Functional improvement following implantation of a microstructured, type-I collagen scaffold into experimental injuries of the adult rat spinal cord

brain research 1585 (2014) 37–50

Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

Research Report

Functional improvement following implantation of a microstructured, type-I collagen scaffold into experimental injuries of the adult rat spinal cord Haktan Altinovaa,b,h,n, Sven Mo¨llersa,c, Tobias Fu¨hrmanna,d, Ronald Deumense,h,i, Ahmet Bozkurtf,h,i, Ingo Heschelg, Leon H.H. Olde Daminkg, Frank Schu¨gnerg, Joachim Weish,i, Gary A. Brooka,h,i a

Department of Neurology, Uniklinik Aachen, Aachen, Germany Department of Neurosurgery, Evangelic Hospital Bethel, Bielefeld, Germany c RNL Europe GmbH, Kleinmachnow, Germany d Donnelly Centre for Cellular & Biomolecular Research, University of Toronto, ON, Canada e Institute of Neuroscience, Université Catholique de Louvain, Brussels, Belgium f Department of Plastic Surgery, Reconstructive and Hand Surgery, Burn Centre, Uniklinik Aachen, Aachen, Germany g Matricel GmbH, Herzogenrath, Germany h Institute for Neuropathology, Uniklinik Aachen, Aachen, Germany i Jülich-Aachen Research Alliance – Translational Brain Medicine (JARA Brain), Germany b

art i cle i nfo

ab st rac t

Article history:

The formation of cystic cavitation following severe spinal cord injury (SCI) constitutes one

Accepted 14 August 2014

of the major barriers to successful axonal regeneration and tissue repair. The development

Available online 3 September 2014

of bioengineered scaffolds that assist in the bridging of such lesion-induced gaps may

Keywords:

contribute to the formulation of combination strategies aimed at promoting functional

Biomaterials

tissue repair. Our previous in vitro investigations have demonstrated the directed axon

Collagen

regeneration and glial migration supporting properties of microstructured collagen scaffold

Scaffold

that had been engineered to possess mechanical properties similar to those of spinal cord

Nerve guide

tissues. Here, the effect of implanting the longitudinally orientated scaffold into unilateral

Spinal cord injury

resection injuries (2 mm long) of the mid-cervical lateral funiculus of adult rats has been

Functional recovery

investigated using behavioural and correlative morphological techniques. The resection injuries caused an immediate and long lasting (up to 12 weeks post injury) deficit of food pellet retrieval by the ipsilateral forepaw. Implantation of the orientated collagen scaffold promoted a significant improvement in pellet retrieval by the ipsilateral forepaw at 6 weeks which continued to improve up to 12 weeks post injury. In contrast, implantation of a non-orientated gelatine scaffold did not result in significant functional improvement. Surprisingly, the improved motor performance was not correlated with the regeneration of

n Corresponding author at: Haktan Altinova, Department of Neurosurgery, Evangelic Hospital Bethel, Kantensiek 11, 33617 Bielefeld, Germany. Tel.: þ49 241 8088861. E-mail address: [email protected] (H. Altinova).

http://dx.doi.org/10.1016/j.brainres.2014.08.041 0006-8993/& 2014 Elsevier B.V. All rights reserved.

38

brain research 1585 (2014) 37–50

lesioned axons through the implanted scaffold. This observation supports the notion that biomaterials may support functional recovery by mechanisms other than simple bridging of the lesion site, such as the local sprouting of injured, or even non-injured fibres. & 2014 Elsevier B.V. All rights reserved.

1.

Introduction

Severe SCI results in the loss of motor, sensory and autonomic function, largely due to the disconnection of projection neurons from their target neurons (Schwab and Bartholdi, 1996). The dramatic changes at the lesion site following acute SCI include the development of molecular and physical barriers (scarring), as well as the formation of cystic cavitations, resulting in abortive sprouting by long distance projection neurons (Eng, 1985; Fawcett and Asher, 1999; Hagg and Oudega, 2006; Hall and Springer, 2004; Morgenstern et al., 2002; Profyris et al., 2004). The presence of large cystic cavities, combined with the loss of the normal orientated geometry of spinal cord white matter tracts at the lesion site have been reported to be detrimental to axonal regeneration (Pettigrew and Crutcher, 1999; Pettigrew et al., 2001). Although a large number of experimental strategies have been developed that are capable of inducing some degree of functional recovery (Bauchet et al., 2009; Deumens et al., 2005; Schwab et al., 2006; Thuret et al., 2006), it is likely that the most effective treatment will involve a combination of approaches (Lu and Tuszynski, 2008), including the implantation of a scaffold that is capable of restoring the normal orientated geometry of the damaged fibre tracts. A vast array of engineered materials (based on either natural or synthetic polymers) has been developed to act as scaffolds that are capable of supporting axon regeneration (Führmann, 2011; Schmidt and Leach, 2003), however, the ideal bridging material has yet to be identified. Collagens have proved to be popular natural polymers in bioengineering due to their biocompatibility, non-toxic degradation products, and the relative ease of controlling their shape and structure (Burke et al., 1981; Pachence, 1996; Schoof et al., 2001; Yannas et al., 1982). Over recent years, our in vitro studies have demonstrated the cytocompatibility and axon growth promoting properties of an orientated microstructured porcine type-I collagen scaffold. The scaffold has been demonstrated to support cell attachment, proliferation and orientated migration of a range of central- and peripheral nervous system-related glia, including astrocytes, olfactory ensheathing cells, Schwann cells as well as microglia and macrophages. Furthermore, orientated axonal growth has been demonstrated by sensory neurons from dorsal root ganglia (DRG), by spinal cord ventral horn motoneurons and by differentiated human neuroblastoma cell line (SH-SY5Y) (Bozkurt et al., 2007, 2009; Gerardo-Nava et al., 2014; Mollers et al., 2009). The framework of longitudinally orientated channels of the hydrophilic scaffold with visco-elastic properties similar to those of mammalian spinal cord (Ozawa et al., 2001; Tunturi, 1978) are widely regarded as ideal physical characteristics for the promotion of graft-host integration (e.g. (Saglam et al., 2013)) and tissue repair following traumatic injury.

The use of such collagen in bridging materials in experimental models of SCI (as well as peripheral nerve injury) has taken various forms, including hollow conduits, spongefilled conduits, gels, extruded filaments and orientated microporous scaffolds (Bozkurt et al., 2012; Cholas et al., 2012a, 2012b; Deumens et al., 2010; Joosten et al., 1995; Marchand et al., 1993; Yoshii et al., 2003a, 2003b). The hollow collagen conduits and the orientated microporous scaffolds have proven successful in supporting axon regeneration across 10–20 mm gaps of the lesioned PNS (Bozkurt et al., 2012; Chamberlain et al., 1998). Similarly structured scaffolds composed of a mixture of collagen (either bovine or porcine type-I collagen) and glycosaminoglycan (chondroitin sulphate proteoglycan) have been used in attempts to promote functional tissue repair following implantation into 5 mm full resection injuries of the adult rat thoracic spinal cord (Cholas et al., 2012a, 2012b), or into hemi-resection injuries of the thoracic spinal cord (Cholas et al., 2012a). Using a subjective (modified Tarlov) behavioural test, a small but statistically significant improvement of function was noted for the implantation of the naïve scaffolds, as well as those associated with chondroitinase ABC or with mesenchymal stromal cells in the hemi-resection model of SCI, but not in the complete spinal cord transection model (Cholas et al., 2012a, 2012b; Joosten et al., 1995). Our own attempts to implant cell seeded- or naïve orientated collagen scaffolds into low thoracic hemi-resection injuries of the adult rat spinal cord had demonstrated no improvement (Deumens et al., 2013). However, our earlier observations and those of others have suggested that the hemi-resection model at low thoracic spinal cord levels may not be the ideal experimental model for biomaterial-based bridging strategies: local inflammation and distortion of the implanted scaffold may lead to reduced graft-host integration (Deumens et al., 2013), also the natural tendency for plastic reorganisation of spared or uninjured populations of descending axons (observed after partial thoracic SCI) may confound the search for axon regeneration-mediated functional recovery (Ballermann and Fouad, 2006; Fouad and Tse, 2008). The present investigation focuses on the implantation of a longitudinally orientated microstructured collagen scaffold into a unilateral 2 mm resection injury of the lateral funiculus at the mid-cervical level of the adult rat spinal cord. An objective, quantitative approach to assess behavioural recovery was performed by determining the lesion-induced deficits in food pellet retrieval of the ipsilateral forepaw, revealing a progressive, implant-mediated improvement in performance over a period of 3 month. No such recovery was noted in the control groups that were subjected to a lesion without implantation, or were implanted with a non-orientated gelatine sponge.

