Acta Oto-Laryngologica
ISSN: 0001-6489 (Print) 1651-2251 (Online) Journal homepage: http://www.tandfonline.com/loi/ioto20
Retinal Ganglion Cells Related to Optokinetic Nystagmus in the Rat Isao Kato, Tomoyuki Okada, Shoji Watanabe, Shigeki Sato, Tamotsu Urushibata And & Isamu Takeyama To cite this article: Isao Kato, Tomoyuki Okada, Shoji Watanabe, Shigeki Sato, Tamotsu Urushibata And & Isamu Takeyama (1992) Retinal Ganglion Cells Related to Optokinetic Nystagmus in the Rat, Acta Oto-Laryngologica, 112:3, 421-428 To link to this article: http://dx.doi.org/10.3109/00016489209137422
Published online: 08 Jul 2009.
Submit your article to this journal
Article views: 6
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ioto20 Download by: [137.189.171.235]
Date: 16 March 2016, At: 06:53
Acta Otolaryngol (Stockh) 1992; 112: 421-428
Retinal Ganglion Cells Related to Optokinetic Nystagmus in the Rat ISAO KATO, TOMOYUKI OKADA, SHOJI WATANABE, SHIGEKI SATO, TAMOTSU URUSHIBATA and ISAMU TAKEYAMA From the Deparlment of Otarhinolatyngolagv, St. Marianna University School of Medicine. Kawasaki, Japan
Acta Oto-Laryngologica 1992.112:421-428.
&to I, Okada T, Watanabe S, Sat0 S, Urushibata T, Takeyama I. Retinal ganglion cells related to optokinetic nystagmus in the rat. Acta Otolaryngol (Stockh) 1992; 112: 421-428. The nucleus of the optic tract (NOT) in the pretectum is the visuo-motor relay between the retina and preoculomotor structure in the pathway conveying signals responsible for optokinetic nystagmus (OKN) in both afoveate and foveate animals. Unilateral lesions of the NOT abolish OKN toward the side ipsilateral to the lesion. However, what parts of the retina and what kinds of retinal ganglion cells project their fibers into the NOT are still unknown. To examine this, horseradish peroxidase conjugated with wheatgerm agglutinin was injected into the NOT of the rat. Labeled retinal ganglion cells were diffusely documented in the contralateral side, whereas those situated only in the lower temporal crescent, were found in the ipsilateral side. Ganglion cells dominated in the range of small cells, i.e., W cells. Therefore, ganglion cells responsible for OKN are believed to be related to W cells. However, medium-sized and large cells were definitely labeled in the rat, suggesting possible participation of these cells in OKN. Key words: pigmented rat, retinal ganglion cells. nucleus of the optic tract, OKN.
INTRODUCTION Recent neurophysiological and ablation studies have disclosed that the nucleus of the optic tract (NOT) is the visuo-motor relay between the retina and preoculomotor structures in the pathway mediating optokinetic reflex in rats (l), rabbits (2), cats (3) and monkeys (4, 5). Afoveate animals such as rats and rabbits have predominantly crossed retinofugal fiber projection to the NOT. Unilateral lesions of the NOT abolish optokinetic nystagmus (OKN) in any stimulus mode during monocular stimulation of the contralateral eye (1,2). In cats, which have both crossed and uncrossed retinofugal fibers and produce OKN of almost equal magnitude in both directions, unilateral lesions of the NOT likewise abolish OKN in response to surround motion toward the lesioned side (6). In monkeys, which have the pursuit system, the same was true with a rapid rise kept intact (4). However, it is still open to discussion from what parts of the retina it stems and what kinds of retinal ganglion cells project their fibers into the NOT. To achieve this goal, we used pigmented rats, because these afoveate animals have a fundamentally old OKN system and none of the OKN are changed in ablation of either cerebrum or cerebellum or both, and only the crossed temporonasal system generates significant OKN responses (6).
MATERIAL AND METHODS Twenty-one female Long-Evans rats weighing about 250-270 g were used for the present experiment, and 12 female Wistar rats as controls.
Retinal ganglion cells In the first part of the present experiment, pigmented rats were anesthetized with intrapentoneal injection of pentobarbital sodium (20 m&) and mounted on a stereotaxic apparatus. For surgical procedures, 1% xylocaine was used topically. A lateral incision sufficiently large
Acta Oto-Laryngologica 1992.112:421-428.
422 I. Karo et al.
ActD Otolaryogol (Stockb) I I2
Fig. 1. A photomicrograph of the ganglion cell layers. (A) Large, medium-sized and small cells are labeled 1 , 2, and 3, respectively;(B)Sagittal incision of the retina. Labeled cells are located in the ganglion cell layer. Each horizontal bar indicates 100 pm.
to expose the I& optic nerve just posterior to an eye ball was made. For massive injections of horseradish peroxidase, a microsyringe of 10 was stereotaxically located so as to have the tip penetrate the optic nerve, and a 10% solution of horseradish peroxidase conjugated with wheatgerm agglutinin (WGA-HRP) was injected into the nerve through a microsyringe under an operating microscope. After a survival time of about 48 h the animals were reanesthetized with an overdose of pentobarbital sodium and sacrificed by vascular perfusion of 0.9% NaCl followed by 3 1 of 1 % paraformaldehyde and 2 % glutaraldehyde in 0. I M phosphate buffer pH 7.4,650 mOsm. After completing perfusion, the full extent of the retina, up to the ora serata, was dissected free. One large incision was made at the top of the eye ball to facilitate correct alignment of the preparation on the slide, and to avoid folding, several radial incisions were made in the periphery, if necessary. Thus, whole mount preparations of the retina were prepared. Shrinkage in the retinal area after the dehydration processes was estimated at 5-1096 (7). Therefore, no correction was made for cell soma. Soma sizes of ganglion cells were measured in photomicrographs magnified 250 times (Fig. 1 A). Roughly estimated, large cells had bodies with a diameter of about 30 pm; medium-sized cells, with a diameter of about 15 wm; small cells had a diameter of about 5 pm, and, furthermore, these cells were precisely confirmed to be located in the retina (Fig. 1 B).
