Optic axons regenerate into sciatic nerve isografts only in the presence of Schwann cells

Brain Research Bulletin,

0361-9230188 $3.00 + .OO

Vol. 20, pp. 223-231. bl Pergamon Press plc, 1988. Printed in the U.S.A.

Optic Axons Regenerate Into Sciatic Nerve Isografts Only in the Presence of Schwann Cells h/l. BERRY,* L. REES,* S. HALL,* AND J. SIEVERSt

P. YIU*

Anatomy Departments *Guy’s Hospital Medicul School, London SE1 9RT, UK tUniversity of Kiel, Olshausen Str. 40, D-2300, Kiel, FRG Received 8 September

1987

BERRY, M., L. REBS, S. HALL, P. YIU AND J. SIEVERS. Optic axnns regenerate into sciatic nerve isografts only in oj’Schwann cells. BRAIN RES BULL 20(2) 223-231, 1988.-Optic axons regenerate into normal but not acellular peripheral nerve (PN) grafts. The first axons penetrate the PN graft before 5 days and grow inside the basal lamina tubes amongst the Schwann cells. By 30 days, 4% of the surviving retinal ganglion cells (RGC) regenerate axons for at least 10 mm into the PN graft. Laminin rich basal lamina tubes persist in the acellular PN transplants but only a few axons penetrate the most proximal parts of the tubes by 5 days and none grow farther into the graft by 30 days. RCC counts demonstrate that 34% of the normal RGC population survive 30 days after anastomosing a normal PN to the transected optic nerve. After anastomosing acellular PN grafts, 25% of RGCs survive compared with 10% after optic nerve section. These findings demonstrate that laminin does not promote regeneration of axons and that Schwann cells play the primary role of offering trophic support and even a substrate for growth. RGC survival is also enhanced by PN grafts even when Schwann cells are absent. This latter result suggests that RGC survival is promoted by a trophic substance released from axons and/or Schwann cells in the PN grafts which survives the thawin~freezing procedure (used to kill the Schwann cells) and is active in the grafts in the immediate post operative period.

the presence

Optic axons

Regeneration

Sciatic nerve isografts

Schwann cells

stance which may not be related to the axon growth promoting factor since it is active in both the cellular and aceliular

most fibres in the central nervous system (CNS), retinal ganglion cell (RGC) axons will not regenerate if damaged in the optic nerve [ 12,521, but are exceptional in exhibiting good, although non-directed, regrowth when injured in the retina [42]. Regeneration of axons in the completely transected optic nerve is promoted by anastomosing a segment of sciatic nerve to the cut stump, whilst the regrowth of severed retinal fibres becomes directed into peripheral nerves (PN) implanted into a retinal lesion [6-S, 591. We have investigated the source of possible PN-derived trophic and tropic factors [50] (which could support and direct RGC axonal regeneration, and maintain the viability of RGCs) by anastomosing segments of sciatic nerve, with and without their complement of Schwann cells, to the distal stumps of completely transected optic nerves. We report here that RGC axons only grow into PN grafts if the latter contain Schwann cells. Thus, regenerating CNS axons do not penetrate acellular PN transplants in which basal lamina tubes, expressing the neurite adhesion molecule laminin [26], remain; a finding which fails to support the assertion that laminin promotes axon regene~tion [2?, 36, 371. Moreover, our results show that PN also contain a RGC viability subLIKE

PN grafts. METHOD

Throughout, the optic nerve stump is labelled according to the convention that the end connected to the retina is proximal and that connected to the brain is distal. Animals

WAG albino rats (250 g) of either sex were used throughout, and each experimental group composed 6 animals (Table 1). PN

Grafts

Complete sectioning of the optic nerve was achieved by removing a 3 mm length, 2-3 mm from the globe (Fig. IB). A 15-20 mm segment of sciatic nerve was then anastomosed to the distal optic nerve stump, as previously described by us

223

224

BERRY, REES, HALL, YIU AND SIEVERS TABLE TREATMENT

1

GROUPS A-H

Duration Post Lesion

Parameter

Treatment

Preparation

Analysis

A. Normal total number of RGCs in retina

Optic nerve crush

HRP to crush site

HRP development of retinal whole mount for total RGC count

36 hours

Cresyl violet staining of whole mounted retina

Total neuron count in retina; B-A=amacrine count

-

B. Normal total number of amacrine cells C. HRP diffusion control

Sciatic nerve segment grafted to proximal end of cut optic nerve

HRP to crush site of sciatic nerve graft 10 mm from anastomosis

HRP development of retinal whole mount to assess HRP diffusion

36 hours

D.

