Purdue University Purdue e-Pubs

Open Access Dissertations Theses and Dissertations

Fall 2014 Acrolein as a novel therapeutic target for spinal cord injury induced neuropathic pain Jonghyuck Park Purdue University

Follow this and additional works at: https://docs.lib.purdue.edu/open_access_dissertations Part of the Biology Commons, Biomedical Engineering and Bioengineering Commons, and the Neuroscience and Neurobiology Commons

Recommended Citation Park, Jonghyuck, "Acrolein as a novel therapeutic target for spinal cord injury induced neuropathic pain" (2014). Open Access Dissertations. 342. https://docs.lib.purdue.edu/open_access_dissertations/342

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for additional information. PURDUE UNIVERSITY GRADUATE SCHOOL Thesis/Dissertation Acceptance

Jonghyuck Park

ACROLEIN AS A NOVEL THERAPEUTIC TARGET FOR SPINAL CORD INJURY INDUCED NEUROPATHIC PAIN

Doctor of Philosophy

Riyi Shi

Eric Nauman

Kevin Otto

Bradley Duerstock

To the best of my knowledge and as understood by the student in the Thesis/Dissertation Agreement, Publication Delay, and Certification/Disclaimer (Graduate School Form 32), this thesis/dissertation adheres to the provisions of Purdue University’s “Policy on Integrity in Research” and the use of copyrighted material.

Riyi Shi James Leary 12/01/2014

i

ACROLEIN AS A NOVEL THERAPEUTIC TARGET FOR SPINAL CORD

INJURY INDUCED NEUROPATHIC PAIN

A Dissertation

Submitted to the Faculty

of

Purdue University

by

Jonghyuck Park

In Partial Fulfillment of the

Requirements for the Degree

of

Doctor of Philosophy

December 2014

Purdue University

West Lafayette, Indiana

ii

To my parents, in-laws, my lovely wife Jinah Kim

and my little angel Irene Park for their endless love and support.

사랑하는 나의 아내 김진아,

나의 작은 천사 박신우, 그리고 양가 부모님께 감사와 사랑의 마음을 담아 받칩니다

iii

ACKNOWLEDGEMENTS

First of all, I would like to thank my academic advisor, Dr. Riyi Shi, for his excellent guidance and support. With his enthusiasm and great efforts, I could broaden my scientific horizon. During my Ph.D. journey and still now, he has been teaching me how to be an independent scientist.

In addition, I would like to acknowledge my committee members, Dr. Kevin Otto, Dr. Eric Nauman and Dr. Bradley Duerstock for their serving on my committee and advices for my projects and research. Without their efforts, this dissertation could not have been successful.

Beyond my committee members, my dissertation would not be possible without my colleagues. I would like to thank Dr. Bruce Cooper and Ms. Amber Jannasch for their technical supports when I use the LC-MS-MS device and also would like to thank Dr. Fletcher White and Dr. Michael Due at Indiana University Medical School for their fantastic electrophysiology data and analysis. My lab mates at Dr. Shi’s lab, especially, Dr. Lingxing Zheng who is going to be a daddy soon, without his dot blotting and many biochemical data, I could not finish my Ph.D. dissertation. Many thanks go to one of my lab mates Glen Acosta who always helps me out for ordering, productive scientific discussions, especially, I thank him for his generosity because he makes me realize that I am a very good racquetball player.

I also would like to acknowledge Dr. Yongdoo Park who was my academic advisor during my master program at Korea University Medical School. I really appreciate his advises both personally and professionally, as well as inspiration for

iv

the last ten years. He is still teaching me to be a good scientist as well as good father, consciously and unconsciously.

For my financial and administrative support, I would like to extend my acknowledge Sandy May at Weldon School of Biomedical Engineering, Sue Wood at Interdisciplinary Biomedical Sciences, Dr. Riyi Shi and Department of Biological Sciences.

Finally, I would like to acknowledge my parents, in-laws and family for their endless support and love during my Ph.D. studies. My lovely wife Jinah Kim, as a best friend and as a lifetime companion, she always supports me and makes me happy and comfortable. My little girl Irene Park, she completes Jinah and me and makes me laugh when I am stressed out. Lastly, I thank everyone who has motivated me and has believed in me throughout my academic career.

v

TABLE OF CONTENTS

Page LIST OF FIGURES……………………………………………………………..……..viii LIST OF ABBREVIATIONS…………………………………………………..……….x ABSTRACT………………………………………………………………………….....xi

CHAPTER 1. INTRODUCTION ...... 1

1.1 Introduction of Spinal Cord Injury ...... 1

1.2 The Role of Oxidative Stress after SCI ...... 2

1.3 The Origin of Acrolein and Role of Acrolein in Spinal Cord Injury ...... 4

1.4 Acrolein Detection Techniques ...... 6

1.5 Acrolein Scavenging ...... 11

1.6 Classification of Pain ...... 16

1.6.1 Nociceptive Pain ...... 16

1.6.2 Inflammatory Pain ...... 16

1.6.3 Neuropathic Pain ...... 17

1.7 Chronic Neuropathic Pain after Traumatic Spinal Cord Injury ...... 17

1.8 The Role of Reactive Oxygen Species in Neuropathic Pain ...... 19

1.9 Acrolein-mediated Sensory Dysfunction after SCI ...... 20

1.10 General Characteristic of TRPA1 Channel and Its Activation by Acrolein .....23

1.11 Summary and Conclusion ...... 25

CHAPTER 2. NEUROPROTECTIVE ROLE OF ACROLEIN SCAVENGER IN

SPINAL CORD INJURY RAT MODEL ...... 26

2.1 Introduction ...... 26

2.2 Material and Methods ...... 28

2.3 Results ...... 34

2.4 Discussion ...... 49

vi

Page

CHAPTER 3. ACROLEIN INVOLVEMENT IN SENSORY AND BEHAVIORAL

HYPERSENSITIVITY FOLLOWING SPINAL CORD INJURY ...... 53

3.1 Introduction ...... 53

3.2 Materials and Methods ...... 55

3.3 Results ...... 60

3.4 Discussion ...... 74

3.5 Conclusion ...... 76

CHAPTER 4. THE EFFECT OF TRPA1 CHANNEL IN THE PERIPHERAL

TERNIMAL AFTER SPINAL CORD INJURY ...... 77

4.1 Background and Research Scope of This Study ...... 77

4.2 Spinal Cord Injury-induced Neuropathic Pain Mechanical Allodynia ...... 80

4.3 Changes in TRPA1 Gene Expression Level after SCI ...... 82

4.4 Persistent Elevation of Acrolein Level and Onset of Neuropathic Pain

Behaviors after SCI ...... 87

4.5 Gene Expression Level and Behavioral Alternations by Pro-nociceptive Factor

Acrolein ...... 89

4.6 Acrolein Increases the Peripheral Tterminal Sensitivity Fllowing SCI ...... 94

4.7 Summary and Discussion ...... 97

CHAPTER 5. THE EFFECT OF PHENELZINE AS AN ALTERNATIVE TO A

HYDRALAZINE ...... 99

5.1 Background and Research Scope of This Study ...... 99

5.2 Phenelzine Treatment Reduced 3-HPMA Production ...... 100

5.3 Acrolein Scavenging by Phenelzine after SCI ...... 102

5.4 Analgesic Effect of Phenelzine on Neuropathic Pain after SCI ...... 106

5.5 Summary and Discussion ...... 110

CHAPTER 6. THE EFFECT OF POTASSIUM CHANNEL BLOCKER ON SCI-

INDUCED MOTOR DEFICIT AND NEUROPATHIC PAIN ...... 113

6.1 Background and Rsearch Sope of This Study ...... 113

6.2 The Effects of Potassium Channel Blockers on Neuropathic Pain ...... 114

vii

Page

6.3 The Effects of Potassium Channel Blockers on Motor Function Deficits ...... 119

6.4 Summary and Discussion ...... 123

CHAPTER 7. SUMMARY AND CONCLUSION ...... 125

CHAPTER 8. FUTURE STUDIES ...... 131 LIST OF REFERENCES ...... 136 VITA…………………………………………………………………………………..151 PUBLICATIONS...... 152

viii

LIST OF FIGURES

Figure ...... Page Figure 1. Structure of Acrolein ...... 4 Figure 2. Vicious Cycle of Oxidative Stress by Acrolein after Traumatic SCI ...... 6 Figure 3. Detection of Acrolein-lysine Level through Immunoblotting Method ...... 8 Figure 4. Detection of 3-HPMA Level through LC/MS/MS Method ...... 10 Figure 5. Hydralazine Structure and Its Reaction with Free Acrolein ...... 12 Figure 6. Chemical Structure of Acrolein Scavengers...... 15 Figure 7. Elevation of Acrolein Level in Spinal Cord Based on the SCI Severity ...... 35 Figure 8. Correlation of 3-HPMA Level with SCI Severity ...... 37 Figure 9. Detection Hydralazine Level in the CNS after Systemic Application ...... 38 Figure 10. Hydralazine can Reduce the Acrolein-lysnie Adduct Level after SCI ...... 40 Figure 11. Urine 3-HPMA Level Reduction by Hydralazine ...... 41 Figure 12. The Effect of Hydralazine on Cyst Area Reduction after SCI ...... 43 Figure 13. Systemic Application of Hydralazine Increased Motor Function after SCI .... 44 Figure 14. Hydralazine Alleviated Mechanical Threshold after SCI ...... 46 Figure 15. Acrolein can Induce Neurological Deficits without Mechanical Injury ...... 48 Figure 16. Elevation of Acrolein-Lysine Adduct Level after SCI ...... 62 Figure 17. TRPA1 Gene Expression Level in DRG Cells ...... 63 Figure 18. Calcium Imaging of TRPA1 Signaling via Acrolein in DRG ...... 64 Figure 19. SCI-mediated Current Threshold to Generate Action Potential ...... 67 Figure 20. Acrolein-Induced Sensory Neuron’s Excitability after SCI ...... 68 Figure 21. Micro-injected Acrolein Produced Neuropathic Pain Behaviors ...... 71 Figure 22. The Effect of Hydralazine on Pain Behaviors and Acrolein Level after SCI . 72 Figure 23. Delayed Application of Hydralazine on Neuropathic Pain after SCI ...... 73

ix

Figure ...... Page Figure 24. Mechanical Sensory Hypersensitivity after Moderate Contusive SCI ...... 81 Figure 25. The Elevation of TRPA1 mRNA Level 1 Week and 2 Weeks after SCI ...... 84 Figure 26. The Effect of Hydralazein on TRPA1 mRNA Level 7 Days after SCI ...... 85 Figure 27. The Effect of Hydralazein on TRPA1 mRNA Level 14 Days after SCI ...... 86 Figure 28. Elevation of Acroein-lysine Adducts in Spinal Cord and 3-HPMA after SCI 88 Figure 29. The Effect of Micro-Acrolein Injection on TRPA1 mRNA Level in Tissues 90 Figure 30. Micro-injected Acrolein Promoted the Mechanical and Cold Allodynia ...... 92 Figure 31. The Motor Function Deficits by Micro-injected Acrolein ...... 93 Figure 32. Sites for Acrolein Injection and Mechanical Allodynia Von Frey Test ...... 95 Figure 33. The Effect of Peripheral Injection of Acrolein in Control and SCI group ...... 96 Figure 34. The Reduction of Urine 3-HPMA by Acrolein Scavengers ...... 101 Figure 35. Reduction of Acrolein-lysine Adduct Level by Acrolein Scavengers ...... 104 Figure 36. 3-HPMA Level in 1 Day after SCI with various Dosages of Acrolein Scavengers ...... 105 Figure 37. Phenelzine Reduced the SCI-induced Mechanical Allodynia ...... 108 Figure 38. Phenelzine Reduced SCI-induced Neuropathic Pain Cold Allodynia ...... 109 Figure 39. Potassium Channel Blockers Reduced Mechanical Allodynia after SCI ...... 117 Figure 40. Potassium Channel Blockers Reduced Cold Allodynia after SCI ...... 118 Figure 41. The Effect of Potassium Channel Blocker on Motor Deficit after SCI ...... 121 Figure 42. Potassium Channel Blockers Promoted the Motor Recovery after SCI ...... 122

x

LIST OF ABBREVIATIONS

3-HPMA: 3-hydroxypropylmercapturic acid

4-AP: 4-aminopyridine

4-AP-3-MeOH: 4-aminopyridine-3-methanol

BBB Scale: Basso, Beattie and Bresnahan Scale

BSA: Bovine Serum Albumin

CNS: Central Nerve System

GC: Gas Chromatography

GHS: Glutathione

HPLC: High Performance Liquid Chromatography

IP Injection: Intraperitoneal injection

LC: Liquid Chromatography

MS: Mass Spectrometry

NYU impactor: New York University impactor

PBS: Phosphate Buffered Saline

PUFA: polyunsaturated fatty acid

ROS: reactive oxygen species

SCI: Spinal Cord Injury

xi

ABSTRACT

Park, Jonghyuck. Ph.D., Purdue University, December 2014. Acrolein as a Novel Therapeutic Target for Spinal Cord Injury Induced Neuropathic Pain. Major Professor: Riyi Shi, M.D., Ph.D.

Despite years of research, post-spinal cord injury (SCI) chronic neuropathic pain remains refractory to treatment and drastically impairs quality of life for SCI victims beyond paralysis. Although inflammation and free radicals contribute to neuropathic pain in SCI, the mechanism is not completely clear. We have recently demonstrated that acrolein, a product and catalyst of lipid peroxidation, induces a vicious cycle of oxidative stress, amplifying its effects and perpetuating oxidative stress and inflammation. In the current study, we have confirmed that acrolein is elevated significantly at least two weeks post- SCI which coincides with the emergence of hyperalgesia (mechanical, cold and thermal). Furthermore, anti-acrolein treatment hydalazine can reverse pain behavior. Consistent with this, acrolein is a known ligand that directly excites transient receptor potential ankyrin 1 receptor (TRPA1) in DRG nociceptive neurons to transmit pain sensation. In addition, we have observed a significant increase of monocyte chemoattractant protein-1 (MCP1), a pro-inflammatory chemokine that is also known be increased upon acrolein stimulation and capable of sensitizing TRPA1 following SCI. Similarly, we observed a heightened excitatory response of DRG sensory neurons to current stimulation in the presence of acrolein, indicating an enhanced sensitivity of DRG cells to acrolein post SCI. In summary, acrolein may play an important role in the post SCI hyperalgesia through greater direct binding to TRPA1 and enhanced sensitivity of DRG to acrolein via MCP1-mediated pathway

1

CHAPTER 1. INTRODUCTION

1.1 Introduction of Spinal Cord Injury

According to the National Spinal Cord Injury Statistical Center, there are approximately 400,000 spinal cord injured patients in the United States with over 14,000 new injuries occurring each year. Traumatic spinal cord injury (SCI) could result in irreversible never injury which leads to both motor and sensory malfunction. In addition to the motor and sensory dysfunction, SCI patients can also suffer from various levels of neuropathic pain, bladder reflexive dysfunction, urinary tract infection, renal malfunction, loss of respiratory function, loss of sexual function, pneumonia, deep vein thrombosis, pulmonary embolism, and septicemia. These SCI-induced disorders have significantly negative effects on their quality of life. In addition, SCI results in significant financial impact. Based on data from the National Spinal Cord Injury Statistical Center, the estimated lifetime costs for a 25 years old tetraplegia patient is over $3 millions.

Despite years of study to improve the functional recovery and the quality of life, lack of effective therapeutic targets result in a frustrating condition to treat. High doses of methylprednisolone have been using as a standard care for SCI clinical trials because of its antioxidant properties (Bracken et al., 1997; Bracken et al., 1998; Bracken and Holford, 2002). However, recent clinical trials have failed to improve functional recovery after SCI with methylprednisolone treatment. Based on research, this treatment could lead to respiratory tract infection, urinary tract infection, pulmonary embolism, and gastrointestinal problems (George, 1995; Matsumoto et al., 2001; Lee et al., 2007; Suberviola et al., 2008b). Even though treatment with methylprednisolone may not be effect way to improve SCI patients’ symptoms, these clinical applications demonstrated

2 that attenuating oxidative stress, as a one of the therapeutic strategies, could be a effective way to improve the motor and sensory dysfunction following SCI.

1.2 The Role of Oxidative Stress after SCI

One of the critical features of SCI pathophysiology is the secondary cascade events after primary traumatic SCI, which occur in the hours, days, and weeks including ischemia/reperfusion injury, inflammation, oxidative stress, and glutamate excitotoxicity (Braughler and Hall, 1989b; Juurlink and Paterson, 1998b; Hall and Springer, 2004). The secondary cascade events result in spreading of additional damage throughout the spinal cord and contributing to the tissue degeneration and functional loss (Hall and Springer, 2004; Huang and Huang, 2006).

As a key factor of secondary cascade events, oxidative stress plays a critical role in mediating motor and sensory dysfunction after SCI. In addition, oxidative stress has been implicated in various neurodegenerative diseases (Hall, 1989; Hall and Braughler, 1993a; Smith et al., 1999; Seiler, 2000). Traumatic mechanical SCI injury to the membrane induces the disruption the intracellular calcium and sodium rise, and potassium falls which contribute to depolarization leading to intracellular level of calcium and sodium increase more. Furthermore, increase level of calcium could result in mitochondria injury and an increase in production and release of reactive oxygen species (ROS) (Hall, 1996; Xiong et al., 2007).

Traumatic primary SCI induces blood supply disruption which results in ischemia- reperfusion secondary injury. Shortage of oxygen and glucose may lead to impaired ATP production, mitochondrial damage, and generation and leakage of ROS from mitochondria (Braughler and Hall, 1989a; Hall and Braughler, 1993b). The generation of xanthine oxidase has been found to produce superoxide after ischemia-reperfusion damage (Braughler and Hall, 1989a; Hall, 1996; Juurlink and Paterson, 1998a; Kumar et al., 2004). In addition, recruitment of neutrophils to the tissue damage site and their

3 activations results in the increase in generation of ROS (Braughler and Hall, 1989a; Lewen et al., 2000).

The central nerve system (CNS) is especially very sensitive to oxidative stress because of its unique properties. Particularly, the CNS, compared to the other tissue, contains large proportion of phospholipids membrane and poly unsaturated fatty acids, which are specifically vulnerable to peroxidation (Hall and Braughler, 1993b; Hall, 1996; Halliwell and Gutteridge, 1999). Furthermore, compared to other types of cells, there are a number of mitochondria that could be a target and source of ROS. Under the high level of oxidative stress, the endogenous antioxidants such as glutathione, vitamin E, and ascorbic acid could be overwhelmed because the level of antioxidants may not be compensated due to rapid consumption. Even though many of research groups have been demonstrated that oxidative stress plays an important role in the pathogenesis after SCI, clinical trials with free radical scavengers have failed to promote functional recovery after traumatic SCI (George et al., 1995; Bracken et al., 1997; Bracken et al., 1998; Suberviola et al., 2008a). Therefore, we need to understand the mechanisms of oxidative stress and investigate the novel and effective therapeutic targets for SCI.

In addition to SCI, many of byproducts of oxidative stress have been implicated in numerous neurodegenerative diseases (Liu and Ji; Yowtak et al.; Hensley et al., 2000; Vincent et al., 2002; Naoi et al., 2005; Schwartz et al., 2008). The research for byproducts of oxidative stress-mediated neuronal diseases has been an intense area to prevent and slow down the development of various neurodegenerative diseases. However, clinical trials with free radical scavengers have not produced meaningful outcome similar to SCI. In this sense, this study could have critical applications for not only SCI patients but also for other neurodegenerative diseases.

4

1.3 The Origin of Acrolein and Role of Acrolein in Spinal Cord Injury

Acrolein is a highly reactive α,β-unsaturated aldehyde (Figure 1.) that is an environmental pollutant and byproduct of lipid peroxidation (LPO) after traumatic injury to spinal cord (Uchida, 1999a; Lovell et al., 2000; O'Brien et al., 2005). Acrolein can originate via endogenous pathway from LPO of poly unsaturated fatty acid in cell membrane, carbohydrates and amino acids. In addition, generation of acrolein occurs through exogenously such as incomplete combustion of organics, manufacturing processes, and cigarette smoke (Witz, 1989; Esterbauer et al., 1991a; Ghilarducci and Tjeerdema, 1995; Kirkham et al., 2003; Hamann et al., 2008b).

Figure1. Structure of Acrolein

5

In addition, because acrolein can act as byproducts and inducers of LPO, it has been implicated in many neurological diseases (Shi et al.; Horton et al., 1999; Lovell et al., 2000; Waxman, 2001; Luo and Shi, 2004; Luo et al., 2005a; Liu-Snyder et al., 2006b; Due et al., 2014). Acrolein has been shown to be the most reactive and abundant electrophile byproducts of LPO compared to much studied ROS. In addition, it can directly bind to proteins, DNA, and endogenous antioxidant glutathione (GSH) and reacts with GSH 100-150 times faster than another type of LPO byproducts such as 4- hydroxynonenal (4-HNE) (Witz, 1989; Esterbauer et al., 1991a; Hall and Braughler, 1993b; Ghilarducci and Tjeerdema, 1995; Uchida et al., 1998; Halliwell and Gutteridge, 1999; Uchida, 1999b). Furthermore, half life of acrolein is significantly longer than that of other transient ROS. Acrolein-protein and acrolein-GSH adducts also have much longer half-lives than those of free acrolein and significantly reactive (Adams and Klaidman, 1993; Burcham et al., 2004; Kaminskas et al., 2004b). In addition, acrolein can react with the thiol group of cysteine residue, the ε-amino group of lysine and histidine residues to generate reactive carbonyl-retaining adducts through the Michael addition (Esterbauer et al., 1991a; Burcham et al., 2004; Kaminskas et al., 2004a; Burcham and Pyke, 2006). Particularly, binding between acrolein and cystein via Michael addition is reversible so that acrolein can break from acrolein-cystein adduct and allow it to continue to react with other proteins. Acrolein can also bind to necleophilic DNA bases to generate exocyclic adducts (Kehrer and Biswal, 2000a). Through this pathway, acrolein can induce vicious cycle of oxidative stress and amplify its effect to many biochemical processes (Figure 2).

6

Figure2. Vicious Cycle of Oxidative Stress by Acrolein after Traumatic SCI.

1.4 Acrolein Detection Techniques

Because of reactive and volatility nature of acrolein, the detection of free acrolein is technically challenging. In addition, acrolein is able to undergo spontaneous polymerization, reduction, and oxidation because there are carbonyl and vinyl functional groups on the molecule (Shi et al., 2011). In this sense, based on the current researches and understating of characteristics of acrolein, it is reasonable to anticipate that majority of endogenous acrolein would exist in acrolein-protein adducts or be metabolized by endogenous antioxidants such as GSH and vitamin E.

Detecting acrolein-lysine adduct level by using immunoblotting methods has been using in many rat SCI model (Uchida, 1999a; Hamann et al., 2008b; Hamann and Shi, 2009; Park et al., 2013) (Figure 3). By using these techniques, many researchers could quantify the in vivo acrolein adducts level directly at the tissue of interest. In addition,

7 with the acrolein standard, actual in vivo acrolein level could be estimated. Although immunoblotting methods are very sensitive for in vivo acrolein level detection, using anti-acrolein antibodies to detect acrolein-lysine adduct only represent a portion of the total generated acrolein, which could always underestimate the actual acrolein level in the tissue. In addition, acrolein normally reacts with endogenous antioxidant GSH first, which results in a relative small amount of acrolein would bind to protein in normal condition of cell. Under this condition, immunoblotting methods could overlook the in vivo level of acrolein. Lastly, these methods are highly invasive because animals should be sacrificed to collect tissues which significantly limit their applications for the clinical trial. Therefore, a non-invasive method for detecting acrolein or acrolein metabolites would be effective for the investigation of acrolein-mediated neurodegenerative diseases and for the clinical application.

8

Figure 3. Detection of Acrolein-lysine Level through Immunoblotting Method. Top: artificially acrolein standard were prepared with different concentrations of acroelin. The monoclonal anti-acrolein antibody was used for acrolein standard samples. Bottom: based on the optical density of prepared acrolein samples, sample standard curve was calculated and this standard curve could be used for acrolein level quantification.

9

Non-invasive acrolein level measurement through acrolein metabolite in urine quantification has been reported (Zheng et al.; Lovell et al., 2001b; Abraham et al., 2011; Zheng et al., 2013) (Figure 4). Rather than invasive acrolein measuring via immunoblotting, this method was considered to be an indirect indicator of the in vivo acrolein level. In addition, detecting acrolein metabolite through collecting urine makes it particularly promise for the clinical applications. Specifically, 3-hydroxypropyl mercapturic acid (3-HPMA) has been selected as a major acrolein metabolite in urine (Park et al., 2000; Yan et al., 2010). Chemically, 3-HPMA is very stable and be kept for months in the freezer without sample deterioration. Furthermore, binding ratio between acrolein and GSH is 1:1, which indicates that 3-HPMA is a suitable biomarker to quantify the in vivo level of acrolein after SCI or acrolein-mediated neurodegenerative diseases (Schettgen et al., 2008; Yan et al., 2010; Park et al., 2013). The acrolein metabolite 3-HPMA in urine could be detected via liquid chropatography (LC) and mass spectroscopy (MS), which is an effective technique because of being performed very quickly.

10

Figure 4. Detection of 3-HPMA Level through LC/MS/MS Method The reaction scheme between acrolein and endogenous antioxidant GSH (A). The result of LC/MS/MS of acrolein metabolite 3-HPMA level detection. d3-3-HPMA was used as an internal control (B).

11

However, there are difference acrolein levels between 3-HPMA measuring and immunoblotting acrolein-lysine adduct levels (Park et al., 2013; Zheng et al., 2013). These discrepancies between two results could be explained through first, limited diffusion characteristics of acrolein into the blood stream. Second, immunoblotting is tissue and local specific, but 3-HPMA urine level was detected via systemically; this increased level cannot be considered to be a specific tissue. So, using these two different methods could contribute to a more precise detection level of acrolein following SCI.

1.5 Acrolein Scavenging

Because acrolein plays a critical role in secondary cascade events after traumatic primary SCI, acrolein scavenging would be an effective method to investigate the role of acrolein in SCI and other neurodegenerative diseases, and to improve motor and sensory dysfunction after SCI.

Anti-hypertensive medication hydralazine, which is the FDA approved drug, has been shown to its acrolein scavenging properties (Burcham et al., 2000; Burcham et al., 2002; Burcham et al., 2004). Hydralazine is a strong nucleophile and it has a hydrazide functional group. Through hydrazide group, hydralazine can bind to and neutralize acrolein and acrolein-protein adducts (Burcham et al., 2004; Kaminskas et al., 2004a, b; Burcham and Pyke, 2006) (structure and reactions between hyalazine and free acrolein shown in figure 5). Hydralazine reacts with free acrolein and acrolein-protein adduct in an equimolar manner to neutralize in a cell free system, so equimolar concentration of hydralazine could eliminate nearly whole acrolein (Zhu et al.; Burcham et al., 2000; Burcham et al., 2002; Hamann et al., 2008a; Hamann and Shi, 2009).

12

Figure 5. Hydralazine Structure and Its Reaction with Free Acrolein

13

The reaction products between hydralazine and acrolein were identified as (1E)- acrylaldehyde phthalazin-1-ylhydrazone (E-APH) and (1Z)-acrylaldehyde phthalazin-1- ylhydrazone (Z-APH)(Zhu et al., 2011). E-APH is the primary reaction product, which can significantly attenuate the acrolein-induced cellular toxicity and be excreted from body safely.

Based on previous studies, hydralazine can prevent acrolein-induced PC-12 cells’ death and injuries (Liu-Snyder et al., 2006a; Liu-Snyder et al., 2006b). In addition, hydralazine also prevents acrolein-mediated allyl alcohol-induced hepatotoxicity in in vitro hepatocytes (Burcham et al., 2000; Burcham et al., 2004). In addition, hydralazine can reduce the acrolein-mediated membrane damage, oxidative stress, and loss of compound action potential (Hamann et al., 2008a; Hamann and Shi, 2009). Furthermore, the isolated spinal cord ex vivo study showed that hydralazine can alleviate compression- induced the level of acrolein and acrolein-protein adducts in spinal cord (Hamann et al., 2008a).

Recent research has demonstrated that hydralazine could reach therapeutic concentration in 2 hours by systemic injection (Wang et al., 2011). The data has demonstrated that the level of hydralazine in spinal cord, brain and kidney were higher than minimum level of hydralazine which could alleviate the acrolein-mediated cellular death. Because a specific gravity value of spinal cord tissue is 1, the hydralazine level in the spinal cord can be calculated about 20 μmol/L which is in the range of therapeutic concentrations, 25 μmol/L (Liu-Snyder et al., 2006a). In addition, compared to bioavailability of oral application, which is about 40%, bioavailability through systemic application is about 90% (Hamann et al., 2008a). Taken all together, these results supported that systemic application of hydralazine may be an effective treatment for traumatic SCI and acrolein-induced neurodegenerative diseases via scavenging acrolein.

One of limitations of hydralazine is its hypertensive effect. Especially, vasodilatory effect by hydralazine could be life threatening side effect to SCI patients, as

14 they would be get neurogenic shock (Tuncel and Ram, 2003). Short half life of hydralazine is also another limitation for clinical application of hydralazine. Based on previous study, half life of hydralazine in vivo is only about 1 hour to 2 hours (Reece, 1981), which is significantly shorter than expected oxidative stress and acrolein production time window after SCI. However, many of in vitro studies using hydralazine have demonstrated that acrolein scavenging is an effective ways for intervention to oxidative stress. Because the hydrazine functional group is in charge of scavenging the acrolein, developing or investigating an alternative acrolein scavengers without antihypertensive effect would be needed.

There are other molecules that have a hydrazine functional group like hydralazine. Their chemical structure can be seen in figure 6. Phenelzine has a hydrazine functional group which can react with acrolein equimolar manner like hydralazine (Wood et al., 2006a). In addition, phenelzine is not a vasodilator so it can be used at higher doses than hydralazine safely and as an alternative to hydralazine. Furthermore, its half life is much longer than that of hydralazine so it can provide additional neuro-protection effect after SCI (Shi et al., 2011; Zheng et al., 2013). Even though phenelzine is the FDA approved anti-depressant, the safety of phenelzine as an acrolein scavenger for clinical application should be established through further research. Another possible acrolein scavenger is dihydralazine which has two hydrazine functional groups. Therefore, it could potentially be twice as efficient as hydralazine’s acrolein scavenging (Kaminskas et al., 2004a). Dihydralazine has been shown to alleviate the acrolein level in a cell-free system and also prevent acrolein-induced release of lactate dehydrogenase (LDH) in in vitro hepatocytes study (Kaminskas et al., 2004b). Taken together, although phenelzine and dihydralazine can be considered as an acrolein scavenger alternative to hydralazine, researches for their safety and efficacy should be done first.

