizing time constant (Zl) of 1.22 ms, and respective coefficients (C Oand C~ ) of 5.61 and 5.15. From equation (2), a dendrite to soma conductance ratio (p) of 4.13 was calculated. An electrotonic length constant (L) of 1.07 was then estimated with equation 3. The mean values obtained for t0, ~], p and L, are given in Table 3 for all cells analysed. Repetitive firing

The repetitive firing of interneurons in response to long (700 ms) depolarizing current pulses of increasing intensity was examined in nine cells. In all interneurons, the frequency of APs diminished

BURST

C

"d" SPIKES

i

-74

-78

-,4 150 rns

>_ 30 ms

-8,

_1!

150ms

-7,

30ms

Fig. 3. Spontaneous activity of interneurons. (A) Example of a cell, spontaneously firing single APs at RMP (A1; firing frequency 9.4 Hz). Membrane hyperpolarization uncovered spontaneous, small amplitude, EPSPs (A2). (B) In a different interneuron, bursts of APs (or single APs) occurred spontaneously at RMP (BI). Burst morphology is shown at a faster time base in B2. Large amplitude (>10mV) spontaneous EPSPs can be seen in the background. (C) In a different cell, small amplitude spikes ("d" spikes) were spontaneously produced (RMP - 6 7 mV). These were fast rising and of short duration. Their amplitude increased with membrane hyperpolarization ( - 7 4 and -78mV), until they were prevented from occurring ( - 90 mV).

354

J.-C. LACAILLEand S. WILLIAMS A

A

10

AVERAGE (4)

> >A

-65

-65 ~ k ~ _ ~ . _ / ~ 8

E

z

_o

I

~

~ ........

[15mV ~ 1.5 nA 30 ms

~ = ~

"--

6

z

= ¢5 z

4

< 1-

2¸ . . . . .

DI+I'~mDW$I+

mm.

. . . .

.... . - . " I 5

0

15 mV

I

I

I

1 5 nA

10

>

CURRENT (nA) -0.8

I

-06

-04 ,

I

_o

.(

0

~°

Vma x

Vss

-/~'~"

~.J~

-

2

-10 -20

-30 -40

.

-50

E

v LU

< "fO

(.9 < k..J o>

' " ' ~

' ..........

-

0.1-

0

~

i

i

1

5

10

15

210

2;

30

TIME (ms)

(adapted) during these long depolarizations (Fig. 6). As shown in Fig. 6, spike frequency adaptation was present at all current intensities. In eight of nine cells, however, the adaptation was never complete. The percentage change in spike frequency was measured as

(ffinal/finitial))< 100

•,, .._...~_.. . . . .

\o

0.03

Fig. 4. V-I relation in interneuron. (A) Superimposed responses to hyperpolarizing current pulses of different intensities. Time-dependent inward rectification (sag in the response) became more evident with increasing current intensities. Maximum (limax, open square) and steady-state (Vss, filled square) voltages were measured at times indicated by the symbols and plotted as a function of current in B. Note anodal break excitation upon removal of hyperpolarizing current (spikes are truncated). (B) V-I graph for Vm,~ (open squares) and Vs~ (filled square) from responses of interneuron in A. Line represents the linear equation for Vm,xvalues. Throughout the voltage range, V-1 relation is linear for Vr~x. For Vss, in the range 0 to -0.25 nA, the V-I relation appears linear, but in the range -0.25 to - 1 . 0 n A , deviation from the linear equation suggests inward rectification.

=

'

?t°o

(5

-60

% of control

~% \ %

1

z

~'~

•

~":CZ~=<.~

z

0

o PEELED "E1

3

E -02 ,

1 3o

%

>

-1 0

_" ".._.". "_..'_-.'Z_ I .... I 20 25

I 15

B

75 ms

B

I 10

Fig. 5. Cable properties of interneuron. (A) Averaged response (n = 4) to 100 ms, -0.25 nA current pulse, shown in the inset. Charging function (Vr - V) of the averaged hyperpolarizing transient graphed against time. (B) Semi log plot of the charging function of A (filled squares) with the fitted exponential function (line) to obtain z0 (9.7ms). Deviation from the exponential function in the early part of the charging function indicated the presence of another faster component. Subtracting the exponential function from the charging function, "peeled" this component (open square). From the exponential function fitted to the peeled component (line), z~ (I .22 ms) was obtained.

