Involvement of ON and OFF retinal channels in the ...

3 downloads 0 Views 846KB Size Report
stimulus velocity and high spatial frequency (DeMarco et al.,. 1987). Nevertheless, the role of the retinal message in the frog. OKN generation is further ...
Visual Neuroscience (1989), 2, 357-365. Printed in the USA. Copyright © 1989 Cambridge University Press 0952-5238/89 $5.00 + .00

Involvement of ON and OFF retinal channels in the eye and head horizontal optokinetic nystagmus of the frog

Y.H. YUCEL, B. JARDON, AND N. BONAVENTURE Departement de Neurophysiologie et de Biologie des Comportements, Centre de Neurochimie du CNRS, Strasbourg, France (RECEIVED February 23, 1988; ACCEPTED September 28, 1988)

Abstract

The specific role of ON and OFF retinal information channels in the generation of the horizontal optokinetic nystagmus (OKN) of the frog was studied. Coil recordings of monocular eye and head OKN were obtained before and after intravitreal injection of two drugs that block either ON or OFF channels. The intravitreal injection of 2-amino-4-phosphonobutyrate (APB), a glutamate analog that selectively blocks the ON retinal channel, strongly reduced or even cancelled the monocular OKN of the head and of the eye. The intravitreal injection of another glutamate analog, the cis-2,3-piperidine dicarboxylic acid (PDA) that especially blocks the OFF retinal channel, did not affect the gain velocity of the slow phase of both the horizontal monocular head and eye OKN, for low stimulus velocities. Our results suggest that the retinal ON information channel, but not the OFF channel, is involved in the generation of the slow phase of the OKN of the frog, at least at low drum velocities. Keywords: Optokinetic nystagmus, Slow phase, Retinal input, ON channel, OFF channel, Receptive field, 2-amino-4-phosphonobutyrate, cis-2,3-piperidine dicarboxylic acid, Frog Introduction

The optokinetic nystagmus (OKN) is a visuomotor reflex permitting, with the vestibuloocular reflex, the stability of the image on the retina. OKN is evoked in experimental conditions by the movement of the visual environment. It is composed of slow phases interrupted by resetting fast phases. The retinal input is the first step in the basic OKN network. The ganglion cells transmit visual information from the retina to the mesencephalic structures responsible for OKN. Direction-selective cells were found in these structures in all of the species studied: frog (Kondrashev & Orlov, 1976; Katte & Hoffmann, 1980; Cochran et al., 1984; Gruberg & Grasse, 1984), salamander (Manteuffel, 1984), pigeon (Gioanni et al., 1984), rabbit (Collewijn, 1975; Simpson et al., 1979), and cat (Hoffmann & Schoppmann, 1981). For many years, it has been argued that ON center directionselective ganglion cells provide the major, if not the sole, retinal input to the accessory optic system (AOS) in the rabbit. This hypothesis is based on the similarities between the response properties of the direction-selective neurons in both the retina and the medial terminal nucleus of the AOS in this speReprint requests to: Y.H. Yucel, Departement de Neurophysiologie et de Biologie des Comportements, Centre de Neurochimie du CNRS, 5 rue Blaise Pascal-67084, Strasbourg, France.

cies (Oyster & Barlow, 1967; Simpson et al., 1979; Buhl & Peichl, 1986). In the cat, electrophysiological recordings have identified the specific retinal ganglion cells projecting to the nucleus of the optic tract (NOT) as W cells; among them almost half were ON-center direction-selective cells (Hoffmann & Stone, 1985). However, according to these authors, this retinal input is not sufficient to account for the specific properties of the NOT neurons (concerning their temporo-nasal direction selectivity, their large receptor field, and their ability to respond to very high velocities of the stimulus). Therefore, Hoffmann and his collaborators demonstrated the major role played by a cortical input to the NOT in the generation of OKN in the cat (Hoffmann & Stone, 1985; Hoffmann, 1986). In the frog, the mesencephalic structures (pretectum and nucleus of the basal optic root (nBOR)) responsible for OKN receive visual information directly from the retina and mainly from the central retina for the nucleus mesencephali lentiformis (Montgomery et al., 1981, 1985). ON, OFF, and ON-OFF ganglion cells project to the direction-selective cells in the nBOR, which mediates vertical OKN (Lazar et al., 1983; Gruberg & Grasse, 1984). Although direction-selective retinal ganglion cells were described (Backstrom et al., 1978; Watanabe & Murakami, 1984), little is known about the types of ganglion cells projecting to the pretectum mediating horizontal OKN (Montgomery et al., 1982). Taking into account the data obtained in higher

357 Downloaded from http:/www.cambridge.org/core. University of Toronto, on 11 Nov 2016 at 16:31:25, subject to the Cambridge Core terms of use, available at http:/www.cambridge.org/core/terms. http://dx.doi.org/10.1017/S0952523800002169

