Investigating Neuroprotectants for the Treatment of Chemotherapy-Induced Peripheral Neuropathy

Home , Amiloride, Benzamil, Resatorvid

THE UNIVERSITY OF NEW SOUTH WALES

Translational Neuroscience Facility School of Medical Sciences Faculty of Medicine UNSW Australia

A thesis in fulfilment of the requirements for the degree of: Masters

Investigating Neuroprotectants for the

Treatment of Chemotherapy-Induced

Peripheral Neuropathy

by Munawwar Abdulla Supervisor: Dr. Gila Moalem-Taylor Co-supervisor: Dr. Justin Lees Co-supervisor: Dr. Patsie Pollie

August, 2018

i The University of New South Wales Thesis/Dissertation Sheet

Surname or Family name: Abdulla First name/s: Munawwar

Abbreviation for degree as given in the University calendar: MSc

School: School of Medical Sciences Faculty: Medicine

Title: Investigating neuroprotectants for the treatment of chemotherapy-induced peripheral neuropathy

Abstract 350 words maximum:

Chemotherapy-induced peripheral neuropathy (CIPN) is a debilitating and dose-limiting side effect of many chemotherapy regimens and is becoming a more prevalent issue as the longevity of cancer patients continues to increase. Sensory symptoms include neuropathic pain, paraesthesia, and numbness and usually spread in a glove-and-stocking distribution. At present, there are no effective medications to treat or prevent CIPN, and the mechanisms by which symptoms are induced have not been fully elucidated. Paclitaxel (PTX) is a commonly used chemotherapeutic that induces neuropathy in a high percentage of patients. The broad aim of this thesis was to establish a model of CIPN in vitro and in vivo and test clinically approved drugs for potential neuroprotective effects. Since the dorsal root ganglion (DRG) has been implicated in CIPN, we cultured dissociated primary DRG neurons from 5-week-old C57BL/6 mice and treated them with PTX to observe neurotoxic effects. We found a marked reduction of neurite outgrowth per neuron and significant morphological changes. Next, we tested several drugs selected from the literature (ibudilast, nicotinamide mononucleotide, resatorvid, amiloride, duloxetine, safinamide) for their potential neuroprotective effects in this model and of those, found that amiloride moderately but significantly prevented the reduction of neurite outgrowth at a certain concentration, although could be harmful if the concentration was too high. We then established a chronic CIPN model in C57BL/6 mice using 6 injections of PTX over a two-week period. PTX-treated mice developed mechanical allodynia, an increase in acetylated tubulin representing damage in the sciatic nerve, increased macrophage presence in the DRG and increased glial cell activation in the spinal cord. Amiloride given at 5 mg/kg, 2 hrs before each PTX treatment was found to have moderate but significant effect in ameliorating mechanical allodynia and reducing astrogliosis in the spinal cord. Taken together, the experiments in this thesis provide in vitro and in vivo models to test drugs against PTX-induced neurotoxicity and present some evidence of amiloride’s potential use to treat PTX-induced peripheral neuropathy. However, further work is required to understand the mechanisms underlying amiloride effects in CIPN, and its drug safety profile in models of cancer.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350-word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

…………………………………….. …………………………………….. ……………………………………..

Signature Witness Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

i Copyright Statement ‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350-word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed: ……………………………………..

Date: ……………………………………..

Authenticity Statement ‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital formal.’

Signed: ……………………………………..

Date: ……………………………………..

ii Originality Statement ‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by other, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistant from others in the projects design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed: ……………………………………..

Date: ……………………………………..

iii Acknowledgments I would first like to thank my supervisor, Dr Gila Moalem-Taylor, for letting me into her lab during Honours year and allowing me to continue through to Masters. Without her guidance this thesis would not have been possible, nor would I have been able to navigate the world of research as a new initiate. Her advice and comments have boosted my confidence and communication skills, which have helped me in both my professional and personal life.

I would like to thank Dr Justin Lees, without whom I would have been very much lost indeed. His support and guidance in the lab and office have been indispensable and he very well may have taught me most of the practical skills I have. Thank you for letting me ask dumb questions and being so involved in this project, and for being a great mentor despite my vocal and your auricular inadequacies. I could not have asked for better or more supportive supervisors.

My co-supervisor, Dr Patsie Pollie – we could not work together too much but I appreciated running into you and giving updates and planning potential studies.

I am grateful to all members of my lab group for helping me with my experiments and for making this degree bearable. I have learned many things from meeting all of you and overall it has been a fascinating and educational experience.

I want to thank all my friends, old and new, who have pushed me to be my best person and grounded me in times of need. They have helped me grow as a person through chocolates and chai, shared experiences and explorations of the unknown, and have been instrumental in shaping my choices in life. I am forever grateful for their attention and their willingness to kick me into the sunlight, literally and figuratively.

Of course, none of this would be possible without my parents, brothers and family (and my extended family, the Uyghur community). Their constant stream of support, prayers and food have kept me physically and spiritually nourished during these at-times- difficult years and I am forever grateful. Distance does not make a difference when family is always in the group chat and the letter box. I do not have the words to describe the all-encompassing type of support I have received, and how much I will do my best to give back soon. I am always grateful for the many shoulders I can metaphorically lean on.

I would also like to acknowledge my three grandparents who passed away during my time at UNSW; they always believed in me and I will do my best to never let them down. Bismillāhi r-raḥmāni r-raḥīm.

Abstract Chemotherapy-induced peripheral neuropathy (CIPN) is a debilitating and dose-limiting side effect of many chemotherapy regimens and is becoming a more prevalent issue as the longevity of cancer patients continues to increase. Sensory symptoms include neuropathic pain, paraesthesia and numbness, and usually spread in a glove-and- stocking distribution. At present, there are no effective medications to treat or prevent CIPN, and the mechanisms by which symptoms are induced have not been fully elucidated. Paclitaxel (PTX) is a commonly used chemotherapeutic that induces neuropathy in a high percentage of patients. The broad aim of this thesis was to establish a model of CIPN in vitro and in vivo and test clinically approved drugs for potential neuroprotective effects. Since the dorsal root ganglion (DRG) has been implicated in CIPN, we cultured dissociated primary DRG neurons from 5-week-old C57BL/6 mice and treated them with PTX to observe neurotoxic effects. We found a marked reduction of neurite outgrowth per neuron and significant morphological changes. Next, we tested several drugs selected from the literature (ibudilast, nicotinamide mononucleotide, resatorvid, amiloride, duloxetine, safinamide) for their potential neuroprotective effects in this model and of those, found that amiloride moderately but significantly prevented the reduction of neurite outgrowth at a certain concentration, although could be harmful if the concentration was too high. We then established a chronic CIPN model in C57BL/6 mice using 6 injections of PTX over a two-week period. PTX-treated mice developed mechanical allodynia, an increase in acetylated tubulin representing damage in the sciatic nerve, increased macrophage presence in the DRG and increased glial cell activation in the spinal cord. Amiloride given at 5 mg/kg, 2 hrs before each PTX treatment was found to have moderate but significant effect in ameliorating mechanical allodynia and reducing astrogliosis in the spinal cord. Taken together, the experiments in this thesis provide in vitro and in vivo models to test drugs against PTX-induced neurotoxicity and present some evidence of amiloride’s potential use to treat PTX-induced peripheral neuropathy. However, further work is required to understand the mechanisms underlying amiloride effects in CIPN, and its drug safety profile in models of cancer.

Preface

vii

Publications arising from this thesis i. Parts of the introductory chapter (Sections 1.3.5, 1.3.10) have been drafted by Munawwar Abdulla, and published as: Lees JG, Makker PGS, Tonkin RS, Abdulla M, Park SB, Goldstein D, Moalem-Taylor G (2017). Immune mediated processes implicated in chemotherapy-induced peripheral neuropathy. European Journal of Cancer 1-18

Conference proceedings arising fully from this thesis i. Toxicity of Paclitaxel on Dissociated Primary Dorsal Root Ganglion Neurons In Vitro Abdulla, M; Lees, JG; Barkl-Luke, M; Goldstein, D; Moalem-Taylor, G Poster Presentation Australasian Neuroscience Society (ANS) 37th Annual Scientific Meeting, Sydney, Australia, 2017

Conference proceedings arising partially from this thesis i. Mouse Models for Identifying Neuroprotectants in Chemotherapy-Induced Peripheral Neuropathy. Lees, JG; Abdulla, M; Barkl-Luke M; Makker, P; Goldstein, D; Park, S; Moalem-Taylor, G Poster Presentation Australasian Neuroscience Society (ANS) 37th Annual Scientific Meeting, Sydney, Australia, 2017

ii. Dorsal Root Ganglion Explants as an In Vitro Model for Oxaliplatin-Induced Peripheral Neuropathy. Barkl-Luke, M; Abdulla, M; Livni, L; Goldstein, D; Moalem-Taylor, G; Lees, JG Poster Presentation Australasian Neuroscience Society (ANS) 37th Annual Scientific Meeting, Sydney, Australia, 2017

viii

Support i. Research Training Program (RTP) stipend (2016-2018) ii. Cancer Institute NSW Translational Program Grant [ID # 14/TPG/1-05]

Table of Contents

Authenticity Statement ...... ii

Originality Statement...... iii

Acknowledgments ...... iv

Abstract ...... vi

Preface ...... vii

Publications arising from this thesis ...... viii

Conference proceedings arising fully from this thesis ...... viii

Conference proceedings arising partially from this thesis ...... viii

Support ...... ix

Index of Figures ...... xiv

Index of Tables ...... xv

Abbreviations ...... xvi

Chapter 1: General Introduction ...... 1

1.1 Pain ...... 2 1.1.1 Neuropathic Pain...... 2 1.1.2 Neuropathy...... 2

1.2 Chemotherapy ...... 3 1.2.1 Chemotherapy Induced Peripheral Neuropathy ...... 4 1.2.3 Paclitaxel ...... 5

1.3 Effects of Paclitaxel in animal models ...... 14 1.3.1 Behaviour ...... 14 1.3.2 Nerve conduction ...... 15 1.3.3 Axonal degeneration and neuronal effects...... 15 1.3.4 Schwann Cells...... 17

1.3.5 Neuroinflammation and Cytokines ...... 18 1.3.6 Receptors ...... 19 1.3.7 Sodium Channels ...... 20 1.3.8 Calcium Channels ...... 21 1.3.9 Mitochondria ...... 22 1.3.10 Central Nervous System (CNS) ...... 24

1.4 Effects of PTX in vitro ...... 27

1.5 CIPN management ...... 28

1.6 Summary ...... 30

Thesis Aims ...... 31

Chapter 2: Optimisation of in vitro sensory neuronal culturing model and testing of potential neuroprotectants against PTX-induced neurotoxicity ...... 32

2.1 Introduction ...... 33 2.1.1 Ibudilast ...... 37 2.1.2 Nicotinamide mononucleotide (NMN) ...... 38 2.1.3 Resatorvid (TAK-242) ...... 39 2.1.4 Amiloride ...... 39 2.1.5 Duloxetine ...... 40 2.1.6 Safinamide ...... 40

2.2 Materials and Methods ...... 42 2.2.1 Animals ...... 42 2.2.2 Dissociated DRG neuron culture ...... 42 2.2.4 Drug administration ...... 45 2.2.5 Immunohistochemistry ...... 45 2.2.6 Imaging ...... 46 2.2.7 Analysis ...... 46 2.2.8 Statistics ...... 47

2.3 Results ...... 48 2.3.1 PTX effects in established culture model of dissociated primary DRG neurons ...... 50

2.3.2 PTX effects in short-term model of dissociated primary DRG neurons...... 48 2.3.3 Effects of drug candidates on neurite outgrowth in the short-term model of PTX treatment ...... 50

2.4 Discussion ...... 57

2.5 Conclusion ...... 62

Chapter 3: Developing an in vivo animal model of PTX-induced peripheral neuropathy and testing the effects of amiloride ...... 63

3.1 Introduction ...... 64

3.2 Materials and Methods ...... 70 3.2.1 Animals ...... 70 3.2.2 Drug administration ...... 70 3.2.3 Behaviour testing ...... 70 3.2.4 Tissue Collection ...... 73 3.2.5 Immunohistochemistry ...... 73 3.2.6 Image Analysis ...... 74 3.2.7 Transmission Electron Microscopy (TEM) ...... 75 3.2.8 Statistics ...... 75

3.3 Results ...... 76 3.3.1 Effects of PTX treatment on C57BL/6 mice ...... 76 3.3.2 Behavioural effects of amiloride...... 80 3.3.3 Molecular effects of amiloride on nervous system tissues ...... 83

3.4 Discussion ...... 88

3.5 Conclusion ...... 94

Chapter 4: Discussion and Conclusions ...... 95

4.1 Overview ...... 96

4.2 CIPN models – in vitro and in vivo ...... 96

4.3 Effects of Amiloride ...... 98

4.4 Translational Capacity and Future Directions ...... 99

4.5 Conclusion ...... 102 xii

Supplementary Data ...... 103

References ...... 104

xiii

Index of Figures

Fig. 1.1. Chemical structure of PTX ...... 5

Fig. 1.2. PTX binding to β-tubulin on the interior of the microtubule ...... 6

Fig. 1.3. Effects of PTX on various tissue based on murine models ...... 28

Fig. 2.1. Cytoskeletal organization in growth cones versus retraction bulbs ...... 33

Fig. 2.2. NeuralMetrics program settings ...... 45

Fig. 2.3. Sample analysis of neurite outgrowth ...... 46

Fig. 2.4. PTX does not alter neurite outgrowth in established model of sensory neurons

...... 48

Fig. 2.5. PTX causes morphological changes (formation of retraction bulbs) in established model of sensory neurons ...... 49

Fig. 2.6. PTX induces a reduction of neurite outgrowth in acutely treated primary neuronal culture ...... 51

Fig. 2.7. The effect of NGF on short-term model ...... 53

Fig. 2.8. Amiloride had significant protective effects among various potential neuroprotectants tested in the short-term model of PTX toxicity in dissociated primary DRG neuronal cultures ...... 55

Fig. 3.1. Sample scoring for MGS ...... 70

Fig. 3.2. Effects of PTX on body weight and exploratory behaviours ...... 75

Fig. 3.3. Effects of PTX on mechanical pain hypersensitivity ...... 76

Fig. 3.4. No difference in g-ratio between small myelinated fibres in vehicle and PTX- treated groups ...... 77

Fig. 3.5. No difference between ratio of normal to abnormal mitochondria in myelinated fibres between vehicle and PTX-treated mice ...... 77

Fig. 3.6. No effects of amiloride on body weight ...... 79

Fig. 3.7. No effects of PTX and amiloride treatment on thermal hyperalgesia ...... 79

Fig. 3.8. No effects of PTX and amiloride treatment on facial grimacing ...... 80

xiv

Fig. 3.9. Amiloride significantly reduced mechanical allodynia in PTX-treated mice 80

Fig. 3.10. Amiloride had no effect on the increase of acetylated tubulin staining in PTX treated mice ...... 82

Fig. 3.11. CGRP immunoreactivity in the sciatic nerve and spinal cord ...... 83

Fig. 3.12. Iba-1 immunoreactivity in the dorsal horn of the spinal cord ...... 84

Fig. 3.13. GFAP immunoreactivity in the dorsal horn of the spinal cord...... 85

Fig. 5.1. Macrophages in the DRG of PTX-treated mice ...... 100

Index of Tables

Table 1.1. Neuropathy Terms ...... 3

Table 1.2. Animal models of CIPN using Paclitaxel...... 8

Table 2.1. Optimisation of cell culture conditions ...... 41

Table 3.1. Drugs that have shown protective effects against PTX pathology in animal studies ...... 63

Table 3.2. Experimental Timeline ...... 71

Abbreviations 5-HD 5-Hydroxydecanoate

5-HT Serotonin

AMPK 5' AMP-activated Protein Kinase

AK295 α-ketoamide Calpain Inhibitor

ALCAR Acetyl-L-Carnitine

Ara-C Cytosine Arabinoside

ASIC Acid-Sensing Ion Channel

ATP Adenosine Triphosphate

ATF3 AMP-Dependent Transcription Factor

ANOVA Analysis of Variance

ACC Anterior Cingulate Cortex

AMI Amiloride

B27 B-27 plus Supplement

BBB Blood Brain Barrier

BDNF Brain-Derived Neurotrophic Factor

BMSC Bone Marrow Stromal Cells

BSA Bovine Serum Albumin

CCI Chronic Constriction Injury

CCR2 C-C Chemokine Receptor Type 2

CCL2 Chemokine C-C Motif Ligand 2

CD Cumulative Dose

CD11B Cluster of Differentiation Molecule 11B

CD38 Cyclic ADP Ribose Hydrolase

CGRP Calcitonin Gene-Related Peptide

CIPN Chemotherapy Induced Peripheral Neuropathy

xvi

CNS Central Nervous System

DMA 5-(N,N-Dimethyl) Amiloride Hydrochloride

DMSO Dimethyl Sulfoxide

DNA Deoxyribonucleic Acid

DRG Dorsal Root Ganglia

DUL Duloxetine

EAE Experimental Autoimmune Encephalomyelitis

EB3 End Binding Protein 3

ER Endoplasmic Reticulum

FDA USA Food and Drug Administration

GFAP Glial Fibrillary Acidic Protein

IASP International A Society of Pain

Iba-1 Ionized Calcium-Binding Adapter Molecule 1

IB4 Isolectin B4

IBU Ibudilast

IENF Intraepidermal Nerve Fibres

IHC Immunohistochemistry iNos Inducible Nitric Oxide Synthase

InsP3R Inositol Trisphosphate Receptor

IL Interleukin i.p. Intraperitoneal i.v. Intravenous

L-15 Leibovitz's L-15 Medium

LD50 Lethal Dose, 50%

LPS-RS Lipopolysaccharide from Rhodobacter sphaeroides

MGS Mouse Grimace Scale

xvii

MIF Migration Inhibitory Factor

MMP-3 Matrix Metalloproteinase-3

MTD Maximum Tolerated Dose mPTP Mitochondrial Permeability Transition Pore

NA Noradrenaline

NAD Nicotinamide Adenine Dinucleotide

Navs Voltage-Gated Sodium Channels

NBA Neurobasal-A Medium

NCS-1 Neuronal Calcium Sensor-1

NCV Nerve Conduction Velocity

NF200 Monoclonal Anti-Neurofilament 200

NF-κB Nuclear factor-κB

NGF Nerve Growth Factor

NHE Sodium-Hydrogen Exchanger

NMN Nicotinamide Mononucleotide

NO Nitric Oxide

NP Neuropathic Pain

OCT Optimal Cutting Temperature Compound

P-APS Paclitaxel-Acute Pain Syndrome

PARP1 Poly(ADP-Ribose) Polymerase 1

PAS Photobeam Activity System

PBS Phosphate Buffered Solution

PD Parkinson’s Disease

PDE Phosphodiesterase

PDL Poly-D-Lysine

PFA Paraformaldehyde

xviii

PGP9.5 Protein Gene Product 9.5

PI Propidium Iodide

PNS Peripheral Nervous System

PPAR-α Peroxisome Proliferator-Activated Receptor Alpha

P/S Penicillin Streptomycin

PTX Paclitaxel

QOL Quality of Life rHuEPO Exogenous Recombinant Human Erythropoietin

RO Reverse Osmosis

ROS Reactive Oxygen Species

RT Room Temperature

SCI Spinal Cord Injury

S1P Ceramide-Sphingosine 1-Phosphate

S1PR1 S1P Receptor Subtype 1

σ1R Sigma-1 Receptor

SAF Safinamide s.c. Subcutaneous

SEM Standard Error of the Mean

SGC Satellite Glial Cell

SIRT1 Deacetylase Sirtuin1

SNAP Sensory Nerve Action Potential

SNL Spinal Nerve Ligation

SNRI Serotonin and Noradrenaline Reuptake Inhibitor

TBI Traumatic Brain Injury

TEM Transmission Electron Microscopy

TGF-β1 Transforming Growth Factor Beta 1

xix

TLR4 Toll-Like Receptor 4

TNF-α Tumour Necrosis Factor-α

TRPA1 Transient Receptor Potential Ankyrin 1

TRPV Transient receptor potential Vanilloid

TTX Tetrodoxin

UV Ultraviolet

VEH Vehicle

WT Wild Type

Chapter 1: General Introduction

1.1 Pain Pain is defined as “an unpleasant sensory and emotional experience associated with actual or potential tissue damage” by the International Association for the Study of Pain (Iasp-pain.org, 2017). Nociceptive pain is the body’s protective response to a potentially harmful stimulus and it is a subjective experience, which can be influenced by psychological and social factors (Linton and Shaw, 2011, Main, 2013, Edwards et al., 2016). Although pain is a major symptom in numerous pathological conditions, there are distinct types and many potential causes of pain; this thesis will focus on neuropathic pain caused by chemotherapy.

1.1.1 Neuropathic Pain Neuropathic pain (NP) is “pain arising as a direct consequence of a lesion or disease affecting the somatosensory system” (Iasp-pain.org, 2017). It is often caused by nervous system pathology, such as autoimmune diseases, various infections or toxins, diabetes, nerve injury or compression, and other disease or trauma states. Symptoms of NP include allodynia, hyperalgesia, paraesthesia and spontaneous pain and may continue as a chronic pain state. Reports estimate the prevalence of chronic NP to be anywhere between 0.9% (Gore et al., 2007) to 17.9% (Toth et al., 2009) globally and the incidence rate of NP is estimated to be 8.2 per 1000 person–years (Van Hecke et al., 2014). In Australia, a quarter of patients who visit general practices have been recorded as living with chronic pain, and of this population chronic NP is estimated to be approximately 5% (Henderson et al., 2016). NP in patients with cancer ranges from 19% to 39.1% (Bennett et al., 2012) although these estimates are increased if the patient is undergoing chemotherapy treatment. Pain has significant effects on the economy due to increased health costs and decreased productivity, and has the most significant effect on quality of life (QOL) of any disease (Henschke et al., 2015).

1.1.2 Neuropathy Peripheral neuropathy is a common neurologic problem that results from damage or disease affecting nerves. Neuropathy has diverse forms and presentations, including focal neuropathy affecting a single nerve (e.g. carpal tunnel syndrome) or polyneuropathy affecting several nerves, and various aetiologies such as autonomic 2 neuropathy due to damage to nerves that control autonomic functions, and diabetic neuropathy due to nerve damage caused by diabetes. Clinical diagnosis occurs following neurological examination, careful review of medical history and selected tests, including blood tests, urine analysis, x-rays, nerve conduction studies, electromyographies, skin biopsies and genetics tests for neuropathies that may be hereditary. Peripheral nerves consist of sensory, motor, and autonomic fibres, and accordingly symptoms vary based on the relative involvement of each type and of large and small fibres. Peripheral neuropathies are associated with common neuropathic symptoms, which are outlined in Table 1.1.

Term Definition Allodynia Pain due to non-noxious stimuli Dysesthesia Abnormal sensations from normal stimuli Hypoesthesia Decreased sensation Hyperalgesia Increased perception of pain Hyperaesthesia Increased perception of sensory stimuli Hyperpathia Reduced sensations with increased hyperalgesia or allodynia Hypoalgesia Reduced perception of pain Paraesthesia Abnormal sensations such as pins and needles, tingling, prickling, burning, or reduced or complete loss of sensation. Table 1.1. Terms used to define various neuropathic symptoms of peripheral neuropathy

1.2 Chemotherapy Chemotherapy is one of the most common forms of cancer treatment. Depending on the type of cancer, chemotherapy drugs can be given alone or in conjunction with surgery or radiation, or in certain drug combinations for acute treatment or on a long-term basis, either through injections, orally or applied as a cream. Chemotherapeutics can be broadly categorised as alkylating agents, plant alkaloids, anti-tumour antibiotics, antimetabolites, topoisomerase inhibitors and mitotic inhibitors. Although chemotherapy drugs kill rapidly dividing cancer cells, they also harm healthy cells

3 causing short- and long-term side effects such as neuropathy, nausea, fatigue, muscle weakness, hair loss, loss of appetite and increased risk of infection or bruising. One of the most common side-effects of chemotherapy is chemotherapy-induced peripheral neuropathy (CIPN), which is the focus of this thesis.

1.2.1 Chemotherapy Induced Peripheral Neuropathy CIPN is a common, debilitating and dose-limiting side effect of many chemotherapy regimens (Gutiérrez-Gutiérrez et al., 2010, Pachman et al., 2011). With the increase in the number of cancer diagnoses and improved longevity of patients, the long-term effects of cancer and its treatments have become a more prevalent issue, particularly in relation to QOL deficits (Mols et al., 2014, Ezendam et al., 2014). CIPN can begin during treatment and either resolves, or continues for months or even years after cessation thus becoming a chronic issue (Hershman et al., 2014). CIPN is associated with neuropathic symptoms, which can be disabling and commonly include forms of sensory loss (e.g. numbness) and tingling at the extremities that spreads in a “glove and stocking” distribution (Pachman et al., 2011), suggesting the longer axons are more at risk (Benbow et al., 2016). CIPN affects 30-68% of patients dependent on the type of drug, the duration of administration, cumulative dose and pre-existing health conditions (Beijers et al., 2012, Seretny et al., 2014). The exact prevalence is difficult to ascertain as there is no gold standard to assess CIPN (Alberti and Cavaletti, 2015), and some studies indicate that up to 80% of patients experience some form of neuropathic pain (Sisignano et al., 2014). A systematic review and meta-analysis indicates that the prevalence of CIPN reduces to 60% at 3 months after cessation of treatment and 30% at 6 months or more (Seretny et al., 2014). There are currently no cures or effective treatment options for CIPN and its associated neuropathic pain (Pachman et al., 2011), and often chemotherapy is simply stopped after symptoms occur to prevent more damage (Sisignano et al., 2014, Benbow et al., 2016). Chemotherapeutics usually implicated in CIPN are taxanes, platinum derivatives, vinca alkaloids, proteasome inhibitors, epothilones and thalidomide (Carozzi et al., 2015). This study focuses on one particular taxane, Paclitaxel (PTX), a potent chemotherapy drug associated with neurotoxicity and the development of CIPN.

1.2.3 Paclitaxel PTX is an antineoplastic drug first characterised in 1971 as taxol (Wani et al., 1971) and later trademarked as Taxol by Bristol Myers-Squibb, a US pharmaceutical manufacturer. It belongs to the terpenoid class of natural products as it is derived from taxadiene, which is a diterpene (Roberts, 2007) (Figure 1.1). It consists of a four- membered oxetane ring and a complex ester side chain (Lanzotti, 2013). PTX was initially developed from the bark of the Pacific yew tree, Taxus brevifolia. Currently, it is commercially available through semi-synthesis of products such as 10- deacetylbaccatin from plants such as the European yew (Taxus baccata), necessary because extracting PTX from the Pacific yew requires killing the tree and is not sustainable (Weaver, 2014).

Figure has been removed due to copyright restrictions.

Figure 1.1. Chemical structure of PTX (PubChem, 2005)

PTX can be used to prevent restenosis of coronary and peripheral stents as it can prevent the growth of scar tissues (Scheller et al., 2006). However, it is more commonly used as a chemotherapeutic agent effective against solid tumours such as head and neck, breast, ovarian, and non-small cell lung cancer and AIDS-related Kaposi sarcoma (Weaver, 2014).

Currently, PTX is given on its own or in conjunction with other chemotherapy drugs for 3-12 cycles of 80 mg/m2 - 175 mg/m2 depending on the type of cancer and whether other drugs are included in the regimen. The total clinical dosage is up to 1050 mg/m2 per treatment round (O'Shaughnessy, 2016). PTX is usually delivered through intravenous injections for a systemic, rather than local effect. The most common side 5 effects include neuropathy, neuropathic pain, nausea, neutropenia, vomiting, appetite loss and change in blood pressure or heart rate (Carozzi et al., 2015). Cases of motor weakness and cranial nerve palsies have been recorded in some patients (Rowinsky et al., 1993, Lee et al., 1999). Treatment may also cause a severe allergic reaction. Some of these side effects may be due to the castor oil derivative (Cremaphor EL) and ethanol mix PTX requires for injection, since PTX is a hydrophobic substance (Gelderblom et al., 2001). To combat this, there has been an increased amount of research into Cremaphor-free or water-soluble forms of PTX such as Abraxane (Green et al., 2006). Nevertheless, PTX is still used as a front-line drug for chemotherapy treatment (Kumar et al., 2010). It is listed in the World Health Organisation’s Model List of Essential Medicines, and is now “…one of the most highly prescribed chemotherapy drugs for cancer” (NIH, 2015).

PTX is understood to cause apoptosis in cancer cells through stabilisation of microtubules by binding to β-tubulin subunits (Figure 1.2), effectively preventing depolymerisation and blocking mitosis (Schiff et al., 1979, Manfredi et al., 1982), although it has been suggested that anti-tumour activity is due to the induction of multipolar divisions rather than mitotic arrest (Weaver, 2014). PTX may also disrupt axonal transport, causing apoptosis through downstream effects such as mitochondrial damage (Carozzi et al., 2015). Furthermore, PTX sensitises cancer cells to radiation therapy, enhancing the effectiveness of cancer treatment schedules (Pizzorno et al., 2008).

Figure has been removed due to copyright restrictions.

Figure 1.2. PTX binding to β-tubulin on the interior of the microtubule (Gornstein and Schwarz, 2014).

While its anti-tumour effects are potent, it is also known to have a cumulative neurotoxicity that increases with each cycle of treatment (van Gerven et al., 1994). The incidence of neuropathy from PTX increases based on dosage ranging from 22% to approximately 100%, beginning at a dose of 250 mg/m2 (Polomano et al., 2001, Argyriou et al., 2014). Differences between patients depend on dose, treatment schedule, infusion-time, age, prior health conditions and perhaps genetics (Baldwin et al., 2012, Hertz et al., 2013, Hertz et al., 2014, Tanabe et al., 2017). Patients with higher pain responses from their first dose are more likely to develop chronic neuropathy (Loprinzi et al., 2011). Chronic CIPN is three times more likely to occur after treatment if symptoms are experienced during treatment (Reyes-Gibby et al., 2009). PTX treatment can produce an acute pain state (peaking at 3 days after initiation of chemotherapy) called PTX acute pain syndrome (P-APS). Clinical studies suggest that those patients who develop P-APS do not necessarily develop CIPN, but those who do tend to have increased severity compared to patients who do not develop P-APS (Brown et al., 2016). It has been suggested that P-APS is caused by a sensitisation of nociceptors or the spinothalamic system based on clinical observations (Loprinzi et al., 2007) and it is associated with the development of CIPN (Reeves et al., 2011). Treatment is often stopped soon after neuropathy is diagnosed, however chronicity (pain lasting more than 3 months) is still a common debilitating problem (Polomano et al., 2001).

Many studies using rodent models of PTX-induced neuropathy have been conducted to understand the mechanisms by which PTX causes neurotoxicity and neuropathic pain. Table 1.2 is a list of the different animal models and the observed symptoms that have been used in numerous studies of PTX-induced peripheral neuropathy.

Drug Regimen Animal Strain & Study Symptoms Intraperitoneal (i.p.) injections Mice 1x injection M, ddY, 20-25 g, 7 wks 24hrs: mechanical allodynia and hyperalgesia peak Cumulative dose (CD): 0.1, (Hidaka et al., 2009) No mechanical hyperalgesia for 0.1-5 mg/kg 1, 5, 10, 20, 30, or 40 mg/kg 1x injection M, C57BL/6, 6 wks D14: mechanical allodynia peak CD: 5 mg/kg (Gauchan et al., 2009) 1x injection M, C57BL/6, 25 g D2-20: mechanical allodynia CD: 6 mg/kg (Materazzi et al., 2012) D4-12: cold allodynia 2x injections M, C57BL/6J, Rag1-/-, 8-12 wks D3: mechanical allodynia D0, 2 (Krukowski et al., 2016) D12: WT recovery CD: 4 mg/kg D21: Rag1-/- recovery 3x injections F, Trpv4−/−, 20-25 g 1 week: hyperalgesia in WT mice but not in TRPV4 knockout mice D0, 7, 14 (Matsumura et al., 2014) CD: 30 mg/kg 4x injections M & F, 129P3, A, AKR, D7-21: mechanical allodynia D1, 3, 5, 7 C3H/He, C57BL/6, C57BL/10, D10: thermal hyperalgesia (but not in C57BL/6) CD: 4 mg/kg CBA, DBA/2, RIIIS, and SM (all J) D15: cold allodynia (Smith et al., 2004) Males exhibited more pain than females (mechanical allodynia). Mechanical allodynia higher in DBA/2, lower in C57BL/6. M, C57BL/6, 22 g D3/4: mechanical allodynia (Boehmerle et al., 2014) Reduced stance phase of hind paws, reduced NCV, no effect on rotarod.

4x injections M &F CB2KO with C576BL/6J D4: mechanical allodynia D1, 3, 5, 7 background and CB1KO with CD1 D4: cold allodynia background, 25-33 g CD: 16 mg/kg (Deng et al., 2015) 4x injections F, C57BL/6, 7 wks D14: thermal hypoalgesia D1, 3, 5, 7 (Mo et al., 2012) CD: 18 mg/kg 4x injections F, BALB/c, 19-21 g D4: no DRG pathology, reduced caudal NCV but not digital NCV. D1, 8, 15, 22 (Carozzi et al., 2010) Sciatic nerve myelinated fibre degeneration. CD: 200 or 280 mg/kg 5x injections F CD-1, 25-30 g Peak— D1, 2, 3, 4, 5 (Nieto et al., 2008) D7: Heat hyperalgesia CD: 10 mg/kg D10: mechanical allodynia D10-14: cold allodynia

Long lasting (2-4 weeks) M, C57BL/6, CD1, 8-10 wks D7: mechanical allodynia (Costa et al., 2011) D7: thermal hyperalgesia M, ddY, 23-25 g D3-15: mechanical allodynia (Katsuyama et al., 2013) M, CD1, Group (G)1: 31 days, G2: D7: mechanical allodynia 3-4 months, G3: 12-13 months D7: thermal hyperalgesia (Ruiz-Medina et al., 2013) F, BALB/c, 8-12 wks, 20-30 g D7: chemical nociception

(Masocha, 2014) D7: thermal hyperalgesia 7x injections C57BL/6J, 12 wks Wk 3: mechanical allodynia D1, 3, 5, 7, 9, 11, 13 (Mao-Ying et al., 2014) CD: 70 mg/kg 12x injections C57BL/6J, 9 wks D14: mechanical allodynia D1, 3, 5, 8, 10, 12, 15, 17, (Huehnchen et al., 2013) Reduced caudal nerve SNAP 19, 22, 24, 26 CD: 240 mg/kg Rats 1x injection M, Sprague-Dawley, 180-200 g 1 injection— CD: 16 or 32 mg/kg (Authier et al., 2000) D4-9: mechanical allodynia in 16 mg/kg D4-14: mechanical allodynia in 32 mg/kg 5x injections D4-7: thermal hypoalgesia in 32 mg/kg D1, 7, 14, 21, 28 5 injections— CD: 80 mg/kg Weight loss and 2 mortalities 2nd to 5th injections: thermal hypoalgesia 1 week after 5th injection: decreased nerve conduction velocity 2x injections M, Sprague-Dawley, 340-390 g D10: mechanical hyperalgesia D1, 3 (Nishida et al., 2008) IENF loss, increased ATF3 in DRG CD: 32 mg/kg 3x injections M, Sprague-Dawley, 220-280 g D4-12: mechanical allodynia D1, 4, 7 (Liu et al., 2010) CD: 24 mg/kg 4x injections M, Sprague-Dawley D5: Mechanical allodynia for 4 mg/kg group 10

D1, 3, 5, 7 (Polomano et al., 2001) D16: Mechanical allodynia for 2 and 8 mg/kg groups CD: 2, 4, or 8 mg/kg D10-28: Cold allodynia D12-28: Heat hyperalgesia for 2 mg/kg group (Studies use 8 mg/kg unless Developed endoneurial oedema, but no degeneration of axons or otherwise stated) structural abnormality of sciatic nerve, DRG, dorsal and ventral roots or spinal cord. M, Sprague-Dawley, 250-300 g D16-41: Mechanical hypersensitivity (peak D26-29, declining by (Flatters et al., 2006) D41) M, Sprague-Dawley, 250-300 g D15-155: Mechanical allodynia (Flatters and Bennett, 2006) D13-52: Cold allodynia Heat hyperalgesia: mild, transient D7 & 27: swollen mitochondria in C-fibres and myelinated axons. No changes in ATF3 in DRG. No axon degeneration. M, Sprague-Dawley, 325-375 g D14-21: mechanical allodynia (lasted 6 weeks in 4 mg/kg, 16 weeks (Ledeboer et al., 2007b) for 8 mg/kg) M, Sprague-Dawley, 200-300 g D7: no pain (Siau et al., 2006) D27: Mechanical hyperalgesia peak D160: resolved IENF degeneration M, Sprague-Dawley D7-29: mechanical hyperalgesia (Cata et al., 2008) D7-45: mechanical allodynia D7-50: thermal hyperalgesia M, Sprague-Dawley, 200-220 g D12-25: mechanical allodynia (Janes et al., 2014a) No thermal hyperalgesia.