15

Figure 6. Chemical Structure of Acrolein Scavengers Hydralazine, dihydralazine, and phenelzine (left to right)

16

1.6 Classification of Pain

1.6.1 Nociceptive Pain Nociceptive pain is an early-warning physiological alarm system which is mediated by high-threshold unmyelinated C or thinly myelinated Aδ primary sensory neurons (Woolf, 2007; Woolf and Ma, 2007). This is the normal sensory reaction when we sense noxious stimulus such as something too hot, cold, or sharp. Only the intense stimuli (noxious stimuli) that are higher than threshold can activate nociceptive sensory neuron (Basbaum et al., 2009) and once the noxious stimuli are disappeared, the nociceptive pain would also be removed. Nociceptive sensory neurons express specific transducers, mainly transient receptor potential (TRP) channels on C or/and Aδ primary sensory, which can transmit noxious stimuli into electrical activity (Clapham, 2003; Tai et al., 2008).

1.6.2 Inflammatory Pain Inflammatory pain is responsible for responding to tissue damage and the subsequent immune reaction. This pain is also protective like nociceptive pain in order to aid healing and repair of the damaged tissues by creating the hypersensitivity to avoid physical contact and movement (Sandhir et al.; Marchand et al., 2005; Lucas et al., 2006). Under this condition, normally non-painful factors become painful factors (allodynia) so, response to normal stimuli and noxious stimuli are exaggerated and lasted (Marchand et al., 2005; Lucas et al., 2006). In addition, damaged tissues and immune reaction to infection can result in release of various chemical mediators which create an inflammatory soup (histamine, serotonin, ATP and bradykinin) (Gilden, 2005; Marchand et al., 2005; Moalem and Tracey, 2006; Ren and Dubner, 2008). These inflammatory soups can activate and modify the responsiveness of nociceptor afferents to stimuli (Glezer et al., 2007). Typically, inflammatory pain is removed by resolving the initial tissues damage.

17

1.6.3 Neuropathic Pain Neuropathic pain is not adaptive and protective, but maladaptive which results from both lesion and diseases to the central nervous system (CNS). In addition, neuropathic pain can result in both neurological disorders and pain (Scholz and Woolf, 2007). Under this condition, nociceptive pathway could be changed by peripheral sensitization and central sensitization so that normally non-painful factors produce the pain (allodynia) and response to noxious stimuli could be enhanced (hyperalgesia) (Zimmermann, 2001; Campbell and Meyer, 2006; Costigan et al., 2009a). Peripheral sensitization indicates that a reduction in threshold and an increase in sensitivity at the nociceptor in peripheral terminal ending. It can result in the pain hypersensitivity at the tissue damage and inflammation (Zhuo, 2007; Costigan et al., 2009a). Central sensitization is an increase in the excitability of sensory neuron in the CNS, which results from damage or/and diseases. Under the central sensitization, normal sensory information begins to generate abnormal responses so, low-threshold primary afferent sensory neurons which are activated by innoxious stimuli can activate neurons in the spinal cord and/or the brainstem that normally respond to only noxious stimuli (Zimmermann, 2001; Zhuo, 2007; Costigan et al., 2009a). In addition, after damage to the CNS, the immune cells are activated in the CNS, which increase the release of cytokines and also lead to central sensitization (Costigan et al., 2009b; White et al., 2009).

1.7 Chronic Neuropathic Pain after Traumatic Spinal Cord Injury

Chronic neuropathic pain is commonly found in SCI patients. Based on the previous studies, traumatic SCI could lead to a chronic neuropathic pain within months (Richards et al., 1980) and more than 80% of SCI patients suffer from significant pain which is felt as burning and stabbing (Siddall et al., 2003). As a matter of fact, beyond paralysis, chronic neuropathic pain further drastically impairs the quality of life for SCI victims. Despite years of research, this type of clinical pain remains refractory to treatment. As a result, chronic neuropathic pain could be prolonged for several decades after SCI and leads to depression and suicide (Segatore, 1994; Cairns et al., 1996). There

18 are several possible mechanisms that can explain the generation of chronic neuropathic pain after SCI (Scholz and Woolf, 2002; Basbaum et al., 2009; Costigan et al., 2009a; Milligan and Watkins, 2009).

First, after neuronal damage, the central terminal of non-nociceptive primary afferents Aβ fibers could make new synaptical connections with nociceptive neurons in laminae I and II. This re-connection and sprouting in the nociceptive pathway can result in persistent pain hypersensitivity

Second, changing in glutamatergic neuron and N-methyl-D-aspartate (NMDA) receptor mediated sensitization. Under the normal condition, pain is transmitted by release of glutamate from the central ending of nociceptor neurons, which is generating excitatory postsynaptic currents (EPSCs) in secondary order neuron in the spinal cord dorsal horn interneurons. However, under the neuropathic pain condition, increased release of glutamatergic neurotransmitter from central ending of nociceptors would sufficiently depolarize and activate postsynaptic NMDA receptors. As a result, an increased in calcium influx can activate calcium-dependent signaling transductions such as MAPK, PKC and PI3K, which could keep activating the dorsal horn pain transmission resulting in exacerbating sensitivity to noxious stimuli.

Third, after neuronal damage, the loss of function of GABAergic and/or glycinergic inhibitory interneurons could result in increased sensitivity. Under normal physiological condition, inhibitory interneurons continuously release inhibitory neurotransmitters such as Gamma-aminobutyric acid (GABA) and/or glycine to alleviate the excitability of lamina I interneurons and modulate pain transmission. However, under the setting of neuropathic pain physiology condition, there are loss of inhibitory effects (disinhibition), which contribute to the generation of allodynia and hyperalgesia. Disinhibition can make non-nociceptive myelinated Aβ primary afferent to involve in pain transmission circuitry, so normally innoxious stimuli are perceived as painful stimuli.

19

Lastly, glial cells such as microglia and astrocytes also can lead to chronic neuropathic pain after SCI. Under normal conditions, microglias act as resident microphages in the CNS. After injury within hours, microglias accumulate in the superficial dorsal horn and they release various cytokines and chemokines which can induce the central sensitization and persistent pain. However, in contrast to microglia, an activation of astrocyte is normally delayed and prolonged up to several months, which indicate that astrocytes may play an important role for the maintenance of persistent pain and induction of central sensitization after SCI.

1.8 The Role of Reactive Oxygen Species in Neuropathic Pain

Recent studies have shown that ROS are also implicated in neuroapthic pain (Yowtak et al.; Kim et al., 2004; Park et al., 2006; Gao et al., 2007). Kim et al showed that the nerve ligation induced-neuropathic pain allodynia significantly reversed after treatment with various ROS scavengers (Kim et al., 2004; Kim et al., 2006). This data has demonstrated that ROS could play a critical role in the maintenance of neuropathic pain. As we already discussed about the role of NMDA receptors in central sensitization after SCI, an increased in the phosphorylation of NMDA receptor in the spinal dorsal horn is a one of the key factors for inducing the neuropathic pain and central sensitization. In addition, previous study has showed that oxidative stress in the spinal cord also contributed to maintaining a high level of phosphorylation of NMDA receptors in dorsal horn which results in central sensitization. Furthermore, elevated levels of ROS in the spinal cord can activate various protein kinases such as PKA and PKC, which again elevate the level of phosphorylation of NMDA receptor. In addition, Gao et al have demonstrated that the level of activated NMDA receptors in the dorsal horn and neuropathic pain behaviors were significantly attenuated by systemic application of ROS scavenger in neuropathic pain animal model (Gao et al., 2007). These studies indicate that ROS are implicated in phosphorylation of NMDA receptor which results in neuropathic pain behavior after SCI, and ROS scavengers can be used as analgesic by blocking ROS-related signaling pathways after SCI.

20

Again, another important mechanism that induces the chronic neuropathic pain was loss of inhibitory function after neuronal damage (in section 1.7). Several recent studies have demonstrated a sensitivity of GABA neurons to oxidative stress. Virgili et al showed that an antioxidant resveratrol, which was found in red wine could promote neuroprotective effect in vivo because it could selectively alleviate the reduction in GABA synthesizing enzyme levels in rat (Virgili and Contestabile, 2000). In addition, Chen et al have demonstrated that superoxide, which is a type of ROS, could lead to presynaptic inhibition of GABA release to postsynaptic neuron (Chen and Pan, 2007). This study indicated the vulnerability of GABA neuron to oxidative stress. One recent study showed that increase levels of spinal ROS can promote neuropathic pain behaviors such as mechanical hyperalgesia in normal mice, which was very similar to the pain behaviors in peripheral nerve injury-induced neuropathic pain animal model. In addition, they showed that ROS were critically involved in reducing presynaptic inhibitory GABA neurotransmitter release in neuropathic pain animal model. However, when they treated them with ROS scavengers, ROS-mediated effects were reversed. These data indicate that ROS promote the pathogenesis of neuropathic pain following SCI by interfering with the inhibitory function of the spinal GABAergic system, and ROS scavengers could effectively attenuate neuropathic pain behaviors, which are achieved through the restore of GABA-mediated inhibitory system. Furthermore, many of byproducts of oxidative stress could activate TRP channels on the nociceptive sensory neuron, which transmit the pain signaling to the CNS (Bautista et al., 2006). We will discuss about this in sections 1.9 and 1.10 more specifically.

1.9 Acrolein-mediated Sensory Dysfunction after SCI

Although many of studies have demonstrated the role of ROS in neuropathic pain following SCI (Kim et al., 2004; Chen and Pan, 2007; Gao et al., 2007; Yowtak et al., 2011), There are a few studies about acrolein and acrolein-protein adduct-mediated neuropathic pain after SCI. As a self-regenerative molecule, acrolein can generate additional acrolein and promote oxidative stress even with micro-molar levels of acrolein,

21 which indicate that acrolein may promote pathogenesis of neuropathic pain through oxidative stress after SCI (Luo and Shi, 2004).

Many studies have shown that acrolein, is a neuropathic pain inducer after SCI through the transient receptor potential ankyrin 1 (TRPA1) (Bautista et al., 2006; Nilius et al., 2007; Andersson et al., 2008; Bessac et al., 2008). Specifically, TRPA1 channel act as sensor for byproducts of oxidative stress-derived molecules and of metabolisms. As a TRPA1 agonist, acrolein can bind to TRPA1 directly and activate TRPA1 channel (Bautista et al., 2006). Based on previous studies, gene expression level of TRPA1 is increased in the spinal dorsal horn and dorsal root ganglia (DRG) after SCI (Venkatachalam and Montell, 2007; Due et al., 2014).

Due et al. have demonstrated that mechanical and cold allodynia could be observed in 2 weeks after SCI and lasted several weeks (Due et al., 2014). Also, data have shown that there was increased in expression level of acrolein at least 2 weeks after SCI, which was in coincidence with neuropathic pain mechanical and cold allodynia (Due et al., 2014; Park et al., 2014). Furthermore, Due and colleagues have shown that expression of TRPA1 messenger RNA (mRNA) level was increased seven days after SCI using lumbar level of DRG cells, which indicated that acrolein could contribute to generation of neuropathic pain even its level returned to normal physiological concentration after SCI. Furthermore, acrolein also acts as pro-inflammatory factor so that they can stimulate the release of chemokines after neuronal damage (Boddeke, 2001; Kirkham et al., 2003; Hulsebosch et al., 2009). Previous study has also shown that chemokine which was produced by pro-inflammatory factors such as acrolein could activate TRPA1 channel (Jung et al., 2008). In this sense, acrolein, both as a TRPA1 agonist and as a pro- inflammatory factor, it can activate TRPA1 channel directly and indirectly through chemokine-medicated signaling pathways.

An incidence of spontaneous activity and an increase in overall excitability state of DRG sensory neuron also are responsible for emergence and maintenance of neuropathic

22 pain after SCI other than acrolein level increase (Due et al., 2014). In this study, the data indicated that there was a decrease in the current threshold, which is minimum current to generate single action potential, in small and medium diameter of sensory neuron after SCI. In addition, there was an increase in number of DRG cells that responded to acrolein following SCI. Furthermore, acrolein could increase the excitability of sensory neuron at single DRG cell level, and acrolein-mediated sensory excitability was greater in the SCI group compared to the sham injury control group. The effect of acrolein in increase of sensory neuron’s excitability was further supported by Park and colleagues’ data. They have shown that exogenous injected acrolein into the spinal cord could promote not only the mechanical allodynia, but motor function and neuronal tissue deficits (Park et al., 2014). Taken results of electrophysiology and in vivo acrolein injection together, acrolein can be considered as an endogenous pro-nociceptive factor after SCI.

Based on current understanding and results from research, acrolein could be an effective therapeutic target as a pro-nociceptive and pro-inflammatory factor. The involvement of acrolein in chronic neuropathic pain after SCI was further demonstrated using acrolein scavenger hydralazine studies (Due et al., 2014; Park et al., 2014). The data showed that systemic application of hydralazine significantly attenuated mechanical and cold allodynia after SCI. The benefits of hydralazine as an analgesic lasted for several weeks and even worked two weeks delayed hydralazine application. This data indicated that acrolein scavenger hydralazine could prevent acrolein binding to TRPA1 channel and alleviate the acrolein-mediated inflammatory events after SCI.

These recent studies have demonstrated that acrolein could be a key therapeutic target in the generation of neuropathic pain after SCI, and acrolein scavenger such as hydralazine, could be an effective therapeutic strategy to attenuate neuropathic pain for promoting the SCI victims’ quality of life.

23

1.10 General Characteristic of TRPA1 Channel and Its Activation by Acrolein

TRPA1 channel is a non-selective cation calcium ion channels and consists of six trnasmembrane segments (Doerner et al., 2007; Andre et al., 2009; Patapoutian et al., 2009). The N-terminal and C-terminal of TRPA1 channel are located in the intracellular side of membrane. Especially, the long N-terminal contains up to 18 ankyrin repeated domains which are involved in the activation of TRPA1 channel. On this ankyrin repeated domain, there is EF-hand domain where calcium ions can bind to and activate TRPA1 channel. So, calcium ions play a critical role in TRPA1 functioning (Clapham et al., 2001; Bandell et al., 2007). In addition, N-terminal cystein residues are responsible for electrophile-mediated TRPA1 channel activation.

The gene expression level of TRPA1 was increased in the spinal dorsal horn and DRG after SCI (Due et al., 2014). TRPA1 channels have a wide pore which is stimulus dependent so that it can regulate the calcium ions permeability (Bandell et al., 2007). In this sense, SCI-induced neuropathic pain can increase not only the expression level of TRPA1 channels, but the pore size, which lead to influx of much larger calcium ions compared to normal physiological condition. Furthermore, TRPA1 channels are expressed in C and Aδ primary afferent fiber and many of TRPA1-expression neurons are peptidergic neurons which contribute to release of neurotransmitters to postsynaptic neurons (Bautista et al., 2006). So, TRPA1 channel on the central ending of primary afferent neuron in the spinal dorsal horn might play an important role in the maintenance and transmission of pain signaling.

Based on previous studies, TRPA1 channel acts as a sensor for endogenous byproducts of oxidative stress-mediated substances (Andersson et al., 2008). These byproducts such as acrolein, 4-HNE and H2O2 can directly activate TRPA1 channel because of the electrophilic properties of compounds. These electrophilic compounds react with nucleophilic thiol group of cystein and ε-amino group of lysine residues in the N-terminal of TRPA1 channel via covalent modification. In addition, the reaction

24 between acrolein and thiol group of cystein is reversible so acrolein can break from cystein-acrolein adduct and bind to other proteins and/or DNA. Under normal physiological pH and temperature, covalent modification on the channel lasts for hours which contribute to conformational change of TRPA1 channel. Therefore, membrane permeability is the one of the critical fundamental requirements for TRPA1 channel’s agonist (Bautista et al.; Bautista et al., 2006; Andersson et al., 2008; Bessac et al., 2008). Another possible activation of TRPA1 channel by byproducts of oxidative stress is the formation of disulfide bond via oxidation by hydrogen peroxide and again, this modification of N-terminal cystein in TRPA1 results in wide the pore size of the channel (Bautista et al., 2006; Nilius et al., 2007; Patapoutian et al., 2009).

Other non-electrophilic TRPA1 agonists such as menthol, camphor and icilin do not bind to cystein or lysine residue covalently because of their chemical structure (Bandell et al., 2004; Bandell et al., 2007). These agonists can stimulate TRPA1 channel through a temporal and reversible Fischer lock and key principle. In general, anesthetic agents such as lidocaine can promote the activation and sensitization of TRPA1 channel (Matta et al., 2008; Piao et al., 2009). Normally these anesthetic agents depress the CNS, but also activate peripheral nociceptive sensory neurons by activating TRPA1 channel. Furthermore, these effects could integrate with tissue damage during the surgical operation, which could result in an increase in post-operative pain and inflammation.

TRPA1 channel also can be activated indirectly through G-protein coupled receptor (GPCR)-mediated pathway (Bautista et al., 2006). For example, TRPA1 can be activated by pro-inflammatory factor bradykinin (BK)-induced signaling pathway. When BK binds to its receptor (bradykinin 2 receptor), intracellular calcium level is increased through a phospholipase C (PLC)-mediated down-stream signaling. These calcium ions could bind to EF-hand domain of N-terminal of TRPA1 and lead to activation of TRPA1 channel. In addition, TRPA1 can be also activated via transient receptor potential vanilloid1 (TRPV1). TRPA1 channel is most likely expressed with TRPV1 channel. TRPV1 can be sensitized by binding of its agonists such as capsaicin or/and by secondary messengers

25 such as PKC. In addition, heat thermal stimulus can activate the TRPV1 channel. TRPV1 channel activation lead to an increase in influx of calcium ions into the cytoplasm, and these calcium ions could bind to TRPA1 and results in activation of TRPA1 channel. Therefore, TRPA1 gates in response to both calcium ions release from intracellular calcium ion stores and influx calcium ions after gating TRPV1 channel(Bautista et al.; Hinman et al., 2006; Andersson et al., 2008; Bessac and Jordt, 2008).

1.11 Summary and Conclusion

There is strong evidence that acrolein acts as key factor in secondary cascade events after primary traumatic SCI. With its reactive, volatile, long half-life and ability to generate protein adducts nature, acrolein has been associated with many different types of neurological diseases.

In addition, as a strong pro-nociceptive agonist, acrolein can promote sensory neurons’ excitability, and reduction current threshold of sensory neuron after SCI. Although, more studies are needed to understand the mechanisms about elevation of sensitivity on nociceptive sensory neuron via TRPA1 channel or/and acrolein alone, many of studied have demonstrated that acrolein could be a valuable therapeutic target to improve the SCI-induced neuropathic pain and neurological deficits.

Many of recent studies have demonstrated that acrolein scavenger hydralazine can attenuate acrolein-mediated cellular death and toxicity. In addition, hydralazine can decrease the acrolein-protein adduct level and offer the neuroprotective effects in in vivo SCI animal model. Because half-life of hydralazine is only a few hours (2hours), it might be possible that frequent application of hydralazine would provide additional neuroprotective effects. Taken all together, acrolein-scavenging strategies might be a novel and potential way to improve quality of patients’ life, who are suffering from SCI and acrolein-mediated neurological diseases.

26

CHAPTER 2. NEUROPROTECTIVE ROLE OF ACROLEIN SCAVENGER IN SPINAL CORD INJURY RAT MODEL

2.1 Introduction

Secondary cascade biochemical events after primary traumatic SCI can result in spreading the additional damages throughout the spinal cord and amplify the effects of traumatic SCI.

Post-traumatic oxidative stress, as a key factor of secondary cascade, plays a critical role in the pathogenesis and in mediating functional loss of SCI (Braughler and Hall, 1989b). However, treatment with free radical scavengers have been failed to produce effective results on clinical trials. We have been demonstrated that acrolein, byproduct of LPO and α,β-unsaturated aldehyde, plays an important role in neuronal cell death and secondary cascade events after traumatic injury (Shi and Luo, 2006; Hamann et al., 2008b; Hamann and Shi, 2009; Shi et al., 2011). Basically, acrolein is very reactive and volatile molecule, which concentration is increased in the spinal cord after SCI. In addition, with its significantly long half-life, it can stay within biological system compared to other types of ROS (Esterbauer et al., 1991a; Ghilarducci and Tjeerdema, 1995; Picklo and Montine, 2001; Shi et al., 2002a; Luo and Shi, 2004; Liu-Snyder et al., 2006b; Shi and Luo, 2006). Furthermore, acrolein can bind to proteins, DNA, and lipid directly and induce the formation of acrolein-protein adducts (Kehrer and Biswal, 2000b; Stevens and Maier, 2008). In addition, with acrolein’s neurotoxicity nature, many studies have been also shown that acrolein has been implicated in many of neurodegenerative diseases such as degrading myelin (Luo et al., 2005a; Liu-Snyder et al., 2006b).

27

Therefore, acrolein can induce a vicious cycle of oxidative stress and amplifies its effects, and also it is responsible for spreading of additional damage throughout the spinal cord tissue after traumatic SCI.

Because of important pathological role of acrolein, the detection and measurement techniques of acrolein are very important to investigate the acrolein-mediated neurodegenerative diseases. Recent study has shown that stable acrolein specific metabolite 3-HPMA was generated when acrolein reacted with endogenous antioxidant GSH, which could be measured in urine (Carmella et al., 2007; Schettgen et al., 2008). Because of its non-invasive method, measuring 3-HPMA has been using to estimate the level of acrolein such as acrolein intake via smoke and food (Schettgen et al., 2008).

Based on current understanding of acrolein, we hypothesize that attenuation of acrolein might reduce neurological damage and improve the functional recovery after SCI. To investigate this hypothesis, and to demonstrate the role of acrolein in SCI, we have investigated that the acrolein scavenger, hydralazine, could mitigate acrolein- mediated cell death (Liu-Snyder et al., 2006b), as well as in isolated spinal cord (Hamann et al., 2008a). In addition, hydralazine could prevent acrolein-mediated toxicity in hepatocytes via in vitro test and delayed onset of multiple sclerosis (MS)-induced motor neuron dysfunction in mice (Leung et al., 2011b). Although there are many of in vitro studies that have investigated the effect of acrolein-scavenger hydralazine via in vitro test, anyone hasn’t evaluated the effect of acrolein scavenger therapy using in vivo SCI animal model. Without this information, the effect and therapeutic value of acrolein for treatment for SCI patients remain uncertain.

In this sense, the primary goal of this chapter is to ascertain the pathological role of acrolein in the CNS and the effect of acrolein scavenger hydralazine in live SCI animal model. Through this study, we also investigate whether the FDA-approved hypertensive medication hydralazine can attenuate acrolein-mediated tissue damage and promote functional recovery.

28

2.2 Material and Methods

Animal

Body weight between 210-230 grams male Sprague-Dawley rats were used for this study. Rats were housed and handled in compliance with the Purdue University Animal Care and Use Committee guidelines. For the acclimation, the animals were kept at least one week before surgery.

Contused- spinal cord injury rat model

Animals were anesthetized with a ketamine (80 mg/kg) and xylazine (10 mg/kg) mixture by an intraperitoneal (IP) injection. A dorsal surface of spinal cord at the T-10 level was exposed by removing spinous process and vertebrae lamina. We made two different severities contusion models by using New York University (NYU) impactor. Basically, NYU impactor has a 10-gram of weight rod with 1.5 mm. By dropping this weight rod from different height, we can generate different severity of contused-SCI model. In this study, we dropped the weight rod form 25mm onto the intact T-10 level of spinal cord to generate moderate contused-SCI model. For severe contused-SCI model, we dropped it onto the spinal cord from 37.5mm. The sham control was operated using only a laminectomy at the T-10 level. For the post-surgical care, the animals were placed on a heating pad and performed manual bladder expression daily two times until the reflexive control bladder function was observed, and to prevent dehydration, 3.0 cc of saline was injected through subcutaneously.

Hydralazine treatment

The hydralazine hydrochloride (Sigma, St. Louis, MO, USA) was dissolved in sterilized phosphate buffered saline (PBS). 5mg/kg of hydralazine solution was injected via I.P. injection daily for two weeks, right after SCI within 5 minutes. For

29 immunoblotting study, hydralazine solution was administrated by I.P. injection right after surgery and 24 hours after SCI. Then, animals were euthanized two hours injection to collect spinal cord injury sites. The dosage of hydralazine solution 5mg/kg was decided based on previous study (Zheng et al., 2013). They have demonstrated that 5mg/kg of hydralazine showed not only the effectiveness of acrolein reduction, but the minimal influence of blood pressure to the animals.

Quantification of hydralazine in the central nervous system

Paper spray mass spectrometry was used to quantify the concentration of hydralazine in the spinal cord, and brain. To quantify the concentration of hydralazine in the tissues, first, we collected tissues from the untreated animals, which were spiked artificially with different concentration of hydralazine from 15.63 ng/ml to 2000 ng/ml. For the internal control, the fixed concentration of nicotine was used. Based on these data, we constructed a calibration curve. In the experiment group, saline was injected as a control and 5mg/kg of hydralazine was injected then, animals were sacrificed two hours after injection. Collected samples were punched and transferred onto the paper for paper spray mass spectrometry analysis and measurement of hydralazine level. The final concentration of hydralazine was calculated by using calibration curve fits.

Urine collection

For the urine collection from the animals, we used the standard metabolic cages. Drinking water and urine collection were separated to avoid urine dilution by water source. During the urine collection, food was also provided. To collected enough urine for the study, 0.5 cc of saline was injected via subcutaneously at each collection time. Sometimes to collect sufficient urine, we performed manually press the rat bladder.

30

Creatinine measurement

We measured the creatinine level with a creatinine (urinary) assay kit (Cayman Chemical Company, Item No. 500701). Original urine sample were diluted for 12 and 24 times before the creatinine assay. According to the assay manual, all required agents were prepared. Then, all diluted urine samples and creatinine standards were incubated on the shaker with the alkaline picrate solution for 15 mins. Using standard spectrophotometry, the samples were read at 492 nm then, 5 ul of acid solution were added to all samples after initial reading. The plate also was placed on the shaker for 15 mins. The differences wave absorbance between the initial reading and final reading were used for analysis.

3-hydroxypropyl mercapturic acid (3-HPMA) Analysis in Urine

3-HPMA was analyzed in urine according to Eckert et al (Eckert et al., 2010). For using LC/MS/MS solid phase extraction with isolute ENV+ cartridge (Biotage, Charlotte, NC) was used to prepare each sample. Each cartridge was washed with 1ml of methanol, 1 ml of water, and 1 ml of 0.1% formic acid in water. A volume of 500 µl deuterated 3- HPMA (d3-3-HPMA) were spiled with 200ng, which was then mixed with 500 µl of 50mM ammonium formate and 10 µL of undiluted formic acid. This mixture was loaded into the ENV+ cartridges. After the samples loading, using 1 ml of 0.1% formic acid, and 1 ml of 10% methanol/90% 0.1% formic acid in water, washed each cartridge twice. Next, under the nitrogen gas, the cartridges were dried for 30mins, then, using three volumes of 600 µl methanol with 2% formic acid, all cartridges were eluted. The eluates were dried in the evaporate device. Right before loading the samples into the LC/MS/MS, each sample was reconstituted in 100 µl of 0.1% formic acid. The LC/MS/MS using and data analysis were performed based on Zheng et al (Zheng et al., 2013).

31

Locomotor behavioral test

The functional motor neuron recovery was investigated by using the Basso, Beattie and Bresnahan (BBB) Locomotor Rating Scale (Basso et al., 1996). The BBB locomotor scores are based on the ability of SCI rat’s hind limb movement. For example, BBB score 0 means that there is no observable movement of rat’s hind limb, and 21, which is maximum score, means that movement of normal rat. For the test, the rats were placed in an open field for 5 mins for the acclimation, then observe the movement of left and right side of hind limb separately. BBB score were assessed at day one and weekly for four weeks after SCI.

Neuropathic pain assessment

50% up-down method was used for mechanical allodynia test (Chaplan et al., 1994). Briefly, SCI animals were placed in a plastic acrylic box on the top of a metal mesh. For the acclimation, left them alone at least 10 mins then test was performed by using von Frey filaments. As mechanical stimuli, we used calibrated von Frey filaments which range from 0.4, 0.6, 1.0, 2.0, 4.0, 6.0, 8.0 and 15.0 grams (stoelting, Wood Dale, IL, USA). Initially, 4 grams of von Frey filament was applied perpendicularly with a sufficient bending force to the plantar surface of the hind-limb for 3-5 sec then removed. A brisk movement and/or licking and/or biting were considered as a positive reaction, then a lower von Frey filament was applied to the same spot. If there was no positive response, the next greater filament was presented. Different stimulus of filaments were applied at least 1 min interval.

Isolation of spinal cord

Initially, animals were anesthetized deeply with a mixture of ketamone (80mg/kg) and xylazine (10mg/kg) via I.P. injection. Then, they were perfused with oxygenated cold

Kreb’s solution (124mM NaCl, 2mM KCl, 1.24 mM KH2PO4, 26 mM NaHCO3, and 10

32

mM ascorbic acid, 1.3 mM MgSO4, 1.2 mM CaCl2, and 10 mM glucose). To isolate spinal cord, first, the whole vertebral column of rat was removed and we performed a dorsal laminectomy along the vertebral column. Collected spinal cord was cut into 1 cm long for immnoblotting for acrolein level detection and for histological analysis.

Artificial acrolein standard preparation

For the immunoblotting, the artificial acroelin standard was prepared. Initially, rat albumin (0.5 ml of 2mg/ml, CalBioChem) solution and acrolein (0.5ml of 100 mM, Sigma) solution were mixed together, which was incubated for 4 hours at 37 degree C. Using 0.1% BSA, we performed serial dilution up to 10000 fold with the acrolein concentration ranging from 5uM (or 100ng/ml) to 25nM (or 0.5ng/ml). The standard smaples were loaded and measured, then we measured the band intensities by using Image J (NIH). Based on these bands density, we constructed the calibration acrolein standard curve. So, acrolein level in the interest samples can be measured from this fitted standard curve.

