Field of the Invention

The present invention relates to a medical device useful as a graft in repairing injured nerve tissues. It is a graft device composed of a collagen filter coated with neurite outgrowth promoting substances. The invention also comp- rises the preparation of the said device, and its use as a graft for repairing injured nerve tissues.

Description of the Prior Art

At present the conventional suturing of the nerves, known as neuroraphy (Sunderland 1978) is the method of choise used in repairing injured nerves. As suturing of the se¬ vered nerves is a technically demanding approach that may yield to poor results due to a variety of factors, alter¬ native methods are being searched for. It is known that adult nerve tissue reacts to trauma by fibrosis. Therefore, formation of a connective tissue scar in a resutured nerve may be a problem. Furthermore, as resuturing also causes severe autotomy, this method of operation may both increase pain and obstruct the growth of regenerating nerve fibre bundles.

Alternatives to substitute the conventional suturing method have been suggested by various investigators. For example, Narakas (1988) used glueing of the nerve stumps together with a fibrin glue.

Various tubulization techniques have also been disclosed. These methods apply tubular grafts or implants fabricated from various biological materials. These grafts bring the stumps of the severed nerve close to each other as well as

provide a suitable environment for nerve regeneration. The¬ se grafts may also contain additives, such as growth fac¬ tors.

The PCT application WO 84/03035 discloses a longitudinally openable, porous, rough-surfaced tube of a natural or synthetic absorbable polymer. This polymer is said to support nerve regeneration as a tube but not as a membrane wrapped around the nerve. The tube is compared with resutu- ring, which is, however, done without sufficient numbers of animals to allow a reliable interpretation of the data. It is described that the nerve becomes bonded to the porous inner surface of the tube. It remains, however, unclear if the graft remains in its place merely due to this effect.

European patent application No. 277 678 discloses a graft for use in reconstructive surgery to repair e.g. damaged nerve tissue. The graft comprises porous matrix of an organic polymeric material incorporating optionally addi- tives, such as growth factors. The application does not, however, contain any information about the use of the said graft.

PCT-application WO 89/10725 discloses a method for produ- cing a biodegradable polymer having a preferentially oriented pore structure, and a method for using this poly¬ mer to regenerate damaged nerve tissue.

In European patent application No. 286 284 medical devices for use in regenerating nerves are disclosed. The said devices are in form of tubular semipermeable conduits which have smooth inner skins and which can be formed from va¬ rious polymeric materials. However, the tests described in the application include merely procedures wherein the conduits are implanted by suturing them in their place. It is also indicated that a device could be used in a sheet form. No evidence is, however, given that such a sheet

would stay in its place merely by friction. Two further applications of the same inventors, WO 90/05552 and WO 90/05490, disclose nerve guidance channels containing diffusible nerve growth-inducing active factors, and electrically-charged nerve guidance channels, respectively.

Navarro et al . (1991) disclose a comparative study of nerve regeneration using collagen conduits optionally coated with laminin. The conduits must, however, be sutured in their place. Furthermore, the results obtained showed that lami¬ nin did not exert any positive effect in the reinnervation.

Most of the above mentioned prior art tubes or conduits must be secured in their position by suturing. Thus the advantages which may be achieved using these tubular grafts are reduced due to the fact that the suturation is necessa¬ ry. Suturation tends to cause trauma which is known to stimulate the growth of fibroblasts and other connective tissue components, and increase the painful dysesthesias due to nerve manipulation.

Thus, the prior art does not provide a method that would be an improvement as compared to the methods in use at pre¬ sent. Furthermore, the degree of the regeneration is un- clear in most cases, as only few investigators have inclu¬ ded an adequate control, e.g. resuturing, to compare the efficiencies of their resuturing applied in restorative surgery.

The PCT application WO 84/03035 is the only one in which resuturing is replaced by a graft. However, this method does not significantly reduce autotomy as compared to the resuturing. Furthermore, regeneration was only assessed by histological examination, without testing of a functional recovery.

