- Email: [email protected]
Optic nerve regeneration by peripheral nerve transplant
Neuroscience Research, Suppl. 13 (1990) $24-$30 Elsevier Scientific Publishers Ireland Ltd.
$24
NEURES 00356
Optic nerve regeneration by peripheral nerve transplant Yutaka Fukuda
1, Hitoshi
Sasaki 3, Eijiro A d a c h i 2, Tetsu I n o u e 4 a n d K a t s u k o Morigiwa 1
1 Department of Physiology and 2 Department of Anatomy, Osaka University Medical School, s Department of Physiology, Hyogo Medical College and 4 Section of Ophthalmology, Izumiohtsu City Hospital, Osaka (Japan) Key words: Optic nerve; Regeneration; Retinal ganglion cells; Transplantation; Basal lamina; Schwann cells; Laminin; Adult mammals
SUMMARY We studied the morphology of regenerated retinal ganglion cells and their axons in adult rodents after axotomy and autologous transplantation of the sciatic nerve. Regenerated ganglion cells, backlabeled with rhodamine dextran, were of similar size to or larger than those of intact cells in control animals. Dendrites and occasionally axons as well showed abnormal morphologies in most cells, though some cells appeared quite normal. Cross-sections of the regenerated axons, observed by electron microscopy, were always attached to either the Schwann cell cytoplasm or the basal lamina. The immunoreactive structures to anti-laminin antibody were quite irregular in the cross-sectioned graft and, compared with those of the intact sciatic nerve, they were generally smaller. Their appearance closely resembled that of the basal lamina in the graft observed by elecron microscopy. These observations, taken together, suggest that the laminin-rich basal laminae of Schwann cells are essentially important for the regeneration of retinal axons in adult rodents.
INTRODUCTION
Since the pioneering work by Aguayo and his co-workers, it is now known that, once damaged, CNS neurons can regenerate their axons when the peripheral nerve has been transplanted. For example, in the rat and hamsters transected optic nerve fibers can regrow lengthy axons in the peripheral nerve graft and when properly guided their terminals make normal synapses with target cells in the superior colliculus 1,7,8. Furthermore, these synapses have recently been shown to function in the way that reinnervated superior colliculus cells respond to a flash of light applied to the operated eye 4. As compared with visual responses of normal superior colliculus cells, however, visual responses of the reinnervated cells are still weak. This may be partly related to some morphological or functional abnormalities of regenerated ganglion, cells besides their small number (less than 10% of total ganglion cells) 7,8. Here we report some light-microscopic observations of the regenerated ganglion cells in whole-mount retina and light- and electron-microscopic observations of cross-sections of the graft wherein regenerated axons grew. Correspondence: Y. Fukuda, Department of Physiology, Osaka University Medical School, Osaka 530, Japan. 0168-0102/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
$25
MATERIALSAND METHODS Adult golden hamsters 6-8 weeks old or Wistar rats 8-9 weeks old were anesthetized with 4% chloral hydrate (1 ml/100 g body wt.). Under aseptic conditions the left retina was lesioned in the rats and the left optic nerve was transected in the hamsters. In both cases autologous transplantation of the right sciatic nerve was made into the lesioned retina or the cut end of the optic nerve. The graft was laid on the skull and its distal end was tightened to the temporal muscle 7,8.
Retrograde labeling of regenerated ganglion cells At 2-4 months after grafting, granules of tetramethylrhodamine dextran (Molecular Probes) were applied to the graft at 1-2 cm from the eye in operated hamsters and rats. Three to 5 days later the animals were perfused with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4), the operated eye was excised and the retina was mounted whole. Retrogradely labeled ganglion cells were observed under an epifluorescent microscope (Optiphoto, Nikon) excited at 546 nm by G-filter (Nikon). The locations of labeled cells were plotted with the aid of a microscanner (Sapporo) and later replotted in the outline of the whole mount. Soma diameter was directly measured for each cell under the epifluorescent microscope using the lattice attached to the eye-piece.