Immunoblotting

Spinal cord segments which include damage site were incubated with a 1% Triton solution with the corresponding amount of Protease Inhibitor Cocktails (Sigma Aldrich), then homogenized with sonicator for 10 sec. The homogenized tissue suspension was incubated in the ice for 1 hour then, centrifuged at 4 °C at 13500g for 30 mins. Using BCA protein assay kit, we measured the total protein concentration from each sample for the equal sample loading. 200μg of samples were transfer to the nitrocellulose membrane by using a Bio-Dot SF Microfiltration Apparatus (Bio-Rad, Hercules, CA, USA). Next this transferred membrane was blocked in 0.2% casein and 0.1% Tween 20 in PBS blocking buffer for 1 hour and incubated with monoclonal mouse anti-acrolein antibody (ABCAM) for at least 18 hours at 4 °C. An anti-acrolein antibody was diluted at a ratio of 1:1000 in blocking buffer with 2% goat serum and 0.025% sodium azide. The

33 membrane was washed 4 times in blocking buffer for 4 mins then, the membrane was transferred to 1:10000 alkaline phosphatase conjugated goat anti-mouse IgG solution for another hour (VECTASTAIN ABC-AmP Kit). For the final washes, we used blocking buffer and 0.1% Tween 20 in Tris-buffered saline then, the membrane was exposed to substrate of the ABC-AMP kit and visualized by chemiluminescence. The band density was assessed by Image J (NIH) and using an arbitrary unit for the expression.

Histology of rat spinal cord injury

For the tissue collection, initially we perfused animals with oxygenated cold Kreb’s solution then followed by 10% formalin. The removed spinal cord was into 10% formalin for the post fixation for 24 hours then, embedded in paraffin. Tissues were cut with 10 µm thickness through horizontal section and stained with luxol fast blue and cresyl violet. The cyste area was quantified by measuring the percent area of the cyst relative to the rest tissue area on the section, then we took an average two adjacent sections at the spinal cord mid-line. The area was compared between hydralazine treatment and non-treatment group with a Student’s T-test.

Acrolein microinjection into spinal cord and histological analysis

After deep anesthesia, we performed laminectomy at T-10 level to expose intact spinal cord. The micro-pipettes were pulled and beveled for the insertion into the spinal cord. Micro-pipette was loaded with saline as a control and acrolein in saline as an experimental group. The solution was injected using a three-way valve and a syringe under the negative pressure. For the injection, we used the PMI-100 oressure micro- injector (Dagan Corp., Minneapolis, MN). 1.6 µl of saline was injected into rat’s left side of spinal cord as a control and 1.6 µmol of acrolein in saline was injected into rat’s right side of spinal cord. Specifically, each solution was injected 0.6 mm lateral to the spinal cord mid-line and 1.2 mm ventral to the spinal cord surface at T-10 level. For the post-

34 surgical care, rat was placed on the heating pad and anti-biotic ointment was applied to the incision site.

For the histological analysis, we performed hematoxylin and eosin staining on the spinal cord 2 weeks after surgery. Paraffin embedded spinal cord samples were cut into longitudinally with 15 µm thickness by using AO829 microtome (American Optical, Buffalo, NY). After staining with hematoxylin and eosin, images were acquired by Olympus microscope at 4X and followed by Image J for the analysis.

Statistical analysis

For the data comparison between two groups, we performed Student’s t-test method and for the more than two variables, we used ANOVA and Tukey test. If the p value is smaller than 0.05, we considered the data is statistically different. All the averages were expressed in mean േ SEM.

2.3 Results

The increase of acrolein level after SCI

We have measured the acrolein level after SCI by using immunoblotting. The data have shown that acrolein level was significantly increased in moderate and severe SCI compared to the acrolein level in sham control group at both 1 day and 14 days after SCI (Figure 7). Specifically, the levels of acrolein in sham, moderate, and severe SCI were 9.1±5.3, 31.1± 4.0, and 42.7±3.1 respectively. There were significantly difference in acrolein level between sham and moderate SCI also between sham and sever SCI group at 1 day after SCI. Similar to this, acrolein levels in 14 days after SCI were 8.5±3.6, 15.3±4.0, and 34.7±3.6 respectively. There were statistically differences in acrolein level between sham and severe SCI group.

35

Figure 7. Elevation of Acrolein Level in Spinal Cord Based on the SCI Severity.

Acrolein-lysine adduct level based on the SCI severity was detected by using immunoblotting at different time points (1 day and 14 days after SCI). The bar graph showed that acrolein levels were significantly increased in moderate and severe SCI compared to sham control group at both 1 day and 14 days after SCI. In addition, acrolein level was correlated with SCI severity which indicated that sever SCI showed the highest level of acrolein. This data also indicated that acrolein level was persisted at least 14 days following SCI. For the comparison, acrolein arbitrary unit in moderate and severe were 31.09± 4.0 and 42.69±3.13, which were significantly higher than that of shame (9.13±5.29, p <0.005) at day 1. Also, acrolein level in severe SCI at day 14 was 34.73±3.64, which was also significantly higher than that of sham (8.53±3.57, p<0.005). There was no difference between moderate and sham at 14 days after SCI. For the data analysis, one way ANOVA and post hoc tests were used. N=6 in all groups.

36

Elevation of urine 3-HPMA levels from SCI

To determine the endogenously generated acrolein level after SCI, we measured the 3-HPMA level using rat urine samples from different SCI severity groups. Similar to acrolein level in different SCI severity group, 3-HPMA levels also were correlated with SCI severity (Figure 8). Specifically, we collected rat urine 24 hours after SCI and 3-HPMA levels were 1.99±0.21, 3.51±0.24, and 5.51±1.13 μg/mg in sham, moderate and severe SCI group. These data showed that 3-HPMA level of moderate SCI was higher than that of sham (p<0.05) and 3-HPMA level of severe group was also higher than that of moderate SCI (p<0.05). Taken together, elevation of 3-HMPA level also correlated with SCI severity.

The level of hydralazine in CNS following IP injection

Before test the benefit of acrolein scavenger hydralazine, we investigated whether it could reach therapeutic level in the CNS after I.P injection. According to the previous study, half-life of hydralazine was about 2 hours, so we euthanized the animals 2 hours after hydralazine systemic application (5mg/kg) and then, the brain and spinal cord was removed. Hydralazine level was measured by using paper spray mass spectrometer. The results data has shown in figure 9. First of all, there were significant levels of hydralazine in the CNS within two hours after systemic application. Specifically, the detected level of hydralazine in the brain and spinal cord were 2.92 ± 0.88 and 4.42 ± 1.11 µg/g (29.47 ± 4.65 and 19.87 ± 4.42 µM), while, there was no detected hydralazine concentration in control group. Based on previous in vitro test, minimal level of hydralazine to attenuate acrolein-mediated cellular death was no higher than 20 µM. So, this data indicated that hydralazine could reach the CNS using systemic application and its concentration was enough to provide the neuroprotection effect.

37

Figure 8. Correlation of 3-HPMA Level with SCI Severity

The urine samples from sham control, moderate SCI, and severe SCI were collected 24 hours after SCI for the 3-HPMA study. Similar to acrolein level, 3-HPMA level also correlated with SCI severity. The 3-HPMA levels were 1.99±0.21, 3.51±0.24, and 5.51±1.13 μg/mg respectively. Specifically, 3-HPMA level in severe SCI showed the highest level compared to any other groups. N=4-6 for each group. For the (P < 0.05 כ) statistical analysis, one way ANOVA and Turkey test were used

38

Figure 9. Detection Hydralazine Level in the CNS after Systemic Application.

A) 5mg/kg of hydralazine solution was injected through systemically and same amount of saline was used for the control group. Two hours after injection, the brain and spinal cord tissues were removed to determine level of hydralazine. B) hydralazine level was determined using paper spray mass spectrometer. The data showed that level of hydralazine in the CNS were 29.47 ± 4.65 and 19.87 ± 4.42 µM respectively. No hydralazine level was detected in saline control group. This data demonstrate that systemic application of hydralazine could reach the CNS within two hours and could provide neuroprotective benefits. N=4 in all groups.

39

Acrolein scavenger hydralazine can effectively decrease acrolein level in SCI animal model

We have confirmed that hydralazine could reach the CNS with the therapeutic level. Next, to test if the acrolein scavenger hydralazine could reduce the acrolein level after SCI, we injected 5mg/kg of hydralazine right after SCI surgery within 5 mins and gave the second injection two hours before the animals were euthanized. As shown in figure 10, acrolein scavenger hydralazine could significantly reduce the acrolein-lysine adduct level after SCI. The data indicated that in both moderate SCI and severe SCI, the levels of acrolein were reduced by more than 50% after hydralazine application. This data supported that acrolein scavenger hydralazine could reduce the acrolein level after SCI effectively.

Acrolein scavenger hydralazine can decrease the urine 3-HPMA level after SCI

Again, urine 3-HPMA level was measured to confirm the effect of hydralazine as an acrolein scavenger. For this study, 5mg/kg of hydralazine was injected for seven days daily and urine also collected every 24 hours for seven days. Initially, 3-HPMA level was significantly increased in SCI group in the first three days after operation (Figure 11). Specifically, 3-HPMA levels were 3.51±0.24, 2.96 ±0.20, and 2.49±0.12 μg/mg, respectively. However, 4 days post injury, the level of 3-HPMA were as low as that of control. Hydralazine treatment group showed that the level of 3-HPMA at one day after SCI was significantly lower than that of SCI only group. However, there was no difference between hydralazine treatment group and SCI only group in the remaining period.

40

Figure 10. Hydralazine can Reduce the Acrolein-lysnie Adduct Level after SCI

Using immunoblotting, we measured the acrolein-lysine adduct level after 5mg/kg of hydralazine injection. Hydralazine was injected two times before collecting spinal cord (right after SCI within 5 mins and 2 hours before euthanizing). The data results showed that the level of acrolein-lysine adduct was significantly decreased from 30.35+5.34 to 8.94+7.44 in moderate SCI rats and from 43.45+4.34 to 22.45+3.56 in severe SCI rats. For the band densities’ analysis, Image J was used. N=6 in all groups and p<0.05 in both moderate and severe groups between hydralazine treatment and SCI only group.

41

Figure 11. Urine 3-HPMA Level Reduction by Hydralazine

To determine effect of hydralazine, urine 3-HPMA was detected from SCI rat for seven days. Urine was collected every 24 hours for seven days and 5mg/kg of hydralazine was injected daily. 3-HPMA level was significantly decreased 1 day after P < 0.05 when compared to control; #: P :כ) hydralazine injection. N=4 in each group < 0.05 when compared to SCI only (Day 1); ANOVA)

42

Hydralazine significantly reduces tissue damage after SCI

We measured the cyst area to determine whether the hydralazine could reduce the neurological damage after SCI. To quantify the area of cyst, we took an average two adjacent longitudinal sections at the spinal cord mid-line and stained by luxol fast blue method (Figure 12). There was no cyst area in control group, however, cyst covered about 37.99% of the tissue on SCI group. When we treated with hydralazine, the cyst area was significantly decreased (from 37.99% to 11.39%). This data demonstrated that acrolein scavenger hydralazine could reduce area of cyst formation after SCI.

Hydralazine improves locomotor function recovery in rat SCI

The locomotor function recovery was assessed by using BBB scale one day after SCI and once a week for four weeks (Figure 13). As a base line, BBB score in whole groups showed 21 points. However, one day after SCI, all animals scored 0 which indicated no movement. For the treatment group, 5mg/kg of hydralazine was injected every day for 14 days. The BBB score in hydralazine injection group was significantly increased compared to SCI only group starting at seven days after hydralazine injection. Significant differences BBB score were maintained until 4 weeks. Specifically, the BBB score in SCI with hydralazine treatment group was 16.12 ± 1.03, and BBB score in SIC only group showed 12.44 ± 1.17 which was significantly lower than that of hydralazine treatment group (p<0.05). The BBB score in Sham injury group was 18.45±1.05 one day after SCI, but one week after SCI, we observed normal locomotor function (BBB score 21). During the test periods, there was no motor deficit behavior in control group.

43

Figure 12. The Effect of Hydralazine on Cyst Area Reduction after SCI

Cyst area was quantified 4 weeks after SCI using saline injected uninjured spinal cord, SCI only spinal cord, and SCI with hydralazine treatment group. A-C) Luxol fast blue staining showed the size of cyst area in the spinal cord. (A) uninjured saline injection site, (B) SCI only, and (C) SCI with hydralazine injection spinal cord. The cyste area in hydralazine treatment group was significantly smaller than SCI only group (11.39±5.44% vs. 37.99±4.32%, p < 0.01, student t test). N=4 in each group.

44

Figure 13. Systemic Application of Hydralazine Increased Motor Function Recovery after SCI

SCI animals’ locomotor function was assessed by BBB score. Before SCI surgery, all group showed normal gaits (21 points). One day after SCI, BBB score in all groups were 0 except sham injury group. Acrolein scavenger hydralazine (5mg/kg) was injected for 14 days daily. One week after SCI, the BBB score in hydralazine treatment group was statistically higher than SCI only group and this trend kept maintained until 4 weeks. Specifically, the BBB score in hydralazine treatment group was 16.12 ± 1.03 and that in SCI only was 12.44 ± 1.17 (p<0.05). There was no abnormal motor function in control group during the test period. N=10-13 in each group.

45

Hydralazine attenuated acute neuropathic pain behavior following SCI

In addition to the functional benefits of hydralazine on motor function recovery, we also tested the effect of acrolein scavenger hydralazine on sensory dysfunction in SCI. Specifically, we investigated whether hydralazine could attenuate the neuropathic pain behavior through reduction in acrolein level after SCI. For this, again, 5mg/kg of hydralazine was injected systemically for two weeks every day. Neuropathic pain behavior, allodynia, was assessed by using von Frey filaments. SCI animals’ mechanical threshold was measured one day before SCI as a base line, and started at 2 weeks after SCI (Figure 14). Normally, mechanical allodynia test could be started when the SCI animals could support their body weight via their hindlimb and figure 14 showed that two week after SCI, mechanical allodynia was well established. Based on 50% up-down method, the maximum threshold is 15.0g and minimum threshold is 0.25g (Chaplan et al., 1994). The base lines of control, sham, SCI only, and SCI with hydralazine treatment were 14.02 ± 0.8 g, 14.65 ± 0.3 g, 14.3 ± 0.7 g and 13.92 ±1.0 g, respectively. During the test periods, there were no changes of mechanical threshold in control and sham injury groups. However, mechanical threshold in SCI only group was significantly lower than that in SCI with hydralazine treatment groups at two weeks after SCI. In addition, neuropathic pain behavior was significantly decreased in SCI with hydralazine treatment group during test period compared to SCI only group. Specifically, mechanical threshold in SCI only was 8.1 ± 1.2 g at two weeks after SCI, which was significantly higher than that in SCI with treatment group, which was 3.71 ± 0.28 g. We could check similar mechanical allodynia reductions in hydralazine treatment group during the test. Basically, the low mechanical threshold means less sensitive to mechanical stimulus. So, this data demonstrated that acrolein scavenger hydralazine could attenuate the neuropathic pain allodynia through removing acrolein levels after SCI.

46

Figure 14. Hydralazine Alleviated Mechanical Threshold after SCI

Systemic application of hydralazine alleviated neuropathic pain allodynia after SCI. No neuropathic pain behaviors were observed in control and sham. However, mechanical threshold in SCI only group was significantly lower than that in sham group during the test period. Mechanical threshold in hydralazine treatment was significantly higher than that in SCI only. This data showed that acrolein scavenger hydralazine can attenuate neuropathic pain after SCI. # p<0.05, ## p<0.01 for comparison between sham and SCI. * p<0.05, ** p<0.01 for comparison between SCI+HZ and SCI. Control: N=10, SCI: N=13, SCI+HZ: N=13, Sham: N=6

47

Acrolein micro-injection into the spinal cord resulted in neurological motor function deficits and tissue damage

Next, we investigated whether the acorlein itself could induce the neurological deficits such as motor dysfunction and neural tissue damage without traumatic injury. For this, we injected a certain amount of acrolein into the animal’s right side of spinal cord, and the same amount of saline was injected into the animal’s left side of spinal cord directly (Figure 15). For the motor function evaluation, we performed BBB locomotor test every 2 days for 14 days on each right and left hind limb. As shown in figure 15B, BBB score, the motor function of contralateral side of acrolein injection side was significantly lower than saline injection site, which was persistent during the test period. This data showed that acrolein could produce motor function deficit.

In addition, acrolein alone can produce the neuronal damage on tissue. The eosinophilic lesion area was surrounded by notable glial scar tissue with a number of small nucleus cells. However, there was no glial scar formation on saline injection site of spinal cord.

48

Figure 15. Acrolein can Induce Neurological Deficits without Mechanical Injury

A) Acrolein and saline injection site into the spinal cord and the tissue section plan for histology in (C). B) Micro-injected acrolein into the spinal cord could induce the motor function deficits. Motor function was evaluated via BBB score, which was significantly lower in acrolein injection site compared to that of saline injection site. p < 0.01 for comparison ככ .This trend lasted for 2 weeks. The values are mean±SEM between right side (acrolein-injected) and left side (saline injected), Student’s t-test. N=6 in each group. C) H and E staining was used for histological analysis. There was no morphological in saline injection site, while there was massive glial scar formation in acrolein injection site. The inset indicated the enlarged acrolein injected area which showed the detailed cellular damage by acrolein. Scale bar in C is 1 mm.

49

2.4 Discussion

In this current study, we have demonstrated that the acrolein level was significantly increased in contusion rat SCI model and elevated acrolein could be decreased via acrolein scavenger hydralazine treatment. In addition to removing acrolein, systemic application of hydralazine also could lead to a decrease in the tissue damage, locomotor dysfunction, and neuropathic pain behaviors. Furthermore, acrolein itself could result in neurological damage and motor function deficits without any mechanical traumatic injury, which were very similar to the SCI-induced neurological disorders such as tissue damage and motor neuron malfunction. These data were supported by many of previous studies that acrolein scavenger hydralazine could reduce acrolein-mediated cellular death via in vitro study (Shi and Blight, 1996; Liu-Snyder et al., 2006b; Hamann et al., 2008b; Hamann et al., 2008a). Therefore, current study demonstrated that acrolein plays an important role in the pathogenesis of spinal cord mechanical injury.

This study has showed that acrolein level increased in contused spinal cord injury rat model. This is the first research that proved the elevation of acrolein in the SCI rat, even though there was previous research showing the increased acrolein level in compression-injured spinal cord from guinea pig (Luo et al., 2005b). In addition, we have demonstrated that acrolein level was significantly increased even two weeks after SCI (Figure 7), while Luo et al. showed the elevated acrolein level in only one week time period in guinea pig spinal cord. In addition, our data showed that the maximal increase of acrolein after rat spinal cord injury was 300%, however, increased level of acrolein in guinea pig spinal cord was 60% based on their ELISA data . There are some different between two studies because of using different animal species and different type of injury model (contusion vs. compression SCI model). Also, there were differences in measuring techniques (immunoblotting vs. ELISA). However, both of studies showed that the acorlein reached the highest level 24 hours after SCI. So, these two studies demonstrated that the level of acrolein after SCI peak at one day after SCI.

50

By using paper spray mass spectrometer (Wang et al., 2011), we have demonstrated that systemically injected hydralazine could reach the brain and spinal cord within two hours. The data showed that the concentration of hydralazine in the CNS was 20 μM in the spinal cord and 30 μM in the brain. Based on the previous study, the minimal level of hydralazine to mitigate the acrolein-mediated neuronal cell death was no more than 15 μM (Liu-Snyder et al., 2006b), which was within hydralazine concentration range in the spinal cord after systemic application. Many of studies have shown that acrolein scavenger hydralazine could attenuate neuronal damage by using neural cell culture, and ex vivo studies using isolated spinal cord (Liu-Snyder et al., 2006b; Hamann et al., 2008b; Hamann et al., 2008a; Hamann and Shi, 2009). This study is the first one that demonstrates the benefits of acrolein scavenger hydralazine as a neuroprotective agent via in vivo study. The estimated hydralazine concentration through ex vivo or in vitro cannot be applied to the in vivo study because of its complexity. Therefore, spray mass spectrometer data provided the valuable aspects for hydralazine application to the in vivo study.

Our data have demonstrated that systemic application of hydralazine could reduce acrolein level in the spinal cord following SCI by 50-70% and level of 3- HPMA in urine. This data supported that the benefit of hydralazine as a neuroprotective agent not only in vitro, but in vivo. Because we have injected hydralazine once a day and half-life of hydralazine is only a few hours, more frequent application of hydralazine would provide more benefits in attenuate acrolein level follow SCI.

We have demonstrated that hydralazine also could reduce the formation of cyst after SCI, which is a severe form of tissue damage and known as syringomyelia in traumatic spinal cord (Nurick et al., 1970; Mudge et al., 1984; Schurch et al., 1996; Rooney et al., 2009). This cyst area (Figure 12) would be the scar tissue such as seen in figure 15. Therefore, the reduction of cyst area by application of hydralazine could lead to the reduction of scar formation eventually, which acts as mechanical and

51 chemical barrier for axon elongation after SCI (Giger et al.; Nguyen et al., 2009; Park et al., 2010).

In addition to the reduction of cyst area, our data demonstrated that hydralazine application could not only improve locomotor dysfunction (Figure 13), but attenuate neuropathic pain mechanical allodynia after SCI (Figure 14). Neuropathic pain is the one of major symptoms that at least 80% of SCI patients are suffering from this neurological disease (Hulsebosch et al., 2009). Acrolein is also known as pro- inflammatory factor and stimulates the generation of various inflammatory compounds which are responsible for neuropathic and inflammatory pain after SCI (Facchinetti et al., 2007; Moretto et al., 2009). Especially, inflammation is the one of the main reasons that contribute to onset and maintenance of neuropathic pain after SCI (Schnell et al., 1999; Jones et al., 2005; White et al., 2005; Trivedi et al., 2006; White et al., 2007; Donnelly and Popovich, 2008; Hulsebosch, 2008). Therefore, it is reasonable to estimate that acroelin-mediated inflammation could result in onset of neuropathic pain. In this sense, acrolein scavenger hydralazine could attenuate the neuropathic pain behavior by reduction of inflammation reaction through acrolein removing. Taken together, acrolein scavenger hydralazine could be an effective analgesic to reduce neuropathic pain behavior, and also could be an effective treatment to improve the motor function recovery for SCI patients.

The results of this study demonstrated that actolein could be a valuable therapeutic target for neurological diseases and its scavenger hydralazine has a potential value as analgesic to mitigate the neuropathic pain, and agent to improve the motor function recovery after SCI. In spite of many of benefits of hydralazine, hydralazine has some limitations as an acrolein scavenger for SCI patients. Basically, hydralazine is the FDA-approved hypertensive mediation. However, hypotension effect by hydralazine could be a life threatening side effect to especially SCI patients because of neurogenic shock. In this sense, the other acrolein scavengers, which do not have a hypotension effect, are needed to be studied for SCI patients. FDA- approved antidepressant phenelzine could be an alternative to hydralazine because it

52 is not a vasodilator and its half-life is much longer than that of hydralazine (Cole and Weiner, 1960; Wood et al., 2006a). Therefore, it could provide more neuroprotective effect to SCI patients.

Through this study, we have demonstrated that the pathological role of acrolein in SCI and the benefits of acrolein scavenger hydralazine on sensory and motor function abnormalities. Since crlinical trials with ROS scavengers have failed to produce neuroprotection effect to SCI patients because of its various side effects (Constantini and Young, 1994; Chen et al., 1996), acrolein scavenging could be a novel and effective way to remove the oxidative stress and promote neuronal recovery (Anderson et al., 1994; Behrmann et al., 1994; Bartholdi and Schwab, 1995; George et al., 1995; Pereira et al., 2009). In addition, acrolein scavenging therapy also would provide neuroprotective effect to not only SCI patients, but other neuronal disorder patients, especially, acrolein-related neurodegenerative diseases such as multiple sclerosis (Smith et al., 1999; Leung et al., 2011b; Shi et al., 2011), Alzheimer’s disease (Lovell et al., 2001a; Montine et al., 2002). Therefore, acrolein scavenging has the valuable potential to improve the quality of patients’ life.

53

CHAPTER 3. ACROLEIN INVOLVEMENT IN SENSORY AND BEHAVIORAL HYPERSENSITIVITY FOLLOWING SPINAL CORD INJURY

3.1 Introduction

Chronic neuropathic pain after traumatic SCI significantly impairs the quality of SCI patients’. As a matter of fact, beyond paralysis, chronic neuropathic pain further drastically impairs the quality of life for SCI victims. Despite years of research, this type of clinical pain remains refractory to treatment. As a result, chronic neuropathic pain could be prolonged for several decades after SCI and lead to depression and suicide (Segatore, 1994; Cairns et al., 1996).

Based on previous study, at least 80% of SCI patients are suffering from neuropathic pain, which is felt like burning, stabbing and electric shock like (Hulsebosch, 2008; Hulsebosch et al., 2009), and felt through the whole body, but patients describe the location of pain is below or near the injury site level (Siddall et al., 2003). In the rodent SCI-induced neuropathic pain model, the possible explanations for the pain at the below injury level are, first, after SCI there could be anatomical changes. For example, the central terminal of non-nociceptive primary afferents Aβ fibers could make new synaptical connection with nociceptive neurons in laminae I and II. These anatomical changes after SCI can cause the persistent hypersensitivity. Second, the changing in excitatory interneuron after SCI. Under the neuropathic pain condition, there could be an increased in release of excitatory neurotransmitter such as glutamate, which can bind to its receptor NMDA. As a result, increased in calcium ion influx via NMDA receptor can activate calcium dependent signaling pathway such as MAPK, PKC and PI3K, which can

54 maintain the activation of pain transmission in dorsal horn and lead to hypersensitivity to the stimuli. Third, there could be an interruption in the inhibitory interneurons such as GABAergic and/or glycinergic inhibitory interneurons after neuronal damage (disinhibition). Therefore, loss of inhibitory effect can result in perceiving normally in- noxious stimuli as painful stimuli, which can contribute to the generation of neuropathic pain after SCI. Lastly, glial cells can also lead to neuropathic pain after SCI. After SCI, glial cells release various cytokines, which can bind to excitatory interneuron, and lead to signal transduction down-stream to stimulate the central sensitization (Bruce et al., 2002; Hains et al., 2003; Lu et al., 2008; You et al., 2008; Bedi et al., 2010; Wang et al., 2012).

Many of studies have demonstrated that the possible mechanisms for generation of neuropathic pain after neurological damage, which including ionic channel imbalances, release of pro-inflammatory factors, anatomical changes, and malfunction of neuro-glial cells. However, the possible role of byproduct of LPO, acrolein, and acrolein-protein adducts as important factors have been underestimated.

Acrolein, α,β-unsaturated aldehyde, is a byproduct of LPO (Esterbauer et al., 1991b; Hamann and Shi, 2009; Shi et al., 2011), and transient receptor potential ankyrin 1 (TRPA1) channel specific agonist, which are most likely expressed on peptidergic nociceptive sensory neuron (Bautista et al., 2006). Acrolein has been implicated in many of neurodegenerative disease with its reactive and long half-life properties, which could induce the vicious cycle of oxidative stress and enhance its effect (Shi et al., 2011; Wang et al., 2011). As a specific agonist for TRPA1 channel and pro-inflammatory agent, increase level of acrolein could lead to the generation of neuropathic and inflammatory pain after neuronal damage (Story et al., 2003; Bautista et al., 2006; del Camino et al., 2010). One of the previous studies showed that persistent elevated acrolein level at least two weeks after SCI and it could result in onset of neuropathic pain allodynia after SCI (Park et al., 2014).

55

The FDA-approved hypertensive medication hydralazine has shown that its ability to react with acrolein and acrolein-protein adduct and neutralized them (Burcham and Pyke, 2006). Based on previous studies, acrolein scavenger hydralazine can mitigate acrolein-medicated cellular, cytotoxicity (Hamann et al., 2008a; Hamann and Shi, 2009) and damaged spinal cord by compression injury (Luo et al., 2005b). Although we do not know yet the degree to which acrolein involve in SCI-induced neuropathic pain behavior, acrolein scavenger hydralazine may provide effective therapeutic strategy for people who are suffering from neuropathic pain after SCI through scavenging acrolein and acrolein- protein adducts. The goal of this chapter is to investigate the hypothesis that acrolein can increase excitability of SCI-derived DRG sensory neurons, and hydralazine can attenuate the neuropathic pain behavior via SCI-induced acrolein and acrolein-protein adducts scavenging.

3.2 Materials and Methods

Animal

See Chapter 2

Contusion spinal cord injury rat model

See Chapter 2

Hydralazine application

See Chapter 2

Neuropathic pain mechanical allodynia test

See Chapter 2

56

Acrolein micro-injection

See Chapter 2

Isolation of spinal cord

See Chapter 2

Protein immunoblotting

See Chapter 2

Neuropathic pain cold allodynia test

Cold allodynia was measured using 100% acetone-induced cooling test. Animals were placed on the metal mesh in the transparent plastic box for at least 10 mins for acclimation. Then, 0.05 ml of acetone was applied to the animals’ plantar paw skin from a distance of 2 mm from paw surface. The acetone was applied 5 times to each paw skin and took an average a withdrawal frequency. To prevent from sensitivity to cold stimulus, we waited 5 min in between tests. The brisk withdrawal movement and/or licking and/or biting behaviors are considered to be a positive response.

For the neuropathic pain behavioral tests, to minimize the stress to the animals, the least stressful factor was applied first. Therefore, in this study, the order of the pain behavioral test was mechanical allodynia von Frey filament assay and then cold allodynia acetone-induced cold test. The tests order was maintained throughout the experimental period for all animals. In addition, we performed two different behavioral tests at interval of, at least, 20 min.

57

RNA isolation and cDNA synthesis

For the real time PCR (RT-PCR), we collected L1-L6 DRG cells from sham and SCI animals 1 week after SCI. Total RNA was isolated from samples using 1 ml of Trizol reagent (Sigma-Aldrich, St. Louis, MO). Isolated RNA was extracted by chloroform and precipitated by isopropanol. RNA concentration was determined using NanoDrop 2000c (Thermo Scientific, DE, USA). Then, the extracted RNA was dissolved in 50ul of RNase-free distilled water and stored at -80Ԩ deep freezer until in use. cDNA was synthesized by iScript™ cDNA Synthesis kit manual guide (BIO-RAD,170-8890,USA) using one microgram of total RNA from each sample.