Many factors that promote nerve growth are known in the art. One of the factors, known to enhance outgrowth of axons of cultured neurons is laminin. Laminin is a 1000 kDa extracellular matrix protein that promotes neurite out- growth in vitro (Baron van Evercooren et al. 1982), and correlates with regenerative CNS systems in vivo (Liesi, 1985). As this protein is produced by the Schwann cells, it has been applied in some recent descriptive studies to promote peripheral nerve regeneration. Laminin has been used in the form of a gel (Madison et al. 1985), a soluble agent (Muller et al. 1987, Muller 1988) or a molecule attached to silicone fibres (Yoshii et al. 1985). In recent studies a 12 amino acid peptide from the C-terminal region of the B2-chain of mouse laminin has been shown to promote neurite outgrowth of cultured neurons in both soluble and substrate bound forms (Liesi et al. 1989).

Description of the Invention

It has now been discovered that functional in vivo re¬ generation of nerve tissue can be achieved by using a graft in which pure laminin is attached to a collagen filter. It is also shown that a 10 amino acid peptide from the neurite outgrowth promoting domain of mouse laminin, its human analogue, or even an antibody raised against the said domain have the same effect.

The present invention thus provides a new kind of device for use as a graft in nerve repair, the device comprising a collagen filter coated with at least one of the neurite outgrowth promoting substances selected from the group consisting of laminin, a peptide of a neurite outgrowth promoting domain of laminin, and an antibody raised against said domain of laminin.

A preferable peptide for the device of this invention is the 10 amino acid peptide of a neurite outgrowth promoting domain of the B2 chain of mouse laminin, or its human analogue.

The graft of the invention, comprising pure protein or peptide, is sticky when wetted and remains in its place like a "fly-paper". Thus, use of this graft is rapid and minimizes manipulation of the nerve as no suturing is required.

Collagen used in this invention may be any type I collagen, but preferably type I collagen suitable for medical use. In the present experiments, clinically acceptable bovine tendon type I collagen was used as a back-bone for the filters.

The laminin used in the grafts of the experiments is pu¬ rified from the mouse EHS tumor using published methods (Timpl et al . , 1979). Human laminin or recombinant mamma¬ lian laminin may also be used.

The 12 amino acid neurite outgrowth promoting peptide from the C-terminal region of the B2-chain of mouse laminin was identified by Liesi et al. 1989. The published peptide has an amino acid sequence Arg-Asn-Ile-Ala-Glu-Ile-Ile-Lys-Asp- Ile-Glv-Cvs. For this invention the 10 amino acid form, excluding the two last amino acids, is preferable. For the purposes of this invention the 10 amino acid peptide is called p20. Thus, this peptide can be prepared according to Liesi et al . 1989, except that Gly-Cys present in the published form can be omitted.

The corresponding human analogue of the said mouse neurite outgrowth promoting peptide has an amino acid sequence Arg- Asp-Ile-Glu-Glu-Ile-Met-Lys-Asp-Ile. This peptide can be produced as the mouse peptide and it is called p20-HU.

In a preferred embodiment of the present invention the peptides are prepared in the C-terminal acid form.

The antibodies against the neurite outgrowth promoting pep- tide of the mouse B2 chain is produced in rabbits by met¬ hods known as such, i.e. by immunizing the animals with sequential injections of the peptide antigen.

The method of fabricating the device of the invention is also described. In short, soluble collagen is spread onto an aluminium block having suitable dimensions, another block is located onto the said first block, and even pressure is subjected to the material in the apparatus. The apparatus is quick frozen on dry ice whereby the collagen layer in it also freezes. The apparatus is then located in a lyofilizer whereby the material forms a filter. A layer of laminin and/or desired peptides or antibodies is conse¬ cutively pressed on top of the collagen filter.

Alternatively, grafts of the invention can be fabricated by coating commercially available collagen filters (Helisert V301 or V166, Helitrex, Princeton, NJ, USA) with laminin and/or the desired peptides or antibodies using the above described apparatus.