Electron-microscopic examinations of the cross-section of the graft The grafted animals were perfused with 4% paraformaldehyde in 0.1 M phosphatebuffered saline (pH 7.4) (PBS) under anesthesia. Autografted sciatic nerves were removed and immersed in a mixture of 0.2% picric acid, 4% paraformaldehyde and 0.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) and kept at 4 ° C for 2 h. They were postfixed with 2% OsO4 in 0.1 M phosphate buffer at 4 ° C for another 2 h. They were then dehydrated with a series of graded alcohol and embedded in epoxy resin. Thick sections, ca. 1/~m, were cut and stained with a toluidine blue solution and examined with a light microscope. Thin sections were cut and stained with uranyl acetate and lead citrate. They were examined with an electron microscope (H-7000, Hitachi).
Indirect immunofluorescence At 8-12 weeks post-grafting the hamsters were perfused with 4% paraformaldehyde, to examine the distribution of laminin in the cross-section of the graft. Intact optic nerves and sciatic nerves were also obtained from control hamsters similarly perfused with 4% paraformaldehyde. Sample specimens were washed with PBS 3 times for 5 min each and immersed in PBS containing 4% glycine to avoid non-specific binding of antibodies. These sections were air-dried and rehydrated in PBS. They were incubated with 0.01 m g / m l rabbit anti-laminin antibody (Transformation Research) for 1 h and with 0.44 mg/ml FITC-labelled anti-rabbit IgG (Organon Teknika) for 1 h at room temperature in a humidified chamber. Sections were washed with PBS 3 times for 15 min in total at each step. They were coverslipped and examined with an epi-illuminated fluorescent microscope (Zeiss). RESULTS
Morphology of regenerated ganglion cells In hamsters where the graft was made into the transected optic nerve, retrogradely labeled cells were widely distributed across the whole mount. The labeled cells of the rat retina were distributed, as expected, in a fan-shaped area which extended from the lesioned and grafted site of the retina towards the periphery. In 2 grafted hamsters and 3
S26
A
B Hamster
N
Rat
12o. Control
Control ~ = 1 3.5±3.61
40! o
40
~ = 16.3±4.17
20 . _I ...... I0
I.I .............. 20 30 40
o 50 (Pm)
Regenerated
o , .-.-.i! ........ I0
20
50 (l~m)
40
SOMA DIAMETER
n=92
N
HI.,._,, .......... 30
.!..= . . . . . . . . . . . . . . . . . . 20 30 40
Regenerated n=218
N
.... I : 10
o 50
. _,1
i',
I0
20
.....
. .......... 30
40
50
SOMA DIAMETER
Fig. 1. Soma size histograms of regenerated ganglion cells back-labeled with tetramethylrhodamine dextrans. (A) Data from the hamster retina, where the graft was applied to the transected optic nerve. (B) Data from the rat retina where the graft was applied to the lesioned retina. In each case regenerated cells are compared with those from the control animal. Note that regenerated cells tend to be larger than the control only in the hamster retina.
grafted rats, the soma size of labeled ganglion cells was measured and compared with that obtained from normal control animals. As Figure 1A indicates, the soma size of regenerated ganglion cells tends to be larger, on average, than that in normal hamsters. All the data were sampled from various locations of the retina to exclude biased sampling in terms of centro-periphery. In the rats with the graft into the retina (as shown in Fig. 1B), the soma size distribution of the ganglion cells covers almost the same range of that of the non-operated retinas. Retrograde labeling with tetramethylrhodamine dextran occasionally revealed distal as well as proximal dendrites of regenerated ganglion cells. Although a limited number of these cells showed normal dendritic patterns, most cells were abnormal in having fewer, thinner and curved dendrites. Some cells also showed axonal abnormality, such as emanating from the secondary dendrites or taking a meandering course or a long detour to the optic disk. Detailed comparisons of the dendritic or axonal morphologies of these regenerated ganglion cells are in progress using intracellular injections of Lucifer Yellow (Tauchi et al., in preparation).