Real-time PCR

Primers were designed for RT-PCR based on the previous study (Nozawa et al., 2009). To recognize the trpa1 channel 5′-TCCTATACTGGAAGCAGCGA-3′, and 5′- CTCCTGATTGCCATCGACT-3′; 18S was used as an internal control 5′- CGGCTACCACATCCAAGGAA-3′ and 5′-GCTGGAATTACCGCGGCT-3′. The accumulation of PCR products were measured by the level of iQ™SYBR Green Supermix (BIO-RAD, 170-8880, USA) fluorescence following manufacturer’s manual on a 7300 Read Time PCR System (Applied Biosystems). The interested gene expression level (trpa1) was normalized by the expression of 18s expression. Relative quantification was calculated as X = 2 – ΔΔCt, where ΔΔCt = ΔE−ΔCt and ΔE = Ct exp – Ct18s, ΔCt = Ct control – Ct 18s (Livak and Schmittgen, 2001). Data were then normalized to the average of the control groups.

Dissociation of DRG cell and preparation

Based on the previous study, the L3-L6 DRG cells were dissociated acutely (Ma and LaMotte, 2005). We collected DRG cells from sham control and SCI group 3 weeks after SCI. These DRG cells were treated with collagenase A and collagenase D in HBSS

58 for 20 min (1 mg/ml; Roche Applied Science, Indianapolis, IN), then treated by papain (30 U/ml, Worthington Biochemical, Lakewood, NJ) in HBSS containing 0.5 mM EDTA and cysteine at 35 °C. The cells were placed in culture media, which contains 1 mg/ml bovine serum albumin (BSA), trypsin inhibitor (Worthington Biochemical, Lakewood, NJ), Ham’s F-12 mixture and DMEM, 10% fetal bovine serum (FBS), penicillin and streptomycin (100 µg/ml and 100 U/ml) and N2 (Life Technologies). The cells were placed on coverslips which coated with poly-L lysine and laminin (BD Biosciences) and incubated for 2-3 hours, then additional culture media was provided to the wells. Cells were cultured for 12-15 hours at 37°C with 5% CO2.

Intracellular Ca2+ imaging

To investigate the intracellular Ca2+ imaging, DRG cells were in a balanced sterile salt solution (BSS) [NaCl (140 mM), Hepes (10 mM), CaCl2 (2 mM), MgCl2 (1 mM), glucose (10 mM), KCl (5 mM)] for 25 min at room temperature with Fura-2 AM (3 µM, Molecular Probes/Invitrogen Corporation, Carlsbad, CA). After rinsing with the BBS, the cells were placed onto the inverted microscope chamber. MetaFluor software (Molecular Devices Corporation, Downington, PA) and digital video microfluorometry with an intensified CCD camera coupled to a microscope were used for measuring Intracellular calcium level. We selected the Fura-2 (340/380 nm) excitation wavelengths by a filter changer, and a 150W xeonnon arc lamp were used for cell illumination. Before adding acrolein, sterile solution was added to the cells, and if cell responded to buffer only, this cell was not used for neuronal responsive counts. It there was no response to buffer alone, 250 µM of acrolein was applied into the coverslip directly. We waited for 2 min then washed out acrolein solution, if there was no response. After washing out acrolein, 3 μM of capsaicin was applied.

59

Electrophysiology

Electrophysiology data recoding were performed 12-18 hours after cell dissociation. DRG-contained coverslips were move to a recoding chamber which is on the stage of an inverted microscope (Nikon Eclipse Ti, Nikon Instruments Inc., Melville, NY). A bath solution (NaCl 120mM, KCl 3mM, CaCl2 1mM, MgCl2 1mM, Hepes 10mM, Glucose 10mM) was used for chamber perfusion, and adjusted to a pH 7.4 with an osmolarity of 300 mosM. All electrophysiology data were obtained at room temperature. Electrodes for intracellular recording were made from borosilicate glass (World Precision Instruments, Sarasota, FL) and pulled on a Flaming/Brown micropipette puller (P-98, Sutter Instruments, Novato, CA). Then, filling with 1.0 M KCl (impedance: 40-80 MΩ), electrodes were positioned by a micromanipulator (Newport Corporation, Irvine, CA). For bridge-balance the electrode resistance, we used -0.1 nA current injection. The small (≤ 30 µm), medium (31–45 µm) and large (≥ 45 µm) diameter neurons were used for electrophysiological recordings with continuous current cpamp in bridge mode using an AxoClamp-2B amplifier, stored digitally via Digidata 1322A interface, and analyzed offline with pClamp 9 software (Axon Instruments, Union City, CA). The neuronal cells that showing a resting membrane potential (RMP) more negative than -45 mV were considered to be a positive data. For the isolated neuron studies, we used a continuous recording for 1 min without any stimulus. Action potential data were obtained by injecting current steps of 1 sec duration via the intracellular recoding electrode with 0.1 nA increase. This was maintained until generating one or more action potentials, or reaching 2 nA.

The current threshold, which is the minimum injected to evoke a single action potential was assessed to investigated the effect of SCI itself. During the test, if there is spontaneous discharge, the sensory neuron was considered as spontaneous activity. To measure the excitability from small and medium diameter sensory neurons, 1s current pulses were injected into the soma every 30 sec. Current was adjusted for eliciting 3-4 action potentials per current injection under baseline conditions. 250 µM of acrolein was

60 added to the coverslip after three control current injections, which was continued every 30 sec. We measured the neuronal excitability by comparing the number of action potentials per injected current before and after adding acrolein.

Statistical analysis

One-way ANOVA was used to identify the source of significant interactions at each time point, followed by Tukey post hoc tests. For comparing only two conditions, The Student T-test was used. Yates-corrected chi-square test was used for differences in the incidence of neurons responding to acrolein. For the RT-PCR data analysis, un-paired t-test was used. All statistical analyses were performed using IBM-SPSS Statistics version 19.0 (SPSS inc., an IBM company, Chicago, IL, USA). P < 0.05 was considered statistically significant and the averages were expressed in mean + SEM.

3.3 Results

Persistent elevation of acrolein protein level after traumatic SCI

The acrolein-lysine expression level was measured by using immunoblotting (Figure 16). We collected about 1 cm of spinal cord tissue including damage site from animals 1 day after and 2 weeks after SCI. Acrolein-lysine expression level in spinal cord was significantly increased at both 1 day and 2 weeks after SCI. Specifically, the acrolein-lysine level in 1 day after SCI was 39.43 ± 2.46 and that in 2 weeks after SCI was 30.38 ± 2.71, while sham control group showed 9.68 ± 4.67, P < 0.001. This data supported that acrolein-lysine adducts level was persisted at least 2 weeks after SCI, which was significantly higher than that of control.

61

TRPA1 mRNA gene expression level in the DRG after SCI

To investigate the potential effect of acrolein on generation of neuropathic pain after SCI, we measured TRPA1 mRNA expression level via Real Time PCR (RT-PCR). For this study, we collected L1-L6 DRG cells from sham control and SCI 1 week after SCI (Figure 17). The data showed that TRPA1 mRNA level was significantly increased in DRG cells (2.03±0.11 fold) compared to the sham control group in 1 week after SCI. This data demonstrated that SCI could affect the expression level of TRPA1 channel, which may be related to increased-acrolein level.

Increase of [Ca2+]i in DRG sensory cells by acrolein after SCI

We have demonstrated that there were increase in acrolein level in spinal cord tissue and TRPA1 gene expression level SCI. Next, we investigated whether the exogenous acrolein can induce the increase of intracellular calcium level in DRG cells, which were collected from sham and SCI animals (Figure 18). The data indicated that about 38% of DRG sensory cells from sham control responded to acrolein alone, while about 14% of DRG cell from sham control responded to capsaicin and acrolein mixture. However, the percentage of DRG sensory cells that responded to acrolein from SCI groups was 58% (54/93) this was significantly higher than that in sham group (P < 0.005, Chi-square with Yates correction). There was no change in percentage of cell that responded to acrolein and capsaicin mixture compared to those in sham control group (17%). In addition, there was no change in the number of cells that responded to the capsaicin alone between sham control group (31%, 32/103) and SCI group (28%, 26/93 p > 0.05). This data demonstrated that, after SCI, there was an increased number of DRG sensory neurons, which are sensitive to acrolein.

62

Figure 16. Elevation of Acrolein-Lysine Adduct Level after SCI

Immunodotblotting data indicated that SCI could result in the increase of acrolein-lysine adduct level in the spinal cord both 1 day and 2 weeks after SCI. Even though the acrolein-lysine level reached the highest points 1 day after SCI, its expression level was persisted 2 weeks after SCI. The acrolein-lysine adduct levels were 9.68 ± 4.67, 39.43 ± 2.46, and 30.38 ± 2.71 in sham, 1 day after SCI, and 2 weeks after SCI respectively. All value were expressed as mean ± SEM and N=6 in each group.

63

Figure 17. TRPA1 Gene Expression Level in DRG Cells

Relative TRPA1 gene expression level was quantified by using RT-PCR. We collected L1-L6 DRG cells 1 week after SCI. The interested gene level was normalized by the expression level of internal control (18s) and compared to the cycle threshold (Ct) value of 18s. This data indicated that there was about two folds increase in TRPA1 gene expression level in DRG cells from SCI. Paired t-test was used for the statistical analysis and N=4 in each group.

64

Figure 18. Calcium Imaging of TRPA1 Signaling via Acrolein in DRG

A). Representative recording of a transient intracellular calcium increase in a DRG sensory neuron dissociated from a rat, following SCI, in response to acrolein (250 µM). The cells were loaded with Fura-2 (3 µM, Molecular Probes/Invitrogen Corporation, Carlsbad, CA). Intracellular calcium was measured by digital video microfluorometry. B). Calcium imaging data indicating the percentage of neurons that respond to acrolein and capsaicin with an intracellular calcium flux in naive and SCI dissociated DRG sensory neurons. Following acrolein application, capsaicin (TRPV1 agonist) was added to further characterize the imaged cells as nociceptive neurons. In naïve rats, 38% of neurons exhibited an acrolein-induced [Ca2+]i flux (n=103) where approximately one third of the neurons also exhibited a capsaicin-induced [Ca2+]i flux (36%). In rats that were subjected to SCI, a large percentage of neurons displayed an acrolein-induced [Ca2+]i flux (58%, n=93). This is significantly higher than what is observed in naïve rats (∗ P < 0.005, Chi-square with Yates correction). Among the SCI rats that responded to acrolein, 29% of them also exhibited a capsaicin-induced [Ca2+]i influx. Therefore, the proportion of capsaicin responsive cells among acrolein responsive cells is largely unchanged (P > 0.05).

65

SCI can decrease the current thresholds in DRG sensory neurons

Next, we tested if SCI itself can affect the sensory neuron’s activation state. For the Electrophysiological current clamp recordings, we used DRG cells from SCI and sham control group (Figure 19). To get the current threshold (CT), the minimum current was injected to evoke single action potential. The amount of current to generate single action potential in small diameter sensory neurons from SCI group was significantly lower than that in sham control group. It were specifically 0.9 ± 0.1 nA for sham injury and 0.4 ± 0.1 nA for SCI (p<0.05). In addition, medium diameter from SCI group also showed a significantly reduced amount of current that was needed to generate single action potential compared to DRG sensory neurons from sham control group (1.1 ± 0.1 nA for sham injury vs. 0.6 ± 0.1 nA for SCI, p<0.05). However, there were no differences in amount of current using large diameter of sensory neurons between SCI and sham control group. Specifically, the mean current threshold for large DRG neurons for SCI group was 2.10 ± 0.33 nA (n = 15) and 1.99 ± 0.17 nA for sham-injury animals (n =20). This data demonstrated that SCI alone can change and increase the overall state of small and medium diameter sensory neurons.

Increased excitability of DRG neuron after a combination of SCI and acrolein.

Through this study, we have demonstrated that there were an increased number of cells that responded to acrolein after SCI, and SCI alone could reduce the current threshold to evoke single action potential in DRG sensory neuron. This time, we investigated whether the acrolein could increase the sensitivity of single sensory neuron. For this test, we collected DRG cells from sham control and SCI group. We tested sensory cells’ response using sharp electrodes in current clamp mode to investigate the degree to which acrolein could induce the excitability in single sensory neuron level. During the study, acrolein did not induce the spontaneous activity in all tested neurons. We could measure the increased excitability in small and medium diameter of sensory

66 neurons when compared with baseline levels in DRG sensory neurons from sham control and SCI group, after combination of repeated current injections with acrolein adding (Figure 20). Specifically, the mean number of action potential by injecting minimum current in sensory neuron from sham control group was 1.9 ± 0.1 APs (n=12) for control and 4.7 ± 1.0 APs following acrolein application (n=3) (Figure 20B). However, the average number of action potentials which can be generated minimal current injection in sensory neuron from SCI group was 2.3 ± 0.4 APs for control (n=14) and 8.0 ± 0.9 APs after adding acrolein (n=5) (Figure 20B). In both sham control and SCI group, application of acrolein could increase the sensory neurons’ excitability compared to controls (p<0.05). Furthermore, the increased excitability by acrolein in sensory neuron was greater in SCI group compared with sham control (Figure 20B). This data indicated that acrolein could increase the neuronal cells’ sensitivity, and acrolein-mediated neuronal excitability was significantly enhanced in SCI group.

67

Figure 19. SCI-mediated Current Threshold to Generate Action Potential

SCI alone can decrease the current threshold (CT), which is minimal current to generate single action potential in small and medium diameter sensory neurons. Lumbar DRG cells from sham and SCI group were used for current clamp recording. A). CT value in small diameter sensory neuron was reduced in SCI compared to DRG cells from sham group (A2 vs. A1). The current threshold was 0.9 ± 0.1 nA for sham injury and 0.4 ± 0.1 nA for SCI, p<0.05 (B). C) CT value in medium sensory neuron also decreased following SCI (C1 vs. C2). Specifically, amount of injected current was 1.1 ± 0.1 nA for sham injury and 0.6 ± 0.1 nA for SCI, p<0.05. There was no difference in CT using large diameter of sensory neuron between sham and SCI group. The number means the number of cells. Student’s t-test was used for the statistical analysis.

68

Figure 20. Acrolein-Induced Sensory Neuron’s Excitability after SCI.

The number of action potentials by injected minimal current was measured to investigate the acrolein-mediated excitability 2 weeks after SCI. We used small and medium diameter of DRG cell from L3-L6 for the current clamp recordings. Using a 1 second depolarizing current injection (ranging from 0.1 to 2.0 nA depending on the cell) every 30 seconds, two or four action potentials were generated. A) Representative recording for application of 250 µM acrolein could increase the number of generation of action potentials in DRG sensory neurons from SCI group. B) Acrolein application can significantly increase the DRG neurons’ excitability in both sham control and SCI group (n=12 for sham injury controls, and n=14 for SCI controls, ANOVA *p<0.05,). Furthermore, the acrolein-induced sensitivity was greater in SCI compared to sham group.

69

Micro-injected acrolein into the spinal cord produces neuropathic pain behavior.

We investigated whether the acrolein alone could generate the neuropathic pain- like behavior without any mechanical injury. We injected a small amount of acrolein into the rat’s spinal cord dorsal horn directly on animal’s right side of spinal cord. For the control, the same amount of saline was injected on the left side of spinal cord. The data showed that micro-injected acrolein into the spinal cord could produce the mechanical allodynia and cold allodynia on the side of injection site (Figure 21). However, application of saline to the contralateral side did not show any specific neuropathic pain behaviors.

Hydralazine can attenuate the neuropathic pain behavior after SCI

Next, we tested the analgesic effects of acrolein scavenger hydralazine following SCI. 5mg/kg of hydralazine was injected via I.P injection everyday for two weeks. The results data showed that hydralazine could significantly alleviate the mechanical and cold allodynia compared to hydralazine non-treatment group during the experiment period (Figure 22A and 22B). Furthermore, the acrolein-lysine adduct level was significantly decreased in hydralazine treatment group when compared to the SCI only group (Figure 22C, p<0.05). This data demonstrated that acrolein scavenger hydralazine could attenuate neuropathic pain behaviors after SCI via acrolein scavenging way.

Delayed application of hydralazine reduces neuropathic pain after SCI

So far, to investigate the analgesic effect of hydralazine, we have injected hydralazine right after SCI within 5 min. However, clinically, applying medication right after SCI is impossible. In this sense, we tested whether the delayed application of acrolein scavenger hydralazine could attenuate neuropathic pain behaviors after SCI. So, the same amount of hydralazine (5mg/kg) was injected 2 weeks after SCI for 5 weeks,

70 then we did mechanical and cold allodynia tests every week until 7 weeks after SCI. The data indicated that the delayed beneficial effects of hydralazine also altered neuropathic pain behaviors after SCI (Figure 23A). This treatment paradigm significantly attenuate mechanical allodynia from 35 days to the end of period (p<0.05), and cold allodynia from 21 days to end of behavioral test (Figure 23B, p<0.05).

71

Figure 21. Micro-injected Acrolein Produced Neuropathic Pain Behaviors

The same amount of acrolein (40nmol, 1.6 µL) and saline (1.6 µL) was injected into the spinal cord directly, which could change the responses to mechanical and cold stimulus. A) Mechanical threshold in acrolein injection side significantly lower than that in saline injection side up to 10 days. B) Withdrawal frequency to 100% acetone also significantly increase compared to saline injection side. These data showed that acrolein alone could induce the neuropathic pain behaviors without any traumatic damage to the spinal cord. One way ANOVA and Tukey’s test were used for statistical analysis. (∗ P < 0.05 when compared to control group, n=4 in all groups). All data was expressed as mean + SEM.

72

Figure 22. The Effect of Hydralazine on Pain Behaviors and Acrolein Level after SCI

A and B) Daily application of hydralazine for two weeks could attenuate the SCI-induced neuropathic pain mechanical and cold allodynia. On days from 14 to 28 after SCI, only SCI group was sensitive to mechanical and cold stimulus compared to sham control and SCI treatment with hydralazine (# P < 0.05 when compared to the sham injured control and ∗ P < 0.05 when compared to SCI alone). C) Acrolein-lysine adduct level was significantly reduced by acrolein scavenger hydralazine in one day after SCI. (n=6 in each condition). (∗P < 0.05, ∗∗P < 0.01, ANOVA). All data (A-C) were expressed in mean + SEM.

73

Figure 23. Delayed Application of Hydralazine on Neuropathic Pain after SCI

To investigate the effect of delayed application of hydralazine, 5mg/kg of hydralazine was injected 2 weeks after SCI for 5 weeks every day. Mechanical and cold allodynia were assessed weekly until 49 days after SCI. A) The mechanical threshold in hydralazine treatment group significantly lower than SCI only at 35 days after SCI until 49 days after SCI. B) Like mechanical allodynia, withdrawal frequency to acetone in hydralazine treatment group was significantly higher than that in SCI only at 21 days after SCI. These data showed that delayed application of hydralazine also attenuated SCI- induced hypersensitivity. (∗ P < 0.05 when compared to SCI group, n=4 for all groups). All data were expressed as mean + SEM.

74

3.4 Discussion

This study has demonstrated that SCI-injuced acrolein level could be persisted at least two weeks at the damaged spinal cord tissue. This SCI-induced generation of acrolein correlated with the onset of neuropathic pain mechanical and cold allodynia. In addition, our data demonstrated that SCI could induce a reduction of current threshold to generate action potential and an increase in the number of DRG sensory cells that responded to acrolein. In addition, exogenously injected acrolein into the spinal cord could produce the neuropathic pain mechanical and cold allodynia without any traumatic mechanical damage to the spinal cord. Furthermore, application of acrolein to DRG sensory neurons could increase the excitability of small and medium diameter of sensory neurons following SCI. Taken together, these data strongly support that SCI-induced acrolein plays a critical role in generation of neuropathic pain behavior as a pro- nociceptive factor.

Through this study, we observed increased level of acrolein after SCI may implicate in generation of neuropathic pain by increasing sensory neurons’ excitability in lumbar level of DRG. Because of acrolein’s nature such as diffusion cross membrane and circulation through the system, elevated acrolein in thoracic level of spinal cord could spread out throughout the spinal cord via subarachnoid space (Luo et al., 2005b). In addition because of the permeability of the DRG cells, SCI-induced acrolein could directly increase the sensitivity of sensory neurons (Devor, 1999; Abram et al., 2006; Puljak et al., 2009).

As a pro-nociceptive factor, acrolein can induce an increase of influx of calcium ion into the cell through TRPA1, which can also activate phospholipase C (PLC) signal transduction (Bautista et al., 2006; Dai et al., 2007). TRPA1 channel can also promote the pain sensitivity through other stimuli (Lennertz et al., 2012; Zhao et al., 2012; Perin- Martins et al., 2013).

75

There is little information about the changes in TRP channel’s expression level after SCI. Our data showed that there was an increase in the expression level of TRPA1 gene after SCI. In this sense, TRPA1 channel could be a potential target for analgesic medication. According to previous studies, TRPA1 antagonists could be applied as an effective analgesic by blocking the pain transmission (Eid et al., 2008; Andre et al., 2009; Bessac et al., 2009; Kerstein et al., 2009). However, TRPA1 antagonists for SCI-induced neuropathic pain as analgesic remained uncertain and many of these antagonists are needed to be studied further for clinical application.

In this sense, as an endogenous TRPA1 agonist, acrolein could be a more reliable therapeutic target after SCI. Many of studies have shown that acrolein has been implicated in many types of neurodegenerative diseases by damaging to the CNS, proteins, lipids and DNA (Shi et al., 2002b; Hamann et al., 2008b; Leung et al., 2011c; Shi et al., 2011). With acrolein’s nature and as a pro-inflammatory factor, it can contribute to the generation of chronic neuropathic pain after SCI.

The role of acrolein in generation of neuropathic pain was assessed with acrolein scavenger hydralazine. The data showed that systemic application of hydralazine after SCI could reduce not only the acrolein-lysine adduct level, but also neuropathic pain mechanical and cold allodynia. In addition, delayed application of hydralazine also showed its analgesic effect after SCI. However, we also need to consider about the role of hydralazine on nervous system. As the FDA approved hypertensive medication, hydralazine promotes the increase of cGMP level in smooth muscle resulting in relaxation of blood vessels. However, one previous study has demonstrated that cGMP- PKG signaling pathway agonist could promote the sensory neuron’s excitability via in vitro experiments (Song et al., 2006; Zheng et al., 2007). Based on these results, there is a possibility that hydralazine application could also promote the neuropathic pain-like behaviors such as mechanical and cold allodynia in sham control group. But, we did not check any hydralazine-induced neuropathic pain like behavior in both hydralazine

76 treatment groups. Therefore, an analgesic effect of hydralazine is correlated with scavenging of acrolein after SCI.

There could be other TRP channels that respond to cold stimulus resulting in SCI- induced neuropathic pain other than TRPA1. TRPM8 is the sub-family of TRP channels, which is activated by cold and cold effect-generated molecules such as menthol and icilin (McKemy et al., 2002; Patapoutian et al., 2003; Patapoutian et al., 2009). In addition, some of TRPA1 agonists have been shown to activate TRPM8 (Story et al., 2003). According to the previous studies, TRPM8 has been implicated in sciatic nerve injury- induced sensitivity to cold stimulus and tissue inflammation (Colburn et al., 2007; Knowlton et al., 2010). However, recent studies have demonstrated that TRPM8 gene expression level in lumbar DRG cells in significantly lower than TRPA1 gene expression level under naïve conditions (Vandewauw et al., 2013). Furthermore, acrolein is known as TRPA1 specific agonist which support that the analgesic effect of hydralazine attenuates acrolein-induced neuropathic pain, which was mediated via TRPA1 not TRPM8.

3.5 Conclusion

Our data supports that acrolein can enhance the neuronal sensitivity after SCI, which is known as pro-nociceptive TRPA1 agonist and pro-inflammatory factor. With evidence that elevation of acrolein level after SCI until two weeks, acrolein could be a valuable therapeutic target for neuropathic pain. In addition, DRG neurons remain hypersensitivity to stimuli because of persistent up-regulation of TRPA1 channel level after SCI. In this sense, anti-acrolein strategy using hydralazine could be an effective way to attenuate not only the initial state (acute) of neuropathic pain, but the chronic stage of neuropathic pain following SCI.

77

CHAPTER 4. THE EFFECT OF TRPA1 CHANNEL IN THE PERIPHERAL TERMINAL AFTER SPINAL CORD INJURY

4.1 Background and Research Scope of This Study

Noxious mechanical and thermal stimuli can act on the peripheral terminals of primary sensory afferent nerve. The conversion of mechanical, chemical and thermal stimuli into electrical signal and signal transmission are conducted by transient receptor potential (TRP) channels (Puljak et al., 2009). Basically, TRPs are non-selective cation calcium ion permeable channels which comprise six transmembrane segments (S1-S6) and ion permeable pore is located between S5-S6 (Bandell et al., 2007; Nilius et al., 2012; Bautista et al., 2013).

The transient receptor potential ankyrin 1 (TRPA1) channel is a member of the TRP channel family and also calcium permeable non-specific cation channel. As the only TRPA subfamily member, TRPA1 channels are expressed in both neuronal cells such as unmyelinated primary sensory neuron C fiber, DRG cell body and trigeminal ganglia neurons and non-neuronal tissues such as respiratory tract, gastrointestinal tract, bladder, and keratinocytes (Patapoutian and Macpherson, 2006; Patapoutian et al., 2009; Nilius et al., 2012; Bautista et al., 2013). Based on channel’s ability to respond to noxious stimulus, pain sensing channels can be divided into unimodal, which are activated by a unique stimulus and polymodal, which are stimulated by various stimulus such as environmental irritants, mechanical, and thermal stimuli (Dhaka et al., 2006). TRPA1 channels are activated in a polymodal fashions. That is, TRPA1 can be activated by environmental irritants (acrolein), thermal sensation (cold), pro-inflammatory factors (bradykinin) and mechanical sensation (mechanical). The N-terminal and C-terminal of TRPA1 channel

78 are located in the intracellular part. Specifically, the long ankyrin-repeated domain at N- terminal plays a critical role for activation of TRPA1 channel. EF-hand motif on ankyrin- repeated domain where calcium ion can bind to and activate TRPA1 channel, and cystein and lysine neucleophilic amino acid on here, which are the targets for electrophilic TRPA1 agonists contribute to gate the TRPA1 channel as well (Clapham et al., 2001; Bandell et al., 2007).

According to the previous study, TRPA1 channel is known for endogenous byproducts of oxidative stress process such as lipid peroxidation (LPO) (Andersson et al., 2008). Byproduct of LPO after neuronal damage, acrolein is known as TRPA1 specific agonist because of its electrophilic properties. With these properties, β-carbon of the a,β- unsaturated aldehyde, acrolein can bind to nucleophilic thiol group of cystein and ε- animo group of lysine residues via covalent modification. Especially, reaction between acrolein and cystein is reversible, so acrolein can break from acrolein-cystein adduct and react with other proteins and DNA (Hinman et al., 2006; Bandell et al., 2007; Nilius et al., 2007; Andrade et al., 2008; Andre et al., 2009; Patapoutian et al., 2009). This covalent modification between TRPA1 channel and electrophiles contribute to gate of TRPA1 channel and transmit pain signaling.

TRPA1 channels are also activated by non-electrophile agonists G-protein coupled receptor indirect pathway (Bautista et al., 2005; Bautista et al., 2006). Bradykinin (BK) is known as pro-inflammatory factor. When it binds to its receptor bradykinin 2 receptor (B2R), it can activate phospholipase C (PLC) signaling and promote the release of calcium ions from intracellular calcium store. Released-calcium ions can bind to EF-hand domain which is located on N-terminal of TRPA1 activate TRPA1 channel. Acrolein is also known as pro-inflammatory factor (Bautista et al., 2006). As an inflammatory factor, acrolein has been shown to stimulate the release of chemokines such as monocyte chemotactic protein-1 (MCP-1 or CCL2) (Knerlich-Lukoschus et al., 2008; Knerlich- Lukoschus et al., 2011). In the nervous system, MCP-1 is expressed in nociceptive sensory neurons which are densely colocalized with TRPA1 channels (Abbadie, 2005;

79

Abbadie et al., 2009). In addition, MCP-1 is released at the postsynaptic spinal dorsal horn to promote the activity of neuronal glial cells that express MCP-1 receptor CCR2 (White et al., 2005; White et al., 2009). Based on previous study, acrolein-induced MCP- 1 and CCR2 activation can result in induce the neuropathic pain-like behaviors through transactivation of TRPA1 channel (Knerlich-Lukoschus et al., 2010). In this sense, with acrolein’s diffusive nature, stable and prolonged presence in the body, acrolein can serve as a pro-inflammatory factor that contributes to generation of chronic neuropathic pain after SCI by either directly or indirectly.

TRPA1 pore which is located between S5-S6 is stimulus dependent. Under the neuropathic pain condition, it has a wide size of pore compared to normal physiological condition, which induces the large number of calcium ion influx (Patapoutian and Macpherson, 2006; Bandell et al., 2007; Patapoutian et al., 2009; Nilius et al., 2012). In chapter 3, we have demonstrated that the expression level of TRPA1 mRNA was increased in the DRG after SCI. Based on this result, we can speculate that pain can promote not only the TRPA1 channel expression, but also the pore size of TRPA1 channel. As we already mentioned above, TRPA1 channels are expressed on nociceptive C fiber, specifically, peptidergic primary afferent neuron. So, when TRPA1 channels are activated, they can activate the releases of nociceptive neurotransmitters such as substance P and glutamate to higher order sensory neurons (Bautista et al., 2006; Macpherson et al., 2007). In this sense, TRPA1 channels on the central ending of primary afferent neurons in the spinal dorsal horn may play a critical role in the generation and maintenance of pain signaling. However, there is little information about the role of TRPA1 channels on the peripheral terminal of primary afferent sensory fiber. Therefore, main goals of this study are to demonstrate the role and effect of TRPA1 channels in the peripheral terminal of primary afferent nerve after SCI, and to investigate the effect of acrolein after SCI whether it can promote the increase of sensitivity of responsiveness at the peripheral terminal nerve. Lastly, we will investigate that the scavenging strategy using acrolein scavenger hydralazine to check whether hydralazine treatment could alleviate the TRPA1 expression level with various animal tissues such as DRG, dorsal

80 horn of spinal cord and paw skin at 1 week and 2 weeks after SCI, and attenuate the neuropathic pain behavior after SCI.

