University of Veterinary Medicine Hannover

Hannover Medical School Institute of Neuroanatomy

Peripheral nerve regeneration using hollow and enriched chitosan-based guidance conduits

Thesis

submitted in partial fulfillment of the requirements for the degree of

-Doctor rerum naturalium-

(Dr. rer. nat.)

by

Cora Meyer born in Bremen

Hannover, Germany 2014

Supervisor: Prof. Dr. C. Grothe Supervision group: Prof. Dr. C. Grothe

Prof. Dr. R. Gerardy-Schahn Prof. Dr. K. Schwabe

1st Evaluation: Prof. Dr. C. Grothe Hannover Medical School

Institute of Neuroanatomy Carl-Neuberg-Straße 1 30625 Hannover Prof. Dr. R. Gerardy-Schahn

Hannover Medical School Institute for Cellular Chemistry Carl-Neuberg-Straße 1 30625 Hannover

Prof. Dr. K. Schwabe Hannover Medical School

Department of Neurosurgery Carl-Neuberg-Straße 1 30625 Hannover

2nd Evaluation: Esther Udina i Bonet, MH, PhD Universitat Autònoma Barcelona Department of Cellular Biology, Physiology and Immunology Campus UAB 08193 Bellaterra (Cerdayola del Vallès)

Day of final exam: 10.10.2014

Parts of the thesis have been published previously:

Kirsten Haastert-Talini, Stefano Geuna, Lars Dahlin, Cora Meyer, Lena Stenberg, Thomas Freier, Claudia Heimann, Christina Barwig, Luis F.V. Pinto, Stafania Raimondo, Giovanna Gambarotta, Silvina Ribeiro Samy, Nuno Sousa, Antonio J. Salgado, Andreas Ratzka, Sandra Wrobel, Claudia Grothe (2013).

“Chitosan tubes of varying degrees of acetylation for bridging peripheral nerve defects”. Biomaterials, 2013, 34, 9886-9904.

Parts of this thesis have been submitted previously:

Cora Meyer*, Sandra Wrobel*, Stefania Raimondo, Claudia Heimann, Abraham Shahar, Stefano Geuna, Claudia Grothe, Kirsten Haastert-Talini.

“Engineered Schwann cells delivering fibroblast growth factor-2 increase peripheral nerve regeneration through hydrogel enriched chitosan tubes”. [submitted July 2014] * Both authors contributed equally to the work and share first authorship.

Poster presentations:

Cora Meyer, Kirsten Haastert-Talini, Sandra Worbel, Andreas Ratzka, and Claudia Grothe. “Inflammatory and physiological tissue responses following nerve reconstruction with chitosan tubes of different degrees of acetylation”. 2nd International Symposium on Peripheral Nerve Regeneration, Turin, Italy, January 23-25, 2014. Cora Meyer, Kirsten Haastert-Talini, Sandra Wrobel, and Claudia Grothe. “Addressing the long gap: Chitosan nerve guides supplemented with genetically modified Schwann cells”. 5th Lübeck Regenerative Medicine Symposium, Lübeck, Germany, June 26-27, 2014.

Sponsorship:

This work has received funding from the European Community’s Seventh Framework Programme (FP7-HEALTH-2011) under grant agreement n° 278612 (BIOHYBRID).

Dedicated to my parents

Table of contents

I

Table of Contents Abbreviations ............................................................................................................... IV List of figures .............................................................................................................. VII List of tables ................................................................................................................ IX Summary ..................................................................................................................... XI Zusammenfassung ..................................................................................................... XV 1 Introduction ............................................................................................................ - 1 - 

1.1 Clinical relevance of peripheral nerve injuries (PNIs) .................................... - 1 - 1.2 Organization of the peripheral nervous system (PNS) ................................... - 2 - 1.3 Classification of peripheral nerve injuries (PNIs) ........................................... - 4 - 1.4 Degeneration and regeneration processes following injury ........................... - 5 - 

1.4.1 Neuronal response ................................................................................... - 6 - 1.4.2 Changes in the distal environment and axonal regeneration .................. - 7 - 

1.5 Surgical repair strategies and their limitations ............................................... - 9 - 1.5.1 End-to-end repair ..................................................................................... - 9 - 1.5.2 Autologous nerve graft (ANG) ............................................................... - 10 - 1.5.3 Allograft .................................................................................................. - 10 - 1.5.4 CE and FDA Approved Conduits ........................................................... - 11 - 

1.5.4.1. Non-resorbable conduits ................................................................ - 13 - 1.5.4.2 Type I collagen ................................................................................ - 13 - 1.5.4.3 Porcine small intestinal submucosa (SIS) ....................................... - 13 - 1.5.4.4 Polyglycolic acid (PGA) ................................................................... - 14 - 1.5.4.5 Poly D,L lactide-co-ɛ-caprolactone (PCL) ....................................... - 14 - 

1.6 Considerations to design the future ideal conduit ........................................ - 14 - 1.6.1 Biocompatible and biodegradable polymers .......................................... - 15 - 

1.6.1.1 Chitosan .......................................................................................... - 15 - 1.6.2 Luminal fillers ......................................................................................... - 17 - 

1.6.2.1 NVR-Gel-A hyaluronic and laminin-based 3D guidance structure .. - 18 - 1.6.3 Incorporation of support cells ................................................................. - 19 - 

1.6.3.1 Schwann Cells (SCs) ...................................................................... - 19 - 1.6.4 Incorporation of neurotrophic factors (NTFs) ......................................... - 20 - 

1.6.4.1 Glial cell-derived neurotrophic factor (GDNF) ................................. - 20 - 1.6.4.2 Fibroblast growth factor (FGF)-2 ..................................................... - 21 - 

1.7 Tissue responses following implantation of medical devices ....................... - 22 - 1.8 The sciatic nerve model ................................................................................ - 24 - 1.9 Aim of the study ............................................................................................ - 26 - 

2. Material ............................................................................................................... - 29 - 3. Methods .............................................................................................................. - 33 - 

3.1 Animals ......................................................................................................... - 33 - 3.2 Experimental design and groups .................................................................. - 33 - 

3.2.1-a Project 1-Short-term study .................................................................. - 33 - 3.2.1-b Project 1-Long-term study .................................................................. - 34 - 3.2.2 Project 2 ................................................................................................. - 35 - 

3.4 Anesthesia and surgical procedures ............................................................ - 36 - 3.5. Functional evaluation of nerve regeneration. .............................................. - 38 - 

3.5.1 Static Sciatic Index (SSI) ....................................................................... - 38 - 3.5.2 Von Frey test .......................................................................................... - 39 - 3.5.3 Electrophysiology ................................................................................... - 40 - 

Table of contents

II

3.5.3.1 Serial non-invasive electrodiagnostic recordings ............................ - 40 - 3.5.3.2 Final invasive recordings ................................................................. - 42 - 

3.5.4 Muscle weight ........................................................................................ - 43 - 3.5.5 Autotomy score ...................................................................................... - 43 - 

3.6 (Immuno-) histological evaluation of nerve regeneration ............................. - 44 - 3.6.1 Assessment of nerve tissue ................................................................... - 44 - 

3.6.1.1 Project 1-short-term study ............................................................... - 44 - 3.6.1.2 Project 1-long-term study ................................................................ - 46 - 3.6.1.3 Project 2 .......................................................................................... - 47 - 

3.6.2 Assessment of fibrous capsule .............................................................. - 48 - 3.6.2.1 Project 1 .......................................................................................... - 48 - 3.6.2.2 Project 2 .......................................................................................... - 50 - 

3.7 Molecular and biochemical evaluation of nerve regeneration ...................... - 51 - 3.7.1 Quantitative real-time polymerase chain reaction (qRT-PCR) .............. - 51 - 3.7.2 Western Blot........................................................................................... - 52 - 

3.8 Analysis of tube properties after explantation .............................................. - 52 - 3.9 Statistical analysis ........................................................................................ - 53 - 

4 Results-Project 1 ................................................................................................. - 55 - 4.1 Gene expression of regeneration-associated proteins ................................ - 55 - 4.2 Matrix formation and presence of brain-derived neurotrophic factor (BDNF)- 56 - 4.2 Static Sciatic Index (SSI) .............................................................................. - 57 - 4.3 Electrophysiological assessment of muscle reinnervation ........................... - 59 - 4.4 Calculation of the lower limb muscle weight-ratio ........................................ - 67 - 4.6 Conduit properties upon explantation and tissue regeneration .................... - 67 - 4.5 Inflammatory tissue response ...................................................................... - 71 - 

4.5.1 Connective tissue ................................................................................... - 71 - 4.5.2 Nerve tissue ........................................................................................... - 77 - 

5 Results-Project 2 ................................................................................................. - 83 - 5.2 Mechanical pain threshold assessment ....................................................... - 83 - 5.1 Static Sciatic Index (SSI) .............................................................................. - 86 - 5.3 Electrophysiological assessment of muscle reinnervation ........................... - 87 - 5.4 Calculation of the lower limb muscle weight-ratio ........................................ - 95 - 5.3 Autotomy score ............................................................................................. - 95 - 5.5 Macroscopic analysis of the explanted nerve tissue .................................... - 96 - 5.6 Inflammatory tissue response ...................................................................... - 97 - 

5.6.1 Connective tissue ................................................................................... - 97 - 5.6.2 Regenerated tissue .............................................................................. - 101 - 

6 Discussion ......................................................................................................... - 107 - 6.1 Project 1 ...................................................................................................... - 108 - 

6.1.1 Regulation of regeneration associated proteins following nerve lesion and repair ............................................................................................................. - 108 - 6.1.2 Matrix formation and brain-derived neurotrophic factor (BDNF) presence in the tube content ............................................................................................ - 109 - 6.1.3 Reinnervation of distal muscle target................................................... - 111 - 

6.1.3.1 Behavior ........................................................................................ - 111 - 6.1.3.2 Electrophysiological recordings ..................................................... - 113 - 6.1.3.3 Morphological evaluation .............................................................. - 115 - 6.1.3.4 Muscle weight ................................................................................ - 116 - 

6.1.4 Foreign body response (FBR) upon chitosan implantation degradation- 117 - 

Table of contents

III

6.1.5 Concluding remarks-Chances of hollow chitosan conduits on the existing market ........................................................................................................... - 121 - 

6.2 Project 2 ...................................................................................................... - 124 - 6.2.1 Schwann cells (SCs) as vehicles for gene delivery ............................. - 124 - 6.2.2 Recovery of mechanical pain sensibility .............................................. - 125 - 6.2.3 Fibroblast growth factor (FGF)-218kDa partly allows reinnervation of distal muscle targets and functional recovery ........................................................ - 128 - 

6.2.3.1 Behavior ........................................................................................ - 128 - 6.2.3.2 Electrophysiological recordings ..................................................... - 128 - 

6.2.4 The difficulty in finding the right luminal filler for axonal guidance ...... - 131 - 6.2.5 No enhanced inflammatory response elicited by (modified) Schwann cells (SCs) in NVR Gel .......................................................................................... - 133 - 6.2.6 Concluding remarks and considerations regarding future conduits for long nerve defects ................................................................................................. - 133 - 

7 References ........................................................................................................ - 136 - 8. Affidavit ............................................................................................................. - 150 - 9. Acknowledgements .......................................................................................... - 152 - 

Abbreviations

IV

Abbreviations ADSC Adipose-derived stem cell

ANOVA Analysis of variance

ANG Autologous nerve graft

ANH Adjacent neuropathic hyperalgesia

AUC Area under the curve

AxL Axon loss

BDNF Brain-derived neurotrophic factor

BMSC Bone marrow-derived mesenchymal stem cell

CE Conformité Européenne

CMAP Compound muscle action potential

CNS Central nervous system

CNTF Cilliary neurotrophic factor

DA Degree of acetylation

DAPI 4‘, 6-diamidino-2-phenylindole

DRG Dorsal root ganglion

ECM Extracellular matrix

EG Europäische Gemeinschaft

EV Empty vector

FBGC Foreign body giant cell

FBR Foreign body response

FDA Food and Drug Administration

FGF Fibroblast growth factor

GAP-43 Growth-associated protein-43

Gapdh Glyceraldehyde-3-phosphate dehydrogenase

GDNF Glial cell-derived growth factor

GRAS Generally Recognized As Safe

HA Hyaluronic acid

HE Hematoxylin eosin

IHC Immunohistochemistry

IL Interleukin

i.m. Intramusculary

Abbreviations

V

i.p. Intraperitonealy

ITS Intermediate toe spread

ITSF Intermediate toe spread factor

L Lesion

MAG Myelin-associated glycoprotein

MGC Multinucleated giant cell

MHH Medizinische Hochschule Hannover

mRNA Messenger ribonucleic acid

N Non-lesion

NCAM Neural cell adhesion molecule

NCV Nerve conduction velocity

neo Neonatal

NF Neurofilament

NGF Nerve growth factor

NGS Normal goat serum

nt Non-transfected

NT Neurotrophin

NTF Neurotrophic factor

NVR Neural and vascular reconstruction

OEC Olfactory ensheating cell

P0 Myelin protein zero

PBS Phosphate buffered saline

PCL Poly(D,L lactide-co-ɛ-caprolactone)

PCLF Poly(caprolactone fumarat)

PFA Paraformaldehyde

PGA Polyglycolic acid

PL Plantar

PLGA Poly(lactic-co-glycolic acid)

PLLA Poly(L-lactic acid) PNI Peripheral nerve injury

PNS Peripheral nervous system

Ppia Peptidylpropyl isomerase

PSA Polysialic acid

PVA Polyvinyl alcohol

Abbreviations

VI

qRT-PCR Quantitative real-time polymerase chain reaction

R Receptor

RIPA Radioimmunoprecipitation

RT Room temperature

RT-PCR Real time polymerase chain reaction

s.c. Subcutaneous

SC Schwann cell

SD Sunderland degree

SEM Standard error of means

SFI Sciatic function index

SIS Small intestinal submucosa

SSI Static Sciatic Index

TA Tibialis anterior

UAB Autonomous University of Barcelona

ULUND Lund University

UMINHO University of Minho

UNITO University of Turin

tf Transfected

TrkB Tyrosine kinase receptor B

TS Toe spread

TSF Toe spread factor

US United States

List of figures

VII

List of figures Introduction Figure 1: Anatomical organization of the peripheral nervous system. ..................... - 3 - Figure 2: Degenerative and regenerative events following injuries to the mammalian PNS. ......................................................................................................................... - 7 - Figure 3: Regeneration processes across a 10 mm gap within a hollow nerve guidance conduit. ................................................................................................... - 12 - Figure 4: Chemical structure of chitin and chitosan. .............................................. - 16 - Figure 5: Formation of foreign body giant cells (FBGCs). ..................................... - 24 - Figure 6: The sciatic nerve and its innervation targets. ......................................... - 25 -  Methods Figure 7: Experimental design for the long-term study (13 weeks) using hollow chitosan conduits in a 10 mm gap (project 1). ....................................................... - 35 - Figure 8: Experimental design for the long-term study (17 weeks) using enriched tubes in a 15 mm gap (project 2). .......................................................................... - 36 - Figure 9: Chitosan conduit implantation. ................................................................ - 38 - Figure 10: Static Sciatic Index (SSI) ...................................................................... - 39 - Figure 11: Von Frey ................................................................................................ - 40 - Figure 12: Serial non-invasive electrodiagnostic recordings. ................................ - 42 - Figure 13: Autotomy Score. ................................................................................... - 44 - Figure 14: Processing of the nerve tissue gathered 5 and 18 days after surgery (project 1). .............................................................................................................. - 45 - Figure 15: Processing of the regenerated nerve tissue 13 weeks after implantation (project 1). .............................................................................................................. - 46 - Figure 16: Illustration of fibrous capsule analysis. ................................................. - 50 - Figure 17: Regulation of mRNA levels in spinal cord. ........................................... - 55 - Figure 18: Endogenous brain-derived neurotrophic factor (BDNF) detection in tube content. ................................................................................................................... - 56 -  Results-Project 1 Figure 19: Static Sciatic Index (SSI). ..................................................................... - 58 - Figure 20: Serial recordings of CMAPs. ................................................................. - 60 - Figure 21: Recovery of the CMAP amplitude. ........................................................ - 62 - Figure 22: Axon Loss. ............................................................................................ - 64 - Figure 23: NCV-ratio. ............................................................................................. - 65 - Figure 24: Analysis of the maximal CMAP. ............................................................ - 66 - Figure 25: Muscle weight determination. ............................................................... - 67 - Figure 26: Macroscopic properties of DAIII tubes. ................................................. - 68 - Figure 27: Regenerated nerve tissue upon explantation. ...................................... - 69 - Figure 28: Fibrous capsule optical density. ............................................................ - 71 - Figure 29: Fibrous capsule analysis. ..................................................................... - 72 - Figure 30: Presence of activated macrophages in the capsule surrounding the conduits. ................................................................................................................. - 73 - Figure 31: Collagen distribution within capsule.. ................................................... - 75 - 

List of figures

VIII

Figure 32: Vascularization of the capsule surrounding the conduit. ...................... - 76 - Figure 33: Number of MGCs and presence of chitosan inclusions 18 days after tube implantation within the regenerated nerve tissue.. ................................................ - 78 - Figure 34: Presence of MGCs in the regenerated nerve tissue 13 weeks after surgery. ................................................................................................................... - 79 - Figure 35: Scattered chitosan inclusions in the nerve tissue of the DAIII group. .. - 80 - Figure 36: Activated macrophages in the proximal nerve stump after 18 days. .... - 81 -  Results-Project 2 Figure 37: Mechanical pain threshold (lateral) ....................................................... - 84 - Figure 38: Mechanical pain threshold.(medial) ...................................................... - 85 - Figure 39: Static Sciatic Index (SSI). ..................................................................... - 87 - Figure 40: Serial recordings of CMAPs. ................................................................. - 89 - Figure 41: Amplitude-ratio from CMAPs recorded from the tibialis anterior (TA) muscle. ................................................................................................................... - 91 - Figure 42: Amplitude-ratio from CMAPs recorded from the plantar (PL) muscle. . - 92 - Figure 43: Muscle weight-ratio. .............................................................................. - 95 - Figure 44: Autotomy Score. ................................................................................... - 96 - Figure 45: Representative photographs of explanted specimens following a 4 month observation period. ................................................................................................. - 97 - Figure 46: Fibrous capsule analysis. ..................................................................... - 98 - Figure 47: Presence of activated macrophages in the capsule surrounding the conduits. ................................................................................................................. - 99 - Figure 48: Collagen distribution within capsule.................................................... - 100 - Figure 49: Vascularization of the capsule surrounding the conduit. .................... - 101 - Figure 50: Regeneration across the gap. ............................................................. - 103 - Figure 51: Failed regeneration. ............................................................................ - 104 - Figure 52: Cross-section of an enriched chitosan conduit and regenerated nerve tissue 17 weeks following implantation. ............................................................... - 105 - 

List of tables

IX

List of tables Methods Table 1: Overview of the collected tissue samples in the short-term study of project 1. ................................................................................................................. - 34 -  Results-Project 1 Table 2: Listing of animals per group, which demonstrated evocable CMAPs in the course of the study. ................................................................................................ - 59 - Table 3: Overview of animal numbers showing regenerated nerve tissue 13 weeks after chitosan conduit implantation. ....................................................................... - 70 -  Results-Project 2 Table 4: Listing of animals per group, which demonstrated evocable compound muscle action potentials in the course of the study. .............................................. - 88 - Table 5: NCV, AxL, and amplitude-ratio ................................................................ - 94 - Table 6: Tissue regeneration seen upon explantation. .......................................... - 96 - Table 7: Quantity of MGCs and activated macrophages within the (regenerated) nerve tissue. ......................................................................................................... - 102 - 

X

Summary

XI

Summary Peripheral nerve regeneration using hollow and enriched chitosan-based guidance conduits Cora Meyer Peripheral nerve injuries (PNIs) represent a global problem, which mainly affects

young men in their work productive years. Following crush injuries, patients will likely

experience a full functional recovery. If the nerve is completely transected, on the

other hand, surgical intervention is needed and the outcome is highly variable.

Cases, which do not allow tensionless end-to-end repair, require a bridging strategy

to overcome the gap. As a clinical standard therapy, surgeons in such situations

usually harvest a negligible sensory nerve and use it to reconnect the two nerve

stumps. This method, though, has disadvantages including the need of a second

surgery, limited amount of available tissue as well as mismatch issues between the

donor nerve and the affected nerve. In addition, only about 50 % of the patients

treated with autologous nerve grafts (ANGs) are reported to have a successful

recovery. Due to these reasons, alternative strategies are needed to bridge nerve

defects. Various synthetic as well as biologically-derived materials have already

been tested and some have been approved by the "Food and Drug Administration"

and "Conformité Européenne" to be marketed as nerve guidance conduits for small

gaps, but none of these conduits is completely accepted in the clinics.

Aim of the present study was to test hollow chitosan-based nerve guidance conduits

of 3 different degrees of acetylation (DAI-III) for their possible application in

treatment of small gaps (e.g. in digital nerves) (project 1). Therefore, short (5 days

and 18 days) and long-term (13 weeks) studies were conducted. In addition, bridging

of long gaps (e.g. in brachial plexus injuries) was addressed by introducing luminal

fillers into the hollow chitosan conduits, which should, amongst others, support

axonal elongation (project 2). Genetically modified neonatal Schwann cells (SCs)

over-expressing either glial cell-derived neurotrophic factor (GDNF) or the low

molecular weight isoform of fibroblast growth factor-2 (FGF-218kDa) were suspended

in a hyaluronic acid and laminin containing gel (NVR-Gel), which was then injected

Summary

XII

into the chitosan scaffold. Empty vector-transfected SCs as well as non-transfected

SCs served as controls besides the reversed ANG in this study of 17 weeks

duration. For both objectives, the nerve guidance conduits were tested in the rat

sciatic nerve model (10 mm or 15 mm gap), which is frequently used to evaluate

peripheral nerve reconstruction. This animal model offers the opportunity to assess

motor and sensory recovery by means of electrophysiological recordings, Static

Sciatic Index, and the von Frey test. Furthermore, the tissue response in presence of

the chitosan implants was analyzed to exclude the possibility of a strong

inflammatory response.

Project 1 led to very encouraging results. Hollow chitosan conduits of all DAs

allowed peripheral nerve regeneration in a similar manner as ANGs across a 10 mm

nerve defect. Motor recovery was observed in almost all animals and the analysis of

the inflammatory response revealed a clear reduction in the number of activated

macrophages and multinucleated giant cells (MGCs) in the fibrous capsule, which

had developed around the conduits over time. In addition, less macrophages were

found after 13 weeks than 18 days in the regenerated nerve tissue, while the number

of MGCs remained comparably equal and low. Noteworthy, no changes in the

regulation of selected regeneration-associated proteins was detected as a response

towards the conduits in dorsal root ganglia or the spinal cord 5 and 18 days after

conduit implantation. Morphometric data, however, revealed increased sprouting of

axons in the presence of DAI conduits, while DAIII conduits demonstrated a fast

degradation rate. Upon explantation, the latter were partly broken and displayed

fissures and cracks. As a consequence of these observations, DAII conduits were

chosen for further experiments, since these conduits were most promising to support

peripheral nerve regeneration.

For project 2, DAII conduits were enriched and tested in a critical 15 mm defect. At

the end of the study, the group that had received ANGs was superior compared to

the experimental groups that had received chitosan conduits containing enriched gel.

Solely, the FGF-218kDa group showed nerve tissue regeneration in 4 out of 7 animals,

which also displayed motor recovery; while it was only one animal from the GDNF

and non-transfected SCs group, respectively. These results indicate that the

introduction of gel as a luminal filler might not be the best choice, since it possibly

Summary

XIII

acts as a physical impediment to peripheral nerve regeneration and only FGF-218kDa

seems to partially compensate for this negative effect.

Reflecting the findings of both studies it can be concluded that DAII chitosan-based

conduits represent an alternative to the marketed conduits for treatment of small

gaps. Besides the promising results obtained in this study, it is especially appealing

for clinical use because of its transparency. The latter property facilitates the

handling procedure for the surgeon and reduces the time needed for correct

positioning of the implant. Regarding treatment of long gaps, however, the search for

an optimal luminal filler / guidance structure continuous and further experiments are

indispensable. Alternatives, which guide the regrowing axons without filling the whole

luminal space, such as highly porous sponges, membranes, and fibers, are probably

better suited for this task than gel.

XIV

Zusammenfassung

XV

Zusammenfassung

Periphere Nervenregeneration mit leeren und angereicherten Chitosan-basierten Leitschienen Cora Meyer Periphere Nervenverletzungen stellen ein globales Problem dar, von welchem

insbesondere junge Männer in ihren arbeitsfähigen Jahren betroffen sind. Im Zuge

von Quetschungen ist eine Wiederherstellung der Funktionalität wahrscheinlich,

während die Heilungschancen nach kompletten Nervendurchtrennungen variieren

und ein chirurgischer Eingriff erforderlich ist. Falls ein spannungsfreies Verbinden

der beiden Nervenstümpfe nicht möglich ist, ist der Einsatz von Nervenleitschienen

unerlässlich. In der Klinik wird in diesen Fällen in der Regel ein autologes

Nerventransplantat verwendet, welches von einem funktionell entbehrlichen

sensiblen Nerven gewonnen wird. Diese Methode hat jedoch Nachteile, da eine

zweite Operation nötig ist, das potenziell zu entnehmende Nervengewebe begrenzt

ist und es zu Inkompatibilitäten zwischen dem Spender- und Empfängernerv

kommen kann. Des Weiteren wird lediglich in der Hälfte der Fälle, welche mit einem

Autotransplantat behandelt werden, eine erfolgreiche Regeneration beobachtet.

Folglich werden alternative Überbrückungsstrategien für Nervenlücken benötigt.

Hierfür wurden bereits verschiedene Synthetik- als auch Biomaterial-basierte

Nervenleitschienen getestet und zum Teil auch von der „Food and Drug

Administration“ (FDA) sowie „Conformité Européenne“, zu Deutsch „Europäische

Gemeinschaft“, zur Behandlung am Menschen zugelassen. Allerdings hat bisher

keine dieser Nervenleitschienen eine breite Akzeptanz in der klinischen Anwendung

gefunden.

Folglich war es ein Ziel dieser Arbeit leere Chitosan-basierte Nervenleitschienen

verschiedener Acetylierungsgrade (DAI-III) für die Überbrückung von kleinen

Nervendefekten, wie sie z.B. bei Handnerven häufig vorkommen, zu testen. Zu

diesem Zwecke wurden Kurzzeit- und Langzeitstudien durchgeführt über Zeiträume

von 5 und 18 Tagen, sowie 13 Wochen (Projekt 1). Des Weiteren wurden Chitosan

Röhrchen mit einem Hyaluronsäure- und Laminin-basiertem Gel (NVR-Gel) befüllt, in

Zusammenfassung

XVI

welchem genetisch veränderte neonatale Schwann Zellen suspendiert waren

(Projekt 2). Letztere wurden dahingehend modifiziert, dass sie entweder eine

erhöhte Expression von GDNF (glial cell-derived neurotrophic factor) oder der 18kDa

FGF (fibroblast growth factor)-2-Isoform aufwiesen. Neben dem reversen autologen

Nerveninterponats wurden physiologische sowie mit einem Leervektor-transfizierte

Schwann Zellen (in Gel) als Kontrollen transplantiert. Mit der Ergänzung von

intraluminalen Leitstrukturen und Schwann Zellen sollte die Behandlung von

größeren Nervendefekten, wie sie z.B. für den Plexus brachialis beschrieben sind,

angesprochen werden, da diese die Regeneration potenziell unterstützen können.

Für beide Projekte wurde das Modell des Nervus ischiadicus bei der Ratte (10 mm

oder 15 mm Lücke) gewählt, welches als Standard-Modell zur Untersuchung

peripherer Nervenregeneration gilt. Es bietet die Möglichkeit, die motorische als

auch sensorische Regeneration u.a. durch elektrophysiologische Messungen, die

Feststellung des SSIs (Static Sciatic Index) als auch durch den von Frey Test zu

beurteilen. Im Rahmen dieser Arbeit wurde ferner die Gewebereaktion auf das

Implantieren der Chitosan-basierten Nervenleitschienen untersucht, um eine starke

Abwehrreaktion auszuschließen.

Die Ergebnisse, die im Projekt 1 erzeugt wurden waren vielversprechend und eine

erfolgreiche, mit dem Autotransplantat vergleichbare, Nervenregeneration konnte für

Chitosan Röhrchen aller Acetylierungsgrade, bei einer Überbrückungsdistanz von

10 mm, festgestellt werden. Bei nahezu allen Tieren konnte die Reinnervation

distaler Muskeln nachgewiesen werden und die Immunantwort zeigte eine deutliche

Reduktion im Zeitverlauf. Letzteres äußerte sich in einer Abnahme aktivierter

Makrophagen und multinukleärer Riesenzellen in der fibrösen Kapsel, welche sich

um die Chitosan Röhrchen gebildet hatte. Zusätzlich konnte eine Verringerung der

Makrophagen im Regenerat beobachtet werde, während die Anzahl der

multinukleären Riesenzellen hier im Zeitraum von 18 Tagen bis 13 Wochen

unverändert gering blieb. Es ist ferner von Bedeutung, dass die Regulation von

beispielhaft ausgewählten Regenerations-assoziierten Proteinen in Spinalganglien

und dem Rückenmark durch die Implantation der Nervenleitschienen nach 5 und 18

Tagen nicht beeinflusst wurde. Eine morphometrische Analyse der distalen

Nervenabschnitte zeigte jedoch ein erhöhtes Maß an axonalen Verzweigungen für

die DAI Gruppe, während die DAIII Röhrchen bereits nach 18 Tagen und verstärkt

Zusammenfassung

XVII

nach 13 Wochen Risse und Bruchstellen aufwiesen und zum Teil gänzlich

auseinandergebrochen waren. Folglich wurden DAII Nervenleitschienen für weitere

Experimente ausgewählt, da sie die periphere Nervenregeneration am besten

unterstützen.

Für Projekt 2 wurden die angereicherten Nervenleitschienen für die Überbrückung

einer Distanz von 15 mm getestet. Bis zum Ende der Studie wurden mit dem

autologen Nerventransplantat konsequent die besten Ergebnisse erzielt. Lediglich

die Gruppe, welche FGF-218kDa überexprimierende Schwann Zellen transplantiert

bekommen hatte, zeigte Regeneration in 4 von 7 Tieren. In diesen konnte auch eine

partielle Wiederherstellung der Funktionalität gezeigt werden, während dies nur für

ein Tier der GDNF sowie physiologischen Schwann Zell Gruppe zutraf. Diese

Ergebnisse implizieren, dass die Nutzung von Gel zur Befüllung von

Nervenleitschienen nicht die ideale Wahl ist. Möglicherweise fungiert es als

physische Barriere und verhindert somit einen erfolgreichen Regenerationsprozess.

FGF-218kDa scheint jedoch einen kompensatorischen Effekt zu haben und das

axonale Wachstum, auch durch das Gel hindurch, zu ermöglichen.

Aus den Resultaten beider Projekte lässt sich folgern, dass DAII Chitosan-basierte

Röhrchen eine Alternative für die bereits erhältlichen Nervenleitschienen darstellen

um kleine Nervendefekte zu überbrücken. Neben den vielversprechenden

Ergebnissen im Rahmen von Projekt 1 besticht dieses Röhrchen durch seine

Transparenz. Diese Eigenschaft macht es für den klinischen Einsatz besonders

attraktiv, da so die Handhabung für den Chirurgen erleichtert wird und eine schnelle

und korrekte Positionierung des Implantates möglich ist. Bezüglich der Behandlung

von großen Nervendefekten geht die Suche nach einer idealen, intraluminalen

Leitstruktur / Füllung weiter und zusätzliche Experimente sind unerlässlich.

Alternativen, welche die Axone leiten ohne den Raum im Röhrchen komplett zu

füllen, wie z.B. sehr poröse Schwämme, Membranen oder Fasern, sind demnach

vermutlich besser für diese Aufgabenstellung geeignet als Gele.

XVIII

Introduction

- 1 -

1 Introduction 1.1 Clinical relevance of peripheral nerve injuries (PNIs) In the last decades, the understanding of biological mechanisms underlying

peripheral nerve injuries (PNIs) and the following regeneration process has

significantly increased. Nevertheless, surgical reconstruction is still challenging and

the clinical results have not substantially improved in the same time frame

(Lundborg, 2000). Despite modern microsurgery techniques that are characterized

by a high level of accuracy, adult patients are likely to suffer from functional

impairment due to incomplete or incorrect target reinnervation (Cobianchi et al.,

2013; Frostick et al., 1998; Navarro et al., 2007; Svennigsen and Dahlin, 2013).

Resulting symptoms include poor sensibility, deficient motor function, cold

intolerance, and neuropathic pain. These are consequences that often add up to a

reduced quality of life for the patient (Lundborg, 2000; Svennigsen and Dahlin,

2013). Children, on the other hand, who e.g. suffer from birth-related brachial plexus

injuries, show a better clinical recovery. Although the mechanisms causing these

differences are not yet understood, it has been proposed that young patients hold a

better cerebral plasticity (cortical reorganization) (Chemnitz et al., 2013; Frykman,

1976; Laurent and Lee, 1994; Svennigsen and Dahlin, 2013; Tajima and Imai, 1989).

According to that, previous existing, but inert connections get activated and new

connections are established via sprouting (Chen et al., 2002). Furthermore, a loss of

myelinated and unmyelinated nerve fibers and a decline in the expression as well as

transport of cytoskeletal proteins has been reported in elderly subjects, thus

explaining the better regeneration seen in children (Verdu et al., 2000).

Worldwide more than one million people are affected by PNIs. In the United States

(US) alone more than 200.000 surgeries on PNIs are performed per year resulting in

five million disability days (Daly et al., 2012; Freier et al., 2005a; Kehoe et al., 2012).

Associated health care costs total approximately $150 billion in the US and therewith

represent a major cost factor (Apel et al., 2008). Additionally, in Europe more than

300.000 PNIs are reported annually (Ciardelli and Chiono, 2006). Causes for PNIs

are manifold and surgical reconstruction may be very challenging. Working in a

military environment for one can result in battlefield wounds, of which approximately

14–22 % include injuries to the upper extremities (e.g. gunshot injuries of the

Introduction

- 2 -

brachial plexus) (Gaudet et al., 2011; Samadian et al., 2009; Waldram, 2003). Due to

the long history of war inflicted injuries, crucial progress in repairing PNIs has been

made e.g. during the American Civil War, World War I, and World War II (Friedman

et al., 2009; Robinson, 2000). Also in daily life situations (during work and leisure

time), however, injuries to the peripheral nervous system (PNS) are common and the

most prominent injuries include those to digital nerves and the brachial plexus

(Ciaramitaro et al., 2010; Fox and Mackinnon, 2011; Svennigsen and Dahlin, 2013;

Thorsén et al., 2012). Mostly young men at productive age suffer from PNIs following

vehicle accidents, but also domestic violence or sport activities represent major

causes for peripheral nerve trauma (Ciaramitaro et al., 2010; Ijkema-Paassen et al.,

2004; Thorsén et al., 2012; Waldram, 2003). Additionally, nerve damage is the result

of conditions like the carpal tunnel syndrome or diabetes (Lekholm et al., 2001;

Moore et al., 2009).

1.2 Organization of the peripheral nervous system (PNS) With multicellular life developing 1 billion years ago, it was about 500 million years

later that the first simple nervous tissue appeared in the phylum Cnidaria

(Grimmelikhuijzen and Westfall, 1995; Selden and Nudds, 2004; Westfall, 1996).

Since then, the complexity has increased and today the nervous system of

vertebrates is organized into the central nervous system (CNS) and PNS.

Approximately 86 billion neurons form the CNS, which is comprised of the brain and

its extension, the spinal cord (Azevedo et al., 2009; Waldram, 2003). The PNS is

composed of cranial nerves, spinal nerves including their roots, peripheral nerves,

and the autonomic nervous system. The soma of motor neurons of the PNS are

located in the anterior horn of the spinal cord and send their axons out to the

periphery to terminate on motor end plates in skeletal muscles. Sensory fibers, with

their cell bodies situated in the dorsal root ganglia (DRGs), convey information from

their pain, thermal, tactile, and stretch receptors, which are found e.g. in skin and

muscles to the CNS (Waldram, 2003).

The functional unit of a peripheral nerve is the axon together with its Schwann cell

(SC). This unit is surrounded by different layers of connective tissue (Fig. 1). The

endoneurium encircles the axons and provides a packaging for nerve fibers, while

Introduction

- 3 -

the perineurium defines fascicles that consist of a bundle of nerve fibers and protects

the nerve from mechanical pressure. Nerve fascicles are then again embedded in a

loose connective tissue, the epineurium (Brushart, 2011a; Waldram, 2003). SCs

spiral around axons and myelinate them in a 1:1 relationship. Approximately, 95 % of

the complete length of an axon is myelinated and only interrupted by nodes of

Ranvier (spaces between adjacent SCs). These sites enable saltatory conduction

due to membrane specializations like a concentrated appearance of voltage-gated

sodium channels (Boiko et al., 2003; Brushart, 2011a).

Figure 1: Anatomical organization of the peripheral nervous system. Somata of sensory neurons are located in DRGs, while those of motor neurons are found in the anterior horn of the spinal cord. Neurons of both modalities send their processes to peripheral targets as part of peripheral nerves. Each peripheral nerve is comprised of various fascicles, which are enclosed by the epineurium, a loose connective tissue layer. Every single fascicle is then again surrounded by another layer of connective tissue, the perineurium, and contains many axons. Each axon is further encircled by SCs, which produce a myelin sheath, as well as by a third connective tissue layer, the endoneurium (the figure was made based on illustrations of (Scholz et al., 2011; Sierpinski Hill, 2009).

The PNS displays a diverse array of axons that vary in their size and end organ and

are characterized by varying nerve conduction velocities (NCV). The latter was used

to subdivide axons into three groups. Group A represents efferent and afferent

myelinated axons of the somatic nervous system that forward information most

Introduction

- 4 -

rapidly. Within this group, axons are further defined and subdivided into alpha

(70-120 m/s), beta (30-70 m/s), gamma (15-30 m/s), and delta (12-30 m/s)

subpopulations according to their NCV (Brushart, 2011a). The largest and fastest of

these neurons, the alpha motoneurons, terminate on extrafusal muscle fibers, while

smaller and slower conducting gamma motoneurons specifically innervate intrafusal

muscle fibers (Manuel and Zytnicki, 2011). Furthermore, neurons of the alpha and

beta subpopulation forward signals from cutaneous and subcutaneous

mechanoreceptors, while gamma neurons take part in sensation of temperature and

pain. A second group is represented by B-fibers, which represent myelinated

preganglionic autonomic fibers. The third group are C-fibers, which are slow

conducting (<2 m/s), unmyelinated axons. Like some A-gamma neurons, they

mediate nociceptive pain and account for 75 % of the total axon number in

cutaneous nerves and their respective DRGs. In general, one single myelinating SC

wraps one large caliber axon (> 2µm) or one non-myelinating SC ensheathes

multiple (up to 100) small unmyelinated axons (C-fibers), thus forming so called

Remak bundles (Berger et al., 2014; Brushart, 2011a; Hall, 2005; Heavner and de

Jong, 1974; Jungnickel et al., 2010; Murinson et al., 2005; Sun et al., 2012).

