Reengineering Butyrylcholinesterase for the Catalytic Degradation of Organophosphorus Compounds DISSERTATION Presented in Partia

Home , Butyrylcholinesterase, VE (nerve agent)

Reengineering Butyrylcholinesterase for the Catalytic Degradation of Organophosphorus Compounds

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Kevin Garrett McGarry, Jr.

Ohio State University Biochemistry Program

The Ohio State University

2019

Dissertation Committee

David W. Wood, Ph.D.

Thomas J. Magliery, Ph.D.

Hannah S. Shafaat, Ph.D.

Patrice P. Hamel, Ph.D.

Christopher M. Hadad, Ph.D.

Copyrighted by

Kevin Garrett McGarry, Jr.

2019

Abstract

Chemical warfare nerve agents (CWNAs) present a global threat to both military and civilian populations. The acute toxicity of CWNAs stems from their ability to strongly inhibit cholinesterases – specifically acetylcholinesterase (AChE). This inhibition can lead to uncontrolled cholinergic cellular signaling, resulting in a cholinergic crisis and, ultimately, death. While the current FDA-approved standard of care is moderately effective when administered early, development of novel treatment strategies is necessary and was the goal that was set forth upon joining Dr. Wood’s laboratory six years ago.

Butyrylcholinesterase (BChE) is an enzyme which displays a high degree of structural homology to AChE. Unlike AChE, BChE appears to be a non-essential enzyme. In vivo, BChE primarily serves as a bioscavenger of toxic esters due to its ability to accommodate a wide variety of substrates within its active site. Like AChE,

BChE is readily inhibited by CWNAs. Due to its high affinity for binding CWNAs and that null-BChE yields no apparent health effects, exogenous BChE has been explored as a candidate therapeutic for CWNA intoxication. Despite years of research, minimal strides have been made to develop BChE (or any other enzyme) as a therapeutically relevant catalytic bioscavenger of CWNAs. BChE is, however, in early clinical trials as a ii

stoichiometric bioscavenger of CWNAs. Unfortunately, as a stoichiometric bioscavenger, large quantities of the protein must be administered to combat CWNA toxicity.

Throughout this work are described various platforms to produce recombinant butyrylcholinesterase; an enzyme comparison study across multiple, commonly-used large animal models for organophosphate (OP) research; ultimately, culminating in previously unidentified mutations of BChE that confer catalytic degradation of the

CWNA, sarin. These exciting mutations, along with corresponding future efforts, may finally lead to a novel, catalytic therapeutic to combat CWNA intoxication.

iii

Dedication

I would like to dedicate this dissertation to some very special people in my life:

First, I dedicate this work to my wife, Stephanie and our new son, Riley. Riley, you don’t understand this now, but one day, I hope that the work I’ve done here makes you proud of your daddy. Steph, thank you for supporting me in my desire to return to school, dealing with me being in the lab late in the evenings, having to come in to the lab with me before and/or after football games, or even when I have been home and I am working or thinking about science, as well as the many “Fend for Yourself (FFYN)” dinner nights as a result! Over these last few months, as well, I truly appreciate all you do as a new mother! Riley and I are so lucky to have you! Steph, I could not have done this without you being there for me and listening to all of my scientific ramblings – even if you didn’t always understand it! I love you.

Next, I would like to recognize my parents, Kathy and Kevin. Mom and Dad, you always believed in me and supported me in everything I have done, and, to this day, you continue to challenge me to do better in everything. Mom and Dad, your faith in me never wavered, and for that I am eternally grateful. Mom, thank you for always proofreading my writings -- from grade school through this dissertation. (In theory, this should be the last one!). iv

My brother, Brian, who, for the past 34+ years I have competed with in everything. The drive to challenge and better each other from the basketball court on mom and dad’s driveway, to our education, careers, and lives in general is something I continue to grow from daily. Continue challenging me, Brian, it is appreciated.

I would also like to dedicate this work to Dr. Juan Alfonzo who, in 2004, decided to take a very persistent undergraduate into his lab (if only so I would quit asking if I could join his lab). Juan, what I learned in your lab lit the research fire within me and has ultimately led to my career in science. The friendship we still have to this day is a bond that will last forever.

Finally, I would like to dedicate this dissertation to my adviser, Dr. David Wood.

I truly appreciate Dr. Wood’s openness to try something that, to our knowledge, has never been done before in the sciences at Ohio State; nor is it something that most professors would want to undertake. Without Dr. Wood’s willingness to take a chance on me and allowing me to join his lab while still maintaining full-time employment at

Battelle, this would never have been possible, and for that I am forever grateful. Dr.

Wood even established a co-appointment so that he could serve as my advisor and for this I am especially grateful. Thank you.

Acknowledgments

I would like to acknowledge my co-workers at Battelle who have supported me throughout these last six years. I would especially like to recognize Dr. Jill Harvilchuck who initially approached me about the idea of returning to graduate school to pursue a doctorate in biochemistry. I would also like to recognize Tom Snider, who recently passed away but was always willing to lend an ear and to provide valuable insight into anything I was working on or just wanted to discuss. Dr. Bob Moyer, thank you for your guidance during your time at Battelle and willingness to listen to the often unsolicited

OSU research news I inundated you with over the years. To all of the members of the biochemistry team (past and present) at Battelle who have helped me out along the way to get where I am today, thank you.

Lastly, I would like to acknowledge all of my committee members (both past and present), your input and guidance over the last several years has led to some very interesting ideas, discussions, and scientific breakthroughs. I have learned a great deal from you, and I hope that the relationships we have developed will continue as scientific collaborations over the course of our careers. Thank you all.

Vita

2000 – 2004...... Bachelor of Science in Microbiology

The Ohio State University, Columbus, OH

2005 – 2007...... Master of Science in Microbiology

The Ohio State University, Columbus, OH

2007 – Present ...... Researcher in Health, Highly Toxic Materials

Battelle Biomedical Research Center,

West Jefferson, OH

2013 – Present ...... Graduate Student, Candidate, Doctor of Philosophy

Ohio State Biochemistry Program

The Ohio State University, Columbus, OH

vii

Publications

Moyer, R. A., McGarry, K. G., Jr., Babin, M. C., Platoff, G. E., Jr., Jett, D. A., and

Yeung, D. T. (2018) Kinetic analysis of oxime-assisted reactivation of human, Guinea pig, and rat acetylcholinesterase inhibited by the organophosphorus pesticide metabolite phorate oxon (PHO), Pestic Biochem Physiol 145, 93-99.

Snider, T. H., McGarry, K. G., Jr., Babin, M. C., Jett, D. A., Platoff, G. E., Jr., and

Yeung, D. T. (2016) Acute toxicity of phorate oxon by oral gavage in the Sprague-

Dawley rat, Fundam Toxicol Sci 3, 195-204.

Brittain, M. K., McGarry, K. G., Moyer, R. A., Babin, M. C., Jett, D. A., Platoff, G. E.,

Jr., and Yeung, D. T. (2016) Efficacy of Recommended Prehospital Human Equivalent

Doses of Atropine and Pralidoxime Against the Toxic Effects of Carbamate Poisoning in the Hartley Guinea Pig, International journal of toxicology 35, 344-357.

McGarry, K. G., Bartlett, R. A., Machesky, N. J., Snider, T. H., Moyer, R. A., Yeung, D.

T., and Brittain, M. K. (2013) Evaluation of HemogloBind treatment for preparation of samples for cholinesterase analysis, Adv Biosci Biotechnol 4, 1020-1023.

viii

McGarry, K. G., Walker, S. E., Wang, H., and Fredrick, K. (2005) Destabilization of the

P site codon-anticodon helix results from movement of tRNA into the P/E hybrid state within the ribosome, Molecular cell 20, 613-622.

Fields of Study

Major Field: Ohio State Biochemistry Program

Table of Contents

Abstract ...... ii

Dedication ...... iv

Acknowledgments...... vi

Vita ...... vii

Publications ...... viii

List of Tables ...... xv

List of Figures ...... xvi

Chapter 1. Introduction ...... 1

1.1 Organophosphates ...... 1

1.2 Cholinergic Neurotransmission ...... 2

1.3 Inhibition of AChE ...... 4

1.4 Current Treatment Strategies for OP Intoxication ...... 4

1.5 Butyrylcholinesterase ...... 5

1.6 BChE as a Therapeutic Bioscavenger ...... 6

1.7 Additional Bioscavengers of OPs ...... 7

Chapter 2. Platforms for the Synthesis of Recombinant BChE ...... 9

2.1 Introduction ...... 9

2.2 Production of BChE using in vitro transcription and translation ...... 10

2.3 Production of BChE in Escherechia coli ...... 13

2.4 Production of BChE using mammalian cell culture ...... 15

2.4.1 BChE Expression Vectors...... 15

2.4.2 Mutagenesis ...... 15

2.4.3 Expression of BChE in Mammalian Cell Culture ...... 18

2.5 Plant and Algae-Based Methods for the Production of BChE ...... 19

2.5.1 Introduction ...... 19

2.5.2 BChE Production in Tobacco ...... 20

2.5.3 BChE Production in Chlamydomonas reinhardtii ...... 21

Chapter 3. Characterization of Cholinesterases from Multiple Large Animal Species for

Medical Countermeasure Development against Chemical Warfare Nerve Agents ...... 25

Preface...... 25

3.1 Introduction ...... 26

3.2 Materials and Methods ...... 29

3.2.1 Chemicals...... 29

3.2.2 Enzyme preparation ...... 30

3.2.3 Enzyme activity determination ...... 30

3.2.4 AChE and BChE inhibition assays ...... 31

3.2.5 Oxime reactivation of AChE and BChE ...... 31

3.2.6 Aging...... 32

3.3 Results ...... 32

3.3.1 Cholinesterase Levels in Circulation ...... 32

3.3.2 Enzyme Inhibition ...... 34

3.3.3 Aging...... 36

3.3.4 Reactivation – EC50 Determination ...... 39

3.4 Discussion ...... 42

3.5 Funding ...... 49

3.6 Acknowledgements ...... 49

Chapter 4. A Human-Porcine BChE Hybrid Enzyme ...... 51

4.1 Introduction ...... 51

4.2 Materials and Methods ...... 55

4.2.1 Chemicals ...... 55

4.2.2 BLAST® sequence alignments ...... 56

xii

4.2.3 Plasmid construction ...... 56

4.2.4 Enzyme production and purification ...... 57

4.2.5 Protein quantification ...... 58

4.2.6 Western blotting ...... 58

4.2.7 Silver staining ...... 59

4.2.8 Enzyme activity assays ...... 60

4.2.9 Enzyme inhibition ...... 61

4.2.10 Inhibition rate constant determination ...... 61

4.2.11 Aging...... 62

4.2.12 GC-HRMS ...... 62

4.2.13 LC-MS/MS ...... 65

4.2.14 Density functional theory calculations...... 68

4.2.15 Protein preparation for molecular dynamics simulations ...... 68

4.2.16 Molecular dynamics (MD) simulations ...... 70

4.2.17 Clustering Analysis ...... 71

4.3 Results ...... 71

4.3.1 Sequence Alignment ...... 71

4.3.2 Purification of Recombinant BChEs ...... 72

xiii

4.3.3 Inhibition of Recombinant BChEs by Sarin ...... 77

4.3.4 Aging and Spontaneous Reactivation of Recombinant BChEs ...... 81

4.3.5 Molecular dynamics (MD) simulations of WT and hybrid BChE before and

after inhibition with sarin ...... 86

4.4 Discussion ...... 93

4.5 Chapter Acknowledgements ...... 96

Chapter 5. Ongoing and Future BChE Efforts ...... 97

5.1 BChE Metalloenzymes ...... 97

5.2 An Intein:BChE Fusion Protein ...... 103

Bibliography ...... 109

Appendix A. Plasmid Maps ...... 121

Appendix B: Supplemental Information for Human-Porcine Hybrid BChE ...... 138

B-1: Supplemental Information for Molecular Dynamics Simulations ...... 144

xiv

List of Tables

Table 1. Nerve Agent Inhibition Results ...... 35

Table 2. Aging Half-Times ...... 37

Table 3. EC50 (µM) Determination of 2-PAM for Multiple Species versus

VX or Sarin ...... 40

Table 4.GC-HRMS Parameters ...... 64

Table 5. LC-MS/MS Parameters...... 67

Table 6. ki calculations for the inhibition of WT and recombinant BChEs by sarin ...... 80

Table 7. Reactivation potential percentages for different species ...... 92

Table 8. Wild-Type BChE and a BChE Metalloenzyme vs.

Paraxon in the presence of Zn2+ ...... 102

Table 9. Measured Concentrations of Sarin and IMPA using LC-MS/MS ...... 142

Table 10. Measured Concentrations of sarin over time using GC-HRMS ...... 143

List of Figures

Figure 1. Cholinergic Neurotransmission ...... 3

Figure 2. In vitro transcription and translation ...... 11

Figure 3. IVT Expression of BChE...... 12

Figure 4. Ellman’s Activity Assay ...... 13

Figure 5. Silver stains of the expression and purification of BChE expressed in E. coli ...... 15

Figure 6. Individual circulating cholinesterase levels...... 33

Figure 7. Aging and/or spontaneous reactivation of sarin-inhibited cholinesterases...... 38

Figure 8. Reactivation of VX-inhibited AChE and BChE by 2-PAM...... 41

Figure 9. Sequence alignment comparison of human BChE vs. porcine (Sus scrofa) BChE...... 72

Figure 10. Anti-Histidine Western blot following the expression and purification of indicated BChE constructs ...... 74

Figure 11. Anti-BChE Western Blot following the purification of recombinant BChE. . 75

Figure 12. Silver stain following the expression and purification of human WT BChE

(left) and human porcinated BChE (right)...... 76

xvi

Figure 13. Substrate titration comparison of WT BChE vs. human-porcine hybrid BChE...... 77

Figure 14. Inhibition of multiple recombinant BChEs by sarin...... 79

Figure 15. Catalytic degradation of sarin by human-porcine hybrid BChE...... 84

Figure 16. Extracted Ion Chromatograms for Sarin and IMPA...... 85

Figure 17. Proposed pathways for spontaneous hydrolysis by conserved water molecules with sarin-inhibited BChE ...... 92

Figure 18. Various OPH and BChE model protein structures...... 100

Figure 19. Process workflow for a split-intein affinity capture and purification...... 106

Figure 20. Silver stain of HEK293E expression of WT-BChE and SA-mBChE-His6. .. 108

Figure 21. pT7CFE1-BChE (WT)-His6 ...... 122

Figure 22. pET- BChE-His6 ...... 123

Figure 23. pTT- BChE-His6 ...... 124

Figure 24. pTT-BChE-His6 (without restriction enzymes) ...... 125

HN Figure 25. pTT-IK-NpuC -mBChE-His6 ...... 126

Figure 26. pRcCMV-BChE-His6 ...... 127

Figure 27. pRcCMV-BChE-His6 (without restriction enzymes) ...... 128

Figure 28. CGT8288-BChE (WT - Tobacco)-His6 ...... 129

Figure 29. AKK1448-BChE (WT BChE - Tobacco)-His6 ...... 130

Figure 30. AKK1472-BChE (WT BChE - Tobacco)-His6 ...... 131

Figure 31. pOpt_CCA_gLuc_Hyg[7646] ...... 132

xvii

Figure 32. pOpt_CCA_gLuc_Paro[7647] ...... 133

Figure 33. pOpt_CCA_BChE (without Intron)_Hyg...... 134

Figure 34. pOpt_CCA_BChE (without Intron)_Hyg (with restriction enzymes)...... 135

Figure 35. pOpt_CCA_BChE (with Intron)_Hyg ...... 136

Figure 36. pOpt_CCA_BChE (with Intron)_Hyg (with restriction enzymes) ...... 137

Figure 37. Sequence alignment comparison of human BChE vs.

BChE from various species...... 138

Figure 38. 48-Hr Aging Experiment Results ...... 141

Figure 39. Cluster analysis by RMSD for the native forms of BChE...... 145

Figure 40. RMSF analysis for the native forms of BChE ...... 146

Figure 41. Active site distances native forms of BChE ...... 148

Figure 42. Distances of the acyl loop and the site of the mutation for the native forms of BChE ...... 150

Figure 43. Distances and angle of Trp82 in the -loop of the native form ...... 152

Figure 44. Illustration of “acyl-pocket encroachment” in the HID form of the native enzyme...... 153

Figure 45. Illustration of active site “climbing” in the HIE form of the native enzyme ...... 154

Figure 46. Cluster analysis by RMSD for the 2XQJ inhibited forms of BChE ...... 155

Figure 47. Cluster analysis by RMSD for the 2XQK inhibited forms of BChE ...... 156

Figure 48. RMSF analysis for the 2XQJ inhibited forms of BChE ...... 157

xviii

Figure 49. RMSF analysis for the 2XQK inhibited forms of BChE ...... 158

Figure 50. Active site distances for the 2XQJ inhibited forms of BChE ...... 160

Figure 51. Active site distances for the 2XQK inhibited forms of BChE ...... 161

Figure 52. Distances of the acyl loop and the site of the mutation for the

2XQJ inhibited forms of BChE ...... 163

Figure 53. Distances of the acyl loop and the site of the mutation for the

2XQK inhibited forms of BChE ...... 164

Figure 54. Distances and angle of Trp82 in the -loop of the 2XQJ inhibited form ..... 166

Figure 55. Distances and angle of Trp82 in the -loop of the 2XQK inhibited form .... 167

Figure 56. Water interaction distances for the 2XQJ inhibited forms of BChE ...... 169

Figure 57. Water interaction distances for the 2XQK inhibited forms of BChE ...... 170

Figure 58. Example of an “active” water for Pathway 2 ...... 171

Figure 59. Example of an “active” water for Pathways 1 and 2 ...... 172

Figure 60. Illustration of the water pocket formed in between Glu197 and phosphylated Ser198 ...... 173

xix

Chapter 1. Introduction

1.1 Organophosphates

From accidental exposure and poisoning of agricultural workers, to acts of terrorism, to the battlefield, organophosphates (OPs) continue to pose a threat to both military and civilian populations worldwide. Chemical warfare nerve agents (CWNAs) such as sarin and VX were first introduced during World War II and the years that followed. CWNAs are some of the most potent neurotoxicants known to mankind and they belong to this class of chemicals known as OPs. Many commonly used pesticides also belong to this class of chemicals; however, they are generally much less potent than

CWNAs. It is estimated that as many as 200,000 people in the developing world die annually from acute OP poisoning although most of these deaths are deliberate suicides 1,

2. In Japan, however, two sarin attacks by the Aum Shinrikyo religious cult in 1994 and

1995 resulted in the injury of approximately 1,000 people, with an estimated 5,500 people potentially exposed 3, 4. Furthermore, CWNAs have been recently used by the

Syrian regime on its own people during an ongoing civil war 5 as well as in the assassination of Kim Jong-Nam, Kim Jong-Un’s half-brother, in the Kuala Lumpur

airport 6. These reports suggest a continuing need for safe and effective treatments for

OP exposures.

1.2 Cholinergic Neurotransmission

It is widely understood that the primary mechanism of OP toxicity is due to inhibition of acetylcholinesterase (AChE) via phosphylation of the active serine site 7-10.

It is this phosphylation event that can have devastating consequences to a poisoned individual. In cellular signaling involving cholinergic neurons, the neurotransmitter acetylcholine (ACh) is packaged and released into the synaptic cleft where it binds to and activates ACh-receptors, continuing a downstream signaling event. To attenuate the signaling event, AChE hydrolyes excess ACh, thereby terminating neuronal signaling

(Figure 1. Cholinergic Neurotransmission). As such, AChE is an essential enzyme involved in the regulation of neuronal cellular signaling.