In this study, we assessed the involvement of acroelin in neuropathic pain after SCI by using immunoblotting and urine 3-HPMA techniques and investigated the effect of TRPA1 channels on peripheral terminal of a primary afferent nerve after SCI by using RT-PCR with various tissues and time points.

4.2 Spinal Cord Injury-induced Neuropathic Pain Mechanical Allodynia

Through chapter 2 and 3, we have demonstrated that spinal cord injury could lead to neuropathic pain behaviors. Here we again confirmed that NYU impactor induced contusion injury at spinal cord T-10 level resulted in neuropathic pain mechanical allodynia (figure 24). 2 weeks after SCI, the mechanical threshold in SCI group was significantly lower than that in sham control group.

81

Figure 24. Mechanical Sensory Hypersensitivity after Moderate Contusive SCI

Mechanical sensory status was determined using Von Frey filament performed before the injury, and weekly post injury starting two weeks post SCI for up to 5 weeks. Note that there were no difference in mechanical threshold in sham injury and SCI groups before the injury. However, starting from 14 days post injury, the average values of the mechanical threshold in SCI group were significantly reduced compared to sham group in the 2-5 weeks post SCI. N=10~13 in each group. **P < 0.01. One-way ANOVA and Tukey tests were used for statistical analysis. All values were expressed as mean ±SEM

82

4.3 Changes in TRPA1 Gene Expression Level after SCI

In chapter 3, we have demonstrated that there was increased in TRPA mRNA gene expression level in lumbar level of DRG 1 week after SCI. In this chapter, we assessed TRPA1 gene expression level with various tissues such as DRG, dorsal horn of spinal cord and hind limb paw skin in 1 week and 2 weeks after SCI. For the RT-PCR, we collected L1-L6 DRG cells, dorsal horn of spinal cord including T-10 injury site, and hind-limb paw skin from sham control group and SCI group. As the result is shown in figure 25, the TRPA1 gene expression level was significantly increased in all tissues at 1 week and 2 weeks after SCI, however, there was no difference between 1 week and 2 weeks after SCI in dorsal horn of spinal cord, DRG, and paw skin. Specifically, TRPA1 gene expression level in spinal dorsal horn was 2.58±0.16 for 1 week and 2.57±0.43 for 2 weeks after SCI (p>0.05), in DRG (2.03±0.22 vs 1.87±0.21, p>0.05), and in paw skin (41.45±2.99 vs 38.48±2.02, p>0.05). This data indicated that TRPA1 gene level could be persisted at least 2 weeks after SCI.

Next, 5mg/kg of acrolein scavenger was injected for 1 week and 2 weeks after SCI daily to investigate the benefits of hydralazine on reduction of TRPA1 channel gene expression level in various tissues after SCI. Figure 26 showed the gene expression level 1 week after SCI in dorsal horn, DRG, and paw skin. The data indicated that systemic application of hydralazine could attenuate gene expression level in all tissues. Specifically, the gene expression level induction folds were reduced from 2.6±0.15 to 1.0±0.13 in dorsal horn of spinal cord, from 2.0±0.22 to 1.2±0.20 in DRG, and from 41.4±2.98 to 0.60±0.11 in hind limb paw skin. Furthermore, figure 27 showed that even 2 weeks after SCI, TRPA1 levels in SCI group were significantly higher than those in sham control. Specifically, Gene expression level 2 weeks after SCI were significantly increased in dorsal horn (2.57±0.45), DRG (1.27±0.22) and paw skin (38.48±2.0) following SCI. However, acrolein hydralazine could significantly reduce the TRPA1 gene expression level dorsal horn (0.71±0.10), DRG (0.59±0.01) and paw skin (15.25±3.12). These data demonstrated that, combined with immunoblotting and urine 3-

83

HPMA results, acrolein was involved in an increase in TRPA1 gene expression level in both peripheral and central terminal of a primary afferent nerve especially, SCI could contribute to the increase in expression level of TRPA1 at the peripheral terminal of primary sensory neurons. In addition, as a TRPA1 channel specific agonist, acrolein could be a promising therapeutic target for neuropathic pain treatment after SCI.

84

Figure 25. The Elevation of TRPA1 mRNA Level 1 Week and 2 Weeks after SCI

Bar graph indicates the RT-PCR relative quantification of TRPA1 mRNA examined 1 and 2 weeks post SCI at three tissue types: Dorsal horn (1 cm long including T10), DRG (L1-L6), and paw skin (1 cm x 1cm). Gene expression was normalized by the expression of 18 s and compared to the cycle threshold value for 18 s of tissue mRNA post SCI. There was no difference of RT-PCR mRNA level of TRPA1 between two time points (1 week vs 2 weeks) in each tissue type. However, significant difference were detected regarding the RT-PCR mRNA level of TRPA1 between SCI and sham group in three tissue types at both 1 week and three week. Note near forty fold increase of TRPA1 mRNA in paw skin presented 1 and 2 weeks post SCI. ∗ P < 0.05, ∗∗∗ P < 0.005 when compared to sham. One-way ANOVA and Tukey test. All values were expressed as mean ± SEM.

85

Figure 26. The Effect of Hydralazein on TRPA1 mRNA Level 7 Days after SCI

The suppression of TRPA1 mRNA level by acrolein scavenger hydralazine in spinal dorsal horn (1 cm long including T10), dorsal root ganglia (DRG, L1-L6), and paw skin assessed 7 days after SCI. Specifically, the TRPA1 mRNA levels were significantly increased in dorsal horn, DRG, and paw skin following SCI (P < 0.05 in dorsal horn and DRG, and P < 0.001 in paw skin group when compared to sham). However, this elevated TRPA1 mRNA was significantly attenuated in all three tissue types with the continuous daily IP injection of hydralazine for a week post SCI. (∗ P<0.05, ∗∗∗ P<0.005 when compared to SCI alone, One-way ANOVA and Tukey test). N=4~8 in each group. All values were expressed as mean ±SEM.

86

Figure 27. The Effect of Hydralazein on TRPA1 mRNA Level 14 Days after SCI.

The suppression of TRPA1 mRNA level by acrolein scavenger hydralazine in spinal dorsal horn (1 cm long including T10), dorsal root ganglia (DRG L1-L6), and paw skin assessed 2 weeks after SCI. Specifically, the TRPA1 mRNA levels were significantly increased 14 days post SCI in dorsal horn, DRG, and paw skin following SCI (P < 0.05 in dorsal horn, DRG, and P < 0.001 in paw skin group when compared to sham). However, this elevated TRPA1 mRNA was significantly attenuated in all three tissue types with the continuous daily IP injection of hydralazine for 2 weeks post SCI. (∗ P<0.05, One-way ANOVA and Tukey test). N=4~8 in each group. All values were expressed as mean ±SEM.

87

4.4 Persistent Elevation of Acrolein Level and Onset of Neuropathic Pain Behaviors after SCI

Through chapters 2 and 3, we have shown that there were elevated level of acrolein and acrolein metabolite 3-HPMA level after SCI. In addition, we have demonstrated that as a pro-nociceptive factor acrolein may correlate with onset of neuropathic pain mechanical and cold allodynia and enhance neuronal sensitization after SCI. Furthermore, when we treated with acrolein scavenger hydalazine, it could not only attenuate acrolein- lysine adduct level and neuropathic pain mechanical and cold allodynia, but also promote locomotor functional recovery following SCI. These data indicated that considering neurotoxicity of acrolein, anti-acrolein therapy could be an effective treatment to alleviate neurological deficits and enhance functional recovery for SCI patients. Here, we again confirmed our theory by measuring acrolein level via immunoblotting, urine 3-HPMA and neuropathic pain behavior after SCI (Figure 28).

88

Figure 28. Elevation of Acroein-lysine Adducts in Spinal Cord and 3-HPMA after SCI.

A) Bar graph depicts the elevation of acrolein-lysine adduct level in spinal cord tissue after contusive spinal cord injury (SCI) at 1 day and 2 weeks post injury. Measurements through dot immunoblotting indicated that SCI was associated with significant elevation of acrolein-lysine adduct 1 day following SCI compared to sham injury group. Persistent elevations of acrolein-lysine adduct were observed for at least 2 weeks after SCI. Top: photograph shows a representative blot for each conditions. Bottom: Bar graph depicts the quantification of the bands density by Image J. N=6~8 for each group. ∗ P < 0.05, ∗∗ P < 0.01. One-way ANOVA and Tukey test. B) Elevated level of 3-HPMA, an acrolein metabolite, in urine after SCI in rats (3.5 ± 0.24 μg/mg), when compared to sham injured group (1.9±0.21 μg/mg). Rat urine samples were collected one day after SCI to determine 3-HPMA level. N=4~6 in each group. ∗P < 0.05. unpaired t-test. All values were expressed as mean ± SEM.

89

4.5 Gene Expression Level and Behavioral Alternations by Pro-nociceptive Factor Acrolein

In chapter 2 and 3, we have shown that acrolein alone could lead to neuronal tissue damage, motor and sensory function deficits without any traumatic damage to spinal cord. And also, the data indicated that SCI could result in the increase of TRPA1 mRNA expression level in spinal dorsal horn, DRG, and paw skin 1 week and 2 weeks after SCI.

In this chapter, we investigated whether acrolein itself could promote the elevation of TRPA1 gene expression levels in dorsal horn of spinal cord, DRG, and paw skin without any traumatic injury. We also have tested if acrolein could promote generation of neurological deficits such as neuropathic pain and locomotor dysfunction. Initially, certain amount of acrolein (40nmol, 1.6 µl) was injected into the T-10 level of spinal cord directly which was the same location as we described in chapter 2. Again, same amount of saline was injected into the animals’ left side of spinal cord as a control. The gene expression level was assessed using RT-PCR and tissues samples were collected 1 week after acrolein injection. Figure 29 showed the result of relative TRPA1 mRNA gene expression level in DRG, spinal cord dorsal horn, and paw skin. As the data indicated, TRPA1 gene expression levels were significantly increased in all tissues 1 week after acrolein micro-injection. Specifically, the increased gene expression levels were 83.4±5.9 folds (DRG), 56.9±8.4 folds (spinal dorsal horn) and 76.3±6.4 folds (paw skin). This data demonstrated that acrolein alone could result in elevation of TRPA1 gene expression level without any traumatic SCI. This data also further supported that acrolein played a critical role in an elevation of TRPA1 gene expression level in both peripheral and central terminal of a primary afferent nerve, and it could be the valuable therapeutic target for SCI-induced neuropathic pain patients.

90

Figure 29. The Effect of Micro-Acrolein Injection on TRPA1 mRNA Level in Tissues

Relative TRPA1 mRNA expression level was assessed in DRG, spinal dorsal horn and paw skin after direct injecting acrolein (40nmol, 1.6 µl) into spinal cord. For the control, the same amount of saline was injected into the animals left side of spinal cord. Compared to saline injection site, the gene expression levels were significantly increased in DRG (83.4±5.9), (spinal dorsal horn) 56.9±8.4, and (paw skin) 76.3±6.4 folds. This data indicated that acrolein could induce the gene expression level changes in the central and peripheral terminal of primary afferent without any traumatic mechanical injury. N=4 in each group. #P<0.005 All values were expressed as mean ±SEM

91

Next, we tested, again, the ability of acrolein as a pro-nociceptive factor. This time, the experiment period has been extended from 10 days (in chapter 3) to 14 days. To examine the effect of acrolein on neuropathic pain generation, we performed mechanical and cold allodynia neuropathic pain tests after acrolein micro-injection every 2 days until 14 days (Figure 30). As figure 30 indicated, acrolein micro-injection into the spinal cord promoted significant mechanical and cold hypersensitivity on the injection site, whereas saline injection to the contralateral side led to the decreased or absence of hypersensitivity response to mechanical and cold stimuli. Furthermore, micro-injected acrolein-induced neuropathic pain behaviors were maintained at least 2 weeks. This data supported that acrolein could generate the neuropathic pain mechanical and cold allodynia without any traumatic mechanical injury to the spinal cord.

In addition to the ability of acrolein on generation of sensory abnormalities, we tested if acrolein alone could promote the locomotor dysfunction without any mechanical injury. For this study, the experiment period also has been extended from 10 days (in chapter 2) to 14 days. We examined the animals’ locomotor ability of hind limb every two days for 14 days. As shown in figure 31, we found that the micro-injection of acrolein could produce significant locomotor malfunction on the side of acrolein injection, while there were little changes or a decreased in BBB score until first week, but it came back to normal BBB score (21) in 6 days after acrolein injection. This data suggested that acrolein alone could induce the motor deficits with no traumatic injury.

92

Figure 30. Micro-injected Acrolein Promoted the Mechanical and Cold Allodynia

We tested if acrolein itself could generate neuropathic pain without any traumatic injury to the spinal cord. Using micro-pump device, 40nmol, 1.6 µl of acrolein was injected into the right side of spinal cord and the same amount of saline was used for a control into the left side of spinal cord at T-10 level. Micro-injected acrolein produced the significantly changes in withdrawal frequency to cold stimuli (A) and mechanical threshold (B). N=6 in each group *P <0.05, **P<0.01 compared to control group. ANOVA and Tukey test were used for statistical analysis. All values were expressed as mean ±SEM.

93

Figure 31. The Motor Function Deficits by Micro-injected Acrolein

The changes of Basso, Beattie and Bresnahan (BBB) scores for acrolein (right, 40 nmol dissolved in 1.6 µL)) and saline (left) injections over the course of 2 weeks. Notice the significant reduction of the BBB score on the side of acrolein injection compared to the contralateral or saline injected side. **p < 0.01, *P < 0.05 for comparison between right side (acrolein injected) and left side (saline injected). One-way ANOVA then Tukey test. n = 6 in each group. The values are mean ± SEM.

94

4.6 Acrolein Increases the Peripheral Terminal Sensitivity Following SCI

Next, we have tested whether the acrolein could increase the peripheral terminal sensitivity following SCI. For this, initially, a volume of 25 µl of acrolein (625nmol) was injected into the intradermal plantar region of the right hind foot in sham control group and SCI group. Therefore, there were four different groups; group 1: normal rats, group 2: normal rats + acrolein injection, group 3: SCI rats, and group 4: SCI rats + acrolein injection. For the acrolein injection, a 30 gauge needle attached to a Hamilton Syringe was used (figure 32). A successful injection was considered if there was about 2 mm in diameter of bleb formation at the injection site and needle insertion site was pressed for 1 min to prevent from leakage of acrolein solution after removal of the needle. An hour after acrolein injection, for the mechanical hypersensitivity test, animals were placed inside a plastic box. Mechanical threshold was measured using von Frey filaments up- down method. As shown in figure 33A, peripherally injected acrolein resulted in the neuropathic pain-like behavior in control group. In addition, acrolein-mediated neuropathic pain behaviors were enhanced in SCI group (figure 33B). In both groups, intradermally injected acrolein contributed to the reduction of mechanical threshold from 14.26±0.28 g to 10.65±1.22 g in control group (p<0.05), and from 5.42±0.41 g to 0.82±0.24 g in SCI group (p<0.01). However, acrolein-mediated neuropathic pain effect was enhanced in SCI group. Specifically, the percentage of change of mechanical threshold after acrolein injection was 25.53±10.21% in sham control group and 82.44±1.21% in SCI group (figure 33B). This data demonstrated that intradermally injected acrolein could generate neuropathic pain-like behavior in control group and acrolein-mediated effect was greater in SCI group compared to control group.

95

Figure 32. Sites for Acrolein Injection and Mechanical Allodynia Von Frey Test

For the acrolein injection (25 µl, 625nmol), was injected using 30 gauge of needle. Initially, the needle was inserted at the site of A and solution was injected at the site of B. For the mechanical hypersensitivity test, von Frey filaments were applied to the site C.

96

Figure 33. The Effect of Peripheral Injection of Acrolein in Control and SCI group

Peripheral paw injection of acrolein induced neuropathic pain-like behavior and its potentiation after SCI. Mechanical sensitivity was assessed using Von Frey filament before and after the injection of acrolein at intraplantar paw. (A) Bar graph shows that mechanical threshold in both sham and SCI groups can be reduced significantly following acrolein injection (∗ P < 0.05, ∗∗ P < 0.01, when compared to pre-acrolein, One-way ANOVA then Tukey test). (B) The percent of change of mechanical threshold in SCI following acrolein injection was significantly greater compared that in sham group (P < 0.0001) N = 5-8 in all groups. unpaired t-test. The values are mean ± SEM

97

4.7 Summary and Discussion

In this chapter, our data strongly supported that elevation of acrolein was correlated with establishment and maintenance of neuropathic pain after SCI. In addition, RT-PCR data showed that TRPA1 gene expression level could be persisted at least 2 weeks in the various tissues after SCI not only at the central ending of a primary sensory nerve (spinal dorsal horn), but also at peripheral terminal (paw skin) of a primary afferent. This data indicated that SCI can lead to the increase the gene expression level in the nervous system. However, when we treated SCI animals with acrolein scavenger hydralazine everyday, it could alleviate the expression level of TRPA1 gene in dorsal horn, DRG, and paw skin in 1 week and 2 weeks after SCI, which indicated that SCI-induced acrolein may contribute to the up-regulation of TRPA1 gene expression level in the nervous system after SCI. This is further supported by section 4.4. In here, micro-injected acrolein could up-regulate the gene expression level without any mechanical injury at the spinal cord. The data showed that the gene expression levels in acrolein injected side were significantly higher than that in saline injection side. Animal behavioral tests on acrolein micro-injected group also demonstrated that acrolein could lead to sensory and motor abnormalities with no mechanical damage. Taken together, acrolein, as a TRPA1 channel specific agonist, could be a promising therapeutic target for neuropathic pain treatment after SCI.

Especially, out RT-PCR data indicated that increased TRPA1 gene expression level at the peripheral terminal of primary afferent nerve may contribute to the hypersensitivity to the stimuli after SCI. According to the previous studies (Nilius et al., 2007; Anand et al., 2008; Andre et al., 2009), TRPA1 channels were also expressed at the keratinocytes at the peripheral nerve. Our RT-PCR data figure 25, showed that the TRPA1 gene expression level in paw skin was much higher than that in any other tissues at 1 week and 2 weeks after SCI (41.4±2.98 folds in skin, 2.6±0.15 folds in dorsal horn, and 2.0±0.22 in DRG at 1 week after SCI and 38.48±2.0 folds in paw skin, 2.57±0.45 folds in dorsal horn, and 1.27±0.22 folds in DRG at 2 weeks after SCI). This data indicated that neuropathic

98 pain mechanical and cold allodynia following SCI most likely due to the up-regulation of TRPA1 expression level in peripheral terminal of primary afferent sensory neurons. This is also further supported by two animal behavioral in vivo tests. First of all, as shown in figure 29, micro-injected acrolein directly into the spinal cord caused the up-regulation of TRPA1 gene expression level at the spinal dorsal horn, DRG, and skin. In addition, it induced neuropathic pain mechanical and cold allodynia. This acrolein-induced hypersensitivity to the mechanical and cold stimuli may be mainly caused by up- regulation of TRPA1 gene expression at the peripheral terminal of primary sensory neurons. Secondly, data from figure 33 supported this theory. Figure 33 showed that acrolein intradermal injection could generate the neuropathic pain-like behavior in control animal group. However, acrolein-mediated neuropathic pain behaviors were enhanced in SCI in case of mechanical threshold. This behavioral phenomenon may be due to the increase of TRPA1 expression level at the peripheral ending of primary sensory neuron after SCI as we have already seen in figure 25 and 26.

99

CHAPTER 5. THE EFFECT OF PHENELZINE AS AN ALTERNATIVE TO A HYDRALAZINE

5.1 Background and Research Scope of This Study

The majority of researches on the acrolein scavengers have used the hydralazine, because its acrolein scavenging capacity was much greater than other well known aldehyde scavengers such as methoxyamine, aminoguanidine, pyridoxamine, and carnosine (Burcham et al., 2002; Burcham et al., 2004; Kaminskas et al., 2004b). Compared to hydralazine, the other acrolein scavengers such as phenelzine have less attention, even though it has also acrolein-scavenging capabilities (Burcham et al., 2002; Wood et al., 2006b).

Through previous study, our in vitro experiment results have demonstrated that the ability of phenelzine to protect against acrolein-mediated cell death was very similar to that of hydralazine, and its protection capability seemed to increase with higher concentration. In addition, although there was no difference statistically between phenelzine and hydralazine, phenelzine could provide slightly better protection against free acrolein. Furthermore, cell viability data with phenelzine and hydralazine in uninjured cells demonstrated that phenelzine also provided the cellular protection through scavenging free radicals, which were produced by normal cells.

Although hydralazine has benefits for acrolein scavenging, it has also some limitations for in vivo treatment. Basically, hydralazine is the FDA-approved hypertension medication, however, anti-hypertensive effect could be the life threatening side effect to, especially, SCI patients, which could cause the neurogenic shock. In this sense, although many of studies have demonstrated that the effect of phenelzine as an

100 acrolein scavenger via in vitro experiment, anyone hasn’t evaluated the efficacy of acrolein scavenger phenelzine in SCI animal model. Therefore, the goal of these experiments is to investigate the efficacy of acrolein-scavenger phenelzine as an althernative to hydralazine at in vivo SCI animal model.

5.2 Phenelzine Treatment Reduced 3-HPMA Production

To investigate phenelzine as an acrolein scavenger, acrolein pre-incubated with saline (0.7 mg/kg), saline + acrolein + hydralazine (5mg/kg), and saline + acrolein + phenelzine (5mg/kg) were injected via I.P. For the acrolein scavenging, we waited for 60 min before injecting solution to the animals and 24 hour after injection, we collected urine from animals. For the control group, only saline was injected. As indicated in figure 34, such treatment significantly attenuated the 3-HPMA levels in urine. Specifically, the average urine 3-HPMA levels in control, acrolein + saline, acrolein + saline +hydralazine, and acrolein + saline + phenelzine were 2.06+0.06, 76.79±1.45, 6.84±0.23, and 5.03±0.86 μg/mg respectively. This result showed that injection of acrolein containing acrolein scavengers such as hydralazine and phenelzine could reduce the level of 3- HPMA in urine compared to the acrolein only group (p<0.001). This result clearly indicated that the reduction of 3-HPMA was correlated with acrolein scavenging by hydralazine and phenelzine. In addition, when we compared 3-HPMA levels in hydralazine and phenelzine treatment group, there was significantly difference between these groups. That is, the 3-HPMA level in phenelzine treatment group was significantly lower than that in hydralazine treatment group (p<0.05). Therefore, capability of phenelzine as an acrolein scavenger may be more effective than hydralazine.

101

Figure 34. The Reduction of Urine 3-HPMA by Acrolein Scavengers

For the 3-HPMA urine test, we collected urine 24 hours after I.P. injection. The 3-HPMA level in acrolein only injection group was significantly increased following injection. However, the elevated 3-HPMA level was decreased by more than 90% (from 76.79±1.45 to 6.84±0.23, and from76.79±1.45 to 5.03±0.86 μg/mg). (# P < 0.001 when P < 0.001 when compared to acrolein only, ANOVA). N=4-5 in :כ ;compared to control each group.

102

5.3 Acrolein Scavenging by Phenelzine after SCI

To investigate the efficacy of phenelzine as an acrolein scavenger after SCI, we performed immunoblotting to measure the endogenous acrolein-lysine adduct level, and LC/MS/MS to detect acrolein metabolite urine 3-HPMA following SCI. First of all, we generated moderate SCI animal models using NYU impactor by dropping its weight rod from 25mm onto the intact spinal cord at T-10 level. Then, 5mg/kg of phenelzine was injected via I.P right after SCI within 3 min. Phenelzine solution was injected 2 times before euthanizing animals and 1 cm of spinal cord including T-10 level damage site was collected for immunoblotting experiment. In this experiment, we had three groups; 1) SCI only, 2) SCI + phenelzine (5mg/kg), and 3) SCI + hydralazine (5mg/kg). For the comparison, the same amount of hydralazine (5mg/kg) was also injected 2 times before collecting spinal cord tissue samples.

Based on constructed acrolein standard curve in figure 3, we detected acrolein- lysine adducts levels after SCI in SCI, SCI+phenelzine, and SCI+hydralazine groups. As figure 35 indicated, the acrolein-lysine adduct level in acrolein scavenger treatment groups was significantly lower than that in SCI only group (p<0.005). Specifically, the acrolein levels were 31.10 ± 4.21 in SCI only, 3.46 ± 1.98 in SCI + phenelzine injection group, and 8.70 ± 1.45 in SCI + hydralazine injection group. Especially, acrolein level in phenelzine injection group was statistically lower than that in hydralazine injection group (3.46 ± 1.98 in SCI + phenelzine injection vs. 8.70 ± 1.45 in SCI + hydralazine injection, p<0.05). This data supported that phenelzine could alleviate endogenous acrolein-lysine adduct level following SCI and it may provide more acrolein scavenging effects.

In addition to immonoblotting, we also measured the acrolein metabolite 3-HPMA from urine with various dosages of acrolein scavengers in 1 day after SCI. In this study, we had 7 experiment groups; A: control, B: SCI post operative day (POD) 1, C: SCI POD1 + hydralazein (5mg/kg), D: SCI POD1 + hydralazine (25mg/kg), E: SCI POD1+phenelzine (5mg/kg), F: SCI POD1+phenelzine (15mg/kg), and G: SCI

103

POD1+phenelzine (60mg/kg). As figure 36 indicated, 3-HPMA level in SCI POD1 (B) was significantly higher than that in control (A) (3.51 ± 0.24 vs. 1.98 ± 0.21, p<0.05). The 3-HPMA levels in hydralazine treatment groups (C (5mg/kg) and D (25mg/kg)) showed that there were reductions of 3-HPMA levels in group C and D compared to SCI only group (B) (3.08 ± 0.22 in C vs. 3.51 ± 0.24 in B, p<0.05 and 2.22 ± 0.19 in D vs. 3.51 ± 0.24 in B, p<0.01). The 3-HPMA levels in phenelzine injection groups were 2.84 ± 0.25 in E (phenelzine (5mg/kg)), 1.86 ± 0.13 in F (phenelzine (15mg/kg)), and 0.83 ± 0.17 in G (phenelzine (60mg/kg)). All 3-HPMA levels in these groups were statistically lower than that in SCI POD1 group (B vs. E, p<0.05, B vs. F, p<0.01, B vs. G, p<0.005). In this result, there was no difference between C and E also there was statistically no difference between D and E. However, although there was no difference between D (hydralazine 25mg/kg) and E (phenelzine 5mg/kg), the 3-HPMA level in group E was as low as group D was. In addition, 3-HPMA level in group F (phenelzine 15mg/kg) was significantly lower than that in group D (hydralazine 25 mg/kg) (p<0.05). Taken these immunoblotting and 3-HPMA urine data together, phenelzine could be an effective acrolein scavenger compared to hydralazine.

104

Figure 35. Reduction of Acrolein-lysine Adduct Level by Acrolein Scavengers

Using immunoblotting, we measured the acrolein-lysine adduct levels from 1 day after SCI animals. Acrolein level were 31.10 ± 4.21 in SCI only, 3.46 ± 1.98 in SCI + phenelzine injection group, and 8.70 ± 1.45 in SCI + hydralazine injection group. Especially, acrolein level in phenelzine injection group was significantly lower than that in hydralazine injection group. N=4-6 in each group, ANOVA and Tukey tests for the statistical analysis. #p<0.05 when compared to hydralazine injection group and *p<0.01 when compared to SCI only.

105

Figure 36. 3-HPMA Level in 1 Day after SCI with various Dosages of Acrolein Scavengers

For the 3-HPMA measurement, we have 7 different groups. A: control, B: SCI post- operative day (POD)1, C: SCI POD1 + hydralazein (5mg/kg), D: SCI POD1 + hydralazine (25mg/kg), E: SCI POD1+phenelzine (5mg/kg), F: SCI POD1+phenelzine (15mg/kg), and G: SCI POD1+phenelzine (60mg/kg). All 3-HPMA levels in phenelzine injection groups were statistically lower than that in SCI POD1 group (B vs. E, p<0.05, B vs. F, p<0.01, B vs. G, p<0.005). In addition, 3-HPMA level in group F was significantly lower than that in group D (p<0.05). This data indicated that phenelzine might be a better acrolein scavenger compared to hydralazine. N= 4-6 in each group. ANOVA and Tukey tests were used for statistical analysis. Urine from animals was collected 24 hours after SCI.

106

5.4 Analgesic Effect of Phenelzine on Neuropathic Pain after SCI

The benefit of phenelzine as an analgesic was investigated on SCI-induced neuropathic pain animal model. Initially, we created the moderate contused-SCI animal model by using NYU impactor. Based on 3-HPMA result, 15mg/kg of phenelzine was administrated for 2 weeks daily via systemically. For the comparison, hydralazine (5mg/kg) was injected for 2 weeks every day. Therefore, in this study, there were four different groups; 1) Sham control, 2) SCI only, 3) SCI + hydralazine (5mg/kg), and 4) SCI + phenelzine (15mg/kg). For the behavioral tests, we performed the mechanical and cold allodynia tests. For the experimental set up, we followed the materials and methods part in chapter 2 and chapter 3.

Figure 37 showed mechanical threshold on various groups. Initially, the base lines were 15 grams in sham control, 14.23 ± 0.56 g in SCI only, 15g in SCI + phenelzine injection, and 15g in SCI + hydralazine injection group. However, from 2 weeks after treatment, mechanical threshold in acrolein-scavenger groups were significantly higher than that in SCI only group, which were lasted until the end of experiment periods. Specifically, mechanical threshold in phenelzine injection group was significantly higher than that in hydralazine treatment group 2 weeks after injection (14.44 ± 0.44g vs. 7.02 ± 0.48g, p<0.005). This trends were maintained until 4 weeks (8.58 ± 0.42g vs. 6.55 ± 0.76g in 3 weeks after injecting (p<0.05), and 8.27 ± 0.43g vs. 6.40 ± 0.66g 4 weeks after injecting (p<0.05)). During the test periods, there was no significant change in sham control group. Also, mechanical threshold in SCI only group showed consistently hypersensitive to mechanical stimulus during the test 2.48 ± 0.73g in 2 weeks, 2.55 ± 0.48g in 3weeks, and 2.61 ± 0.78g in 4 weeks after SCI. This data indicated that acrolein scavenger phenelzine could attenuate the SCI-induced neuropathic pain mechanical allodynia and its efficacy was better than hydralazine’s.