4x injections M, Wistar, 200-250 g D5: mechanical allodynia D1, 3, 5, 7 (Pascual et al., 2010) D18: thermal hyperalgesia CD: 4 mg/kg Jugular vein injections Mice 3x injections F, Wlds & C57BL/6, 8 wks Rotarod: 50% reduction in ability to stay on rotarod D1, 3, 5 (Wang et al., 2002) CD: 180 mg/kg F, C57BL/6, 8 wks Rotarod: 60% reduction in ability to stay on rotarod (Wang et al., 2004) 3 weeks: loss of myelinated fibres Intravenous (i.v.) injections Mice 3x injections F, AJ mice, 8 wks 2 weeks: reduced intraepidermal innervation, sensory axonal D1, 3, 5 (Melli et al., 2006) degeneration. No loss of axons in sciatic nerve, no increase in microtubule accumulation. CD: 75 mg/kg M, AJ, 15-20 g, 6 wks 4 weeks: thermal hyperalgesia (Park et al., 2015) Reduced SNAP amplitude, unmyelinated axon numbers in IENF, increased detyrosinated tubulin in sural nerves. 4x injections F, C57BL/6 & Mito-CFP/COX8A, D7: prolonged sensory nerve conduction latency D1, 8, 15, 22 4-5 wks D14: decreased sensory nerve amplitude CD: 200 mg/kg (Bobylev et al., 2015) D14: cold allodynia

Smaller unmyelinated sensory nerve fibre degeneration. 6x injections F, BALB/c, 7-8 wks MTD for PTX defined as 30 mg/kg per individual injection (total 180 D1, 3, 5, 8, 10, 12 (Wozniak et al., 2011) mg/kg).

CD: 120, 150, 180, 210, F, BALB/c, 7-8 wks (At MTD): Axonal loss, axonal degeneration, increased abundance of 240, or 270 mg/kg (Benbow et al., 2016) non-neuronal cells in sciatic nerve, some tubulin accumulation and acetylation, neurodegeneration (disruption of normal myelin morphology). M & F, C57Bl/6, 3-4 months (At MTD): axonal degeneration in spinal cord, sciatic nerve (large and (Tasnim et al., 2016) medium fibre classes). Rats 2x injections M, Sprague Dawley, 250-275 g D10: reduced coordination and speed on rotarod. D0, 3 (Peters et al., 2007b) D10: mechanical allodynia CD: 36 mg/kg D10: cold hyperalgesia D1-10: Increased ATF3 in DRG and Schwann cells No thermal hyperalgesia Subcutaneous (s.c.) injections Mice 1x injection M, CD-1 Swiss, 8-10 wks D1-7: cold hyperalgesia CD: 10 mg/kg (Pevida et al., 2013) Table 1.2. Animal models of CIPN using Paclitaxel. Most studies used Paclitaxel dissolved in 100% ethanol with Cremaphor and saline while a few used pre-made Taxol.

1.3 Effects of Paclitaxel in animal models

1.3.1 Behaviour All studies listed in Table 1.2 have shown rodents survive without any major weight loss or motor deficit when the dose is below the maximum tolerated dose (MTD), which was suggested to be 6 injections of 30 mg/kg, for a cumulative dose of 180 mg/kg over a period of two weeks in mice (Wozniak et al., 2011). Mice given doses of PTX greater than the MTD (from 35-45 mg/kg) were shown to have significant weight loss and hind limb nerve malfunction after the second dose. Indeed, most studies that observed behavioural phenotypes prescribed dosages well below MTD, and variations in cumulative dose seemed to have more of an effect on the onset and prolongation of pain. For example, a study that compared two doses of PTX found the higher dose induced a longer-lasting neuropathy (Ledeboer et al., 2007b). The testing methods as well as the behavioural outcomes often vary among studies, but in general PTX was shown to cause mechanical allodynia, mechanical hyperalgesia, and thermal hyperalgesia depending on animal model and drug schedule. Reduced speed and coordination on the rotarod test, a test that measures endurance, balance, grip strength and motor coordination, was observed in some instances (Wang et al., 2002, Peters et al., 2007b). However, others showed no differences in rotarod but instead a marked gait alteration (Boehmerle et al., 2014) and still others showed there were no changes in movement or activity when the dose was below MTD (Table 1.2). In regards to thermal hyperalgesia, there is some conjecture as to whether PTX affects hot or cold sensitivity (Peters et al., 2007b, Flatters and Bennett, 2006, Nieto et al., 2008, Deng et al., 2015, Smith et al., 2004, Materazzi et al., 2012) or any thermal sensitivity at all (Janes et al., 2014a) and this may be due to differences in techniques used to measure these changes in CIPN models. Some studies also noted thermal hypoalgesia rather than hyperalgesia (Park et al., 2015), while others did not measure thermal sensitivity, relying solely on mechanical allodynia for behavioural neuropathic assessment.

Dosage varied from one large, single injection to consecutive or alternate days of low doses over a period of 1 to 3 weeks (Table 1.2). While most injected PTX intraperitoneally, a few studies described intravenous injections, and some also used

14 jugular vein or subcutaneous injections. The onset of pain varied from almost immediate signs (Materazzi et al., 2012, Pevida et al., 2013), to peak one or two weeks after the first injection (Nieto et al., 2008, Flatters et al., 2006). Some models reported alleviation of pain after cessation of treatment while others reported chronicity until endpoint of up to 160 days (Siau et al., 2006, Flatters and Bennett, 2006). The differences in species, age, sex, dosage and treatment schedules seem to have significant effects on some phenotypes of CIPN, some of which will be discussed in the following sections, indicating the importance of the model chosen for testing.

1.3.2 Nerve conduction Nerve conduction studies are one way patients can be diagnosed with CIPN through the direct measurement of large motor and sensory nerve fibres (Velasco et al., 2014, Grisold et al., 2012, Argyriou et al., 2005), and this method is often used to study patients undergoing various clinical trials (Sundar et al., 2016, Velasco et al., 2017). In mice treated at MTD there was significant reduction in caudal nerve conduction velocity (NCV) and amplitude (though not at 0.5 of MTD) and significant reduction in digital nerve amplitude (until 0.75 MTD) (Wozniak et al., 2011). A low cumulative dose of 4 mg/kg of PTX caused reduced sensory nerve action potential (SNAP) amplitude in the caudal nerve and decreased NCV during PTX treatment but not after cessation (Boehmerle et al., 2014). Although reports of differences in nerve conductance following PTX treatment are varied, especially in low-dose regimens, it can be an important avenue of investigation, especially as an objective measure of nerve function that counters the variability of mice behavioural studies.

1.3.3 Axonal degeneration and neuronal effects PTX promotes microtubule assembly and stabilisation in rapidly dividing cells (Schiff et al., 1979, Schiff and Horwitz, 1980) by binding to the β-tubulin and preventing the depolymerisation of α- and β-tubulins (Manfredi et al., 1982), effectively blocking cell division and inducing apoptosis. However, non-dividing neuronal cells in the periphery are also affected (Park et al., 2011). Microtubules are essential for maintenance of elongated neuronal morphology and axonal transport, and the movement of cargo between cell bodies and distal axons (Carlson and Ocean, 2011). Disruption to these 15 functions in neurons may have a causative role in axon degeneration and CIPN, as PTX has been shown to accumulate in the dorsal root ganglia (DRG), dorsal and ventral roots, and sciatic nerves (Gornstein and Schwarz, 2014).

At MTD, there appears to be severe axonal degeneration in both large- and small- diameter fibres (milder in 0.5 and 0.75 MTD), DRG neuronal degeneration, and axonal degeneration of proximal axons (from 0.5 MTD) (Wozniak et al., 2011). A later study using the MTD model found axonal degeneration of both central and peripheral DRG axons, affecting mostly large- and medium-sized fibres, without significant damage to the DRG neural cell body (Tasnim et al., 2016). Axonal degeneration was also found in the sciatic nerve, more so in the distal than in the proximal regions although present in both (Tasnim et al., 2016). These areas were also correlated with macrophage infiltration indicating a neuroimmune response.

There have been reports of increased expression of α-tubulin and tubulin acetylation (a marker for increased microtubule stability) in the sciatic nerves, as well as a mild suppression of end binding protein (EB)3, a protein associated with microtubule growth, at MTD (Benbow et al., 2016). Axonal degeneration and the upregulation of detyrosinated tubulins was decreased in PTX-treated DRG in vitro, and epidermal denervation in the footpads was reduced in vivo with exogenous recombinant human erythropoietin (rHuEPO), suggesting the promotion of normal microtubule assembly dynamics in the DRG sensory neurons and associated axons could prevent CIPN (Melli et al., 2006). Interestingly, an exercise regimen in mouse models was found to reduce some of the neuropathic effects of PTX, including reduced detyrosinated tubulins (Park et al., 2015). One study suggests that neurotoxicity may be due to inhibition of fast axonal transport in particular (LaPointe et al., 2013). It was also shown that WldS mice, who are resistant to axonal degeneration, were completely protected from the neuropathic effects of PTX (Wang et al., 2002).

However, axonal degeneration is not always necessary for the induction of pain; rodents treated with low doses of PTX exhibited pain hypersensitivity without any effect on large nerve fibres. For example, significant decreases in the number of intraepidermal 16 nerve fibres (IENF) (Siau et al., 2006, Liu et al., 2010, Melli et al., 2006), or the presence of endoneurial oedema in the sciatic nerves were sufficient to cause painful peripheral neuropathy (Polomano et al., 2001) without marked axonal degeneration. The latter study also found no structural changes in the spinal cord, sciatic nerve or DRG. One study focusing on IENFs in the hind paws of PTX-treated rats found a partial reduction of IENFs and a significant increase in Langerhans cells, suggesting a role for immune cells (Siau et al., 2006). Rats treated with 4 i.p. injections of 2 mg/kg PTX, a relatively low dose, developed a painful neuropathy that peaked at D27 and resolved at D155, but exhibited no axonal degeneration, normal myelin structure, normal myelinated axon and C-fibre density, and normal cyclic AMP-dependent transcription factor activating transcription factor 3 (ATF3; a marker of neuronal damage) expression in DRG neurons (Flatters and Bennett, 2006). The only significant difference was an increase in swollen and vacuolated mitochondria in C-fibres and myelinated axons, which also resolved by D160, indicating that mitochondrial dysfunction, rather than axonal degeneration, could be responsible for painful neuropathy.

1.3.4 Schwann Cells Recent studies have looked at Schwann cells as a potential target of PTX-induced neuropathy. Schwann cells are a type of glial cells in the peripheral nervous system that have a role in guiding regrowth of peripheral axons and myelin formation, and are important for axon maintenance (Imai et al., 2017). Primary Schwann cells cultured from sciatic nerves of neonatal rats exposed to PTX exhibited marked morphological changes, cytotoxicity and increased dedifferentiation in both in vitro and in vivo experiments, but no changes in mitochondria function (Imai et al., 2017). However, previous studies have found changes in Schwann cell mitochondria (Melli et al., 2008), perhaps because the concentration of PTX used was much higher (approximately 0.29 μM) than the former experiment (0.01 µM), indicating that damage to Schwann cells may occur before damage to their mitochondria. Schwann cells could have a role in inflammation, as they may be a source of pro-inflammatory cytokines like tumour necrosis factor (TNF)-α (Hama et al., 2012). However, studies that treated rodents with a lower dose of PTX did not find any changes in Schwann cells or Schwann cell mitochondria (Nieto et al., 2014, Xiao et al., 2011). Nevertheless, Schwann cell damage in high dose therapies presents an interesting target for CIPN therapy and may explain 17 some of the discrepancies between studies that observed variable damage to axons or myelin.

1.3.5 Neuroinflammation and Cytokines It was previously suggested that PTX has immune system-modulating effects that further promoted the anti-tumour effects already seen from microtubule stabilisation (Chan and Yang, 2000). However, the following studies suggest that these immune modulating effects may be the primary cause of CIPN.

PTX induces the release of inflammatory cytokines such as TNF-α and interleukin (IL)- 1 in the DRG, and intrathecal treatment with IL-10, an anti-inflammatory cytokine, or IL-1αRA, an IL-1 antagonist, reverses mechanical allodynia (Ledeboer et al., 2007b, Krukowski et al., 2016). It has also been shown that CD8+ T cells, which increase the expression of IL-10 receptor 1 in the DRG of PTX-treated mice, are required for the resolution of CIPN in animal models (Krukowski et al., 2016). The increase in cytokines may be due to the increase of activated macrophages in the DRG (Ledeboer et al., 2007b, Peters et al., 2007a, Nishida et al., 2008, Liu et al., 2010, Zhang et al., 2016), perhaps attracted by the damage to the cells in DRG as demonstrated by an upregulation of ATF3 (Peters et al., 2007a, Liu et al., 2010). The macrophages may be activated by the upregulation of matrix metalloproteinase-3 (MMP-3) in large neurons in the DRG of PTX-treated animals (Nishida et al., 2008) which can contribute to neuronal sensitisation (Sisignano et al., 2014). Depleting macrophages using liposome- encapsulated clodronate (Clophosome) reduces TNF-α expression in the DRG, some behavioural phenotypes of CIPN, and IENF loss (Zhang et al., 2016). Macrophages also infiltrate degenerated fibres of the sciatic nerves in PTX-treated mice (Tasnim et al., 2016). Macrophages, and by extension inflammatory cytokines, can be potential targets for neuroprotection. Certainly, minocycline, a microglial inhibitor, has previously been found to reduce mechanical allodynia and thermal hyperalgesia in PTX-induced neuropathy both alone (Masocha, 2014, Cata et al., 2008, Burgos et al., 2012, Liu et al., 2010) and when used in conjunction with indomethacin (Parvathy and Masocha, 2015). Minocycline treatment also protects from the loss of IENFs (Liu et al., 2010). Thalidomide, thought to reduce TNF-α and inhibit pro-inflammatory cytokines, reduced

18 mechanical allodynia but not thermal hyperalgesia in rats (Cata et al., 2008) suggesting its selective nature of action. A study showed that intrathecal injection of bone marrow stromal cells (BMSC) produced a long term reduction of neuropathic pain, most likely due to the migration of BMSCs to the DRG and transforming growth factor-β1 (TGF- β1) secretion, which may suppress macrophage/microglial activation in PTX-treated mice (Huh et al., 2016).

1.3.6 Receptors The upregulation of certain receptors may have a contributing role in CIPN, or may be used as an indicator for CIPN. C-C chemokine receptor type 2 (CCR2) are upregulated in sensory neurons after PTX treatment, and blocking chemokine (C-C motif) ligand 2 (CCL2)/CCR2 in the DRG reduced CIPN phenotypes, further suggesting the involvement of inflammatory processes (Sisignano et al., 2014, Zhang et al., 2013). Toll-like receptor 4 (TLR4), a transmembrane protein involved in intracellular signalling pathways that lead to inflammation, has been shown to be activated in the DRG (Barajon et al., 2009) and spinal cord along with the increase in CCL2 (Li et al., 2015, Zhang et al., 2016). Treatment with TLR4 antagonist LPS-RS reduced CCL2 expression, macrophage infiltration, loss of IENFs, and mechanical hypersensitivity (Zhang et al., 2016). Transient receptor potential vanilloid subtype 1 (TRPV1), which is often associated with nociception was also activated after TLR4 activation (Li et al., 2015). Blocking TLR4 after PTX treatment inhibited mechanical allodynia (Li et al., 2014) and the increase of TRPV1 sensitisation (Li et al., 2015). Interestingly, some studies suggest that TRPV1 is mostly involved in cold allodynia rather than mechanical allodynia, at least in peripheral nerve injury models of neuropathic pain (Cobos et al., 2018). The study further suggests that treatment of cold allodynia and mechanical allodynia may require different strategies based on the idea that these are mediated by separate mechanisms, i.e. tactile allodynia requires an immune response which may in turn activate different fibres that converge in the CNS and result in allodynia, whereas cold allodynia is a result of sensory neuronal processes, particularly TRPV1 lineage nociceptors. How this translates to some PTX-treated animals expressing cold allodynia while others do not is an interesting field to consider; one likelihood is through T-type calcium channels, which will be discussed in section 1.3.8.

Transient receptor potential vanilloid 4 (TRPV4) is increased in the DRG after PTX treatment (Matsumura et al., 2014), and has been shown to be essential for mechanical hyperalgesia (Alessandri-Haber et al., 2008). Transient receptor potential ankyrin 1 (TRPA1), involved in mechanical and cold allodynia, is activated indirectly through oxidative stress, and joint treatment with TRPV4 and TRPA1 antagonists abated mechanical and cold allodynia induced by PTX treatment (Materazzi et al., 2012). Two other drugs that exhibited pain prevention through antioxidant effects and TRPV4 suppression were Goshajinkigan, a traditional Japanese medicine (Matsumura et al., 2014), and glutathione (Materazzi et al., 2012). However, there may be complications in using this pathway to treat neuropathy. Oxidative stress has been shown to help kill cancer cells (Alexandre et al., 2007) and antioxidants seem to reduce the toxicity of PTX on breast cancer cell lines (Hadzic et al., 2010), so attempting to reduce neurotoxicity through reduction of reactive oxygen species (ROS) may have a negative effect on chemotherapy.

1.3.7 Sodium Channels Voltage-gated sodium channels (Navs) have a role in pain sensitisation, and each of the currently identified subtypes (Nav1.1-Nav1.9) have distinct expression patterns in the DRG (Cummins et al., 2000). In particular, Nav1.7 gain-of-function mutations have been associated with DRG hyperexcitability and pain through the increased generation of action potentials (Dib-Hajj et al., 2007, Dib-Hajj et al., 2009). A recent study demonstrated that treatment of dissociated human DRG neuron cultures with PTX caused an increased expression of Nav1.7 and increased transient sodium currents and action potential firing frequency in small diameter neurons (Chang et al., 2017). Another study found that the increase in Nav1.7 was primarily in the small DRG neuron somata, particularly calcitonin gene-related peptide (CGRP)- and TRPV1-expressing cells (Li et al., 2017a). Blocking the channel in vitro using ProTx II, a selective Nav1.7 channel blocker, stopped spontaneous action potential firing, and after intrathecal administration in vivo attenuated mechanical hypersensitivity (Li et al., 2017a). Nav1.7 is considered a tetrodotoxin (TTX)-sensitive channel, and an in vivo study found that certain doses of subcutaneous TTX in PTX-treated mice alleviated mechanical and

20 thermal hyperalgesia, and could also prevent the establishment of mechanical and cold allodynia but not heat hyperalgesia (Nieto et al., 2008). Alternatively, changes in sodium conductance could be the result of altered potassium currents; an in vitro study found PTX to inhibit Kv2.1 voltage-dependent K+ currents (IKur) and therefore PTX could potentially be a Kv2.1 channel blocker (Kitamura et al., 2015). Kv2 downregulation in DRG of nerve injured rats was correlated to mechanical and thermal hypersensitivity (Tsantoulas et al., 2014). The downregulation of voltage-gated potassium channels together with the upregulation of sodium channels in PTX-treated DRG have been suggested to play a role in PTX CIPN by inducing hyperexcitability and spontaneous activity (Zhang and Dougherty, 2014, Boyette-Davis et al., 2015).

1.3.8 Calcium Channels As mentioned previously, PTX seems to affect small DRG neurons, and one indicator of this could be the increased expression of Cav3.2, a protein expressed in T-type (low- voltage activated) calcium channels, which in one study was colocalised with CGRP (peptidergic C-fibre marker)- and isolectin IB4 (non-petidergic C-fibre marker)- positive DRG neurons, but not neurofilament 200 (myelinated large fibre marker) neurons or glial fibrillary acidic protein (GFAP; activated astrocyte cell marker)+ cells (Rose et al., 2013, Li et al., 2017b, Xu et al., 2015, Ruscheweyh et al., 2007). Conversely, others have found they are more specifically found in TrkB-positive Aδ- low-threshold mechanoreceptors (Aδ-LTMRs) and Ret-positive/IB4-negative C- LTMRs which are involved in perception of light-touch and noxious cold and chemical sensations, and not in Aβ fibres (François et al., 2015) thereby proposing a mechanism for cold allodynia that does not involve TRPV1 (Samour et al., 2015) or CGRP. Cav3.2 was shown to have a role in both mechanical and cold allodynia through its expression in C-LTMRs in peripheral neuropathy (François et al., 2015). Cav3.2 was also shown to be co-localised with TLR4, and treatment with TLR4 antagonist TAK-242 before PTX treatment (but not after) was shown to prevent mechanical allodynia in rats (Li et al., 2017b). These findings provide basis for T-type channel antagonists, and specifically

Cav3.2 blockers, to be a good candidate for treatment in neuropathic pain.

PTX was also found to increase cytosolic calcium levels and signalling through the binding of cytoplasmic calcium binding proteins (NCS-1) to triphosphate receptors (InsP3R) and activation of calpain (Mo et al., 2012). Increased calpain activity eventually degrades the calcium binding protein and reduces calcium signalling, which may have a role in neuropathy. Both ibudilast and lithium, which bind to NCS-1 and inhibit InsP3R sensitisation, were shown to prevent the initial excess in calcium signalling and the subsequent chronic decrease of calcium signalling, as well as thermal hyperalgesia in PTX-treated mice (Mo et al., 2012). Systemically blocking calpains using α-ketoamide calpain inhibitor AK295 reduced axonal degeneration and protected from PTX-induced neuropathy in mice (Wang et al., 2004). Another drug that has shown neuropathic pain ameliorating effects is gabapentin (Matsumoto et al., 2006, Gauchan et al., 2009, Xiao et al., 2007, Huehnchen et al., 2013). Gabapentin is thought to modulate Caα2δ-1, a subunit of a voltage-gated calcium channel important in synaptic neurotransmission, which is upregulated in DRG neurons (Matsumoto et al., 2006) or the spinal cord (Gauchan et al., 2009, Xiao et al., 2007) in CIPN. More recently, Minoxidil, a potassium channel opener, has been shown to have a protective effect on cultured neurons by preventing the increase of intracellular Ca2+ (Chen et al., 2017). The drug also alleviated mechanical allodynia, prevented thermal hyperalgesia and reduced the infiltration of pro-inflammatory M1 macrophages in mice treated with Minoxidil prior to PTX treatment (Chen et al., 2017).

1.3.9 Mitochondria It has been suggested that a major effect of changes in ion channels and disrupted axonal transport is damage to mitochondria. In vivo studies of mitochondria in rats treated with PTX found them to be swollen and vacuolated in C-fibres and myelinated axons (Flatters and Bennett, 2006), suggesting a link between mitochondrial damage and pain. Swollen and vacuolated mitochondria were also found in myelinated and unmyelinated sensory axons of the dorsal root, but not the ventral root motor axons (Xiao et al., 2011) in agreement with most studies that find no damage to locomotion or motor control in PTX-treated animals. The authors suggest this may be because sensory neurons in the DRG were exposed to PTX for much longer and at greater concentrations than motor neuron cell bodies protected by the blood-spinal cord barrier in the ventral horn of the spinal cord. Mitochondrial damage is also suggested to be caused by a 22 reduction of mitochondrial mRNA transportation due to damaged axons (Bobylev et al., 2015). PTX also caused deficits in mitochondrial ATP production in the sciatic nerves of rats, along with mitochondrial swelling before the onset of pain symptoms (Zheng et al., 2011). These changes were not seen in rats treated with acetyl-L-carnitine (ALCAR), an amino acid derivative that is essential in transporting certain fatty acids into mitochondria (Zheng et al., 2011). The protective effects of ALCAR were found to be specific to C-fibre mitochondria and not A-fibre mitochondria (Jin et al., 2008), although it was found to block spontaneous discharging of both A- and C-fibres in a different study (Xiao and Bennett, 2008). Interestingly, another study found that PTX produced swollen and vacuolated mitochondria in A-fibres and not C-fibres of the saphenous nerve (Nieto et al., 2014), showing some discrepancies in the literature. Nevertheless, it can be suggested that following PTX treatment mitotoxicity has a role in the development of CIPN.

Energy deficits in the form of reduced ATP levels, decreased maximal respiration and spare reserve capacity, and increased glycolytic function were observed in the DRG during pain onset in rats treated with PTX (Duggett et al., 2017). This was provided as evidence for the nature of mitochondrial dysfunction in PTX treatment and a cause of pain through the energy deficit of DRG, although none of these molecular changes were replicated in in vitro treatment of DRG neurons. This may be due to the method through which the cells were analysed, as other groups have found reduced cellular ATP levels in PTX-treated DRG neurons, correlated with axonal degeneration (Cetinkaya-Fisgin et al., 2016).

Another mechanism through which mitochondria are affected could be oxidative stress and the production of ROS in neurons (Duggett et al., 2016). MitoVitE, a mitochondria- targeted antioxidant, was found to reduce oxidative stress and mitochondrial damage from PTX in vitro and reduce mechanical hypersensitivity in rats (McCormick et al., 2016). The same group later found that melatonin, a well-known antioxidant, could reduce mechanical hypersensitivity in PTX-treated rats and prevented the reduction of mitochondrial membrane potential and metabolic rate in in vitro studies using an immortalised DRG neuronal stem cell line (Galley et al., 2017). Certain complexes in

23 the electron-transport chain are known to produce ROS as a result of electron transfer (Turrens and Boveris, 1980, Turrens et al., 1985); antimycin A, a complex III inhibitor, and rotenone, a complex I inhibitor, were found to reverse or prevent PTX-induced mechanical hypersensitivity in rats, respectively, suggesting that activity in the electron- transport chain could be a potential target for treatment (Griffiths and Flatters, 2015).

Furthermore, PTX’s effect on mitochondria could be due to the association of tubulin with mitochondrial permeability transition pore (mPTP) and the mitochondrial membrane (Carré et al., 2002), thereby affecting mitochondria both directly and indirectly. This poses an interesting question as to whether a neuroprotectant or mitochondrial protectant alone will be enough to prevent neuropathic pain. PTX induces the over-activation of mPTP which results in swelling, a release of cytochrome c, and apoptosis in vitro (André et al., 2000), and an active mPTP has been suggested to be key in causing axonal degeneration (Barrientos et al., 2011). It also causes Ca2+ release from intracellular Ca2+ stores and rapidly reduces mitochondrial membrane potential (Kidd et al., 2002). Increased calcium could be a potential cause of neuronal excitability and spontaneous pain states. Blocking the mPTP has been shown to stop axonal degeneration (Barrientos et al., 2011) and treating with calcium chelating agents has reduced pain in some animal models (Siau and Bennett, 2006). 5-Hydroxydecanoate (5- HD), a mitochondrial K+ channel antagonist, was shown to help reduce pain through its regulation of calcium in DRG neurons (Chen et al., 2015). Treatment with sigma-1 receptor (σ1R) agonist BD-1063, which is involved in mitochondrial calcium homeostasis, was also shown to prevent mitochondrial changes and pain hypersensitivity in PTX-treated mice (Nieto et al., 2014).

1.3.10 Central Nervous System (CNS) PTX is detectable at very low concentrations in the CNS after intravenous administration and does not readily cross the blood brain barrier (BBB) (Gornstein and Schwarz, 2014, Fellner et al., 2002). Nevertheless, some studies suggest it may have an effect in the spinal cord and brain, particularly through activation of non-neuronal cells. This is most likely a secondary effect resulting from its peripheral activity rather than any direct effect on CNS cells. One study found CNS effects were restricted to the

24 dorsal column of the spinal cord in the form of degenerated myelin axons which persisted for at least 2 months, indicating PTX affects ascending sensory axons (Tasnim et al., 2016).

Glial cells within the CNS play a pivotal role in the pathology of NP (Pevida et al., 2013, Ruiz-Medina et al., 2013). There is some debate as to whether astrocytes or microglia play a role in the initiation versus the maintenance of neuropathic pain. In a study using Sprague-Dawley rats treated with PTX, astrocytes, but not microglia, were found to be activated in the spinal cord (Zhang et al., 2012). However, one study showed an increase in both microglia and astrocytes in the dorsal horn of young (31 days) and aged (12-13 months) mice treated with PTX, although less so in adult mice (3-4 months) (Ruiz-Medina et al., 2013). These findings indicate variability in gliosis due to age and species-related differences. Interestingly, the higher glial response was correlated with a higher pain response in the mice that exhibited gliosis (Ruiz-Medina et al., 2013). An earlier study conducted on Sprague-Dawley rats found an increase in microglial activation in the dorsal horn of the spinal cord at an early time point (Peters et al., 2007a). Furthermore, during PTX-induced thermal hyperalgesia, an increase in astrocytes was found in the anterior cingulate cortex (ACC), an area in the brain involved in pain processing (Masocha, 2015).

Minocycline treatment led to attenuation of both microglia and astrocyte levels and alleviated CIPN (Burgos et al., 2012), and reduced cluster of differentiation molecule 11B (CD11B; a macrophage marker), GFAP and inducible nitric oxide synthase (iNOS) overexpression. Additionally, ibudilast, a phosphodiesterase inhibitor that reduces proinflammatory cytokine production from microglia, was shown to alleviate allodynia in a rat model of CIPN (Ledeboer et al., 2006). PTX was also found to increase levels of CCL2 and active CCR2 in the spinal cord, which correlated with cold hypersensitivity and microglial activation, and intrathecal injection of anti-CCL2 antibodies suppressed cold hyperalgesia and microglial activation (Pevida et al., 2013). Although there are some contrasting studies on whether astrocytes or microglia are activated in the CNS during PTX-induced pain, glial or astrocyte blockers could be a potential target for pain attenuation.

TLR4 has also been found to be upregulated in the spinal cord of PTX-treated rats, co- localised with astrocytes, but not microglia or neurons, and lasting for two weeks after which TLR4 levels had returned to normal in the DRG (Li et al., 2014). Intrathecal treatment with TLR4 antagonist LPS-RS transiently reversed mechanical allodynia, although systemic administration also produced the same behavioural phenotype with a reduction in TLR4 in the DRG (Li et al., 2014).

Another pathway that may cause CIPN through spinal neuroinflammation is the ceramide to ceramide-sphingosine 1-phosphate (S1P) pathway, which has previously been implicated in the development of neuropathic pain and is also important in regulating cell fate (Janes et al., 2014b). The development of PTX-induced neuropathy was associated with S1P receptor subtype 1 (S1PR1)-dependent neuroinflammatory processes in the spinal dorsal horn, which includes the activation of NFκB, TNF-α, and

IL-1β. Intrathecal injection of S1PR1-antagonist W146 reduced neuroinflammatory processes, increased anti-neuroinflammatory cytokines IL-10 and IL-4, and reversed neuropathic pain in PTX-treated rats. The study concluded that the SIP/S1PR1 axis may be a potential target for CIPN treatment and does not influence the anti-cancer effects of chemotherapy (Janes et al., 2014b).

1.4 Effects of PTX in vitro In vitro studies have shown that PTX directly targets neuronal axons more than cell bodies, causing axon fragmentation, swelling or degeneration (Yang et al., 2009, Gornstein and Schwarz, 2014) and large fibres are the most affected of these (Dougherty et al., 2004). While some groups have shown PTX inhibits neurite outgrowth in DRG neurons in a dose dependant manner (Chen et al., 2015, Chen et al., 2017), others have seen no difference in overall axonal outgrowth, but have instead shown an increased number of retraction bulbs (characteristic swellings at the axon tips) (Gornstein and Schwarz, 2017). The discrepancies in the above studies may be due in part to the time-point at which the cells were treated, with the former being treated at day (D)1 for 24 hrs and the latter at D4 or D6 for 48-72 hrs. Axonal growth was shown to be inhibited within a few hours of PTX treatment, followed by the formation of retraction bulbs, which contained disorganised and deacytelated microtubules as well as an accumulation of mitochondria (Gornstein and Schwarz, 2017). Axonal degeneration is also linked to increased detyrosinated tubulins, another marker for microtubule stability (Melli et al., 2006).

While most evidence suggest that cell bodies are much less affected by PTX than axons (Gornstein and Schwarz, 2017, Yang et al., 2009), recent studies postulate that, considering the different dynamics of microtubules in the axons and cells bodies, microtubule-stabilising agents such as PTX may simply have differential effects in these areas (Benbow et al., 2017). PTX caused increased ATF3-stained nuclei and increased acetylated tubulin, signifying greater neuronal stress and cytoskeletal stabilisation. Cell bodies have also been observed to enlarge in PTX-treated neuronal cell cultures (Guo et al., 2017). Differences in observations are also likely due to treatment time points and method of culture.

Although in vitro testing can be very useful for testing mechanisms or conducting high- throughput testing of specific effects, there are some shortfalls to this method. For example, a recent study showed significant energy reduction in the DRG in PTX-treated rats during pain onset, but no effect was found in cultured rat DRG neurons (Duggett et al., 2017). As previously mentioned however, other studies have found some ATP

27 reduction in culture (Cetinkaya-Fisgin et al., 2016), demonstrating the importance of finding a particular method of culturing, treating, and assessment that is physiologically relevant. There are also likely to be differences between rodent and human DRG cultures. A recent study comparing cultured normal and PTX-treated human and mouse DRG neurons found discrepancies between Nav expressions, in which there were much higher expressions of Nav1.7 to Nav1.8 in humans and vice versa in mice (Chang et al., 2017). In addition, PTX was found to cause an increase in Nav1.7 expression, but not Nav1.8 in these neuronal cultures (Chang et al., 2017). Therefore, the clinical relevance of findings from in vitro primary cultures of rodent sensory neurons must be interpreted with caution.

1.5 CIPN management At present, there is no preventative or curative treatment for CIPN, and the current methods of managing CIPN are based on treatment dose reduction or discontinuation, which may ultimately affect overall survival (Brewer et al., 2016). While many interventions have been studied for prevention and treatment, only duloxetine (an anti- depressant agent) has proven moderately efficacious at reducing neuropathic pain symptoms in CIPN (Smith et al., 2013) and is the only US Food and Drug Administration (FDA)-approved drug for treating this condition. Duloxetine is a serotonin–norepinephrine reuptake inhibitor (known by the brand name Cymbalta) that has been shown to have some efficacy in ameliorating or delaying the onset of neuropathic pain due to diabetes (Goldstein et al., 2005) and PTX- or oxaliplatin- induced NP (Smith et al., 2013, Hirayama et al., 2015, Wang et al., 2017), even in different forms of PTX treatment, such as PTX in conjunction with carboplatin or epirubicin (Otake et al., 2015) or nab-PTX (Abraxane) with gemcitabine (Suzuki et al., 2017). However, considering the lack of effective therapies, novel approaches, such as neuroprotective drugs, for the management of CIPN are needed.

As with all potential CIPN therapies, there is concern that blocking the action of the chemotherapeutic, which for PTX is microtubule stabilisation and mitotic arrest, will render PTX ineffective against the cancer. This does not mean drugs that have these effects should not be tested; recent studies seem to show that microtubule stabilisation is

28 not the only way tumour cells are killed, and instead propose a mechanism where cells form abnormal spindles and induce multipolar divisions where daughter cells die due to loss of essential chromosomes (Weaver, 2014, Zasadil et al., 2014). Since the mechanisms of PTX are still being elucidated, neuroprotective candidates with established mechanisms may help to better understand PTX’s mechanisms of action, whether they prove to be clinically relevant or not. The mechanisms outlined in this review are summarised in Figure 1.3.

Figure 1.3. Effects of PTX on various tissue based on murine models. PTX causes apoptosis in tumour cells through two possible mechanisms: 1. Mitotic arrest due to depolymerisation of tubulins, or 2. Multipolar divisions leading to loss of essential chromosomes in daughter cells. PTX has been shown to cause inflammation in the spinal cord and areas of the brain. Peripheral inflammation also seems to cause muscle degradation in some instances. In peripheral nerves such as the sciatic nerve, there is an increase in axonal degeneration and decrease in neuronal transport. A loss of IENFs has been shown to cause neuropathic pain without the degradation of larger nerve 29 fibres. In the DRG as a whole, there is an increase in inflammation due to inflammatory cytokines, infiltration of immune cells, and over-activation of certain receptors involved in pain pathways. The mitochondria has been shown to be damaged in various cells or axons, with increased mPTP activation and calcium output, and decreased ATP production. These actions in the peripheral nerves, DRG and mitochondria all directly or indirectly induce CIPN.

