Axonal growth within poly (2-hydroxyethyl methacrylate) sponges infiltrated with Schwann cells and...

BRAIN RESEARCH

E L S E V I E R Brain Research 671 (1995) 119-130

Research report

Axonal growth within poly (2-hydroxyethyl methacrylate) sponges infiltrated with Schwann cells and implanted into the lesioned rat optic

tract

Giles W. Plant a, Alan R. Harvey a,,, Traian V. Chirila b a Department qfAnatomy and Human Biology, The University of Western Australia, Nedlands, Perth, WA 6009, Australia

b Lions Eye Institute, Nedlands, Perth, WA 6009, Australia

Accepted 25 October 1994

Abstract

Porous hydrophilic sponges made from 2-hydroxyethyl methacrylate (HEMA) have a number of possible biomedical applications. We have investigated whether these poly(HEMA) hydrogels, when coated with collagen and infiltrated in vitro with cultured Schwann cells, can be implanted into the lesioned optic tract and act as prosthetic bridges to promote axonal regeneration. Nineteen rats (20-21 days old) were given hydrogel/Schwann cell implants. No obvious toxic effects were seen, either to the transplanted glia or in the adjacent host tissue. Schwann cells survived the implantation technique and were immunopositive for the low affinity nerve growth factor receptor, S100 and laminin. Immunohistochemical studies showed that host non-neuronal cells (astrocytes, oligodendroglia and macrophages) migrated into the implanted hydrogels. Astrocytes were the most frequently obserled host cell in the polymer bridges. RT97-positive axons were seen in about two thirds of the implants. The axons were closely associated with transplanted Schwann cells and, in some cases, host glia (astrocytes). Individual axons regrowing within the implanted hydrogels could be traced for up to 900/xm, showing that there was continuity in the network of channels within the polymer scaffold. Axons did not appear to be myelinated by either Schwann cells or by migrated host oligodendroglia. In three rats, anterograde tracing with WGA/HRP failed to demonstrate the presence of retinal axons within the hydrogels. The data indicate that poly(HEMA) hydrogels containing Schwann cells have the potential to provide a stable three-dimensional scaffold which is capable of supporting axonal regeneration in the damaged CNS.

Keywords: Hydrogel; Schwann cell; Regeneration; Rat visual system; Astrocyte; Immunohistochemistry

1. Introduction

Hydrogels are natural or synthetic water swollen polymers and consist of a three-dimensional network of macromolecules. There are several classes of hydrogels which have different characteristics, such as their porosity and hydrophilicity. One type of hydrogel is the hydrophilic sponge which is produced by phase-sep- aration polymerization of 2-hydroxyethyl methacrylate (HEMA) in excess water. These sponges are non-trans- parent (white), porous materials. Some potential biomedical applications were suggested in a number of early studies in animal models [6,14,15,57,60,63], how-

* Corresponding author. Fax: (61) (9) 380 1051. E-mail: [email protected]

0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0006-8993(94)01312-8

ever in recent years poly(HEMA) sponges have gener- ally not elicited the same interest as the homogeneous, t ransparent poly(HEMA) hydrogels. However, the use of poly(HEMA) sponge material has been revived in conjunction with their use as elements in a new type of artificial cornea [18-20,23]. Such sponges, when prepared in the presence of at least 75% water in the polymerization mixture, displayed interconnected pores larger than 10 /zm in diameter.

The porous nature of phase-separated poly(HEMA) hydrogels makes them permeable to low molecular weight molecules, gases and nutrients; when combined with natural polymers (collagen) they provide a material which is both mechanically stable and biologically acceptable to the host tissue. Collagen or other materials such as laminin added to the hydrogel act as a bioadhesive substrate which allows the at tachment of

120 G.W. Plant et al. /Brain Research 671 (1995) 119-130

cells to the surfaces of the polymer matrix [21]. Experi- ments in vivo using these sponges (rabbit) demon- strated massive biocolonization through fibroblastic invasion and growth, as well as neovascularization, both subcutaneously and in the corneal tissue [19,23]. An in vitro method using cultured human fibroblasts was also designed in order to assess the cellular penetration and proliferation in sponges, and the results indicated that the pore size is critical to cellular invasion [19]. This parameter can be conveniently controlled by varying the water content in the polymerization mixture.

In the present study, we have tested whether poly(HEMA) sponges may be a candidate for use as prosthetic bridges in the repair of the damaged central nervous system (CNS). Lesioned axons of mammalian CNS neurons show some ability to regrow over short distances [10,22,30,32,39]. In recent years a number of experimental approaches have been used in an attempt to provide an environment in the CNS which is more conducive to regenerative growth. These approaches include the implantation of fetal neurons, adult peripheral nerves, collagen matrices, hydrogels or cultured glial cells (Schwann cells or astrocytes) [1,9,17,24,34- 35,46,50,51,53,54,59,63,69,73,75]. Considerable success has been achieved with the use of peripheral nerves, their associated cells and matrices. The most important component of the peripheral nerve is the Schwann cell and its associated extracellular matrix molecules such as laminin. These matrix products along with the large number of growth factors released by the Schwann cells are important in the regeneration of CNS axons [2,3,41,42,49,52,70]. Studies in the visual pathways of the brain have shown that Schwann cells promote the survival of retinal ganglion cells (RGC) and the growth of their axons. This has been shown in vitro [4,43] and in vivo [9,11,68,73].

Most regeneration studies in the visual system have concentrated on using peripheral nerves or associated material in the region of the retina or optic nerve. In contrast, our laboratory has examined whether fetal grafts [39] and cultured glial cells [17,36,37] are able to stimulate regrowth of RGC axons at more central sites; specifically within the brachial region of the optic tract between the dorsal lateral geniculate nucleus (dLGN) and the superior colliculus (SC). The most recent work involved the use of polycarbonate tubes filled with purified cultures of Schwann cells [37]. This method was used in an attempt to provide a more controlled environment for transplanted cells but still allow exchange of bioactive molecules needed for Schwann cell survival and proliferation. At short survival times the cells supported axonal regrowth, but over an extended period there was an apparent collapse of the cellular architecture and the contents formed a dense cord in the centre of the tube.

To provide a more stable scaffold for the regenera-

tion of RGC axons, we have now used hydrogel sponges produced from poly(HEMA) with 12 /zm pores and infiltrated them with a purified population of Schwann cells. The advantages of using these gels are: (1) me- chanical stability, (2) their highly porous nature, allow- ing the movement (in and out) of soluble factors such as nutrients and growth promoting factors, (3) biocom- patibility and (4) their ability to entrap neuronal supporting cells (Schwann cells) within the structure. Hy- drogel sponges containing Schwann cells will hopefully encourage better deposition of extracellular matrix molecules in conjunction with the release of cellular- derived growth factors. This will supply the lesioned tract with a bridge more conducive to the regeneration of axons. In this first study using poly (HEMA) sponges in the CNS, we have qualitatively assessed the efficacy of this approach by using monoclonal and polyclonal antibodies and the indirect immunofluorescence technique. Axons regrowing within the hydrogels were stained using the antibody RT97 (which recognizes neurofilaments); the cellular content of implants was examined using antibodies specific for the low affinity nerve growth factor receptor (LNGFR), S-100, glial fibrillary acidic protein (GFAP), ED1, carbonic anhy- drase II (CAll), P0 and proteolipid protein (PLP). Three animals were processed for anterograde HRP tracing in an attempt to demonstrate the presence of RGC axons within the implanted hydrogels.