brain research 1585 (2014) 37–50

Fig. 1 – Schematic illustration of experimental design. Lateral funiculotomy of C3-C4 spinal cord, followed by (I) no repair, i.e. lesion only, (II) implantation of non-orientated gelatine scaffold, or (III) implantation of an orientated microstructured collagen scaffold. At 12 weeks post injury/ implantation, tissue were fixed by transcardial perfusion with 4% PFA and tissue blocks were longitudinally cryosectioned for immunohistochemical analysis.

2.

Results

2.1.

Behavioural analysis – food pellet retrieval

Animals showed a progressive improvement in food pellet retrieval during the training period and demonstrated stables values after 5–6 weeks (data not shown). Quantification of their performance at this time indicated a small, but statistically significant difference between right and left forepaw performance: more pellets being taken by the left forepaw than the right (left: 16.4171.03, right: 13.8570.033, Student's t-test: p¼ 0.033). The lateral funiculotomy (resection of 2 mm of the left lateral spinal cord tissue) resulted in an immediate reduction of ipsi-lateral forepaw function in all experimental groups (Fig. 2). The food pellet retrieval performance of the ipsilateral forepaw remained low in the lesion-only group for the full duration of the investigation, achieving a maximal score of 2.671.7 by 12 weeks post operation (wpo). However, lesioned animals receiving orientated collagen scaffold implants demonstrated an early and statistically significant recovery of performance (already detectable by 6 wpo, p¼ 0.024). Improvement in food pellet retrieval continued between 6 and 12 wpo (p ¼0.016). However, despite substantial recovery

39

Fig. 2 – Functional analysis as indicated by food pellet retrieval using the Montoya staircase device. Pre-operative baseline data indicated no statistically significant differences in left forelimb pellet retrieval between all three experimental groups (each p40.05). At 6 weeks post injury/ implantation, only the naïve, orientated collagen scaffold implanted group showed a statistically significant functional improvement in comparison to the lesion only group (p¼ 0.024). Pellet retrieval continued to improve between 6 and 12 weeks post implantation in the orientated collagen scaffold group (p¼ 0.009 compared to the lesion only group), whereas implantation of the non-orientated gelatine scaffold, showed no significant improvement (p¼ 0.056 compared to the lesion only group). Values are given as mean7SEM.

of function, the performance was still below that of the prelesion, baseline values (p¼ 0.023, Fig. 2). Lesioned animals receiving non-orientated gelatine implants (i.e. group II, acting as a control implant material) also showed a trend for recovery of function but these values did not reach the level of statistical significance when compared with the untreated, lesion-only group I (p¼ 0.056, Fig. 2).

2.2.

Morphological analysis – immunohistochemistry

The lesion/implantation sites were processed for immunohistochemistry to demonstrate host axonal responses to the resection injury, to the implanted scaffolds and the degree of graft-host integration. The gap caused by the resection of the lateral spinal cord in the lesion-only group (group I) had, in most cases, been occupied by migrating cells that formed a dense scar-like tissue (Fig. 3A). The DAPI-labelled nuclei of the cells demonstrated no particular orientation within the connective tissue scar but some small areas of cystic cavitation could be observed (e.g. asterisk, Fig. 3A), which in a few instances occupied most of the lesion site (asterisk, Fig. 3B, C). Intensely GFAP-immunoreactive astrocytes and their processes could be observed at and around the lesion edge, presenting a rather ragged interface (Fig. 3A). Some NF 200positive axons could be seen following a convoluted, nonorientated trajectory within the scar-like tissue (small arrows, Fig. 3A). At the cranial and caudal edges of the lesion site, NF 200 and GAP-43 could clearly be seen, some of which demonstrated the rounded appearance of retraction bulbs (see arrow, Fig. 3B, C). Bridging the lesion site with the orientated collagen scaffolds (group III) was also followed by a phase of extensive

40

brain research 1585 (2014) 37–50

Fig. 3 – Immunohistochemistry of the lesion site at 12 weeks following injury and no repair (lesion only control group). (A): Dense host cell migration into the lesion site, as revealed by DAPI-labelled nuclei, demonstrates the formation of disorganised connective tissue scarring. The presence of small, fluid filled cystic cavities can also be seen (e.g. asterisk). Occasional, randomly orientated NF 200-positive axons can be seen within the lesion (small arrows), whereas the GFAP-positive reactive astrocytes from the ragged edge around the lesion site. (B): NF 200-positive axonal profiles at the lesion edge. Numerous bulbous profiles can be identified, some of which (are also immunoreactive for GAP-43 (white arrow)). (C): GAP-43immunoreactivity of axonal profiles at the lesion edge. (D): Combined image of NF 200 (red) and GAP-43 (green) immunoreactivity. Scale bars, (A): 100 lm, (B–D): 20 lm.

host cell migration. However, the density of the migrating cells appeared somewhat lower than that observed in the lesion-only group (compare Figs. 3 and 4A). Furthermore, the migrating cells formed columns along the longitudinal axis of the microporous framework within the scaffold (e.g. asterisk, Fig. 4A and at higher magnification in Fig. 4E). The oval shaped, DAPI-labelled nuclei gave the impression of groups of overlapping cells (arrows, Fig. 4E) that followed the walls of the collagen scaffold. The lumen of many of the orientated channels within the scaffolds (asterisks, Fig. 4E) appeared to be free of cells. In addition to the organised columns of migrating cells within the orientated collagen scaffold, a bridge of densely packed, longitudinally orientated cells also developed outside the scaffold, along its lateral-most edge (e.g. arrow, Fig. 4A). This newly formed tissue bridge appeared to be at the interface between the lateral edge of the implant and the inner surface of the dura mater (which was often lost from most sections during tissue processing; however see dura mater in Fig. 5C). Numerous NF 200-positive axons could often be seen coursing along this tissue bridge (arrow, Fig. 4B, D). These NF 200-positive axons extended over substantial

distances, sometimes completely spanning the entire outer surface of the scaffold. However, only occasional axons could be seen within the orientated collagen scaffold itself (e.g. arrows, Fig. 5A). GFAP-immunoreactivity demonstrated the strongly stained reactive astrocytes and their processes around the implant (Fig. 5B). Such GFAP-positive astrocytic processes extended for only short distances along the outer, lateral edge of the scaffold (if at all) or occasionally entered the scaffold perimeter for distances of 200–300 mm. The limited penetration of the collagen scaffold by GFAPpositive astrocytic profiles (e.g. Fig. 6B) was also associated with some limited penetration by MAP-2a/b-positive dendritic processes (Fig. 6A and C). The growing dendritic profiles appeared to follow the processes of the reactive astrocytes at the graft-host interface (e.g. arrows, Fig. 6). Implantation of the control, non-orientated gelatine scaffold (group II) also resulted in a dense host cell migration into the non-orientated internal porous framework of the scaffold (Fig. 7A). Similar to the collagen implants, scattered, individual GFAP-positive cells and processes could be observed following the lateral tissue bridge for short distances (Fig. 7B).

brain research 1585 (2014) 37–50

41

Fig. 4 – Immunofluorescence of the lesion site at 12 weeks after implantation of the longitudinally microstructured collagen scaffold. (A): DAPI nuclear counterstain demonstrates host cell migration into the scaffold, forming longitudinally orientated columns (e.g. asterisk, also shown at higher magnification in Fig. 4E). A newly formed tissue bridge at the lateral edge of the implanted scaffold can also be seen (arrow). (B): NF 200-positive axon regeneration (arrow) following the trajectory of the newly formed tissue bridge. (C): Phase contrast demonstrates the outline of the implanted scaffold (highlighted by the dotted line) as well as the newly formed tissue bridge at the lateral edge of the implant (arrow). (D): Overlay of phase contrast and NF 200 immunohistochemistry. (E): Migrating host DAPI-labelled cells forming orientated columns (arrows) of cells. The lumen of the orientated microchannels within the scaffold (asterisks) often remained devoid of cells and were presumably fluid filled. Scale bar: 50 lm.

