H. Sasaki a, P. Coffey b, M.P. Villegas-Perez c, M. Vidal-Sanz c, M.J. Young d , R.D. Lund d and Y. Fukuda a

a Department of Physiology, Osaka University School of Medicine, 2-2, Yamadaoka, Suita, Osaka 565, Japan
b Department of Psychology, University of Sheffield, Western Bank, Sheffield S10 2UR, ENGLAND
c Laboratorio de Optalmologia Experimental Facultad de Medicina, Universidad de Murcia, 30100 Espinardo, Murcia, SPAIN
d Institute of Ophthalmology, University of London, Bath Street, London EC1V 9EL, ENGLAND

Address correspondence to :
Hitoshi Sasaki Department of Physiology, Osaka University Medical School, 2-2, Yamadaoka, Suita, Osaka 565 Japan

Tel: 011-81-6-879-3611
FAX : 011-81-6-879-3619
E-mail: h.sasaki@physiol.com

Abstract
Peripheral nerve (PN) was grafted to sectioned optic nerve and was bridged to the superior colliculus in adult rats. To test functional recovery of restored retinocollicular pathway, we examined cortical EEG and behavioral arousal responses to light stimuli. In 8 of10 recording trials of the 6 PN grafted rats and in all of 8 trials in 5 normal rats, EEGs showed desynchronization to light stimuli. On the other hand, after bilateral sections of the optic nerve EEG desynchronization to light disappeared while it was induced by a white noise. Mean threshold duration of light for EEG desynchronization was significantly longer in the PN grafted rats (440 .0 ms) than that of normal rats (172.5 ms). In 3 of 6 trials in PN grafted rats, and in 4 of 8 trials in normal rats, EEG desynchronization elicited by light stimulus was accompanied by behavioral arousal responses. While no behavioral arousal could be induced by light in the three blind rats. These results strongly suggest that visual information processed through the restored retinocollicular pathway was further transmitted to the cerebral cortices and ultimately resulted in behavioral arousal of the animals.

Keywords:

Axonal Regeneration; Peripheral Nerve Graft; EEG Arousal; Orienting Response; Superior Colliculus; Visual System

INTRODUCTION

Autologus peripheral nerve (PN) grafts lead regeneration of retinal ganglion cell (RGC) axons, and when their axons are guided to the superior colliculus (SC) their terminals can make functional synapses with target neurons in the SC in adult rodents [13,16,10]. From previous studies it has been shown that a light stimulus could evoke both pre- and postsynaptic responses in the SC of the PN grafted animals [2,3,9]. Furthermore, light-dark discrimination learning and pupillary light reflex have been shown to recover, at least to some extent, in the PN grafted animals [5,8,15]. In the present study, we attempted to show further arousal responses induced by light stimuli in the PN grafted rats by recording cortical EEGs and monitoring body movements. With these simple monitorings we expected to test whether visual perception has recovered or not in the PN grafted rats under more natural conditions than conditioned discrimination learning we used previously.
According to Dean et al. (1984)[1], the cortical EEGs could not be desynchronized by photic stimuli in rats after lesion of the SC but not after the visual cortex lesion. Thus, the light-induced cortical EEG desynchronization is a proper measure for functional evaluation of retinal afferents to the SC which is essential for cortical arousal by visual inputs.

Animals

Thirteen hooded rats (Long-Evans) were used in the present experiment. A segment of PN was grafted to the sectioned optic nerve in 7 rats at 7-8 weeks of age, and other 6 rats served as controls. Procedures for the transplantation were similar as described before [16]. In brief, under anesthesia, left optic nerve was cut at its exit. A stump of the common peroneal nerve was excised from the the same animal. The cutting end of the optic nerve was sutured with the PN graft by a 10-0 string. Four to 8 weeks after the first operation, animals were reanesthetized and distal end of the graft was inserted into the SC and the pretectum, and the right optic nerve was cut to facilitate synaptic formation for the regenerated RGC fibers in the SC. Five to 7 months after the second operation, under anesthesia with avertin (tribromoethanol 230 mg/Kg, i.p.), steel screw electrodes (1.5 mm diameter, and 8 mm long) were implanted into the skull over the sensorimotor cortex (2 mm anterior and 4 mm lateral to bregma) and the frontal sinus (6 mm anterior to the bregma). In some animals, another screw electrode was also implanted into the visual cortex (2 mm anterior and 2 mm lateral to lambda). These electrodes were connected to a miniature socket and were glued on the skull using dental cements. For the control rats, the same electrodes were implanted at 8-10 weeks of age. In three rats, after two sessions of EEG recording, the optic nerve was cut bilaterally under anesthesia with avertin. After one week of recovery, the same test was repeated.

