ABSTRACT
Unilateral left tectal ablation was carried out in Xenopus between stage 48 and 1 month after metamorphosis. Six to 12 weeks after metamorphosis the retinal projection from the right eye was examined with the use of [3H]proline autoradiography. The autoradiographs indicated that optic fibres whose tectal target was destroyed recrossed to the ipsilateral tectum and basal optic nucleus via the posterior and pretectal commissures. No anomalous recrossing occurred if the tectal ablation was carried out at stage 58 or later. The aberrant optic fibres were restricted to the rostrolateral and central parts of the ipsilateral tectum and they terminated in a discontinuous manner. It is concluded that available surfaces serving as contact guidance cues are needed to direct aberrant optic fibres to the ipsilateral tectum.
INTRODUCTION
Neurohistological studies in newborn hamster have demonstrated an anomalous retinocollicular projection following unilateral destruction of the superior colliculus (Schneider, 1973). Retinal fibres destined for the ablated contralateral terminal area, after recrossing the midline, innervated the superficial layer of the ipsilateral superior colliculus. After unilateral lesion of the right tectum with both optic nerves remaining intact, an anomalous retinal projection to the ipsilateral tectum has also been shown in adult goldfish (Sharma, 1973).
In adult Anura the effect of partial tectal removal on the formation of the retinotectal projection has been studied in detail with electrophysiological recording techniques. Following tectal ablations and without optic nerve regeneration part of the retinal projection, corresponding to the ablated area, could not be found on the residual tectum (Straznicky, 1973; Meyer & Sperry, 1973). Similar results have also been reported after partial tectal ablation in larval Xenopus (Straznicky, Gaze & Keating, 1971). Recent findings in Rana pipiens indicate that at least some of the optic fibres whose terminal areas were removed, may be relocated in the residual tectum, though their position is abnormal with respect to the retinal topographic map (Udin, 1977). In these experiments, however, the intact tectum was not examined for aberrant retinal fibres from the ipsilateral eye nor were their actual terminations determined morphologically. Mammalian results have shown that an anomalous ipsilateral retinocollicular projection can be induced only in the neonate and not in the adult (So & Schneider, 1978). These observations may indicate that the induction of an aberrant retinal projection in Anura by tectal ablation is also age dependent. We here report autoradiographic evidence that (a) in Xenopus following unilateral tectal ablation without optic nerve section, retinal fibres grow into the ipsilateral tectum; and (b) abnormal ipsilateral retinotectal projections do not appear if the tectal ablation is performed at stage 58 or later.
MATERIALS AND METHODS
The South African clawed-toed frog Xenopus laevis was used in this study. Animals at larval stages 48, 50, 52, 55, 58, 65 (after Nieuwkoop & Faber, 1956) and 1 month after metamorphosis were anaesthetized with MS222 (Tricaine methanesulphonate, Sandoz), by immersion in a 0-1% aqueous solution. The left optic tectum was approached by an incision in the skin and in the cartilaginous skull. After exposure, the left tectum was cut along its medial, lateral, rostral and caudal boundaries down to the optic ventricle and the excised tissue was removed by suction. The skull patch was replaced and the skin approximated and sealed with Ethicon (isobutyl-2-cyanoacrilate monomer) tissue adhesive. [3H]Proline autoradiography and on some operated animals electrophysiological studies were carried out between 6 and 12 weeks after metamorphosis. In other animals (operated at stage 52) the autoradiographic studies were performed between 4 and 30 days after tectal ablation.
Autoradiography
Twenty-four hours before sacrifice 4-10μCi [3H]proline (specific activity 24 Ci/mmol, Amersham) varying according to the age of the animals, was injected into the posterior chamber of the right eye. The brains were fixed in Bouins’ solution and embedded in paraffin. Serial sections (10 μm) in the transverse plane were dipped in Ilford K2 emulsion and developed after 2 weeks storage in light tight boxes at 4°C. Slides were counterstained with Harris’s haematoxylin. For control purposes isotope was injected into the right eye of four normal animals.