NOT identification Animals were anesthetized with intraperitoneal injection of urethane (1.2 &) and achloralose (30m&). Animals were mounted on a stereotaxic apparatus. The dorsal aspect of the pretectum and the superior colliculus was exposed after suctioning the cerebral hemisphere under an operating microscope. A glass capillary electrode filled with 3 M KCI was inserted
Rat retinal ganglion cells related to OKN
Acta Oto-Laryngologica 1992.112:421-428.
Acta Otolaryngol (Stockh) 112
caudal
-
Fig. 2. HRP injection site of pigmented rat 19 (PR 19). (A) Densely stained area is restricted to the NOT and diffusion spreads to adjacent areas. Arrow indicates the injection site; (B) Schematic representation of the injection site. Horizontal sections are rostrocaudal from above to below at 500 pm. Black areas indicate densely stained sites which include NOT; hatched areas, diffusion of HRP. Diffusion covered OPN and PPN. Each horizontal bar indicates I mm.
into the anterolateral border of the superior colliculus. The NOT was identified by recording a unit responding to lights which were held by a hand from the outside of a shielded room and moved slowly in the horizontal direction. Units in the NOT definitely responded in a direction-selective way, and in the right site, 0.01 pl of 10%WGA-HRP solution was injected mechanically into the NOT through a micropipette connected tightly with a microsyringe. After a survival time of about 48 h, the animals were reanesthetized deeply and perfused with 0.9% saline followed by fixation. Injection sites of the brain and the retina were processed for histochemical demonstration of WGA-HRP by the tetramethyl benzidine (TMB)protocol (8) and the Hanker-Yeates method (9), respectively. Subsequently, the injection site was counterstained with 1.0% neutral red.
Soma area measurements Photomicrographs of retinal ganglion cells suggest that a three-group classification could be possible in the rat as well (Fig. I A). However, the soma size of these cells varied depending upon whether the dendrites of the cells were stained or not. Therefore, cell somata of labeled cells were photographed on a negative film on which alphabetical letters were printed and they were used as a landmark to identify the sites of labeled ganglion cells. The ganglion cells were enlarged precisely 250 times their actual size. Photographed cell soma areas were measured by a tracing device and compared with each other in great detail. Ganglion cell density In order to clarify whether the distribution of retinal ganglion cells varies in different regions within the retina, ganglion cell density was established in the central, temporal, nasal, dorsal and ventral regions of the retina. Ganglion cells ipsilateral to the injection site were observed
423
10
(rim )
-L
25-
50
50-
75
75
- 100
100125
t25
- 150
150115
175 200
-
Acta Oto-Laryngologica 1992.112:421-428.
i
200 - 225 225 250 250 215 275 300 300 325 325 - 350 350 375 375 - 400
-
20
30
40
50
60
70
80
90
Cell density
' 49
Fig. 3. Cell density per unit area ( I .O mm2) in each representative area and cell distribution in the ventral area contralateral to the injection site. Labeled cells were almost evenly distributed except in the ventral area. Cell decrease was noticed in the ventral area. Soma areas of labeled cells predominated in the range of small cells of between 25 and 50 pm2 in soma areas. The same was true of other areas.
only in the temporoventral region. These regions were arbitrarily selected in the areas where labeled cells were evenly distributed in 1.O mm'. Ganglion cell density was counted per unit area (1.0 mm2). Cell soma larger than 6 bm in diameter and containing irregularly stained Nissl granules were judged as ganglion cells (10). RESULTS
The conclusions of this study were drawn on the basis of 6 (out of 21) rats, and 15 rats were excluded due to an inadequate injection site of WGA-HRP or large diffusion. In rat 19 (PR 19) processed with the TMB method, a densely stained area was surrounded by a faint one (Fig. 2). The densely stained area was determined as the injection site, whereas the faint one was due to diffusion of WGA-HRP, in agreement with the result reported by Mesulam (8). Enlarged drawings of histological sections were made with the aid of a camera lucida. In rats 13, 18, 19, and 21, the injection site of HRP was restricted to the NOT and diffusion of HRP extended slightly to the posterior pretectal nucleus (PPN). In the remaining two rats (7, I2), the injection site was localized in the NOT and diffusion of HRP encrouched on the olivary pretectal nucleus (OPN), and the PPN as well as the medial border of the medial geniculate body.
Ganglion cell density Ganglion cell densities were counted in the whole mount preparation of the right retina contralateral to the injection site (Fig. 3). Numerals in the figure are all labeled cells sampled per unit areas (1 .O mm'). In rat 19, ganglion cells were most densely labeled in the dorsal area (D) (Fig. 3). In the remaining 5 rats, however, there was no clear tendency for the density of ganglion cells to increase in any particular area except in the ventral area (V). In the ventral area, ganglion cells were least labeled. In the left retina ipsilateral to the injection site, ganglion cells were labeled only in the temporoventral area. Labeled cells per unit area were 379, which
Rat retinal ganglion cells related to OKN 425
Acta Otolaryngol (Stockh) 112
Soma area
(am‘)
Acta Oto-Laryngologica 1992.112:421-428.
0-
25
25-
50
50-
75
10
PR 19 20
30
40
%
50
60
70
80
90
Fig. 4. Cell density per unit area and cell distribution in the ventral area ipsilateral to the injection si?e. Labeled cells were observed only in the temporoventral area. Cell density in this area increased. In soma areas the larger cells of more than 100 pnz increased.
was 7.6 times larger than in the corresponding contralateral retina (Fig. 4). The same trends in ganglion cell density were apparent in the other retinas studied.