In contrast, many longitudinally orientated NF-positive and GAP-43-positive profiles could be observed along the tissue bridge, passing around the implant (e.g. arrow, Fig. 7C) while a random, non-orientated pattern of growth could be seen at

the rostral and caudal interfaces of the sponge. The development of the lateral tissue bridge around the implants, along which regenerating NF 200- and GAP-43-positive profiles could be observed, was a phenomenon that was only

42

brain research 1585 (2014) 37–50

Fig. 5 – Limited penetration of regenerating axons into the microstructured collagen scaffold at 12 weeks post implantation. (A): Longitudinally orientated NF 200-positive axons (arrows) extend within the scaffold, following the axis of the scaffold's micropores. (B): Strongly GFAP-immunopositive reactive astrocytic profiles (arrows) close to the implant-host interface, some of which also penetrate the scaffold for a limited distance. (C): DAPI-stained host nuclei demonstrate the formation of columns of cells within the scaffold. A remnant of the host dura mater is indicated by the arrow. (D): Dark-field microscopy also demonstrates the borders of the implant (dashed line) and the intimate contact between graft and host. Scale bar: 100 lm.

detected in the 2 implantation groups (Group II and Group III). No similarly positioned lateral tissue bridges could be detected in any of the lesion only (Group I) animals. Although some orientated growth of axons was observed within naïve collagen scaffolds (group III), the total number of NF 200-positive axons within the scaffold was not significantly different to that observed within control groups I and II (lesion-only and non-orientated gelatine respectively, each p ¼ 0.9) (Fig. 8A). However, significantly more GAP-43positive profiles could be detected in the implanted orientated collagen scaffolds (Group III) compared with lesiononly (Group I, p ¼0.03) and non-orientated gelatine implants (Group II, p ¼0.03, but no significant difference between groups I and II, p ¼0.7, Fig. 8B). Interestingly, the degree of reactive astrocytosis around the lesion or implantation sites, as indicated by the intensity of GFAP immunoreactivity, was significantly reduced following implantation of the orientated collagen scaffold (Group III) in comparison to the lesion-only animals (Group I, p ¼0.03) and to nonorientated gelatine implanted animals (Group II, p ¼0.03), with no differences being detected between Group I and II (p ¼0.34, Fig. 8C).

3.

Discussion

Apart from injury-induced secondary tissue degeneration, the presence of growth inhibitory molecules and the development of reactive astrocytic- and connective tissue scarring, the generation of fluid-filled cystic cavities acts as a significant barrier to axon regeneration and functional tissue repair (Bartholdi and Schwab, 1998; Eng, 1985; Faulkner et al., 2004; Fawcett and Asher, 1999; Shearer and Fawcett, 2001; Silver and Miller, 2004; Stichel and Muller, 1998a, 1998b; Tator, 2002). Implantable biomaterials that support the bridging of lesion-induced cavities, by promoting host cell migration and axonal growth across tissue gaps, may prove useful in the development of combinational therapeutic strategies in regenerative medicine. The present study has demonstrated the beneficial functional effects of implanting biomaterial scaffolds into acute unilateral resection injuries of the C3-C4 adult rat spinal cord white matter. Significant improvement of ipsi-lateral forelimb function was observed after implantation of the longitudinally microstructured collagen scaffold (group III). A trend for improvement was also observed following implantation of the non-orientated gelatine

brain research 1585 (2014) 37–50

43

Fig. 6 – Novel growth of dendritic profiles into the implanted orientated collagen scaffolds. (A): MAP-2a/b-positive dendritic profiles (arrows) can clearly be seen crossing the graft-host interface and extending along the axis dictated by the longitudinally orientated collagen micropores. (B): Intense host GFAP-positive astrocytic process (arrows) also extend beyond the graft-host interface. (C): Overlay demonstrating the close association between the novel dendritic growth and host astrocytic profiles. No such pattern of host integration into the scaffold was observed following implantation of the nonorientated gelatine matrix. Scale bar: 50 lm.

scaffold (control group II), however, the values from this group failed to reach the level of statistical significance. It is probable that, although 10 animals were operated in each group, the relatively high rate of animal exclusion (e.g. for inappropriate, bilateral, functional deficits after injury) resulted in a final n for group II (i.e. n¼ 5) that was insufficient to demonstrate statistical significance. Future studies should include a higher n value to ensure greater statistical power. During the training phase, prior to surgery/implantation, a degree of “left-handedness” of animals was observed when comparing the number of pellets retrieved by the left and right forepaws. Others have investigated fore-paw preference in laboratory animals and reported a tendency for right forepaw preference, with no sex-related differences being observed (Guven et al., 2003; Pence, 2002). The present data suggests that female inbred Lewis rats show a different pattern of handedness. The left-sided C3-C4 lateral funiculotomy of the spinal cord induced a marked and permanent reduction of skilled ipsi-lateral fore-paw function. A similar but less severe deficit in food pellet retrieval function has also been demonstrated following dorso-lateral rat spinal cord funiculotomy. Such lesions were reported to affect the

control of distal fore-limb musculature, resulting in a substantial reduction of digit flexion (Schrimsher and Reier, 1993). It is possible that the more severe deficits observed in the present investigation are due to the larger lesions induced at C3-C4 (the lateral funiculotomies causing partial destruction of the ventro-lateral funiculus as well as of the dorso-lateral funiculus). The rationale of implanting the orientated collagen scaffold into the resection injuries was to determine the extent to which cell-scaffold interactions observed in vitro reflect the events that take place in the more challenging in vivo environment, as well as promoting tissue repair by directed host axon regeneration across the lesion. Similarly structured collagen scaffolds have previously been used in attempts to promote functional tissue repair following spinal cord injury (Cholas et al., 2012a, 2012b; Spilker et al., 2001). In recent developments, naïve (or non-seeded) collagen scaffolds with an orientated microporous framework, as well as scaffolds that were seeded with a range of axon growth promoting cell types (i.e. olfactory ensheathing cells, neural- and bone marrow-derived mesenchymal stem cells), or other potentially therapeutic agents (i.e. laminin, recombinant soluble

44

brain research 1585 (2014) 37–50

Fig. 7 – Implantation of the non-orientated gelatine scaffold also induced the formation of a tissue bridge at the lateral edge of the implant. (A): DAPI-staining reveals the non-orientated migration of host cells into the scaffold as well as at the lateral border of the implant (arrow). (B): Densely packed, intensely GFAP-immunoreactive astroglia close to the graft-host interface (indicated by the dotted line). Some orientated astrocytic processes extend a short distance along the newly formed lateral tissue bridge (arrow). (C): GAP-43 immunoreactive profiles coursing along the lateral tissue bridge with little penetration into the non-orientated gelatine implant itself. (D): Overlay showing the spatial relationship between the reactive GFAP-positive host astrocytes and the GAP-43-positive host axons and Schwann cells along the lateral tissue bridge. Scale bar: 100 lm.