358

vertebrates, we supposed that the retinal input to the frog's pretectum is probably made to a large extent by the ON center ganglion cells. In order to study the contribution of this retinal ON input in the generation of OKN, we used the possibility of selectively blocking the ON responses in the retina by injecting APB into the vitreous body of the eye. Recent studies have indeed shown that two structural glutamate analogs selectively block ON or OFF retinal channels. The first, 2-amino-4-phosphonobutyric acid (APB) could act as a glutamate agonist blocking the ON response at the kainate receptor located on ON bipolar cells, while having minor effects on horizontal cells, OFF bipolars, and photoreceptors. Consequently, ON discharges of third-order neurons were also blocked as well as ON component of ON-OFF amacrine and ON-OFF ganglion cells (Slaughter & Miller, 1981; Shiells et al., 1981; Neal et al., 1981; Schiller, 1982; Bolz et al., 1984; Miller & Slaughter, 1986; Sucher et al., 1986). The second, cis-2,3-piperidine dicarboxylic acid (PDA), considered as a nonselective antagonist of the excitatory amino-acid receptors in the spinal cord (Davies et al., 1982) could act upon the kainate receptors (APB insensitive), located on the OFF bipolar and horizontal cells (Cunningham & Neal, 1985; Miller & Slaughter, 1986). PDA also depresses the activation of the ON and OFF ganglion cells, whose membrane displays the three types of glutamate receptors, without affecting the activity of ON bipolar and of photoreceptor cells (Miller & Slaughter, 1986). It was already shown that the suppression of the ON retinal channel by APB entails the suppression of the OKN of the injected eye in the rabbit (Knapp & Ariel, 1984), in the turtle (Ariel, 1987), and in the frog (Bonaventure et al., 1987). In a recent electrophysiological study on the effects of APB and PDA on transient ganglion cell responses, we confirmed the ON or OFF blocking properties of these drugs in the frog in vivo (Jardon et al., 1989); from this study we were able to select the efficient drug concentrations blocking relatively selectively ON or OFF retinal channels. Thus, the aim of the present paper was to analyze the respective participation of ON and OFF retinal channels in the generation of the slow phase of horizontal OKN. Materials and Methods In adult frogs (Rana esculenta), two methods were used: (1) monocular horizontal head OKN observation, and (2) recording of eye or of head OKN by coil technique, before and after the intravitreal drug injection. Stimulation To evoke horizontal head and eye OKN, the same stimulation conditions were used. The frogs were stimulated by an optokinetic drum rotating at various constant speeds. The animals were placed in the drum (300 mm in diameter and 450 mm in height) with alternative black and white vertical stripes distributed equally on its inner surface (10-mm wide). The drum could be rotated clockwise and counterclockwise at regular speeds varying between 0.4 deg/s and 50 deg/s by means of an electronic control system. Room illumination was kept constant at 80 lux, at the level of the frog's eye.

Y.H. Yticel, B. Jardon, and N. Bonaventure Head OKN observations One hour after suturing the lids of one eye, unrestrained animals were placed in the drum. The rotation speed was regularly increased and the drum alternatively rotated clockwise and counterclockwise. The presence of horizontal head OKN was assessed by visual inspection. For the drum rotation speeds of optimal efficiency, frogs showed classical OKN that disappeared when the drum speed was increased. OKN performances were measured by the extinction speed of OKN, defined as the maximal speed of the drum still provoking OKN, with both slow phases and resetting fast phases. Eye and head OKN recording by coil technique In order to record eye OKN, a magnetic coil system as described by Koch (1977) was used. One pair of coils (200 mm of diameter) carrying a current of 50-kHZ frequency generates a homogenous magnetic field. These coils were mounted on an immobile platform. The sensing coil, fixed on the eyeball and oriented perpendicularly to the interaural axis, was placed in the center of the magnetic field. The voltage in the sensing coil, induced by a horizontal angular displacement, was amplified, rectified, filtered, and recorded on a paper recorder (BBC). Frogs were prepared under MS-222 (Tricain, Sandoz) anesthesia: a nut wasfixedto the skull by means of dental cement, and three metal screws (1 mm in diameter) were implanted in the dorsal skull. In order to observe and record monocular OKN, the lids of one eye were sutured, while the sclera of the other was exposed by removing the superior eyelid. After overnight recovery from anesthesia, the frog's head was restrained by attaching the nut fixed on the skull to the bar of the apparatus. The fixed head position was 15 deg nose up (Dieringer & Precht, 1982). The small sensing coil (1 mg, 75 turns, Sokymat) was perpendicularly secured under local anesthesia on the sclera with a drop of glue just before the experiment. The electrical connection between this coil and the detection equipment was made by a pair of flexible isolated wires. The animal was placed in the optokinetic drum on an immobile platform. To record head OKN, the frog's hind legs were restrained by a plaster on the immobile platform, and the sensing coil fixed to the skull under anesthesia the day before, was placed at the center of the magnetic field. Before each recording, the system was calibrated: the linear relationship between the angular displacement of the sensing coil (in the center of the magnetic field) and the induced voltage was verified. Then the voltage induced in the sensing coil by an angular displacement of 1 deg was calculated. The speed of the slow phase was measured from eye-movement tracings, after the elimination of resetting fast phases. The velocity gain was defined as the ratio of the speed of the slow phase to the speed of the drum. The drugs (provided by Sigma) were diluted in phosphatebuffered saline (PBS) (10 mM) and prepared daily (pH = 7.3). The concentrations used were 1 mM, 2 mM, 5 mM, 10 mM, and 18 mM for APB and 2 mM, 5 mM, 10 mM, and 15 mM for PDA. Thirty microliters of each drug were intravitreally injected by means of a microsyringe in the open eye under local anesthesia (Cebesine, Chauvin Blache). In some experiments, the drug was injected into the closed eye in order to detect an

Downloaded from http:/www.cambridge.org/core. University of Toronto, on 11 Nov 2016 at 16:31:25, subject to the Cambridge Core terms of use, available at http:/www.cambridge.org/core/terms. http://dx.doi.org/10.1017/S0952523800002169

Involvement of retinal channels in frog OKN

359

eventual central effect, as it was observed with GABA antagonists (Bonaventure et al., 1983). Head OKN observation and eye OKN recording were carried out before administration of the drug, as well as 15 min, 30 min, 45 min, 1 h, and 5 h afterwards. For statistical treatment, the standard deviation was noted in parentheses after the median value, and we used the Wilcoxon signed-ranks test (Conover, 1971).