Figure 1. Cholinergic Neurotransmission (Reproduced from Katzung 2001 11)

1.3 Inhibition of AChE

Inhibition of AChE results in the accumulation of ACh, leading to continual stimulation of ACh receptors and may ultimately lead to cholinergic crisis. Symptoms depend on toxic compound dose, route of exposure, and the type of receptor that is being stimulated. During cholinergic crisis, the continuous stimulation of cholinergic neurons and/or skeletal muscles may result in seizures, paralysis, and, ultimately, death typically results from one or more respiratory factors, including paralysis of the respiratory muscles, increased bronchial secretions, and depression of the brain’s respiratory center 1.

1.4 Current Treatment Strategies for OP Intoxication

The current FDA-approved/fielded standard of care for OP poisoning includes the use of three drugs: 1) A reversible, muscarinic acetylcholine receptor antagonist slows down signal transmission by blocking acetylcholine receptors (e.g., atropine); 2) An

AChE reactivator (e.g., pralidoxome) is used to reactivate inhibited AChE through the removal of the bound OP compound prior to aging and irreversible (until recently) AChE inactivation; 3) A benzodiazepine (e.g., diazepam) is used to control seizures 1.

Unfortunately, these aforementioned treatment strategies for the treatment of acute OP poisoning are insufficient and do not lend themselves to prophylactic applications as this approach is not always effective, and atropine and diazepam, in particular, can produce significant unwanted side effects. Further, the relatively short

serum half-lives of these compounds (on the order of hours) makes them impractical for applications where extended protection against potential exposures is desired (e.g., battlefield or agricultural field applications). The route of exposure, dosage, and characteristics of the OP compounds themselves all play a role in the effectiveness of treatment strategies 12. Therefore, as discussed above, the continued threat of CWNA exposure, accidental poisoning of agricultural workers, and the deliberate suicide attempts demand the development of more effective countermeasures for both acute clinical cases and extended periods of risk for OP exposure.

1.5 Butyrylcholinesterase

Butyrylcholinesterase (BChE), also known as pseudocholinesterase or plasma cholinesterase, is an enzyme that shares approximately 50% amino acid identity to AChE within a species and a high degree of structural homology between species 13-15. BChE is believed to have arisen early in the evolution of vertebrates due to a gene duplication event and the structural similarities of the two enzymes are striking 16. BChE is synthesized in the liver as a heavily glycosylated homotetramer, and, in humans, is present in relatively large amounts in the plasma (3-9mg/L). Each subunit includes nine

N-glycosylation sites, and glycosylation constitutes approximately 25% of its 340 kDa mass 17-19. Native BChE can be purified from human plasma or Cohn Fraction IV20 and has also been produced through recombinant expression in various expression systems including, but not limited to, mammalian cell culture (CHO and HEK293 cells 17, 21-24.) 5

and insect cells25. Interestingly, the exact function of BChE remains unknown. For years, the primary focus on BChE involved its function in the degradation of the anesthetic, succinylcholine. As some individuals are incapable of metabolizing this drug during surgery, life-threatening apnea can occur. It was later determined that BChE was the primary enzyme in the hydrolysis of succinylcholine and that patients who were sensitive to the drug possessed variants of the enzyme with reduced activity towards its degradation26-29. Additionally, BChE is thought to play a role in a number of biological processes including growth and development, neuronal cellular development, neuronal signaling, synaptic transmission 26, 30, and it appears to have a role in bronchial airway smooth muscle function 31, 32. Furthermore, BChE acts as a scavenger and detoxifier of toxic esters (including OP cholinergic toxicants).

1.6 BChE as a Therapeutic Bioscavenger

BChE has been shown to provide prophylactic protection against OP exposure 22,

33-38. Importantly, for use as a potential therapeutic, administered BChE also exhibits a serum half-life of 10 to 14 days. However, while BChE provides hope as a potential therapeutic bioscavenger for OPs, it does have limits. Unfortunately, native BChE only serves as a stoichiometric scavenger, whereby both the BChE and OP molecules are inactivated following binding. Thus, large amounts of BChE must be administered, on the order of 24.2 mg/kg, to provide protection from the toxic effects of VX and other

OPs22. The standard method for the purification of human BChE requires large volumes 6

of human plasma where BChE is affinity-purified on procainamide resins. Based on typical BChE concentrations in human blood, up to 800 liters of human donor blood must be processed to treat a single adult male for VX exposure.

As can be imagined, a catalytic version of the BChE enzyme with the ability to enzymatically degrade OP compounds would provide a major advance in this field.

Although this goal is ambitious, there is some evidence that it will be possible. For example, a specific mutation, G117H, has been shown to confer measurable organophosphorus hydrolase (OPH) activity to the BChE enzyme although its catalytic efficiency is not sufficient for use as a therapy for OP exposure39. Based on molecular dynamics studies, it has been postulated that the OPH activity of the G117H mutant stems from the coordination of a water molecule in the oxyanion hole40. These studies, when combined with additional research on the ability of oxime compounds to reactivate

BChE following OP binding, suggest that this enzyme may be tractable for conversion from its native BChE activity to a more therapeutically relevant OPH activity.

1.7 Additional Bioscavengers of OPs

There have also been examples where protein engineering of the naturally occurring bacterial (Psuedomonas) enzyme organophosphorus hydrolase (OPH) has been successful in enhancing catalytic degradation of OPs41, 42. Highly active OPH enzymes can detoxify paraoxon at diffusion-controlled rates and have a broad substrate scope across many neurotoxic compounds. The active site contains a dinuclear zinc center that 7

can be replaced by redox-inactive Co2+, Cd2+, Mn2+, or Ni2+ ions. The proposed mechanism of hydrolysis involves initiation via a nucleophilic bridging water molecule, though modifications of the metal-binding site indicate activity can be retained with only one metal center in the protein43. Furthermore, recent work redesigning the mononuclear mouse adenosine deaminase confirmed that OPH activity is not strictly dependent on a binuclear active site44. Despite their high activity, the bacterial origin of OPH enzymes renders them poor candidates for therapeutics due to potential immunogenicity and low expected bioavailability. Conversely, BChE has high bioavailability, but no metal binding capability within the active site, and thus, while the wild-type (WT) enzyme stoichiometrically binds OPs, it exhibits no catalytic OPH activity.

Chapter 2. Platforms for the Synthesis of Recombinant BChE

2.1 Introduction

As discussed in Chapter 1, OP compounds bind and inhibit the acetylcholinesterase (AChE) enzyme, which can lead to cholinergic crisis and eventual death in the case of severe acute exposures. Low level chronic exposures have been associated with additional health effects, although direct links remain unclear45. BChE, also known as pseudocholinesterase or plasma cholinesterase, is an enzyme that shares approximately 50% amino acid identity and a high degree of structural homology to

AChE31. Importantly, BChE acts as a scavenger and detoxifier of toxic esters (including

OP cholinergic toxicants) and has been shown to provide prophylactic protection against

OP exposure46. BChE is normally synthesized in the human liver as a heavily glycosylated homotetramer, with nine N-linked glycosylation sites on each monomer17-19.

The structural complexity of BChE makes it a difficult target for recombinant production, which has hampered BChE research as well as the manufacture of BChE as an effective therapeutic. Native BChE is currently purified from expired human donor blood, but low-level recombinant expression in mammalian cells as well as transgenic tobacco has been used. Importantly, BChE only serves as a stoichiometric scavenger for

OPs in the bloodstream, whereby both the BChE and OP molecules are “inactivated” by

OP binding. Thus, large amounts of BChE must be administered, requiring up to 800 liters of human donor blood to be processed to prophylactically treat a single adult for a potential VX exposure. These activity and production issues must be addressed before

BChE can be used therapeutically and that was the overarching goal of the work described herein.

2.2 Production of BChE using in vitro transcription and translation

As a result of a DARPA funded project that was ongoing when I first joined the Dr. Wood’s laboratory, we sought to express BChE using the Thermo Scientific™ 1-Step Human

Coupled In Vitro Protein Expression Kits. The manufacturer has claimed that these kits are able to express full-length, fully glycosylated, functional proteins in a short amount of time

(90 minutes for 100 µg/mL) using the unique HeLa or CHO cell lysate-based protein expression systems when combined with the optimized pT7CFE1 Expression Vector

(Figure 2).

Figure 2. In vitro transcription and translation

As such, for expression in the in vitro transcription and translation system, BChE cDNA was purchased from MyBiosource and cloned into the pT7CFE1 expression vector using

PacI and XhoI restriction sites (Appendix A-1). Expression using the in vitro transcription and translation kits was performed per manufacturer’s instructions. BChE was purified using Ni-NTA immobilized metal affinity chromatography (IMAC). While production of non-glycosylated BChE appears to have been successful – as evidenced by the band present at 68 kDa, (Figure 3). Unfortunately, a fully glycosylated enzyme has a molecular weight of ~85 kDa, and this was obviously absent from this expression system.

Despite the fact that a fully glycosylated protein was not produced, enzymatic activity was assessed. To perform this assay, briefly, expression samples were analyzed as described using the adapted Ellman’s activity assay47, 48 and summarized in Figure 4 below. Briefly, using a 96-well optical bottom plate test sample was diluted in a 1X PBS.

Ellman’s Reagent [DTNB; 5,5′-dithiobis (2-nitrobenzoic acid)] was added to a final concentration of 500 µM. Butyrylthiocholine iodide was used as the reaction substrate

and added to a final concentration of 3 mM immediately prior to analysis. Upon addition of the substrate, the plate was analyzed using a Synergy™ HTX (BioTek) microplate reader and kinetic data were obtained at 412 nm every 15-20 seconds for a period of 10 minutes. Using Beer’s Law (A=εLC), where ε is the extinction coefficient of DTNB

(13,600M-1 cm-1) and L is 1 cm path length, the enzyme activity in µmoles min-1 mL-1

(U/mL) for each sample was calculated. Unfortunately, the protein was also inactive, and due to this observation, in vitro transcription and translation was not pursued further.

Figure 3. IVT Expression of BChE.Panel A displays an anti-histidine Western Blot of two elutions (E1, E2) of two independent clones. It is unclear what caused the smearing in Lane 3 (E1, clone 2). Panel B displays a silver stain of the purification of an IVT- expressed BChE. Load order indicated is expression reaction, clarified expression reaction, flow-through, wash 1, wash 2, elution 1, and elution 2. The arrow indicates the presence of the expected 68 kDa, non-glycosylated BChE elution number 2.

Figure 4. Ellman’s Activity Assay Reproduced from McGarry et al.48

2.3 Production of BChE in Escherechia coli

In order to successfully clone human BChE into the desired vector for

Escherechia coli expression, a silent mutation within the BChE signal peptide was first introduced to remove the NdeI site that was present. Human BChE was subsequently cloned into a pET expression vector using NdeI and XhoI restriction sites (Appendix A-

2).

It is understood that expression of a properly glycosylated mammalian protein in

E. coli is not possible. It should be noted that full glycosylation does not appear to be required for BChE activity17. However, as mature BChE also possesses multiple disulfide bonds, in order to attempt to obtain a properly folded protein, two different strains of E. coli (New England Biolabs (NEB); SHuffle® T7 Competent K12 and B)

which constitutively expresses a chromosomal copy of the disufide bond isomerase DsbC were used to express BChE. DsbC promotes the correction of mis-oxidized proteins into their correct form49, 50. The cytoplasmic DsbC is also a chaperone that can assist in the folding of proteins that do not require disulfide bonds51.

BChE was expressed in both SHuffle® T7 Competent E. coli K12 or SHuffle®

T7 Competent E. coli Express B strains. Briefly, 200 mL of cells were cultured in Luria-

Bertani (LB) broth and grown to an OD600 of 0.814 (E. coli K12) and 0.560 (E. coli B) respectively. Cells were then induced with 0.4 mM isopropyl β-D-1- thiogalactopyranoside (IPTG) and grown for 21 hours at 16°C. Cells were lysed by sonication, the lysate was clarified, and the BChE was purified using IMAC. As observed in Figure 5, non-glycosylated BChE was expressed in both SHuffle® K12 and

Express B strains of E. coli with K12 yielding slightly greater amounts of the purified protein. Unfortunately, neither strain of E. coli was capable of yielding active BChE and as such, this mode of expression was not pursued any further.

Figure 5. Silver stains of the expression and purification of BChE expressed in E. coli SHuffle® T7 Competent E. coli K12 (left) or SHuffle® T7 Competent E. coli Express B strains (right).

2.4 Production of BChE using mammalian cell culture

2.4.1 BChE Expression Vectors

Human BChE was cloned into the mammalian pTT expression vector52 using the

HindIII and AfeI restriction sites (Appendix A-3). An additional expression vector, pRcCMV-BChE was obtained through the generous gift of Dr. Oksana Lockridge

(University of Nebraska). A 6x-histidine tag was added to the mBChE protein for ease of purification. These two vectors were used for the expressions described herein.

2.4.2 Mutagenesis

All mutations were created using a method similar to Agilent Technologies’

QuikChange site-directed mutagenesis kit. The method designed and used for the 15

creation of these mutations is described below for reference by future students and technical staff.

Mutagenesis Reaction (50 µL)

10 µL 5X Q5® Reaction Buffer

10 µL High GC Enhancer

1 µL Template DNA (200 ng/µL – isolated from a dam+ strain of E. coli)

1 µL Forward mutagenesis primer (Designed using Agilent’s primer design tool; 125 ng/µL)

1 µL Reverse mutagenesis primer (Designed using Agilent’s primer design tool; 125 ng/µL)

1 µL 10 mM Deoxynucleotide solution (dNTPs)

1 µL Q5® High Fidelity DNA Polymerase (New England Biolabs)

25 µL dH2O

A Polymerase chain reaction (PCR) was subsequently performed using the above reaction mix using the following cycling parameters:

1. 95°C, 2 minutes

2. 95°C, 20 seconds

3. 60°C, 15 seconds

4. 72°C, 4 minutes (30 seconds/kb of template DNA)

5. Repeat steps 2-4, 20 times.

6. 72°C, 5 minutes to complete elongation.

7. Hold at 2-8°C or room temperature

Once the PCR had completed, 5 µL of the reaction was run on a qualitative agarose gel (typically 0.8% for the 8 kB plasmids used for these mutagenesis reactions was sufficient) to ensure that the PCR was successful. To determine if a

PCR was successful, a mutagenesis reaction will yield a much more intense band as compared to the template DNA control.

If the PCR was successful, 10 µL of each mutagenesis reaction was digested with

1 µL of DpnI overnight at 37°C to remove all methylated parental template DNA, thereby selecting for only the mutated DNA. This step is not to be shortened as complete removal of the parent DNA is essential for the success of this procedure.

Once DpnI digestion is completed, 5-10 µL of the digestion reaction was transformed into competent E. coli cells (in almost all cases, Z-competent DH5α was used) and plated onto appropriate antibiotic-containing agar plates and grown overnight at

37°C. The following day, colonies were cultured in 5 mL LB broth containing antibiotic, DNA was isolated by mini-prep, and mutagenization of the DNA was confirmed by commercial sequencing.

2.4.3 Expression of BChE in Mammalian Cell Culture

Expression of multiple recombinant proteins using mammalian cell culture and the specific method to do so is discussed in greater detail in Chapter 4; however, several results are worth mentioning here in regards to the cell lines used throughout this project.

Human embryonic kidney cells, ebna strain (HEK293E), Chinese hamster ovary (CHO), and human hepatocellular carcinoma cells (HuH-7) were all explored for their potential utility in the production of recombinant BChE. Both HEK293E and CHO cells are widely used for the expression of various recombinant proteins, including BChE. HuH-7, however, was explored because BChE is naturally produced in the liver53, and, as such, it was hypothesized that a hepatocyte line may produce greater protein yields. While all three cell lines were capable of successfully producing active BChE; HEK293E and CHO were far superior to HuH-7 (HEK293E + pTT expression vector > CHO + pRcCMV expression vector > HuH7 + pTT or pRcCMV expression vectors). HEK293E yielded

64.5 – 80.3 µg/mL (~1.5 mL total of purified protein) from 40 mL expressions. This equates to 108.6 µg per 40 mL (2.7 mg/L). Therefore, this transient expression system produces greater yields than the 1 mg/L in CHO cells that has been reported but not the 4 mg/L reported for insect cells25. As such, all constructs for the bulk of the work described herein were produced (or reproduced) in the pTT expression vector for expression using the HEK293E cell line.

Notes and lessons learned regarding HuH-7 for future lab members, however, should be discussed. First, HuH-7 was difficult to culture and required working under 18

biological safety level 2 (BSL-2) handling conditions due to the presence of the HCV replicon. After collaborating with Anthony Stephenson (Zucai Suo laboratory, formerly of The Ohio State University) it was determined that HuH-7 requires specialized growth media supplemented with non-essential amino acids. Additionally, transfection using the common transfection reagents Lipofectamine™ 2000 and Lipofectamine™ 3000 proved difficult. Additionally, the pTT vector was not suitable for transfection into HuH-7.

Transfection of the pRcCMV-BChE-His6 into HuH-7 resulted in BChE yields and activity that appeared to be ~30% of HEK293E expression. Because of these difficulties and the undue requirement of working under BSL-2 conditions, culture and protein expression involving HuH-7 was suspended indefinitely.

2.5 Plant and Algae-Based Methods for the Production of BChE

2.5.1 Introduction

While mammalian expression was successful in producing recombinant BChE, our lab desired to pursue an even more efficient production system for our engineered and wild-type BChEs. To do this, our lab submitted a SEED grant to the Center for

Applied Plant Sciences (CAPS). The goal of our proposal was such that might produce large amounts of BChE at a low cost using plant cells for heterologous expression with a self-cleaving tag method for highly efficient purification. In developing the interdisciplinary team to accomplish this work, we identified a number of candidate plant

and algae cell systems to diversify our ability to produce BChE variants. Further, our discussions with faculty at Ohio State Wooster suggested that transgenic plants expressing a catalytic BChE (or even organophosphohydrolase; OPH) may have applications in environmental decontamination, as well as impacts on food safety.

Thus, our rationale was to combine expertise in protein engineering and protein purification with expertise in recombinant protein expression in plants to accomplish the goal of engineering and producing sufficient quantities of a fully catalytic BChE. Based on previously reported successes (outside of OSU) in expressing BChE in transgenic tobacco54, we sought to leverage the expertise of the CAPS in the area of plant-based protein expression to accelerate this work. Additionally, as mentioned above, the ability to produce a catalytic BChE in plants might have significant impacts for food safety and environmental remediation, thus supporting the goals of CAPS as well.

2.5.2 BChE Production in Tobacco

The BChE gene was successfully codon optimized and cloned into the CGT8288 shuttle vector (Appendix A-8) by Genescript using the NcoI and SacI restriction sites.

The BChE gene was subsequently subcloned into the AscI sites in the AKK-1448

(Appendix A-9) and AKK-1472 (Appendix A-10) binary vectors. Leaf infiltration experiments and transgenic plants were created by Dr. Christopher Taylor (Ohio State

University - Wooster) using Agrobacterium transformation as described55, 56.

Successful expression was confirmed by observing the presence of green fluorescent protein (GFP) under the microscope of the plants transformed with AKK-

1448. Transgenic tobacco leaves were subsequently homogenized to 0.2 g/mL using a

Polytron® homogenizer in ice cold 1X PBS, pH 7.4. Debris was pelleted at 20,000 x g for 2 minutes. The supernatant was passed through a 0.2 µm filter to remove any residual debris and to ensure sterility. Activity was assessed using the Ellman’s based assay as described previously. Active BChE was indeed obtained. However, as the CAPS project has since closed, a full characterization was not performed and all remaining leaves are stored at -70°C in the Wood laboratory. It should also be noted that Dr. Taylor indicated that stable plant expression of BChE was achieved and seeds were obtained and are stored in the Taylor lab. This expression system, was successful and should be considered in the future as a potential source for the production of BChE.