A process for the repair of an injured nerve is also described, wherein the graft defined above is provided, the said graft is wrapped around the stumps of the injured nerve, and the nerve is allowed to regenerate.

Brief description of the drawings

Fig. 1 Recovery of muscle strength 4 months after surgery. The laminin graft of the invention (gr) and resuturation

(su) are compared with each other. The mean tension (+/-

S.E.M. ) of a single twitch (A and B), the tetanic tension

(C), and weight of the gastrocnemius-soleus muscles (D) in the laminin grafted (gr), and neuroraphy treated (su) rats four months after surgery. In both groups the mean weight is referred to the mean value of the unoperated side (100 %).

Fig. 2 The degree of autotomy using the laminin graft vs. resuturation. The mean (+/- S.E.M. ) autotomy score of the laminin grafted (gr), and neuroraphy treated (su) rats during the follow-up time of 16 weeks. A significant diffe¬ rence between the groups is indicated by an asterisk (p <0.05; Mann-Whitney U-test).

Fig. 3 Double immunocytochemistry for the NGF receptor (A,C,E,G) and neurofilaments (B,D,F,H) in the proximal stumps of the resutured (A-B), or laminin (C-D), its neu¬ rite outgrowth domain (E-F), and collagen (G-H) grafted nerves 20 days after operation.

Fig. 4 Double immunocytochemistry for the NGF receptor (A,C,E) and neurofilaments (B,D,F) in the proximal areas of the resutured (A-B), and laminin (C-D) or its neurite outgrowth promoting peptide (E-F) grafted nerves 20 days after operation.

Fig. 5 NGF receptor immunoreactivity in the distal stump areas (A-B) of the resutured (A) or laminin grafted (B) animals, and double immunostaining for NGF receptor (C,E), and neurofilaments (D,F) in the distal stumps of the neuri- te outgrowth promoting peptide grafted animals 20 days after operation.

Fig. 6 Immunocytochemical demonstration of the neurite outgrowth promoting domain of the B2 chain of laminin (A,C,D), and its free peptide antigen (B) in the peptide grafted animals in the graft area (A,B,D), and in the proximal stump (C). In A-B, the same photographic field is

immunostained with an antibody that recognizes both the free decapeptide as well as this domain in the B2 chain of laminin.

Fig. 7 The ability of the p20-HU to promote neurite out¬ growth in vitro . A comparison between the percentages of neurons with long neurites on a laminin substrate with (left bar) or without (right bar) the peptide p20-HU in¬ cluded into the culture medium.

Fig. 8 The ability of the antibody against the p20 to promote neurite outgrowth in vivo is verified. In A, the preimmune serum of the rabbit immunized with p20 is used. In B, the antibody to p20 is applied onto the filter. In C, the resuturing of the nerve.

Description of the preferred embodiments

The ability of the materials described in this invention to support peripheral nerve regeneration was evaluated by electrophysiological, behavioral and immunocytochemical methods. The methods were applied as described in the experimental section.

Electrophysiological and behavioral tests

Transection of the sciatic nerve abolishes the responses of the gastrocnemius-soleus muscles to electrical stimulation of the sciatic nerve in control experiments. Four months after surgery, there was no statistical difference in the twitch tension, tetanic tension, and the muscle mass bet¬ ween groups treated with laminin grafting or resuturing (Fig. 1, Table 1). The 10 amino acid neurite outgrowth promoting peptide of the mouse laminin had a similar effect ( able 2) . The experiment for Table 2 was as in Table 1. However, the number of animals (n=3) are not sufficient to

perform statistical analysis. The growth promoting property of p20 is clearly demonstrated. In the laminin grafted rats, the autotomy scoring was significantly lower than in the rats with resutured nerves (Fig. 2).

Table 1. The mean (+/- S.D. ) twitch tension, tetanic ten¬ sion, and weights of the gastrocnemius-soleus muscles in the graft-treated and neuroraphy-treated rats four months after surgery. The statistical significance was calculated between the operated and unoperated sides using Student's t-test.