Ganglion cell soma area In order to clarify whether the distribution of retinal ganglion cells varies depending upon the different regions of the retina, frequency distribution was established depending upon some areas of retinal ganglion cells in each unit area. In the ventral area in Fig. 3, soma areas of labeled ganglion cells contralateral to the injection site predominated in the range between 25 and 50 pm2. Next to this range, the range between 0 and 25 pm2 was followed. Labeled cells greater than 100 pm2 in soma areas were in the minority. In the remaining prescribed areas, the distribution of ganglion cell soma areas showed a similar tendency, although cell density varied according to area (Fig. 3). In the retina ipsilateral to the injection site, however, soma areas predominated in the range between 0 and 25 pm2 which decreased slowly, and soma areas greater than 100 pm2appeared to increase in frequency, which was in marked contrast to the contralateral retina (Fig. 4). Fig. 5 shows a photomicrograph of labeled cells of the ventral areas both ipsilateral and contralateral to the injection sites. The increase of cell density and soma areas markedly predominated on the ipsilateral side. Table I tabulates labeled large cells greater than 200 Fm2 in soma areas either contralateral or ipsilateral to the injection site, and albino rats served as controls. In ventral retinas contralateral to the injection site, none of the large cells were labeled either in pigmented or albino rats, whereas in those ipsilateral to the injection site of the pigmented rats, about 7% of large cells were labeled in the prescribed area. In the albino rats, the relative frequencies can be recognized in Table I. In other parts of the retina, large cells were distributed sporadically all over the retina without any special predominance. DISCUSSION The concept of a three-group classification of the retinal ganglion cells in the cat has recently been established and is based on differences in receptive field properties and morphology of
426 I . Kato et a!.
Acta Oto-Laryngologica 1992.112:421-428.
Ips vent
Acta Otolaryngol (Stockh) 112
PR 19
Ctr vent
Fig. 5. A photomicrograph of labeled cells in ventral (venr)areas both ipsilateral (Ips)and contralateral (Ctr)to the injection site. Large,medium-sized and small cells are indicated as 1,2, and 3, respectively. An increase of cell density and soma areas are noticed in the ipsilateral side (Ips venr). Horizontal bar indicates 100pm.
the ganglion cells, measurement of axonal conduction velocities, and central distribution of axons (1 1-14). In the present experiment, the three-group classification of the retinal ganglion cells could be applied to the rat in addition to being based on soma sizes and areas (Figs. 1 A, 5). which is compatible with results of other experiments (10, 15). Dense WGA-HRP-stained areas were restricted to the NOT and diffusion of HRP extended slightly to the OPN and/or the PPN in representative cases. A recent physiological study has disclosed that luminance and darkness detector cells are to be found in the OPN and the PPN in the rat, both of which receive direct visual inputs (16). In the present experiment, we could not specify the retinal ganglion cells solely related to the NOT, even though contamination was very slight. Retrogradely-labeled retinal ganglion cells recognized by injecting WGA-HRP in the NOT
Table I. Cell distribution of large cells, greater than 200 pm2 in soma areas Large cells distributed sporadically all over the retina except in the ventral area contralateral (Ctr) to the injection site. No large cells were labeled in either pigmented (PR) of albino rats (AR), and those ipsilateral (Ips) to the injection site increased by about 7% in PR and by 2% in AR
L-NOT Rat no. Inj. site
AR36
AR38
PRl2
PR19
Ctr
Temporal
0 0 1.50
0.1 I 0 0.11
2.35 0
0.86 0.46
0.30
Ips
Central DoMl NaJal ventral Ventral
0 0 2.33
0 1.10
0.90 0 0 7.06
0.64 4.65 0 7.65
Acta Oto-Laryngologica 1992.112:421-428.
Acta Otolaryngol (Stockh) 112
Rat retinal ganglion cells related to OKN 427
dominated in the range of small cells, i.e., W cells. Therefore, retinal ganglion cells responsible for OKN are assumed to be related to W cells. This is in agreement with results in the cat (17). However, in contrast to the cat, medium-sized, and large cells were definitely labeled in the rat, suggesting possible participation of these cells in OKN (1 8). We used albino rats as controls. OKN cannot be elicited in albino rats. Similarly, vestibular nuclear neurons of the horizontal canal system fail to respond to optokinetic stimulation in these animals (19). There have been comparative studies comparing several morphological characteristics between albino and pigmented rats (20). At present, it is not known why the OKN system in the albino rat is so poorly developed. Judging from cell density and distribution of retinal ganglion cells studied, we could not find any abnormalities characteristic of the albino rat (21). In rats injected with WGA-HRP in the pretectum, the ipsilateral distribution of ganglion cells with pretectoretinal projection fibers was confined to the lower temporal crescent. The situation was quite similar to results reported in normal hooded rats (15,22) and in albino rats (7). In the lower temporal crescent of the retina we found that the density of ipsilaterally projecting ganglion cells increased approximately 7.6 times as compared with that in the contralateral side, being accompanied by an increase of the soma areas of the ipsilateral side (Figs. 3 and 4). An increase of cell density and soma areas in ipsilaterally projecting ganglion cells was also observed in normal rats (7). In parallel with this phenomenon, however, a contralateral decrease of labeled cells in the ventral retina of the albino rat, and more marked in the pigmented rat, proved significant (Table I), because WGA-HRP injection in both the superior colliculus and the lateral geniculate body in the control albino rats produced no decrease in the contralateral, ventral retina. Concerning this point, we have no data to warrant further discussion. Labeled cells in the retina ipsilateral t o the NOT deserve some mention. Temporoventral retina corresponds to the upper visual field in front of the nose. The ganglion cells giving rise to the uncrossed optic fibers occupy about 40" of the temporal retina, corresponding to the binocular overlap in the visual field (23,24). In fact, rats with chiasmal splitting anteroposteriorly, leaving only uncrossed fibers intact, were able to relearn intensity discrimination and orientation discrimination, although optokinetic following was abolished (24). It has been speculated that OKN is a reflex to uniform movement over a large part of the visual field that must include the periphery. This speculation, however, seems unlikely, because in the ipsilatera1 temporal retina of the cat, only 13 labeled cells could produce OKN under monocular conditions (17). One of the possible causes is that units in the NOT in the rats have a very low resting rate, so that optokinetic modulation of discharge becomes effectively unidirectional (1) and/or cortical links are essential in producing OKN to nasotemporal stimulation, because removal of bilateral visual cortices and one optic tract or early monocular deprivation in the cat results in unidirectional OKN to monocular stimuli similar to normal rats (6).
