- Email: [email protected]
Outgrowth and directional specificity of fibers from embryonic retinal transplants in the chick optic tectum
Developmental Brain Research, 32 (1987) 161-179 Elsevier
161
BRD 50515
Research Reports
Outgrowth and directional specificity of fibers from embryonic retinal transplants in the chick optic tectum Solon Thanos and Dieter Dtitting Max-Planck-lnstitut fiir Entwicklungsbiologie, Tiibingen (F.R. G.) (Accepted 9 September 1986)
Key words: Chick retinotectal system; Transplant; Neuronal specificity
Retinal pieces taken from known positions of 6-day chick embryos were vitally labeled with the fluorescent dye Rhodamine-B-isothiocyanate (RITC). They were then transferred onto the surfaces of optic tecta following early bilateral removal of the embryo's optic vesicles. One to 5 days after transplantation the tecta were fixed and transplants that issued fibers were examined on tectal wholemounts or were sectioned and viewed with a fluorescence microscope. Retinal fibers growing out from transplants on day E6 tecta showed a capacity for changing their initial outgrowth directions and for reorienting themselves towards their specific retinotopic projection area. Frequently, changes in growth direction appeared in a right-angled pattern. The capacity for turning was strongest for fibers of nasal retinal origin, less strong for fibers of temporal origin, and occurred rarely but unquantifiably in the case of fibers of ventral retinal origin. Fibers of all investigated retinal quadrants were found to reach their corresponding projection areas and to arborize there, that is, fibers of nasal retinal transplants in the posterior tectum, of temporal transplants in the anterior tectum, and of ventral transplants in the dorsal tectum. Furthermore, once in their target region, the fibers left the outer layer of the tectum and turned, again in right angles, to invade deeper layers. Capacity of fibers to turn towards their projection area was not observed for fibers issued from transplants placed on the tectum later than day E8. We suggest that there is a specific guidance of retinal axons on the tectum.
INTRODUCTION The t o p o g r a p h i c p r o j e c t i o n of the retina onto the v e r t e b r a t e tectum is organized roughly linearly and in a mirror-image fashion preserving the spatial relationships within the visual field. Possible mechanisms governing d e v e l o p m e n t of this orderly projection have been the subject of several investigations in the past. E x p e r i m e n t a l a p p r o a c h e s in fish and amphibian visual systems, from which most concepts are derived, were insufficient for the less accessible avian visual system. The few d a t a on the mechanisms involved in the d e v e l o p m e n t of retinotectal projection in birds derive from newly d e v e l o p e d n e u r o a n a t o m i cal labeling techniques. Besides being m a p p e d electrophysiologicallyn, the chick retinotectal system
was investigated neuroanatomically after partial ablation of the retina 1336'32, after a n t e r o g r a d e degeneration l°'16, by local staining procedures 37'38, and in transplantation experiments of embryonic retinal pieces onto the tectum 14"19. The results of the transplantation experiments of D e L o n g and C o u i o m b r e 14 and that of G o l d b e r g 19 display an inconsistency with respect to the specific guidance of outgrowing fibers. According to D e L o n g and C o u l o m b r e the quadrants of 4-day-old embryonic retina, transplanted onto 6day-old tectum, issued fibers that showed a specific orientation according to the functional retinotopicity. In a repetition of these experiments, G o l d b e r g found that outgrowing fibers followed the same straight courses taken by carbon particles implanted at c o m p a r a b l e positions in control embryos. In other
Correspondence: S. Thanos, Max-Planck-Institut fiir Entwicklungsbiologie, Abteilung Physikalische Biologie, Spemannstr. 35, D7400 Tfibingen, F.R.G. 0165-3806/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
162 words, retinal fibers did not show specific orientation toward their appropriate tectal quadrants. These reports brought some confusion into the discussion whether and to what degree growing optic axons are specifically guided. In the meantime transplantation approaches were reported from other systems, such as the Xenopus is and the rat visual system 23'29'33, the basal ganglia 1'5, cortex °'24'27, locus coeruleus 4 and cerebellum 21. In many of these experiments, fibers growing out from transplanted embryonic tissue showed specific recognition of the appropriate projection area, and in some cases formed significant functional connections within this area 4-6'23. New techniques have been added to the older histological procedures, in particular neuronal labeling with fluorescent dyes transported anterogradely 7'9'37-41. The present investigation analyzes the pathways of retinal fibers growing out on chick tectum. Small pieces of 6-day retina from known positions within the eye that show a capacity to issue fibers in vitro 7'9'20 w e r e vitally labeled with RITC and grafted directly onto the tectal surface (underneath the pial membrane) of 6-12-day-old anophthalmic host embryos. We were interested in the following questions: (1) Do outgrowing fibers from retinal transplants follow specific routes on the tectal surface? (2) Would retinal fibers arrive at their retinotopically correct projection field? (3) Would they recognize this field? (4) Is there a 'critical period' in embryonic development during which time fibers are guided specifically? MATERIALS AND METHODS
Host embryos Fertilized chicken eggs (Hiline Hisex, White Leghorn) were incubated at 38 °C for 38-43 h in a forced-draft incubator. The eggs were windowed after removal of 2 ml of albumen through a hole near the egg's tip. The embryos were staged and (if necessary) further incubated until stage 11 (12-13 somites) of the Hamburger-Hamilton scale = was reached. After injecting a small amount of Pelikan drawing ink (No. 17, black, India) beneath the head region of the embryo, the two optic vesicles were cut out by means of electrolytically sharpened tungsten needles (NS-86 tungsten wire, ~ 0.25 mm, obtained
from GTE Sylvania Licht G m b H , Bad Hersfeld, electropointed in a solution containing 71 g NaNO 2 and 34 g K O H in 100 ml distilled water, fine hooks produced by means of forceps). We chose the bilateral enucleation because it ensures that neither contranor ipsilateral retinal fibers invade the exposed tectal lobe, contrary to an exclusive removal of one eye vesicle which results in an enhanced ipsilateral retinal innervation of this tectum 34'39. The eggs were returned to the incubator until embryonic day E3 (stage 20). Then the contents of the eggs were transferred into plastic dishes for further development as described elsewhere 2'37'3s. In some embryos at stage 11, not only the optic vesicles were excised. In addition, we excised the right mesencephalic alar plate 3 and reinserted the rectangular piece of tissue after turning it by 180° . In this way, the antero-posterior and the dorso-ventral axes of the alar plate tissue were reversed. A normal right tectum results if the rotated piece is excised precisely along the dorsal midline and the right side of the neural tube, along the line that separates the upper alar plate from the lower basal plate. According to unpublished results (D. Dfitting), [3H]thymidine incorporation into operated tecta, sliced at days E6 to E8 into horizontal strips from the anterior to the posterior pole, revealed the normal gradient of proliferation with incorporation increasing towards the posterior pole.
Preparation of donor retina The left or right retina of a donor embryo (day E6) was dissected out completely and flattened on a piece of Sartorius membrane filter (type SM, black) in a serum-free Eagle's medium (without Ca and Mg) 9'19. The spread retina was incubated in 4 ml Eagle's medium containing about 150pl of an aqueous RITC solution (0.1-0.3%, Sigma R 1755) for 10 min at room temperature with occasional shaking. The filter with the retina was then washed by dipping it 4x into fresh Eagle's medium without RITC. The filter was transferred into Eagle's medium containing 10% fetal calf serum + Ca/Mg, and was incubated in a Heraeus cell culture incubator for 30-60 rain at 30 °C and 4% CO 2. A drawing of the retina by means of a camera lucida allowed us to determine the retinal axes (provided the optic fissure was left intact).
163
Transplantation procedure The chlorioallantoic membrane over the exposed tectum of the eyeless host embryo (in the plastic dish) was opened for the transplantation. A small piece of retinal tissue (0.02-0.03 mm 2) from a known position with respect to the dorsoventral and nasotemporal axes was transferred under the pial membrane of a defined position on the host tectum by means of a glass capillary. Transplantation success was controlled under an improvised fluorescence dissection scope (Zeiss). All manipulations were performed at 37 °C under sterile conditions.
Microscopy The transplanted embryos were further incubated for 1-5 days, the tecta dissected out, freed of all meninges, and fixed for at least 1 day in 4% paraformaldehyde in phosphate buffer (pH 7.2). The tecta were cut into a dorsal and a ventral half, and the two halves were mounted on a gelatine-coated glass slide in a solution of 2% n-propyl-gallate (Merck), 8% phosphate buffer (pH 7.2) and 90% glycerol. Mounted tectal halves were viewed with epifluorescence using a Zeiss universal microscope. The wavelength of the incident light was 520 nm, fluorescence was observed at 590 nm. Cryostat sections (20-40/~m) were prepared from representative specimens to control the position of transplants in relation to tectal depth and the layer(s) in which fibers from the transplant grow out preferentially. We examined the host tectal surfaces (days E 4 El4) in the absence of any visual input following bilateral ablation of eye primordia at early stages by scanning electron microscopy (SEM). The tecta were fixed overnight in fixative containing 2.5% glutaraldehyde in 0.1 M phosphate buffer at pH 7.2, postfixed in 1% buffered osmium tetroxide for 30 min and dehydrated in a series of ascending ethanol concentrations. They were then treated according to Krayanek 26. The results of these controls confirmed Krayanek's studies according to which the tectal surface does not consist of an oriented extracellular matrix 26. These experiments are not further described in the present report. In the same way we prepared representative tecta carrying a retinal transplant to control the graft's position on the tectal surface.
RESULTS The transplantations performed between days E6 and E12 are divided into 3 groups. Group 1 contains transplantations at days E 6 - E 7 with postoperative survival of 1-3 days. Out of more than 80 transplanted embryos, 49 had a visible transplant with fibers grown after that time. In this group we analysed the direction of initial growth as well as corrective phenomena. Group 2 contains successful transplantations (28 cases) at day E6 with postoperative survival of 3.5-5 days. In this group of transplants we investigated the ability of fibers to find their normal termination field and their capacity to recognize it. Group 3 summarizes all successful experiments (8 cases) performed between days E8 and E12. In this group we investigated the ability of axons to recognize tectal cues after day E8. The postoperative survival varied between i and 4 days. Known in our system were (1) the positional origin of the retinal graft, (2) the position on the tectum which received the transplant, and (3) the approximate size of the transplant and the age of donor and recipient. Unknown was the orientation of the transplant on the tectum with respect to its original retinal axes. The following properties of outgrowing axons were scored quantitatively as far as possible, otherwise qualitatively, using approximate number of outgrowing fibers, their lengths (in particular those of longest fibers), direction of initial outgrowth, depth of graft in relation to tectal tissue as revealed by microscopic focusing, fasciculation, capacity of branching, correction of initial routes and finally the fibers' fate within the projection area and within inappropriate tectal areas. Independent of their preferential growth direction, the number of outgrowing fibers varied between a few and hundreds of axons depending on the size of the transplant, although we tried to transfer pieces of roughly the same size. Thus, in some cases we observed only a few outgrowing fibers, in others not a single one. The quantity of fibers seemed to be independent of the recipient's age, but dependent on the stage of retinal differentiation which is not uniformly advanced at day E6. The nasoventral retinal quadrant differentiates with a time delay of 12-24 h in relation to temporal and dorsal retinal regions 15.