Twitch tension regenerated (stim. 0.01 ras) intact

Twitch tension regenerated (stim. 0.10 ras) intact

Tetanic tension regenerated (stim. 50 Hz, intact 0.1 ms)

Muscle weight regenerated intact

Table 2. The mean (+/- S.D. ) tetanic tension, and weights of the gastrocnemius-soleus muscles in the p20-graft-trea- ted and laminin graft-treated rats four months after sur¬ gery.

Immunocytochemistry of the transected nerves

In short term experiments, a graft of laminin, collagen alone, or a graft comprising the mouse neurite outgrowth promoting peptide p20, its human analogue or antibody against the mouse peptide were inserted to join a 10 mm long gap between the stumps of a transected sciatic nerve, immunohistochemistry was applied to verify nerve regenera¬ tion, and resuturing of the cutted piece of nerve served as a control. The grafts supported a short term regeneration of the sciatic nerve across the gap as effectively as resuturing. This was apparent by the demonstration of outgrowth of neurofilament positive nerve fibres from the proximal stump into the graft, and to the distal stupm area (Fig. 3-5). This nerve fibre outgrowth occured within 20 days after the operation, whereas in the collagen grafted animals the neurites did not grow into the graft areas.

As the low affinity NGF receptor has been linked to pe- ripheral nerve regeneration (Taniuchi et al . , 1986; Heumann et al., 1987), we used antibodies directed against this receptor to monitor regeneration in grafted or resutured sciatic nerves.

In the resutured nerves, the NGF receptor was weakly ex¬ pressed by groups of the presumptive Schwann cells in the proximal nerve stump (Fig. 3). In 3A, the NGF receptor is expressed by scattered Schwann cells in the nerve, and neurofilament immunoreactive nerve fibre bundles grow in between the NGF receptor positive areas (3B). In 3C-D, a similar distribution of the NGF receptor is apparent in the laminin grafted animal. In 3E-F, the proximal stump of a neurite outgrowth promoting domain grafted animal show scattered NGF receptor immunoreactivity (E), and neurofila- ment positive nerve fibre bundles (F). In 3G, the NGF receptor immunoreactivity is markedly increased in the collagen grafted animal. In 3H, double immunostaining shows

neurofilament positive nerve fibre bundles slightly swollen and disorganized.

No NGF receptor expression was seen inside the resutured piece of the nerve (Fig. 4). In 4A-B, the resutured area of the cut piece of the nerve shows little NGF receptor acti¬ vity (4A). The neurofilament staining reveals slightly disorganized, but massive outgrowth of nerve fibres. In 4C- D, a similar distribution is observed in a laminin grafted animal. Note that the nerve fibres seem to be thinner and more organized than in 4B. In 4E-F, the graft with the neurite outgrowth promoting peptide expresses low levels of the NGF receptor (4E), and shows neurofilament positive nerve fibre bundles (4F) that resemble those in 4B.

In the distal stump area, the NGF receptor expressed at high levels by all Schwann cells (Fig. 5). In 5A, intense immunoreactivity for the NGF receptors is shown in all the Schwann cells. In 5B, a similar distribution of the NGF immunoreactivity is demonstrated in the laminin grafted nerve. In 5C, the NGF receptor expression in the distal stump is elevated also in the peptide grafted nerve. In 5D, double immunocytochemical demonstration of neurofilaments reveals massive regeneration. In 5E, the Schwann cells are immunoreactive for the NGF receptor (5E), and double im¬ munocytochemistry for neurofilaments (5F) shows that the Schwann cells are in close contact with regenerating nerve fibres that have varicosities (5F).

A pattern of the low affinity NGF receptor expression was thus detected in the nerves restored using grafts of la¬ minin or the neurite outgrowth domain of the B2 chain of laminin. The NGF receptor activity was highest in the distal stump area, low in the graft and scattered in the proximal stump area. In contrast, the collagen graft that did not support regeneration, induced Schwann cells to

express high levels of the NGF receptor in the proximal stump.