ACKNOWLEDGEMENTS We would like to thank Mr J. P. Barron for his help in revising the manuscript and Miss K. Hasegawa for her secretarial help. This work was supported by the science research promotion fund of the Japan Private School Promotion Foundation.
REFERENCES 1. Cazin L, precht W,Lannou J. Pathways mediating optokinetic responses of vestibular nucleus neurons
in the rat. F'tliigen Arch 1980; 384: 19-29. 2. Colkwijn, H. Oculomotor areas in the rabbit's midbrain and pretectum. J Neurobiol 1975; 6 3-22.
428 I. Kato et al.
Acta Otolaryngol (Stockh) 112
Acta Oto-Laryngologica 1992.112:421-428.
3. Precht W, Strata P. On the pathway mediating optokinetic responses in vestibular nuclear neurons. Neurosci 1980; 5: 777-87.
4. Kato I, Harada K, Hasegawa T, Ikarashi T. Role of the nucleus of the optic tract of monkeys in op&okineticnystagmus and optokinetic after-nystagmus. Brain Res 1988; 4 7 4 16-26. 5. Mustari MJ,Fuchs AF. Discharge patterns of neurons in the pretectal nucleus of the optic tract (NOT) in the behaving primate. J Neurophysiol 1990; 6 4 77-90. 6. Precht W.Visual-vestibular interaction in vestibular neurons: Functional pathway organization. Ann NY Acad Sci 1981; 374 230-48. 7. Hsiao C-F, Fukuda Y. Plastic changes in the distribution and soma size of retinal ganglion cells after neonatal monocular enucleation in rats. Brain Res 1984; 301: 1-12. 8. Mesulam M-M.Tetramethyl benzidine for horseradish peroxidase neurochemistry: A non-carcinogenic blue reaction-product with superior sensitivity for visualizing neural afferents and efferents. J Histochem cytochem 1978; 2 6 106-17. 9. Hanker JS,Yates PE, Metz CB, Rustioni A. A new specific, sensitive and non-carcinogenic reagent for the demonstration of horse-radish peroxidax. J Histochem 1977; 9 789-92. 10. Fukuda Y. A three-group classification of rat retinal ganglion cells: Histological and physiological studies. Brain Res 1977; I 1 9 327-44. 11. Cleland BG, Dubin MW, Levick WR. Sustained and transient neurones in the cat’s retina and lateral geniculate nucleus. J Physiol (Lond) 1971; 217: 479-96. 12. Boycott BB, Wassle H. The morphological types of ganglion cells of the domestic cat’s retina. J Physiol (Lond) 1974; 240: 397419. 13. Hoffmann K-P. Conduction velocity in pathways from retina to superior colliculus in the cat: A correlation with receptive field properties. J Neurophysiol 1973; 36: 409-24. 14. Stone J, Fukuda Y.Properties of cat retinal ganglion cells: A comparison of W-cells with X- and Y-cells. J Neurophysiol 1974; 3 7 72248. 15. Lund RD, Land PW, Boles J. Normal and abnormal uncrossed retinotectal pathways in rats: An HRP study in adults. J Comp Neurol 1980; 189: 71 1-20. 16. Clark RJ, Ikeda H. Luminance and darkness detectors in the olivary and posterior pretectal nuclei and their relationship to the pupillary light reflex in the rat. Exp Brain Res 1985; 57: 224-32. 17. Ballas I, Hoffmann K-P, Wagner H-J. Retinal projection to the nucleus of the optic tract in the cat as revealed by retrograde transport of horseradish peroxidase. Neurosci Lett 1981; 26: 197-202. 18. Sawai H, Fukuda Y, Wakakuwa K. Axonal projections of X-cells to the superior colliculus and to the nucleus of the optic tract in cats. Brain Res 1985; 341: 1-6. 19. Precht W,Cazin L. Functional deficits in the optokinetic system of albino rats. Exp Brain Res 1979; 37: 183-6. 20. Sugimoto T, Fukuda Y, Wakakuwa K.Quantitative analysis of a cross-sectional area of the optic nerve: A comparison between albino and pigmented rats. Exp Brain Res 1984; 5 4 266-74. 21. Urushibata T, Kato I, Okada T, Takeyama I. Distribution of retinal ganglion cells projecting into the nucleus ofthe optic tract in rat. Adv Otoryinolaryngol 1988; 41: 95-7. 22. Cowey A, Perry VH. The projection of the temporal retina in rats, studied by retrograde transport of horseradish peroxidase. Exp Brain Res 1979; 35: 457-64. 23. Adams AD, Forrester JM.The projection of the rat’s visual field on the cerebral cortex. Q J Exp Physiol 1968; 53: 327-36. 24. Cowey A, Franzini C. The retinal origin of uncrossed optic nerve fibers in rats and their role in visual discrimination. Exp Brain Res 1979; 35: 443-55.