1.6 Summary As cancer treatments and longevity of cancer patients improve, more survivors are affected with the detrimental long-term effects of certain cancer treatments, such as CIPN. PTX is a commonly used chemotherapeutic that can induce both short- and long- term painful neuropathy which can be dose-limiting and debilitating, thereby hindering treatment of cancer. Cures, preventative treatments, or interventions for symptom improvement for those who develop CIPN are inadequate. One way of providing a quick transition from the bench to bedside is repurposing drugs that are already approved for use in other conditions. Based on the current literature, we have screened several clinically approved drugs (ibudilast, nicotinamide mononucleotide, resatorvid, amiloride, duloxetine, safinamide) in an in vitro dissociated DRG neuronal model of PTX neurotoxicity (Chapter 2). We then created an animal model of PTX-induced peripheral neuropathy and tested the effects of a selected neuroprotectant in PTX- treated mice (Chapter 3).

Thesis Aims

Aim 1 (Chapter 2):

i) Establish a DRG primary neuronal culture model whereby the neurotoxic effects of paclitaxel can be effectively assessed in vitro ii) Test the effects of a range of drug candidates on axonal outgrowth in paclitaxel-treated primary neuronal cells

Aim 2 (Chapter 3):

i) Establish a chronic animal model of paclitaxel-induced peripheral neuropathy and assess phenotypic changes in the nervous system ii) Test the best candidate drug identified in Aim 1 in the animal model

Hypothesis: PTX induces neuronal changes both in vitro and in vivo, which can be prevented by selected repurposed drugs.

Chapter 2: Optimisation of in vitro sensory neuronal culturing model and testing of potential neuroprotectants against PTX-induced neurotoxicity

Chapter 2: Optimisation of in vitro sensory neuronal culturing model and testing of potential neuroprotectants against PTX-induced neurotoxicity Contributions

Munawwar Abdulla designed experiments, performed the cell preparation, drug administration, immunohistochemistry, image analysis, and drafted the chapter. Justin G Lees was involved in the conception of the study, assisted with interpretation of data and revision of chapter, and conducted the imaging of the cells. Gila Moalem-Taylor conceived and designed the study, and assisted with revision of the chapter.

2.1 Introduction

PTX-induced peripheral neuropathy is a severe side-effect of cancer treatment that can result in dose reduction, delays, or complete cessation of chemotherapy (Brewer et al., 2016). Affected patients report sensory symptoms such as paraesthesia, dysesthesia, numbness, electric shock-like sensation, hyperalgesia and allodynia. The DRG, containing the cell bodies of peripheral sensory neurons and their associated axons are often demonstrated to be adversely affected by PTX treatment in preclinical CIPN models (Argyriou et al., 2014). However, the mechanisms by which PTX causes neuropathy are not completely understood, and there are no medications currently available for prevention or treatment of this condition.

In vitro studies have shown that following treatment of DRG neurons with PTX there is an inhibition of neurite outgrowth (Chen et al., 2015, Chen et al., 2017) with significant decrease in the number of neurons that grow neurites and the number and length of neurites (Ustinova et al., 2013), neurite retraction (Chen et al., 2016), reduced axon length (Cetinkaya-Fisgin et al., 2016) and the formation of retraction bulbs at the ends of axons (Gornstein and Schwarz, 2017). Other changes include mitochondrial impairment (Melli et al., 2008), increased expression of TRPA1 and TRPV4 due to the release of TNF-α from Satellite glial cells (SGCs) (Cetinkaya-Fisgin et al., 2016), reduced ATP levels (Cetinkaya-Fisgin et al., 2016), increased glutathione peroxidase (Duggett et al., 2016) and spontaneous discharges from DRG neurons (Krukowski et al., 2016). Thus, axonal effects are clearly evident in in vitro studies (Yang et al., 2009, Gornstein and Schwarz, 2017). Others have also demonstrated changes in the cell bodies of PTX-treated DRG neurons, including signs of stress, injury and microtubule dysregulation (Benbow et al., 2017), as well as increased soma size (Guo et al., 2017). Despite these cell body effects, there is little debate that axons are vulnerable to PTX- induced damage.

Another interesting observation is that PTX causes the formation of retraction bulbs at the tips of growing axons (Gornstein and Schwarz, 2017). These are spherical structures containing disorganised stable microtubules and mislocalised dynamic microtubules

(Ertürk et al., 2007). Unlike growth cones, retraction bulbs grow larger over time and can be caused by microtubule destabilisation in vitro (Ertürk et al., 2007) or by abnormal cytoskeletal dynamics (Blanquie and Bradke, 2018) and generally prevents regeneration of axons. Figure 2.1 presents the differences between growth cones, which are sometimes formed after nerve injury to promote regeneration and retraction bulbs. Interestingly, low doses of PTX have been shown to promote the formation of growth cones in the CNS through its microtubule-stabilising effects (Sengottuvel et al., 2011). The study showed that at doses between 0.5-3 nM, PTX increased neurite extension in retinal cell cultures, but at 10-50 nM the neurite lengths were reduced compared to control. This suggests that microtubule hyperstabilisation can also cause the formation of retraction bulbs (Gornstein and Schwarz, 2017).

Figure has been removed due to copyright restrictions.

Figure 2.1. Cytoskeletal organization in growth cones versus retraction bulbs. (a) Growth cones display three distinct regions: the centre contains microtubules which emerge from the axonal shaft, the periphery contains a dense network of actin filaments, and the transition domain borders the two regions and allows for control of various organelles from reaching the periphery. (b) In the retraction bulb, the separation between the three regions is lost, and microtubules are depolymerized, disorganized, or do not reach the axon tip. Taken from (Blanquie and Bradke, 2018).

Despite causing prevention of axonal outgrowth and retraction bulb formation, PTX does not cause cell death or axon fragmentation at low doses (around 10 nM) (Gornstein 34 and Schwarz, 2017). Higher doses of PTX starting at 10 µM induce the dying back of neurites, as well as non-neuronal cell processes (Guo et al., 2017). A recent study implicates a Bclw-IP3R1-dependent cascade in the causation of axon degeneration, whereby PTX damages axonal trafficking through microtubule stabilisation, thereby reducing levels of Bclw, which no longer binds to IP3R1 to prevent axon degeneration (Pease-Raissi et al., 2017). The study also finds mitochondrial dysfunction and calcium dysregulation to be characteristic of this model of axon degeneration. This hypothesis contradicts another recent study which postulates that microtubule stabilisation itself, and not the problems arising from axonal transportation or any tubulin post-translational modification, are the cause of CIPN (Gornstein and Schwarz, 2017). The conflicting ideas within the literature could be due to the different models used to test the neurotoxic effects of PTX, where lower doses cause less damage than higher doses. Indeed, the former study used 30 nM – 1.2 µM, whereas the latter used 10 – 50 nM. As mentioned previously, lower doses of PTX are found to enhance neurite outgrowth in vitro when grown in the presence of an inhibitory environment (Ertürk et al., 2007) or to have no effect at all at doses below 1 nM (Ustinova et al., 2013). In addition, these studies also differed in the growth factors they used for neuronal cultures; Nerve growth factor (NGF), glial cell line-derived neurotrophic factor (GDNF) and L-glutamine (Gornstein and Schwarz, 2017) compared to GlutaMAX, glucose and brain-derived neurotrophic factor (BDNF)+NGF (as well as cytosine arabinoside (Ara-C), used in cultures to reduce the number of non-neuronal cells) (Pease-Raissi et al., 2017). Thus, inconsistencies in the culture media and culturing environment make it difficult to directly compare published data.

In vivo studies have also demonstrated axonal degeneration in large- or medium-sized fibres (Tasnim et al., 2016) and small fibres (Bobylev et al., 2015), as well as the formation of retraction bulbs (Ertürk et al., 2007) following PTX treatment. Furthermore, PTX was shown to promote inflammation in the DRG, including increased release of inflammatory cytokines TNF-α and IL-1 (Ledeboer et al., 2007b) and increased macrophage infiltration (Ledeboer et al., 2007b, Peters et al., 2007a, Nishida et al., 2008, Liu et al., 2010, Zhang et al., 2016). Other changes in the DRG include an upregulation of matrix metalloproteinase-3 (MMP-3) (Nishida et al., 2008), increased TLR4 (Wu et al., 2015), increased CCL2/CCR2 (Sisignano et al., 2014, 35

Zhang et al., 2013), increased TRPV4 (Alessandri-Haber et al., 2008, Matsumura et al., 2014), upregulation of certain voltage-gated calcium channels (Matsumoto et al., 2006), increased low-voltage–activated calcium channels (T-type; Cav3.2) co-localised with TLR4 (Li et al., 2017b) and a decrease in substance P (Apfel et al., 1991). Some of these effects were taken into consideration when examining the literature for potential neuroprotective drug candidates in this thesis.

Drugs that have shown efficacy in preventing axonal degeneration in vitro are minoxidil, a potassium channel opener (Chen et al., 2017), recombinant human erythropoietin (rHuEPO) through downregulation of detyrosinated tubulin (Melli et al., 2006) and 5-hydroxydecanoate (5-HD), a mitochondrial K+ channel antagonist, through regulating calcium homeostasis (Chen et al., 2015). The latter study also found mild effects of amiloride, glyburide, gingerol, minoxidil sulfate, chlormezanone, hinokitiol, and propyl gallate in PTX-treated DRG cultures (Chen et al., 2015) although they did not conduct follow-up studies of these drugs. NGF is part of the neurotrophin family and is known to have a trophic and maintenance role in various neuronal and immune/glial cells in the nervous system (Aloe et al., 2012). NGF was found to protect cells from the reduction of substance P after PTX treatment (Apfel et al., 1991) and be neuroprotective against vincristine-, PTX-, and cisplatin-induced neuropathy demonstrated by measuring neurite outgrowth of intact rat DRG grown in vitro (Konings et al., 1994). In dissociated primary DRG neuron cultures, NGF has been suggested to reverse PTX-induced hyposensitivity as measured by stimulated CGRP release, although it had no effect on neurite outgrowth (Pittman et al., 2016). Nevertheless, NGF-treated neurons had slightly enhanced neurite outgrowth and branching than untreated neurons (Pittman et al., 2016). Studies looking at other chemotherapeutics have suggested that NGF can prevent cisplatin-induced neuropathy in an experimental model (Alberts and Noel, 1995), protect from cisplatin-induced cell death in DRG neurons and PC12 cells (Fischer et al., 2001, McDonald and Windebank, 2002), and protect from cisplatin and vincristine neurotoxicity in dissociated rat DRG neurons following pre-treatment with NGF for 24 hours (Malgrange et al., 1994). More recently, NGF has been found to be upregulated in chemotherapy patients that develop CIPN (Velasco et al., 2017). Whether NGF is a precursor to pain or simply correlated

36 with pain is yet to be elucidated, however its in vitro neuroprotective effects seem worthwhile to study.

To screen drugs that can be quickly placed in clinical trials, various clinically approved drug candidates were chosen based on the literature, so they could be repurposed if found to have protective effects against PTX-induced neurotoxicity. Here, candidate drug selection strategy was carried out based on those drugs that have shown potential clinical efficacy in neuropathic pain states or neurodegenerative disorders and had been approved for human clinical trials. Drugs that had previously been used in animal studies of PTX were also selected, such as Duloxetine, Ibudilast and TAK-242. NMN, Safinamide, and Amiloride were chosen through searching “pain”, “neuropathy” or “neuropathic pain” in the DrugBank database (Wishart et al., 2018) and search results were filtered for drugs that were already clinically approved.

2.1.1 Ibudilast Ibudilast is a registered drug used to treat bronchial asthma, cerebrovascular disorders, post-stroke dizziness and some ocular allergies due to its various effects on platelet aggregation, smooth muscle contractility and cerebral blood flow (Ledeboer et al., 2007a). It is a non-selective cyclic AMP phosphodiesterase (PDE) inhibitor (Souness et al., 1994) and a macrophage migration inhibitory factor (MIF) inhibitor (Ellis et al., 2014). In regards to inflammatory pain, PDEs have been found to promote glial activation (Zhang et al., 2002) and MIF is a proinflammatory cytokine that stimulates the release of IL-1 and TNF-α (Toh et al., 2006). Furthermore, ibudilast has been shown to suppress nitric oxide (NO), ROS, IL-1β, IL-6 and TNF-α, and enhance IL-10 production in activated microglia, thereby protecting neurons from cell death through activation of anti-inflammatory signalling pathways in cell cultures (Mizuno et al., 2004, Suzumura et al., 1999). It is also suggested to be a TLR4 receptor antagonist (Ruiz-Pérez et al., 2016). Ibudilast may therefore be an appropriate neuroprotectant for PTX-induced peripheral neuropathy due to the finding that PTX interacts with TLR4. Moreover, ibudilast has been shown to reverse mechanical allodynia in other neuropathic pain models such as chronic constriction injury (CCI) of the sciatic nerve and spinal nerve ligation (SNL) (Ledeboer et al., 2006, Hama et al., 2012) and central

37 neuropathic pain (Ellis et al., 2014). One study found ibudilast and lithium to prevent PTX-induced peripheral neuropathy development in female mice (Mo et al., 2012) and another found ibudilast efficacy in a rat model (Ledeboer et al., 2006). Ibudilast can also cross the BBB (Ledeboer et al., 2007a), and although PTX does not have a substantial direct effect on the CNS, there are changes in glial activation in the spinal cord and supraspinal areas (Masocha, 2015), which could be alleviated by ibudilast. Of course, these effects cannot be seen in vitro, however we decided to include ibudilast in the study since it also interacts with TLR4 possibly expressed on neurons, and thereby may alter PTX-induced neurotoxicity.

2.1.2 Nicotinamide mononucleotide (NMN) Nicotinamide mononucleotide (NMN) is a nicotinamide adenine dinucleotide (NAD) precursor, and has been shown to increase levels of NAD+ (Picciotto et al., 2016). NAD+ is a signalling molecule involved in energy metabolism, DNA repair, and telomere maintenance (Belenky et al., 2007). It is a cofactor for deacetylase sirtuin1 (SIRT1), ADP-ribosyl cyclase (CD38), poly(ADP-ribose) polymerase 1 (PARP1) and is essential for enzymatic reactions in the mitochondria (Long et al., 2015). A reduction of NAD+ is associated with decreased mitochondrial activity, cellular respiration and ATP production, perhaps resulting in apoptosis (Long et al., 2015, de Picciotto et al., 2014) and it is thought to be associated with ageing and obesity (Uddin et al., 2016, Gomes et al., 2013, Picciotto et al., 2016, Wu and Sinclair, 2016). As such, NMN is currently being tested for treatments regarding high-fat diet induced obesity and diabetes (Uddin et al., 2016), age-related cellular degeneration or arterial dysfunction (Picciotto et al., 2016, Wu and Sinclair, 2016), ischemia (both heart and brain) (Yamamoto et al., 2014, Park et al., 2016) and Alzheimer’s disease (Long et al., 2015). In addition to its potential protective effects in neurodegenerative diseases, activation of AMPK/SIRT1 pathways has been shown to have anti-allodynic effects in other models of neuropathic pain such as CCI (Gui et al., 2015). NMN and NAD+ may also have a role in axonal or Wallerian degeneration (Di Stefano et al., 2015). Since PTX has damaging effects on the mitochondria and may specifically target axons, NMN has potential therapeutic effects in this model of CIPN. Other advantages are that the dose and regimen for therapeutic effects in mice are known and can be provided through oral consumption rather than injections (Mills et al., 2016, Das et al., 2017). 38

2.1.3 Resatorvid (TAK-242) TAK-242 inhibits TLR4 activation by binding to intracellular domain sites and disrupting interactions between adaptor molecules and TLR4 (Matsunaga et al., 2011). It was first shown to suppress the production of NO and inflammatory cytokines from macrophages and monocytes (Ii et al., 2006) and has since been shown to have anti- inflammatory effects on animal models with stress-induced neuroinflammation (Garate et al., 2014), lung injury and atherosclerosis (Ni et al., 2013), aldosterone-induced cardiac and renal injury (Zhang et al., 2015), cerebral ischemia/reperfusion injury (Hua et al., 2015), sepsis when co-administered with ceftazidime (Takashima et al., 2009), endotoxin shock (Sha et al., 2007), polymicrobial sepsis (Sha et al., 2011), kidney injury from endotoxemic sheep (Fenhammar et al., 2011) and endotoxaemia in guinea pigs with sepsis (Kuno et al., 2009). The drug has been used in clinical trials to prevent severe sepsis due to its role in suppressing inflammatory cytokine expression (Rice et al., 2010). TAK-242 failed to suppress cytokine activation in sepsis patients, however the drug was shown to be well tolerated in humans (Rice et al., 2010). Since TLR4 and inflammatory cytokines have been implicated in some pain states, it would be interesting to see if TAK-242 had a neuroprotective effect. Indeed, it was used in a recent study that showed that rats pre-treated with TAK-242 did not develop mechanical allodynia after PTX treatment, although it had no pain ameliorating effects when administered after treatment (Li et al., 2017b). TAK-242 has also been shown to cross the BBB, and remain in the serum for at least 24 hrs in mice studies (Hua et al., 2015). Neurons isolated from TLR4 knockout mice were somewhat protected as well, as they exhibited normalised numbers of neurites per cell after treatment with PTX, although there was no significant protection of the length of neurites (Ustinova et al., 2013).

2.1.4 Amiloride Amiloride is a weak, proton-gated cation channel inhibitor and potassium sparing diuretic (Waldmann et al., 1997, Frelin et al., 1988, Kleyman and Cragoe, 1988) that is currently used to treat congestive heart failure or hypertension (Drugbank.ca, 2005). More recently, acid-sensing ion channel (ASIC) blockers such as amiloride have been shown to have an effect on pain sensitivity following peripheral nerve injury (Jeong et al., 2013) and vincristine-induced neuropathic pain (Muthuraman et al., 2008), as well as nociception induced by injection of chemical algogens (Córdova et al., 2011), 39 formalin (Kolasani et al., 2016, Rocha-González et al., 2009), complete Freud’s adjuvant (Dube et al., 2005), serotonin and capsaicin (Rocha-González et al., 2009). ASICs are known to modulate pain signals at both central and peripheral levels (Deval et al., 2010). ASICs are found in DRG neurons and spinal cord dorsal horn, and therefore could be a potential target in CIPN (Dube et al., 2005, Jeong et al., 2013).

2.1.5 Duloxetine Duloxetine is a serotonin (5-HT) and noradrenaline (NA) reuptake inhibitor (SNRI) used to treat major depressive disorder and generalised anxiety disorder (Wohlreich et al., 2007, Carter and McCormack, 2009). It has also shown efficacy in treating diabetic neuropathic pain and fibromyalgia (Lunn et al., 2014), as well as CIPN. Indeed, duloxetine has already undergone clinical trials for treating CIPN with positive results in a proportion of patients, reducing neuropathic pain (Smith et al., 2013, Hirayama et al., 2015, Wang et al., 2017), delaying onset of pain (Suzuki et al., 2017) and decreasing numbness and tingling (Smith et al., 2013). One study found that duloxetine inhibits P2X4 receptors, which are ATP-gated non-selective cation channels that are activated in spinal microglia during neuropathic pain, suggesting that duloxetine could be reducing pain through glial mechanisms as well as through the inhibition of 5-HT and NA reuptake (Yamashita et al., 2016). Duloxetine reduced mechanical hyperalgesia in rats with SNL, which was reversed through the intrathecal injection of α2-adrenoceptor antagonist (Ito et al., 2018), indicating a different potential mechanism of action with regards to pain that involves spinal NA. Few studies have been conducted to understand how duloxetine affects neuropathic pain, or its direct effect on sensory neurons.

2.1.6 Safinamide Safinamide is most commonly known as an anticonvulsant used for the treatment of Parkinson’s disease (PD) due to its selective inhibition of monoamine oxidase (MAO)- B, which is considered a dopaminergic replacement therapy (Deeks, 2015, Fariello et al., 1998). However, it can also inhibit voltage-gated sodium and N-type calcium channels and glutamate release and is known to have neuroprotective effects (Caccia et al., 2006, Morari et al., 2018). One clinical study has found that patients with PD given safinamide had reduced pain and used less analgesic treatment than those who were not 40 treated with safinamide (Cattaneo et al., 2017). Considering safinamide’s effect on sodium and calcium channels, its general neuroprotective effect in the brain and a clinical study showing reduction of pain in PD patients, safinamide has potential for treatment of patients with CIPN.

While these drugs have all demonstrated some effects on neuropathic pain in vivo, their direct effects on sensory neurons have not been investigated. Therefore, this study aimed to first establish a DRG primary neuron culture model, whereby the effects of PTX can be adequately and robustly observed in vitro, and secondly to test the effects of a range of neuroprotective drugs on PTX-treated primary neuronal cells.

2.2 Materials and Methods

2.2.1 Animals Five to six-week-old male C57BL/6 (WT) mice (Australian BioResources, Moss Vale, NSW, Australia) were used for culture preparations. Mice were group-housed with water and food ad-libitum and maintained on a 12:12 hour light/dark cycle at constant room temperature and humidity levels. All animal experiments were approved by the Animal Care and Ethics Committee of the University of New South Wales, UNSW Sydney, Australia.

2.2.2 Dissociated DRG neuron culture A range of different purification, plating and culturing methods were assessed before a final protocol was decided upon, as outlined in Table 2.1. Cells were grown for up to a week per protocol while observed daily under a bright-field microscope for detection of changes in morphology. Based on these observations, cells had started growing axons by 24 hours and reached peak axonal extension by ~72 hours. Therefore, day 3 (D3) was chosen as endpoint for our acute treatment short-term model, and D4 for the established treatment model. Morphological changes in cells treated with PTX were evident by D3 for the short-term model and by D4 for the established model. The general protocol for isolation and growth of primary neuronal culture was adapted from (Lee and Levine, 2015, Lee et al., 2013).

Testing Condition Outcome Media NBA + B27 + P/S + Ara-C Ara-C killed supporting cells but also hindered growth of neurons NBA + B27 + P/S + Ara-C + Ara-C continued to be deleterious NGF + GlutaMAX NBA + B27 + P/S + NGF + Growth factors allowed axons to grow GlutaMAX rapidly and for a sustained period * NBA + B27 + P/S Promoted good axon growth for the 5- day experimental period

Purification No purification (collagenase, Collagenase and trypsin remove some trypsin, 70 micron filter) fibroblasts 30 minute adhesion to tissue A lot of debris and non-neuronal cells culture plate (with NBA) still visible 30 minute adhesion to tissue No significant difference culture plate (with serum included) BSA cushion Cell population too low to do testing Percoll gradient (2.6%) Cell population too low * Percoll gradient (30% and 60% Removed myelin and Schwann cell layers of 90% Percoll) debris, some supporting cells still present but much clearer culture. Plating 20 mm glass coverslips (4-8), Cells scattered, difficult to control in petri dish, 200-250 µl, then environment flooded petri dish with media 20 mm glass coverslips (4-8) in Easier to control individual 24 well plate, 200-250 µl then environments, cells prefer edges to 1 ml media centre of coverslip 12 mm glass coverslips (6-12) Lower density, less growth in 24 well plate, 150-200 µl then 1 ml media 12 mm glass coverslips, 6, 24- Denser and more even cell population well plate, 150 µl then 850 µl media 12 or 24 wells in 96- well plate, 1 mouse DRG is enough for 12 wells 50 µl coat and seed, flood with rather than 24, however cells adhere to another 50 µl edges of the well * 12 wells in 96- well plate, coat Higher density, cells clustered in and seed only the centre with 5- centre, can immune-stain directly onto 10 µl then flood with approx. plate 90 µl 12 coverslips in 24-well plate, Higher density, used for coat with 100 µl PDL and 5-7 immunohistochemistry (IHC) studies µl laminin, flood with approx. that could not be imaged in a 96-well 90 µl plate Table 2.1. Optimisation of cell culture conditions. * indicates the preferred protocols used in this chapter. Cells were grown for up to a week.

2.2.2.1 96-well plate preparation 96-well plates were sterilised under UV light then coated with ~6.5 µl of poly-D-lysine (PDL) (Sigma) (0.1 mg/ml in filtered MilliQ water) for 15 minutes. This volume of PDL covered a central circular area in the middle of the 96 well plate, thereby coating only a region where cells are forced to grow in the centre of the well and are available for microscope imaging. The PDL was then aspirated and wells were washed with MilliQ water and left at room temperature (RT) to dry overnight. On the day of the dissection, laminin (Life Technologies) was thawed in 4oC and coated onto the coverslips (1.2 mg/ml) and left at RT until required, at which time it was aspirated before cells were seeded. The laminin covered approximately the same area as the PDL in the centre of the well.

2.2.2.2 Cell preparation Mice were anaesthetised using isoflurane (Delvet, Seven Hills, NSW, Australia), then decapitated, and skin, muscle, and bone were removed to expose the spinal cord and harvest DRG (approximately 15-20 per mouse). DRG were placed in 1x Phosphate Buffered Solution (PBS) on ice and trimmed. Trimmed DRG were then placed in a 1 ml Eppendorf with PBS until they had settled to the bottom, then PBS was replaced with 10 µg/ml collagenase (from Clostridium histolyticum, Type IV, Life Technologies) in filtered Neurobasal-A Medium (NBA) (Life Technologies) and placed in 37oC water bath for 1 hr. The collagenase was then replaced with 0.5% trypsin (Life Technologies) and placed in 37oC water bath for 15 mins. The trypsin was replaced with NBA media supplemented with 1% penicillin/streptomycin (P/S) (Life Technologies) and 2% B27 (Life Technologies) and triturated ~20 times using a 1 ml pipette. Cells were filtered through a 70-micron sterile nylon cell strainer (Fisherbrand) and centrifuged (500 g, 3 mins, RT), then resuspended in Leibovitz's L-15 Medium (L-15) (Life Technologies) and purified in a Percoll (GE technologies) gradient. The Percoll solution was made by adding 9 ml of Percoll to 1 ml of 10x PBS, and from that a 30% and 60% Percoll solution was made with 1x PBS. The Percoll gradient was made by layering 3 ml of 30% Percoll above 3ml of 60% Percoll in a conical 15 ml tube. After centrifugation (800 g, 20 mins, 4oC), the top 2-3 ml containing myelin debris was removed, and the next 2-3 ml containing cells were placed in a new conical tube. Cells were centrifuged (800 g, 20 mins, 4oC) and washed once with L-15, then with supplemented NBA (200 g, 44

10 mins, 4oC), and subsequently seeded in 6 µl onto the central coated section of a 96- well plate and placed in 37oC to adhere for one hour. Wells were then flooded with 90 µl of the remaining media per well.

2.2.4 Drug administration PTX (CAS 33069-62-4, Tocris) was dissolved in DMSO with gentle warming and kept at -20oC in aliquots of 5 µl until required. PTX was diluted in culture media at concentrations of 10 nM, 100 nM and 250 nM for various experiments. TAK-242 (CAS 243984-11-4, Calbiochem, Millipore) was dissolved in DMSO and kept at -80oC until required. Amiloride (CAS 2016-88-8, A7410 Sigma) was dissolved in DMSO and kept at -20oC until required. Ibudilast (CAS 50847-11-5, I0157 Sigma), and Duloxetine (CAS 136434-34-9, SML0474 Sigma) were dissolved in DMSO and kept at 4oC until required. NMN (gift from Dr Lindsey Wu) was dissolved in NBA and kept at 4oC. Safinamide (CAS 202825-46-5, SML0025 Sigma) was dissolved in MilliQ water and kept at 4oC.

2.2.4.1 Treatment Strategy For the established model, cells were treated with PTX at 72 hours after plating and fixed after 24 hours of PTX treatment. For the short-term model, cells were treated with PTX at 24 hours after plating and fixed after 48 hours of PTX treatment. During drug testing, the short-term model was used with PTX at a concentration of 10 nM, which has been used in other studies with no significant effect on cell viability (Gobrecht et al., 2016, Ustinova et al., 2013) and cells were treated with the selected neuroprotective candidate drug 2 hours prior to PTX treatment at three concentrations (low, medium and high) based on findings from the literature.

2.2.5 Immunohistochemistry Cells were fixed using 4% paraformaldehyde (PFA) (BDH Laboratory Supplies) at 4oC for 10 minutes then washed with PBS. Cells were permeabilised using 0.1% Triton-X (Sigma) for 30 mins then blocked with 5% donkey serum with 0.1% Tween (Sigma) for 30 mins. The cells were then washed once with 0.1% Tween in PBS (PBS-T) then

45 incubated with primary antibody (Monoclonal Anti-β-Tubulin III (neuronal) antibody produced in mouse, 1:1,000, T8578 Sigma) diluted in PBS-T + 2% donkey serum for 1 hr at RT in the dark. After washing, cells were incubated in secondary antibody (Alexa Fluor546 conjugated donkey anti-rabbit, 1:1000, Life Technologies) diluted in PBS-T + 2% donkey serum for 30 mins at RT in the dark. Once washed, cells in 96-well plate were incubated in Hoechst 33342 solution (1:20,000, Life Technologies) for 15 mins in the dark, then washed once again and left in 100 µl PBS for imaging.

2.2.6 Imaging A Leica SP8 confocal was used with 10x objective selected for imaging βIII-tubulin staining of axons. During confocal microscopy sessions, 16 images were taken from each well of the 96 well plate in an automated system and analysed individually or stitched together depending on the type of analysis. Cell bodies and neurites were differentiated using an automated plugin in FIJI called NeuralMetrics (Pani et al., 2014).

2.2.7 Analysis The NeuralMetrics FIJI plugin (Pani et al., 2014) was used to automatically count number of neurons and measure the total area of axonal outgrowth based on βIII-tubulin staining, thereby eliminating investigator bias. Area of axons per neuron was taken into consideration to measure effects of PTX. NeuralMetrics was installed with the ImageScience update on FIJI. 16 images per well were examined separately. Images with no neurons were removed from analysis for an average number of 10 images per well. Program settings were adjusted between experiments to the image or staining quality, but generally remained the same (Figure 2.2). Once set, images were automatically analysed through the program (Figure 2.3). Cell bodies were counted by Figure 2.2. hand using the multi-point tool to double check the program’s NeuralMetrics program settings accuracy. Images with no axonal outgrowth were given the value “0” for area of axon per neuron. Value of neurite area per neuron

46 from each well were presented as an average of all images per well. Retraction bulbs were counted manually using the multi-point tool in ImageJ.

A B C

Figure 2.3. Sample analysis of neurite outgrowth. (A) Sample image of cells grown for 3 days and stained for βIII-tubulin. (B) The NeuralMetrics program is able to trace cell bodies based on input shape and size, and count the number of cells and area per image. (C) Neurites are then traced, and cell body area is subtracted from total neurite area to achieve neurite area per image. Parameters are kept the same per image to account for oversights that the program may make. 9-15 images were analysed per well depending on well density and the number of cells, then area of neurite per neuron was averaged per well. Scale bars removed for analysis.

2.2.8 Statistics GraphPad Prism 7.02 was used for analysis. For differences between treatment groups in dose response and drug effect experiments, an ordinary one-way ANOVA with Tukey’s multiple comparisons test or Dunnett’s multiple comparisons test (α=0.05) comparing to vehicle was used.

2.3 Results

2.3.1 PTX effects in short-term model of dissociated primary DRG neurons To determine the effects of PTX on primary DRG neuronal culture during the initial growth stages, we treated cells at 24 hrs after seeding with low, medium, and high concentrations of PTX or vehicle control (DMSO at the same concentration of the highest concentration of PTX). At 24 hrs later, PTX caused an obvious obstruction of axonal outgrowth (Figure 2.4) and visibly altered axonal ends, as evinced by large swelling and shorter axons compared to control. For the next experiments, PTX at the lowest tested concentration (10 nM) was used as significant difference was observed (P<0.001). While testing lower concentrations of PTX may have been useful, we found that the replication of the results from the literature was satisfactory for this study. Given more time a more thorough study elucidating the EC50 of PTX in this model could be an important avenue to explore.

A ) 2 .5

e B

a r

e 2 .0

n /

o 1 .5

i 1 .0

e l

t * * * i

a * * * * * * * *

m r

e 0 .5

n ( 0 .0

l o M M M tr n n n n 0 0 0 o 1 0 5 C 1 2

C D E

Figure 2.4. PTX induces a reduction of neurite outgrowth in acutely treated primary neuronal culture. (A) PTX treatment at low, medium, and high concentrations were given at 24 hrs for 48 hrs. Control group was treated with DMSO equivalent to the highest concentration of PTX. n=6 wells per treatment group; experiment was repeated twice and results were normalised to the average of each experiment before combining. Images were analysed for neurite area per neuron (~8 images per well) then averaged for one value per well. All treated groups had significantly lower neurite outgrowth than control group using one-way ANOVA, and Tukey’s multiple comparison’s test, ***P<0.001, ****P<0.0001. Data are expressed as mean ± SEM. Representative images of control group (B), 10 nM PTX treatment (C), 100 nM PTX treatment (D), and 250 nM PTX treatment (E). Scale bar: 100 µm.

2.3.2 PTX effects in established culture model of dissociated primary DRG neurons To determine the effect of PTX on neurite outgrowth in an established model of sensory neurons, we treated primary DRG neuronal culture from 5 to 6-week C57BL/6 mice with low, medium, and high concentrations of PTX or vehicle control (DMSO at the same concentration of the highest concentration of PTX) after a 3-day period of axon establishment. Following 24 hours of treatment with PTX, we did not observe any axonal degradation (Figure 2.5). However, there was an observable trend towards a reduction in neurite outgrowth in cells treated with 100 nM and 250 nM.

A previous study demonstrated the formation of retraction bulbs in PTX-treated neurons (Gornstein and Schwarz, 2017). To determine whether PTX causes any morphological changes in our established model of sensory neurons, we tested for the presence of retraction bulbs in the same experiment. Although PTX did not cause significant degradation in established neuronal cultures, a significant increase in the numbers of retraction bulbs was observed following treatment with all tested concentrations of PTX (Figure 2.6).

1 .5 B

)

e c

v 1 .0

s i

e 0 .5

n n ( 0 .0

l o M M M tr n n n n 0 0 0 o 1 0 5 C 1 2

P T X tre a tm e n t

C D E

Figure 2.5. PTX does not alter neurite outgrowth in established model of sensory neurons. (A) PTX treatment at low, medium, and high concentrations were given at 72 hrs for 24 hrs. n=6 wells per treatment group; experiment was repeated twice and results were normalised to the average of each experiment before combining. Images were analysed for neurite area per neuron (~8 images per well) then averaged for one value per well. No significant difference was found between groups using one-way ANOVA, Tukey’s multiple comparison’s test. Data are expressed as mean ± SEM. Representative images of Control (B), 10 nM PTX treatment (C), 100 nM PTX treatment (D), and 250 nM PTX treatment (E). Scale bar: 100 µm.

A 4 * * *

n * * B C o

r * * u

e 3

b l

u 2

o n

v a

0 l o M M M tr n n n n 0 0 0 o 1 0 5 C 1 2

P T X tre a tm e n t

Figure 2.6. PTX causes morphological changes (formation of retraction bulbs) in established model of sensory neurons. (A) The number of retraction bulbs were counted based on morphology. n=6 wells per group, average number of bulbs per neuron per well from two separate experiments. PTX treatment given at 72 hrs for 24 hrs. Control: DMSO equivalent to 250 nM condition. All wells treated with PTX had significantly higher numbers of retraction bulbs per neuron compared to control wells. **p<0.01, ***p<0.001 one-way ANOVA, Tukey’s multiple comparison’s test. Data are expressed as mean ± SEM. (B) Representative image of control well (red arrows indicate bulbs, images zoomed in from Figure 2.5 data). (C) Representative image of cells treated with 10 nM PTX.