Apparatus

A glass chamber (50 cm x 25 cm x 25 cm) in an isolated room was used for EEG recordings. Top of the chamber was covered by a translucent acrylic glass to prevent heat transmission from the light source. Room temperature was maintained at 25-30｡C using a fun heater which also served for making a continuous background masking noise (58 dB in the chamber). Light stimulus was presented by a halogen tube (300 W) located at 60 cm above the floor of chamber. Tone burst was generated by a white noise generator (Type 419C, Dawe Instruments, set to 20K Hz bandwidth). Through an audio-amplifier (SU 3400, Technics) and a frequency equalizer (Model 31-9082, Genexxa), the tone was presented with a loud speaker (8.6 cm, 35 ohm, 2W) which was mounted at 50 cm above the floor of chamber. The sound pressure level was set at 20dB above the background. All the stimulus presentation was controlled using a logic control unit (Campden Instrument Ltd.). EEG was recorded with an AC amplifier (NL105, Neurolog) with a high-cut filter set at 0.1KHz (NL115, Neurolog). The signals were fed into an A/D converter (MacLab, AD Instruments) together with signals for triggering the stimulus. Usually, cortical EEGs were recorded on a computer (Macintosh II, Apple) for 10 sec before and after the stimulus presentation, with a sampling rate of 200 ms/div. The EEGs were also monitored continuously on two oscilloscopes (D63, Telequiptment and V-211, Hitachi) with different sweep speeds.

Procedures

After 3-7 days of recovery from the operation, the animals were put in the test chamber. After habituation to the chamber for 30 min EEG recordings started. When high amplitude low frequency waves (1-3 Hz, 100-200 uV) continued at least for 30 sec the light stimulus was presented to test whether the EEGs showed desynchronized or not. A criterion for EEG desynchronization was fulfilled when the slow wave EEGs changed into low amplitude higher frequency waves (8-30 Hz, 20-50 uV) within 5 sec after the stimulus offset and the low amplitude waves lasted at least for 3 sec. To measure threshold duration of light stimulus to induce EEG desynchronization, we started with 20 msec light and gradually increased stimulus duration by 40 msec step up to 740 msec until desynchronization was elicited. The interval between trials was set at least for 2 min. More than 5 min after cessation of the trials with light stimuli we then performed the same test using auditory stimuli.

EEG Recordings

One of 6 normal rats and one of the 6 PN grafted rats failed to show desynchronized EEGs both to visual and auditory stimuli even with maximum durations available. These two rats were excluded from the following quantitative analysis. Test of the EEG arousal response to visual stimuli was repeated twice with a delay at least 7 days in most of the normal, blind and PN grafted rats. One PN grafted rat failed to show any arousal response to both auditory and visual stimuli in the second test though it revealed EEG desynchronization in the first test, so this rat was also excluded from the data summarized in Table 1.

Behavioral Recordings

Behavior of the experimental animals was continuously monitored with a video camera. The recorded video-tape was converted from PAL to U-matic and images were enhanced by increasing contrasts. The movements of rats induced by the light were analyzed by replaying each video field step by step with a 1/60 sec interval. Some flames were printed on a paper using a video printer (UP-3000, Sony). Behavioral arousal was defined as an occurrence of movement of some body parts within 10 sec after the stimulus offset. If any part of the animal, for example a tip of the nose or the tail, showed movements to the stimulus, the horizontal and vertical distances from the lower left corner of the video field to a marked point of the animal were measured on a video monitor (PVM-9220, Sony) and they were plotted as a function of time (see Fig. 2).

EEG Results

A typical example of the EEG desynchronization induced by light in a normal rat is shown in Fig.1A. Immediately after light presentation of 200 msec duration, slow wave EEGs changed into low amplitude higher frequency waves (around 10 Hz and more) in the sensorimotor cortex. The EEG desynchronization was observed simultaneously in the visual and sensorimotor cortices with similar time course, although it was more clearly observed in the sensorimotor cortex. An auditory stimulus consistently elicited apparent EEG desynchronization. These findings in normal rats were quite similar to those reported by Dean et al. (1984) [1]. As shown in Fig 1B, a similar light stimulus with duration of 660 msec also elicited the EEG desynchronization in one PN grafted rat. Number of animals and number of trials of light-induced EEG desynchronization are listed in Table 1. The desynchronization was elicited in all 8 recording trials of 5 normal rats. The desynchronization was repeatedly observed in the same rats after 1 week, thus confirming reliability of this test. After bilateral optic nerve sections in three of normal control rats, no desynchronization could be observed, while the tone stimulus still elicited desynchronization. On the other hand, in 8 of 10 recording trials (80.0%) in 6 PN grafted rats, light stimulus elicited EEG desynchronization as in normal rats.