Manuscript received April 5, 1991; accepted July 23, 1991 Address for correspondence: I. Kato, Department of Otolaryngology, St. Marianna University School of Medicine. Sugao 2-1 6-1, Miyamae-ku, Kawasaki, 216 Japan
ISSN: 0001-6489 (Print) 1651-2251 (Online) Journal homepage: http://www.tandfonline.com/loi/ioto20
Retinal Ganglion Cells Related to Optokinetic Nystagmus in the Rat Isao Kato, Tomoyuki Okada, Shoji Watanabe, Shigeki Sato, Tamotsu Urushibata And & Isamu Takeyama To cite this article: Isao Kato, Tomoyuki Okada, Shoji Watanabe, Shigeki Sato, Tamotsu Urushibata And & Isamu Takeyama (1992) Retinal Ganglion Cells Related to Optokinetic Nystagmus in the Rat, Acta Oto-Laryngologica, 112:3, 421-428 To link to this article: http://dx.doi.org/10.3109/00016489209137422
Published online: 08 Jul 2009.
Submit your article to this journal
Article views: 6
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ioto20 Download by: [137.189.171.235]
Date: 16 March 2016, At: 06:53
Acta Otolaryngol (Stockh) 1992; 112: 421-428
Retinal Ganglion Cells Related to Optokinetic Nystagmus in the Rat ISAO KATO, TOMOYUKI OKADA, SHOJI WATANABE, SHIGEKI SATO, TAMOTSU URUSHIBATA and ISAMU TAKEYAMA From the Deparlment of Otarhinolatyngolagv, St. Marianna University School of Medicine. Kawasaki, Japan
Acta Oto-Laryngologica 1992.112:421-428.
&to I, Okada T, Watanabe S, Sat0 S, Urushibata T, Takeyama I. Retinal ganglion cells related to optokinetic nystagmus in the rat. Acta Otolaryngol (Stockh) 1992; 112: 421-428. The nucleus of the optic tract (NOT) in the pretectum is the visuo-motor relay between the retina and preoculomotor structure in the pathway conveying signals responsible for optokinetic nystagmus (OKN) in both afoveate and foveate animals. Unilateral lesions of the NOT abolish OKN toward the side ipsilateral to the lesion. However, what parts of the retina and what kinds of retinal ganglion cells project their fibers into the NOT are still unknown. To examine this, horseradish peroxidase conjugated with wheatgerm agglutinin was injected into the NOT of the rat. Labeled retinal ganglion cells were diffusely documented in the contralateral side, whereas those situated only in the lower temporal crescent, were found in the ipsilateral side. Ganglion cells dominated in the range of small cells, i.e., W cells. Therefore, ganglion cells responsible for OKN are believed to be related to W cells. However, medium-sized and large cells were definitely labeled in the rat, suggesting possible participation of these cells in OKN. Key words: pigmented rat, retinal ganglion cells. nucleus of the optic tract, OKN.
INTRODUCTION Recent neurophysiological and ablation studies have disclosed that the nucleus of the optic tract (NOT) is the visuo-motor relay between the retina and preoculomotor structures in the pathway mediating optokinetic reflex in rats (l), rabbits (2), cats (3) and monkeys (4, 5). Afoveate animals such as rats and rabbits have predominantly crossed retinofugal fiber projection to the NOT. Unilateral lesions of the NOT abolish optokinetic nystagmus (OKN) in any stimulus mode during monocular stimulation of the contralateral eye (1,2). In cats, which have both crossed and uncrossed retinofugal fibers and produce OKN of almost equal magnitude in both directions, unilateral lesions of the NOT likewise abolish OKN in response to surround motion toward the lesioned side (6). In monkeys, which have the pursuit system, the same was true with a rapid rise kept intact (4). However, it is still open to discussion from what parts of the retina it stems and what kinds of retinal ganglion cells project their fibers into the NOT. To achieve this goal, we used pigmented rats, because these afoveate animals have a fundamentally old OKN system and none of the OKN are changed in ablation of either cerebrum or cerebellum or both, and only the crossed temporonasal system generates significant OKN responses (6).
MATERIAL AND METHODS Twenty-one female Long-Evans rats weighing about 250-270 g were used for the present experiment, and 12 female Wistar rats as controls.
Retinal ganglion cells In the first part of the present experiment, pigmented rats were anesthetized with intrapentoneal injection of pentobarbital sodium (20 m&) and mounted on a stereotaxic apparatus. For surgical procedures, 1% xylocaine was used topically. A lateral incision sufficiently large
Acta Oto-Laryngologica 1992.112:421-428.
422 I. Karo et al.
ActD Otolaryogol (Stockb) I I2
Fig. 1. A photomicrograph of the ganglion cell layers. (A) Large, medium-sized and small cells are labeled 1 , 2, and 3, respectively;(B)Sagittal incision of the retina. Labeled cells are located in the ganglion cell layer. Each horizontal bar indicates 100 pm.
to expose the I& optic nerve just posterior to an eye ball was made. For massive injections of horseradish peroxidase, a microsyringe of 10 was stereotaxically located so as to have the tip penetrate the optic nerve, and a 10% solution of horseradish peroxidase conjugated with wheatgerm agglutinin (WGA-HRP) was injected into the nerve through a microsyringe under an operating microscope. After a survival time of about 48 h the animals were reanesthetized with an overdose of pentobarbital sodium and sacrificed by vascular perfusion of 0.9% NaCl followed by 3 1 of 1 % paraformaldehyde and 2 % glutaraldehyde in 0. I M phosphate buffer pH 7.4,650 mOsm. After completing perfusion, the full extent of the retina, up to the ora serata, was dissected free. One large incision was made at the top of the eye ball to facilitate correct alignment of the preparation on the slide, and to avoid folding, several radial incisions were made in the periphery, if necessary. Thus, whole mount preparations of the retina were prepared. Shrinkage in the retinal area after the dehydration processes was estimated at 5-1096 (7). Therefore, no correction was made for cell soma. Soma sizes of ganglion cells were measured in photomicrographs magnified 250 times (Fig. 1 A). Roughly estimated, large cells had bodies with a diameter of about 30 pm; medium-sized cells, with a diameter of about 15 wm; small cells had a diameter of about 5 pm, and, furthermore, these cells were precisely confirmed to be located in the retina (Fig. 1 B).