2.5.3 BChE Production in Chlamydomonas reinhardtii

It must be noted that the following sections were adapted from an experimental write-up that was provided and optimized by Andrew Castonguay (Ph.D. candidate in the laboratory of Dr. Patrice Hamel). The accompanying experiments were performed with extensive assistance of Andrew Castonguay as well as part of the collaboration established under the CAPS SEED grant.

2.5.3.1 Strains and Growth conditions

Transformation of the Chlamydomonas reinhardtii uvm4 strain was performed using the glass bead method described by Kindle et. al57 with several noteworthy modifications. Briefly, uvm4 culture for transformation was grown in liquid tris acetate phosphate (TAP) media supplemented with arginine (Arg, 0.4 mg/ml) and sorbitol (Sorb,

40 mM) under constant light (30 µmol photons), with shaking (180 rpm, 25°C) for 3-5 days until reaching mid-log phase (1-5x106 cells/ml). Solid TAP agar plates (1.5% w/v) were used to maintain the uvm4 strain and were refreshed monthly for selection and propagation of potential transformants. Cells were collected by centrifugation at 1,500 x g for 5 minutes at room temperature and the pellet was resuspended to a final cell density of 1x108 cells/ml in TAP+Arg+Sorb. The uvm4 strain is already a cell wall deficient strain (confirmed visually at 100x magnification by treatment with 0.1% triton x 100) and therefore, no autolysin treatment was necessary to remove the cell wall prior to transformation.

2.5.3.2 Transformation of Chlamydomonas

The BChE gene for the transformation of Chlamydomonas was “designed” by Dr.

Patrice Hamel whereby not only was the gene codon-optimized for Chlamydomonas expression, but a consenus intron was inserted into one of the constructs in hopes of increasing the likelihood of success integration and protein expression. Each of the two

BChE genes (with and without the intron) were synthesized by Genewiz and delivered in a pUC57 vector. Two produce the four candidate constructs, the BChE genes were

subcloned into the EcoRI and NsiI restriction sites of the expression vectors pOpt_CCA_gLuc_Hyg (Appendix A-11) or pOpt_CCA_gLuc_Paro (Appendix A-12) to produce pOpt_CCA_BChE_Hyg (Appendices A-13 – A-16) or pOpt_CCA_BChE_Paro..

In order to transform Chlamydomonas, the transformation reaction mixture was prepared as follows using sterile 1.5 mL microcentrifuge tubes, in order, the following was added: (1) sterile 300 µg glass beads, (2) 100 µl 20% polyethylene glycol (PEG

8000), (3) 300 µL cells at 1x108 cells/mL, (4) 2.5 µL filter-sterilized sheared herring sperm DNA (10 mg/mL), and (5) transforming, plasmid DNA (1-5 µg, linearized with

ScaI). Transformation reactions were then vortexed at max speed for 30 seconds and 300

µL was transferred immediately to 125 mL Erlenmeyer flasks containing 30 mL

TAP+Arg+Sorb to recover overnight (16-24 hrs with shaking at 180 rpm, 25°C, with constant illumination at 30 µmol photons) prior to transfer to selective media. After the recovery incubation, transformant cultures were pelleted by centrifugation at 1,500 x g for 5 minutes at room temperature and resuspended in 1 mL TAP+Arg+Sorb. 500µL cell suspension was then aseptically plated using glass spreader onto solid media containing the appropriate selective antibiotic. Minimum sensitivity of the uvm4 recipient strain to each antibiotic (paromomycin sulfate, Pm and hygromycin B, HygB) was previously determined by ten-fold dilution series on solid media containing increasing concentrations (2.5-25 µg/mL Pm and 5-25 µg/mL HygB) of each respective antibiotic. uvm4 was sensitive at the lowest concentrations for each antibiotic but as a precaution, a

range of concentrations were used for selection of transformant colonies (2.5 – 15 µg/mL

Pm and 10-25 µg/mL HygB). Potential transformants were then selected from plates with antibiotic concentrations for which there were minimal background colonies in the negative (no DNA) control transformation reactions. These were subsequently refreshed onto solid media containing 25 µg/mL of each respective antibiotic. As the expression of the transgene should be linked to that of the selectable marker, only those colonies that survived on plates containing higher concentrations of antibiotic were selected to potentially enrich for transformants also expressing high levels of the transgene.

Using these optimized conditions, hundreds of transformants from each construct were obtained. ~100 of each construct were selected for further screening. This process is ongoing in the laboratory now and optimism is high that one (or several) of these transformants will successfully yield a secreted, active BChE protein.

Chapter 3. Characterization of Cholinesterases from Multiple Large Animal Species for Medical Countermeasure Development against Chemical Warfare Nerve Agents

It must be noted that the following chapter has been adapted and reproduced in its entirety from the manuscript submitted for publication to Toxicological Sciences on

March 1, 2019. The revised version of the manuscript was accepted for publication on

April 15, 2019 and will be in press after the due date for upload of this dissertation.

Characterization of Cholinesterases from Multiple Large Animal Species for Medical Countermeasure Development against Chemical Warfare Nerve Agents

1,2*McGarry, K.G., 1Schill, K.E., 1Winters, T.P., 1Lemmon, E.E., 1Sabourin, C.L., 1Harvilchuck, J.A., 1Moyer, R.A

1Battelle Memorial Institute, Columbus, OH United States 2Department of Chemical and Biomolecular Engineering, The Ohio State University, Columbus, OH

*Corresponding Author

Preface

The work described within this section was derived from, and funded by, an institutional research and development project (IR&D) that was proposed and led by 25

Kevin McGarry while he maintained employment at the Battelle Memorial Institute and pursued a doctorate at the Ohio State University. This work provided the foundational knowledge and was instrumental in the development and engineering of the patent- pending novel human-porcine BChE hybrid enzyme discussed in Chapter 4. The work described in this chapter has been captured in the corresponding manuscript published alongside the human-porcine hybrid BChE manuscript and reproduced here.

3.1 Introduction

Organophosphorus (OP) cholinesterase inhibitors, including chemical warfare nerve agents (CWNAs) such as sarin and VX, continue to be a global threat against both military personnel and civilian populations. Inhibition of the enzyme acetylcholinesterase

(AChE) via phosphylation of the active serine site is the primary mechanism of toxicity of CWNAs. Inhibition of AChE prevents it from carrying out its normal physiological function (i.e., hydrolysis of the neurotransmitter acetylcholine). The resulting accumulation of acetylcholine leads to hyperstimulation of acetylcholine receptors at the neuromuscular junction and cholinergic neuronal synapses, which can trigger a cholinergic crisis that may ultimately result in respiratory failure and death 1. In the U.S., pharmacotherapy for OP-induced cholinergic effects typically includes the muscarinic antagonist atropine along with an oxime reactivator, the latter of which removes the OP from the enzyme 58-60. If needed, a benzodiazepine (e.g. diazepam) may also be administered to control seizures. 26

OPs (including CWNAs) also bind to and inhibit butyrylcholinesterase (BChE), a related enzyme that is sequentially and structurally similar to AChE [51-54% sequence identity in mammalian species; 24] that can also hydrolyze acetylcholine. The normal physiological function of BChE is not clear and it is not essential for life, as evidenced by the existence of genetic variations that result in a lack of BChE activity in apparently healthy people 61. Further evidence for the lack of a critical function of BChE may be found in mice, where genetic ablation of the BCHE gene yields healthy animals with no obvious phenotype 62. Therefore, it is not surprising that inhibition of BChE does not cause toxicity in humans or laboratory animals, and BChE’s role in the biological response to OP poisoning is primarily that of a bioscavenger that stoichiometrically binds the compounds with high affinity, effectively removing them from circulation. BChE is a soluble protein that is found in circulation and this trait, along with its capacity for binding CWNAs, has led to the ongoing development of exogenous BChE as a medical countermeasure (MCM) for OP poisoning 63.

Though effective, the currently approved MCMs for CWNA poisoning have significant limitations. For instance, the poor blood brain barrier permeability of most oximes restricts their capacity for reactivating inhibited AChE in the CNS, and the efficacy of individual oximes is highly dependent on the identity of the OP and the time elapsed between exposure and treatment 64. Efforts to develop improved MCMs are complicated by interspecies variability among other factors; the efficacy of individual oximes has been found to vary considerably between species depending on the,

experimental model and specific OP tested 65. Efforts to develop improved MCMs are ongoing in laboratories across the world, and selection of appropriate animal models is one of the major challenges faced by researchers. In the U.S., the Food and Drug

Administration’s Animal Rule permits the approval of MCMs for use in humans based on the results of animal efficacy studies when human efficacy studies would not be ethical and/or feasible (e.g., CWNA poisoning) 66, 67. It is generally recommended that MCM efficacy be demonstrated in more than one species that is expected to react with a response predictive of the human response 67. Due to their close evolutionary relationship and similarities to humans, nonhuman primates (esp. rhesus macaques) have historically been the large animal model of choice for CWNA MCM efficacy studies. However, the swine model (esp. Göttingen minipig), has been explored recently as an alternate/supplemental model to the nonhuman primate for CWNA MCM efficacy studies

68-70.

The expression and activity levels of AChE and BChE in the animal model are important factors to consider when extrapolating the results of CWNA MCM animal efficacy studies to humans. Therefore, it is not surprising that comparisons of the properties of AChE and BChE in common small animal models and humans are readily found in the literature 71. Although similar comparisons for large animal models have been reported 72-74, they are less common and generally more limited in scope, which confounds efforts to compare large animal species directly. To enable a direct comparison of the properties of AChE and BChE in relevant large animal models and

humans, the authors set out to elucidate the basal activity of each enzyme in circulation as well as the properties of inhibition, reactivation and aging in humans, Yorkshire swine,

Göttingen minipigs, African green monkeys, cynomolgus macaques, and both Indian and

Chinese origin rhesus macaques.

3.2 Materials and Methods

3.2.1 Chemicals.

5,5-dithio-bis-2-nitrobenzoic acid (DTNB), Acetylthiocholine iodide (ATC), S- butyrylthiocholine iodide (BTC), phosphate buffered saline (PBS), and pralidoxime chloride (2-PAM) were obtained from Sigma Aldrich (St. Louis, Missouri). All tests involving CWNAs were performed at the Battelle Biomedical Research Center (BBRC) located in West Jefferson, Ohio. The BBRC is certified to work with chemical surety material under a Provisioning Agreement with oversight by the U.S. Army Materiel

Command (AMC; Provisioning Agreement Battelle-1). Wherever applicable and required, the reporting requirements for this agreement were followed. All quantities of sarin and VX used during this testing were synthesized at Battelle’s Hazardous Materials

Research Center (HMRC) under Chemical Weapons Convention program guidelines, with accountability through the U.S. AMC. The sarin and VX originated from the same synthesis lots. All sarin and VX were stored in accordance with BBRC security and

CWA storage policies until needed for testing. To preserve purity, sarin and VX were stored in individual vials until needed for testing. 29

3.2.2 Enzyme preparation

Whole blood was obtained from either BioreclamationIVT (New York, New

York) or The Ohio State University (Columbus, Ohio). All blood samples were collected using Na-heparin as the anticoagulant. For the evaluation of basal AChE and BChE activity levels in the circulation, whole blood samples from 10 male and 10 female individuals from each species (i.e., human, Indian origin rhesus macaque, Chinese origin rhesus macaque, Yorkshire swine, Göttingen minipig, cynomolgus macaque, and African green monkey) was used. For the evaluation of inhibition, reactivation, and aging, whole blood samples were pooled from at least 5 males from each species. Red blood cell

(RBC) membranes and plasma were then prepared from each sample using methods optimized for the preparation of RBCs for kinetics experiments that were described previously 48, 75.

3.2.3 Enzyme activity determination

AChE and BChE activity were determined using a spectrophotometric assay as described previously 47, 48. Unless otherwise stated in the methods for each specific set of experiments, the final concentrations of the subtrates and indicator were as follows: 1 mM ATC (for RBC membranes), 3 mM BTC (for plasma), and 0.5 mM DTNB (for both

RBC membranes and plasma). All experiments were performed at 37°C in PBS.

Absorbance readings were recorded every 20 seconds over a period of 10 minutes at 412 nm using a BioTek® Synergy™ HTX Multi-Mode Microplate Reader. Absorbance

readings were initiated immediately after the addition of the substrate (ATC or BTC) for all experiments. Active enzyme units were calculated using Beer’s Law where, A = Ɛlc.

3.2.4 AChE and BChE inhibition assays

Concentration-response curves for each combination of matrix/species/CWNA were generated using CWNA concentrations ranging from 1.00E-04 M to 1.79E-12 M.

Each reaction mixture was incubated for five minutes at 37° C prior to the addition of substrate and DTNB. IC50 and IC90 values were calculated by nonlinear regression in

GraphPad Prism® 7. For the determination of inhibition rate constants (ki), the IC90 concentration of each CWNA was incubated with each combination of matrix/species at

37°C for various time intervals ranging from 0-15 min prior to the addition of substrate.

12 The ki values were calculated as reported previously .

3.2.5 Oxime reactivation of AChE and BChE

RBC membranes and plasma samples were incubated at 37°C with the experimentally determined IC90 of VX or sarin for 15 minutes. Following incubation, unbound VX or sarin was removed by passing the reaction mixture through 7K MWCO

Zeba spin desalting columns (Thermo Scientific; Waltham, MA). Following removal of unbound agent, agent-inhibited AChE and BChE were incubated at 37°C with 500 µM

DTNB for 5 min. Next, 2-PAM or vehicle was added to the reaction mixture,

immediately followed by addition of substrate. For all experiments that included 2-PAM, background resulting from oximolysis was accounted for during calculations.

3.2.6 Aging

RBC membranes and plasma samples were incubated at 37°C with the experimentally determined IC90 of VX or sarin for 15 minutes. Unbound VX or sarin was removed by passing the reaction mixture through 7K MWCO Zeba spin desalting columns (Thermo Scientific). Next, the inhibited enzyme was incubated at 37°C for various time intervals (0-48 hr) before 2-PAM (50 µM) was added and enzyme activity was determined. The rate constants for aging (ka) and spontaneous reactivation (ks) were calculated using a nonlinear regression model in GraphPad Prism® 7.

3.3 Results

3.3.1 Cholinesterase Levels in Circulation

Compared to humans, AChE activity levels in circulation were similar in the nonhuman primates and significantly lower in swine (Figure 6). For BChE, only Chinese origin rhesus macaque yielded results that were not significantly different from the human results (Figure 6). Basal BChE activity in the plasma of Indian origin and cynomolgus macaques was significantly higher than that observed in humans, while

African green monkeys and both swine displayed significantly lower values. The

contribution of sex to AChE or BChE activity within each species was also evaluated, and significant differences were identified in Indian origin rhesus macaques for AChE and cynomolgus macaques for BChE (males displayed higher values in both cases;

Figure 6). When both enzymes are taken into account, the Chinese origin rhesus macaque displayed circulating ChE levels most similar to those observed in humans.

Figure 6. Individual circulating cholinesterase levels Whole blood was obtained from 10 male and 10 female animals per species and processed to RBC membranes and plasma as described. Panel A displays AChE activity from the RBC membrane preparations of each individual using acetylthiocholine iodide as the substrate. Panel B displays BChE activity from the plasma preparations of each individual using butyrylthiocholine iodide as the substrate. Each point represents the mean activity from a single sample measured in triplicate in two independent experiments. A two-way Anova with a Tukey’s multiple comparisons test was utilized to assess statistical significance. P < 0.05 - 0.01 (*), P < 0.001 - 0.01 (**), P < 0.001 - 0.0001 (***), P < 0.0001 (****).

3.3.2 Enzyme Inhibition

Results of the enzyme inhibition and inhibition kinetics experiments are shown in Table

1. In all species, AChE was more susceptible than BChE to inhibition by sarin and VX.

Greater interspecies variability was observed for BChE (> 10-fold variance between some ki values for the same nerve agent between species) than for AChE (< 5-fold variance between any two ki values for the same nerve agent between species). Of the species tested, human AChE and BChE were the most susceptible to inhibition by both nerve agents, although the ki and IC50 values were generally similar between humans and all of the nonhuman primates for AChE. In general, the enzyme inhibition results obtained for rhesus macaques were the most similar to human, and the results obtained for swine were the least similar.

Table 1. Nerve Agent Inhibition Results

Agent: VX Agent: Sarin

Species 1 BChE AChE BChE

-1 -1 -1 -1 -1 -1 -1 -1 IC50 (M) ki (M min )] IC50 (M) ki (M min )] IC50 (M) ki (M min )] IC50 (M) ki (M min )]

Göttingen Mini 4.80 x 10- 1.42 ± 0.11 x 0.25 ± 0.06 x 1.24 ± 0.08 x 35.1 x 10-9 5.72 x 10-9 21.7 x 10-9 0.06 ± 0.01 x 107 Pig 9 107 107 107

Yorkshire Swine 5.27 x 10- 1.06 ± 0.10 x 0.28 ± 0.04 x 0.85 ± 0.05 x 32.1 x 10-9 2.07 x 10-9 21.8 x 10-9 0.14 ± 0.02 x 107 (Weanling) 9 107 107 107

Indian Origin 1.82 x 10- 2.56 ± 0.26 x 1.36 ± 0.25 x 1.81 ± 0.12 x 32.4 x 10-9 0.80 x 10-9 8.61 x 10-9 0.50 ± 0.06 x 107 Rhesus Macaque 9 107 107 107

Chinese Origin 1.89 x 10- 2.97 ± 0.28 x 1.44 ± 0.27 x 1.52 ± 0.18 x 40.0 x 10-9 0.59 x 10-9 6.74 x 10-9 0.57 ± 0.09 x 107 Rhesus Macaque 9 107 107 107

Cynomolgus 1.49 x 10- 2.97 ± 0.21 x 0.15 ± 0.02 x 1.81 ± 0.15 x 40.4 x 10-9 0.79 x 10-9 6.59 x 10-9 0.70 ± 0.06 x 107 Macaque 9 107 107 107

African Green 1.95 x 10- 2.65 ± 0.18 x 0.13 ± 0.01 x 1.70 ± 0.12 x 40.6 x 10-9 1.68 x 10-9 19.1 x 10-9 0.34 ± 0.04 x 107 Monkey 9 107 107 107

1.86 x 10- 4.03 ± 0.31 x 2.39 ± 0.43 x 2.00 ± 0.16 x Human 15.4 x 10-9 0.52 x 10-9 0.80 x 10-9 0.83 ± 0.14 x 107 9 107 107 107 Values represent mean ± SEM measured in triplicate in at least two independent experiments. 35

3.3.3 Aging

Cholinesterases may become refractory to reactivation following a dealkylation reaction which occurs after the covalent bonding of an OP to the active site serine

(termed “aging”). This dealkylated, phosphylated serine adduct is no longer reactivatable by oxime nucleophiles due to the resulting bond strength of the OP to the serine hydroxyl group and the local charges within the active site – specifically that the protonated histidine residue which constitutes of one-third of the catalytic triad, is no longer able to serve as a general base to assist in nucleophilic reactivation of the serine residue. The details of this aging reaction are well summarized by Nachon and colleagues 76. It should be noted, that the aging rate is a property specific to each individual OP and enzyme and not all OPs undergo an aging reaction.

Aging half-times (t1/2) of sarin-inhibited AChE ranged from 5 hours for human

AChE to 9.2 hours for Göttingen minipig AChE (Table 2). Except for human BChE, which displayed an aging t1/2 of 10.6 hours, no aging was observed for sarin-inhibited

BChE. As expected, aging of VX-inhibited AChE or BChE was not observed within 48 hours for any of the species tested (data not shown).