Light- and electron-microscopic observations of the regenerated axons In toluidine-blue-stained specimens large numbers of degenerated profiles were observed in the cross-section of the graft. Among them there were clusters of Schwann cells with apparently normally myelinated axons. Under electron microscopy many unmyelinated axons, small in diameter, were observed surrounded by Schwann cell processes and their basal laminae. N o neuronal process was observed alone in the endoneural space. On occasion, as illustrated in Figure 2, multiple uumyelinated axons were surrounded by a single Schwann cell. Usually neuropils were completely wrapped with Schwann cell processes. Some axons, however, were directly apposed to the basal lamina (indicated by arrows in Fig. 2). Nuclei of Schwann cells were frequently distorted in profile. In this manner Schwann cell processes and basal laminae were both quite irregular in the
$27
Fig. 2. Electron micrographs of cross-sections of regenerated axons in the graft. Note 3 unmyelinated axons ( * ) which are incompletely wrapped by Schwarm cell processes and directly apposed to their basal lamina (arrow heads). An apparently normal large myelinated axon is seen at the upper left comer. S = Schwann cell. Calibration bar = 1/Lm. x 28 000. cross-section of the graft. There were some apparently normally myelinated axons which were surrounded by single Schwann cell cytoplasms. On the other hand, more often we observed smaller axons individually wrapped by a myelin sheath that was disproportionately thick. Moreover, we encountered some small axons which could not fill up the inner space made with the myelin sheath. Some of these myelin sheaths had deteriorated and appeared to be in the course of adjusting their size to that of the regenerated axon.
Immunoreactioity to anti-laminin antibody Figure 3 shows the distribution of laminin on a cross-section of the the grafted sciatic nerve, the optic nerve, and the intact sciatic nerve. In the cross-section of the graft, specific immunofluorescence was observed as round or spherical profiles with irregular contours in the endoneurium (Fig. 3A). Blood vessels were also intensely labeled (arrows in Fig. 3A). This pattern of laminin-immunoreactive structures appears to correspond well with the irregular arrangement of basal laminae observed above b y electron microscopy. As has been reported by other authors 6, no specific fluorescence was observed on cross-sections of the mature optic nerve except for the periphery of blood vessels and the perineurium (arrows and p in Fig. 3B). In the cross-sections of the intact sciatic nerve, we observed laminin-immunoreactive structures with regular round or spherical profiles in the endoneural space (Fig. 3A). DISCUSSION By grafting the autologous sciatic nerve to the lesioned retina or the transected optic nerve we have succeeded in regenerating axons of retinal ganglion cells in rats and
t~
rO
S29 hamsters. However, we f o u n d that these regenerated ganglion cells revealed various abnormalities in soma size and dendritic morphology. Especially in the case of optic nerve transection, we observed that the s o m a size was larger in regenerated than in control retinal ganglion cells 7,8. This p r o b a b l y reflects stronger R N A synthesis during the phase of axonal regeneration as has been shown for retinal ganglion cells in vertebrates 3. W h e n a retinal lesion was m a d e and grafted into the rat retina, there was n o difference between the soma size of regenerated and intact ganglion cells. In such an operation only a small n u m b e r of cells regenerated and we m a y have missed larger regenerated cells by retrograde labelling. Besides enlargement of their soma size m a n y ganglion cells reveal quite an a b n o r m a l pattern of dendritic m o r p h o l o g y such as very thin elongated curved dendrites. W h a t kinds of neural inputs are making synapses with these regenerated dendrites in the inner plexiform layer? W e also noted that the dendritic m o r p h o l o g y of some ganglion cells appears quite normal. A n interesting question arises as to whether these apparently n o r m a l dendrites are the result of complete regeneration or of ganglion cells resistance to axonal severance. A t present we do n o t k n o w the m e c h a n i s m underlying the regenerative process of axons after sciatic nerve transplantation. The Schwann cells have been implicated to play a major role in axonal regeneration of the C N S neurons 5. F r o m electron-microscopic and immunofluorescent observations in the present study, it is quite possible that laminin attached to the basal lamina of Schwann cells acts as a cell-adhesion molecule and a neurite-promoting factor as well as for regenerating retinal axons. O n the other hand, although a recent study has indicated the developmental loss of ganglion cell responses to laminin 2, some other substances expressed or p r o d u c e d b y S c h w a n n cells m a y be m o r e essential for the regeneration of retinal axons in the adult. F u r t h e r studies are definitely needed to examine these possibilities b y using b o t h in vivo and in vitro preparations. ACKNOWLEDGEMENT The authors thank Dr. K.-F. So for his kind guidance in the p r e p a r a t i o n of transplanted animals. REFERENCES 1 Carter, D.A., Bray, G. and Aguayo, A.J., Regenerated retinal ganglion cell axons can form well-differentiated synapses in the superior colliculus of adult hamsters, J. Neurosci., 9 (1990) 4042-4050. 2 Cohen, J., Bume, J.F., Winter, J. and Bartlett, P., Retinal ganglion cell lose response to laminin with maturation, Nature (Lond.), 322 (1986) 465-467. 3 Dokas, L.A., Kohsaka, S., Burrell, H.R. and Agranoff, B.W., Uridine metabolism in the goldfish retina during optic nerve regeneration: whole mount studies, J. Neurochem., 36 (1981) 1160-1165. 4 Keirstead, S., Rasminsky, M., Fukuda, Y., Carter, D. and Aguayo, A.J., Electrophysiologlc responses in hamster superior colliculus evoked by regenerating retinal axons, Science, 246 (1989) 255-257. 5 Keynes, R.J., Schwann cells during development and regeneration: leaders or followers7 Trends Neurosci., 10 (1987) 137-139.