NOT identification Animals were anesthetized with intraperitoneal injection of urethane (1.2 &) and achloralose (30m&). Animals were mounted on a stereotaxic apparatus. The dorsal aspect of the pretectum and the superior colliculus was exposed after suctioning the cerebral hemisphere under an operating microscope. A glass capillary electrode filled with 3 M KCI was inserted
Rat retinal ganglion cells related to OKN
Acta Oto-Laryngologica 1992.112:421-428.
Acta Otolaryngol (Stockh) 112
caudal
-
Fig. 2. HRP injection site of pigmented rat 19 (PR 19). (A) Densely stained area is restricted to the NOT and diffusion spreads to adjacent areas. Arrow indicates the injection site; (B) Schematic representation of the injection site. Horizontal sections are rostrocaudal from above to below at 500 pm. Black areas indicate densely stained sites which include NOT; hatched areas, diffusion of HRP. Diffusion covered OPN and PPN. Each horizontal bar indicates I mm.
into the anterolateral border of the superior colliculus. The NOT was identified by recording a unit responding to lights which were held by a hand from the outside of a shielded room and moved slowly in the horizontal direction. Units in the NOT definitely responded in a direction-selective way, and in the right site, 0.01 pl of 10%WGA-HRP solution was injected mechanically into the NOT through a micropipette connected tightly with a microsyringe. After a survival time of about 48 h, the animals were reanesthetized deeply and perfused with 0.9% saline followed by fixation. Injection sites of the brain and the retina were processed for histochemical demonstration of WGA-HRP by the tetramethyl benzidine (TMB)protocol (8) and the Hanker-Yeates method (9), respectively. Subsequently, the injection site was counterstained with 1.0% neutral red.
Soma area measurements Photomicrographs of retinal ganglion cells suggest that a three-group classification could be possible in the rat as well (Fig. I A). However, the soma size of these cells varied depending upon whether the dendrites of the cells were stained or not. Therefore, cell somata of labeled cells were photographed on a negative film on which alphabetical letters were printed and they were used as a landmark to identify the sites of labeled ganglion cells. The ganglion cells were enlarged precisely 250 times their actual size. Photographed cell soma areas were measured by a tracing device and compared with each other in great detail. Ganglion cell density In order to clarify whether the distribution of retinal ganglion cells varies in different regions within the retina, ganglion cell density was established in the central, temporal, nasal, dorsal and ventral regions of the retina. Ganglion cells ipsilateral to the injection site were observed
423
10
(rim )
-L
25-
50
50-
75
75
- 100
100125
t25
- 150
150115
175 200
-
Acta Oto-Laryngologica 1992.112:421-428.
i
200 - 225 225 250 250 215 275 300 300 325 325 - 350 350 375 375 - 400
-
20
30
40
50
60
70
80
90
Cell density
' 49
Fig. 3. Cell density per unit area ( I .O mm2) in each representative area and cell distribution in the ventral area contralateral to the injection site. Labeled cells were almost evenly distributed except in the ventral area. Cell decrease was noticed in the ventral area. Soma areas of labeled cells predominated in the range of small cells of between 25 and 50 pm2 in soma areas. The same was true of other areas.
only in the temporoventral region. These regions were arbitrarily selected in the areas where labeled cells were evenly distributed in 1.O mm'. Ganglion cell density was counted per unit area (1.0 mm2). Cell soma larger than 6 bm in diameter and containing irregularly stained Nissl granules were judged as ganglion cells (10). RESULTS
The conclusions of this study were drawn on the basis of 6 (out of 21) rats, and 15 rats were excluded due to an inadequate injection site of WGA-HRP or large diffusion. In rat 19 (PR 19) processed with the TMB method, a densely stained area was surrounded by a faint one (Fig. 2). The densely stained area was determined as the injection site, whereas the faint one was due to diffusion of WGA-HRP, in agreement with the result reported by Mesulam (8). Enlarged drawings of histological sections were made with the aid of a camera lucida. In rats 13, 18, 19, and 21, the injection site of HRP was restricted to the NOT and diffusion of HRP extended slightly to the posterior pretectal nucleus (PPN). In the remaining two rats (7, I2), the injection site was localized in the NOT and diffusion of HRP encrouched on the olivary pretectal nucleus (OPN), and the PPN as well as the medial border of the medial geniculate body.
Ganglion cell density Ganglion cell densities were counted in the whole mount preparation of the right retina contralateral to the injection site (Fig. 3). Numerals in the figure are all labeled cells sampled per unit areas (1 .O mm'). In rat 19, ganglion cells were most densely labeled in the dorsal area (D) (Fig. 3). In the remaining 5 rats, however, there was no clear tendency for the density of ganglion cells to increase in any particular area except in the ventral area (V). In the ventral area, ganglion cells were least labeled. In the left retina ipsilateral to the injection site, ganglion cells were labeled only in the temporoventral area. Labeled cells per unit area were 379, which
Rat retinal ganglion cells related to OKN 425
Acta Otolaryngol (Stockh) 112
Soma area
(am‘)
Acta Oto-Laryngologica 1992.112:421-428.
0-
25
25-
50
50-
75
10
PR 19 20
30
40
%
50
60
70
80
90
Fig. 4. Cell density per unit area and cell distribution in the ventral area ipsilateral to the injection si?e. Labeled cells were observed only in the temporoventral area. Cell density in this area increased. In soma areas the larger cells of more than 100 pnz increased.
was 7.6 times larger than in the corresponding contralateral retina (Fig. 4). The same trends in ganglion cell density were apparent in the other retinas studied.