The frequency-current ( f - l ) relationship was examined for different ISis (Fig. 6). In six of nine cells, the slope of f - I curve for the first ISI gave a primary firing range of 361 + 143 Hz/nA. In three cells, primary (606+__329Hz/nA) and secondary (165 ___12 Hz/nA) firing ranges were observed. For the pooled data (nine cells), a mean primary firing range of 442 + 2 3 4 H z / n A was obtained for ISI no. 1. In the adapted state (n = 9 cells, median ISI = ISI no. 60), the slope of the f - I curve was 117 ___68 Hz/nA.

(4)

where f~,,l and finitial are, respectively, the frequency during the last and first interspike intervals (ISis). The change in spike frequency was to 34.3 + 7.7% of control (n = 8 cells). For a given cell, there was relatively little variability in spike frequency adaptation at the different current intensities (mean S.D. = 15.0%).

DISCUSSION

Using intracellular recording and staining, the properties of interneurons in O - A were examined to assess their (dis)similarity with pyramidal cells and other interneuron types (B-cells and S-cells) in the rat CAI region.

O-A interneuron membrane properties

355

Table 3. Cable properties of interneurons in oriens alveus (n = 8) Mean S.D. Range

Zo

~,

to/r,

p

L

7.79 2.27 5.2-12.0

1.33 0.57 0.59 2.23

6.32 1.66 3.8-8.9

2.70 1.38 0.44 4.40

1.13 0.12 0.96 1.31

Morphology

of cellular morphology, V-cells are significantly different from other types of physiologically-identified interneurons, B-cells ~v'~8'4~'4`"and S-cells, m8"2225 particularly in terms of pattern of dendritic arborizations.

LY-filled interneuron somata in O - A were multipolar and dendritic processes projected horizontally in O - A along the alveus and vertically toward stratum lacunosum-moleculare. A similar morphology of physiologically-characterized interneurons in O - A was observed previously in guinea-pig. 24 These cells correspond to the cells with horizontal axon described by R a m o n y Cajal 3s and to the polygonal cells with ascending axon of Lorente de No. 3° However, this nomenclature is confusing since another cell type with dendrites restricted to stratum oriens is referred to by CajaP 8 as cells with ascending axon and by Lorente de No 3° as cells with horizontal axon. To avoid this confusion, we will refer to the interneurons we have recorded from as vertical cells (V-cells) for their ascending (vertical) dendrites. Interneurons with similar morphology and distribution to V-cells, have been reported to be immunoreactive for G A B A and somatostatin. 23"45 Thus, the V-cells we have recorded from may be GABAergic interneurons which co-localize the peptide somatostatin. In terms

Membrane properties Intrinsic properties of V-cells were generally similar to those previously reported in guinea-pig. 24 Values obtained for R M P ( - 6 2 mV), R~, (58 MI1) and AP amplitude (61 mV) were slightly larger perhaps reflecting lesser injury. All cells displayed a fAHP and 14 of 15 cells exhibited a sAHP following trains of APs. These properties are similar to those previously reported for hippocampal B -7'17'2a'42 and S-cells. 2s Except for the presence of a sAHP, these properties are also comparable to those observed in neocortical fast-spiking non-pyramidal cells. 3-' In contrast, in pyramidal cells, afterpotentials consist of a fAHP, a depolarizing afterpotential, a medium duration A H P and a sAHP. 39"~

JlJJJJJ:!JJ,i_ 15•mY ~ 1.4 rtA 150m$

j=

350~

• 0 125nA

• 0.00 nA

• t0nA

o 0 2 7 5 nA

o 0.75 nA

= 1.8 nA

300

25O

35O

• kSl #~

•

•

•

• ISl #1

. ooo

.ore

Lu

Oo 5O"

•

lee~e

.

.......

•co••oct.•a• o

.... ...°,..

aaaDD°OaooDaa = • • • • • I 100

I 200

, 0 o ~ . o• oo=~

.:.