Fig. 8 – Quantification of axon growth (NF 200) and combined axon/Schwann cell growth (GAB-43) into the lesion/implantation site, as well as of the extent of reactive astrogliosis at the lesion/implantation edge. (A): No significant differences were detected in the number of NF 200-positive axons penetrating the lesion site of non-implanted, non-orientated gelatineimplanted or orientated collagen scaffold implanted lesion sites (p40.05 for all comparisons). (B): The orientated collagen scaffold induced significantly more ingrowth by GAP-43 immunoreactive profiles (axons and de-differentiated Schwann cells) than was observed in the non-implanted (po0.05) or non-orientated gelatine-implanted groups (po0.05). No differences were found between the lesion only and non-orientated gelatine-implanted groups (p40.05). (C): Implantation of the collagen scaffold also resulted in a reduction of GFAP-positive astrogliosis at- and around the lesion/implantation edge in comparison to the lesion only and non-orientated gelatine-implanted groups (po0.05). No differences were found between the lesion only and non-orientated gelatine-implanted groups (p40.05). Values are given as medians and lower/upper quartiles.

brain research 1585 (2014) 37–50

Nogo receptor and chondroitinase ABC) have been implanted into thoracic spinal cord injury models of the adult rat (Cholas et al., 2012a, 2012b; Deumens et al., 2013; Spilker et al., 1997). A number of issues arose in those studies that may have interfered with the demonstration of the beneficial effects of implanted scaffolds. The extensive laminectomy and complete spinal cord transection model employed by Cholas and colleagues was reported to result in compression of the lesion/implantation site by surrounding tissues (Cholas et al., 2012b). Such deleterious compression at the lesion/ implantation site is also likely to have occurred in the unilateral thoracic spinal cord hemisection model which we adopted earlier for the implantation of a cylindrical scaffold into a semi-cylindrical lesion cavity (Deumens et al., 2013). However, neither beneficial nor detrimental effects of collagen scaffold implantation could be observed (Deumens et al., 2013). Similarly, no beneficial functional effect was reported 6 weeks after implantation of the collagen scaffold (with or without ECM, growth factor or stem cell supplements) across the complete thoracic spinal cord transection injury. This apparent lack of effect may, in part, have been due to the caps of scar tissue that formed over the rostral and caudal stumps of the lesioned spinal cord (Cholas et al., 2012b). An unilateral 3 mm hemi-resection injury model was also adopted by Cholas and colleagues into which a semicylindrical collagen scaffold was implanted, however, following the use of relatively short survival times, the authors reported no statistically significant differences between the mean scores of all groups at 1 and 4 weeks post implantation (Cholas et al., 2012a). However, some of the experimental groups demonstrated a small but statistically significant within-group improvement of function (assessed by a modified Tarlov score) over the 4 week period. In the present study, the use of the unilateral C3-C4 lateral funiculotomy model and survival times of 12 weeks has allowed the clear demonstration of functional improvement of the ipsilateral forelimb following the implantation of the naïve, non-seeded scaffold. The use of the Montoya food pellet retrieval test also allowed for the generation of quantifiable and objective data based on the number of food pellets retrieved by the animals within a strictly defined time span. The present investigation also used the implantation of non-orientated gelatine, with its non-orientated sponge-like microporosity, as a control scaffold. The immunohistochemical studies demonstrated that, in some instances, the resection injuries contained little or no cellular content and had likely developed into areas of fluid-filled cavitation. However, most of the lesion sites contained a dense cellular infiltrate. Earlier investigations have shown the development of similar dense connective tissue scarring (Cholas et al., 2012b; Spilker et al., 1997), and have demonstrated a range of cell types migrating into the lesion sites, including monocytes and macrophages, ependymal cells, Schwann cells, endothelial cells and leptomeningeal fibroblasts (Brook et al., 1998b, 1999, 2000; Cholas et al., 2012b; Guth et al., 1994, 1999; Pasterkamp et al., 1999). The ability of axons to regenerate into such spinal cord lesions depends on the cellular composition of the forming connective tissue scar as well as on the relative balance of

45

axon growth-promoting and growth-inhibitory molecules expressed in- and around the lesion site (Fitch and Silver, 2008; Pernet and Schwab, 2012). The presence of the rounded NF 200-positive and GAP-43-positive axonal processes around the edge of the lesions is likely to represent dystrophic growth cones: indicators of axon regeneration failure (Tom et al., 2004). The few regenerating neurofilament-positive axons that extended beyond the intensely GFAP-positive reactive astrocytes at the lesion edge were clearly randomly orientated within the connective tissue scar. Immunofluorescence was used to assess the relationship between improved fore-limb function and axonal regrowth across the lesion following implantation of the orientated collagen scaffold in the present study. Although substantial host cell infiltration could be observed within the implanted scaffold, much of which clearly followed the longitudinal axis of the microstructured framework, little or no axonal regrowth could be observed within the scaffold. The orientation of migrating host cells showed a striking similarity to that already documented following the seeding of a range of glial cell types (i.e. Schwann cells, astrocytes and olfactory ensheathing cells) into the scaffold in vitro (Bozkurt et al., 2007, 2009; Mollers et al., 2009) or following the implantation of orientated agarose scaffolds into the lesioned spinal cord (Stokols et al., 2006; Stokols and Tuszynski, 2006). The relatively few axonal profiles that crossed the host-implant interface adopted a longitudinal orientation, similar to that of migrating host cells, but extended no further than a few hundred microns in length. The most impressive axonal response by far was that observed outside the scaffold, between the lateral-most edge of the implant and dura mater. It is unlikely that this lateral tissue bridge, observed in many of the implanted animals, was due to tissue sparing since particular attention had been paid during the initial surgical procedure to ensure that no residual lateral white matter remained at the lesion site. Therefore, it is reasonable to conclude that the numerous NF 200-positive and GAP-43positive profiles observed within the newly formed tissue bridges included regenerating axons (Steward et al., 2003). This region was also rich in densely packed, longitudinally orientated host cells (as indicated by shape and direction of the ovoid, DAPI-labelled nuclei). An unequivocal identification of the cellular composition of this lateral tissue bridge was not performed in the present investigation, however, the intense cellular density and GAP-43 immunoreactivity suggests that many of the cells were Schwann cells. In addition to regenerating axons, de-differentiated Schwann cells are known to express high levels of GAP-43 protein (Plantinga et al., 1993; Scherer et al., 1994) and are capable of migrating from adjacent damaged spinal nerve roots into spinal cord lesions (Brook et al., 1998a, 2000; Guth et al., 1994, 1999). Furthermore, the interface between dura mater and an implanted, non-cell seeded poly-ß-hydroxybutyrate scaffold has also been reported to be a region associated numerous low affinity nerve growth factor receptor-positive Schwann cells and regenerating axons (Novikova et al., 2008). The lateral-most edge of implanted materials (natural or synthetic) appears to be a significant region for regeneration, as also observed by others (Bunge, 2002). After implantation of a