Ganglion cell soma area In order to clarify whether the distribution of retinal ganglion cells varies depending upon the different regions of the retina, frequency distribution was established depending upon some areas of retinal ganglion cells in each unit area. In the ventral area in Fig. 3, soma areas of labeled ganglion cells contralateral to the injection site predominated in the range between 25 and 50 pm2. Next to this range, the range between 0 and 25 pm2 was followed. Labeled cells greater than 100 pm2 in soma areas were in the minority. In the remaining prescribed areas, the distribution of ganglion cell soma areas showed a similar tendency, although cell density varied according to area (Fig. 3). In the retina ipsilateral to the injection site, however, soma areas predominated in the range between 0 and 25 pm2 which decreased slowly, and soma areas greater than 100 pm2appeared to increase in frequency, which was in marked contrast to the contralateral retina (Fig. 4). Fig. 5 shows a photomicrograph of labeled cells of the ventral areas both ipsilateral and contralateral to the injection sites. The increase of cell density and soma areas markedly predominated on the ipsilateral side. Table I tabulates labeled large cells greater than 200 Fm2 in soma areas either contralateral or ipsilateral to the injection site, and albino rats served as controls. In ventral retinas contralateral to the injection site, none of the large cells were labeled either in pigmented or albino rats, whereas in those ipsilateral to the injection site of the pigmented rats, about 7% of large cells were labeled in the prescribed area. In the albino rats, the relative frequencies can be recognized in Table I. In other parts of the retina, large cells were distributed sporadically all over the retina without any special predominance. DISCUSSION The concept of a three-group classification of the retinal ganglion cells in the cat has recently been established and is based on differences in receptive field properties and morphology of
426 I . Kato et a!.
Acta Oto-Laryngologica 1992.112:421-428.
Ips vent
Acta Otolaryngol (Stockh) 112
PR 19
Ctr vent
Fig. 5. A photomicrograph of labeled cells in ventral (venr)areas both ipsilateral (Ips)and contralateral (Ctr)to the injection site. Large,medium-sized and small cells are indicated as 1,2, and 3, respectively. An increase of cell density and soma areas are noticed in the ipsilateral side (Ips venr). Horizontal bar indicates 100pm.
the ganglion cells, measurement of axonal conduction velocities, and central distribution of axons (1 1-14). In the present experiment, the three-group classification of the retinal ganglion cells could be applied to the rat in addition to being based on soma sizes and areas (Figs. 1 A, 5). which is compatible with results of other experiments (10, 15). Dense WGA-HRP-stained areas were restricted to the NOT and diffusion of HRP extended slightly to the OPN and/or the PPN in representative cases. A recent physiological study has disclosed that luminance and darkness detector cells are to be found in the OPN and the PPN in the rat, both of which receive direct visual inputs (16). In the present experiment, we could not specify the retinal ganglion cells solely related to the NOT, even though contamination was very slight. Retrogradely-labeled retinal ganglion cells recognized by injecting WGA-HRP in the NOT
Table I. Cell distribution of large cells, greater than 200 pm2 in soma areas Large cells distributed sporadically all over the retina except in the ventral area contralateral (Ctr) to the injection site. No large cells were labeled in either pigmented (PR) of albino rats (AR), and those ipsilateral (Ips) to the injection site increased by about 7% in PR and by 2% in AR
L-NOT Rat no. Inj. site
AR36
AR38
PRl2
PR19
Ctr
Temporal
0 0 1.50
0.1 I 0 0.11
2.35 0
0.86 0.46
0.30
Ips
Central DoMl NaJal ventral Ventral
0 0 2.33
0 1.10
0.90 0 0 7.06
0.64 4.65 0 7.65
Acta Oto-Laryngologica 1992.112:421-428.
Acta Otolaryngol (Stockh) 112
Rat retinal ganglion cells related to OKN 427
dominated in the range of small cells, i.e., W cells. Therefore, retinal ganglion cells responsible for OKN are assumed to be related to W cells. This is in agreement with results in the cat (17). However, in contrast to the cat, medium-sized, and large cells were definitely labeled in the rat, suggesting possible participation of these cells in OKN (1 8). We used albino rats as controls. OKN cannot be elicited in albino rats. Similarly, vestibular nuclear neurons of the horizontal canal system fail to respond to optokinetic stimulation in these animals (19). There have been comparative studies comparing several morphological characteristics between albino and pigmented rats (20). At present, it is not known why the OKN system in the albino rat is so poorly developed. Judging from cell density and distribution of retinal ganglion cells studied, we could not find any abnormalities characteristic of the albino rat (21). In rats injected with WGA-HRP in the pretectum, the ipsilateral distribution of ganglion cells with pretectoretinal projection fibers was confined to the lower temporal crescent. The situation was quite similar to results reported in normal hooded rats (15,22) and in albino rats (7). In the lower temporal crescent of the retina we found that the density of ipsilaterally projecting ganglion cells increased approximately 7.6 times as compared with that in the contralateral side, being accompanied by an increase of the soma areas of the ipsilateral side (Figs. 3 and 4). An increase of cell density and soma areas in ipsilaterally projecting ganglion cells was also observed in normal rats (7). In parallel with this phenomenon, however, a contralateral decrease of labeled cells in the ventral retina of the albino rat, and more marked in the pigmented rat, proved significant (Table I), because WGA-HRP injection in both the superior colliculus and the lateral geniculate body in the control albino rats produced no decrease in the contralateral, ventral retina. Concerning this point, we have no data to warrant further discussion. Labeled cells in the retina ipsilateral t o the NOT deserve some mention. Temporoventral retina corresponds to the upper visual field in front of the nose. The ganglion cells giving rise to the uncrossed optic fibers occupy about 40" of the temporal retina, corresponding to the binocular overlap in the visual field (23,24). In fact, rats with chiasmal splitting anteroposteriorly, leaving only uncrossed fibers intact, were able to relearn intensity discrimination and orientation discrimination, although optokinetic following was abolished (24). It has been speculated that OKN is a reflex to uniform movement over a large part of the visual field that must include the periphery. This speculation, however, seems unlikely, because in the ipsilatera1 temporal retina of the cat, only 13 labeled cells could produce OKN under monocular conditions (17). One of the possible causes is that units in the NOT in the rats have a very low resting rate, so that optokinetic modulation of discharge becomes effectively unidirectional (1) and/or cortical links are essential in producing OKN to nasotemporal stimulation, because removal of bilateral visual cortices and one optic tract or early monocular deprivation in the cat results in unidirectional OKN to monocular stimuli similar to normal rats (6).