Effects on head OKN of an intravitreal injection of APB into the open eye After intravitreal injection of APB into the open eye, head OKN evoked by a stimulus in T-N direction was strongly reduced or even totally suppressed (n = 31; Fig. 1). At low concentrations, the extinction speed was significantly reduced: for instance, 30 min after injection of 1 mM (n = 7) and 2 mM (n = 8) of APB it dropped from 14.6 (±1.42) to 2.34 deg/s (±1.43) and to 1.35 deg/s (+1.72), respectively. At 5 mM (n = 12) and at higher concentrations, OKN was totally cancelled 15 min after injection (Fig. 1), and the duration of this depressant effect was roughly correlated with the concentration of APB injected intravitreally. The APB effect was reversible, and total recovery occurred within 24 h. When the drum rota-

Results

The effects of APB and PDA on head OKN* Control monocular OKN (a) Before injection. In monocular vision, the frogs displayed an OKN when stimulated in the temporo-nasal (T-N) direction for the open eye. The average extinction speed of monocular OKN was 14.6 deg/s (±1.42). When the stimulus was in the naso-temporal (N-T) direction, OKN was generally not observed; only 2 frogs among 60 reacted to the N-T stimulation with an extinction speed of 3 deg/s.

Table 1. Velocity gain of monocular head OKN before and after injection of PBS, APB, or PDA at different concentrations, with a drum speed of 3 deg/s Velocity gain Injection type

(b) After injection. After intravitreal injection of PBS (which is the vehicle) into the open eye, no significant modification of head OKN was observed (« = 3; Fig. 1).

II

25

l

0.425

0.382

0.41 0.475 0.373

0 0

APB

1 mM 5 mM 10 mM

0.054

PDA

0.53 0.49 0.44 0.64

5 mM 10 mM 12.5 mM 15 mM

0.44 0.41 0.13 0.1

Each line corresponds to one experimental situation with one frog only.

APB imM

PBS

One hour after

PBS

* Recording of monocular head OKN: Only a few recordings were achieved. After APB injection, the velocity gain was strongly reduced or totally suppressed —the suppressing effect of APB lasting at least 90 min for each of the drum speeds used. After administration of PDA, the velocity gain was only slightly reduced at low concentrations (2 mM, 5 mM, and 10 mM), while it was strongly depressed, at higher concentrations (12.5 mM and 15 mM). The results, summarized in Table 1, reinforced those obtained through the observation of head OKN extinction speed when the animal was unrestrained.

30

Before

APB

APB 2mAA

5 mM

I

20

i

15

5

10

O UJ ui Q.

o

-

\

[

1

[J

-\

»

5

z 2 20

I—1

1

PDA 2mM

i____l

\ \-A—\—

1

PDA 5mM

^ •

-

*

^

- - ^ i - i

. PDA lOmM

PDA

1I 1

TIME

2

ismM

3

(Hours)

Fig. 1. Mean values of extinction speed of monocular head OKN (in deg/s) evoked by a temporo-nasal stimulation, after drug injection into the open eye. The drug concentration is given at the top of each graph. The first value of each graph indicates the extinction speed before injection. Injection of PBS (n = 3); injection of APB: 1 mM (n = 7), 2 mM (n = 8), 5 mM (n = 12); injection of PDA: 2 mM (n = 3), 5 mM (n = 6), 10 mM (n = 8), 15 mM (n = 4). Arrows indicate the injection time. The vertical bar indicates the standard deviation.

Downloaded from http:/www.cambridge.org/core. University of Toronto, on 11 Nov 2016 at 16:31:25, subject to the Cambridge Core terms of use, available at http:/www.cambridge.org/core/terms. http://dx.doi.org/10.1017/S0952523800002169

Y.H. Yucel, B. Jardon, and N. Bonaventure

360

9 deg/s) in T-N as well as in N-T directions, before and after drug injection.

tion was in the N-T direction, no head OKN was observed after APB injection, whatever the injected concentration. Effects on head OKN of an intravitreal injection of PDA into the open eye After intravitreal injection of PDA into the open eye, head OKN was never suppressed at concentrations of 2 mM, 5 mM, and 10 mM. PDA at 2 mM (n = 3) had no effect, but a small decrease of the extinction speed of head OKN occurred after the injection of 5 and 10 mM. Only at a very high concentration of 15 mM (« = 4), the average extinction speed was strongly reduced to 1.5 deg/s (±2.4) 30 min after injection (Fig. 1). Moreover, at all concentrations of PDA, and for all drum speeds used, we generally observed head-resetting fast phases; but it was also observed that some animals blinked their eyelids during the head-resetting fast phases; others did not display this fast OKN component, and they moved their hind legs to adjust the body position. When the stimulation was in the N-T direction, no OKN was observed as it was already the case before injection. The effects of APB and PDA on monocular horizontal eye OKN Eye movements evoked by optokinetic stimulation were recorded at four constant speeds (0.5 deg/s, 3 deg/s, 6 deg/s, and

Control recordings (a) Before injection. Before injection, in a monocular viewing condition, the frog's eye followed predominantly the stripes moving in the T-N direction. Stimulation in the N-T direction was significantly less efficient as seen in Fig. 2A. For T-N stimulation, and after a short delay (about 0.5 s), the eye movements reached their maximal velocity. After the interruption of the slow phase by a resetting fast phase, the velocity of the following slow phase increased and sometimes reached its maximal velocity once again (n = 11; Fig. 2A). The average velocity gain measured with drum speed of 0.5 deg/s was 0.47 (±0.14); it progressively decreased with stimulus velocity, reaching 0.09 (±0.03) when the drum speed was 9 deg/s (Fig. 3). On the other hand, the average frequency of resetting fast phases, which was 1.64 (±1.15) when the drum was rotating at 0.5 deg/s, increased to 3.9 (±1.50) and to 4.18 (±2.74) with drum speeds at 3 deg/s and 6 deg/s, respectively. Then, the average frequency decreased to 2.9 (±1.83), when the drum speed was 9 deg/s. For N-T stimulation, the frog's eye followed the stripes very slowly (for instance, 0.28 deg/s (±0.13) for a drum speed of 3 deg/s), then the animal's eye remained in an eccentric posi-