164 According to earlier in vitro investigations, the optimal age to prepare retinal explants is day E6 (refs. 7, 9 and 20). For this reason we transplanted day E6 retinal tissue. Fibers growing out from transplants of various retinal regions did not differ in their morphology. They did not grow out preferentially as a single fascicle, as described for day E4 grafts placed mainly on top of the pial membrane ~4'19. In our experiments, fibers grew out of the transplant in more than one direction, often radially in multiple directions (Fig. 1). Two types of fibers could be identified: (1) short fibers (the majority) with lengths up to 0.5 mm growing randomly around the transplant, forming fascicles up to 10 fibers and branching shortly after outgrowth; and (2) straight growing axons, thin, non-branching, non-fasciculating with lengths up to 4 ram. The long fibers are almost certainly ganglion cell axons, the
short fibers dendrites or processes from ganglion cells and other types of retinal cells. With retinal explants of the same embryonic age, Halfter et al. z~ demonstrated by means of retrograde H R P transport that the long fibers in in vitro preparations derive from cells in the ganglion cell layer. In our analysis we counted only fibers (axons) longer than 0.5 mm which did not remain near the graft.
Initial outgrowth of fibers The earliest outgrowing fibers could be detected about 12 h after transplantation. The direction of initial outgrowth depended on the age of the host tecturn, that is, the younger the tectal tissue, the higher the fraction of fibers growing in dorsal or ventral directions. In tecta older than day E8 the fibers do not seem to have a preference for the dorsal or ventral direction, they grow out randomly in a radial fashion.
Fig. 1. Fluorescence micrograph showing a fluorescing nasal retinal piece grafted on the anterior tectum at day E8 and observed at day El0. Position of the transplant is shown on the schematic embryo head in the upper left corner. Fibers grow out almost radially without a clear preference. Scale bar = 100/~m.
165 TABLE I Quantification of the data obtained from grafting experiments on day E6 with short postsurgical survival The first line shows the retinal origin of the grafts, the second line shows the tectal position to which the retinal grafts were transferred, with numbers of experiments in parentheses. The lines 'Initial outgrowth direction' give the numbers of experiments in which long axons (see text) grew out initially in anterior (A), posterior (P), dorsal (D), ventral (V) or radially in random (0) directions. The last lines analyze the final directions into which fibers preferentially grew after correcting their route. Details are given in the text. Retinal origin
N (22)
Tectal position Initial outgrowth direction Final directional preference Yes
A(13) A(0) P(0) O(13)
P(9) A(0) P(0) 0(9)
P (18) A(3) P(0) O(15)
A(8) A(5) P(0) 0(3)
D(3) V(0) D(0) 0(3)
V(13) V(6) O(0) 0(7)
V(18) V(7) D(4) 0(7)
D(1) V(0) D(1) 0(0)
P(9) A(0) 0(4)
A(0) P(0) 0(9)
A(13) P(0) 0(5)
P (0) A(5) 0(3)
V(0) D(0) 0(3)
D(0) V(6) 0(7)
D(6) V(7) 0(5)
V(0) D(1) 0(0)
No
T (26)
D (16)
V (19)
O
/
I
Fig. 2. Fluorescence photomicrograph (a), accompanying drawing (b) and schematic drawing of head (c) showing the position of a nasal transplant (Exp. No. HTr13) on the anterior tectum as well as the direction and corrective behavior of outgrowing fibers. Axons, initially oriented in ventral direction, turn in posterior direction by nearly right-angled routes. Some of them are marked by arrows in drawing (b). A, anterior; D, dorsal; P, posterior; Tel, forebrain; Tec, tectum; Tr, transplant. Asterisk in c marks the position of absent left eye. Scale bars: 100/~m for (a) and (b), 1 mm for (c).
166 Fig, 1 p r e s e n t s such an e x p e r i m e n t at days E 8 - E 1 0 .
o b s e r v e d a differential b e h a v i o r of fibers e m e r g i n g f r o m the graft.
Corrective behavior of fibers D e p e n d i n g on the t r a n s p l a n t ' s retinal origin, we
Nasal transplants'.
N i n e nasal transplants placed at
day E6 o n t o the p o s t e r i o r t e c t u m ( p r o j e c t i o n area) is-
Fig. 3. Fluorescence photomicrographs showing a nasal transplant (Expt. HTr 143) on the anterior tectum, a: fibers leaving the transplant are oriented either directly to the posterior part of the tectum or in dorsal direction. From this population of dorsally oriented axons, several turn, by sharp right angles (small arrows), to posterior direction, b: higher-power magnification of the region marked by asterisk in (a) and (b). c: drawing showing the position of the transplant on the anterior rectum as well as the direction of outgrowth and correction as described above. Abbreviations as in Fig. 2. Scale bars: 100urn for (a) and (b), ! mm for (cL
167 sued longer fibers either in ventral and dorsal directions or randomly without any preferred direction (Table I, column N/P). In contrast, fibers from 9 of 13 nasal grafts onto the anterior tectum (Table I, column N/A) issued fibers that showed a strong capacity to turn in posterior direction (towards their normal projection area). These fibers often deviated from their initial direction of growth by making nearly right-angled turns (Figs. 2, 3, and see Fig. 8B). In 4 cases, the outgrowth was not oriented specifically. After turning, axons grew straight to the posterior tectum without straying or searching on the tectal surface. Nasal fibers turning away from their projection area were observed rarely. Within the outgrowing fiber population the proportion of turning axons varied between a few percent and about 50% of the outgrowing fibers. Figs. 2 and 3 represent two examples of high-rate correction. The grafting of nasal retinal pieces (7 cases) onto the anterior part of tecta, derived from 180° rotations of the right mesencephalic alar plate at stage 11, did not reveal any differences in the behavior of the outgrowing fibers (data not shown), compared to those
described above. This suggests that the embryological axes are irreversibly fixed some time after stage 11. Transplants originating from temporal retina, which would project onto the anterior tectum, issued fibers which showed a specific reorientation and corrective behavior opposite to those shown for nasal fibers. Out of 8 temporal transplants onto the anterior tectum (the projection area) 5 issued fibers which remained within the anterior tectal half and preferred to grow into regions more anterior than the graft's position. The remaining 3 grafts issued axons in dorsal and ventral directions (Table I, column T/A). Out of 18 temporal transplants on the posterior tectum, 3 issued fibers which grew directly towards the anterior tectum. Of the other 15, which issued axons in dorsal and ventral directions or radially, 10 corrected their initial outgrowth direction towards their anterior target (Table I, column T/P). Figs. 4 - 6 show 3 temporal transplants on the posterior tectum which demonstrate the ability of some axons to turn abruptly towards their target. The proportion of turning fibers in each transplant varies substantially. For example, in
Fig. 4. Fluorescence photomicrograph (a) and accompanying schematic drawing of the recipient head (b) showing one small temporal transplant (Expt. HTr 122) on the posterior tectum with only one outgrowing fiber which turns to anterior direction. The growth cone of the axon is on the left side of the figure. Abbreviations as in Fig. 2. Scale bars: 100ktm for (a) and 1 mm for (b).
168
Fig. 5. High-power fluorescence photomicrograph (a) and schematic head (b) showing several fascicles (big arrows) growing out from a temporal transplant on the posterior tectum (Expt. HTr 99). Single fibers (small arrows) leave the initial ventral growth direction (shown in (b)) and grow in anterior direction. Scale bars: 100 ftm for (a), I mm for (b).
169
6
b
Fig. 6. The path of fibers growing out from one temporal transplant (Expt. HTr 108) on the posterior dorsal tectum, a: fluorescence photomicrograph showing the more or less parallel growth of fibers in dorsal direction (drawn in c). b: drawing from photomicrographic prints: the bundle grows over a substantial distance (about 1.5 mm) straight in dorsal direction before several axons leave and turn in anterior direction (arrows). Scale bar = 100 um.
Fig. 4 the only outgrowing axon turns, in Fig. 5 several axons turn in anterior direction. In comparison to nasal transplants the total number of temporal axons turning to anterior direction was lower. Fig. 8A presents schematically 7 transplantations of temporal retinal pieces onto the posterior tectum. In 9 experiments (5 nasal and 4 temporal retinal grafts) we investigated the trajectory of axons growing out of transplants placed in the middle of the ventral tectal half, in a position that bisects the anteroposterior axis. Surprisingly, none of these transplants issued axons which corrected their path. Axons grew out radially or in dorsoventral directions up to the tectal border where they turned mainly in posterior directions. Table II summarizes the results of these experiments showing the tectal region on which grafts were
placed, as well as the directions of outgrowth. Similar observations (with respect to initial outgrowth and correction capacity) have been made for axons growing out of either ventral or dorsal retinal transplants which in most cases were placed in the middle of the anteroposterior axis, but either on the ventral or on the dorsal tectal half (see below).
Dorsal and ventral retinal pieces were usually grafted onto the ventral half of the rectum, because the dorsal rectum was less accessible at days E 6 - E 7 . Out of 13 dorsal transplants onto the ventral tectum, 6 issued fibers oriented radially which remained mainly within this tectal region (Table I, column D/V, and see Fig. 8D). In t h e remaining 7 transplants, fibers grew out randomly. None of them grew preferentially to the dorsal tectum. In 3 cases in which dorsal transplants were successfully placed on
170
TABLE II
Grafting experiments performed into the middle of the A/P-axis Five nasal and 4 temporal retinal grafts were positioned within the region marked by the black circle on the schematic tectal surface. Thick arrows towards the dorsal and ventral direction show the initial orientation of the majority of axons. Once arriving at the dorsal or ventral tectal border, most of them turned in a posterior, some of them in an anterior direction (two arrows vs one). A restricted radial outgrowth from the area of transplantation is indicated by small arrows on the black circle. The last line negates the question of correction lor both groups of grafts. RETINAL ORIGIN
NASAL(5)
TEMPORAL(4) D
GRAFTING POSIT.
V CORRECTION ?
NO
NO
the dorsal tectal half, fibers grew out without a preferential direction (Table I, column D/D). Transplants from the ventral retina showed a similar pattern of fiber outgrowth c o m p a r e d with dorsal transplants (Fig. 8C). Their fibers p r e f e r r e d to grow out radially. Out of 18 ventral transplants positioned onto the ventral tectum, 4 issued axons in a dorsal, 7 in a ventral direction and 7 did not prefer any direction (Table 1, column V/V). Fiber turning, as described for nasal and temporal fibers, was o b s e r v e d in two of the ventral transplants; if it occurred, fibers were directed towards their corresponding projection area (Table I, column V/V). Fig. 7 shows an example of one of these experiments with ventral fibers turning in dorsal direction. Drawings of fiber pathways from other transplants are shown in Fig. 8C, D. Only one ventral transplant was successfully placed onto the dorsal tectum. Its fibers r e m a i n e d within the dorsal tectal half (Table I, column V/D).