ACKNOWLEDGEMENTS We would like to thank Mr J. P. Barron for his help in revising the manuscript and Miss K. Hasegawa for her secretarial help. This work was supported by the science research promotion fund of the Japan Private School Promotion Foundation.
REFERENCES 1. Cazin L, precht W,Lannou J. Pathways mediating optokinetic responses of vestibular nucleus neurons
in the rat. F'tliigen Arch 1980; 384: 19-29. 2. Colkwijn, H. Oculomotor areas in the rabbit's midbrain and pretectum. J Neurobiol 1975; 6 3-22.
428 I. Kato et al.
Acta Otolaryngol (Stockh) 112
Acta Oto-Laryngologica 1992.112:421-428.
3. Precht W, Strata P. On the pathway mediating optokinetic responses in vestibular nuclear neurons. Neurosci 1980; 5: 777-87.
4. Kato I, Harada K, Hasegawa T, Ikarashi T. Role of the nucleus of the optic tract of monkeys in op&okineticnystagmus and optokinetic after-nystagmus. Brain Res 1988; 4 7 4 16-26. 5. Mustari MJ,Fuchs AF. Discharge patterns of neurons in the pretectal nucleus of the optic tract (NOT) in the behaving primate. J Neurophysiol 1990; 6 4 77-90. 6. Precht W.Visual-vestibular interaction in vestibular neurons: Functional pathway organization. Ann NY Acad Sci 1981; 374 230-48. 7. Hsiao C-F, Fukuda Y. Plastic changes in the distribution and soma size of retinal ganglion cells after neonatal monocular enucleation in rats. Brain Res 1984; 301: 1-12. 8. Mesulam M-M.Tetramethyl benzidine for horseradish peroxidase neurochemistry: A non-carcinogenic blue reaction-product with superior sensitivity for visualizing neural afferents and efferents. J Histochem cytochem 1978; 2 6 106-17. 9. Hanker JS,Yates PE, Metz CB, Rustioni A. A new specific, sensitive and non-carcinogenic reagent for the demonstration of horse-radish peroxidax. J Histochem 1977; 9 789-92. 10. Fukuda Y. A three-group classification of rat retinal ganglion cells: Histological and physiological studies. Brain Res 1977; I 1 9 327-44. 11. Cleland BG, Dubin MW, Levick WR. Sustained and transient neurones in the cat’s retina and lateral geniculate nucleus. J Physiol (Lond) 1971; 217: 479-96. 12. Boycott BB, Wassle H. The morphological types of ganglion cells of the domestic cat’s retina. J Physiol (Lond) 1974; 240: 397419. 13. Hoffmann K-P. Conduction velocity in pathways from retina to superior colliculus in the cat: A correlation with receptive field properties. J Neurophysiol 1973; 36: 409-24. 14. Stone J, Fukuda Y.Properties of cat retinal ganglion cells: A comparison of W-cells with X- and Y-cells. J Neurophysiol 1974; 3 7 72248. 15. Lund RD, Land PW, Boles J. Normal and abnormal uncrossed retinotectal pathways in rats: An HRP study in adults. J Comp Neurol 1980; 189: 71 1-20. 16. Clark RJ, Ikeda H. Luminance and darkness detectors in the olivary and posterior pretectal nuclei and their relationship to the pupillary light reflex in the rat. Exp Brain Res 1985; 57: 224-32. 17. Ballas I, Hoffmann K-P, Wagner H-J. Retinal projection to the nucleus of the optic tract in the cat as revealed by retrograde transport of horseradish peroxidase. Neurosci Lett 1981; 26: 197-202. 18. Sawai H, Fukuda Y, Wakakuwa K. Axonal projections of X-cells to the superior colliculus and to the nucleus of the optic tract in cats. Brain Res 1985; 341: 1-6. 19. Precht W,Cazin L. Functional deficits in the optokinetic system of albino rats. Exp Brain Res 1979; 37: 183-6. 20. Sugimoto T, Fukuda Y, Wakakuwa K.Quantitative analysis of a cross-sectional area of the optic nerve: A comparison between albino and pigmented rats. Exp Brain Res 1984; 5 4 266-74. 21. Urushibata T, Kato I, Okada T, Takeyama I. Distribution of retinal ganglion cells projecting into the nucleus ofthe optic tract in rat. Adv Otoryinolaryngol 1988; 41: 95-7. 22. Cowey A, Perry VH. The projection of the temporal retina in rats, studied by retrograde transport of horseradish peroxidase. Exp Brain Res 1979; 35: 457-64. 23. Adams AD, Forrester JM.The projection of the rat’s visual field on the cerebral cortex. Q J Exp Physiol 1968; 53: 327-36. 24. Cowey A, Franzini C. The retinal origin of uncrossed optic nerve fibers in rats and their role in visual discrimination. Exp Brain Res 1979; 35: 443-55.
Manuscript received April 5, 1991; accepted July 23, 1991 Address for correspondence: I. Kato, Department of Otolaryngology, St. Marianna University School of Medicine. Sugao 2-1 6-1, Miyamae-ku, Kawasaki, 216 Japan