46

brain research 1585 (2014) 37–50

2-hydroxypropyl methacrylamide hydrogel either alone or seeded with mesenchymal stem cells, NF 160-positive axons were reported to have grown along the outer edge of the implants. Furthermore, Schwann cells originating from the spinal root entry zone infiltrated the implants in close association with axons (Hejcl et al., 2010). Implantation of a similar hydrogel into T10 hemisection injuries resulted in numerous NF-positive axons being detected within, but also particularly at the edge of the matrices (Pertici et al., 2013). Compared with our observation these lateral tissue bridges appear more unorientated and discontinuous. Immunohistochemistry and quantification of the present investigation also suggested that substantially more GAP-43-positive profiles could be observed within the collagen scaffold than NF 200positive profiles. This may indicate that NF 200-/GAP-43positive axons as well GAP-43-positive Schwann cells have infiltrated into the scaffold. Host astrocytes demonstrated minimal penetration or migration into the implanted orientated collagen scaffolds, however, this is in contrast to the migration reported by others into implanted empty collagen tubes or tubes containing the orientated collagen-glycosaminoglycan scaffold (Spilker et al., 1997). This would suggest that the presence of the chondroitin 6 sulphate proteoglycan within the scaffold framework (developed by Spilker and colleagues) was important for promoting host astrocyte integration into the scaffold. In the present study, GFAP-immunoreactivity also indicated that host astrocytes contributed minimally to the cellularity of the lateral tissue bridge, with only occasional fine, longitudinally orientated process showing some degree of mixing with the cells that coursed along the lateral margin of the implant. This observation may also, indirectly, support the notion that the lateral tissue bridge was largely in composed of Schwann cells (Aguayo et al., 1981). Although Schwann cells and astrocytes are known to have mutually repulsive characteristics and occupy anatomically distinct territories (Adcock et al., 2004; Fraher, 1997), microtransplantation of Schwann cell columns into host central nervous system (CNS) parenchyma has revealed the capacity of these cells to undergo a surprising degree of intimate intermingling, and to support axonal regeneration (Brook et al., 1994, 2001). Similarly, others have shown that implantation of genetically engineered fibroblasts (another cell type sharing mutually repulsive properties with astrocytes) into CNS parenchyma can induce cooperative interactions with host astrocytes in the support of axonal growth (Kawaja and Gage, 1991). It is possible that the activated, GFAP-positive astrocytic processes seen extending into the lateral tissue bridge, around the implanted scaffold, may reflect such Schwann cell-astrocyte interactions. Surprisingly, the lateral margin of the control, nonorientated gelatine implant also demonstrated the longitudinally orientated tissue bridge. As mentioned earlier, these implants did not result in any statistically significant improvement of function. It is possible therefore, that the NF 200- and GAP-43-positive axons growing around the lateral margin of the implanted scaffolds may not be responsible for the improved behavioural performance. However, it remains to be determined whether different axonal phenotypes are capable of extending in the environments

surrounding the orientated collagen and non-orientated gelatine implants. In this context, it was also surprising to observe that only collagen implants were capable of supporting the (limited) growth of host MAP2a/b-positive dendrites into the scaffolds. Although most studies on the use of biomaterials for nervous tissue repair focus on the extent of axonal growth and integration within the implanted scaffold, this is to the best of our knowledge, the first demonstration of attempted integration by host dendritic profiles. This observation clearly supports the notion that the ability of the implanted scaffold to interact with the surrounding host tissues depends on its molecular composition and its microstructure. Overall, the immunohistochemical data has demonstrated that although substantial axonal and glial (including astroglial) penetration of the scaffold has been described in in vitro investigations (Bozkurt et al., 2007, 2009; GerardoNava et al., 2014; Mollers et al., 2009), no such behaviour was found following implantation into the lesioned spinal cord. It is clear that the use of in vitro investigations, although of substantial importance for studying issues of cytocompatibility and particular cell-substrate interactions, have limited predictive potential when considering implantation into complex tissues such as CNS. The confounding issues of local bleeding, inflammation and reactive scar formation around the implant obviously influence the extent and type of cell-substrate interactions that will take place. Indeed, it is likely that the reactive nature of the host astroglial response was responsible for the limited penetration of astroglial processes and the lack of axon regeneration into the implanted scaffold. The presence of scar-associated molecules at- and around the lesion or implant-host interface act as potent inhibitors to host axon regeneration (Fawcett et al., 2012; Silver and Miller, 2004). The present data also suggest that mechanisms, other than directed host axonal regeneration through the implanted scaffold may be responsible for the observed functional improvement. Functional recovery in the absence of graft or implant-mediated regeneration of the originally cut fibres has been reported by others (Toft et al., 2007; Yamamoto et al., 2009). Compensatory mechanisms brought about by the local sprouting of injured or un-injured nerve fibres should be considered in future experiments. The sprouting of intact fibres to reinnervate partially denervated territories has been clearly demonstrated following the neutralisation of myelin associated Nogo-A (Bareyre et al., 2004; Raineteau et al., 2002). It may be that the implanted collagen scaffold induced, as yet, undefined changes in the CNS neuropil that supported enhanced local sprouting. Remarkably, incomplete injuries to the adult rat spinal cord have been demonstrated to result in the spontaneous formation of novel propriospinal connections that could relay information to deafferented target neurons, effectively forming new neuronal circuits by-pass the lesion site (Bareyre et al., 2004).

4.

Conclusion

The present study emphasizes the notion that bridging a lesion-induced gap after SCI may lead to a host tissue

brain research 1585 (2014) 37–50

response in the form of orientated axonal regrowth following an unusual trajectory, around the implanted scaffold, or the activation of other, as yet, undefined tissue repair mechanisms leading to functional recovery. The mechanisms by which implanted biomaterials interact with host tissues are thus likely to be more complex than currently appreciated. Nonetheless, the ability of a naïve (non-cell seeded) orientated collagen scaffold to promote a significant level of functional improvement following SCI is an encouraging observation. The controlled modification of the properties or contents of such implantable scaffold may highlight, as yet, poorly appreciated mechanisms of tissue repair and will likely contribute to the development of more sophisticated and targeted combination strategies for the treatment of traumatic CNS injuries.

5.

Experimental procedure

5.1.

Experimental animals

The experiments were performed on adult female Lewis rats (weight 180–200 g at the start of experiments, Charles River, Germany). The animal care and experimental procedures were carried out in accordance with the guidelines of the German and EU animal protection statutes. In all cases, every attempt was made to minimise the number of animals used in the investigations, as well as any pain or discomfort. Animals were housed under temperature controlled conditions at 2171 1C, with a normal 12:12 h light/dark cycle with ad libitum access to water. A diet restriction protocol (12 g food/rat/day) was used during the pre-operative period of staircase training (see below) and throughout the study.

5.1.1.

Functional analysis

Animals were trained to retrieve food pellets from the Perspex Montoya stairwell device (Camden Instruments, London, UK) as described previously (Montoya et al., 1991). Animals were trained for a minimum of 3 times per week for up to 6 weeks prior to surgery, during which time they were placed in the device for 15 minutes and were allowed to search for the food pellets. The stairwell encourages rats to make coordinated reaching and grasping movements to successfully retrieve the food pellets. It is particularly well suited for animals being subjected to unilateral lesions because the device prevents the use of the unaffected forepaw for pellet retrieval from the side of the affected forepaw. Each hollowed step of the stairwell was filled with 4  4 mg pellets of food (Lohmann Research Equipment, Castrop-Rauxel, Germany).

5.2.

Surgical procedure

A sub-cutaneous injection with Buprenorphine (Temgesic 0.1 mg/kg body weight; Schering-Plough, Utrecht, The Netherlands) was administered to all animals 30–60 min prior to surgery. Anaesthesia was induced by inhalation of a 4-5% mixture of isoflurane in air, and maintained at 2% isoflurane using a U-400 anaesthesia unit (Agntho's, Lidingö, Sweden). The drying of the eyes was prevented by applying a small amount of ophthalmic ointment prior to surgery and body

47

temperature maintained using a thermostatically controlled heating pad. The neck and shoulders of the rats were shaved and disinfected before making a midline incision of the skin. The dorsal surfaces of the C3-C4 vertebrae were exposed by blunt dissection of the overlying muscles, and a laminectomy performed to expose the underlying spinal cord. Using an operating microscope, a small window (or flap) in the dura was opened and a 2 mm long, left-sided lateral funiculotomy was performed using micro-scissors. After establishing haemostasis, completeness of the lateral funiculotomy was verified microscopically following aspiration of the lesion site. Care was taken to prevent damage to major blood vessels. The lesion site was either left untreated or filled with appropriately shaped pieces of non-orientated gelatine scaffold or orientated microstructured collagen scaffold (see schematic illustration, Fig. 1). The dura mater was then closed using 10/0 single stitch sutures (Ethicon Inc., Somerville), followed by re-adaption of the overlying muscle layers and closure of the skin using 4/0 single stitches (Prolenes, Ethicon Inc., Somerville). Animals were divided into three groups: (I) those receiving no implant (n¼ 10), (II) those receiving a non-orientated gelatine scaffold (Spongostan, Johnson & Johnson) implant, (n¼ 10), and (III) those receiving the orientated microstructured collagen scaffold (Matricel GmbH, Herzogenrath, Germany; (Mollers et al., 2009)), (n¼ 10). Due to insufficient task learning, surgical complications or inappropriate paw deficits, the final animal numbers used in the investigation were n¼ 5 for groups I and II and n¼ 7 for group III. Food pellet retrieval tests for the left forepaw (ipsilateral to the spinal cord lesion) were obtained at 6 and 12 weeks post-surgery (wps).