APB

CONTROL

PDA

TNNT

TN_ NT,

NT_

TN

NT

Fig. 2. Coil recording of ocular OKN evoked by constant drum speeds in a monocular viewing condition. A, eye recordings at different drum speeds before drug injection; B, after intravitreal injection of APB (5 mM); the eye OKN was totally suppressed; C, after PDA (5 mM); the eye was blocked in an eccentric position and no resetting fast phase was observed. OKN are recorded in the same frog in (A) and in (C). The direction of the stimulus (T-N for temporo-nasal and N-T for naso-temporal directions) is indicated on the left of each recording. The drum speeds are indicated at the left of the figure. Arrows point to onset and to stop of the stimulation. Calibration: the vertical bar corresponds to an angular displacement of 1 deg and the horizontal bar indicates a duration of 10 s.

Downloaded from http:/www.cambridge.org/core. University of Toronto, on 11 Nov 2016 at 16:31:25, subject to the Cambridge Core terms of use, available at http:/www.cambridge.org/core/terms. http://dx.doi.org/10.1017/S0952523800002169

361

Involvement of retinal channels in frog OKN

PBS

T«-N

.5

.5

6

3

.5

DRUM

.5

3

SPEED

6

9

('/s )

Fig. 3. Mean values of velocity gain of monocular eye OKN before (white circles), 1 h (dark circles), and 5 h (dark triangles) after drug injection into the open eye. The velocity gain of monocular eye OKN is plotted on the ordinate and the drum speed (in deg/s) on the abcissa. The OKN gain in response to T-N stimulus is drawn on the right and the OKN gain in response to N-T stimulus on the left of each graph. The intravitreal injection of PBS (control) did not affect the OKN velocity gain, whatever the drum speed. One hour after intravitreal injection of APB (5 mM) (n = 11), a strong decrease in the velocity gain was observed; partial recovery was seen 5 h later, for T-N as well as for N-T stimulations. One hour after intravitreal injection of PDA (5 mM) (n = 1!), the velocity gain was not modified, at least for the lower drum speeds. For a drum speed of 9 deg/s, the velocity gain was significantly reduced. The vertical bar indicates the standard deviation.

tion; the eye velocity was nil and no resetting fast phase was observed (Fig. 2A). These observations were made with the four drum speeds tested. Moreover, the velocity gain of the slow phase decreased when the drum speed increased as shown in Fig. 3. The difference between the velocity gain of OKN evoked by T-N stimulation and that evoked by N-T stimulation, is significant for the four drum speeds mentioned, the eye predominantly following the pattern motion in the T-N direction. The frequency of resetting fast phases evoked by a T-N stimulation was also significantly higher than that provoked by an N-T one, for each of the four drum speeds used. (b) After injection of PBS into the open eye. In this condition, no significant modification of the ocular OKN was noticed: the velocity of the slow phases and their gain, as well as the frequency of resetting fast phases, were identical to those recorded before injection (n = 3; Fig. 3). Effects on monocular eye OKN of an intravitreal injection of APB into the open eye As in the control, few eye blinks and no other spontaneous eye movement were recorded after intraocular administration of 5 mM APB into the open eye. For T-N stimulation, and a few minutes after injection, the eye OKN was strongly reduced, or even suppressed. Whatever the drum speed, the eye followed the stripes very slowly or did not move at all (Fig. 2B). With a drum speed of 3 deg/s, the velocity of the slow phase dropped to zero 5 min after injection, and remained at zero when tested 15 min, 30 min, and 45 min later. One hour after injection, the OKN was still suppressed, or its parameters remained strongly reduced. In these conditions, the velocity gain was significantly reduced as shown in Fig. 3 (n = 11) and Fig. 4 (/i = 6) {W< 0.005). Five hours after APB injection, a slight recovery was observed; the velocity gain was higher than that measured at 15 min, 30 min, 45 min, and 60 min after injection, but still remained very low

(Fig. 3). No eye movement was evoked at the end of stimulation either by extinguishing the light or by stopping the drum. For N-T stimulation, the velocity of the eye slow phase was strongly and significantly reduced as shown in Fig. 2B, Fig. 3 (n = 11), and Fig. 4 (n = 6) (W< 0.005). Effects on monocular eye OKN of an intravitreal injection of PDA (5 mM) into the open eye During the optokinetic stimulation in the T-N direction, the eye followed the stripes for a few seconds, and after that, the eye remained blocked in spite of its eccentric position. Thus, no eye resetting fast phase was observed (Fig. 2C) (n = 11). In these conditions, the velocity gain was not modified for the three lower drum speeds used, whatever the delay after the injection (W> 0.05) (Fig. 3 and Fig. 4). On the contrary, a significant decrease in the velocity gain was measured for the highest drum speed used (9 deg/s) (W < 0.005). Moreover, the frequency of the resetting fast phases was strongly reduced or even cancelled, at all drum speeds used. But these effects of PDA were reversible, and less than 24 h after injection, the frequency of the resetting fast phases reached the value measured before drug injection. For N-T stimulation and whatever the drum speed used, PDA had no or little effect on the velocity gain of monocular eye OKN (Figs. 2C, 3, and 4) (« = 11), and as it was already the case before injection, no resetting fast phase was noticed. Thus, an intravitreal injection of PDA (5 mM) into the open eye seems to have no effect on the OKN slow phase following T-N or N-T stimulation, for low stimulus velocities. But it strongly reduced the frequency of the eye-resetting fast phases, even cancelled them, when the head was restrained. Effects on monocular eye OKN of an intravitreal injection of APB (5 mM) or PDA (5 mM) into the closed eye After injection of APB (n = 10) or PDA (n = 10) into the closed eye, no effect on monocular eye OKN was observed