Electrophysiology
The retinal projections from the right eye to the residual contralateral and intact ipsilateral tecta were assessed according to the method described previously (Straznicky, 1976). In short, a lacquer-insulated tungsten micro-electrode was placed at regularly spaced tectal positions and the corresponding visual receptive fields, from which multi-unit evoked action potentials could be obtained, were mapped.
RESULTS
Forty operated animals with longer and 24 animals with shorter survival times have been included in this report. The attempted tectal removal in 40 animals with longer survival times was complete in 24, partial in 12 and unsuccessful in 4 cases (Table 1). Autoradiographic studies have shown that in 19 animals, where the tectal ablation was performed between stages 48 and 55, that is from early to midlarval life, the right intact tectum contained autoradiographic silver grains indicative of the presence of optic fibres from the ipsilateral eye. Following isotope administration in animals with later operations (between stage 58 and 1 month after metamorphosis), silver grain deposition was found only in normal optic fibre receiving areas, i.e. bilaterally in the thalamic visual centres, in the contralateral basal optic nucleus (BON) and in the contralateral residual tectum (Fig. 1C).
The results of tectal ablation carried out at different larval stages and after metamorphosis in 40 animals

(A-E) Darkfield photomicrographs of the tectodiencephalic border of animals with left tectal removal following [3H]proline injection into the right eye. On each of these and the following photomicrographs, the contralateral half is on the right and the ipsilateral half of the brain is on the left side and the bar represents 500 μm. (A) The animal was operated at stage 55 and killed after metamorphosis. The pretectal commissure and the ipsilateral tectum are heavily labelled. (B) The animal was operated at stage 50 and killed after metamorphosis. The posterior commissure is heavily labelled. (C, E) The animal was operated at stage 52 and killed after metamorphosis. Notice in (C) that optic fibres (arrow) are descending towards the accessory optic tract leading to the BON. Section in (E) is 200 μm caudal from the section shown in (C), both contra- and ipsilateral (arrow) BON are labelled. (D) The animal was operated at stage 52 and killed 12 days later. Recrossing optic fibres are labelled (arrow).
Fig. 2. Holmes’ silver impregnation on normal tadpole at stage 50 at the level of the developing posterior commissure. Arrow indicates the point at which commissural fibres intermingle with optic fibres, directed towards the tectum.
(A-E) Darkfield photomicrographs of the tectodiencephalic border of animals with left tectal removal following [3H]proline injection into the right eye. On each of these and the following photomicrographs, the contralateral half is on the right and the ipsilateral half of the brain is on the left side and the bar represents 500 μm. (A) The animal was operated at stage 55 and killed after metamorphosis. The pretectal commissure and the ipsilateral tectum are heavily labelled. (B) The animal was operated at stage 50 and killed after metamorphosis. The posterior commissure is heavily labelled. (C, E) The animal was operated at stage 52 and killed after metamorphosis. Notice in (C) that optic fibres (arrow) are descending towards the accessory optic tract leading to the BON. Section in (E) is 200 μm caudal from the section shown in (C), both contra- and ipsilateral (arrow) BON are labelled. (D) The animal was operated at stage 52 and killed 12 days later. Recrossing optic fibres are labelled (arrow).
Fig. 2. Holmes’ silver impregnation on normal tadpole at stage 50 at the level of the developing posterior commissure. Arrow indicates the point at which commissural fibres intermingle with optic fibres, directed towards the tectum.