As the neuronal regeneration with a graft of the neurite outgrowth promoting domain of the B2 chain of laminin equals to that with resuturing, we wanted to investigate the distribution of this peptide antigen, and the B2 chain of laminin after the nerve reconstruction by the peptide grafting. These studies were carried out using either antibodies that recognize both the B2 chain of laminin and the free decapeptide itself (MurtomSki et al . , 1992), or with antibodies that are only directed against this domain in the B2 chain of laminin. We found that antibodies that recognize also the free peptide, could demonstrate punctate deposits of the peptide antigen on the Schwann cell surfa¬ ces as well as their intracytoplasmic immunoreactivity for the B2 chain of laminin (Fig.6 A-B). In A, the Schwann cells are shown to express increased levels of the B2 chain. The punctate deposits of the neurite outgrowth promoting peptide antigen of laminin were present in the proximal stump, and in the grafted areas (6B), but were absent from the distal nerve (not shown). In contrast, those antibodies that only recognized this domain in the native laminin, showed the B2 chain immunoreactivity of the Schwann cells, but no extracellular punctate deposits (6 C- D). In the proximal stumps of the peptide grafted nerves (6C), the B2 chain immunoreactivity was present in a few scattered Schwann cells, whereas this immunoreactivity was markedly increased in the proximal graft areas (6D), and in the distal stump of the peptide grafted nerves (not shown). In the laminin grafted and resutured nerves the B2 chain immunoreactivity was identical to that of the peptide grafted nerves (Fig.6), whereas the collagen grafted nerves showed low levels of the B2 chain immunoreactivity in the proximal graft areas (not shown).

The ability of an antibody against the p20 to promote neurite outgrowth in vivo was also verified. A 1 cm gap was reconnected with a graft containing 200 μl of 10 μg/ml concentration of the preimmune serum (Fig. 8 A), or the antibody against p20 (8B) . Resuturing of the cut 1 cm piece of the nerve (8C) was used as a control. The 200 kDa neuro¬ filament protein is demonstrated in the distal part of the nerve using monoclonal antibodies (Boehringer, Germany, RT 97). Note that in 8A no nerve fibres are growing. The only neurofilament positive structures are accumulations of depri, typical in Wallerian degeneration.

Promoting the neurite outgrowth in vitro

The ability of the p20-HU to promote neurite outgrowth was investigated as described in the experimental section. It was shown that addition of 100 μg/ml of the peptide p20-HU into the culture medium of cerebellar neurons grown on a laminin substrate support significantly the neurite out- growth of cerebellar neurons p<0.005 (Fig. 7).

These results show, for the first time, that laminin sup¬ ports functional recovery after a peripheral nerve injury. As the short term recovery with the mouse neurite outgrowth promoting peptide, its human analogue or antibody equal that induced by laminin, this domain may be a regeneration promoting domain of laminin in vivo.

Lastly, the significant reduction of autotomy in the la- minin grafts of operated animals indicates that this method of nerve reconstruction may be widely applicable in nerve surgery in humans.

Our results provide the first direct evidence for the functional role of laminin in nerve regeneration in vivo .

Laminin and its neurite outgrowth promoting domain may also function as both soluble and substrate bound trophic fac-

tors in a regenerating peripheral nerve. This idea is supported by the three lines of evidence:

1) decrease in NGFr accompanies rapid regeneration 2) the peptide antigen is expressed at increased levels on Schwann cell surfaces

3) the antibody against the neurite outgrowth promoting peptide seems to bind the peptide in its place and there¬ fore increase its local concentration.