• Fig. 3. Immunofluorescence micrographs of grafted sciatic nerve (A), optic nerve (B) and sciatic nerve (C). The sections were incubated with rabbit anti-laminin antibody and with anti-rabbit IgG. (A) In the cross-sectioned graft, specific fluorescence is seen as small round or spherical profiles in the endoneural space, probably corresponding to the basal lamina of Schwarm cells ((2). Blood vessels are also strongly immunoreactive (arrows). (B) Specific fluorescence is seen only along the blood vessels (arrows) and in the perineurium (p) and none in endoneural space. (C) Bright fluorescence=is seen as round or spherical profiles of regular size in the endoneural space, corresponding to the basal laminae of Schwann cells. × 145.
S30 6 McLoon, S.C., McLoon, L.K., Palm, S.L. and Furcht, L.T., Transient expression of laminin in the optic nerve of the developing rat, J. Neurosci., 8 (1988) 1981-1990. 7 So, K.F. and Aguayo, A.J., Lengthy regrowth of cut axons from ganglion cells after peripheral nerve transplantation into the retina of adult rats, Brain Res., 328 (1985) 349-354. 8 Vidal-Sanz, M., Bray, G.M., Villegas-Perez, M.P., Thanos, S. and Aguayo, A.J., Axonal regeneration and synapse formation in the superior colliculus by retinal ganglion cells in the adult rat., J. Neurosci., 7 (1987) 2894-2909.
$24
NEURES 00356
Optic nerve regeneration by peripheral nerve transplant Yutaka Fukuda
1, Hitoshi
Sasaki 3, Eijiro A d a c h i 2, Tetsu I n o u e 4 a n d K a t s u k o Morigiwa 1
1 Department of Physiology and 2 Department of Anatomy, Osaka University Medical School, s Department of Physiology, Hyogo Medical College and 4 Section of Ophthalmology, Izumiohtsu City Hospital, Osaka (Japan) Key words: Optic nerve; Regeneration; Retinal ganglion cells; Transplantation; Basal lamina; Schwann cells; Laminin; Adult mammals
SUMMARY We studied the morphology of regenerated retinal ganglion cells and their axons in adult rodents after axotomy and autologous transplantation of the sciatic nerve. Regenerated ganglion cells, backlabeled with rhodamine dextran, were of similar size to or larger than those of intact cells in control animals. Dendrites and occasionally axons as well showed abnormal morphologies in most cells, though some cells appeared quite normal. Cross-sections of the regenerated axons, observed by electron microscopy, were always attached to either the Schwann cell cytoplasm or the basal lamina. The immunoreactive structures to anti-laminin antibody were quite irregular in the cross-sectioned graft and, compared with those of the intact sciatic nerve, they were generally smaller. Their appearance closely resembled that of the basal lamina in the graft observed by elecron microscopy. These observations, taken together, suggest that the laminin-rich basal laminae of Schwann cells are essentially important for the regeneration of retinal axons in adult rodents.