1.3 Classification of peripheral nerve injuries (PNIs) Mechanisms that cause PNIs are diverse and include crush trauma, laceration,

stretching, and compression. Accordingly, the degree of the resulting PNIs is

variable as is the prognosis for the functional recovery. In 1943, based on the

severity of the lesion, Seddon described three types of injuries: neurapraxia,

axonotmesis and neurotmesis (Seddon, 1942). Sunderland later established a more

detailed classification system (Sunderland, 1951). Neurapraxia, or Sunderland

degree (SD) I, is the mildest type of injury, describing cases, in which the continuity

of the nerve trunk is preserved, but the nerve conduction is blocked at the site of the

injury. In these cases, Wallerian Degeneration does not occur and the segment

distally to the injury continuous to conduct. Full recovery is usually seen within 3 to 4

months with almost the same number of regenerated axons reaching the target

organs (Dubový, 2011; Flores et al., 2000; Isaacs, 2010; Marshall Devor and Ruth

Govrin-Lippmann, 1979; Sunderland, 1990; Waldram, 2003). Axonotmesis (SD II),

on the other hand, is characterized by Wallerian degeneration and loss of

Introduction

- 5 -

spontaneous muscle activity for about three weeks. Axons are disrupted, but the

surrounding connective tissue (epineurium, perineurium, and endoneurium) stays

intact. Therefore, sprouting axons can send their axons through their original

endoneurial tubes and spontaneous recovery is possible. The recovery time

depends on the distance between the injury and distal target organs, with axons

regrowing at a rate of 1-1.5 mm per day (depending on nerve and progress of

reinnervation). If not only the axon is damaged, but also layers of the surrounding

connective tissue, the injury is classified as neurotmesis (Flores et al., 2000; Isaacs,

2010; Seddon et al., 1943; Waldram, 2003). Sunderland further divided injuries of

this level according to the degree of connective tissue disruption (Sunderland, 1990,

1951). SD III indicates that aside from the axons also the endoneurium is affected

and has collapsed. Due to the missing guidance of this layer regenerating axons

have difficulties to find their way and tend to grow together with scar tissue.

Nevertheless, regeneration is possible, but unspecific reinnervation may result in

abnormal motor and sensory recovery. In injuries of SD IV, the epineurium is the

only structure that remains intact and spontaneous recovery does not occur. A

complete transection of the nerve is the highest degree according to Sunderland (SD

V) (Allodi et al., 2012; Flores et al., 2000; Sunderland, 1990, 1951). Surgical

intervention is needed if a patient suffers from an injury that is classified as SD IV or

SD V accordingly, but functional recovery is often incomplete. Furthermore, patients

frequently have to deal with pain and dysthesia due to the formation of neuroma that

are composed of immature axons and connective tissue (Allodi et al., 2012; Isaacs,

2010; Mourad et al., 2001; Navarro et al., 2007; Svennigsen and Dahlin, 2013).

1.4 Degeneration and regeneration processes following injury Unlike the CNS, the PNS has an intrinsic capacity to regenerate and reinnervate

distal target organs following injuries. Neurons perform a switch in function, thus

changing from a transmission state to a growth state (Allodi et al., 2012). Many

axons are able to assemble a new growth cone within hours after nerve crush, while

about 50 % undergo apoptosis (Bradke et al., 2012; Dubový, 2011; Törnqvist and

Aldskogius, 1994). At the same time, the creation of a nerve regeneration friendly

environment (see below) distally to the lesions starts with a process called Wallerian

Introduction

- 6 -

degeneration, if the lesion is at least classified as SD two (Flores et al., 2000;

Liebermann, 1971).

1.4.1 Neuronal response Following nerve injury, morphological changes occur in the cell bodies of the affected

neurons. The localization of the nucleus is altered to a more eccentric position and

its shape appears deformed (Fig. 2). Swelling of the cell body and dissolution of the

Nissl bodies (chromatolysis) are further characteristics that define neurons after

transection (Gaudet et al., 2011; Kreutzberg, 1995; Liebermann, 1971; Navarro et

al., 2007; Pabari et al., 2011).

The axonal regeneration process does not directly start at the level of the lesion, but

rather a few internodal segments more proximal, while the immediately affected

segments undergo degeneration (Raimondo et al., 2011). The neuronal reaction

after transection of the axons is characterized by an increased production of

components relevant for the assembly of growth cones as well as structural

components. Tubulin, actin, and growth-associated protein 43 (GAP-43) are e.g.

reported to be increasingly synthesized and transported, while proteins required for

neurotransmitter generation are down-regulated (Brushart, 2011b; Navarro et al.,

2007; Raimondo et al., 2011). The signals that trigger this shift from a “transmitting”

to a “regenerative” state of the neuron include various mechanisms. For example,

following axonal disruption, and preceding sealing of the axon, an influx of

extracellular calcium and sodium causes a burst of action potentials. Furthermore,

the retrograde transport is interrupted providing the cell body with negative and

positive signals. One negative signal e.g. is the decrease in retrograde transported

nerve growth factor (NGF), while contact with trophic molecules like e.g. cilliary

neurotrophic factor (CNTF) serves as a positive signal (Allodi et al., 2012; Hall, 2005;

Kirsch et al., 2003; Navarro et al., 2007).

Introduction

- 7 -

Figure 2: Degenerative and regenerative events following injuries to the mammalian PNS. (A-B) Following PNIs, degeneration occurs in the directly affected internodal segments of the proximal nerve stump and in the distal nerve stump. Following a brief time period, which allows e.g. influx of extracellular calcium that triggers, amongst others, a burst of action potentials and a subsequent switch to a regenerative state of the neuron, the tip of the proximal nerve stump gets sealed within hours. Myelinating SCs start to release their myelin and the axoplasm undergoes degradation. The rapid inflammatory response includes resident macrophages and invading neutrophils, (C) while recruited macrophages and T-lymphocytes arrive at the lesion site within 2 days and infiltrate the complete distal stump within 3 to 7 days after nerve injury. (E) Dedifferentiated SCs then start to form the bands of Büngner along which axonal sprouts are guided to peripheral target organs. (F) Successful reinnervation ultimately leads to retraction of unsuccessful axon sprouts, reversal of muscle atrophy, chromatolysis as well as the formation of mature SC-axon interactions. The latter are characterized by thinner myelin sheaths and shorter internodal spacing (Deumens et al., 2010; Gaudet et al., 2011; Hall, 2005; Nectow et al., 2011). The figure was made based on illustrations of (Deumens et al., 2010; Gaudet et al., 2011).

1.4.2 Changes in the distal environment and axonal regeneration Distally to the injury site, a degenerative process called Wallerian degeneration

starts initiated by the detachment of axons. Subsequently, axoplasm, axolemma, and

myelin sheaths immediately start to degrade. Calcium dependent proteolytic

enzymes like calpains are mediators causing the breakdown of the cytoskeleton, a

process that starts within 2 days in rodents and within seven days in humans

Introduction

- 8 -

(Dubový, 2011; Gaudet et al., 2011; Hall, 2005). Meanwhile, SCs stop to produce

myelin proteins like myelin basic protein or myelin protein zero (P0), enhance

neurotrophic factor (NTF) and GAP-43 production, and start to proliferate within their

basal lamina tubes. Within 48 h, the myelin sheaths have been segmented into short

pieces. During the first days, SCs play a major role in fragmentation and

phagocytosis of myelin debris. Resident macrophages, which represent about 4-9 %

of the endoneurial cellular population in a normal nerve, then directly begin to

remove axonal as well as myelin debris that contains neurite outgrowth inhibitors like

e.g. myelin-associated glycoprotein (MAG). Production of cytokines and chemokines

by SCs and resident macrophages leads, amongst others, to an accumulation of

blood-derived monocytes within 4 days after injury (Brushart, 2011c; Dubový, 2011;

Freier et al., 2005a; Gaudet et al., 2011; Hall, 2005; Hirata and Kawabuchi, 2002;

Ide, 1996; Stoll et al., 2002). Following differentiation into macrophages, they support

the myelin clearance as well as cytokine production, thereby facilitating SC migration

and axonal growth. Their accumulation reaches a peak 2 weeks after injury with an

150-fold increase in number (Brushart, 2011c). In case of contact with MAG on the

surface of newly formed myelin, recruited macrophages exit from the basal lamina to

re-enter the circulation (Gaudet et al., 2011). Usually, the identification of resident

and invading macrophages is possible, because resident macrophages can be

immunocytochemically recognized with ED-2 antibody and invading macrophages

display immunoreactivity for ED-1 antibody. Following nerve injury, however,

resident macrophages start to proliferate and show ED-1 immunoreactivity making

these two populations indistinguishable (Dubový, 2011). Aside from SCs and

macrophages, neutrophils phagocytize debris in the early phase of Wallerian

degeneration, but do not survive this act and undergo apoptosis. T-lymphocytes, on

the other hand, are mainly crucial during the later phase of axonal regeneration by

producing pro- and anti-inflammatory cytokines (Gaudet et al., 2011; Stoll et al.,

2002).

With the assembly of the growth cone progressing, SCs start to align and form the

bands of Büngner. They serve as a guidance structure for the regrowing axons

towards peripheral reinnervation targets (Gaudet et al., 2011; Pabari et al., 2011;

Svennigsen and Dahlin, 2013). Upon contact, SCs start to remyelinate the

regenerating axons. The morphology of these newly remyelinated axons is

Introduction

- 9 -

characterized by thinner myelin sheaths and shorter internodes compared to the

proximal nerve segment causing a slower nerve NCV (Hall, 2005; Ide, 1996;

Svennigsen and Dahlin, 2013). After weeks to months, the regenerating nerve is

characterized by numerous small nerve bundles called mini-fascicles. In the process

of approaching the periphery, the number of bundles decreases resulting in the

maturation of only the axons that reach distal targets (Freier et al., 2005a).

1.5 Surgical repair strategies and their limitations The severity of peripheral nerve injuries decides if and what kind of surgery is

necessary. Clean nerve transections should be surgically treated as soon as

possible to facilitate an easier approximation and alignment of the two nerve

segments. This, however, is only possible, if the reconnection can be performed

without causing any tension. Short gaps, on the other hand, are usually treated with

commercially available conduits (limited to defects not exceeding 3 cm in humans or

1 cm in rats) or autologous nerve grafts (ANGs), which are also preferentially used

to overcome longer gaps (Bell and Haycock, 2011; Daly et al., 2012; de Ruiter et al.,

2009; Isaacs, 2010; Nectow et al., 2011; Pfister et al., 2011).

1.5.1 End-to-end repair Minor nerve injuries that include no or only very small gaps (not greater than

0.5-2 cm) are treated by reconnecting the two nerve ends, as long as tension-free

repair is possible (Bell and Haycock, 2011; Daly et al., 2012). After removing the

scarred part of the affected nerve tissue, fascicular patterns and longitudinal surface

vessels are consulted as markers to guarantee an accurate alignment. The

reconnection can be achieved with epineurial sutures, but approximation of deep

fascicles is not ensured with this method. Some surgeons therefore align individual

fascicles by suturing their perineurium. This, though, implicates the risk of

misdirecting all axons, if the repair is not performed accurately. Grouped fascicular

alignment, on the other hand, is less problematic. None of these suture methods,

however, has been proven to produce superior results (Isaacs, 2010; Pabari et al.,

2010).

Introduction

- 10 -

1.5.2 Autologous nerve graft (ANG) Defects that cannot be reconnected by end-to-end repair without causing tension

are usually treated with ANGs. This standard therapy includes the harvest of nerve

tissue from the sural nerve or other dispensable sensory nerves like the medial and

lateral antebrachial cutaneous nerves or the superficial radial nerve. The advantage

of these nerves is that they are easily accessible and provide the opportunity to

obtain long nerve grafts (Isaacs, 2010; Pabari et al., 2010). Injuries of small nerves

are usually treated with a single nerve graft, while larger nerves require the use of

cable grafts. In the latter case, multiple grafts are used as interponates to overcome

the gap (Isaacs, 2010). Studies have, however, shown, that motor nerve grafts

support regeneration of motor neurons more effectively, a problem, which probably

occurs due to mismatch issues regarding axonal size, distribution and alignment.

Furthermore, SCs of different nerve modalities express varying phenotypes and

produce modality specific growth factors (Daly et al., 2012; Höke et al., 2006;

Isaacs, 2010; Nichols et al., 2004; Pabari et al., 2010). Additional disadvantages of

this method include donor site morbidity, the need of a second surgery, and the

limited amount of nerve tissue that is available. Furthermore, a prolonged

inflammatory reaction is caused as a result of secondary removal of axonal and

myelin debris. Nevertheless, ANGs have proven to be suitable for nerve gaps of up

to 5 cm length with a success rate of about 50 % (Bell and Haycock, 2011; Daly et

al., 2012; Deumens et al., 2010; Lin and Marra, 2012).

1.5.3 Allograft Since treatment of PNIs with ANGs involves disadvantages like limited amount of

available tissue and donor site morbidity, allografts represent a fitting alternative

strategy to overcome nerve gaps. These grafts can be gained from donor or

cadaveric nerve tissue and provide guidance structures as well as donor SCs. The

required systemic immunosuppression for 18 to 24 months, however, limits the

clinical application of this method, which has so far led to variable results (Bell and

Haycock, 2011; Kehoe et al., 2012; Mackinnon et al., 2001; Pabari et al., 2010; W.

Ray, 2010). Mackinnon and colleagues reported on 10 patients that showed motor

and sensory recovery after nerve repair with cadaveric allografts, while one patient

suffered from rejection of the graft following subtherapeutic immunosuppression

(Mackinnon et al., 2001).

Introduction

- 11 -

To avoid immunosuppression, allografts derived from human nerve tissue have been

decellularized by AxoGen Inc. (Alachua, FL). This process preserves the structure of

the extracellular matrix while clearing cells, cellular debris, and proteins like

chondroitin sulfate proteoglycans that are known to act as neurite outgrowth

inhibitors (axogenic.com, 2014; Kehoe et al., 2012). Decellularized allografts from

AxoGen have received approval from the “Food and Drug Administration” (FDA) and

return of sensation was reported in gaps of up to 3 cm length, while no rejection was

observed (Karabekmez et al., 2009; Kehoe et al., 2012). A multicenter clinical trial

study confirmed these findings with meaningful recovery seen in 87.3 % of the

patients suffering from injuries of up to 5 cm length in sensory, motor, and mixed

nerves (Brooks et al., 2012)

1.5.4 CE and FDA Approved Conduits As an alternative to the standard therapy, short gaps can be treated with nerve

conduits to allow tensionless nerve repair of e.g. digital nerves, since they have been

shown to provide a permissive environment for regrowing axons (Fig. 3). Additionally, this method is time saving, a factor that is important for clinical

application (de Ruiter, 2009; Meek and Coert, 2008; Pabari et al., 2010; Williams et

al., 1983). At first, mostly silicone conduits had been tested, but up to the present

various different conduits have now been experimentally investigated and some

have been cleared for marketing in the US and Europe (de Ruiter, 2009; Meek and

Coert, 2008). To assure that only safe medical devices enter the market, the FDA, a

regulatory system in the US, is responsible to keep the risk minimal for patients (IOM

(Institute of Medicine), 2012). Therefore, manufacturers have to demonstrate safety

and efficacy of high risk devices. The requirements that need be fulfilled, however,

vary depending on the classification of the device. The classification ranges from

low-risk devices in class I (e.g. stethoscopes) to medium-risk class II devices

(computed tomographic scanners and nerve guidance conduits like Neurolac) and

high-risk class III products such as deep-brain stimulators, but also all devices

containing chitin and its derivatives. With the same aim as the FDA, the “Conformité

Européenne” (CE) is responsible for the European market. While low-risk devices

only have to be declared, more complex devices are in need to be cleared by a

Introduction

- 12 -

notified body (specialists in evaluating different products) (Kehoe et al., 2012;

Kramer et al., 2010; Strusycyzk and Struszczyk, 2007).

Figure 3: Regeneration processes across a 10 mm gap within a hollow nerve guidance conduit. Regeneration within a hollow nerve guidance conduit can be divided into five phases: (A) the initial phase is characterized by an influx of fluid that contains NTFs, inflammatory cells, and extracellular matrix precursor molecules. (B) Subsequently, a fibrin cable forms within one week after injury, which connects the two nerve stumps. (C) SCs then start to migrate along the fibrin cable and proliferate to form the bands of Büngner during the second week of regeneration before (D) axonal sprouts can cross the defect in the axonal phase. During this phase, it is likely that the fibrin cable has already vanished. (E) Ultimately, mature SCs start to remyelinate larger regenerated axons (Belkas et al., 2004a; Daly et al., 2012; de Ruiter, 2009; Lundborg et al., 1981; Madison et al., 1987). The figure was made based on illustrations of (Belkas et al., 2004a; Daly et al., 2012).

Introduction

- 13 -

1.5.4.1. Non-resorbable conduits

Salubridge™ and SaluTunnel™ (both Salumedica LLC; Atlanta, GA) are marketed

worldwide and are made of polyvinyl alcohol (PVA) hydrogel. While Salubridge™

serves as a conduit to enable axonal growth across a nerve gap; SaluTunnel™ is

indicated for injuries without substantial loss of nerve tissue. These products are the

only synthetic non-resorbable conduits marketed, but so far no published data from

clinical studies are available (Kehoe et al., 2012).

1.5.4.2 Type I collagen

Advantages of collagen include its adhesive properties for different cell types, its

fibrillar structure, which is preserved during the manufacturing process, and the fact

that it can be easily isolated and purified. There are various conduits based on this

major extracellular matrix (ECM) component. NeuraGen® (Integra Life Sciences

Corporation; Plainsboro, NJ) is reported to be as effective as ANGs and sensory and

motor recovery were equivalent to results with direct suture in a multicenter study

when the nerve gap was 6 mm or less (Archibald et al., 1991; Boeckstyns et al.,

2013; Kehoe et al., 2012). NeuraWrap™ is another product from Integra that has an

identical material composition and is serving as a wrap (Kehoe et al., 2012). Just

recently, Georgiou and colleagues could show robust neuronal regeneration across

a 15 mm gap within 8 weeks when using this wrap in combination with rods of rolled-

up sheets and aligned SCs as fillers (Georgiou et al., 2013). Furthermore,

NeuroMatrix™ and Neuroflex™ (Collagen Matrix Inc., Franklin Lakes, NJ) are also

collagen-based conduits. The latter is highly flexible and has been created to bridge

nerve defects of up to 2.5 cm length. NeuroMend™ is another product of this

company with the same material compositions, but neither of the three devices has

been tested in published animal or clinical studies (Kehoe et al., 2012).

1.5.4.3 Porcine small intestinal submucosa (SIS)

AxoGuard™ nerve connector and AxoGuard™ nerve protector (AxoGen, Alachua,

FL) are based on small intestinal submucosa (SIS), which is a cell-free collagen

matrix that provides a good mechanical support and supports remodeling of the host

tissue by accelerating new vessel growth, cell proliferation, and differentiation

(Kehoe et al., 2012).

Introduction

- 14 -

1.5.4.4 Polyglycolic acid (PGA)

Neurotube® (Synovis Micro Companies Alliance; Birmingham; AL) is a synthetic

porous nerve conduit made of polyglycolic acid (PGA), which allows the infiltration of

oxygen. Results from animal as well as clinical studies vary, but its usage has shown

efficacy in defects up to 3 cm. A disadvantage, however, is the high rate of

degradation and presence of acidic degradation products (Kehoe et al., 2012; Meek

and Coert, 2008).

1.5.4.5 Poly D,L lactide-co-ɛ-caprolactone (PCL)

Neurolac® (Polyganics, Groningen, NL) is the only transparent nerve guidance

conduit, a property that facilitates the correct positioning of the nerve stumps in the

device, and offers the opportunity for the surgeon to remove eventual blood clots that

could hamper nerve regeneration (de Ruiter, 2009; Kehoe et al., 2012;

Polyganics.com, 2014). On the other hand, it is inflexible, highly rigid, thus causing

needle breakage, and swelling of the conduit. Additionally, neuroma formation as

well as foreign body reactions have been described (Kehoe et al., 2012; Meek and

Coert, 2008; Meek and Den Dunnen, 2009).

1.6 Considerations to design the future ideal conduit Approved nerve guidance conduits are so far limited to bridge nerve defects of up to

3 cm length and are associated with insufficient functional recovery. Thus, to

overcome these limitations, new strategies are needed (Bell and Haycock, 2011;

Daly et al., 2012; de Ruiter et al., 2009; Gu et al., 2011). There are various design

possibilities of conduits that can influence axonal regeneration. Future conduits

should not only have mechanical strength, while still being suturable, but should also

be made of biocompatible as well as biodegradable materials. The latter allows

degradation at a reasonable rate, if adjustments are undertaken properly, to avoid

inflammatory reactions. The device should further be flexible without having the

tendency of swelling to avoid compression of regenerated axons. Additionally, the

wall should be semi-permeable to allow exchange of oxygen and nutrients, but at the

same time ingrowth of fibrous tissue should be prevented and NTFs should be kept

inside the conduit. Introduction of 3D guidance structures such as fibers, sponges,

membranes or hydrogels inside the tube could further support axonal outgrowth and

Introduction

- 15 -

SC migration and are thus promising tools to promote regeneration. The same

expectations are held for the incorporation of support cells and NTFs. Also, the

surface structure was found to be important. Smooth surfaces enable the growth of a

free floating nerve cable, while rough surfaces cause a complete covering of the

conduit (Clements et al., 2009; Daly et al., 2012; de Ruiter, 2009; Huang and Huang,

2006; Jiang et al., 2010; Kehoe et al., 2012; Mukhatyar et al., 2014; Pabari et al.,

2011). Also, since the ultimate goal is to market the designed device for clinical

usage, it is important that the end product is easy to produce, to sterilize, and offers

simple handling for the surgeon. The latter includes transparency of the material,

since this characteristic facilitates correct positioning of the conduit (Chalfoun et al.,

2006; Daly et al., 2012; Polyganics.com, 2014)

1.6.1 Biocompatible and biodegradable polymers Biocompatibility and biodegradability are important properties of all materials

included in a nerve conduit. Assessment of biocompatibility is e.g. performed with

regard to blood compatibility (haemolysis, destruction of blood components,

coagulation, and thrombus formation are limiting factors) and histocompatibility

(material should be non-toxic and not elicit immunological rejection responses from

the surrounding tissue). Additionally, degradation in a controlled manner is important

to avoid secondary injuries to the newly grown nerve tissue (Daly et al., 2012).

Otherwise, a second surgery might be necessary to remove the material since its

permanent presence can cause chronic tissue responses or nerve compression.

Moreover, it would be good to match degradation kinetics to the axonal regeneration

rate (Daly et al., 2012; Jiang et al., 2010; Pabari et al., 2011).

Materials that fulfill these requirements include synthetic materials like aliphatic

polyesters (e.g. polycaprolactone) and naturally-derived biopolymers. The latter

includes ECM molecules such as collagen, laminin, and fibronectin, but also silk

fibroin and polysaccharides like alginate or chitin and chitosan (Daly et al., 2012).

1.6.1.1 Chitosan

Chitin is a natural polymer that can be gained from the exoskeleton of arthropods,

cuticles of insects, walls of fungi, and shells of crustaceans (e.g. the deep-water

shrimp Pandalus borealis). It is the second most abundant organic resource on earth

after cellulose (Freier et al., 2005c; Kardas et al., 2012; Singh and Ray, 2000). Intra-

Introduction

- 16 -

and intermolecular hydrogen bonds between the polymer chains lead to a high

insolubility of chitin (Kardas et al., 2012). Full or partly deacetylation of chitin

converts the material into chitosan (Fig. 4), which is insoluble in water and neutral or

basic pH (due to free amino groups), while an acidic pH results in protonation of the

amino functions, thus facilitating solubility in water (Freier et al., 2005a; Kardas et al.,

2012; Singh and Ray, 2000). Chitin, however, degrades faster than chitosan, is more

compatible with blood, and activates fever macrophages (Freier et al., 2005b).

Figure 4: Chemical structure of chitin and chitosan. Chitosan can be gained from chitin through alkaline hydrolysis of the N-acetyl groups. Chitosan degradation occurs by non-specific enzymes such as lysozymes, chitinases, cellulases or hemicellulases, proteases, lipases, ß-1,3-1,4 glucanases as well as chitosanases (Raafat and Sahl, 2009).

Chitosan is comprised of ß(1-4) linked D-glucosamine and N-acetyl-D-glucosamine

subunits with a degree of acetylation (DA) between 0 % and approximately 60 %

(Freier et al., 2005b). Degradation of chitosan occurs, amongst others, by lysozyme,

an enzyme that is present in the human body and has the ability to hydrolyze the

ß(1 -4) linkages between D-glucosamine and N-acetyl-D-glucosamine in chitin and

chitosan. The rate of degradation is thereby related to the distribution of the N-acetyl

groups and to the DA, with higher DAs resulting in a more rapid degradation (Han et

al., 2011; Kean and Thanou, 2010). Chitosan has been investigated for various

biomedical applications due to its biodegradability, but also biocompatibility, low

Introduction

- 17 -

toxicity, and molecular similarity to glycosaminoglycans that are part of the basal

membrane and ECM (Freier et al., 2005a; Siemionow et al., 2010). Furthermore,

chitosan is known to accelerate wound healing and shows antimicrobial activity.

While Chitosan is not cleared as “Generally Recognized As Safe (GRAS)” by the

FDA, it has been approved for some medical applications. Accordingly, several

wound dressings that are based on chitosan are commercially available and it is e.g.

used in food preservation and dentistry as well as being considered as a potential

carrier in drug delivery systems (Itoh et al., 2003; Raafat and Sahl, 2009; Singh and

Ray, 2000). Chitosan has further been studied for peripheral nerve reconstruction

and is thought to be a promising candidate for tissue engineering applications. In

vitro studies revealed that chitosan membranes allow SC adhesion, survival,

migration, and proliferation as well as neuronal survival and differentiation (Simões

et al., 2011; Wang et al., 2005; Yuan et al., 2004). Chitosan has already been tested

alone or in combination with other materials for its possible usage in peripheral nerve

repair. Successful bridging of small gaps (8-12 mm) has repeatedly been reported

with hollow chitosan or chitosan combined conduits (Ao et al., 2011; Han et al.,

2010; Patel et al., 2006; Raisi et al., 2012; Simões et al., 2010; Xie et al., 2008; Xu et

al., 2011; Zheng and Cui, 2012). Treatment of 15 mm gaps with chitosan containing

conduits has also been described and chitosan/PGA and chitosan/poly(lactic-co-

glycolic acid ) (PLGA)-based conduits led to encouraging results bridging 30, 50, or

60 mm defects in dog or rhesus monkey nerve injury models (Ding et al., 2010; Hu et

al., 2013, 2009; Itoh et al., 2003; Wang et al., 2005; Xue et al., 2012).

1.6.2 Luminal fillers Since regeneration following treatment with conduits so far is limited to small gaps,

modifications of hollow tubes are considered that might enhance the success of

target reinnervation in cases of short and long gaps by stabilizing the unformed or

incomplete fibrin cable. Various approaches to incorporate intraluminal guidance

structures have been investigated including multichannel conduits or the introduction

of filaments, sponges, or gels such as collagen- or laminin-containing gels into the

scaffold (Bell and Haycock, 2011; Daly et al., 2012; de Ruiter et al., 2009; Gu et al.,

2011).

Introduction

- 18 -

1.6.2.1 NVR-Gel-A hyaluronic and laminin-based 3D guidance structure

Filling the conduit with ECM components like collagen, fibronectin, laminin, or

hyaluronic acid (HA) is thought to facilitate early ingrowth of axons and support cell

migration (de Ruiter et al., 2009; Jiang et al., 2010; Khaing and Schmidt, 2012; Lin

and Marra, 2012). These components can serve as a basis for hydrogels, which are

injectable and gelling fillers that can also serve as carriers for cells and growth

factors. Typical hydrogels are cross-linked hydrophilic water-soluble polymer

networks, which are used in various medical devices including bone, tissue and

nerve repair (Lin and Marra, 2012).

NVR (Neural and Vascular Reconstruction)-Gel is comprised of HA combined with

laminin (Rochkind et al., 2009; Shahar et al., 2001). HA is a non-sulfated

glycosaminoglycan, a natural repeating disaccharide that is ubiquitous in the human

body (e.g. in the vitreous of the eye and connective tissue). It is e.g. involved in the

regulation of tissue hydration, water transport, inflammation (it has an

anti-inflammatory effect), and lubrication of joints. Accordingly, medical applications

of this hydrophilic molecule have been clinically used for 30 years and include the

supplementation of joint fluid in patients with arthritis and its use as wound dressing.

Moreover, it is biocompatible and non-immunogenic, reduces nerve sensitivity

associated with pain, and regulates cell proliferation, migration, and differentiation.

HA is highly negatively charged and therefore cell adhesive components must be

added such as ECM components like laminin (Burdick and Prestwich, 2011; de

Ruiter et al., 2009; Khaing and Schmidt, 2012; Necas et al., 2008). Laminin, a

trimeric glycoprotein, is part of the basal lamina and promotes neurite outgrowth and

elongation. Like HA, it also has a positive effect on SC proliferation and migration

and additionally provides support for SC adherence. Overall, laminin acts as a cue

that guides regrowing axons and thus has the ability to promote nerve regeneration

(Belkas et al., 2004a; Daly et al., 2012; de Ruiter, 2009; Jiang et al., 2010; Swindle-

Reilly et al., 2012). Importantly, the laminin included in the NVR-Gel is laminin

1,while laminin 5 has been associated with carcinogenesis (Shahar et al., 2001;

Waterman et al., 2007).

Introduction

- 19 -

1.6.3 Incorporation of support cells The addition of support cells to different luminal fillers inside the nerve conduit has

been widely studied and is a favorable tool to enhance peripheral nerve regeneration

across long gaps (de Ruiter, 2009; Pabari et al., 2011). The most promising

candidates to be used as cellular components in an enriched conduit include

adipose-derived stem cells (ADSCs), olfactory ensheating cells (OECs), bone

marrow-derived mesenchymal stem cells (BMSCs), and SCs. (Ao et al., 2011; Bell

and Haycock, 2011; Di Summa et al., 2011; di Summa et al., 2010; Jiang et al.,

2010; Radtke et al., 2011; Wei et al., 2011; Xue et al., 2012; ; Yuan, 2004 et al.,

2004; Zheng and Cui, 2012).

1.6.3.1 Schwann Cells (SCs)

SCs are the main glial cells of the vertebrate PNS and are known for their essential

role during peripheral nerve regeneration. Aside from expressing a myelinating or

unmyelinating phenotype, sensory and motor phenotypes have been described for

SCs. They express different molecular markers, thereby creating modality specific

pathways and facilitating correct target reinnervation. The most prominent example

is the L2/HNK1 carbohydrate epitope, which is only expressed by SCs encircling

motor axons, while neural cell adhesion molecule (NCAM)-positive cells are

associated with sensory axons. L2/HNK1, however, is down-regulated following

injury and reemerges upon contact with regenerating axons (Allodi et al., 2012; Höke

et al., 2006; Nichols et al., 2004).

Following PNIs, SCs switch their phenotype to a regenerative state and start to

proliferate (reaching up to 4-17 times higher numbers compared to intact nerve) and

migrate (Daly et al., 2012; Gu et al., 2011; Haastert et al., 2006b; Madduri and

Gander, 2010). Previously, it has been shown that incorporation of SCs improves

peripheral nerve regeneration across a 2 cm gap in the rat median nerve and across

a distance of 18 mm in the sciatic nerve (Ansselin et al., 1997; Brushart, 2011a;

Sinis et al., 2005). Autologous SCs were further reported to allow regeneration over

a 6 cm gap in the peroneal nerve of rabbits, if applied in combination with Matrigel

within a gluteal vein (Strauch et al., 2001). SCs provide a substratum consisting of

structural and adhesive ECM molecules for axonal outgrowth and secrete a cocktail

of NTFs, thus creating the ideal environment for regeneration (Huang and Huang,

Introduction

- 20 -

2006). Depending on the modality of the affected nerve and the phenotype of the

SCs, the composition of produced NTFs varies. While SCs of the sensory system

produce high levels of NGF, brain derived neurotrophic factor (BDNF), insulin like

growth factor, and erythropoietin, SCs of the motor system secrete glial cell-derived

growth factor (GDNF) (Gu et al., 2011; Madduri and Gander, 2010). Low and high

molecular isoforms of fibroblast growth factor (FGF)-2 are also up-regulated by SCs

at the lesion site and differentially promote peripheral nerve regeneration (Grothe et

al., 2001; Haastert et al., 2006a). SC seeded chitosan conduits have already been

tested and were successful in improving regeneration (Ao et al., 2011).

1.6.4 Incorporation of neurotrophic factors (NTFs) Since the Nobel prize-awarded discovery of NGF and its role as a target-derived

factor for the survival and maintenance of neurons during development by Rita

Levi-Montalcini and Viktor Hamburger, half a century ago, multiple families of NTFs

that are part of various biological activities have been identified. NGF, BDNF,

neurotrophin (NT)-3, NT-4/5, and NT-6 are part of the NT-family, CNTF belongs to a

neuropoietic family, and GDNF and FGF are eponyms for their respective family. In

the last decades, other sources of trophic support have been identified such as SCs

and the nerve regenerating promoting effects of these factors have been

investigated (Jiang et al., 2010; Levi-Montalcini and Angeletti, 1968; Levi-Montalcini,

1987; Oppenheim and Bartheld, 2008). For example, null mutants of NGF or its

receptor trkA suffered from loss of sympathetic and some small sensory neurons,

while NT-3 and trkC absence induced loss of large diameter neurons (Oppenheim

and von Bartheld, 2008; Snider, 1994). Furthermore, controlled release (e.g. via

genetically engineered SCs) of NTFs promotes outgrowth and survival of regrowing

axons and is therefore an option to improve nerve regeneration across long gaps

(Daly et al., 2012; Pfister et al., 2007; Timmer et al., 2003).

1.6.4.1 Glial cell-derived neurotrophic factor (GDNF)

GDNF, a member of the tumor necrosis factor-ß superfamily, has been widely

investigated and is mainly associated with being a survival factor for mesencephalic

dopaminergic cells and motor neurons as well as for some sensory and autonomic

neurons (Boyd and Gordon, 2003a; Brushart, 2011a; Höke et al., 2002; Jiang et al.,

2010; Oppenheim et al., 1995; Yan et al., 1995). It is retrogradely transported to

Introduction

- 21 -

motor and sensory neurons and is up-regulated in DRGs as well as ventral roots

following injury (Höke et al., 2006, 2002). Signaling occurs, amongst others, via

receptor complexes that consist of a GFR-α1 and ret subunit, both of which are

up-regulated in axotomized motoneurons, while GDNF messenger ribonucleic acid

(mRNA) has not been described in regular or lesioned motoneurons. The latter,

however, as well as the protein, have been found to be increased in SCs of the

sciatic nerve following injury peaking at 7 days (Boyd and Gordon, 2003a; Höke et

al., 2002). Synthetic polymer conduits sustainingly releasing GDNF have further

been shown to allow regeneration of motor and sensory neurons over major gaps of

15 mm length (Fine et al., 2002; Kokai et al., 2011). Moreover, blending of chitosan

conduits with laminin and GDNF as well as delivery of the latter via transfected SCs

in a silicone tube has led to an enhanced functional recovery over a 10 mm gap (Li

et al., 2006; Patel et al., 2007). The number of myelinated axons has further been

reported to be greater with GDNF released from a rod than with NGF or NT-3 and

axonal elongation promoting effects have been reported following local

administration (Barras et al., 2002; Fine et al., 2002; Madduri et al., 2009).

Additionally, impaired regeneration following chronic axotomy has been associated

with a decline of GDNF expression by chronically denervated SCs (Höke et al.,

2002). In correspondence, GDNF administration via microspheres or pumps has

been shown in some studies to only be effective following nerve repair after a

2 month period of chronic axotomy, but fails to support regeneration in cases of

immediate nerve repair (Boyd and Gordon, 2003a, 2003b; Wood et al., 2013b).

Furthermore, intraperitoneal (i.p.) injection of GDNF affected SCs by inducing an

increased proliferation of unmyelinated SCs as well as myelination of small

unmyelinated SCs (Höke et al., 2003).

1.6.4.2 Fibroblast growth factor (FGF)-2

The vertebrate FGF family consists of 22 members that have been grouped into 7

subfamilies. These factors take part in regulating cell proliferation, migration, and

differentiation during development and play a role in tissue repair following injury.

Signaling of these NTFs takes place via binding to high-affinity tyrosine

transmembrane receptors (R) 1-4 and heparin or heparin sulfate proteoglycans

(Dorey and Amaya, 2010; Grothe and Nikkhah, 2001; Ornitz and Itoh, 2001;

Woodbury and Ikezu, 2014).

Introduction

- 22 -

FGF-2 has been detected in DRGs during development and adulthood and has been

reported to be up-regulated following nerve injuries. Also, SCs as well macrophages

express FGF-2 and its receptors at the lesion site (Grothe and Nikkhah, 2001;

Grothe et al., 2001, 1996). Furthermore, FGF-2 as well as the FGFR1-3 are usually

expressed at low levels in adult sciatic nerve, but are subjected to an up-regulation

after nerve crush (Grothe et al., 2001). Administration of exogenous FGF-2 to the

lesion site has been found to prevent cell death of sensory neurons and promote

neurite outgrowth and vascularization of regrowing axons (Aebischer et al., 1989;

Danielsen et al., 1989; Grothe and Nikkhah, 2001). Different isoforms of FGF-2

result from the alternate start codons with the additional amino-terminal sequence of

the high-molecular-weight isoforms containing a nuclear-localization signal (Grothe

and Nikkhah, 2001; Ornitz and Itoh, 2001; Woodbury and Ikezu, 2014). Following

nerve transection, these isoforms are differently regulated with an overall increase of

all three isoforms in the proximal nerve stump (higher up-regulation of 18-and 21-

kDa FGF-2 isoform than 23-kDa isoform), while a decrease has been described in

the distal stump after 5 days using RT-PCR (real-time polymerase chain reaction),

and no FGF-2 protein has been found with Western Blot analysis after 7 days. In the

case of a nerve crush, however, an up-regulation of all three isoforms is seen in the

proximal and distal nerve segment (Brushart, 2011a; Meisinger and Grothe, 2002).

In accordance with the differing patterns of up-regulation of these isoforms, it was

reported that they have different effects on peripheral nerve regeneration. While

FGF-218kDa inhibits myelination, but allows functional motor recovery, FGF-221/23kDa

has been associated with promotion of early sensory target reinnervation over long

gaps of 15 mm (Haastert et al., 2006a). Furthermore, it was recently shown that

FGF-218kselectively promotes neurite outgrowth of spinal motor neurons in vitro

(Allodi et al., 2013).