• • DaO~DO°D°. • •

3

;o

,;0

TtME (ms)

.....

• •

~ * * •

oOoo~ooO

eeeeeeeeeeee• o . o ~ = •

-"

ooa0~

" ~

o ,00

.

.

,00 , ,

o

. *

1 o

5O

g

~

, w

o 0.

.

.

.

.

.

.

.

0.2 0.4 0.6 0.8 1 0 ~ 2 1.4 1 5 1,8 CURRENT (hA)

. O0 .

0.1

. 0.2 . 0.3 . 0,4

0.5 0.6 0.7 0.8

C U R R E N T (nA)

Fig. 6. Repetitive firing of interneuron in O-A. (A) Interneuron responses to long depolarizing current pulses of increasing intensity (0.125-1.0nA; membrane potential -64mV). Note the progressive lengthening of the ISI during these responses. (B) Graph of frequency of firing vs time during depolarizing current injections, at intensities indicated, for responses of interneuron of A. At all current intensities, the frequency of firing decreases during the pulses (spike frequency adaptation). (C) Graph of frequency of firing as a function of injected current ( f 1) for the first ISI (ISI no. 1) and for an ISI later during the pulse (ISI no. 60) with their respective linear equations (lines). For this cell, the f - I relation consisted of a single firing range both early and late in the pulses. ( D ) f l Graph from another interneuron, for which the/ I relation for ISI no. I was best described by two linear equations, suggesting the presence of primary and secondary firing ranges early in the pulses.

356

J.-C. LACAILLEand S. WILLIAMS

We have observed the presence of time dependent inward rectification during hyperpolarizing responses in the majority of V-cells. This type of rectification is prominent in S-cells18'25 and is also present, but to a lesser extent, in pyramidal cells.4° Such rectification has not been reported in hippocampal B-cells42 or in fast-spiking non-pyramidal cells of neocortex. 32 Anodal break excitation was also present either in the form of APs or of a sub-threshold depolarization. The cellular events underlying this anodal break excitation are probably different from the low threshold calcium spike described in thalamic neurons ~4 since in V-cells the mode of firing did not change from sustained to bursting with membrane hyperpolarization.

Spontaneous activity AP threshold for these cells was - 5 4 mV and, in combination with the presence of spontaneous EPSPs, usually resulted in spontaneous activity. This type of spontaneous activity is similar to the spontaneous activity of B-cells42 but unlike S-cells, in which spontaneous activity is almost absent. 25 Two classes of V-cells were distinguished based on their type of spontaneous activity: cells firing single APs and cells firing high frequency bursts of APs. Since bursting responses were associated with more marked spontaneous EPSPs and since intrinsic properties of the two subtypes of interneurons did not differ significantly, network properties are probably responsible for these two subtypes of firing behaviour of V-cells. This distinction between types of V-cells is potentially important since their respective postsynaptic action must reflect their frequency of firing and, therefore, may be different.

Action potential duration The mean AP duration (at base) of V-cells was 1.15 ms, similar to the duration previously reported for interneurons in O A of guinea pigfl4 This is greater than for B-cell APs (duration approximately 0.8 ms) 7'24'42 and less than for S-cell APs (duration 2.0 ms). 25 It is thus possible that two types of interneurons have been sampled in O - A (B-cell type and S-cell type), and that the resulting mean AP duration falls between the duration of each type. However, this seems unlikely since intrinsic properties characteristic of S-cells (absence of spontaneous activity and presence of voltage-dependent mode of firing25) were not seen in O - A interneurons. Since the range of AP duration included small values typical of B-cells (0.5-0.8 ms), it is possible that the recordings with longer duration of APs reflected dendritic penetrations of interneurons in O~A. Dendritic penetration of B-cells would, however, be much less frequent. B-cell recordings are obtained from an area of the slice (strata pyramidale and oriens border) where somata are found but where dendritic arborizations are not extensive. ~8'4~ The presence of significant spontaneous synaptic activity and the

elevated spontaneous firing rate, also suggest that V-cells belong in the same physiological class with B-cells but not with S-cells.