Optic nerve section

Complete transection of optic nerve 2 mm from disc

Cresyl violet staining of whole mounted retina

Total neuron count in retina, less amacrine count in B=total RGC survival

30 days

E.

Sciatic nerve graft

Anastomosis of sciatic nerve graft to proximal end of cut optic nerve

HRP to crush site of sciatic nerve graft 10 mm from anastomosis

Total HRP-filled RGC count in retina=total number of RGC regenerating axons 10 mm into sciatic nerve graft

30 days

F. Sciatic nerve graft

Anastomosis of sciatic nerve graft to proximal end of cut optic nerve

Cresyl violet staining of whole mounted retina

Total neuron count in retina, less amacrine count in B=total RGC survival

30 days

G. Acellular sciatic graft

Anastomosis of acellular sciatic nerve graft to proximal end of cut optic nerve

HRP to crush site of sciatic nerve graft 10 mm from anastomosis

Total HRP-filled RGC count in retina=total number of RGC regenerating axons 10 mm into acellular sciatic graft

30 days

H. Acellular sciatic graft

Anastomosis of acellular sciatic nerve graft to proximal end of cut optic nerve

Cresyl violet staining of whole mounted retina

Total neuron count in retina, less amacrine count in B= total RGC survival

30 days

Each group contained 6 animals. [f&8]. In one group, all Schwann cells were killed in the sciatic nerve graft, prior to transplantation, by repeatedly freezing (X 5 to -25°C) and thawing [28]. Viability of Schwann cells after this regime was checked using the trypan blue exclusion test [23]; all Schwann cells stained with trypan blue after the freezing/thawing procedure. In a further group of 6 rats, the optic nerve was transected and the cut ends reapproximated by sutures without inserting a PN graft (Fig. 1A). Anterograde

and Retrograde

Tracing

At 30 days post-lesion (dpl), 0.5 ~1 of a 4% solution of HRP-WGA (Sigma L 2384) was slowly infused using a micropipette into a crushed region of the sciatic nerve grafts, 10 mm from the optic nerve head (Fig. lB), and the site of anastomosis and the retina were both examined 36 hr later for the presence of HRP-WGA filed axons and gahglion cells. The HRP was developed using the technique of Mesulam [43]. The fluorescent tracer rhodamine B isothiocyanate (RITC-B, Sigma; R-1755) was prepared according to the method of Thanos and Bonhoeffer 1601, and 5

~1 was injected into the vitreous of the operated eye in 6 animals at 5, 10, 15 and 30 dpl. Regeneration across approximated stumps of the optic nerve and also into the different PN grafts was studied under green light using 10-20 pm frozen sections of 4% paraformaldehyde fixed material. Histology and tmmunohistochemistt-y

The anastomotic site was examined at 5, 10, 15 and 30 dpl. Tissue was embedded in polyester wax and alternate 5 pm sections processed for immunocytological demonstration of either laminin, using rabbit antisera to mouse laminin (Bethesda Research Laboratories, BRL); or axons, using both a mouse monoclonal anti-neurofilament antibody (dil. l/3; Clone No. 97 producing antibody against 200K Da subunit of rat brain neurofilament, Dr. B. Anderton, St. George’s Hospital, London) and the silver physical developer method described by Kieman [33]. The immunofluorescence demonstrated in rats with laminin antisera, used at l/100 dilution, was completely suppressed by pre-incubation with purified laminin (BRL) from EHS mouse sarcoma.

225

OPTIC AXON REGENERATION

injection P-WGA

site of anastomosis

dorsal

FIG. 1. Illustration of the method of (A) optic nerve section; (B) grafting of a PN to the proximal stump of the optic nerve; and (C) distribution of sampling graticule at identified points over the surface of a whole mounted retina. Electron Microscopy

A further 2 groups, each containing 8 animals, were used for electronmicroscopical examination of the anastomosis after receiving either normal or acellular sciatic nerve grafts. At 5, 10, 15 and 30 dpl, 2 rats were perfused with 2.5% glutaraldehyde in 0.1 M phosphate buffer. The site of anastomosis was post-fixed in 1% osmium tetroxide, divided into 2 mm segments in a proximo-distal sequence and embedded in TAAB resin. One pm sections were stained with toluidine blue and examined light microscopically. Ultrathin sections were examined in a Hitachi HU12A electron microscope. Retinal Cell Counting Counting technique. The total population of neurons in whole mounted retinae was estimated using a sampling technique [6-8). A dorsal suture was placed in the sclera of each eye during removal so as to identify temporal, nasal, dorsal and ventral quadrants on whole mounts. Counts were made in a sampling graticule (22,500 pm’) placed at 3 sites along the four radii which bisect each quadrant at points R, 2R/3 and R/3 (where R=O radius of the retina, Fig. 1C). A MOP videoplan (Kontron) image analysis system was used to record both the number of neurons and their surface area. Glia and endothelial cells were excluded. The frequency histograms of the cell profiles were plotted on histograms at size intervals of 5.0 pm2. The displaced amacrine cell frequency histogram of sizes was estimated by subtracting the normal HRP-WGA filled RGC size profile from that of the neuron profile, calculated from estimating cell sizes in the cresyl violet stained retinae.