Fate o f fibers in their projection areas To answer the question whether fibers arrive in their projection area, and to examine the fate of fibers after they arrived there, we cultivated 28 successfully transplanted embryos for longer periods of time, i.e. 3 . 5 - 5 days after transplantation at day E6. Arborization was a first criterion to define a fiber as
'terminating'. A second criterion was the ability of fibers to leave the layer of initial outgrowth to invade the underlying tectal layers, as they do in normal development of the projection 2~-35. One difficulty in evaluating such experiments is the e n o r m o u s length of axons (sometimes 3 - 4 mm) and poor fluorescence intensity, particularly in the middle third of the axon's length. Axons of all retinal regions (ventral, nasal and temporal) were able to reach their specific projection areas. The higher the degree of pathway correction, the more fibers from the transplants were later found in the a p p r o p r i a t e tectal projection area. That is, in most cases with nasal transplants on the anterior tecturn we found axons terminating on the posterior tecturn. Fig. 9 shows nasal axons arborizing on the posterior tectum 4 days after transplantation. Fibers of temporal transplants placed onto the posterior tecturn were found less often within the anterior rectum (data not shown). Axons of ventral transplants placed onto the ventral tectum rarely t e r m i n a t e d on the dorsal rectum. Fig. 10 presents a rare case in which several ventral axons stop and branch on the dorsal rectum. The routes of these terminating axons could not be followed over their entire length from the transplant (big arrow in Fig. 10c) to the dorsal rectum (small arrow). Only 3 dorsal transplants were successfully placed onto the dorsal tectum; outgrowing fibers could not be followed to the ventral tectum. Fig. 11 shows a cryostat coronal section through a tectum with fluorescent fiber bundles growing from a nasal transplant within the most superficial tectal layer. The relative positions of fibers correspond to the normal stratum opticum which is absent in these experiments because of early eye removal. In another experiment with a nasal transplant, fibers leaving this layer can be seen in Fig. 12. Note here the sharp fiber turns which are similar to those shown for normal retinotectal axons 28'35. With increasing time after transplantation, an increasing number of degenerating fibers is found on the tectum. W e defined as degenerating those axons which lost their h o m o g e n e o u s l y distributed R I T C fluorescence along their length and possessed shrinking growth cones with retracted processes. Fibers in advanced stages of d e g e n e r a t i o n showed decomposition with irregularly distributed fluorescent varicoses akmg their length and loss of typical growth cone
171
Fig. 7. Trajectories of two turning axons from a ventral transplant (Expt. HTr 190) on the ventral tectum (position of the transplant is shown in the accompanying head drawing and in Fig. 8D, its retinal origin (f) is shown in Fig. 8C. The initial posteroventral direction changes to a clear dorsal one. Scale bars: 100pm for (a) and 1 mm for (b). morphology (Thanos et al., unpublished observations). At 4 - 5 days after transplantation, the time of longest observation, healthy axons have been found almost exclusively within their projection area. Usually the RITC labeling was not strong enough to reveal the whole pathway of these axons. In other tectal areas fibers were already degenerated (data not shown).
Transplantation on days E8-E12 Eight successful transplantations of nasal retina (day E6) onto anterior tectum of days E 8 - E 1 0 revealed a different picture compared with grafting onto tecta of days E6-E7. Fibers growing out after day E8 did not show a specific orientation on the tectum, nor a capacity for correcting initial routes. The observed radial or at least multidirectional outgrowth lacked any preferential orientation. Fig. 1
presents a nasal transplant grafted at day E8 and incubated up to day El0. Transplantations at day E l 0 were terminated not later than day El2, since scanning electron microscope studies of uninnervated tectal surfaces (days E4-E14; Diitting and Thanos, unpublished results) showed significant degenerative changes from day El3 onwards. The analysis of younger uninnervated tectal surfaces by scanning electron microscopy did not reveal any preformed channels that might have been used by the outgrowing fibers. DISCUSSION Embryonic retinal tissue of defined positional origin, labeled with the fluorescent dye RITC, was grafted into defined positions of the tectal surface of eyeless embryos that did not receive any prior input
a
\.\..
J4_~
4
~ ~ H T r ~
,2) HTr28
i
HTr 96
HTr 78
',\
,,
/
/
HTr123
RETINA
HTr14
-
RETINA
HTr 117
i O
~, //' . / ~'HTr 13 e
tel
rpos
ven.
HTr 120
HTr 119
I
Fig. 8. Schematic drawings summarizing the results of selected experiments of the 4 groups of transplantations mentioned. A: 7 examples of temporal retinal transplants on the posterior tectum, a - g mark the origin of the transferred retinal pieces in the accompanying schematic retina. Black arrows (straight and curved) indicate the directions of outgrowth and correction, respectively. Big arrows (not in all drawings) show preferential or fasciculated outgrowth. Asterisks indicate the approximate projection area. B: 5 examples of nasal transplants located on the anterior tectum. Meaning of arrows and small letters as in A. In A and B the transplants were placed close to the bisection line (see E). C: 3 ventral retinal pieces (d, g, h) grafted on the ventral tectal half. D: two dorsal grafts onto dorsal tectum (c, e) and two dorsal grafts (b, f) onto the ventral tectum. Retinal origin as marked on C). E: the orientation of the mounted tectal halves (left) and spread retinas (right) for all groups. The tectum was bisected along the antero-posterior axis into a dorsal half (left) and a ventral half (right). ant, anterior; dor, dorsal; nas, nasal: pos, posterior; tern, temporal; ven, ventral.
173
Fig. 9. Fluorescence photomicrographs showing arborizing nasal axons within the posterior tectum (compare with Fig. 8B). a: one axon that just started to arborize, b and c: two clearly arborizing axons as photographed on a tectal w h o l e - m o u n t with microscopic focusing. Note the presence of a growth cone with various processes (arrows) on each of the axonal branches, Scale bar = 100 p m .
174
~.
Fig. 10. Fluorescence photomicrograph (a) and accompanying drawing (b) shows axons from ventral transplant on ventral tectum arborizing within the dorsal tectum (position shown in c). Several axons reaching this region on straight routes (arrows in a and b) seem to stop in a relatively small region (about 200/~m diameter), and arborize there. Abbreviations, see Fig. 2. Scale bars: 100 btm for (a) and (b), 1 mm for (c).
175
Fig. 11. a: bright-light photomicrograph showing a cryostat section through a recipient tectum carrying a nasal retinal transplant on the anterior tectum (not visible on this section), b: fluorescence photomicrograph of the same section showing the position of the fluorescent fibers (arrows) in a distance of about 2 mm posterior to the grafts. This position of fibers corresponds to the stratum opticum in normal embryos. SC, stratum centrale; SP, stratum periventriculare; SGFS, stratum griseum et fibrosum superficiale; Vent, ventricle. Scale bar = 50am.
176
@ Q
•
•
Q O
O
t
e
O
O
o
0
Fig. 12. High-powerfluorescence photomicrograph(a) and accompanyingdrawing (b) showing 4 fluorescing axons within the outer layer (marked with dotted line in b) of the tectum. Two of these fibers turn sharply (arrows) towards deeper tectal regions (bottom of the pictures). On the right edge one fiber seems to start branching. Scale bar = 50 pm. of optic fibers. Retinal axons issued from the grafts traversed the tectal surface, and, depending on their origin, were able to correct an initially 'wrong' direction of outgrowth. In many cases the fibers finally reached their correct retinotopic projection area as indicated by strong arborization and invasion of deeper tectal layers. Usually we grafted day E6 retinal tissue onto day E 6 - E 7 tectal surfaces and terminated the transplanted embryos after 1-3 days of further incubation. Between days E6 and El2 the rectum matures in the anterior to posterior direction 12'25'28'3°'31which implies that grafts onto the thicker anterior wall of the tectum are easier to perform than those onto the thinner posterior tectal roof. Grafts of nasal retina upon the anterior tectum, that send their fibers to the posterior tectum, have a greater chance of suc-
cess than those of temporal retina onto the posterior tectum, the fibers of which are destined for the anterior rectum. In vitro experiments 2° showed that day E6 retinal tissue is suited particularly well for an optimal outgrowth of fibers. The fluorescent vital labeling with RITC provides advantages and disadvantages 37'3s. The dye accumulates near the growth cone of a fiber and is depleted from the middle of long axons. This can make tracking fibers over long distances difficult. At high magnification, many details of growth cones, terminal arbors and branches along the axon can be seen. The staining, unfortunately, is not bright enough to allow a view of complete projection fields at lower magnification. The poor quality of some RITC batches and a sometimes observed high background fluorescence impede the fiber tracing. At high magnification it is
177 almost impossible to focus on a longer section of the axon, which means that photographs have to be made on different levels of the tectal surface. Likely due to leakage of the dye from the axon, the technique does not allow observations exceeding 4-5 days after grafting. There are two earlier reports that described transplantations of chick embryonic retinal tissue onto the tectum 14'19. Their results are contradictory. With a new technique we tried to get more information about the mechanisms guiding retinal axons growing on the tectum. The following technical modifications improved the transplantation system. (a) Chick embryos growing in plastic dishes 2'37'38 made the tectal surface more easily accessible to local manipulations. In the old reports 14'19, transplantations were performed mainly within the 'central' third of the tectal surface which is the only region well accessible in embi'yos developing in ovo. (b) Transplants were placed (with fine glass capillaries) directly beneath the pial membrane on the tectal surface on which retinal fibers normally grow. According to the previous reports, outgrowing fibers traversed an empty space as a well defined fascicle before penetrating the pia to get access to the tectal surface 14. This implies that in these experiments transplants were placed on top of the pia. In our experiments, fibers, having direct access to the tectal surface, grew out in many directions from the graft and did not form one single fascicle as reported earlier 14'19. Our results are in agreement with the supposed existence of specific guidance cues on the tectum 14. In addition they provide information about the mode of correction. Goldberg's results, which show only fibers running to the posterior tectum, could be explained by the appearance of an ipsilateral retinotectal projection during development 34, 38,39. This is more prominent after unilateral enucleation during early development 39. To avoid any visual innervation of the tectum, we removed both eye cups during early development. Analysis of fiber routes with respect to the transplant's origin in the retina and to the transplant's location on the tectum revealed a differential behavior of outgrowing fibers. Fibers of nasal and temporal origin can clearly recognize their position along the anteroposterior tectal axis. As a consequence, they are able to correct the direction of their outgrowth towards their appropriate projection area. Nasal fibers
seem to have the strongest capacity to correct their routes. A difficulty in the interpretation of these resuits is the observation that neither nasal nor temporal fibers orient themselves retinotopically correctly if the corresponding transplant was placed on the ventral tectum half-way between the anterior and posterior margin. One explanation might be found in the specific distribution of guiding cues on the rectum. Fibers of dorsal and ventral origin showed a weak capacity of correcting their initial outgrowth direction. Most of them traversed long distances on the tectal surface randomly; sharp, right-angled turns were rarely found in these cases. As data with longer survival showed, fibers reached their normal projection area in all groups of experiments. Once this goal was achieved, fibers arborized extensively within this area, and in some cases (shown by sectioning the tectum) left the superficial layer of growth to invade underlying tectal layers in which synaptogenesis takes place in normal development 12'31. Synaptogenesis cannot be observed with the techniques used in our experiments. There seems to be a contradiction between the observed low capacity for correction along the dorsoventral axis and the fact that after long-time survival some axons terminate in appropriate areas along this axis. This could be selective survival of those axons which arrived in the proper target region rather than guided outgrowth. The different routes of these fibers were only partially visible by the technique used in our experiments. Outgrowing fibers communicate with unknown tectal elements, and might orient themselves according to the information they obtain 17'36. The presence of position-specific information was clearly shown in the developing chick visual system 8,9m. Fiber tracing by anterogradely transported RITC in vivo was combined with immunological perturbation of retinal fasciculation by means of antibodies against the neural cell-adhesion molecule (N-CAM). This provided information about the corrective behavior of fibers41. Antibodies injected into the embryonic eye at day E4 appeared to prevent proper fasciculation of axons near the fissure. The resulting misrouted fibers remained ectopic up to the tectum and there, at first, they grew parallel to the normally routed fibers. Then several axons (particularly those growing in close contact to the tectal surface) turned towards
178 their target area. This turning occurred at the correct
which we grafted nasal retina between days E8 and
position along the growth axis (anteroposterior) and
El(). Transplantations of temporal retina onto poste-
resulted in a correction of the dorsoventral position
rior rectum of days E 8 - E 1 0 (two cases) did not re-
of axons. Similar results were obtained from experi-
veal a clear result and therefore cannot be evaluated
ments in which the pathway of first fibers invading a
presently. According to our results, positional cues
given part of the tectum became deflected by an
on the u n i n n e r v a t e d tectal surface might guide the
obstacle during development (Thanos and Bonhoef-
first ingrowing fibers of a normal projection. Proba-
fer, in press). Deflected fibers soon corrected their
bly these cues are not available to fibers arriving later in development, which could be guided to the re-
direction of growth and curved smoothly towards their target, suggesting that they sense directional cues on the virgin tectum with respect to both the an-
maining free synaptic space by the projections established earlier in development.