Downloaded from http:/www.cambridge.org/core. University of Toronto, on 11 Nov 2016 at 16:31:25, subject to the Cambridge Core terms of use, available at http:/www.cambridge.org/core/terms. http://dx.doi.org/10.1017/S0952523800002169

Y.H. Yiicel, B. Jardon, and N. Bonaventure

362

APB 5mM

0.4

PDA 5mM

0.3 TN

0.2 0.1

1015

30

45

60min

5h

1015

30

45

60min

T IME Fig. 4. Time course of the monocular OKN velocity gain at a drum speed of 3 deg/s, after intravitreal injection of 5 mM APB (n = 6) or of 5 mM PDA (« = 6) into the open eye. Arrows indicate the injection time. White circles stand for mean values of the OKN velocity gain for T-N stimulation, dark circles for N-T stimulation. The vertical bar indicates the standard deviation.

(Fig. 5): the velocity gain was similar to that observed before injection of APB or of PDA. Conclusion and discussion Our results, obtained in the monocular viewing condition, confirmed the clear directional asymmetry of eye and head OKN of the frog observed by Birukow (1937), and recorded by Dieringer and Precht (1982) and Dieringer et al. (1983). Moreover, the coil recording technique allowed us to show that the N-T component that is not visible by the naked eye actually exists especially in the eye OKN. We have shown that in the frog an intravitreal injection of APB into the open eye strongly reduced, or even suppressed both horizontal head and eye OKN. This depressant effect of APB is observed for stimulations in the T-N, as well as in the N-T directions and at all drum speeds used. Injected at low

concentrations into the open eye, PDA did not modify the velocity gain of eye and head OKN slow phase, at least for low drum velocities; at higher concentrations, PDA provoked a decrease in the OKN slow phase, but never its total suppression. These results must be related to observations recently made in our laboratory concerning the modification of the retinal message after intravitreal injection of these drugs. Briefly summarized, these results are the following: when intravitreally injected into the frog in vivo, APB and PDA showed opposite effects on ON and OFF retinal channels. APB abolished ON responses in the ERG as well as in transient ganglion cell activity, while increasing OFF responses; at the same time, the receptive-field area measured at OFF was enlarged, this enlargement was due to the suppression of the inhibition exerted by the surround upon the center of the receptive field. On the contrary, PDA abolished OFF responses in the ERG as well as in transient ganglion cell activity; it provoked an increase in the

GAIN APB 5mM

T-».N 0.2

0.1

.5

.5

6

9

DRUM Fig. 5. Mean values of velocity gain of monocular eye OKN, before (white circles) and 1 h (dark circles) after injection of APB (n = 10) or PDA (n = 10) into the closed eye. The velocity gain is plotted on the ordinate, the drum speed (in deg/s) on the abcissa. OKN gain in response to T-N stimulus is drawn on the right and OKN gain in response to N-T stimulus on the left of each graph. The OKN velocity gain was not significantly modified in these conditions.

Downloaded from http:/www.cambridge.org/core. University of Toronto, on 11 Nov 2016 at 16:31:25, subject to the Cambridge Core terms of use, available at http:/www.cambridge.org/core/terms. http://dx.doi.org/10.1017/S0952523800002169

Involvement of retinal channels in frog OKN ON response of the ERG, while decreasing the ganglion cell sensitivity at ON, probably by blocking the synaptic connections from ON bipolar to ganglion and to amacrine cells. The transient ganglion cell receptive-field area, measured at ON, was not modified but the surround inhibition (which exists in the control) was suppressed, probably by a blocking effect upon the glutamate receptors of horizontal cells (Jardon et al., 1989). Some indications lead us to believe that there is a direct relationship between the effects recorded at the retinal level and those observed in OKN. Indeed, efficient drug concentrations were similar in both types of experiments; in the same way that the duration of the APB and PDA depressant effect was analogous at retinal level as well as in OKN. Moreover, intraperitoneal administration of APB (« = 5) and PDA (n = 5) at concentrations previously used had no effect either on contralateral ganglion cell ON responses, or on head or eye OKN. In addition, intravitreal injection of APB or PDA into the closed eye did not modify either ganglion cell activity of the other eye, or the OKN triggered by this open eye. This fact indicates that APB and PDA injected into the open eye did not act on central structures, but only at the retinal level. In a previous study, we have shown that the modifications in the frog OKN, observed after intravitreal injection of GABA antagonists, resulted from alterations of the retinal message, but also from a direct effect of the drug on central structures; indeed, in monocular viewing conditions, the intraperitoneal administration of bicuculline or of picrotoxin or their intravitreal injection into the closed eye had provoked modifications in the OKN triggered by the open eye (Bonaventure et al., 1983). Similar effects were never observed after injection of PDA or APB into the closed eye. Thus, OKN modifications seem to be mainly related to alterations of the ganglion cell message. It is then tempting to suggest that, in the frog, the retinal ON channel (or part of it) blocked by APB conveys the retinal input responsible for horizontal head and eye OKN, but this correlation must be done cautiously; in the frog, APB totally blocked ON ganglion cell responses (and the ON component of the transient ganglion cells responses) while increasing the OFF component as observed in the cat (Bolz et al., 1984; Chen & Linsenmeier, 1987). However, we have no precise indication about the effect of APB or PDA on the other types of ganglion cells such as the sustained cells or dim cells (groups I and IV) according to the classification of Maturana et al. (1960), (Gordon & Hood, 1976). On the other hand, an injection of APB (5 mM) provoked an enlargement of the OFF receptive field of the transient ganglion cell, this increase being due to the suppression of the inhibition normally exerted by the surround upon the center of the field. It can be noticed that in the monkey (Schiller, 1982), the center surround organization of the OFF-center cells was unaffected by APB, suggesting that in this species the ON and OFF systems do not interact significantly at this level. On the contrary, we have shown that both systems in the frog do interact already in the retina. Therefore, the abolition of eye and head OKN after APB may be correlated in the frog with the abolition of the ON responses, but also with the modification of the spatial organization of the OFF receptive field, probably by alterations of contrast sensitivity. It is wellknown that the spatial sensitivity to contrast is directly related to the receptive-field area: the smaller the receptive field, the higher the contrast sensitivity (Barlow et al., 1957). Indeed after APB, a decrease of the contrast sensitivity was observed in the monkey (Schiller et al., 1986; Smith et al., 1987).