2. Materials and methods

HEMA was supplied by Ubichem Ltd., UK. Ethy- lene dimethacrylate, used as a crosslinking agent, was supplied by Tokyo Kasei Kogyo Co., Japan. Aqueous solutions (6% wt) of ammonium persulfate and sodium metabisulfite were used as initiators for polymerization. Sterile collagen type I was isolated from tendons of Wistar rat tails. A solution (3.7 mg/ml) was prepared in 0.04 M acetic acid (pH 2.9).

2.1. Synthesis o f sponges

The synthesis of hydrophilic poly(HEMA) sponges was described in detail elsewhere [19,20]. For the present experiments, a sponge was produced by polymeriz- ing HEMA in the presence of water (ratio HEMA/water was 20/80 by weight), 0.5% wt crosslinking agent and 0.12% wt initiator. The mixture was distributed in polypropylene moulds fitted in a mould- ing device designed by one of us (TVC). After replac- ing the air with nitrogen, the sealed unit was placed in a water bath provided with a microprocessor con- troller. A temperature programme was run in the bath, consisting of three 10-h steps at 30°C, 40°C and 50°C,

G. W. Plant et al. / Brain Research 671 (1995) 119-130 121

respectively. On completion of the cycle, the poly(HEMA) sponge buttons were removed from moulds and placed in deionized water, where they were stored for 2 weeks before sterilization, with at least two daily water exchanges. In a fully hydrated state, this sponge has an equilibrium water content of 73% wt, a permeability coefficient of 3.22 x 10-11, and pore diameters between 10 and 30/xm [19].

2.2. Impregnation with collagen

Sponge specimens were sterilized in an autoclave for 20 minutes at 130°C, then stirred for 24 h at room temperature in a sterile solution of collagen type 1. After specimens were saturated with collagen, they were removed and placed in DMEM at 37°C for 1 h to allow the crosslinking of collagen.

2.3. Schwann cell cultures

Schwann cells were prepared from sciatic nerves of 3-day-old rats (PVG/c). The method has been described elsewhere [17] and is modified from Brockes et al [13]. Nerves were desheathed, chopped and dissociated in HEPES-buffered modified Eagles medium (HMEM) (CSL) which contained 0.15% collagenase (Sigma) and 0.1% trypsin (Sigma). Following tritura- tion through needles, dissociated cells and any remain- ing clumps were plated on poly-L-Lysine (PLL) coated flasks and maintained in Dulbecco's modified Eagles medium (DMEM) containing 10% fetal calf serum (FCS) (Gibco). After a period of 2-3 days in vitro (DIV), ceils were treated with cytosine arabinoside (Sigma) for 24 h (concentration of 10 -5 M). The cells were then treated witlh an antibody to Thy-l.1 (1:5 dilution, Serotec) for 30 min followed by guinea pig complement (1:20 dilution, CSL) for 20 min. Both procedures help to purify cultures by eliminating Thy-1 expressing cells, i.e. fibroblasts [13].

Purified Schwann ce][l cultures were then induced to proliferate by the addition of 5 /zM forskolin (Sigma) and 0.4% glial growth factor-like conditioned medium (GGF-CM) (gift of Dr. E. Walls and Dr. P. Stroobant, Ludwig Institute for Cancer Research, London) [25]. After 5-11 DIV, Schwann cells (93-96% pure) were resuspended in a small volume of HMEM. In all cases the Schwann cells were labelled with a fluorescent dye bisbenzimide (Hoechst 33342) (Sigma). Schwann cells were exposed to the dye (10 /zg/ml) for 3 min and were carefully washed 3 x in HMEM prior to trypsinization and incubation with the sponges. Small pieces of dried sponge were placed in medium containing a high density of Schwann cells (5 x 105 cells in 3-5 /xl) with the aim of drawing the cells into the polymer structure as it rehydrated. After rehydration the pieces of hydrogel measured approximately 1 to 1.5 mm in

length, 0.5 to 1 mm in width and about 0.5 mm in depth.

2.4. Implantation of sponges with or without Schwann cells

Sponges containing Schwann cells were maintained in culture overnight (DMEM + 10%FCS) prior to implantation. All 19 rats (20-21 days old PVG/c) that received Schwann cell/polymer implants were anaesthetized by an intraperitoneal injection of 2,2,2-tri- bromoethanol (250 mg/kg). The lesion and general implantation procedures have been described in detail elsewhere [17,37]. Briefly, in each animal a small part of the left cerebral hemisphere was removed to expose the left brachial area and rostral SC. Sponges were inserted in a rostral caudal direction in lesion cavities made in the superior brachium, between the dLGN and SC. They thus formed a bridge between the lesioned optic tract and its midbrain target. Three rats were lesioned and received implants of hydrogels that were not impregnated with Schwann cells.

2.5. Immunohistochemical analysis

Sixteen rats with Schwann cell/hydrogel implants were examined 63-193 days after surgery. Animals were deeply anaesthetized (Nembutal, i.p.) and perfused with 1% sodium nitrite wash followed by 4% paraformaldehyde in Sorensons phosphate buffer (pH 7.4). The three rats which received cell-free implants were perfused 113 days after surgery. Brains were postfixed for 2 h and then placed in 30% sucrose overnight (cryoprotection). Frozen sections (40 /zm) were cut and sections containing sponges were rinsed in Dulbecco's phosphate buffered saline (DPBS) for 10 min prior to immunohistochemistry. An indirect im- munolabelling method was employed using antisera diluted in DPBS containing 0.1% bovine serum albu- min (BSA) to reduce background labelling, and 0.2% Triton to assist antibody penetration. Incubation was overnight at 4°C.