The normal frequency distribution of displaced amacrine cell sizes was then subtracted from the distribution of neuron sizes in each experimental group to obtain the frequency distribution of different sizes of RGC in the experimental animals assuming that amacrine cell size is unaffected by optic nerve section. The number of cells regenerating axons into grafts was estimated directly, after injecting the graft with HRP-WGA, by counting all retrogradely filled RGC over the entire retinal surface of the whole mounts. Totul RGC population in normal rats. HRP-WGA was injected into the crush site of the optic nerve of normal animals (Table 1) and 36 hr later the HRP-WGA was developed [43] in the whole mounted retinae. A count of all retrogradely filled cells is an estimate of the total population of RGCs. Displaced amacrine cell population in normul rats. After perfusing animals (Table 1) with 4% paraformaldehyde the total number of neurons in the RGC layer of the retina was counted in cresyl violet stained whole retinae mounted onto gelatin subbed slides. The number of displaced amacrine cells was estimated by subtracting the total RGC count (obtained by the method outlined above) from the total (cresyl violet stained) neuron count. Total RGC population in experimental rats. In the experimental groups, the total number of RGCs in each retina is underestimated since not all RGCs regenerate their axons. Thus, to estimate the total number of RGCs surviving in the retinae of experimental animals, cresyl violet stained material was used. It was assumed that the number of displaced amacrine cells was unaffected by optic nerve section (since these cells do not project axons into the optic nerve) and thus the total number of RGCs in experimental animals was computed by subtracting the normal displaced amacrine cell number from the total neuron count in the cresyl violet stained experimental retinae. Number of RGCs regenerating axons. The number of RGCs regenerating axons was estimated by counting the total number of retrogradely HRP-WGA filled RGCs in whole retinal mounts after injecting HRP-WGA into a crush site of the sciatic nerve graft 10 mm from the site of anastomosis with the proximal stump of the optic nerve. RESULTS

Our estimate of the mean total number of HRP-WGA filled ganglion cells in the normal retina of 111,206+-3,290 (S.E.) is in good agreement with both the estimate of the total number of axons in the rat optic nerve and with the number of RGCs in the ganglion cell layer [39, 48, 491. The mean total displaced amacrine count of 107,945?1,764 also corroborates the finding that displaced amacrine cells in the rat represent approximately 50% of the neurons in the ganglion cell layer [39,48,49]. The mean total number of RGCs in the retinae of rats with normal sciatic grafts, (37,658+6,329), acellular sciatic grafts (27,851+3,794) and after optic nerve section (10,743?4,058), represents a survival of 34%, 25% and 10% respectively. Of the 34% RGCs remaining in rats receiving normal PN grafts, only 4% (1,529? 125) grow their axons 10 mm along the PN to the site of HRP-WGA injection. No regeneration occurs into either acellular PN grafts or into the distal optic nerve. Frequency histograms (Fig. 2) of the size classes of RGCs show a huge loss of large and medium sized cells in all groups, which is almost total both after optic nerve section and after acellular PN grafting. In

226

BERRY, REES, HALL, YIU AND SIEVERS normal 6000 4000 2000 0 6000

optic nerve section

Boo0 2000 z

0

s g 8000t

I-11 acehiar

normal

sciatic

sciatic

100

anastomosis

anastomosis

150

area of ganglion

200

250

300

cells tp&

FIG. 2. (A) Frequency histogram of different sizes of RGC in the retinae of normal rats, those with complete optic nerve section and those with either normal or acellular sciatic nerve grafts. Size interval in all histograms is 5.0 pm* and each represents the means (*SE in A) of 6 animals. In normal animals, the number of HRP-tilled RGCs in each size group is subtracted from the total retinal neuron count to obtain the number ofdisplacedamacrine cells at each size interval. This mean size distribution of amacrine cells is subtracted from the mean total retinal neuron count of experimental animals to obtain the mean total number of RGCs and their size profile in each experimental group. (B) HRP-WGA filled ganglion cells in the retina of normal rat. (C) HRP-filled ganglion cells in the retina of a rat 30 days after anastomosing a normal sciatic nerve segment to the cut end of the optic nerve (marker in B and C=lOO pm).