Most interestingly, however, was the observation of spontaneous reactivation of porcine BChE inhibited by sarin with a t1/2 of approximately 2 hours (Figure 7). This was not observed with VX. To the authors’ knowledge, this spontaneous reactivation of porcine BChE following inhibition by sarin has not been reported previously.

Table 2. Aging Half-Times Aging Half-Time (Hours)

Species Agent: Sarin

AChE BChE

Göttingen Minipig 9.2 NA

Yorkshire Swine 6.3 NA (Weanling)

Indian Origin 7.4 NA Rhesus Macaque

Chinese Origin 5.3 NA Rhesus Macaque

Cynomolgus 5.4 NA Macaque

African Green 8.2 NA Monkey

Human 5.0 10.6

NA: No aging was observed within the 48-hour experimental timeframe.

Values represent mean ± SEM measured in triplicate in at least two independent experiments.

Figure 7. Aging and/or spontaneous reactivation of sarin-inhibited cholinesterases. Activity of sarin-inhibited AChE or BChE in the presence or absence of 2-PAM. Sarin- inhibited RBC membranes from human (A), Yorkshire swine (B) or sarin-inhibited plasma preparations from human (C), Yorkshire swine (D) were incubated at 37°C until 2-PAM was added and the enzyme activity measured at the times indicated. Data represent mean ± SEM performed in triplicate. Aging half times were calculated using a non-linear regression model. For simplicity, only the sarin results are displayed above since aging was not observed with VX. Further, only human and one swine model are displayed as all of the primates evaluated yielded similar results and the Göttingen minipig displayed results similar to the Yorkshire swine. The calculated spontaneous reactivation of plasma BChE following a sarin challenge was 5.0 hours for the Yorkshire swine (6.7 hours for the Göttingen minipig).

3.3.4 Reactivation – EC50 Determination

The effective concentration of 2-PAM which reactivated each inhibited enzyme preparation to 50% activity of the unchallenged control (EC50) under the same experimental conditions was calculated (Table 3 and Figure 8). There was considerable variability in EC50 values between species, especially for BChE. AChE derived from

Indian origin rhesus macaque was the most prone to reactivation by 2-PAM. For BChE, the lowest EC50 values were observed in the Yorkshire swine and Göttingen minipigs

(Table 3 and Figure 8). It is worth noting that a greater degree of reactivation of VX- inhibited porcine BChE was achieved compared to the other species, and sarin-inhibited human BChE clustered with porcine BChE in displaying a greater degree of reactivation

(Figure 8).

Table 3. EC50 (µM) Determination of 2-PAM for Multiple Species versus VX or Sarin Agent: VX Agent: Sarin

Species AChE BChE AChE BChE

Yorkshire 177.8 ± 31.4 25.3 ± 3.4 120.9 ± 9.2 60.5 ± 7.7 Swine

Göttingen 236.6 ± 28.6 30.6 ± 4.1 113.2 ± 11.4 57.2 ± 13.5 Minipig

Indian Origin

Rhesus 52.3 ± 3.6 490.4 ± 40.6 40.5 ± 4.2 1142.0 ± 92.2

Macaque

Chinese Origin

Rhesus 95.2 ± 30.9 732.1 ± 51.0 59.1 ± 8.3 775.7 ± 99.0

Macaque

Cynomolgus 204.8 ± 30.2 559.7 ± 27.7 57.2 ± 7.0 3024.0 ± 928.0 Macaque

African Green 221.6 ± 22.1 277.0 ± 25.7 171.9 ± 13.9 582.0 ± 107.1 Monkey

Human 167.1 ± 17.4 268.9 ± 8.2 212.3 ± 54.6 71.2 ± 5.7

Values represent mean ± SEM measured in triplicate in at least two independent experiments.

E C 5 0 : B C h E v s . V X E C 5 0 : A C h E v s . V X 1 7 5 1 7 5

1 5 0

l H u m a n

1 5 0 H u m a n

f r

1 2 5 t

n C h in e s e R h e s u s M a c a q u e o

Y o rk s h ire y

1 2 5

i In d ia n R h e s u s M a c a q u e

C G o ttin g e n M in i P ig

d 1 0 0

t c

d 1 0 0

e A

In d ia n R h e s u s M a c a q u e g C y n o m o lg u s M a c a q u e

n e

E 7 5

l h

e 7 5 l

l C h in e s e R h e s u s M a c a q u e

h A fric a n G re e n M o n k e y

C y n o m o lg u s M a c a q u e c 5 0

c 5 0 Y o rk s h ire

% U

U A fric a n G re e n M o n k e y G o ttin g e n M in i P ig 2 5 2 5

0 0 0 1 0 0 2 0 0 3 0 0 0 1 0 0 2 0 0 3 0 0 [2 P A M C l], M [2 -P A M C l], M n  3 n  3

E C 5 0 : B C h E v s . G B E C 5 0 : A C h E v s . G B 1 7 5 1 7 5

1 5 0

l H u m a n

1 5 0 H u m a n

f r

1 2 5 t

n C h in e s e R h e s u s M a c a q u e o

Y o rk s h ire y

1 2 5

i In d ia n R h e s u s M a c a q u e

C G o ttin g e n M in i P ig

d 1 0 0

t c

d 1 0 0

e A

In d ia n R h e s u s M a c a q u e g C y n o m o lg u s M a c a q u e

n e

E 7 5

l h

e 7 5 l

l C h in e s e R h e s u s M a c a q u e

h A fric a n G re e n M o n k e y

C y n o m o lg u s M a c a q u e c 5 0

c 5 0 Y o rk s h ire

% U

U A fric a n G re e n M o n k e y G o ttin g e n M in i P ig 2 5 2 5

0 0 0 1 0 0 2 0 0 3 0 0 0 1 0 0 2 0 0 3 0 0 [2 P A M C l], M [2 -P A M C l], M n  3 n  3

Figure 8. Reactivation of VX-inhibited AChE and BChE by 2-PAM. Oxime reactivation of VX-inhibited AChE (A) and BChE (B) or sarin-inhibited AChE (C) and BChE (D) was performed as described. Background due to oximolysis was accounted for in each calculation. Data represent mean ± SEM performed of at least three independent experiments performed in triplicate. 41

3.4 Discussion

The characterization and selection of animal models for conditions that affect humans are critical factors for drug development, especially drugs developed under the

FDA’s Animal Rule. Both small and large animal models for human therapeutic development have their place in the field and understanding the limitations of a given model is imperative for the interpretation of the results generated. For MCMs seeking approval under the FDA’s Animal Rule, it is generally recommended that MCM efficacy be demonstrated in more than one species that is expected to react with a response predictive of the human response 67. NHPs are an accepted large animal model for biomedical research due to their similarities to humans, and rhesus macaques have been the most frequently used large animal model for advanced development of CWNA

MCMs. However, there are a number of important factors that are unrelated to biology or the relevance of the model that do not favor the continued use of NHPs (particularly

Indian origin rhesus macaques) for CWNA MCM research. These factors include availability, public perception, political pressure, and the high cost associated with working with NHPs due to procurement and specialized housing, animal husbandry, and training requirements. In light of these factors, swine models (esp. Göttingen minipig) have been explored as an alternative to the rhesus macaque as a large animal model for

CWNA MCM efficacy evaluation.

Swine have been used in biomedical research for decades, and many of their organ systems (esp. skin) share extensive similarities with humans. For a comprehensive 42

evaluation of important considerations for the use of swine models for OP poisoning, the reader is referred to the excellent review by Dorandeu and colleagues 77. References discussed herein are restricted to those that are especially relevant to the results presented in the current manuscript. Porcine models have been shown to be well suited as human surrogates for skin permeability 78, 79 and decontamination studies 80, 81. In studies that are relevant to CWNA research, swine were utilized for the evaluation of the dermal penetration and pharmacokinetics of cholinesterase-inhibiting pesticides 82, 83. More recently, the toxicity of the CWNAs sarin and VX were characterized in Göttingen minipigs 68, 84, 85, and efficacy and bioavailability of CWNA MCMs have also been evaluated in Göttingen minipigs 70, 86. However, the limitations of the swine model related to CWNA MCM research must be considered. Basal levels of circulating esterases, which may be an especially relevant consideration for research using cholinesterase inhibitors (e.g., CWNAs and OP pesticides), are substantially different in swine vs. humans. Compared to humans, minipigs have ~10-fold higher levels of total esterase activity in the blood 87. Conversely, swine have been shown to have much lower constitutive levels of BChE activity in the plasma 74, 88, which is corroborated by the data reported in Figure 6 of the current manuscript.

In general, the results reported in Table 1 indicate that human AChE is more susceptible to inhibition by sarin and VX than AChE derived from other species, and inhibition parameters were more similar between human and NHP AChE than between human and swine AChE. This observation is aligned with similar results reported

previously by other groups. For instance, Aurbek and colleagues found human AChE to be more sensitive than swine AChE to inhibition by VX, VR, and Chinese VX 73, and similar results were reported for human vs. minipig AChE inhibited by paraoxon, VX, and VR 74. The only exception was cyclosarin, for which the inhibition rate constants were slightly higher for swine AChE (4.84 x 108) vs. human AChE (4.21 x 108) 74. Both articles also reported longer aging half times for swine vs. human AChE 73, 74, which is corroborated by the data shown in Table 2 of the current manuscript. As shown in Table

3 and Figure 8. Reactivation of VX-inhibited AChE and BChE by 2-PAM, 2-PAM mediated reactivation of inhibited AChE varied depending on the CWNA and the species, although it can be stated that rhesus macaque AChE was the most prone to reactivation for both CWNAs. Similar studies reported in the literature have also given mixed results, which may not be surprising given that the rate and extent of reactivation is dependent on several factors that are themselves dependent on the oxime, agent, and species of origin of the enzyme 65. Herkert and colleagues reported the following for

AChE derived from the species indicated and inhibited with sarin or paraoxon (rank ordered from greatest to lowest reactivation rate): rhesus macaque > human > swine for sarin- inhibited AChE reactivated with obidoxime or HI-6; swine > human > rhesus macaque for paraoxon-inhibited AChE reactivated with obidoxime; rhesus macaque > swine > human for paraoxon-inhibited AChE reactivated with HI-6 89. In a related study that utilized the oxime MMB-4, five different CWNAs, and AChE derived from humans, swine, and cynomolgus macaques, the reactivation rate constants reported for human and

cynomolgus macaques were generally similar to each other and higher than those reported for swine AChE 90. Finally, Luo and colleagues concluded that reactivation and aging were similar for AChE derived from humans and African green monkeys as well as rhesus and cynomolgus macaques tested using three different CWNAs and four different oximes 72

In the results reported herein, there was much greater BChE variability between species than with AChE. This is not surprising, given that BChE is not essential for survival 61, 62 and therefore can be presumed to have been under less selective pressure throughout evolutionary history. As mentioned above, the observation that swine have lower basal levels of BChE in circulation than humans is consistent with previous reports

74, 88. In contrast to AChE, reports describing in vitro inhibition, reactivation, and aging of

BChE are relatively rare in the open literature 91, 92, and the author is not aware of any published comparisons of these parameters using BChE derived from multiple large animals. For all species, and for both CWNAs tested, BChE was less susceptible to inhibition than AChE, with differences in ki values between the two enzymes ranging from ~2-fold for human up to ~20-fold for swine (Table 1). This conflicts with data presented by Bartling and colleagues, who reported identical ki values for human AChE and BChE vs. sarin (i.e., 3.2 x 107 M-1 min-1) 91. The results reported in the current manuscript with sarin were 2.00 x 107 M-1 min-1 for human AChE and 0.83 x 107 M-1

-1 min for human BChE (Table 1). Given that ki values may vary by several orders of magnitude depending on the enzyme and the inhibitor and that the experiments were

conducted in separate laboratories using similar but not identical methods, the differences could be considered to be relatively small.

Reactivation results for BChE were even more variable between species than for

AChE, and BChE derived from humans and NHPs was less prone to reactivation than

AChE with one exception; namely human AChE/BChE inhibited with sarin (Table 3). In contrast, BChE was more prone to reactivation than AChE inhibited with either CWNA for swine. It must be noted that the reactivation results obtained for sarin-inhibited AChE and BChE, where the latter was more prone to reactivation by 2-PAM in the swine model, may appear to conflict with previous publications that reported the opposite 91, 92.

One plausible explanation is that the differences stem from different binding affinities for

2-PAM with BChE rather than the rate itself. Molecular dynamics simulations to deduce the reasoning for these differences are ongoing. It is also useful to consider variations in methods and analysis to explain the different results. The reactivation rate constants reported in the aforementioned references reflect the affinity and reactivity of an oxime toward OP-inhibited ChE. A discussion of reactivation kinetics is beyond the scope of the current manuscript; the reader is referred to excellent articles by Worek and colleagues for a thorough explanation of the reactivation rate constant 12, 93. For the experiments reported herein, enzyme activity was measured for 10 minutes following the addition of

2-PAM to inhibited AChE or BChE. This enabled calculation of the EC50, the concentration required to achieve half of the maximal response. Since the toxic effects of

CWNAs occur rapidly in vivo and the MCMs must also act rapidly in order to be

effective, it was decided that this relatively short time window is physiologically relevant and sufficient for a head-to-head comparison of the effectiveness of 2-PAM using multiple enzymes and CWNAs under identical experimental conditions. The EC50 values given in Table 3 do not allow for direct estimation of the reactivation rate, but they do permit a direct comparison between the species, enzymes, and CWNAs tested.

Furthermore, the results shown graphically in Figure 8 reveal the maximum level of reactivation achieved, which is an important consideration for the effectiveness of an oxime.

The robust spontaneous reactivation of sarin-inhibited BChE derived from

Yorkshire swine and Göttingen minipigs (Figure 7) was initially perplexing. One potential explanation for this is that sarin binds porcine BChE reversibly and the elevated levels of total esterase observed in swine 87 contribute to the degradation of sarin.

However, multiple experiments with recombinant BChE have excluded this as a likely explanation, and the reader is referred to the accompanying manuscript (Chapter 4) for the results of these experiments and a discussion of the putative mechanisms underlying this phenomenon (McGarry et al., 2019, in review). Furthermore, Brazzolotto and colleagues reported spontaneous reactivation of the S-enantiomer of VX using BChE purified from pig plasma 94. While this spontaneous reactivation was not observed in this study, this may be attributable to the different experimental methods that were employed, since Brazzolotto et al. conducted their experiments using purified porcine BChE and individual enantiomers of VX 94. Another unexpected observation was that 2-PAM-

assisted reactivation of VX-inhibited porcine BChE approached ~150% of the unchallenged control (Figure 8). An explanation for this result is elusive. It is unlikely to be an experimental artifact since it was species-specific. One could speculate that this result might stem from a substrate activation-like event which has been reported previously for human BChE with the butyrylthiocholine substrate 39. Exactly how 2-

PAM, VX, and the enzyme may be interacting to elicit this result, however, requires additional investigation. Molecular dynamics simulations are ongoing with our collaborators in an attempt to determine the cause of this porcine-specific reactivation event.

The toxic effects of CWNAs are mediated primarily through the inhibition of

AChE, and BChE is known to bind CWNAs and effectively remove them from circulation. Therefore, the properties of both enzymes in laboratory animals are critical factors for selecting an appropriate animal model and for interpreting the results of toxicology and MCM efficacy studies performed in animals in order to extrapolate the results to humans. To address perceived knowledge gaps in relevant large animal models, the authors evaluated the basal AChE and BChE activity levels in circulation as well as the inhibition, aging, and reactivation for both enzymes derived from humans and several large animal models under consistent experimental conditions. Some surprising observations regarding reactivation of porcine BChE are described above and explored further in the accompanying manuscript (McGarry et al., 2019, in review; i.e., Chapter 4).

The results described herein indicate that the properties of AChE and BChE are more

similar between NHPs (esp. rhesus macaques) and humans than between Yorkshire swine or Göttingen minipigs and humans. This conclusion is corroborated by previous reports

72-74, 90. Of course, it must be acknowledged that many other factors should be considered for animal model selection and data interpretation (e.g., absorption, distribution, metabolism, excretion, organ(s) involved in route of exposure), but these are beyond the scope of the current study. Of the species tested, the cholinesterase profiles of rhesus macaques were the most similar to humans, which suggests that the interactions of

AChE/BChE, CWNA, and a reactivator (e.g., 2-PAM) in CWNA MCM efficacy studies performed using rhesus macaques will most closely mimic the same interactions expected in humans.

3.5 Funding

Financial support for these studies was provided by Battelle’s Independent

Research and Development program.

3.6 Acknowledgements

The authors would like to acknowledge the efforts of the Battelle chemistry team, especially Brent McCracken, Beth Reed, and Meredith Andrews for their contribution to these experiments. Additionally, the authors would like to thank John Mitchell for his efforts in reviewing the data, as well as Kemla Siddoway, Christopher Hadad (The Ohio

State University), and Remy Lalisse (The Ohio State University) for their technical input and assistance in reviewing the manuscript.

Chapter 4. A Human-Porcine BChE Hybrid Enzyme

The following chapter and appendices were submitted for publication in

Toxicological Sciences in March 2019. The title of this submission and the list of authors is indicated below. The work has been been adapted and reproduced in its entirety. The effort and input of all of the authors are greatly appreciated.

A Novel, Human-Porcine Hybrid Butyrylcholinesterase Catalytically Degrades the Chemical Warfare Nerve Agent, Sarin

1,2*McGarry, Kevin G., 3Lalisse, Remy F., 1Moyer, Robert A.,1Johnson, Kristyn M., 2Tallan, Alexi M., 1Winters, Tyson P., 4McElroy, Craig A., 1Lemmon, Erin E.,3Shafaat, Hannah S., 3Marguet, Sean C., 2Fan, Yamin, 2Deal, Aniliese R., 1Harvilchuck, Jill A., 3Hadad, Christopher M., 2Wood, David W.

1Battelle Memorial Institute, Columbus, OH 2Department of Chemical and Biomolecular Engineering, The Ohio State University, Columbus, OH 3Department of Chemistry and Biochemistry, The Ohio State University, Columbus, OH 4College of Pharmacy, The Ohio State University, Columbus, OH *Corresponding Author

4.1 Introduction

Organophosphorus (OP) chemical warfare nerve agents (CWNAs), such as sarin and VX, are some of the most potent neurotoxicants known. Many pesticides also belong 51

to this class of chemicals but are generally much less toxic. The primary mechanism of

OP toxicity arises from inhibition of acetylcholinesterase (AChE) via phosphylation of the active serine site 7-10. This phosphylation event can have devastating consequences to a poisoned individual.

In cellular signaling involving cholinergic neurons, the neurotransmitter acetylcholine (ACh) is packaged and released into the synaptic cleft where it binds to and activates ACh-receptors, continuing a downstream signaling event. To attenuate the signaling event, AChE hydrolyzes excess ACh, thereby terminating neuronal signaling.

Thus, AChE is an essential regulator of neuronal cellular signaling. Inhibition of AChE results in the accumulation of ACh, leading to over-stimulation of ACh receptors and may ultimately lead to cholinergic crisis. The continuous stimulation of cholinergic neurons and/or skeletal muscles may result in seizures, paralysis, and ultimately respiratory failure and death 1. The current FDA-approved/fielded standard of care for OP poisoning includes the use of three drugs: 1) a muscarinic antagonist (e.g., atropine); 2) an AChE reactivator (e.g., pralidoxome); and 3) a benzodiazepine (e.g. diazepam), to control seizures.