teroposterior and dorsoventral axes. This is confirmed by the results of the present study.
ACKNOWLEDGEMENTS
According to the results reported here, positional cues are available to retinal fibers on days E 6 - E 7 , when the normal projection has started to invade the anterior tectum. From late day E8 onwards these cues seem to be unavailable to fibers growing from the transplants, at least in the anterior tectum onto
REFERENCES 1 Aguayo, A.J., BjOrklund, A., Stenevi, U. and Carlstedt, T., Fetal mesencephalic neurons survive and extend long axons across peripheral nervous system grafts inserted into the adult rat striatum, Neurosci. Lett., 45 (1984) 53-58. 2 Auerbach, R., Kubai, L., Kinghton, D. and Folkman, J., A simple procedure for long-term cultivation of chicken embryos, Dev. Biol., 41 (1974) 391-394. 3 Birge, W., An analysis of differentiation and regulation in the mesencephalon of the chick embryo, Am. J. Anat., 104 (1959) 431-463. 4 Bj6rklund, A., Segal, M. and Stenevi, U., Functional reinnervation of rat hippocampus by locus coeruleus implants, Brain Res., 170 (1979) 409-426. 5 BjOrklund, A., Schmidt, R.H. and Stenevi, U., Functional reinnervation of theneostriatum in the adult rat by use of intraparenchymal grafting of dissociated cell suspensions from the substantia nigra, Cell Tissue Res., 212 (1980) 39-45. 6 Bj6rklund, A. and Stenevi, U., Intracerebral neural implants: neuronal replacement and reconstruction of damaged circuitries, Annu. Rev. Neurosci., 7 (1984) 279-308. 7 Bonhoeffer, F. and Huf, J., Recognition of cell types by axonal growth cones in vitro, Nature (London), 288 (1980) 162-164. 8 Bonhoeffer, F. and Gierer, A., How do retinal axons find their targets on the tectum? TINS 7 (1984) 378-381. 9 Bonhoeffer, F. and Huf, J., In vitro experiments on axon guidance demonstrating an anterior-posterior gradient on the tectum, EMBO J., 1 (1982) 427-431. 10 Carton, H.C. and Appel, S.H., Biochemical studies of transneuronal degeneration: the effects of enucleation on the biochemical maturation of the chick optic rectum, Brain Res., 67 (1974) 289-306. 11 Clarke, P.G.H. and Whitteridge, D., The projection of the
We are greatly indebted to Drs. F. Bonhoeffer and A. Gierer for supporting this work and for critical discussions during the preparation of the manuscript. We thank Dr. J. Raper for a critical reading and Mrs. K. Ralinofsky for typing the manuscript.
retina including the 'red area' onto the optic tectum of the pigeon, Quart. J. Exp. Physiol., 61 (1976) 351-358. 12 Crossland, W.J., Cowan, W.M. and Rogers, J.P., Studies on the development of the chick optic tectum. IV. An autoradiographic study of the development of retinotectal connections, Brain Res., 91 (1975) 1-13. 13 DeLong, G.R. and Coulombre, A.J., Development of the retinotectal projection in the chick embryo, Exp. Neurol., 13 (1965) 351-363. 14 DeLong, G.R. and Coulombre, A.J., The specificity of retinotectal connections studied by retinal grafts onto the optic tectum in chick embryos, Dev. Biol., 16 (1967) 513-531. 15 Diitting, D., Gierer, A. and Hansmann, G., Self-renewal of stem cells and differentiation of nerve cells in the developing chick retina, Dev. Brain Res., 10 (1983) 21-32. 16 Ehrlich, D. and Mark, R., Topography of primary visual centres in the brain of the chick, Gallus gallus, J. Comp. Neurol., 223 (1984) 611-625. 17 Gierer, A., Development of projections between areas of the nervous system, Biol. Cybern., 42 (1981) 69-78. 18 Giorgi, P.P. and Van der Loos, H., Axons from eyes grafted in Xenopus can grow into the spinal cord and reach the optic tectum, Nature (London), 275 (1978) 746-748. 19 Goldberg, S., Studies on the mechanics of development of the visual pathways in the chick embryo, Dev. Biol., 36 (1974) 24-43. 20 Halfter, W., Claviez, M. and Schwarz, U., Preferential adhesion of tectal membranes to anterior embryonic chick retina neurites, Nature (London), 292 (1981) 67-70. 21 Hallas, B.H., Oblinger, M.M. and Das, G.D., Heterotopic neural transplants in the cerebellum of the rat: their afferents, Brain Res., 196 (1980) 242-246. 22 Hamburger, V. and Hamilton, H., A series of normal stages in the development of the chick embryo, J. Morphol., 88 (1951) 49-92. 23 Harvey, A.R., Golden, G.T. and Lund, R.D., Transplan-
179 tation of tectal tissue in rats, Exp. Brain Res., 47 (1982) 437-445. 24 Jaeger, C.B. and Lund, R.D., Transplantation of embryonic occipital cortex to the textal region of newborn rats: a light microscopic study of organization and connectivity of the transplants, J. Cornp. Neurol., 194 (1980) 571-597. 25 Kelly, J.P. and Cowan, W.M., Studies on the development of the chick optic system. III. Effects of early eye removal, Brain Res., 28 (1972) 263-288. 26 Krayanek, S., Structure and orientation of extracellular matrix in developing chick optic tectum, Anat. Rec., 197 (1980) 95-109. 27 Labbe, R., Firl, A., Jr., Mufson, E.J. and Stein, D.G., Fetal brain transplants: reduction of cognitive deficits in rats with frontal cortex lesions, Science, 221 (1983) 470-472. 28 LaVail, J.H. and Cowan, W.M., The development of the chick optic tectum. I. Normal morphology and cytoarchitectonic development, Brain Res., 28 (1971) 391-419. 29 Lund, R.D. and Hauschka, S.D., Transplanted neural tissue develops connections with host rat brain, Science, 193 (1976) 582-584. 30 Mathers, L.H. and Ostrach, L.H., A critical period in the development of tectal neurons in the chick as revealed by early enucleation, Brain Res., 170 (1979) 219-230. 3! McGraw, C.F. and McLaughlin, B.J., Fine structural studies of synaptogenesis in the superficial layers of the chick optic tectum, J. Neurocytol., 9 (1980) 79-93. 32 McLoon, S.C., Alterations in precision of the crossed retinotectal projection during chick development, Science, 215 (1982) 1418-1420. 33 McLoon, S.C. and Lund, R.D., Specific projections of reti-
na transplanted to rat brain, Exp. Brain Res., 45 (1980) 277-284. 34 McLoon, S.C. and Lund, R.D., Transient retinofugal pathways in the developing chick, Exp. Brain Res., 45 (1982) 277-284. 35 Rager, G., von Oeynhausen, B., Ingrowth and ramification of retinal fibers in the developing optic tectum of the chick embryo, Exp. Brain Res., 35 (1979) 213-217. 36 Sperry, R.W., Chemoaffinity in the orderly growth of fiber patterns and connections, Proc. Natl. Acad. Sci. U.S.A., 50 (1963) 703-710. 37 Thanos, S., Fluoreszenzmikroskopische Untersuchungen der Entwicklung und des Verlaufs Retinotektaler Fasern beim Hahner-Ernbryo, Ph.D. Thesis, 1983, TiJbingen. 38 Thanos, S. and Bonhoeffer, F., Investigations on development and topographic order of retinotectal axons: anterograde and retrograde staining of axons and perikarya with rhodamine in vivo, J. Comp. Neurol., 219 (1983) 420-430. 39 Thanos, S., Fujisawa, H. and Bonhoeffer, F., Elimination of ipsilateral retinotectal projections in monoophthalmic chick embryos, Neurosci. Left., 44 (1984) 142-148. 40 Thanos, S. and Bonhoeffer, F., Development of the transient ipsilateral retinotectal projection in the chick embryo: a numerical fluorescence-microscopic analysis, J. Comp. Neurol., 224 (1984) 407-414. 41 Thanos, S., Bonhoeffer, S. and Rutishauser, U., Fiber-fiber interaction and tectal cues influence the development of the chicken retinotectal projection, Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 1906-1910.