Monoclonal anti-neurofilament antibody RT97 (Boehringer Mannheim) was used at a dilution of 1/100 (as a marker for axons); polyclonal anti-laminin (Be- thesda Research Laboratories) at 1/200 (as a marker for Schwann cells [44], astrocytes [7] and new blood vessel growth); polyclonal anti-glial fibrillary acidic protein (GFAP) (Dakopatts) at 1/200 (as a marker for astrocytes and non-myelinating Schwann cells); polyclonal anti-S100 (Dakopatts) at 1/200 (as a marker for Schwann cells [44] and astrocytes [8]); monoclonal (192- Ig) anti-low affinity nerve growth factor receptor (LNGFR) (gift from Dr. Eugene M. Johnson Jr., Wash- ington University School of Medicine) at 1/250 (as a marker for Schwann cells [44]), ED1 (Serotec) at 1/200 (as a marker for maerophages [26]), CAII (Binding


Site) at 1 /200 (as a marker for ol igodendrogl ia [40]), ant i -P 0 (gift from Dr. B. Trapp, Johns Hopkins, Balti- more) at 1 /500 (as a marker for per ipheral myelin), an t i -PLP (gift from Dr. N.A. Gregson, UMDS, Guy 's campus, London) at 1 /250 (as a marker for central myelin). After the pr imary ant ibody incubat ion, sec- t ions were washed in DPBS 3 x 5 min and then incu-

I

bated in a secondary ant ibody for 2 h (gently shaken)

at room tempera ture . The secondary ant ibodies were ei ther an t i -mouse F I T C ant ibody or ant i - rabbi t T R I T C . Double labell ing with RT97 and ant i - laminin , or RT97 and a n t i - G F A P was carried out by the s imul taneous addi t ion of both ant ibodies to the sections. Ant i - rabb i t T R I T C and ant i -mouse F I T C were also added simulta-

Fig. 1. A,B: immunofluorescence photomicrographs (same field) showing Hoechst (A) and LNGFR (B) staining of cultured Schwann cells. C: Schwann cells showing Hoechst labelling inside the poly(HEMA) sponge at 24 hours in vitro. D: LNGFR ÷ Sehwann cells at the surface of an implanted sponge. E,F: Hoechst ÷ Schwann cells within the sponges in vivo (E) and the corresponding laminin immunoreactivity (E). The arrowheads show labelled nuclei. G: S100 + cells within a sponge and in the adjacent host tissue. H,I: two examples of CAll + oligodendroglia within implanted hydrogel sponges. J,K: GFAP + astrocytes within the sponge structures (J), some associated with blood vessels (K). In K, the large arrow points to an astrocyte with endfeet (small arrows) abutting a blood vessel (by). P, poly(HEMA) sponge; H, host tissue. Bars: A-C = 50 ~m; D = 10/zm; E,F = 25/xm; G = 100/~m; H,I = 10 p,m; J,K = 25 ~tm.

G.W. Plant et al./Brain Research 671 (1995) 119-130 123

Table 1 Immunohistochemical analysis

Number Host age at t ime Post-op Number of implants Number of complete Number with of rats of surgery survival attached to dlGN bridges (dLGN-SC) RT97 + axons

With Schwann cells 16 20-21 days 63-193 days 13 11 11 No Schwann cells 3 20-21 days 113 days 2 0 0

neously to the double labelled sections. Once stained the sections were washed in DPBS (3 × 5 min) mounted onto subbed slides and allowed to adhere, care being taken not to let the sponge pieces dry out (which would cause the structure to collapse). The slides were then coverslipped with Citifluor (an anti-quenching agent).

2.6. Anterograde tracing

Three rats with Schwann cel l /hydrogel implants into the optic tract were assessed for retinal axon regrowth 93-100 days after surgery. The animals were anaesthetized with halothane and their right eye in- jected with 2 - 3 / z l of a mixture of 0.75% wheat germ agglut inin/horseradis :h peroxidase ( W G A / H R P ) (Sigma) and 15% HRP (Boehringer) in 1% dimethyl sulfoxide. Rats were perfused 24 h later with 2% glutaraldehyde and 0.4% paraformaldehyde in 0.1 M phosphate buffer (pH ;'.4). Fixed brains were encapsu- lated in albumen-gelatin mixture, placed into 30% sucrose overnight and 40/xm frozen sagittal sections cut. Sections were reacted with tetramethylbenzidine to show anterograde HRP labelled retinal axons.

2.7. Semi-quantitative analysis

The number and distribution of axons (RT97+), astrocytes (GFAP ÷) and macrophages (ED1 ÷) within the implanted hydrogels and in surrounding host tissue was examined. Three iimmunohistochemically stained 40 /zm sections per implant were analyzed, each section being separated by 40/.tm. The length of individual axons within the hydrogel, the furthest distance axons were seen from the nearest CNS neuropil and the extent of host astrocyte migration into the polymers was measured (using an eyepiece graticule: 100 divi- sions = 235 ~ m at x 50). The volume of hydrogel containing axons or astrocytes was then assessed as a percentage of the estimated total hydrogel volume.

3. Results

3.1. Schwann cell purity

Schwann cells used for transplantation within the hydrogels were labelled with Hoechst 33342, which showed no toxic effects and clearly labelled the cell

nuclei (Fig. 1A). The cultures were stained immunocy- tochemically with an antibody to LN G F R (Fig. 1B). Cultures were also stained with anti Thy-l.1 to show the level of any fibroblast contamination. The cultures used in the sponges were approximately 93-96% pure. Fig. 1C shows Hoechst labelled Schwann ceils within the sponge structure after 1 DIV and prior to transplantation.

3.2. Hydrogels (no Schwann cells)

Three rats which received a suction lesion of the left optic tract followed by a transplant of poly(HEMA) hydrogel which did not contain purified Schwann cells showed no evidence of regenerating RT97-positive (RT97 ÷) axons. The position of the hydrogel sponges was not ideal; while two of the three implants were attached to the dLGN in no case did the gels form a complete bridge across the lesion site. Surrounding the implants were ED1 ÷ macrophages but few were seen within the implanted polymer structure. The ED1 ÷ macrophages had a foamy appearance presumably due to the debris produced from the lesion. The number of ED1 + macrophages surrounding the implant was between 77-105 (per 40 /zm section). GFAP immunohistochemistry revealed intense staining of the host astrocytic population surrounding the implants and also a small number of reactive astrocytes which had migrated into the poly(HEMA) sponges. Anti-laminin immunohistochemistry showed the presence of new blood vessels within the poly(HEMA) hydrogels and also stained the reactive astrocytes which had migrated into the sponges. A small number of CAll + oligodendroglia were seen within the polymer implants.

3.3. Hydrogels containing Schwann cells

The 12 /zm pore sponges coated with collagen I proved to be biocompatible and were not rejected by the host tissue. Sponges showed no sign of collapse and degradation even after a period of 193 days in vivo. The size of the implanted hydrogels - - as measured from the tissue sections - - ranged from 0.9 to 1.5 mm in length, 0.5 to 0.9 mm in width and 0.3 to 0.6 mm in depth (average volume approximately 0.3 mm3). Im- planted polymers containing Schwann cells were attached to the dLGN in 81% of the animals. In 68% of animals with hydrogel implants, the polymer formed a


complete bridge from the dLGN to the target SC (Table 1).