Butyrylcholinesterase (BChE) exhibits a high degree of structural similarity and sequence homology to AChE 24, 95, and, like AChE, can hydrolyze the neurotransmitter acetylcholine. Notably, BChE is also inhibited by CWNAs. BChE is believed to have arisen early in the evolution of vertebrates due to a gene duplication event, and the structural similarities of the two enzymes are striking 16. Most notably, the catalytic triad

of BChE, like AChE, resides at the bottom of a 20 Å deep gorge that is lined with aromatic amino acid residues. However, while AChE has fourteen aromatic residues lining its gorge which facilitates pi-pi stacking interactions, BChE retains only eight 15.

Interestingly, the endogenous function of BChE remains largely unknown as individuals deficient in active BChE appear to lead normal, healthy lives 61. BChE, however, is thought to serve in a number of biological processes, including neuronal cellular development and neuronal signaling and also appears to have a role in bronchial airway smooth muscle function 31, 32.

One significant function of BChE is the ability to serve as a bioscavenger of toxic esters. The difference in the volume of the active sites between the two enzymes allows

BChE to hydrolyze a wider variety of substrates and thus serve in this capacity 26.

The use of exogenous human BChE as a prophylactic bioscavenger for CWNA poisoning has been evaluated in animal studies 31. Indeed, exogenous BChE is protective against

OP poisoning, although large amounts of BChE are required – for example, more than 20 mg/kg of BChE was required to protect guinea pigs against a 2.5x LD50 percutaneous challenge of VX 35, 96. With these promising results, BChE is being evaluated in early clinical trials. However, the binding of CWNAs to BChE is stoichiometric and irreversible (i.e., one molecule of BChE is required to remove one molecule of CWNA).

Thus, the required dose of BChE imposes a challenge for OP protection due to limitations on how much exogenous protein may be administered safely. The volume of human

serum required for purification and the cost associated with the purification pose other challenges for BChE as a medical countermeasure 20, 31, 97-99.

Mutagenesis of BChE to create a catalytic or self-reactivating bioscavenger has been extensively explored 97. Lockridge and colleagues discovered that a point mutation

(G117H) within the oxyanion hole confers organophosphorus hydrolase characteristics to

BChE 39. This discovery led to two decades of research regarding the creation and optimization of catalytic and pseudocatalytic (oxime-assisted) bioscavengers based on wild-type human BChE. Relatively recent developments involving several species and various mutations indicates that increased flexibility within the acyl loop of both pig

BChE and bovinated-human BChE appears to promote auto-reactivation following inhibition by VX or chlorpyrifos oxon 94, 100-102. To date, however, neither the G117H mutation alone, nor in combination with other mutations, has yielded an enzyme that

5 -1 -1 displays sufficient activity required for a catalytic bioscavenger (i.e. kcat/KM > 10 M s ;

98 ) while retaining sufficient binding affinity to surpass the effectiveness of wildtype

BChE as a medical countermeasure.

Herein, motivated by the catalytic activity of porcine BChE, we implemented some specific modifications to the human isoform, and then expressed this novel human- porcine hybrid BChE. The resulting enzyme appears to have the remarkable ability to catalytically degrade the CWNA sarin, while retaining near wild-type binding affinity.

4.2 Materials and Methods

4.2.1 Chemicals

5,5-dithio-bis-2-nitrobenzoic acid (DTNB), acetylthiocholine iodide (ATC), S- butyrylthiocholine iodide (BTC), phosphate buffered saline (PBS), 3-(N-morpholino) propanesulfonic acid, 4-morpholinepropanesulfonic acid (MOPS), and pralidoxime chloride (2-PAM Cl) were obtained from Sigma Aldrich (St. Louis, MO). All tests involving authentic chemical agents were performed at the Battelle Biomedical Research

Center (BBRC) located in West Jefferson, Ohio. The BBRC is certified to work with chemical surety material under a Provisioning Agreement with oversight by the U.S.

Army Materiel Command (AMC; Provisioning Agreement Battelle-1). Wherever applicable and required, the reporting requirements for this agreement were followed. All quantities of sarin used during this testing were synthesized at Battelle’s Hazardous

Materials Research Center (HMRC) under Chemical Weapons Convention program guidelines, with accountability through the U.S. AMC. The sarin originated from the same synthesis lot. All sarin was stored in accordance with BBRC security and CWA storage policies until needed for testing. To preserve CWA purity, sarin was stored in individual vials until needed for testing.

4.2.2 BLAST® sequence alignments

BLAST® Sequence Alignments. Using the National Center for Biotechnology

Information (NCBI) website, a BLAST® sequence alignment was performed to compare the amino acid sequences of BChE from five species. The species that were compared and their corresponding accession numbers were as follows: Human (Homo sapiens;

P06276.1), Pig (Sus scrofa; NP_001344438.1), African Green Monkey (Chlorocebus sabaeus; XP_007970394.1), Rhesus Macaque (Macaca mulatta; XP_002808379.2), and the Crab-eating / Cynomolgus macaque (Macaca fascicularis; NP_001306299.1).

4.2.3 Plasmid construction

The gene encoding human BChE was purchased from MyBiosource and cloned into the pTT expression vector 52 using the HindIII and AfeI restriction sites. This vector was selected for its high level of gene transfer and protein expression in transiently transfected human embryonic kidney, HEK 293 EBNA1 cells (HEK293E; ATCC: CRL-

10852)103. As described earlier in Chapter 2, site directed mutagenesis was performed using a modified QuikChange® reaction whereby Q5® High-Fidelity DNA Polymerase,

5X Q5 Reaction Buffer, and Q5 High GC Enhancer supplement (New England BioLab

Cat. No. M0491S) were utilized. The QuikChange® primer design tool (available at

Agilent’s website; https://www.chem.agilent.com/store/primerDesignProgram.jsp) and cycling conditions described within the QuikChange® protocol were used as described.

4.2.4 Enzyme production and purification

As briefly introduced in Chapter 2, multiple recombinant BChEs were transiently expressed and purified in HEK293E cells. Briefly, HEK293E cells were grown in 75 cm2 tissue culture flasks with vent caps containing 20 mL of Dulbecco's Modified Eagle's medium supplemented with 10% fetal bovine serum and 8 mM L-glutamine to ~70% confluence. Separate flasks (two per construct, 40 mL total) were transiently transfected using a Lipofectamine® 3000 Reagent kit (Invitrogen) with 20 µg DNA. Expression media was collected 7 days post-transfection and centrifuged at 100x g for 5 minutes to remove cellular debris. The resulting supernatant was diluted ≥ 2-fold in 1X Phosphate

Buffered Saline, pH 7.4 + 500 mM NaCl + 10 mM imidazole and subsequently purified.

Briefly, using a gravity column, the diluted supernatant was passed over an 800 µL bed volume of Ni-NTA resin (Thermo Fisher Scientific) equilibrated with 1X Phosphate

Buffered Saline, pH 7.4 + 500 mM NaCl + 10 mM imidazole. Columns were washed once with ten column volumes (CVs) 1X Phosphate Buffered Saline, pH 7.4 + 500 mM

NaCl + 10 mM imidazole and once with 1X Phosphate Buffered Saline, pH 7.4 + 500 mM NaCl + 25 mM imidazole. The protein was eluted with 1X Phosphate Buffered

Saline, pH 7.4 + 500 mM NaCl + 100 mM imidazole and 1 column volume (CV) fractions were collected. Fraction #2 of each purification was confirmed via activity assays and protein gel analysis to contain the highest amount of the desired product and therefore was used for each construct in the subsequent experiments.

4.2.5 Protein quantification

To determine the amount of purified protein produced from the transient transfection of HEK293E, a bicinchoninic acid assay (BCA) was used. Briefly, using a

Micro BCA™ Protein Assay Kit (Thermo Fisher Scientific), working reagent and BCA standards were prepared per manufacturer instructions. Experimental protein samples were diluted 1:10 in 1X PBS, pH 7.4. Each test plate was prepared by adding 150 µL of standard or protein samples to each appropriate well, followed by the addition of 150 µL of working reagent. The test plate was then incubated at 37°C for 120 ± 5 minutes with gentle shaking. At the end of the incubation period, the plate was cooled for a minimum of 5 minutes at room temperature. The plate was analyzed by measuring the absorbance at 562 nm within 15 minutes of the incubation stop time using a BioTek® Synergy™

HTX Multi-Mode Microplate Reader. Unknown protein concentrations were calculated by interpolating the values from the standard curve.

4.2.6 Western blotting

To confirm the presence of the various, purified BChEs, anti-histidine Western blotting was performed. Experiments were performed by first denaturing and separating the second elution from the purification aliquots under denaturing conditions using a 4-

12% gradient polyacrylamide gel (Thermo Fisher Scientific). Transfer to the polyvinylidene difluoride (PVDF) membrane was performed using an iBlot™ 2 (Life

Technologies) per manufacturer’s instructions. Immunoblotting was performed as follows: Following transfer, the membrane was rinsed in PBS, pH 7.4 + 0.1% Tween-20

(PBST) and blocked overnight at 2-8°C with 5% non-fat milk (prepared in PBST). The membrane was rinsed three times with PBST and the primary antibody (mouse anti- histidine IgG; Invitrogen Cat. No. MA1-21315) was prepared in 1% milk in PBST at a

1:10,000 dilution and added to the membrane. The membrane was incubated in the presence of the primary antibody for > 3 hours at 2-8°C. The membrane was rinsed three times with PBST and the secondary antibody (Horseradish peroxidase conjugated Goat anti-mouse IgG; Invitrogen Cat. No. 62-6520) was prepared in 1% milk in PBST at a

1:20,000 dilution and added to the membrane. The membrane was incubated in the presence of the secondary antibody for 1 hour at room temperature. The membrane was rinsed three times with PBST. Finally enhanced chemiluminescent (ECL) substrate

(Thermo Fisher Scientific) was added to the membrane and incubated at room temperature for 5 minutes prior to imaging.

4.2.7 Silver staining

To observe the purity at each step of the immobilized metal affinity chromatography purification scheme, each aliquot from the various purification steps was denatured and then separated under denaturing conditions using a 6% polyacrylamide gel.

To visualize the protein bands, a silver stain was performed. To perform the silver stain, each gel was rinsed in deionized water following completion of the PAGE. Fixation 59

solution (30% ethanol, 10% acetic acid) was then added to each of the gels and incubated at room temperature for > 18 hours with mild shaking. Gels were then rinsed four times

(twice in 20% ethanol and twice in deionized water) for 10 minutes each rinse.

Sensitizing solution (0.8 mM Sodium thiosulfate) was added to each gel and incubated with mild shaking for 1 minute at room temperature. Gels were rinsed twice in deionized water for 1 minute to remove excess sensitizing solution. The gels were then stained with

12 mM (2.04 mg/mL) AgNO3 for 20-120 minutes. The gels were again briefly rinsed with deionized water (10-30 seconds). Developer solution (3% w/v potassium carbonate,

10% w/v sodium thiosulfate, and 0.025% formalin) was added and the gels were allowed to react until bands became clearly visible (~5-10 minutes). The reactions were stopped using stop solution (4% w/v Tris Base, 2% v/v acetic acid) prior to imaging.

4.2.8 Enzyme activity assays

The rate of hydrolysis of varying concentrations of the chromogenic substrate, butyrylthiocholine iodide (BTC), was evaluated for the recombinant BChE enzymes in 10 mM MOPS buffer, pH 7.4 (unless otherwise indicated). The final BTC concentration was

3 mM, except for the substrate titration experiments, in which case the final BTC concentrations ranged from 7.8 µM to 10 mM. A concentration of 0.50 mM DTNB was used for all enzyme activity assays. BChE activity was determined spectrophotometrically using a method similar to that described previously 47.

Absorbance readings were captured at 412 nm using a BioTek® Synergy™ HTX Multi- 60

Mode Microplate Reader and enzyme activity was calculated using Beer’s Law, where A

= εlc. All experiments were performed at 37°C.

4.2.9 Enzyme inhibition

Enzyme inhibition was studied using the potent chemical warfare nerve agent, sarin. Recombinant BChEs were inhibited with varying concentrations of sarin ranging from 1.00 x 10-4 M to 1.79 x 10-12 M in PBS (pH 7.4) or in 10 mM MOPS buffer (pH 7.4) to produce an inhibition curve. After the addition of agent, samples were incubated at

37°C for > 5 minutes. Enzymatic activity was monitored as described above. The concentrations which inhibited 50% and 90% of enzymatic activity (IC50 and IC90 respectively) were calculated using GraphPad Prism® 7.

4.2.10 Inhibition rate constant determination

Recombinant BChE samples were incubated with the experimentally determined

IC90 sarin concentration at pH 7.4 and 37°C for various time intervals (0-15 minutes) prior to assessing enzymatic activity as described above. The inhibition rate constant, ki, was calculated as described previously 12.

4.2.11 Aging

Recombinant BChE was incubated at 37°C with the experimentally determined

IC90 of sarin for 15 minutes at pH 7.4. Unbound nerve agent was removed by passing the reaction mixture through 7K MWCO Zeba™ spin desalting columns (Thermo Fisher

Scientific Cat. No. 89894). Following removal of excess, unbound nerve agent, the inhibited enzymes were incubated at 37°C with an aliquot being removed at various time intervals (0-48 hr). 2-PAM Cl (50 µM) was added and enzymatic activity was determined using the Ellman’s method described above.

4.2.12 GC-HRMS

To generate samples for GC-HRMS analysis, porcinated human BChE was challenged with 10, 167, or 500-fold molar excess sarin at 37°C and aliquots were removed at T = 0, 2, and 5 hours post-challenge and precipitated with ethyl acetate.

Samples were stored at -70°C prior to analysis. The sarin analysis was performed using a gas chromatographic (GC) system coupled with a Thermo DFS™ magnetic sector high resolution mass spectrometer (HRMS). Ionization was accomplished in electron impact

(EI) ionization mode. The ion calibration gas was PFTBA, and the lock mass used was

99.99306 Da. The outer source of the MS was maintained at 250°C and the injection volume was 2 µL. The GC column was a CP Sil-5 column with a 0.25-mm internal diameter and 1.0 µm film thickness (Agilent, Santa Clara, CA, USA). Helium carrier gas

was maintained at a constant flow of 1.0 mL/min, and the transfer line was set at 280°C.

The GC program began at 50°C for 2 min, ramped to 70°C at 20°C /min and was held for

4.5 min, ramped to 80°C at 4°C /min, then ramped to 280°C at 30°C/min, where it was held for 5 min. The total GC run time was 21.1 min. Table 4 contains a summary of the instrumental parameters and a list of the masses that were monitored during the analysis.

Working stock solutions of sarin were prepared from the sarin stock solution in ethyl acetate for use in preparation of calibration standards and quality control (QC) samples. A calibration (standard) curve was analyzed at concentrations of 0.2, 0.4, 1.0,

2.0, 5.0 and 20.0 ng/mL. A weighted linear regression curve using 1/x as the weighting factor, with x being the concentration of sarin spiked in ng/mL, and y being the sarin peak area, was used to calculate the correlation coefficient (r = 0.971). The formula for linear regression was used as follows: y = mx + b.

Table 4.GC-HRMS Parameters Mass Spectrometer Thermo DFS GC Thermo Trace Ultra w/ TriPlus Autosampler Data Acquisition Software Xcalibur 2.0 Data Reduction Software Waters MassLynx 4.0 Column Agilent CP Sil-5 - 30m x 0.25mm I.D. x 1.0µm film Temperature Program Rate (°C/sec) Temperature (°C) Hold (min) 0 50 2.0 20 70 4.5 4 80 0 30 280 5 Total Run Time 21.1 min Carrier Gas / flow Helium / 1.0 mL/min constant flow Injection Mode Splitless Injector Temperature 250 °C Split Flow / Splitless Time 30 mL/min / 1.0 min Injection Volume 2 µL Autosampler Temperature Ambient Transfer Line Temperature 280 °C Ion Source Electron Ionization (40 eV) Polarity Positive Source Temperature 250 °C Scan Type MID – Lock Mass Acquisition Time 6.5-10.0 min Monitored Ions Compound Mass (Da) Approximate Retention Time (minutes) Sarin Primary 99.00057 7.88 Sarin Secondary 125.01622 7.88 PFTBA Lock Mass 99.99306 NA PFTBA Calibration 130.99147 NA

4.2.13 LC-MS/MS

The analysis of sarin and its hydrolysis product, isopropyl methylphosphonic acid

(IMPA), was performed using an ultra-pressure liquid chromatographic (UPLC) system coupled with a Waters TQ-XS quadrupole mass spectrometer (MS/MS). Data was collected with MassLynx 4.2 (Waters, Milford, MA, USA) using multiple reaction monitoring (MRM) for each of the ion transitions summarized in Table 5. The electrospray ionization source was operated in positive ion mode and under the following parameters: curtain gas = nitrogen at 1,000 L/hr; collision gas = argon at 0.15 mL/min; ion spray voltage = 1 kV; temperature = 110oC. The collision energy and cone voltage were optimized for each transition and are reported in Table 5, along with the rest of the instrumental parameters. The timepoint sample aliquots (15 µL) were separated by reversed-phase chromatography using a Phenomenex Prodigy ODS HPLC column.

Mobile phases were 0.1% formic acid in HPLC grade water (mobile phase A) and 0.1% formic acid in acetonitrile (mobile phase B). The gradient profile used is summarized in

Table 5 at a flow rate of 0.4 mL/min. Extracted ion chromatograms were assessed on

MassLynx 4.2 for correct analyte peak shape, retention time, and manually integrated when needed.

A sarin working solution was used to prepare a sarin stock solution in Milli-Q water. A 1000 µg/mL standard of IMPA in methanol was purchased from Cerilliant

(Round Rock, TX, USA). A working stock solution of sarin and IMPA was prepared from the stock solutions in Milli-Q water for use in preparation of calibration standards 65

and quality control (QC) samples. A calibration (standard) curve was analyzed at concentrations of 0.1, 0.5, 1.0, 5.0, 50, 100, 500 and 1,000 ng/mL, with the exception of sarin which had a limit of quantitation (LOQ) of 0.5 ng/mL, not 0.1 ng/mL. A weighted linear regression curve using 1/x2 as the weighting factor, with x being the concentration of sarin/IMPA spiked in ng/mL, and y being the sarin/IMPA peak area, was used to calculate the correlation coefficient (r = 0.996). The formula for linear regression was used as follows: y = x2 + mx + b

Table 5. LC-MS/MS Parameters Tandem Mass Waters Xevo TQ-XS Spectrometer UPLC Waters Acquity Data Acquisition Waters MassLynx 4.2 Software MS Source Electrospray, positive ion mode Column Phenomenex Prodigy ODS (3) 100Å 3µm 2.0 x 100mm Column Temperature 30°C Mobile Phase A 0.1% Formic Acid in Milli-Q Water Mobile Phase B 0.1% Formic Acid in Acetonitrile Gradient Profile Flow Rate, Time, min %B Curve mL/min 0 10 0.3 - 1 10 0.3 6 5 75 0.3 6 5.01 75 0.4 6 8 75 0.4 6 8.01 10 0.4 6 10 10 0.3 6 Injection Volume 15 µL Capillary 1 kV Source Temperature 110°C Desolvation, nebulizer Nitrogen @ 1,000 L/hr and 300°C gas Collision gas Argon @ 0.15 mL/min Mass Resolution Unit in both quadrupoles Run Time Approximately 10 minutes Monitored Ions Precursor Product Mass Collision Compound Cone (V) Mass (Da) (Da) (eV) GB 141 20 99 10 79 20 IMPA 139 14 97 8 47 20 MPA 97 34 79 13

4.2.14 Density functional theory calculations

A non-standard residue calculation was performed for the inhibited form of the phosphylated serine after inhibition with sarin. Geometry optimization was done with the hybrid exchange functional and the Lee–Yang–Parr correlation functional (B3LYP)104-106 in combination with the cc-PVTZ basis set107 and implicit solvation for water using the

IEF-PCM solvation method108-110. At the same level of theory, Merz-Kollman electrostatic potential (ESP) calculations were performed to determine the atomic charges of all atoms. The B3LYP/cc-PVTZ method correlates with the charge calculations of the standard residues in the AMBER ff03 force field111, 112.

