Elsevier

Biomaterials

Volume 28, Issue 36, December 2007, Pages 5498-5508
Biomaterials

Long-term neurite orientation on astrocyte monolayers aligned by microtopography

https://doi.org/10.1016/j.biomaterials.2007.08.034Get rights and content

Abstract

After spinal cord injury neuronal connections are not easily re-established. Success has been hampered by the lack of orientation of neurites inside scar tissue and a lack of neurites crossing out of the site of injury. Oriented scaffolds in biodegradable polymers could be an excellent way to support both the orientation of neurites within the injury site as well as aiding their crossing out of the lesion. To establish the validity of using grooved micro-topography in polycaprolactone in combination with glia we have studied the long-term (3 weeks) orientation of neuronal cells on monolayers of astrocytes on the top of grooved topographies of various dimensions. We find that neurites are significantly aligned by groove/ridge type topographies which are “buried” under a monolayer of astrocytes for up to 3 weeks. This alignment is significantly lower than that of neurites growing directly on the topography, but these neurons do not survive on the poly-l-lysine coated polymer for more than a week. The alignment of neurites on the astrocyte layer to the underlying topography decreases over time, and with groove width. Topographies with 12.5 or 25 μm lateral dimension appear optimal for the long-term alignment and can support myelination. We have shown for the first time that micro-topography can act through an overlaid astrocyte layer and results in aligned neurites in long-term culture and that these can be myelinated by endogenous oligodendrocytes.

Introduction

It is well known that the adult spinal cord does not spontaneously regenerate following injury. However, some degree of axonal re-growth into a lesion has been shown after the transplantation of glial cells such as astrocytes, Schwann cells, olfactory ensheathing cells and other permissive cellular grafts [1], [2], [3], [4], [5], [6], [7], [8]. These cells appear to suppress the inhibitory microenvironment in and around the lesion and thus making it more attractive for axons to enter the lesion site. However, there is little evidence that axons are able to exit the graft and re-connect with neurons on the opposite side to any significant extent [9]. The regeneration process appears to be a chaotic, unorganised process and often functional neural transmission and behaviour recovers only partially after transplantation [2], [3], [4], [6].

Recently three-dimensional artificial scaffolds have been used as a strategy to increase axonal guidance in spinal cord lesions. Here, the intrinsic ability of the axons to follow topographic and chemical cues is exploited to guide them into and out of the implant. Topographic guidance of neurite outgrowth has been explored in vitro with culture substrates that contained well-defined micropatterned features. In these short-term investigations (2 days) it has been shown that chicken dorsal root ganglion (DRG) axons align parallel to the axis of a single scratch on glass coverslips (0.1–2 μm in depth and diameter) [10]. Also Xenopus neurites grow parallel to the grooves on quartz (14 nm deep and 1 μm wide), whereas rat hippocampal neurites were seen to extend perpendicular to the direction of the same groove size [11]. Alignment of adult rat hippocampal neurons on polystyrene microgrooves with width of 16 μm and depth of 4 μm has also been demonstrated at least after 2 days in vitro [12]. Moreover, neurons are not very likely to be the first cells to encounter an implant as any topography may be covered and obscured by glia rendering this intervention useless. To date, there has been no information in the literature evaluating the effectiveness of different groove sizes for neurite alignment in conjunction with glia, together with any long-term changes in orientation occurring thereafter.

Alternative or in addition to scaffolds to support regeneration several cell-based therapies have been tried in vivo. Poly(d,l)-lactide foams seeded with either olfactory ensheathing cells and olfactory nerve fibroblasts, or with Schwann cells were transplanted into a spinal cord injury site [13], [14]. However, due to inadequate adhesion and survival of supporting cells axonal regeneration did not occur to any great extent, indicating that these types of scaffold/cell combinations were not sufficient enough to stimulate axonal re-growth. Thus, integrating the biological influence of glia cells together with the physical guiding cues of microgrooves might be a better option than using either one individually in order to create a permissive environment for re-growth and the retention of the “proper” orientation of neurites.

As there is a distinct lack of evidence for a truly permissive construct we explored the use of single groove/ridge topography to aid, and maintain neurite alignment over more than 3 weeks. As healing of nerve injuries requires more than a few months we chose a slow degrading polymer ε-polycaprolactone (PCL) as material for our constructs. As astrocytes are potential candidates to promote axonal outgrowth in vitro and in vivo [4], [15] we populated groove/ridge patterned PCL substrates with a confluent layer of astrocytes and followed neurite alignment over 3 weeks. In addition, to confirm that grooves and ridges did not compromise the myelinating capability of these axons, we examined myelination by the endogenous oligodendrocytes.

Section snippets

Fabrication of microgrooved PCL constructs

Sheets of PCL (Mw 65 000, CAS 24980-41-4, Aldrich, Poole, UK) was prepared by first washing PCL pellets in methanol on a shaker, followed by drying, and then creating a sheet by melting the pellets under pressure between two glass plates at 70 °C for 2 h. The PCL sheet was stored at room temperature (RT) in the dark.

Microgrooved constructs was prepared by hot embossing as described by Gadegaard et al. [16]. Briefly a sandwich of a topographic master (27×75 mm2) followed by PCL sheet (10×30 mm2), and

Neuronal characteristics

The viability of neurons and their ability to extend neurites on PCL substrates coated with PLL or with a layer of astrocytes was evaluated after 1–3 weeks in vitro as shown in Fig. 1(A) and (B). On PLL-coated PCL substrates, the cultures, after 1 week were characterised by small clusters of neurons from which neurites extended. After 2 weeks (results not shown) the neurites appeared thinner and decreased in numbers. The number of neuronal cell bodies also decreased. Following 3 weeks in

Discussion

The main aim of the present study was to assess if micro-engineered PCL substrates could promote the survival, orientated extension and myelination of CNS neurites generated form the spinal cord. On PLL-coated PCL constructs without a layer of astrocytes, neurites survived for up to a week although at a lower density than on constructs covered with astrocytes. After 3 weeks in culture no neurites were observed if the constructs lacked astrocytes. This is in agreement with Bender et al. [23],

Acknowledgements

M.R. and M.O.R. would like to thank the EPSRC for partial financial support (GR/S/134415/01). S.C.B. and A.S. would like to thank the Multiple Sclerosis Society of Great Britain and Northern Ireland for their support. T.A. was an University of Glasgow M.Sc. by Research student and K.K. was a B.Sc. (Med Sci) student. We would also like to thank the scorers for their contribution.

References (42)

Cited by (58)

  • Spinal cord injury

    2020, Principles of Tissue Engineering
  • 3.17 Materials to control and measure cell function

    2017, Comprehensive Biomaterials II
View all citing articles on Scopus
View full text