Cable properties Exponential analysis of hyperpolarizing transient responses of V-cells gave an average measure of membrane time constant of 6.55ms. Similar measures of membrane time constant were obtained by other means for interneurons in O - A (5.6 ms) and for B-cells (5.8 ms). 24 Slower membrane time constants have been reported for S-cells (8.6 ms). 25 CAI pyramidal cell membrane time constant measures, obtained from exponential analysis, were also slower, in the order of 15-18 ms. 8'53 Based on the equivalent cylinder model, 8,36'37,53two cable parameters were estimated from the analysis of the slope of the charging function. An average dendrite to soma conductance ratio (p) of 2.70 and an average electrotonic length (L) of 1.1 were found. These measures suggest that V-cells are electrically compact, with their dendrites terminating approximately at an electrotonic distance of one length constant from the soma. The dendrite to soma conductance ratio suggests significant conductances in dendritic regions. The reported electrotonic parameters of CA1 pyramidal cells do not differ greatly from those of V-cells. They are also electrically compact (L = 0.9) but their dendrite to soma conductance ratio is lower (1.2). 8'53 In view of the electrical compactness of V-cells, single electrode voltageclamp techniques may be applicable, within their constraints, ~5 to these interneurons.

Repetitive firing behaviour Analysis of firing behaviour during long depolarizing current pulses, indicated that V-cells can sustain high frequencies of firing. From the slope of the f - I relation for their first ISI, we observed a mean primary firing range of 442 Hz/nA for these interneutons. However, for ISis later during the pulses, the slope was decreased to an average of 117 Hz/nA (ISI no. 60). For B-cells, a high-gain f - I slope with a single component for ISI no. 1 has also been reported. 42 Similar elevated primary f - I slopes (549 Hz/nA for ISI no. 1) have been reported for neocortical fast-spiking non-pyramidal cells.32 In contrast, t h e f - I slope of CA I pyramidal cells (ISI no. 1) showed a slow primary (29 Hz/nA), a rapid secondary (286 Hz/nA) and a slow tertiary (90 Hz/nA) range. 29 Thus, even the steepest firing range of CA1 pyramidal cells is lower than the primary range of V-cells. In the "adapted state", the firing frequency of V-cells (117 Hz/nA) is also higher than for pyramidal cells (20-34 Hz/nA). 29 As evidenced by the reduced.fi-I slope at late ISis, V-cells displayed spike frequency adaptation during long depolarizing pulses. The adaptation was quite marked, On average, the frequency of the last ISI was 34% of the frequency of the first ISI. Frequency of

O-A interneuron membrane properties firing diminished rapidly during the initial moments and then continuously during the remainder of the depolarizing pulse. Given that a sAHP is present in V-cells and that a sAHP underlies adaptation in other cells, 3j it is not surprising that V-cells also undergo spike frequency adaptation. In CAI pyramidal cells, spike frequency adaptation is present but much more prominently: in most pyramidal cells, following the initial period of spiking, there is a complete pause in the firing of APs and a subsequent return of regular firing, z9'31 Previous reports have emphasized the relative lack of spike frequency adaptation in hippocampal interneurons as a characteristic non-pyramidal cell feature. 7"24"25"42However, careful analysis indicates that significant adaptation, although not complete, is present in interneurons. Therefore, it is rather the sustained higher frequency of firing which may physiologically differentiate non-pyramidal from pyramidal cells. Functional implications

Analyses of intrinsic membrane properties so far suggest that V-cells belong to the same physiological class of interneurons as B-cells. 7'24'42 However it is possible that some of the newly described physiological properties of V-cells (spontaneous bursting, cable properties, repetitive firing behaviour), when assessed in B-cells, would reveal physiological differences between these interneuron types. Their reported local circuit interactions with pyramidal cells also appear similar: excitation of interneurons by pyramidal cells and inhibition of pyramidal cells by V-cells and B-cells. 2°'24 However, differences in synaptic properties have been described between V-cells and B-cells: stimulation of fibres in the alveus evoked powerful

357

bursting responses with transient spike inactivation in V-cells but not in B-cellsfl Such differences in responsivity to afferents may be related to their respective cellular architecture. V-cell dendrites project along and into the alveus, 23"24whereas few B-cell dendrites are found in this area. :'18'4~ Thus, their spatial arrangement and their synaptic connectivity, rather than their membrane properties, may shape the respective influence of V-cells and B-cells in the hippocampus. In contrast, distinct membrane properties may additionally alter the influence of S-cells.