1.7 Tissue responses following implantation of medical devices Implantation of biomaterial devices causes a series of tissue responses within 2 to 4

weeks after surgery including blood-material interactions, provisional matrix

formation, acute (less than a week) and chronic inflammation (usually of short

duration) as well as fibrous capsule development. The foreign body response (FBR)

is part of the chronic inflammation and includes monocytes, macrophages, and the

Introduction

- 23 -

formation of multinucleated giant cells (MGCs) at tissue/material interfaces. The

latter separate the FBR from granulation tissue, which is characterized by the

presence of macrophages, fibroblasts, and capillaries, and which represents a

precursor to fibrous capsule formation. Two types of MGCs have been described

and are known as Langhans-type cells and foreign body giant cells (FBGCs). Cells

of the Langhans-type are thought to be characteristic for sarcoidosis and

tuberculosis and feature up to 20 nuclei in their periphery, which are arranged in a

circular or semi-circular “horse-shoe” pattern. FBGCs, on the other hand, have been

shown to form in the presence of foreign bodies like medical devices and can contain

tens to hundreds nuclei, which are irregularly distributed. Both types, however, are

formed from blood monocytes. While it has been suggested that these variants of

MGCs are identical except for their nuclei arrangement, it has more recently been

stated that they fulfill different functions. Furthermore, it was demonstrated that their

formation is induced by differential cytokine signals (McNally and Anderson, 2011;

Okamoto et al., 2003).

Following medical device implantation, an infection is likely, if the acute and chronic

inflammation persists for more than three weeks. Depending on the surface

topography of the implanted material, the degree of the FBR varies, with rough

surfaces provoking a broader reaction. Regarding formation of FBGCs it is thought

that they originate from biomaterial surface adherent macrophages that undergo

fusion (Fig.5). Although they might continue to surround the device for the duration

of implantation, it is not known whether FBGCs remain activated or not. Fusion,

however, is reported to be, amongst others, mediated via mannose receptors

following an up-regulation that is initiated by Interleukin (IL)-4 and IL-13. Usually

macrophages are able to phagocytize small particles (< 5 µm) while larger particles

result in the formation of MGCs (Anderson, 2001; Anderson et al., 2008; Brodbeck

and Anderson, 2009; McNally and Anderson, 2011).

Introduction

- 24 -

Figure 5: Formation of foreign body giants cells (FBGCs). Blood-derived monocytes transform to biomaterial adherent monocytes/macrophages and ultimately to FBGCs at the tissue/biomaterial interface. A group of adhesion receptors (integrins) enables the binding to certain blood proteins that have been adsorbed to biomaterials as well as to ECM molecules such as laminin. Adhesion success is thought to be necessary for FBGC formation (Anderson, 2001). The figure was made based on illustrations of (Anderson, 2001; McNally and Anderson, 2011).

1.8 The sciatic nerve model The rat sciatic nerve model is frequently used to determine the experimental

outcome of peripheral nerve reconstruction (Angius et al., 2012; Swett et al., 1991).

This mixed nerve is formed by spinal nerves L3-L6 and contains sensory and motor

axons. Accordingly, it is involved in movement, proprioreception, skin sensation as

well as smooth muscle functions (Swett et al., 1986; Wang et al., 2005). The ~2000

motoneurons and ~11,000 sensory neurons that form the sciatic nerve mainly reside

in L4 and L5 DRGs (only some are found in L3 and even less in L6). The main

branches of the sciatic nerve include the tibial nerve, common peroneal nerve, and

caudal sural cutaneous nerve (Fig. 6). 96 % of the axons of the sural nerve originate

in DRGs, while only 4 % belong to motoneurons. The common peroneal nerve is the

second largest branch of the sciatic nerve containing a third of all motoneurons

found in the sciatic nerve as well as mainly large diameter sensory neurons. The

largest branch, however, is the tibial nerve, which holds about half of the

motoneurons of the sciatic nerve and also most sensory neurons (~ 4700) (Maciel et

al., 2013; Pardo et al., 2012; Rigaud et al., 2008; Swett et al., 1991, 1986; Ueda,

2006; Wang et al., 2005).

Introduction

- 25 -

Figure 6: The sciatic nerve and its innervation targets. The sciatic nerve contains axons from motoneurons and sensory neurons and is formed mainly by spinal nerves L3-L6. It runs parallel to the femur and splits into three main branches: the tibial nerve, common peroneal nerve, and caudal sural cutaneous nerve. The deep peroneal nerve, a branch from the common peroneal nerve, innervates, amongst other, the tibialis anterior (TA) muscle, while the lateral plantar branch of the tibial nerve supplies the short flexor muscles of digits (plantar (PL) muscles). Innervation of the lateral plantar surface occurs via the caudal sural cutaneous nerve and lateral plantar nerve, while the medial plantar surface is innervated via a branch from the saphenous nerve and a branch from the medial plantar nerve (Gelderd and Chopin, 1977; Pardo et al., 2012; Popesko et al., 1990; Swett et al., 1986).

Evaluation of muscle reinnervation following nerve repair in this model can be

evaluated with regard to motor recovery via electrophysiological recordings of

compound muscle action potentials (CMAPs). This method allows, amongst other,

Introduction

- 26 -

the determination of the NCV. Recordings are performed from target muscles of the

sciatic nerve like the gastrocnemius muscle, tibialis anterior (TA) (synonym in

animals: cranial tibial) muscle, and short flexor muscles of digits, which is here

referred to as plantar muscle (PL). The branch that innervates the TA muscle is the

deep peroneal nerve, which arises from the common peroneal nerve, while

innervation of the PL muscle occurs via the muscular branches of the tibial nerve

(Nickel et al., 2004; Popesko et al., 1990). Another widely used method to assess

muscle reinnervation is the Static Sciatic Index (SSI), which is calculated by

measuring the distances between toes. Following sciatic nerve injuries, the toe

positions are altered due to the missing innervation of the muscles (Bervar, 2000).

The sciatic nerve model is further suitable to evaluate sensory recovery using

mechanical or thermal algesimeter tests (Cobianchi et al., 2013; Popesko et al.,

1990).

1.9 Aim of the study Although, ANGs continue to be the clinical standard therapy to bridge peripheral

nerve defects their usage comes with disadvantages (donor site morbidity, second

surgery, time consuming, and mismatch issues). Various nerve guidance conduits

have already been approved to be officially marketed, but none of the products fulfils

the requirements to prevail on the market. Therefore, alternative materials are

needed that facilitate peripheral nerve regeneration and keep the FBR minimal.

Addressing this issue, hollow chitosan conduits of 3 different DAs were tested in a

10 mm rat sciatic nerve model regarding their possible usage to reconstruct small

nerve defects (e.g. digital nerve defects). Short-term studies over 5 and 18 days

were performed as well as a long-term study (13 weeks) and analyzed with regard to

various parameters. Following an implantation period of 18 days and 13 week, the

inflammatory response was assessed through an (immuno-) histological analysis of

the fibrous capsule that had formed around the conduits as well as of nerve tissue.

Moreover, the physiological regulation of regeneration-associated proteins in DRGs,

nerve stumps, and the lumbar spinal cord region was analyzed by quantitative

(q)RT-PCR as part of the short-term studies. Furthermore, the functional recovery

was investigated via electrophysiological recordings and determination of the SSI in

Introduction

- 27 -

course of the long-term study. Mechanical tube properties were further noted for all

studies to ensure stability of the conduits.

Since treatment of long nerve defects (medial-, and ulnar nerves, brachial plexus

injuries) is thought to require a 3D guidance structure within conduits that provides

support for migrating SCs and axonal sprouts a second study was conducted.

Chitosan conduits filled with enriched NVR-Gel containing genetically modified SCs

either over-expressing GDNF or FGF-218kDa, were tested in a 15 mm rat sciatic nerve

model for a time period of 17 weeks. The functional outcome was determined

through electrophysiological recordings and SSI, while sensory recovery was further

assessed by determination of the mechanical pain threshold. Additionally, the

inflammatory response after an implantation time of 17 weeks was analyzed.

- 28 -

Materials

- 29 -

2. Material Materials Manufacturer 1 ml single-use syringe Injekt® Braun, Germany 5 ml single-use syringe Injekt® Braun, Germany Adson forceps World Precision Instruments, Germany Anatomical forceps World Precision Instruments, Germany AxioVision Zeiss, Germany Bipolar steel hook electrode Medizintechnik MHH, Germany

Chitosan hollow tubes DAI-III Medovent GmbH, Germany & Altakitin S.A, Portugal

Cover slip (Superfrost Plus®) Thermo Scientific, Gerhard Menzel GmbH, Germany

Dumont forceps #5 World Precision Instruments, Germany Dumont forceps #55 World Precision Instruments, Germany Ethilon 4-0 Ethicon®, Belgium Ethilon 9-0 Ethicon®, USA

Glass slide (Menzel-Gläser) Thermo Scientific, Gerhard Menzel GmbH, Germany

Iris scissors World Precision Instruments, Germany iScript Kit BioRad, Germany Micro scissor World Precision Instruments, Germany Mowiol® Calbiochem, Germany Needle electrodes Natus Europe GmbH, Germany Nitrocellulose membrane (Amersham Hybond ECL) GE Healthcare, UK NVR Gel N.V.R. LABS LTD., Israel Raucodrape® Lohmann & Rauscher, Germany Photoshop® CS2 Adobe® Systems Inc., USA Polysorb™ 3-0 Syneture™ Tyco Healthcare, USA Rats Janvier, France Razor Kai Co., LTD, Japan Rneasy Plus Micro Kit 50 Quiagen®, Germany Scalpel No 21 Feather® Safety Razor Co. LTD., Japan Thermostatic blanket (Bosotherm 1500) Bosch & Sohn GmbH, Germany Canon MV 20i, Digital video camcorder Canon, Japan

Instruments & Software Manufacturer Electronic von Frey anesthesiometer IITC Life Sciences, USA CellP software Olympus, Germany CellSense dimensions software Olympus, Germany CellSense entry software Olympus, Germany Citadel tissue processor 2000 Thermo Scientific, USA GraphPad PRISM® 6.0 GraphPad Software, Inc., USA ImageJ software National Institutes of Health, USA Microscope (Olympus IX70) Olympus, Germany Microscope (Olympus BX51) Olympus, Germany

Materials

- 30 -

Microscope (Olympus BX53) Olympus, Germany Microtome RM 2155 Leica, Germany StepOnePlus instrument Applied Biosystems®, Germany StepOne™ software Applied Biosystems®, Germany Toshiba Tecra 8200 Toshiba, Japan

Portable electrodiagnostic device Medtronic functional diagnostics A/S, Denmark

Antibodies Manufacturer Dilution Western Blot α-BDNF Santa Cruz, Germany (sc-546) 1:1000 Primary antibodies - Immunhistochemistry

α-NF200 Sigma-Aldrich®, Germany (N4142) 1:500

α-ED1 Serotec, UK (MCA275R) 1:1000 α-GAP43 Millipore™, Germany (AB5220) 1:200 Secondary antibodies - Immunhistochemistry ALEXA 488 α-rabbit Invitrogen, Germany (A11034) 1:1000 ALEXA 555 α-mouse Invitrogen, Germany (A21422) 1:1000

Chemical Manufacturer 4', 6-diamidin-2-phenylindol (DAPI) Sigma-Aldrich®, Germany (1:2000) Acetic acid Merck, Germany Acetic fuchsin Merck, Germany Amitripyline hydrochloride (Amitriptylin®) Neuraxpharm, Germany Bovine serum albumin (BSA) PAA, Germany Bepanthen® Bayer Vital GmbH, Germany Brain derived nuerotrophic factor (BDNF) PeproTech Inc, USA Braunol® B. Braun Meisungen AG, Germany Buprenorphine (Buprenovet®) Bayer Vital GmbH, Germany Butorphanol tartrate (Torbugesic®) Pfitzer GmbH, Germany Chloral hydrate Sigma-Aldrich®, Germany Sodium citrate tribasic dihydrate (0.01M, pH 6.0) Sigma-Aldrich®, Germany Corbit-balsam Hecht, Germany Distilled water (H2O) Millipore™, Germany Eosin Roth®, Germany Ethanol Mallinckrodt Baker, Germany Hematoxylin Roth®, Germany

Light green Chroma Gesellschaft, Schmidt & Co., Germany

Liquid nitrogen Linde Gas Therapeutics, Germany Milk powder Heirler, Germany Normal goat serum (NGS) Gibco, Germany

Materials

- 31 -

Orcein Merck, Germany Paraformaldehyd (PFA) Sigma-Aldrich®, Germany Phosphate buffered saline (PBS) Biochrom, Germany Rabbit serum Sigma-Aldrich®, Germany Sodium chloride Braun, Germany Sodium dodecyl sulfate (SDS) Roth®, Germany SYBR-Green PCR Master Mix Applied Biosystems®, Germany Triton-X-100 Roche, Germany Wolframato phosphoric acid Merck, Germany

- 32 -

Methods

- 33 -

3. Methods 3.1 Animals All animal experiments followed the German Animal Protection Act and were

approved by the local animal protection committee (Bezirksregierung LAVES,

Hannover, Germany, AZ 33.12-4202-04-12/0816). Adult female Wistar rats

(220-250 g) were obtained from Janvier (France) and housed in groups of 4 under

standard conditions (room temperature (RT) 22 ± 2 °C; humidity 55 ± 5 %; 14 h

light/10 h dark cycle). Food and water was available ad libitum for the complete

duration of the experiments.

3.2 Experimental design and groups Hollow chitosan conduits (inner diameter: 2.1 mm, wall thickness: 0.3 mm) of

different DAs were tested in short and long-term studies (project 1 / 10 mm gap).

Furthermore, a second long-term study was performed with chitosan conduits

containing enriched NVR-Gel. This gel is comprised of extracellular matrix

components such as HA (high molecular weight, 1 %) and laminin 1 (10 ng/ml) and

was used as a matrix to deliver genetically modified SCs, either overexpressing

GDNF or FGF-218kDa, to the injury site (Rochkind et al., 2009; Shahar et al., 2001)

(project 2 / 15 mm gap). The experimental procedures for both projects are

separately described in the following text.

3.2.1-a Project 1-Short-term study A total of 36 animals out of 44 were subjected to sciatic nerve surgery and a 10 mm

gap was introduced. Subsequently, nerve reconstruction was performed with hollow

chitosan conduits of a low (2 %), medium (5 %), or high (20 %) DA (n = 12,

respectively). The remaining eight animals served as controls. Following an

implantation time of 18 days, the fibrous capsule that had developed around the

tubes as well as the proximal tissue were harvested and embedded in Paraffin for

(immuno-) histological analyses. Furthermore, conspicuous features of the conduits

upon explantation were noted regarding their mechanical properties. Additionally,

DRGs, nerve stumps, and the lumbar spinal cord region were harvested at both time

points and snap frozen for qRT-PCR, while the tube contents were collected and

used for Western Blot analysis (Tab. 1).

Methods

- 34 -

Table 1: Overview of the collected tissue samples in the short-term study of project 1. n.a. = non analyzed; n.av. = not available

3.2.1-b Project 1-Long-term study Sixty-eight rats were included in this study and a 10 mm gap was introduced into the

sciatic nerve of all animals. Overall 51 of these defects were treated with hollow

chitosan conduits of the three different DAs (see 3.2.1-a), while ANGs were used to

reconnect the two nerve stumps in the remaining cases (n = 17, respectively) (Fig. 7). Tests for functional evaluation of the nerve reconstruction were performed

4, 9, 12, and 13 weeks after surgery (SSI, electrodiagnostic recordings). On the final

day of observation, animals were sacrificed and the regenerated nerve tissue and

fibrous capsules were harvested for (immuno-) histological and morphometric

analyses. Additionally, the muscle weight of the TA muscle and PL muscle was

determined as were the mechanical tube properties upon explantation.

Methods

- 35 -

Figure 7: Experimental design for the long-term study (13 weeks) using hollow chitosan conduits in a 10 mm gap (project 1). (A) Illustration of nerve reconstruction with hollow chitosan conduits in the 10 mm rat sciatic nerve model. (B-D) Schematic illustration of the three experimental groups that were treated with chitosan conduits of different DAs (DAI = 2%, DAII = 5%, DAIII = 20%) after introduction of 10 mm gap into the sciatic nerve as well as of (E) the control group that received reversed ANGs.

3.2.2 Project 2 For the second project a 15 mm, gap was introduced into the sciatic nerve of 40

female Wistar rats and the nerve was reconstructed using chitosan conduits

enriched with NVR-Gel containing 1) non-transfected neonatal SCs (nt-neoSCs) 2)

SCs that were transfected with 3) an empty vector (tf-SCs-EV), 3) GDNF (tf-neoSCs-

GDNF), or 4) FGF-218kDa (tf-neoSCs-FGF-218kDa) (n = 8, respectively). The

preparation of the conduits was performed by Sandra Wrobel (MSc, Medizinisiche

Hochschule Hannover (MHH) Neuroanatomy), and is described in detail in her PhD-

thesis. In short, three days prior to transplantation 2.5 x 106 neoSCs (neoSCs; 8.-9

passage; purity: 75-90 %) were genetically modified by AMAXA nucleofection

(transfection rate: 73-80 %). Afterwards, the cells were allowed to recover in culture

before a total of 1 x 106 t-SCs as well as nt-SCs were mixed with 0.5 % NVR-Gel

(110 µl 1st series and 130 µl 2nd series) and placed inside each chitosan conduit.

24 h following conduit preparation, surgery was performed and the conduits were

used to bridge the nerve defects. Over-expression of GDNF and FGF-218kDa was

confirmed by Western Blot analyses by (see Sandra Wrobel, PhD-thesis). In control

animals the nerve defect was bridged with 5) ANGs (n = 8) (Fig. 8). Tests regarding

Methods

- 36 -

sensory and motor recovery were performed 4, 9, 12, and 16 weeks after surgery by

means of SSI and mechanical pain threshold assessment. Furthermore,

electrodiagnostic measurements were performed following an implantation time of 9,

12, and 16 weeks. On the final day of observation, animals were sacrificed and the

regenerated nerve tissues (if regeneration was successful) as well as the fibrous

capsule that had formed around the conduits were harvested for (immuno-)

histological analysis. In cases of failed or thin tissue regeneration proximal nerve

stumps or total samples (nerve stumps + conduit (+ fibrous capsule)) were

embedded. Additionally, the muscle weight of the TA muscle and gastrocnemius

muscle was determined as were the mechanical tube properties upon explantation.

Schematic illustration of the four experimental groups that were treated with chitosan conduits (DA = 5 %) enriched with NVR-Gel containing either (A) nt-neoSCs, SCs that were transfected with (B) an EV, (C) GDNF, or (D) FGF-218kDa. (E) ANGs served as controls.

3.4 Anesthesia and surgical procedures During surgery and electrodiagnostic recordings, the rats were in general

anesthetized by an i.p. injection of chloral hydrate (370 mg/kg) and intramuscularly

(i.m.) applied buprenorphine (0.045 mg/kg, Buprenovet®) (project 1 & 2). On the two

days following the initial surgery, the animals also received butorphanol tartrate

(0.5 mg/kg, Torbugesic ®) via subcutaneous injection (s.c.) as pain relief (project

Figure 8: Experimental design for the long-term study (17 weeks) using enriched tubes in a 15 mmgap (project 2).

Methods

- 37 -

1-short-term study and project 2). Additionally, amitriptyline hydrochloride

(9 mg/kg/day, Amitriptylin®) was administered through the drinking water, starting

two weeks prior to the surgery and was given for the time period of the experiments

to minimize autotomy events (project 2) (Navarro et al., 1994).

Animals were prepared for surgery by shaving the left hind leg and disinfecting the

skin. During surgery, to keep the body temperature stable, the rats were placed on a

thermostatic blanket. Aseptic techniques were further used to ensure sterility. A skin

incision was made with a surgical blade along the femur, and the gluteus and biceps

femoris muscles were separated by sharp and blunt dissection to expose the sciatic

nerve. The wound was kept open with the help of small retractors. The sciatic nerve,

exposed at the mid-thigh level, was transected with a single micro scissors cut at a

constant point (6 mm from the exit of the gluteus muscle). For tube transplantation,

5 mm of the sciatic nerve end were removed and 14 (Fig. 9) or 19 mm long conduits

were used to bridge the nerve defects leaving a 10 mm (project 1) or 15 mm gap

(project 2) between the two nerve stumps, respectively. Control animals of the

long-term studies received reversed ANGs of 10 (projects 1) or 15 mm (project 2)

length. Therefore, a sciatic nerve piece of the desired length was cut out, reversed

(distal-proximal), and flipped 90° before 3 sutures (9-0) were placed 120° apart from

each other. In control animals of the 5 and 18 days short-term studies (project 1), the

nerve defect was not bridged, but the proximal and distal nerve ends were folded

over and sutured to themselves after removal of the 5 mm nerve piece (Haastert-

Talini et al., 2013). The critical steps of the surgeries were conducted by Prof.

Haastert-Talini (MHH Neuroanatomy), while it was my responsibility to expose the

sciatic nerve and close the wound. Jennifer Metzen (biology lab assistant, MHH

Neuroanatomy), was further in charge of all aspects regarding anesthesia.

Methods

- 38 -

Figure 9: Chitosan conduit implantation. Representative photograph showing a hollow chitosan conduit bridging a 10 mm sciatic nerve defect.

3.5. Functional evaluation of nerve regeneration. During the observation periods of both long-term studies (project 1 and 2), the

progress of functional recovery was monitored by means of SSI, electrophysiological

recordings (serial non-invasive and final invasive), and calculation of the lower limb

muscle weight ratio. Additionally, in the scope of project 2, the von Frey test was

performed to assess sensory recovery, while cases of autotomy were noted using an

autotomy score. The investigators were blinded to the conditions of sciatic nerve

reconstruction applied to the animals.

3.5.1 Static Sciatic Index (SSI) To assess motor recovery after nerve repair the SSI was determined. This index is

based on changes in the hind paw posture that occur following sciatic nerve injury

due to missing innervation. In case of successful regeneration changes of the

posture gradually decrease. For project 1 SSI analysis was conducted before

surgery and during weeks 1, 4, 9, and 12 following surgery. In the course of project

2, however, this test was additionally performed 16 after nerve repair.

For this test, a plastic box was placed on a glass table and rats were transferred

inside the box (Fig. 10). A webcam (Canon MV 20i) placed under the glass table and

a notebook (Toshiba Tecra 8200) allowed the image acquisition of the paw positions.

These images were then exported to a freely available image-editing program

(AxioVision, Zeiss, Jena, Germany). The distance between toes 1 and 5 (toe spread,

TS) as well as toes 2 and 4 (intermediate toe spread, ITS) was measured for the

lesioned (L) and non-lesioned (N) hind paws. The SSI was then calculated using the

Methods

- 39 -

following formulas, which were introduced by Bervar (Bervar, 2005, 2000) (TSF = toe

spread factor; ITSF = intermediate toe spread factor):

TSF= (L-TS - N-TS) / N-TS

ITSF= (L-ITS - N-ITS) / N-ITS

SSI= (108.44 * TSF) + (31.85 * ITSF) - 5.49

The setup for the SSI includes (A) a webcam placed under a glass table, on which the animals are positioned so that (B) images can be taken of the paw positions. For calculation of the SSI, the toe spread (TS, marked in pink) and intermediate toe spread (ITS, marked in green) are measured. Animals that displayed autotomy were not included in the test and account for changes in numbers of evaluated animals.

3.5.2 Von Frey test An electronic von Frey algesimeter was used to assess the mechanical pain

threshold to determine sensory recovery following nerve reconstruction. Therefore,

plastic compartments were positioned on a metallic mesh stand and animals were

placed inside the boxes and allowed to acclimate to the surrounding (15 minutes)

(Fig. 11). Following habituation, a blunted needle connected to a force sensor in a

hand-held probe was used to stimulate the medial and lateral site in the intact and

injured paw. Pressure was increased gradually and in a slow manner until the animal

withdrew the tested paw. The applied force needed to induce paw lifting was

Figure 10: Static SciaticIndex (SSI).

Methods

- 40 -

displayed as the weight-force (in grams). A cut-off value was set to 40 g, when it was

observed that the paw was lifted simply by the hand-held probe and not as a result

from an active response from the animal or in cases where no paw lifting at all was

seen. The interval between the stimuli was at least 3 min. The paw withdrawal

pressure was recorded five times for each site and the three steadiest values were

used for calculation of a mean that was then used for statistical analysis. The test

was performed three times before surgery for habituation and then in weeks 1, 4, 9,

13, and 16 following surgery.

Figure 11: Von Frey. Sensory recovery was assessed under usage of the von Frey algesimeter. (A) The plastic chambers, in which the animals were placed and allowed to habituate for 15-30 minutes. Following habituation, a B) hand held probe that consists of a force sensor and a blunted needle was used to stimulate the lateral and medial side of the paws, which can be seen through the (C) holey metal plate.

3.5.3 Electrophysiology 3.5.3.1 Serial non-invasive electrodiagnostic recordings

For evaluation of ongoing muscle reinnervation in the long-term studies, a portable

electrodiagnostic device (Keypoint Portable) was used and recordings were

performed in weeks 4, 9, and 13 for project 1 and in weeks 9, 12, and 16 for project

2. Therefore, animals were anesthetized as described above and fixated with the

hind limbs extended on a metal plate on top of a thermostatic blanket. Only during

recordings, the thermostatic blanket was turned off to avoid interfering signals. The

sciatic nerve was stimulated with single electrical pulses (100 μs duration and

supramaximal intensity) delivered by monopolar needles (30 G, diameter 0.3 mm,

length 10 mm), which were percutaneously placed either at the sciatic notch, thus

Methods

- 41 -

proximal to the injury (proximal stimulation: cathode directly at sciatic notch, anode

1 cm more proximal) or in the popliteal fossa (distal stimulation) (Fig. 12). The

CMAPs of the TA and PL muscles were recorded by means of another pair of

monopolar needles inserted across the skin on the muscle belly. To ensure

reproducibility, the recording needles were placed with the help of anatomical

landmarks to secure the same placement on all animals. The active recording

electrode was placed subcutaneously at the first third of the distance between knee

and ankle on the TA muscle, or at the third metatarsal space for PL muscle

recordings. The reference needle electrode was inserted at the distal phalange of the

second or fourth toe. A ground needle electrode was inserted into the skin. Taking

into account the latency difference of the two stimulation points and the distance

between them, the electrodiagnostic device calculated automatically the NCV

(Haastert-Talini et al., 2013). For each animal, four recordings per time point and

recording site (TA, PL muscles of both body sides) were performed and means were

used for further calculations. A NCV-ratio was calculated according to the following

equation (L = recorded values at one time point on the lesioned site, NM = mean of all

recorded values from the non-lesioned site obtained during the complete follow-up

period):

NCV-ratio = L/NM

Furthermore, the amplitude and the area under the curve (AUC) of the negative

CMAP peaks were analyzed. For evaluation of the amplitude, a ratio was calculated

according to the concept described for the NCV. AUC-values of the CMAPs were

used to calculate the percentage of axon loss (AxL) according to the following

equation:

AxL [%] = NM-L/NM x 100

Methods

- 42 -

Figure 12: Serial non-invasive electrodiagnostic recordings. To evaluate the functional outcome electrophysiological recordings from the TA and PL muscle were conducted. (A) For non-invasive recordings, grounding, reference, and stimulation electrodes were placed at defined landmarks. Furthermore, recording electrodes were positioned either on the muscle belly of the (B) TA or (C) PL muscle.

Regarding recordings from the TA muscle, it is important to note that potential

artifacts from neighboring muscles were seen and early regeneration signals partly

fused together with these artifacts. If the correct signal was not clearly identifiable, it

was not analyzed. The recordings were performed together with Prof. Haastert-Talini

(MHH Neuroanatomy).

3.5.3.2 Final invasive recordings

On the last day of observation, animals of the long-term studies were once more

anesthetized and the sciatic nerve was exposed consecutively on the lesioned and

non-lesioned side. The recording electrodes were placed as described above. Using

a bipolar steel hook electrode the sciatic nerves were directly stimulated proximally

and distally to the transplants. Single electrical pulses (100 μs duration) with

Methods

- 43 -

gradually increased intensity (not exceeding 8 mA) were applied in order to evaluate

the maximal CMAP. The current required to elicit the maximal CMAP was also

determined as was the latency of these signals. Furthermore, the NCV as well as

AUC were calculated by stimulating the sciatic nerve, proximal and distal to the

injury, with a current intensity of 1 mA (Haastert-Talini et al., 2013). The NCV-ratio,

AxL, and amplitude-ratio were then calculated using the equations noted under

3.5.3.1, but only values obtained during the invasive recordings were taken into

account for calculations.

3.5.4 Muscle weight After completing the long-term experiments, the animals were sacrificed and the

gastrocnemius and TA muscle of both hind limbs were excised and weighted in order

to determine the muscle weight ratio ([g]ipsilateral/[g]contralateral) (Haastert-Talini et

al., 2013).

3.5.5 Autotomy score Events of autotomy were reported using an autotomy score based on a point-

system. As part of this system one point per automutilation site (nails, distal

phalanges, proximal phalanges, and plantar pads) was counted (Fig. 13).

Methods

- 44 -

Figure 13: Autotomy Score. (A-B) Nails, the proximal and distal phalanges from the animals toes (U1-U5) as well as the plantar pads (indicated by asterisks) were taken into account for the determination of the autotomy score.

3.6 (Immuno-) histological evaluation of nerve regeneration 3.6.1 Assessment of nerve tissue 3.6.1.1 Project 1-short-term study

After an implantation time of 18 days, the transplants were harvested together with

the matrix that had formed inside the tube. While the tube content was used for

Western Blot analysis (see 3.7.2), the proximal nerve stumps were fixated in 4 %

paraformaldehyde (PFA) overnight at 4°C. Afterwards, samples were

paraffin-embedded with the help of a Citadel tissue-embedding-automatic unit and

serial 7 µm cross-sections were prepared (Fig. 14). Blind-coded sample sections

(from 3 randomly chosen animals per conduit group as well as from 2 random control

animals), taken from the distal end and approximately 390 µm more proximal, were

subjected to hematoxylin-eosin (HE)-staining in order to determine the number of

MGCs per mm2. Therefore, sections were examined manually under light microscopy

(Olympus BX53) and overview photomicrographs were used for measurements of

the area.

Methods

- 45 -

Figure 14: Processing of the nerve tissue gathered 5 and 18 days after surgery (project 1). After 5 and 18 days, a matrix had formed inside the tube, but no regenerated nerve cable was found. Therefore, the proximal nerve stump was paraffin-embedded and 7 µm sections were prepared for (immuno-) histological analyses.

Additional blind coded sample sections (from the same animals), taken at defined

distances in distal-proximal direction (0 µm, 72 µm, 152 µm, 320 µm, 392 µm and

472 µm), were stained for activated macrophages (ED1) and GAP-43. Following a

pre-incubation step in citrate buffer (heating until boiling then 9 min at 95°C and

30 min at RT), the sections were incubated in phosphate buffered saline (PBS) +

5 % rabbit serum as a blocking step prior to incubation with primary mouse α-ED1

antibody (1:1000, in blocking) at 4°C overnight. The next day, sections were washed

with PBS (3x) before being incubated with ALEXA 555-conjugated secondary goat

α-mouse antibody (1:1000, in blocking) for 1h at RT. A second blocking step in 0.3 %

Triton + 3 % normal goat serum (NGS) followed before rabbit α-GAP43 (1:200; in

blocking) was applied overnight at 4°C. As a next step, the sections were washed in

PBS (3x) and incubated with ALEXA 488-conjugated goat α-rabbit antibody (1:1000,

in blocking) for 1h at RT. The sections were finally counterstained with the nuclear

dye 4’, 6’-diamino-2-phenylindole (DAPI, 1:2000 in PBS) in PBS for 2 min at RT and

then mounted with Mowiol. For quantification of the number of activated ED1+

macrophages/mm2, seven photomicrographs per section were randomly taken (20x

magnification) almost covering the whole cross section and counting was performed

using ImageJ software.

Methods

- 46 -

3.6.1.2 Project 1-long-term study

On the last day of observation, the transplants with the adjacent nerve tissue were

harvested and the regenerated nerve tissue was extracted from the chitosan tubes.

Afterwards, the regenerated nerve tissue was separated into a proximal and a distal

part by a cut 1 mm distal to the distal suture (Fig. 15). Five mm of the distal segment

were then prepared for the stereological analysis and shipped to our partner

laboratory at the University of Turin (UNITO) (for methods and results see (Haastert-

Talini et al., 2013). The proximal regenerated nerve tissue was subjected to fixation

in 4 % PFA overnight at 4°C. Next, paraffin-embedding was performed in a Citadel

tissue-embedding-automatic unit and serial 7 µm cross-sections were prepared.

Two blind-coded sample sections (from randomly chosen animals; n = 5,

respectively) taken from the distal end as well as 320 µm more proximal were

subjected to HE-staining in order to examine the number of MGCs per mm2.

Sections were examined manually under light microscopy (Olympus BX53).

Overview photomicrographs were used for measurements of the area.

Figure 15: Processing of the regenerated nerve tissue 13 weeks after implantation (project 1). 13 weeks after surgery, the regenerated nerve tissue was exposed and prepared for further analysis. Therefore, a cut was made 1 mm behind the distal suture and the proximal part was embedded in paraffin for (immuno-) histological analysis, while the distal part was sent to our partner laboratory (UNITO, Italy) for morphometric evaluation. From the proximal nerve tissue serial 7 µm sections were prepared (in distal to proximal direction, covering a distance of approximately 1000 µm) and subjected to different stainings.

Additional blind-coded sample sections (from randomly chosen animals; n = 8,

respectively), taken at defined distances in distal-proximal direction, were subjected

Methods

- 47 -

to double-immunohistochemistry (IHC) for activated macrophages (ED1) and

neurofilament (NF) 200. Quantification of the number of activated macrophages/mm2

was performed by Sandra Wrobel (MSc, MHH Neuroanatomy) (for methods and

results see (Haastert-Talini et al., 2013)).

3.6.1.3 Project 2

Following an implantation time of 17 weeks, the transplants were exposed and the

nerve tissue was cut 2 mm distal to the distal suture site, thus leaving a proximal as

well as distal segment. In accordance to project 1, 5 mm of the distal segment were

prepared for morphometric analysis and sent to our partner laboratory (UNITO). In

case of clear tissue regeneration, the chitosan scaffold was carefully removed and

the proximal tissue sample was subjected to fixation in 4% PFA (overnight at 4°) for

subsequent paraffin-embedding and (immuno-) histological analysis. Furthermore, in

case of failed regeneration single proximal nerve stumps were prepared for

histological analysis. Samples displaying thin regeneration as well as one sample of

successful regeneration, and representative examples of failed regeneration were

embedded together with the chitosan conduit, distal nerve stump, and, if applicable,

together with the surrounding connective tissue.

Seven µm cross-sections were prepared from samples showing clear tissue

regeneration as well as of representative proximal nerve stumps gained from cases

of failed regeneration. The sectioning was undertaken in distal-proximal direction as

described for project 1 and covered a distance of approximately 600 µm. After

discarding a 1.5 mm piece of the sample a second series of 600 µm was cut and

gathered for histological analysis. Two sections of each series were HE-stained and

evaluated with regard to the number of MGCs, tissue thickness, and area as

described for project 1 (long-term study). Adjacent sections were subjected to

ED1-staining, to enable counting of activated macrophages, and NF200 staining.

ED1-staining staining, however, was performed according to a new staining protocol

compared to project 1. Therefore, paraffin cross-sections were incubated in PBS with

3 % milk powder and 0.5 % Triton X 100 for blocking before incubation with primary

mouse α-ED1 antibody (1:1000, in blocking solution) at 4°C overnight. Following

three washing steps in PBS, incubation with Alexa 555-conjugated secondary goat

α-mouse antibody (1:1000, in blocking) for 1h at RT was performed. Afterwards, a

Methods

- 48 -

second blocking step with PBS containing 3 % milk powder and 0.5 % triton X-100

and subsequent overnight incubation with primary rabbit α-NF200 antibody was

conducted. Next, three washing steps followed, before incubation with Alexa

488-conjugated secondary goat α-rabbit antibody (1:1000, in blocking) for 1h at RT

and counterstaining with DAPI (1:2000, in PBS) was performed. Finally, sections

were mounted using Mowiol. Five images (20x magnification) were acquired per

section or less if the whole section was already covered and counting of activated

macrophages was performed with the help of ImageJ software by Natalie Schecker

(internship, MHH Neuroanatomy).

Exemplary samples of thin as well as failed tissue regeneration were cut in the

middle of the tube and sections were gathered covering a distance of about 300 µm.

Staining for activated macrophages and axons was performed as described above.

3.6.2 Assessment of fibrous capsule Eighteen days and 13 weeks (project 1) as well as 17 weeks (project 2) after

surgery, a fibrous capsule / connective tissue had formed around the conduit, which

was macroscopically and histologically analyzed. For the macroscopic evaluation,

the tissue was dissected, spread on a scale paper glass plate, and photographs

were taken of every sample. From the photographs, three classes of macroscopic

connective tissue optical density were defined as translucent, translucent-opaque,

and opaque (only project 1). Afterwards, samples of the connective tissue (n = 4-6)

were folded and sutured at the proximal end so that the chitosan conduit contact

surfaces were facing each other. The tissue was then fixed overnight in 4 % PFA

and paraffin-embedded. For histological analysis, blind-coded 7 µm paraffin sections

were cut from the distal end of the connective tissue and subjected to HE-, Trichrom-

, or Orcein-staining as well as to IHC for activated macrophages.

3.6.2.1 Project 1

In 4-6 HE-stained sections per sample, covering a distal-proximal distance of about

400 µm, the connective tissue area and thickness (at 4 randomly chosen points)

were determined. Light microscopic evaluation (Olympus BX53 and Olympus BX51)

was aided by the programs CellSense Dimension and CellSense Entry and overview

photomicrographs were used for measurements (Fig. 16). Furthermore, the number

Methods

- 49 -

of MGCs per mm2 within the former contact area with the chitosan conduits (18 days)

or in the whole section (13 weeks) was manually quantified. Trichrom- and

Orcein-stainings were performed to visualize the distribution of collagen and elastic

fibers, which are amongst others found in the wall of vessels (Ushiki, 2002). For the

Trichrom-staining, sections were subjected to hematoxylin, before being washed-

offed under running water. Afterwards, 20 ml solution containing acetic acid (1 ml in

99 ml aqua dest) and acetic fuchsin (0.5 g, in 100 ml 1 % acetic acid) were mixed

with 10 ml light green (1 g in 100 ml 1% acetic acid) and 10 ml wolframato

phosphoric acid (1 g in 100 ml H2O) and the slides subjected to this solution.

Following a washing step in aqua dest, the slides were subjected to 1 % acetic acid,

before being pressed with filter paper, dehydrated, and mounted with corbit-balsam.

For orcein-staining, slides were exposed to a staining solution (for 45 min) containing

orcein (1 g in 99 ml 70 % ethanol) mixed with 25 % hydrochloric acid. Following a

washing step in aqua dest the slides were placed in 96 %, 100 % ethanol, and xylol,

consecutively, before being mounted with corbit-balsam. Representative

photomicropgraphs were taken using the programs CellSense Dimension and

CellSense Entry

IHC was performed to estimate the number of ED1 stained cells (n = 4-5 / group).