363

We have shown that APB abolished the frog OKN for the four speeds tested, when the same optokinetic pattern was used (black and white stripes of 10-mm width). It would be interesting to study the OKN modifications after APB with stimulus pattern of different spatial frequencies in relation to the modifications of the ganglion cell receptive-field area. Indeed, in the goldfish, it was shown that APB suppresses OKN for high stimulus velocity and high spatial frequency (DeMarco et al., 1987). Nevertheless, the role of the retinal message in the frog OKN generation is further supported by the experiments in which the OFF retinal information channel was suppressed by PDA. At the lowest concentrations used (1 mM and 2 mM), PDA did not modify the OKN or the ON activity of the transient ganglion cells, while partially decreasing the OFF responses. At higher concentrations (10 mM and 15 mM), PDA provoked the total disappearance of the activity at OFF, and a decrease of the ON responses, related to the concentration of the injected drug. The receptive-field area was not modified, but the inhibitory surround was abolished. In these conditions, the extinction speed of head OKN was significantly reduced, as was the velocity gain of head and eye OKN. This could be related to the reduction of the ON responses. It must be noticed that OKN as ON responses were never totally suppressed by PDA. Therefore, in spite of the total abolition of the OFF channel, and of the surround inhibition upon the center of the ON receptive field after PDA of 5 mM, the retinal information concerning the detection of direction and speed of the stimulus necessary to generate the OKN slow phases was still conveyed to the mesencephalic structures, at least for lower drum speeds. On the other hand, it seems that for higher drum speeds, the OFF channel could be involved in the generation of the OKN slow phase. In the frog, it is well-known that the pretectal nuclei and the nBOR, described in detail (Montgomery et al., 1981; 1985), mediate the OKN (Montgomery et al., 1982; Lazar et al. 1983) and contain direction-selective cells (Kondrashev & Orlov, 1976; Katte & Hoffmann, 1980; Cochran et al., 1984; Gruberg & Grasse, 1984). Our results suggest that the retinal input to these mesencephalic structures is made to a large extent by ON ganglion cells. But we have no evidence whether these ganglion cells are also direction-selective ones (Backstrom et al., 1978; Watanabe & Murakami, 1984). These results agree with the hypothesis that in the rabbit, the mesencephalic structures, i.e. the medial terminal nucleus of the AOS and the nucleus of the optic tract (NOT), responsible for OKN and containing direction-selective cells, mainly receive ON direction-selective ganglion cells (Oyster & Barlow, 1967; Oyster et al., 1972; Collewijn, 1975; Simpson et al., 1979; Buhl & Peichl, 1986). The observation of Rademaker and Ter Braak (1948) was consistent with this hypothesis: the inner surface of an optokinetic drum being half white, half black, and only the white leading edge, and not the black, was able to trigger an OKN in the rabbit. APB totally abolished the horizontal OKN not only in the frog, but also in the rabbit and in the turtle (Knapp & Ariel, 1984; Ariel, 1987). But in the cat, the results of similar experiments were slightly different: an intravitreal administration of APB (Hoffmann, 1986) did not suppress the OKN: only the velocity gain was slightly reduced, especially at a high velocity range. In the cat, it was shown by the same author that a relevant cortical output for OKN was more efficient than the direct retinal ON message to the NOT. The NOT neurons are driven

Downloaded from http:/www.cambridge.org/core. University of Toronto, on 11 Nov 2016 at 16:31:25, subject to the Cambridge Core terms of use, available at http:/www.cambridge.org/core/terms. http://dx.doi.org/10.1017/S0952523800002169

364

by the direct retinal input alone, only during the first few months of life (Van Hof-Van Duin, 1978; Mallach et al., 1981; Hoffmann, 1983). It could then be interesting to examine if APB would have a more depressant effect on the kitten's OKN during this short period. To conclude, it seems that the spatio-temporal modification of the ganglion cell activity, observed after an intravitreal injection of APB, could be responsible for the modifications observed in the frog OKN. Our results strongly suggest that the ON retinal response with its spatial organization in center-surround arrangement, blocked by APB and decreased by high concentrations of PDA, conveys the retinal input generating the head and eye OKN. On the contrary, the OFF retinal channel blocked by PDA does not seem to be involved in the generation of the OKN slow phase in the frog, for lower drum velocities. Moreover, it should be noted that eye-resetting fast phases were suppressed after PDA injection when the animal's head was restrained. This point will be separately examined in an other paper (Yucel et al., 1989). Acknowledgments We wish to thank Professor Dr. N. Dieringer, Institute of Physiology, University of Munich, for helpful criticism and advice; E. Dreyfus for the electronics; A. Guigue and S. Stein for the mechanical setup; and G. Rudolf for iconography. This study was supported by a grant of the Fondation pour la Recherche Medicale Francaise.

References ARIEL, M. (1987). Synaptic drugs injected into the vitreous affect the retinal control of turtle eye movements. 17th Annual MeetingSociety of Neuroscience Abstracts 13, Part 3, 1053. BACKSTROM, A.-C, HEMILA, S. & REUTER, T. (1978). Directional selec-

tivity and color coding in the frog retina. Medical Biology 56, 7283.