2.3.3 Effects of drug candidates on neurite outgrowth in the short-term model of PTX treatment Potential neuroprotective drug candidates were selected if they were: a) clinically approved or in late clinical trials, b) had shown efficacy in alleviating other pain states, especially neuropathic pain, or c) had a proposed mechanism that could be neuroprotective with regards to PTX’s effect on DRG neurons. Since NGF is known to promote neurite outgrowth and is usually used in the culture media of sensory neurons, we tested its effects in our DRG neuronal cultures as a potential positive control (Figure 2.7). We found that although NGF allowed for a trend towards greater neurite outgrowth on its own (at 500 ng/ml), it had no significant effect on preventing PTX- induced neurotoxicity in this model at 50 ng/ml or 500 ng/ml. There were no significant differences in neurite outgrowth between PTX-treated cells without NGF and PTX- treated cells with NGF. We elected not to use NGF as a growth factor in our model as DRG neurons are not likely to have a high concentration of NGF available to them in vivo and cells seemed to grow well enough without its addition.

Of the tested drugs, amiloride proved to have modest protective effects at 100 µM, showing a significant increase in neurite outgrowth when compared to the PTX treated group (P<0.05). Amiloride had no significant effect at 10 µM, but significantly reduced neurite outgrowth at 500 µM (P<0.05) compared to PTX treated cells as well as vehicle group (P<0.0005) (Figure 2.8 F) indicating a toxic damaging effect at high concentration. The 100 µM concentration was considered a “high” concentration in previous studies which showed amiloride to have an effect on cultured rat DRG (Liu et al., 2004, Gschossmann et al., 2000). TAK-242 (0.1 µM, 1 µM, 10 µM), NMN (100 µM, 200 µM, 500 µM), safinamide (20 µM, 2 µM, 0.2 µM), duloxetine (5 nM, 10 nM, 100 nM) and ibudilast (1 µM, 10 µM, 100 µM) had no significant effects on neurite outgrowth at the tested concentrations (Figure 2.8).

A N G F B C

2 .5 n

o 2 .0

e n

/ 1 .5

r a

1 .0

r u

e 0 .5 n

0 .0

H H F F F E E G G G V V N / / N N H X H L E T / / /H V P H X X E T T V P P

T r e a tm e n t

Figure 2.7. The effect of NGF on short-term model. (A) NGF on its own induced an increase in neurite outgrowth that approached significance (P=0.0586) when compared to cells not treated with NGF. L: 50 ng/ml, H: 500 ng/ml, VEH: DMSO. n=3-4 per treatment group. Significance was not found between any groups when conducting a one-way ANOVA with Dunnett’s multiple comparisons test, however when the VEH/H NGF group was removed, the PTX group was significantly different to VEH (P=0.0089) and so were those treated with NGF and PTX (P<0.05). Data are expressed as mean ± SEM. (B) Representative image of VEH/VEH. (C) Representative image of VEH/H NGF. Scale bar: 100 µm. VEH=vehicle; L=low (concentration); H=high.

A T A K -2 4 2 B N M N C S a fin a m id e

2 .0 2 .0 2 .0

r r 1 .5 r 1 .5

1 .5 u

e e e * 1 .0 1 .0 r r 1 .0 r * * *

* * a a * * *

* * a

* * *

e e

e * *

r r r 0 .5 0 .5 u

u 0 .5

n n

0 .0 0 .0 0 .0

H H F F F F H H K K K K H H A E A N N N N E E A A A E A A A E E V V S S S S V V T T T T V M M M M / / / / / /V N H L H H L N N N H X / / M /H X / / M /H H X E T / E T / E T /H /L M H H X X X H X X X / / V P E T T V P E T T V P H X X T T T X V P P P V P P P E T T V P P P T r e a tm e n t T r e a tm e n t T r e a tm e n t E D D u lo x e tin e Ib u d ila s t F A m ilo rid e

2 .0 2 .0 2 .5

o o

o 2 .0

r r

1 .5 r 1 .5

e n

n #

/ /

/ 1 .5

a a

a * * * * * *

e e

1 .0 e * r 1 .0 r * r

a * * a

* a * *

* *

* 1 .0

e e

e * * * *

i r * * * * r

0 .5 r u 0 .5 u #

u # e

e 0 .5

e n n * * * * n * * * *

0 .0 0 .0 0 .0 I I I I H H L L L L H H H H U U U U M M M E E U U U U E E E M V E IB B B B V A A A A / /V D D D D V V I I I / /V / / L H X H L H H L H H X /H / M H / / M / H X / / /M / / / E T / E T X E T H X X V P H X X X H X X V P T X E T T V P E T T E T T T T V P P V P P P V P P P P T r e a tm e n t T r e a tm e n t T r e a tm e n t

G H I

J K L

Figure 2.8. Amiloride had significant protective effects among various potential neuroprotectants tested in the short-term model of PTX-induced neurotoxicity in dissociated primary DRG neuronal cultures. Cells were treated with candidate drugs 2 hours before PTX treatment in the short-term model. Drugs were tested at low (L), medium (M), and high (H) concentrations against 10 nM PTX. Vehicle for each drug 55 was set at H concentration for each condition. Each treatment group had n=3-4 wells per experiment. (A) TAK-242 had no significant effect on neurite outgrowth as all PTX- treated wells were not significantly different from each other. TAK-242 on its own had no effect on neurite outgrowth. VEH: DMSO, L: 0.1 µM, M: 1 µM, H: 10 µM. (B) NMN had no significant effect against PTX or on its own. VEH: NBA media, L: 100 µM, M: 200 µM, H: 500 µM. (C) Safinamide had no significant effect. VEH: media, L: 20 µM, M: 2 µM, H: 0.2 µM. (D) Duloxetine had no significant effect. VEH: DMSO, L: 5 nM, M: 10 nM, H: 100 nM. (E) Ibudilast had no significant effect. VEH: DMSO, L: 1 µM, M: 10 µM, H: 100 µM. (F) Amiloride at M concentration was significantly different to the PTX group (P<0.05) as well as the VEH group (P=0.0007). At H concentration in normal and PTX-treated conditions, AMI treated cells had significantly lower neurite outgrowth than the PTX group (P<0.05). Although AMI at L concentration showed a trend of increased neurite outgrowth, it did not reach significance (P=0.1124). L: 10 µM, M: 100 µM, H: 500 µM. Amiloride experiment was repeated twice more; n=9 per treatment group. One-way ANOVA, Dunnett’s multiple comparison’s test. Data are expressed as mean ± SEM. All values are normalised to the average of each experiment. *p<0.05, **p<0.005, ***p<0.001 (difference to VEH/VEH). #p<0.05 (difference to PTX /VEH). Representative images from amiloride experiment of VEH/VEH (G), PTX/VEH(H), VEH/H AMI (I), PTX/L AMI (J), PTX/M AMI (K), and PTX/H AMI (L). Scale bar: 100 µm. VEH=vehicle; AMI=amiloride.

2.4 Discussion In this study, we established a relatively streamlined method to assess the effects of PTX and potential neuroprotectant drug candidates in vitro using primary dissociated DRG neurons and developed a semi-automated method of analysing axonal outgrowth to reduce experimenter bias. We found that PTX caused the prevention of axonal outgrowth when cells were treated before establishment of axons and caused the formation of retraction bulbs when cells were treated after the establishment of axons, and may have led to evidence of axonal degeneration if cell were exposed to a higher concentration of PTX or were grown for a longer period. PTX-treated neurons also exhibited morphological changes such as swollen cell bodies and stunted axonal endings. Further analyses looking into the number of processes per neuron or axonal swelling could not be undertaken due to time constraints, however other studies have found changes in these areas as well, including an increased number of short processes and decreased number of developed neurites (Ustinova et al., 2013).

The short term DRG neuronal model of PTX neurotoxicity was chosen for drug testing because the effects of PTX were robust and easily measurable in an automated manner after a 3-day experiment using a low-concentration PTX treatment (10nM). Using this concentration of PTX was preferable as axons were not completely prevented from growing despite a significant reduction in axonal area, and cells were able to survive based on observations with live cell imaging microscopy (data not shown), and as outlined in previous studies that used similar concentrations of PTX (Gobrecht et al., 2016, Ustinova et al., 2013). A recent study conducted on DRG neurons treated with 300 nM PTX for five days showed no change in cell viability (Pittman et al., 2016). Future studies could encompass cell viability studies to reinforce these observations.

In our experiments, we did not use NGF or any other neurotrophic factors such as BDNF. This was to prevent any masking of potential beneficial effects of the candidate drugs we were testing in our assessment of neuroprotective drugs (Apfel et al., 1991), or to prevent any potential adverse interactions with the tested drugs. Interestingly, recent studies have found that NGF could be a biomarker for neuropathic pain in CIPN patients as levels of NGF correlated with the presence and severity of neuropathic pain

(Velasco et al., 2017). Moreover, in our settings NGF was not neuroprotective against PTX-induced reduction of neurite outgrowth. Further enquiry into the role NGF plays both in vitro and in vivo in CIPN is certainly of interest.

Although we did not analyse the specific cell types affected by PTX in our neuronal cultures, previous reports in in vivo models show that PTX affects large (Aβ)- or medium (Aδ)-sized myelinated fibres more than C-fibres (Xu et al., 2015, Nieto et al., 2014, Boehmerle et al., 2014), with regards to neuropathic pain development. However, other studies report axonal degeneration in both small and large fibres (Wozniak et al., 2011), spontaneous discharge from both A- and C-fibres (Xiao and Bennett, 2008) or changes in the mitochondria of only C-fibres (Jin et al., 2008) in response to PTX treatment. Future studies looking into characterisation of which cells types are represented in culture and which are more affected by PTX could be useful for finding better drug candidates for neuroprotection. The type of cells represented in culture could also be a limitation that would need to be further addressed. Other studies have also found changes in ATP levels in PTX-treated culture (Chen et al., 2015), another interesting avenue to pursue in future studies. In addition, PTX may have an effect on supporting cells in the DRG as well, such as SGCs and other resident cells. Therefore, a whole-DRG culture may provide some interesting points of analysis.

Of the drug candidates tested in this thesis, amiloride at a certain concentration was found to have a moderate protective effect on PTX-induced reduction of axonal outgrowth, although was found to be neurotoxic at a higher concentration. This is in line with previous work that found amiloride to have a moderate protective effect against PTX (Chen et al., 2015), although their study focused more on the effects of 5-HD rather than amiloride, and used a different model of PTX toxicity where dissociated DRG neurons were treated with 1 µmol/L of PTX for 24 hrs, presumably after the establishment of axons. Very few in vitro studies have been conducted on the effects of amiloride on neurons, so this may be a novel area to expand upon.

Amiloride is a degenerin/epithelial Na+ channel (DEG/ENaC) inhibitor, which form non-voltage gated, amiloride-sensitive cation channels (Ben-Shahar, 2011). A major 58 branch of the DEG/ENaC superfamily are ASICs, which are more often found in neuronal tissues, compared to ENaC-coding genes which are more often found in epithelial tissues (Ben-Shahar, 2011) and is the focus of amiloride’s current use as a potassium sparing diuretic (Teiwes and Toto, 2007). ASICs are located in both the central and peripheral nervous system, and many subtypes are found in the DRG (Dube et al., 2005) and nerve endings in the periphery (Price et al., 2001). Six ASIC subtypes have been described thus far (ASIC1a, ASIC1b, ASIC2a, ASIC2b, ASIC3, and ASIC4) and ASICs have been implicated in pain, particularly ASIC1a, ASIC1b and ASIC3, usually due to acidosis and inflammation (Wemmie et al., 2013, Mamet et al., 2002, Voilley et al., 2001). A study using ASIC3 knockout animals found that although ASICs may not directly transduce painful stimulus, they are involved in mechanical allodynia and inflammatory pain (Mogil et al., 2005) and others have found that the upregulation of ASIC1a in the spinal dorsal horn is required for Freund's adjuvant (CFA)-induced hypersensitivity (Duan et al., 2007). Interestingly, ASIC3 and ASIC1a has been found to be present in small DRG neurons such as nociceptors, and ASIC1a was found to be upregulated in larger Aβ fibres during inflammation (Voilley et al., 2001). Future studies may test the cell or fibre types affected by amiloride in vitro.

Amiloride is also known to inhibit sodium-potassium exchange (Vidt, 1981), the sodium-hydrogen exchanger (NHE) pathway (Liang et al., 2015), voltage-gated sodium channels (Kleyman and Cragoe, 1988) and Ca2+-permeable acid-sensing ion channels (Xiong et al., 2006). These multi-factorial effects may explain why the high concentration of amiloride was neurotoxic in the PTX-treated primary neuron cultures. A low concentration was found to have no protective effect, but a median concentration was moderately protective. Previous studies have shown the duality of amiloride’s effects in vitro as well – treatment with 20 µM amiloride on primary glial cells from C57BL/6 mice brains prevented endoplasmic reticulum (ER) stress-induced changes as well as cell death, whereas treatment with 200 µM amiloride increased cell death through ER stress (Hosoi et al., 2010). The concentration-dependent effects of amiloride highlight the importance of choosing the right dose for treatment.

The mechanisms behind the dose-dependent protective effects of amiloride against PTX-induced reduction in neurite outgrowth were not pursued in this study; instead, due to time constraints, the drug was taken to an animal model of CIPN for testing (Chapter 3). It is interesting to note that safinamide, which is also a sodium and calcium channel blocker with neuroprotective effects, did not have any effect on the treated cells. Although we treated cells at the recommended concentrations based on the literature, it is possible that a higher concentration of safinamide may have proved effective. Future studies using calcium imaging or ATP assays could help unravel the underlying mechanisms of amiloride effects. Additionally, future studies can investigate various types of channel blockers that have already been deemed safe to use clinically in neurodegenerative diseases for their effects against PTX-induced neurotoxicity.

Of course, differences between in vitro, in vivo and clinical studies were again represented here when duloxetine failed to show a neuroprotective effect on primary dissociated neuronal axons despite having pain ameliorating effects in patients. Although these results were expected, since duloxetine most likely has analgesic effects relating to central processing rather than axonal protection, it highlights one limitation of in vitro studies, in that it does not consider drugs that have other beneficial systemic effects not directly affecting axonal outgrowth from cultured neurons. Nevertheless, we believe this is an efficient way to screen neuroprotective drug candidates before moving to expensive and time consuming in vivo tests. We also show for the first time, to our knowledge, that duloxetine does not have a direct beneficial effect on neuronal axons following PTX treatment, at least in this model. TAK-242 and ibudilast, both of which are TLR4 channel blockers, also did not have a neuroprotective effect on treated cells. Perhaps the positive in vivo effects of these two drugs are based more so on altering macrophage responses or other neuroinflammatory processes rather than a direct effect on neurons, despite TLR4 having been shown to be upregulated in the DRG in correlation with pain (Li et al., 2014). NMN is known to have a protective effect on mitochondria; however, this was not evident in the dissociated neuronal cultures. This result suggests that NMN was unable to overcome PTX-induced mitotoxicity at this concentration. Nevertheless, cells treated with NMN alone had similar axonal outgrowth to controls, indicating that NMN at the selected concentrations has no toxic effects on neurons. Conversely, studies showing that the accumulation of NMN after 60 neuronal injury promotes axon degeneration (Di Stefano et al., 2015), although the exact mechanisms of degeneration are apparently still being investigated (Fukuda et al., 2017) and some studies have shown protective effects of NMN (Wang et al., 2015).

2.5 Conclusion This study optimised a method of measuring PTX toxicity in a short-term axonal growth model of dissociated DRG neurons that can be used to measure various effects and mechanisms of PTX and test potential neuroprotectants before moving into an animal model. We also measured the effects of PTX after the establishment of neurites and found morphological changes in the form of retraction bulbs formation but did not expand on this model due to time constraints. Lastly, our results identified amiloride as a potential neuroprotective candidate to be tested in in vivo model of PTX-induced peripheral neuropathy (Chapter 3).

Chapter 3: Developing an in vivo animal model of PTX-induced peripheral neuropathy and testing the effects of amiloride

Chapter 3: Developing an in vivo animal model of PTX-induced peripheral neuropathy and testing the effects of amiloride Contributions

Munawwar Abdulla designed experiments, performed animal injections, behavioural tests, dissections, cryosections, immunohistochemistry, microscopy, analysis and drafted the chapter. Justin G Lees was involved in the conception of the study, assisted with experiment design as well as confocal microscopy, dissections, MGS scoring, interpretation of data and revision of chapter. Samuel S Duffy provided training for conducting MGS tests. Mallory Barkl-Luke assisted with dissections. Isabel Morrow from the Mark Wainwright Analytical Centre - UNSW conducted the sciatic nerve TEM imaging. Gila Moalem-Taylor conceived and designed the study and assisted with revision of the chapter.

3.1 Introduction In cancer patients, PTX chemotherapy is often given in doses of 175 mg/m² intravenously over 3 hrs every 3 weeks, four times, but can range from infusions of 80- 250 mg/m² given over 1-96 hrs, weekly to every 3 weeks, depending on the type and severity of cancer (Eisenhauer et al., 1994, Seidman et al., 2008, Smith et al., 1999, PDR.net, 2018, Polomano et al., 2001, Brewer et al., 2016). Based on the common development of CIPN in PTX-treated patients, many in vivo models have been created using different rodent species, cumulative doses and methods of treatment, as outlined in Table 1.2, Chapter 1. Here, we aimed to establish in our group a new in vivo model of PTX treatment that would elicit chronic CIPN in C57BL/6 mice using i.p. injections of PTX. Table 1.2 summarises 12 models that used i.p. injections in mice, of which 9 studies used C57BL/6 mice and most demonstrated the development of mechanical allodynia; two studies demonstrated thermal hyperalgesia and two cold allodynia. Considering the variability in the literature, the behavioural studies conducted here were based on the most commonly used methods of testing peripheral neuropathy, and neuropathic pain in general.

As mentioned previously, several studies have found drugs that alleviated neuropathic pain in animal models, which have been used to understand the molecular effects of PTX in vivo. These drugs are listed in Table 3.1 below with their behavioural and molecular effects. Almost all of them were able to alleviate mechanical allodynia in the animal model, but not many have made it to clinical trials, and those that have are yet to be approved for clinical use (see chapter 4 for discussion). In this study, amiloride was tested for its effects in in vivo model of PTX-induced neuropathy.

Drug Effect Studies 5-HD Alleviated mechanical allodynia, reduced damage to non-myelinated (Chen et al., 2015) and smaller myelinated fibres, rescued dysregulation of intracellular calcium homeostasis. ALCAR Alleviated mechanical allodynia, prevented increase in incidence of (Jin et al., 2008, Zheng et swollen and vacuolated C-fibre mitochondria, decrease spontaneous al., 2011, Flatters et al., discharge incidence of A- and C-fibres, provided mitochondrial 2006, Xiao and Bennett, protection 2008) Antimycin A Alleviated mechanical allodynia (Griffiths and Flatters, 2015) BMSCs Alleviated mechanical allodynia (Huh et al., 2016) Calcium chelating agents Alleviated mechanical allodynia and thermal hyperalgesia (Siau and Bennett, 2006) Cannabinoid agonist (WIN 55,212- Alleviated mechanical allodynia and thermal hyperalgesia, prevented (Burgos et al., 2012) 2) microglia and astrocyte increase in spinal cord, prevented early production of IL1β, IL-6 and TNF-α. Cannabinoid receptor 2 agonist Alleviated mechanical allodynia (Deng et al., 2015) (AM1710) Ethosuximide (T-type calcium Alleviated mechanical allodynia and cold allodynia (Flatters and Bennett, channel blocker) 2004) Exercise Partially alleviate thermal hyperalgesia, protected from IENF density (Park et al., 2015) reduction in plantar surface of hind paw, normalised increased detyrosinated tubulin. Fluocinolone acetonide Alleviated thermal hyperalgesia, protected from IENF density reduction (Cetinkaya-Fisgin et al., in plantar surface of hind paw. 2016).

Gabapentin Alleviated mechanical allodynia and thermal hyperalgesia, improved (Matsumoto et al., 2006, altered gait parameters, inhibited A-fibre sensitisation, reduced α2δ-1 Gauchan et al., 2009, Xiao subunit in voltage-gated calcium channels in spinal dorsal horn et al., 2007, Huehnchen et al., 2013) Gabapentin & Synergistic anti-allodynic effect through PPAR-α receptors, possibly by (Donvito et al., 2016) Palmitoylethanolamide preventing the activation of the NFκB signalling pathway. (endogenous fatty acid amide) Glutathione Inhibited calcitonin gene-related peptide (CGRP) release. (Materazzi et al., 2012) Goshajinkigan Alleviated mechanical allodynia, protected from mitochondrial swelling, (Matsumura et al., 2014) nucleus degeneration and increase in TRPV4 gene expression Ibudilast Prevented and alleviated mechanical allodynia (Ledeboer et al., 2006) Ibudilast & Lithium Protected from thermal hyperalgesia and sensory-motor deficits, allowed (Mo et al., 2012) for higher doses of PTX, protected from cardiac abnormalities, prevented decrease in calcium through disrupting interaction of PTX, NCS-1 & InsP3R, protected from axonal demyelination IL-10 and IL-1αRA Alleviated mechanical allodynia, increased IL-10 and decreased IL-1, (Ledeboer et al., 2007b) TNF, and CD11b mRNA levels in lumbar DRG

Melatonin Prevented mechanical allodynia, reduced elevated 8-isoprostane F2 α (Galley et al., 2017) levels in peripheral nerves and the reduction in C-fibre activity- dependent slowing Minocycline Alleviated mechanical allodynia and hyperalgesia, prevented (Cata et al., 2008, Liu et hyposensitivity to chemical nociception, inhibited loss of IENF, al., 2010, Burgos et al., prevented macrophage infiltration, prevented astrocyte activation in 2012, Zhang et al., 2012, spinal dorsal horn Masocha, 2014)

Minocycline & Indomethacin Attenuated thermal hyperalgesia, possibly through cannabinoid receptor (Parvathy and Masocha, mechanisms 2015) Minoxidil Alleviated mechanical allodynia and protected from thermal (Chen et al., 2017) insensitivity through neuroinflammation suppression, rescued axon loss, attenuated decrease in g-ratio of small fibres and decreased the number of atypical mitochondria in sciatic nerve Milnacipran (serotonin reuptake Alleviated mechanical allodynia after repeated administration (Katsuyama et al., 2013) inhibitor) Metformin (anti-diabetic drug) Alleviated mechanical allodynia (Mao-Ying et al., 2014) MitoVitE Alleviated mechanical allodynia, prevented loss of mitochondrial (McCormick et al., 2016) function (metabolic activity) and glutathione in DRG cells Morphine Alleviated mechanical allodynia and thermal hyperalgesia or had a (Flatters and Bennett, partial (50%) alleviation of mechanical allodynia 2004, Pascual et al., 2010) rHuEPO Protects against axonal degeneration (Melli et al., 2006, Yang et al., 2009)

σ1R agonist BD-1063 Prevent mechanical allodynia and cold allodynia, protect from (Nieto et al., 2014) mitochondrial changes

TAK-242 Prevented mechanical hypersensitivity when given before, but not after (Li et al., 2017b) PTX

Tetrodoxin Low doses alleviated mechanical allodynia and cold allodynia, perhaps (Nieto et al., 2008) through sodium channel effects

Thalidomide Alleviated mechanical allodynia and hyperalgesia (Cata et al., 2008)

TLR4 antagonists (LPS-RS) Alleviated mechanical allodynia, blocked the increase of TLR4 and (Zhang et al., 2016, Li et MyD88 in DRG (but not spinal cord), reduced MCP-1 expression, al., 2014) blocked macrophage increase in DRG, prevented loss of IENFs

TRPA1 antagonist (HC-030031) & Alleviated mechanical allodynia and some cold allodynia (Materazzi et al., 2012) TRPV4 antagonist (HC-067047) Table 3.1. Drugs that have shown protective effects against PTX pathology in animal studies.

Amiloride is known by its brand name Midamor and is a diuretic that can be taken orally to treat high blood pressure or congestive heart failure due to its potassium sparing effects. However, it is also considered an ASIC blocker and has been shown to have anti-epileptic effects in rats, possibly through general inhibition of ASICs (N'Gouemo, 2008) or deactivation of ASIC1a and ASIC3 (Liang et al., 2015). Rats treated with high doses of 100 or 200 mg/kg were shown to suppress limbic seizures in a third of those treated, and pre-treatment at lower doses could delay onset (N'Gouemo, 2008). Additionally, amiloride was found to have neuroprotective effects in experimental autoimmune encephalomyelitis (EAE), an animal model of multiple sclerosis, through its effect as an ASIC1 blocker (Friese et al., 2007) as well as in spinal cord injury (SCI) (Durham-Lee et al., 2011), where daily amiloride treatment improved hind limb locomotor activity. In pain conditions such as chronic muscle-related hyperalgesia (Sluka et al., 2003) and localised cutaneous pain produced by acidic solution infusion (Ugawa et al., 2002), amiloride relieved pain through ASIC3 blocking, although not in severely acidic conditions. Amiloride has also shown anti-allodynic effects in neuropathic pain conditions such as peripheral nerve injury and vincristine- induced pain in mice (Muthuraman et al., 2008), where pain alleviation was accompanied by a reduction in calcium levels and oxidative stress, as well as in nerve- injured rats (Jeong et al., 2013), where elevated ASIC3 levels were reduced in the spinal cord. Considering these previous studies and the neuroprotective effects found in vitro (Chapter 2), we hypothesised that amiloride will have positive effects in PTX-induced peripheral neuropathy. Amiloride is also an ideal drug candidate as it is well-tolerated in humans, has a plasma half-life of 6-9 hrs and an oral LD50 of 56 mg/kg in mice (Drugbank.ca, 2005). It has been shown to have protective effects in animals treated with doses ranging from a subcutaneous injection of 0.1-10 µM in rats (Rocha-González et al., 2009) to single i.p. injections of up to 100 mg/kg in mice (Córdova et al., 2011), perhaps demonstrating that the method of administration is important both for function and for survivability. Thus, in this study, we aimed to (i) establish a chronic animal model of paclitaxel-induced peripheral neuropathy and assess phenotypic changes in the nervous system; and (ii) test the effects of treatment with amiloride in the animal model.

3.2 Materials and Methods

3.2.1 Animals Seven to eight-week-old male C57BL/6 (WT) mice (Australian BioResources, Moss Vale, NSW, Australia) were used. Mice were group-housed with water and food ad- libitum and maintained on a 12:12 hour light/dark cycle at constant room temperature and humidity levels. All animal experiments were approved by the Animal Care and Ethics Committee of the University of New South Wales, UNSW Sydney, Australia.

3.2.2 Drug administration PTX (Tocris) was dissolved in ethanol and kept at -80oC until ready for administration, where it was mixed in Cremaphor EL (Sigma) and saline. PTX was administered to mice on days 0, 2, 4, 7, 9 and 11 using i.p. injections of 5 mg/kg in approximately 50 µl vehicle for a total cumulative dose of 30 mg/kg. A saline vehicle control was used for comparison. Amiloride (Sigma) was dissolved in DMSO (Sigma) at a concentration of 100 mM and administered 2 hrs prior to each PTX administration by i.p. injections of 5 mg/kg in approximately 60 µl saline vehicle for a total cumulative dose of 30 mg/kg. An equivalent dose of DMSO in saline was used as control.

3.2.3 Behaviour testing All behavioural assessments were carried out blinded to treatment group unless otherwise stated. For the von Frey and Hargreaves tests, mice were placed randomly on the apparatus, so the observer was not aware which mouse was in which group. For the hot plate test, the observer was blinded to the treatment groups during video analysis. For MGS, photo analysis was carried out by an experimenter unaware of treatment groups. Mice were habituated to the room and various apparatus before any testing was conducted unless otherwise stated. Mice were habituated to the von Frey and Hargreaves apparatus for an hour before testing, and to the hot plate (at RT) and MGS setup for 5-10 mins individually. Behavioural tests were carried out approximately a week after the final injection for 2 weeks (Table 3.2); testing for von Frey was conducted at D18, 21, 23 and 25 for the first experiment and D23 and 25 were chosen

70 for testing in the amiloride experiment. The thermal tests were carried out over D19 and 23. Locomotion tests were carried out at D15 and D24, and MGS measurements were taken at D27, shortly before mice were culled at 4 weeks.

Experimental Timeline D0 – 1st D1 D2 – 2nd D3 D4 – 3rd D5 D6 injection injection injection D7 – 4th D8 D9 – 5th D10 D11 – 6th D12 D13 injection injection injection D14 D15 – D16 D17 D18 – von D19 – D20 Open Field Frey thermal test D21 – von D22 D23 – von D24 – D25 – von D26 D27 – MGS Frey Frey (only Open Field Frey (only AMI first study), study) thermal test Table 3.2. Timeline of the various behavioural assessments.

Locomotion test An open field Photobeam Activity System (PAS) (SD Instruments, San Diego, USA) was used to monitor the locomotion of mice on Perspex covered flooring inside a 40 cm x 40 cm enclosure. Total number of nose pokes and rearing were tested using laser detection beams and analysed using PAS software for 5 minutes per animal. Arena was cleaned with 70% ethanol and allowed to evaporate between each test. Mice were placed in the middle of the box and measurements were taken as soon as the laser was first broken. Mice were not habituated to the apparatus beforehand.

Mechanical Allodynia Von Frey up-down method: von Frey filaments of increasing rigidity were aimed at the centre of the plantar surface of the hind paw and gently pushed upwards to assess mechanical allodynia, measured as the maximal force (in grams) required for paw withdrawal in response to the mechanical stimulus. Six measurements were taken per hind paw and the paw withdrawal threshold was calculated. This technique was adopted from a widely used method (Chaplan et al., 1994).

Von Frey filament response method: A selected filament was aimed at the centre of the plantar surface of the hind paw and gently pushed upwards for 3-5 seconds and 71 withdrawal response was recorded. A response was counted as 1 and no response was counted as 0. After 5 trials, the numbers of responses were totalled for a score of 0-5.

Von Frey alternative scoring method: Alternatively, a selected filament (0.4g) was aimed at the centre of the plantar surface of the hind paw and gently pushed upwards to assess mechanical allodynia. A score of 0 was given for no response, 1 for a minimal response, and 2 for a maximal response. After 6 measurements were taken from each hind paw, the score was presented as an average.

Thermal Hyperalgesia Hargreaves method: Thermal hyperalgesia was assessed by exposing the mid-plantar region of the hind paw to a beam of radiant heat through a transparent Perspex surface using a plantar analgesia meter for paw stimulation (Ugo Basile). The withdrawal latency to the heat stimulus was recorded from both left and right hind paws, as the time taken from the onset of the thermal stimulus to withdrawal of the hind paw from the heat source. Separate measurements were taken for each hind paw and averaged for two values per animal. Thermal hyperalgesia was measured as paw withdrawal latency (in seconds) in response to the thermal stimulus. A cut-off time of 20 seconds was applied to prevent tissue damage.

Hot Plate: Hot plate (Ugo Basile) was set at 55oC and each mouse was placed in a chamber and timed until showing signs of paw licking or jumping or until a cut-off time of 10 seconds. This was repeated three times per mouse in intervals of at least 30 mins. Thermal hyperalgesia was measured as average paw withdrawal latency (in seconds) in response to the thermal stimulus. Video recording was taken for careful analysis.

Mouse Grimace Scale (MGS) The method for MGS was based on (Langford et al., 2010). Briefly, mice were placed in a confined chamber with glass on either side so that two cameras could record at once. Mice were habituated to the apparatus for 5-10 mins each before testing day. On testing day, mice were recorded for approximately 10 mins each. A still image at every minute mark +/- 10 s was taken using VLC media player. If neither camera caught a clear image at this mark, then no image was taken. The images were analysed by a blinded experimenter by giving a score of 0, 1 or 2 for severity of orbital tightening, nose bulge, cheek bulge, and ear position (Figure 3.1). The scores were then calculated by averaging the scores of these four parameters. 72

Figure has been removed due to copyright restrictions.

Figure 3.1. Sample scoring for MGS (Langford et al., 2010)

3.2.4 Tissue Collection Mice were culled 4 weeks after the first injection by anaesthetising with isoflurane in oxygen (Delvet, Seven Hills, NSW, Australia), and decapitating. Sciatic nerve, lumbar DRG, and spinal cords were carefully dissected and then placed in formalin (Formalin solution, neutral buffered, 10%; Sigma-Aldrich) at 4oC overnight. Tissue was then transferred to 30% sucrose solution and stored at 4oC until required for IHC analysis. Alternative 2 mm sections of sciatic nerve were placed in PBS and transferred to transmission electron microscopy (TEM) fixative as described in section 3.2.7.

3.2.5 Immunohistochemistry Fixed tissue was embedded in OCT (Scigen) at -20oC and cryosectioned at 10-20 µm. Once dried overnight, they were stained or stored in -20oC. Cryosectioned DRG, spinal

73 cord and sciatic nerve tissues were thawed, rinsed for 10 mins in ethanol, washed in water and then PBS. Tissues were then permeabilised in 0.1% Triton X-100 for 30 mins, rinsed with PBS-T, and then blocked with 0.1% Tween and 5% BSA (Sigma) or donkey or goat serum for 30 minutes depending on the experiment. They were then incubated in primary antibody (Anti-Iba-1 produced in rabbit, 1:1000, 019-19741 Wako Chemicals; Anti-GFAP produced in chicken, 1:500, ab4674 Abcam; Monoclonal Anti- Acetylated Tubulin produced in mouse, 1:1000, T7451 Sigma; Anti-CGRP produced in goat, 1:500, ab36001 Abcam) for 1 hr, washed, then incubated in secondary antibody (Alexa Fluor488 conjugated donkey anti-goat, 1:1000, A-11055 Life Technologies; Alexa Fluor546 conjugated donkey anti-rabbit, 1:1000, A-10040 Life Technologies; Alexa Fluor647 conjugated donkey anti-rabbit, 1:1000, A-31573 Life Technologies; Alexa Fluor488 conjugated donkey anti-mouse, 1:1000, A-21202 Life Technologies; Alexa Fluor488 AffiniPure donkey anti-chicken, 1:1000, 703-545-155 Jackson ImmunoResearch) for 30 mins in the dark, and washed. Coverslips were placed on top with Prolong Gold (Life Technologies). Slides were dried at RT then edges were sealed with nail polish. Alternatively, tissues were permeabilised and blocked with 5% donkey serum, 0.2% Tween-20 and 0.3% Triton X-100 for 1 hr in a closed, moist box, then incubated overnight in 4oC in primary antibody with 5% donkey serum and 0.3% Triton X-100. The next day, after washing, slides were incubated with secondary antibody in 5% donkey serum and 0.3% Triton X-100 for 1 hr in a closed moist box before being treated with ProLong Gold and cover-slipped.

3.2.6 Image Analysis Cryosectioned DRG, spinal cord and sciatic nerves were imaged using an Olympus BX51 + DP73-535 fluorescent microscope or Leica SP8 confocal microscope. One image was taken per DRG at 40x objective and images were quantified by counting the number of macrophages per area. Two images per spinal cord section were taken at 20x objective (left and right dorsal horn) and analysed by tracing the dorsal horn and quantifying the percent area (%area) stained by setting a threshold for staining intensity in ImageJ. Approximately three images per sciatic nerve were taken at 20x objective and also analysed by quantifying %area stained. For images taken on the confocal microscope (20x objective), z-stacks were taken with 2µm intervals through the entire

74 section and layers were stacked before analysis using ImageJ (Images to Stack>Z Project at Max Intensity).

3.2.7 Transmission Electron Microscopy (TEM) Sciatic nerves were dissected from mice and placed into PBS wash buffer for ~1-2 minutes. A small section (1-2 mm) from the mid-thigh portion of the sciatic nerve was placed in fixative of 2.5% glutaraldehyde in 0.1M Sodium Cacodylate Buffer for ~3 hours at RT. Sections were then transferred to 0.1M Sodium Cacodylate Buffer and stored at 4ºC prior to post-fixing and sectioning. Post-fixing, sectioning and imaging were performed by staff at the UNSW - Mark Wainwright Analytical Centre - Electron Microscope Unit using a JEOL 1400 TEM. For myelination g-ratio quantification; 4 images from 3 separate sections from each of 3 sciatic nerves were imaged at 2000x magnification for each treatment group. For mitochondria analysis, 9 images from 3 separate sections from each of 3 sciatic nerves were imaged at 30,000x magnification for each group. Sciatic nerve myelinated axon images were analysed using the gRatio plugin on ImageJ. Mitochondria were analysed by counting the number of vacuolated mitochondria and creating a ratio between normal and abnormal mitochondria.

3.2.8 Statistics Statistical analyses were carried out using GraphPad Prism 7.02 software. Behavioural data were analysed using 2-way ANOVA, followed by Sidak’s multiple comparisons test. MGS and most molecular data were analysed using 1-way ANOVA, followed by Tukey’s multiple comparisons test. TEM results were analysed using a 2-way Student’s t-test. α=0.05 was used for all tests.