Cellular content and immunohistochemical charac- terization. Fluorescence microscopy was used to iden- tify Hoechst labelled Schwann cells which had been entrapped within the poly(HEMA) hydrogels. The number of entrapped Hoechst + Schwann cells varied between each implanted hydrogel; many but not all of these cells were immunopositive for LNGFR, S100, or laminin. LNGFR expression was evident on Schwann cells entrapped within the hydrogel up to approxi-

mately 120 days in vivo (Fig. 1D). After this time, either LNGFR expression was too low for detection by this method or the expression was lost. LNGFR immunoreactivity was particularly evident on Schwann cells located close to the surface of the transplanted hydrogel sponges where many of the surviving Schwann cells were found in this transplantation paradigm. LNGFR ÷ staining could be seen surrounding blood vessels within the sponges and around blood vessels in the host near to the polymer implant. The pattern of LNGFR staining within the hydrogel sponges (up to a

Fig. 2. A: anti laminin staining of a poly(HEMA) sponge (P) initially infiltrated with cultured Schwann cells, 85 days post implantation. The arrow shows the junction between the polymer and host dLGN. B: same field showing RT97 ÷ axons within the hydrogel structure. C: higher power view of B showing RT97 ÷ axons running in various orientations within the sponge. Individual axons could be traced for considerable distances (arrows). D: hydrogel lying within the lesioned optic tract containing GFAP + astrocytes (163 days post implantation). Note the intense GFAP immunoreactivity in the host brain adjacent to this polymer and the non-homogeneous distribution of astrocytes within the hydrogel. This implant did not contain any RT97 + axons, dLGN, dorsal lateral geniculate nucleus; SC, superior colliculus; H, host tissue. Bars: A and B = 200 /zm; C = 50 p~m, D = 200/xm.

G.W. Plant et a l . / Brain Research 671 (1995) 119-130 125

period of 120 days in vivo) corresponded to the uptake of Schwann cells into the hydrated sponge ie. non-uni- form distribution. Ent rapped Hoechst-labelled Schwann cells expressed laminin (Fig. 1E,F). This expression on the Hoechst labelled cells resembled that seen on Schwann cells within polycarbonate tubes implanted into the CNS [37]. Note however that S100 and anti-laminin expression was not solely confined to Schwann cells but was also found on astrocytes (see below).

The most intense GFAP immunoreactivity was usually seen around the periphery of the implanted sponges abutting the host tissue; this was particularly obvious in 5 animals (Fig. 2D). GFAP + astrocytes were also seen throughout the structure of the poly(HEMA) hydrogels. These astrocytes had extensive processes and appeared normal in their morphology (Fig. 1J and K, Fig. 2D). Astrocytes were sometimes seen to be closely associated with blood vessels (bv) within the transplanted matrix (Fig. 1K). In polymers containing relatively large numbers of GFAP + astrocytes, the distribution was not homogeneous and the glial cells tended to be in clumps (Fig. 2D). The volume of hydrogel which contained host astrocytes varied. Six animals with hydrogel/Schwann cell implants contained large numbers of GFAP ÷ astrocytes (occupying approximately 50-60% of the total polymer matrix volume). Five hydrogel implants contained fewer astrocytes (occupying about 30% of the hydrogel volume). GFAP ÷ astrocytes were fewer in areas which contained large numbers of Hoechst labelled Schwann cells; where Schwann cell numbers were lower, astrocytes could be seen in close proximity and in larger numbers.

Anti-laminin staining showed a pattern of intense immunoreactivity around the periphery of the hydrogel sponges. Laminin expression was seen in association with astrocytes both within and immediately surrounding the hydrogel implant. Laminin ÷ structures could also be seen throughout the whole matrix network. Fig. 2A shows a sponge stained intensely with laminin. The large arrow indicates the attachment of the polymer to the dLGN. Some staining was clearly indicative of new blood vessel growth; note the dLGN stained very intensely.

Sections stained with anti-S100 showed the presence of immunopositive Schwann cells but this antibody also showed the presence of reactive astrocytes [8] within the transplanted hydrogel. S100 staining was very intense (Fig. 1G) and in many instances was similar to that seen in GFAP stained sections. CAll stained sections showed a normal distribution of oligodendrocytes in the host brain surrounding the implant. In the hydrogel sponges, CAll staining was relatively weak and only small numbers of CAll ÷ oligodendroglia were seen to have migrated into the polymers (Fig. 1H and I).

The number and distribution of macrophages surrounding the implanted hydrogel was also ascertained using the antibody ED1. ED1 ÷ macrophages were seen in all the experimental animals containing Schwann cell implants. The number of ED1 ÷ ceils surrounding the implants varied across the spectrum of animals processed, from as few as 39 to as many as 306 per 40 /~m section. Overall the number of ED1 ÷ macrophages surrounding the Schwann cell containing hydrogels was higher than the number seen in the hydrogels not containing Schwann cells. Particularly high numbers of ED1 ÷ macrophages were also seen in animals that showed high expression of GFAP ÷ astrocytes surrounding the implant.

Axonal growth within poly(HEMA) sponges containing Schwann cells. Immunohistochemical analysis of transplanted sponges infiltrated with Schwann cells showed that these implants provided a suitable matrix for axonal regrowth. Eleven of the 16 implants (68%) contained RT97 ÷ axons (Table 1). In the example shown in Fig. 2B and C, RT97 ÷ axons can be seen in various orientations throughout the matrix of the implanted sponge. The hydrogels in the five animals which did not contain RT97 ÷ axons were surrounded by the densest astroglial reaction (e.g. Fig. 2D).

The number of axons in hydrogels impregnated with Schwann cells varied from animal to animal and with the age of the implant. The most extensive growth was seen at 85 days. The density of axons (occupying approximately 60-70% of the total hydrogel volume) was relatively high in this example (Fig. 2B,C). Other animals did not have such a high density of axons within the hydrogels, nonetheless axons were found in about 5-35% of the total hydrogel volume (mean = 25%). In RT97-immunostained sections of the hydrogel implants, individual regenerating axons could be followed for considerable distances (see example arrowed in Fig. 2C). The longest individual axon was traced for 900 /zm. RT97 ÷ axons were found up to a distance of 500 /xm from the nearest host brain region.

Axons regenerating within the hydrogel were not only found in areas which contained Schwann cells but they were also seen in regions where astrocytes were present. Regenerating RT97 ÷ axons found within the structure of the hydrogel sponges were not immuno- reactive for P0 and were therefore unlikely to be myelinated by Schwann cells. Some sections were stained for proteolipid protein (PLP) but no evidence of central myelin was found. In sections double labelled with RT97 and anti-GFAP, RT97 ÷ regenerating axons, when associated with astrocytes, grew close to or on their surfaces. The axons in close association with astrocytes were in transplants that had a reduced astrocytic response surrounding the implant. The low astrocytic activity (low expression of GFAP) around the implant perhaps indicated a reduced scarring and


therefore better access for regenerating axons. In two animals, axons were present in membranes associated with the surfaces of implants. These membranes were GFAP ÷ and laminin ÷.