We also performed the cold allodynia test using 100% acetone. Similar to chapter 3, we applied 100% of acetone onto the plantar surface of SCI animals 5 times per each side

107 of hind limb and expressed it as withdrawal frequency to cold stimulus. Therefore, higher frequency means that animal was sensitive to acetone-generated cold effect. As indicated in figure 38, 2 weeks after SCI, neuropathic pain cold allodynia was well established on SCI animal group. In addition, during the whole experiment period, the withdrawal frequency in phenelzine and hydralaznie treatment groups was significantly lower than that in SCI only group. Especially, in 2 weeks, the frequency in phenelzine injection group was statistically lower than that in hydralaznie injection group (2.50 ± 0.56% vs. 27.65 ± 5.94%, p<0.005). However, there was no difference between these groups during last of periods. This data also supported that phenelzine could alleviate the SCI-induced neuropathic pain cold allodynia after SCI, and although benefit of phenelzine was only 2 weeks time point, phenelzine may provide better neuroprotective effects in the initial stage of neuropathic pain on cold allodynia.

108

Figure 37. Phenelzine Reduced the SCI-induced Mechanical Allodynia

15mg/kg of phenelzine was injected into the SCI animals everyday for two weeks. The mechanical threshold was assessed by von Frey filaments up-down methods. As data indicated, mechanical thresholds in phenelzine and hydralazine injection were significantly higher than that in SCI only during the experiment periods. Especially, mechanical threshold in phenelzine injection group was higher than that in hydralazine injection group during the whole experiment periods. This data demonstrated that phenelzine could attenuated SCI-indiced neuropathic pain mechanical allodynia and its effect may be better than hydralazine. N=4-13 in each group, ANOVA and Tukey tests were used for statistical analysis. **p<0.005, *p<0.05 compared to SCI +HZ, #p<0.01 compared to SCI only.

109

Figure 38. Phenelzine Reduced SCI-induced Neuropathic Pain Cold Allodynia

Withdrawal frequency to 100% acetone was assessed to investigate the effect of phenelzine on cold allodynia. For 2 weeks every day, 15mg/kg of phenelzine was injected into the SCI animal via I.P. During the experiment periods, frequency in acrolein scavengers treatment groups was lower than that of SCI only group. At 2 weeks after injection, phenelzine injection group showed the less sensitivity to cold stimulus than hydralazine injection group. This data indicated that phenelzine also could attenuate SCI- induced cold allodynia. N=4-13 in each group, ANOVA and Tukey tests were used. **p<0.005 compared to phenelzine injection group, &p<0.001 compared to SCI only, and #p<0.005 compared to hydralazine injection group.

110

5.5 Summary and Discussion

Throughout this dissertation, we have used hydralazine as an acrolein-scavenger to alleviate acrolein-mediated neuronal damage, motor and sensory abnormalities following SCI. Hydralazine is the FDA-approved vasodilator. However, anti-hypertensive effect could be a life threatening side effect to SCI patients because of neurogenic shock. Therefore, with this information, someone might point out limitations of our studies (using hydralazine as an acrolein scavenger for neuroprotection after SCI). However, through the entire of these studies, the dosage of hydralazine was 5mg/kg, which was significantly lower than the suggested upper limitation for human patients with safe dosage, which is 7.5 mg/kg (Hamann and Shi, 2009). In addition, metabolite rate of rodents is much faster than that of human. Therefore, even though higher dosage of hydralazine could be metabolized very fast. This is supported by our data. In figure 36, we have injected 25mg/kg of hydralazine into group D for 3-HPMA detection. Even though the administered hydralazine (25mg/kg) was much higher than suggested upper limitation, we did not see any hypotensive side effect on SCI animals. However, although any hypotensive side effect by hydralazine has not been observed through this study, it is important to point out some limitations of hydralazine for clinical application.

Fortunately, some research groups identified another acrolein scavenger phenezine which could alleviate acrolein-mediated cellular toxicity (Parent et al., 2002; Kaminskas et al., 2004b; Wood et al., 2006b). Phenelzine is a monoamine oxidase inhibitor which has a hydrazine functional group like hydralazine (Wood et al., 2006b). Therefore, phenelzine could be an acrolein-scavenger and, like hydralazine, it may react with acrolein 1:1 molecular ratio. However, compared to hydralazine, phenelzine could offer the some advantages. First, the half-life of phenelzine is much longer than that of hydralazin. Based on previous study, the half-life of phenelzine was 11.6 hours, which was much longer than half-life of hydralazine (30 min to 1 hour) (Reece, 1981). This aspect is very important, since oxidative stress and acrolein production contribute to the secondary cascade events after primary traumatic SCI, which are occurred in the hours,

111 days, and weeks. Our data also supported this notation, through chapter 2, 3, and 4, we have demonstrated that acrolein-lysine adduct level was persisted at least 2 weeks after SCI (Due et al., 2014; Park et al., 2014). Therefore, with longer half-life of phenelzine, it may provide additional neuroprotective effect via acrolein scavenging after SCI. Second, phenelzine is not a vasodilator thus, it does not offer hypertensive side effect to SCI patients. Based on our data in figure 36, we injected 60mg/kg of phenelzine into the SCI animal and no anti-hypertensive effects were observed. In this sense, with phenelzine’s long half-life and non-hypertensive effect, phenelzine may offer additional neruoprotective role to the CNS following initial trauma. Lastly, phenelzine is the FDA- approved anti-depressant. Therefore, it could bring additional benefits to SCI patients who are suffering from chronic neuropathic pain and post traumatic stress after SCI.

The benefits of phenelzine were further supported by previous in vitro data. Based on in vitro tests, acrolein scavenging effects of phenelzin were very similar to hydralazine’s with concentration dependent way. Acrolein-treated cell viability test with various acrolein scavengers such as phenelzine and hydralazine indicated that the phenelzine had an ability to protect against acrolein-mediated cell death, which was statistically no different from hydralazine treatment group. In addition, cell viability was also improved in uninjured cells by treating phenelzine, since it could remove the endogenous acrolein produced by PC12 cell, and also the free radicals such as DPPH which were produced by normal cells.

In addition to the in vitro results, the benefits of phenelzine as an acrolein scavenger were supported by our in vivo SCI animal study. The immunoblotting and urine 3-HPMA data results demonstrated that phenelzine could result in the reduction of acrolein levels after SCI (Figures 35 and 36). Especially, these data showed that phenelzine could be a better acrolein scavenger than hydralazine is, since the acrolein- lysine adduct level in phenelzine group was significantly lower than that in hydralazine. In addition, although there was no different of 3-HPMA levels between 25mg/kg of hydralazine injection group and 5mg/kg of phenelzine injection group, 3-HPMA level in

112

5 mg/kg of phenelzine injection group was as low as 25mg/kg of hydralazine injection group was. Furthermore, the 3-HPMA level in 15mg/kg of phenelzine injection group was significantly lower than that in 25mg/kg of hydralazine’s.

The benefits of phenelzine on sensory neurons’ abnormalities following SCI were investigated using mechanical and cold allodynia. The mechanical allodynia data (Figure 37) indicated that phenelzine also could attenuate the hypersensitivity to mechanical stimulus. Especially, mechanical threshold in phenelzine injection group was similar to that in sham control, and statistically higher than hydralazine’s during the experiment period. Withdrawal frequency to cold stimulus data (Figure 38) also demonstrated the analgesic benefits of phenelzine after SCI. Especially, at 2 weeks after SCI, the frequency of phenelzine injection group was lower than that of hydralazine. Taken these results together, with its long half-life and non-hypotension effects, phenelzine could be used an acrolein scavenger with high dosage as an alternative to hydralazine. However, phenelzine has been implicated in a wide range of side effects such as urinary retention, muscle tremors, food restriction, and drug interaction (Verrilli et al., 1987; Stonnington et al., 1988; Song et al., 2010). Therefore, additional research for phenelzine should be performed. In addition, the developing novel drug delivery system also should be continued to improve the drug’s scavenging capability.

113

CHAPTER 6. THE EFFECT OF POTASSIUM CHANNEL BLOCKER ON SCI- INDUCED MOTOR DEFICIT AND NEUROPATHIC PAIN

6.1 Background and Research Scope of This Study

Traumatic SCI causes myelin damage mainly in white matter, which contribute to expose the past potassium channels at the juxtaparanodal regions (Shi and Blight, 1996, 1997; Leung et al., 2011a). Under normal physiological condition, juxtaparanodal regions are covered by myelin, which prevent from potassium ions leaking and allow it to depolarize to generate action potential (Chiu and Ritchie, 1980; Haghighi et al., 1995; Sun et al., 2010; Leung et al., 2011a). Since the type of SCI is most likely contused and/or compressed injury (Blight, 1983; Blight and Decrescito, 1986; Hayes et al., 1993; Kakulas, 1999), myelin damage in white matter of spinal cord leads to the potassium efflux and causes the action potential failure. Furthermore, acrolein generation resulting from inflammation has been known to play a critical role in demyelination after neuronal damage (Smith et al., 1999; Gilgun-Sherki et al., 2004). In this sense, blocking the potassium channels could be an effective way to re-generate action potential after SCI (Waxman, 1989; Blight, 1991; Shi and Blight, 1997).

4-aminopyridine (4-AP) has been identified as a fast potassium channel blocker, which could restore axonal conduction after SCI (Blight and Decrescito, 1986; Haghighi et al., 1995). The effect of 4-AP as an effective potassium channel blocker was supported by recent approval by the FDA for promoting motor dysfunction in multiple sclerosis (MS) via restoring axonal action potential. However, there are some limitations of using 4-AP for clinical applications because of its side effects such as dizziness, nausea, and

114 seizures (Stork and Hoffman, 1994; Potter et al., 1998; Pena and Tapia, 1999; Korenke et al., 2008). In this sense, a developing an alternative potassium channel blocker to 4-AP is required.

Our lab has been focusing on a new type of fast potassium channel blocker 4- aminopyridine-3-methanol (4-AP-3-MeOH). Many of previous researches have been shown that this compound could block potassium channels and restore axonal conduction to similar level of normal axonal conduction in various neurodegenerative disease models such as MS, and SCI where the myelin damage is occurred (Sun et al., 2010; Leung et al., 2011a). Furthermore, the effective lowest concentration of 4AP-3MeOH for mechanically injured spinal cord is between 0.01 and 0.1 μM, which is significantly lower than that of 4-AP (between 0.1 and 1 μM) (Shi and Blight, 1997; Sun et al., 2010). However, the effects of 4-AP and 4AP-3MeOH have not been investigated in SCI and SCI-induced neuropathic pain animal model. Therefore, in the current study, we test whether potassium channel blockers can enhance the SCI-induced motor function deficits, and attenuate SCI-induced neuropathic pain.

6.2 The Effects of Potassium Channel Blockers on Neuropathic Pain

Initially, we have tested the effect of potassium channel blockers such as 4-AP and 4AP-3MeOH on SCI-induced neuropathic pain model. We generated contused-SCI model by using NYU impactor. Two weeks after SCI, two different types of potassium channel blockers were used for this study; 4-AP and 4AP-3MeOH injection groups. Based on the previous study (Sun et al., 2010), since the effect of potassium channel blockers is transient, we performed mechanical and cold allodynia tests every one hour after injection until 6 hours, and did the same tests at 24 hours after injection. The same amount (1mg/kg) of potassium channel blockers (4-AP and 4AP-3MeOH) was injected via systemically.

115

Figure 39 showed the mechanical threshold on the various groups. Before injection, the mechanical thresholds in 4-AP, 4AP-3MeOH, and SCI were 2.99 ± 0.51g, 2.67 ± 0.44g, and 2.47 ± 0.14g respectively. However, from at 2 hours after injection to at 5 hours after injection, the mechanical threshold in potassium channel blocker injection groups were significantly higher than that in SCI only group. Especially, at 3 hours after injection, mechanical threshold in 4AP-3MeOH was 7.41 ± 0.51g, which was statistically higher than that in 4-AP group (4.98 ± 0.61g, p<0.01). The significant differences between 4-AP and 4AP-3MeOH were maintained by 5 hours after injection (7.23 ± 0.73g vs. 4.45 ± 0.22g at 4 hours after injection, p<0.01, and 5.65 ± 0.44g vs. 3.51 ± 0.30g at 5 hours after injection, p<0.01). This data showed that potassium channel blocker could attenuate neuropathic pain mechanical allodynia after SCI, and the analgesic effect of 4AP-3MeOH may be better than 4-AP’s. However, the mechanical threshold in whole groups got closer to their initial values, which indicated that the effect of potassium channel blocker was transient.

Withdrawal frequency to cold stimulus is shown in figure 40. The base lines of withdrawal frequency were 52.50 ± 4.78% in 4-AP injection group, 50.00 ± 3.13% in 4AP-3MeOH injection group, and 55.50 ± 3.53% respectively, while 7.21 ± 3.21% in uninjured control group. Very similar to mechanical allodynia, potassium channel blockers could alleviate withdrawal frequency to 100% acetone stimulus from at 1 hour after injection to at 4 hours after injection. In addition, the efficacy of 4AP-3MeOH, in terms of analgesic, was further supported by this data. Specifically, at 1 hour after injection, frequency in 4AP-3MeOH injection group was 14.51 ± 4.33%, while in 4-AP injection group, 35.00 ± 2.89% (p<0.01). (At 2 hours after injection, 15.50 ± 2.33% vs. 27.50 ± 2.89% (p<0.01), at 3 hours after injection, 20.50 ± 4.33% vs. 32.50 ± 2.55% (p<0.01), and at 4 hours after injection, 30.50 ± 2.21% vs. 42.50 ± 2.55% (p<0.01)). Again, very similar to mechanical threshold, after 24 hours injection, cold allodynia closed to base line except 4AP-3MeOH, which showed a lower frequency compared to 4- AP injection group (p<0.05). This data also showed that potassium channel blockers

116 could reduce SCI-induced cold allodynia, and 4AP-3MeOH might have better analgesic capability compared to 4-AP.

117

Figure 39. Potassium Channel Blockers Reduced Mechanical Allodynia after SCI

1mg/kg of 4-AP and 4AP-3MeOH were injected systemically 2 weeks after SCI. To test the mechanical threshold, we performed von Frey filaments test. The data showed that two different types of potassium channel blockers could reduce the mechanical threshold from 2 hours to 5 hours after injection, but its effect was not maintained beyond 24 hours after injection except 4AP-3MeOH injection group. N=6-8 in each group. ANOVA and Tukey tests were used for statistical analysis. **p<0.01 compared to 4-AP injection group, #p<0.01 compared to SCI only group, and &p<0.05 compared to SCI only group.

118

Figure 40. Potassium Channel Blockers Reduced Cold Allodynia after SCI

To investigate the effect of potassium channel blockers on cold allodynia after SCI, we applied 100% acetone onto the animals’ hind limb plantar surface. Very similar to mechanical threshold, potassium channel blockers reduced the withdrawal frequency to acetone during some time points (from at 1 hour to at 4 hours after injection). Still, 24 hours after injection, 4AP-3MeOH injection group showed less sensitivity to cold stimulus compared to 4-AP injection group. N=6-8 in each group, ANOVA and Tukey tests were used for analysis. *p<0.05, **p<0.01 compared to 4-AP injection group, #p<0.01 compared to SCI only group.

119

6.3 The Effect of Potassium Channel Blockers on Motor Function Deficits

To investigate the effect of potassium channel blockers on motor function recovery after SCI, we administered 2 different dosages of 4AP-3MeOH (1mg/kg and 5mg/kg) and 1mg/kg of 4-AP 2 weeks after SCI. Then, 30min after injection, we performed the rotarod tests. Rotarod test is commonly used for investigation of motor function coordination to evaluate the effect of medication on neurogenerative diseases. Previous study showed that gradually increased rotation speed resulted in the greater sensitivity of the test (Jones and Roberts, 1968). Therefore, the rotation speed was gradually increased by 3 rotations per minute (RPM) every 10 s until it reached 30 RPM and 180 s. Before SCI, all the animals have passed this training, and we did not use the animals that did not pass the training. When the animals completely fall off from the device, or spin around twice on the device, we stopped the test and recorded the RPM and time on the rotarod. These values indicated the animal’s capability of remaining on the spinning rod and motor function recovery by injected medication.

Figure 41 showed the time on rotarod of 5 different groups; normal control (A), SCI only (B), 5mg/kg of 4AP-3MeOH (C), 1mg/kg of 4AP-3MeOH (D), and 1mg/kg of 4-AP (E). There were no significant differences between SCI only and 1mg/kg of 4-AP and 4AP-3MeOH groups (47.30 ± 3.69 sec. in SCI, 46.33 ± 2.01 sec. in 1mg/kg of 4AP- 3MeOH, and 45.55 ± 2.84 sec. in 1mg/kg of 4-AP). However, when we treated 5mg/kg of 4AP-3MeOH, the time on rotarod was significantly longer than any other groups. We did not investigate the effect of 5mg/kg of 4-AP on motor function recovery, since the animals showed the seizure-like activity after injection. We also checked the rotation per minutes (RPM) at the same time. As figure 42 indicated, there were no differences between potassium channel blocker injection groups and SCI only group. Specifically, the RPM in SCI only group was 9.68 ± 0.51 RPM, 9.58 ± 0.48 RPM in 1mg/kg of 4AP- 3MeOH, and 9.47 ± 0.39 RPM in 1mg/kg of 4-AP. However, the RPM in 5mg/kg of 4AP-3MeOH showed 9.58 ± 0.48 RPM, which was statistically different compared to other groups. This rotarod data demonstrated that 5mg/kg of 4AP-3MeOH could promote

120 the motor function after SCI, and high level of 4AP-3MeOH could be used safely compared to 4-AP.

121

Figure 41. The Effect of Potassium Channel Blocker on Motor Deficit after SCI

To assess the effect of potassium channel blockers on motor function, two differences dosages of 4AP-3MeOH (1mg/kg and 5mg/kg), and 1mg/kg of 4-AP were injected systemically (A: normal control, B: SCI only, C: 4AP-3MeOH (5mg/kg), D: 4AP- 3MeOH (1mg/kg), and E: 4-AP (1mg/kg)). The time on device in 5mg/kg of 4AP- 3MeOH was longer than any other groups. This data showed the high level of 4AP- 3MeOH could promote motor function after SCI. N=6-8 in each group. ANOVA and Tukey tests were used for statistical analysis.*p<0.05

122

Figure 42. Potassium Channel Blockers Promoted the Motor Recovery after SCI.

The animals’ motor function recovery after SCI was assessed by measuring the rotation per minute on the rotarod (A: normal control, B: SCI only, C: 4AP-3MeOH (5mg/kg), D: 4AP-3MeOH (1mg/kg), and E: 4-AP (1mg/kg)). Similar to time on the rotarod, RPM in 5mg/kg of 4AP-3MeOH was higher than any other groups. N= 6-8 in each group, ANOVA and Tukey tests were used for the statistical analysis, *p<0.05

123

6.4 Summary and Discussion

Through this study, we have demonstrated that potassium channels blockers could attenuate SCI-induced neuropathic pain behaviors (figures 39 and 40). Especially, in terms of analgesic, the effect of 4AP-3MeOH was better than that of 4-AP. In addition, although there were no difference between SCI only and treatment groups under the same dosage of potassium channel blockers, higher dosage of 4AP-3MeOH (5mg/kg) was effective to improve the motor function after SCI (figures 41 and 42). These behavior tests demonstrated that potassium channel blockers could not only attenuate neuropathic pain behaviors, but promote motor function recovery after SCI. This notion is further supported by previous studies (Van Diemen et al., 1993; Wu et al., 2009). Furthermore, this data showed that the efficacy and therapeutic value of 4AP-3MeOH. Even though we did not present the data, 5mg/kg of 4AP-3MeOH also led to attenuation of neuropathic pain mechanical and cold allodynia, and its efficacy was better than 1mg/kg of 4AP- 3MeOH’s. It was unable to test with 5mg/kg of 4-AP, since the animals showed the seizure-like activity after injection, but this action went away in 2 hours after injection because of the transient property of potassium channel blocks (Sun et al., 2010). Furthermore, high dosage of 4AP-3MeOH (5mg/kg) also led to better results compared to 1mg/kg of 4AP-3MeOH and 4-AP. These data supported that 4AP-3MeOH could be used as an effective potassium channel blocker with high dosage.

Two major differences between 4AP-3MeOH and 4-AP led to different results in this experiment. First of all, the effective lowest concentration of 4AP-3MeOH (between 0.01 and 0.1 μM) to recover axonal action potential is significantly lower than that of 4- AP (between 0.1 and 1 μM) (Shi and Blight, 1997; Sun et al., 2010). Secondly, based on the ex vivo studies, when they treated with 4AP-3MeOH, axon can restore its action potentials, which was similar to axons’ action potentials way under normal condition (Jensen and Shi, 2003; Sun et al., 2010). In addition, 4AP-3MeOH can also significantly enhance axonal action potential conduction in EAE mouse (Leung et al., 2011a). Based

124 on these mechanisms, high dosage of 4AP-3MeOH may induce the motor function recovery after SCI.

The hyperexcitability of primary afferent neuron after SCI results in neuropathic pain allodynia. Although many of voltage-gated channels are implicated in transmission of noxious stimuli to higher order of neuron and hyperexcitability, potassium channels play an important role in transmission of pain information by regulating membrane potential and generating action potential. In addition, neuronal damage leads to the opening of potassium channels, which results in hyperpolarization of cell membrane (Ficker and Heinemann, 1992; Wu and Saggau, 1997; Takeda et al., 2011). Therefore, potassium channels have been focused on as a therapeutic target for neuropathic pain treatment after neuronal damage. This notion is well supported by our data (figures 39 and 40). These data indicated that potassium channel blockers could attenuate neuropathic pain mechanical and cold allodynia following SCI. Specifically, analgesic effect of 4AP-3MeOH was better than that of 4-AP. Taken together, based on these data using SCI-induced neuropathic pain animal model, potassium channel blockers can not only alleviate neuropathic pain, but promote motor function recovery following SCI. Especially, 4AP-3MeOH may be an effective potassium channel blocker as an alternative to 4-AP in enhancing the sensory and motor dysfunction recovery after neuronal damage. However, the working mechanisms of potassium channel blockers on the sensory neuron for neuropathic pain, and on the motor neuron for motor function deficits after SCI are further studied.

125

CHAPTER 7. SUMMARY AND CONCLUSION

Traumatic spinal cord injury (SCI) induces motor and sensory dysfunction, bladder reflexive dysfunction, urinary tract infection, renal malfunction, loss of respiratory function, loss of sexual function, and pneumonia. Beyond these SCI-induced malfunctions, neuropathic pain after SCI has significant negative effects on patients’ quality of life. In spite of research to improve and recover functional deficits after SCI, insufficient therapeutic targets lead to a frustrating condition to treat. Recently, antioxidants have attracted researchers’ attention to alleviate the SCI-induced malfunctions. However, recent clinical trials using these medications have failed to improve functional recovery after SCI. Despite of failing the clinical trials, this study indicated that attenuating oxidative stress by targeting a neurotoxin, acrolein, could be an effective way to improve the SCI-induced dysfunctions.

As the strongest electrophile and the most reactive aldehydes, acrolein has been implicated in various neurodegenerative diseases (Hensley et al., 2000; Vincent et al., 2002; Naoi et al., 2005; Liu-Snyder et al., 2006b; Schwartz et al., 2008). Acrolein can originate through exogenously through environmental sources such as air pollution and cigarette smoke, and endogenously through the breakdown of carbohydrates and amino acids (Witz, 1989; Ghilarducci and Tjeerdema, 1995; Hamann et al., 2008b). In addition to its high reactivity, the half-life of acrolein is much longer than that of the other ROS. Acrolein can bind to protein and DNA, forming acrolein-protein adducts, which have half-lives that are even longer than that of free acrolein. Furthermore, acrolein can bind to endogenous anti-oxidant GSH and react with it much faster than other types of oxidative stress-inducing molecules. With these unique properties, acrolein can induce the vicious cycle of oxidative stress and amplify its effects after SCI.

126

Because acrolein plays an important role in pathogenesis after traumatic SCI, we have focused on acrolein scavenging as the most important means of intervention to further degeneration after SCI. The FDA-approved hypertensive medication hydralazine has shown its acrolein-scavenging effect with the hydrazine functional group (Burcham et al., 2000; Burcham et al., 2002; Burcham et al., 2004). In addition, because hydralazine can react with free acrolein and acrolein-protein adducts in an equimolar manner, equimolar concentration of hydralazine can neutralize acrolein entirely (Burcham et al., 2002; Burcham et al., 2004; Hamann et al., 2008a; Hamann and Shi, 2009). Many previous studies have showed acrolein-scavenging effects of hydralazine via in vitro tests (Liu and Wood; Burcham and Pyke, 2006; Liu-Snyder et al., 2006b). However, to our knowledge, this is the first study that hydralazine has been shown to reduce the acrolein levels and acrolein-induced neuropathic pain through in vivo animal SCI models.

The main goals of chapter 2 were to ascertain the pathological role of acrolein in the CNS, and the effects of acrolein scavenging via hydralazine in an in vivo SCI animal model. Therefore, we hypothesized, first, the level of acrolein would depend on the SCI severity, second, suppressing acrolein would reduce the neuronal damage and promote functional recovery, and third, acrolein itself would generate the neurological deficits. Figures 7 indicated that the elevations of acrolein levels were correlated with SCI severity. Specifically, the elevated level of acrolein persisted for at least two weeks after both moderate and severe SCI groups. This was further supported by urine 3-HPMA results (Figure 8). With a recently developed technique, LC/MS/MS, we measured the concentration of the acrolein metabolite, 3-HPMA, in animals’ urine samples, of which levels were also correlated with SCI severity. Before testing the benefits of acrolein scavenging by hydralazine, we investigated whether it could reach the therapeutic level within the CNS. Figure 9 showed that systemically injected hydralazine was detected in concentrations higher than the minimal level needed to reduce acrolein-mediated damage. With this information, we injected 5 mg/kg of hydralazine after SCI, and figures 10 and 11 showed that hydralazine could reduce the acrolein level. In addition to reducing acrolein level after SCI, we also studied if hydralazine treatment could not only reduce

127 the neuronal tissue damage, but also improve locomotor functional recovery after SCI (Figures 12 and 13). These results may support that elevated acrolein after SCI lead to neuronal tissue damage and motor dysfunction, and by scavenging acrolein through hydralazine, we can attenuate the generation of cysts formation and malfunction of motor neuron. This notion is again supported by figure 15. In this data, micro-injected acrolein into the spinal cord induced the neuronal tissue damage and motor function deficits without any traumatic injury. Through this study, we showed that, first, acrolein level is correlated with SCI severity, second, systemic application of hydralazine can reduce acrolein levels, neuronal damage, and promote motor function recovery, and third, considering the role of acrolein after SCI, anti-acrolein therapy could be an effective treatment to improve motor dysfunction and neuronal damage.

As a TRPA1 channel specific agonist and an inflammatory factor, acrolein plays an important role in neuropathic pain pathogenesis after SCI. In chapter 2, figure 14, showed that persistent acrolein level elevation was correlated with the onset and establishment of neuropathic pain after SCI. In addition to increased levels of acrolein for two weeks after SCI (figure 16), we demonstrated that the TRPA1 gene expression level was up-regulated in various tissues such as dorsal horn of spinal cord, DRGs, and hind- limb paw skin one week and two weeks after SCI (figures 17 and 25). In addition, figures 14 and 22 showed that there was an increase in sensitivity in mechanical and cold stimuli after SCI, furthermore, micro-injected acrolein resulted in neuropathic pain-like mechanical and cold allodynia behaviors without any traumatic injury to the spinal cord (figures 21 and 30). Acrolein-induced neuropathic pain generation was further supported by up-regulation of gene expression level of TRPA1 (figure 29). However, as figures 26 and 27 indicated, hydralazine could down-regulate the TRPA1 gene expression level one and two weeks after SCI in various tissues such as dorsal horn, DRG and paw skin. Therefore, the DRG neurons may remain hypersensitive to its effects because of acrolein- mediated persistent up-regulation of TRPA1 after SCI.

128

Electrophysiological data demonstrated the effects of acrolein and SCI on cellular hypersensitivity. As figure 18 indicated, acrolein could increase the intracellular calcium levels after SCI and there was an increased in number of cells that responded to acrolein after SCI. Furthermore, SCI alone could increase the excitability of nociceptors (small and medium diameter of sensory neurons). In figure 19, we investigated the magnitude of current thresholds to generate the single action potentials, and the data showed that SCI could decrease the current threshold on sensory neurons. In addition, acrolein could increase the number of action potentials on sham and SCI groups (figure 20) and acrolein-mediated cellular excitability was enhanced after SCI. This notion is also supported by figure 33. Intradermally injected acrolein could increase the mechanical sensitivity to von Frey filaments in the sham control group, but acrolein-mediated mechanical sensitivity was enhanced in SCI group.

Taken together, these data showed that acrolein could increase neuronal sensitization, and its effect could be enhanced after SCI. In addition, considering the persisted elevation of acrolein for two weeks and the onset of neuropathic pain after SCI, acrolein could be a valuable therapeutic target for neuropathic pain. Therefore, scavenging acrolein by hydralazine may be an effective way to reduce not only acute neuropathic pain, but also chronic stages of neuropathic pain after SCI (figure 23).

With the anti-hypertensive property of hydralazine, it has number of limitations for the SCI patients. The major drawback of hydralazine is hypotensive effect to SCI patients, which could be a life threatening side effect because it could lead to neurogenic shock. In addition, with its short half-life, it has to be administrated more frequently for the neuro-protection. Recently, a monoamine oxidase inhibitor, phenelzine, was identified as a possible acrolein scavenger. Since it has a hydrazine functional group like hydralazine, it is reasonable to anticipate that it may react with acrolein 1:1 molecular ratio. However, Reece and colleagues showed that the half-life of phenelzine (11.6 hours) was significantly longer than that of hydralazine (Reece, 1981). Normally, secondary cascade events including oxidative stress and acrolein-mediated damage to the spinal

129 cord after primary injury were observed in the hours to days after injury. Therefore, phenelzine’s long half-life property may inhibit acrolein-mediated additional damage after primary injury. Furthermore, as the FDA-approved anti-depressant and non- hypertensive medication, phenelzine could be used with a higher dosage without severe side effects (figure 36), and provide additional neuroprotective role for the treatment of SCI-induced neuropathic pain behaviors (figures 37 and 38).