3.3 Results

3.3.1 Effects of PTX treatment on C57BL/6 mice Mice treated with a cumulative dose of 30 mg/kg of PTX (6 injections of 5mg/kg over 2 weeks) did not have significant weight loss and continued to gain weight at the same rate as control mice (Figure 3.2 A). Rearing and Nose Poke behaviours were evaluated in the open field apparatus to test for sickness signs in the form of loss of exploratory behaviour. Again, no differences were found between groups (Figure 3.2 B-C).

To test the effects of PTX treatment on pain hypersensitivity, we measured mechanical allodynia. PTX caused mechanical allodynia, as demonstrated by both the up-down method and the von Frey filament response method (Figure 3.3). Figure 3.3 A shows that mice exhibited a significant decrease in paw withdrawal threshold relative to baseline on days 18, 23 and 25 following PTX treatment (P<0.05) using the up-down method. In the filament response method, the most effective filament was 0.4 g showing a significant increase (P<0.05) in paw responses on day 21, and a slightly modified version of this method (alternative scoring method) was used for subsequent studies due to its time efficiency, considering it appeared to give similar results; the scoring method is currently the preferred method to test mechanical allodynia for CIPN studies in our lab. Mechanical allodynia appeared to rise to a peak in the 2 weeks following the final injection, and mice were culled at 4 weeks from first injection while they were still exhibiting signs of allodynia. Due to constraints on time and resources, we were unable to conduct a more thorough observation of mechanical allodynia across a longer period to see the full extent of allodynia; further study of this parameter in our PTX model is desirable.

TEM images were taken of the sciatic nerves of treated and non-treated mice at endpoint to analyse changes in myelination using g-ratio. Since previous studies had measured both large and small myelinated fibres and had only found differences in small myelinated fibres that were less than 5 µm in axon diameter (Chen et al., 2015), we chose to measure only the g-ratio of small myelinated fibres in each image. However, no differences were found between groups (P=0.5936) (Figure 3.4). We also 76 decided to count the ratio of “normal” to “abnormal” mitochondria based on morphology in 30,000x magnification and found no differences between groups (P=0.9729) (Figure 3.5).

A % W e ig h t C h a n g e a fte r In je c tio n

1 1 0

V E H e

g P T X

n a

h 1 0 5 In je c tio n s

g i

e 1 0 0

9 5

0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 1 3 5 D D D D D D D D D D 1 1 1 1 1 1 1 1 1 2 2 2 D D D D D D D D D D D D

T im e p o in t

B C

O p e n F ie ld - R e a rin g O p e n F ie ld - N o s e P o k e

3 0 V E H e 5 0

k V E H

n i

p P T X r

P T X a

e 4 0

r o

2 0

e 3 0

o 2 0

r 1 0

r b

e 1 0

m N 0 u

N 0

L 5 4 L 5 4 B 1 2 1 2 D D B D D

Figure 3.2. Effects of PTX on body weight and exploratory behaviours. (A) No changes in weight gain were found between PTX-treated and control mice. No differences between groups were found in rearing behaviour (B) or in nose poke behaviour (C), and in general both groups showed less interest in exploratory behaviour after the first exposure. Two-way ANOVA, followed by Sidak’s multiple comparisons test. Data are presented as mean ± SEM; n=6 per group.

A B M e c h a n ic a l A llo d y n ia M e c h a n ic a l A llo d y n ia (D 2 1 )

P a c lita xe l 5

l f

o 2 S a lin e V E H o

w *

a 4 s

s P T X

e s

1 p

a s

m ** 2

t i

** v 1

w * A

0 0

L 8 3 5 1 2 2 7 .4 .0 B .0 0 1 D D D 0 F ila m e n t (g )

Figure 3.3. Effects of PTX on mechanical pain hypersensitivity. (A) Up-down method. Baseline was measured twice prior to injections, and the results averaged per hind paw for a total of 12 measurements per group. Results from tested timepoints were normalised to each paw’s baseline score. Mechanical allodynia was apparent in PTX- treated mice from D18 (P<0.005), although the variation on that day was high in the sham group, perhaps due to the proximity of the last injection. There was a significant difference in paw withdrawal threshold on D23 (P<0.005) and D25 (P<0.05) in PTX- treated mice as compared to vehicle-treated mice. Two-way ANOVA, followed by Sidak’s multiple comparisons test. (B) The von Frey response method on D21 showed mechanical allodynia was most apparent when using the 0.4 g filament (P<0.05); Data presented as mean ± SEM; n=6 per group. VEH=vehicle.

A G -ra tio B C 0 .8

0 .6

0 .4

0 .2

0 .0

H X E T V P

Figure 3.4. No difference in g-ratio between small myelinated fibres in vehicle and PTX-treated groups. (A) 10 fibres 5 µm or smaller were chosen from each TEM image at random and g-ratio (the ratio of the inner to the outer diameter of the myelin sheath, indicating axonal myelination) was automatically measured using the g-ratio plugin on ImageJ. n=3-6 images per section, 3 sections per animal, 3 animals per group (~360 fibres per group). No significant difference was found between groups using 2-tailed Student’s t-test. Data presented as mean ± SEM. Representative images of VEH group (B), and PTX group (C). Scale bar: 10 µm.

A M ito c h o n d ria M o r p h o lo g y B C

1 .5

)

o n

b 1 .0

a :

m r

o 0 .5 n

( D

a R 0 .0

H X E T V P

Figure 3.5. No difference between ratio of normal to abnormal mitochondria in myelinated fibres between vehicle and PTX-treated mice. (A) Mitochondria from TEM images were analysed by counting the ratio of normal to abnormal mitochondria. Abnormal mitochondria were defined as those that had vacuolation or swelling. No significant differences were found between groups using two-tailed Student’s t-test. Data presented as mean ± SEM; n=9-11 images of individual fibres per section, 3 sections per animal, 3 animals per group. Mean number of mitochondria per animal was ~150. (B) A representative image of 30,000x magnification of nerve fibre. (C) Example of “normal” mitochondria. (D) Example of “abnormal” mitochondria. Scale bar: 0.5 µm.

3.3.2 Behavioural effects of amiloride Animals treated with amiloride prior to each chemotherapy injection did not show obvious signs of weakness, or any observable sickness during the time tested. Rate of weight gain was not significantly different to control group (Figure 3.6). To test potential effects of PTX and amiloride on thermal pain, we assessed heat hyperalgesia using the hot plate method in mice treated with PTX, PTX and amiloride, and vehicle controls. No differences were found between groups at any time point at 55oC (Figure 3.7), and earlier tests using 51oC or the Hargreaves method also showed no difference between PTX-treated mice and vehicle-treated mice (data not shown). To assess the potential effects of PTX and amiloride on spontaneous pain, we used the MGS in mice treated with PTX, PTX and amiloride, and vehicle controls. No differences were found between any groups at the tested time point (D27) (Figure 3.8).

To test whether treatment with amiloride affects pain hypersensitivity in PTX-treated mice, mechanical allodynia was tested at two-time points. D18 time point was removed due to high variability, especially in the sham group, most likely due to proximity to the last injection, a phenomenon which we have previously observed in other cohorts. All groups had increased sensitivity; again, we believe this is due to repeated injections and handling. As expected, the PTX group had significantly increased mechanical allodynia (P<0.005) compared to the vehicle control group. Importantly, however, amiloride treated mice also had significantly reduced mechanical allodynia (P<0.05) compared to the PTX group (Figure 3.9).

% W e ig h t C h a n g e

1 2 0 V E H /V E H

1 1 5 P T X /V E H

P T X /A M I 1 1 0 In je c tio n s 1 0 5

1 0 0

9 5

0 1 2 3 4 6 7 8 9 0 1 2 3 4 9 1 3 5 7 8 D D D D D D D D D 1 1 1 1 1 1 2 2 2 2 2 D D D D D D D D D D D

Figure 3.6. No effects of amiloride on body weight. Amiloride (AMI) and its vehicle DMSO (VEH) had no effects on weight gain. Mice were injected with AMI or VEH two hours prior to PTX, so all received 2 injections for a total of 12 injections over 2 weeks. Two-way ANOVA, followed by Sidak’s multiple comparisons test. Data presented as mean ± SEM; n=8 per group.

H o t P la te (5 5 o C )

1 0

V E H /V E H

) s

( P T X /V E H

e P T X /A M I

e 5

e R

L 9 3 B 1 2 D D

Figure 3.7. No effects of PTX and amiloride treatment on thermal hyperalgesia. No differences between groups were found at 55oC for any time point. Cut-off time for testing was 10 s to prevent tissue damage. Two-way ANOVA, followed by Sidak’s multiple comparisons test. Data presented as mean ± SEM; n=8 per group.

M o u s e G rim a c e S c a le

0 .4

e 0 .3

e 0 .2

e v

A 0 .1

0 .0

I H H E E M V A / /V / H X X E T T V P P

Figure 3.8. No effects of PTX and amiloride treatment on facial grimacing. Mice were assessed for orbital tightening, nose bulge, cheek bulge, and ear position and given a score of 0, 1 or 2. No differences between groups were found at D27. Mean score for all groups were approximately 0.3. Ordinary one-way ANOVA, followed by Tukey's multiple comparisons test. Data presented as mean ± SEM; n=8 per group.

M e c h a n ic a l A llo d y n ia (0 .4 g )

V E H /V E H

f o

* * P T X /V E H r * *

e 1 .0 s b P T X /A M I

e # m

s #

a r

r 0 .5

v A

L 1 5 B 2 2 D D

Figure 3.9. Amiloride significantly reduced mechanical allodynia in PTX-treated mice. Mice were assessed for mechanical allodynia using the von Frey alternative scoring method with the 0.4 g filament. The PTX/VEH group showed significant difference to VEHVEH group at D21 and D25 (**P<0.005) although no difference was observed between VEH/VEH and PTX/AMI. The PTX/AMI group showed significant difference to the PTX/VEH group on D21 and D25 (#P<0.05). Two-way ANOVA, followed by Sidak's multiple comparisons test. Data presented as mean ± SEM; n=8 per group.

3.3.3 Molecular effects of amiloride on nervous system tissues Acetylated tubulin staining was used as a marker of nerve damage and was assessed in the sciatic nerve in mice treated with PTX, PTX and amiloride, and vehicle controls. Although PTX-treated mice showed increased staining as compared to vehicle control mice (P<0.0002), amiloride had no significant effect on this alteration in tubulin (Figure 3.10). We also stained for CGRP in the sciatic nerve and spinal cord to test for changes in small fibre sensory nerves. No differences were found in either tissue; however CGRP immunoreactivity in the spinal cord approached significance, with PTX-treated group showing mildly elevated levels (Figure 3.11 B).

To assess microgliosis, we carried out IHC for Iba-1 (a marker of microglia/ macrophages) in the spinal cord of mice treated with PTX, PTX and amiloride, and controls (Figure 3.12). Iba-1 immunoreactivity was expressed as percentage area stained in the spinal dorsal horn. While there was a significant increase (P=0.0292) in Iba-1 expression after PTX treatment, the amiloride group was not significantly different from the PTX group (P=0.4299). Interestingly, amiloride was also not significantly different to the control group (P=0.3037).

To assess astrogliosis, we carried out IHC for GFAP (a marker of astrocytes) in the spinal cord of mice treated with PTX, PTX and amiloride, and controls (Figure 3.13). GFAP expression was measured as a percentage area in the dorsal horn of the spinal cord. PTX-treated mice had significantly (P=0.0048) increased levels of GFAP expression as compared to controls, and PTX and amiloride-treated mice had significantly (P=0.0262) lower levels of GFAP expression, as compared to PTX-treated mice.

A A c e ty la te d T u b u lin B C in S c ia tic N e rv e

2 .0 a

e * * * * r 1 .5

a * * *

e 1 .0

a m

r 0 .5

o N

0 .0

I H H E E M V A / /V / H X X E T T V P P

Figure 3.10. Amiloride had no effect on the increase of acetylated tubulin staining in PTX -treated mice. Sciatic nerves from endpoint were stained for acetylated tubulin and analysed using ImageJ for %area stained. (A) Results shown were taken from two separate experiments with 3-6 images taken at random parts of each sciatic nerve and %area stain averaged for 1 value per mouse. n=14 for PTX/VEH and VEH/VEH, n=8 for PTX/AMI. PTX/VEH (P<0.0001) and PTX/AMI (P<0.0002) groups had significantly increased staining compared to VEH/VEH group; however PTX/AMI group was not significantly different to PTX/VEH. One-way ANOVA, followed by Tukey’s multiple comparisons test. Data presented as mean ± SEM. Representative images of VEH/VEH (B) and PTX/VEH (C) sciatic nerves. Scale bar: 50 µm.

A C G R P in S c ia tic N e rv e B C G R P in S p in a l C o rd

0 .4 2 .0

0 .3 1 .5

a 1 .0

a 0 .2

% % 0 .1 0 .5

0 .0 0 .0 I I H H H E M H M E E E V V /A V A / / / /V / H X X H X X E T T E T T V P P V P P

C D E

F G H

Figure 3.11. CGRP immunoreactivity in the sciatic nerve and spinal cord. (A) Sciatic nerves were stained for CGRP and analysed for %area stained using ImageJ. No significant differences were found between groups. 2-4 images per sciatic nerve, 1 sciatic nerve per mouse (n=8 mice per group). (B) Spinal dorsal horn (as in Figure 3.13A) was analysed for %area stained, two images (left and right dorsal horn) per mouse. No significant differences were found between CGRP staining using one-way ANOVA, followed by Tukey’s multiple comparisons test. However, an unpaired Student’s t-test between VEH/VEH and PTX/VEH groups approached significance, with p=0.0535. Data presented as mean ± SEM. Representative images of VEH/VEH (C), PTX/VEH (D) and PTX/AMI (E) sciatic nerves. Representative images of VEH/VEH (F), PTX/VEH (G) and PTX/AMI (F) spinal dorsal horns. Scale bar: 50 µm.

A Ib a -1 in S p in a l C o rd

1 .0 *

0 .8

i a

t 0 .6

a e

r 0 .4

a % 0 .2

0 .0 I H H E E M V A / /V / H X X E T T P V P

B C D

Figure 3.12. Iba-1 immunoreactivity in the dorsal horn of the spinal cord. (A) Spinal dorsal horn areas (as in Figure 3.13A) were analysed for %area stained using ImageJ and a set threshold. Two images (left and right dorsal horn) per mouse (n=8 per group). Iba-1 expression was significantly increased in the spinal dorsal horn after PTX treatment (P<0.05) as compared to vehicle-controls. Amiloride treatment group was not significantly different from either PTX or VEH treatment groups. One-way ANOVA, Tukey's multiple comparisons test. Data are expressed as mean ± SEM. Representative images of VEH/VEH (B), PTX/VEH (C) and PTX/AMI (D) treated mice. Scale bar: 50 µm.

G F A P in S p in a l C o rd A B 0 .2 5 * *

0 .2 0

i a

t 0 .1 5 s

a e

r 0 .1 0

a % 0 .0 5

0 .0 0

I H H E E M V A / /V / H X X E T T V P P

C D E

Figure 3.13. GFAP immunoreactivity in the dorsal horn of the spinal cord. (A) Spinal dorsal horn areas were analysed for %area stained using ImageJ and a set threshold. 2 images (left and right dorsal horn) per mouse (n=8 per group). (B) PTX-treated mice had significantly higher GFAP staining than VEH/VEH group (P<0.005). AMI-treated mice had decreased levels of GFAP compared to PTX/VEH group (P<0.05) and this was not significantly different to VEH/VEH group (P=0.7425). One-way ANOVA, followed by Tukey's multiple comparisons test. Data are expressed as mean ± SEM. Representative images of VEH/VEH (C), PTX/VEH (D) and PTX/AMI (E). Scale bar: (A) 100 µm; (C-E) 50 µm.

3.4 Discussion In this study, we established a chronic CIPN model treating male C57BL/6 mice with 6 injections of PTX and a cumulative dose of 30mg/kg, which mimics human chemotherapy cycles and dose. Although other preclinical PTX CIPN models have previously been described, we decided to establish a model based on the clinical treatment of breast cancer patients who generally receive 6 cycles of 175 mg/m2 during treatment. A single injection of 5 mg/kg is equivalent to ~185 mg/m2 (in humans the average is 175 mg/m2) and a 30 mg/kg cumulative dose is equivalent to ~1100 kg/m2 (Nair and Jacob, 2016) in humans which is the normal range where patients develop CIPN (Argyriou et al., 2007). Our PTX-treated mice exhibited mechanical allodynia but not thermal hyperalgesia, increased acetylated tubulin staining, representing damage in the sciatic nerve, increased microgliosis (Iba-1 staining) and increased astrogliosis (GFAP staining) in the spinal cord. There are several limitations for behavioural studies in mice which will be addressed in the following sections.

Although there have been studies that found thermal hyperalgesia to be present in PTX- treated C57BL/6 mice (Costa et al., 2011, Mo et al., 2012), there are also those who have not (Smith et al., 2004) in agreement with our findings in this study. This discrepancy could have been due to a variation in technique, dosage or strain of mice. Our study used two common methods to test thermal hyperalgesia; the Hargreaves paw withdrawal method and the hot plate method. We found it was difficult to gain accurate results from the Hargreaves apparatus due to the positioning of the mice’s feet or the collection of urine in the chambers. Hargreaves is also used to test hind paws individually to test ipsilateral and contralateral effects of certain neuropathy models, and since we did not aim to test these measures, the hot plate method was employed. Nevertheless, no significant differences were found in our cohort. Perhaps testing with a different temperature or having a longer cut-off time would yield different results. Testing a different strain of mice using this treatment regimen may also be a future avenue for exploration. Indeed, different behavioural responses have been demonstrated in different strains of mice, for example, although Smith et al. (2004) did not find thermal hyperalgesia in PTX-treated C57BL/6 mice, they did find it in nine other strains of mice tested in the same study.

In this study, we found that the PTX-treated mice develop significant mechanical pain hypersensitivity. There are multiple methods to test for mechanical allodynia and interpretation of responses can be subjective and dependent on the examiner’s experience (Vrinten and Hamers, 2003). Responses may also be affected by the amount of time spent habituating the mice, room temperature, humidity, time of day, time from treatment and other such factors, which need to be carefully considered (Vrinten and Hamers, 2003). Although these conditions were kept as constant as possible, there may have been some unknown factors that introduced variability in this study. We confirmed the development of mechanical allodynia in PTX-treated mice using several methods, based on the use of von Frey filaments (Deuis et al., 2017). In some cases, we found large variation in the sham group, especially if behaviour testing was carried out too close to the final injection point; therefore, we decided to wait a week after the final injection before re-habituating them to the apparatus and conducting behavioural tests. Given more time, a careful temporal analysis of the exact timeline for mechanical allodynia in this model would be useful to understand behavioural variability and inform future test. As the variations could be due to the number of injections, perhaps a lower number of injections with a higher concentration of PTX may prove to be a better model if testing is required during the treatment period. We used a 0.4 g von Frey filament and tested repeated response rate because our study showed that this filament produced effective differences between behavioural groups. Von Frey testing can be influenced by investigator bias, and although we were blinded while carrying out these experiments, perhaps future tests can use an automated von Frey testing (Campana and Rimondini, 2015), though it is fair to say some automated methods may come with their own set of problems (Nirogi et al., 2012). Nevertheless, a less subjective way to test for pain would be a great avenue of study, although von Frey is still considered the gold standard for murine behaviour testing (Deuis et al., 2017). In addition, we did attempt to use the MGS to test for signs of spontaneous pain in CIPN mice but found no differences between groups. MGS has been successfully used by our group to measure spontaneous pain in EAE (Duffy et al., 2016) and by others in SCI mouse models (Wu et al., 2016). Our results suggest that CIPN does not cause facial grimacing at the tested time point in our model of PTX-induced peripheral neuropathy in C57BL/6 mice.

With regards to molecular changes, there was an increase in acetylated tubulin staining in the sciatic nerve, in line with previous studies (Benbow et al., 2016). In the spinal cord we found an increase in both Iba-1 and GFAP staining, representing an increase in both microglia and astrocyte activation. As discussed in Chapter 1, there isn’t yet consensus as to whether there is an increase in glial activation, and if so, which cell type is most responsible for pain signalling in CIPN. Our study suggests that both glial cell types are upregulated (Ruiz-Medina et al., 2013), and since amiloride seemed to reduce only the expression of GFAP, the study points to a greater role of spinal astrocytes than microglia in PTX-induced neuropathic pain (Zhang et al., 2012). Although there have been studies that found no differences in GFAP expression in the spinal cord (Pevida et al., 2013, Ledeboer et al., 2007b), this could be explained by using different animal strains or age, as differential astrocytic upregulation was demonstrated between young and adult mice (Ruiz-Medina et al., 2013). An increase in GFAP staining has been seen in other animal studies of CIPN, including cisplatin (Krukowski et al., 2017), vinctristine (Ji et al., 2013), bortezomib (Robinson and Dougherty, 2015) and oxaliplatin (Mannelli et al., 2014). These studies generally agreed that astrocytic activation has a key role in pain modulation, and it is interesting that amiloride treatment in this study ameliorated pain to some extent and reduced GFAP expression in the dorsal horn of the spinal cord. We found a trend towards increased CGRP staining in the spinal cord of PTX-treated mice. CGRP is found in C- and Aδ- sensory nerve fibres, or small unmyelinated or thinly myelinated fibres, respectively (Iyengar et al., 2017). Perhaps changes in the periphery had downstream effects in causing the slight upregulation, as well as the demonstrated activation of astrocytes and microglia in the dorsal horn of the spinal cord.

While we conducted TEM imaging and analysis on the sciatic nerves in the hope of finding some measurable effects, we did not find a difference in sciatic nerve myelination and mitochondria in PTX-treated mice. G-ratios are widely used to measure axonal myelination, and previous studies using similar models have identified a decrease in g-ratio of fibres smaller than 5 µm (Chen et al., 2015). The study used 4.5 mg/kg PTX over 4 injections, similar to our method, but used female C57BL/6 mice instead of male. Although it is possible that there may have been a methodological problem in our study, as this was the first time our lab had attempted to use TEM 90 analysis, it could also be that our model does not affect g-ratio, or may have an effect in different types of fibres. Fibres smaller than 5 µm tend to be C- and Aδ-fibres, which have little myelin; therefore, the differences between affected fibres may have been small and therefore a larger sample size may be required. Other studies have shown axonal degeneration and myelin fragmentation but did not observe a difference in g- ratio, although they did not differentiate between small and large fibres (Benbow et al., 2016). Another study found smaller, unmyelinated sensory nerve fibres to show signs of degeneration as well as vacuolised mitochondria (Bobylev et al., 2015). One study did find axonal demyelination and degeneration of C-fibres in the sciatic nerve, although they did not do this by measuring g-ratio (Mo et al., 2012). Perhaps using a different method of testing or focusing on unmyelinated fibres would provide different results in our study. Another possibility is sex differences; all the above mentioned studies used female mice, and studies have shown that cold allodynia, which is due to small fibres (Serra et al., 2009), is more pronounced in female mice than in males (Naji-Esfahani et al., 2016). Interestingly, mice (Naji-Esfahani et al., 2016) and rats (Hwang et al., 2012) do not show sex differences in mechanical allodynia.

Unfortunately, we were not able to test DRG and footpads for injury and inflammation markers such as ATF3, PGP9.5 (IENF fibres) or Iba-1. In a previous cohort of mice we did in fact find greater macrophage infiltration in the DRG of PTX-treated mice compared to control (supplementary Figure 5.1), in accordance with previous studies (Tasnim et al., 2016, Ledeboer et al., 2007b, Peters et al., 2007a, Nishida et al., 2008, Liu et al., 2010, Zhang et al., 2016). However, we were not able to test the amiloride- treated cohort for all these tissues markers due to time constraints.

Amiloride has been shown to have neuroprotective effects in various animal models such as peripheral nerve injury, where i.t. injections reduced mechanical allodynia and inhibited ASIC3 increase in the spinal dorsal horn (Jeong et al., 2013), EAE, where treatment reduced clinical disease severity in EAE mice by blocking ASIC1 expression (Friese et al., 2007) and SCI, where i.p. injections improved locomotor activity and reduced myelin loss (Durham-Lee et al., 2011) and has had analgesic effects in several models of nociception and neuropathic pain (Córdova et al., 2011, Kolasani et al., 2016,

Rocha-González et al., 2009, Muthuraman et al., 2008, Sluka et al., 2003). In this study, amiloride was injected two hours before each PTX treatment to replicate a “preventative” method of treatment, similar to the way we tested amiloride in vitro. Future studies may investigate whether amiloride can have ameliorating effects after establishment of pain. Our study showed that amiloride-treated mice had significantly lower scores on mechanical allodynia tests than PTX-treated mice. Interestingly, the only difference we found in the cellular analysis in the tissues tested was a difference in astrogliosis (GFAP staining) in the spinal cord. As mentioned previously, amiloride is an ASIC inhibitor. It was previously thought that ASICs were not expressed in GFAP+ cells (Johnson et al., 2001), however, a more recent study showed astrocytes to express ASIC1, ASIC2a, ASIC3, and TRPV1 (Huang et al., 2010). Hence, it is possible that amiloride had a direct interaction with astrocytes. There have been studies indicating that amiloride does not cross the BBB due to its hydrophilic nature (Tai and Truong, 2013), however, the study still found that amiloride was able to reduce global cerebral hypoxia-induced neurodegeneration and therefore have central effects. Indeed, animal studies using i.p. injections of amiloride have found positive effects in SCI for hind limb motor recovery (Durham-Lee et al., 2011). The study also found that amiloride reduced myelin loss in injured animals, perhaps through the protection of oligodendrocytes. A later study using an SCI rat model found that amiloride actually protects oligodendrocyte precursor cells from ER stress (Kuroiwa et al., 2014) and inhibits cell death in the spinal cord. Amiloride’s myelin-sparing effects have also been found beneficial in models of multiple sclerosis (Vergo et al., 2011). These studies indicate that amiloride does have some central effects.

There have been very few studies investigating amiloride’s effects on astrocytes specifically; amiloride was found to reverse the swelling and increase sodium uptake in methylmercury-treated astrocytes (Aschner et al., 1998, Vitarella et al., 1996) and was found to prevent the reduction of GFAP-stained cells as well as neuronal degeneration in the hippocampus of a traumatic brain injury (TBI) rat models (Zhao et al., 2008). It seems that amiloride protects astrocytes from damage by reducing or blocking sodium or calcium exchange, but it is difficult to know how this translates to reduced astrocyte activation. It could also be that amiloride ameliorated pain hypersensitivity through other mechanisms and the reduced astrocyte activation is a secondary bystander effect. 92

The mechanisms underlying amiloride effects in this study remain unknown but could be due to ASIC inhibition. Exposure to low pH conditions is known to produce pain and this may be mediated through ASIC3 (formerly known as DRASIC i.e. Dorsal Root Acid Sensing Ion Channel), a type of ASIC expressed in large-diameter mechanoreceptors and small-diameter nociceptors in the DRG as well as nerve endings in the periphery (Price et al., 2001). Loss of ASIC3 function via amiloride treatment or gene knockout prevented mechanical hyperalgesia after intramuscular acid injections in mice (Sluka et al., 2003) and sensory neurons from ASIC3 knockout mice did not respond to acidic stimuli (Price et al., 2001) indicating that it is necessary for the induction of pain. Amiloride was able to reduce neuropathic pain in spinal nerve-ligated mice, which was correlated with a decrease in ASIC3 expression in the spinal cord (Jeong et al., 2013) providing further evidence that amiloride’s effects may be central rather than peripheral. It has been shown that ASICs are also activated by acidosis produced by inflammation; ASIC1 was found to contribute to axonal degeneration in the CNS after induction of EAE, and treatment with amiloride was shown to preserve axons from damage in inflammatory lesions (Friese et al., 2007), although CNS inflammation remained at the same levels in ASIC1 knockout mice. A study that found amiloride reduced pain in vincristine-treated mice suggested that it may be due to the inhibition of Na+/Ca2+ and Na+/H+ exchangers (NHE) and their effect on calcium levels and oxidative stress (Muthuraman et al., 2008). Other Na+ channel modulators have been found to have analgesic effects in a tibial nerve injury model of neuropathic pain (Wang et al., 2011). Perhaps amiloride affects a pathway where PTX causes pain through neuroinflammation. In any case, there is sufficient data here to recommend a repeat of this study in additional animal models of CIPN, as well as further investigation of the mechanisms behind the pain ameliorating effects of amiloride in vitro and in vivo.

3.5 Conclusion This study designed a chronic model of PTX-induced peripheral neuropathy that exhibited behavioural signs of mechanical pain hypersensitivity and molecular signs of neuropathy. Furthermore, this study tested a pre-screened, clinically approved drug (amiloride) to investigate its potential as a neuroprotectant in vivo. Amiloride was found to have anti-allodynic effects and significantly reduced astrogliosis in the dorsal horn of the spinal cord. Further study is required to confirm these results.

Chapter 4: Discussion and Conclusions

4.1 Overview In this thesis, we established an in vitro (Chapter 2) and an in vivo (Chapter 3) model of PTX-induced peripheral neuropathy as well as tested neuroprotective drug candidates. PTX-induced neuropathy and neuropathic pain affects a significant number of cancer patients and there are currently few options for treatment (Brewer et al., 2016). Since repurposing drugs is a good way to streamline treatments to patients, we focused on testing drugs that were already clinically approved and which had some potential neuroprotective mechanisms. Our results from the in vitro testing provided interesting data on the protective effects of amiloride, which prompted us to test the drug in vivo in PTX-induced peripheral neuropathy. The in vivo study incorporated various behavioural tests, including the gold-standard method of mechanical allodynia and the MGS, which has not been undertaken before in PTX-treated mice, and provided some interesting data with regards to the beneficial effects of amiloride treatment.

4.2 CIPN models – in vitro and in vivo There are some challenges in developing a testable model of CIPN for PTX. Although many studies have characterised mechanisms by which PTX causes neuropathic pain, there is not yet full consensus in the literature and it is likely that there are a number of contributing factors. Because of this, there may be risks to focusing a model on only one proposed mechanism of damage, such as axonal outgrowth in the in vitro model. However, we chose this property because it was practical and reliably quantifiable, and it is directly related to the consensus mechanism by which PTX cause cytotoxicity. Our results in the in vitro study provide further evidence that PTX is neurotoxic in primary DRG neuron cultures, predominantly affecting the axons, but also inducing some swelling in the cell bodies. Our in vitro testing was performed at one of the lowest doses (10 nM PTX) cited in the literature (Gobrecht et al., 2016, Ustinova et al., 2013), thereby minimising off-target damage and more severe effects that can occur at higher dosages.

PTX-induced neuropathic pain in mice is also a challenge to model as there is substantial variation in the literature regarding behavioural and molecular outcomes and it is dependent on species, dose and time schedule. The only consensus finding is that

PTX causes mechanical allodynia, although the methods through which this is tested is varied; therefore, we carried out as many different behavioural tests as we could considering the time and resources available. Despite this, we also found a change in mechanical allodynia only. Future studies could test cold allodynia, which is uncommon but has been demonstrated in PTX-treated animals (Materazzi et al., 2012, Smith et al., 2004, Polomano et al., 2001, Nieto et al., 2008, Deng et al., 2015) or different methods of measuring thermal hyperalgesia, such as the tail flick test (D'Amour and Smith, 1941). Although rotarod measurements have not shown changes in PTX-treated mice, automated gait analysis techniques such as the Catwalk method have shown changes correlated with mechanical allodynia and reduced caudal nerve SNAP in higher-dose models (Huehnchen et al., 2013). Furthermore, electrophysiological studies on mice treated with PTX have found a decrease in amplitude and conduction velocity in sensory compound action potentials (Leandri et al., 2012), although not in C57BL/6 mice, and also reduced sensory NCV in both caudal and digital nerves but only at MTD (Wozniak et al., 2018). Being able to objectively assess neuropathy using electrophysiology would bring preclinical studies into line with the gold standard for clinical assessment and address the concern of using evoked pain to measure painful neuropathies in animal studies (Deuis et al., 2017).

Our study used a 6- injection schedule (Wozniak et al., 2011) with i.p. injections of 5mg/kg for a total of 30 mg/kg, which is equivalent to a cumulative dose that produces CIPN in patients (>1000 mg/m2) (Argyriou et al., 2007). Since PTX induces a transitory pain (P-APS) as well as a chronic neuropathic pain, we wanted to test the chronic effects of PTX and therefore started assessments a week after the end of chemotherapy, at which stage our mice had also settled down following the disruption of handling and injections. Our results provided evidence that PTX may cause neuroinflammation through the upregulation of astrocytes and microglia in the spinal cord and cause the upregulation of acetylated tubulin in the sciatic nerve. We did not find any differences in the sciatic nerve with regards to CGRP staining, or myelination of small myelinated nerve fibres, nor did we find a difference in the mitochondria structure within myelinated nerve fibres. Our next analyses will characterise the large, myelinated nerve fibres that could be affected (Boehmerle et al., 2014) as well as the small unmyelinated nerve fibres. A substantial shortcoming of this thesis is that there wasn’t available time 97 to conduct analyses of the IENFs in the footpads and perform physiological and molecular analysis of DRG dissected from the mice as part of the in vivo studies

4.3 Effects of Amiloride The results presented in this thesis provide some evidence for the use of amiloride as an analgesic agent. We found the reduction of pain in the animal study coincided with a reduction of elevated GFAP levels (indicating an increase in astrocyte activation) in the spinal cord of PTX-treated mice; perhaps amiloride has a role in alleviating pain through central effects, or through secondary effects via glial cells in the periphery. It is possible that the mild neuroprotection we saw in vitro may have been due to the drug’s beneficial effect on any supporting cells which remained in the neuronal culture following purification, such as SGCs. Our earliest tests of neuronal cultures, during the optimisation stage, had found that PTX significantly reduced the number of supporting cells, and other studies have found PTX has an effect on glial cells, such as increasing TNF-α release and expression (Wu et al., 2015), GFAP staining in the SGCs of DRG in treated mice, possibly through increased gap-junction coupling (Warwick and Hanani, 2013), or increasing infiltration of Schwann cells in sciatic nerve of mice treated at MTD (Wozniak et al., 2018). Previous studies have also implicated glial cells as an important factor for pain signalling in CIPN (Boyette-Davis et al., 2015). Further study looking into the effects of PTX and amiloride on SGCs could prove interesting.

An important aspect of CIPN treatment is making sure the drug does not disrupt the cancer therapy. The effects of amiloride on cancer cells lines or tumour-bearing mice should also be tested for interaction effects. Amiloride has previously been shown to have some inhibitory effect in certain tumour cell lines in vitro (Sparks et al., 1982) and inhibits tumour growth and proliferation in mice (Sparks et al., 1983). It has also had some anti-tumour effects in animal models of gastric cancer (Tatsuta et al., 1993), colon cancer (Tatsuta et al., 1995), malignant glioma cell lines (Hegde et al., 2004), oesophageal cancer in vitro (Guan et al., 2014), multiple myeloma (Rojas et al., 2017), the formation of breast cancer metastases (Evans and Sloan-Stakleff, 2000, Evans and Sloan Stakleff, 2004) and lung or pulmonary metastasis (Kellen et al., 1988, Evans and Sloan-Stakleff, 1998). Amiloride has been shown to sensitise pancreatic cancer cell

98 lines to other potential cancer treatments (Zheng et al., 2015) and has synergistic antimyeloma effects with four other drugs (Rojas et al., 2017). With regards to interactions between PTX and amiloride, the amiloride analogue, 5-(N,N-Dimethyl) amiloride hydrochloride (DMA), was found to produce a synergistic effect with PTX in causing apoptosis in breast cancer cell lines (Reshkin et al., 2003) perhaps through its effects on Na+/H+ exchanger, NHE1. NHE1 is thought to have a role in breast cancer metastasis and PTX-driven apoptosis (Amith and Fliegel, 2013), so amiloride at its most optimistic could provide synergistic effects with PTX as well as reduce PTX-induced neuropathy.