4.2.15 Protein preparation for molecular dynamics simulations

For the native enzyme, a crystal structure of human BChE (PDB: 1P0I)15 was the starting point. Missing residues, including D378, D379, and N455, were added according to the FASTA sequence and implemented manually using MODELLER113. Missing hydrogens were added using AMBER’s xleap114, 115 module. The protonation states of the titratable residues were determined using the PDB: 2pqr116 utility for a pH of 7.4 and several chloride anions were added to neutralize the system. For the inhibited form of

BChE by sarin, three different starting points were used. A forward-engineered inhibited structure was created from the native crystal structure of BChE (PDB: 1P0I)15. The native structure was edited in UCSF Chimera117 to reflect the inhibited form of sarin and

missing hydrogens were added with AMBER’s xleap114, 115 module. Two other preparations of the inhibited structure were done with a crystal structure of human BChE inhibited by enantiopure isomers of (R)-VX (PDB: 2XQJ) and of (S)-VX (PDB:

2XQK)118. Two missing residues at the N-terminus were omitted in the protein preparation from the VX-inhibited enantiomers. The original protein structures were edited to reflect the inhibited form of sarin at the active site serine residue. The partial atomic charges for the atoms for the phosphorylated inhibited of the catalytic serine residue were calculated using the RESP protocol implemented in the Antechamber18 module in the AMBER 18 package after the previously described ESP calculations. The protonation states of the titratable residues were determined using the pdb2pqr utility for a pH of 7.4116. All missing hydrogen atoms were added and several chlorides were added to neutralize the system.

For the protein preparations described above, the same was done for the porcinated form of BChE. The mutations for the Y282N/G283H/T284M/P285L native and inhibited forms were made using the most probable Drunback rotamer119 as predicted by UCSF Chimera117. At the site of the G283H mutation, three possible different protonation states of  (HIE),  (HID), and doubly protonated (HIP) histidine were considered.

4.2.16 Molecular dynamics (MD) simulations

All MD simulations were performed using the AMBER ff03 force field111, 112 in the presence of explicit water molecules, using a TIP3P120 representation. The structures were solvated in an octahedral water box and water molecules were added until about 12

Å away from the enzyme. The MD protocol involved a two-step minimization, a heating step, an equilibration step, and finally a production MD simulation. In the minimization procedure, the solvent and ions were first minimized followed by the entire system for

2500 steps. Before starting the equilibration and production MD simulations, the temperature was raised from 0 to 300 K with a small force constant on the enzyme to resist any drastic changes. The system was then subjected to a 40 ns equilibration step.

Normally equilibration steps are on the ~10 ps time scale, but in order to ensure a constant backbone root mean square deviation (RMSD), especially for the inhibited forms, a longer equilibration was used. Finally, each system was then subjected to a production MD simulation for 50 ns. The number of atoms, temperature, and pressure were kept constant for these NPT MD simulations, and periodic boundary conditions were used. During the production simulation, the time step was set to 2 fs over a 50 ns production step resulting in 25,000 frames for analysis. All distances and RMSD values were measured using AMBER’s CPPTRAJ and PYTRAJ modules121.

4.2.17 Clustering Analysis

To decrease the amount of structures for analysis, a clustering procedure for each frame was implemented using AMBER’s CPPTRAJ module121. Similar RMSD values of the protein backbone were grouped in several clusters and a centroid was chosen as a

“representative” structure for each cluster. In order to get an accurate representation of all of the possible conformations of the enzyme, five structures were chosen for analysis that represent close to 100% of the 50 ns MD simulation. In the supporting information,

RMSD plots are included in order to illustrate several clusters and how they were grouped by similar RMSD values.

4.3 Results

4.3.1 Sequence Alignment

To ascertain a possible mechanism for the unexpected result described in Chapter

3 (i.e. that porcine BChE can reactivate following exposure to sarin), a sequence alignment was conducted comparing Sus scrofa BChE with that of BChE from several primates (Appendix B-1). While it is expected that multiple amino acids between species would differ, a stretch of four amino acids at positions 282-285, when compared with human, were different in an otherwise conserved region of the enzyme (Figure 9). Two of these residues (284 & 285) are located within the aminoacyl binding pocket of the enzyme, while the other two residues are immediately adjacent to the pocket. While other

differences were observed, those differences were not explored for their importance in reactivity in this work.

Pig 301 LQNEVFVVPNHMLLSVNFGPTVDGDFLTDLPDTLLQLGQFKKTQILVGVNKDEGTAFLVY 360 Human 301 .L..A....YGTP...... M..I..E...... 360 Figure 9. Sequence alignment comparison of human BChE vs. porcine (Sus scrofa) BChE The red letters indicate differences between the two species. The highlighted region represents the acyl binding pocket of the enzyme.

4.3.2 Purification of Recombinant BChEs

As described above, multiple recombinant BChEs variants were transiently expressed in HEK293E cells and purified from the cell culture supernatant. To assess purity and to ensure the presence of the proteins of interest, silver staining of the purifications of two of the constructs as well as an anti-histidine Western blot of each construct were performed under denaturing conditions. As observed in Figure 10 and

Figure 11 below, the histidine-tagged BChEs are clearly present as a prominent band at

~85 kDa – indicative of the expected size of the fully glycosylated human BChE. The identity of the copurifying, ~130 kDa, anti-histidine reactive protein in the human WT, human-porcine hybrid, and human G117H variants (Figure 10) was not determined. An anti-BChE western blot was also performed against the WT and human-porcine hybrid enzymes and confirmed the presence of the BChE (data not shown); however, the antibody that was used appears to have lower sensitivity to the human-porcine hybrid enzyme. Although purification using procainamide resin was attempted, it yielded 72

varying levels of success depending upon the construct; therefore, immobilized metal affinity chromatography (IMAC) purification was used for the purification of all of the constructs discussed herein. In particular, Elution 2 (as shown in Figure 12) was used for all subsequent experiments, despite the observation of impurities in the silver stained gel.

Figure 10. Anti-Histidine Western blot following the expression and purification of indicated BChE constructs

Figure 11. Anti-BChE Western Blot following the purification of recombinant BChE.Human WT BChE (left) and human-porcine hybrid BChE (right) are displayed. 75

Figure 12. Silver stain following the expression and purification of human WT BChE (left) and human porcinated BChE (right).

To ensure that the purified WT and human-porcine hybrid enzymes catalyzed cleavage of the substrate butyrylthiocholine (BTC) in a similar manner, a substrate titration experiment was performed. As seen in Figure 13, while both the KM (WT: 456 µM vs. hybrid: 292 µM) and the Vmax (WT: 20.5 vs. hybrid: 8.78), of the hybrid enzyme are lower than that of the WT, the ability to catalyze BTC hydrolysis is not greatly reduced.

Figure 13. Substrate titration comparison of WT BChE vs. human-porcine hybrid BChE.

4.3.3 Inhibition of Recombinant BChEs by Sarin

Each recombinant enzyme was challenged with varying concentrations of sarin and assessed for residual activity. As observed in Figure 14, similar levels of inhibition are observed in two “clusters” of enzymes. The human WT, porcine WT, and human- porcine hybrid BChE all display similar IC50 values (5.8 nM, 23.0 nM, and 103.2 nM, respectively). Additionally, as controls, and in an attempt to improve the human-porcine 77

hybrid BChE, variations of BChE were created using mutations that have been previously identified and characterized in depth 32, 36, 39, 97, 98. Interestingly, each of these enzymes

(human G117H, human-porcine hybrid with G117H, and human-porcine hybrid with

G117H, E197Q) cluster together with a much higher IC50 than that of the other three enzymes (45.6 µM, 24.3 µM, and 17.7 µM, respectively). High substrate affinity is a desirable component for all candidate therapeutic bioscavengers, and included in this

39 figure as a point of reference are the reported Kd for the G117H mutation as well as the estimated intravenous dose corresponding to 50% lethality (IV LD50) of sarin for a 70 kg

122 human male . It should be noted that the estimated IV LD50 was converted to the corresponding molar concentration in the blood based on the published results using the estimated blood volume of a 70 kg human male to be 5.250 L. The kcat and Kd/Km has been a point of emphasis for years regarding the G117H mutation’s inability to efficiently catalyze the hydrolysis of nerve agents. As shown below, the driving property behind the ineffectiveness of G117H likely stems from a low affinity of the nerve agent for this enzyme. On the contrary, human WT BChE (which is currently in clinical trials as a stoichiometric bioscavenger), the porcine WT BChE, and the human-porcine hybrid

BChE enzymes all display much higher affinity for sarin and are readily inhibited by the agent. Furthermore, the rate of inhibition (ki) of each of the recombinant enzymes are similar (; n = 6). Additionally, the recombinant enzymes, with the exception of recombinant human WT BChE (which is an order of magnitude lower), display a

comparable ki to that of the plasma counterparts as described in Chapter 3 and reproduced here.

Figure 14. Inhibition of multiple recombinant BChEs by sarin. As a reference, the estimated IV LD50 and the Kd of the G117H BChE variant are included. Each point represents mean ± SEM of triplicate measurements.

Table 6. ki calculations for the inhibition of WT and recombinant BChEs by sarin

Gottingen Yorkshire Recombinant Recombinant Human Recombinant Mini Pig Swine Human WT Human-Porcine (Plasma) Pig WT BChE (Plasma) (Plasma) BChE Hybrid BChE

Mean k 8.30 ± 1.41 x 0.63 ± 0.11 x 1.36 ± 0.19 x 0.45 ± 0.03 x 0.97 ± 0.17 x i 2.31 ± 0.43 x 106 (M-1min-1) 106 106 106 106 106

4.3.4 Aging and Spontaneous Reactivation of Recombinant BChEs

As observed in Figure 15A, the human-porcine hybrid BChE displays spontaneous reactivation with a t1/2 of 2.4 hours and a maximal reactivation of approximately 75% of the activity of the unchallenged control. The addition of 2-PAM

Cl results in an additional 15-30% activity. These experiments were also carried out in multiple, separate experiments to 48 hours with similar results (Appendix B-2). The 22- hour experiment is shown below as the time points utilized for this experiment were also analyzed by mass spectrometry – an analysis that was not performed in the 48-hour experiments.

Figure 15B (and Appendix B-3) shows the measured concentrations of IMPA

(ng/mL), the postulated primary degradation product of sarin, in both the human wild type and human-porcine hybrid BChE enzymes that were challenged with sarin. The samples that were frozen at -70°C were thawed and aliquoted for analysis without any additional sample manipulation. The reported concentrations are in ng/mL and were calculated using an eight-point standard calibration curve that went from 0.1 to 1,000 ng/mL. The targeted LC-MS/MS method that was used monitored the IMPA 139 > 97 ion transition and the determined concentrations are plotted (Figure 15B and Appendix

B-3). The concentrations of IMPA increased from 0.31 to 0.66 ng/mL in the first 8 hours for the human-porcine hybrid BChE, indicative of the degradation of sarin. In the human

WT BChE, an increase in IMPA was observed, but to a much lower extent, from 0.17 to

0.21 ng/mL in the first 8 hours. The LC-MS/MS also monitored for the presence of sarin 81

in these samples. The detection limit for sarin was 0.5 ng/mL and, as expected, it was not detected at any of the timepoints because all sarin is presumably covalently bound to the protein. Additionally, included in Figure B, a plot of the sarin-only control in buffer alone is plotted on the right-hand axis to observe the natural hydrolysis reaction over time

(the concentration of IMPA increased from 122 to 295 ng/mL in this case). Much greater

IMPA concentrations were observed from the enzyme-free controls due to the higher initial, unbound concentrations.

An overlay of the extracted ion chromatograms for a 10 ng/mL solvent standard of IMPA (137 > 97) and sarin (141 > 99) can be seen in Figure 16A. The observed retention times for the two ions were 1.56 min for IMPA and 3.66 min for sarin. Figure

16B shows a magnified region of the chromatogram (1.2 – 2.0 min) with all of the timepoints plotted for the human-porcine hybrid BChE samples. The IMPA peak areas

(~1.56 min) increase with each timepoint, indicating the degradation of sarin over time.

In addition to the LC-MS/MS data, GC-HRMS data obtained from dilutions following a 500-fold molar excess challenge indicated a slight difference in the kinetic profile of the degradation of sarin over time in the presence of the porcinated-human hybrid enzyme when compared to the no enzyme control (Appendix B-4). It should be noted that unlike the aforementioned aging experiments, excess, unbound agent was not removed using spin columns in this experiment. As such, the corresponding activity results indicated that in the presence of 500-fold molar excess sarin, the hybrid BChE was effectively rendered inactive for the duration of the experiment with enzymatic

activity only increasing by 1% over the 5-hour timeframe. Following a challenge using a

10-fold molar excess of sarin, enzymatic activity was beginning to recover, albeit slowly

(3.5% total increase, 13.3% at 2 hours to 16.8% at 5 hours post-challenge, data not shown).

R e c o m b in a n t B C h E s v s . S a rin S p o n ta n e o u s R e a c tiv a tio n

1 0 0

y H u m a n (W T ) B C h E

o t

i 7 5

i H u m a n (W T ) B C h E + 5 0 µ M 2 P A M -C l

d c

e H u m a n (P o rc in a te d ) B C h E

n 5 0 E

e H u m a n (P o rc in a te d ) B C h E + 5 0 µ M 2 P A M -C l

c 2 5

n U

0 0 4 8 1 2 1 6 2 0 2 4 T im e (H rs ) n = 3

B D e g r a d a tio n o f S a r in

H u m a n (W T ) 1 .0 4 0 0 E

n B C h E

[

I H u m a n

M L 0 .8

m 3 0 0 (P o rc in a te d ) B C h E

P m

e /

- g 0 .6 F E n z y m e -F re e

] n

, C o n tro l

e ,

2 0 0 n ]

0 .4 C

m P

L M

I 1 0 0 [ 0 .2 t

l 0 .0 0 0 4 8 1 2 1 6 2 0 2 4 T im e (H r s ) n = 1

Figure 15. Catalytic degradation of sarin by human-porcine hybrid BChE. A) Percent activity/reactivation of both wild-type and porcinated BChE measured in presence or absence of the oxime reactivator, 2 PAM-Cl. The enzymes were subjected to the aging experiments as described previously. Following the removal of unbound agent using spin columns, reactions were incubated from 0 to 22 hours at 37°C prior to the addition of 2 PAM-Cl and activity was determined using the Ellman’s method as described. Error represents ± the standard error of the mean. B) Concentration of IMPA, ng/mL, measured by LC-MS/MS analysis at 0, 1, 2, 4, 6, and 22 hours using aliquots from inhibition reaction described in (A). The concentration of IMPA that was detected in a sarin control (i.e. in MOPS buffer alone) is plotted at the various time points on the right y-axis to show the degradation sarin in the absence of enzyme.

8  1 0 6 IM P A

6  1 0 6

r A

4  1 0 6

k a

e S a rin P 2  1 0 6

0 1 .5 2 .0 2 .5 3 .0 3 .5 4 .0 R e te n tio n T im e (m in )

T - 2 2 H rs 3  1 0 6 T - 6 H rs

a T - 4 H rs e

r 2  1 0 6 A

T - 2 H rs

k a

e T - 1 H rs

P 6 1  1 0 T - 0 H rs

0 1 .2 1 .4 1 .6 1 .8 2 .0 R e te n tio n T im e (m in )

Figure 16. Extracted Ion Chromatograms for Sarin and IMPA. Standard sarin (retention time = 3.66 min) and IMPA (retention time = 1.56 min) controls are displayed in panel A showing the experimentally observed peaks. These peaks are displayed at the expected retention times for these two compounds based off previous experience with these analytes using this LC-MS/MS method. Panel B displays a zoomed in capture of the IMPA chromatogram at each time point from 0 to 22 hours post challenge. As can clearly be seen, the abundance of IMPA increases over time, corresponding to the reactivation of the inhibited enzyme.

4.3.5 Molecular dynamics (MD) simulations of WT and hybrid BChE before and after

inhibition with sarin

As to gain a better understanding for the potential of active WT and potential for enzyme reactivation, computational tools were employed by Remy Lalisse. The WT enzyme was evaluated and inhibited forms. The Y282N/G283H/T284M/P285L variant was also evaluated for 50 ns, taking into consideration three different possible protonation states of  (HIE),  (HID), and double protonated (HIP) histidine at the site of the G283H mutation. (Significant additional data is included in Appendix B-5.)

The active site distances of the WT enzyme and the three different mutant structures are evaluated in Appendix B-5. Active site distances that were evaluated for the native forms of the enzyme include the O of Ser198 and the hydrogens from the backbone amides of Gly116, Gly117, and Ala199 (oxyanion hole), the interactions of H of Ser198 with the N of His438 (SerH-HisN) and oxyanion of Glu197 (SerH-

Glu197), the H interactions of His438 (Glu325 and Glu441), the interaction of the backbone amide to Glu197, and the measurement of the proximity of the acyl-pocket residues Leu286 and Phe329 to the active site (Leu286-HisN and Phe329-HisN). To evaluate the impact at the site of the mutation, important interaction distances of Tyr282,

Pro285, and Tyr332 were measured for the WT enzyme and distances of His283, Leu285, and Tyr332 were measured for the mutated enzyme.

The active site distances of the WT and HID forms of the native enzyme were observed to be quite similar; however, the oxyanion hole created by the O of Ser198 and the hydrogens from the backbone amides of Gly116, Gly117, and Ala199 are much shorter in the HID form. The decrease in the distance of the oxyanion hole is possibly as a result of the encroachment of Leu286 and Phe329 into the active site (shown in

Appendix B-5; Supplemental Figure S6). Leu 286 is adjacent to the 4-residue mutation, and the observed encroachment may be a result of the P285L mutation. Tyr282 and the backbone of Pro285 of the WT form seems to “anchor” the position of the α-helix that includes several important residues, such as Tyr332, Phe329, and Glu325. Leu285 seems to “push” the α-helix and allows Leu286 to move into the active site. In the WT form of the enzyme, the α-helix containing Tyr332, Phe329, and Glu325 is situated very close to the acyl loop. The HID form of the His283 mutation is situated above the acyl-loop and the -hydrogen shows a small interaction with the alcohol group of Tyr332 as it is pushed away from interaction with backbone of Leu285.

The active site interactions of the HIE and HIP forms of the His283 hybrid enzyme were quite different from the WT enzyme. Noticeable differences were observed in the lack of interaction between the H of Ser198 and the oxyanion of Glu197 as well as the weak interaction of the H of Ser198 and HisN. In the HIE form of the His283 mutation, the catalytic His438 moves away from the active site and into the active site gorge where there is little to no interaction with the H of Ser198. As a result, for the hybrid enzyme, the catalytic triad transforms into a catalytic dyad. The active site climb

can also be observed by the very close distance of ~7.5 angstroms to the Trp82 residue of the -loop (shown in Appendix B-5; Supplemental Figure S5). As a result of the HIE mutation, there is a very strong interaction between the -hydrogen of His283 and the carbonyl backbones of Pro230 and Trp231 siting at the bottom of the acyl-binding pocket. Tyr332 lacks the same strong interaction with the Pro285 backbone and enters the active site to interact with the catalytic His438 after climbing into the active site gorge.

The HIP form of the native His438 enzyme performs a similar “climbing” of the active site, except not nearly as high of a climb, as evidence by the Trp82 distances sitting at

~10 Å. Instead, the catalytic His438 sits just above Ser198 and forms a catalytic dyad by a strong hydrogen bonding interaction of H of His438 and Glu197.