rats given intact PN grafts the few surviving large and medium sized cells are qualitatively obvious and this trend is

clearly seen in the histogram (Fig. 2). RITC-B anterograde tracing (Fig. 3) and anti-neurofdament immunofluorescence detected a few retinal RGC fibres at the site of anastomosis at 5 dpl after optic nerve section or after anastomosis of an acellular PN to the cut end of the optic nerve. In those animals receiving normal PN grafts, optic fibres had already penetrated the implant for up to 1 mm. At 10 and 15 dpi, deep penetration had occurred into the normal PN grafts (for distances beyond 5 mm) and, by 30 dpl, large numbers of RGC axons were seen streaming across ‘the anastomosis (Fig. 3A). Qualitatively, the density of axons decreased along the PN suggesting that RGC axons were either entering the grafts at

different times or were growing at different rates along the grafts. Few axons penetrated the acellular PN implants at 5, 10 and 15 dpt, and by 30 dpl only a small number of short varicose axons remained at the borders of the graft (Fig. 3C). No axons bridged the approximated cut stumps of the optic nerve at any of the sampling times although, from 10 dpl, many RGC axons invaded the scar tissue at the anastomosis and ramified within the connective tissue about the optic nerve (Fig. 3B). The findings shown in Fig. 2A, B and C were replicated by results with anti-neuro~ament immunocytochemistry. Laminin-rich Schwann cell basal lamina tubes were abundant in both acellular and normal sciatic nerve grafts, at all times including 30 dpl. Axons only grew within the basal lamina tubes of normal implants (Fig. 4) in

OPTIC AXON REGENERATION

FIG. 3. (A) Anastomotic site between optic nerve (0) and normal sciatic nerve (ns) showing RITC-B anterograde labelling of RGC axons passing mto the PN segment from the optic nerve. (B) Anastomotic site between the cut ends of the optic nerve (ds, distal segment; ps, proximal segment) showing RITC-B anterograde labelling of RGC axons. NO axons pass from the proximal segment of the optic nerve into the distal segment, but fibres do course freely within the connective tissue scar and the dural sheath of the optic nerve. (C) Anastomotic site between optic nerve (0) and acellular sciatic nerve (as) showing RITC-B anterograde Iabelling of RGC axons. Note that the axons cross the scar tissue but only small numbers ramify for short distances within the acellular PN segment. All labelled tibres are of RGC origin. Apparent lateral entry of fibres into the PN in 3A on the distal side of the scar is caused by tibre clustering in the basal lamina tubes in the proximal part of the PN before entry into the less disorganised distal part of the PN (30 dpl; magni~cation x 110, A and B; x220, C).

association with Schwann cells. Occasional naked fibres were seen within basal lamina tubes confined to the site of anastomosis in acellular grafts at 15 and 30 dpl. Astrocytes and axons formed compound islands of tissue at the anastomosis and no Schwann cells invaded the optic nerve. DISCUSSION

Our counts of the total number of RGCs in the retinae of ex~~mental rats are obtained by an indirect technique in which the normal displaced amacrine count is subtracted from the actual total neuron count on the assumption that experimental manipulations do not affect the number of displaced amacrine cells [47]. In the absence of a direct estimate of the number of displaced amacrine cells in our experimental retinae, we are unable to test this assumption. However, the size of the amacrine population cannot increase, indeed, it might contract because, despite the absence of an extra-retinal projection, some displaced amacrine cells might die after optic nerve section, possibly because of synaptic reorganisation in the retina. Thus, our RGC counts might be underestimates, but not overestimates, of the true RGC number in optic nerve-sectioned and grafted rats. RGC viability in the presence of normal PN and acellular PN grafts is enhanced significantly @
and recently killed Schwann cells and/or isolated PN axon segments, of a factor capable of promoting RGC survival akin to that released from fetal spinal cord transplants [9]. Such a factor is different from that enhancing regeneration, since it is effective both in acellular and normal PN grafts alike. The persistence of RGCs after optic nerve section alone may indicate (a) the presence of serum derived growth factor [31,32]; (b) that glial cells in the distal optic nerve stump produce an RGC survival factor; (c) that surviving RGC cells possess coltaterals ramifying within the retina [29] which maintain the flow of trophic substances from their targets. The number of RGCs surviving after optic nerve section in our experiments on albino WAG rats is much less than that seen by Misantone et al. [45], in hooded rats after intracranial optic nerve crush. This difference may reflect inter-strain variation, or it may be related to the use of dBering sites and methods of expe~mental lesioning: in both experiments the reduction in cell size is of the same magnitude. RGC size reduction, after optic nerve section, may be attributed to either cell shrinkage [20,4.5] and/or differential loss of large cells [2]. Absolute counts, over all cell size ranges, indicate that large RGCs are lost and medium and small RGCs shrink. It is interesting that normal PN grafting has little effect upon these somatic reactions although the greater survival of RGCs is both qualitatively and quantitatively obvious [6-81. The present quantitative data define the paucity of the regeneration response, already alluded to by others [I, 4, 14,