We show here that nerve fascicles of a small diameter can be reconstructed fastly with this new bioresorbable mate¬ rial. The functional results of nerve regeneration are as good as with the resuturation. Furthermore, our technique is easier and faster to learn and apply than conventional neuroraphy or a repair with fibrin glue (Narakas, 1988). Manipulation of the injured nerve is also reduced to mi¬ nimum in our graft-model. At present, our method is es¬ pecially suitable for reconstruction of distal peripheral nerves (for example nerves of the fingers) or for pediatric nerve surgery.

The incidence of autotomy is generally high in the rats that undergo conventional nerve repair by resuturing (Wall et al. (1979), De Medinaceli et al. (1982) and Terzis and Smith (1986)). Thus the ability of the laminin graft to decrease autotomy is of great clinical importance. One additional advantage of our method is that unlike suturing the graft of the invention causes no interpositioning and scarring between the stumps. Furthermore, the graft is faster to apply and the amount of stimulation of the transected proximal stump may therefore be smaller than with the conventional methods. It has been previously suggested that availability of neurotrophic substances correlates negatively with autotomy (Blumenkopf and Lippman 1991). Therefore it is possible that laminin, or its pepti¬ des that may also act as neurotrophic substances (Liesi

1990), applied locally to the site of nerve trauma, de¬ crease autotomy.

This invention is further illustrated by the following examples and experiments.

Animals used in the experiments

Female Han-Wistar rats (age 3 months, weights 250-300 g) were used in all experiments. The rats were housed in groups of six, having food and water ad libitum. The animal house had lights on 6.00-18.00 h, and humidity in the room was 35-55percent. All the experimentation was performed under an approved animal licence from the Provincial Go- vernment of Uusimaa, Finland.

Example 1: Preparation of the graft device

a. A graft was produced by first laying down 1 mg/ml con- centration of bovine tendon collagen on an aluminium block having a space of about 0.3 mm thick and 1 cm wide. This was closed by another aluminium block having lines of about 0.3 mm thick and 0.1 mm wide crossing each other. These two aluminium blocks were placed towards one another and a constant pressure was created by pressing them firmly against each other. The apparatus containing the bovine tendon collagen was then quick frozen in powdered dry ice, and the material was lyophilized overnight in a standard lyophilizer.

The mouse EHS-tumor laminin was purified according to published methods (Timpl et al., 1979), and 400 μl of 100 μg/ml concentration of laminin sprayed onto the collagen filter placed in the same device on dry ice. The aluminium block was sealed again with its upper part after which the frozen materials were lyophilized.

The mouse peptide p20 or its human analogue were prepared in the C-terminal acid form as described (Liesi et al., 1989) or supplied by Multiple Peptide Systems, Inc. (San Diego, CA). These materials were applied at 100 μg/ml concentration. Each filter obtained 200 μl of such solu¬ tion.

Rabbit antibodies to the mouse peptide p20 were prepared as described (Murtomaki et al., 1992), i.e. by immunizing rabbits with 10 sequential injections of a 100 μg/injection of the 10 amino acid peptide antigen, and the IgG fraction purified by a protein A Sepharose column. The antibody obtained (anti-1543) was applied on top of the collagen filter at a final concentration of 10 μg/ml.

b. The other grafts used in the described experiments were prepared by lyophilizing the described concentrations of laminin, the mouse peptide p20 or its human analogue or antibodies to the mouse peptide on top of the collagen filters purchased by Helitrex.

Example 2: Nerve regeneration in vivo

Methods Experiment 1.: The rats were anesthetized with pentobar- bital (50 mg/kg; Orion, Finland), and their right sciatic nerves were exposed under aseptic conditions. The nerves were transected with a pair of eyesurgeon's scissors at the midtight. Then the nerves were resutured with 3-4 pe- rineural 10-0 monofilament sutures (n = 6, Dermalon, USA atraumatic needle) under a stereomicroscope (Wild, Switzer¬ land) or reconnected with a moistened laminin graft which was wrapped around the stumps of the transected nerve (n = 6, 4 mm margin over the both stumps). The anatomic orienta- tion was restored by careful observation of fascicular and vascular anatomy of the transected nerve before the re-

construction. After the reconstruction the wound was closed in layers with 3-0 silk sutures.