Table 1 also summarizes mean threshold durations of light for EEG desynchronization. The mean threshold with S.D. was 172.5 ｱ126.46 msec in the normal control rats while it was significantly longer in the PN grafted rats (440 ｱ 270.45 msec, Mann-Whitney U test, p=.0106). In three blind rats the threshold could not be identified because even with longest duration of light we used, these animals did not show any sign of desynchronization. The latency of EEG desynchronization was similar in normal and PN grafted rats (1.08 ｱ0.98 sec and 1.34 ｱ0.51 sec, respectively).

Then we studied behavioral changes induced by the visual stimulus. In several cases EEG desynchronization was followed by movements such as head rising, head turning, sway of body, flip of a tail or standing up. An example of the behavioral arousal response in a PN grafted rat is shown in Fig. 2. In this case, a tail flip was observed after the light presentation. Similar behavioral arousal responses were observed in other PN grafted and normal rats as summarized in Table 1. With all 4 normal rats tested, in a half of 8 trials (50%), behavioral arousal was elicited by the light stimulation. Mean threshold of light duration with S.D. was 230.0 ｱ 157.90 msec. Similarly, in a half of 6 trials (50%) or in a half of 4 PN grafted rats, the light stimulation caused behavioral responses. Mean ｱ S.D. threshold of light duration was 393.3 ｱ 333.07 msec. Following bilateral sections of the optic nerve in three normal control rats, no behavioral arousal occurred to the light stimuli.

Behavioral Results

In the present study, we have shown that after light stimulation PN grafted rats revealed not only EEG desynchronization but also behavioral arousal responses. In normal rats similar visual stimuli of shorter durations always induced arousal responses in both EEG recording and behavioral monitoring. After bilateral sections of the optic nerve, neither EEG desynchronization nor behavioral arousal was elicited by the same light stimulation. Thus the present results in PN grafted rats clearly show that these animals can perceive visual signals coming through the PN graft.

DISCUSSIONS

One might argue, however, that the EEG desynchronization or behavioral arousal was elicited by non-specific cues, such as auditory noise, or heat emission accompanied with the light presentation. However, this possibility can be excluded because all of three blind rats showed neither EEG changes nor behavioral arousal to the light, while they all had shown definite EEG desynchronization to the same light before the optic nerve sections. There might be another argument that the EEG or behavioral change has occurred spontaneously. A probability of such contamination of spontaneous arousal would be extremely low because of the following reasons. Firstly, EEG desynchronization elicited by the stimulus always occurred abruptly. Such a sudden change of EEGs was different from a pattern of spontaneous EEG arousal, which was more gradual. And secondly, preferential occurrence of EEG arousal only to the tone in all blind rats can not be explained only by the spontaneous changes, since the probability of contamination can be assumed to be similar for both test sessions with visual and auditory stimuli. Some of normal rats in the present experiment, showed the head turning to the light stimulus. Sensory stimulus-induced EEG desynchronization have been regarded as a component of orienting responses to the stimulus [14]. Indeed, animals can respond with EEG desynchronization selectively to a novel stimulus presented during slow wave sleep [12]. Thus the EEG and behavioral arousal response seemed to be one of orienting responses. On the other hand, it is generally thought that the orienting response is closely related to function of the SC [11]. For example, lesions of the SC in rats produce visual neglect [4], and disappearance of the cortical EEG desynchronization [1]. In the present study, the EEG and behavioral arousal responses could be observed in more than 50% of the recording in the PN grafted rats. These data suggest that in the PN grafted rats, orienting responses, as a main function of the SC, has recovered.

In previous studies the pupillary light reflex could be recorded in the rats with restored retinopretectal pathway by PN graft [15]. Furthermore, we have shown functional recovery of the restored retinocollicular pathway using a light-dark discrimination task in hamsters [8]. Similar light-dark discrimination have been reported in the PN grafted rats [5]. All of these studies strongly suggest that the restored visual pathway by the PN graft is functional at least to detect visual signals. The present data confirm these previous findings and further show that the visual information processed in the SC can be transmitted to the cerebral cortices and utilized for visual orienting behavior. Although it is most conceivable that the EEG desynchronization was induced by afferent inputs from the SC to the brain stem reticular activating system, which in turn to the cerebral cortices via non-specific thalamic nuclei [6, 7], the precise pathway responsible for the EEG desynchronization to visual stimulus is still unknown [1]. That is certainly a subject of future studies using both anatomical and electrophysiological techniques in PN grafted rats.