(A) Abnormal optic fibre pathways
In normal Xenopus the optic fibres after crossing the midline in the chiasma course in the optic tract dorsocaudally along the margin of the diencephalon. At the level of the posterior commissure the optic tract divides into medial and lateral branches for entry into the tectum. A small component of the retinal projection, from the peripheral temporo-ventral retina, terminates in ipsilateral thalamic visual centres. In animals with left tectal removal between stages 48 and 55 heavy silver grain deposition was observed either over the posterior commissure (Fig. 1B) or over the pretectal commissure (Fig. 1A). In some animals both commissures were labelled. These findings clearly show that optic fibres recrossed the midline and entered the ipsilateral side of the brain via the pretectal and posterior commissures which normally do not carry optic fibres. We did not find any indication that fibres from the operated side entered the ipsilateral tectum through the intertectal commissure, along the tectal midline (Fig. 3B, D). In addition to the anomalous tectal projection the ipsilateral BON, normally receiving contralateral retinal input only, became also innervated (Fig. 1E). Because of the very few ipsilateral fibres reaching the BON their anomalous pathway was difficult to establish. It appears, however, that aberrant optic fibres from the ipsilateral pretectal area descend down to the ventral diencephalon (Fig. 1C) in order to join the accessory optic tract coursing from the chiasma to the BON. The very heavy labelling of the pretectal area on the side of the lesion (Figs. 1A and 3 A) on the other hand indicates that a substantial part of the retinal projection, displaced by the ablation, may have been arrested between the posterior thalamus and the residual rostral tectum. In contrast to this heavy labelling of the pretectal area the density of silver grain deposition over the contralateral thalamic visual nuclei was normal.
(A-D) Light field photomicrographs taken from [3H]prolineautoradiographs. (A, B) Animal operated at stage 48 and killed after metamorphosis. The pretectal area on the contralateral side (A) is heavily labelled. Recrossing optic fibres are present in the posterior commissure (arrow). About 250μm caudal from the section shown in (A) the contralateral part of the ipsilateral tectum is labelled (B). (C) Animal with partial tectal ablation at stage 65 and sacrificed 12 weeks after operation. Only the contralateral residual tectum is labelled. (D) Animal operated at stage 55 and killed after metamorphosis. Arrows indicate areas of high silver grain density in the ipsilateral tectum. Calibration is the same for all micrographs.
(A-D) Light field photomicrographs taken from [3H]prolineautoradiographs. (A, B) Animal operated at stage 48 and killed after metamorphosis. The pretectal area on the contralateral side (A) is heavily labelled. Recrossing optic fibres are present in the posterior commissure (arrow). About 250μm caudal from the section shown in (A) the contralateral part of the ipsilateral tectum is labelled (B). (C) Animal with partial tectal ablation at stage 65 and sacrificed 12 weeks after operation. Only the contralateral residual tectum is labelled. (D) Animal operated at stage 55 and killed after metamorphosis. Arrows indicate areas of high silver grain density in the ipsilateral tectum. Calibration is the same for all micrographs.
(B) Anomalous tectal termination of ipsilateral optic fibres
Following [3H]proline injection in normal animals the superficial layer of the contralateral tectum, receiving the optic fibre input, is evenly labelled across its whole rostrocaudal and mediolateral extents, and there is no direct input to the ipsilateral tectum. In those operated animals in which there was autoradiographic evidence of a projection to the ipsilateral tectum, grains were localized to the superficial layer of the ipsilateral tectum, though consistently the tectal surface was partially and very unevenly covered. In fact, apart from two animals (Fig. 3D), where more than half of the tectal surface was covered, only the rostro-lateral and central parts of the tectum revealed the presence of silver grains (Fig. 3B). In three animals, representing typical cases, the total extent of the ipsilateral retinal projection has been reconstructed on the dorsal tectal surface by plotting the area of silver grain deposition in the tectum in 100 μm steps. These results show that the heavy silver grain deposition, marking the terminal arborization of optic fibres, is abruptly followed by areas where very light or no silver grain deposition is present (Fig. 4A-C). The patchy deposition of silver grains mediolaterally and to a lesser extent rostrocaudally, is quite apparent in Fig. 4B, reconstructed from an animal with a tectal ablation at stage 55.
The distribution of aberrant optic fibres in the ipsilateral tectum. The curving tectal surface is flattened out two-dimensionally. Hatched area of the tecta corresponds to the termination of aberrant optic fibres as revealed by [3H]-proline autoradiography. The right side of the drawings show the extent of the residual tectum (shaded area). Diagram A was drawn from an animal whose autoradiograph is shown in Fig. 3B. The level of the autoradiograph is indicated with a broken line. Diagram B was reconstructed from an animal whose autoradiograph is shown in Fig. 3D. The level of the autoradiogram is indicated with a broken line. Diagram C was drawn from an animal operated at stage 52 and killed after metamorphosis.