4.4 Translational Capacity and Future Directions The premise of this study was to provide a quick transition between in vitro, in vivo, and clinical trials to provide a viable option of treatment for those experiencing CIPN. Amiloride is sold commercially as Midamor, Amizide, or Moduretic, among other names and is relatively cheap, with each 5 mg unit costing between 50 to 80 cents as of 2018. When orally ingested, its half-life in humans is 10 hours and it is not metabolised in mammals (Sunkara, 2017). In patients with cancer, amiloride has been tested to prevent electrolyte wasting seen in those being treated with amphotericin B deoxycholate, used to treat fungal infections, with no adverse effects (Bearden and Muncey, 2001). It has also been tested in older patients to treat hypertension, and showed no effect on the risk of developing cancer over a 5-year period (Lindholm et al., 2001). However, in general, not many studies have been carried out to look specifically at amiloride’s effects on cancer or chemotherapy in patients. The side-effects of amiloride include moderate weight loss, hyperkalemia and electrolyte imbalance due its diuretic and potassium-sparing effects (Durham-Lee et al., 2011, Drugbank.ca, 2005). However, it is often noted as well-tolerated when consumed properly, and most general side effects are associated with diuresis (medicines.org.uk, 2009). An important detail to note is that amiloride is approved for use orally in humans (at 5-10 mg per day) whereas this study provided amiloride through i.p. injections. What this means for the suggested dose and mode of delivery is yet to be understood. It seems likely that a higher dose than what is currently given may be necessary; an in vitro dose of 100 μM is quite high, since in humans the plasma concentration has been suggested to sit at 30-50 μg/L, while the authors equate 50 μM in vitro to be approximately 13 mg/L (Skinner et al., 2013). 99

Indeed, a dose of 5 mg/kg in mice is also much higher than what is provided orally in humans, however intravenous amiloride has not been clinically tested to our knowledge. As previously mentioned, amiloride seems to have a different LD50 when given through different means, considering the oral LD50 of mice sits at 56 mg/kg while there are some studies that have successfully treated mice with 100 mg/kg amiloride i.p. (Córdova et al., 2011). There is also no data on human overdose. Perhaps this will be a new area to explore if clinical trials with orally ingested amiloride fail to have an effect on CIPN.

Of the drugs that have had positive effects in animal models, ALCAR seems to have had pain ameliorating effects in some patients with neuropathy (Bianchi et al., 2005, Ellithy et al., 2014, Maestri et al., 2005) or may worsen symptoms in others (Hershman et al., 2013). A pilot study for minocycline has found some promising results for the reduction of P-APS but not the prevention of CIPN (Pachman et al., 2016) and further studies are currently being conducted. Pregabalin may have reduced some neuropathy- induced numbness, but had no effect on pain (Shinde et al., 2016). Glutamate (Loven et al., 2009), glutathione (Leal et al., 2014), recombinant human leukemia inhibitory factor (rhuLIF) (Davis et al., 2005), and 2% ketamine plus 4% amitriptyline (KA) cream (Gewandter et al., 2014) were shown to have no effect on patients receiving PTX. Vitamin E seemed to protect from peripheral nerve damage in one study (Argyriou et al., 2006), but not in a later one that used the same dose (Kottschade et al., 2011). Milnacipran was shown to have adverse effects on a patient with CIPN (Nakagawa et al., 2011). That same case study showed gabapentin was not effective in relieving other symptoms of neuropathy. One group has suggested some positive results from inducing hypothermia in breast cancer patients while they received PTX treatment (Sundar et al., 2016) or similarly, regional cooling of hands and feet (Sato et al., 2016), however further testing is still required. Electroacupuncture was found to have no effect in clinical trials either (Rostock et al., 2013) even though it had alleviated mechanical allodynia in rats treated with PTX (Meng et al., 2011). Omega-3 fatty acids seemed to reduce the incidence of CIPN when given in conjunction with PTX, but did not reduce the severity of neuropathy in those who did develop CIPN (Esfahani et al., 2014). Since many animal studies were unable to be translated to clinical practice, it is certainly possible that amiloride will also prove untranslatable. 100

There are several potential future directions of this thesis. Certainly, amiloride will need to be tested again, perhaps in a dose-response study. We could also test the drug in female mice, mice strains with greater susceptibility to the effects of PTX, or in a different species, such as rats which are also commonly used to assess CIPN. Testing amiloride’s effects through oral ingestion is also a potential avenue of study considering it is taken orally in its current use. Nevertheless, although we have hypothesised some of the mechanisms through which amiloride may be acting, we cannot know for certain until further investigations are completed, and at this point it is unclear whether amiloride will have translatable effects in clinical studies.

101

4.5 Conclusion This thesis provides evidence for the use of amiloride as a potential treatment for PTX- induced neuropathic pain, with protective effects against reduction of axonal outgrowth in dissociated DRG neuronal cultures, as well as in ameliorating mechanical pain hypersensitivity and spinal astrogliosis, observed in mouse behaviour and tissue analysis. Due to time constraints, we could not do multiple repetitions of the animal study, nor could we test amiloride on cancer cell lines or animals with tumours to further test interactions between amiloride, PTX and cancer. Although mechanisms were hypothesised from these results, we were not able to test these hypotheses. Nevertheless, we provide evidence for a potential repurposed drug candidate, amiloride, for the treatment of neuropathic pain due to PTX, and suggest further avenues of exploration.

102

Supplementary Data

5.1. PTX causes an increased infiltration of macrophages in the DRG

A M a c ro p h a g e s in filtr a tio n in to th e D R G B C

1 5

e 1 0 **

a h

p 5

a m 0 H X E T V P

Figure 5.1. Macrophages in the DRG of PTX-treated mice. DRG were stained with Iba-1 for macrophages and imaged at 40x. ImageJ software was used to trace the area of the DRG, then macrophage nuclei were counted per area. (A) Results presented as number of macrophages/area*10^5. PTX-treated mice had significantly more Iba-1 staining than control mice (P<0.005). Unpaired two-tailed Student’s t-test. Data presented as mean ± SEM; n=6 mice per group (3 DRG per mouse). (B) Representative image of VEH. (C) Representative image of PTX. Scale bar: 20 µm.

103

References

ALBERTI, P. & CAVALETTI, G. 2015. Chemotherapy-Induced Peripheral Neurotoxicity (CIPN): When Epidemiological Issues Come before the Cure. International Journal of Cancer Research and Prevention, 8, 467. ALBERTS, D. S. & NOEL, J. K. 1995. Cisplatin-associated neurotoxicity: can it be prevented? Anticancer Drugs, 6, 369-83. ALESSANDRI-HABER, N., DINA, O. A., JOSEPH, E. K., REICHLING, D. B. & LEVINE, J. D. 2008. Interaction of transient receptor potential vanilloid 4, integrin, and SRC tyrosine kinase in mechanical hyperalgesia. The Journal of Neuroscience, 28, 1046-1057. ALEXANDRE, J., HU, Y., LU, W., PELICANO, H. & HUANG, P. 2007. Novel action of paclitaxel against cancer cells: bystander effect mediated by reactive oxygen species. Cancer research, 67, 3512-3517. ALOE, L., ROCCO, M. L., BIANCHI, P. & MANNI, L. 2012. Nerve growth factor: from the early discoveries to the potential clinical use. Journal of Translational Medicine, 10, 239-239. AMITH, S. R. & FLIEGEL, L. 2013. Regulation of the Na+/H+ Exchanger (NHE1) in Breast Cancer Metastasis. Cancer Research, 73, 1259-1264. ANDRÉ, N., BRAGUER, D., BRASSEUR, G., GONÇALVES, A., LEMESLE- MEUNIER, D., GUISE, S., JORDAN, M. A. & BRIAND, C. 2000. Paclitaxel Induces Release of Cytochrome from Mitochondria Isolated from Human Neuroblastoma Cells. Cancer Research, 60, 5349-5353. APFEL, S. C., LIPTON, R. B., AREZZO, J. C. & KESSLER, J. A. 1991. Nerve growth factor prevents toxic neuropathy in mice. Annals of neurology, 29, 87-90. ARGYRIOU, A. A., CHRONI, E., KOUTRAS, A., ICONOMOU, G., PAPAPETROPOULOS, S., POLYCHRONOPOULOS, P. & KALOFONOS, H. P. 2006. Preventing Paclitaxel-Induced Peripheral Neuropathy: A Phase II Trial of Vitamin E Supplementation. Journal of Pain and Symptom Management, 32, 237-244. ARGYRIOU, A. A., KYRITSIS, A. P., MAKATSORIS, T. & KALOFONOS, H. P. 2014. Chemotherapy-induced peripheral neuropathy in adults: a comprehensive update of the literature. Cancer Management and Research, 6, 135-147. 104

ARGYRIOU, A. A., POLYCHRONOPOULOS, P., KOUTRAS, A., ICONOMOU, G., ICONOMOU, A., KALOFONOS, H. P. & CHRONI, E. 2005. Peripheral neuropathy induced by administration of cisplatin- and paclitaxel-based chemotherapy. Could it be predicted? Support Care Cancer, 13, 647-51. ARGYRIOU, A. A., POLYCHRONOPOULOS, P., KOUTRAS, A., XIROS, N., PETSAS, T., ARGYRIOU, K., KALOFONOS, H. P. & CHRONI, E. 2007. Clinical and electrophysiological features of peripheral neuropathy induced by administration of cisplatin plus paclitaxel-based chemotherapy. European Journal of Cancer Care, 16, 231-7. ASCHNER, M., VITARELLA, D., ALLEN, J. W., CONKLIN, D. R. & COWAN, K. S. 1998. Methylmercury-induced astrocytic swelling is associated with activation of the Na+/H+ antiporter, and is fully reversed by amiloride. Brain Research, 799, 207-214. AUTHIER, N., GILLET, J. P., FIALIP, J., ESCHALIER, A. & COUDORE, F. 2000. Description of a short-term Taxol®-induced nociceptive neuropathy in rats. Brain Research, 887, 239-249. BALDWIN, R. M., OWZAR, K., ZEMBUTSU, H., CHHIBBER, A., KUBO, M., JIANG, C., WATSON, D., ECLOV, R. J., MEFFORD, J., MCLEOD, H. L., FRIEDMAN, P. N., HUDIS, C. A., WINER, E. P., JORGENSON, E. M., WITTE, J. S., SHULMAN, L. N., NAKAMURA, Y., RATAIN, M. J. & KROETZ, D. L. 2012. A Genome-Wide Association Study Identifies Novel Loci for Paclitaxel-Induced Sensory Peripheral Neuropathy in CALGB 40101. Clinical Cancer Research, 18, 5099-5109. BARRIENTOS, S. A., MARTINEZ, N. W., YOO, S., JARA, J. S., ZAMORANO, S., HETZ, C., TWISS, J. L. & ALVAREZ, J. 2011. Axonal degeneration is mediated by the mitochondrial permeability transition pore. The Journal of Neuroscience, 31, 966-978. BEARDEN, D. T. & MUNCEY, L. A. 2001. The effect of amiloride on amphotericin B-induced hypokalaemia. Journal of Antimicrobial Chemotherapy, 48, 109-111. BEIJERS, A., JONGEN, J. & VREUGDENHIL, G. 2012. Chemotherapy-induced neurotoxicity: the value of neuroprotective strategies. The Netherlands Journal of Medicine, 70, 18-25.

105

BELENKY, P., BOGAN, K. L. & BRENNER, C. 2007. NAD+ metabolism in health and disease. Trends in Biochemical Sciences, 32, 12-19. BEN-SHAHAR, Y. 2011. Sensory Functions for Degenerin/Epithelial Sodium Channels (DEG/ENaC). In: FRIEDMANN, T., DUNLAP, J. C. & GOODWIN, S. F. (eds.) Advances in Genetics. Academic Press. BENBOW, S. J., COOK, B. M., REIFERT, J., WOZNIAK, K. M., SLUSHER, B. S., LITTLEFIELD, B. A., WILSON, L., JORDAN, M. A. & FEINSTEIN, S. C. 2016. Effects of Paclitaxel and Eribulin in Mouse Sciatic Nerve: A Microtubule- Based Rationale for the Differential Induction of Chemotherapy-Induced Peripheral Neuropathy. Neurotoxicity Research, 29, 299-313. BENBOW, S. J., WOZNIAK, K. M., KULESH, B., SAVAGE, A., SLUSHER, B. S., LITTLEFIELD, B. A., JORDAN, M. A., WILSON, L. & FEINSTEIN, S. C. 2017. Microtubule-Targeting Agents Eribulin and Paclitaxel Differentially Affect Neuronal Cell Bodies in Chemotherapy-Induced Peripheral Neuropathy. Neurotoxicity Research, 32, 151-162. BENNETT, M. I., RAYMENT, C., HJERMSTAD, M., AASS, N., CARACENI, A. & KAASA, S. 2012. Prevalence and aetiology of neuropathic pain in cancer patients: a systematic review. Pain, 153, 359-65. BIANCHI, G., VITALI, G., CARACENI, A., RAVAGLIA, S., CAPRI, G., CUNDARI, S., ZANNA, C. & GIANNI, L. 2005. Symptomatic and neurophysiological responses of paclitaxel- or cisplatin-induced neuropathy to oral acetyl-l- carnitine. European Journal of Cancer, 41, 1746-1750. BLANQUIE, O. & BRADKE, F. 2018. Cytoskeleton dynamics in axon regeneration. Current Opinion in Neurobiology, 51, 60-69. BOBYLEV, I., JOSHI, A. R., BARHAM, M., RITTER, C., NEISS, W. F., HÖKE, A. & LEHMANN, H. C. 2015. Paclitaxel inhibits mRNA transport in axons. Neurobiology of Disease, 82, 321-331. BOEHMERLE, W., HUEHNCHEN, P., PERUZZARO, S., BALKAYA, M. & ENDRES, M. 2014. Electrophysiological, behavioral and histological characterization of paclitaxel, cisplatin, vincristine and bortezomib-induced neuropathy in C57Bl/6 mice. Scientific Reports, 4.

106

BOYETTE-DAVIS, J. A., WALTERS, E. T. & DOUGHERTY, P. M. 2015. Mechanisms involved in the development of chemotherapy-induced neuropathy. Pain Management, 5, 285-296. BREWER, J. R., MORRISON, G., DOLAN, M. E. & FLEMING, G. F. 2016. Chemotherapy-induced peripheral neuropathy: Current status and progress. Gynecologic Oncology, 140, 176-183. BROWN, S., SMITH, G., SINGLETON, R. & KOLB, N. 2016. The Clinical Phenotype and Neuropathic Outcomes of Paclitaxel-Acute Pain Syndrome (P4.233). Neurology, 86. BURGOS, E., GOMEZ-NICOLA, D., PASCUAL, D., MARTIN, M. I., NIETO- SAMPEDRO, M. & GOICOECHEA, C. 2012. Cannabinoid agonist WIN 55,212-2 prevents the development of paclitaxel-induced peripheral neuropathy in rats. Possible involvement of spinal glial cells. European Journal of Pharmacology, 682, 62-72. CACCIA, C., MAJ, R., CALABRESI, M., MAESTRONI, S., FARAVELLI, L., CURATOLO, L., SALVATI, P. & FARIELLO, R. G. 2006. Safinamide: From molecular targets to a new anti-Parkinson drug. Neurology, 67, S18-S23. CAMPANA, G. & RIMONDINI, R. 2015. Mechanical Nociception Measurement in Mice and Rats with Automated Von Frey Equipment. In: SPAMPINATO, S. M. (ed.) Opioid Receptors: Methods and Protocols. New York, NY: Springer New York. CARLSON, K. & OCEAN, A. J. 2011. Peripheral Neuropathy with Microtubule- Targeting Agents: Occurrence and Management Approach. Clinical Breast Cancer, 11, 73-81. CAROZZI, V. A., CANTA, A. & CHIORAZZI, A. 2015. Chemotherapy-induced peripheral neuropathy: What do we know about mechanisms? Neuroscience Letters, 596, 90-107. CAROZZI, V. A., CANTA, A., OGGIONI, N., SALA, B., CHIORAZZI, A., MEREGALLI, C., BOSSI, M., MARMIROLI, P. & CAVALETTI, G. 2010. Neurophysiological and neuropathological characterization of new murine models of chemotherapy-induced chronic peripheral neuropathies. Experimental Neurology, 226, 301-309.

107

CARRÉ, M., ANDRÉ, N., CARLES, G., BORGHI, H., BRICHESE, L., BRIAND, C. & BRAGUER, D. 2002. Tubulin is an inherent component of mitochondrial membranes that interacts with the voltage-dependent anion channel. Journal of Biological Chemistry, 277, 33664-33669. CARTER, N. J. & MCCORMACK, P. L. 2009. Duloxetine: a review of its use in the treatment of generalized anxiety disorder. CNS Drugs, 23, 523-41. CATA, J. P., WENG, H. R. & DOUGHERTY, P. M. 2008. The effects of thalidomide and minocycline on taxol-induced hyperalgesia in rats. Brain Research, 1229, 100-110. CATTANEO, C., BARONE, P., BONIZZONI, E. & SARDINA, M. 2017. Effects of safinamide on pain in fluctuating Parkinson’s disease patients: a post-hoc analysis. Journal of Parkinson's Disease, 7, 95-101. CETINKAYA-FISGIN, A., JOO, M. G., PING, X., THAKOR, N. V., OZTURK, C., HOKE, A. & YANG, I. H. 2016. Identification of fluocinolone acetonide to prevent paclitaxel-induced peripheral neuropathy. Journal of the Peripheral Nervous System, 21, 128-133. CHAN, O. T. & YANG, L.-X. 2000. The immunological effects of taxanes. Cancer Immunology, Immunotherapy, 49, 181-185. CHANG, W., BERTA, T., KIM, Y. H., LEE, S., LEE, S.-Y. & JI, R.-R. 2017. Expression and Role of Voltage-Gated Sodium Channels in Human Dorsal Root Ganglion Neurons with Special Focus on Nav1.7, Species Differences, and Regulation by Paclitaxel. Neuroscience Bulletin. CHAPLAN, S. R., BACH, F. W., POGREL, J. W., CHUNG, J. M. & YAKSH, T. L. 1994. Quantitative assessment of tactile allodynia in the rat paw. Journal of Neuroscience Methods, 53, 55-63. CHEN, C., BAI, X., BI, Y., LIU, G., LI, H., LIU, Z. & LIU, H. 2016. Insulin-like growth factor-1 attenuates apoptosis and protects neurochemical phenotypes of dorsal root ganglion neurons with paclitaxel-induced neurotoxicity in vitro. Nutritional Neuroscience, 1-14. CHEN, L.-H., SUN, Y.-T., CHEN, Y.-F., LEE, M.-Y., CHANG, L.-Y., CHANG, J.-Y. & SHEN, M.-R. 2015. Integrating Image-Based High-Content Screening with Mouse Models Identifies 5-Hydroxydecanoate as a Neuroprotective Drug for Paclitaxel-Induced Neuropathy. Molecular Cancer Therapeutics, 14, 2206-2214.

108

CHEN, Y., CHEN, L., YEH, Y., WU, P., CHEN, Y., CHANG, L., CHANG, J. & SHEN, M. 2017. Minoxidil is a potential neuroprotective drug for paclitaxel- induced peripheral neuropathy. Scientific Reports, 7, 45366. COBOS, E. J., NICKERSON, C. A., GAO, F., CHANDRAN, V., BRAVO- CAPARRÓS, I., GONZÁLEZ-CANO, R., RIVA, P., ANDREWS, N. A., LATREMOLIERE, A. & SEEHUS, C. R. 2018. Mechanistic Differences in Neuropathic Pain Modalities Revealed by Correlating Behavior with Global Expression Profiling. Cell Reports, 22, 1301-1312. CÓRDOVA, M. M., WERNER, M. F. D. P., SILVA, M. D. D., RUANI, A. P., PIZZOLATTI, M. G. & SANTOS, A. R. S. 2011. Further antinociceptive effects of myricitrin in chemical models of overt nociception in mice. Neuroscience Letters, 495, 173-177. COSTA, R., MOTTA, E. M., DUTRA, R. C., MANJAVACHI, M. N., BENTO, A. F., MALINSKY, F. R., PESQUERO, J. B. & CALIXTO, J. B. 2011. Anti- nociceptive effect of kinin B(1) and B(2) receptor antagonists on peripheral neuropathy induced by paclitaxel in mice. British Journal of Pharmacology, 164, 681-693. CUMMINS, T. R., DIB-HAJJ, S. D., BLACK, J. A. & WAXMAN, S. G. 2000. Sodium channels and the molecular pathophysiology of pain. Progress in Brain Research. Elsevier. D'AMOUR, F. E. & SMITH, D. L. 1941. A method for determining loss of pain sensation. Journal of Pharmacology and Experimental Therapeutics, 72, 74-79. DAS, A., SINCLAIR, D., BONKOWSKI, M., JOHNSTON, I. & WU, L. 2017. Extension of physical endurance and protection against physical, chemical and radiological trauma by NAD+ precursors. Journal of Science and Medicine in Sport, 20, S165-S166. DAVIS, I. D., KIERS, L., MACGREGOR, L., QUINN, M., AREZZO, J., GREEN, M., ROSENTHAL, M., CHIA, M., MICHAEL, M., BARTLEY, P., HARRISON, L. & DALY, M. 2005. A Randomized, Double-Blinded, Placebo-Controlled Phase II Trial of Recombinant Human Leukemia Inhibitory Factor (rhuLIF, Emfilermin, AM424) to Prevent Chemotherapy-Induced Peripheral Neuropathy. Clinical Cancer Research, 11, 1890-1898.

109

DE PICCIOTTO, N., GANO, L., JOHNSON, L., ENG, J., WEBER, S., IMAI, S.-I., MILLS, K., SEALS, D. & SINDLER, A. 2014. Nicotinamide mononucleotide supplementation reverses vascular endothelial dysfunction and large elastic artery stiffness in old mice (698.10). The FASEB Journal, 28. DEEKS, E. D. 2015. Safinamide: First Global Approval. Drugs, 75, 705-711. DENG, L., GUINDON, J., CORNETT, B. L., MAKRIYANNIS, A., MACKIE, K. & HOHMANN, A. G. 2015. Chronic Cannabinoid Receptor 2 Activation Reverses Paclitaxel Neuropathy Without Tolerance or Cannabinoid Receptor 1– Dependent Withdrawal. Biological Psychiatry, 77, 475-487. DEUIS, J. R., DVORAKOVA, L. S. & VETTER, I. 2017. Methods Used to Evaluate Pain Behaviors in Rodents. Frontiers in Molecular Neuroscience, 10, 284. DEVAL, E., GASULL, X., NOËL, J., SALINAS, M., BARON, A., DIOCHOT, S. & LINGUEGLIA, E. 2010. Acid-Sensing Ion Channels (ASICs): Pharmacology and implication in pain. Pharmacology & Therapeutics, 128, 549-558. DI STEFANO, M., NASCIMENTO-FERREIRA, I., ORSOMANDO, G., MORI, V., GILLEY, J., BROWN, R., JANECKOVA, L., VARGAS, M. E., WORRELL, L. A., LORETO, A., TICKLE, J., PATRICK, J., WEBSTER, J. R. M., MARANGONI, M., CARPI, F. M., PUCCIARELLI, S., ROSSI, F., MENG, W., SAGASTI, A., RIBCHESTER, R. R., MAGNI, G., COLEMAN, M. P. & CONFORTI, L. 2015. A rise in NAD precursor nicotinamide mononucleotide (NMN) after injury promotes axon degeneration. Cell Death & Differentiation, 22, 731-742. DIB-HAJJ, S. D., BINSHTOK, A. M., CUMMINS, T. R., JARVIS, M. F., SAMAD, T. & ZIMMERMANN, K. 2009. Voltage-gated sodium channels in pain states: role in pathophysiology and targets for treatment. Brain Research Reviews, 60, 65-83. DIB-HAJJ, S. D., CUMMINS, T. R., BLACK, J. A. & WAXMAN, S. G. 2007. From genes to pain: Nav1. 7 and human pain disorders. Trends in Neurosciences, 30, 555-563. DONVITO, G., WILKERSON, J. L., DAMAJ, M. I. & LICHTMAN, A. H. 2016. Palmitoylethanolamide Reverses Paclitaxel-Induced Allodynia in Mice. Journal of Pharmacology and Experimental Therapeutics, 359, 310-318.

110

DOUGHERTY, P. M., CATA, J. P., CORDELLA, J. V., BURTON, A. & WENG, H. R. 2004. Taxol induced sensory disturbance is characterized by preferential impairment of myelinated fiber function in cancer patients. Pain, 109. DRUGBANK.CA. 2005. Drugbank: Amiloride (DB00594) [Online]. Available: https://www.drugbank.ca/drugs/DB00594 [Accessed 5 Nov 2017]. DUAN, B., WU, L. J., YU, Y. Q., DING, Y., JING, L., XU, L., CHEN, J. & XU, T. L. 2007. Upregulation of acid-sensing ion channel ASIC1a in spinal dorsal horn neurons contributes to inflammatory pain hypersensitivity. The Journal of Neuroscience, 27, 11139-48. DUBE, G., LEHTO, S. G., BREESE, N. M., BAKER, S. J., WANG, X., MATULENKO, M. A., HONORÉ, P., STEWART, A. O., MORELAND, R. B. & BRIONI, J. D. 2005. Electrophysiological and in vivo characterization of A- 317567, a novel blocker of acid sensing ion channels. Pain, 117, 88-96. DUFFY, S. S., PERERA, C. J., MAKKER, P. G. S., LEES, J. G., CARRIVE, P. & MOALEM-TAYLOR, G. 2016. Peripheral and Central Neuroinflammatory Changes and Pain Behaviors in an Animal Model of Multiple Sclerosis. Frontiers in Immunology, 7. DUGGETT, N. A., GRIFFITHS, L. A. & FLATTERS, S. J. 2017. Paclitaxel-induced painful neuropathy is associated with changes in mitochondrial bioenergetics, glycolysis, and an energy deficit in dorsal root ganglia neurons. Pain, 158, 1499. DUGGETT, N. A., GRIFFITHS, L. A., MCKENNA, O. E., DE SANTIS, V., YONGSANGUANCHAI, N., MOKORI, E. B. & FLATTERS, S. J. 2016. Oxidative stress in the development, maintenance and resolution of paclitaxel- induced painful neuropathy. Neuroscience, 333, 13-26. DURHAM-LEE, J. C., MOKKAPATI, V. U. L., JOHNSON, K. M. & NESIC, O. 2011. Amiloride improves locomotor recovery after spinal cord injury. Journal of neurotrauma, 28, 1319-1326. EDWARDS, R. R., DWORKIN, R. H., SULLIVAN, M. D., TURK, D. C. & WASAN, A. D. 2016. The Role of Psychosocial Processes in the Development and Maintenance of Chronic Pain. The Journal of Pain, 17, T70-T92. EISENHAUER, E. A., TEN BOKKEL HUININK, W., SWENERTON, K. D., GIANNI, L., MYLES, J., VAN DER BURG, M., KERR, I., VERMORKEN, J. B., BUSER, K. & COLOMBO, N. 1994. European-Canadian randomized trial

111

of paclitaxel in relapsed ovarian cancer: high-dose versus low-dose and long versus short infusion. Journal of Clinical Oncology, 12, 2654-2666. ELLIS, A., WIESELER, J., FAVRET, J., JOHNSON, K. W., RICE, K. C., MAIER, S. F., FALCI, S. & WATKINS, L. R. 2014. Systemic Administration of Propentofylline, Ibudilast, and (+)-Naltrexone Each Reverses Mechanical Allodynia in a Novel Rat Model of Central Neuropathic Pain. The Journal of Pain, 15, 407-421. ELLITHY, M., GHALI, R., ELGHAMRY, W., ELTAEB, S. & ELHOSEINY, K. 2014. The Effect of Acetyl-L-Carnitine in Reducing the Taxanes Induced Neuropathy. Neuro-oncology, 16, ii6. ERTÜRK, A., HELLAL, F., ENES, J. & BRADKE, F. 2007. Disorganized Microtubules Underlie the Formation of Retraction Bulbs and the Failure of Axonal Regeneration. The Journal of Neuroscience, 27, 9169-9180. ESFAHANI, A., GHOREISHI, Z., EFTEKHARSADAT, B. & JAFARABADI, M. A. 2014. The effects of omega-3 fatty acids on chemotherapy-induced neuropathy and inflammation in patients with breast cancer. Journal of Clinical Oncology, 32, 1039-1039. EVANS, D. M. & SLOAN-STAKLEFF, K. D. 1998. Maximum effect of urokinase plasminogen activator inhibitors in the control of invasion and metastasis of rat mammary cancer. Invasion Metastasis, 18, 252-60. EVANS, D. M. & SLOAN-STAKLEFF, K. D. 2000. Suppression of the invasive capacity of human breast cancer cells by inhibition of urokinase plasminogen activator via amiloride and B428. The American Journal of Surgery, 66, 460-4. EVANS, D. M. & SLOAN STAKLEFF, K. D. 2004. Control of pulmonary metastases of rat mammary cancer by inhibition of uPA and COX-2, singly and in combination. Clinical & Experimental Metastasis, 21, 339-46. EZENDAM, N. P. M., PIJLMAN, B., BHUGWANDASS, C., PRUIJT, J. F. M., MOLS, F., VOS, M. C., PIJNENBORG, J. M. A. & VAN DE POLL-FRANSE, L. V. 2014. Chemotherapy-induced peripheral neuropathy and its impact on health-related quality of life among ovarian cancer survivors: Results from the population-based PROFILES registry. Gynecologic Oncology, 135, 510-517. FARIELLO, R. G., MCARTHUR, R. A., BONSIGNORI, A., CERVINI, M. A., MAJ, R., MARRARI, P., PEVARELLO, P., WOLF, H. H., WOODHEAD, J. W.,

112

WHITE, H. S., VARASI, M., SALVATI, P. & POST, C. 1998. Preclinical Evaluation of PNU-151774E as a Novel Anticonvulsant. Journal of Pharmacology and Experimental Therapeutics, 285, 397-403. FELLNER, S., BAUER, B., MILLER, D. S., SCHAFFRIK, M., FANKHÄNEL, M., SPRUß, T., BERNHARDT, G., GRAEFF, C., FÄRBER, L., GSCHAIDMEIER, H., BUSCHAUER, A. & FRICKER, G. 2002. Transport of paclitaxel (Taxol) across the blood-brain barrier in vitro and in vivo. The Journal of Clinical Investigation, 110, 1309-1318. FENHAMMAR, M. D. J., RUNDGREN, M. D. P. D. M., FORESTIER, M. D. J., KALMAN, M. D. P. D. S., ERIKSSON, O. D. P. D. S. & FRITHIOF, M. D. P. D. R. 2011. Toll-Like Receptor 4 Inhibitor TAK-242 Attenuates Acute Kidney Injury in Endotoxemic Sheep. Anesthesiology, 114, 1130-1137. FISCHER, S. J., PODRATZ, J. L. & WINDEBANK, A. J. 2001. Nerve growth factor rescue of cisplatin neurotoxicity is mediated through the high affinity receptor: studies in PC12 cells and p75 null mouse dorsal root ganglia. Neuroscience Letters, 308, 1-4. FLATTERS, S. J., XIAO, W. H. & BENNETT, G. J. 2006. Acetyl-L-carnitine prevents and reduces paclitaxel-induced painful peripheral neuropathy. Neuroscience Letters, 397. FLATTERS, S. J. L. & BENNETT, G. J. 2004. Ethosuximide reverses paclitaxel- and vincristine-induced painful peripheral neuropathy. Pain, 109, 150-161. FLATTERS, S. J. L. & BENNETT, G. J. 2006. Studies of peripheral sensory nerves in paclitaxel-induced painful peripheral neuropathy: Evidence for mitochondrial dysfunction. Pain, 122, 245-257. FRANÇOIS, A., SCHÜETTER, N., LAFFRAY, S., SANGUESA, J., PIZZOCCARO, A., DUBEL, S., MANTILLERI, A., NARGEOT, J., NOËL, J., WOOD, JOHN N., MOQRICH, A., PONGS, O. & BOURINET, E. 2015. The Low- Threshold Calcium Channel Cav3.2 Determines Low-Threshold Mechanoreceptor Function. Cell Reports, 10, 370-382. FRELIN, C., BARBRY, P., VIGNE, P., CHASSANDE, O., CRAGOE, E. J. & LAZDUNSKI, M. 1988. Amiloride and its analogs as tools to inhibit Na+ transport via the Na+ channel, the Na+/H+ antiport and the Na+/Ca2+ exchanger. Biochimie, 70, 1285-1290.

113

FRIESE, M. A., CRANER, M. J., ETZENSPERGER, R., VERGO, S., WEMMIE, J. A., WELSH, M. J., VINCENT, A. & FUGGER, L. 2007. Acid-sensing ion channel- 1 contributes to axonal degeneration in autoimmune inflammation of the central nervous system. Nature Medicine, 13, 1483. FUKUDA, Y., LI, Y. & SEGAL, R. A. 2017. A Mechanistic understanding of axon degeneration in chemotherapy-induced peripheral neuropathy. Frontiers in Neuroscience, 11, 481. GALLEY, H. F., MCCORMICK, B., WILSON, K. L., LOWES, D. A., COLVIN, L. & TORSNEY, C. 2017. Melatonin limits paclitaxel-induced mitochondrial dysfunction in vitro and protects against paclitaxel-induced neuropathic pain in the rat. Journal of Pineal Research, 63, e12444. GARATE, I., GARCIA-BUENO, B., MADRIGAL, J. L., CASO, J. R., ALOU, L., GOMEZ-LUS, M. L. & LEZA, J. C. 2014. Toll-like 4 receptor inhibitor TAK- 242 decreases neuroinflammation in rat brain frontal cortex after stress. Journal of Neuroinflammation, 11, 8. GAUCHAN, P., ANDOH, T., IKEDA, K., FUJITA, M., SASAKI, A., KATO, A. & KURAISHI, Y. 2009. Mechanical Allodynia Induced by Paclitaxel, Oxaliplatin and Vincristine: Different Effectiveness of Gabapentin and Different Expression of Voltage-Dependent Calcium Channel a2d-1 Subunit. Biological and Pharmaceutical Bulletin, 32, 732-734. GELDERBLOM, H., VERWEIJ, J., NOOTER, K. & SPARREBOOM, A. 2001. Cremophor EL: the drawbacks and advantages of vehicle selection for drug formulation. European Journal of Cancer, 37, 1590-1598. GEWANDTER, J. S., MOHILE, S. G., HECKLER, C. E., RYAN, J. L., KIRSHNER, J. J., FLYNN, P. J., HOPKINS, J. O. & MORROW, G. R. 2014. A phase III randomized, placebo-controlled study of topical amitriptyline and ketamine for chemotherapy-induced peripheral neuropathy (CIPN): a University of Rochester CCOP study of 462 cancer survivors. Supportive Care in Cancer, 22, 1807- 1814. GOBRECHT, P., ANDREADAKI, A., DIEKMANN, H., HESKAMP, A., LEIBINGER, M. & FISCHER, D. 2016. Promotion of Functional Nerve Regeneration by Inhibition of Microtubule Detyrosination. The Journal of Neuroscience, 36, 3890-3902.

114

GOLDSTEIN, D. J., LU, Y., DETKE, M. J., LEE, T. C. & IYENGAR, S. 2005. Duloxetine vs. placebo in patients with painful diabetic neuropathy. Pain, 116, 109-18. GOMES, A. P., PRICE, N. L., LING, A. J., MOSLEHI, J. J., MONTGOMERY, M. K., RAJMAN, L., WHITE, J. P., TEODORO, J. S., WRANN, C. D., HUBBARD, B. P., MERCKEN, E. M., PALMEIRA, C. M., DE CABO, R., ROLO, A. P., TURNER, N., BELL, E. L. & SINCLAIR, D. A. 2013. Declining NAD(+) induces a pseudohypoxic state disrupting nuclear-mitochondrial communication during aging. Cell, 155, 1624-38. GORE, M., DUKES, E., ROWBOTHAM, D. J., TAI, K.-S. & LESLIE, D. 2007. Clinical characteristics and pain management among patients with painful peripheral neuropathic disorders in general practice settings. European Journal of Pain, 11, 652-664. GORNSTEIN, E. & SCHWARZ, T. L. 2014. The paradox of paclitaxel neurotoxicity: Mechanisms and unanswered questions. Neuropharmacology, 76, Part A, 175- 183. GORNSTEIN, E. L. & SCHWARZ, T. L. 2017. Neurotoxic mechanisms of paclitaxel are local to the distal axon and independent of transport defects. Experimental Neurology, 288, 153-166. GREEN, M. R., MANIKHAS, G. M., ORLOV, S., AFANASYEV, B., MAKHSON, A. M., BHAR, P. & HAWKINS, M. J. 2006. Abraxane®, a novel Cremophor®- free, albumin-bound particle form of paclitaxel for the treatment of advanced non-small-cell lung cancer. Annals of Oncology, 17, 1263-1268. GRIFFITHS, L. A. & FLATTERS, S. J. 2015. Pharmacological Modulation of the Mitochondrial Electron Transport Chain in Paclitaxel-Induced Painful Peripheral Neuropathy. The Journal of Pain, 16, 981-94. GRISOLD, W., CAVALETTI, G. & WINDEBANK, A. J. 2012. Peripheral neuropathies from chemotherapeutics and targeted agents: diagnosis, treatment, and prevention. Neuro-oncology, 14, iv45-iv54. GSCHOSSMANN, J. M., CHABAN, V. V., MCROBERTS, J. A., RAYBOULD, H. E., YOUNG, S. H., ENNES, H. S., LEMBO, T. & MAYER, E. A. 2000. Mechanical activation of dorsal root ganglion cells in vitro: comparison with capsaicin and modulation by κ-opioids. Brain Research, 856, 101-110.