BERRY, REES, HALL, YIU AND SIEVERS

FIG. 4. (A) Anti-laminin immunofluorescence in a normal sciatic nerve segment 30 days after anastomosis to an optic nerve-magnification x 110. (B) Anti-laminin immunotluorescence in an acellular sciatic nerve segment 30 days after anastomosis to an optic nerve-magnification x 110. (C) Transverse section through a normal PN graft, 0.5 cm distal to the point of anastomosis with an optic nerve, at 15 dpl, showing endoneu~um containing numerous basal lamina tubes in which Schwann celts (S) or Schwann cell processes (sp) and RGC axon sprouts (a) are present. One neurite is undergoing remyehnation @I). x6300. (D) Transverse section through an acellular PN graft, 0.5 cm distal to the point of anastomosis with an optic nerve, at 15 dpl, showing endoneurium containing a number of crenulated, collapsed and empty Schwann cell basal lamina tubes (arrows). Debris is present in macrophages, one of which is associated with a breach of the tubal wall (*). Negatively stained collagen is abundant between these structures. x8500.

22, 53, 551. Although RITC-B anterograde staining demonstrates huge numbers of fibres crossing the anastomotic site, there is a discrepancy between numbers of HRP-filled RGCs (4%) and viable RGCs (34%). Since no axons were seen to leave the PN nerve graft along its length, the latter observation might be explained if optic fibres regenerate at different

rates so that only the fastest had reached the site of HRP injection by 30 dpl. Nevertheless, our positive findings of regeneration into sciatic nerve grafts contrasts with those of Richardson et al. [54] who observed iittle regeneration into PN segments grafted into the optic nerve intracranially. They did not anastomose the two nerves and reported poor

229

OPTIC AXON REGENERATION approximation of the graft with the proximal stump of the optic nerve, which probably accounts for the failure of optic axons to grow into the grafts. In our experiments, the cut stumps were closely applied and immobilised by suture: necrosis was not observed at any time after operation. Our observations that optic axons do not penetrate acellular PN grafts suggests that Schwann cells play an essential role in the regeneration of central axons into PN grafts and that laminin, present in the basal lamina tubes of the PN, is not the neurotactic molecule as suggested by Liesi [36,37] and Hopkins et al. [27]. Although basal lamina components may be degraded either by the freeze/thawing procedure or by enzymes released from Schwann cells or by infiltrating macrophages [40] (as in degenerating myofibrils [21]), lamininrich, structurally intact, basal lamina tubes survive within our acellular grafts. Nonetheless, RGC axons do not regenerate along these channels. A basal lamina substrate may be used by pioneering axons in order to establish early connections, for example, in the optic system of fish [ 15, 16, 581 and birds [ 131, but is not essential for axonal regeneration after injury in the mature animal [13,16]. In the chick, this transient sensitivity to laminin by early growing axons is correlated with the simultaneous expression of laminin receptors on these axons [ 131. In mammals, in which reactive astrocytes produce laminin [37,38], and secrete the molecule into the basal lamina of the reconstituted glia limitans [5,35], axonal regeneration fails despite the laminin-rich environment created within the area of injury. Moreover, in a recent study of regeneration of CNS cholinergic axons in viva, laminin alone was found to be an insufficient substrate for axonal regeneration [34]. In experimental models where PN re-innervation of acellular grafts or regeneration chambers has been analyzed [3, 10,23,24,41,46,51,61], Schwann cells have been described either co-migrating with, or more typically, preceding the outgrowing neurites. Since axons have rarely been found in isolation within basal lamina tubes, it is reasonable to assume that, in the PNS, the primary event in the natural his-

tory of regeneration into acellular grafts is the migration of Schwann cells from the intact stump into the graft, and that axon invasion is secondary, perhaps organised by the products of secretion of the new population of Schwann cells. Even though axons that enter pre-existing basal lamina scaffolds in either muscle [18, 19, 30, 56, 571 or acellular nerve [23,28] do so exclusively along the inner aspects of the basal laminae (the lamina lucida, where laminin is concentrated [ll], although not necessarily accessible [62]), it is their association with Schwann cells that is the significant factor, rather than any possible trophic influence of the basal lamina per se [27,37]. Inhibition of Schwann cell mitosis [25] in a proximal PN stump anastomosed to an acellular PN segment [24], prevents both Schwann cell and axon invasion into the graft for 2-3 weeks, until Schwann cell replication recommences. Thereafter, Schwann cells first invade the empty basal lamina tubes of the grafted acellular PN segment, followed by axons which grow in from the proximal stump. In the experiments described here, no Schwann cells are available to repopulate the acellular PN grafts and hence, even though laminin is present, regeneration of CNS axons does not occur into acellular PN grafts. These findings raise interesting points of speculation about the role of Schwann cells in regeneration. Clearly, Schwann cells have an affinity for the inner aspect of basal lamina tubes, seeking out and preferentially moving over this surface. Since the latter activity of Schwann cells appears to be a primary event in regeneration, it seems unnecessary to propose that regenerating axons replicate this but rather that growing fibres preferentially adhere to Schwann cells which provide trophic support and guidance into and along the basal lamina tubes. ACKNOWLEDGEMENTS