After the surgery the rats were observed for autotomy, a behavior which is related to painful dysesthesias (Wall et al . 1979), weekly during a period of four months. The scoring was performed as described earlier (Kauppila and Pertovaara 1991): every autotomized phalanx meant one score. The maximum score of a rat was nine if metatarsal bones were affected (the sciatic nerve innervates three lateral digits) . In order to avoid genetic factors to interfere with the results of the functional tests, the rats were chosen and housed so that there was a similar amount of rats of the same parents in both experimental and housing groups. Mann-Whitney U-test was applied for tes¬ ting.

Four months after surgery the rats were anesthetized with pentobarbital (60 mg/kg). The restoration of muscle rein- nervation was evaluated by comparing the amplitude of the maximal twitch and tetanic tensions of the operated ex¬ tensor muscles of the ankle with those of the contralateral unoperated control side, (Brunetti et al. 1987). In the end of the tests the rats were sacrified with an overdose of pentobarbital, the muscles were dissected free and weighed immediately (Sartorium, USA). The statistical analysis of the data was carried out with the Student's t-test.

Experiment 2. : The surgery was carried out as described in the first experiment except that a 10 mm gap was created by removing a 10 mm piece of the nerve. In the control group the removed stump was resutured with microsutures. In experimental groups a laminin graft or a collagen substra¬ tum alone were placed to bridge the gap. In further similar experiments a graft with peptide p20 and an antibody raised against p20 were used.

20 days after surgery the rats (n = 3 in each time and treatment group) were killed with an overdose of pento¬ barbital. The nerves were dissected free and cut in 10 μm cryostat sections. Immunocytochemistry was performed as described (Liesi and Silver, 1988). Briefly, the cryostat sections were fixed in 0.4% p-benzoquinone in PBS for 15 min, washed in PBS and dehydrated and rehydrated as desc¬ ribed (Liesi and Silver, 1988). Rabbit antibodies to neu¬ rofilaments (Dahl and Bignami, 1977), the neurite outgrowth promoting domain of laminin (Murtomaki et al . , 1992), or mouse monoclonal antibodies to NGF receptor (Yan and Johns- son, 1988) were applied at 1:1000, 10 μg/ml and 1:2000 dilutions, respectively. After an overnight incubation with the first antibodies, the sections were washed in PBS and exposed to second antibodies for 1 hour.

Example 3: Promoting the neurite outgrowth of cerebellar neurons in vitro

The ability of the p20-HU to promote neurite outgrowth was investigated in vitro by comparing the numbers of neurons with long neurites (>10 times the cell soma) on a laminin substrate with or without 100 μg/ml of the peptide p20-HU included in the culture medium. In short, cerebellar tissue from 5 day old rats was dissected aceptically, and the cells dissociated as described (Liesi et al., 1989). They were plated on LAMININ substrate at a density of 1 x 10 ⁵ and cultured overnight in a serum-free RPM medium. The p20- HU was added to the cells simultaneously with the plating.

REFERENCES

Baron van Evercooren, A., Kleinman, H.K., Ohno, S., Ma- rangos, P., Schwartz, J.P. and Dubois-Dalq, M.E., Nerve growth factor, laminin and fibronectin promote neurite growth in human fetal sensory ganglia cultures. J. Neu- rosci. Res. 8 (1981) 179-194

Blumenkopf, B. and Lippman, J.J., Studies in autotomy: its pathophysiology and usefulness as a model of chronic pain. Pain 45 (1991) 203-209

Brunetti, 0., Carobi, C. and Pazzaglia, U. , Influence of atrophy on the efficiency of muscle reinnervation. Exp. Neurol. 96 (1987) 248-252

Dahl, D, and Bignami, A. Preparation of antisera to neu¬ rofilament protein from chicken brain and human sciatic nerve. J. Comp. Neurol. 176 (1977) 645-658 de Medinaceli, L., Freed, W.J., and Wyatt, R.J., An index of the functional condition of rat sciatic nerve based on measurements made from walking traces. Exp. Neurol. 77 (1982) 634-643