115

GUAN, B., HOQUE, A. & XU, X. 2014. Amiloride and guggulsterone suppression of esophageal cancer cell growth in vitro and in nude mouse xenografts. Frontiers in Biology, 9, 75-81. GUI, Y., LI, A., CHEN, F., ZHOU, H., TANG, Y., CHEN, L., CHEN, S. & DUAN, S. 2015. Involvement of AMPK/SIRT1 pathway in anti-allodynic effect of troxerutin in CCI-induced neuropathic pain. European Journal of Pharmacology, 769, 234-241. GUO, L., HAMRE, I. I. I. J., ELDRIDGE, S., BEHRSING, H. P., CUTULI, F. M., MUSSIO, J. & DAVIS, M. 2017. Multiparametric Image Analysis of Rat Dorsal Root Ganglion Cultures to Evaluate Peripheral Neuropathy-Inducing Chemotherapeutics. Toxicological Sciences, 156, 275-288. GUTIÉRREZ-GUTIÉRREZ, G., SERENO, M., MIRALLES, A., CASADO-SÁENZ, E. & GUTIÉRREZ-RIVAS, E. 2010. Chemotherapy-induced peripheral neuropathy: clinical features, diagnosis, prevention and treatment strategies. Clinical and Translational Oncology, 12, 81-91. HADZIC, T., AYKIN-BURNS, N., ZHU, Y., COLEMAN, M. C., LEICK, K., JACOBSON, G. M. & SPITZ, D. R. 2010. Paclitaxel combined with inhibitors of glucose and hydroperoxide metabolism enhances breast cancer cell killing via H 2 O 2-mediated oxidative stress. Free Radical Biology and Medicine, 48, 1024-1033. HAMA, A. T., BROADHEAD, A., LORRAIN, D. S. & SAGEN, J. 2012. The antinociceptive effect of the asthma drug ibudilast in rat models of peripheral and central neuropathic pain. Journal of Neurotrauma, 29, 600-610. HEGDE, M., ROSCOE, J., CALA, P. & GORIN, F. 2004. Amiloride kills malignant glioma cells independent of its inhibition of the sodium-hydrogen exchanger. Journal of Pharmacology and Experimental Therapeutics, 310, 67-74. HENDERSON, J., POLLACK, A. J., PAN, Y. & MILLER, G. C. 2016. Neuropathic and non-neuropathic chronic pain at GP encounters: Prevalence, patient characteristics, suffering and pregabalin use. Australian Family Physician, 45, 783. HENSCHKE, N., KAMPER, S. J. & MAHER, C. G. The epidemiology and economic consequences of pain. Mayo Clinic Proceedings, 2015. Elsevier, 139-147.

116

HERSHMAN, D. L., LACCHETTI, C., DWORKIN, R. H., SMITH, E. M. L., BLEEKER, J., CAVALETTI, G., CHAUHAN, C., GAVIN, P., LAVINO, A., LUSTBERG, M. B., PAICE, J., SCHNEIDER, B., SMITH, M. L., SMITH, T., TERSTRIEP, S., WAGNER-JOHNSTON, N., BAK, K. & LOPRINZI, C. L. 2014. Prevention and Management of Chemotherapy-Induced Peripheral Neuropathy in Survivors of Adult Cancers: American Society of Clinical Oncology Clinical Practice Guideline. Journal of Clinical Oncology, 32, 1941- 1967. HERSHMAN, D. L., UNGER, J. M., CREW, K. D., MINASIAN, L. M., AWAD, D., MOINPOUR, C. M., HANSEN, L., LEW, D. L., GREENLEE, H., FEHRENBACHER, L., WADE, J. L., WONG, S.-F., HORTOBAGYI, G. N., MEYSKENS, F. L. & ALBAIN, K. S. 2013. Randomized Double-Blind Placebo-Controlled Trial of Acetyl-L-Carnitine for the Prevention of Taxane- Induced Neuropathy in Women Undergoing Adjuvant Breast Cancer Therapy. Journal of Clinical Oncology, 31, 2627-2633. HERTZ, D. L., ROY, S., JACK, J., MOTSINGER-REIF, A. A., DROBISH, A., CLARK, L. S., CAREY, L. A., DEES, E. C. & MCLEOD, H. L. 2014. Genetic heterogeneity beyond CYP2C8*3 does not explain differential sensitivity to paclitaxel-induced neuropathy. Breast Cancer Research and Treatment, 145, 245-54. HERTZ, D. L., ROY, S., MOTSINGER-REIF, A. A., DROBISH, A., CLARK, L. S., MCLEOD, H. L., CAREY, L. A. & DEES, E. C. 2013. CYP2C8*3 increases risk of neuropathy in breast cancer patients treated with paclitaxel. Annals of Oncology, 24, 1472-1478. HIDAKA, T., SHIMA, T., NAGIRA, K., IEKI, M., NAKAMURA, T., AONO, Y., KURAISHI, Y., ARAI, T. & SAITO, S. 2009. Herbal medicine Shakuyaku- kanzo-to reduces paclitaxel-induced painful peripheral neuropathy in mice. European Journal of Pain, 13, 22-27. HIRAYAMA, Y., ISHITANI, K., SATO, Y., IYAMA, S., TAKADA, K., MURASE, K., KURODA, H., NAGAMACHI, Y., KONUMA, Y., FUJIMI, A., SAGAWA, T., ONO, K., HORIGUCHI, H., TERUI, T., KOIKE, K., KUSAKABE, T., SATO, T., TAKIMOTO, R., KOBUNE, M. & KATO, J. 2015. Effect of duloxetine in Japanese patients with chemotherapy-induced peripheral

117

neuropathy: a pilot randomized trial. International Journal of Clinical Oncology, 20, 866-871. HOSOI, T., KUME, A., OTANI, K., OBA, T. & OZAWA, K. 2010. A unique modulator of endoplasmic reticulum stress-signalling pathways: the novel pharmacological properties of amiloride in glial cells. British Journal of Pharmacology, 159, 428-437. HUA, F., TANG, H., WANG, J., PRUNTY, M. C., HUA, X., SAYEED, I. & STEIN, D. G. 2015. TAK-242, an Antagonist for Toll-like Receptor 4, Protects against Acute Cerebral Ischemia/Reperfusion Injury in Mice. Journal of Cerebral Blood Flow & Metabolism, 35, 536-542. HUANG, C., HU, Z. L., WU, W. N., YU, D. F., XIONG, Q. J., SONG, J. R., SHU, Q., FU, H., WANG, F. & CHEN, J. G. 2010. Existence and distinction of acid- evoked currents in rat astrocytes. Glia, 58, 1415-1424. HUEHNCHEN, P., BOEHMERLE, W. & ENDRES, M. 2013. Assessment of Paclitaxel Induced Sensory Polyneuropathy with “Catwalk” Automated Gait Analysis in Mice. PLoS ONE, 8, e76772. HUH, Y., CHEN, G. & JI, R. 2016. Intrathecally injected bone marrow stromal cells produce sustained neuropathic pain reduction in paclitaxel-treated mice. The Journal of Pain, 17, S78. HWANG, B.-Y., KIM, E.-S., KIM, C.-H., KWON, J.-Y. & KIM, H.-K. 2012. Gender differences in paclitaxel-induced neuropathic pain behavior and analgesic response in rats. Korean Journal of Anesthesiology, 62, 66-72. IASP-PAIN.ORG. 2017. IASP Taxonomy [Online]. International Association for the Study of Pain. Available: https://www.iasp-pain.org/Taxonomy [Accessed 5 May 2017]. II, M., MATSUNAGA, N., HAZEKI, K., NAKAMURA, K., TAKASHIMA, K., SEYA, T., HAZEKI, O., KITAZAKI, T. & IIZAWA, Y. 2006. A novel cyclohexene derivative, ethyl (6R)-6-[N-(2-Chloro-4- fluorophenyl)sulfamoyl]cyclohex-1-ene-1-carboxylate (TAK-242), selectively inhibits toll-like receptor 4-mediated cytokine production through suppression of intracellular signaling. Molecular Pharmacology, 69, 1288-1295. IMAI, S., KOYANAGI, M., AZIMI, Z., NAKAZATO, Y., MATSUMOTO, M., OGIHARA, T., YONEZAWA, A., OMURA, T., NAKAGAWA, S. &

118

WAKATSUKI, S. 2017. Taxanes and platinum derivatives impair Schwann cells via distinct mechanisms. Scientific Reports, 7, 5947. ITO, S., SUTO, T., SAITO, S. & OBATA, H. 2018. Repeated Administration of Duloxetine Suppresses Neuropathic Pain by Accumulating Effects of Noradrenaline in the Spinal Cord. Anesthesia & Analgesia, 126, 298-307. IYENGAR, S., OSSIPOV, M. H. & JOHNSON, K. W. 2017. The role of calcitonin gene–related peptide in peripheral and central pain mechanisms including migraine. Pain, 158, 543-559. JANES, K., ESPOSITO, E., DOYLE, T., CUZZOCREA, S., TOSH, D. K., JACOBSON, K. A. & SALVEMINI, D. 2014a. A3 adenosine receptor agonist prevents the development of paclitaxel-induced neuropathic pain by modulating spinal glial-restricted redox-dependent signaling pathways. Pain, 155, 2560- 2567. JANES, K., LITTLE, J. W., LI, C., BRYANT, L., CHEN, C., CHEN, Z., KAMOCKI, K., DOYLE, T., SNIDER, A., ESPOSITO, E., CUZZOCREA, S., BIEBERICH, E., OBEID, L., PETRACHE, I., NICOL, G., NEUMANN, W. L. & SALVEMINI, D. 2014b. The Development and Maintenance of Paclitaxel- induced Neuropathic Pain Require Activation of the Sphingosine 1-Phosphate Receptor Subtype 1. The Journal of Biological Chemistry, 289, 21082-21097. JEONG, S., LEE, S. H., KIM, Y. O. & YOON, M. H. 2013. Antinociceptive effects of amiloride and benzamil in neuropathic pain model rats. Journal of Korean Medical Science, 28, 1238-1243. JI, X.-T., QIAN, N.-S., ZHANG, T., LI, J.-M., LI, X.-K., WANG, P., ZHAO, D.-S., HUANG, G., ZHANG, L., FEI, Z., JIA, D. & NIU, L. 2013. Spinal Astrocytic Activation Contributes to Mechanical Allodynia in a Rat Chemotherapy-Induced Neuropathic Pain Model. PLoS ONE, 8, e60733. JIN, H. W., FLATTERS, S. J. L., XIAO, W. H., MULHERN, H. L. & BENNETT, G. J. 2008. Prevention of paclitaxel-evoked painful peripheral neuropathy by acetyl- L-carnitine: Effects on axonal mitochondria, sensory nerve fiber terminal arbors, and cutaneous Langerhans cells. Experimental Neurology, 210, 229-237. JOHNSON, M. B., JIN, K., MINAMI, M., CHEN, D. & SIMON, R. P. 2001. Global Ischemia Induces Expression of Acid-Sensing Ion Channel 2a in Rat Brain. Journal of Cerebral Blood Flow & Metabolism, 21, 734-740.

119

KATSUYAMA, S., SATO, K., YAGI, T., KISHIKAWA, Y. & NAKAMURA, H. 2013. Effects of repeated milnacipran and fluvoxamine treatment on mechanical allodynia in a mouse paclitaxel-induced neuropathic pain model. Biomedical Research, 34, 105-111. KELLEN, J. A., MIRAKIAN, A. & KOLIN, A. 1988. Antimetastatic effect of amiloride in an animal tumour model. Anticancer Research, 8, 1373-6. KIDD, J. F., PILKINGTON, M. F., SCHELL, M. J., FOGARTY, K. E., SKEPPER, J. N., TAYLOR, C. W. & THORN, P. 2002. Paclitaxel affects cytosolic calcium signals by opening the mitochondrial permeability transition pore. Journal of Biological Chemistry, 277, 6504-6510. KITAMURA, N., SAKAMOTO, K., ONO, T. & KIMURA, J. 2015. The Inhibitory Effect of Paclitaxel on (KV2.1) K(+) Current in H9c2 Cells. Fukushima Journal of Medical Science, 61, 47-53. KLEYMAN, T. R. & CRAGOE, E. J., JR. 1988. Amiloride and its analogs as tools in the study of ion transport. The Journal of Membrane Biology, 105, 1-21. KOLASANI, B. P., PONNALURI, R. R. & MUDIUM, R. 2016. Study of analgesic effect of amiloride compared to aspirin in chemically induced pain models in rats and mice. National Journal of Physiology, Pharmacy and Pharmacology, 6, 544-549. KONINGS, P. N. M., KAROLIEN MAKKINK, W., VAN DELFT, A. M. L. & RUIGT, G. S. F. 1994. Reversal by NGF of cytostatic drug-induced reduction of neurite outgrowth in rat dorsal root ganglia in vitro. Brain Research, 640, 195-204. KOTTSCHADE, L. A., SLOAN, J. A., MAZURCZAK, M. A., JOHNSON, D. B., MURPHY, B. P., ROWLAND, K. M., SMITH, D. A., BERG, A. R., STELLA, P. J. & LOPRINZI, C. L. 2011. The use of vitamin E for the prevention of chemotherapy-induced peripheral neuropathy: results of a randomized phase III clinical trial. Supportive Care in Cancer, 19, 1769-1777. KRUKOWSKI, K., EIJKELKAMP, N., LAUMET, G., HACK, C. E., LI, Y., DOUGHERTY, P. M., HEIJNEN, C. J. & KAVELAARS, A. 2016. CD8+ T Cells and Endogenous IL-10 Are Required for Resolution of Chemotherapy- Induced Neuropathic Pain. The Journal of Neuroscience, 36, 11074-11083. KRUKOWSKI, K., MA, J., GOLONZHKA, O., LAUMET, G. O., GUTTI, T., VAN DUZER, J. H., MAZITSCHEK, R., JARPE, M. B., HEIJNEN, C. J. &

120

KAVELAARS, A. 2017. HDAC6 inhibition effectively reverses chemotherapy- induced peripheral neuropathy. Pain, 158, 1126-1137. KUMAR, S., MAHDI, H., BRYANT, C., SHAH, J. P., GARG, G. & MUNKARAH, A. 2010. Clinical trials and progress with paclitaxel in ovarian cancer. International Journal of Women's Health, 2, 411-427. KUNO, M., NEMOTO, K., NINOMIYA, N., INAGAKI, E., KUBOTA, M., MATSUMOTO, T. & YOKOTA, H. 2009. THE NOVEL SELECTIVE TOLL- LIKE RECEPTOR 4 SIGNAL TRANSDUCTION INHIBITOR TAK-242 PREVENTS ENDOTOXAEMIA IN CONSCIOUS GUINEA-PIGS. Clinical and Experimental Pharmacology and Physiology, 36, 589-593. KUROIWA, M., WATANABE, M., KATOH, H., SUYAMA, K., MATSUYAMA, D., IMAI, T. & MOCHIDA, J. 2014. Effect of amiloride on endoplasmic reticulum stress response in the injured spinal cord of rats. European Journal of Neuroscience, 40, 3120-3127. LANGFORD, D. J., BAILEY, A. L., CHANDA, M. L., CLARKE, S. E., DRUMMOND, T. E., ECHOLS, S., GLICK, S., INGRAO, J., KLASSEN- ROSS, T., LACROIX-FRALISH, M. L., MATSUMIYA, L., SORGE, R. E., SOTOCINAL, S. G., TABAKA, J. M., WONG, D., VAN DEN MAAGDENBERG, A. M., FERRARI, M. D., CRAIG, K. D. & MOGIL, J. S. 2010. Coding of facial expressions of pain in the laboratory mouse. Nature Methods, 7, 447-9. LANZOTTI, V. 2013. Diterpenes for Therapeutic Use. In: RAMAWAT, G. K. & MÉRILLON, J. M. (eds.) Natural Products: Phytochemistry, Botany and Metabolism of Alkaloids, Phenolics and Terpenes. Berlin, Heidelberg: Springer Berlin Heidelberg. LAPOINTE, N. E., MORFINI, G., BRADY, S. T., FEINSTEIN, S. C., WILSON, L. & JORDAN, M. A. 2013. Effects of eribulin, vincristine, paclitaxel and ixabepilone on fast axonal transport and kinesin-1 driven microtubule gliding: Implications for chemotherapy-induced peripheral neuropathy. NeuroToxicology, 37, 231-239. LEAL, A. D., QIN, R., ATHERTON, P. J., HALUSKA, P., BEHRENS, R. J., TIBER, C. H., WATANABOONYAKHET, P., WEISS, M., ADAMS, P. T., DOCKTER, T. J. & LOPRINZI, C. L. 2014. North Central Cancer Treatment

121

Group/Alliance trial N08CA—the use of glutathione for prevention of paclitaxel/carboplatin‐induced peripheral neuropathy: A phase 3 randomized, double‐blind, placebo‐controlled study. Cancer, 120, 1890-1897. LEANDRI, M., GHIGNOTTI, M., EMIONITE, L., LEANDRI, S. & CILLI, M. 2012. Electrophysiological features of the mouse tail nerves and their changes in chemotherapy induced peripheral neuropathy (CIPN). Journal of Neuroscience Methods, 209, 403-409. LEDEBOER, A., HUTCHINSON, M. R., WATKINS, L. R. & JOHNSON, K. W. 2007a. Ibudilast (AV-411). Expert Opinion on Investigational Drugs, 16, 935- 950. LEDEBOER, A., JEKICH, B. M., SLOANE, E. M., MAHONEY, J. H., LANGER, S. J., MILLIGAN, E. D., MARTIN, D., MAIER, S. F., JOHNSON, K. W., LEINWAND, L. A., CHAVEZ, R. A. & WATKINS, L. R. 2007b. Intrathecal interleukin-10 gene therapy attenuates paclitaxel-induced mechanical allodynia and proinflammatory cytokine expression in dorsal root ganglia in rats. Brain, Behavior, and Immunity, 21, 686-698. LEDEBOER, A., LIU, T., SHUMILLA, J. A., MAHONEY, J. H., VIJAY, S., GROSS, M. I., VARGAS, J. A., SULTZBAUGH, L., CLAYPOOL, M. D. & SANFTNER, L. M. 2006. The glial modulatory drug AV411 attenuates mechanical allodynia in rat models of neuropathic pain. Neuron Glia Biology, 2, 279-291. LEE, R., OSTER, M., BALMACEDA, C., HESDORFFER, C., VAHDAT, L. & PAPADOPOULOS, K. 1999. Bilateral facial nerve palsy secondary to the administration of high-dose paclitaxel. Annals of Oncology, 10, 1245-1247. LEE, S. I. & LEVINE, J. 2015. Isolation and Growth of Adult Mouse Dorsal Root Ganglia Neurons. Bio-protocol, 5, e1601. LEE, S. I., ZHANG, W., RAVI, M., WESCHENFELDER, M., BASTMEYER, M. & LEVINE, J. M. 2013. Atypical Protein Kinase C and Par3 Are Required for Proteoglycan-Induced Axon Growth Inhibition. The Journal of Neuroscience, 33, 2541-2554. LI, Y., ADAMEK, P., ZHANG, H., TATSUI, C. E., RHINES, L. D., MROZKOVA, P., LI, Q., KOSTURAKIS, A. K., CASSIDY, R. M., HARRISON, D. S., CATA, J. P., SAPIRE, K., ZHANG, H., KENNAMER-CHAPMAN, R. M., JAWAD, A.

122

B., GHETTI, A., YAN, J., PALECEK, J. & DOUGHERTY, P. M. 2015. The Cancer Chemotherapeutic Paclitaxel Increases Human and Rodent Sensory Neuron Responses to TRPV1 by Activation of TLR4. The Journal of Neuroscience, 35, 13487-13500. LI, Y., NORTH, R. Y., RHINES, L. D., TATSUI, C. E., RAO, G., EDWARDS, D. D., CASSIDY, R. M., HARRISON, D. S., JOHANSSON, C. A. & ZHANG, H. 2017a. DRG Voltage-gated Sodium Channel 1.7 is Up-regulated in Paclitaxel- Induced Neuropathy in Rats and in Humans with Neuropathic Pain. Journal of Neuroscience, 0899-17. LI, Y., TATSUI, C. E., RHINES, L. D., NORTH, R. Y., HARRISON, D. S., CASSIDY, R. M., JOHANSSON, C. A., KOSTURAKIS, A. K., EDWARDS, D. D., ZHANG, H. & DOUGHERTY, P. M. 2017b. Dorsal root ganglion neurons become hyperexcitable and increase expression of voltage-gated T-type calcium channels (Cav3.2) in paclitaxel-induced peripheral neuropathy. Pain, 158, 417-429. LI, Y., ZHANG, H., ZHANG, H., KOSTURAKIS, A. K., JAWAD, A. B. & DOUGHERTY, P. M. 2014. Toll-Like Receptor 4 Signaling Contributes to Paclitaxel-Induced Peripheral Neuropathy. The Journal of Pain, 15, 712-725. LIANG, J. J., HUANG, L. F., CHEN, X. M., PAN, S. Q., LU, Z. N. & XIAO, Z. M. 2015. Amiloride suppresses pilocarpine-induced seizures via ASICs other than NHE in rats. International Journal of Clinical and Experimental Pathology, 8, 14507-14513. LINDHOLM, L. H., ANDERSON, H., EKBOM, T., HANSSON, L., LANKE, J., DAHLÖF, B., DE FAIRE, U., FORSÉN, K., HEDNER, T., LINJER, E., SCHERSTÉN, B., WESTER, P. O. & MÖLLER, T. R. 2001. Relation between drug treatment and cancer In hypertenslves in the Swedish Trial in Old Patients with Hypertension 2: a 5-year, prospective, randomised, controlled trial. The Lancet, 358, 539-544. LINTON, S. J. & SHAW, W. S. 2011. Impact of Psychological Factors in the Experience of Pain. Physical Therapy, 91, 700-711. LIU, C. C., LU, N., CUI, Y., YANG, T., ZHAO, Z. Q., XIN, W. J. & LIU, X. G. 2010. Prevention of Paclitaxel-induced allodynia by Minocycline: Effect on loss of

123

peripheral nerve fibers and infiltration of macrophages in rats. Molecular Pain, 6, 76-76. LIU, M., WILLMOTT, N., MICHAEL, G. & PRIESTLEY, J. 2004. Differential pH and capsaicin responses of Griffonia simplicifolia IB4 (IB4)-positive and IB4- negative small sensory neurons. Neuroscience, 127, 659-672. LONG, A. N., OWENS, K., SCHLAPPAL, A. E., KRISTIAN, T., FISHMAN, P. S. & SCHUH, R. A. 2015. Effect of nicotinamide mononucleotide on brain mitochondrial respiratory deficits in an Alzheimer’s disease-relevant murine model. BMC Neurology, 15, 19. LOPRINZI, C. L., MADDOCKS-CHRISTIANSON, K., WOLF, S. L., RAO, R. D., DYCK, P. J. B., MANTYH, P. & DYCK, P. J. 2007. The Paclitaxel Acute Pain Syndrome: Sensitization of Nociceptors as the Putative Mechanism. The Cancer Journal, 13, 399-403. LOPRINZI, C. L., REEVES, B. N., DAKHIL, S. R., SLOAN, J. A., WOLF, S. L., BURGER, K. N., KAMAL, A., LE-LINDQWISTER, N. A., SOORI, G. S., JASLOWSKI, A. J., NOVOTNY, P. J. & LACHANCE, D. H. 2011. Natural History of Paclitaxel-Associated Acute Pain Syndrome: Prospective Cohort Study NCCTG N08C1. Journal of Clinical Oncology, 29, 1472-1478. LOVEN, D., LEVAVI, H., SABACH, G., ZART, R., ANDRAS, M., FISHMAN, A., KARMON, Y., LEVI, T., DABBY, R. & GADOTH, N. 2009. Long-term glutamate supplementation failed to protect against peripheral neurotoxicity of paclitaxel. European Journal of Cancer Care, 18, 78-83. LUNN, M. P. T., HUGHES, R. A. C. & WIFFEN, P. J. 2014. Duloxetine for treating painful neuropathy, chronic pain or fibromyalgia. Cochrane Database of Systematic Reviews. MAESTRI, A., DE PASQUALE CERATTI, A., CUNDARI, S., ZANNA, C., CORTESI, E. & CRINÒ, L. 2005. A pilot study on the effect of acetyl-L- carnitine in paclitaxel-and cisplatin-induced peripheral neuropathy. Tumori, 91, 135-138. MAIN, C. J. 2013. The importance of psychosocial influences on chronic pain. Pain Management, 3, 455-466. MALGRANGE, B., DELRÉE, P., RIGO, J. M., BARON, H. & MOONEN, G. 1994. Image analysis of neuritic regeneration by adult rat dorsal root ganglion neurons

124

in culture: quantification of the neurotoxicity of anticancer agents and of its prevention by nerve growth factor or basic fibroblast growth factor but not brain-derived neurotrophic factor or neurotrophin-3. Journal of Neuroscience Methods, 53, 111-122. MAMET, J., BARON, A., LAZDUNSKI, M. & VOILLEY, N. 2002. ProInflammatory Mediators, Stimulators of Sensory Neuron Excitability via the Expression of Acid-Sensing Ion Channels. The Journal of Neuroscience, 22, 10662-10670. MANFREDI, J. J., PARNESS, J. & HORWITZ, S. B. 1982. Taxol binds to cellular microtubules. The Journal of Cell Biology, 94, 688-696. MANNELLI, L. D. C., PACINI, A., MATERA, C., ZANARDELLI, M., MELLO, T., DE AMICI, M., DALLANOCE, C. & GHELARDINI, C. 2014. Involvement of α7 nAChR subtype in rat oxaliplatin-induced neuropathy: effects of selective activation. Neuropharmacology, 79, 37-48. MAO-YING, Q. L., KAVELAARS, A., KRUKOWSKI, K., HUO, X. J., ZHOU, W., PRICE, T. J., CLEELAND, C. & HEIJNEN, C. J. 2014. The Anti-Diabetic Drug Metformin Protects against Chemotherapy-Induced Peripheral Neuropathy in a Mouse Model. PLoS One, 9, e100701. MASOCHA, W. 2014. Paclitaxel-induced hyposensitivity to nociceptive chemical stimulation in mice can be prevented by treatment with minocycline. Scientific Reports, 4, 6719. MASOCHA, W. 2015. Astrocyte activation in the anterior cingulate cortex and altered glutamatergic gene expression during paclitaxel-induced neuropathic pain in mice. PeerJ, 3, e1350. MATERAZZI, S., FUSI, C., BENEMEI, S., PEDRETTI, P., PATACCHINI, R., NILIUS, B., PRENEN, J., CREMINON, C., GEPPETTI, P. & NASSINI, R. 2012. TRPA1 and TRPV4 mediate paclitaxel-induced peripheral neuropathy in mice via a glutathione-sensitive mechanism. Pflügers Archiv: European Journal of Physiology, 463, 561-569. MATSUMOTO, M., INOUE, M., HALD, A., XIE, W. & UEDA, H. 2006. Inhibition of paclitaxel-induced A-fiber hypersensitization by gabapentin. Journal of Pharmacology and Experimental Therapeutics, 318, 735-740. MATSUMURA, Y., YOKOYAMA, Y., HIRAKAWA, H., SHIGETO, T., FUTAGAMI, M. & MIZUNUMA, H. 2014. The prophylactic effects of a

125

traditional Japanese medicine, goshajinkigan, on paclitaxel-induced peripheral neuropathy and its mechanism of action. Molecular Pain, 10, 61. MATSUNAGA, N., TSUCHIMORI, N., MATSUMOTO, T. & II, M. 2011. TAK-242 (Resatorvid), a Small-Molecule Inhibitor of Toll-Like Receptor (TLR)4 Signaling, Binds Selectively to TLR4 and Interferes with Interactions between TLR4 and Its Adaptor Molecules. Molecular Pharmacology, 79, 34-41. MCCORMICK, B., LOWES, D. A., COLVIN, L., TORSNEY, C. & GALLEY, H. F. 2016. MitoVitE, a mitochondria-targeted antioxidant, limits paclitaxel-induced oxidative stress and mitochondrial damage in vitro, and paclitaxel-induced mechanical hypersensitivity in a rat pain model. British Journal of Anaesthesia, 117, 659-666. MCDONALD, E. S. & WINDEBANK, A. J. 2002. Cisplatin-Induced Apoptosis of DRG Neurons Involves Bax Redistribution and Cytochrome c Release But Not fas Receptor Signaling. Neurobiology of Disease, 9, 220-233. MEDICINES.ORG.UK. 2009. Amiloride 5 mg Tablets BP [Online]. emc: Accord Healthcare Limited. Available: https://www.medicines.org.uk/emc/product/6020/smpc [Accessed 18 Jul 2018]. MELLI, G., JACK, C., LAMBRINOS, G. L., RINGKAMP, M. & HÖKE, A. 2006. Erythropoietin protects sensory axons against paclitaxel-induced distal degeneration. Neurobiology of Disease, 24, 525-530. MELLI, G., TAIANA, M., CAMOZZI, F., TRIOLO, D., PODINI, P., QUATTRINI, A., TARONI, F. & LAURIA, G. 2008. Alpha-lipoic acid prevents mitochondrial damage and neurotoxicity in experimental chemotherapy neuropathy. Experimental Neurology, 214, 276-284. MENG, X., ZHANG, Y., LI, A., XIN, J., LAO, L., REN, K., BERMAN, B. M., TAN, M. & ZHANG, R. X. 2011. The effects of opioid receptor antagonists on electroacupuncture-produced anti-allodynia/hyperalgesia in rats with paclitaxel- evoked peripheral neuropathy. Brain Research, 1414, 58-65. MILLS, K. F., YOSHIDA, S., STEIN, L. R., GROZIO, A., KUBOTA, S., SASAKI, Y., REDPATH, P., MIGAUD, M. E., APTE, R. S., UCHIDA, K., YOSHINO, J. & IMAI, S. I. 2016. Long-Term Administration of Nicotinamide Mononucleotide Mitigates Age-Associated Physiological Decline in Mice. Cell Metabolism, 24, 795-806.

126

MIZUNO, T., KUROTANI, T., KOMATSU, Y., KAWANOKUCHI, J., KATO, H., MITSUMA, N. & SUZUMURA, A. 2004. Neuroprotective role of phosphodiesterase inhibitor ibudilast on neuronal cell death induced by activated microglia. Neuropharmacology, 46, 404-411. MO, M., ERDELYI, I., SZIGETI-BUCK, K., BENBOW, J. H. & EHRLICH, B. E. 2012. Prevention of paclitaxel-induced peripheral neuropathy by lithium pretreatment. The FASEB Journal, 26, 4696-4709. MOGIL, J. S., BREESE, N. M., WITTY, M. F., RITCHIE, J., RAINVILLE, M. L., ASE, A., ABBADI, N., STUCKY, C. L. & SÉGUÉLA, P. 2005. Transgenic Expression of a Dominant-Negative ASIC3 Subunit Leads to Increased Sensitivity to Mechanical and Inflammatory Stimuli. The Journal of Neuroscience, 25, 9893-9901. MOLS, F., BEIJERS, T., VREUGDENHIL, G. & VAN DE POLL-FRANSE, L. 2014. Chemotherapy-induced peripheral neuropathy and its association with quality of life: a systematic review. Supportive Care in Cancer, 22, 2261-2269. MORARI, M., BRUGNOLI, A., PISANÒ, C. A., NOVELLO, S., CACCIA, C., MELLONI, E., PADOANI, G., VAILATI, S. & SARDINA, M. 2018. Safinamide Differentially Modulates In Vivo Glutamate and GABA Release in the Rat Hippocampus and Basal Ganglia. Journal of Pharmacology and Experimental Therapeutics, 364, 198-206. MUTHURAMAN, A., JAGGI, A. S., SINGH, N. & SINGH, D. 2008. Ameliorative effects of amiloride and pralidoxime in chronic constriction injury and vincristine induced painful neuropathy in rats. European Journal of Pharmacology, 587, 104-111. N'GOUEMO, P. 2008. Amiloride delays the onset of pilocarpine-induced seizures in rats. Brain Research, 1222, 230-232. NAIR, A. B. & JACOB, S. 2016. A simple practice guide for dose conversion between animals and human. Journal of Basic and Clinical Pharmacy, 7, 27. NAJI-ESFAHANI, H., VASEGHI, G., SAFAEIAN, L., PILEHVARIAN, A. A., ABED, A. & RAFIEIAN-KOPAEI, M. 2016. Gender differences in a mouse model of chemotherapy-induced neuropathic pain. Laboratory Animals, 50, 15- 20.

127

NAKAGAWA, S., OKAMOTO, Y., TSUNETO, S., TANIMUKAI, H., GOYA, S., MATSUDA, Y., OONO, Y., TSUGANE, M. & UEJIMA, E. 2011. Can milnacipran used for neuropathic pain in patients with advanced cancer cause neuromuscular and somatosensory disorders? Journal of Palliative Medicine, 14, 403-405. NI, J. Q., OUYANG, Q., LIN, L., HUANG, Z., LU, H., CHEN, X., LIN, H., WANG, Z., XU, D. & ZHANG, Y. 2013. Role of toll-like receptor 4 on lupus lung injury and atherosclerosis in LPS-challenge ApoE(-)/(-) mice. Clinical & Developmental Immunology, 2013, 476856. NIETO, F. R., CENDÁN, C. M., CAÑIZARES, F. J., CUBERO, M. A., VELA, J. M., FERNÁNDEZ-SEGURA, E. & BAEYENS, J. M. 2014. Genetic inactivation and pharmacological blockade of sigma-1 receptors prevent paclitaxel-induced sensory-nerve mitochondrial abnormalities and neuropathic pain in mice. Molecular Pain, 10, 1-12. NIETO, F. R., ENTRENA, J. M., CENDÁN, C. M., POZO, E. D., VELA, J. M. & BAEYENS, J. M. 2008. Tetrodotoxin inhibits the development and expression of neuropathic pain induced by paclitaxel in mice. Pain, 137, 520-531. NIH. 2015. A Story of Discovery: Natural Compound Helps Treat Breast and Ovarian Cancers Available: http://www.cancer.gov/research/progress/discovery/taxol. NIROGI, R., GOURA, V., SHANMUGANATHAN, D., JAYARAJAN, P. & ABRAHAM, R. 2012. Comparison of manual and automated filaments for evaluation of neuropathic pain behavior in rats. Journal of Pharmacological and Toxicological Methods, 66, 8-13. NISHIDA, K., KUCHIIWA, S., OISO, S., FUTAGAWA, T., MASUDA, S., TAKEDA, Y. & YAMADA, K. 2008. Up-regulation of matrix metalloproteinase-3 in the dorsal root ganglion of rats with paclitaxel-induced neuropathy. Cancer Science, 99, 1618-1625. O'SHAUGHNESSY, J. 2016. Breast Cancer (Invasive) Treatment Regimens. Available: http://www.cancertherapyadvisor.com/breast-cancer/breast-cancer-invasive- treatment-regimens/article/218154/. OTAKE, A., YOSHINO, K., UEDA, Y., SAWADA, K., MABUCHI, S., KIMURA, T., KOBAYASHI, E., ISOBE, A., EGAWA-TAKATA, T., MATSUZAKI, S., FUJITA, M. & KIMURA, T. 2015. Usefulness of Duloxetine for Paclitaxel-

128

Induced Peripheral Neuropathy Treatment in Gynecological Cancer Patients. Anticancer Research, 35, 359-363. PACHMAN, D. R., BARTON, D. L., WATSON, J. C. & LOPRINZI, C. L. 2011. Chemotherapy-Induced Peripheral Neuropathy: Prevention and Treatment. Clinical Pharmacology & Therapeutics, 90, 377-387. PACHMAN, D. R., DOCKTER, T. J., ZEKAN, P. J., FRUTH, B., RUDDY, K. J., TA, L. E., DENTCHEV, T., LE-LINDQWISTER, N., SIKOV, W. M. & LOPRINZI, C. L. 2016. Pilot study of minocycline for the prevention of paclitaxel- associated neuropathy: ACCRU RU221408I. Journal of Clinical Oncology, 34, 220-220. PANI, G., DE VOS, W. H., SAMARI, N., DE SAINT-GEORGES, L., BAATOUT, S., VAN OOSTVELDT, P. & BENOTMANE, M. A. 2014. MorphoNeuroNet: An automated method for dense neurite network analysis. Cytometry Part A, 85, 188-199. PARK, J. H., LONG, A., OWENS, K. & KRISTIAN, T. 2016. Nicotinamide mononucleotide inhibits post-ischemic NAD+ degradation and dramatically ameliorates brain damage following global cerebral ischemia. Neurobiology of Disease, 95, 102-110. PARK, J. S., KIM, S. & HOKE, A. 2015. An exercise regimen prevents development paclitaxel induced peripheral neuropathy in a mouse model. Journal of the Peripheral Nervous System, 20, 7-14. PARK, S. B., LIN, C. S. Y., KRISHNAN, A. V., FRIEDLANDER, M. L., LEWIS, C. R. & KIERNAN, M. C. 2011. Early, progressive, and sustained dysfunction of sensory axons underlies paclitaxel-induced neuropathy. Muscle and Nerve, 43, 367-374. PARVATHY, S. S. & MASOCHA, W. 2015. Coadministration of indomethacin and minocycline attenuates established paclitaxel-induced neuropathic thermal hyperalgesia: Involvement of cannabinoid CB(1) receptors. Scientific Reports, 5, 10541. PASCUAL, D., GOICOECHEA, C., BURGOS, E. & MARTÍN, M. I. 2010. Antinociceptive effect of three common analgesic drugs on peripheral neuropathy induced by paclitaxel in rats. Pharmacology Biochemistry and Behavior, 95, 331-337.