Retinal axonal growth (hydrogels + Schwann cells). Three animals (93-100 days post surgery) were studied to ascertain if any of the regenerating axons were retinal ganglion cell axons. Two of these implants were attached to the dLGN and formed complete bridges across the lesion site, however neither implant contained HRP labelled retinal axons inside the polymer matrix. Retinal axons were, however, observed to regrow for 600-700/xm in membranes which formed on the dorsal surface of the implanted hydrogels.

4. Discussion

It is only recently that neuroscientists have begun to investigate the potential use of hydrogels to assist in the reconstruction of damaged CNS tissue and tracts. Implants of cell-free hydrogels made from poly(glyceryl methacrylate) (poly (GMA)) and impregnated with collagen have been shown to support host cellular infiltra- tion and the growth of damaged axons [74]. Such gels produced little inflammatory response and proved to be compatible with host tissue.

The new developments reported here involved the use of a different type of hydrogel sponge and the incorporation of cultured Schwann cells into the polymer matrix prior to implantation into the CNS. We have shown that poly(HEMA)/collagen sponges provided a three-dimensional network which possessed a highly porous, mechanically stable structure. The hydrogels were able to house a purified population of glial cells and provided a scaffold for regenerative growth of axons in the lesioned rat optic tract.

4.1. The presence of migratory host cells within implanted hydrogels

Implanted poly(HEMA) hydrogels contained cell types derived from the host including astrocytes, macrophages and oligodendroglia. Cellular migration into poly(GMA) hydrogels has also been reported; addition of polar groups to these polymers resulted in a decrease in astrocytic scarring as well as increased cellular migration into the transplanted gel [74]. Mi- grating astrocytes were usually accompanied by new blood vessel formation [74]. Cationic hydrogels were also shown to allow greater deposition of extracellular matrix products than did anionic gels [74,76]. These results along with those of other investigators have shown that modification of hydrogels, either chemically or structurally will affect the degree of cell adhesion and proliferation [35,48,56,77] and neurite growth [16].

Incorporation of biological macromolecules or functional groups into polymer matrices can give the materials an appropriate environment for tissue ingrowth [48,65-67,71].

The type of cells migrating into poly(GMA) hydrogels was not known although they were assumed to be astrocytes [74,75]. In the present study we have shown that GFAP + astrocytes are the main migratory cell type which populate poly(HEMA) sponges from the surrounding host tissue. Small numbers of CA/I + oligodendroglia and ED1 ÷ macrophages were also present in hydrogels. The migration of these host cell types into implanted polymers occurred irrespective of whether the poly(HEMA) sponges were initially seeded in vitro with Schwann cells, however the number of host non-neuronal cells appeared to be less in the initially cell-free implants. There were also fewer ED1 ÷ cells surrounding these cell-free hydrogels.

4.2. Regrowth of axons within hydrogels containing Schwann cells

Previous experiments have shown that Schwann cells placed onto or within polymers and then transplanted into the CNS can support the regrowth of lesioned axons [17,36,37,54]. In our early studies using nitrocellulose implants it was shown that retinal axons re- grew up to a distance of 1120/xm, but the number of regrowing axons was limited [17]. These nitrocellulose implants provided a stable two-dimensional framework but gave little or no protection to the peripheral glia placed on the polymers and implanted within the damaged CNS environment. Indeed, in mice surgical models using a Y-chromosome specific marker, it has been shown that there is extensive migration and intermixing of donor and host glia on such nitrocellulose implants [38]. After implantation into the rat optic tract, many oligodendroglia were found on the polymer surface and myelination of the regrowing axons was found to be mostly CNS in origin [17]. These migrated oligodendroglia may have provided a major environmental ob- stacle to the regenerating axons via their expression of membrane inhibitory factors [45,62,64].

In later studies, we attempted to provide a less exposed environment for the transplanted Schwann cells by placing them within polycarbonate tubes prior to implantation into the rat optic tract [37]. These tubes with 12 /~m pores still allowed the exchange of bioactive molecules and nutrients. Using this method, extensive axon growth was seen in the short term, but there were fewer axons seen at longer survival periods, perhaps because the cellular material inside the tube was unstable and over time collapsed to form a dense cord in the center of the tube [37]. A recent report by Montgomery and Robson [54] also using polycarbonate tubes filled with Schwann cells showed that CNS axons

G.W. Plant et aL /Brain Research 671 (1995) 119-130 127

could be induced to regrow when the polymers were transplanted to the thalamus. The contents of the tube formed a dense network covered by a perineurial sheath which resembled the structure of a peripheral nerve. Regenerating axons were closely associated with the Schwann cells and were myelinated by them.

In the present study, Schwann cells entrapped within poly(HEMA) hydrogel sponges provided a favourable milieu for axonal growth within the lesioned optic tract. Axonal growth was present in 68% of the poly(HEMA) hydrogels processed at different survival times, however the degree of axonal regrowth varied, perhaps due to the number of surviving Schwann cells and their distribution within the polymers. The maxi-, mum distance that an individual RT97 ÷ axon was traced within a section of Schwann cell/hydrogel polymer was 900 ~m. This is a considerable distance and shows that there was a significant degree of continuity in the porous network within the hydrogel polymer implant. Most regenerating axons were found to have grown for a distance of between 200 and 500/zm from the nearest adjacent r~;gion of host neuropil.

The exact origin of the RT97 ÷ regenerating axons within the implant remains unknown. The axons could have originated from the dLGN, the floor of the lesion (thalamus) or some may have originated from the SC end of the polymer. In a small number (3) of animals we were unable to dletect retinal axons within the hydrogels. Other authors have shown that the distance between the neuronal cell body and the location of the growth promoting graft or implant affects the degree of axonal regeneration [2',7,58]. In the present paradigm the Schwann cells within the polymer would be in closer proximity to the thalamic or tectal neurons than the more distant RGCs and therefore might exert a stronger growth promoting effect on these local neurons. Retrograde labelling of regenerating axons within these polymers is currently being investigated to ascertain their exact origin.

The number of surviving Schwann cells appeared to be lower than that seen in polycarbonate tubes [37]. In hydrogel implants, the morphology and expression of specific antigens by Schwann cells was consistent with that described by other authors [12,28,31]. LNGFR immunoreactivity seen within or surrounding blood vessels in the hydrogells was similar to the pattern of staining after transplantation of Schwann cells to the fimbria and hippocampus [12]. The finding that Schwann cells congregated around blood vessels is also similar to that seen by Baron-Van Evercooren et al. [5], when they transplanted Hoechst-labelled Schwann cells to the developing mouse brain. As in the study of Brook et al. [12] we no~!ed an apparent loss of LNGFR expression on Schwann cells at long survival times. This may reflect cell death or transplanted Schwann cells may have come into contact with host central

axons and hence down regulated the expression of LNGFR [44,47,72]. However, if the latter explanation is correct, the lack of P0 staining suggests that the grafted Schwann cells had either lost their ability to myelinate, or the diameter of the axons or signals emanating from the axons were not sufficient to trigger myelinogenesis (cf. [37]).