We are grateful to Kevin Fitzpatrick for photography, Mass N’jie and Andrew Kent for technical help and Margaret Collins and Cecilia Rodrigues for secretarial assistance. This work was supported by the MRC, Wellcome Trust and EEC.

REFERENCES 1. Aguayo, A. J., A. Bjorklund, U. Stenevi and T. Carlsted. Fetal mesencephalic neurons survive and extend long axons across peripheral nervous system grafts inserted into the adult rat striatum. Nrurosci Left 45: 53-58, 1984. 2. Allcutt, D., M. Berry and J. Sievers. A quantitative comparison of the reaction of retinal ganglion cells to optic nerve crush in neonatal and adult mice. Bruin Rrs 16: 219230, 1984. 3. Anderson, P. N., J. Mitchell, D. Mayor and V. V. Stauber. An ultrastructural study of the early stages of axonal regeneration through rat nerve grafts. Neuroputhol Neurohiol 9: 455-466, 1983. 4. Benfey, M. and A. J. Aguayo. Extensive elongation of axons from rat brain into peripheral nerve grafts. Nature 296: 15&152, 1982. 5. Bernstein, J. J., R. Getz, M. Jefferson and M. Kelemen. Astrocytes secrete basal lamina after hemisection of rat spinal cord. Brain Res 327: 135-141, 1985. 6. Berry, M., L. Rees and J. Sievers. Regeneration of axons in the mammalian visual system. Exp Bruin Res Suppl 13: 1833, 1986. 7. Berry, M., L. Rees and J. Sievers. Unequivocal regeneration of rat optic nerve axons into sciatic nerve isografts. In: Neural Trunsplantation and Regeneration, edited by-G. D. Das and R. B. Wallace. New York: Sminaer Verlag. 1986. on. 6>79. 8. Berry, M., S. M. Hall, E.-L. Rees, P. %u and j: Sievers. The role of basal lamina in axon regeneration. In: MesenchymeEpithrlial Interactions in Neural Development. NATO ASI Series, Vol H5, edited by J. R. Wolf, J. Sievers and M. Berry.

Berlin: Springer-Verlag,

1987, pp. 361-383.

9. Bregman, B. S. and P. J. Reier. Neural tissue transplants rescue axotomised rubrospinal cells from retrograde death. J Camp Neurol 244: 8695,

1986.

10. Bunge, R. P. Contribution of tissue culture studies to our understanding of basic processes of peripheral nerve regeneration. In: Post Traumatic Peripheral Nerve Regeneration. Experimental Basis and Clinicul Implications, edited by A. Gorio. H. Millesi and S. Migrino. New York: Raven Press: 1982, pp. 105-113. 11. Carbonetto, S. The extracellular matrix of the nervous system. Trends Neurosci

7: 382-387,

1984.

12. Cattaneo, D. I fenomeni degenerativi e regenerativi nelle vie visive in sequito a lesioni del nervo ottico. Riv Patol Nerv Ment 28: 61-l 18, 1923.

13. Cohen, J., J. F. Burne, C. McKinlay and J. Winter. The role of laminin and the laminin/tibronectin receptor complex in the outgrowth of retinal ganglion cell axons. Dev Biol 123: 407-418, 1987. 14. David, S. and A. J. Aguayo. Axonal regeneration after crush injury of rat central nervous system tibres innervating nerve grafts. J Neurocytol 14: l-12, 1985. IS. Easter, S. S., Jr., A. C. Rusoff and P. E. Kish. The growth and organisation of the optic nerve and tract in juvenile and adult goldfish. J Neurosci 1: 793-811. 1981. 16. Easter, S. S., Jr., B. Bratton and S. S. Scherer. Growth related order of the retinal tibre layer in goldfish. J Neurosci 4: 21732190, 1984.