161
BRD 50515
Research Reports
Outgrowth and directional specificity of fibers from embryonic retinal transplants in the chick optic tectum Solon Thanos and Dieter Dtitting Max-Planck-lnstitut fiir Entwicklungsbiologie, Tiibingen (F.R. G.) (Accepted 9 September 1986)
Key words: Chick retinotectal system; Transplant; Neuronal specificity
Retinal pieces taken from known positions of 6-day chick embryos were vitally labeled with the fluorescent dye Rhodamine-B-isothiocyanate (RITC). They were then transferred onto the surfaces of optic tecta following early bilateral removal of the embryo's optic vesicles. One to 5 days after transplantation the tecta were fixed and transplants that issued fibers were examined on tectal wholemounts or were sectioned and viewed with a fluorescence microscope. Retinal fibers growing out from transplants on day E6 tecta showed a capacity for changing their initial outgrowth directions and for reorienting themselves towards their specific retinotopic projection area. Frequently, changes in growth direction appeared in a right-angled pattern. The capacity for turning was strongest for fibers of nasal retinal origin, less strong for fibers of temporal origin, and occurred rarely but unquantifiably in the case of fibers of ventral retinal origin. Fibers of all investigated retinal quadrants were found to reach their corresponding projection areas and to arborize there, that is, fibers of nasal retinal transplants in the posterior tectum, of temporal transplants in the anterior tectum, and of ventral transplants in the dorsal tectum. Furthermore, once in their target region, the fibers left the outer layer of the tectum and turned, again in right angles, to invade deeper layers. Capacity of fibers to turn towards their projection area was not observed for fibers issued from transplants placed on the tectum later than day E8. We suggest that there is a specific guidance of retinal axons on the tectum.
INTRODUCTION The t o p o g r a p h i c p r o j e c t i o n of the retina onto the v e r t e b r a t e tectum is organized roughly linearly and in a mirror-image fashion preserving the spatial relationships within the visual field. Possible mechanisms governing d e v e l o p m e n t of this orderly projection have been the subject of several investigations in the past. E x p e r i m e n t a l a p p r o a c h e s in fish and amphibian visual systems, from which most concepts are derived, were insufficient for the less accessible avian visual system. The few d a t a on the mechanisms involved in the d e v e l o p m e n t of retinotectal projection in birds derive from newly d e v e l o p e d n e u r o a n a t o m i cal labeling techniques. Besides being m a p p e d electrophysiologicallyn, the chick retinotectal system
was investigated neuroanatomically after partial ablation of the retina 1336'32, after a n t e r o g r a d e degeneration l°'16, by local staining procedures 37'38, and in transplantation experiments of embryonic retinal pieces onto the tectum 14"19. The results of the transplantation experiments of D e L o n g and C o u i o m b r e 14 and that of G o l d b e r g 19 display an inconsistency with respect to the specific guidance of outgrowing fibers. According to D e L o n g and C o u l o m b r e the quadrants of 4-day-old embryonic retina, transplanted onto 6day-old tectum, issued fibers that showed a specific orientation according to the functional retinotopicity. In a repetition of these experiments, G o l d b e r g found that outgrowing fibers followed the same straight courses taken by carbon particles implanted at c o m p a r a b l e positions in control embryos. In other
Correspondence: S. Thanos, Max-Planck-Institut fiir Entwicklungsbiologie, Abteilung Physikalische Biologie, Spemannstr. 35, D7400 Tfibingen, F.R.G. 0165-3806/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
162 words, retinal fibers did not show specific orientation toward their appropriate tectal quadrants. These reports brought some confusion into the discussion whether and to what degree growing optic axons are specifically guided. In the meantime transplantation approaches were reported from other systems, such as the Xenopus is and the rat visual system 23'29'33, the basal ganglia 1'5, cortex °'24'27, locus coeruleus 4 and cerebellum 21. In many of these experiments, fibers growing out from transplanted embryonic tissue showed specific recognition of the appropriate projection area, and in some cases formed significant functional connections within this area 4-6'23. New techniques have been added to the older histological procedures, in particular neuronal labeling with fluorescent dyes transported anterogradely 7'9'37-41. The present investigation analyzes the pathways of retinal fibers growing out on chick tectum. Small pieces of 6-day retina from known positions within the eye that show a capacity to issue fibers in vitro 7'9'20 w e r e vitally labeled with RITC and grafted directly onto the tectal surface (underneath the pial membrane) of 6-12-day-old anophthalmic host embryos. We were interested in the following questions: (1) Do outgrowing fibers from retinal transplants follow specific routes on the tectal surface? (2) Would retinal fibers arrive at their retinotopically correct projection field? (3) Would they recognize this field? (4) Is there a 'critical period' in embryonic development during which time fibers are guided specifically? MATERIALS AND METHODS
Host embryos Fertilized chicken eggs (Hiline Hisex, White Leghorn) were incubated at 38 °C for 38-43 h in a forced-draft incubator. The eggs were windowed after removal of 2 ml of albumen through a hole near the egg's tip. The embryos were staged and (if necessary) further incubated until stage 11 (12-13 somites) of the Hamburger-Hamilton scale = was reached. After injecting a small amount of Pelikan drawing ink (No. 17, black, India) beneath the head region of the embryo, the two optic vesicles were cut out by means of electrolytically sharpened tungsten needles (NS-86 tungsten wire, ~ 0.25 mm, obtained
from GTE Sylvania Licht G m b H , Bad Hersfeld, electropointed in a solution containing 71 g NaNO 2 and 34 g K O H in 100 ml distilled water, fine hooks produced by means of forceps). We chose the bilateral enucleation because it ensures that neither contranor ipsilateral retinal fibers invade the exposed tectal lobe, contrary to an exclusive removal of one eye vesicle which results in an enhanced ipsilateral retinal innervation of this tectum 34'39. The eggs were returned to the incubator until embryonic day E3 (stage 20). Then the contents of the eggs were transferred into plastic dishes for further development as described elsewhere 2'37'3s. In some embryos at stage 11, not only the optic vesicles were excised. In addition, we excised the right mesencephalic alar plate 3 and reinserted the rectangular piece of tissue after turning it by 180° . In this way, the antero-posterior and the dorso-ventral axes of the alar plate tissue were reversed. A normal right tectum results if the rotated piece is excised precisely along the dorsal midline and the right side of the neural tube, along the line that separates the upper alar plate from the lower basal plate. According to unpublished results (D. Dfitting), [3H]thymidine incorporation into operated tecta, sliced at days E6 to E8 into horizontal strips from the anterior to the posterior pole, revealed the normal gradient of proliferation with incorporation increasing towards the posterior pole.
Preparation of donor retina The left or right retina of a donor embryo (day E6) was dissected out completely and flattened on a piece of Sartorius membrane filter (type SM, black) in a serum-free Eagle's medium (without Ca and Mg) 9'19. The spread retina was incubated in 4 ml Eagle's medium containing about 150pl of an aqueous RITC solution (0.1-0.3%, Sigma R 1755) for 10 min at room temperature with occasional shaking. The filter with the retina was then washed by dipping it 4x into fresh Eagle's medium without RITC. The filter was transferred into Eagle's medium containing 10% fetal calf serum + Ca/Mg, and was incubated in a Heraeus cell culture incubator for 30-60 rain at 30 °C and 4% CO 2. A drawing of the retina by means of a camera lucida allowed us to determine the retinal axes (provided the optic fissure was left intact).
163
Transplantation procedure The chlorioallantoic membrane over the exposed tectum of the eyeless host embryo (in the plastic dish) was opened for the transplantation. A small piece of retinal tissue (0.02-0.03 mm 2) from a known position with respect to the dorsoventral and nasotemporal axes was transferred under the pial membrane of a defined position on the host tectum by means of a glass capillary. Transplantation success was controlled under an improvised fluorescence dissection scope (Zeiss). All manipulations were performed at 37 °C under sterile conditions.
Microscopy The transplanted embryos were further incubated for 1-5 days, the tecta dissected out, freed of all meninges, and fixed for at least 1 day in 4% paraformaldehyde in phosphate buffer (pH 7.2). The tecta were cut into a dorsal and a ventral half, and the two halves were mounted on a gelatine-coated glass slide in a solution of 2% n-propyl-gallate (Merck), 8% phosphate buffer (pH 7.2) and 90% glycerol. Mounted tectal halves were viewed with epifluorescence using a Zeiss universal microscope. The wavelength of the incident light was 520 nm, fluorescence was observed at 590 nm. Cryostat sections (20-40/~m) were prepared from representative specimens to control the position of transplants in relation to tectal depth and the layer(s) in which fibers from the transplant grow out preferentially. We examined the host tectal surfaces (days E 4 El4) in the absence of any visual input following bilateral ablation of eye primordia at early stages by scanning electron microscopy (SEM). The tecta were fixed overnight in fixative containing 2.5% glutaraldehyde in 0.1 M phosphate buffer at pH 7.2, postfixed in 1% buffered osmium tetroxide for 30 min and dehydrated in a series of ascending ethanol concentrations. They were then treated according to Krayanek 26. The results of these controls confirmed Krayanek's studies according to which the tectal surface does not consist of an oriented extracellular matrix 26. These experiments are not further described in the present report. In the same way we prepared representative tecta carrying a retinal transplant to control the graft's position on the tectal surface.
RESULTS The transplantations performed between days E6 and E12 are divided into 3 groups. Group 1 contains transplantations at days E 6 - E 7 with postoperative survival of 1-3 days. Out of more than 80 transplanted embryos, 49 had a visible transplant with fibers grown after that time. In this group we analysed the direction of initial growth as well as corrective phenomena. Group 2 contains successful transplantations (28 cases) at day E6 with postoperative survival of 3.5-5 days. In this group of transplants we investigated the ability of fibers to find their normal termination field and their capacity to recognize it. Group 3 summarizes all successful experiments (8 cases) performed between days E8 and E12. In this group we investigated the ability of axons to recognize tectal cues after day E8. The postoperative survival varied between i and 4 days. Known in our system were (1) the positional origin of the retinal graft, (2) the position on the tectum which received the transplant, and (3) the approximate size of the transplant and the age of donor and recipient. Unknown was the orientation of the transplant on the tectum with respect to its original retinal axes. The following properties of outgrowing axons were scored quantitatively as far as possible, otherwise qualitatively, using approximate number of outgrowing fibers, their lengths (in particular those of longest fibers), direction of initial outgrowth, depth of graft in relation to tectal tissue as revealed by microscopic focusing, fasciculation, capacity of branching, correction of initial routes and finally the fibers' fate within the projection area and within inappropriate tectal areas. Independent of their preferential growth direction, the number of outgrowing fibers varied between a few and hundreds of axons depending on the size of the transplant, although we tried to transfer pieces of roughly the same size. Thus, in some cases we observed only a few outgrowing fibers, in others not a single one. The quantity of fibers seemed to be independent of the recipient's age, but dependent on the stage of retinal differentiation which is not uniformly advanced at day E6. The nasoventral retinal quadrant differentiates with a time delay of 12-24 h in relation to temporal and dorsal retinal regions 15.