No fibres were seen in animals in which there appeared to be a dense and extensive astrocytic scar (high GFAP immunoreactivity) at the interface between the implant and host brain. These scars have been suggested to form a barrier for any regenerating fibres [57]. Lower levels of GFAP expression seen in the host brain surrounding the implants in other animals did not appear to hinder the entry of regenerating axons. Indeed, astrocytes found within the confines of the hydrogel sponges were seen in close association with RT97 + axons and may have provided a surface suitable for regeneration of axons. Reactive astrocytes found within the hydrogels compare to those seen on the surface of nitrocellulose membranes inserted into the rat cerebral cortex [61]. The presence of regenerating axons in close proximity to astrocytes in the Schwann cell containing hydrogels is of interest because regrowth was not observed in the small number of con- trol (initially cell-free) hydrogels. Perhaps the presence of Schwann cells provided an initial environment which facilitated the entry of axons and the growth supporting process was then taken over by the migrated host astrocytes. Whether the support for regenerating axons is in the nature of trophic factors or other substrate molecules or both is uncertain. It will be of interest to increase t h e number of Schwann cells within the hydrogels and determine whether, in this circumstance, astrocytes are also seen in close association with regenerating axons. Interestingly, the presence of Schwann cells in significant numbers in tubular implants in the peripheral nervous system has recently been shown to modify an astrocytic environment, by reducing the inhibitory effects of the CNS glia [34].

4.3. Conclusion

The combination of hydrogels with cultured glial cells has provided us with a new type of prosthetic device which can be manipulated in vitro, prior to transplantation into the CNS. This initial study has shown the efficacy of using such a paradigm but fur- ther work is necessary to optimize the procedure. Im- proving Schwann cell survival within the gels, enhanc- ing the relationship between regenerating axons and the entrapped supporting cells, and providing a more oriented trabecular network within the hydrogel scaffold are three such steps to follow. Work underway involves the introduction of additional extracellular matrix products into the hydrogel structure. Collagen 1


was used in this paradigm but this is not necessarily an ideal substrate for Schwann cells transplanted into the CNS due to its inhibitory effect on proliferation [29]; note, however that Schwann cells must stop proliferat- ing before myelinogenesis can procede [44]. Other pos- sibilities for improving cell survival and axonal regrowth are the chemical attachment of extracellular matrix sequences to the hydrogel structures them- selves, or the active polymerization of hydrogels around living cells in vitro prior to implantation into the CNS. Such studies are in progress [55].

Acknowledgements

We wish to thank Natalie Symons, Margaret Scoones, Yi-Chi Chen, Dawn Thompson-Wallis and Laszlo Bubrik for technical assistance. Supported by grants from NHMRC, and the Arnold Yeldham and Mary Raine Medical Research Foundation (UWA).

References

[1] Aguayo, A., David, S., Richardson, P. and Bray, G., Axonal elongation in peripheral and central nervous system transplants, Adv. Cell Neurobiol., 3 (1982) 215-234.

[2] Ard, M.D., Bunge, R.P. and Bunge, M.B., Comparison of the Schwann cell surface and Schwann cell extracellular matrix as promoters of neurite growth, J. Neurocytol., 16 (1987) 539-555.

[3] Assouline, J.G., Bosch, P., Lim, R., Kim, I.S., Jensen, R. and Pantazis, N.J., Rat astrocytes and Schwann cells in culture synthesise nerve growth factor-like promoting factors, Dev. Brain Res., 31 (1987) 103-118.

[4] Baehr, M. and Bunge, R.P., Functional status influences the ability of Schwann cells to support adult rat retinal ganglion cell survival and axonal growth, Exp. Neurol., 106 (1989) 27-40.

[5] Baron-Van Evercooren, A., Clerin-Duhamel, E., Lapie, P., Gansmiiller, A., Lachapelle, F. and Gumpel, M., The fate of Schwann cells transplanted in the brain during development, Dev. Neurosci., 14 (1992) 73-84.

[6] Barvic, M., Kliment, K. and Zavadil, M., Biologic properties and possible uses of polymer-like sponges, J. Biomed. Mater. Res., 1 (1967) 313-323.

[7] Bernstein, J.J., Getz, R., Jefferson, M. and Keleman, M., Astro- cytes secrete basal lamina after hemisection of rat spinal cord, Brain Res., 327 (1985) 135-141.

[8] Berry, M., Hall, S., Follows, R., Rees, L., Gregson, N. and Sievers, J., Response of axons and glia at the site of anastomosis between the optic nerve and cellular or acellular nerve grafts, J. Neurocytol., 17 (1988) 727-744.

[9] Berry, M., Rees, L., Hall, S., Yiu, P. and Sievers, J., Optic axons regenerate into sciatic nerve isografts only in the presence of Schwann cells, Brain Res. Bull., 20 (1988) 223-231.

[10] Bjorklund, A. and Stenevi, U., Regeneration of monomaminer- gic and cholinergic neurons in the mammalian central nervous system, Physiol. Rev., 59 (1979) 62-100.

[11] Bray, G.M. and Aguayo, A.J., Exploring the capacity of CNS neurons to survive injury, regrow axons and form new synapses in adult mammals. In F.J. Seil (Ed.), Neural Regeneration and Transplantation, Frontiers of Clinical Neuroscience, Vol. 61, Liss, New York, 1989. pp. 67-78.

[12] Brook, G.A., Lawrence, J.M. and Raisman, G., Morphology and migration of cultured Schwann cells transplanted into the fimbria and hippocampus in adult rats, Glia, 9 (1993) 292-304.

[13] Brockes, J.P., Fields, K.L. and Raft, M.C., Studies on cultured rat Schwann cells. I. Establishment of purified populations from cultures of peripheral nerves, Brain Res., 165 (1979) 105-118.

[14] Calnan, J.S., Assessment of biological properties of implants before their clinical use, Proc. Roy. Soc. Med., 63 (1970) 1115- 1118.

[15] Calnan, J.S., Pflug, J.J., Chhabra, A.S. and Raghupati, N., Clinical and experimental studies of polyhydroxyethyl- methacrylate gel ('Hydron') for reconstructive surgery, Br. J. Plast. Surg., 24 (1971) 113-124.

[16] Carbonetto, S.T., Gruver, M.M. and Turner, D.C., Nerve fiber growth on defined hydrogel substrates, Science, 216 (1982) 897- 899.

[17] Chen, M., Harvey, A.R. and Dyson, S.E., Regrowth of lesioned retinal axons associated with the transplantation of Schwann cells to the brachial region of the rat optic tract, Restor. Neurol. Neurosci., 2 (1991) 233-248.

[18] Chen, Y.C., Chirila, T.V. and Russo, A.V., Hydrophilic sponges based on 2-hydroxyethyl methacrylate. II. Effect of monomer mixture composition on the equilibrium water content and swelling behaviour, Mater. Forum., 17 (1993) 57-65.