CONCLUSION Vertical cells in stratum oriens-alveus region displayed electrophysiological properties relatively similar to those of B-cells and in some respect unlike those of S-cells. The AP duration of V-cells was longer, however, than the duration of B-cell APs. This difference may reflect occasional dendritic penetrations of V-cells. Other hitherto undescribed properties of V-cells are their two types of spontaneous synaptic activity (single or high frequency bursts of APs), their electrical cable properties, and their typical repetitive firing behaviour (adapting, high frequency responses). Acknowledgements--This study was supported by grants

from Fonds de la Recherche en Sant6 du Quebec, the Medical Research Council of Canada, the Savoy Foundation and the Banting Foundation. J.-C.L. was an FRSQ Scholar and a Sloan Research Fellow. S.W. was supported by a postgraduate fellowship from the Savoy Foundation. The authors wish to thank C. Gauthier and D. Cyr for graphic and photographic assistance, L. Imbeault for secretarial assistance, Dr J.-P. Raynauld for loan of equipment and Dr J.-M. Peyronnard for use of microscope facilities.

REFERENCES

1. Alger B. E. (1984) Characteristics of a slow hyperpolarizing synaptic potential in rat hippocampal pyramidal cells in vitro. J. Neurophysiol. 52, 892-910. 2. Alger B. E. (1985) GABA and glycine: postsynaptic actions. In Neurotransmitter Actions in the Vertebrate Nervous System (eds Rogawski M. A. and Barker J. L.), pp. 33 69. Plenum Press, New York. 3. Alger B. E. and Nicoll R. A. (1982) Feedforward dendritic inhibition in rat hippocampal pyramidal cells studied in vitro. J. Physiol., Lond. 328, 105 123. 4. Alger B. E. and Nicoll R. A. (1982) Pharmacological evidence for two kinds of GABA receptors on rat hippocampal pyramidal cells studied in vitro. J. Physiol., Lond. 328, 125 141. 5. Andersen P., Eccles J. C. and L~yning Y. (1964) Location of postsynaptic inhibitory synapses on hippocampal pyramids. J. Neurophysiol. 27, 592 607. 6. Andersen P., Eccles J. C. and L~yning Y. (1964) Pathway of postsynaptic inhibition in the hippocampus. J. Neurophysiol. 27, 608-619. 7. Ashwood T. J., Lancaster B. and Wheal H. V. (1984) In vivo and in vitro studies on putative interneurones in the rat hippocampus: possible mediators of feed-forward inhibition. Brain Res. 293, 279 291. 8. Brown T. H., Fricke R. A. and Perkel D. H. (1981) Passive electrical constants in three classes of hippocampal neurons. J. Neurophysiol. 46, 812-827. 9. Collingridge G. L., Gage P. W. and Robertson B. (1984) Inhibitory post-synaptic currents in rat hippocampal CAI neurones. J. Physiol., Lond. 356, 551 564. 10. Dutar P. and Nicoll R. A. (1988) A physiological role for GABA B receptors in the central nervous system. Nature 332, 156- 158. 11. Dutar P. and Nicoll R. A. (1988) Pre- and postsynaptic GABA s receptors in the hippocampus have different pharmacological properties. Neuron 1, 585-591. 12. Gahwiler B. H. and Brown D. A. (1985) GABAa-receptor activated K* current in voltage-clamped CA3 pyramidal cells in hippocampal cultures. Proc. natn. Acad. Sci. U.S.A. 82, 1558 1562. 13. Gray R. and Johnston D. (1985) Rectification of single GABA-gated chloride channels in adult hippocampal neurons. J. Neurophysiol. 54, 134 142.