Therefore, two blind-coded 7 µm paraffin cross sections, approximately 320 µm

apart from each other, were stained for ED1 as described above (see 3.6.1.1) and

after incubation with goat α-mouse antibody the sections were counterstained

directly with DAPI and mounted with Mowiol. Two photomicrographs each, were

taken from an area with minimal as well as maximal ED1 signal per section (20x

magnification) using an IX70 microscope and CellP software. The number of ED1+

cells was counted as number/mm2 with the help of ImageJ software. The nuclear

DAPI counterstain was used to clearly identify ED1+ cells (Haastert-Talini et al.,

2013).

Methods

- 50 -

Figure 16: Illustration of fibrous capsule analysis. 18 days and 13 weeks after implantation, connective tissue had developed around the implanted scaffolds. (A) The area of the connective tissue was determined in HE-stained 7 µm paraffin cross-sections. In these sections, the inner surfaces of the connective tissue (former contact area with chitosan conduits) are facing each other. (B) Furthermore, the thickness of the connective tissue was measured at 4 randomly chosen locations. (C, D) After 18 days, MGCs were further quantified within the contact area with the chitosan conduits, while for the long-term studies the whole tissue area was taken into account. Scale bars: 200 µm, 200 µm, 10 µm, a 5 µm.

3.6.2.2 Project 2

For the second project, 7 µm thick sections were cut from the distal end of the

samples and four sections within a distance of 2 cm were subjected to HE-staining

and analyzed with regard to number of MGCs, tissue thickness and area as

described above. Also, Trichrom- and Orcein-stainings were conducted according

the procedures noted for project 1.

Additional two sections obtained from the distal end, covering a distance of

approximately 300 µm, were stained for activated macrophages using staining

protocol stated under 3.6.1.3. Following the incubation step with Alexa 555

conjugated secondary goat α-mouse antibody (1:1000, in blocking) and subsequent

counterstaining for DAPI, however, samples were mounted with Mowiol. Four

photomicrographs per section were acquired (Olympus IX70) and cell counting was

performed as described before using ImageJ software as described for project 1.

Methods

- 51 -

3.7 Molecular and biochemical evaluation of nerve regeneration 3.7.1 Quantitative real-time polymerase chain reaction (qRT-PCR) The ipsi- and contralateral lumbar region of the spinal cord of animals subjected to

the 5 and 18 days observation period were dissected and snap frozen in liquid

nitrogen (5 days: DAI, DAII, DAIII n = 3 each, control n=4; 18 days: DAI + DAIII n = 3

each, DAII + control n=4 each). Tissue samples of each animal were homogenized

and total RNA was extracted according to the manufacturer’s instructions (RNeasy

Plus Micro Kit 50). The eluate of 14 μl was used completely for cDNA synthesis

using the iScript Kit (BioRad, Germany). Primer sequences were as previously

stated for BDNF, GAP-43, and NGF (Ratzka et al., 2012). Further primer sequences

were as follows: FGF-2-F: 5’-GAACCGGTACCTGGCTATGA-3’; FGF-2-R: 5’-

CCAGGCGTTCAAAGAAGAAA-3’; IL-6-F: 5’- CGTTTCTACCTGGAGTTTGTGAAG-

3’; IL-6-R: 5’- GGAAGTTGGGGTAGGAAGGAC-3’; tyrosine kinase receptor B

(TrkB)-F: 5’- CCCAATTGTGGTCTGCCG-3’; TrkB-R: 5’-

CTTCCCTTCCTCCACCGTG-3‘. Equal PCR efficiency was certified by serial cDNA

dilutions and estimated to be 100 %. qRT-PCR was performed with Power SYBR-

Green PCR Master Mix on a StepOnePlus instrument as described before (Ratzka et

al., 2012). Fold changes in mRNA levels are stated in comparison to the levels

measured on the respective day at the contralateral side of the control animals.

Calculation was done by using the 2(-∆∆Ct) method and normalized to the

housekeeping genes glyceraldehyde-3-phosphate dehydrogenase (Gapdh)-F: 5′-

GAACATCATCCCTGCATCCA-3′, Gapdha-R: 5′-CCAGTGAGCTTCCCGTTCA-3′;

and peptidylprolyl isomerase (Ppia)-F: 5’-TGTGCCAGGGTGGTGACTT-3’; Ppia-R:

5’- TCAAATTTCTCTCCGTAGATGGACTT-3’) (Haastert-Talini et al., 2013; Rumpel

et al., 2013).

Furthermore, DRGs (L3-L5) were gathered for qRT-PCR to investigate expression

pattern of the noted factors as well as of IL-6. This work was performed by Dr.

Andreas Ratzka (MHH Neuroanatomy) and methods as well as results can be

looked up in (Haastert-Talini et al., 2013). Also, proximal and distal nerve stumps

were collected, but processing of the samples was not successful likely due to an

insufficient material amount.

Methods

- 52 -

3.7.2 Western Blot Conduit contents were collected 5 and 18 days after nerve reconstruction and pooled

from 5-6 animals per DA, respectively, to verify the presence of BDNF. Therefore,

samples were homogenized in 40 μl Radioimmunoprecipitation assay (RIPA) buffer

according to (Ratzka et al., 2012). Samples were then boiled in Laemli buffer (5 min)

and separated by sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis

(PAGE) (12 % gel), before being transferred (50 μg) electrophoretically to

nitrocellulose membranes with 4 ng recombinant BDNF serving as positive control.

Membranes were then probed with α-BDNF antibody (1:1000) and analyzed

according to (Ratzka et al., 2012) (Haastert-Talini et al., 2013). This work was

performed by Kerstin Kuhlemann (VMTA, MHH Germany).

3.8 Analysis of tube properties after explantation On day 5 and 18 after nerve reconstruction (project 1 short-term study), the chitosan

tubes were explanted. Within 18 days, a fibrous capsule had developed, which was

macroscopically classified and processed for histological analysis (see 3.6.2). On

day 5 and day 18, representative conduits and their contents were stored for

Western Blot analysis of the content (see 3.7.2). Therefore, the middle parts of the

tubes were harvested by cutting of the proximal and distal end of the tubes at the

suture sites. Also, upon explantation in week 13 (long-term studies) representative

conduits were explanted, dissected from surrounding tissue, and stored in 0.9 %

NaCl solution for further analysis of biomaterial properties. The assets of the

conduits, the connective tissue as well as the macroscopic nerve cable peculiarities

were then noted and connective tissue samples were processed for histological

evaluation. Furthermore, tube samples (before and after nerve reconstruction) were

analyzed by Scanning electron microscopic or biochemical tests (n = 3 / per group)

in our partner laboratory at the University of Minho (UMINHO) (for methods and

results see (Haastert-Talini et al., 2013)). For project 2, again the conduit and

connective tissue properties were noted in the scope of the explantation procedure,

before being prepared for further analysis.

Methods

- 53 -

3.9 Statistical analysis The data obtained from the various functional and molecular tests as well as the

(immuno-) histological analysis were subjected to One-way or Two-way ANOVA

(Analysis of Variance) followed by Dunn’s or Tukey’s multiple comparison test

(GraphPad Prism 6). In general non-parametric statistical approaches were preferred

due to the significant deviation from normality. Furthermore, a Chi-square test was

performed, if applicable. Results are presented as means ± standard error of the

mean (SEM) and p-values were set to < 0.05, < 0.01, and < 0.001 as level of

significance.

- 54 -

Results-Project 1

- 55 -

4 Results-Project 1 4.1 Gene expression of regeneration-associated proteins qRT-PCR was performed to assure that chitosan does not influence the

physiological regulation of NTFs (NGF, FGF-2, BDNF), the cytokine IL-6, BDNF-

receptor TrkB, and GAP-43 following nerve lesion. Therefore, the lumbar region of

the spinal cord was explanted from representative animals of all experimental groups

5 and 18 days after nerve injury and mRNA was extracted. All data were normalized

to the values measured on the respective day for the contralateral nerve of the

control animals.

NGF and IL-6 expression was below detection limit and could not be quantified. No

significant up- or down-regulations of the additional analyzed factors were seen in

the ipsilateral spinal cord of all chitosan conduit grafted animals as well as the

animals of the control group, when compared to the expression levels seen in the

contralateral spinal cord of the control animals 5 and 18 days after surgery,

respectively (Fig. 17).

Figure 17: Regulation of mRNA levels in spinal cord. 5 and 18 days after conduit implantation or control treatment, the lumbar region of spinal cords were harvested and total mRNA was extracted. qRT-PCR was conducted to determine gene expression of representative NTFs and regeneration-associated proteins but did not reveal any significant differences between all experimental groups with regard to their regulation following nerve injury. Values were normalized to the non-lesioned site of the control group. DAI, DAII, DAIII n= 3 each; control n=4. Results were tested for significance (p < 0.05) using Two-way ANOVA.

Results-Project 1

- 56 -

In addition, there were no significant changes in the expression patterns of the

analyzed factors between day 5 and 18. Thus, it can be concluded that implanted

chitosan conduits do not influence the physiological regulation of NTFs following

nerve injury.

4.2 Matrix formation and presence of brain-derived neurotrophic factor (BDNF) Five days after implantation, the content of the tube appeared jellylike, while matrix

formation was seen after an observation period of 18 days (Fig. 18 A-C). This matrix

was rather translucent and partly displayed revascularization (small vessels traveling

along). Furthermore, blood accumulations could be seen accompanying the matrix.

To detect the relative amounts of secreted BDNF into the chitosan conduit the

content with the formed matrix was collected and Western Blot analysis performed.

For both time points, 5 as well as 18 days after nerve reconstruction, pre-mature and

mature BDNF could be detected in the pooled conduit content with no obvious

differences between the groups of different DAs (Fig. 18 D).

Figure 18: Endogenous brain-derived neurotrophic factor (BDNF) detection in tube content. (A-C) Eighteen days after nerve reconstruction a matrix had formed inside the implanted tubes with a jellylike surrounding. The complete tube content was collected and (D) endogenous BDNF was detected to be present in the direct environment of the regenerating nerve tissue (Haastert-Talini et al., 2013).

Results-Project 1

- 57 -

4.2 Static Sciatic Index (SSI) Motor recovery was assessed through determination of the SSI before as wells as 1,

4, 9, and 12 weeks after nerve transection and subsequent nerve reconstruction.

Prior to surgery, healthy animals displayed a SSI around ± 0 (ANG -2.07 ± 1.75; DAI

-7.30 ± 1.71; DAII -4.62 ± 1.72; DAIII -7.96 ± 1.65) indicating that there was no

impairment in the position of the animals paws. One week following nerve lesion and

nerve repair with ANGs or chitosan conduits, the distance between the toes was

altered resulting in a significant decrease of the index for all groups (ANG -73.94 ±

3.21; DAI -74.35 ± 2.84; DAII -76.14 ± 3.31; DAIII -70.72 ± 5.89; (Fig. 19). For the

duration of the complete observation period there was no significant improvement of

the SSI. For the ANG group the SSI mean value was still -54.07 ± 9.07, for the DAI

group -66.12 ± 6.19, for the DAII group -77.92 ± 4.76, and for the DAIII group it was -

75.70 ± 5.38. In addition to the maintained impairment over time, we did not detect

any significant SSI-value differences between the experimental groups.

Results-Project 1

- 58 -

Figure 19: Static Sciatic Index (SSI). (A) Representative photomicrograph of the plantar surface of the animals’ paws before surgery. The toe spread and intermediate toe spread of both hind paws is similar and indicates that there was no impairment of the paw posture prior to sciatic nerve lesion. (B) One week following nerve reconstruction, a change in the posture of the paw is visible as indicated by the narrowed TS and ITS on the operated left side. (C) Evaluation of motor recovery using the SSI did not reveal any significant differences between animals that were treated with the clinical standard therapy, the ANG, and animals that had received hollow chitosan conduits of different DAs for nerve reconstruction. Variances as well as a decrease in animal numbers over time and per group are the result from autotomy and cases of early death due to repeated anesthesia. Results were tested for significance (p< 0.05) using Two-way ANOVA, followed by Tukey’s multiple comparison. Triple asterisk indicating difference to post nerve lesion values within the same group, p< 0.001. (Haastert-Talini et al., 2013)

Results-Project 1

- 59 -

4.3 Electrophysiological assessment of muscle reinnervation Reinnervation of the distal muscle targets was investigated by means of non-

invasive electrodiagnostic recordings 4, 9, and 12 weeks after nerve reconstruction

from the TA and PL muscles. Additionally, a final invasive-recording was performed

prior to tissue harvest in week 13.

First non-invasive recordings after 4 weeks did not show any evocable CMAPs. After

9 weeks, 8 out of 13 animals that had received ANGs displayed TA muscle

reinnervation, while this was only the case for single animals of the experimental

groups that had received chitosan conduits of different DAs for nerve reconstruction.

With a delay of additional three weeks, however, an increased number of animals of

conduit repair groups also demonstrated successful functional regeneration, which

was then confirmed with the results obtained during invasive recordings (Tab. 2). Measurements performed on the PL muscle yielded similar results, with the number

of chitosan conduit treated animals showing recordable CMAPs, reaching a level

comparable to that of ANG animals 13 weeks after implantation.

Table 2: Listing of animals per group, which demonstrated evocable CMAPs in the course of the study.

The shape and amplitude of the recorded CMAPs changed in the course of the

study. Four weeks following surgery, either no signals were detected or artifacts

were displayed when recordings were performed from the TA muscle (Fig. 20). These multiphasic artifacts could have been caused through interference from

neighboring muscles or they represent early reinnervation, which is characterized by

asynchronous firing of only a few motor units (Fugleholm et al., 2000; Navarro and

Udina, 2009; Rupp et al., 2007). Solely recordings with amplitudes clearly above

baseline were evaluated to exclude the chance of false positive results. During next

measurements, however, CMAPs were evident from both muscles, which were then

Results-Project 1

- 60 -

analyzed with regard to amplitude (indicating recruited axons / muscle fibers), AUC

(indicating AxL) and NCV.

Figure 20: Serial recordings of CMAPs. Recovery of CMAPs recorded from (A) TA and (B) PL muscles. With progressing reinnervation and a growing number of available muscle units as well as improving synchronization the amplitudes of the signal gradually increased. Note the changing scale for recordings from the lesioned and healthy side at the left margin.

Results-Project 1

- 61 -

For amplitude analysis the maximal amplitude of the M wave was measured and

results are expressed as a ratio (0 = no signal, no recruited axons; 1 = as many

recruited axons as on the non-lesioned side). In case of inclusion of “0” values

resulting from unsuccessful CMAP recordings from the TA muscle, an increase in

amplitude was seen for all experimental groups from week 9 (TA: ANG 0.18 ± 0.04;

DAI 0.05 ± 0.03; DAII 0.03 ± 0.02; DAIII 0.08 ± 0.03) to 12 (TA: ANG 0.36 ± 0.02;

DAI 0.29 ± 0.04; DAII 0.25 ± 0.05; DAIII 0.31 ± 0.04) that was statistically significant

(p< 0.05 for ANG; P< 0.001 for chitosan reconstructed groups). When disregarding

these values, on the other hand, the increase from week 9 to 12 is not significant

(Fig. 21 A). For the CMAPs obtained from the PL muscle no significant increase in

amplitude size could be found with or without inclusion of “0” values (Fig. 21 C). Additionally, no significant differences were seen between all experimental groups,

including the ANG group.

Results-Project 1

- 62 -

Figure 21: Recovery of the CMAP amplitude. Calculated amplitude-ratio from CMAPs recorded from (A, B) TA and (C, D) PL muscle in the course of (A, C) non-invasive recordings performed 9 as well as 12 weeks after conduit implantation and during (B, D) final invasive recordings. No significant differences between the experimental groups or between the results obtained during the different test intervals were found. Note that “0” values are not part of this scheme. Results were tested for significance (p < 0.05) using (A, C) Two- or (B, D) One-way ANOVA.

Invasive recordings performed directly before explantation, confirmed the

functionality of the regenerated nerve tissue and the amplitude-ratio was again

calculated. With regard to results obtained from TA muscle recordings the chitosan

conduit treated animals displayed amplitudes similar to the ANG group (Fig. 21 B), which had a size of about 30 % of the amplitude recorded from fully innervated

contralateral muscles. PL muscle CMAP amplitudes only reached around 20 % of

healthy control values when grafting was undertaken with chitosan conduits, while

the percentage was 39 % for ANGs (Fig. 21 D). This comparatively high percentage,

however, results from varying values within the ANG group, since one animal

Results-Project 1

- 63 -

responded with an almost complete amplitude recovery during stimulation and a

second animal even displayed a higher amplitude size on the lesioned than the

non-lesioned side. Excluding these two animals reduces the average amplitude-ratio

to 0.2 ± 0.04 (~20 %) and consequently to the same level seen with chitosan conduit

grafting.

Results obtained from evaluation of the AUC (AxL) of the CMAP are in line with the

findings of the amplitude analysis. A significant increase in axon numbers is seen for

ANG (P< 0.05), DAI (P< 0.001) and DAII animals (P< 0.01) in case of “100 %” value

inclusion (which indicates complete AxL) from week 9 (TA: ANG 84.48 ± 3.94; DAI

98.10 ± 1.90; DAII 95.76 ± 2.96) to 12 (TA: ANG 63.24 ± 2.34; DAI 69.76 ± 4.04;

DAII 72.59 ± 5.79). Due to a high SEM, however, this is not the case for the DAIII

group. Statistical analysis only of values from positive recordings, on the other hand,

did not reveal a significant increase in axon number over time for any of the groups

(Fig. 22 A). For PL muscle recordings, no significant change in AxL was seen over

time regardless of “100 %” value inclusion or exclusion (Fig. 22 C). Recordings from

direct nerve stimulation confirm the comparable functional recovery of all

experimental groups in terms of TA and PL muscle reinnervation (Fig. 22 B, D). Additional exclusion of the values from the two animals of the ANG group, which

showed a gain in axon number compared to the healthy control nerve when

recordings were undertaken from the PL muscle, shifts the average AxL in this group

to 56.84 ± 5.07, thus approximating the AxL seen with chitosan conduit treatment.

Differences between the experimental groups were not found, regardless of the

investigated time point.

Results-Project 1

- 64 -

Figure 22: Axon Loss. AxL in the sciatic nerve of the injured hind limb was calculated for (A, C) non-invasive and (C, D) invasive measurements from the (A, B) TA and (C, D) PL muscle. No significant differences between the experimental groups or different time intervals were seen. Note that “100 %” values are not part of this scheme. Results were tested for significance (p< 0.05) using (A, C) Two- or (B, D) One-way ANOVA.

The NCV was automatically calculated by the recording device based on the

different latencies obtained from proximal and distal stimulation of the sciatic nerve.

Recordings from the TA muscle revealed a significant increase of NCV from week 9

(TA: ANG 0.2 ± 0.05; DAI 0.02 ± 0.02; DAII 0.04 ± 0.03; DAIII 0.11 ± 0.04) to 12 (TA:

ANG 0.4 ± 0.03; DAI 0.3 ±0.03; DAII 0.26 ± 0.05; DAIII 0.31 ± 0.04) when “0” values

are taken into account for analysis. This, however, is not the case if these values are

excluded. As previously described for the amplitude and the AxL, no change in NCV

is seen for recordings from the PL muscle, in the course of non-invasive

measurements. Also, no differences among the groups were found during

non-invasive as well as invasive recordings for both recording sites (Fig. 23).

Results-Project 1

- 65 -

Figure 23: NCV-ratio. Calculation of the NCV-ratio with respect to values obtained from the contralateral healthy site. Recordings were performed from the (A, B) TA and (C, D) PL muscle. No significant differences between the groups or testing intervals were found. Results were tested for significance (p < 0.05) using (A, C) Two- or (B, D) One Way ANOVA. (Haastert-Talini et al., 2013)

On the final day of observation, we further analyzed the maximal CMAP with regard

to amplitude size, latency, and current intensity needed to elicit it (Fig. 24). The

amplitude data of the maximal CMAP mirror the results from the invasive recordings,

which were performed with a steady current of 1 mV and are described above.

Transduction of the signals to the distal target muscles was significantly slower for all

groups compared to the healthy contralateral side. Furthermore, it was a higher

current required to produce the maximal CMAP than on the right, non-lesioned side.

An exception represents the result obtained of the DAII group when recordings were

undertaken from the PL muscle. None of these investigated parameters, however,

Results-Project 1

- 66 -

revealed significant differences between the experimental groups including the ANG

group.

Figure 24: Analysis of the maximal CMAP. The maximal CMAP recorded during final invasive recordings from (A, C, E) TA and (B, D, F) PL muscle was analyzed with regard to its (A, B) amplitude, (C, D) latency and (E, F) required current intensity. Significant differences to the healthy contralateral side are marked by asterisk (* =p< 0.05; ** = p< =.01; *** = P < 0.001). Results were tested for significance using One-way ANOVA followed by Tukey’s multiple comparison.

Results-Project 1

- 67 -

4.4 Calculation of the lower limb muscle weight-ratio At the end of the observation period, the TA and gastrocnemius muscles were

harvested to assess muscle atrophy. The harvested tissue from the lesioned side

was macroscopically smaller compared to the non-lesioned side, but the extent of

visible atrophy differed for single animals (Fig. 25 A-C). Nevertheless, there were no

significant differences detected for either muscle between the experimental groups

including the ANG group (Fig. 25 D, C).

Figure 25: Muscle weight determination. Representative photographs of harvested muscle tissue from animals of the (A) ANG, (B) DAII, and (C) DAIII group. No statistical differences (P< 0.05) between the calculated muscle weight ratios could be found for the (D) TA or (E) gastrocnemius muscle using One-way ANOVA (P < 0.05). (Haastert-Talini et al., 2013)

4.6 Conduit properties upon explantation and tissue regeneration As part of the short- as well as long-term studies the regenerated nerve tissue was

cut out as a final step. In case of nerve reconstruction using chitosan conduits, this

Results-Project 1

- 68 -

procedure was conducted with the conduit still around the bridged distance. Five

days after conduit implantation, no macroscopic differences between individual

conduits were visible, while after 18 days, DAIII chitosan conduits already showed

signs of beginning degradation. In 2 out of 6 animals within this group small

longitudinal fissures were seen and another sample displayed a rough surface. The

degradation of DAIII conduits was further in progress after 13 weeks, when 4 out of

15 samples demonstrated longitudinal fissures, 8 out of 15 samples broke apart

upon nerve tissue harvest, and the other three samples were longitudinally broken

and compressed (Fig. 26).

Figure 26: Macroscopic properties of DAIII tubes. Representative photographs showing chitosan conduits of a high DA and regenerated nerve tissue within the bridging material. 13 weeks after surgery, the implanted conduits displayed fissures, crack lines and the tendency to break apart upon explantation (A-D). (A) As a result, the regenerated nerve tissue was partly secondary injured and appeared cleaved.

DAI and DAII conduits, on the other hand, featured a regular shape and firmness

(13/16 and 11/15 respectively) and only in few cases a slightly bent shape (3 within

each group) or a minor compression (DAII: 1/15) was observed, both observations

were accompanied by regular firmness of the sample (Haastert-Talini et al., 2013).

Results-Project 1

- 69 -

Regenerated nerve tissue (Fig. 27) was found in all grafted animals. Only 1 of 15

DAII animals displayed a thin cable (Tab. 3) and the regenerated tissue in 2 out of

16 DAIII animals was, most likely, secondary injured due the early degradation

process of these conduits.

Figure 27: Regenerated nerve tissue upon explantation. Representative photographs showing the regenerated nerve tissue upon explantation from animals of the (A) ANG, (B) DAI, (C) DAII, and (D) DAIII group.

Results-Project 1

- 70 -

Table 3: Overview of animal numbers showing regenerated nerve tissue 13 weeks after chitosan conduit implantation.

Results-Project 1

- 71 -

4.5 Inflammatory tissue response 4.5.1 Connective tissue A thin fibrous capsule had developed around the conduits following an implantation

time of 18 days and 13 weeks. Directly upon explantation, this tissue was analyzed

with regard to its optical density (Fig. 28) and a significant increase was seen for the

DAIII group.

Figure 28: Fibrous capsule optical density. 18 days and 13 weeks after tube implantation a fibrous capsule had formed around the bridging material, which was divided into (A) three levels of optical density. Statistical analysis revealed a significant higher optical density for tissue gathered from DAIII animals. Results were tested for significance using Chi-Square test, asterisk indicating differences between DAIII- tubes and others, p< 0.05 (Haastert-Talini et al., 2013).

Additionally, to analyze the FBR, the number of MGCs within this tissue was

determined. No significant differences between the groups were detected, but a

decrease of these cells was seen over time for all groups. After 13 weeks, only 2

samples from DAI animals featured single MGCs (Fig. 29). Furthermore, the

thickness as well as area of the tissue was measured, but no differences were seen

between the groups or investigation time points.

Results-Project 1

- 72 -

Figure 29: Fibrous capsule analysis. (A) Representative photomicrograph of a MGC. (B) No significant difference was found regarding the presence of MGCs in the connective tissue around the different conduits, but a decrease of this cell over time was observed. Significant analysis of results from measurements of the (C) thickness and (D) area of the tissue did not detect significant differences. Results were tested for significance (p< 0.05) using (C, D) Two- or (B) One-way ANOVA.

To further investigate the tissue / material response, ED1 / DAPI counterstained

sections were analyzed with regard to activated macrophages (Fig. 30). A significant

decrease in number of these phagocytic cells per mm2 was seen from 18 days (DAI

886.61 ± 41.87; DAII 760.00 ± 91.86; DAIII 1027.95 ± 101.95) to 13 weeks (DAI

153.35 ± 26.24; DAII 194.44 ± 32.74; DAIII 194.99 ± 35.83), indicating a reduction in

the inflammatory tissue response towards the implanted material. Although an

increased number of activated macrophages are present in the tissue around DAIII

Results-Project 1

- 73 -

tubes 18 days following nerve reconstruction, the differences between the groups

are not significantly different, independently from the investigated time point.

Figure 30: Presence of activated macrophages in the capsule surrounding the conduits. (A) Connective tissue formed around the implanted nerve guidance devices and was (B) cut longitudinally and folded with the areas formerly lying against the conduit facing each other. (C) Numbers of activated macrophages within the tissue. No significant differences were found between the experimental groups, but a decrease was seen over time. Results were tested for significance (p< 0.05) using Two-way ANOVA, Tukey’s multiple comparisons, triple asterisk indicating difference to 18 day values, p< 0.001. (D) Representative photographs of an ED1- immunopositive/DAPI counterstained section 18 days after conduit implantation and (E, F) activated macrophages in detail.

Furthermore, additional sections were investigated for elastic fibers (Orcein-staining),

and collagen (Trichrom-staining). After 18 days, a granulation tissue was seen with

many cells, but only a limited share of collagen and vessels. By 13 weeks, this

granulation tissue had been replaced by a vascular and collagenous fibrous capsule.

Thereby, collagen fibers were facing the conduit and the blood supply was located in

Results-Project 1

- 74 -

the layer fronting the surrounding tissue (Fig. 31, 32). Also, only few cells were seen

compared to 18 days and elastic fibers were confined to the blood vessel walls. The

portion of dense connective tissue, characterized by collagen, and loose connective

tissue, including the passing vessels, varied between samples, but certain patterns

could not be assigned to specific groups.

Results-Project 1

- 75 -

Figure 31: Collagen distribution within capsule. (A, D) 18 days after nerve reconstruction, granulation tissue had formed around the chitosan conduits, which did not feature vessels, only little collagen, but many cells. (B, C, E, F) After 13 weeks, however the collagen portion in the tissue increased (marked by light green color) as revealed by Trichrom-staining. Collagen fibers were mainly traveling in close proximity to the chitosan conduit. * = collagen. Scale bars: 200 µm, 100 µm, 200 µm, 20 µm¸20 µm, and 20 µm.

Results-Project 1

- 76 -

Figure 32: Vascularization of the tissue surrounding the conduit. (A, D) The granulation tissue that had formed 18 days after conduit implantation, did not display vessels, (B, C, E, F) while after 13 weeks, they were abundant in the loose tissue layer averted from the material. Orcein-staining revealed elastic fibers, which were confined to vessel walls. Scale bars: 200 µm, 100 µm, 100 µm, 20 µm¸20 µm, and 20 µm.

Results-Project 1

- 77 -

4.5.2 Nerve tissue The nerve tissue harvested in the final day of observation was analyzed with regard

to the FBR. Therefore, the presence of MGCs within the nerve tissue was

investigated (Fig. 33). After 18 days, an unstable matrix had formed inside the tube,

consequently only the proximal nerve stump without regenerated nerve tissue was

prepared for histological analysis (Fig 33 A, B). Only few MGCs were found in the

tissue and no differences between the groups were detected (Fig. 33 C). In line with

the conduit properties observed for the DAIII group upon explantation scattered

chitosan inclusions were discovered in the sections of two out of three animals of this

group (Fig. 33 D, E, G, H), while this was not the case for the DAI group, and only

one single chitosan piece was seen in the periphery of an explanted sample from a

DAII animal (Fig 33 I).

Results-Project 1

- 78 -

Figure 33: Number of MGCs and presence of chitosan inclusions 18 days after tube implantation within the regenerated nerve tissue. The proximal nerve stump of explanted samples was subjected to histological analysis with regard to the presence of MGCs. (C) No statistical differences in the number of MGCs between the experimental groups were revealed 18 days after surgery. Results were tested for significance using One-way ANOVA (p< 0.05). Scattered chitosan inclusions were found in two samples of the (D, E, G, H) DAIII groups, while only one single piece was present in the tissue surrounding a DAII conduit. Abbreviations: C= chitosan; MGC= multinucleated giant cell.

After 13 weeks, regenerated nerve tissue was present inside the tube and sections

were prepared starting 1 mm behind the distal suture (Fig. 34 A, B). In line with

results from the 18 day short-term study no significant differences between the

Results-Project 1

- 79 -

experimental groups were seen with regard to the number of MGCs within the tissue

(Fig. 34 C). The high SEM for the DAII conduit group is the result of one animal

displaying a high number of these cells (7 MGCs / mm2), while 3 out of 5 did not

show any MGCs, and samples of one animal featured only 1 MGC per mm2.

Figure 34: Presence of MGCs in the regenerated nerve tissue 13 weeks after surgery. (A, B) the proximal nerve stump of explanted samples was subjected to histological analysis with regard to the presence of MGCs. (C) No statistical differences in the number of MGCs between the experimental groups were found. Results were tested for significance using One-way ANOVA (p< 0.05).

Scattered chitosan inclusions, however, were again found in samples obtained from

DAIII animals (in 5/5 representative animals subjected to the analysis)

(Fig. 35 G, H). This observation was not made for the other two groups.

Results-Project 1

- 80 -

Figure 35: Scattered chitosan inclusions in the nerve tissue of the DAIII group. (A-I) Representative photomicrographs of sections obtained from the regenerated nerve tissue of the (A, B) ANG group and newly grown tissue found inside (C, D) DAI, (E, F) DAII, or (G, H) DAIII conduits. Sections obtained from DAIII animals displayed chitosan inclusions. Furthermore, the thickness of the connective tissue surrounding the axons was increased compared to the other groups. Results regarding the latter were obtained by Sandra Wrobel (M.Sc., MHH Neuroanatomy) and can be looked up in (Haastert-Talini et al., 2013). Abbreviation: C= chitosan.

Consecutive sections from the 18 day study were counterstained for ED1

immunopositive cells and DAPI to assess the number of activated macrophages in

Results-Project 1

- 81 -

the regenerated nerve tissue (Fig 36). Between the experimental groups no

significant differences were found.

Figure 36: Activated macrophages in the proximal nerve stump after 18 days. (A) Sections were prepared form the proximal nerve stump and staining was performed for (B) activated macrophages. (C) Between the grafted groups no significant difference in the number of these cells was found. Results were tested for significance using One Way ANOVA, not including the ANG group (P < 0.05).

Also, sections of regenerated nerve tissue obtained 13 weeks after surgery, were

investigated for activated macrophages. This analysis was performed by Sandra

Wrobel (M.Sc., MHH Neuroanatomy) and can be looked up in (Haastert-Talini et al.,

2013).

- 82 -

Results-Project 2

- 83 -

5 Results-Project 2 5.2 Mechanical pain threshold assessment The second study was performed with the aim to address bridging of long gaps by

introducing luminal fillers into the conduit. Therefore, chitosan conduits were

enriched with NVR-Gel containing genetically modified SCs over-expressing either

GDNF or FGF-218kDa. Investigation parameters were broadened compared to project

1, thus sensory recovery was assessed through determination of the nociceptive

pain threshold in the lateral and medial side of the hind paws. One and 4 weeks

following nerve reconstruction, no withdrawal responses were seen when the lateral

side was stimulated. Five weeks later, all animals of the ANG group showed some

recovery (12.64 ± 1.26 g), but the pressure required to elicit a response was

significantly decreased compared to the values obtained from stimulation of the

contralateral side (20.16 ± 0.30 g) (Fig. 37). An observation, which has, amongst

others, already been described in previous studies by our partner laboratory at the

autonomous University of Barcelona (UAB) (Cobianchi et al., 2014, 2013). On the

other hand, only single animals of the nt-neoSCs (18.93 g), tf-neoSCs-GDNF (20.93

g), and tf-neoSCs-FGF-218kDa (15.60 g) groups demonstrated signs of reinnervation.

In the following weeks, a higher pain threshold was measured for animals of the

ANG group (week 12: 21.44 ± 0.54; week 16: 23.10 ± 0.62) that did not differ

significantly from healthy control values of the contralateral side (week 12: 15.15 ±

1.59; week 16: 17.76 ± 1.28). Until the end of the observation period, only one more

animals of the tf-neoSCs-FGF-218kDa group responded to stimulation of the lateral

side of the paw, while most animals of the groups that had received enriched

chitosan conduits for nerve repair did not show withdrawal responses.

Results-Project 2

- 84 -

Stim

ulat

ion

of t

he l

ater

al t

errit

ory

reve

aled

alm

ost

sole

rei

nner

vatio

n in

the

AN

G g

roup

, w

hile

onl

y si

ngle

ani

mal

s of

the

ex

perim

enta

l gr

oups

dem

onst

rate

d a

with

draw

al r

espo

nse.

Gre

en=

tissu

e re

gene

ratio

n se

en u

pon

expl

anta

tion;

Pur

ple=

thi

n tis

sue

rege

nera

tion

seen

upo

n ex

plan

tatio

n; R

ed=

no t

issu

e re

gene

ratio

n se

en u

pon

expl

anta

tion.

For

sta

tistic

al a

naly

sis

all

anim

als

wer

e in

clud

ed a

nd in

cas

e of

no

resp

onse

cut

off

forc

e va

lues

(40

g) w

ere

set.

Res

ults

wer

e te

sted

for

sign

ifica

nce

(p<

0.05

) us

ing

Two-

way

AN

OVA

, Tuk

ey’s

mul

tiple

com

paris

ons.

* p

< 0.

05; *

** p

< 0.

001

vs v

alue

s ob

tain

ed fr

om th

e co

ntra

late

ral

side

; # p

< 0.

05, #

# p<

0.0

1, #

## p

< 0.

001

vs 7

day

s an

d 4

wee

ks, ‡

p<

0.05

, ‡‡

p< 0

.01,

‡‡‡

p<

0.00

1 vs

fello

w e

xper

imen

tal

grou

ps a

t the

sam

e tim

e po

int, ₸

p<0.

05 v

s tf-

neoS

Cs-

EV

gro

up a

t the

sam

e tim

e po

int.

Fi

gure

37:

Mec

hani

cal p

ain

thre

shol

d (la

tera

l).

Results-Project 2

- 85 -

Figu

re 3

8: M

echa

nica

l pai

n th

resh

old

(med

ial).

W

ithdr

awal

res

pons

es w

ere

seen

for

all a

nim

als

whe

n st

imul

atio

n w

as p

erfo

rmed

on

the

med

ial p

aw s

ide,

but

the

forc

e ne

eded

to

elic

it th

e re

actio

n w

as s

igni

fican

tly r

educ

ed c

ompa

red

to c

ontro

l val

ues

durin

g ev

ery

test

ing

inte

rval

. G

reen

= tis

sue

rege

nera

tion

seen

upo

n ex

plan

tatio

n; P

urpl

e= t

hin

tissu

e re

gene

ratio

n se

en u

pon

expl

anta

tion;

Red

= no

tis

sue

rege

nera

tion

seen

upo

n ex

plan

tatio

n. F

or s

tatis

tical

ana

lysi

s al

l ani

mal

s w

ere

incl

uded

and

in c

ase

of n

o si

gnal

det

ectio

n cu

t off

forc

e va

lues

(40g

) wer

e se

t. R

esul

ts w

ere

test

ed f

or s

igni

fican

ce (

p <

0.05

) us

ing

Two-

way

AN

OVA

, Tu

key’

s m

ultip

le c

ompa

rison

s.**

* p

< 0.

001

vs v

alue

s ob

tain

ed fr

om th

e co

ntra

late

ral s

ide.

Results-Project 2

- 86 -

The pain threshold measured in response to stimulation of the medial paw side, a

territory that is also partly innervated by the saphenous nerve, was significantly

decreased following nerve repair (ANG: 11.98 ± 0.60 g; nt-neoSCs 11.65 ± 1.14 g; tf-

neoSCs-EV 12.86 ± 0.85 g; tf-neoSCs-GDNF 13.27 ± 1.45 g; tf-neoSCs-FGF218kDa

12.35 ± 1.11 g) compared to values obtained from the contralateral side (18.54 ±

0.42) (Fig. 38). For the complete duration of the study, however, the animals

continued to show signs of hyperalgesia on the lesioned side and the measured pain

threshold remained significantly decreased. ₸

5.1 Static Sciatic Index (SSI) Following nerve reconstruction, using enriched chitosan conduits or ANGs, motor

recovery was analyzed at various time points via SSI as seen for project 1.

Pre-surgery values confirmed that there was no paw position impairment before

implantation (ANG -1.90 ± 2.93; nt-neoSCs -3.15 ± 2.83; tf-neoSCs-EV -0.64 ± 3.11;

tf-neoSCs-GDNF -2.25 ± 2.30; tf-neoSCs-FGF-218kDa -3.73 ± 2.50) (Fig. 39). Shortly

after lesion and subsequent reconstruction, however, the SSI dropped to around

-100 % (ANG -95.69 ± 1.97; nt-neoSCs -96.91 ± 2.32; tf-neoSCs-EV -92.70 ± 1.45;

tf-neoSCs-GDNF -100.52 ± 3.64; tf-neoSCs-FGF-218kDa -94.37 ± 1.58) due to the

missing innervation from the sciatic nerve branches.

Results-Project 2

- 87 -

Figure 39: Results of the calculation of the Static Sciatic Index (SSI). Evaluation of distal target reinnervation using SSI did not reveal any significant difference between animals that were treated with ANGs, and animals that received enriched chitosan conduits for nerve reconstruction. Results were tested for significance (p < 0.05) using Two-way ANOVA, Tukey’s multiple comparisons, ***= difference to post nerve lesion values, p < 0.001, and #= difference to 4 weeks post lesion values of tf-neoSCs-EV, p < 0.05.