[19] Chirila, T.V., Constable, I.J., Crawford, G.J., Vijayasekaran, S., Thompson, D.E., Chen, Y.C., Fletcher, W.A. and Griffin, B.J., Poly(2-hydroxyethyl methacrylate) sponges as implant materials: in vivo and in vitro evaluation of cellular invasion, Biomaterials, 14 (1993) 26-38.

[20] Chirila, T.V., Chen, Y.C., Griffen, B.J. and Constable, l.J., Hydrophilic sponges based on 2-hydroxyethyl methacrylate. I. Effect of monomer mixture composition on the pore size, Polym. International, 32 (1993) 221-232.

[21] Civerchia-Perez, L., Faris, B., La Pointe, G., Beldras, J., Lei- bowitz, H. and Franzblau, C., Use of collagen-methacrylate hydrogels for cell growth, Proc. Natl. Acad. Sci., 77 (1980) 2064-2068.

[22] Cotman, C.W. and Nadler, J.R., Reactive synaptogenesis in the hippocampus. In C.W. Cotman (Ed.), Neuronal plasticity., Raven Press, New York. 1978, pp. 227-272.

[23] Crawford, G.J., Constable, I.J., Chirila, T.V., Vijayasekaran, S. and Thompson, D.E., Tissue interaction with hydrogel sponges implanted in the rabbit cornea, Cornea, 12 (1993) 348-357.

[24] David, S. and Aguayo, A., Axonal elongation into peripheral nervous system 'bridges' after central nervous system injury in adult rats, Science, 214 (1981) 931-933.

[25] Davis, J.B. and Stroobant, P., Platelet-derived growth factors and fibroblast growth factors are mitogens for rat Schwann cells, J. Cell Biol., 110 (1990) 1353-1360.

[26] Dijkstra, C.D., Dopp, E.A., Joling, P. and Kraal, G., The het- erogeneity of mononuclear phagocytes in lymphoid organs: dis- tinct macrophage subpopulations recognized by monoclonal antibodies ED1, ED2 and ED3, Immunology, 54 (1985) 589-599.

[27] Doster, S.K., Lozano, A.M., Aguayo, A.J. and WiUard, M.B., Expression of the growth associated protein GAP-43 in adult rat retinal ganglion cells following axonal injury, Neuron, 6 (1991) 635-647.

[28] Dubois-Dalq, M., Rentier, B., Baron Van Evercooren, A. and Burge, B.W., Structure and behaviour of rat primary and secondary Schwann cells in vitro, Exp. Cell Res., 131 (1981) 283-297.

[29] Eccleston, P.A., Mirsky, R. and Jessen, K.R., Typed collagen preparations inhibit DNA synthesis in glial cells of the peripheral nervous system, Exp. Cell Res., 182 (1989) 173-185.

[30] Feringa, E.R., Vahlsing, H.L. and Smith, B.E., Retrograde transport in cortico-spinal neurons after spinal cord transection, Neurology, 33 (1983) 478-482.

[31] Ferrari, G., Fabris, M., Polato, P., Skaper, S.D., Fiori, M.G. and

G.W. Plant et al. / Brain Research 671 (1995) 119-130 129

Yan, Q., Rat NGF receptor is recognized by tumor-associated antigen monoclonal antibody 217c, Exp. Neurol., 112 (1991) 183-194.

[32] Gage, F.H., Wictorin, K., Fischer, W., Williams, L.R., Varon, S. and Bjorklund, A., Retrograde cell changes in medial septum and diagonal band following fimbria-fornix transection: quantitative temporal analysis, Neuroscience, 19 (1986) 241-255.

[33] Gelderd, J.B., Evaluation of blood vessel and neurite growth into a collagen matrix placed within a surgically created gap in rat spinal cord, Brain Res., 511 (1990) 80-92.

[34] Guenard, V., Aebischer, P. and Bunge, R.P., The astrocyte inhibition of peripheral nerve regeneration is reversed by Schwann cells, Exp. Ne~trol., 126 (1994) 44-60.

[35] Harkes, G., Dankent, J. and Feijen, J., Adhesion of Escherichia Coli onto a series of 9oly(methacrylates) differing in hydro- phobicity and charge. In Heimke, G., Saltesz, U. and Lee, A.J.C., (Eds.), Clinical Implant Materials. Advances in Biomate- rial, 9 (1990) 111-116.

[36] Harvey, A.R., Chen, M. and Dyson, S.E., Glial cells transplanted to the rat optic tract, Ann. NYAcad. Sci., 633 (1991) 573-576.

[37] Harvey, A.R., Chen, M., Plant, G.W. and Dyson, S.E., Re- growth of axons within $chwann cell filled polycarbonate tubes implanted into the damaged optic tract and cerebral cortex of rats, Restor. Neurol. Ne~trosci., 6 (1994) 221-237.

[38] Harvey, A.R., Fan, Y., Connor, t A.M., Grounds, M.D. and Beilharz, M.W., The migration and intermixing of donor and host glia on nitrocellulose polymers implanted into cortical lesion cavities in adult mice and rats, Int. J. Dev. Neurosci., 11 (1993) 569-581.

[39] Harvey, A.R., Gan, S.K. and Pauken, J.M., Fetal tectal or cortical tissue transplanted into brachial lesion cavities in rats: influence on the regrowth of host retinal axons, J. Comp. Neurol., 263 (1987) 126-136.

[40] Harvey, A.R., Plant, G.W. and Kent, A.P., The distribution of astrocytes, oligodendroglia and myelin in normal and transplanted rat superior colliculus: an immunohistochemical study, J. Neural Transplant Plast., 4 (1993) 1-14.

[41] Hefti, F., Nerve growth factor promotes survival of septal cholinergic neurons after fimbrial transections, J. Neurosci., 6 (1986) 2155-2162.

[42] Heumann, R., Lindholm, D., Bandtlow, C., Meyer, M., Radecker M.J., Misko, T.P., Shooter, E. and Thoenen, H., Differential regulation of mRNA encoding nerve growth factor receptor in rat sciatic nerve during development, degeneration, and regeneration: role of macrophages, Proc. Natl. Acad. Sci., 84 (1987) 8735-8739.

[43] Hopkins, J.M. and Bunge, R.M., Regeneration of axons from rat retinal ganglion cells on cultured Schwann cells is not dependent on basal lamina, Glia, 4 (1991) 46-55.

[44] Jessen, K.R. and Mirs~r, R., Schwann cell precursors and their development, Glia, 4 (1991) 185-194.

[45] Kapfhammer, J., Schwab, M.E. and Schneider, G.E., Antibody neutralization of neurite growth inhibitors from oligodendrocytes results in expanded pattern of postnataUy spouting retinocollicular axons, J. Neurosci., 12 (1992) 2112-2119.

[46] Kromer, L.F, Cornbrooks and C.J., Transplants of Schwann cell cultures promote axonal regeneration in the adult mammalian brain, Proc. Natl. Acad. Sci., 82 (1985) 6330-6334.