164 According to earlier in vitro investigations, the optimal age to prepare retinal explants is day E6 (refs. 7, 9 and 20). For this reason we transplanted day E6 retinal tissue. Fibers growing out from transplants of various retinal regions did not differ in their morphology. They did not grow out preferentially as a single fascicle, as described for day E4 grafts placed mainly on top of the pial membrane ~4'19. In our experiments, fibers grew out of the transplant in more than one direction, often radially in multiple directions (Fig. 1). Two types of fibers could be identified: (1) short fibers (the majority) with lengths up to 0.5 mm growing randomly around the transplant, forming fascicles up to 10 fibers and branching shortly after outgrowth; and (2) straight growing axons, thin, non-branching, non-fasciculating with lengths up to 4 ram. The long fibers are almost certainly ganglion cell axons, the
short fibers dendrites or processes from ganglion cells and other types of retinal cells. With retinal explants of the same embryonic age, Halfter et al. z~ demonstrated by means of retrograde H R P transport that the long fibers in in vitro preparations derive from cells in the ganglion cell layer. In our analysis we counted only fibers (axons) longer than 0.5 mm which did not remain near the graft.
Initial outgrowth of fibers The earliest outgrowing fibers could be detected about 12 h after transplantation. The direction of initial outgrowth depended on the age of the host tecturn, that is, the younger the tectal tissue, the higher the fraction of fibers growing in dorsal or ventral directions. In tecta older than day E8 the fibers do not seem to have a preference for the dorsal or ventral direction, they grow out randomly in a radial fashion.
Fig. 1. Fluorescence micrograph showing a fluorescing nasal retinal piece grafted on the anterior tectum at day E8 and observed at day El0. Position of the transplant is shown on the schematic embryo head in the upper left corner. Fibers grow out almost radially without a clear preference. Scale bar = 100/~m.
165 TABLE I Quantification of the data obtained from grafting experiments on day E6 with short postsurgical survival The first line shows the retinal origin of the grafts, the second line shows the tectal position to which the retinal grafts were transferred, with numbers of experiments in parentheses. The lines 'Initial outgrowth direction' give the numbers of experiments in which long axons (see text) grew out initially in anterior (A), posterior (P), dorsal (D), ventral (V) or radially in random (0) directions. The last lines analyze the final directions into which fibers preferentially grew after correcting their route. Details are given in the text. Retinal origin
N (22)
Tectal position Initial outgrowth direction Final directional preference Yes
A(13) A(0) P(0) O(13)
P(9) A(0) P(0) 0(9)
P (18) A(3) P(0) O(15)
A(8) A(5) P(0) 0(3)
D(3) V(0) D(0) 0(3)
V(13) V(6) O(0) 0(7)
V(18) V(7) D(4) 0(7)
D(1) V(0) D(1) 0(0)
P(9) A(0) 0(4)
A(0) P(0) 0(9)
A(13) P(0) 0(5)
P (0) A(5) 0(3)
V(0) D(0) 0(3)
D(0) V(6) 0(7)
D(6) V(7) 0(5)
V(0) D(1) 0(0)
No
T (26)
D (16)
V (19)
O
/
I
Fig. 2. Fluorescence photomicrograph (a), accompanying drawing (b) and schematic drawing of head (c) showing the position of a nasal transplant (Exp. No. HTr13) on the anterior tectum as well as the direction and corrective behavior of outgrowing fibers. Axons, initially oriented in ventral direction, turn in posterior direction by nearly right-angled routes. Some of them are marked by arrows in drawing (b). A, anterior; D, dorsal; P, posterior; Tel, forebrain; Tec, tectum; Tr, transplant. Asterisk in c marks the position of absent left eye. Scale bars: 100/~m for (a) and (b), 1 mm for (c).
166 Fig, 1 p r e s e n t s such an e x p e r i m e n t at days E 8 - E 1 0 .
o b s e r v e d a differential b e h a v i o r of fibers e m e r g i n g f r o m the graft.
Corrective behavior of fibers D e p e n d i n g on the t r a n s p l a n t ' s retinal origin, we
Nasal transplants'.
N i n e nasal transplants placed at
day E6 o n t o the p o s t e r i o r t e c t u m ( p r o j e c t i o n area) is-
Fig. 3. Fluorescence photomicrographs showing a nasal transplant (Expt. HTr 143) on the anterior tectum, a: fibers leaving the transplant are oriented either directly to the posterior part of the tectum or in dorsal direction. From this population of dorsally oriented axons, several turn, by sharp right angles (small arrows), to posterior direction, b: higher-power magnification of the region marked by asterisk in (a) and (b). c: drawing showing the position of the transplant on the anterior rectum as well as the direction of outgrowth and correction as described above. Abbreviations as in Fig. 2. Scale bars: 100urn for (a) and (b), ! mm for (cL
167 sued longer fibers either in ventral and dorsal directions or randomly without any preferred direction (Table I, column N/P). In contrast, fibers from 9 of 13 nasal grafts onto the anterior tectum (Table I, column N/A) issued fibers that showed a strong capacity to turn in posterior direction (towards their normal projection area). These fibers often deviated from their initial direction of growth by making nearly right-angled turns (Figs. 2, 3, and see Fig. 8B). In 4 cases, the outgrowth was not oriented specifically. After turning, axons grew straight to the posterior tectum without straying or searching on the tectal surface. Nasal fibers turning away from their projection area were observed rarely. Within the outgrowing fiber population the proportion of turning axons varied between a few percent and about 50% of the outgrowing fibers. Figs. 2 and 3 represent two examples of high-rate correction. The grafting of nasal retinal pieces (7 cases) onto the anterior part of tecta, derived from 180° rotations of the right mesencephalic alar plate at stage 11, did not reveal any differences in the behavior of the outgrowing fibers (data not shown), compared to those
described above. This suggests that the embryological axes are irreversibly fixed some time after stage 11. Transplants originating from temporal retina, which would project onto the anterior tectum, issued fibers which showed a specific reorientation and corrective behavior opposite to those shown for nasal fibers. Out of 8 temporal transplants onto the anterior tectum (the projection area) 5 issued fibers which remained within the anterior tectal half and preferred to grow into regions more anterior than the graft's position. The remaining 3 grafts issued axons in dorsal and ventral directions (Table I, column T/A). Out of 18 temporal transplants on the posterior tectum, 3 issued fibers which grew directly towards the anterior tectum. Of the other 15, which issued axons in dorsal and ventral directions or radially, 10 corrected their initial outgrowth direction towards their anterior target (Table I, column T/P). Figs. 4 - 6 show 3 temporal transplants on the posterior tectum which demonstrate the ability of some axons to turn abruptly towards their target. The proportion of turning fibers in each transplant varies substantially. For example, in
Fig. 4. Fluorescence photomicrograph (a) and accompanying schematic drawing of the recipient head (b) showing one small temporal transplant (Expt. HTr 122) on the posterior tectum with only one outgrowing fiber which turns to anterior direction. The growth cone of the axon is on the left side of the figure. Abbreviations as in Fig. 2. Scale bars: 100ktm for (a) and 1 mm for (b).
168
Fig. 5. High-power fluorescence photomicrograph (a) and schematic head (b) showing several fascicles (big arrows) growing out from a temporal transplant on the posterior tectum (Expt. HTr 99). Single fibers (small arrows) leave the initial ventral growth direction (shown in (b)) and grow in anterior direction. Scale bars: 100 ftm for (a), I mm for (b).
169
6
b
Fig. 6. The path of fibers growing out from one temporal transplant (Expt. HTr 108) on the posterior dorsal tectum, a: fluorescence photomicrograph showing the more or less parallel growth of fibers in dorsal direction (drawn in c). b: drawing from photomicrographic prints: the bundle grows over a substantial distance (about 1.5 mm) straight in dorsal direction before several axons leave and turn in anterior direction (arrows). Scale bar = 100 um.
Fig. 4 the only outgrowing axon turns, in Fig. 5 several axons turn in anterior direction. In comparison to nasal transplants the total number of temporal axons turning to anterior direction was lower. Fig. 8A presents schematically 7 transplantations of temporal retinal pieces onto the posterior tectum. In 9 experiments (5 nasal and 4 temporal retinal grafts) we investigated the trajectory of axons growing out of transplants placed in the middle of the ventral tectal half, in a position that bisects the anteroposterior axis. Surprisingly, none of these transplants issued axons which corrected their path. Axons grew out radially or in dorsoventral directions up to the tectal border where they turned mainly in posterior directions. Table II summarizes the results of these experiments showing the tectal region on which grafts were
placed, as well as the directions of outgrowth. Similar observations (with respect to initial outgrowth and correction capacity) have been made for axons growing out of either ventral or dorsal retinal transplants which in most cases were placed in the middle of the anteroposterior axis, but either on the ventral or on the dorsal tectal half (see below).
Dorsal and ventral retinal pieces were usually grafted onto the ventral half of the rectum, because the dorsal rectum was less accessible at days E 6 - E 7 . Out of 13 dorsal transplants onto the ventral tectum, 6 issued fibers oriented radially which remained mainly within this tectal region (Table I, column D/V, and see Fig. 8D). In t h e remaining 7 transplants, fibers grew out randomly. None of them grew preferentially to the dorsal tectum. In 3 cases in which dorsal transplants were successfully placed on
170
TABLE II
Grafting experiments performed into the middle of the A/P-axis Five nasal and 4 temporal retinal grafts were positioned within the region marked by the black circle on the schematic tectal surface. Thick arrows towards the dorsal and ventral direction show the initial orientation of the majority of axons. Once arriving at the dorsal or ventral tectal border, most of them turned in a posterior, some of them in an anterior direction (two arrows vs one). A restricted radial outgrowth from the area of transplantation is indicated by small arrows on the black circle. The last line negates the question of correction lor both groups of grafts. RETINAL ORIGIN
NASAL(5)
TEMPORAL(4) D
GRAFTING POSIT.
V CORRECTION ?
NO
NO
the dorsal tectal half, fibers grew out without a preferential direction (Table I, column D/D). Transplants from the ventral retina showed a similar pattern of fiber outgrowth c o m p a r e d with dorsal transplants (Fig. 8C). Their fibers p r e f e r r e d to grow out radially. Out of 18 ventral transplants positioned onto the ventral tectum, 4 issued axons in a dorsal, 7 in a ventral direction and 7 did not prefer any direction (Table 1, column V/V). Fiber turning, as described for nasal and temporal fibers, was o b s e r v e d in two of the ventral transplants; if it occurred, fibers were directed towards their corresponding projection area (Table I, column V/V). Fig. 7 shows an example of one of these experiments with ventral fibers turning in dorsal direction. Drawings of fiber pathways from other transplants are shown in Fig. 8C, D. Only one ventral transplant was successfully placed onto the dorsal tectum. Its fibers r e m a i n e d within the dorsal tectal half (Table I, column V/D).