Four month later, there was no significant improvement regarding distal muscle

target reinnervation assessed by SSI for all experimental groups when compared to

values obtained 1 week after lesion. For the ANG group the SSI mean value was still

-83.64 ± 11.03, for nt-neoSCs -80.08 ± 5.69, for tf-neoSCs-EV -74.40 ± 2.71, for

tf-neoSCs-GDNF -80.32 ± 5.01 and for tf-neoSCs-FGF218kDa -92.03 ± 5.02.

5.3 Electrophysiological assessment of muscle reinnervation Motor recovery was further evaluated by means of non-invasive nerve conduction

tests 9, 13, and, 17 weeks after sciatic nerve lesion and subsequent repair, as well

as by additional final invasive recordings. Measurements performed at the earliest

time point revealed reinnervation of the TA and PL muscle for almost all animals of

the ANG group, while this was only seen for single animals of the nt-neoSCs and

tf-neoSCs-EV group. Four weeks later, all animals of the ANG group showed

evocable CMAPs, while the signal detected for the animal of the tf-neoSCs-EV group

after 9 weeks was not confirmed. Instead one animal of the tf-neoSCs-FGF-218kDa

group showed signs of reinnervation. On the last day of observation, however,

Results-Project 2

- 88 -

signals were recorded from 4 out of 7 animals of the tf-neoSCs-FGF-218kDa group

when recordings were undertaken from the TA muscle (Tab. 4). Table 4: Listing of animals per group, which demonstrated evocable CMAPs in the course of the study.

These findings were confirmed during final invasive recordings. One of these

animals further displayed PL muscle reinnervation. Exemplary CMAPs recorded from

the TA and PL muscle during the different recording sessions are depicted in Fig. 40.

Results-Project 2

- 89 -

Figure 40: Serial recordings of CMAPs. Recovery of CMAPs recorded from (A) TA and (B) PL muscle following nerve repair of a 15 mm gap with either an ANG or enriched chitosan conduit containing FGF-218kDa over-expressing SCs. Note the changing scale for recordings from the lesioned and healthy side at the left margin.

Results-Project 2

- 90 -

The data gained from the recordings were used to calculate the NCV, AUC (AxL),

and amplitude-ratio, with the latter two providing information with regard to axon loss

or recruited axons, respectively.

At the end of the observation period, solely tf-neoSCs-FGF-218kDa treatment allowed

TA muscle reinnervation that was comparable to results seen following ANG

treatment (Fig. 41). The latter, however, enabled recovery of the amplitude size to

almost 50 % of contralateral control values, while the FGF-218kDa group reached

13 %. Regarding PL muscle reinnervation ANG treatment was superior to all groups,

since only one animal of the tf-neoSCs-FGF-218kDa group showed CMAPs during the

final invasive recording (Fig. 42). Also, single animals of the nt-neoSCs and

tf-neoSCs-GDNF groups displayed PL muscle CMAPS during non-invasive

recordings, but these signals were not stable and could not be confirmed during

invasive recordings.

Results-Project 2

- 91 -

Non

-inva

sive

reco

rdin

gs fr

om T

A m

uscl

es re

veal

ed s

igni

fican

t diff

eren

ces

betw

een

the

AN

G g

roup

and

all

grou

ps th

at h

ad

been

tre

ated

with

eith

er o

f th

e en

riche

d ch

itosa

n co

ndui

ts.

Add

ition

ally

, so

lely

tre

atm

ent

with

the

gol

d st

anda

rd le

d to

a

sign

ifica

nt im

prov

emen

t of m

uscl

e re

inne

rvat

ion

over

tim

e (T

wo-

way

AN

OV

A fo

llow

ed b

y Tu

key’

s m

ultip

le c

ompa

rison

, # <

0.

05 v

s A

NG

13

and

17 w

eeks

; *

p< 0

.05

vs A

NG

gro

up o

f th

e sa

me

wee

k).

Fina

l inv

asiv

e re

cord

ings

17

wee

ks a

fter

surg

ery,

how

ever

, re

veal

ed s

igni

fican

t di

ffere

nces

bet

wee

n th

e pe

rform

ance

of

the

ANG

gro

up a

nd t

he a

nim

als

treat

ed

with

the

diff

eren

t va

riant

s of

enr

iche

d co

ndui

ts,

exce

pt f

or t

he g

roup

tha

t ha

d re

ceiv

ed t

f-neo

SC

s-FG

F-218

kDa t

reat

men

t (K

rusk

al-W

allis

test

follo

wed

by

Dun

n’s

mul

tiple

com

paris

on, *

p<

0.05

vs

AN

G g

roup

of t

he s

ame

wee

k. G

reen

= tis

sue

rege

nera

tion

seen

up

on

expl

anta

tion;

P

urpl

e=

thin

tis

sue

rege

nera

tion

seen

up

on

expl

anta

tion;

R

ed=

no

tissu

e re

gene

ratio

n se

en u

pon

expl

anta

tion.

For

sta

tistic

al a

naly

ses

all a

nim

als

wer

e in

clud

ed a

nd in

cas

e of

no

sign

al d

etec

tion

“0” v

alue

s w

ere

set.

Figu

re 4

1: A

mpl

itude

-rat

io fr

om C

MA

Ps re

cord

ed fr

om th

e tib

ialis

ant

erio

r (TA

) mus

cle.

Results-Project 2

- 92 -

PL

mus

cle

rein

nerv

atio

n fo

llow

ing

treat

men

t w

ith A

NG

s w

as s

uper

ior

to a

ll ot

her

treat

men

t st

rate

gies

at

ever

y re

cord

ing

time

poin

t. (N

on-in

vasi

ve: T

wo-

way

AN

OV

A fo

llow

ed b

y Tu

key’

s m

ultip

le c

ompa

rison

, # <

0.0

5 vs

AN

G 1

3 an

d 17

wee

ks; #

# <

0.05

vs

AN

G 1

3 an

d 17

wee

ks, *

p<

0.05

vs

ANG

gro

up o

f the

sam

e w

eek;

Inva

sive

: Kru

skal

-W

allis

test

follo

wed

by

Dun

n’s

mul

tiple

com

paris

on, *

< 0

.05

vs A

NG

gro

up o

f th

e sa

me

wee

k). (

B)

Gre

en=

tissu

e re

gene

ratio

n se

en u

pon

expl

anta

tion;

Pur

ple=

thi

n tis

sue

rege

nera

tion

seen

upo

n ex

plan

tatio

n; R

ed=

no t

issu

e re

gene

ratio

n se

en u

pon

expl

anta

tion.

For

sta

tistic

al a

naly

ses

all

anim

als

wer

e in

clud

ed a

nd i

n ca

se o

f no

sig

nal

dete

ctio

n “0

” val

ues

wer

e se

t.

Figu

re 4

2: A

mpl

itude

-rat

io fr

om C

MA

Ps re

cord

ed fr

om th

e pl

anta

r (PL

) mus

cle.

Results-Project 2

- 93 -

Results of NCV- and AxL-calculation support the findings of the amplitude analysis.

During non-invasive TA muscle recordings, animals of the ANG group showed

significant faster NCVs and more available axons compared to the other

experimental groups (Tab. 5). In the course of invasive tests, however, no significant

differences were found between the NVC and AxL calculated for ANG and tf-

neoSCs-FGF-218kDa treated animals. An increase in NCV over time was seen solely

for the ANG group when recordings were undertaken from the PL muscle.

Furthermore, a significant decrease of AxL over time was seen for the ANG (TA and

PL) and FGF-218kDa (TA) groups.

Results-Project 2

- 94 -

Ta

ble

5: N

CV,

AxL

, and

am

plitu

de-r

atio

Te

stin

g fo

r sig

nific

ance

(p<

0.05

) was

per

form

ed u

sing

Kru

skal

-Wal

lis te

st o

r Tw

o-w

ay A

NO

VA, f

ollo

wed

by

Tuke

y’s

or D

unn’

s m

ultip

le c

ompa

rison

. *p

< 0.

05 v

s fe

llow

exp

erim

enta

l gro

ups,

**p

< 0.

01 v

s fe

llow

exp

erim

enta

l gro

ups,

***

p< 0

.001

vs

fello

w e

xper

imen

tal g

roup

s, #

# p<

0.0

1 vs

AN

G 1

7 w

eeks

, ##

# p<

0.0

01 v

s A

NG

17

wee

ks,

§ p<

0.0

5 vs

AN

G 1

3 w

eeks

, §§

p<

0.01

vs

ANG

13

wee

ks,

$$ p

<0.0

1 vs

nt-n

eoS

Cs,

tf-n

eoS

Cs-

EV

, tf-

neoS

Cs-

GD

NF,

$$$

p<

0.00

1 vs

nt-n

eoSC

s, tf

-neo

SC

s-EV

, tf-n

eoSC

s-G

DN

F

Results-Project 2

- 95 -

5.4 Calculation of the lower limb muscle weight-ratio The muscle weight-ratio was calculated subsequently to harvest of the TA and

gastrocnemius muscles. Solely nerve reconstruction using enriched chitosan

conduits featuring SCs that over-expressed FGF-218kDa led to a muscle weight that

did not differ significantly from that seen after ANG treatment although an optical

difference was still visible (Fig. 43).

Figure 43: Muscle weight-ratio. The TA as well as gastrocnemius muscle weights were determined upon explantation of the nerve tissue specimens. Representative photomicrographs showing cut muscles from animals that demonstrated successful tissue regeneration: (A) ANG group, (B) FGF-218kDa over-expressing group. (C) Harvested muscles from an animal without tissue regeneration that had been treated with a conduit containing non-transfected SCs. (D, E) Significant differences were found comparing the ANG group with all experimental groups, except for the group that had received conduits with SCs over-expressing FGF-218kDa, although an optical difference was seen (Kruskal-Wallis test, followed by Dunn’s multiple comparison, *< 0.05 vs ANG group). Green= tissue regeneration seen upon explantation; Purple= thin tissue regeneration seen upon explantation; Red= no tissue regeneration seen upon explantation.

5.3 Autotomy score The degree of automutilation was regularly assessed using an autotomy score.

Predominantly animals that had received enriched chitosan conduits for peripheral

Results-Project 2

- 96 -

nerve reconstruction displayed automutilation, while this was only case for one

animal of the ANG group. Thereby, animals that showed tissue regeneration as well

as animal with failed regeneration after 17 weeks were affected. Animals of the

tf-neoSCs-GDNF group, however, demonstrated highest ratings on the score

(Fig. 44).

Figure 44: Autotomy Score. Animals predominantly displayed no automutilation or mild toe nail biting (1 point), but some cases of higher degrees of automutilation were also seen. For the duration of the experiment, animals of the tf-neoSCs-GDNF group demonstrated the highest score. Green= tissue regeneration seen upon explantation; Purple= thin tissue regeneration seen upon explantation; Red= no tissue regeneration seen upon explantation. Results were tested for significance (p< 0.05) using Two Way ANOVA.

5.5 Macroscopic analysis of the explanted nerve tissue Following final functional tests, regenerated nerve tissue was collected if regeneration had occurred. All animals of the ANG group showed regeneration (Tab. 6, Fig. 45A), while this was only the case for single animals of the experimental groups. Solely the tf-neoSCs-FGF-218kDa treatment group stood out within the conduit groups with 4 out of 7 animals showing clear tissue regeneration (Fig. 45B), and the remaining animals of the group displaying thin tissue regeneration. Table 6: Tissue regeneration seen upon explantation.

Results-Project 2

- 97 -

Figure 45: Representative photographs of explanted specimens following a 4 month observation period. (A) Successful tissue regeneration following ANG treatment. (B) Tissue regeneration through a chitosan conduit filled with NVR-Gel and SCs overexpressing FGF-218kDa. (C) Thin tissue regeneration (arrows) through a chitosan conduit filled with NVR-Gel and empty vector transfected SCs. (D) Failed regeneration through a chitosan conduit filled with NVR-Gel and SCs over-expressing GDNF.

5.6 Inflammatory tissue response 5.6.1 Connective tissue The fibrous capsule that had developed around the conduits was harvested and

analyzed with regard to MGCs, its thickness, and area. Only single MGCs were

found in one out of four animals of the tf-neoSCs-EV group and in two out of six

animals of the tf-neoSCs-FGF-218kDa group. Furthermore, no significant differences

were seen in the thickness of the connective tissue as well as its area when

comparing samples collected from the different experimental groups (Fig. 46).

Results-Project 2

- 98 -

Figure 46: Fibrous capsule analysis. Connective tissue that had developed around the chitosan tubes was analyzed with regard to (A, B) presence of MGCs (arrows), its (C) thickness, and (D) area. No significant differences between the experimental groups were detected (One way ANOVA; p < 0.05).

To further analyze the inflammatory response staining for activated macrophages

(ED1+) was performed on consecutive sections, but counting of these cells also did

not reveal significant differences between the groups (Fig. 47).

Results-Project 2

- 99 -

Figure 47: Presence of activated macrophages in the capsule surrounding the conduits. (A) Sections of connective tissue surrounding the enriched chitosan conduits were stained for activated macrophages. Some vessels also appear in red as indicated by arrows. (B) Detail of activated macrophages. (C) No significant differences were found between the experimental groups. Results were tested for significance (p< 0.05) using Kruskal-Wallis test.

Further sections were examined with regard to collagen formation and elastic fiber

presence. As seen with hollow chitosan conduits (project 1), collagen formed directly

at the material / tissue interface, while vascularization of the tissue occurred in the

outer, more loose layers of the capsule. The samples displayed variations regarding

the content of collagen and loose connective tissue (FIG. 48, 49), but these

observations could not be assigned to specific groups. Also, single samples of the

experimental groups (nf-neoSCs, tf-neoSCs-EV, and tf-neoSCs-GDNF each one

sample) were characterized by the presence of numerous cells in the direct contact

area with the material as depicted in Fig. 49D and H.

Results-Project 2

- 100 -

Figure 48: Collagen distribution within capsule. (A-C) Collagen bands (light green) are seen in the direct contact area with the chitosan as revealed by TriChrom-staining. (E-G) Details of the contact areas. (D) Representative overview photomicrograph of a sample displaying numerous cells compared to other capsules. (H) Detail of the same sample, which also featured a high number of activated macrophages when stained for ED1.

Results-Project 2

- 101 -

Figure 49: Vascularization of the capsule surrounding the conduit. (A-D) Orcein-staining revealed the presence of elastic fibers in connection with vessels in the outer layers of the connective tissue that had formed around the implant. (E-H) Details of vessels showing elastic fibers as part of the vessel wall.

5.6.2 Regenerated tissue Samples of successful tissue regeneration were subjected to HE- and ED1-staining

and analyzed towards the presence of MGCs and activated macrophages,

respectively. Only single MGCs were detected in the samples, except for one animal

of the FGF-218kDa group. In this case a higher number of these cells were found in

two out of four analyzed sections, while no MGCs were seen in the other two

sections. The same animal also displayed the highest number of macrophages when

compared to other samples of successful regeneration (Tab. 7).

Results-Project 2

- 102 -

Table 7: Quantity of MGCs and activated macrophages within the (regenerated) nerve tissue. *= cases of failed regeneration where the proximal nerve stump was analyzed. n.a. = not analyzed.

Presence of axons that completed regeneration across the 15 mm gap was

confirmed via NF200 staining and can exemplary be seen in Fig. 50, while Fig. 51

shows cases of failed regeneration. Additionally, a representative cross-section of

regenerated nerve tissue inside a chitosan conduit is displayed in Fig. 52.

Results-Project 2

- 103 -

Figure 50: Regeneration across the gap. (A) Schematic illustration of the approximate localization of the stained sections from regenerated samples obtained from animals of the (C, D) ANG and (E, F) FGF-218kDa group. (C, E) Sections within the bridged distance (orange box) and (D, F) close to the distal suture site (pink box). (B) Representative staining of a healthy control nerve.

Results-Project 2

- 104 -

Figure 51: Failed regeneration. Schematic illustration of the approximate localization of the stained sections from samples obtained from animals of the (B, C) nt-neoSCs and (D, E) tf-neoSCs-GDNF groups. (C, E) Sections close to the suture site (pink box) and (B, D) 1.5 mm more proximal (orange box) were stained for activated macrophages and neurofilament. (C, F,). No or only very few axons were seen directly at the entering site into the chitosan conduits, however, sections of a more proximal region stained positive for neurofilament.

Results-Project 2

- 105 -

Figure 52: Cross-section of an enriched chitosan conduit and regenerated nerve tissue 17 weeks following implantation. Representative photomicrograph of a cross-section obtained 17 weeks after surgery from an FGF-218kDa animal, which also demonstrated functional recovery by means of electrophysiological recordings.

- 106 -

Discussion

- 107 -

6 Discussion PNIs have been a challenging research field in the last century and represent a

global clinical problem. Although, there has been great progress in development of

nerve guidance conduits based on synthetic and non-synthetic materials, the

functional outcome for patients is often unsatisfactory. Consequently, the ANG

remains the clinical standard therapy. Despite refinement of this technique, however,

bridging nerve defects with dispensable sensory nerves does also not guarantee a

full recovery. Furthermore, this method is associated with disadvantages like a

limited availability of autologous nerve tissue and mismatch issues. Consequently,

alternatives have to be created for the currently available bridging strategies that

allow improvement of peripheral nerve regeneration across small gaps, but also

address critical gaps.

As part of this thesis, we tested hollow chitosan conduits of three different DAs

(DAI= 1%, DAII= 5%, DAIII= 20%) for peripheral nerve repair of small gaps

(Project 1). Therefore, we introduced a 10 mm defect into the sciatic nerve of rats

and evaluated the performance of the chitosan conduits with regard to functional

outcome and foreign body response elicited through the implantation of the material.

Both issues were addressed in a long-term study (13 weeks) with ANGs serving as

positive control. Secondary, short-term-studies were conducted to look at the

regulation of regeneration associated proteins as well as at the tissue response

shortly upon nerve lesion in the presence of chitosan (5 days and 18 days).

In addition, as a second project of my thesis, we focused on NVR-Gel as a matrix

inside the tube, which contained genetically modified SCs over-expressing GDNF or

FGF-218kDa to improve peripheral nerve regeneration across critical defects

(Project 2). Therefore, a 15 mm gap was bridged with enriched chitosan conduits

and the functional recovery and device / tissue reaction was analyzed in a long-term

study of 17 weeks duration.

Discussion-Project 1

- 108 -

6.1 Project 1 6.1.1 Regulation of regeneration associated proteins following nerve lesion and repair In response to nerve damage, various NTFs and cytokines as well as their receptors

are subjected to putative expression changes in a time and spatial manner. Neurons

and SCs up-regulate these survival and axonal outgrowth promoting factors and

thereby provide the basis for successful nerve regeneration (Boyd and Gordon,

2003a; Gordon, 2009). It is therefore of importance that the physiological regulation

is not disturbed as a result of material implantation. Proximal and distal nerve

stumps, the lumbar region of the spinal cord as well as DRGs (L3-L5) were collected

5 and 18 days after nerve lesion and repair with chitosan conduits. Using qRT-PCR,

the expression level of NTFs (NGF, FGF-2, BDNF), the cytokine IL-6, the BDNF-

receptor TrkB, and GAP-43 were determined and normalized to values obtained

from the contralateral side of control animals. Analysis of the nerve stumps,

however, was not successful, most likely because the distance (5 mm) that was

harvested was insufficient.

NGF as well as IL-6 expression was below detection limit in the spinal cord. The

former is in accordance with reports that NGF is neither expressed by motoneurons

in a detectable amount nor up-regulated upon injury (Boyd and Gordon, 2003a;

Josephson et al., 2001). Another study, on the other hand, has demonstrated NGF

expression in ventral spinal cords, but no regulation in any direction was revealed 1

or 2 days after injury (Cobianchi et al., 2013). IL-6 transcripts have previously been

reported to be immediately up-regulated by axotomized motoneurons, peaking at

24 h before returning to undetectable baseline expression within 21 days after nerve

injury (Kiefer et al., 1993). Also, an increase of IL-6 mRNA levels has been reported

by SCs of ventral roots after 4 and 7 days using an in vitro model that mimics

degeneration (Allodi et al., 2013). BDNF, FGF-2, GAP-43, and trkB were all neither

significantly up- nor down-regulated in the spinal cord and also no differences

between the experimental groups including the control group were seen. Previous

results regarding BDNF expression in the ventral horn have revealed that there is no

up-regulation 1 or 3 days following nerve injury (Cobianchi et al., 2013). But also an

up-regulation in the spinal cord 2 days after injury has been reported before

Discussion-Project 1

- 109 -

subsequent down-regulation to below baseline levels by 2 weeks. Additionally,

GAP-43 has already been shown to not be significantly up- or down-regulated

(Haastert-Talini et al., 2011). Allodi et al. (2013) further described an up-regulation of

FGF-2 by SCs of ventral roots in the same study set-up as noted for IL-6.

Results addressing expression levels of the same factors in DRGs following nerve

repair with chitosan conduits were obtained by Dr. Ratzka (MHH, Institute of

Neuroanatomy) (Haastert-Talini et al., 2013). The data show that regardless of the

time point (5 days or 18 days) NGF, FGF-2, IL-6, BDNF. and GAP-43 were all

up-regulated in the control group as well as in the conduit treated groups, while no

change in expression was seen for TrkB. BDNF has been previously described to be

up-regulated 1 to 3 days after injury, before returning to baseline levels by 14 days

(Cobianchi et al., 2013; Haastert-Talini et al., 2011). An in vitro model mimicking

degeneration after injury has revealed IL-6 up-regulation in dorsal roots to a similar

extent as seen in this study after 4 days before decreasing again following a 7 days

culture period (Allodi et al., 2013). An increase of GAP-43 transcripts has already

been described 2 days after injury, while baseline levels were reported after 2 weeks

and only in case of electrical stimulation did the levels remain significantly higher

(Haastert-Talini et al., 2011). Furthermore, up-regulation of mRNA levels of NGF and

FGF-2 in DRGs have already been stated (Cobianchi et al., 2013; Grothe et al.,

1996; Meisinger and Grothe, 2002).

Since the regulation patterns of the analyzed growth factors are comparable in

conduit treated animals as well as control animals, it can be concluded that chitosan

does not affect these molecular processes.

6.1.2 Matrix formation and brain-derived neurotrophic factor (BDNF) presence in the tube content Following an implantation time of 18 days, a matrix formation was seen in our

short-term study, which was not further analyzed, since we collected the entire tube

content for Western Blot analysis. Our partner laboratory at the Lund University

(ULUND), however, examined the matrix in a separate study (see (Haastert-Talini et

al., 2013)). They evaluated the formed matrix after 21 days with regard to its

thickness, axonal outgrowth, and presence of SCs. The latter were seen within the

Discussion-Project 1

- 110 -

matrix along the complete conduits of different DAs, but DAI as well as DAII tubes

contained less apoptotic SCs in the distal region than DAIII conduits. Analyses of

matrix thickness and axonal outgrowth did not show significant differences between

the tubes, but in all experimental groups axons had almost completed the growth

across the 10 mm defect. This kind of matrix formation has previously been

described as an important phase on the way to successful regeneration. One week

after conduit implantation, formation of a fibrin cable has been reported, along which,

amongst others, SCs begin to migrate within the second week from the proximal and

distal nerve stumps, thereby aligning to form the bands of Büngner. Subsequently,

axonal sprouts enter the conduit 7-14 days after repair and use the matrix as a

guidance structure while crossing the defect. The fibrin cable itself, however, is

believed to vanish after fulfilling its duty during cell migration (Belkas et al., 2004a;

Dahlin et al., 1995; Daly et al., 2012; Lundborg et al., 1981; Madison et al., 1987;

Seckel, 1990; Williams et al., 1983).

From the matrix formation seen regardless of the DA it can be concluded that the

early regenerative processes described in hollow conduits are allowed in the

presence of chitosan as well. The reason for the higher number of apoptotic SCs in

the presence of chitosan of a high DA (about 20 %) is not clear. A good

biocompatibility of chitosan membranes and SCs has been shown by our laboratory

together with a partner laboratory (UMINHO, Portugal) in vitro, however, the highest

DA tested was 5 % (Wrobel et al., 2014). Other studies undertaking SC seeding on

chitosan membranes and fibers could further demonstrate viability and migration if

the DA was around 8 % (Yang et al., 2009; Yuan et al., 2004). Interestingly, Wenling

et al. have reported that chitosan films with a DA lower than 10 % are a more

suitable substrate for spreading and proliferation of SCs (Wenling et al., 2005). While

a DA of 20 % resulted in a spherical shape of the SCs, they displayed a polar

morphology on lower DAs of e.g. 5 %. Additionally, more SCs attached on films of a

low DA. This is in agreement with observations that cells in general prefer hydrophilic

surfaces with positively charged amine groups, which are properties that both

account for chitosan with low DAs, while chitin is highly hydrophobic and displays

less amine groups (Aranaz et al., 2009; Dutta et al., 2004; Ma et al., 2002; Tomihata

and Ikada, 1997). The cause for the change in morphology, however, is not known

Discussion-Project 1

- 111 -

as well as what kind of physiological changes cause an increase in apoptosis in the

presence of chitosan of a high DA.

Furthermore, BDNF was chosen as an exemplary NTF to verify its secretion into the

conduit after nerve repair. Five and 18 days following surgery, BDNF was detected in

the pooled tube content with no difference between the DAs. mRNA level of BDNF

have been reported to increase 3 days after sciatic nerve transection before reaching

maximal levels 3-4 weeks later (Boyd and Gordon, 2003a; Meyer et al., 1992). With

regard to protein levels in the tube content, however, no such increase could be

seen from day 5 to 18 in our study, but an increase in proBDNF levels. The presence

of BDNF in the tube matrix is of importance since it has e.g. been reported to be

anterogradely transported in sensory neurons, thus accumulating in the tip of the

proximal nerve stump following nerve injury, before probably being released to

support prevention of axotomy induced death (Tonra et al., 1998). Moreover, BDNF

has been shown to enhance motoneuron outgrowth in vitro using an outgrowth

assay of spinal cord slices in a collagen matrix (Allodi et al., 2011).

6.1.3 Reinnervation of distal muscle target Following PNIs, functional recovery occurs if injured neurons successfully send their

axons out and connect to distal targets subsequently recreating motor endplates and

reestablishing appropriate reinnervations. Improvement of functionality is the most

important factor for the patient (Fugleholm et al., 2000; Valero-Cabré et al., 2001).

Therefore, to evaluate the progress regarding this parameter, motor recovery was

assessed by means of SSI and electrophysiological recordings.

6.1.3.1 Behavior

Since the introduction of the SSI by Bervar, when it was described as a time-saving

alternative to the sciatic function index (SFI), this method has been widely used in

our and other laboratories (Bervar, 2000; Bozkurt et al., 2011, 2008; Dornseifer et

al., 2011; Korte et al., 2011; Luís et al., 2008). In the present study, however, we

could not detect significant improvement over time regarding motor recovery using

the SSI, while it was demonstrated and reflected in the electrophysiological results.

Such inconsistency of results, obtained by both parameters, has been described

before. In a previous study, it has been shown that SSI and electrophysiological

results only correlate in cases of nerve crush, while following gap introduction and

Discussion-Project 1

- 112 -

subsequent regeneration, false negative results are reported (Korte et al., 2011).

This phenomenon has already been stated for the SFI when results did not correlate

with histological findings (Shenaq et al., 1989). Accordingly, the usefulness of this

evaluation method has to be questioned with regard to future experiments that

investigate treatment strategies in connection with nerve transection.

Importantly, however, it is known that the functional outcome following nerve

transection using nerve conduits or ANG treatment is often unsatisfactory for the

patient (Isaacs, 2010). This can be attributed to the loss of endoneurial tubes as

guidance structures upon nerve transection and the consequence that axons need to

find their way to and through the distal nerve stump to ultimately reestablish

connection with correct target organs (Bozkurt et al., 2008; Fugleholm et al., 2000;

Ransom, 2012; Thomas, 1989; Valero-Cabré et al., 2004). Unfortunately,

electrophysiological as well as histological data do not provide information regarding

accurateness of target reinnervation. Accordingly, quantitative target reinnervation

and remyelination represent no guarantee of a successful complex functional

recovery. Therefore, the inconsistency of results obtained from SSI and

electrophysiological as well as histological investigations can be explained by

quantitative, but not appropriate, reinnervation of distal muscle targets (Fugleholm et

al., 2000; Navarro and Udina, 2009; Sarikcioglu et al., 2009; Valero-Cabré et al.,

2001).

To allow a better functional evaluation of peripheral nerve regeneration, various

studies have tested alternatives to the SSI and SFI in the last decades, thereby

focusing on assessment of locomotor function by video gait analysis (Costa et al.,

2009). This method allows the determination of the angle formed by the knee, ankle,

and feet. Therefore, animals are trained to walk on a treadmill. Following sciatic

nerve injury, a decrease of the angles is observed, which then gradually increases

again with proceeding regeneration. Also, different gait cycle phases (e.g. initial

contact, toe off, mid swing) can be evaluated. Importantly, this method has been

suggested to be more sensitive than SFI in detecting motor recovery (De Boer et al.,

2012; Kokai et al., 2011; Lin et al., 1996; Patel et al., 2007). Furthermore, Bozkurt et

al. described an automatic gait analysis system (“CatWalk”), which allows the

quantification of dynamic parameters such as intensity of paw prints. In a similar

Discussion-Project 1

- 113 -

manner, another study used a Tescan pressure sensor grid to determine timing and

contact variables (Boyd et al., 2007; Bozkurt et al., 2008). Thereby, they could show

that animals decrease contact pressure on the injured hind limb following sciatic

nerve injury, while the contact force on the contralateral side increases. Still, finding

the appropriate evaluation method for functional recovery remains challenging.

Another possibility is the use of a forelimb rather than the sciatic nerve rat model

(Costa et al., 2009). The median nerve model e.g. offers the opportunity to perform

electrophysiological recordings, but also grip strength tests can be carried out. For

this test the rat / mouse can be held by the tail, while grasping a bar and the maximal

force generated when pulling the tail is recorded (Park and Höke, 2014).

Furthermore, a skilled walking test can be executed, during which animals have to

walk across a horizontal ladder with variable spacing of the rugs (Costa et al., 2009;

Metz and Whishaw, 2009) Wang et al. additionally reported the possibility to assess

wrist kinetics using a 2D motion analysis (Wang et al., 2008).

6.1.3.2 Electrophysiological recordings

Nevertheless, electrophysiological recordings and histological evaluation are

indispensable and provide useful information regarding the progress of regeneration

as well as regarding the amount of axons reaching distal muscle. Thereby, repeated

measurements can be performed since non-invasive recordings are not harmful to

the regeneration process and animals only have to be anesthetized (Navarro and

Udina, 2009). Four weeks after surgery, however, we did not see any CMAPs. This

observation is in line with reports that axons reach the distal end of a conduit (10 mm

gap) approximately after 3 weeks, a time frame that was also reported by our partner

laboratory (ULUND), who described axonal outgrowth into the chitosan tube for

about 7.9-8.6 mm after 3 weeks (Haastert-Talini et al., 2013; Williams et al., 1983).

Axons regrow with a rate of 1-3 mm per day. But due to retrograde axonal

degeneration as well as staggered growth at the suture site, a process that is

associated with progressive reinnervation of appropriate muscle pathways by motor

axons, the distance of the defect is not overcome earlier (Al-Majed et al., 2000; Daly

et al., 2012; Gordon, 2009; Gordon et al., 2003; Huelsenbeck et al., 2012; Kimura,

1984). Second measurements after 9 weeks, on the other hand, revealed

reinnervation for most of the animals that had received an ANG (TA: 62 %, PL:

80 %), while this was only the case for 6–29 % (TA) or 33-50 % (PL) of the

Discussion-Project 1

- 114 -

experimental groups. Last non-invasive as well as final invasive recordings

demonstrated successful reinnervation also in a higher number of animals that had

received chitosan conduits as treatment and which approached the level of

reinnervation seen with ANG treatment. The delay seen in nerve repair using hollow

chitosan conduits might be due to missing guidance structures for the distance of the

defect. Axonal elongation following ANG repair is eventually faster since the tissue

serves as a substratum for the growth cones and the SCs of this denervating

segment take part in the creation of a regeneration promoting microenvironment

(Wang et al., 2001). CMAPs selectively give information regarding function and

regeneration of α-motoneurons (Navarro and Udina, 2009). Using chitosan conduits

of different DAs, nerve reconstruction was comparably successful with respect to

skeletal muscle reinnervation by these motoneurons as seen with ANGs. No

differences were found within the treatment groups with regard to NCV, AxL (AUC),

or amplitude, but the level of healthy control values was not reached. This

observation is in line with previous results for treatment of nerve defects of 10-

14 mm length using hollow guidance conduits based on materials like PLGA / PCL,

PLGA / Pluronic F127, silk fibroin, collagen, but also chitin or chitosan (Dornseifer et

al., 2011; Oh et al., 2008; Panseri et al., 2008; Yang et al., 2007). The amplitude of

the measured CMAP provides information about the number of regrown motor axons

as well as regarding the size of the motor unit. But, a full amplitude recovery does

not necessarily mean that regeneration of all motoneurons was successful. Some

axons fail at reaching distal muscle targets and subsequently allow regenerating

axons to branch, thereby increasing the degree of reinnervation (Archibald et al.,

1991; Fugleholm et al., 2000; Navarro and Udina, 2009; Pfister et al., 2011; Valero-

Cabré et al., 2001).

The observation that measurable PL innervation occurred earlier than TA innervation

is hard to explain since CMAPs are expected to occur first in more proximal target

muscles (Navarro and Udina, 2009; Valero-Cabré et al., 2001). TA and PL muscles

are innervated by different branches of the sciatic nerve, but still it would be more

reasonable to find a first detectable signal in the more proximal muscle. A stronger

recovery of the TA amplitude compared to the PL amplitude as seen in this study

has already been reported before and can likely be explained by less axons reaching

the more distal muscle (Valero-Cabré et al., 2004, 2001). The latency, with which the

Discussion-Project 1

- 115 -

signals arrived at the muscles, was as expected higher for recordings from the more

distal target (PL) and reflects the conduction time of the largest recruited motor fibers

and therefore the degree of myelination and the size of the axons (Archibald et al.,

1991; Navarro and Udina, 2009; Valero-Cabré et al., 2001). Additionally, no

significant differences between the experimental groups including the control group

(ANG) were found. Contralateral intact control values, though, were superior to all

results obtained by the different bridging strategies like seen before in similar

experimental settings (Mancuso et al., 2011; Valero-Cabré et al., 2001). It has

already been previously noted that the latency in general is prolonged for a long time

after nerve lesion and differences are rarely detected between treatment groups.

This is an observation that also accounts for NCVs (Navarro and Udina, 2009;

Valero-Cabré et al., 2001). Furthermore, the current needed to elicit the maximal

CMAP when recordings were undertaken from the TA muscle was not different

between the treatment groups but differed to contralateral healthy control values.

Interestingly, only for PL recordings and only with DAII conduits, the current needed

to elicit the maximal CMAP was comparable to contralateral control values.

6.1.3.3 Morphological evaluation

Distal nerve stumps harvested upon explantation were sent to our partner laboratory

(UNITO, Italy) and subjected to morphometric analysis. Results are published in

Haastert-Talini et al. (2013) and show that in general the total number of myelinated

nerve fibers in all groups is higher than on the contralateral intact control side, but

only DAI treatment resulted in significantly more fibers in comparison to the ANG

group. The total numbers of myelinated fibers in healthy and ANG samples reported

are in line with morphometric results described before (Daly et al., 2013; de Ruiter et

al., 2008; Haastert-Talini et al., 2013; Luis et al., 2007). For hollow tube treatment,

however, lower numbers have been reported in former studies using e.g. PLGA or

poly (caprolactone fumarat) (PCLF) conduits (Daly et al., 2013; de Ruiter et al.,

2008), while other studies have stated similar numbers using PLGA or PCL conduits

(Luis et al., 2007; Varejão et al., 2003) as bridging material for 10 mm gaps.

Nevertheless, excessive sprouting as seen with DAI conduits is undesirable. Shortly

after lesion, lateral budding starts close to the injury site resulting in up to 25

branches per axon. While branches are pruned off with time, this is eventually not

the case for all of them. Consequently, single axons are often misguided and

Discussion-Project 1

- 116 -

reinnervate several muscles via their branches, thereby often eliciting antagonizing

effects. Also, muscles (hyperinnervation) as well as motor endplates

(polyinnervation) are targeted by two or more motoneurons. Altogether, these

processes increase the chance of disturbed motor coordination (Ijkema-Paassen et

al., 2002; Klimaschewski et al., 2013; Valero-Cabré et al., 2004, 2001). The reason

for increased sprouting in the presence of chitosan with a low DA, however, is not

clear. Investigation of axon and fiber diameters as well as myelin thickness revealed

differences to healthy values but no differences were seen among the treatment

groups, while no differences at all were detected regarding the g-ratio (axon

diameter / fiber diameter). Comparable results regarding these parameters have

been already reported bridging 10-14 mm gaps amongst others with chitosan

conduits, but also collagen, silk fibroin, PCL, PCLF and PLGA (Bini et al., 2004; Daly

et al., 2013; de Ruiter et al., 2008; Dornseifer et al., 2011; Varejão et al., 2003; Wang

et al., 2010, 2001; Yang et al., 2007).

6.1.3.4 Muscle weight

To further assess the reinnervation of distal muscle targets, the TA and

gastrocnemius muscles were harvested and their weight determined. No differences

were found between the experimental groups indicating comparable reinnervation

following ANG and chitosan conduit implantation. Nevertheless, none of the groups

reached healthy control values. The latter is in accordance with previous studies

testing PGA, PLGA, PCLF, collagen, but also chitosan conduits in 10-14 mm gap

models, while in some cases the conduits could not reach ANG levels (Alluin et al.,

2009; Daly et al., 2013; Dornseifer et al., 2011; Patel et al., 2006; Raisi et al., 2012;

Waitayawinyu et al., 2007).

Altogether, the functional, morphometric, and muscle weight data point out that

chitosan conduits of all DAs allow comparable results as the standard therapy

(ANG). But, an important finding is that DAI conduits seem to encourage excessive

sprouting and are therefore not convenient for further experiments. Also, healthy

control values have not been reached and a recovery time of 13 weeks was not

enough for nerve maturation to pre-surgery levels. This discrepancy, however, has

frequently been reported for hollow conduits of varying materials including those that

have received FDA and/or CE approval and is thought to only be overcome with the

Discussion-Project 1

- 117 -

right intraluminal guidance structure or filling materials that feature appropriate

textures and cues (Daly et al., 2013; Oh et al., 2008; Panseri et al., 2008;

Waitayawinyu et al., 2007; Yang et al., 2009).

6.1.4 Foreign body response (FBR) upon chitosan implantation degradation Chitosan has already been widely investigated for its use in tissue engineering, but

limitations were encountered due to its poor mechanical properties (Freier et al.,

2005c; Madihally and Matthew, 1999; Siemionow et al., 2010; Xie et al., 2008; Yang

et al., 2004). Notably techniques, however, have been adjusted by our partner

Medovent GmbH that make it possible to control the mechanical properties (Freier et

al., 2005c). 18 days, but especially 13 weeks after implantation, however, fissures

and cracks were displayed by the chitosan conduits with the highest DA (20%).