[47] Lemke, G, and Chao, M., Axons regulate Schwann cell expression of the major myelin and NGF receptor genes, Develop- ment, 102 (1988) 499-5(14.

[48] Lydon, M.J., Minett, T.W. and Tighe, B.J., Cellular interactions with synthetic polymer surfaces in culture, Biomaterials, 6 (1985) 396-402.

[49] Manthorpe, M., Rudge, J. and Varon, S., Astroglial cell contri- butions to neuronal su~rival and neuritic growth. In S. Fedoroff

and A. Vernadakis (Eds.), Astrocytes, Vol. 2, Academic Press, New York, 1986, pp. 315-376.

[50] Marchand, R. and Woerly, S., Transected spinal cords grafted with in situ self assembled collagen matrices, Neuroscience, 36 (1990) 45-60.

[51] Marchand, R., Woerly, S., Bertrand, L. and Valdes, N., Evalua- tion of two cross linked collagen gels implanted in the transected spinal cord, Brain Res. Bull., 30 (1993) 415-422.

[52] Meyer, M., Matsuoko, I., Wetmore, C., Olson, L. and Thoenen, H., Enhanced synthesis of brain-derived neurotrophic factor in the lesioned peripheral nerve: different mechanisms are respon- sible for the regulation of BDNF and NGF mRNA, J. Cell. Biol., 119 (1992) 45-54.

[53] Montero-Menei, C.N., Pouplard-Barthelaix, A., Gumpel, M. and Baron-Van Evercooren, A., Pure Schwann cell suspension grafts promote regeneration of the lesioned septo-hippocampal cholinergic pathway, Brain Res., 570 (1992) 198-208.

[54] Montgomery, C.T. and Robson, J.A., Implants of cultured Schwann cells support axonal growth in the central nervous system of adult rats, Exp. Neurol., 122 (1993) 107-124.

[55] Plant, G.W., Harvey, A.R. and Woerly, S.W. The entrapment of glial cells into Poly[N-(2-hydroxypropyl) methacrylamide] polymer matrices: a potential tool for transplantation, Proc. Aust. Neurosci. Soc., 5 (1994) 260.

[56] Pokharna, H., Zhong, Y. and Smith, D.J., Copolymers of hydroxyethyl methacrylate with quadrol methacrylate and with various aminoalkyl methacrylamides as fibroblast substrata, J. Bioact. Comp. Polymers, 5 (1990) 42-52.

[57] Reier, P.J., Stensaas, L.J. and Guth, L., The astrocytic scar as an impediment to regeneration in the central nervous system. In C.C. Kao, R.P. Bunge and P.J. Reier (Eds.), Spinal Cord Recon- struction, Raven Press, New York, 1983, pp 163-195.

[58] Richardson, P.M., Issa, V.M. and Shemie, S., Regeneration and retrograde degeneration of axons in the rat optic nerve, J. Neurocytol., 11 (1982) 949-966.

[59] Richardson, P.M., McGuiness, U.M. and Aguayo, A.J., Axons from CNS neurons regenerate into PNS grafts, Nature, 284 (1980) 264-265.

[60] Rubin, R.M. and Marshall, J.L., Porous hydrophilic polymers: good and bad news in the orthopedic application of cruciate ligament substitution, J. Biomed. Mater. Res., 9 (1975) 375-380.

[61] Rudge, J.S. and Silver, J., Inhibition of neurite outgrowth on astroglial scars in vitro, J. Neurosci., 10 (1990) 3594-3603.

[62] Schnell, L. and Schwab, M.E., Sprouting and regeneration of lesioned corticospinal tract fibres in the adult rat spinal cord, Eur. J. Neurosci., 5 (1993) 1156-1171.

[63] Schreyer, D.J. and Jones, E.J., Growth of corticospinal axons on prosthetic substrates introduced into the spinal cord of neonatal rats, Dev. Brain Res., 35 (1987) 291-299.

[64] Schwab, M.E. and Caroni, P., Oligodendrocytes and CNS myelin are nonpermissive substrates for neurite growth and fibroblast spreading in vitro, J. Neurosci., 8 (1988) 2381-2393.

[65] Shimizu, Y. Myamoto, Y. and Teramastu, T., Studies on com- posites of collagen and a synthetic polymer, Biota. Med. Devices. Artif. Organs, 6 (1978) 375-391.

[66] Smahel, J., Moserova, J. and Behounkova, E., Tissue reaction after subcutaneous implantation of Hydron sponge, Acta Chir. Plast., 13 (1971) 193-202.

[67] Smetana, K., Vacik, J., Sonckova, Z. and Sulc, J., The influence of hydrogel functional groups on cell behaviour, J. Biomed. Mat. Res., 24 (1990) 463-470.

[68] 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.

[69] Stevenson, J., Growth of optic tract axons in nerve grafts in hamsters, Exp. Neurol., 87 (1985) 446-457.

[70] Stockli, K.A., Lottspeich, F., Sendtner, M., Maisiakowski, P.,

130 G.W. Plant et aL /Brain Research 671 (1995) 119-130

Carroll, P., Gotz, R., Lindholm, D. and Thoenen, H., Molecular cloning expression and regional distribution of rat ciliary neurotrophic factor, Nature, 342 (1989) 920-923.

[71] Stol, M., Tolar, M. and Adam, M., Poly(2-hydroxyethyl methacrylate)-collagen components which promote muscle differentiation in vitro, Biomaterials, 6 (1985) 193-197.

[72] Taniuchi, M., Clark, B.H., Schweitzer, J.B. and Johnson, E.M., Expression of nerve growth factor receptors by Schwann cells of axotomized peripheral nerves: ultrastructural location, suppres- sion by axonal contact and binding properties, J. Neurosci., 8 (1988) 664-681.

[73] 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.

[74] Woerly, S., Marchand, R. and Lavallee, C., Interactions of co-polymeric poly(glyceryl-methacrylate)-collagen hydrogels with neural tissue: effects of structure and polar groups, Biomateri- als, 12 (1991) 197-203.

[75] Woerly, S. and Marchand, R., Collagen-chondroitin-6-sulfate hydrogel implants in CNS lesion cavities favor glial repair, the differentiation of co-implanted neurons and the growth of axons, Restor. Neurol. Neurosci., 3 (1991) 95-99.

[76] Woerly, S., Lavallee, C. and Marchand, R., Intracerebral implantation of ionic synthetic hydrogels: effect of polar substrata on astrocytosis and axons, J. Neur. Transplant Plast., 3 (1992) 21-34.

[77] Yoshii, F. and Kaetsu, I., Cell culture on polymers prepared by radiation induced polymerization of various glass forming monomers, Appl. Biochem. Biotechnol., 8 (1983) 115-126.

Axonal growth within poly (2-hydroxyethyl methacrylate) sponges infiltrated with Schwann cells and...

Documents

Transcript of Axonal growth within poly (2-hydroxyethyl methacrylate) sponges infiltrated with Schwann cells and...