230

BERRY,

17. Easter,

S. S., Jr. Retinal axons and the basal lamina. In:

Mesenchyme-Epithelial Interactions in Neural Development. NATO ASI Series. Vol H5. edited bv J. R. Wolf. J. Sievers and

M. Berry. Berlin:‘Springer-Verlag, 1987, pp. 385-396. 18. Glasby, M. A., S. E. Gschmeissner, R. J. I. Hitchcock and C. L.-H. Huang. The dependence of nerve regeneration through muscle grafts in the rat on the availability and orientation of basement membrane. J Neurocytol 15: 497-510, 1986. 19. Glasby, M. A., S. E. Gschmeissner, R. J. I. Hitchcock, C. L.-H. Huang and B. A. de Souza. A comparison of nerve regeneration through nerve and muscle grafts in rat sciatic nerve. Neuroorthopedics 2: 21-28, 1986. 20. Graftstein, B. and N. A. Ingoglia. Intracranial transection of the optic nerve in adult mice: preliminary observations. Exp Neural

76: 318-330, 1982. 21. Gulati, A. K., A. H. Reddi and A. A. Zalewski. Changes in the basement membrane zone components during skeletal muscle tibre degeneration and regeneration. J Celi Biol 97: 957-962, 1983. 22. Haftek,

J. and P. K. Thomas. Electron-microscope observations on the effects of localized crush injuries on the connective tissues of peripheral nerve. J Anat 103: 233-243, 1968. 23. Hall, S. M. Regeneration in cellular and acellular autografts in the peripheral nervous system. Neuropathol Appl Neurobiol 12: 27-46, 1986. 24. Hall, S. M. The effect of inhibiting Schwann cell mitosis on the

re-innervation of acellular autografts in the peripheral nervous system. Neuropathol Appl Neurobiol 12: 401-414, 1986. 25. Hall, S. M. and N. A. Gregson. The effects of mitomycin C on the process of remyelination in the mammalian peripheral nervous svstem. Neurooathol ADDI Neurobiol 1: 65-78, 1977. 26. Hammarback, J. A:, S. L. Palm, L. T. Furcht and P. C. Letourneau. Guidance of neurite outgrowth by pathways of substratum-absorbed laminin. J Neurosci Res 13: 213-220, 1985. 27. Hopkins, J. M., T. S. Ford-Holevinski,

J. P. McCoy and B. W. Agranoff. Laminin and optic nerve regeneration in the goldfish.

J Neurosci 5: 3030-3038, 1985. 28. Ide, C., K. Tohyama, R. Yokota, T. Nitatori and S. Onodera. Schwann cell basal lamina and nerve regeneration. Brain Res 283: 61-75, 1976. 29. Kaufman, H. E. and T. J. Zimmerman. Evidence for two retinal control loops. Curr Top Ophthalmol 6: 111-121, 1979.

YIU AND SIEVERS

80: 45%470,

1917.

47 Osborne, N. N. and V. H. Perry. Effect of neonatal optic nerve transection on some classes of amacrine cells in the rat retina. Brain Res 343: 230-235, 1975. 48. Perry, V. H. Evidence for an amacrine cell system in the ganglion cell layer of the rat retina. Neuroscience 5: 931-944, 1981. 49. Perry, V. H., Z. Henderson and R. Linden. Postnatal changes in

50.

51.

275-281, 1984. 31. Kieman, J. A. An explanation

52,

1984.

HALL,

39. Linden, R. and V. H. Perry. Ganglion cell death within the developing retina: a regulatory role for retinal dendrites. Ncurrjscience 7: 2813-2827, 1982. 40. Liotta, L. A., R. H. Goldfarb, R. Brundage, G. P. Siegal, V. Terranova and S. Garbisa. Effect of plasminogen activator (urokinase), plasmin, and thrombin on glycoprotein and collagenous components of basement membrane. Canc,c,r Rrs 41: 46294636, 1981. 41 Lundborg, G., L. B. Dahlin, N. P. Danielsen, H. A. Hansson and K. Larsson. Reorganisation and orientation of regenerating nerve libres, perineurium, and epineurium in preformed mesothelial tubes-an experimental study on sciatic nerve grafts. J Neurosci Res 4: 265-281, 1981. 42. McConnell, P. and M. Berry. Regeneration of ganglion cell axons in the adult mouse retina. Brain Res 241: 362-365, 1982. 43. Mesulam, M-M. Principles of horseradish peroxidase neurohistochemistry and their applications for tracing neural pathways-axonal transport enzyme histochemistry and light microscopical analysis. In: Tracing Neural Connections with Horseradish Peroxidase, edited by M-M. Mesulam. Chichester: Wiley, 1982, pp. 1-152. 44. Miller, N. M. and M. Oberdorfer. Neuronal and neuroglial responses following retinal lesions in the neonatal rats. .I Camp Nemo1 202: 493-504, 1981. 45. Misantone, L. J., M. Gershenbaum and M. Murray. Viability of ganglion cells after optic nerve crush in rats. J Neurocyfoi 13: 449-465, 1984. 46. Nageotte, J. Sur la greffe des tissus morts et en particulier sur la reperation des pertes de substance des nerfs a l’aide de greffons nerveux conserves dans l’alcool. Comp Rend Sot Bioi (Paris)