Y.H. Yiicel, B. Jardon, and N. Bonaventure CUNNINGHAM, J.R. & NEAL, M.J. (1985). Effect of excitatory amino acid on gamma-aminobutyric acid release from frog horizontal cells. Journal of Physiology 362, 51-67. DAVIES, J., EVANS, R.H., JONES, A.W., SMITH, D.A.S. & WATKINS,

J.C. (1982). Differential activation and blockade of excitatory amino-acid receptors in the mammalian and amphibian central nervous systems. Comparative Biochemistry and Physiology 27, 211224. DEMARCO, P.J., NUSSDORF, J.D., BROCKMAN, D.A. & POWERS, M.K.

(1987). The ON channel is necessary for optokinetic nystagmus at high spatio-temporal frequency in goldfish. 17th Annual Meeting— Society of Neuroscience Abstracts 13, Part 3, 1053. DIERINGER, N. & PRECHT, W. (1982). Compensatory head and eye movement in the frog and their contribution to stabilization of gaze. Experimental Brain Research 47, 394-406. DIERINGER, N., COCHRAN, S.L. & PRECHT, W. (1983). Differences in

the central organization of gaze stabilizing reflexes between frog and turtle. Journal of Comparative Physiology 153, 495-508. GIOANNI, H., REY, J., VILLALOBOS, J. & DALBERA, A. (1984). Single

unit activity in the nucleus of the basal optic root (nBOR) during optokinetic, vestibular, and visuo-vestibular stimulations in the alert pigeon (Columbia livia). Experimental Brain Research 57, 49-60. GORDON, J. & HOOD, D.C. (1976). Anatomy and physiology of the frog retina. In The Amphibian Visual System. A Multidisciplinary Approach, ed. FITE, K.V., pp. 29-86. New York, San Francisco, London: Academic Press. GRUBERG, E.R. & GRASSE, K.L. (1984). Basal optic complex in the frog (Rana pipiens): a physiological and HRP study. Journal of Neurophysiology 51, 998-1010. HOFFMANN, K.-P. (1983). Effects of early monocular deprivation on visual input to cat nucleus of the optic tract. Experimental Brain Research 51, 236-246. HOFFMANN, K.-P. (1986). Visual inputs relevant for OKN in mammals. In The Oculomotor and Skeletal Motor Systems: Differences and Similarities: Progress in Brain Research, Vol.64, ed. FREUND, H.-J., BUTTNER, U., COHEN, B. & NOTH, J., pp. 75-84. Amsterdam, New

York, Oxford: Elsevier. HOFFMANN, K.-P. & SCHOPPMANN, A. (1981). A quantitative analysis of the direction-specific responses of neurons in the cat's nucleus of the optic tract. Experimental Brain Research 42, 146-157. HOFFMANN, K.-P. & STONE, J. (1985). Retinal input to the nucleus of the optic tract of the cat assessed by antidromic activation of ganglion cells. Experimental Brain Research 59, 395-403.

BARLOW, H.B., FITZHUG, R. & KUFFLER, S.W. (1957). Change of orga-

JARDON, B., YUCEL, Y.H. & BONAVENTURE, N. (1989). Role of the

nization in the receptive fields of the cat's retina during dark adaptation. Journal of Physiology 137, 338-354. BIRUKOW, G. (1937). Untersuchungen iiber den optischen Drehnystagmus iiber die Sehscharfe des Grasfrosches (Rana temporaria). Zeitschrift fiir Vergleichende Physiologie 25, 92-142.

glutamatergic system in the separation of the ON and OFF channels in the frog retina; possible modulation by the cholinergic and glycinergic systems. European Journal of Pharmacology. KATTE, O. & HOFFMANN, K.-P. (1980). Direction-specific neurons in the pretectum of the frog (Rana esculenta). Journal of Comparative Physiology 140, 53-57. KNAPP, A.G. & ARIEL, M. (1984). Selective blockade of retinal ON channel eliminates horizontal optokinetic nystagmus in rabbits. The Annual Meeting of the Association of Research in Vision and in Ophthalmology 25, 229. KOCH, U.T. (1977). A miniature movement detector applied to recording of wing beats in Locusta. Fortschritte der Zoologie 24, 327-332. KONDRASHEV, S.L. & ORLOV, O.Y. (1976). Direction-sensitive neurons in the frog visual system. Neirofiziologiya 8, 196-198.

BOLZ, J., WASSLE, H. & THIER, P. (1984). Pharmacological modula-

tion of ON and OFF ganglion cells in the cat retina. Neuroscience 12, 875-885. BONAVENTURE, N., WIOLAND, N. & BIGENWALD, J. (1983). Involvement

of GABAergic mechanisms in the optokinetic nystagmus of the frog. Experimental Brain Research 50, 433-441. BONAVENTURE, N., YUCEL, Y.H. & JARDON, B. (1987). Effects of the

glutamate antagonists, 2-4 amino-phosphonobutyric acid (APB) and cis-2,3,-piperidine dicarboxylic acid (PDA), on ERG, retinal ONOFF ganglion cell responses and optokinetic nystagmus (OKN) in the frog. The Second World Congress of Neuroscience (IBRO), p. 731. BUHL, E.H. & PEICHL, L. (1986). Morphology of rabbit retinal ganglion cells projecting to the medial terminal nucleus of AOS. Journal of Comparative Neurology 253, 163-174. CHEN, E.P. & LINSENMELER, R.A. (1987). The effects of APB on the

contrast sensitivity of cat retinal X cells. 17th Annual Meeting — Society of Neuroscience Abstracts 13, Part 2, 381. COCHRAN, S.L., DIERINGER, N. & PRECHT, W. (1984). Basic optoki-

netic ocular reflex pathways in the frog. Journal of Neuroscience 4, 43-57. COLLEWLTN, H. (1975). Direction-selective units in the rabbit's nucleus of the optic tract. Brain Research 100, 489-508. CONOVER, W.J. (1971). The use of ranks. In Practical Nonparametric Statistics, pp. 203-216. New York, London, Sydney, Toronto: Wiley.