INTRODUCTION
Since the pioneering work by Aguayo and his co-workers, it is now known that, once damaged, CNS neurons can regenerate their axons when the peripheral nerve has been transplanted. For example, in the rat and hamsters transected optic nerve fibers can regrow lengthy axons in the peripheral nerve graft and when properly guided their terminals make normal synapses with target cells in the superior colliculus 1,7,8. Furthermore, these synapses have recently been shown to function in the way that reinnervated superior colliculus cells respond to a flash of light applied to the operated eye 4. As compared with visual responses of normal superior colliculus cells, however, visual responses of the reinnervated cells are still weak. This may be partly related to some morphological or functional abnormalities of regenerated ganglion, cells besides their small number (less than 10% of total ganglion cells) 7,8. Here we report some light-microscopic observations of the regenerated ganglion cells in whole-mount retina and light- and electron-microscopic observations of cross-sections of the graft wherein regenerated axons grew. Correspondence: Y. Fukuda, Department of Physiology, Osaka University Medical School, Osaka 530, Japan. 0168-0102/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
$25
MATERIALSAND METHODS Adult golden hamsters 6-8 weeks old or Wistar rats 8-9 weeks old were anesthetized with 4% chloral hydrate (1 ml/100 g body wt.). Under aseptic conditions the left retina was lesioned in the rats and the left optic nerve was transected in the hamsters. In both cases autologous transplantation of the right sciatic nerve was made into the lesioned retina or the cut end of the optic nerve. The graft was laid on the skull and its distal end was tightened to the temporal muscle 7,8.
Retrograde labeling of regenerated ganglion cells At 2-4 months after grafting, granules of tetramethylrhodamine dextran (Molecular Probes) were applied to the graft at 1-2 cm from the eye in operated hamsters and rats. Three to 5 days later the animals were perfused with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4), the operated eye was excised and the retina was mounted whole. Retrogradely labeled ganglion cells were observed under an epifluorescent microscope (Optiphoto, Nikon) excited at 546 nm by G-filter (Nikon). The locations of labeled cells were plotted with the aid of a microscanner (Sapporo) and later replotted in the outline of the whole mount. Soma diameter was directly measured for each cell under the epifluorescent microscope using the lattice attached to the eye-piece.
Electron-microscopic examinations of the cross-section of the graft The grafted animals were perfused with 4% paraformaldehyde in 0.1 M phosphatebuffered saline (pH 7.4) (PBS) under anesthesia. Autografted sciatic nerves were removed and immersed in a mixture of 0.2% picric acid, 4% paraformaldehyde and 0.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) and kept at 4 ° C for 2 h. They were postfixed with 2% OsO4 in 0.1 M phosphate buffer at 4 ° C for another 2 h. They were then dehydrated with a series of graded alcohol and embedded in epoxy resin. Thick sections, ca. 1/~m, were cut and stained with a toluidine blue solution and examined with a light microscope. Thin sections were cut and stained with uranyl acetate and lead citrate. They were examined with an electron microscope (H-7000, Hitachi).
Indirect immunofluorescence At 8-12 weeks post-grafting the hamsters were perfused with 4% paraformaldehyde, to examine the distribution of laminin in the cross-section of the graft. Intact optic nerves and sciatic nerves were also obtained from control hamsters similarly perfused with 4% paraformaldehyde. Sample specimens were washed with PBS 3 times for 5 min each and immersed in PBS containing 4% glycine to avoid non-specific binding of antibodies. These sections were air-dried and rehydrated in PBS. They were incubated with 0.01 m g / m l rabbit anti-laminin antibody (Transformation Research) for 1 h and with 0.44 mg/ml FITC-labelled anti-rabbit IgG (Organon Teknika) for 1 h at room temperature in a humidified chamber. Sections were washed with PBS 3 times for 15 min in total at each step. They were coverslipped and examined with an epi-illuminated fluorescent microscope (Zeiss). RESULTS
Morphology of regenerated ganglion cells In hamsters where the graft was made into the transected optic nerve, retrogradely labeled cells were widely distributed across the whole mount. The labeled cells of the rat retina were distributed, as expected, in a fan-shaped area which extended from the lesioned and grafted site of the retina towards the periphery. In 2 grafted hamsters and 3
S26
A
B Hamster
N
Rat
12o. Control
Control ~ = 1 3.5±3.61
40! o
40
~ = 16.3±4.17
20 . _I ...... I0
I.I .............. 20 30 40
o 50 (Pm)
Regenerated
o , .-.-.i! ........ I0
20
50 (l~m)
40
SOMA DIAMETER
n=92
N
HI.,._,, .......... 30
.!..= . . . . . . . . . . . . . . . . . . 20 30 40
Regenerated n=218
N
.... I : 10
o 50
. _,1
i',
I0
20
.....