30. Keynes, R. J., W. G. Hopkins and C. L.-H. Huang. Regeneration of mouse peripheral nerves in degenerating skeletal muscle: guidance by residual fibre basement membrane. Brain Res 295: of axonal regeneration in peripheral nerves and its failure in the central nervous system. Med Hvootheses 4: 15-26, 1978. 32. K&man, J. A. Hypotheses concerned with axonal regeneration in the mammalian nervous system. Biol Rev 54: 155-159, 1979. Methods: 33. Kieman, J. A. Histological and Histochemical Theory and Practice. Oxford: Pergamon Press, 1981, pp. 272273. 34. Kromer, L. F. and C. J. Combrooks. Transplants of Schwann cell cultures promote axonal regeneration in the adult mammalian brain. P&c Nat1 Acad Sci USA 82: 6330-6334, 1985. 35. Kruger. S.. J. Sievers. C. Hansen, M. Sadler and M. Berry. Three morphologically’ distinct types of interface develop between adult host and fetal brain transplants: implication for scar formation in the adult central nervous system. J Comp Neurol 249: 103-116, 1986. 36. Liesi, P., D. Dahl and A. Vaheri. Neurons cultured from developing rat brain attach and spread preferentially to laminin. J Neurosci Res 11: 241-251, 1984. 37. Liesi, P. Laminin-immunoreactive glia distinguish regenerative adult CNS systems from non-regenerative ones. EMBO J 4: 2505-2511, 1985. 38. Liesi, P., S. Kaakkola, D. Dahl and A. Vaheri. Laminin is induced in astrocytes of adult brain by injury. EMBO J 3: 683-686,

REES,

53.

retinal ganglion cell and optic axon population in the pigmented rat. J Comp Neural 219: 356-368, 1983. Politis, M. J., K. Ederle and P. S. Spencer. Tropism in nerve regeneration in viva. Attraction of regenerating axons by diffusable factors derived from cells in distal stumps of transected peripheral nerves. Bruin Res 253: l-12, 1982. Pollard, J. D. and J. G. McLeod. Fresh and predegenerate nerve allografts and isografts in Trembler mice. Muscle Nerve 4: 274281, 1981. Ramon Y Cajal, S. Degeneration and Regeneration in the Nervous Svstem. London: Oxford University Press, 1929, pp. 583-596. Richardson, P. M., V. M. McGuiness and A. J. Aguayo. Axons from CNS neurones regenerate into PNS grafts. Nature 284: 264-265.

1980.

54. Richardson, P. M., V. M. K. lssa and S. Shemie. Regeneration and retrograde degeneration of axons in the rat optic nerve. J Neurocytol 11: 949-966, 1982. 55. Richardson, P. M., V. M. K. Issa and A. J. Aguayo. Regeneration of long axons in the rat. J Neurocytol 13: 165-182, 1984. 56. Sanes, J. R. Roles of extracellular matrix in neural development. Am Rev Physiol 45: 581-600, 1985. 57. Saws. J. R.. L. M. Marshall and U. J. McMahan. Reinnervation of muscle fiber basal lamina after removal of myofibers. Differentiation of regenerating axons at original synaptic sites. J Cell Biol 78: 176198,

1978.

58. Scherer, S. S. and S. S. Easter. Degenerative and regenerative changes in the trochlear nerve of goldfish. J Neurocytol 13: 519-565,

1984.

59. So, K-F. and A. J. Aguayo. Lengthy regrowth of cut axons from ganglion cells after peripheral nerve transplantation into the retina of adult rats. Brain Res 328: 349-354, 1985. 60. Thanos, S. and F. Bonhoeffer. Investigations on the development and topographic order of retinotectal axons and perikarya with rhodamine in viva. .I Comp Neurol 219: 420-430, 1983.

OPTIC AXON REGENERATION 61. Williams, L. R., H. C. Powell, G. Lundborg and S. Varon. Competence of nerve tissue as distal insert promoting nerve regeneration in a silicone chamber. Brain RPS 293: 201-211, 1984.

231

62. Yurchenco, P. D., E. C. Tsilibary, A. S. Charonis and H. Furthamayr. Models for the self-assembly of basement membrane. J Histochrm Cytochrm 34: 93-102, 1986.