5.3.

Tissue processing and double immunofluorescence

At 12 wps, animals received an intraperitoneal overdose of sodium pentobarbital (150 mg/kg body weight) and were transcardially perfused with 50 ml saline followed by 300 ml paraformaldehyde (PFA) in 0.1 M phosphate buffered saline (PBS, pH 7.4). The spinal cord was carefully dissected, removed from the spinal canal, and post-fixed for up to 24 h in 4% PFA. Samples were then cryoprotected (immersion in 20% sucrose in 0.1 M PBS) for at least three days at 41C. Segments of spinal cord (1 cm long), centred around the lesion site, were snap-frozen using powdered dry-ice and horizontal cryosections (30 μm thick) were prepared (e.g. Fig. 1). Sections were serially mounted onto adjacent SuperFrost-Plus slides (such that sections on the same slide were obtained from tissue 300 μm apart) and stored at -80 1C for later immunohistochemical processing. A detailed description of the immunohistochemical procedure has been published elsewhere (Brook et al., 2001). Briefly, sections were incubated overnight at room temperature with the following primary antibodies: monoclonal anti-growth-associated-protein-43 (GAP-43, 1:2.500, Chemicon), anti-phosphorylated neurofilament 200 kDa (NF 200, Clone NE14, 1:5.000, Sigma), anti-microtubule-associated-protein 2aþb (MAP2, 1:1.000, Sigma), polyclonal anti-glial fibrillary acidic protein (GFAP, 1:2.000, DAKO), and polyclonal anti-GAP-43 (1:2.500, Chemicon). The next day, sections were washed and incubated in fluorochrome–conjugated secondary antibodies: Alexa

48

brain research 1585 (2014) 37–50

488-conjugated goat anti-rabbit IgG and Alexa-594 conjugated goat anti-mouse IgG (1:500, Molecular Probes) for 2 h at room temperature. Sections were counterstained with the nuclear marker 40 ,6-diamidino-2-phenylindole (DAPI, 1 μg/ml, Invitrogen), cover-slipped using Fluoprep media (Bio Mérieux) and observed using a Zeiss Axioplan epi-fluorescence microscope connected to a Zeiss AxioVision CCD camera. Images were processed and stored using the Zeiss AxioVision 3.1 software.

5.3.1.

Quantification of immunohistochemistry

Three sections (i.e. from the dorsal, middle, ventral part of the lesion site/implant) per spinal cord were examined for quantification. The number of NF 200-positive axons and GAP-43-positive profiles that crossed a reference line (drawn perpendicular to the axis of the spinal cord) in the cranial portion (150 mm from the lesion/implant-host interface), middle, and caudal parts (150 mm from the lesion/implant-host interface) of the lesion or implant. Furthermore, ImageJs (Image Processing and Analysis in Java) was used to determine the percentage of the photographic fields of interest that were occupied by GFAP-immunoreactive astroglia (as an indicator of reactive astrogliosis).

5.4.

Statistical analyses

Forelimb function tests and morphological data were analysed and presented in graphical form using Graph Pad Prism, version 4 (San Diego, CA, USA). To evaluate the presence of any right- or left forelimb preference in the food pellet retrieval test, a paired Student's t-test was used. Furthermore, at three specific time-points (i.e. pre-operative baseline, 6 wps, 12 wps), an unpaired Mann Whitney U test was used to determine any statistical differences of medians of left forelimb function between the groups. Comparisons within a group between different timepoints were assessed by using a paired Wilcoxon signed rank test. The unpaired Mann Whitney U test was used to assess the number of NF 200-positive and GAP-43-positive profiles within the lesions and implants, as well as the intensity of the GFAP-immunoreactivity at the lesion/implant edge. Behavioural data presented are mean7standard error of the mean (SEM) (Fig. 2). All graphical data of the morphological analyses presented are medians with their lower and upper quartiles (Fig. 8). A p-value of 0.05 was considered as statistically significant.

Acknowledgement This work was supported by the Deutsches Zentrum für Luftund Raumfahrt e.V. Projektträger des BMBF Gesundheitsforschung (01GN0109) and the EC FP6 project RESCUE (LSHBCT-2005-518233). Support from the Interdisciplinary Centre for Clinical Research “IZKF – BIOMAT” within the faculty of Medicine at the RWTH Aachen University is also gratefully acknowledged.

r e f e r e n c e s

Adcock, K.H., Brown, D.J., Shearer, M.C., Shewan, D., Schachner, M., Smith, G.M., Geller, H.M., Fawcett, J.W., 2004. Axon behaviour at Schwann cell – astrocyte boundaries: manipulation of axon signalling pathways and the neural adhesion molecule L1 can enable axons to cross. Eur. J. Neurosci. 20, 1425–1435. Aguayo, A.J., David, S., Bray, G.M., 1981. Influences of the glial environment on the elongation of axons after injury: transplantation studies in adult rodents. J. Exp. Biol. 95, 231–240. Ballermann, M., Fouad, K., 2006. Spontaneous locomotor recovery in spinal cord injured rats is accompanied by anatomical plasticity of reticulospinal fibers. Eur. J. Neurosci. 23, 1988–1996. Bareyre, F.M., Kerschensteiner, M., Raineteau, O., Mettenleiter, T. C., Weinmann, O., Schwab, M.E., 2004. The injured spinal cord spontaneously forms a new intraspinal circuit in adult rats. Nat. Neurosci. 7, 269–277. Bartholdi, D., Schwab, M.E., 1998. Oligodendroglial reaction following spinal cord injury in rat: transient upregulation of MBP mRNA. Glia 23, 278–284. Bauchet, L., Lonjon, N., Perrin, F.E., Gilbert, C., Privat, A., Fattal, C., 2009. Strategies for spinal cord repair after injury: a review of the literature and information. Ann. Phys. Rehabil. Med. 52, 330–351. Bozkurt, A., Brook, G.A., Moellers, S., Lassner, F., Sellhaus, B., Weis, J., Woeltje, M., Tank, J., Beckmann, C., Fuchs, P., Damink, L.O., Schugner, F., Heschel, I., Pallua, N., 2007. In vitro assessment of axonal growth using dorsal root ganglia explants in a novel three-dimensional collagen matrix. Tissue Eng.. Bozkurt, A., Deumens, R., Beckmann, C., Olde Damink, L., Schugner, F., Heschel, I., Sellhaus, B., Weis, J., Jahnen-Dechent, W., Brook, G.A., Pallua, N., 2009. In vitro cell alignment obtained with a Schwann cell enriched microstructured nerve guide with longitudinal guidance channels. Biomaterials 30, 169–179. Bozkurt, A., Lassner, F., O’Dey, D., Deumens, R., Bocker, A., Schwendt, T., Janzen, C., Suschek, C.V., Tolba, R., Kobayashi, E., Sellhaus, B., Tholl, S., Eummelen, L., Schugner, F., Damink, L.O., Weis, J., Brook, G.A., Pallua, N., 2012. The role of microstructured and interconnected pore channels in a collagen-based nerve guide on axonal regeneration in peripheral nerves. Biomaterials 33, 1363–1375. Brook, G.A., Lawrence, J.M., Shah, B., Raisman, G., 1994. Extrusion transplantation of Schwann cells into the adult rat thalamus induces directional host axon growth. Exp. Neurol. 126, 31–43. Brook, G.A., Plate, D., Franzen, R., Martin, D., Moonen, G., Schoenen, J., Schmitt, A.B., Noth, J., Nacimiento, W., 1998a. Spontaneous longitudinally orientated axonal regeneration is associated with the Schwann cell framework within the lesion site following spinal cord compression injury of the rat. J. Neurosci. Res. 53, 51–65. Brook, G.A., Schmitt, A.B., Nacimiento, W., Weis, J., Schroder, J.M., Noth, J., 1998b. Distribution of B-50(GAP-43) mRNA and protein in the normal adult human spinal cord. Acta Neuropathol (Berl) 95, 378–386. Brook, G.A., Perez-Bouza, A., Noth, J., Nacimiento, W., 1999. Astrocytes re-express nestin in deafferented target territories of the adult rat hippocampus. Neuroreport 10, 1007–1011. Brook, G.A., Houweling, D.A., Gieling, R.G., Hermanns, T., Joosten, E.A., Bar, D.P., Gispen, W.H., Schmitt, A.B., Leprince, P., Noth, J., Nacimiento, W., 2000. Attempted endogenous tissue repair following experimental spinal cord injury in the rat:

brain research 1585 (2014) 37–50

involvement of cell adhesion molecules L1 and NCAM?. Eur. J. Neurosci 12, 3224–3238. Brook, G.A., Lawrence, J.M., Raisman, G., 2001. Columns of Schwann cells extruded into the CNS induce in-growth of astrocytes to form organized new glial pathways. Glia 33, 118–130. Bunge, M.B., 2002. Bridging the transected or contused adult rat spinal cord with Schwann cell and olfactory ensheathing glia transplants. Prog. Brain Res. 137, 275–282. Burke, J.F., Yannas, I.V., Quinby Jr., W.C., Bondoc, C.C., Jung, W.K., 1981. Successful use of a physiologically acceptable artificial skin in the treatment of extensive burn injury. Ann. Surg. 194, 413–428. Chamberlain, L.J., Yannas, I.V., Hsu, H.P., Strichartz, G., Spector, M., 1998. Collagen-GAG substrate enhances the quality of nerve regeneration through collagen tubes up to level of autograft. Exp. Neurol. 154, 315–329. Cholas, R., Hsu, H.P., Spector, M., 2012a. Collagen scaffolds incorporating select therapeutic agents to facilitate a reparative response in a standardized hemiresection defect in the rat spinal cord. Tissue Eng. Part A. 18, 2158–2172. Cholas, R.H., Hsu, H.P., Spector, M., 2012b. The reparative response to cross-linked collagen-based scaffolds in a rat spinal cord gap model. Biomaterials 33, 2050–2059. Deumens, R., Koopmans, G.C., Joosten, E.A., 2005. Regeneration of descending axon tracts after spinal cord injury. Prog Neurobiol. 77, 57–89. Deumens, R., Bozkurt, A., Meek, M.F., Marcus, M.A., Joosten, E.A., Weis, J., Brook, G.A., 2010. Repairing injured peripheral nerves: Bridging the gap. Prog. Neurobiol. 92, 245–276. Deumens, R., Van Gorp, S.F., Bozkurt, A., Beckmann, C., Fuhrmann, T., Montzka, K., Tolba, R., Kobayashi, E., Heschel, I., Weis, J., Brook, G.A., 2013. Motor outcome and allodynia are largely unaffected by novel olfactory ensheathing cell grafts to repair low-thoracic lesion gaps in the adult rat spinal cord. Behav. Brain Res. 237, 185–189. Eng, L.F., 1985. Glial fibrillary acidic protein (GFAP): the major protein of glial intermediate filaments in differentiated astrocytes. J. Neuroimmunol. 8, 203–214. Faulkner, J.R., Herrmann, J.E., Woo, M.J., Tansey, K.E., Doan, N.B., Sofroniew, M.V., 2004. Reactive astrocytes protect tissue and preserve function after spinal cord injury. J. Neurosci. 24, 2143–2155. Fawcett, J.W., Asher, R.A., 1999. The glial scar and central nervous system repair. Brain Res. Bull. 49, 377–391. Fawcett, J.W., Schwab, M.E., Montani, L., Brazda, N., Muller, H.W., 2012. Defeating inhibition of regeneration by scar and myelin components. Handb. Clin. Neurol 109, 503–522. Fitch, M.T., Silver, J., 2008. CNS injury, glial scars, and inflammation: Inhibitory extracellular matrices and regeneration failure. Exp. Neurol. 209, 294–301. Fouad, K., Tse, A., 2008. Adaptive changes in the injured spinal cord and their role in promoting functional recovery. Neurol. Res. 30, 17–27. Fraher, J.P., 1997. Axon-glial relationships in early CNS-PNS transitional zone development: an ultrastructural study. J. Neurocytol. 26, 41–52. Fu¨hrmann, T., Gerardo-Nava, G., Brook, G.A., 2011. Tissue engineering: central nervous system. Tissue Eng. From Lab to Clinic, 221–244. Gerardo-Nava, J., Hodde, D., Katona, I., Bozkurt, A., Grehl, T., Steinbusch, H.W., Weis, J., Brook, G.A., 2014. Spinal cord organotypic slice cultures for the study of regenerating motor axon interactions with 3D scaffolds. Biomaterials 35, 4288–4296. Guth, L., Zhang, Z., DiProspero, N.A., Joubin, K., Fitch, M.T., 1994. Spinal cord injury in the rat: treatment with bacterial

49

lipopolysaccharide and indomethacin enhances cellular repair and locomotor function. Exp. Neurol. 126, 76–87. Guth, L., Zhang, Z., Steward, O., 1999. The unique histopathological responses of the injured spinal cord. Implications for neuroprotective therapy. Ann. N.Y. Acad. Sci. 890, 366–384. Guven, M., Elalmis, D.D., Binokay, S., Tan, U., 2003. Populationlevel right-paw preference in rats assessed by a new computerized food-reaching test. Int. J. Neurosci. 113, 1675–1689. Hagg, T., Oudega, M., 2006. Degenerative and spontaneous regenerative processes after spinal cord injury. J. Neurotrauma. 23, 264–280. Hall, E.D., Springer, J.E., 2004. Neuroprotection and acute spinal cord injury: a reappraisal. NeuroRx 1, 80–100. Hejcl, A., Sedy, J., Kapcalova, M., Toro, D.A., Amemori, T., Lesny, P., Likavcanova-Masinova, K., Krumbholcova, E., Pradny, M., Michalek, J., Burian, M., Hajek, M., Jendelova, P., Sykova, E., 2010. HPMA-RGD hydrogels seeded with mesenchymal stem cells improve functional outcome in chronic spinal cord injury. Stem. Cells Dev. 19, 1535–1546. Joosten, E.A., Bar, P.R., Gispen, W.H., 1995. Collagen implants and cortico-spinal axonal growth after mid-thoracic spinal cord lesion in the adult rat. J. Neurosci. Res. 41, 481–490. Kawaja, M.D., Gage, F.H., 1991. Reactive astrocytes are substrates for the growth of adult CNS axons in the presence of elevated levels of nerve growth factor. Neuron 7, 1019–1030. Lu, P., Tuszynski, M.H., 2008. Growth factors and combinatorial therapies for CNS regeneration. Exp. Neurol. 209, 313–320. Marchand, R., Woerly, S., Bertrand, L., Valdes, N., 1993. Evaluation of two cross-linked collagen gels implanted in the transected spinal cord. Brain Res. Bull. 30, 415–422. Mollers, S., Heschel, I., Damink, L.H., Schugner, F., Deumens, R., Muller, B., Bozkurt, A., Nava, J.G., Noth, J., Brook, G.A., 2009. Cytocompatibility of a novel, longitudinally microstructured collagen scaffold intended for nerve tissue repair. Tissue Eng. A 15, 461–472. Montoya, C.P., Campbell-Hope, L.J., Pemberton, K.D., Dunnett, S. B., 1991. The “staircase test”: a measure of independent forelimb reaching and grasping abilities in rats. J. Neurosci. Meth. 36, 219–228. Morgenstern, D.A., Asher, R.A., Fawcett, J.W., 2002. Chondroitin sulphate proteoglycans in the CNS injury response. Prog. Brain Res. 137, 313–332. Novikova, L.N., Pettersson, J., Brohlin, M., Wiberg, M., Novikov, L. N., 2008. Biodegradable poly-beta-hydroxybutyrate scaffold seeded with Schwann cells to promote spinal cord repair. Biomaterials 29, 1198–1206. Ozawa, H., Matsumoto, T., Ohashi, T., Sato, M., Kokubun, S., 2001. Comparison of spinal cord gray matter and white matter softness: measurement by pipette aspiration method. J. Neurosurg. 95, 221–224. Pachence, J.M., 1996. Collagen-based devices for soft tissue repair. J. Biomed. Mater. Res. 33, 35–40. Pasterkamp, R.J., Giger, R.J., Ruitenberg, M.J., Holtmaat, A.J., De Wit, J., De Winter, F., Verhaagen, J., 1999. Expression of the gene encoding the chemorepellent semaphorin III is induced in the fibroblast component of neural scar tissue formed following injuries of adult but not neonatal CNS. Mol. Cell Neurosci. 13, 143–166. Pence, S., 2002. Paw preference in rats. J. Basic Clin. Physiol. Pharmacol. 13, 41–49. Pernet, V., Schwab, M.E., 2012. The role of Nogo-A in axonal plasticity, regrowth and repair. Cell Tissue Res. 349, 97–104. Pertici, V., Amendola, J., Laurin, J., Gigmes, D., Madaschi, L., Carelli, S., Marqueste, T., Gorio, A., Decherchi, P., 2013. The use of poly(N-[2-hydroxypropyl]-methacrylamide) hydrogel to repair a T10 spinal cord hemisection in rat: a behavioural,