Based on previous studies, most SCIs are contused and/or compressed injury, which causes myelin damage mainly in white matter leading to exposure of potassium channels at the juxtaparanodal regions (Blight, 1983; Blight and Decrescito, 1986; Hayes et al., 1993; Leung et al., 2011a). Efflux of potassium ions results in slowing or the complete blocking of axonal conduction. Potassium channels play an important role in transmission of noxious stimuli to the higher order of neuron by regulating membrane potential and generating action potential. Therefore, many of researchers have focused on potassium channels as a potential therapeutic target for neuropathic pain after neuronal damage. So, initially, we tested if potassium channel blockers (4-AP and 4AP-3MeOH) could attenuate neuropathic pain mechanical and cold allodynia after SCI. As figures 39 and 40 indicated, both potassium channel blockers could reduce neuropathic pain behaviors, but the analgesic effect of 4AP-3MeOH was better compared to 4AP. We also checked the benefits of potassium channel blockers on the motor function deficits after SCI. Even though 1mg/kg of 4AP and 4AP-3MeOH did not show any effect on motor function recovery, a higher dosage of 5 mg/kg of 4AP-3MeOH did indeed show motor function improvement after SCI (figures 41 and 42). Taken together, potassium channel blockers could not only attenuate the neuropathic pain behavior, but also improve the motor function recovery after SCI. However, as we have already mentioned above, 5mg/kg of 4-AP induced the seizure-like behavior on animals, even though this activity did not last more than 2 hours because of its transient property. So, when we consider the efficacy and safety, 4AP-3MeOH would provide additional advantages to improve motor and sensory dysfunctions. For the future, we need to investigate various factors that could affect the efficacy and potency of potassium channel blockers for neurodegenerative

130 disease patients. In addition, we also need to investigate the working mechanisms of potassium channel blockers to reduce the side effects of these medications.

131

CHAPTER 8. FUTURE STUDIES

We have demonstrated the potential and pathophysiological values of acrolein as a therapeutic target for SCI-induced neuropathic pain. Our data supported that acrolein scavengers may be a novel analgesic to intervene neuropathic pain after SCI.

In particular, the immunoblotting and urine 3-HPMA results showed that elevation of acrolein level correlated with SCI severity and, based on the immunoblotting result, its level could be persisted at least two weeks after SCI. In addition, systemic application of acrolein scavengers, hydralazine and phenelzine, could decrease accumulation of acrolein-lysine adducts and 3-HPMA levels following contusion SCI. Although we have shown the elevation of acrolein level after SCI, it has been uncertain what types of cells produce and accumulate acrolein after neuronal damage. Many studies have demonstrated the accumulation of acrolein-protein adducts through immunohistochemistry (Luo et al., 2007; Ismahil et al., 2011). However, additional studies should be performed to investigate where acrolein is produced and accumulated. For this, we may use specific markers such as GFAP for astrocyte, DAPI for nuclei, and MBP (myelin basic protein) for myelin. This experiment would provide additional information about the exact location where the acrolein-protein adducts is formed. This information could be utilized to develop the efficient acrolein scavengers as well.

Neuropathic pain data showed that, as a pro-nociceptive factor, SCI-induced formation of acrolein correlated with the onset of neuropathic pain behaviors. This is supported through reduction of pain behaviors by using acrolein scavengers, such as hydralazine and phenelzine, after SCI. Furthermore, electrophysiology data indicated that acrolein could enhance the sensory neuron’s excitability after SCI and even sensory neuronal cells from sham control group. This phenomenon could be explained by our 132

gene expression data. The data showed that the elevation and maintenance of DRG sensory neurons’ excitability may be mediated by up-regulation of TRPA1 gene expression level, which is also correlated with elevation of acrolein level after SCI. An increase in TRPA1 gene expression level persisted for at least two weeks after SCI and its levels were decreased by scavenging the SCI-induced acrolein via hydralazine. Therefore, targeting acrolein for an analgesic effect would be a promising approach to attenuate neuropathic pain behaviors after SCI. However, we cannot rule out the possibility that generation of neuropathic pain by other acrolein-mediated pathways. As we already mentioned through this study, acrolein is known as pro-inflammatory factor. Based on previous studies, acrolein has been shown to stimulate the release of chemokines such as monocyte chemotactic protein-1 (MCP-1 or CCL2) after SCI (Knerlich-Lukoschus et al., 2010; Knerlich-Lukoschus et al., 2011b). In DRG, MCP-1 is expressed in small and medium diameter of sensory neurons which are also colocalized extensively with TRPA1 channels (Abbadie, 2005). In addition, MCP-1 is released at the spinal dorsal horn to modulate the activity of post synaptic neurons and glial cells, which express the MCP-1 specific receptor CCR-2 (White et al., 2009). A recent study showed that the activation of CCR-2 by MCP-1, which was activated by acrolein, could elicit membrane depolarization, trigger action potentials and sensitize nociceptors through the transactivation of TRPA1 channels to generate neuropathic pain after injury (Knerlich- Lukoschus et al., 2010). In this sense, based on reactive nature of acrolein, coupled with its diffusive, stable, and prolonged presence in the body, acrolein would play an important role in an inflammation-associated pathway that leads to the chronic nature of neuropathic pain. Therefore, the exact mechanisms and involvement of acrolein on the immune response-mediated generation of neuropathic pain after SCI should be performed. In line with immune system-evoked neuropathic pain, we also have to further investigate the balance between neuropathic pain generation and neuro-inflammatory response after injury. Under the normal physiological condition chemokines usually have positive effects on limitation of cellular and tissue damage, and they are well controlled by a complex network. However under neuronal damage conditions, this complex mechanism 133

is also damaged, and is insufficient to control inflammation. Therefore, the inflammatory response itself could result in a various range of chronic neuronal diseases including neuropathic pain after SCI. So, further studies are needed to demonstrate the mechanisms keeping a balance between neuro-inflammation and chronic neuronal diseases such as neuropathic pain.

As a non-selective cation channel, TRPA1 channels play an important role in transmitting pain signaling to higher order neurons. Based on our preliminary data, it is well known that electrophilic reactive compounds such as acrolein can activate TRPA1 channel after SCI. In addition, we have demonstrated that acrolein itself could increase the excitability of DRG sensory neuron, which is enhanced after SCI. Furthermore, our data indicated that SCI itself could affect the overall activity of sensory neurons’ sensitivity in nociceptor neurons (c and Aδ primary afferent neurons). Together with immunodotblotting results, our data demonstrated that formation of acrolein after SCI was correlated with the enhancement of neuronal sensitization and elevation of TRPA1 gene expression level after SCI. As we have already mentioned in chapter 1, many of research articles demonstrated the various mechanisms about the activation of TRPA1 channels after tissue damages. However, it is still unclear how this channel is deactivated. Additionally, we still do not know the exact downstream biochemical signaling pathway of the channel for subsequent activation after deactivation. Therefore, the logical next study would be to investigate the mechanisms of TRPA1 desensitization after its activation. Based on previous studies (Nilius et al.; Bandell et al., 2007), the binding between acrolein and cysteine residues by Michael addition is reversible, while the other covalent modifications are not. This property could be a key mechanism for re-activation after deactivation of TRPA1 channel and vice versa. Therefore, initially, we should focus on reversible reactions between the TRPA1 channel and its agonists.

As an additional approach to the TRPA1 channel desensitization study, we also should check the effect and role of TRPA1 channel specific antagonists, such as HC- 134

030031. Based on previous studies, oral administration of HC-030031 exhibited a significant decrease in neuropathic pain-like behavior in inflammatory and neuropathic pain animal models compared to TRPA1 antagonist non-treatment group (Eid et al., 2008). This study showed the potential analgesic effect of TRPA1 specific antagonist on neuropathic pain model. However, possible working mechanisms of non-electrophilic TRPA1 channel antagonists, such as HC-030031, have not been investigated. In addition, some research articles have shown that some electrophilic TRPA1 antagonist compounds, such as AP-18 and A-967079, can act as both an agonists and antagonists depending on their concentration (Garcia-Anoveros and Nagata, 2007; Chen et al., 2011). Therefore, without exact working mechanisms of TRPA1 antagonists, the potential value of TRPA1 antagonists as therapeutic agents remains uncertain. Thus, we need to investigate the mechanisms and optimal working concentrations of various TRPA1 specific antagonists, either electrophilic or non-electrophilic, through in vitro tests and in vivo disease animal models.

Although several unique properties of the TRPA1 channel make it an attractive potential therapeutic target to treat neuropathic pain after injury, there are potential disadvantages using TRPA1 channel for treatment. First of all, because of the wide spread presence of TRPA1 channel in various tissues, such as bladder, skin cell, GI track, nervous system, and respiratory track, there could be unwanted side effects when we use TRPA1 channel-specific medications. Second, the level of correlation between human TRPA1 and rodents TRPA1 is only 79%. Therefore, pre-clinical experimental results with TRPA1 antagonists or agonists in rodent model would not act in the same way in the human. In this sense, although the amino acid sequences in the N-terminal of TRPA1 or in the EF-hand domain share some similarities, the critical amino acid such as cysteine and lysine in the N-terminal differ between species. This difference could lead to changing in the affinity of the agonists or antagonists to the TRPA1 channel. To overcome this barrier, using a transgenic animal model which expresses the human TRPA1 channel could be an effective way to reduce such discrepancies. In addition, by 135

using this model, pre-clinical experiments can provide valuable information for developing effective analgesic compounds.

One of the challenging problems of pharmacological treatment is the delivery of compound to specific cell types. As we already mentioned above, the wide spread of expression of TRPA1 channel in nociceptors makes it an excellent candidate for therapeutic target. As a non-selective cation channel, TRPA1 channel’s pore size is stimulus- and time- dependent. This feature of the TRPA1 channel can be utilized to increase the entry of small membrane-impermeable cationic compounds into the target sensory neuron to block the pain transmission. To accomplish this aim, we may encapsulate therapeutic compounds in a drug delivery vehicle. By using a drug delivery vehicle, we can not only improve a compound’s specificity for targeted cells, but prevent undesirable side effects of TRPA1 channel.

LIST OF REFERENCES

136

LIST OF REFERENCES

Abraham K, Andres S, Palavinskas R, Berg K, Appel KE, Lampen A (2011) Toxicology and risk assessment of acrolein in food. Mol Nutr Food Res 55:1277-1290. Abram SE, Yi J, Fuchs A, Hogan QH (2006) Permeability of injured and intact peripheral nerves and dorsal root ganglia. Anesthesiology 105:146-153. Adams JD, Jr., Klaidman LK (1993) Acrolein-induced oxygen radical formation. Free Radical Biology & Medicine 15:187-193. Anand U, Otto WR, Facer P, Zebda N, Selmer I, Gunthorpe MJ, Chessell IP, Sinisi M, Birch R, Anand P (2008) TRPA1 receptor localisation in the human peripheral nervous system and functional studies in cultured human and rat sensory neurons. Neurosci Lett 438:221-227. Anderson DK, Dugan LL, Means ED, Horrocks LA (1994) Methylprednisolone and membrane properties of primary cultures of mouse spinal cord. Brain Res 637:119-125. Andersson DA, Gentry C, Moss S, Bevan S (2008) Transient receptor potential A1 is a sensory receptor for multiple products of oxidative stress. J Neurosci 28:2485- 2494. Andrade EL, Luiz AP, Ferreira J, Calixto JB (2008) Pronociceptive response elicited by TRPA1 receptor activation in mice. Neuroscience 152:511-520. Andre E, Gatti R, Trevisani M, Preti D, Baraldi PG, Patacchini R, Geppetti P (2009) Transient receptor potential ankyrin receptor 1 is a novel target for pro-tussive agents. Br J Pharmacol 158:1621-1628. Bandell M, Macpherson LJ, Patapoutian A (2007) From chills to chilis: mechanisms for thermosensation and chemesthesis via thermoTRPs. Curr Opin Neurobiol 17:490-497. Bandell M, Story GM, Hwang SW, Viswanath V, Eid SR, Petrus MJ, Earley TJ, Patapoutian A (2004) Noxious cold ion channel TRPA1 is activated by pungent compounds and bradykinin. Neuron 41:849-857. Bartholdi D, Schwab ME (1995) Methylprednisolone inhibits early inflammatory processes but not ischemic cell death after experimental spinal cord lesion in the rat. Brain Res 672:177-186. Basbaum AI, Bautista DM, Scherrer G, Julius D (2009) Cellular and molecular mechanisms of pain. Cell 139:267-284. Basso DM, Beattie MS, Bresnahan JC (1996) Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight-drop device versus transection. Exp Neurol 139:244-256.

137

Bautista DM, Pellegrino M, Tsunozaki M TRPA1: A Gatekeeper for Inflammation. Annu Rev Physiol 75:181-200. Bautista DM, Pellegrino M, Tsunozaki M (2013) TRPA1: A gatekeeper for inflammation. Annu Rev Physiol 75:181-200. Bautista DM, Movahed P, Hinman A, Axelsson HE, Sterner O, Hogestatt ED, Julius D, Jordt SE, Zygmunt PM (2005) Pungent products from garlic activate the sensory ion channel TRPA1. Proc Natl Acad Sci U S A 102:12248-12252. Bautista DM, Jordt SE, Nikai T, Tsuruda PR, Read AJ, Poblete J, Yamoah EN, Basbaum AI, Julius D (2006) TRPA1 mediates the inflammatory actions of environmental irritants and proalgesic agents. Cell 124:1269-1282. Bedi SS, Yang Q, Crook RJ, Du J, Wu Z, Fishman HM, Grill RJ, Carlton SM, Walters ET (2010) Chronic spontaneous activity generated in the somata of primary nociceptors is associated with pain-related behavior after spinal cord injury. J Neurosci 30:14870-14882. Behrmann DL, Bresnahan JC, Beattie MS (1994) Modeling of acute spinal cord injury in the rat: Neuroprotection and enhanced recovery with methylprednisolone, U- 74006F and YM-14673. Exp Neurol 126:61-75. Bessac BF, Jordt SE (2008) Breathtaking TRP channels: TRPA1 and TRPV1 in airway chemosensation and reflex control. Physiology (Bethesda) 23:360-370. Bessac BF, Sivula M, von Hehn CA, Escalera J, Cohn L, Jordt SE (2008) TRPA1 is a major oxidant sensor in murine airway sensory neurons. J Clin Invest 118:1899- 1910. Bessac BF, Sivula M, von Hehn CA, Caceres AI, Escalera J, Jordt SE (2009) Transient receptor potential ankyrin 1 antagonists block the noxious effects of toxic industrial isocyanates and tear gases. FASEB J 23:1102-1114. Blight AR (1983) Axonal physiology of chronic spinal cord injury in the cat: intracellular recording in vitro. Neuroscience 10:1471-1486. Blight AR (1991) Morphometric analysis of blood vessels in chronic experimental spinal cord injury: hypervascularity and recovery of function. J Neurol Sci 106:158-174. Blight AR, Decrescito V (1986) Morphometric analysis of experimental spinal cord injury in the cat: the relation of injury intensity to survival of myelinated axons. Neuroscience 19:321-341. Boddeke EW (2001) Involvement of chemokines in pain. Eur J Pharmacol 429:115-119. Bracken MB, Shepard MJ, Holford TR, Leo-Summers L, Aldrich EF, Fazl M, Fehlings MG, Herr DL, Hitchon PW, Marshall LF, Nockels RP, Pascale V, Perot PL, Jr., Piepmeier J, Sonntag VK, Wagner F, Wilberger JE, Winn HR, Young W (1998) Methylprednisolone or tirilazad mesylate administration after acute spinal cord injury: 1-year follow up. Results of the third National Acute Spinal Cord Injury randomized controlled trial. Journal of Neurosurgery 89:699-706. Braughler JM, Hall ED (1989a) Central nervous system trauma and stroke. I. Biochemical considerations for oxygen radical formation and lipid peroxidation. Free Radic Biol Med 6:289-301. Braughler JM, Hall ED (1989b) Central nervous system trauma and stroke I. Biochemical considerations for oxygen radical formation and lipid peroxidation. Free Radical BiolMed 6:289-301. 138

Bruce JC, Oatway MA, Weaver LC (2002) Chronic pain after clip-compression injury of the rat spinal cord. Exp Neurol 178:33-48. Burcham PC (2008) Potentialities and PItfalls Accompanying Chemico-Pharmacological Strategies against Endogenous Electrophiles and Cargonyl Stress. Chem Res Toxicol 21:779-786. Burcham PC, Pyke SM (2006) Hydralazine inhibits rapid acrolein-induced protein oligomerization: role of aldehyde scavenging and adduct trapping in cross-link blocking and cytoprotection. Mol Pharmacol 69:1056-1065. Burcham PC, Kerr PG, Fontaine F (2000) The antihypertensive hydralazine is an efficient scavenger of acrolein. Redox Rep 5:47-49. Burcham PC, Kaminskas LM, Fontaine FR, Petersen DR, Pyke SM (2002) Aldehyde- sequestering drugs: tools for studying protein damage by lipid peroxidation products. Toxicology 181-182:229-236. Burcham PC, Fontaine FR, Kaminskas LM, Petersen DR, Pyke SM (2004) Protein adduct-trapping by hydrazinophthalazine drugs: mechanisms of cytoprotection against acrolein-mediated toxicity. Mol Pharmacol 65:655-664. Cairns DM, Adkins RH, Scott MD (1996) Pain and depression in acute traumatic spinal cord injury: origins of chronic problematic pain? Arch Phys Med Rehabil 77:329- 335. Campbell JN, Meyer RA (2006) Mechanisms of neuropathic pain. Neuron 52:77-92. Carmella SG, Chen M, Zhang Y, Zhang S, Hatsukami DK, Hecht SS (2007) Quantitation of acrolein-derived (3-hydroxypropyl)mercapturic acid in human urine by liquid chromatography-atmospheric pressure chemical ionization tandem mass spectrometry: effects of cigarette smoking. Chem Res Toxicol 20:986-990. Chaplan SR, Bach FW, Pogrel JW, Chung JM, Yaksh TL (1994) Quantitative assessment of tactile allodynia in the rat paw. Journal of neuroscience methods 53:55-63. Chen A, Xu XM, Kleitman N, Bunge MB (1996) Methylprednisolone administration improves axonal regeneration into Schwann cell grafts in transected adult rat thoracic spinal cord. Experimental Neurology 138:261-276. Chen J et al. (2011) Selective blockade of TRPA1 channel attenuates pathological pain without altering noxious cold sensation or body temperature regulation. Pain 152:1165-1172. Chen Q, Pan HL (2007) Signaling mechanisms of angiotensin II-induced attenuation of GABAergic input to hypothalamic presympathetic neurons. J Neurophysiol 97:3279-3287. Chiu SY, Ritchie JM (1980) Potassium channels in nodal and internodal axonal membrane of mammalian myelinated fibres. Nature 284:170-171. Clapham DE (2003) TRP channels as cellular sensors. Nature 426:517-524. Clapham DE, Runnels LW, Strubing C (2001) The TRP ion channel family. Nat Rev Neurosci 2:387-396. Colburn RW, Lubin ML, Stone DJ, Jr., Wang Y, Lawrence D, D'Andrea MR, Brandt MR, Liu Y, Flores CM, Qin N (2007) Attenuated cold sensitivity in TRPM8 null mice. Neuron 54:379-386. Cole RA, Weiner MF (1960) Clinical and theoretical observations on phenelzine (nardil), an antidepressant agent. Am J Psychiatry 117:361-362. 139

Constantini S, Young W (1994) The effects of methylprednisolone and the ganglioside GM1 on acute spinal cord injury in rats. J Neurosurg 80:97-111. Costigan M, Scholz J, Woolf CJ (2009a) Neuropathic pain: a maladaptive response of the nervous system to damage. Annu Rev Neurosci 32:1-32. Costigan M, Moss A, Latremoliere A, Johnston C, Verma-Gandhu M, Herbert TA, Barrett L, Brenner GJ, Vardeh D, Woolf CJ, Fitzgerald M (2009b) T-cell infiltration and signaling in the adult dorsal spinal cord is a major contributor to neuropathic pain-like hypersensitivity. J Neurosci 29:14415-14422. Dai Y, Wang S, Tominaga M, Yamamoto S, Fukuoka T, Higashi T, Kobayashi K, Obata K, Yamanaka H, Noguchi K (2007) Sensitization of TRPA1 by PAR2 contributes to the sensation of inflammatory pain. J Clin Invest 117:1979-1987. del Camino D, Murphy S, Heiry M, Barrett LB, Earley TJ, Cook CA, Petrus MJ, Zhao M, D'Amours M, Deering N, Brenner GJ, Costigan M, Hayward NJ, Chong JA, Fanger CM, Woolf CJ, Patapoutian A, Moran MM (2010) TRPA1 contributes to cold hypersensitivity. J Neurosci 30:15165-15174. Devor M (1999) Unexplained peculiarities of the dorsal root ganglion. Pain 82:S27-S35. Dhaka A, Viswanath V, Patapoutian A (2006) Trp ion channels and temperature sensation. Annu Rev Neurosci 29:135-161. Doerner JF, Gisselmann G, Hatt H, Wetzel CH (2007) Transient receptor potential channel A1 is directly gated by calcium ions. J Biol Chem 282:13180-13189. Donnelly DJ, Popovich PG (2008) Inflammation and its role in neuroprotection, axonal regeneration and functional recovery after spinal cord injury. Exp Neurol 209:378-388. Due MR, Park J, Zheng L, Walls M, Allette YM, White FA, Shi R (2014) Acrolein involvement in sensory and behavioral hypersensitivity following spinal cord injury in the rat. J Neurochem 128:776-786. Eckert E, Drexler H, Goen T (2010) Determination of six hydroxyalkyl mercapturic acids in human urine using hydrophilic interaction liquid chromatography with tandem mass spectrometry (HILIC-ESI-MS/MS). J Chromatogr B Analyt Technol Biomed Life Sci 878:2506-2514. Eid SR, Crown ED, Moore EL, Liang HA, Choong KC, Dima S, Henze DA, Kane SA, Urban MO (2008) HC-030031, a TRPA1 selective antagonist, attenuates inflammatory- and neuropathy-induced mechanical hypersensitivity. Mol Pain 4:48. Esterbauer H, Schaur RJ, Zollner H (1991a) Chemistry and biochemistry of 4- hydroxynonenal, malonaldehyde and related aldehydes. Free Radical Biology & Medicine 11:81-128. Esterbauer H, Schaur RJ, Zollner H (1991b) Chemistry and biochemistry of 4- hydroxynonenal, malonaldehyde and related aldehydes. Free Radic Biol Med 11:81-128. Facchinetti F, Amadei F, Geppetti P, Tarantini F, Di Serio C, Dragotto A, Gigli PM, Catinella S, Civelli M, Patacchini R (2007) Alpha,beta-unsaturated aldehydes in cigarette smoke release inflammatory mediators from human macrophages. Am J Respir Cell Mol Biol 37:617-623. 140

Ficker E, Heinemann U (1992) Slow and fast transient potassium currents in cultured rat hippocampal cells. J Physiol 445:431-455. Gao X, Kim HK, Chung JM, Chung K (2007) Reactive oxygen species (ROS) are involved in enhancement of NMDA-receptor phosphorylation in animal models of pain. Pain 131:262-271. Garcia-Anoveros J, Nagata K (2007) Trpa1. Handb Exp Pharmacol:347-362. George ER, Scholten DJ, Buechler CM, Jordan-Tibbs J, Mattice C, Albrecht RM (1995) Failure of methylprednisolone to improve the outcome of spinal cord injuries. Am Surg 61:659-663; discussion 663-654. George ER, D. J. Scholten, C. M. Buechler, J. Jordan-Tibbs, C. Mattice, R. M. Albrecht (1995) Failure of methylprednisolone to improve the outcome of spinal cord injuries. The American Surgeon 61:659-663. Ghilarducci DP, Tjeerdema RS (1995) Fate and effects of acrolein. Rev Environ Contam Toxicol 144:95-146. Giger RJ, Hollis ER, 2nd, Tuszynski MH Guidance molecules in axon regeneration. Cold Spring Harb Perspect Biol 2:a001867. Gilden DH (2005) Infectious causes of multiple sclerosis. Lancet Neurol 4:195-202. Gilgun-Sherki Y, Melamed E, Offen D (2004) The role of oxidative stress in the pathogenesis of multiple sclerosis: the need for effective antioxidant therapy. J Neurol 251:261-268. Glezer I, Simard AR, Rivest S (2007) Neuroprotective role of the innate immune system by microglia. Neuroscience 147:867-883. Haghighi SS, Pugh SL, Perez-Espejo MA, Oro JJ (1995) Effect of 4-aminopyridine in acute spinal cord injury. Surg Neurol 43:443-447. Hains BC, Klein JP, Saab CY, Craner MJ, Black JA, Waxman SG (2003) Upregulation of sodium channel Nav1.3 and functional involvement in neuronal hyperexcitability associated with central neuropathic pain after spinal cord injury. J Neurosci 23:8881-8892. Hall ED (1989) Free radicals and CNS injury. Critical Care Clinics 5:793-805. Hall ED (1996) Lipid peroxidation. Advances in Neurology 71:247-257; discussion 257- 248. Hall ED, Braughler JM (1993a) Free radicals in CNS injury. Res Publ Assoc Res Nerv Ment Dis 71:81-105. Hall ED, Braughler JM (1993b) Free radicals in CNS injury. Research Publications - Association for Research in Nervous & Mental Disease 71:81-105. Hall ED, Springer JE (2004) Neuroprotection and acute spinal cord injury: a reappraisal. NeuroRx 1:80-100. Halliwell B, Gutteridge JMC (1999) Free radicals in biology and medicine. Oxford: Oxford University Press. Hamann K, Shi R (2009) Acrolein scavenging: a potential novel mechanism of attenuating oxidative stress following spinal cord injury. J Neurochem 111:1348- 1356. Hamann K, Nehrt G, Ouyang H, Duerstock B, Shi R (2008a) Hydralazine inhibits compression and acrolein-mediated injuries in ex vivo spinal cord. J Neurochem 104:708-718. 141

Hamann K, Durkes A, Ouyang H, Uchida K, Pond A, Shi R (2008b) Critical role of acrolein in secondary injury following ex vivo spinal cord trauma. J Neurochem 107:712-721. Hayes KC, Hsieh JT, Potter PJ, Wolfe DL, Delaney GA, Blight AR (1993) Effects of induced hypothermia on somatosensory evoked potentials in patients with chronic spinal cord injury. Paraplegia 31:730-741. Hensley K, Robinson KA, Gabbita SP, Salsman S, Floyd RA (2000) Reactive oxygen species, cell signaling, and cell injury. Free Radic Biol Med 28:1456-1462. Hinman A, Chuang HH, Bautista DM, Julius D (2006) TRP channel activation by reversible covalent modification. Proc Natl Acad Sci U S A 103:19564-19568. Horton ND, Biswal SS, Corrigan LL, Bratta J, Kehrer JP (1999) Acrolein Causes Inhibitor κB-independent Decreases in Nuclear Factor Kappa B Activation in Human Lung Adenocarcinoma (A549) Cells. J Biol Chem 247. Huang YC, Huang YY (2006) Biomaterials and strategies for nerve regeneration. Artif Organs 30:514-522. Hulsebosch CE (2008) Gliopathy ensures persistent inflammation and chronic pain after spinal cord injury. Exp Neurol 214:6-9. Hulsebosch CE, Hains BC, Crown ED, Carlton SM (2009) Mechanisms of chronic central neuropathic pain after spinal cord injury. Brain Res Rev 60:202-213. Ismahil MA, Hamid T, Haberzettl P, Gu Y, Chandrasekar B, Srivastava S, Bhatnagar A, Prabhu SD (2011) Chronic oral exposure to the aldehyde pollutant acrolein induces dilated cardiomyopathy. Am J Physiol Heart Circ Physiol 301:H2050- 2060. Jensen JM, Shi R (2003) Effects of 4-aminopyridine on stretched mammalian spinal cord: the role of potassium channels in axonal conduction. J Neurophysiol 90:2334- 2340. Jones BJ, Roberts DJ (1968) The quantiative measurement of motor inco-ordination in naive mice using an acelerating rotarod. J Pharm Pharmacol 20:302-304. Jones TB, McDaniel EE, Popovich PG (2005) Inflammatory-mediated injury and repair in the traumatically injured spinal cord. Current pharmaceutical design 11:1223- 1236. Jung H, Toth PT, White FA, Miller RJ (2008) Monocyte chemoattractant protein-1 functions as a neuromodulator in dorsal root ganglia neurons. J Neurochem 104:254-263. Juurlink BH, Paterson PG (1998a) Review of oxidative stress in brain and spinal cord injury: suggestions for pharmacological and nutritional management strategies. J Spinal Cord Med 21:309-334. Juurlink BH, Paterson PG (1998b) Review of oxidative stress in brain and spinal cord injury: suggestions for pharmacological and nutritional management strategies. Journal of Spinal Cord Medicine 21:309-334. Kakulas BA (1999) A review of the neuropathology of human spinal cord injury with emphasis on special features. J Spinal Cord Med 22:119-124. Kaminskas LM, Pyke SM, Burcham PC (2004a) Reactivity of hydrazinophthalazine drugs with the lipid peroxidation products acrolein and crotonaldehyde. Org Biomol Chem 2:2578-2584. 142