The observed differences in the native enzyme from WT to mutated forms suggests a possible explanation for the decrease in Vmax for the hybrid enzyme. However, due to uncertainties of the protonation states for His283, we are unsure why the KM is lower in the hybrid form of the enzyme.

The inhibited forms of the BChE enzyme were studied with two different starting

PDB structures (2XQJ and 2XQK), which are inhibited forms of the less and more toxic stereoisomers of VX, respectively. To evaluate each active site, similar distances to the native form were measured; however, the distances to the O of phosphylated Ser198 has been split into 4 distances in order to emphasize the different distances of the phosphorus oxygens near the oxyanion hole (Oxyanion-hole and HisH-O1), the bridging oxygen to the isopropyl leaving group (HisH-O2) and the oxygen of the original Ser198 group 88

(HisH-O). As seen in the native form, the active sites are quite similar for the 2XQJ simulations between the WT inhibited and the hybrid HID inhibited forms. A small difference is the decrease of interaction between the oxyanion hole in the HID form, but one does observe a very weak to no interaction of ~5 Å to the HisH. For 2XQJ, the HIE and HIP forms of the mutation lack the strong oxyanion hole interactions of the WT and

HID forms and as a result, the phosphorus oxygens show a much closer interaction to

HisH. In 2XQJ, the movement of the phosphylated Ser198 to His438 allows for a very strong interaction of the Gly115 amide hydrogen backbone to the oxyanion of Glu197 that is maintained throughout the simulation. However, the HID and HIP forms of the inhibited enzyme lack the very strong HisH-Glu197 interactions at His438 that are observed in the HID and WT forms of the inhibited enzyme. In the inhibited forms of

2XQK, the WT and HID forms show similar interactions, but the phosphorus-oxygen distances are now closer for HisH in the WT form, rather than the HID form. The WT form of 2XQK also show mild to strong interactions for the amide hydrogen backbone of

Gly115H to the oxyanion of Glu197.

Distance analysis of the acyl loop revealed differences in the positions of Tyr282,

Pro285, and Tyr332 for the inhibited WT and the positions of His283, Leu285, and

Tyr332 for the inhibited hybrid forms. In the inhibited-WT form of 2XQJ, the hydroxyl group of Tyr282 sits under the acyl-loop and hydrogen bonds to the carbonyl backbone of

Val288, thereby restricting the movement of the acyl loop to the active site. However, the interaction of Tyr282 to the backbone of Val288 is not observed in the 2XQK WT

inhibited form, suggesting that Tyr282 is much more flexible in the inhibited form of the enzyme. The position of Tyr332 is also much more flexible in the inhibited forms of the enzyme, due to the combination of the Leu285 residue as well as the large isopropoxy group of phosphylated Ser198 “pushing” the α-helix of Tyr332, Phe329, and Glu325. In the HIE forms of 2XQK and 2XQJ, Tyr332 enters the active site once more, but His438 does not climb into the active site because of the strong hydrogen bonding interactions between His438 and Glu197, Glu325, and Glu441. For the HIP and HID forms of 2XQJ there is a significant amount of interaction with the hydroxyl group of Ser79 of the - loop, which correlates with the lack of movement of Trp82. However, in the HIP and

HID forms of 2XQK, Tyr332 enters the active site, similar to what happens for the HIE forms of 2XQK and 2XQJ. For the HID and HIP forms of 2XQK and 2XQJ, the His283 mutation sits above the acyl loop near its adjacent residues. For the HIE forms of 2XQK and 2XQJ, the His283 mutation does not sit below the acyl loop, as observed in the HIE form of the native enzyme and lacks any strong hydrogen-bonding interactions even with its adjacent residues.

In order to evaluate the reactivation potential of the inhibited WT and hybrid enzyme, two pathways of spontaneous water hydrolysis were evaluated. The two potential pathways are illustrated in Figure 17, and a number of distances were evaluated to consider either conserved water around the active site residues or potentially catalytic water for reactivation. In pathway #1, it is suggested that if the distance between the oxygen of the closest water to the phosphorus is < 4 Å, the distance between the center of

mass of the hydrogens of the closest water to HisN is < 4 Å, and the distance between

HisH and the oxyanion of Glu197 is < 3 Å, then reactivation is considered possible. In pathway #2, it is suggested that if the distance between the oxygen of the closest water to the phosphorus is < 4 Å and the distance between the center of mass of the hydrogens of the closest water to the oxyanion of Glu197 is < 4 Å, then this route for reactivation is considered possible. The results of this statistical evaluation of pathways #1 and #2 reveal the generally increased potential for water reactivation in the hybrid forms of the enzyme, as shown in Table 7 where the percentage of frames from the MD trajectory has been counted. This is especially true for the HIE form of the mutation for the 2XQK structure where pathway 1 shows potential for reactivation 19% of the time and for pathway 2 shows potential for reactivation 87% of the time. The potential for reactivation correlates quite well with the distance between the backbone amide hydrogens of Gly115H-Glu197 distance as it creates a small pocket that a water molecule can fit in between the phosphorus of phosphylated Ser198 (as shown in Appendix B-5; Supplemental Figures

S20-S22). The reactivity potential can also be amplified by a short distance between

HisH to the oxyanion of Glu197. However, this pathway model is not fully quantitative because the inhibited-WT enzyme of 2XQK shows a larger potential for pathways 1 and

2 than HID.

Figure 17. Proposed pathways for spontaneous hydrolysis by conserved water molecules with sarin-inhibited BChE

Table 7. Reactivation potential percentages for different species for proposed spontaneous reactivation pathways 1 and 2 2XQJ inhibited Pathway 1 (%) Pathway 2 (%) WT 0 2 HID 24 38 HIE 0 69 HIP 0 62 2XQK inhibited Pathway 1 Pathway 2 WT 15 25 HID 3 7 HIE 19 87 HIP 8 51

4.4 Discussion

After observing the unexpected results presented in the corresponding manuscript whereby porcine BChE appears to auto-reactivate following an exposure to sarin

(McGarry et al 2019, in review), we sought to determine the origin of the differences between the porcine BChE and human BChE by synthesizing a human-porcine hybrid

BChE enzyme. It was hypothesized that this hybrid enzyme would be non-immunogenic, allowing for potential use as a therapeutic, while still exhibiting the auto-reactivation capability observed with the porcine BChE. By mutagenizing only four amino acid residues between positions 282 to 285 – two within the aminoacyl binding pocket and two adjacent to it – we were able to engineer a hybrid enzyme that appears to possess the capacity to catalytically degrade sarin.

The four amino acid mutations were Y282N, G283H, T284M, and P285L.

Although other amino acid differences were observed between the porcine and primate sequences, this stretch of four residues distinctly stood out due to its location in and adjacent to the binding pocket and because it was the only region where consecutive amino acid variations were observed. Two of these residues lie within the aminoacyl binding pocket (positions 284 and 285) and the other two are adjacent to the pocket. Of these four residues, only the human enzyme possesses the rigid proline residue at position

285. All other examined species retain a leucine at position 285. Other research groups have studied various residue mutations within the aminoacyl binding pocket as well, specifically looking at P285, and a mutation to this residue alone does not seem to convey 93

the apparent catalytic activity observed here97. As such, one might ask what is different about the interactions with the two residues that lie adjacent to the aminoacyl pocket that allow the spontaneous reactivation to occur. Several hypotheses are discussed below.

Perhaps the simplest explanation for the spontaneous reactivation of the human-porcine

BChE hybrid enzyme is that sarin is not actually covalently bound to the active site serine, but rather sterically occluding the active site. The enzyme then slowly releases the agent, resulting in its natural hydrolysis in the buffer giving the appearance of spontaneous reactivation as the t1/2 of both the reactivation and the non-enzymatic hydrolysis of the agent approximate 2 hrs. However, a transient/noncovalent interaction seems unlikely in light of the results of the inhibition and reactivation experiments.

Specifically, the IC50 value of the human-porcine hybrid enzyme was approximately 23 nM — closer to that of the human wild type BChE IC50 of 5.8 nM than its porcine counterpart (103 nM). Additionally, the hybrid enzyme clusters with the two wild type enzymes, showing relatively high affinity for the nerve agent (Figure 14), whereas the mutations that were introduced previously by Lockridge and colleagues impart a much lower affinity for the agent (~18-45 µM). Furthermore, the fact that 2-PAM Cl appeared to be an effective reactivator of the hybrid enzyme (~15-30% greater than without 2-

PAM Cl; comparable to the ~20% increase observed with human WT BChE) suggests that sarin is covalently bound since it is understood that the oxime reactivates at the catalytic serine via a nucleophilic attack on the methyl phosphonate group of the phosphylated serine residue. Therefore, these inhibition and reactivation data correlate

with the known mechanism of action of the WT enzyme and it seems likely that sarin binds covalently to the active site serine in the hybrid enzyme as well.

The most promising hypothesis is one that is similar to that which has been elucidated with Lockridge’s G117H BChE mutation 39, 40. It is reasonable to speculate that an RNAse A-like mechanism might be the underlying process for the reactivation that is observed. In the human-porcine hybrid BChE, positions 282-285 were mutagenized from YGTP to NHML. The modeling results suggest that these mutations lead to a greater dynamic capacity of the acyl loop which is corroborated by data reported by Brazzolotto and colleagues 94). It is possible that the methionine residue at position

284 perturbs the active site serine such that the histidine at position 283 might activate a water molecule for nucleophilic attack on the bound OP. One could also speculate that the histidine residue at position 283 alone might serve as the nucleophile which reactivates the catalytic serine. Our modeling data suggest that this distance is likely too great for the histidine to catalyze the reaction on its own accord. However, as discussed previously, the mutations in and adjacent to the acyl pocket could lend to a greater flexibility of the loop, thereby accommodating the reactivation event and associated water molecules for the active site pocket 94, 97, 100-102. In particular, Leu285, which is the mutated proline residue from the human isoform, facilitates the structural flexibility that is observed and perhaps leads to enhanced reactivation. The Leu285 residue appears to be more dynamic in the human-porcine hybrid.

Additional experiments and computational modeling, including but not limited to additional mutagenesis and/or molecular docking and QM/MM simulations, will be necessary to confirm the mechanism of action and evaluate the potential of porcine- human-porcine hybrid BChE as a medical countermeasure for OP poisoning. Future efforts could also involve evaluating how commercially available oximes positively or negatively impact the activity and exploring additional molecular simulations to guide subsequent mutagenesis to improve upon the construct created here for greater catalytic or pseudocatalytic activity towards the degradation of chemical warfare nerve agents.

However, the results presented here open the door to a previously unknown, but promising, “next-generation” bioscavenger.

4.5 Chapter Acknowledgements

As mentioned earlier, the above chapter was adapted from the manuscript submitted for publication in Toxicological Sciences. It must be expressly acknowledged that the mass spectrometry experiments, molecular dynamics simulations, and corresponding write-ups were performed and provided by Kristyn Johnson (Battelle) and

Remy Lalisse (Ohio State; graduate student in the laboratory of Dr. Christopher Hadad) respectively. This work would be incomplete without their efforts and their input is greatly appreciated.

Chapter 5. Ongoing and Future BChE Efforts

5.1 BChE Metalloenzymes

As part of multiple R21 grant applications and an Ohio State Center for Applied

Plant Sciences (CAPS) SEED proposal (which was funded), we had set out to engineer a

BChE enzyme that could catalytically hydrolyze OP compounds rather than simply scavenge them. As described in Chapter 4, we accomplished this in collaboration with the

Battelle Memorial Institute. However, another approach that was quite innovative was also pursued and, if possible, this research should continue after this dissertation is finalized. This approach was rooted in the desire to transplant components of the active site of a bacterial organophosphohydrolase (OPH) metalloenzyme into human BChE. The hypothesis of this approach was such that insertion of the metal ion would enhance the nucleophilicity of coordinating water molecules and enable catalytic hydrolysis of OPs in a manner similar to that of naturally occurring OPH. As described earlier (§1.7) the bacterial OPH enzyme active site contains a zinc ion that serves to coordinate a bridging nucleophilic water molecule for catalysis123, 124. As a result, this metalloenzyme is highly processive toward OP compounds. The engineered metal-containing BChE enzyme

would be human-like, with minimal potential for immunogenicity in patients, but would have the ability to effectively degrade OPs in circulation.

To begin the process of rationally engineering a BChE-OPH hybrid metalloenzyme, we initiated a collaboration with Dr. Hannah Shafaat’s laboratory (The

Ohio State University). The (His)2(carboxylato) and (His)3 motifs are common in peptidases and feature an open coordination site for water, which is expected to deprotonate upon binding and replace the histidine nucleophile in the proposed G117H

RNAse A-like mechanism for OP hydrolysis. For development of the novel metalloenzymes, a number of mutations were modelled and designed by Dr. Shafaat to introduce a metal binding site (Figure 18). Initial attempts aimed to reproduce the structure of an engineered, mononuclear OPH site (Figure 18A). The high binding affinity of BChE for its native substrate (Figure 18B) and for OPs is, in part, attributed to shape, volume, and charge complementarity within the substrate gorge. By keeping the residues lining the substrate-binding pocket mostly invariant, it was hypothesized that the high affinity should be retained. For the first mutant BChE, the histidine and glutamate residues from the catalytic triad were retained, and a bulky phenylalanine residue in close spatial proximity (Phe398) was mutated to a histidine residue (Figure 18C). In the second approach, the metal binding site was designed such that it would be situated further within the hydrophobic cleft of the enzyme by introducing a S198H mutation, with the intent of obtaining greater overlap between the phosphorous center and the activated water (Figure 18D). We recognized that this mutation would likely abolish native BChE

function, which it did, but understood that native function is not required for a therapeutic effect. A third approach utilized the G117H mutation along with the F398H mutation as metal-binding ligands. This site is situated higher in the substrate-binding cleft and should prevent OPs from reaching the catalytic triad (Figure 18E). Molecular modeling suggested the geometry of these metal-bound variants would support organophosphate hydrolysis through an RNAse A-like mechanism for OP hydrolysis, as seen in the G117H mutant.

Each of the mutations suggested by Dr. Shafaat were successfully created and transiently expressed in HEK293E cells as described previously. As mentioned, the

S198H mutation abolished all enzymatic activity. All other mutations severely adversely affected native enzyme activity as well, however, the G117H/F398H double mutant maintained approximately 10-20% the level of activity as compared to the WT enzyme.

Additionally, as the goal of these experiments was to generate a metal binding pocket within the enzyme’s active site, it was necessary to evaluate various buffers for their effects on enzymatic activity. PBS, MOPS, Tris, and HEPES were all evaluated with none of the buffers causing reduced activity of WT BChE. Therefore, following the guidance of Tasha Manesis and Dr. Shafaat, we decided to dialyze each of the purified proteins against 50 mM MOPS, pH 8.0 prior to any metal incorporation experiment to ensure the removal of all imidazole. Despite these efforts, we were unable to observe the incorporation of the divalent metal (neither Co2+ nor Ni2+) into any of the mutant BChEs that were created. The reason behind this, however, likely stems from the relatively low

Figure 18. Various OPH and BChE model protein structures. (A) Engineered OPH based on adenosine deaminase (PDB ID 3T1G). (B) Active site of BChE (PDB ID 1P0P) showing the crystal structure-bound butyrylcholine (blue) and butyrylcholine docked using AutoDock Vina (red). (C) – (E) Models of Zn-binding sites in BChE based on PDB structure 1P0P and a model of sarin docked (red). (C) F329D/F398H variant; (D) S198H/F398H variant; (E) G117H/F398H variant. Zn2+ ion shown as white sphere.

100

overall protein concentration and not necessarily the inability of the enzyme to incorporate the metal.

Despite our inability to observe the incorporation of a divalent metal via spectroscopy, we attempted to assess activity. To do this, BChE was transiently expressed in HEK293E using 12-well plates. Supernatant was harvested and incubated at room temperature in 50 mM MOPS buffer, pH 7.0 with or without 10 mM ZnCl2 for an additional 72 hours. The enzymes were diluted 3-fold into their approropriate MOPS buffers and 20 µL was added to the test plate (30-fold final in 200 µL final volume). The plate was pre-warmed to 37°C and varying concentrations of the OP, paraxon, was added.

Substrate (3 mM BTC) and indicator (0.5 mM DTNB) was added and kinteic activity was assessed spectrophotemetrically by obtaining a reading every 15-30 seconds at 412 nM.

Active enzyme units/mL was calculated using Beer’s Law as previously described.

Interestingly, while the presence of the double mutant resulted in ~10% of the activity of the WT enzyme, and the presence of Zn2+ reduced enzymatic activity of both constructs

~2-fold, the G117H/F398H variant appears to be resistant to inhibition by paraxon, even at high (3 mM) concentrations as compared to the WT enzyme (Table 8). These results suggest that OPH activity may be observed in the mutant BChE, or at the very least, a decrease in binding/inhibition of the OP is observed. In light of the spectroscopy and paraoxon-challenge results, once greater protein concentrations can be obtained for all mutants through a stable cell line, or high expressing transient transfection, it is recommended that each of these experiments be revisited.

101

Table 8. Wild-Type BChE and a BChE Metalloenzyme vs. Paraxon in the presence of Zn2+ [Paraoxon], M 0.00E+00 2.86E-09 1.14E-08 4.58E-08 1.83E-07 7.32E-07 2.93E-06 1.17E-05 4.69E-05 1.88E-04 7.50E-04 3.00E-03

WT (-Zn) 100% 4% 6% 4% 3% 3% 2% 3% 3% 3% 3% 7%

WT (+Zn) 100% 5% 6% 4% 3% 2% 2% 3% 2% 2% 3% 5% G117H/F398H 100% 62% 60% 58% 55% 58% 53% 54% 54% 59% 55% 54% (-Zn) G117H/F398H 100% 35% 31% 24% 27% 27% 26% 27% 30% 30% 34% 129% (+Zn)

102

5.2 An Intein:BChE Fusion Protein

The purification of large amounts of BChE is a pressing concern if one wants to pursue BChE as a therapeutic. Its current clinical trials for use as a stoichiometric bioscavenger requires the purification of WT human BChE using procainamide (a cocaine analog) resin from expired human plasma or the Cohn IV fraction of expired human plasma20, 34. It must be noted, that procainamide resin is not effective for the purification of various mutagenized BChE constructs – specifically those nearest the acyl binding pocket of the enzyme. This was unfortunately true for the various mutations created as part of the work described here as well as only the WT and G117H mutations were capable of being purified using the procainamide resin. Alternatively, a new affinity resin, Hupresin®, was recently developed by Emilie David (Chemforase) and has been evaluated for the purification of recombinant BChE125, 126. It must be noted, however, that some mutations may limit the ability of Hupresin® to bind BChE (personal communication with Dr. Emilie David) and, as such, each mutant must be assessed on a case-by-case basis. In collaboration with Dr. Christopher Hadad and Remy Lalisse, evaluation of the potential of utilizing Hupresin® for the purification of our mutant

BChEs using MD simulations has begun.

As a stoichiometric scavenger, in order to obtain the equivalent of one human protective dose, large amounts of BChE must be administered, requiring up to 800 liters of human donor blood be processed to prophylactically treat a single adult for a potential

103

VX exposure. These activity and production issues must be addressed before BChE can ever be used therapeutically in a cost effective manner.

To address the glaring issue of the production and purification of BChE, multiple groups have implemented various approaches as outlined throughout this document. The purpose of this subchapter, however, is to discuss the introductory work our group has begun, in an attempt to develop a novel purification system for recombinant human

BChE.

The novelty of the system we are investigating for use with BChE is one that Dr.