Fate o f fibers in their projection areas To answer the question whether fibers arrive in their projection area, and to examine the fate of fibers after they arrived there, we cultivated 28 successfully transplanted embryos for longer periods of time, i.e. 3 . 5 - 5 days after transplantation at day E6. Arborization was a first criterion to define a fiber as
'terminating'. A second criterion was the ability of fibers to leave the layer of initial outgrowth to invade the underlying tectal layers, as they do in normal development of the projection 2~-35. One difficulty in evaluating such experiments is the e n o r m o u s length of axons (sometimes 3 - 4 mm) and poor fluorescence intensity, particularly in the middle third of the axon's length. Axons of all retinal regions (ventral, nasal and temporal) were able to reach their specific projection areas. The higher the degree of pathway correction, the more fibers from the transplants were later found in the a p p r o p r i a t e tectal projection area. That is, in most cases with nasal transplants on the anterior tecturn we found axons terminating on the posterior tecturn. Fig. 9 shows nasal axons arborizing on the posterior tectum 4 days after transplantation. Fibers of temporal transplants placed onto the posterior tecturn were found less often within the anterior rectum (data not shown). Axons of ventral transplants placed onto the ventral tectum rarely t e r m i n a t e d on the dorsal rectum. Fig. 10 presents a rare case in which several ventral axons stop and branch on the dorsal rectum. The routes of these terminating axons could not be followed over their entire length from the transplant (big arrow in Fig. 10c) to the dorsal rectum (small arrow). Only 3 dorsal transplants were successfully placed onto the dorsal tectum; outgrowing fibers could not be followed to the ventral tectum. Fig. 11 shows a cryostat coronal section through a tectum with fluorescent fiber bundles growing from a nasal transplant within the most superficial tectal layer. The relative positions of fibers correspond to the normal stratum opticum which is absent in these experiments because of early eye removal. In another experiment with a nasal transplant, fibers leaving this layer can be seen in Fig. 12. Note here the sharp fiber turns which are similar to those shown for normal retinotectal axons 28'35. With increasing time after transplantation, an increasing number of degenerating fibers is found on the tectum. W e defined as degenerating those axons which lost their h o m o g e n e o u s l y distributed R I T C fluorescence along their length and possessed shrinking growth cones with retracted processes. Fibers in advanced stages of d e g e n e r a t i o n showed decomposition with irregularly distributed fluorescent varicoses akmg their length and loss of typical growth cone
171
Fig. 7. Trajectories of two turning axons from a ventral transplant (Expt. HTr 190) on the ventral tectum (position of the transplant is shown in the accompanying head drawing and in Fig. 8D, its retinal origin (f) is shown in Fig. 8C. The initial posteroventral direction changes to a clear dorsal one. Scale bars: 100pm for (a) and 1 mm for (b). morphology (Thanos et al., unpublished observations). At 4 - 5 days after transplantation, the time of longest observation, healthy axons have been found almost exclusively within their projection area. Usually the RITC labeling was not strong enough to reveal the whole pathway of these axons. In other tectal areas fibers were already degenerated (data not shown).
Transplantation on days E8-E12 Eight successful transplantations of nasal retina (day E6) onto anterior tectum of days E 8 - E 1 0 revealed a different picture compared with grafting onto tecta of days E6-E7. Fibers growing out after day E8 did not show a specific orientation on the tectum, nor a capacity for correcting initial routes. The observed radial or at least multidirectional outgrowth lacked any preferential orientation. Fig. 1
presents a nasal transplant grafted at day E8 and incubated up to day El0. Transplantations at day E l 0 were terminated not later than day El2, since scanning electron microscope studies of uninnervated tectal surfaces (days E4-E14; Diitting and Thanos, unpublished results) showed significant degenerative changes from day El3 onwards. The analysis of younger uninnervated tectal surfaces by scanning electron microscopy did not reveal any preformed channels that might have been used by the outgrowing fibers. DISCUSSION Embryonic retinal tissue of defined positional origin, labeled with the fluorescent dye RITC, was grafted into defined positions of the tectal surface of eyeless embryos that did not receive any prior input
a
\.\..
J4_~
4
~ ~ H T r ~
,2) HTr28
i
HTr 96
HTr 78
',\
,,
/
/
HTr123
RETINA
HTr14
-
RETINA
HTr 117
i O
~, //' . / ~'HTr 13 e
tel
rpos
ven.
HTr 120
HTr 119
I
Fig. 8. Schematic drawings summarizing the results of selected experiments of the 4 groups of transplantations mentioned. A: 7 examples of temporal retinal transplants on the posterior tectum, a - g mark the origin of the transferred retinal pieces in the accompanying schematic retina. Black arrows (straight and curved) indicate the directions of outgrowth and correction, respectively. Big arrows (not in all drawings) show preferential or fasciculated outgrowth. Asterisks indicate the approximate projection area. B: 5 examples of nasal transplants located on the anterior tectum. Meaning of arrows and small letters as in A. In A and B the transplants were placed close to the bisection line (see E). C: 3 ventral retinal pieces (d, g, h) grafted on the ventral tectal half. D: two dorsal grafts onto dorsal tectum (c, e) and two dorsal grafts (b, f) onto the ventral tectum. Retinal origin as marked on C). E: the orientation of the mounted tectal halves (left) and spread retinas (right) for all groups. The tectum was bisected along the antero-posterior axis into a dorsal half (left) and a ventral half (right). ant, anterior; dor, dorsal; nas, nasal: pos, posterior; tern, temporal; ven, ventral.
173
Fig. 9. Fluorescence photomicrographs showing arborizing nasal axons within the posterior tectum (compare with Fig. 8B). a: one axon that just started to arborize, b and c: two clearly arborizing axons as photographed on a tectal w h o l e - m o u n t with microscopic focusing. Note the presence of a growth cone with various processes (arrows) on each of the axonal branches, Scale bar = 100 p m .
174
~.
Fig. 10. Fluorescence photomicrograph (a) and accompanying drawing (b) shows axons from ventral transplant on ventral tectum arborizing within the dorsal tectum (position shown in c). Several axons reaching this region on straight routes (arrows in a and b) seem to stop in a relatively small region (about 200/~m diameter), and arborize there. Abbreviations, see Fig. 2. Scale bars: 100 btm for (a) and (b), 1 mm for (c).
175
Fig. 11. a: bright-light photomicrograph showing a cryostat section through a recipient tectum carrying a nasal retinal transplant on the anterior tectum (not visible on this section), b: fluorescence photomicrograph of the same section showing the position of the fluorescent fibers (arrows) in a distance of about 2 mm posterior to the grafts. This position of fibers corresponds to the stratum opticum in normal embryos. SC, stratum centrale; SP, stratum periventriculare; SGFS, stratum griseum et fibrosum superficiale; Vent, ventricle. Scale bar = 50am.
176
@ Q
•
•
Q O
O
t
e
O
O
o
0
Fig. 12. High-powerfluorescence photomicrograph(a) and accompanyingdrawing (b) showing 4 fluorescing axons within the outer layer (marked with dotted line in b) of the tectum. Two of these fibers turn sharply (arrows) towards deeper tectal regions (bottom of the pictures). On the right edge one fiber seems to start branching. Scale bar = 50 pm. of optic fibers. Retinal axons issued from the grafts traversed the tectal surface, and, depending on their origin, were able to correct an initially 'wrong' direction of outgrowth. In many cases the fibers finally reached their correct retinotopic projection area as indicated by strong arborization and invasion of deeper tectal layers. Usually we grafted day E6 retinal tissue onto day E 6 - E 7 tectal surfaces and terminated the transplanted embryos after 1-3 days of further incubation. Between days E6 and El2 the rectum matures in the anterior to posterior direction 12'25'28'3°'31which implies that grafts onto the thicker anterior wall of the tectum are easier to perform than those onto the thinner posterior tectal roof. Grafts of nasal retina upon the anterior tectum, that send their fibers to the posterior tectum, have a greater chance of suc-
cess than those of temporal retina onto the posterior tectum, the fibers of which are destined for the anterior rectum. In vitro experiments 2° showed that day E6 retinal tissue is suited particularly well for an optimal outgrowth of fibers. The fluorescent vital labeling with RITC provides advantages and disadvantages 37'3s. The dye accumulates near the growth cone of a fiber and is depleted from the middle of long axons. This can make tracking fibers over long distances difficult. At high magnification, many details of growth cones, terminal arbors and branches along the axon can be seen. The staining, unfortunately, is not bright enough to allow a view of complete projection fields at lower magnification. The poor quality of some RITC batches and a sometimes observed high background fluorescence impede the fiber tracing. At high magnification it is
177 almost impossible to focus on a longer section of the axon, which means that photographs have to be made on different levels of the tectal surface. Likely due to leakage of the dye from the axon, the technique does not allow observations exceeding 4-5 days after grafting. There are two earlier reports that described transplantations of chick embryonic retinal tissue onto the tectum 14'19. Their results are contradictory. With a new technique we tried to get more information about the mechanisms guiding retinal axons growing on the tectum. The following technical modifications improved the transplantation system. (a) Chick embryos growing in plastic dishes 2'37'38 made the tectal surface more easily accessible to local manipulations. In the old reports 14'19, transplantations were performed mainly within the 'central' third of the tectal surface which is the only region well accessible in embi'yos developing in ovo. (b) Transplants were placed (with fine glass capillaries) directly beneath the pial membrane on the tectal surface on which retinal fibers normally grow. According to the previous reports, outgrowing fibers traversed an empty space as a well defined fascicle before penetrating the pia to get access to the tectal surface 14. This implies that in these experiments transplants were placed on top of the pia. In our experiments, fibers, having direct access to the tectal surface, grew out in many directions from the graft and did not form one single fascicle as reported earlier 14'19. Our results are in agreement with the supposed existence of specific guidance cues on the tectum 14. In addition they provide information about the mode of correction. Goldberg's results, which show only fibers running to the posterior tectum, could be explained by the appearance of an ipsilateral retinotectal projection during development 34, 38,39. This is more prominent after unilateral enucleation during early development 39. To avoid any visual innervation of the tectum, we removed both eye cups during early development. Analysis of fiber routes with respect to the transplant's origin in the retina and to the transplant's location on the tectum revealed a differential behavior of outgrowing fibers. Fibers of nasal and temporal origin can clearly recognize their position along the anteroposterior tectal axis. As a consequence, they are able to correct the direction of their outgrowth towards their appropriate projection area. Nasal fibers
seem to have the strongest capacity to correct their routes. A difficulty in the interpretation of these resuits is the observation that neither nasal nor temporal fibers orient themselves retinotopically correctly if the corresponding transplant was placed on the ventral tectum half-way between the anterior and posterior margin. One explanation might be found in the specific distribution of guiding cues on the rectum. Fibers of dorsal and ventral origin showed a weak capacity of correcting their initial outgrowth direction. Most of them traversed long distances on the tectal surface randomly; sharp, right-angled turns were rarely found in these cases. As data with longer survival showed, fibers reached their normal projection area in all groups of experiments. Once this goal was achieved, fibers arborized extensively within this area, and in some cases (shown by sectioning the tectum) left the superficial layer of growth to invade underlying tectal layers in which synaptogenesis takes place in normal development 12'31. Synaptogenesis cannot be observed with the techniques used in our experiments. There seems to be a contradiction between the observed low capacity for correction along the dorsoventral axis and the fact that after long-time survival some axons terminate in appropriate areas along this axis. This could be selective survival of those axons which arrived in the proper target region rather than guided outgrowth. The different routes of these fibers were only partially visible by the technique used in our experiments. Outgrowing fibers communicate with unknown tectal elements, and might orient themselves according to the information they obtain 17'36. The presence of position-specific information was clearly shown in the developing chick visual system 8,9m. Fiber tracing by anterogradely transported RITC in vivo was combined with immunological perturbation of retinal fasciculation by means of antibodies against the neural cell-adhesion molecule (N-CAM). This provided information about the corrective behavior of fibers41. Antibodies injected into the embryonic eye at day E4 appeared to prevent proper fasciculation of axons near the fissure. The resulting misrouted fibers remained ectopic up to the tectum and there, at first, they grew parallel to the normally routed fibers. Then several axons (particularly those growing in close contact to the tectal surface) turned towards
178 their target area. This turning occurred at the correct
which we grafted nasal retina between days E8 and
position along the growth axis (anteroposterior) and
El(). Transplantations of temporal retina onto poste-
resulted in a correction of the dorsoventral position
rior rectum of days E 8 - E 1 0 (two cases) did not re-
of axons. Similar results were obtained from experi-
veal a clear result and therefore cannot be evaluated
ments in which the pathway of first fibers invading a
presently. According to our results, positional cues
given part of the tectum became deflected by an
on the u n i n n e r v a t e d tectal surface might guide the
obstacle during development (Thanos and Bonhoef-
first ingrowing fibers of a normal projection. Proba-
fer, in press). Deflected fibers soon corrected their
bly these cues are not available to fibers arriving later in development, which could be guided to the re-
direction of growth and curved smoothly towards their target, suggesting that they sense directional cues on the virgin tectum with respect to both the an-
maining free synaptic space by the projections established earlier in development.