LAZAR, G., ALKONYI, B. & TOTH, P. (1983). Re-investigation of the role

of the accessory optic system and pretectum in the horizontal optokinetic head nystagmus of the frog. Lesion experiments. Ada Biologica Hungarica 34, 385-393. MALLACH, R., STRONG, N. & VAN SLUYTER, R.C. (1981). Analysis of

monocular optokinetic nystagmus in normal and visually deprived kittens. Brain Research 210, 367-372. MANTEUFFEL, G. (1984). Electrophysiology and anatomy of directionspecific pretectal units in the salamander (Salamandra salamandra). Experimental Brain Research 54, 415-425. MATURANA, H.R., LETTVIN, J.Y., MCCULLOCH, W.S. & PITTS, W.H.

(1960). Anatomy and physiology of vision in the frog (Rana pipiens). Journal of General Physiology 43 (Suppl.), 129-175. MILLER, R.F. & SLAUGHTER, N.M. (1986). Excitatory amino-acid receptors of the retina: diversity of subtypes and conductance mechanisms. Trends in Neuroscience 9, 211-217. MONTGOMERY, N., FITE, K.V. & BENGSTON, L. (1981). The accessory

Downloaded from http:/www.cambridge.org/core. University of Toronto, on 11 Nov 2016 at 16:31:25, subject to the Cambridge Core terms of use, available at http:/www.cambridge.org/core/terms. http://dx.doi.org/10.1017/S0952523800002169

Involvement of retinal channels in frog OKN

365

optic system of Rana pipiens: neuroanatomical connections and intrinsic organization. Journal of Comparative Neurology 203, 595612.

SHELLS, R.A., FALK, G. & NAGHSHTNEH, S. (1981). Action of glutamate

MONTGOMERY, N., FITE, K.V., TAYLOR, M. & BENGSTON, L. (1982).

SIMPSON, J.I., SOODAK, R.E. & HESS, R.E. (1979). The accessory optic

Neuronal correlates of optokinetic nystagmus in the mesencephalon of Rana pipiens: functional analysis. Brain Behavior Evolution 21, 137-153.

system and its relation to the vestibulocerebellum. In Reflex Control of Posture and Movement: Progress in Brain Research, Vol: 50,

MONTGOMERY, N., FITE, K.V. & GRIGONIS, A.M. (1985). The pretec-

tal nucleus Lentiformis Mesencephali of Rana pipiens. Journal of Comparative Neurology 234, 264-275. NEAL, M.J., CUNNINGHAM, J.R., JAMES, T.A., JOSEPH, M. & COLLINS,

J.F. (1981). The effect of 2-amino-4-phosphonobutyrate (APB) on acetylcholine release from rabbit retina: evidence for ON channel input to cholinergic amacrine cells. Neuroscience Letters 26, 301305. OYSTER, C.W. & BARLOW, H.B. (1967). Direction-selective units in rabbit retina: distribution of preferred directions. Science 155, 841-842. OYSTER, C.W., TAKAHAHI, E. & COLLEWUN, H. (1972). Direction-

selective retinal ganglion cells and control of optokinetic nystagmus in the rabbit. Vision Research 12, 183-193. RADEMAKER, G.G.J. & TER BRAAK, J.W.G. (1948). On the central mechanism of some optic reactions. Brain 71, 48-76. SCHILLER, P.H. (1982). Central connections of the retinal ON and OFF pathways. Nature 297, 580-583. SCHILLER, P.H., SANDELL, J.H. & MAUNSELL, J.H.R. (1986). Function

of the ON and OFF channels of the visual system. Nature 322, 824825.

and aspartate analogues on rod horizontal and bipolar cells. Nature 294, 592-594.

GRANIT, R. & POMPEIANO, 0 . , p p . 715-724. Amsterdam, New

York, Oxford, North Holland: Elsevier. SLAUGHTER, M.M. & MILLER, R. (1981). 2-amino-4-phosphonobutyric

acid: a new pharmacological tool for retina research. Science 211, 182-185. SMITH, E.L., DUNCAN, G.C., HARWERTH,

R.S.S. & CRAWFORD,

M.L.J. (1987). Spatial contrast sensitivity deficits produced by ON channel blocked in the Rhesus Monkey. 17th Annual MeetingSociety of Neuroscience Abstracts 13, Part 2, 381. SiicHER, N., PRZYBYZEWSKI, A. & GRUSSER, O.-J. (1986). The effect of amino-phosphonobutyric acid (APB) and of cis-2,3-piperidine dicarboxylic acid (PDA) on flash responses of retinal ganglion cells. In Abstract of Annual Meeting of Neuroscience 13, 381. VAN HOF-VAN DUIN, J. (1978). Direction preference of optokinetic response in monocularly tested normal kittens and light-deprived cats. Archives Italiennes de Biologie 116, 471-477. WATANABE, S. & MURAKAMI, M. (1984). Synaptic mechanisms of directional selectivity in ganglion cells of frog retina as revealed by intracellular recordings. Japanese Journal of Physiology 34, 497-511. YUCEL, Y.H., JARDON, B. & BoNAVENTURE, N. (1989) Is a retinal input involved in the generation of eye-resetting fast phase in the frog eye optokinetic nystagmus? Neuroscience Letters.

Downloaded from http:/www.cambridge.org/core. University of Toronto, on 11 Nov 2016 at 16:31:25, subject to the Cambridge Core terms of use, available at http:/www.cambridge.org/core/terms. http://dx.doi.org/10.1017/S0952523800002169