The distribution of aberrant optic fibres in the ipsilateral tectum. The curving tectal surface is flattened out two-dimensionally. Hatched area of the tecta corresponds to the termination of aberrant optic fibres as revealed by [3H]-proline autoradiography. The right side of the drawings show the extent of the residual tectum (shaded area). Diagram A was drawn from an animal whose autoradiograph is shown in Fig. 3B. The level of the autoradiograph is indicated with a broken line. Diagram B was reconstructed from an animal whose autoradiograph is shown in Fig. 3D. The level of the autoradiogram is indicated with a broken line. Diagram C was drawn from an animal operated at stage 52 and killed after metamorphosis.
(C) The development of the aberrant retinotectal projection
In order to establish when the anomalous projection evolves after tectal ablation the first appearance of silver grains in the commissural pathways and in the ipsilateral tectum was determined in a further 24 animals operated on at stage 52. The earliest time at which it was possible to identify aberrant optic fibres in the pretectal and posterior commissures was 6 days post-operatively (Table 2). Twelve days after tectal ablation the presence of aberrant optic fibres in the ipsilateral tectum can clearly be shown autoradiographically (Fig. 1 D). The amount of silver grains over the commissures and ipsilateral tectum increases considerably with longer survival times and by the time of metamorphic climax (30 days after tectal ablation) the picture is very similar to that described in section A, on older animals. It appears from these cases that from the metamorphic climax onwards very few, if any, aberrant optic fibres manage to recross to the ipsilateral tectum.
(D) Visual field projection to residual tectum
In 12 animals where the tectal ablation was incomplete the retinal projection to the residual tectum was mapped. Since the results of adult tectal lesions on the visual projection has already been published (Straznicky, 1973) only a very brief account will be given. It can be seen in a typical case (Fig. 5) that only the corresponding visual field, in this particular animal the nasal visual field, projects to the residual rostral tectum. Thus about one half of the visual field is not represented on the residual tectum. The results of proline autoradiography on this animal suggest that some of the retinal fibres destined for the operated tectum terminated abnormally in the ipsilateral tectum. In each of the animals with partial tectal ablation, and in a few animals with total tectal ablation, the right tectum was also examined electrophysiologically. However, no evoked action potentials from the right eye were recorded despite the presence of a morphologically identifiable optic fibre projection from this eye.
A diagrammatic representation of the results of electrophysiological mapping of visual field projection. The numbers on the tectum represent electrode positions. The corresponding stimulus positions are indicated on the chart (lower part of the diagram) of the visual field by the same number. Open circles in the ipsilateral tectum indicate electrode penetrations from where no visual responses were obtained through right eye. The left eye during recording was covered with an opaque lid. From the hatched area of the visual field no visual responses were elicited.
A diagrammatic representation of the results of electrophysiological mapping of visual field projection. The numbers on the tectum represent electrode positions. The corresponding stimulus positions are indicated on the chart (lower part of the diagram) of the visual field by the same number. Open circles in the ipsilateral tectum indicate electrode penetrations from where no visual responses were obtained through right eye. The left eye during recording was covered with an opaque lid. From the hatched area of the visual field no visual responses were elicited.
DISCUSSION
Following section of the optic nerve regeneration occurs and optic fibres invade both the contra and ipsilateral tecta (Glastonbury & Straznicky, 1978). Behavioural and electrophysiological studies in Anura indicate that optic fibres may also reach the tectum from completely anomalous entry points, such as the oculomotor root (Gaze, 1959; Hibbard, 1967) or the dorsal surface of the diencephalon (Sharma, 1972). In the present experiments the optic pathway was left intact and only the tectal target area of the retinal projection was removed. Our autoradiographic analysis shows that aberrant optic fibres consistently choose the posterior and pretectal commissures for entry into the ipsilateral tectum (Fig. 6) and that no anomalous optic fibre recrossing occurs following tectal ablation after stage 58.
A schematic drawing of the aberrant optic pathway from the contralateral pretectal area via the posterior and pretectal commissures to the ipsilateral tectum.