Acknowledgements:
We thank to Dr. P. Dean and Dr. P. Redgrave for their useful suggestions for doing present experiment. We also thank to Dr. G.W.M. Westby for EEG recordings, thank to Mr. Len for his helpful technical assistance and to Ms. M. Simkins for electrodes implantation.

REFERENCE
[1] Dean, P., Redgrave, P. and Molton, L., Visual desynchronization of cortical EEG impaired by lesion of superior colliculus in rats. J. Neurophysiol., 52 (1984) 625-637.
[2] Fukuda, Y., Rasminsky,M., Keirstead, S.A., Carter, D. A., Aguayo,A.J. and Vidal-Sanz, M., Reinnervation of adult hamster superior colliculus by regenerating retinal ganglion cell axons. Biomedical Res. (Suppl.), 2 (1989) 81-84.
[3] Keirstead, S.A., Rasminsky,M., Fukuda,Y., Carter, D.A., Aguayo, A.J. and Vidal-Sanz, M., Electrophysiologic responses in hamster superior colliculus evoked by regenerating retinal axons. Science, 246 (1989) 255-257.
[4] Kirvel, R. D. Sensorimotor responsiveness in rats with unilateral superior collicular and amygdaloid lesions. J. Comp. Physiol. Psychol., 89 (1975) 882-891.
[5] Kittlerova P., Regeneration of retinal ganglion cell axons through the sciatic nerve isograft into the superior colliculus - electrophysiological and behavioral study. MSc thesis, Charles University, 1992.
[6] Moruzzi, G. and Magoun, H.W., Brain stem reticular formation and activation of the EEG. Electroenceph. clin. Neurophysiol., 1 (1949) 455-473.
[7] Petrovicky, P., Projections from the tectum mesencephali to the brain stem structures in the rat. Folia Morphol. Praha, 24 (1976) 41- 48.
[8] Sasaki, H., Inoue, T., Fukuda, Y., Iso, H. and Hayashi, Y., Light-dark discrimination after sciatic nerve transplantation to the sectioned optic nerve in adult hamsters. Vision Res., 33 (1993) 877- 880.
[9] Sauve, Y., Sawai, H., Rasminsky, M., Functional synaptic connections made by regenerated retinal ganglion cell axons in the superior colliculus of adult hamsters. J. Neurosci., 15 (1995) 665-75.
[10] Sawai, H., Sugioka, M., Morigiwa, K., Sasaki, H., So, K-F. and Fukuda, Y., Functional and morphological restoration of intracranial brachial lesion of the retinocollicular pathway by peripheral nerve autografts in adult hamsters. Exp. Neurol.,137 (1996) 94-104.
[11] Schneider, G.E., Two visual systems. Science, 163 (1969) 895-902.
[12] Sharpless, S. and Jasper, H. Habituation of the arousal reaction. Brain, 79 (1956) 655-681.
[13] So, K. -F. and Aguayo, A.J., Lengthy regrowth of cut axons from ganglion cells after peripheral nerve transplantation into the retina of adult rats. Brain Res., 328 (1985) 349-354.
[14] Sokolov, Y. N., Perceptian and the Canditioned Ref1ex. New York: Pergamon, 1963.
[15] Thanos, S., Adult retinofugal axons regenerating through peripheral nerve grafts can restore the light-induced pupilloconstriction reflex. Eur. J. Neurosci., 4 (1992) 691- 699.
[16] Vidal-Sanz, M., Bray, G. M.,Villegas-Perez, M.P., Thanos, S. and Aguayo, A. J., Axonal regeneration and synapse formation in the superior colliculus by retinal ganglion cells in the adult rat. J. of Neurosci., 7 (1987) 2894-2909.

LEGENDS
Figure 1. Typical examples of cortical EEG desynchronization to light stimuli in normal control (A) and PN grafted rats (B). EEG was recorded from the sensorimotor cortex in both cases. Light duration were 200 ms (A) and 660 ms (B), respectively. In both cases, immediately after light stimuli high amplitude slow waves disappeared abruptly and low amplitude higher frequency waves became dominant.

Figure 2. Typical example of behavioral arousal to light in a normal PN grafted rat. Four flames were picked up at different time before and after light stimulus. In this trial, light duration of 500 msec (thick bar) presented during slow wave sleep elicited EEG desynchronization (thin horizontal bar), which continued on then behavioral change (tail movement in this case) occurred with latency of 5.2 sec from light onset. Tip of the tail was indicated by arrow in upper four photographs. Changes in position of tail tip were displayed as horizontal (X) and vertical deviations (Y) by measuring distances from the lower left corner of the flame to the tip in sequential video field.

Table 1. Summary of EEG desynchronization and behavioral arousal response to light stimulus.