. .......... 30
40
50
SOMA DIAMETER
Fig. 1. Soma size histograms of regenerated ganglion cells back-labeled with tetramethylrhodamine dextrans. (A) Data from the hamster retina, where the graft was applied to the transected optic nerve. (B) Data from the rat retina where the graft was applied to the lesioned retina. In each case regenerated cells are compared with those from the control animal. Note that regenerated cells tend to be larger than the control only in the hamster retina.
grafted rats, the soma size of labeled ganglion cells was measured and compared with that obtained from normal control animals. As Figure 1A indicates, the soma size of regenerated ganglion cells tends to be larger, on average, than that in normal hamsters. All the data were sampled from various locations of the retina to exclude biased sampling in terms of centro-periphery. In the rats with the graft into the retina (as shown in Fig. 1B), the soma size distribution of the ganglion cells covers almost the same range of that of the non-operated retinas. Retrograde labeling with tetramethylrhodamine dextran occasionally revealed distal as well as proximal dendrites of regenerated ganglion cells. Although a limited number of these cells showed normal dendritic patterns, most cells were abnormal in having fewer, thinner and curved dendrites. Some cells also showed axonal abnormality, such as emanating from the secondary dendrites or taking a meandering course or a long detour to the optic disk. Detailed comparisons of the dendritic or axonal morphologies of these regenerated ganglion cells are in progress using intracellular injections of Lucifer Yellow (Tauchi et al., in preparation).
Light- and electron-microscopic observations of the regenerated axons In toluidine-blue-stained specimens large numbers of degenerated profiles were observed in the cross-section of the graft. Among them there were clusters of Schwann cells with apparently normally myelinated axons. Under electron microscopy many unmyelinated axons, small in diameter, were observed surrounded by Schwann cell processes and their basal laminae. N o neuronal process was observed alone in the endoneural space. On occasion, as illustrated in Figure 2, multiple uumyelinated axons were surrounded by a single Schwann cell. Usually neuropils were completely wrapped with Schwann cell processes. Some axons, however, were directly apposed to the basal lamina (indicated by arrows in Fig. 2). Nuclei of Schwann cells were frequently distorted in profile. In this manner Schwann cell processes and basal laminae were both quite irregular in the
$27
Fig. 2. Electron micrographs of cross-sections of regenerated axons in the graft. Note 3 unmyelinated axons ( * ) which are incompletely wrapped by Schwarm cell processes and directly apposed to their basal lamina (arrow heads). An apparently normal large myelinated axon is seen at the upper left comer. S = Schwann cell. Calibration bar = 1/Lm. x 28 000. cross-section of the graft. There were some apparently normally myelinated axons which were surrounded by single Schwann cell cytoplasms. On the other hand, more often we observed smaller axons individually wrapped by a myelin sheath that was disproportionately thick. Moreover, we encountered some small axons which could not fill up the inner space made with the myelin sheath. Some of these myelin sheaths had deteriorated and appeared to be in the course of adjusting their size to that of the regenerated axon.
Immunoreactioity to anti-laminin antibody Figure 3 shows the distribution of laminin on a cross-section of the the grafted sciatic nerve, the optic nerve, and the intact sciatic nerve. In the cross-section of the graft, specific immunofluorescence was observed as round or spherical profiles with irregular contours in the endoneurium (Fig. 3A). Blood vessels were also intensely labeled (arrows in Fig. 3A). This pattern of laminin-immunoreactive structures appears to correspond well with the irregular arrangement of basal laminae observed above b y electron microscopy. As has been reported by other authors 6, no specific fluorescence was observed on cross-sections of the mature optic nerve except for the periphery of blood vessels and the perineurium (arrows and p in Fig. 3B). In the cross-sections of the intact sciatic nerve, we observed laminin-immunoreactive structures with regular round or spherical profiles in the endoneural space (Fig. 3A). DISCUSSION By grafting the autologous sciatic nerve to the lesioned retina or the transected optic nerve we have succeeded in regenerating axons of retinal ganglion cells in rats and