Kaminskas LM, Pyke SM, Burcham PC (2004b) Strong protein adduct trapping accompanies abolition of acrolein-mediated hepatotoxicity by hydralazine in mice. J Pharmacol Exp Ther 310:1003-1010. Kehrer JP, Biswal SS (2000a) The molecular effects of acrolein. Toxicological Sciences 57:6-15. Kehrer JP, Biswal SS (2000b) The molecular effects of acrolein. Toxicol Sci 57:6-15. Kerstein PC, del Camino D, Moran MM, Stucky CL (2009) Pharmacological blockade of TRPA1 inhibits mechanical firing in nociceptors. Mol Pain 5:19. Kim HK, Park SK, Zhou JL, Taglialatela G, Chung K, Coggeshall RE, Chung JM (2004) Reactive oxygen species (ROS) play an important role in a rat model of neuropathic pain. Pain 111:116-124. Kim HK, Kim JH, Gao X, Zhou JL, Lee I, Chung K, Chung JM (2006) Analgesic effect of vitamin E is mediated by reducing central sensitization in neuropathic pain. Pain 122:53-62. Kirkham PA, Spooner G, Ffoulkes-Jones C, Calvez R (2003) Cigarette smoke triggers macrophage adhesion and activation: role of lipid peroxidation products and scavenger receptor. Free Radic Biol Med 35:697-710. Knerlich-Lukoschus F, von der Ropp-Brenner B, Lucius R, Mehdorn HM, Held-Feindt J (2010) Chemokine expression in the white matter spinal cord precursor niche after force-defined spinal cord contusion injuries in adult rats. Glia 58:916-931. Knerlich-Lukoschus F, von der Ropp-Brenner B, Lucius R, Mehdorn HM, Held-Feindt J (2011a) Spatiotemporal CCR1, CCL3(MIP-1alpha), CXCR4, CXCL12(SDF- 1alpha) expression patterns in a rat spinal cord injury model of posttraumatic neuropathic pain. J Neurosurg Spine 14:583-597. Knerlich-Lukoschus F, Juraschek M, Blomer U, Lucius R, Mehdorn HM, Held-Feindt J (2008) Force-dependent development of neuropathic central pain and time-related CCL2/CCR2 expression after graded spinal cord contusion injuries of the rat. J Neurotrauma 25:427-448. Knerlich-Lukoschus F, Noack M, von der Ropp-Brenner B, Lucius R, Mehdorn HM, Held-Feindt J (2011b) Spinal cord injuries induce changes in CB1 cannabinoid receptor and C-C chemokine expression in brain areas underlying circuitry of chronic pain conditions. J Neurotrauma 28:619-634. Knowlton WM, Bifolck-Fisher A, Bautista DM, McKemy DD (2010) TRPM8, but not TRPA1, is required for neural and behavioral responses to acute noxious cold temperatures and cold-mimetics in vivo. Pain 150:340-350. Korenke AR, Rivey MP, Allington DR (2008) Sustained-release fampridine for symptomatic treatment of multiple sclerosis. Ann Pharmacother 42:1458-1465. Kumar V, Abbas AK, Fausto N (2004) Robbins & Cotran Pathologic Basis of Disease, 7 Edition. Philadelphia: Elsevier. Lee HC, Cho DY, Lee WY, Chuang HC (2007) Pitfalls in treatment of acute cervical spinal cord injury using high-dose methylprednisolone: a retrospect audit of 111 patients. Surgical Neurology 68:S37-41. Lennertz RC, Kossyreva EA, Smith AK, Stucky CL (2012) TRPA1 mediates mechanical sensitization in nociceptors during inflammation. PLoS One 7:e43597. 143

Leung G, Sun W, Brookes S, Smith D, Shi R (2011a) Potassium channel blocker, 4- aminopyridine-3-methanol, restores axonal conduction in spinal cord of an animal model of multiple sclerosis. Exp Neurol 227:232-235. Leung G, Sun W, Zheng L, Brookes S, Tully M, Shi R (2011b) Anti-acrolein treatment improves behavioral outcome and alleviates myelin damage in experimental autoimmune enchephalomyelitis mouse. Neuroscience 173:150-155. Leung G, Sun W, Zheng L, Brookes S, Tully M, Shi R (2011c) Anti-acrolein treatment improves behavioral outcome and alleviates myelin damage in experimental autoimmune encephalomyelitis mouse. Neuroscience 173:150-155. Lewen A, Matz P, Chan PH (2000) Free radical pathways in CNS injury. J Neurotrauma 17:871-890. Liu-Snyder P, Borgens RB, Shi R (2006a) Hydralazine rescues PC12 cells from acrolein- mediated death. J Neurosci Res 84:219-227. Liu-Snyder P, McNally H, Shi R, Borgens RB (2006b) Acrolein-mediated mechanisms of neuronal death. Journal of neuroscience research 84:209-218. Liu M, Wood JN The roles of sodium channels in nociception: implications for mechanisms of neuropathic pain. Pain Med 12 Suppl 3:S93-99. Liu T, Ji RR Oxidative stress induces itch via activation of transient receptor potential subtype ankyrin 1 in mice. Neurosci Bull 28:145-154. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real- time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25:402-408. Lovell MA, Xie C, Markesbery WR (2000) Acrolein, a product of lipid peroxidation, inhibits glucose and glutamate uptake in primary neuronal cultures. Free Radic Biol Med 29:714-720. Lovell MA, Xie C, Markesbery WR (2001a) Acrolein is increased in Alzheimer's disease brain and is toxic to primary hippocampal cultures. Neurobiology of Aging 22:187-194. Lovell MA, Xie C, Markesbery WR (2001b) Acrolein is increased in Alzheimer's disease brain and is toxic to primary hippocampal cultures. Neurobiol Aging 22:187-194. Lu Y, Zheng J, Xiong L, Zimmermann M, Yang J (2008) Spinal cord injury-induced attenuation of GABAergic inhibition in spinal dorsal horn circuits is associated with down-regulation of the chloride transporter KCC2 in rat. J Physiol 586:5701- 5715. Lucas SM, Rothwell NJ, Gibson RM (2006) The role of inflammation in CNS injury and disease. Br J Pharmacol 147 Suppl 1:S232-240. Luo J, Shi R (2004) Acrolein induces axolemmal disruption, oxidative stress, and mitochondrial impairment in spinal cord tissue. Neurochem Int 44:475-486. Luo J, Robinson JP, Shi R (2005a) Acrolein-induced cell death in PC12 cells: role of mitochondria-mediated oxidative stress. Neurochem Int 47:449-457. Luo J, Uchida K, Shi R (2005b) Accumulation of acrolein-protein adducts after traumatic spinal cord injury. Neurochem Res 30:291-295. Luo J, Hill BG, Gu Y, Cai J, Srivastava S, Bhatnagar A, Prabhu SD (2007) Mechanisms of acrolein-induced myocardial dysfunction: implications for environmental and endogenous aldehyde exposure. Am J Physiol Heart Circ Physiol 293:H3673- 3684. 144

Ma C, LaMotte RH (2005) Enhanced excitability of dissociated primary sensory neurons after chronic compression of the dorsal root ganglion in the rat. Pain 113:106-112. Macpherson LJ, Xiao B, Kwan KY, Petrus MJ, Dubin AE, Hwang S, Cravatt B, Corey DP, Patapoutian A (2007) An ion channel essential for sensing chemical damage. J Neurosci 27:11412-11415. Marchand F, Perretti M, McMahon SB (2005) Role of the immune system in chronic pain. Nat Rev Neurosci 6:521-532. Matsumoto T, T. Tamaki, Kawakami M, Yoshida M, Ando H, Yamada H (2001) Early complications of high-dose methylprednisolone sodium succinate treatment in the follow-up of acute cervical spinal cord injury. Spine 26:426-430. Matta JA, Cornett PM, Miyares RL, Abe K, Sahibzada N, Ahern GP (2008) General anesthetics activate a nociceptive ion channel to enhance pain and inflammation. Proc Natl Acad Sci U S A 105:8784-8789. McKemy DD, Neuhausser WM, Julius D (2002) Identification of a cold receptor reveals a general role for TRP channels in thermosensation. Nature 416:52-58. Milligan ED, Watkins LR (2009) Pathological and protective roles of glia in chronic pain. Nat Rev Neurosci 10:23-36. Moalem G, Tracey DJ (2006) Immune and inflammatory mechanisms in neuropathic pain. Brain Res Rev 51:240-264. Montine TJ, Neely MD, Quinn JF, Beal MF, Markesbery WR, Roberts LJ, Morrow JD (2002) Lipid peroxidation in aging brain and Alzheimer's disease. Free Radic Biol Med 33:620-626. Moretto N, Facchinetti F, Southworth T, Civelli M, Singh D, Patacchini R (2009) alpha,beta-Unsaturated aldehydes contained in cigarette smoke elicit IL-8 release in pulmonary cells through mitogen-activated protein kinases. Am J Physiol Lung Cell Mol Physiol 296:L839-848. Mudge K, Van Dolson L, Lake AS (1984) Progressive cystic degeneration of the spinal cord following spinal cord injury. Spine 9:253-255. Naoi M, Maruyama W, Shamoto-Nagai M, Yi H, Akao Y, Tanaka M (2005) Oxidative stress in mitochondria: decision to survival and death of neurons in neurodegenerative disorders. Mol Neurobiol 31:81-93. Nguyen T, Mehta NR, Conant K, Kim KJ, Jones M, Calabresi PA, Melli G, Hoke A, Schnaar RL, Ming GL, Song H, Keswani SC, Griffin JW (2009) Axonal protective effects of the myelin-associated glycoprotein. J Neurosci 29:630-637. Nilius B, Appendino G, Owsianik G The transient receptor potential channel TRPA1: from gene to pathophysiology. Pflugers Arch 464:425-458. Nilius B, Appendino G, Owsianik G (2012) The transient receptor potential channel TRPA1: from gene to pathophysiology. Pflugers Arch 464:425-458. Nilius B, Owsianik G, Voets T, Peters JA (2007) Transient receptor potential cation channels in disease. Physiol Rev 87:165-217. Nozawa K, Kawabata-Shoda E, Doihara H, Kojima R, Okada H, Mochizuki S, Sano Y, Inamura K, Matsushime H, Koizumi T, Yokoyama T, Ito H (2009) TRPA1 regulates gastrointestinal motility through serotonin release from enterochromaffin cells. Proc Natl Acad Sci U S A 106:3408-3413. 145

Nurick S, Russell JA, Deck MD (1970) Cystic degeneration of the spinal cord following spinal cord injury. Brain : a journal of neurology 93:211-222. O'Brien PJ, Siraki AG, Shangari N (2005) Aldehyde sources, metabolism, molecular toxicity mechanisms, and possible effects on human health. Crit Rev Toxicol 35:609-662. Parent MB, Master S, Kashlub S, Baker GB (2002) Effects of the antidepressant/antipanic drug phenelzine and its putative metabolite phenylethylidenehydrazine on extracellular gamma-aminobutyric acid levels in the striatum. Biochem Pharmacol 63:57-64. Park ES, Gao X, Chung JM, Chung K (2006) Levels of mitochondrial reactive oxygen species increase in rat neuropathic spinal dorsal horn neurons. Neurosci Lett 391:108-111. Park J, Lim E, Back S, Na H, Park Y, Sun K (2010) Nerve regeneration following spinal cord injury using matrix metalloproteinase-sensitive, hyaluronic acid-based biomimetic hydrogel scaffold containing brain-derived neurotrophic factor. J Biomed Mater Res A 93:1091-1099. Park J, Zheng L, Marquis A, Walls M, Duerstock B, Pond A, Alvarez SV, He W, Ouyang Z, Shi R (2013) Neuroprotective role of hydralazine in rat spinal cord injury- attenuation of acrolein-mediated damage. J Neurochem. Park J, Zheng L, Marquis A, Walls M, Duerstock B, Pond A, Vega-Alvarez S, Wang H, Ouyang Z, Shi R (2014) Neuroprotective role of hydralazine in rat spinal cord injury-attenuation of acrolein-mediated damage. J Neurochem 129:339-349. Park LC, Calingasan NY, Uchida K, Zhang H, Gibson GE (2000) Metabolic impairment elicits brain cell type-selective changes in oxidative stress and cell death in culture. J Neurochem 74:114-124. Patapoutian A, Macpherson L (2006) Channeling pain. Nat Med 12:506-507. Patapoutian A, Tate S, Woolf CJ (2009) Transient receptor potential channels: targeting pain at the source. Nat Rev Drug Discov 8:55-68. Patapoutian A, Peier AM, Story GM, Viswanath V (2003) ThermoTRP channels and beyond: mechanisms of temperature sensation. Nat Rev Neurosci 4:529-539. Pena F, Tapia R (1999) Relationships among seizures, extracellular amino acid changes, and neurodegeneration induced by 4-aminopyridine in rat hippocampus: a microdialysis and electroencephalographic study. J Neurochem 72:2006-2014. Pereira JE, Costa LM, Cabrita AM, Couto PA, Filipe VM, Magalhaes LG, Fornaro M, Di Scipio F, Geuna S, Mauricio AC, Varejao AS (2009) Methylprednisolone fails to improve functional and histological outcome following spinal cord injury in rats. Exp Neurol 220:71-81. Perin-Martins A, Teixeira JM, Tambeli CH, Parada CA, Fischer L (2013) Mechanisms underlying transient receptor potential ankyrin 1 (TRPA1)-mediated hyperalgesia and edema. J Peripher Nerv Syst 18:62-74. Piao LH, Fujita T, Jiang CY, Liu T, Yue HY, Nakatsuka T, Kumamoto E (2009) TRPA1 activation by lidocaine in nerve terminals results in glutamate release increase. Biochem Biophys Res Commun 379:980-984. Picklo MJ, Montine TJ (2001) Acrolein inhibits respiration in isolated brain mitochondria. Biochim Biophys Acta 1535:145-152. 146

Potter PJ, Hayes KC, Segal JL, Hsieh JT, Brunnemann SR, Delaney GA, Tierney DS, Mason D (1998) Randomized double-blind crossover trial of fampridine-SR (sustained release 4-aminopyridine) in patients with incomplete spinal cord injury. J Neurotrauma 15:837-849. Puljak L, Kojundzic SL, Hogan QH, Sapunar D (2009) Targeted delivery of pharmacological agents into rat dorsal root ganglion. J Neurosci Methods 177:397-402. Reece PA (1981) Hydralazine and related compounds: chemistry, metabolism, and mode of action. Med Res Rev 1:73-96. Ren K, Dubner R (2008) Neuron-glia crosstalk gets serious: role in pain hypersensitivity. Curr Opin Anaesthesiol 21:570-579. Richards JS, Meredith RL, Nepomuceno C, Fine PR, Bennett G (1980) Psycho-social aspects of chronic pain in spinal cord injury. Pain 8:355-366. Rooney GE, Endo T, Ameenuddin S, Chen B, Vaishya S, Gross L, Schiefer TK, Currier BL, Spinner RJ, Yaszemski MJ, Windebank AJ (2009) Importance of the vasculature in cyst formation after spinal cord injury. Journal of neurosurgery Spine 11:432-437. Sandhir R, Gregory E, He YY, Berman NE Upregulation of inflammatory mediators in a model of chronic pain after spinal cord injury. Neurochem Res 36:856-862. Schettgen T, Musiol A, Kraus T (2008) Simultaneous determination of mercapturic acids derived from ethylene oxide (HEMA), propylene oxide (2-HPMA), acrolein (3- HPMA), acrylamide (AAMA) and N,N-dimethylformamide (AMCC) in human urine using liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom 22:2629-2638. Schnell L, Fearn S, Schwab ME, Perry VH, Anthony DC (1999) Cytokine-induced acute inflammation in the brain and spinal cord. Journal of Neuropathology & Experimental Neurology 58:245-254. Scholz J, Woolf CJ (2002) Can we conquer pain? Nat Neurosci 5 Suppl:1062-1067. Scholz J, Woolf CJ (2007) The neuropathic pain triad: neurons, immune cells and glia. Nat Neurosci 10:1361-1368. Schurch B, Wichmann W, Rossier AB (1996) Post-traumatic syringomyelia (cystic myelopathy): A prospective study of 449 patients with spinal cord injury. J Neurol Neurosurg Psychiatry 60:61-67. Schwartz ES, Lee I, Chung K, Chung JM (2008) Oxidative stress in the spinal cord is an important contributor in capsaicin-induced mechanical secondary hyperalgesia in mice. Pain 138:514-524. Segatore M (1994) Understanding chronic pain after spinal cord injury. J Neurosci Nurs 26:230-236. Seiler N (2000) Oxidation of Polyamines and Brain Injury. Neurochemical Research 25:471-490. Shi R, Blight AR (1996) Compression injury of mammalian spinal cord in vitro and the dynamics of action potential conduction failure. J Neurophysiol 76:1572-1580. Shi R, Blight AR (1997) Differential effects of low and high concentrations of 4- aminopyridine on axonal conduction in normal and injured spinal cord. Neuroscience 77:553-562. 147

Shi R, Luo L (2006) The role of acrolein in spinal cord injury. Applied Neurology 2:22- 27. Shi R, Luo J, Peasley MA (2002a) Acrolein inflicts axonal membrane disruption and conduction loss in isolated guinea pig spinal cord. Neuroscience 115:337-340. Shi R, Luo J, Peasley M (2002b) Acrolein inflicts axonal membrane disruption and conduction loss in isolated guinea-pig spinal cord. Neuroscience 115:337-340. Shi R, Rickett T, Sun W (2011) Acrolein-mediated injury in nervous system trauma and diseases. Molecular nutrition & food research 55:1320-1331. Shi Y, Sun W, McBride JJ, Cheng JX, Shi R Acrolein induces myelin damage in mammalian spinal cord. J Neurochem 117:554-564. Siddall PJ, McClelland JM, Rutkowski SB, Cousins MJ (2003) A longitudinal study of the prevalence and characteristics of pain in the first 5 years following spinal cord injury. Pain 103:249-257. Smith KJ, Kapoor R, Felts PA (1999) Demyelination: the role of reactive oxygen and nitrogen species. Brain Pathol 9:69-92. Song MS, Baker GB, Dursun SM, Todd KG (2010) The antidepressant phenelzine protects neurons and astrocytes against formaldehyde-induced toxicity. J Neurochem 114:1405-1413. Song XJ, Wang ZB, Gan Q, Walters ET (2006) cAMP and cGMP contribute to sensory neuron hyperexcitability and hyperalgesia in rats with dorsal root ganglia compression. J Neurophysiol 95:479-492. Stevens JF, Maier CS (2008) Acrolein: sources, metabolism, and biomolecular interactions relevant to human health and disease. Molecular nutrition & food research 52:7-25. Stonnington C, Tecoma ES, Choi DW (1988) Phenelzine toxicity: case report of an electroencephalographically defined biphasic syndrome. J Clin Psychopharmacol 8:382-384. Stork CM, Hoffman RS (1994) Characterization of 4-aminopyridine in overdose. J Toxicol Clin Toxicol 32:583-587. Story GM, Peier AM, Reeve AJ, Eid SR, Mosbacher J, Hricik TR, Earley TJ, Hergarden AC, Andersson DA, Hwang SW, McIntyre P, Jegla T, Bevan S, Patapoutian A (2003) ANKTM1, a TRP-like channel expressed in nociceptive neurons, is activated by cold temperatures. Cell 112:819-829. Suberviola B, González-Castro A, Llorca J, Ortiz-Melón F, Miñambres E (2008a) Early complications of high-dose methylprednisolone in acute spinal cord injury patients. Injury 39:748-752. Suberviola B, Gonzalez-Castro A, Llorca J, Ortiz-Melon F, Minambres E (2008b) Early complications of high-dose methylprednisolone in acute spinal cord injury patients. Injury 39:748-752. Sun W, Smith D, Fu Y, Cheng JX, Bryn S, Borgens R, Shi R (2010) Novel potassium channel blocker, 4-AP-3-MeOH, inhibits fast potassium channels and restores axonal conduction in injured guinea pig spinal cord white matter. J Neurophysiol 103:469-478. Tai C, Zhu S, Zhou N (2008) TRPA1: the central molecule for chemical sensing in pain pathway? J Neurosci 28:1019-1021. 148

Takeda M, Tsuboi Y, Kitagawa J, Nakagawa K, Iwata K, Matsumoto S (2011) Potassium channels as a potential therapeutic target for trigeminal neuropathic and inflammatory pain. Mol Pain 7:5. Trivedi A, Olivas AD, Noble-Haeusslein LJ (2006) Inflammation and Spinal Cord Injury: Infiltrating Leukocytes as Determinants of Injury and Repair Processes. Clinical neuroscience research 6:283-292. Tuncel M, Ram VC (2003) Hypertensive emergencies. Etiology and management. Am J Cardiovasc Drugs 3:21-31. Uchida K (1999a) Current status of acrolein as a lipid peroxidation product. Trends Cardiovasc Med 9:109-113. Uchida K (1999b) Current status of acrolein as a lipid peroxidation product. Trends in Cardiovascular Medicine 9:109-113. Uchida K, Kanematsu M, Morimitsu Y, Osawa T, Noguchi N, Niki E (1998a) Acrolein is a product of lipid peroxidation reaction. Formation of free acrolein and its conjugate with lysine residues in oxidized low density lipoproteins. J Biol Chem 273:16058-16066. Uchida K, Kanematsu M, Sakai K, Matsuda T, Hattori N, Mizuno Y, Suzuki D, Miyata T, Noguchi N, Niki E, Osawa T (1998b) Protein-bound acrolein: potential markers for oxidative stress. Proceedings of the National Academy of Sciences of the United States of America 95:4882-4887. Van Diemen HA, Polman CH, Koetsier JC, Van Loenen AC, Nauta JJ, Bertelsmann FW (1993) 4-Aminopyridine in patients with multiple sclerosis: dosage and serum level related to efficacy and safety. Clin Neuropharmacol 16:195-204. Vandewauw I, Owsianik G, Voets T (2013) Systematic and quantitative mRNA expression analysis of TRP channel genes at the single trigeminal and dorsal root ganglion level in mouse. BMC Neurosci 14:21. Venkatachalam K, Montell C (2007) TRP channels. Annu Rev Biochem 76:387-417. Verrilli MR, Salanga VD, Kozachuk WE, Bennetts M (1987) Phenelzine toxicity responsive to dantrolene. Neurology 37:865-867. Vincent AM, Brownlee M, Russell JW (2002) Oxidative stress and programmed cell death in diabetic neuropathy. Ann N Y Acad Sci 959:368-383. Virgili M, Contestabile A (2000) Partial neuroprotection of in vivo excitotoxic brain damage by chronic administration of the red wine antioxidant agent, trans- resveratrol in rats. Neurosci Lett 281:123-126. Wang H, Manicke NE, Yang Q, Zheng L, Shi R, Cooks RG, Ouyang Z (2011) Direct analysis of biological tissue by paper spray mass spectrometry. Anal Chem 83:1197-1201. Wang HT, Hu Y, Tong D, Huang J, Gu L, Wu XR, Chung FL, Li GM, Tang MS (2012) Effect of carcinogenic acrolein on DNA repair and mutagenic susceptibility. J Biol Chem 287:12379-12386. Waxman SG (1989) Demyelination in spinal cord injury. J Neurol Sci 91:1-14. Waxman SG (2001) Acquired channelopathies in nerve injury and MS. Neurology 56:1621-1627. White FA, Bhangoo SK, Miller RJ (2005) Chemokines: integrators of pain and inflammation. Nature reviews Drug discovery 4:834-844. 149

White FA, Jung H, Miller RJ (2007) Chemokines and the pathophysiology of neuropathic pain. Proc Natl Acad Sci U S A 104:20151-20158. White FA, Feldman P, Miller RJ (2009) Chemokine signaling and the management of neuropathic pain. Mol Interv 9:188-195. Witz G (1989) Biological interactions of alpha,beta-unsaturated aldehydes. Free Radic Biol Med 7:333-349. Wood PL, Khan MA, Moskal JR, Todd KG, Tanay VA, Baker G (2006a) Aldehyde load in ischemia-reperfusion brain injury: neuroprotection by neutralization of reactive aldehydes with phenelzine. Brain Res 1122:184-190. Wood PL, Khan MA, Moskal JR, Todd KG, Tanay VA, Baker G (2006b) Aldehyde load in ischemia-reperfusion brain injury: neuroprotection by neutralization of reactive aldehydes with phenelzine. . Brain Res 1122:184-190. Woolf CJ (2007) Central sensitization: uncovering the relation between pain and plasticity. Anesthesiology 106:864-867. Woolf CJ, Ma Q (2007) Nociceptors--noxious stimulus detectors. Neuron 55:353-364. Wu LG, Saggau P (1997) Presynaptic inhibition of elicited neurotransmitter release. Trends Neurosci 20:204-212. Wu ZZ, Li DP, Chen SR, Pan HL (2009) Aminopyridines potentiate synaptic and neuromuscular transmission by targeting the voltage-activated calcium channel beta subunit. J Biol Chem 284:36453-36461. Xiong Y, Rabchevsky AG, Hall ED (2007) Role of peroxynitrite in secondary oxidative damage after spinal cord injury. J Neurochem 100:639-649. Yan W, Byrd GD, Brown BG, Borgerding MF (2010) Development and validation of a direct LC-MS-MS method to determine the acrolein metabolite 3-HPMA in urine. J Chromatogr Sci 48:194-199. You HJ, Colpaert FC, Arendt-Nielsen L (2008) Long-lasting descending and transitory short-term spinal controls on deep spinal dorsal horn nociceptive-specific neurons in response to persistent nociception. Brain Res Bull 75:34-41. Yowtak J, Lee KY, Kim HY, Wang J, Kim HK, Chung K, Chung JM Reactive oxygen species contribute to neuropathic pain by reducing spinal GABA release. Pain 152:844-852. Yowtak J, Lee KY, Kim HY, Wang J, Kim HK, Chung K, Chung JM (2011) Reactive oxygen species contribute to neuropathic pain by reducing spinal GABA release. Pain 152:844-852. Zhao M, Isami K, Nakamura S, Shirakawa H, Nakagawa T, Kaneko S (2012) Acute cold hypersensitivity characteristically induced by oxaliplatin is caused by the enhanced responsiveness of TRPA1 in mice. Mol Pain 8:55. Zheng JH, Walters ET, Song XJ (2007) Dissociation of dorsal root ganglion neurons induces hyperexcitability that is maintained by increased responsiveness to cAMP and cGMP. J Neurophysiol 97:15-25. Zheng L, Park J, Walls M, Tully M, Jannasch A, Cooper B, Shi R Determination of urine 3-HPMA, a stable acrolein metabolite in a rat model of spinal cord injury. J Neurotrauma 30:1334-1341. 150

Zheng L, Park J, Walls M, Tully M, Jannasch A, Cooper B, Shi R (2013) Determination of urine 3-HPMA, a stable acrolein metabolite in a rat model of spinal cord injury. J Neurotrauma 30:1334-1341. Zhu Q, Sun Z, Jiang Y, Chen F, Wang M Acrolein scavengers: reactivity, mechanism and impact on health. Mol Nutr Food Res 55:1375-1390. Zhu Q, Sun Z, Jiang Y, Chen F, Wang M (2011) Acrolein scavengers: reactivity, mechanism and impact on health. Mol Nutr Food Res 55:1375-1390. Zhuo M (2007) Neuronal mechanism for neuropathic pain. Mol Pain 3:14. Zimmermann M (2001) Pathobiology of neuropathic pain. Eur J Pharmacol 429:23-37

VITA

151

VITA

Jonghyuck Park was born in Seoul, Republic of Korea to Seungkyun Park and Insook Lee, and has one younger sister, Youngok Park. He grew up in Seoul and received his Bachelor of Sciences degree in Life sciences and Biotechnology from Korea University in 2006. Then, he received his Master of Sciences degree in Department of Biomedical Engineering from Korea University Medical School in 2009. During his master program, he was active in interdisciplinary research under Dr. Yongdoo Park’s guidance. In particular, he was interested in neuroscience and tissue engineering research. After obtaining his master degree, he briefly worked for Green Cross Pharmaceutical Co., Ltd. Central Research Center as a research scientist. At Green Cross, as a one of team members of the Animal Cell Culture Lab, he developed the new drug GreenGene, which is the 3rd generation recombinant factor VIII product ever developed for the third time in the world for hemophilia. In August, 2009 he started his studies in the PhD program at Purdue University Weldon School of Biomedical with Lynn fellowship. Under Dr. Riyi Shi’s guidance, he devoted 5 years in investigating the role of acrolein in neuropathic pain after spinal cord injury

PUBLICATIONS

152

PUBLICATIONS

Jonghyuck Park, Lingxing Zheng, Glen Acosta, Sasha M. Vega Alvarez and Riyi Shi ‘Effect of TRPA1 channel at peripheral nerve terminal following SCI’ Prepare for submission

Michael Walls, Nick Race, Glen Acosta, Jonghyuck Park, Lingxing Zheng, Riyi Shi ‘Structural and biochemical abnormalities in the absence of motor deficits in mild blast- induces head trauma’ Journal of Neurosurgery (in press)

Jonghyuck Park, Breanne Muratori and Riyi Shi ‘Acrolein as a novel therapeutic target for motor and sensory deficits in spinal cord injury’ (Review), Neural Regeneration Research 9(7), 677-683, 2014

Jonghyuck Park, Lingxing Zheng, Andrew Marquis, Michael Walls, Brad Duerstock , Amber Pond, Wang He, Zheng Ouyang, and Riyi Shi ‘Neuroprotective role of hydralazine in rat spinal cord injury-attenuation of acrolein-mediated damage’, Journal of Neurochemistry 129(2), 339-349, April 2014

Michael R Due*, Jonghyuck Park*, Lingxing Zheng, Michael Walls, Fletcher White and Riyi Shi, ‘Acrolein involvement in sensory and behavioral hypersensitivity following spinal cord injury in the rat’, Journal of Neurochemistry 128(5), 776– 786, March 2014 *Equal contribution

Wang, He; Ren, Yue; McLuckey, Morgan; Manicke, Nicholas; Park, Jonghyuck; Zheng, Lingxing; Shi, Riyi; Cooks, Robert; Ouyang, Zheng, ‘Direct Quantitative Analysis of Nicotine Alkaloids from Biofluid Samples using Paper Spray Mass Spectrometry’, Analytical Chemistry. November 6, 2013, 85 (23), pp 11540–11544

Lingxing Zheng, Jonghyuck Park, Michael Walls, and Riyi Shi, ‘Determination of urine 3-HPMA, a stable acrolein metabolite in rodent models of spinal cord injury and multiple sclerosis’, Journal of Neurotrauma. August 1, 2013, 30(15): 1334-1341

Jonghyuck Park, Lingxing Zheng, Michael Walls, Bradley Duerstock, Riyi Shi, ‘Hydralazine reduces acrolein concentration accompanied by less tissue damage and enhanced behavioral recovery in a rat spinal cord injury’ Journal of Neurotrauma. August 1, 2013, 30(15): A8-A9 153

Jonghyuck Park, Unjeung Lim, Yongdoo Park, Kyung Sun, ‘Nerve regeneration following spinal cord injury using MMP-sensitive, HA-based biomimetic hydrogel scaffold containing brain-derived neurotrophic factor and human mesenchymal stem cells’, Journal of Biomedical Materials Part A, 2010 Jun 1; 93(3):1091-9