129

PDR.NET. 2018. Paclitaxel - Drug Summary [Online]. US: Physicians' Desk Reference. Available: http://www.pdr.net/drug-summary/Paclitaxel-paclitaxel-1299.4045 [Accessed 11 Jan 2018]. PEASE-RAISSI, S. E., PAZYRA-MURPHY, M. F., LI, Y., WACHTER, F., FUKUDA, Y., FENSTERMACHER, S. J., BARCLAY, L. A., BIRD, G. H., WALENSKY, L. D. & SEGAL, R. A. 2017. Paclitaxel Reduces Axonal Bclw to Initiate IP 3 R1-Dependent Axon Degeneration. Neuron, 96, 373-386. e6. PETERS, C. M., JIMENEZ-ANDRADE, J. M., JONAS, B. M., SEVCIK, M. A., KOEWLER, N. J., GHILARDI, J. R., WONG, G. Y. & MANTYH, P. W. 2007a. Intravenous paclitaxel administration in the rat induces a peripheral sensory neuropathy characterized by macrophage infiltration and injury to sensory neurons and their supporting cells. Experimental Neurology, 203, 42-54. PETERS, C. M., JIMENEZ-ANDRADE, J. M., KUSKOWSKI, M. A., GHILARDI, J. R. & MANTYH, P. W. 2007b. An evolving cellular pathology occurs in dorsal root ganglia, peripheral nerve and spinal cord following intravenous administration of paclitaxel in the rat. Brain Research, 1168, 46-59. PEVIDA, M., LASTRA, A., HIDALGO, A., BAAMONDE, A. & MENÉNDEZ, L. 2013. Spinal CCL2 and microglial activation are involved in paclitaxel-evoked cold hyperalgesia. Brain Research Bulletin, 95, 21-27. PICCIOTTO, N. E., GANO, L. B., JOHNSON, L. C., MARTENS, C. R., SINDLER, A. L., MILLS, K. F., IMAI, S. I. & SEALS, D. R. 2016. Nicotinamide mononucleotide supplementation reverses vascular dysfunction and oxidative stress with aging in mice. Aging Cell, 15, 522-530. PITTMAN, S. K., GRACIAS, N. G. & FEHRENBACHER, J. C. 2016. Nerve growth factor alters microtubule targeting agent-induced neurotransmitter release but not MTA-induced neurite retraction in sensory neurons. Experimental Neurology, 279, 104-115. PIZZORNO, J. E., MURRAY, M. T. & JOINER-BEY, H. 2008. Cancer: Integrated Naturopathic Support. The Clinician's Handbook of Natural Medicine. Churchill Livingstone/Elsevier. POLOMANO, R. C., MANNES, A. J., CLARK, U. S. & BENNETT, G. J. 2001. A painful peripheral neuropathy in the rat produced by the chemotherapeutic drug, paclitaxel. Pain, 94, 293-304.

130

PRICE, M. P., MCILWRATH, S. L., XIE, J., CHENG, C., QIAO, J., TARR, D. E., SLUKA, K. A., BRENNAN, T. J., LEWIN, G. R. & WELSH, M. J. 2001. The DRASIC Cation Channel Contributes to the Detection of Cutaneous Touch and Acid Stimuli in Mice. Neuron, 32, 1071-1083. PUBCHEM 2005. Paclitaxel; CID=36314. 26 March ed. PubChem Compound Database: National Center for Biotechnology Information. REEVES, B., DAKHIL, S., SLOAN, J., BURGER, K., LE-LINDQWISTER, N., SOORI, G., JASLOWSKI, A., KELAGHAN, J., LACHANCE, D. & LOPRINZI, C. Paclitaxel-associated acute pain syndrome (P-APS) and its association on the development of peripheral neuropathy: NCCTG trial N08C1. ASCO Annual Meeting Proceedings, 2011. 9047. RESHKIN, S. J., BELLIZZI, A., CARDONE, R. A., TOMMASINO, M., CASAVOLA, V. & PARADISO, A. 2003. Paclitaxel Induces Apoptosis via Protein Kinase A- and p38 Mitogen-activated Protein-dependent Inhibition of the Na+/H+ Exchanger (NHE) NHE Isoform 1 in Human Breast Cancer Cells. Clinical Cancer Research, 9, 2366-2373. REYES-GIBBY, C. C., MORROW, P. K., BUZDAR, A. & SHETE, S. 2009. Chemotherapy-Induced Peripheral Neuropathy as a Predictor of Neuropathic Pain in Breast Cancer Patients Previously Treated With Paclitaxel. The Journal of Pain, 10, 1146-1150. RICE, T. W., WHEELER, A. P., BERNARD, G. R., VINCENT, J.-L., ANGUS, D. C., AIKAWA, N., DEMEYER, I., SAINATI, S., AMLOT, N., CAO, C., II, M., MATSUDA, H., MOURI, K. & COHEN, J. 2010. A randomized, double-blind, placebo-controlled trial of TAK-242 for the treatment of severe sepsis. Critical Care Medicine, 38, 1685-1694. ROBERTS, S. C. 2007. Production and engineering of terpenoids in plant cell culture. Nature Chemical Biology, 3, 387-395. ROBINSON, C. R. & DOUGHERTY, P. M. 2015. Spinal astrocyte gap junction and glutamate transporter expression contributes to a rat model of bortezomib- induced peripheral neuropathy. Neuroscience, 285, 1-10. ROCHA-GONZÁLEZ, H. I., HERREJON-ABREU, E. B., LÓPEZ-SANTILLÁN, F. J., GARCÍA-LÓPEZ, B. E., MURBARTIÁN, J. & GRANADOS-SOTO, V. 2009.

131

Acid increases inflammatory pain in rats: Effect of local peripheral ASICs inhibitors. European Journal of Pharmacology, 603, 56-61. ROJAS, E. A., CORCHETE, L. A., SAN-SEGUNDO, L., MARTÍNEZ-BLANCH, J. F., CODOÑER, F. M., PAÍNO, T., PUIG, N., GARCÍA-SANZ, R., MATEOS, M. V., OCIO, E. M., MISIEWICZ-KRZEMINSKA, I. & GUTIÉRREZ, N. C. 2017. Amiloride, An Old Diuretic Drug, Is a Potential Therapeutic Agent for Multiple Myeloma. Clinical Cancer Research, 23, 6602-6615. ROSE, K. E., LUNARDI, N., BOSCOLO, A., DONG, X., ERISIR, A., JEVTOVIC- TODOROVIC, V. & TODOROVIC, S. M. 2013. Immunohistological demonstration of CaV3.2 T-type voltage-gated calcium channel expression in soma of dorsal root ganglion neurons and peripheral axons of rat and mouse. Neuroscience, 250, 263-274. ROSTOCK, M., JAROSLAWSKI, K., GUETHLIN, C., LUDTKE, R., SCHRÖDER, S. & BARTSCH, H. H. 2013. Chemotherapy-Induced Peripheral Neuropathy in Cancer Patients: A Four-Arm Randomized Trial on the Effectiveness of Electroacupuncture. Evidence-based Complementary and Alternative Medicine: eCAM, 2013, 349653. ROWINSKY, E., EISENHAUER, E., CHAUDHRY, V., ARBUCK, S. & DONEHOWER, R. Clinical toxicities encountered with paclitaxel (Taxol). Seminars in oncology, 1993. 1-15. RUIZ-MEDINA, J., BAULIES, A., BURA, S. A. & VALVERDE, O. 2013. Paclitaxel- induced neuropathic pain is age dependent and devolves on glial response. European Journal of Pain, 17. RUIZ-PÉREZ, D., BENITO, J., POLO, G., LARGO, C., AGUADO, D., SANZ, L. & DE SEGURA, I. A. G. 2016. The Effects of the Toll-Like Receptor 4 Antagonist, Ibudilast, on Sevoflurane’s Minimum Alveolar Concentration and the Delayed Remifentanil-Induced Increase in the Minimum Alveolar Concentration in Rats. Anesthesia & Analgesia, 122, 1370-1376. RUSCHEWEYH, R., FORSTHUBER, L., SCHOFFNEGGER, D. & SANDKÜHLER, J. 2007. Modification of classical neurochemical markers in identified primary afferent neurons with Aβ-, Aδ-, and C-fibers after chronic constriction injury in mice. The Journal of Comparative Neurology, 502, 325-336.

132

SAMOUR, M. S., NAGI, S. S. & MAHNS, D. A. 2015. Cav3.2-expressing low- threshold C fibres in human hairy skin contribute to cold allodynia—a non- TRPV1- and non-TRPM8-dependent phenomenon. PAIN, 156, 1566-1575. SATO, J., MORI, M., NIHEI, S., KUMAGAI, M., TAKEUCHI, S., KASHIWABA, M. & KUDO, K. 2016. The effectiveness of regional cooling for paclitaxel-induced peripheral neuropathy. Journal of Pharmaceutical Health Care and Sciences, 2, 33. SCHELLER, B., HEHRLEIN, C., BOCKSCH, W., RUTSCH, W., HAGHI, D., DIETZ, U., BÖHM, M. & SPECK, U. 2006. Treatment of coronary in-stent restenosis with a paclitaxel-coated balloon catheter. New England Journal of Medicine, 355, 2113-2124. SCHIFF, P. B., FANT, J. & HORWITZ, S. B. 1979. Promotion of microtubule assembly in vitro by taxol. Nature, 277, 665-7. SCHIFF, P. B. & HORWITZ, S. B. 1980. Taxol stabilizes microtubules in mouse fibroblast cells. Proceedings of the National Academy of Sciences, 77, 1561- 1565. SEIDMAN, A. D., BERRY, D., CIRRINCIONE, C., HARRIS, L., MUSS, H., MARCOM, P. K., GIPSON, G., BURSTEIN, H., LAKE, D. & SHAPIRO, C. L. 2008. Randomized phase III trial of weekly compared with every-3-weeks paclitaxel for metastatic breast cancer, with trastuzumab for all HER-2 overexpressors and random assignment to trastuzumab or not in HER-2 nonoverexpressors: final results of Cancer and Leukemia Group B protocol 9840. Journal of Clinical Oncology, 26, 1642-1649. SENGOTTUVEL, V., LEIBINGER, M., PFREIMER, M., ANDREADAKI, A. & FISCHER, D. 2011. Taxol Facilitates Axon Regeneration in the Mature CNS. The Journal of Neuroscience, 31, 2688-2699. SERETNY, M., CURRIE, G. L., SENA, E. S., RAMNARINE, S., GRANT, R., MACLEOD, M. R., COLVIN, L. A. & FALLON, M. 2014. Incidence, prevalence, and predictors of chemotherapy-induced peripheral neuropathy: A systematic review and meta-analysis. Pain, 155, 2461-2470. SERRA, J., SOLÀ, R., QUILES, C., CASANOVA-MOLLA, J., PASCUAL, V., BOSTOCK, H. & VALLS-SOLÉ, J. 2009. C-nociceptors sensitized to cold in a patient with small-fiber neuropathy and cold allodynia. Pain, 147, 46-53.

133

SHA, T., IIZAWA, Y. & II, M. 2011. Combination of imipenem and TAK-242, a Toll- like receptor 4 signal transduction inhibitor, improves survival in a murine model of polymicrobial sepsis. Shock, 35, 205-9. SHA, T., SUNAMOTO, M., KITAZAKI, T., SATO, J., II, M. & IIZAWA, Y. 2007. Therapeutic effects of TAK-242, a novel selective Toll-like receptor 4 signal transduction inhibitor, in mouse endotoxin shock model. European Journal of Pharmacology, 571, 231-239. SHINDE, S. S., SEISLER, D., SOORI, G., ATHERTON, P. J., PACHMAN, D. R., LAFKY, J., RUDDY, K. J. & LOPRINZI, C. L. 2016. Can pregabalin prevent paclitaxel-associated neuropathy?—An ACCRU pilot trial. Supportive Care in Cancer, 24, 547-553. SIAU, C. & BENNETT, G. J. 2006. Dysregulation of Cellular Calcium Homeostasis in Chemotherapy-Evoked Painful Peripheral Neuropathy. Anesthesia and Analgesia, 102, 1485-1490. SIAU, C., XIAO, W. & BENNETT, G. J. 2006. Paclitaxel- and vincristine-evoked painful peripheral neuropathies: Loss of epidermal innervation and activation of Langerhans cells. Experimental Neurology, 201, 507-514. SISIGNANO, M., BARON, R., SCHOLICH, K. & GEISSLINGER, G. 2014. Mechanism-based treatment for chemotherapy-induced peripheral neuropathic pain. Nature Reviews Neurology, 10, 694-707. SKINNER, J. R., WADDELL-SMITH, K. & ANDERSON, B. J. 2013. To the Editor— Amiloride concentrations in clinical practice. Heart Rhythm, 10, e82-e83. SLUKA, K. A., PRICE, M. P., BREESE, N. M., STUCKY, C. L., WEMMIE, J. A. & WELSH, M. J. 2003. Chronic hyperalgesia induced by repeated acid injections in muscle is abolished by the loss of ASIC3, but not ASIC1. Pain, 106, 229-239. SMITH, E. M. L., PANG, H., CIRRINCIONE, C., FLEISHMAN, S., PASKETT, E. D., AHLES, T., BRESSLER, L. R., FADUL, C. E., KNOX, C. & LE- LINDQWISTER, N. 2013. Effect of duloxetine on pain, function, and quality of life among patients with chemotherapy-induced painful peripheral neuropathy: a randomized clinical trial. JAMA, 309, 1359-1367. SMITH, R. E., BROWN, A. M., MAMOUNAS, E. P., ANDERSON, S. J., LEMBERSKY, B. C., ATKINS, J. H., SHIBATA, H. R., BAEZ, L., DEFUSCO, P. A., DAVILA, E., TIPPING, S. J., BEARDEN, J. D. & THIRLWELL, M. P.

134

1999. Randomized Trial of 3-Hour Versus 24-Hour Infusion of High-Dose Paclitaxel in Patients With Metastatic or Locally Advanced Breast Cancer: National Surgical Adjuvant Breast and Bowel Project Protocol B-26. Journal of Clinical Oncology, 17, 3403-3411. SMITH, S. B., CRAGER, S. E. & MOGIL, J. S. 2004. Paclitaxel-induced neuropathic hypersensitivity in mice: Responses in 10 inbred mouse strains. Life Sciences, 74, 2593-2604. SOUNESS, J. E., VILLAMIL, M. E., SCOTT, L. C., TOMKINSON, A., GIEMBYCZ, M. A. & RAEBURN, D. 1994. Possible role of cyclic AMP phosphodiesterases in the actions of ibudilast on eosinophil thromboxane generation and airways smooth muscle tone. British Journal of Pharmacology, 111, 1081-1088. SPARKS, R., TSO, P. & ARNOLD, E. 1982. Effects of Low Calcium, High Calcium, and Amiloride on Hematopoietic Stem-Cell Proliferation In Vitro. Journal of Cell Biology, 95, 57. SPARKS, R. L., POOL, T. B., SMITH, N. K. & CAMERON, I. L. 1983. Effects of amiloride on tumor growth and intracellular element content of tumor cells in vivo. Cancer Research, 43, 73-77. SUNDAR, R., BANDLA, A., TAN, S. S. H., LIAO, L.-D., KUMARAKULASINGHE, N. B., JEYASEKHARAN, A. D., OW, S. G. W., HO, J., TAN, D. S. P., LIM, J. S. J., VIJAYAN, J., THERIMADASAMY, A. K., HAIROM, Z., ANG, E., ANG, S., THAKOR, N. V., LEE, S.-C. & WILDER-SMITH, E. P. V. 2016. Limb Hypothermia for Preventing Paclitaxel-Induced Peripheral Neuropathy in Breast Cancer Patients: A Pilot Study. Frontiers in Oncology, 6, 274. SUNKARA, P. S. 2017. Novel approaches to cancer chemotherapy, Elsevier. SUZUKI, H., SHINOHARA, A., OHNO, I., MITSUNAGA, S., TAKENO, M. K., FUNAZAKI, H., KIMURA, G., WATANABE, K., UMEMOTO, K., SASAKI, M., TAKAHASHI, H., IMAOKA, H., HASHIMOTO, Y., IKEDA, M. & YAMAGUCHI, M. 2017. The effect of duloxetine on chemotherapy-induced peripheral neuropathy in advanced pancreatic cancer patients receiving gemcitabine plus nab-paclitaxel treatment. Journal of Clinical Oncology, 35, 453.

135

SUZUMURA, A., ITO, A., YOSHIKAWA, M. & SAWADA, M. 1999. Ibudilast suppresses TNFα production by glial cells functioning mainly as type III phosphodiesterase inhibitor in the CNS. Brain Research, 837, 203-212. TAI, K. K. & TRUONG, D. D. 2013. Amiloride but not memantine reduces neurodegeneration, seizures and myoclonic jerks in rats with cardiac arrest- induced global cerebral hypoxia and reperfusion. PloS ONE, 8, e60309. TAKASHIMA, K., MATSUNAGA, N., YOSHIMATSU, M., HAZEKI, K., KAISHO, T., UEKATA, M., HAZEKI, O., AKIRA, S., IIZAWA, Y. & II, M. 2009. Analysis of binding site for the novel small-molecule TLR4 signal transduction inhibitor TAK-242 and its therapeutic effect on mouse sepsis model. British Journal of Pharmacology, 157, 1250-1262. TANABE, Y., SHIMIZU, C., HAMADA, A., HASHIMOTO, K., IKEDA, K., NISHIZAWA, D., HASEGAWA, J., SHIMOMURA, A., OZAKI, Y., TAMURA, N., YAMAMOTO, H., YUNOKAWA, M., YONEMORI, K., TAKANO, T., KAWABATA, H., TAMURA, K. & FUJIWARA, Y. 2017. Paclitaxel-induced sensory peripheral neuropathy is associated with an ABCB1 single nucleotide polymorphism and older age in Japanese. Cancer Chemotherapy and Pharmacology, 79, 1179-1186. TASNIM, A., RAMMELKAMP, Z., SLUSHER, A. B., WOZNIAK, K., SLUSHER, B. S. & FARAH, M. H. 2016. Paclitaxel causes degeneration of both central and peripheral axon branches of dorsal root ganglia in mice. BMC Neuroscience, 17, 47. TATSUTA, M., IISHI, H., BABA, M., UEHARA, H. & NAKAIZUMI, A. 1995. Chemoprevention by amiloride of experimental carcinogenesis in rat colon induced by azoxymethane. Carcinogenesis, 16, 941-2. TATSUTA, M., IISHI, H., BABA, M., UEHARA, H., NAKAIZUMI, A. & TANIGUCHI, H. 1993. Inhibition by amiloride of gastric carcinogenesis induced by N-methyl-N'-nitro-N-nitrosoguanidine in Wistar rats. British Journal of Cancer, 67, 1011-4. TEIWES, J. & TOTO, R. D. 2007. Epithelial Sodium Channel Inhibition in Cardiovascular Disease* A Potential Role for Amiloride. Oxford University Press.

136

TOH, M.-L., AEBERLI, D., LACEY, D., YANG, Y., SANTOS, L. L., CLARKSON, M., SHARMA, L., CLYNE, C. & MORAND, E. F. 2006. Regulation of IL-1 and TNF Receptor Expression and Function by Endogenous Macrophage Migration Inhibitory Factor. The Journal of Immunology, 177, 4818-4825. TOTH, C., LANDER, J. & WIEBE, S. 2009. The prevalence and impact of chronic pain with neuropathic pain symptoms in the general population. Pain Medicine, 10, 918-29. TSANTOULAS, C., ZHU, L., YIP, P., GRIST, J., MICHAEL, G. J. & MCMAHON, S. B. 2014. Kv2 dysfunction after peripheral axotomy enhances sensory neuron responsiveness to sustained input. Experimental Neurology, 251, 115-126. TURRENS, J. F., ALEXANDRE, A. & LEHNINGER, A. L. 1985. Ubisemiquinone is the electron donor for superoxide formation by complex III of heart mitochondria. Archives of Biochemistry and Biophysics, 237, 408-414. TURRENS, J. F. & BOVERIS, A. 1980. Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria. Biochemical Journal, 191, 421- 427. UDDIN, G. M., YOUNGSON, N. A., SINCLAIR, D. A. & MORRIS, M. J. 2016. Head to Head Comparison of Short-Term Treatment with the NAD(+) Precursor Nicotinamide Mononucleotide (NMN) and 6 Weeks of Exercise in Obese Female Mice. Frontiers in Pharmacology, 7, 258. UGAWA, S., UEDA, T., ISHIDA, Y., NISHIGAKI, M., SHIBATA, Y. & SHIMADA, S. 2002. Amiloride-blockable acid-sensing ion channels are leading acid sensors expressed in human nociceptors. The Journal of Clinical Investigation, 110, 1185-1190. USTINOVA, E. E., SHURIN, G. V., GUTKIN, D. W. & SHURIN, M. R. 2013. The Role of TLR4 in the Paclitaxel Effects on Neuronal Growth In Vitro. PLoS ONE, 8, e56886. VAN GERVEN, J. M. A., MOLL, J. W. B., VAN DEN BENT, M. J., BONTENBAL, M., VAN DER BURG, M. E. L., VERWEIJ, J. & VECHT, C. J. 1994. Paclitaxel (Taxol) induces cumulative mild neurotoxicity. European Journal of Cancer, 30, 1074-1077.

137

VAN HECKE, O., AUSTIN, S. K., KHAN, R. A., SMITH, B. & TORRANCE, N. 2014. Neuropathic pain in the general population: a systematic review of epidemiological studies. Pain, 155, 654-662. VELASCO, R., BRUNA, J., BRIANI, C., ARGYRIOU, A. A., CAVALETTI, G., ALBERTI, P., FRIGENI, B., CACCIAVILLANI, M., LONARDI, S. & CORTINOVIS, D. 2014. Early predictors of oxaliplatin-induced cumulative neuropathy in colorectal cancer patients. Journal of Neurology, Neurosurgery, and Psychiatry, 85, 392-398. VELASCO, R., NAVARRO, X., GIL-GIL, M., HERRANDO-GRABULOSA, M., CALLS, A. & BRUNA, J. 2017. Neuropathic Pain and Nerve Growth Factor in Chemotherapy-Induced Peripheral Neuropathy: Prospective Clinical- Pathological Study. Journal of Pain and Symptom Management, 54, 815-825. VERGO, S., CRANER, M. J., ETZENSPERGER, R., ATTFIELD, K., FRIESE, M. A., NEWCOMBE, J., ESIRI, M. & FUGGER, L. 2011. Acid-sensing ion channel 1 is involved in both axonal injury and demyelination in multiple sclerosis and its animal model. Brain, 134, 571-84. VIDT, D. G. 1981. Mechanism of action, pharmacokinetics, adverse effects, and therapeutic uses of amiloride hydrochloride, a new potassium-sparing diuretic. Pharmacotherapy, 1, 179-87. VITARELLA, D., KIMELBERG, H. K. & ASCHNER, M. 1996. Inhibition of regulatory volume decrease in swollen rat primary astrocyte cultures by methylmercury is due to increased amiloride-sensitive Na+ uptake. Brain Research, 732, 169-178. VOILLEY, N., DE WEILLE, J., MAMET, J. & LAZDUNSKI, M. 2001. Nonsteroid Anti-Inflammatory Drugs Inhibit Both the Activity and the Inflammation- Induced Expression of Acid-Sensing Ion Channels in Nociceptors. The Journal of Neuroscience, 21, 8026-8033. VRINTEN, D. H. & HAMERS, F. F. T. 2003. ‘CatWalk’ automated quantitative gait analysis as a novel method to assess mechanical allodynia in the rat; a comparison with von Frey testing. Pain, 102, 203-209. WALDMANN, R., BASSILANA, F., DE WEILLE, J., CHAMPIGNY, G., HEURTEAUX, C. & LAZDUNSKI, M. 1997. Molecular cloning of a non-

138

inactivating proton-gated Na+ channel specific for sensory neurons. Journal of Biological Chemistry, 272, 20975-20978. WANG, J., LI, Q., XU, B., ZHANG, T., CHEN, S. & LUO, Y. 2017. Efficacy and safety of duloxetine in Chinese breast cancer patients with paclitaxel-induced peripheral neuropathy. Chinese Journal of Cancer Research, 29, 411-418. WANG, J. T., MEDRESS, Z. A., VARGAS, M. E. & BARRES, B. A. 2015. Local axonal protection by WldS as revealed by conditional regulation of protein stability. Proceedings of the National Academy of Sciences, 112, 10093-10100. WANG, M. S., DAVIS, A. A., CULVER, D. G. & GLASS, J. D. 2002. Wlds mice are resistant to paclitaxel (taxol) neuropathy. Annals of Neurology, 52, 442-447. WANG, M. S., DAVIS, A. A., CULVER, D. G., WANG, Q., POWERS, J. C. & GLASS, J. D. 2004. Calpain inhibition protects against Taxol‐induced sensory neuropathy. Brain, 127, 671-679. WANG, Y., WILSON, S. M., BRITTAIN, J. M., RIPSCH, M. S., SALOMÉ, C., PARK, K. D., WHITE, F. A., KHANNA, R. & KOHN, H. 2011. Merging structural motifs of functionalized amino acids and α-aminoamides results in novel anticonvulsant compounds with significant effects on slow and fast inactivation of voltage-gated sodium channels and in the treatment of neuropathic pain. ACS Chemical Neuroscience, 2, 317-332. WANI, M. C., TAYLOR, H. L., WALL, M. E., COGGON, P. & MCPHAIL, A. T. 1971. Plant antitumor agents. VI. Isolation and structure of taxol, a novel antileukemic and antitumor agent from Taxus brevifolia. Journal of the American Chemical Society, 93, 2325-2327. WARWICK, R. & HANANI, M. 2013. The contribution of satellite glial cells to chemotherapy‐induced neuropathic pain. European Journal of Pain, 17, 571- 580. WEAVER, B. A. 2014. How Taxol/paclitaxel kills cancer cells. Molecular Biology of the Cell, 25, 2677-2681. WEMMIE, J. A., TAUGHER, R. J. & KREPLE, C. J. 2013. Acid-sensing ion channels in pain and disease. Nature Reviews Neuroscience, 14, 461. WISHART, D. S., FEUNANG, Y. D., GUO, A. C., LO, E. J., MARCU, A., GRANT, J. R., SAJED, T., JOHNSON, D., LI, C., SAYEEDA, Z., ASSEMPOUR, N., IYNKKARAN, I., LIU, Y., MACIEJEWSKI, A., GALE, N., WILSON, A.,

139

CHIN, L., CUMMINGS, R., LE, D., PON, A., KNOX, C. & WILSON, M. 2018. DrugBank 5.0: a major update to the DrugBank database for 2018. Nucleic Acids Research, 46, D1074-d1082. WOHLREICH, M. M., MALLINCKRODT, C. H., PRAKASH, A., WATKIN, J. G. & CARTER, W. P. 2007. Duloxetine for the treatment of major depressive disorder: safety and tolerability associated with dose escalation. Depression and Anxiety, 24, 41-52. WOZNIAK, K. M., NOMOTO, K., LAPIDUS, R. G., WU, Y., CAROZZI, V., CAVALETTI, G., HAYAKAWA, K., HOSOKAWA, S., TOWLE, M. J., LITTLEFIELD, B. A. & SLUSHER, B. S. 2011. Comparison of neuropathy- inducing effects of eribulin mesylate, paclitaxel, and ixabepilone in mice. Cancer Research, 71, 3952-62. WOZNIAK, K. M., VORNOV, J. J., WU, Y., LIU, Y., CAROZZI, V. A., RODRIGUEZ-MENENDEZ, V., BALLARINI, E., ALBERTI, P., POZZI, E., SEMPERBONI, S., COOK, B. M., LITTLEFIELD, B. A., NOMOTO, K., CONDON, K., ECKLEY, S., DESJARDINS, C., WILSON, L., JORDAN, M. A., FEINSTEIN, S. C., CAVALETTI, G., POLYDEFKIS, M. & SLUSHER, B. S. 2018. Peripheral Neuropathy Induced by Microtubule-Targeted Chemotherapies: Insights into Acute Injury and Long-term Recovery. Cancer Research, 78, 817-829. WU, J., ZHAO, Z., ZHU, X., RENN, C. L., DORSEY, S. G. & FADEN, A. I. 2016. Cell cycle inhibition limits development and maintenance of neuropathic pain following spinal cord injury. Pain, 157, 488-503. WU, L. E. & SINCLAIR, D. A. 2016. Restoring stem cells - all you need is NAD+. Cell Research, 26, 971-972. WU, Z., WANG, S., WU, I., MATA, M. & FINK, D. 2015. Activation of TLR‐4 to produce tumour necrosis factor‐α in neuropathic pain caused by paclitaxel. European Journal of Pain, 19, 889-898. XIAO, W., BOROUJERDI, A., BENNETT, G. J. & LUO, Z. D. 2007. Chemotherapy- evoked painful peripheral neuropathy: Analgesic effects of gabapentin and effects on expression of the alpha-2-delta type-1 calcium channel subunit. Neuroscience, 144, 714-720.

140

XIAO, W. H. & BENNETT, G. J. 2008. Chemotherapy-evoked neuropathic pain: Abnormal spontaneous discharge in A-fiber and C-fiber primary afferent neurons and its suppression by acetyl-l-carnitine. Pain, 135, 262-270. XIAO, W. H., ZHENG, H., ZHENG, F. Y., NUYDENS, R., MEERT, T. F. & BENNETT, G. J. 2011. Mitochondrial abnormality in sensory, but not motor, axons in paclitaxel-evoked painful peripheral neuropathy in the rat. Neuroscience, 199, 461-469. XIONG, Z.-G., CHU, X.-P. & SIMON, R. 2006. Ca 2+-permeable acid-sensing ion channels and ischemic brain injury. The Journal of Membrane Biology, 209, 59- 68. XU, Z.-Z., KIM, Y. H., BANG, S., ZHANG, Y., BERTA, T., WANG, F., OH, S. B. & JI, R.-R. 2015. Inhibition of mechanical allodynia in neuropathic pain by TLR5- mediated A-fiber blockade. Nature Medicine, 21, 1326. YAMAMOTO, T., BYUN, J., ZHAI, P., IKEDA, Y., OKA, S. & SADOSHIMA, J. 2014. Nicotinamide Mononucleotide, an Intermediate of NAD+ Synthesis, Protects the Heart from Ischemia and Reperfusion. PLoS ONE, 9, e98972. YAMASHITA, T., YAMAMOTO, S., ZHANG, J., KOMETANI, M., TOMIYAMA, D., KOHNO, K., TOZAKI-SAITOH, H., INOUE, K. & TSUDA, M. 2016. Duloxetine inhibits microglial P2X4 receptor function and alleviates neuropathic pain after peripheral nerve injury. PloS ONE, 11, e0165189. YANG, I. H., SIDDIQUE, R., HOSMANE, S., THAKOR, N. & HÖKE, A. 2009. Compartmentalized microfluidic culture platform to study mechanism of paclitaxel-induced axonal degeneration. Experimental Neurology, 218, 124-128. ZASADIL, L. M., ANDERSEN, K. A., YEUM, D., ROCQUE, G. B., WILKE, L. G., TEVAARWERK, A. J., RAINES, R. T., BURKARD, M. E. & WEAVER, B. A. 2014. Cytotoxicity of Paclitaxel in Breast Cancer Is due to Chromosome Missegregation on Multipolar Spindles. Science Translational Medicine, 6, 229ra43. ZHANG, B., YANG, L., KONISHI, Y., MAEDA, N., SAKANAKA, M. & TANAKA, J. 2002. Suppressive effects of phosphodiesterase type IV inhibitors on rat cultured microglial cells: comparison with other types of cAMP-elevating agents. Neuropharmacology, 42, 262-269.

141

ZHANG, H., BOYETTE-DAVIS, J. A., KOSTURAKIS, A. K., LI, Y., YOON, S. Y., WALTERS, E. T. & DOUGHERTY, P. M. 2013. Induction of monocyte chemoattractant protein-1 (MCP-1) and its receptor CCR2 in primary sensory neurons contributes to paclitaxel-induced peripheral neuropathy. The Journal of Pain, 14. ZHANG, H. & DOUGHERTY, P. M. 2014. Enhanced Excitability of Primary Sensory Neurons and Altered Gene Expression of Neuronal Ion Channels in Dorsal Root Ganglion in Paclitaxel-Induced Peripheral Neuropathy. Anesthesiology, 120, 1463-1475. ZHANG, H., LI, Y., DE CARVALHO-BARBOSA, M., KAVELAARS, A., HEIJNEN, C. J., ALBRECHT, P. J. & DOUGHERTY, P. M. 2016. Dorsal root ganglion infiltration by macrophages contributes to paclitaxel chemotherapy induced peripheral neuropathy. The Journal of Pain. ZHANG, H., YOON, S.-Y., ZHANG, H. & DOUGHERTY, P. M. 2012. Evidence That Spinal Astrocytes but Not Microglia Contribute to the Pathogenesis of Paclitaxel-Induced Painful Neuropathy. The Journal of Pain, 13, 293-303. ZHANG, Y., PENG, W., AO, X., DAI, H., YUAN, L., HUANG, X. & ZHOU, Q. 2015. TAK-242, a Toll-Like Receptor 4 Antagonist, Protects against Aldosterone- Induced Cardiac and Renal Injury. PLoS ONE, 10, e0142456. ZHAO, X., GORIN, F. A., BERMAN, R. F. & LYETH, B. G. 2008. Differential Hippocampal Protection when Blocking Intracellular Sodium and Calcium Entry during Traumatic Brain Injury in Rats. Journal of Neurotrauma, 25, 1195-1205. ZHENG, F. Y., XIAO, W. H. & BENNETT, G. J. 2011. Functional deficits in peripheral nerve mitochondria in rats with paclitaxel- and oxaliplatin-evoked painful peripheral neuropathy. Experimental Neurology, 232. ZHENG, Y.-T., YANG, H.-Y., LI, T., ZHAO, B., SHAO, T.-F., XIANG, X.-Q. & CAI, W.-M. 2015. Amiloride sensitizes human pancreatic cancer cells to erlotinib in vitro through inhibition of the PI3K/AKT signaling pathway. Acta Pharmacologica Sinica, 36, 614.

142