Additionally, it was partly observed that the conduits had completely collapsed and

thereby partly damaged newly grown nerve tissue. An accelerated degradation rate

of these conduits was expected, since lysozymes, the enzymes responsible for the

degradation of chitosan in our body, hydrolyze the ß(1-4) linkages between N-acetyl

glucosamine and glucosamine in chitosan as well as chitin and the degradation rate

is consequently in proportion to the DA (Han et al., 2011; Kean and Thanou, 2010;

Wang et al., 2010; Zhang and Neau, 2001). Previously, Tomihata and Ikada have

reported that about 20 % of chitosan with a DA of 16 % was degraded 12 weeks

upon implantation into a pouch in sub dermal tissue, while almost no weight loss was

seen for chitosan with a DA 10 % (Tomihata and Ikada, 1997). In line with this in our

study, conduits with a middle or low DA (5 and 1 %) did not show any macroscopic

signs of degradation. Interestingly, Wang and colleagues, who also used chitosan

(DA= 6.5 %) as a bridging material, have stated that the conduit had barely degraded

after 12 weeks (Wang et al., 2010). They further reported on a thin fibrous tissue

capsule that had developed around the conduit like we have seen it. Others,

however, have reported complete degradation 3 or 6 months after implanting

chitosan conduits (DA= 17 % or 7.7 %, respectively) with PLGA or PGA filaments

inside the lumen in the dog sciatic nerve model (Shen et al., 2010; Wang et al.,

2005). It has, however, been taken into account that the DA was higher, the

observation time partly longer, and that they used a different and bigger animal

model, which was proposed to maybe influence the degradation rate (Xie et al.,

2008). Xie and colleagues have worked with the same DA, mixed the chitosan with

Discussion-Project 1

- 118 -

polylactic acid, and tested conduits in a rat sciatic nerve model (Xie et al., 2008).

They detected many cracks and described the conduit as coarse 12 weeks after

surgery.

The thin fibrous capsule around the conduits was further analyzed to look at the

inflammatory response towards the chitosan conduits. In general, it is described that

upon implantation of a medical device, plasma proteins (e.g. fibrinogen, fibronectin)

are adsorbed to the material and likely undergo conformational changes. This results

in the exposition of binding sites for immune cells, thus initiating the FBR, which

serves as protection of the body and restructuring of the affected, injured area. Acute

inflammation, which is marked by infiltration of neutrophils within hours to days after

injury, is followed by chronic inflammation. The latter is characterized by the

formation of granulation tissue, a stage of repair and hallmark of healing, which

features macrophages, monocytes, lymphocytes, and collagen as well as

proliferating blood vessels. Additionally, MGCs partly surround the foreign material

through fusion of macrophages with the amount of present cells depending on the

form and topography of the material. In case of successful tissue reconstruction, the

last stage of the FBR is the formation of a collagenous and vascular capsule; if,

however, the immune reaction towards the material is not fading, foreign body

granuloma can occur, which are comprised of modified macrophages (epithelioid

macrophages) (Anderson, 2001; Anderson et al., 2008; Hu et al., 2001; Kisseleva

and Brenner, 2008; Morais et al., 2010; Thevenot et al., 2008). Eighteen days after

implantation of hollow chitosan conduits, the inner side of the formed capsule, which

faced the material surface, featured macrophages as well as few MGCs, a clear sign

of chronic inflammation. Furthermore, collagen forming cells were seen indicating

that tissue remodeling was in process. Interestingly, the MGCs did not contain many

nuclei (less than 20), which were mostly located in the periphery. These

characteristics are known for giant cells of the Langhans-type and are usually

associated with e.g. sarcoidosis and tuberculosis rather than the FBR. The latter

usually includes formation of FBGCs, which feature more and randomly distributed

nuclei. The possibility that Langhans-type cells are precursors of FBGCs has,

however, been discussed (Rhee et al., 1978). Also, following subcutaneous

implantation of cellophane foil, formation of both Langhans-type and foreign body

MGCs has been reported (Smetana, 1987). Moreover, McNally and Anderson noted

Discussion-Project 1

- 119 -

that a gray area exists between these two variants of MGCs and that further studies

are mandatory to elucidate different phenotypes and functions of different types of

MGCs (McNally and Anderson, 2011). Interestingly, following an implantation time of

13 weeks, the number of activated macrophages was significantly decreased, thus

pointing towards a reduction of the inflammatory response. Instead, dense regular

collagen fibers were visible in the contact area with the material, while vessels were

found in the area averted from it. Additionally, only single MGCs were found after 13

weeks in two animals that had received DAI conduits for nerve reconstruction. It was,

however, rather expected that these cells would be present in increased numbers in

the DAIII group, since degradation in this group was already in progress, but this was

not the case. Furthermore, although we could see macroscopic differences in the

optical density of the tissue, no significant differences were seen when it was

measured. The capsule displayed a thickness of 200-300 µm, which is slightly

thicker than known from literature (50-200 µm) following e.g. chitosan implantation in

a subcutaneous air pouch model or nerve reconstruction with poly(2-hydroxyethyl

methacrylate-co-methyl methacrylate) nerve conduits (Barbosa et al., 2010; Belkas

et al., 2005; Morais et al., 2010). On the other hand, it has been also reported that

following subcutaneous implantation of poly(L-lactide acid) (PLLA) the thickness of

the surrounding capsule is thicker during the first 6 month and then decreases to 100

to 200 µm (Mainil-Varlet et al., 1996). Previously, various studies have reported the

lack of a visible inflammation response towards chitosan or chitosan in combination

with other materials like PGA, PLGA (Raisi et al., 2012; Shen et al., 2010, 2009;

Wang et al., 2005). Others, conversely, have described the formation of fibrous

capsules (Simões et al., 2011; Tomihata and Ikada, 1997; Wang et al., 2010).

Simões et al. (2011) subcutaneusly implanted chitosan membranes, which had been

subjected to different drying methods (37°C vs. freeze drying) and ultimately

displayed variances in their porous microstructure. Eight weeks afterwards, they

detected a mild chronic inflammation in form of a collagenous fibrous capsule (37°C)

or a strong granulomatous reaction (freeze drying) depending on the production

procedure. In a similar experimental setting, Tomihata and Ikada described a mild

tissue reaction towards chitosan membranes with varying DAs (DA: 0, 20, 16 or

27 %), while there was strong reaction towards chitin and chitosan with a DA of 31 %

due to rapid degradation (Tomihata and Ikada, 1997). Analogous results were

obtained by Hidaka et al., who saw a mild tissue reaction following chitosan

Discussion-Project 1

- 120 -

membrane implantation with a DA around 0 % or 6 %, while the initial inflammatory

response was stronger with a DA between 20 % and 35 % (Hidaka et al., 1999).

Furthermore, they speculated that the more negatively charged surface of chitosan

with a higher DA lead to an increased plasma protein adhesion, thus triggering the

inflammtion by higher acetylated chitosan derivatives. Another study, in which

chitosan with a DA of 4 % and 15 % was implanted subcutaneously, supported these

findings. Higher recruitment numbers of inflammatory cells to the implant site and

also increased adherance of these cells to the material was reported in the presence

of a higher DA. Furthermore, they also stated that the capsule around 15 % DA

chitosan was thicker (Barbosa et al., 2010). Also, in the presence of other

biodegradable materials, the formation of a fibrous capsule has been reported as

well as formation of MGCs. A study testing PLLA / plasticizer triethylcitrate

membranes reported that MGCs were present in the developed capsule after 7 days

and increased in number up to the 180th day of observation (Maluf-Meiken et al.,

2006). This is contrast to our results, where we saw a decrease or rather none of

these cells at all except for samples of two DAI animals. Other studies implanting

PLLA materials reported the presence of MGCs for weeks to months and also the

eventual formation of a fibrous capsule (Mainil-Varlet et al., 1996; Pistner et al.,

1993).

Although we did not see differences in the inflammatory response between the

experimental groups with regard to analysis of the fibrous capsule that had

developed around the conduits, other observations were made when analyzing the

regenerated nerve tissue. Counting of MGCs also did not reveal differences between

the conduits and also activated macrophages were equally present in the nerve

tissue after 18 days, but 13 weeks after implantation most ED1-positive cells were

found in samples of the ANG and DAIII groups. The latter results were obtained by

Sandra Wrobel ( MSc, Institute of Neuroanatomy, MHH) and can be looked up in

((Haastert-Talini et al., 2013)). This observation can probably be assigned to the

presence of chitosan fragments that were detected in 5 out of 5 analyzed animals in

the DAIII group after 13 weeks. Interestingly, already after 18 days, fragments were

found in this group, supporting the finding of very fast degradation of these conduits

and their risk for newly formed nerve tissue, although those fragments were mainly

localized in the outer tissue layer. Activated macrophages, however, were present in

Discussion-Project 1

- 121 -

higher numbers 18 days after implantation than after 13 weeks indicating a reduction

of the inflammatory response as it was already described for the fibrous capsule.

The great number of macrophages after 18 days, though, is in line with the

understanding that after nerve lesion, nerve stumps are highly invaded by

hematogenous macrophages, which also infiltrate the fibrin matrix from both sides.

They secrete growth factors, pro- as well as anti-inflammatory cytokines, and are

thus of importance for debris clearing, but also for promotion of SC proliferation and

axonal regeneration in early phases of regeneration (Díaz-Flores et al., 1995; Kiefer

et al., 2001; Zochodne and Cheng, 2000).

Altogether, we saw a mild tissue reaction towards chitosan conduits in the formation

of a thin fibrous capsule. This kind of FBR, however, has been frequently described

in studies testing various materials and does not endanger successful nerve tissue

regeneration. The observation of fragments of DAIII conduits in the nerve tissue

already after 18 days as well as after 13 weeks, on the other hand, bears a risk for

the newly formed nerve tissue and is a criterion of exclusion for conduits with a high

DA. Together with the finding of increased sprouting with DAI conduits, which is

undesired since it provokes hyper- and polyinnervation as well as innervation of

wrong targets, it can be concluded that hollow DAII chitosan conduits are the eligible

candidate for bridging small defects.

6.1.5 Concluding remarks-Chances of hollow chitosan conduits on the existing market The market of conduits for peripheral nerve repair is growing and different materials

have already gained approval by the FDA or CE to be used in clinical applications

with NeuraGen® (collagen), Neurotube® (PGA), and Neurolac® (PCL) representing

the mostly described and applied products. Nevertheless, none of the marketed

conduits so far has encountered unlimited acceptance and although it is often

suggested that these conduits can be used for gaps of up to 3 cm length, it was also

proposed that the realistic bridgeable distance is probably less (Isaacs, 2010). Just

recently, Liodaki et al. described insufficient nerve regeneration in four patients using

NeuraGen® (Liodaki et al., 2013). In agreement, successful sensory and motor

recovery was only seen when the defect was 6 mm or less in a multicenter clinical

trial using NeuraGen® (Boeckstyns et al., 2013). Furthermore, case studies revealed

Discussion-Project 1

- 122 -

neuroma formation using this type of conduit (Moore et al., 2009). The same authors

also described a failed case of regeneration through a Neurotube® conduit. In this

case, the conduit was fully degraded after 9 months and a gap defect between the

proximal and distal stump was left. In general, variable results were obtained so far

with this conduit, in total emphasizing that successful regeneration is only seen in

small gaps (Berger et al., 1994; Isaacs, 2010; Kehoe et al., 2012; Meek and Coert,

2008; Navissano et al., 2005). With regard to Neurolac®, complications have been

reported. Meek and den Dunnen e.g. used porous Neurolac® conduits for nerve

reconstruction in rats and described fragmentation as well as severe swelling and

inflammation (Meek and Den Dunnen, 2009). Implantation of less porous conduits

resulted in not as much swelling, but fragments were still found after 16 as well as 24

months, although the material was no longer macroscopically visible at the latter time

point. These fragments, which were accompanied by MGCs and macrophages,

however, where only found in the epineurium of the regenerated nerve and did not

influence the nerve regeneration (Jansen et al., 2004; Meek and Jansen, 2009).

Nevertheless, it was also noted that the conduit might be too stiff, thus causing

needle breakage and problems in case of positioning over joints. Also, in a clinical

case, neuroma formation was described upon collapse of the conduit (Chiriac et al.,

2012; Hernández-Cortés et al., 2010; Kehoe et al., 2012; Meek and Coert, 2008).

Given the limitations of already approved conduits, the here tested hollow chitosan

conduits of a DAII can most likely compete on the market and a first clinical trial is in

planning for bridging of small defects. Advantages of hollow DAII conduits include

their transparency, which so far is only displayed by the Neurolac® conduit, as well

as a mild tissue reaction. Nevertheless, so far we do not know when and in which

time frame the DAII conduits will degrade. Therefore, we currently perform long-term

studies in the rat sciatic nerve model addressing this issue. Furthermore, it is unclear

how the conduit would perform if the nerve defect was localized over a joint, where

kinking could cause damage to the nerve tissue. To conclude, the hollow DAII

chitosan conduit with a DA of 5 % has the best requirements to get a place on the

market of peripheral nerve reconstruction of small defects, but some questions

remain to be answered in the future.

Discussion-Project 1

- 123 -

Discussion-Project 2

- 124 -

6.2 Project 2 6.2.1 Schwann cells (SCs) as vehicles for gene delivery Due to their multifaceted and significant role during peripheral nerve regeneration,

the delivery of NTFs to the site of injury has been extensively studied and different

methods have been applied to make them available in a sustained manner. This

includes e.g. NTF-loaded nerve guidance conduits as well as application via

introduction of microspheres, osmotic minipumps, or matrices such as gels,

sponges, and filaments, but also transplantation of physiological or genetically

modified cells has widely been investigated (Belkas et al., 2004b; Haastert et al.,

2006a; Hadlock et al., 2000; T. Ikeda et al., 2014; Jungnickel et al., 2011; Kokai et

al., 2011, 2010; Madduri et al., 2010; Pabari et al., 2011; Pfister et al., 2007; Timmer

et al., 2003; Wood et al., 2013a). SCs, the glial cells of the PNS, proliferate, loose

their myelin, migrate to the injury site, phagocytize myelin debris, guide regrowing

axons, and represent a major source of NTFs. Accordingly, their transplantation has

been suggested to have a beneficial effect on peripheral nerve regeneration.

Consequently, autologous naïve as well as genetically modified and NTF

over-expressing SCs are candidates for artificial biohybrid nerve devices (Madduri

and Gander, 2010; Pfister et al., 2007; Reichert et al., 1994). Advantages of this

delivery method include long-term availability at the site the injury, the lack of

non-physiological degradation products, and the ability of these cells to respond to

changes in the environment. A drawback regarding SC transplantation, however, is

the ex vivo proliferation and purification procedure of these cells (Di Summa et al.,

2011; Dove, 2002; Madduri and Gander, 2010; Pfister et al., 2007). Therefore,

alternative cell sources including e.g. OECs, BMSCs, and ADSCs have also been

investigated in connection with peripheral nerve repair and revealed promising

results (Ao et al., 2011; Di Summa et al., 2011; di Summa et al., 2010; Gu et al.,

2014; Radtke et al., 2011, 2009; Wei et al., 2011). For example, BMSC-derived SCs

have been reported to allow similar nerve regeneration as seen with sciatic nerve

derived SCs when seeded into a chitosan conduit (Ao et al., 2011). Also, chitosan /

silk fibroin-based nerve grafts enhanced regeneration when ADSCs were included

for treatment of a 10 mm gap in the rat, but ANGs were not included as a positive

control (Wei et al., 2011). In another study, however, results obtained with fibrin

conduits containing ADSCs differentiated into a SC-like phenotype were comparable

Discussion-Project 2

- 125 -

to those with ANGs (Di Summa et al., 2011). Interestingly, in the scope of a short-

term study, it was demonstrated that SCs hold a higher regenerative potential

compared to ADSCs as well as BMSCs (di Summa et al., 2010). Due to their multi-

functional role in the PNS and in response to PNIs, SCs therefore remain ideal

candidates for transplantation experiments (Madduri and Gander, 2010). Collagen

nerve guidance conduits seeded with SCs have already been shown to enhance

nerve regeneration in a 18 mm rat sciatic nerve rat model and autogenous veins

containing SCs performed better in a 40 mm tibial nerve rabbit model than empty

veins, although they could not compete with ANGs (Ansselin et al., 1997; Pfister et

al., 2007; Zhang et al., 2002). Autogenous venous nerve conduits were further

reported to successfully bridge a 6 cm gap in rabbits when SCs in Matrigel were

included, but no functional tests were conducted in this case (Strauch et al., 2001).

Also, Hadlock et al. could show an improved performance of small intestinal

submucosa-based nerve guides by adding SCs, although the loaded conduits did not

reach ANG levels in a 7 mm gap with regard to SFI and extensor postural thrust

testing (Hadlock et al., 2001). In another study, bridging of a 6 mm gap was most

successful with SC-seeded collagen conduits and combined administration of an

immunosuppressant (FK506) (Udina et al., 2004). Also, genetically modified SCs

have already been successfully transplanted and demonstrated to have regeneration

promoting effects (Haastert et al., 2006a; Timmer et al., 2003).

6.2.2 Recovery of mechanical pain sensibility In the scope of the second study, the battery of tests was extended to allow a more

widespread evaluation of the performance of the tested conduits. Therefore, the Von

Frey test was conducted to assess the mechanical pain sensibility since

hyperalgesia (a decrease in pain threshold), has been reported in patients following

nerve lesions (Bennett and Xie, 1988; Bridges et al., 2001; Casals-Díaz et al., 2009;

Cobianchi et al., 2014; Kingery et al., 1994; Sandkühler, 2009). For this test, animals

were placed in plastic compartments on a metallic mesh and medial as well as

lateral paw sites was stimulated with a non-noxious mechanical stimulus to

determine return of sensibility. Basis for this test is that innervation of muscles

proximal to the knee is preserved upon sciatic nerve injury, thus allowing a

withdrawal response in case of touch sensation (Cobianchi et al., 2014). Under

healthy conditions, tibial and sural branches of the sciatic nerve as well as the

Discussion-Project 2

- 126 -

saphenous nerve provide sensory innervation to the medial paw site, while the

lateral area is exclusively innervated by the sciatic nerve (Cobianchi et al., 2013).

Due to these innervation patterns, it is of importance to differentiate between

responses obtained from different paw regions or interpretation of results can be

misleading (Cobianchi et al., 2014).

Following lesion of the sciatic nerve and subsequent repair using either ANGs or

chitosan conduits containing NVR-Gel, which was supplemented with naïve or

genetically modified SCs, respectively, hyperalgesia was seen when stimulating the

medial paw territory on the lesioned side compared to the respective healthy

contralateral side. This observation is in line with results obtained by Cobianchi et al.

(2013), who reported increased sensitivity in the same tests following end-to-end

repair in the rat sciatic nerve model. Also, as a result of nerve transection and

excision of 2 mm from the distal stump mechanical hyperalgesia was observed for

weeks (Casals-Díaz et al., 2009). The reason for the mechanical hyperalgesia after

sciatic nerve injury can be found in collateral sprouting from the intact neighboring

saphenous nerve into the area that was formerly innervated by the sciatic nerve.

This process is known as adjacent neuropathic hyperalgesia (ANH) and is promoted

by NGF secretion by the denervated tissue. In case of successful target-organ

reinnervation through the sciatic nerve, the increased sensitivity might be reversed or

attenuated (Casals-Díaz et al., 2009; Cobianchi et al., 2014, 2013; Diamond et al.,

1987; Kingery and Vallin, 1989; Kingery et al., 1994). With the results obtained in this

study, however, we do not see a reduction of ANH including the ANG group. This is

in agreement with reports that, as a consequence of nerve transection with distal

segments exclusion, ANH is described for at least 12 weeks and normal sensitivity is

not restored for about 2 months after suture repair (Cobianchi et al., 2014, 2013;

Kingery et al., 1994). The persisting hypersensitivity also in case of reinnervation has

further been proposed to be the result of chronic sensitization of the saphenous

nerve pathway including spinal cord circuits (e.g. via unmasking of pre-existing

connections or changes involving inhibitory interneurons) (Casals-Díaz et al., 2009;

Cobianchi et al., 2014; Navarro et al., 2007; Sandkühler, 2009; Sotgiu and Biella,

1995).

Discussion-Project 2

- 127 -

Stimulation of the lateral paw site did not provoke withdrawal response in any of the

tested animals 7 days as well as 4 weeks after nerve lesion indicating complete

denervation of this territory as expected. In all animals of the ANG group, however,

return of sensation was seen after a time period of 9 weeks, but the response could

already be detected upon light stimulation representing hyperalgesia. In similar

manner, single animals of the experimental groups responded to stimulation. One

animal each of the nt-neoSCs and tf-neoSCs-GDNF group as well as a total of 3

animals of the tf-neoSCs-FGF-218kDa group showed signs of sensory recovery. The

general observation that after nerve injury no response is seen is in agreement with

prior findings and can be attributed to the sole innervation of this paw site through

the sciatic nerve in healthy animals. Recent testing of hollow chitosan conduits in a

15 mm gap model by our partner laboratory revealed similar withdrawal responses

(Talk: “Tubulization with chitosan guides for the repair of long gap peripheral nerve

injury in the rat” by Francisco Gonzalez-Perez from UAB at the 2nd International

Symposium on Peripheral Nerve Regeneration in Turin, Italy). Based on experiments

that included cutting of the saphenous nerve, the hypersensitivity in the lateral paw

site after 4 weeks was attributed to the arrival of regenerating axons to this territory

and not ANH (Cobianchi et al., 2014).

Interestingly, Cobianchi et al. could show that after nerve transection, the number of

peptidergic intra-epidermal nerve fibers was reduced in lateral and medial paw

territories (Cobianchi et al., 2014). First regenerating axons were seen reaching the

epidermis 4 weeks following surgery, while massive reinnervation was observed

after 2 months with numbers of peptidergic fibers, which have been described to be

involved in the signaling of noxious pain and maintenance of hypersensitivity,

exceeding healthy control values. Furthermore, they could ascribe the early

hyperalgesia in the medial paw region to saphenous peptidergic fiber sprout growth

in the subepidermal plexus, while the decreased pain threshold seen for stimulation

of the lateral paw region correlated with skin reinnervation.

Altogether, the obtained results show reinnervation of distal target-organs in the

ANG group as indicated by withdrawal responses seen earliest 9 weeks after nerve

injury upon stimulation of the lateral paw territory on the lesioned side. Regarding the

experimental groups, however, only single animals displayed sensory recovery,

Discussion-Project 2

- 128 -

which mainly belonged to the FGF-218kDa group. The animals that responded to

stimulation of the plantar paw surface could later be correlated to show regenerated

nerve tissue, thus supporting the findings in this test.

6.2.3 Fibroblast growth factor (FGF)-218kDa partly allows reinnervation of distal muscle targets and functional recovery 6.2.3.1 Behavior

As already seen with hollow chitosan conduits to bridge short nerve defects in

project 1, we could not detect any motor recovery in the differently treated groups

using the SSI test. Since only single animals of the experimental groups revealed

nerve tissue regeneration upon explantation, this result is not surprising. Aside from

the fact that this test only provides minimal information in case of studies of complete

nerve transection (see 6.1.3.1), another problem is that sometimes it is difficult to

gain data. It is frequently observed that after nerve transection, rats and mice start to

bite off their nails and phalanges, a behavior that is referred to as autotomy (Costa et

al., 2009; Sarikcioglu et al., 2009). Discussions are ongoing if this self-harming

procedure is the response to painful perception in the injured nerve (neuropathic

pain) or to the feeling of insensitiveness and the need to shed off some kind of

foreign body part. During our study, autotomy was first observed 3 weeks after

surgery, a time point, which has previously been noted as a usual starting point

(Sarikcioglu et al., 2009). With advancing observation period the number of animals

that showed signs of autotomy increased. Thereby, it was noticeable that animals of

the ANG, except for one animal, did not display autotomy, while it was frequently

observed across the experimental groups with most severe cases among the GDNF

group. Also, another study reported severe cases of autotomy when fibrin sealant

containing GDNF was locally applied to a transected sciatic nerve (Jubran and

Widenfalk, 2003). This enhanced toe biting within the GDNF group could support the

theory that autotomy is indeed the reaction towards insensitiveness in this body part

since analgetic, neuropathic pain reducing effects of this factor have been reported

(Boucher, 2001; Boucher et al., 2000).

6.2.3.2 Electrophysiological recordings

Also, in project 2, electrophysiological tests were conducted to elucidate functional

recovery. Reinnervation of distal muscle targets and subsequent motor recovery was

Discussion-Project 2

- 129 -

seen for all animals, which had received ANGs for nerve reconstruction. On the other

hand, CMAPs were only detected in single animals of the experimental groups and

here mainly the FGF-218kDa treatment lead to positive recordings. These results are

in line with reports indicating that FGF-218kDa has a promoting effect on neurite

outgrowth of spinal motoneurons in vitro (Allodi et al., 2013). The authors of this

study further suggested that FGF-218kDa likely exhibits its effect during early stages of

nerve repair by encouraging interactions between polysialic acid (PSA)-NCAM and

FGFR-1 with the latter being highly expressed by SCs accompanying motor neurites.

Furthermore, our laboratory could previously show that FGF-2 in general plays an

important role after PNI and different effects can be attributed to the isoforms;

something that is also supported through the dissimilar regulation of these isoforms

with regard to temporal as well as spatial parameters. While the high molecular

isoform, which has already enhanced nerve regeneration across a 15 mm gap, has

mainly been associated with sensory regeneration and has been found to be mostly

up-regulated in DRGs, the low molecular isoform has been demonstrated to allow

motor recovery and is rather prominently seen close to the injury site (Grothe et al.,

2000, 1996; Haastert et al., 2006a; Meisinger and Grothe, 2002; Timmer et al.,

2003). Moreover, Jungnickel et al. could demonstrate that transgenic mice

over-expressing FGF-2 display an increased number of proliferating SCs as well as

more axons 1 week after nerve crush (Jungnickel et al., 2006). An important function

of FGF-2 was suggested during early phases of regeneration by regulation of SCs

differentiation as well as by directly influencing axons (Fujimoto et al., 1997;

Jungnickel et al., 2004). In addition, FGF-2 was shown to be involved, amongst

others, in remyelination, protection of DRGs from lesion-induced death, and injury-

induced neuronal apoptosis. These different observations emphasize the crucial role

of this factor during peripheral nerve regeneration (Grothe et al., 2006; Haastert et

al., 2006a; Jungnickel et al., 2010, 2006, 2004; Otto et al., 1987). Interestingly, Han

et al. (2010) have reported improved functional outcome when chitosan conduits

(DA ~ 16 %) were filled with heparin-incorporated fibrin-fibronectin gel, which served

as a delivery system for FGF-2 compared to control, PBS filled, conduits in a 10 mm

gap (Han et al., 2010). Using the latter, however, they could only report successful

reinnervation in 5 out of 9 animals, while we recorded CMAPs in 13 out of 14

animals that were treated with hollow DAII chitosan conduits during project 1. The

differences regarding the success rate can possibly be attributed to the different DA

Discussion-Project 2

- 130 -

as well as varying production processes / fine tuning methods. Furthermore, FGF-2

has previously promoted nerve tissue regeneration across a 15 mm gap within

4 weeks when it was released from a BSA / ethylene-vinyl acetate based conduit

(Aebischer et al., 1989). Also, Wang et al. have reported enhanced nerve

regeneration across a 15 mm gap when delivering FGF-2 with regard to functional

and morphometric parameters, but the comparison to the ANG was missing (Wang

et al., 2003). Recently, Ikeda et al. (2014) could show that double layered nerve

conduits, consisting of PLA and PCL, in combination with induced pluripotent stem

cell-derived neurospheres and a FGF-drug delivery system accelerated nerve

regeneration in the mice sciatic nerve model compared to conduits alone (M. Ikeda

et al., 2014).

The missing reinnervation seen with GDNF supports former findings that this factor

might possibly be only effective in case of delayed repair and fails to promote

regeneration when administered immediately after nerve transection (Boyd and

Gordon, 2003a, 2003b; Höke et al., 2002; Wood et al., 2013a, 2013b). Moreover, it

has been reported that sustained over-expression of GDNF by SCs trapped axons in

acellular nerve grafts ultimately preventing functional regeneration across a 14 mm

gap in the rat sciatic nerve model (Santosa et al., 2013). On the other hand, it has

been described that GDNF promotes motor as well as sensory neurite outgrowth in

vitro and serves as a protective factor for these neurons (Allodi et al., 2013;

Henderson et al., 1994; Li et al., 1995; Madduri and Gander, 2010; Madduri et al.,

2009). Furthermore, Patel et al. could show that GDNF and laminin blended chitosan

conduits enhanced functional and sensory recovery as assessed by gait analysis,

sensitivity testing, and muscle weight determination in a 10 mm gap (Patel et al.,

2007). Similarly, 7 weeks after nerve repair, Fine et al. have described the formation

of a regeneration cable across a 15 mm gap when GDNF was slowly released (Fine

et al., 2002). That GDNF is an important factor for motoneuron regrowth was

confirmed by the finding that it promotes facial nerve regeneration (Barras et al.,

2002). Li and colleagues could further show enhanced recovery (10 mm gap) by

means of morphometric and electromyography examination compared to hollow

silicone tube (Li et al., 2006). In addition, Kokai and colleagues have reported

functional recovery across a long gap (15 mm) when delivering GDNF to the lesion

site via microspheres (Kokai et al., 2011). After 16 weeks, they saw higher muscle

Discussion-Project 2

- 131 -

twitch force in GDNF treated animals than in negative controls (empty

microspheres), while results were comparable with isografts (nerve graft from rats of

the same strain). Regarding muscle weight, however, the isograft was superior.

Interestingly, all animals that had received GDNF showed nerve tissue regeneration

while this was only the case in 70% of the negative control. Noteworthy is, however,

that they saw a high amount of collagen content within the lumen in both conduit

treated groups and fibers only appeared organized in the GDNF group. This is in

contrast to the tissue regeneration seen in this study. Mainly in the presence of FGF-

218kDa, some regeneration was seen and axons were organized, while collagen

contributed less to the complete nerve section. Furthermore, Chew et al. have

reported functional recovery in 44% of rats that had received fibers releasing GDNF

when bridging a 15 mm gap in the rat, while it was less without the growth factor

(Chew et al., 2007). In this study, however, it is not clear if the failed regeneration

can be explained by the missing effect of GDNF following immediate repair or if the

used luminal filler maybe had a negative effect on the regeneration and only FGF-

218kDa partly enabled regeneration against this obstacle. The idea that the hydrogel

represented some kind of barrier for the regrowing axons is supported by a study

from our partner laboratory that revealed nerve regeneration and functional recovery

across a 15 mm gap in 3 out of seven rats using hollow chitosan conduits (Talk:

“Tubulization with chitosan guides for the repair of long gap peripheral nerve injury in

the rat” by Francisco Gonzalez-Perez from UAB at the 2nd International Symposium

on Peripheral Nerve Regeneration in Turin, Italy).

6.2.4 The difficulty in finding the right luminal filler for axonal guidance In general, it is thought that structures inside the nerve guidance conduits could

facilitate SC migration as well as axonal outgrowth by replacing or supporting the

initially formed fibrin cable (Daly et al., 2012). In this study, however, the impression

is given that a hydrogel as SC carrier may have hampered nerve regeneration.

Possibly, the gel used was too dense, thus preventing the formation of the initial

fibrin matrix as well as subsequent migration of SCs, growth of axonal sprouts, and

diffusion of molecules. Usage of less dense gel, however, is difficult, since it would

leak from the conduit immediately after nerve reconstruction and the transplanted

cells would no longer be present at the site of need. The results of the in vivo study

regarding NVR-Gel do not coincide with in vitro results obtained by Sandra Wrobel

Discussion-Project 2

- 132 -

during her PhD-thesis (Institute of Neuroanatomy, MHH). She could show that

transfected SCs display their bipolar shape within the gel and neurite outgrowth

assays led to promising results. Recently, another in vitro study suggested great

potential of this hydrogel for in vivo applications (Ziv-Polat et al., 2014). Thus,

modifications of the NVR-Gel could potentially enhance its application in vivo, but so

far in vitro and in vivo results are diverging.

While some have used different hydrogels and were able to enhance nerve

regeneration, others have reported impeded regeneration as a result of the

introduction of matrixes (Apel et al., 2008; Hill et al., 2011; Karabekmez et al., 2009;

Pfister et al., 2007; Valentini et al., 1987; Wang et al., 1998; Zheng et al., 2010). It

has e.g. been attempted to bridge a 4 mm rat sciatic nerve gap using collagen- and

laminin-containing gels. Acellular remnants of the gel could still be identified after 12

weeks and less myelinated axons were found compared to saline filled conduits

(Valentini et al., 1987). Consequently, the authors suggested that a faster absorbed

delivery matrix might enable regeneration. Also, a SC-containing

chitosan/glycerol-ß-phosphate disodium salt hydrogel injection into conduits led to

complete failure of nerve regeneration across a 10 mm gap in the rat sciatic nerve

model, while SCs in suspension or culture medium alone allowed regeneration

(Zheng et al., 2010). Interestingly, Madison et al. have demonstrated that

laminin-containing gel initially enhanced nerve regeneration at 2 weeks, but acted

inhibitory at 6 weeks when bridging a 4 mm defect (Madison et al., 1987). The

authors suggested that the observed reversal occurred due to swelling and

decreased luminal space with the laminin gel, something that might have also

happened in this study. Moreover, the implantation of collagen tubes containing

SC-seeded polyglactin filaments in addition to SCs in Matrigel allowed only partial

regeneration, but failed completely regarding motor recovery (Lohmeyer et al.,

2007). It was concluded that the amount of inserted materials possibly acted as a

physical impediment to nerve regeneration or hampered cell function. Noteworthy,

Pfister and colleagues (2007) summarized in a review that growth factor free

matrixes result in poor regeneration, while the addition of these factors generally

mediated a better result than empty tubes. This holds not completely true for the

here tested chitosan conduits, since GDNF did not enhance nerve regeneration

compared to physiological SCs or empty vector transfected SCs in NVR-Gel. The

FGF-218kDa treated animals are the only group showing successful regeneration in

Discussion-Project 2

- 133 -

about 57% giving the impression that this growth factor partly enabled regeneration

even in a non-permissive environment created through the presence of the probably

too dense gel. Interestingly, we found NF200-positive fibers proximal to the proximal

suture in cases of failed regeneration, while this was not the case directly at the

suture site. This observation further supports the idea that axons started to regrow

but did not cross the suture line to grow inside the hydrogel filled conduit.

6.2.5 No enhanced inflammatory response elicited by (modified) Schwann cells (SCs) in NVR Gel The inflammatory response was comparably assessed as described for project 1 by

analyzing the fibrous capsule that had formed around the conduits as well as the

(regenerated) nerve tissue. The determination of the capsules thickness and area as

well as counting of activated macrophages and MGCs resulted in similar results as

seen for hollow chitosan conduits indicating that there was no increased

inflammatory response due to the presence of the NVR Gel or SCs, regardless if the

latter were genetically modified or not. The same observation was made for analysis

of the regenerated nerve tissue or nerve stumps in cases of failed regeneration.

6.2.6 Concluding remarks and considerations regarding future conduits for long nerve defects Studies that address the treatment of long peripheral nerve defects have to take into

account numerous variables that decide whether a conduit will have success in

promoting peripheral nerve regeneration or not. Although the PNS has an intrinsic

capacity to regenerate, problems like misdirection of axons, the deprivation of

regeneration promoting factors (e.g. due to SC denervation), and muscular atrophy

deteriorate the chance of accurate reinnervation and functional recovery limit the

success rate especially across long gaps and in cases of delayed repair (Allodi et al.,

2012; de Ruiter et al., 2008; Gordon, 2009; Gu et al., 2014; Klimaschewski et al.,

2013; Lin et al., 2011; Saito and Dahlin, 2008; Scheib and Höke, 2013)

Intraluminal guidance structures / fillers are thought to be favorable, since they

potentially support SC migration and axonal outgrowth, but as seen in this study,

their introduction can also disturb nerve regeneration (Daly et al., 2012). We

observed complete failure of regeneration in most animals of the enriched conduits

Discussion-Project 2

- 134 -

treated animals except for the FGF-218kDa group and partly only detected very thin

tissue connections between the two nerve stumps, which only occasionally

contained axons. Successful regeneration including functional recovery was only

seen when SCs over-expressing FGF-218kDa were suspended in the gel. Compared

to hollow chitosan conduits, however, which were previously tested in a long gap by

our partner laboratory, the gene delivery did not enhance the rate of successful

regeneration (Talk: “Tubulization with chitosan guides for the repair of long gap

peripheral nerve injury in the rat” by Francisco Gonzalez-Perez from UAB at the 2nd

International Symposium on Peripheral Nerve Regeneration in Turin, Italy). It can be

assumed that the limited success of the enriched nerve conduit is based on the

density of the used gel and that only FGF-218kDa partly promoted axonal outgrowth in

this unfavorable environment. For future experiments, it is therefore of importance to

identify a more suitable delivery method for these genetically modified cells, which

finds a balance in providing a guidance structure without impeding axonal outgrowth

due to unintended barrier functions of the filler (Lin and Marra, 2012). Alternatives

could be very porous sponges and fibers as well as membranes that only cover a

limited area of the conduits cross-section, thus leaving space for regrowing axons

but still offer a guidance surface for growth cones. Furthermore, growth factor

delivery methods are in need of modification in a way to allow a spatial and temporal

dependent release that adjusts to different phases of nerve regeneration. This issue

is, however, very complex, since it has to be determined which factors are needed at

what time point of regeneration as well as in what dose and at what localization, and

most importantly a sophisticated delivery method has to be found that allows this

kind of regulation. Also, different regulation profiles have to be established for nerves

of different modalities, since it was shown that sensory and motor nerves prefer

varying environments and are usually accompanied by SCs of differing phenotypes

(Höke et al., 2006). Furthermore, synergistic effects of different factors have to be

elucidated to increase the regeneration capacity.

- 135 -

References

- 136 -

7 References Aebischer, P., Salessiotis, A.N., Winn, S.R., 1989. Basic fibroblast growth factor released from

synthetic guidance channels facilitates peripheral nerve regeneration across long nerve gaps. Journal of Neuroscience Research 23, 282–89.

Allodi, I., Casals-Díaz, L., Santos-Nogueira, E., Gonzalez-Perez, F., Navarro, X., Udina, E., 2013. Fgf-2 low molecular weight selectively promotes neuritogenesis of motor neurons in vitro. Molecular Neurobiology 47, 770–81.

Allodi, I., Guzmán-Lenis, M.-S., Hernàndez, J., Navarro, X., Udina, E., 2011. In vitro comparison of motor and sensory neuron outgrowth in a 3D collagen matrix. Journal of Neuroscience Methods 198, 53–61.

Allodi, I., Udina, E., Navarro, X., 2012. Specificity of peripheral nerve regeneration: Interactions at the axon level. Progress in Neurobiology 98, 16–37.

Alluin, O., Wittmann, C., Marqueste, T., Chabas, J.-F., Garcia, S., Lavaut, M.-N., Guinard, D., Feron, F., Decherchi, P., 2009. Functional recovery after peripheral nerve injury and implantation of a collagen guide. Biomaterials 30, 363–73.

Al-Majed, A.A., Neumann, C.M., Brushart, T.M., Gordon, T., 2000. Brief electrical stimulation promotes the speed and accuracy of motor axonal regeneration. The Journal of Neuroscience 20, 2602–8.