Studies in hamsters with unilateral ablation of the superior colliculus have shown that aberrant optic axons choose the intertectal commissure to reach their new terminations (Schneider, 1973). It is surprising that this is not true for Xenopus. In the majority of our cases the attempted tectal removal was complete, so that optic fibres from the pretectal area were prevented from reaching the intertectal commissure. In animals where the rostral part of the tectum, with the intertectal commissure, had been spared recrossing optic fibres still favoured the posterior and pretectal commissures. It is conceivable that regenerating optic fibres choose pre-existing commissural pathways instead of establishing new routes to the ipsilateral side. The proximity of the posterior and pretectal commissures to the pretectal area where the truncated optic axons are located is quite apparent. It is also worth noting that the formation of the pretectal and posterior commissures precedes the formation of the intertectal commissure (unpublished observation) and that the former pathways are laid down in midlarval stages (Fig. 2) when the anomalous optic fibre recrossing occurs. Directional growth of optic axons to their tectal termination during regeneration (Attardi & Sperry, 1963) and pathway recognition during development (Straznicky, Gaze & Horder, 1979) have been reported. Our present observations indicate another important factor, that is, the existence of suitable pathways in an advantageous position, which may facilitate optic axons to grow to their alternative termination in the ipsilateral tectum. Apparently, the availability of the suitable pathway may be restricted in time as has been demonstrated in the present experiments. The discrepancy between experiments in adult goldfish reporting ipsilateral optic fibre growth following tectal ablation (Sharma, 1973) and the lack of such aberrant optic fibre growth in late larval and in postmetamorphic Xenopus is not easy to reconcile. In the goldfish experiments the regenerating retinal axons were apparently diverted back to the chiasma from where they grew along the optic tract to the ipsilateral tectum. Our proline autoradiography did not give any indication of the presence of aberrant optic fibres traversing in the chiasma to the ipsilateral tectum. Thus the failure to induce aberrant fibre growth after stage 58 in Xenopus could be due to species differences, in the form of differences in the microstructure of the frog and goldfish brains.
Although some variability in the extent and the location of the aberrant optic fibre projection was observed in the present experiments, fibres consistently terminated in a patchy, uneven manner in the ipsilateral tectum. Discontinuous optic fibre projection is not observed in normal animals, nor has it been reported in adult Xenopus in which the optic nerve has regenerated (Glastonbury & Straznicky, 1978). This patchiness has, however, been seen in tecta receiving direct optic fibre input from both eyes in goldfish (Levine & Jacobson, 1975) and hamster (Jhavesi & Schneider, 1974). It has been suggested (Levine & Jacobson, 1975) that the discontinuity of the aberrant retinal projection may be due to the exclusion of contralateral retinal fibres from parts of the tectum. Since the contralateral retinal projection to the intact tectum has not been examined in the present series of experiments, we cannot decide whether the aberrant optic fibres are superimposed on a continuous retinal projection from the other eye or whether both the aberrant and the normal contralateral projections are discontinuous in a mutually exclusive manner.
In none of the animals subjected to visual field mapping was it possible to demonstrate electrophysiologically the presence of aberrant optic fibres in the ipsilateral tectum or in the contralateral residual tectum. We have shown that a superimposed ipsilateral retinal projection recovers much more slowly than a regenerating contralateral projection (Glastonbury & Straznicky, 1978). The reasons for this phenomenon are not understood. Our autoradiographic evidence suggests that retinal fibres whose targets were removed by late larval or metamorphic tectal ablation remain confined to the residual tectum. This assumption is compatible with recent electrophysiological findings (Udin, 1977). It is, however, impossible to elucidate, without the results of electrophysiological mapping, whether in the case of early and midlarval partial tectal removal the anomalous projection in the ipsilateral tectum arises selectively from the fibres deprived of their tectal targets or whether they represent a random sample of optic fibres from the whole retina.
ACKNOWLEDGEMENTS
The authors would like to thank Ms Theresa Clark for expert histological assistance. This work was supported by a grant from the Australian Research Grants Committee.