Heumann, R. , Korsching, S., Bandtlow, C. and Thoenen, H. Changes of nerve growth factor synthesis in nonneuronal cells in response to sciatic nerve transection. J. Cell Biol. 104 (1987) 1623-1631

Kauppila, T. and Pertovaara, A., Effects of different sensory and behavioral manipulations on autotomy caused by a sciatic lesion in rats. Exp. Neurol. Ill (1991) 128-130 Liesi, P., Laminin immunoreactive glial cells distinguish regenerative CNS systems from nonregenerative ones. EMBO J. 4(1985) 2502-2511.

Liesi, P., Hermojen regeneraatio. Duodecim 106 (1990) 798-807

Liesi, P. and Silver, J. Is astrocyte laminin involved in axon guidance in the mammalian central nervous system? Dev. Biol. 130 (1988) 774-785

Liesi, P., Narvanen, A., Soos, J., Sariola, H., and Snou- nou, G., Identification of a neurite outgrowth promoting domain of laminin using synthetic peptides. FEBS Lett. 244 (1989):141-148.

Madison, R. , Da Silva, C.F., Dikkes, P., Sidman, R.L. and Chiu, T.-H., Increased rate of peripheral nerve regenera¬ tion using bioresorbable nerve guides and laminin con¬ taining gel. Exp. Neurol. 88 (1985) 767-772

Muller, H., Williams, L.R. , and Varon, S., Nerve re¬ generation chamber: Evaluation of exogenous agents applied by multiple injections. Brain Res. 413 (1987) 320-326 Muller, H. , Further evaluation of the effects of laminin, testosterone, ganglioside GMl, and catalase on early growth in rat nerve regeneration chambers. Exp. Neurol. 101 (1988) 228-233 MurtomSki, S., Risteli, J., Risteli, L. , Koivisto, U.-M., Johansson, S. and Liesi, P. Laminin and its neurite out¬ growth promoting domain in the Alzheimer's disease and Down's syndrome patients. J. Neurosci. Res. 32 (1992) 261- 273

Narakas, A., The use of fibrin glue in repair of peripheral nerves. Orthop. Clin. N. Am. 19 (1988) 187-199

Navarro, X., Kennedy, W.R., Stewart, N.J. and Furcht, L.T., Effects of laminin on functional reinnervation of target organs by regenerating axons. NeuroReport 2 (1991) 37-40

Sunderland, S. (ed.), Nerves and Nerve Injuries. 2nd Ed., Churchill-Livingstone, Edinburg, London, 1978

Taniuchi, M. , Clark, H.B. and Johnson, E.M., Jr. Induction of nerve growth factor receptor in Schwann cells after axotomy. Proc. Natl. Acad. Sci. USA 83 (1986) 4094-4098 Terzis, J. and Smith, K.J., Repair of several peripheral nerves: Comparison of "de medinaceli" and standard mic- rosurgical methods. Exp. Neurol. 96 (1987) 672-680

Timpl, R., Rohde, H., Robey, P.G., Rennard, S.I., Foidart, J.M., and Martin, G.R., Laminin - a glycoprotein from basement membranes. J. Biol. Chem. 254 (1979):9933-9937.

Wall, P.D., Devor, M. , Inbal, R. , Scadding, J.W., Sconfeld, D., Seltzer, Z. and Tomkiewicz, M.M. , Autotomy following peripheral nerve lesions: experimental anesthesia dolorosa. Pain 7 (1979) 285-298

Yan, Q. and Johnson, E. An immunohistochemical study of the nerve growth factor receptor in developing rats. J. Neuro- sci. 8 (1988) 3481-3498

Yoshii, S., Yamamuro, T., Ito, S., and Hayashi, M. , In vivo guidance of regenerating nerve by laminin-coated filaments. Exp. Neurol. 96 (1987) 469-473