50

brain research 1585 (2014) 37–50

electrophysiological and anatomical examination. ASN Neuro. 5, 149–166. Pettigrew, D.B., Crutcher, K.A., 1999. White matter of the CNS supports or inhibits neurite outgrowth in vitro depending on geometry. J. Neurosci. 19, 8358–8366. Pettigrew, D.B., Shockley, K.P., Crutcher, K.A., 2001. Disruption of spinal cord white matter and sciatic nerve geometry inhibits axonal growth in vitro in the absence of glial scarring. BMC Neurosci. 2, 8. Plantinga, L.C., Verhaagen, J., Gispen, W.H., 1993. The expression of B-50/GAP-43 in Schwann cells. Ann. N.Y. Acad. Sci. 679, 412–417. Profyris, C., Cheema, S.S., Zang, D., Azari, M.F., Boyle, K., Petratos, S., 2004. Degenerative and regenerative mechanisms governing spinal cord injury. Neurobiol. Dis. 15, 415–436. Raineteau, O., Fouad, K., Bareyre, F.M., Schwab, M.E., 2002. Reorganization of descending motor tracts in the rat spinal cord. Eur. J. Neurosci. 16, 1761–1771. Saglam, A., Perets, A., Canver, A.C., Li, H.L., Kollins, K., Cohen, G., Fischer, I., Lazarovici, P., Lelkes, P.I., 2013. Angioneural crosstalk in scaffolds with oriented microchannels for regenerative spinal cord injury repair. J. Mol. Neurosci. 49, 334–346. Scherer, S.S., Xu, Y.T., Roling, D., Wrabetz, L., Feltri, M.L., Kamholz, J., 1994. Expression of growth-associated protein43 kD in Schwann cells is regulated by axon-Schwann cell interactions and cAMP. J. Neurosci. Res. 38, 575–589. Schmidt, C.E., Leach, J.B., 2003. Neural tissue engineering: strategies for repair and regeneration. Annu. Rev. Biomed. Eng. 5, 293–347. Schoof, H., Apel, J., Heschel, I., Rau, G., 2001. Control of pore structure and size in freeze-dried collagen sponges. J. Biomed. Mater. Res. 58, 352–357. Schrimsher, G.W., Reier, P.J., 1993. Forelimb motor performance following dorsal column, dorsolateral funiculi, or ventrolateral funiculi lesions of the cervical spinal cord in the rat. Exp. Neurol. 120, 264–276. Schwab, J.M., Brechtel, K., Mueller, C.A., Failli, V., Kaps, H.P., Tuli, S.K., Schluesener, H.J., 2006. Experimental strategies to promote spinal cord regeneration–an integrative perspective. Prog. Neurobiol. 78, 91–116. Schwab, M.E., Bartholdi, D., 1996. Degeneration and regeneration of axons in the lesioned spinal cord. Physiol. Rev. 76, 319–370. Shearer, M.C., Fawcett, J.W., 2001. The astrocyte/meningeal cell interface–a barrier to successful nerve regeneration?. Cell Tissue Res. 305, 267–273. Silver, J., Miller, J.H., 2004. Regeneration beyond the glial scar. Nat. Rev. Neurosci. 5, 146–156. Spilker, M.H., Yannas, I.V., Kostyk, S.K., Norregaard, T.V., Hsu, H.P., Spector, M., 2001. The effects of tubulation on healing and

scar formation after transection of the adult rat spinal cord. Restor. Neurol. Neurosci. 18, 23–38. Spilker, M.H., Yannas, H.-P., Hsu, T.V., Norregaard, S.K., Kostyk, Spector, M., 1997. The effects of collagen-based implants on early healing of the adult rat spinal cord. tissue engineering 3, 309–317. Steward, O., Zheng, B., Tessier-Lavigne, M., 2003. False resurrections: distinguishing regenerated from spared axons in the injured central nervous system. J. Comp. Neurol. 459, 1–8. Stichel, C.C., Muller, H.W., 1998a. Experimental strategies to promote axonal regeneration after traumatic central nervous system injury. Prog. Neurobiol. 56, 119–148. Stichel, C.C., Muller, H.W., 1998b. The CNS lesion scar: new vistas on an old regeneration barrier. Cell Tissue Res. 294, 1–9. Stokols, S., Sakamoto, J., Breckon, C., Holt, T., Weiss, J., Tuszynski, M.H., 2006. Templated agarose scaffolds support linear axonal regeneration. Tissue Eng. 12, 2777–2787. Stokols, S., Tuszynski, M.H., 2006. Freeze-dried agarose scaffolds with uniaxial channels stimulate and guide linear axonal growth following spinal cord injury. Biomaterials 27, 443–451. Tator, C.H., 2002. Strategies for recovery and regeneration after brain and spinal cord injury. Inj. Prev. 8 (Suppl 4), IV33–IV36. Thuret, S., Moon, L.D., Gage, F.H., 2006. Therapeutic interventions after spinal cord injury. Nat. Rev. Neurosci. 7, 628–643. Toft, A., Scott, D.T., Barnett, S.C., Riddell, J.S., 2007. Electrophysiological evidence that olfactory cell transplants improve function after spinal cord injury. Brain 130, 970–984. Tom, V.J., Steinmetz, M.P., Miller, J.H., Doller, C.M., Silver, J., 2004. Studies on the development and behavior of the dystrophic growth cone, the hallmark of regeneration failure, in an in vitro model of the glial scar and after spinal cord injury. J. Neurosci. 24, 6531–6539. Tunturi, A.R., 1978. Elasticity of the spinal cord, pia, and denticulate ligament in the dog. J. Neurosurg. 48, 975–979. Yamamoto, M., Raisman, G., Li, D., Li, Y., 2009. Transplanted olfactory mucosal cells restore paw reaching function without regeneration of severed corticospinal tract fibres across the lesion. Brain Res. 1303, 26–31. Yannas, I.V., Burke, J.F., Orgill, D.P., Skrabut, E.M., 1982. Wound tissue can utilize a polymeric template to synthesize a functional extension of skin. Science 215, 174–176. Yoshii, S., Oka, M., Shima, M., Akagi, M., Taniguchi, A., 2003a. Bridging a spinal cord defect using collagen filament. Spine (Phila Pa 1976), 2346–235128, 2346–2351. Yoshii, S., Oka, M., Shima, M., Taniguchi, A., Akagi, M., 2003b. Bridging a 30-mm nerve defect using collagen filaments. J. Biomed. Mater. Res. A 67, 467–474.