Anderson, J.M., 2001. Biological responses to materials. Annual Review of Materials Research 31, 81–110.

Anderson, J.M., Rodriguez, A., Chang, D.T., 2008. Foreign body reaction to biomaterials. Seminars in Immunology 20, 86–100.

Angius, D., Wang, H., Spinner, R.J., Gutierrez-Cotto, Y., Yaszemski, M.J., Windebank, A.J., 2012. A systematic review of animal models used to study nerve regeneration in tissue-engineered scaffolds. Biomaterials 33, 8034–9.

Ansselin, A.D., Fink, T., Davey, D.F., 1997. Peripheral nerve regeneration through nerve guides seeded with adult Schwann cells. Neuropathology and Applied Neurobiology 23, 387–98.

Ao, Q., Fung, C.-K., Tsui, A.Y.-P., Cai, S., Zuo, H.-C., Chan, Y.-S., Shum, D.K.-Y., 2011. The regeneration of transected sciatic nerves of adult rats using chitosan nerve conduits seeded with bone marrow stromal cell-derived Schwann cells. Biomaterials 32, 787–96.

Apel, P.J., Garrett, J.P., Sierpinski, P., Ma, J., Atala, A., Smith, T.L., Koman, L.A., Van Dyke, M.E., 2008. Peripheral nerve regeneration using a keratin-based scaffold: Long-term functional and histological outcomes in a mouse model. Journal of Hand Surgery 33, 1541–7.

Aranaz, I., Mengibar, M., Harris, R., Panos, I., Miralles, B., Acosta, N., Galed, G., Heras, A., 2009. Functional characterization of chitin and chitosan. Current Chemical Biology 3, 203–30.

Archibald, S.J., Krarup, C., Shefner, J., Li, S.T., Madison, R.D., 1991. A collagen-based nerve guide conduit for peripheral nerve repair: An electrophysiological study of nerve regeneration in rodents and nonhuman primates. The Journal of Comparative Neurology 306, 685–96.

axogenic.com, 2014. AxoGen [WWW Document]. Azevedo, F.A.C., Carvalho, L.R.B., Grinberg, L.T., Farfel, J.M., Ferretti, R.E.L., Leite, R.E.P., Jacob

Filho, W., Lent, R., Herculano-Houzel, S., 2009. Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled-up primate brain. The Journal of Comparative Neurology 513, 532–41.

Barbosa, J.N., Amaral, I.F., Aguas, A.P., Barbosa, M.A., 2010. Evaluation of the effect of the degree of acetylation on the inflammatory response to 3D porous chitosan scaffolds. Journal of Biomedical Materials Research Part A 93, 20–8.

Barras, F.M., Pasche, P., Bouche, N., Aebischer, P., Zurn, A.D., 2002. Glial cell line-derived neurotrophic factor released by synthetic guidance channels promotes facial nerve regeneration in the rat. Journal of Neuroscience Research 70, 746–55.

Belkas, J.S., Munro, C.A., Shoichet, M.S., Johnston, M., Midha, R., 2005. Long-term in vivo biomechanical properties and biocompatibility of poly(2-hydroxyethyl methacrylate-co-methyl methacrylate) nerve conduits. Biomaterials 26, 1741–9.

Belkas, J.S., Shoichet, M.S., Midha, R., 2004a. Peripheral nerve regeneration through guidance tubes. Neurological Research 26, 151–60.

Belkas, J.S., Shoichet, M.S., Midha, R., 2004b. Axonal guidance channels in peripheral nerve regeneration. Operative Techniques in Orthopaedics 14, 190–8.

References

- 137 -

Bell, J., Haycock, J.W., 2011. Next generation gerve Guides - materials, fabrication, growth factors and cell delivery. Tissue engineering Part B Reviews 44, 1–38.

Bennett, G.J., Xie, Y.-K., 1988. A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 33, 87–107.

Berger, A., Lassner, F., Schaller, E., 1994. The Dellon tube in injuries of peripheral nerves. Handchirurgie, Mikrochirurgie, plastische Chirurgie 26, 44–7.

Berger, M.J., Hubli, M., Krassioukov, A. V, 2014. Sympathetic skin responses and autonomic dysfunction in spinal cord injury. Journal of neurotrauma Online ahe.

Bervar, M., 2000. Video analysis of standing-an alternative footprint analysis to assess functional loss following injury to the rat sciatic nerve. Journal of Neuroscience Methods 102, 109–16.

Bervar, M., 2005. Effect of weak, interrupted sinusoidal low frequency magnetic field on neural regeneration in rats: functional evaluation. Bioelectromagnetics 26, 351–6.

Bini, T.B., Gao, S., Xu, X., Wang, S., Ramakrishna, S., Leong, K.W., 2004. Peripheral nerve regeneration by microbraided poly(L-lactide-co-glycolide) biodegradable polymer fibers. Journal of Biomedical Materials Research. Part A 68, 286–95.

Boeckstyns, M.E.H., Sørensen, A.I., Viñeta, J.F., Rosén, B., Navarro, X., Archibald, S.J., Valss-Solé, J., Moldovan, M., Krarup, C., 2013. Collagen conduit versus microsurgical neurorrhaphy: 2-year follow-up of a prospective, blinded clinical and electrophysiological multicenter randomized, controlled trial. The Journal of Hand Surgery 38, 2405–11.

Boiko, T., Van Wart, A., Caldwell, J.H., Levinson, S.R., Trimmer, J.S., Matthews, G., 2003. Functional specialization of the axon initial segment by isoform-specific sodium channel targeting. The Journal of Neuroscience 23, 2306–13.

Boucher, T., 2001. Neurotrophic factors and neuropathic pain. Current Opinion in Pharmacology 1, 66–72.

Boucher, T.J., Okuse, K., Bennett, D.L., Munson, J.B., Wood, J.N., McMahon, S.B., 2000. Potent analgesic effects of GDNF in neuropathic pain states. Science 290, 124–7.

Boyd, B.S., Puttlitz, C., Noble-Haeusslein, L.J., John, C.M., Trivedi, A., Topp, K.S., 2007. Deviations in gait pattern in experimental models of hindlimb paresis shown by a novel pressure mapping system. Journal of Neuroscience Research 85, 2272–83.

Boyd, J.G., Gordon, T., 2003a. Neurotrophic factors and their receptors in axonal regeneration and functional recovery after peripheral nerve injury. Molecular Neurobiology 27, 277–323.

Boyd, J.G., Gordon, T., 2003b. Glial cell line-derived neurotrophic factor and brain-derived neurotrophic factor sustain the axonal regeneration of chronically axotomized motoneurons in vivo. Experimental Neurology 183, 610–9.

Bozkurt, A., Deumens, R., Scheffel, J., O’Dey, D.M., Weis, J., Joosten, E.A., Führmann, T., Brook, G.A., Pallua, N., 2008. CatWalk gait analysis in assessment of functional recovery after sciatic nerve injury. Journal of Neuroscience Methods 173, 91–8.

Bozkurt, A., Scheffel, J., Brook, G.A., Joosten, E.A., Suschek, C. V, O’Dey, D.M., Pallua, N., Deumens, R., 2011. Aspects of static and dynamic motor function in peripheral nerve regeneration: SSI and CatWalk gait analysis. Behavioural Brain Research 219, 55–62.

Bradke, F., Fawcett, J.W., Spira, M.E., 2012. Assembly of a new growth cone after axotomy: the precursor to axon regeneration. Nature Reviews. Neuroscience 13, 183–93.

Bridges, D., Ahmad, K., Rice, A.S., 2001. The synthetic cannabinoid WIN55,212-2 attenuates hyperalgesia and allodynia in a rat model of neuropathic pain. British Journal of Pharmacology 133, 586–94.

Brodbeck, W.G., Anderson, J.M., 2009. Giant cell formation and function. Current Opinion in Hematology 16, 53–7.

Brooks, D.N., Weber, R. V, Chao, J.D., Rinker, B.D., Zoldos, J., Robichaux, M.R., Ruggeri, S.B., Anderson, K.A., Bonatz, E.E., Wisotsky, S.M., Cho, M.S., Wilson, C., Cooper, E.O., Ingari, J. V, Safa, B., Parrett, B.M., Buncke, G.M., 2012. Processed nerve allografts for peripheral nerve reconstruction: a multicenter study of utilization and outcomes in sensory, mixed, and motor nerve reconstructions. Microsurgery 32, 1–14.

Brushart, T.M., 2011a. Chapther 1: Peripheral nerve anatomy, in: Nerve Repair. Oxford University Press, pp. 3–18.

Brushart, T.M., 2011b. Chapter 9: Nerve regeneration, in: Nerve Repair. p. 253. Brushart, T.M., 2011c. Chapter 9: Nerve Regeneration, in: Nerve Repair. pp. 283–289. Burdick, J.A., Prestwich, G.D., 2011. Hyaluronic acid hydrogels for biomedical applications. Advanced

Materials 23, H41–56.

References

- 138 -

Casals-Díaz, L., Vivó, M., Navarro, X., 2009. Nociceptive responses and spinal plastic changes of afferent C-fibers in three neuropathic pain models induced by sciatic nerve injury in the rat. Experimental Neurology 217, 84–95.

Chalfoun, C.T., Wirth, G.A., Evans, G.R.D., 2006. Tissue engineered nerve constructs: where do we stand? Journal of Cellular and Molecular Medicine 10, 309–17.

Chemnitz, A., Andersson, G., Rosén, B., Dahlin, L.B., Björkman, A., 2013. Poor electroneurography but excellent hand function 31 years after nerve repair in childhood. Neuroreport 24, 6–9.

Chen, R., Cohen, L.G., Hallett, M., 2002. Nervous system reorganization following injury. Neuroscience 111, 761–73.

Chew, S.Y., Mi, R., Hoke, A., Leong, K.W., 2007. Aligned protein-polymer composite fibers enhance nerve regeneration: A potential tissue-engineering platform. Advanced Functional Materials 17, 1288–96.

Chiriac, S., Facca, S., Diaconu, M., Gouzou, S., Liverneaux, P., 2012. Experience of using the bioresorbable copolyester poly(DL-lactide-ε-caprolactone) nerve conduit guide NeurolacTM for nerve repair in peripheral nerve defects: report on a series of 28 lesions. The Journal of Hand Surgery, European Volume 37, 342–9.

Ciaramitaro, P., Mondelli, M., Logullo, F., Grimaldi, S., Battiston, B., Sard, A., Scarinzi, C., Migliaretti, G., Faccani, G., Cocito, D., 2010. Traumatic peripheral nerve injuries: epidemiological findings, neuropathic pain and quality of life in 158 patients. Journal of the Peripheral Nervous System 15, 120–7.

Ciardelli, G., Chiono, V., 2006. Materials for peripheral nerve regeneration. Macromolecular Bioscience 6, 13–26.

Clements, I.P., Kim, Y., English, A.W., Lu, X., Chung, A., Bellamkonda, R. V, 2009. Thin-film enhanced nerve guidance channels for peripheral nerve repair. Biomaterials 30, 3834–46.

Cobianchi, S., Casals-Diaz, L., Jaramillo, J., Navarro, X., 2013. Differential effects of activity dependent treatments on axonal regeneration and neuropathic pain after peripheral nerve injury. Experimental Neurology 240, 157–67.

Cobianchi, S., de Cruz, J., Navarro, X., 2014. Assessment of sensory thresholds and nociceptive fiber growth after sciatic nerve injury reveals the differential contribution of collateral reinnervation and nerve regeneration to neuropathic pain. Experimental Neurology 255, 1–11.

Costa, L.M., Simões, M.J., Mauricio, A.C., Varejão, S.P., 2009. Methods and protocols in peripheral nerv eregeneration experimental research: Part IV-Kinematic gait analysis to quantify peripheral nerve regeneration in the rat, in: International Review of Neurobiology. pp. 127–39.

Dahlin, L.B., Zhao, Q., Bjursten, L.M., 1995. Nerve regeneration in silicone tubes: distribution of macrophages and interleukin-1β in the formed fibrin matrix. Restorative Neurology and Neuroscience 8, 199–203.

Daly, W., Yao, L., Zeugolis, D., Windebank, A., Pandit, A., 2012. A biomaterials approach to peripheral nerve regeneration: bridging the peripheral nerve gap and enhancing functional recovery. Journal of the Royal Society Interface 9, 202–21.

Daly, W.T., Knight, A.M., Wang, H., de Boer, R., Giusti, G., Dadsetan, M., Spinner, R.J., Yaszemski, M.J., Windebank, A.J., 2013. Comparison and characterization of multiple biomaterial conduits for peripheral nerve repair. Biomaterials 34, 8630–9.

Danielsen, N., Pettmann, B., Vahlsing, H.L., Manthorpe, M., Varon, S., 1989. Fibroblast growth factor effects on peripheral nerve regeneration in a silicone chamber model. Journal of Neuroscience Research 20, 282–89.

De Boer, R., Borntraeger, A., Knight, A.M., Hébert-Blouin, M.-N., Spinner, R.J., Malessy, M.J.A., Yaszemski, M.J., Windebank, A.J., 2012. Short- and long-term peripheral nerve regeneration using a poly-lactic-co-glycolic-acid scaffold containing nerve growth factor and glial cell line-derived neurotrophic factor releasing microspheres. Journal of Biomedical Materials Research Part A 100, 2139–46.

De Ruiter, G.C., Spinner, R.J., Malessy, M.J.A., Moore, M.J., Sorenson, E.J., Currier, B.L., Yaszemski, M.J., Windebank, A.J., 2008. Accuracy of motor axon regeneration across autograft, single-lumen, and multichannel poly(lactic-co-glycolic acid) nerve tubes. Neurosurgery 63, 144–53; discussion 153–5.

De Ruiter, G.C.W., 2009. Misdirection and guidance of regenerating motor axons after experimental nerve injury and repair: Chapter 5: Review of the experimental and clinical literature on nerve tubes for peripheral nerve repair.

De Ruiter, G.C.W., Malessy, M.J.A., Yaszemski, M.J., Windebank, A.J., Spinner, R.J., 2009. Designing ideal conduits for peripheral nerve repair. Neurosurgical Focus 26, E5.

References

- 139 -

Deumens, R., Bozkurt, A., Meek, M.F., Marcus, M.A.E., Joosten, E.A.J., Weis, J., Brook, G.A., 2010. Repairing injured peripheral nerves: Bridging the gap. Progress in Neurobiology 92, 245–76.

Di Summa, P.G., Kalbermatten, D.F., Pralong, E., Raffoul, W., Kingham, P.J., Terenghi, G., 2011. Long-term in vivo regeneration of peripheral nerves through bioengineered nerve grafts. Neuroscience 181, 278–91.

Di Summa, P.G., Kingham, P.J., Raffoul, W., Wiberg, M., Terenghi, G., Kalbermatten, D.F., 2010. Adipose-derived stem cells enhance peripheral nerve regeneration. Journal of Plastic, Reconstructive & Aesthetic Surgery 63, 1544–52.

Diamond, J., Coughlin, M., Macintyre, L., Holmes, M., Visheau, B., 1987. Evidence that endogenous beta nerve growth factor is responsible for the collateral sprouting, but not the regeneration, of nociceptive axons in adult rats. Proceedings of the National Academy of Sciences of the United States of America 84, 6596–600.

Díaz-Flores, L., Gutiérrez, R., Varela, H., Evora, P., Valladares, F., Rodríguez, M., Rancel, N., Alvarez-Argüelles, H., 1995. Contribution of the proximal and distal nerve stumps to peripheral nerve regeneration in silicone chambers. Histology and Histopathology 10, 937–46.

Ding, F., Wu, J., Yang, Y., Hu, W., Zhu, Q., Tang, X., Liu, J., Gu, X., 2010. Use of tissue-engineered nerve grafts consisting of a chitosan/poly(lactic-co-glycolic acid)-based scaffold included with bone marrow mesenchymal cells for bridging 50-mm dog sciatic nerve gaps. Tissue Engineering. Part A 16, 3779–90.

Dorey, K., Amaya, E., 2010. FGF signalling: diverse roles during early vertebrate embryogenesis. Development 137, 3731–42.

Dornseifer, U., Fichter, A.M., Leichtle, S., Wilson, A., Rupp, A., Rodenacker, K., Ninkovic, M., Biemer, E., Machens, H.-G., Matiasek, K., Papadopulos, N.A., 2011. Peripheral nerve reconstruction with collagen tubes filled with denatured autologous muscle tissue in the rat model. Microsurgery 31, 632–41.

Dove, A., 2002. Cell-based therapies go live. Nature Biotechnology 20, 339–43. Dubový, P., 2011. Wallerian degeneration and peripheral nerve conditions for both axonal

regeneration and neuropathic pain induction. Annals of Anatomy 193, 267–75. Dutta, P.K., Dutta, J., Tripathi, V.S., 2004. Chitin and chitosan: Chemistry, properties and

applications. Journal of Scientific & Industrial Research 63, 20–31. Fine, E.G., Decosterd, I., Papaloïzos, M., Zurn, A.D., Aebischer, P., 2002. GDNF and NGF released

by synthetic guidance channels support sciatic nerve regeneration across a long gap. European Journal of Neuroscience 15, 589–601.

Flores, A.J., Lavernia, C.J., Owens, P.W., 2000. Anatomy and physiology of peripheral nerve injury and repair. The Journal Of The American Academy Of Orthopaedic Surgeons 150, 537–41, 551, 550.

Fox, I.K., Mackinnon, S.E., 2011. Adult peripheral nerve disorders: nerve entrapment, repair, transfer, and brachial plexus disorders. Plastic and Reconstructive Surgery 127, 105e–118e.

Freier, T., Jiminez Hamann, M., Katayana, Y., Musoke-Zawedde, P., Piotrowicz, A., Yuan, Y., Shoichet, M.S., 2005a. Chapter 9: Biodegradable polymers in neural tissue engineering, in: Handbook of Biodegradable Polymeric Materials and Their Applications. pp. 1–49.

Freier, T., Koh, H.S., Kazazian, K., Shoichet, M.S., 2005b. Controlling cell adhesion and degradation of chitosan films by N-acetylation. Biomaterials 26, 5872–8.

Freier, T., Montenegro, R., Shan Koh, H., Shoichet, M.S., 2005c. Chitin-based tubes for tissue engineering in the nervous system. Biomaterials 26, 4624–32.

Friedman, A.H., Elias, W.J., Midha, R., 2009. Introduction: peripheral nerve surgery-biology, entrapment, and injuries. Neurosurgical Focus 26, E1.

Frostick, S.P., Yin, Q., Kemp, G.J., 1998. Schwann cells, neurotrophic factors, and peripheral nerve regeneration. Microsurgery 18, 397–405.

Frykman, G.K., 1976. Peripheral nerve injuries in children. The Orthopedic Clinics of North America 7, 701–16.

Fugleholm, K., Schmalbruch, H., Krarup, C., 2000. Post reinnervation maturation of myelinated nerve fibers in the cat tibial nerve: chronic electrophysiological and morphometric studies. Journal of the Peripheral Nervous System 5, 82–95.

Fujimoto, E., Mizoguchi, A., Hanada, K., Yahima, M., Ide, C., 1997. Basic fibroblast growth factor promotes extension of regenerating axons of peripheral nerve. In vivo experiments using a Schwann cell basal lamina tube model. Journal of Neurocytology 26, 511–28.

Gaudet, A.D., Popovich, P.G., Ramer, M.S., 2011. Wallerian degeneration: gaining perspective on inflammatory events after peripheral nerve injury. Journal of Neuroinflammation 8, 110.

References

- 140 -

Gelderd, J.B., Chopin, S.F., 1977. The vertebral level of origin of spinal nerves in the rat. The Anatomical Record 188, 45–7.

Georgiou, M., Bunting, S.C.J., Davies, H.A., Loughlin, A.J., Golding, J.P., Phillips, J.B., 2013. Engineered neural tissue for peripheral nerve repair. Biomaterials 34, 7335–43.

Gonzalez-Perez, F., 2014. Talk (2nd international symposium on peripheral nerve regeneration): “Tubulization with chitosan guides for the repair of long gap peripheral nerve injury in the rat”.

Gordon, T., 2009. The role of neurotrophic factors in nerve regeneration. Neurosurgical Focus 26, E3. Gordon, T., Sulaiman, O., Boyd, J.G., 2003. Experimental strategies to promote functional recovery

after peripheral nerve injuries. Journal of the Peripheral Nervous System 8, 236–50. Grimmelikhuijzen, C.J., Westfall, J.A., 1995. The nervous systems of cnidarians. EXS 72, 7–24. Grothe, C., Haastert, K., Jungnickel, J., 2006. Physiological function and putative therapeutic impact

of the FGF-2 system in peripheral nerve regeneration-lessons from in vivo studies in mice and rats. Brain Research Reviews 51, 293–9.

Grothe, C., Heese, K., Meisinger, C., Wewetzer, K., Kunz, D., Cattini, P., Otten, U., 2000. Expression of interleukin-6 and its receptor in the sciatic nerve and cultured Schwann cells: Relation to 18-kD fibroblast growth factor-2. Brain Research 885, 172–81.

Grothe, C., Meisinger, C., Claus, P., 2001. In vivo expression and localization of the fibroblast growth factor system in the intact and lesioned rat peripheral nerve and spinal ganglia. The Journal of Comparative Neurology 434, 342–57.

Grothe, C., Meisinger, C., Hertenstein, A., Kurz, H., Wewetzer, K., 1996. Expression of fibroblast growth factor-2 and fibroblast growth factor receptor 1 messenger RNAs in spinal ganglia and sciatic nerve: regulation after peripheral nerve lesion. Neuroscience 76, 123–35.

Grothe, C., Nikkhah, G., 2001. The role of basic fibroblast growth factor in peripheral nerve regeneration. Anatomy and Embryology 204, 171–77.

Gu, X., Ding, F., Williams, D.F., 2014. Neural tissue engineering options for peripheral nerve regeneration. Biomaterials 35, 6143–56.

Gu, X., Ding, F., Yang, Y., Liu, J., 2011. Construction of tissue engineered nerve grafts and their application in peripheral nerve regeneration. Progress in Neurobiology 93, 204–30.

Haastert, K., Lipokatic, E., Fischer, M., Timmer, M., Grothe, C., 2006a. Differentially promoted peripheral nerve regeneration by grafted Schwann cells over-expressing different FGF-2 isoforms. Neurobiology of Disease 21, 138–53.

Haastert, K., Mauritz, C., Matthies, C., Grothe, C., 2006b. Autologous adult human Schwann cells genetically modified to provide alternative cellular transplants in peripheral nerve regeneration. Journal of Neurosurgery 104, 778–86.

Haastert-Talini, K., Geuna, S., Dahlin, L.B., Meyer, C., Stenberg, L., Freier, T., Heimann, C., Barwig, C., Pinto, L.F.V., Raimondo, S., Gambarotta, G., Samy, S.R., Sousa, N., Salgado, A.J., Ratzka, A., Wrobel, S., Grothe, C., 2013. Chitosan tubes of varying degrees of acetylation for bridging peripheral nerve defects. Biomaterials 34, 9886–904.

Haastert-Talini, K., Schmitte, R., Korte, N., Klode, D., Ratzka, A., Grothe, C., 2011. Electrical stimulation accelerates axonal and functional peripheral nerve regeneration across long gaps. Journal of Neurotrauma 28, 661–74.

Hadlock, T., Sundback, C., Hunter, D., Cheney, M., Vacanti, J.P., 2000. A polymer foam conduit seeded with Schwann cells promotes guided peripheral nerve regeneration. Tissue Engineering 6, 119–27.

Hadlock, T.A., Sundback, C.A., Hunter, D.A., Vacanti, J.P., Cheney, M.L., 2001. A new artificial nerve graft containing rolled Schwann cell monolayers. Microsurgery 21, 96–101.

Hall, S., 2005. The response to injury in the peripheral nervous system. The Journal of Bone & Joint Surgery 87-B, 1309–19.

Han, H., Ao, Q., Chen, G., Wang, S., Zuo, H., 2010. A novel basic fibroblast growth factor delivery system fabricated with heparin-incorporated fibrin-fibronectin matrices for repairing rat sciatic nerve disruptions. Biotechnology Letters 32, 585–91.

Han, T., New, N., Furuike, T., Tokura, S., Tamura, H., 2011. Methods of N-acetylated chitosan scaffolds and its in vitro biodegradation by lysozyme. Biomedical Science and Engineering 5, 15–23.

Heavner, J.E., de Jong, R.H., 1974. Lidocaine blocking concentrations for B- and C-nerve fibers. Anesthesiology 40, 228–33.

Henderson, C., Phillips, H., Pollock, R., Davies, A., Lemeulle, C., Armanini, M., Simmons, L., Moffet, B., Vandlen, R., Simpson LC Corrected to Simmons, L., Et, A., 1994. GDNF: a potent survival factor for motoneurons present in peripheral nerve and muscle. Science 266, 1062–4.

References

- 141 -

Hernández-Cortés, P., Garrido, J., Cámara, M., Ravassa, F.O., 2010. Failed digital nerve reconstruction by foreign body reaction to Neurolac nerve conduit. Microsurgery 30, 414–6.

Hidaka, Y., Ito, M., Mori, K., Yagasaki, H., Kafrawy, A.H., 1999. Histopathological and immunohistochemical studies of membranes of deacetylated chitin derivatives implanted over rat calvaria. Journal of Biomedical Materials Research 46, 418–23.

Hill, P.S., Apel, P.J., Barnwell, J., Smith, T., Koman, L.A., Atala, A., Van Dyke, M., 2011. Repair of peripheral nerve defects in rabbits using keratin hydrogel scaffolds. Tissue engineering. Part A 17, 1499–505.

Hirata, K., Kawabuchi, M., 2002. Myelin phagocytosis by macrophages and nonmacrophages during Wallerian degeneration. Microscopy research and technique 57, 541–7.

Höke, A., Gordon, T., Zochodne, D.W., Sulaiman, O.A.R., 2002. A decline in glial cell-line-derived neurotrophic factor expression is associated with impaired regeneration after long-term Schwann cell denervation. Experimental Neurology 173, 77–85.

Höke, A., Ho, T., Crawford, T.O., LeBel, C., Hilt, D., Griffin, J.W., 2003. Glial cell line-derived neurotrophic factor alters axon schwann cell units and promotes myelination in unmyelinated nerve fibers. The Journal of Neuroscience 23, 561–7.

Höke, A., Redett, R., Hameed, H., Jari, R., Zhou, C., Li, Z.B., Griffin, J.W., Brushart, T.M., 2006. Schwann cells express motor and sensory phenotypes that regulate axon regeneration. Journal of Neuroscience 26, 9646–55.

Hu, N., Wu, H., Xue, C., Gong, Y., Wu, J., Xiao, Z., Yang, Y., Ding, F., Gu, X., 2013. Long-term outcome of the repair of 50 mm long median nerve defects in rhesus monkeys with marrow mesenchymal stem cells-containing, chitosan-based tissue engineered nerve grafts. Biomaterials 34, 100–11.

Hu, W.J., Eaton, J.W., Ugarova, T.P., Tang, L., 2001. Molecular basis of biomaterial-mediated foreign body reactions. Blood 98, 1231–8.

Hu, X., Huang, J., Ye, Z., Xia, L., Li, M., Lv, B., Shen, X., Luo, Z., 2009. A novel scaffold with longitudinally oriented microchannels promotes peripheral nerve regeneration. Tissue Engineering. Part A 15, 3297–308.

Huang, Y.-C., Huang, Y.-Y., 2006. Biomaterials and strategies for nerve regeneration. Artificial Organs 30, 514–22.

Huelsenbeck, S.C., Rohrbeck, A., Handreck, A., Hellmich, G., Kiaei, E., Roettinger, I., Grothe, C., Just, I., Haastert-Talini, K., 2012. C3 peptide promotes axonal regeneration and functional motor recovery after peripheral nerve injury. Neurotherapeutics 9, 185–98.

Ide, C., 1996. Peripheral nerve regeneration. Neuroscience Research 25, 101–21. Ijkema-Paassen, J., Jansen, K., Gramsbergen, A., Meek, M.F., 2004. Transection of peripheral

nerves, bridging strategies and effect evaluation. Biomaterials 25, 1583–92. Ijkema-Paassen, J., Meek, M.F., Gramsbergen, A., 2002. Reinnervation of muscles after transection

of the sciatic nerve in adult rats. Muscle & Nerve 25, 891–7. Ikeda, M., Uemura, T., Takamatsu, K., Kazuki, K., Tabata, Y., Ikada, Y., Nakamura, H., 2014.

Acceleration of peripheral nerve regeneration using nerve conduits in combination with induced pluripotent stem cell technology and a basic fibroblast growth factor drug delivery system. Journal of Biomedical Materials Research. Part A 102, 1370–8.

Ikeda, T., Ikeda, K., Yamamoto, K., Ishizaki, H., Yoshizawa, Y., Yanagiguchi, K., Yamada, S., Hayashi, Y., 2014. Fabrication and characteristics of chitosan sponge as a tissue engineering scaffold. BioMed Research International 2014. Article ID 786892, 8 pages.

IOM (Institute of Medicine), 2012. Ensuring safe food and medical products through stronger regulatory systems abroad.

Isaacs, J., 2010. Treatment of acute peripheral nerve injuries: current concepts. The Journal of Hand Surgery 35, 491–7.

Itoh, S., Suzuki, M., Yamaguchi, I., Takakuda, K., Kobayashi, H., Shinomiya, K., Tanaka, J., 2003. Development of a nerve scaffold using a tendon chitosan tube. Artificial Organs 27, 1079–88.

Jansen, K., Meek, M.F., van der Werff, J.F.A., van Wachem, P.B., van Luyn, M.J.A., 2004. Long-term regeneration of the rat sciatic nerve through a biodegradable poly(DL-lactide-epsilon-caprolactone) nerve guide: tissue reactions with focus on collagen III/IV reformation. Journal of Biomedical Materials Research. Part A 69, 334–41.

Jiang, X., Lim, S.H., Mao, H.-Q., Chew, S.Y., 2010. Current applications and future perspectives of artificial nerve conduits. Experimental Neurology 223, 86–101.

Josephson, A., Widenfalk, J., Trifunovski, A., Widmer, H.R., Olson, L., Spenger, C., 2001. GDNF and NGF family members and receptors in human fetal and adult spinal cord and dorsal root ganglia. Journal of Comparative Neurology 440, 204–17.

References

- 142 -

Jubran, M., Widenfalk, J., 2003. Repair of peripheral nerve transections with fibrin sealant containing neurotrophic factors. Experimental Neurology 181, 204–12.

Jungnickel, J., Claus, P., Gransalke, K., Timmer, M., Grothe, C., 2004. Targeted disruption of the FGF-2 gene affects the response to peripheral nerve injury. Molecular and Cellular Neurosciences 25, 444–52.

Jungnickel, J., Eckhardt, M., Haastert-Talini, K., Claus, P., Bronzlik, P., Lipokatic-Takacs, E., Maier, H., Gieselmann, V., Grothe, C., 2011. Polysialyltransferase-overexpression in Schwann cells mediates different effects during peripheral nerve regeneration. Glycobiology 22, 107–15.

Jungnickel, J., Haase, K., Konitzer, J., Timmer, M., Grothe, C., 2006. Faster nerve regeneration after sciatic nerve injury in mice over-expressing basic fibroblast growth factor. Journal of Neurobiology 66, 940–8.

Jungnickel, J., Haastert, K., Grzybek, M., Thau, N., Lipokatic-Takacs, E., Ratzka, A., Nölle, A., Claus, P., Grothe, C., 2010. Mice lacking basic fibroblast growth factor showed faster sensory recovery. Experimental Neurology 223, 166–72.

Karabekmez, F.E., Duymaz, A., Moran, S.L., 2009. Early clinical outcomes with the use of decellularized nerve allograft for repair of sensory defects within the hand. Hand 4, 245–9.

Kardas, I., Struszcyk, M.H., Kucharska, M., van den Broek, L.A.M., van Dam, J.E.G., Ciechanska, D., 2012. Chapter 11: Chitin and chitosan as functional biopolymers for industrial applications, in: The European Polysacchride Network of Excellence (EPNOE). pp. 329–73.

Kean, T., Thanou, M., 2010. Biodegradation, biodistribution and toxicity of chitosan. Advanced Drug Delivery Reviews 62, 3–11.

Kehoe, S., Zhang, X.F., Boyd, D., 2012. FDA approved guidance conduits and wraps for peripheral nerve injury: A review of materials and efficacy. Injury 43, 553–72.

Khaing, Z.Z., Schmidt, C.E., 2012. Advances in natural biomaterials for nerve tissue repair. Neuroscience Letters 519, 103–14.

Kiefer, R., Kieseier, B.C., Stoll, G., Hartung, H.P., 2001. The role of macrophages in immune-mediated damage to the peripheral nervous system. Progress in Neurobiology 64, 109–27.

Kiefer, R., Lindholm, D., Kreutzberg, G.W., 1993. Interleukin-6 and transforming growth factor-beta 1 mRNAs are induced in rat facial nucleus following motoneuron axotomy. The European Journal of Neuroscience 5, 775–81.

Kimura, J., 1984. Principles and pitfalls of nerve conduction studies. Annals of neurology 16, 415–29. Kingery, W.S., Lu, J.D., Roffers, J.A., Kell, D.R., 1994. The resolution of neuropathic hyperalgesia

following motor and sensory functional recovery in sciatic axonotmetic mononeuropathies. Pain 58, 157–68.

Kingery, W.S., Vallin, J.A., 1989. The development of chronic mechanical hyperalgesia, autotomy and collateral sprouting following sciatic nerve section in rat. Pain 38, 321–32.

Kirsch, M., Terheggen, U., Hofmann, H.D., 2003. Ciliary neurotrophic factor is an early lesion-induced retrograde signal for axotomized facial motoneurons. Molecular and Cellular Neuroscience 24, 130–8.

Kisseleva, T., Brenner, D.A., 2008. Mechanisms of fibrogenesis. Experimental Biology and Medicine 233, 109–22.

Klimaschewski, L., Hausott, B., Angelov, D.N., 2013. The pros and cons of growth factors and cytokines in peripheral nerve regeneration, in: International Review of Neurobiology. pp. 137– 171.

Kokai, L.E., Bourbeau, D., Weber, D., McAtee, J., Marra, K.G., 2011. Sustained growth factor delivery promotes axonal regeneration in long gap peripheral nerve repair. Tissue Engineering. Part A 17, 1263–75.

Kokai, L.E., Ghaznavi, A.M., Marra, K.G., 2010. Incorporation of double-walled microspheres into polymer nerve guides for the sustained delivery of glial cell line-derived neurotrophic factor. Biomaterials 31, 2313–22.

Korte, N., Schenk, H.C., Grothe, C., Tipold, A., Haastert-Talini, K., 2011. Evaluation of periodic electrodiagnostic measurements to monitor motor recovery after different peripheral nerve lesions in the rat. Muscle & Nerve 44, 63–73.

Kramer, D.B., Xu, S., Kesselheim, A.S., 2010. Regulation of medical devices in the United States and European Union. The New England Jounal of Medicine 1–8.

Kreutzberg, G.W., 1995. Reaction of the neuronal cell body to axonal damage., in: TThe Axon: Structure, Function and Pathophysiology. Oxford University Press, Oxford, pp. 355–374.

Laurent, J.P., Lee, R.T., 1994. Topical Review: Birth-elated upper brachial plexus injuries in infants: operative and nonoperative approaches. Journal of Child Neurology 9, 111–7.

References

- 143 -

Lekholm, C., Sundkvist, G., Lundborg, G., Dahlin, L., 2001. The diabetic hand-complications of diabetes. Lakartidningen 98, 306–12.

Levi-Montalcini, R., 1987. The nerve growth factor: thirty-five years later. The EMBO Journal 6, 1145–54.

Levi-Montalcini, R., Angeletti, P.U., 1968. Nerve growth factor. Physiological Reviews 48, 534–69. Li, L., Wu, W., Lin, L.F., Lei, M., Oppenheim, R.W., Houenou, L.J., 1995. Rescue of adult mouse

motoneurons from injury-induced cell death by glial cell line-derived neurotrophic factor. Proceedings of the National Academy of Sciences of the United States of America 92, 9771–5.

Li, Q., Ping, P., Jiang, H., Liu, K., 2006. Nerve conduit filled with GDNF gene-modified Schwann cells enhances regeneration of the peripheral nerve. Microsurgery 26, 116–21.

Liebermann, A.R., 1971. The axon reaction: A review of the principal features of perikaryal repsonses to axon injury. International Review of Neurobiology 14, 49–124.

Lin, F.M., Pan, Y.C., Hom, C., Sabbahi, M., Shenaq, S., 1996. Ankle stance angle: a functional index for the evaluation of sciatic nerve recovery after complete transection. Journal of Reconstructive Microsurgery 12, 173–7.

Lin, S., Xu, J., Hu, S., Xu, L., Zhang, C., Wang, Y., Gu, Y., 2011. Combined application of neutrophin-3 gene and neural stem cells is ameliorative to delay of denervated skeletal muscular atrophy after tibial nerve transection in rats. Cell Transplantation 20, 381–90.

Lin, Y.-C., Marra, K.G., 2012. Injectable systems and implantable conduits for peripheral nerve repair. Biomedical Materials 7, 9 pages.

Liodaki, E., Bos, I., Lohmeyer, J.A., Senyaman, O., Mauss, K.L., Siemers, F., Mailaender, P., Stang, F., 2013. Removal of collagen nerve conduits (NeuraGen) after unsuccessful implantation: focus on histological findings. Journal of Reconstructive mMcrosurgery 29, 517–22.

Lohmeyer, J.A., Shen, Z.-L., Walter, G.F., Berger, A., 2007. Bridging extended nerve defects with an artifcial nerve graft containing Schwann cells pre-seeded on polyglactin filaments. The International Journal of Artificial Organs 30, 64–74.

Luis, A.L., Rodrigues, J.M., Amado, S., Veloso, A.P., Armada-Da-Silva, P.A.S., Raimondo, S., Fregnan, F., Ferreira, A.J., Lopes, M.A., Santos, J.D., Geuna, S., Varejão, A.S.P., Maurício, A.C., 2007. PLGA 90/10 and caprolactone biodegradable nerve guides for the reconstruction of the rat sciatic nerve. Microsurgery 27, 125–37.

Luís, A.L., Rodrigues, J.M., Geuna, S., Amado, S., Simões, M.J., Fregnan, F., Ferreira, A.J., Veloso, A.P., Armada-da-Silva, P.A.S., Varejão, A.S.P., Maurício, A.C., 2008. Neural cell transplantation effects on sciatic nerve regeneration after a standardized crush injury in the rat. Microsurgery 28, 458–70.

Lundborg, G., 2000. A 25-year perspective of peripheral nerve surgery: Evolving neuroscientific concepts and clinical significance. Journal of Hand Surgery 25, 391–414.

Lundborg, G., Dahlin, L.B., Danielsen, N.P., Hansson, H.A., Larsson, K., 1981. Reorganization and orientation of regenerating nerve fibres, perineurium, and epineurium in preformed mesothelial tubes - an experimental study on the sciatic nerve of rats. Journal of Neuroscience Research 6, 265–81.