teroposterior and dorsoventral axes. This is confirmed by the results of the present study.
ACKNOWLEDGEMENTS
According to the results reported here, positional cues are available to retinal fibers on days E 6 - E 7 , when the normal projection has started to invade the anterior tectum. From late day E8 onwards these cues seem to be unavailable to fibers growing from the transplants, at least in the anterior tectum onto
REFERENCES 1 Aguayo, A.J., BjOrklund, A., Stenevi, U. and Carlstedt, T., Fetal mesencephalic neurons survive and extend long axons across peripheral nervous system grafts inserted into the adult rat striatum, Neurosci. Lett., 45 (1984) 53-58. 2 Auerbach, R., Kubai, L., Kinghton, D. and Folkman, J., A simple procedure for long-term cultivation of chicken embryos, Dev. Biol., 41 (1974) 391-394. 3 Birge, W., An analysis of differentiation and regulation in the mesencephalon of the chick embryo, Am. J. Anat., 104 (1959) 431-463. 4 Bj6rklund, A., Segal, M. and Stenevi, U., Functional reinnervation of rat hippocampus by locus coeruleus implants, Brain Res., 170 (1979) 409-426. 5 BjOrklund, A., Schmidt, R.H. and Stenevi, U., Functional reinnervation of theneostriatum in the adult rat by use of intraparenchymal grafting of dissociated cell suspensions from the substantia nigra, Cell Tissue Res., 212 (1980) 39-45. 6 Bj6rklund, A. and Stenevi, U., Intracerebral neural implants: neuronal replacement and reconstruction of damaged circuitries, Annu. Rev. Neurosci., 7 (1984) 279-308. 7 Bonhoeffer, F. and Huf, J., Recognition of cell types by axonal growth cones in vitro, Nature (London), 288 (1980) 162-164. 8 Bonhoeffer, F. and Gierer, A., How do retinal axons find their targets on the tectum? TINS 7 (1984) 378-381. 9 Bonhoeffer, F. and Huf, J., In vitro experiments on axon guidance demonstrating an anterior-posterior gradient on the tectum, EMBO J., 1 (1982) 427-431. 10 Carton, H.C. and Appel, S.H., Biochemical studies of transneuronal degeneration: the effects of enucleation on the biochemical maturation of the chick optic rectum, Brain Res., 67 (1974) 289-306. 11 Clarke, P.G.H. and Whitteridge, D., The projection of the
We are greatly indebted to Drs. F. Bonhoeffer and A. Gierer for supporting this work and for critical discussions during the preparation of the manuscript. We thank Dr. J. Raper for a critical reading and Mrs. K. Ralinofsky for typing the manuscript.
retina including the 'red area' onto the optic tectum of the pigeon, Quart. J. Exp. Physiol., 61 (1976) 351-358. 12 Crossland, W.J., Cowan, W.M. and Rogers, J.P., Studies on the development of the chick optic tectum. IV. An autoradiographic study of the development of retinotectal connections, Brain Res., 91 (1975) 1-13. 13 DeLong, G.R. and Coulombre, A.J., Development of the retinotectal projection in the chick embryo, Exp. Neurol., 13 (1965) 351-363. 14 DeLong, G.R. and Coulombre, A.J., The specificity of retinotectal connections studied by retinal grafts onto the optic tectum in chick embryos, Dev. Biol., 16 (1967) 513-531. 15 Diitting, D., Gierer, A. and Hansmann, G., Self-renewal of stem cells and differentiation of nerve cells in the developing chick retina, Dev. Brain Res., 10 (1983) 21-32. 16 Ehrlich, D. and Mark, R., Topography of primary visual centres in the brain of the chick, Gallus gallus, J. Comp. Neurol., 223 (1984) 611-625. 17 Gierer, A., Development of projections between areas of the nervous system, Biol. Cybern., 42 (1981) 69-78. 18 Giorgi, P.P. and Van der Loos, H., Axons from eyes grafted in Xenopus can grow into the spinal cord and reach the optic tectum, Nature (London), 275 (1978) 746-748. 19 Goldberg, S., Studies on the mechanics of development of the visual pathways in the chick embryo, Dev. Biol., 36 (1974) 24-43. 20 Halfter, W., Claviez, M. and Schwarz, U., Preferential adhesion of tectal membranes to anterior embryonic chick retina neurites, Nature (London), 292 (1981) 67-70. 21 Hallas, B.H., Oblinger, M.M. and Das, G.D., Heterotopic neural transplants in the cerebellum of the rat: their afferents, Brain Res., 196 (1980) 242-246. 22 Hamburger, V. and Hamilton, H., A series of normal stages in the development of the chick embryo, J. Morphol., 88 (1951) 49-92. 23 Harvey, A.R., Golden, G.T. and Lund, R.D., Transplan-
179 tation of tectal tissue in rats, Exp. Brain Res., 47 (1982) 437-445. 24 Jaeger, C.B. and Lund, R.D., Transplantation of embryonic occipital cortex to the textal region of newborn rats: a light microscopic study of organization and connectivity of the transplants, J. Cornp. Neurol., 194 (1980) 571-597. 25 Kelly, J.P. and Cowan, W.M., Studies on the development of the chick optic system. III. Effects of early eye removal, Brain Res., 28 (1972) 263-288. 26 Krayanek, S., Structure and orientation of extracellular matrix in developing chick optic tectum, Anat. Rec., 197 (1980) 95-109. 27 Labbe, R., Firl, A., Jr., Mufson, E.J. and Stein, D.G., Fetal brain transplants: reduction of cognitive deficits in rats with frontal cortex lesions, Science, 221 (1983) 470-472. 28 LaVail, J.H. and Cowan, W.M., The development of the chick optic tectum. I. Normal morphology and cytoarchitectonic development, Brain Res., 28 (1971) 391-419. 29 Lund, R.D. and Hauschka, S.D., Transplanted neural tissue develops connections with host rat brain, Science, 193 (1976) 582-584. 30 Mathers, L.H. and Ostrach, L.H., A critical period in the development of tectal neurons in the chick as revealed by early enucleation, Brain Res., 170 (1979) 219-230. 3! McGraw, C.F. and McLaughlin, B.J., Fine structural studies of synaptogenesis in the superficial layers of the chick optic tectum, J. Neurocytol., 9 (1980) 79-93. 32 McLoon, S.C., Alterations in precision of the crossed retinotectal projection during chick development, Science, 215 (1982) 1418-1420. 33 McLoon, S.C. and Lund, R.D., Specific projections of reti-
na transplanted to rat brain, Exp. Brain Res., 45 (1980) 277-284. 34 McLoon, S.C. and Lund, R.D., Transient retinofugal pathways in the developing chick, Exp. Brain Res., 45 (1982) 277-284. 35 Rager, G., von Oeynhausen, B., Ingrowth and ramification of retinal fibers in the developing optic tectum of the chick embryo, Exp. Brain Res., 35 (1979) 213-217. 36 Sperry, R.W., Chemoaffinity in the orderly growth of fiber patterns and connections, Proc. Natl. Acad. Sci. U.S.A., 50 (1963) 703-710. 37 Thanos, S., Fluoreszenzmikroskopische Untersuchungen der Entwicklung und des Verlaufs Retinotektaler Fasern beim Hahner-Ernbryo, Ph.D. Thesis, 1983, TiJbingen. 38 Thanos, S. and Bonhoeffer, F., Investigations on development and topographic order of retinotectal axons: anterograde and retrograde staining of axons and perikarya with rhodamine in vivo, J. Comp. Neurol., 219 (1983) 420-430. 39 Thanos, S., Fujisawa, H. and Bonhoeffer, F., Elimination of ipsilateral retinotectal projections in monoophthalmic chick embryos, Neurosci. Left., 44 (1984) 142-148. 40 Thanos, S. and Bonhoeffer, F., Development of the transient ipsilateral retinotectal projection in the chick embryo: a numerical fluorescence-microscopic analysis, J. Comp. Neurol., 224 (1984) 407-414. 41 Thanos, S., Bonhoeffer, S. and Rutishauser, U., Fiber-fiber interaction and tectal cues influence the development of the chicken retinotectal projection, Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 1906-1910.