358

J.-C. LACAILLEand S. WILLIAMS

14. Jahnsen H. and Llinas R. (1984) Electrophysiological properties of guinea pig thalamic neurones: an in vitro study. J. Physiol., Lond. 349, 205-226. 15. Johnston D. and Brown T. H. (1983) Interpretation of voltage-clamp measurements in hippocampal neurons. J. Neurophysiol. 50, 464 486. 16. Kandel E. R., Spencer W. A. and Brinley F. J. (1961) Electrophysiology of hippocampal neurons. I. Sequential invasion and synaptic organization. J. Neurophysiol. 24, 225-242. 17. Kawaguchi Y. and Hama K. (1987) Two subtypes of non-pyramidal cells in rat hippocampal formation identified by intracellular recording and HRP injection. Brain Res. 411, 190-195. 18. Kawaguchi Y. and Hama K. (1988) Physiological heterogeneity of nonpyramidal cells in rat hippocampal CA1 region. Expl Brain Res. 72, 494 502. 19. Knowles W. D., Schneiderman J. H., Wheal H. V., Stafstrom C. E. and Schwartzkroin P. A. (1984) Hyperpolarizing potentials in guinea pig hippocampal CA3 neurons. Cell. molec. Neurobiol. 4, 207-230. 20. Knowles W. D. and Schwartzkroin P. A. (1981) Local circuit synaptic interactions in hippocampal brain slice. J. Neurosci. 1, 318 322. 21. Krnjevic K. (1974) Chemical nature of synaptic transmission in vertebrates. Physiol. Rev. 54, 418-540. 22. Kunkel D. D., Lacaille J.-C. and Schwartzkroin P. A. (1988) Ultrastructure of stratum lacunosum-moleculare interneurons of hippocampal CAI region. Synapse 2, 382-394. 23. Kunkel D. D. and Schwartzkroin P. A. (1988) Ultrastructural characterization and GAD colocalization of somatostatin-like immunoreactive neurons in CAI of rabbit hippocampus. Synapse 2, 371-381. 24. Lacaille J.-C., Mueller A. L., Kunkel D. D. and Schwartzkroin P. A. (1987) Local circuit interactions between oriens alveus interneurons and CAI pyramidal cells in hippocampal slices: electrophysiology and morphology. J. Neurosci. 7, 1979-1993. 25. Lacaille J.-C. and Schwartzkroin P. A. (1988) Stratum lacunosum-moleculare interneurons of hippocampal CA1 region. I. Intracellular response characteristics, synaptic responses, and morphology. J. Neurosci. 8, 1400 1410. 26. Lacaille J.-C. and Schwartzkroin P. A. (1988) Stratum lacunosum-moleculare interneurons of hippocampal CA1 region. !I. Intrasomatic and intradendritic recordings of local circuit synaptic interactions. J. Neurosci. 8, 141l 1424. 27. Lacaille J.-C., Kunkel D. D. and Schwartzkroin P. A. (1989) Electrophysiological and morphological characterization of hippocampal interneurons. In The Hippocampus--New Vistas (eds Chan-Palay V. and Kohler C.), pp. 285 303. Alan Liss, New York. 28. Lacaille J.-C. and Williams S. (1989) Membrane properties of interneurons in stratum oriens/alveus of the CAI region of rat hippocampus. Neurosci. Abstr. 15, 401. 29. Lanthorn T., Storm J. and Andersen P. (1984) Current-to-frequency transduction in CA I hippocampal pyramidal cells: slow prepotentials dominate the primary range firing. Expl Brain Res. 53, 431 443. 30. Lorente de No R. (1934) Studies on the structure of the cerebral cortex. II. Continuation of the study of the ammonic system. J. Psychol. Neurol., Lpz. 46, 113- 177. 31. Madison D. V. and Nicoll R. A. (1984) Control of repetitive discharge of rat CAI pyramidal neurones in vitro. J. Physiol., Lond. 354, 319 331. 32. McCormick D. A., Connors B. W., Lighthall J. W. and Prince D. A. (1985) Comparative electrophysiology of pyramidal and sparsely spiny stellate neurons of the neocortex. J. Neurophysiol. 54, 782-806. 33. Newberry N. R. and Nicoll R. A. (1984) Direct hyperpolarizing action of baclofen on hippocampal pyramidal cells. Nature 308, 450-452. 