Ma, Z., Gao, C., Yuan, J., Ji, J., Gong, Y., Shen, J., 2002. Surface modification of poly-L-lactide by photografting of hydrophilic polymers towards improving its hydrophilicity. Journal of Applied Polymer Science 85, 2163–71.

Maciel, F.O., Viterbo, F., Chinaque, L.D.F.C., Souza, B.M., 2013. Effect of electrical stimulation of the cranial tibial muscle after end-to-side neurorrhaphy of the peroneal nerve in rats. Acta Cirúrgica Brasileira 28, 39–47.

Mackinnon, S.E., Doolabh, V.B., Novak, C.B., Trulock, E.P., 2001. Clinical outcome following nerve allograft transplantation. Plastic and Reconstructive Surgery 107, 1419–29.

Madduri, S., Gander, B., 2010. Schwann cell delivery of neurotrophic factors for peripheral nerve regeneration. Journal of the Peripheral Nervous System 15, 93–103.

Madduri, S., Papaloïzos, M., Gander, B., 2009. Synergistic effect of GDNF and NGF on axonal branching and elongation in vitro. Neuroscience Research 65, 88–97.

Madduri, S., Papaloïzos, M., Gander, B., 2010. Trophically and topographically functionalized silk fibroin nerve conduits for guided peripheral nerve regeneration. Biomaterials 31, 2323–34.

Madihally, S. V., Matthew, H.W.T., 1999. Porous chitosan scaffolds for tissue engineering. Biomaterials 20, 1133–42.

Madison, R.D., da Silva, C., Dikkes, P., Sidman, R.L., Chiu, T.-H., 1987. Peripheral nerve regeneration with entubulation repair: Comparison of biodegradeable nerve guides versus polyethylene tubes and the effects of a laminin-containing gel. Experimental Neurology 95, 378–90.

References

- 144 -

Mainil-Varlet, P., Gogolewski, S., Nieuwenhuis, P., 1996. Long-term soft tissue reaction to various polylactides and their in vivo degradation. Journal of Materials Science: Materials in Medicine 7, 713–21.

Maluf-Meiken, L.C. V, Silva, D.R.M., Duek, E.A.R., Alberto-Rincon, M.C., 2006. Morphometrical analysis of multinucleated giant cells in subdermal implants of poly-lactic acid in rats. Journal of Materials Science: Materials in Medicine 17, 481–5.

Mancuso, R., Santos-Nogueira, E., Osta, R., Navarro, X., 2011. Electrophysiological analysis of a murine model of motoneuron disease. Clinical Neurophysiology 122, 1660–70.

Manuel, M., Zytnicki, D., 2011. Alpha, beta and gamma motoneurons: functional diversity in the motor system’s final pathway. Journal of Integrative Neuroscience 10, 243–76.

Marshall Devor and Ruth Govrin-Lippmann, 1979. Selective regeneration of sensory fibers following nerve crush injury. Experimental Neurology 65, 243–54.

McNally, A.K., Anderson, J.M., 2011. Macrophage fusion and multinucleated giant cells of inflammation. Advances in Experimental Medicine and Biology 713, 97–111.

McNally, A.K., Anderson, J.M., 2011. Foreign body-type multinucleated giant cells induced by interleukin-4 express select lymphocyte co-stimulatory molecules and are phenotypically distinct from osteoclasts and dendritic cells. Experimental and Molecular Pathology 91, 673–81.

Meek, M.F., Coert, J.H., 2008. US Food and Drug Administration/Conformit Europe-approved absorbable nerve conduits for clinical repair of peripheral and cranial nerves. Annals of Plastic Surgery 62, 110–6.

Meek, M.F., Den Dunnen, W.F.A., 2009. Porosity of the wall of a Neurolac nerve conduit hampers nerve regeneration. Microsurgery 29, 473–8.

Meek, M.F., Jansen, K., 2009. Two years after in vivo implantation of poly(DL-lactide-epsilon-caprolactone) nerve guides: has the material finally resorbed? Journal of Biomedical Materials Research Part A 89, 734–8.

Meisinger, C., Grothe, C., 2002. Differential regulation of fibroblast growth factor (FGF)-2 and FGF receptor 1 mRNAs and FGF-2 isoforms in spinal ganglia and sciatic nerve after peripheral nerve lesion. Journal of Neurochemistry 68, 1150–8.

Metz, G.A., Whishaw, I.Q., 2009. The ladder rung walking task: a scoring system and its practical application. Journal of Visualized Experiments 28, 1204.

Meyer, M., Matsuoka, I., Wetmore, C., Olson, L., Thoenen, H., 1992. Enhanced synthesis of brain-derived neurotrophic factor in the lesioned peripheral nerve: different mechanisms are responsible for the regulation of BDNF and NGF mRNA. The Journal of Cell Biology 119, 45–54.

Moore, A.M., Kasukurthi, R., Magill, C.K., Farhadi, H.F., Borschel, G.H., Mackinnon, S.E., 2009. Limitations of conduits in peripheral nerve repairs. Hand 4, 180–6.

Morais, J.M., Papadimitrakopoulos, F., Burgess, D.J., 2010. Biomaterials/tissue interactions: possible solutions to overcome foreign body response. The AAPS Journal 12, 188–96.

Mourad, P.D., Lazar, D.A., Curra, F.P., Mohr, B.C., Andrus, K.C., Avellino, A.M., McNutt, L.D., Crum, L.A., Kliot, M., 2001. Ultrasound accelerates functional recovery after peripheral nerve damage. Neurosurgery 48, 1136–1141.

Mukhatyar, V., Pai, B., Clements, I., Srinivasan, A., Huber, R., Mehta, A., Mukhopadaya, S., Rudra, S., Patel, G., Karumbaiah, L., Bellamkonda, R., 2014. Molecular sequelae of topographically guided peripheral nerve repair. Annals of Biomedical Engineering 42, 1436–55.

Murinson, B.B., Hoffman, P.N., Banihashemi, M.R., Meyer, R.A., Griffin, J.W., 2005. C-fiber (Remak) bundles contain both isolectin B4-binding and calcitonin gene-related peptide-positive axons. The Journal of Comparative Neurology 484, 392–402.

Navarro, X., Butí, M., Verdú, E., 1994. Autotomy prevention by amitriptyline after peripheral nerve section in different strains of mice. Restorative Neurology and Neuroscience 6, 151–7.

Navarro, X., Udina, E., 2009. Chapter 6: Methods and protocols in peripheral nerve regeneration experimental research: part III-electrophysiological evaluation., in: International Review of Neurobiology. pp. 105–26.

Navarro, X., Vivó, M., Valero-Cabré, A., 2007. Neural plasticity after peripheral nerve injury and regeneration. Progress in Neurobiology 82, 163–201.

Navissano, M., Malan, F., Carnino, R., Battiston, B., 2005. Neurotube for facial nerve repair. Microsurgery 25, 268–71.

Necas, J., Bartosikova, L., Brauner, P., Kolar, J., 2008. Hyaluronic acid (hyaluronan): a review. Veterinarni Medicina 53, 397–411.

Nectow, A.R., Marra, K.G., Kaplan, D.L., 2011. Biomaterials for the development of peripheral nerve guidance conduits. Tissue Engineering Part B Reviews 18, 40–50.

References

- 145 -

Nichols, C.M., Brenner, M.J., Fox, I.K., Tung, T.H., Hunter, D.A., Rickman, S.R., Mackinnon, S.E., 2004. Effects of motor versus sensory nerve grafts on peripheral nerve regeneration. Experimental Neurology 190, 347–55.

Nickel, P., Schummer, A., Seiferle, E., 2004. Peripheres Nervensystem, in: Lehrbuch Der Anatomie Der Haustiere Band IV. pp. 228–384.

Oh, S.H., Kim, J.H., Song, K.S., Jeon, B.H., Yoon, J.H., Seo, T.B., Namgung, U., Lee, I.W., Lee, J.H., 2008. Peripheral nerve regeneration within an asymmetrically porous PLGA/Pluronic F127 nerve guide conduit. Biomaterials 29, 1601–9.

Okamoto, H., Mizuno, K., Horio, T., 2003. Langhans-type and foreign-body-type multinucleated giant cells in cutaneous lesions of sarcoidosis. Acta Dermato-Venereologica 83, 171–4.

Oppenheim, R., Bartheld, C. von, 2008. Chapter 19: Programmed cell death and neurotrophic factors, in: Fundamental Neuroscience. pp. 438–444.

Oppenheim, R.W., Houenou, L.J., Johnson, J.E., Lin, L.F., Li, L., Lo, A.C., Newsome, A.L., Prevette, D.M., Wang, S., 1995. Developing motor neurons rescued from programmed and axotomy-induced cell death by GDNF. Nature 373, 344–6.

Oppenheim, R.W., von Bartheld, C.S., 2008. Chapter 19: Programmed cell death and neurotrophic factors, in: Fundamental Neuroscience. pp. 450–452.

Ornitz, D.M., Itoh, N., 2001. Fibroblast growth factors. Genome Biology 2, REVIEW3005. Otto, D., Unsicker, K., Grothe, C., 1987. Pharmacological effects of nerve growth factor and fibroblast

growth factor applied to the transectioned sciatic nerve on neuron death in adult rat dorsal root ganglia. Neuroscience Letters 83, 156–60.

Pabari, A., Yang, S., Mosahebi, A., Seifalian, A., 2011. Recent advances in artificial nerve conduit design: strategies for the delivery of luminal fillers. Journal of Controlled Release 156, 2–10.

Pabari, A., Yang, S., Seifalian, A., Mosahebi, A., 2010. Modern surgical management of peripheral nerve gap. Clinics in Plastic Surgery 63, 1941–8.

Panseri, S., Cunha, C., Lowery, J., Del Carro, U., Taraballi, F., Amadio, S., Vescovi, A., Gelain, F., 2008. Electrospun micro- and nanofiber tubes for functional nervous regeneration in sciatic nerve transections. BMC Biotechnology 8, 39.

Pardo, I.D., Garman, R.H., Weber, K., Bobrowski, W.F., Hardisty, J.F., Morton, D., 2012. Technical guide for nervous system sampling of the cynomolgus monkey for general toxicity studies. Toxicologic Pathology 40, 624–36.

Park, J.-S., Höke, A., 2014. Treadmill exercise induced functional recovery after peripheral nerve repair is associated with increased levels of neurotrophic factors. PloS One 9, e90245.

Patel, M., Mao, L., Wu, B., Vandevord, P.J., 2007. GDNF-chitosan blended nerve guides: a functional study. Journal of Tissue Engineering and Regenerative Medicine 1, 360–7.

Patel, M., Vandevord, P.J., Matthew, H., Wu, B., DeSilva, S., Wooley, P.H., 2006. Video-gait analysis of functional recovery of nerve repaired with chitosan nerve guides. Tissue Engineering 12, 3189–99.

Pfister, B.J., Gordon, T., Loverde, J.R., Kochar, A.S., Mackinnon, S.E., Cullen, D.K., 2011. Biomedical engineering strategies for peripheral nerve repair: Surgical applications, state of the art, and future challenges. Critical Reviews in Biomedical Engineering 39, 81–124.

Pfister, L.A., Papaloïzos, M., Merkle, H.P., Gander, B., 2007. Nerve conduits and growth factor delivery in peripheral nerve repair. Journal of the Peripheral Nervous System 12, 65–82.

Pistner, H., Gutwald, R., Ordung, R., Reuther, J., Mühling, J., 1993. Poly(L-lactide): a long-term degradation study in vivo. I. Biological results. Biomaterials 14, 671–7.

Polyganics.com, 2014. Polyganics-Bioresorbable medical devices [WWW Document]. Popesko, P., Rajtová, V., Horák, J., 1990. A color atlas of anatomy of small laboratory animals.

Volume two: rat, mouse, hamster. Raafat, D., Sahl, H.-G., 2009. Chitosan and its antimicrobial potential-a critical literature survey.

Microbial Biotechnology 2, 186–201. Radtke, C., Aizer, A.A., Agulian, S.K., Lankford, K.L., Vogt, P.M., Kocsis, J.D., 2009. Transplantation

of olfactory ensheathing cells enhances peripheral nerve regeneration after microsurgical nerve repair. Brain Research 1254, 10–7.

Radtke, C., Wewetzer, K., Reimers, K., Vogt, P.M., 2011. Transplantation of olfactory ensheathing cells as adjunct cell therapy for peripheral nerve injury. Cell Transplantation 20, 145–52.

Raimondo, S., Fornaro, M., Tos, P., Battiston, B., Giacobini-Robecchi, M.G., Geuna, S., 2011. Perspectives in regeneration and tissue engineering of peripheral nerves. Annals of Anatomy 193, 334–40.

References

- 146 -

Raisi, A., Azizi, S., Delirezh, N., Heshmatian, B., Amini, K., 2012. Use of chitosan conduit for bridging small-gap peripheral nerve defect in sciatic nerve transection model of rat. Iranian Journal of Veterinary Surgery 5, 89–100.

Ransom, B.R., 2012. Neuroglia (Google eBook). Oxford University Press. Ratzka, A., Kalve, I., Ozer, M., Nobre, A., Wesemann, M., Jungnickel, J., Koster-Patzlaff, C., Baron,

O., Grothe, C., 2012. The co-layer method as an efficient way to genetically modify mesencephalic progenitor cells transplanted into 6-OHDA rat model of Parkinson’s disease. Cell Transplantion 21, 749–62.

Ray, W. Z., Mackinnon, S. E., 2010. Management of nerve gaps: Autografts, allografts, nerve transfers, and end-to-side neurorrhaphy. Experimental Neurology 223, 77–85.

Reichert, F., Saada, A., Rotshenker, S., 1994. Peripheral nerve injury induces Schwann cells to express two macrophage phenotypes: phagocytosis and the galactose-specific lectin MAC-2. The Journal of Neuroscience 14, 3231–45.

Rhee, H.J., Hillebrands, W., Daems, W.T., 1978. Are Langhans giant cells precursors of foreign-body giant cells? Archives of Dermatological Research 263, 13–21.

Rigaud, M., Gemes, G., Barabas, M.-E., Chernoff, D.I., Abram, S.E., Stucky, C.L., Hogan, Q.H., 2008. Species and strain differences in rodent sciatic nerve anatomy: implications for studies of neuropathic pain. Pain 136, 188–201.

Robinson, L.R., 2000. Traumatic injury to peripheral nerves. Muscle & Nerve 23, 863–73. Rochkind, S., El-Ani, D., Nevo, Z., Shahar, A., 2009. Increase of neuronal sprouting and migration

using 780 nm laser phototherapy as procedure for cell therapy. Lasers in Surgery and Medicine 41, 277–81.

Rumpel, R., Alam, M., Klein, A., Oezer, M., Wesemann, M., Jin, X., Krauss, J.K., Schwabe, K., Ratzka, A., Grothe, C., 2013. Neuronal firing activity and gene expression changes in the subthalamic nucleus after transplantation of dopamine neurons in hemiparkinsonian rats. Neurobiology of Disease 59, 230–243.

Rupp, A., Dornseifer, U., Fischer, A., Schmahl, W., Rodenacker, K., Jütting, U., Gais, P., Biemer, E., Papadopulos, N., Matiasek, K., 2007. Electrophysiologic assessment of sciatic nerve regeneration in the rat: surrounding limb muscles feature strongly in recordings from the gastrocnemius muscle. Journal of Neuroscience Methods 166, 266–77.

Saito, H., Dahlin, L.B., 2008. Expression of ATF3 and axonal outgrowth are impaired after delayed nerve repair. BMC Neuroscience 9, 88.

Samadian, M., Rezaee, O., Haddadian, K., Sharifi, G., Abtahi, H., Hamidian, M., Khormaee, F., Sodagari, F., 2009. Gunshot injuries to the brachial plexus during wartime. British Journal of Neurosurgery 23, 165–9.

Sandkühler, J., 2009. Models and mechanisms of hyperalgesia and allodynia. Physiological Reviews 89, 707–58.

Santosa, K.B., Jesuraj, N.J., Viader, A., MacEwan, M., Newton, P., Hunter, D.A., Mackinnon, S.E., Johnson, P.J., 2013. Nerve allografts supplemented with Schwann cells overexpressing glial-cell-line-derived neurotrophic factor. Muscle & Nerve 47, 213–23.

Sarikcioglu, L., Demirel, B.M., Utuk, A., 2009. Walking track analysis: an assessment method for functional recovery after sciatic nerve injury in the rat. Folia Morphologica 68, 1–7.

Scheib, J., Höke, A., 2013. Advances in peripheral nerve regeneration. Nature Reviews. Neurology. 9, 668–76.

Scholz, T., Sumarto, A., Krichevsky, A., Evans, G.R.D., 2011. Neuronal differentiation of human adipose tissue-derived stem cells for peripheral nerve regeneration in vivo. Archives of Surgery 146, 666–74.

Seckel, B.R., 1990. Enhancement of peripheral nerve regeneration. Muscle & Nerve 13, 785–800. Seddon, H.J., 1942. A Classification of Nerve Injuries. British Medical Journal 2, 237–9. Seddon, H.J., Medawar, P.B., Smith, H., 1943. Rate of regeneration of peripheral nerves in man. The

Journal of Physiology 102, 191–215. Selden, P.A., Nudds, J.R., 2004. Chapter 1: Ediacara, in: Evolution of Fossil Ecosystems. p. 10. Shahar, A., Nevo, Z., Rochkind, S., 2001. Cross-linked hyaluronic acid-laminin gels and use thereof in

cell culture and medical implants. Shen, H., Shen, Z.-L., Zhang, P.-H., Chen, N.-L., Wang, Y.-C., Zhang, Z.-F., Jin, Y.-Q., 2010. Ciliary

neurotrophic factor-coated polylactic-polyglycolic acid chitosan nerve conduit promotes peripheral nerve regeneration in canine tibial nerve defect repair. Journal of Biomedical Materials Research. Part B, Applied biomaterials 95, 161–70.

Shen, X., Ruan, J., Zhou, Z., Zeng, Z., Xie, L., 2009. Evaluation of PLGA/chitosan/HA conduits for nerve tissue reconstruction. Journal of Wuhan University of Technology 24, 566–70.

References

- 147 -

Shenaq, J.M., Shenaq, S.M., Spira, M., 1989. Reliability of sciatic function index in assessing nerve regeneration across a 1 cm gap. Microsurgery 10, 214–9.

Siemionow, M., Bozkurt, M., Zor, F., 2010. Regeneration and repair of peripheral nerves with different biomaterials: review. Microsurgery 30, 574–88.

Sierpinski Hill, P., 2009. Regeneration of peripheral nerves using neuroinductive biomaterial scaffolds. Simões, M., Gärtner, A., Shirosaki, Y., Costa, R.G. da, Cortez, P., Gartnër, F., Santos, J., Lopes, M.,

Geuna, S., Varejão, A., Maurício, A., 2011. In vitro and in vivo chitosan for peripheral nerve reconstruction. Acta Medica Portuguesa 24, 43–52.

Simões, M.J., Amado, S., Gärtner, A., Armada-Da-Silva, P.A.S., Raimondo, S., Vieira, M., Luís, A.L., Shirosaki, Y., Veloso, A.P., Santos, J.D., Varejão, A.S.P., Geuna, S., Maurício, A.C., 2010. Use of chitosan scaffolds for repairing rat sciatic nerve defects. Italian Journal of Anatomy and Embryology 115, 190–210.

Singh, D.K., Ray, A.R., 2000. Biomedical applications of chitin, chitosan, and their derivatives. Journal of Macromolecular Science, Part C: Polymer Reviews 40, 69–83.

Sinis, N., Schaller, H.-E., Schulte-Eversum, C., Schlosshauer, B., Doser, M., Dietz, K., Rösner, H., Müller, H.-W., Haerle, M., 2005. Nerve regeneration across a 2-cm gap in the rat median nerve using a resorbable nerve conduit filled with Schwann cells. Journal of Neurosurgery 103, 1067–76.

Smetana, K., 1987. Multinucleate foreign-body giant cell formation. Experimental and Molecular Pathology 46, 258–65.

Snider, W.D., 1994. Functions of the neurotrophins during nervous system development: what the knockouts are teaching us. Cell 77, 627–38.

Sotgiu, M.L., Biella, G., 1995. Spinal expansion of saphenous afferents after sciatic nerve constriction in rats. Neuroreport 6, 2305–8.

Stoll, G., Jander, S., Myers, R.R., 2002. Degeneration and regeneration of the peripheral nervous system: from Augustus Waller’s observations to neuroinflammation. Journal of the peripheral nervous system 7, 13–27.

Strauch, B., Rodriguez, D.M., Diaz, J., Yu, H.L., Kaplan, G., Weinstein, D.E., 2001. Autologous Schwann cells drive regeneration through a 6-cm autogenous venous nerve conduit. Journal of Reconstructive Microsurgery 17, 589–95; discussion 596–7.

Strusycyzk, M.H., Struszczyk, K.J., 2007. Medical applications of chitin and its derivatives, in: Polish Chitin Society. pp. 139– 147.

Sun, W., Miao, B., Wang, X.-C., Duan, J.-H., Wang, W.-T., Kuang, F., Xie, R.-G., Xing, J.-L., Xu, H., Song, X.-J., Luo, C., Hu, S.-J., 2012. Reduced conduction failure of the main axon of polymodal nociceptive C-fibres contributes to painful diabetic neuropathy in rats. Brain 135, 359–75.

Sunderland, S., 1951. A classification of peripheral nerve injuries producing loss of function. Brain 74, 491–516.

Sunderland, S., 1990. The anatomy and physiology of nerve injury. Muscle & Nerve 13, 771–84. Svennigsen, Å., Dahlin, L., 2013. Repair of the Peripheral Nerve-Remyelination that Works. Brain

Sciences 3, 1182–97. Swett, J.E., Torigoe, Y., Elie, V.R., Bourassa, C.M., Miller, P.G., 1991. Sensory neurons of the rat

sciatic nerve. Experimental Neurology 114, 82–103. Swett, J.E., Wikholm, R.P., Blanks, R.H., Swett, A.L., Conley, L.C., 1986. Motoneurons of the rat

sciatic nerve. Experimental Neurology 93, 227–52. Swindle-Reilly, K.E., Papke, J.B., Kutosky, H.P., Throm, A., Hammer, J.A., Harkins, A.B., Willits, R.K.,

2012. The impact of laminin on 3D neurite extension in collagen gels. Journal of Neural Engineering 9, 046007.

Tajima, T., Imai, H., 1989. Results of median nerve repair in children. Microsurgery 10, 145–6. Thevenot, P., Hu, W., Tang, L., 2008. Surface chemistry influences implant biocompatibility. Current

Topics in Medicinal Chemistry 8, 270–80. Thomas, P.K., 1989. Invited review: focal nerve injury: guidance factors during axonal regeneration.

Muscle & Nerve 12, 796–802. Thorsén, F., Rosberg, H.-E., Steen Carlsson, K., Dahlin, L.B., 2012. Digital nerve injuries:

epidemiology, results, costs, and impact on daily life. Journal of Plastic Surgery and Hand Surgery 46, 184–90.

Timmer, M., Robben, S., Müller-Ostermeyer, F., Nikkhah, G., Grothe, C., 2003. Axonal regeneration across long gaps in silicone chambers filled with Schwann cells overexpressing high molecular weight FGF-2. Cell Transplantation 12, 265–77.

Tomihata, K., Ikada, Y., 1997. In vitro and in vivo degradation of films of chitin and its deacetylated derivatives. Biomaterials 18, 567–75.

References

- 148 -

Tonra, J.R., Curtis, R., Wong, V., Cliffer, K.D., Park, J.S., Timmes, A., Nguyen, T., Lindsay, R.M.,

Acheson, A., DiStefano, P.S., 1998. Axotomy upregulates the anterograde transport and expression of brain-derived neurotrophic factor by sensory neurons. The Journal of Neuroscience 18, 4374–83.

Törnqvist, E., Aldskogius, H., 1994. Motoneuron survival is not affected by the proximo-distal level of axotomy but by the possibility of regenerating axons to gain access to the distal nerve stump. Journal of Neuroscience Research 39, 159–65.

Udina, E., Rodríguez, F.J., Verdú, E., Espejo, M., Gold, B.G., Navarro, X., 2004. FK506 enhances regeneration of axons across long peripheral nerve gaps repaired with collagen guides seeded with allogeneic Schwann cells. Glia 47, 120–9.

Ueda, H., 2006. Molecular mechanisms of neuropathic pain-phenotypic switch and initiation mechanisms. Pharmacology and Therapeutics.

Ushiki, T., 2002. Collagen fibers, reticular fibers and elastic fibers. A comprehensive understanding from a morphological viewpoint. Archives of Histology and Cytology 65, 109–26.

Valentini, R.F., Aebischer, P., Winn, S.R., Galletti, P.M., 1987. Collagen- and laminin-containing gels impede peripheral nerve regeneration through semipermeable nerve guidance channels. Experimental Neurology 98, 350–6.

Valero-Cabré, A., Tsironis, K., Skouras, E., Navarro, X., Neiss, W.F., 2004. Peripheral and spinal motor reorganization after nerve injury and repair. Journal of Neurotrauma 21, 95–108.

Valero-Cabré, A., Tsironis, K., Skouras, E., Perego, G., Navarro, X., Neiss, W.F., 2001. Superior muscle reinnervation after autologous nerve graft or poly-L-lactide-epsilon-caprolactone (PLC) tube implantation in comparison to silicone tube repair. Journal of Neuroscience Research 63, 214–23.

Varejão, A.S.P., Cabrita, A.M., Meek, M.F., Fornaro, M., Geuna, S., Giacobini-Robecchi, M.G., 2003. Morphology of nerve fiber regeneration along a biodegradable poly (DLLA-epsilon-CL) nerve guide filled with fresh skeletal muscle. Microsurgery 23, 338–45.

Verdu, E., Ceballos, D., Vilches, J.J., Navarro, X., 2000. Influence of aging on peripheral nerve function and regeneration. Journal of the Peripheral Nervous System 5, 191–208.

Waitayawinyu, T., Parisi, D.M., Miller, B., Luria, S., Morton, H.J., Chin, S.H., Trumble, T.E., 2007. A comparison of polyglycolic acid versus type 1 collagen bioabsorbable nerve conduits in a rat model: an alternative to autografting. The Journal of Hand Surgery 32, 1521–9.

Waldram, M., 2003. Peripheral nerve injuries. Trauma 5, 79–96. Wang, G., Lu, G., Ao, Q., Gong, Y., Zhang, X., 2010. Preparation of cross-linked carboxymethyl

chitosan for repairing sciatic nerve injury in rats. Biotechnology Letters 32, 59–66. Wang, H., Spinner, R.J., Sorenson, E.J., Windebank, A.J., 2008. Measurement of forelimb function by

digital video motion analysis in rat nerve transection models. Journal of the peripheral nervous system : JPNS 13, 92–102.

Wang, K.K., Nemeth, I.R., Seckel, B.R., Chakalis-Haley, D.P., Swann, D.A., Kuo, J.W., Bryan, D.J., Cetrulo, C.L., 1998. Hyaluronic acid enhances peripheral nerve regeneration in vivo. Microsurgery 18, 270–5.

Wang, S., Cai, Q., Hou, J., Bei, J., Zhang, T., Yang, J., Wan, Y., 2003. Acceleration effect of basic fibroblast growth factor on the regeneration of peripheral nerve through a 15-mm gap. Journal of Biomedical Materials Research. Part A 66, 522–31.

Wang, S., Wan, A.C., Xu, X., Gao, S., Mao, H.Q., Leong, K.W., Yu, H., 2001. A new nerve guide conduit material composed of a biodegradable poly(phosphoester). Biomaterials 22, 1157–69.

Wang, X., Hu, W., Cao, Y., Yao, J., Wu, J., Gu, X., 2005. Dog sciatic nerve regeneration across a 30-mm defect bridged by a chitosan/PGA artificial nerve graft. Brain 128, 1897–910.

Waterman, E.A., Sakai, N., Nguyen, N.T., Horst, B.A.J., Veitch, D.P., Dey, C.N., Ortiz-Urda, S., Khavari, P.A., Marinkovich, M.P., 2007. A laminin-collagen complex drives human epidermal carcinogenesis through phosphoinositol-3-kinase activation. Cancer Research 67, 4264–70.

Wei, Y., Gong, K., Zheng, Z., Wang, A., Ao, Q., Gong, Y., Zhang, X., 2011. Chitosan/silk fibroin-based tissue-engineered graft seeded with adipose-derived stem cells enhances nerve regeneration in a rat model. Journal of Materials Science. Materials in Medicine 22, 1947–64.

Wenling, C., Duohui, J., Jiamou, L., Yandao, G., Nanming, Z., Xiufang, Z., 2005. Effects of the degree of deacetylation on the physicochemical properties and Schwann cell affinity of chitosan films. Journal of Biomaterials Applications 20, 157–177.

Westfall, I.A., 1996. Ultrastructure of synapses in the first-evolved nervous systems. Journal of Neurocytology 25, 735–46.

References

- 149 -

Williams, L.R., Longo, F.M., Powell, H.C., Lundborg, G., Varon, S., 1983. Spatial-temporal progress of peripheral nerve regeneration within a silicone chamber: parameters for a bioassay. The Journal of Comparative Neurology 218, 460–70.

Wood, M.D., Gordon, T., Kemp, S.W.P., Liu, E.H., Kim, H., Shoichet, M.S., Borschel, G.H., 2013a. Functional motor recovery is improved due to local placement of GDNF microspheres after delayed nerve repair. Biotechnology and Bioengineering 5, 1272–81.

Wood, M.D., Gordon, T., Kim, H., Szynkaruk, M., Phua, P., Lafontaine, C., Kemp, S.W., Shoichet, M.S., Borschel, G.H., 2013b. Fibrin gels containing GDNF microspheres increase axonal regeneration after delayed peripheral nerve repair. Regenerative Medicine 8, 27–37.

Woodbury, M.E., Ikezu, T., 2014. Fibroblast growth factor-2 signaling in neurogenesis and neurodegeneration. Journal of Neuroimmune Pharmacology 9, 92–101.

Wrobel, S., Serra, S.C., Ribeiro-Samy, S., Sousa, N., Heimann, C., Barwig, C., Grothe, C., Salgado, A.J., Haastert-Talini, K., 2014. In vitro evaluation of cell-seeded chitosan films for peripheral nerve tissue engineering. Tissue Engineering. Part A Epub ahead.

Xie, F., Li, Q.F., Gu, B., Liu, K., Shen, G.X., 2008. In vitro and in vivo evaluation of a biodegradable chitosan-PLA composite peripheral nerve guide conduit material. Microsurgery 28, 471–479.

Xu, H., Yan, Y., Li, S., 2011. PDLLA/chondroitin sulfate/chitosan/NGF conduits for peripheral nerve regeneration. Biomaterials 32, 4506–16.

Xue, C., Hu, N., Gu, Y., Yang, Y., Liu, Y., Liu, J., Ding, F., Gu, X., 2012. Joint use of a chitosan/PLGA scaffold and MSCs to bridge an extra large gap in dog sciatic nerve. Neurorehabilitation and Neural Repair 26, 96–106.

Yan, Q., Matheson, C., Lopez, O.T., 1995. In vivo neurotrophic effects of GDNF on neonatal and adult facial motor neurons. Nature 373, 341–4.

Yang, Y., Ding, F., Wu, J., Hu, W., Liu, W., Liu, J., Gu, X., 2007. Development and evaluation of silk fibroin-based nerve grafts used for peripheral nerve regeneration. Biomaterials 28, 5526–35.

Yang, Y., Gu, X., Tan, R., Hu, W., Wang, X., Zhang, P., Zhang, T., 2004. Fabrication and properties of a porous chitin/chitosan conduit for nerve regeneration. Biotechnology Letters 26, 1793–7.

Yang, Y., Wu, J., Wang, X., Liu, J., Ding, F., Gu, X., 2009. Fabrication and evaluation of chitin-based nerve guidance conduits used to promote peripheral nerve regeneration. Advanced Engineering Materials 11, B209–B218.

Yuan, Y., Zhang, P., Yang, Y., Wang, X., Gu, X., 2004. The interaction of Schwann cells with chitosan membranes and fibers in vitro. Biomaterials 25, 4273–8.

Zhang, F., Blain, B., Beck, J., Zhang, J., Chen, Z., Chen, Z.-W., Lineaweaver, W.C., 2002. Autogenous venous graft with one-stage prepared Schwann cells as a conduit for repair of long segmental nerve defects. Journal of Reconstructive Microsurgery 18, 295–300.

Zhang, H., Neau, S.H., 2001. In vitro degradation of chitosan by a commercial enzyme preparation: effect of molecular weight and degree of deacetylation. Biomaterials 22, 1653–8.

Zheng, L., Ao, Q., Han, H., Zhang, X., Gong, Y., 2010. Evaluation of the chitosan/glycerol-beta-phosphate disodium salt hydrogel application in peripheral nerve regeneration. Biomedical Materials 5, 35003.

Zheng, L., Cui, H.-F., 2012. Enhancement of nerve regeneration along a chitosan conduit combined with bone marrow mesenchymal stem cells. Journal of Materials Science: Materials in Medicine.

Ziv-Polat, O., Shahar, A., Levy, I., Skaat, H., Neumann, S., Fregnan, F., Geuna, S., Grothe, C., Haastert-Talini, K., Margel, S., 2014. The role of neurotrophic factors conjugated to iron oxide nanoparticles in peripheral nerve regeneration: In vitro studies. BioMed Research International. Article ID 267808, 10 Pages.

Zochodne, D.W., Cheng, C., 2000. Neurotrophins and other growth factors in the regenerative milieu of proximal nerve stump tips. Journal of Anatomy 196, 279–83.

Affidavit

- 150 -

8. Affidavit

I herewith declare that I autonomously carried out the PhD-thesis entitled “Peripheral nerve regeneration using hollow and enriched chitosan-based guidance conduits”.

No third party assistance has been used.

I did not receive any assistance in return for payment by consulting agencies or any other person. No one received any kind of payment for direct or indirect assistance in correlation to the content of the submitted thesis.

I conducted the project at the following institution: Institute of Neuroanatomy Hannover Medical School The thesis has not been submitted elsewhere for an exam, as a thesis or for evaluation in a similar context. I hereby affirm the above statements to be complete and true to the best of my knowledge. date, signature

- 151 -

Acknowledgements

- 152 -

9. Acknowledgements The last three years have been a unique journey, which would not have been the

same without the people accompanying me during this time. Therefore, I would like

to thank them in the next lines.

My first thank you belongs to my main supervisor Prof. Dr. Claudia Grothe for

providing me with the opportunity to be a part of this great EU-project and to work in

the research field of peripheral nerve regeneration. Thank you for your scientific

advice and support especially during the final months of this work.

Thank you to my co-supervisors Prof. Dr. Gerardy-Schahn and Prof. Dr. Schwabe for

your support and helpful input during these years. And thank you to Esther Udina,

MH, PhD, for agreeing to be my external referee for this work.

A special thank you to Prof. Dr. Kirsten Haastert-Talini for all the work you put into

the projects, the uncountable hours spend together doing electrophysiological

recordings and surgeries, laughs we shared, and for always making time to discuss

and answer questions.

Thank you to the whole Medovent-Team for allowing me to gain an insight into the

manufacturing of chitosan-based products, introducing me to carnival (Fasching),

Kräppel and for having a great time in Mainz.

Thank you to the whole BIOHYBRID-consortium that allowed me to meet so many

excellent researchers from all over Europe.

I also would like to thank Maike Wesemann, Silvana Taubeler-Gerling and Hildegard

Streich for all the work you did for the BIOHYBRID-project.

A huge thank you to Natascha Heidrich and Silke Fischer for every single section of

the approximately thousands you did in the course of my studies. I think we can fill a

whole room with all the boxes of sections you cut and stained. No kidding!

Acknowledgements

- 153 -

Thank you to Kerstin Kuhlemann for taking me in to your aisle during my long

journey to find my final desk, all the encouraging words, and helpful comments.

Thank, thank, thank you to Jenny Metzen! We spend so much time together looking

at rat feet and I could not imagine another person I wanted to do this with. Also thank

you for transportation help during rain, snow, and wind and bringing a smile to my

face in so many situations. And let’s not forget the great laughs we had over a glass

of wine.

Thank you to my current and past office-co-owners of the last 1 ½ years. Dr. Regina

Rumpel, Dr. Anna Effenberg, Michela Morano, Dr. Alex Klein, and Dr. Andreas

Ratzka. I’m glad I somehow managed to snatch a seat in your office (ok, I kind of

creped in), because I enjoyed so many moments with all of you. A special thank you

for tolerating the looks of my desk(s) during the last months, but I guess this is how it

has to be shortly before thesis-submission. Thank you to Dr. Andreas Ratzka for

helping me with all things regarding PCR and giving me my very own shrimp. Thank

you Dr. Regina Rumpel for sharing a sparkling weekend with me in Berlin and the

love for certain entertainment shows and books. Thank you Dr. Anna Effenberg for

great movie nights and sometimes talking weird, which always causes laughs. Thank

you to Dr. Alex Klein for all the encouragements regarding work and life. Thank you

Michela for being a part of our team for some months and for always having a smile

on your face when entering the office early in the morning.

Dr. Meltem Hohmann, Inga Stockbrügger, Verena Lübben, Michela Morano, Dr.

Benjamin Förthmann, Dr. Niko Hensel, Dr. Regina Rumpel, Sebastian Rademacher,

Sandra Wrobel, Jenny Metzen, Timo Konen, Maria Stößel thank you all for the so

much needed coffee breaks, candy breaks, cake breaks, pizza breaks as well as all

the evenings we spend together. I would never want to exchange this team for

another.

Thank you to Sandra Wrobel and Natalie Schecker for counting macrophages for

hours, days, and months!

Acknowledgements

- 154 -

A huge thank you to Sandra Wrobel for being my BIOHYBRID-partner of three years.

We shared so many moments and experiences in the office, during travels, and

evenings that I will cherish forever. Thank you for a special boat trip in Barcelona,

loving the Sushi-Groupon as much as I do, and watching almost all matches from the

German soccer team on their road to win the World Cup with me.

Thank you to Mariana Behrmann, Jana Wohltmann, and Nora Schaffarczyk for being

close even when there is an ocean between us. Thank you Mariana for the laughs

we shared on Skype and doing a US trip with me that will be a once in a lifetime

experience. Thank you Jana for all the talks we had, being the best friend one can

ask for and for just being who you are. Thank you Nora for all the videos and photos

that keep me up-to-date when you are again enjoying the sun in Florida, they really

have brightened my days.

Thank you to all my friends in Bremen, Jessica Wöhlke, Susanne Adrich, Jule

Hägermann, Mareike Hinze, Sarah Schreck, Julia Bauer, Svenja Ludwig, Tanja

Behnke, Andreas Kerzel, and Till Sonnenberg for being a part of my life since my

childhood or teenage years.

Thank you to Julia Budde for sharing my addiction to concerts and loving as many

music styles as I do. I hope we will attend hundreds more and continue to have great

talks on our ways to the venues.

Last but not least, I would like to thank my parents, Dietlinde and Gerd Meyer, for

their continuous support and being such a big part in my life. Thank you for always

being there, being patient, and sometimes reminding me of all the possibilities life

has to offer. I love you.