t~
rO
S29 hamsters. However, we f o u n d that these regenerated ganglion cells revealed various abnormalities in soma size and dendritic morphology. Especially in the case of optic nerve transection, we observed that the s o m a size was larger in regenerated than in control retinal ganglion cells 7,8. This p r o b a b l y reflects stronger R N A synthesis during the phase of axonal regeneration as has been shown for retinal ganglion cells in vertebrates 3. W h e n a retinal lesion was m a d e and grafted into the rat retina, there was n o difference between the soma size of regenerated and intact ganglion cells. In such an operation only a small n u m b e r of cells regenerated and we m a y have missed larger regenerated cells by retrograde labelling. Besides enlargement of their soma size m a n y ganglion cells reveal quite an a b n o r m a l pattern of dendritic m o r p h o l o g y such as very thin elongated curved dendrites. W h a t kinds of neural inputs are making synapses with these regenerated dendrites in the inner plexiform layer? W e also noted that the dendritic m o r p h o l o g y of some ganglion cells appears quite normal. A n interesting question arises as to whether these apparently n o r m a l dendrites are the result of complete regeneration or of ganglion cells resistance to axonal severance. A t present we do n o t k n o w the m e c h a n i s m underlying the regenerative process of axons after sciatic nerve transplantation. The Schwann cells have been implicated to play a major role in axonal regeneration of the C N S neurons 5. F r o m electron-microscopic and immunofluorescent observations in the present study, it is quite possible that laminin attached to the basal lamina of Schwann cells acts as a cell-adhesion molecule and a neurite-promoting factor as well as for regenerating retinal axons. O n the other hand, although a recent study has indicated the developmental loss of ganglion cell responses to laminin 2, some other substances expressed or p r o d u c e d b y S c h w a n n cells m a y be m o r e essential for the regeneration of retinal axons in the adult. F u r t h e r studies are definitely needed to examine these possibilities b y using b o t h in vivo and in vitro preparations. ACKNOWLEDGEMENT The authors thank Dr. K.-F. So for his kind guidance in the p r e p a r a t i o n of transplanted animals. REFERENCES 1 Carter, D.A., Bray, G. and Aguayo, A.J., Regenerated retinal ganglion cell axons can form well-differentiated synapses in the superior colliculus of adult hamsters, J. Neurosci., 9 (1990) 4042-4050. 2 Cohen, J., Bume, J.F., Winter, J. and Bartlett, P., Retinal ganglion cell lose response to laminin with maturation, Nature (Lond.), 322 (1986) 465-467. 3 Dokas, L.A., Kohsaka, S., Burrell, H.R. and Agranoff, B.W., Uridine metabolism in the goldfish retina during optic nerve regeneration: whole mount studies, J. Neurochem., 36 (1981) 1160-1165. 4 Keirstead, S., Rasminsky, M., Fukuda, Y., Carter, D. and Aguayo, A.J., Electrophysiologlc responses in hamster superior colliculus evoked by regenerating retinal axons, Science, 246 (1989) 255-257. 5 Keynes, R.J., Schwann cells during development and regeneration: leaders or followers7 Trends Neurosci., 10 (1987) 137-139.
• Fig. 3. Immunofluorescence micrographs of grafted sciatic nerve (A), optic nerve (B) and sciatic nerve (C). The sections were incubated with rabbit anti-laminin antibody and with anti-rabbit IgG. (A) In the cross-sectioned graft, specific fluorescence is seen as small round or spherical profiles in the endoneural space, probably corresponding to the basal lamina of Schwarm cells ((2). Blood vessels are also strongly immunoreactive (arrows). (B) Specific fluorescence is seen only along the blood vessels (arrows) and in the perineurium (p) and none in endoneural space. (C) Bright fluorescence=is seen as round or spherical profiles of regular size in the endoneural space, corresponding to the basal laminae of Schwann cells. × 145.
S30 6 McLoon, S.C., McLoon, L.K., Palm, S.L. and Furcht, L.T., Transient expression of laminin in the optic nerve of the developing rat, J. Neurosci., 8 (1988) 1981-1990. 7 So, K.F. and Aguayo, A.J., Lengthy regrowth of cut axons from ganglion cells after peripheral nerve transplantation into the retina of adult rats, Brain Res., 328 (1985) 349-354. 8 Vidal-Sanz, M., Bray, G.M., Villegas-Perez, M.P., Thanos, S. and Aguayo, A.J., Axonal regeneration and synapse formation in the superior colliculus by retinal ganglion cells in the adult rat., J. Neurosci., 7 (1987) 2894-2909.