34. Newberry N. R. and Nicoll R. A. (1985) Comparison of the action of baclofen with 7-aminobutyric acid on rat hippocampal pyramidal cells in vitro. J. Physiol., Lond. 360, 161-185. 35. Prince D. A. and Connors B. W. (1986) Mechanisms of interictal epileptogenesis. Adv. Neurol. 44, 275599. 36. Rall W. (1969) Time constants and electrotonic length of membrane cylinders and neurons. Biophys. J. 9, 1483 1508. 37. Rall W. (1977) Core conductor theory and cable properties of neurons. In Handbook of Physiology, Section I: The Nervous System (ed. Kandel E. R.), Vol. l, pp. 39 97. Williams & Wilkins, Baltimore. 38. Ramon y Cajal S. (1911) Histologie du Systkme Nerveux de l'Homme et des Vertkbr~s. Maloine, Paris. 39. Schwartzkroin P. A. (1975) Characteristics of CAt neurons recorded intracellularly in the hippocampal slice preparation. Brain Res. 85, 423 436. 40. Schwartzkroin P. A. (1977) Further characteristics of hippocampal CAI cells in vitro. Brain Res. 128, 53 68. 41. Schwartzkroin P. A. and Kunkel D. D. (1985) Morphology of identified interneurons in the CAI regions of guinea pig hippocampus. J. comp. Neurol. 232, 205-218. 42. Schwartzkroin P. A. and Mathers L. H. (1978) Physiological and morphological identification of a nonpyramidal hippocampal cell type. Brain Res. 157, 1 10. 43. Schwartzkroin P. A. and Prince D. A. (1980) Changes in excitatory and inhibitory synaptic potentials leading to epileptogenic activity. Brain Res. 183, 169 181. 44. Segal M. and Barker J. L. (1984) Rat hippocampal neurons in culture: voltage-clamp analysis of inhibitory synaptic connections. J. Neurophysiol. 52, 469-487. 45. Somogyi P., Hodgson A. J., Smith A. D., Nunzi M. G., Gorio A. and Wu J. Y. (1984) Different populations of GABAergic neurons in the visual cortex and hippocampus of cat contain somatostatin- or cholecystokinin-immunoreactive material. J. Neurosci. 4, 2590-2603. 46. Spencer W. A. and Kandel E. R. (1961) Electrophysiology of hippocampal neurons. IV. Fast prepotentials. J. Neurophysiol. 24, 272 285. 47. Stewart W. W. (1978) Functional connections between cells as revealed by dye-coupling with a highly fluorescent naphthalimide tracer. Cell 14, 741 759. 48. Storm J. F. (1987) Action potential repolarization and a fast after-hyperpolarization in rat hippocampal pyramidal cells. J. Physiol., Lond. 385, 733 759. 49. Thalmann R. H. and Ayala G. F. (1982) A late increase in potassium conductance follows synaptic stimulation of granule neurons of the dentate gyrus. Neurosci. Lett. 29, 243-248. 50. Traub R. D., Miles R. and Wong R. K. S. (1987) Models of synchronized hippocampal bursts in the presence of inhibition. 1. Single population events. J. Neurophysiol. 58, 739 751.

O A interneuron membrane properties

359

51. Traub R. D., Miles R. and Wong R. K. S. (1989) Model of the origin of rhythmic population oscillations in the hippocampal slice. Science 243, 1319 1325. 52. Traub R. D., Miles R., Wong R. K. S., Schulman L. S. and Schneiderman J. H. (1987) Models of synchronized hippocampal bursts in the presence of inhibition. II. Ongoing spontaneous population events. J. Neurophysiol. 58, 752-764. 53. Turner D. A. (1984) Segmental cable evaluation of somatic transients in hippocampal neurons (CAI, CA3, and dentate). Biophys. J. 46, 73 84. 54. Wong R. K. S. and Prince D. A. (1979) Dendritic mechanisms underlying penicillin-induced epileptiform activity. Science 204, 1228-1231. 55. Wong R. K. S., Prince D. A. and Basbaum A. I. (1979) Intradendritic recordings from hippocampal neurons. Proc. Nam. Acad. Sci. U.S.A. 76, 986 990. (Accepted 15 Februa O, 1990)