Wood established in the late 1990s in the utilization of inteins for the purification of target proteins. Similar to DNA introns, inteins may be defined as self-splicing proteins that occur as in-frame insertions in specific host proteins. During the auto-splicing reaction, an intein excises itself from the precursor protein, while the exteins (flanking regions), join, thereby restoring host gene function127, 128. The beauty of utilizing an intein for the purification of target proteins is that the protein of interest can ultimately be purified without an affinity tag – something that. to date, has not been possible for the mutagenized BChEs.

While Dr. Wood’s work initially focused on the Mycobacterium tuberculosis

(Mtu) recA intein, work in the lab has since shifted to the utilization of Nostoc punctiforme (Npu) DnaE split intein for the on-column purification of fusion proteins129-

131. As described by Cooper et al:

104

The key to the technology is a short, self-cleaving tag that is added

to the N-terminus of the target protein, which is captured via high and

specific affinity to a provided resin. Once the target protein is captured

and purified, a small pH shift triggers self-cleavage of the tag, thus

releasing the tagless target from the capture resin. Importantly, this

cleavage reaction does not require specific residues outside the intein for

activity; therefore, the tagless target protein eluted from the column can

also be traceless—leaving no extraneous residues apart from the native

target.131

This purification scheme is outlined below in Figure 19 and is reproduced in its entirety from Cooper et al.131

105

Figure 19. Process workflow for a split-intein affinity capture and purification. (A) Intein-tagged target protein (POI) precursor is expressed and harvested. (B) The tagged target protein is selectively immobilized onto the capture resin by spontaneous association of the intein fragments. (C) Additional buffer (pH 8.5) is applied to the column to wash host-cell contaminant proteins away from the immobilized target. (D) pH 6.2 buffer is added to induce the cleaving reaction. (E) Column is capped and left to incubate in pH 6.2 buffer for a period of hours. During this time, the intein will cleave the target protein at the NpuC *-POI junction, releasing the target protein into the mobile phase. (F) A small amount additional pH 6.2 elution buffer is applied to the column to elute the liberated target protein at high purity.

106

One caveat of the approach described above is due to the fact that the split intein is in fusion with the N-terminal region of the target protein. This is especially relevant for mammalian proteins which oftentimes require post-translational modifications (i.e. glycosylations in the case of BChE) for proper folding and secretion. The leader sequence drives these post-translational modifications of the immature protein and is ultimately removed during the final maturation process for a given protein. Therefore, when creating a fusion of the split intein to the N-terminus of the target protein, difficulties may arise as a result of the non-native sequence.

In an initial attempt to determine the capacity of BChE for change, the native leader sequence was removed and replaced with the consensus serum albumin (SA) leader sequence. The protein was expressed and purified using IMAC. As observed in

Figure 20 below, the replacement of the native leader sequence with the SA leader sequence had minimal effect on the expression and secretion of the protein. It should be noted that BChE appears to be properly glycosylated as well. If the BChE was improperly or incompletely glycosylated, the 85 kDa band which is observed would appear smaller on the gel as the predicted molecular weight of the non-glycosylated

BChE is 68 kDa. Perhaps most interestingly also was the fact that the enzymatic activity of the two proteins was comparable, indicating that the SA leader sequence was sufficient for both proper glycosylation and proper folding of the protein. With these results, it was determined that the BChE could tolerate additional, non-native mutations at the N-

HN terminus. The next step of this process is to express SA-NpuC -mBChE-His6 in the

107

mammalian cell culture. Work is currently ongoing on this effort to create an intein-

BChE fusion. It is my hope that this work will continue after I have graduated and that a novel expression and purification system for the production of therapeutically relevant

BChEs is achieved.

Figure 20. Silver stain of HEK293E expression of WT-BChE and SA-mBChE-His6. 20 mL of HEK293E cells was transiently transfected using pRcCMV-BChE (WT)-His6 or pRcCMV-SA-mBChE-His6 and purified using IMAC. Samples were analyzed using an 8% polyacrylamide gel under denaturing conditions and subsequently silver stained for visualization.

108

Bibliography

[1] Eddleston, M., Buckley, N. A., Eyer, P., and Dawson, A. H. (2008) Management of acute organophosphorus pesticide poisoning, Lancet 371, 597-607. [2] Eddleston, M., and Phillips, M. R. (2004) Self poisoning with pesticides, BMJ 328, 42-44. [3] Yokoyama, K., Araki, S., Murata, K., Nishikitani, M., Okumura, T., Ishimatsu, S., and Takasu, N. (1998) A preliminary study on delayed vestibulo-cerebellar effects of Tokyo Subway Sarin Poisoning in relation to gender difference: frequency analysis of postural sway, Journal of occupational and environmental medicine / American College of Occupational and Environmental Medicine 40, 17-21. [4] Lee, E. C. (2003) Clinical manifestations of sarin nerve gas exposure, JAMA : the journal of the American Medical Association 290, 659-662. [5] Dolgin, E. (2013) Syrian gas attack reinforces need for better anti-sarin drugs, Nature medicine 19, 1194-1195. [6] Chai, P. R., Boyer, E. W., Al-Nahhas, H., and Erickson, T. B. (2017) Toxic chemical weapons of assassination and warfare: nerve agents VX and sarin, Toxicol Commun 1, 21-23. [7] Marrs, T. C. (1993) Organophosphate poisoning, Pharmacology & therapeutics 58, 51-66. [8] Chambers, H. W. (1992) Organophosphorous Compounds: An Overview, In Organophosphates: Chemistry, Fate, and Effects (Chambers, J. E., and Levi, P. E., Eds.), pp 3-17, Academic Press, Inc., San Diego, CA USA. [9] Wilson, B. W., Hooper, M. J., Hansen, M. E., and Nieberg, P. S. (1992) Reavtivation of Organophosphorous Inhibited AChE with Oximes, In Organophosphates: Chemistry, Fate, and Effects (Chambers, J. E., and Levi, P. E., Eds.), pp 107-137, Academic Press, Inc., San Diego, CA USA. [10] Eto, M. (1974) Organophosphorous Pesticides: Organic and Biological Chemistry, CRC Press, Cleveland, OH USA. [11] Katzung, B. G. (2001) Introduction to autonomic pharmacology, In Basic and clinical pharmacology 8 ed., pp 75-91, The McGraw Hill Companies, Inc., U.S.A. [12] Worek, F., Thiermann, H., Szinicz, L., and Eyer, P. (2004) Kinetic analysis of interactions between human acetylcholinesterase, structurally different organophosphorus compounds and oximes, Biochemical pharmacology 68, 2237- 2248. 109

[13] Sussman, J. L., Harel, M., Frolow, F., Oefner, C., Goldman, A., Toker, L., and Silman, I. (1991) Atomic structure of acetylcholinesterase from Torpedo californica: a prototypic acetylcholine-binding protein, Science 253, 872-879. [14] Pezzementi, L., Nachon, F., and Chatonnet, A. (2011) Evolution of acetylcholinesterase and butyrylcholinesterase in the vertebrates: an atypical butyrylcholinesterase from the Medaka Oryzias latipes, PloS one 6, e17396. [15] Nicolet, Y., Lockridge, O., Masson, P., Fontecilla-Camps, J. C., and Nachon, F. (2003) Crystal structure of human butyrylcholinesterase and of its complexes with substrate and products, The Journal of biological chemistry 278, 41141-41147. [16] Soreq, H., and Zakut, H. (1993) Human Cholinesterases and Anticholinesterases, Academic Press, Inc., San Diego, CA. [17] Nachon, F., Nicolet, Y., Viguie, N., Masson, P., Fontecilla-Camps, J. C., and Lockridge, O. (2002) Engineering of a monomeric and low-glycosylated form of human butyrylcholinesterase: expression, purification, characterization and crystallization, European journal of biochemistry / FEBS 269, 630-637. [18] Lockridge, O., Bartels, C. F., Vaughan, T. A., Wong, C. K., Norton, S. E., and Johnson, L. L. (1987) Complete amino acid sequence of human serum cholinesterase, The Journal of biological chemistry 262, 549-557. [19] Lockridge, O., Eckerson, H. W., and La Du, B. N. (1979) Interchain disulfide bonds and subunit organization in human serum cholinesterase, The Journal of biological chemistry 254, 8324-8330. [20] Saxena, A., Luo, C., and Doctor, B. P. (2008) Developing procedures for the large- scale purification of human serum butyrylcholinesterase, Protein expression and purification 61, 191-196. [21] Kolarich, D., Weber, A., Pabst, M., Stadlmann, J., Teschner, W., Ehrlich, H., Schwarz, H. P., and Altmann, F. (2008) Glycoproteomic characterization of butyrylcholinesterase from human plasma, Proteomics 8, 254-263. [22] Mumford, H., Price, M. E., Cerasoli, D. M., Teschner, W., Ehrlich, H., Schwarz, H. P., and Lenz, D. E. (2010) Efficacy and physiological effects of human butyrylcholinesterase as a post-exposure therapy against percutaneous poisoning by VX in the guinea-pig, Chemico-biological interactions 187, 304-308. [23] Ilyushin, D. G., Smirnov, I. V., Belogurov, A. A., Jr., Dyachenko, I. A., Zharmukhamedova, T., Novozhilova, T. I., Bychikhin, E. A., Serebryakova, M. V., Kharybin, O. N., Murashev, A. N., Anikienko, K. A., Nikolaev, E. N., Ponomarenko, N. A., Genkin, D. D., Blackburn, G. M., Masson, P., and Gabibov, A. G. (2013) Chemical polysialylation of human recombinant butyrylcholinesterase delivers a long-acting bioscavenger for nerve agents in vivo, Proceedings of the National Academy of Sciences of the United States of America 110, 1243-1248. [24] Vellom, D. C., Radic, Z., Li, Y., Pickering, N. A., Camp, S., and Taylor, P. (1993) Amino acid residues controlling acetylcholinesterase and butyrylcholinesterase specificity, Biochemistry 32, 12-17. 110

[25] Brazzolotto, X., Wandhammer, M., Ronco, C., Trovaslet, M., Jean, L., Lockridge, O., Renard, P. Y., and Nachon, F. (2012) Human butyrylcholinesterase produced in insect cells: huprine-based affinity purification and crystal structure, The FEBS journal 279, 2905-2916. [26] Darvesh, S., Hopkins, D. A., and Geula, C. (2003) Neurobiology of butyrylcholinesterase, Nature reviews. Neuroscience 4, 131-138. [27] Evans, F. T., Gray, P. W., Lehmann, H., and Silk, E. (1952) Sensitivity to succinylcholine in relation to serum-cholinesterase, Lancet 1, 1229-1230. [28] Kalow, W., and Staron, N. (1957) On distribution and inheritance of atypical forms of human serum cholinesterase, as indicated by dibucaine numbers, Can J Biochem Physiol 35, 1305-1320. [29] Lehmann, H., and Ryan, E. (1956) The familial incidence of low pseudocholinesterase level, Lancet 271, 124. [30] Lockridge, O., Duysen, E., and Masson, P. (2010) Butyrylcholinesterase: Overview, Structure, and Function, In Anticholinesterase Pesticides: Metabolism, Neurotoxicity, and Epidemiology (Satoh, T., and Gupta, R., Eds.), pp 25-41, John Wiley and Sons, Inc., Hoboken, N.J. [31] Lockridge, O. (2015) Review of human butyrylcholinesterase structure, function, genetic variants, history of use in the clinic, and potential therapeutic uses, Pharmacology & therapeutics 148, 34-46. [32] Masson, P., and Lockridge, O. (2010) Butyrylcholinesterase for protection from organophosphorus poisons: catalytic complexities and hysteretic behavior, Archives of biochemistry and biophysics 494, 107-120. [33] Allon, N., Raveh, L., Gilat, E., Cohen, E., Grunwald, J., and Ashani, Y. (1998) Prophylaxis against soman inhalation toxicity in guinea pigs by pretreatment alone with human serum butyrylcholinesterase, Toxicological sciences : an official journal of the Society of Toxicology 43, 121-128. [34] Raveh, L., Grauer, E., Grunwald, J., Cohen, E., and Ashani, Y. (1997) The stoichiometry of protection against soman and VX toxicity in monkeys pretreated with human butyrylcholinesterase, Toxicology and applied pharmacology 145, 43-53. [35] Mumford, H., M, E. P., Lenz, D. E., and Cerasoli, D. M. (2011) Post-exposure therapy with human butyrylcholinesterase following percutaneous VX challenge in guinea pigs, Clin Toxicol (Phila) 49, 287-297. [36] Nachon, F., Carletti, E., Wandhammer, M., Nicolet, Y., Schopfer, L. M., Masson, P., and Lockridge, O. (2011) X-ray crystallographic snapshots of reaction intermediates in the G117H mutant of human butyrylcholinesterase, a nerve agent target engineered into a catalytic bioscavenger, The Biochemical journal 434, 73- 82. [37] Raveh, L., Grunwald, J., Marcus, D., Papier, Y., Cohen, E., and Ashani, Y. (1993) Human butyrylcholinesterase as a general prophylactic antidote for nerve agent

111

toxicity. In vitro and in vivo quantitative characterization, Biochemical pharmacology 45, 2465-2474. [38] Wang, Y., Boeck, A. T., Duysen, E. G., Van Keuren, M., Saunders, T. L., and Lockridge, O. (2004) Resistance to organophosphorus agent toxicity in transgenic mice expressing the G117H mutant of human butyrylcholinesterase, Toxicology and applied pharmacology 196, 356-366. [39] Lockridge, O., Blong, R. M., Masson, P., Froment, M. T., Millard, C. B., and Broomfield, C. A. (1997) A single amino acid substitution, Gly117His, confers phosphotriesterase (organophosphorus acid anhydride hydrolase) activity on human butyrylcholinesterase, Biochemistry 36, 786-795. [40] Vyas, S., Beck, J. M., Xia, S., Zhang, J., and Hadad, C. M. (2010) Butyrylcholinesterase and G116H, G116S, G117H, G117N, E197Q and G117H/E197Q mutants: a molecular dynamics study, Chemico-biological interactions 187, 241-245. [41] Hawwa, R., Larsen, S. D., Ratia, K., and Mesecar, A. D. (2009) Structure-based and random mutagenesis approaches increase the organophosphate-degrading activity of a phosphotriesterase homologue from Deinococcus radiodurans, Journal of molecular biology 393, 36-57. [42] Poyot, T., Nachon, F., Froment, M. T., Loiodice, M., Wieseler, S., Schopfer, L. M., Lockridge, O., and Masson, P. (2006) Mutant of Bungarus fasciatus acetylcholinesterase with low affinity and low hydrolase activity toward organophosphorus esters, Biochimica et biophysica acta 1764, 1470-1478. [43] diSioudi, B., Grimsley, J. K., Lai, K., and Wild, J. R. (1999) Modification of near active site residues in organophosphorus hydrolase reduces metal stoichiometry and alters substrate specificity, Biochemistry 38, 2866-2872. [44] Khare, S. D., Kipnis, Y., Greisen, P., Jr., Takeuchi, R., Ashani, Y., Goldsmith, M., Song, Y., Gallaher, J. L., Silman, I., Leader, H., Sussman, J. L., Stoddard, B. L., Tawfik, D. S., and Baker, D. (2012) Computational redesign of a mononuclear zinc metalloenzyme for organophosphate hydrolysis, Nature chemical biology 8, 294-300. [45] Burns, C. J., McIntosh, L. J., Mink, P. J., Jurek, A. M., and Li, A. A. (2013) Pesticide Exposure and Neurodevelopmental Outcomes: Review of the Epidemiologic and Animal Studies (vol 16, pg 127, 2013), J Toxicol Env Heal B 16, 395-398. [46] Doctor, B. P., and Saxena, A. (2005) Bioscavengers for the protection of humans against organophosphate toxicity, Chemico-biological interactions 157-158, 167- 171. [47] Ellman, G. L., Courtney, K. D., Andres, V., Jr., and Feather-Stone, R. M. (1961) A new and rapid colorimetric determination of acetylcholinesterase activity, Biochemical pharmacology 7, 88-95. [48] McGarry, K. G., Bartlett, R. A., Machesky, N. J., Snider, T. H., Moyer, R. A., Yeung, D. T., and Brittain, M. K. (2013) Evaluation of HemogloBind treatment 112

for preparation of samples for cholinesterase analysis, Adv Biosci Biotechnol 4, 1020-1023. [49] Bessette, P. H., Aslund, F., Beckwith, J., and Georgiou, G. (1999) Efficient folding of proteins with multiple disulfide bonds in the Escherichia coli cytoplasm, Proceedings of the National Academy of Sciences of the United States of America 96, 13703-13708. [50] Levy, R., Weiss, R., Chen, G., Iverson, B. L., and Georgiou, G. (2001) Production of correctly folded Fab antibody fragment in the cytoplasm of Escherichia coli trxB gor mutants via the coexpression of molecular chaperones, Protein expression and purification 23, 338-347. [51] Chen, J., Song, J. L., Zhang, S., Wang, Y., Cui, D. F., and Wang, C. C. (1999) Chaperone activity of DsbC, The Journal of biological chemistry 274, 19601- 19605. [52] Durocher, Y., Perret, S., and Kamen, A. (2002) High-level and high-throughput recombinant protein production by transient transfection of suspension-growing human 293-EBNA1 cells, Nucleic acids research 30, E9. [53] Li, B., Stribley, J. A., Ticu, A., Xie, W., Schopfer, L. M., Hammond, P., Brimijoin, S., Hinrichs, S. H., and Lockridge, O. (2000) Abundant tissue butyrylcholinesterase and its possible function in the acetylcholinesterase knockout mouse, Journal of neurochemistry 75, 1320-1331. [54] Larrimore, K. E., Barcus, M., Kannan, L., Gao, Y., Zhan, C. G., Brimijoin, S., and Mor, T. (2013) Plants as a source of butyrylcholinesterase variants designed for enhanced cocaine hydrolase activity, Chemico-biological interactions 203, 217- 220. [55] Barghchi, M. (1995) High-frequency and efficient Agrobacterium-mediated transformation of Arabidopsis thaliana ecotypes "C24" and "Landsberg erecta" using Agrobacterium tumefaciens, Methods Mol Biol 44, 135-147. [56] Krenek, P., Samajova, O., Luptovciak, I., Doskocilova, A., Komis, G., and Samaj, J. (2015) Transient plant transformation mediated by Agrobacterium tumefaciens: Principles, methods and applications, Biotechnol Adv 33, 1024-1042. [57] Kindle, K. L. (1990) High-frequency nuclear transformation of Chlamydomonas reinhardtii, Proceedings of the National Academy of Sciences of the United States of America 87, 1228-1232. [58] Cannard, K. (2006) The acute treatment of nerve agent exposure, Journal of the neurological sciences 249, 86-94. [59] Eyer, P. (2003) The role of oximes in the management of organophosphorus pesticide poisoning, Toxicological reviews 22, 165-190. [60] Thiermann, H., Worek, F., and Kehe, K. (2013) Limitations and challenges in treatment of acute chemical warfare agent poisoning, Chemico-biological interactions 206, 435-443.

113

[61] Manoharan, I., Boopathy, R., Darvesh, S., and Lockridge, O. (2007) A medical health report on individuals with silent butyrylcholinesterase in the Vysya community of India, Clin Chim Acta 378, 128-135. [62] Li, B., Duysen, E. G., Carlson, M., and Lockridge, O. (2008) The butyrylcholinesterase knockout mouse as a model for human butyrylcholinesterase deficiency, The Journal of pharmacology and experimental therapeutics 324, 1146-1154. [63] Reed, B. A., Sabourin, C. L., and Lenz, D. E. (2017) Human butyrylcholinesterase efficacy against nerve agent exposure, J Biochem Mol Toxicol 31. [64] Lorke, D. E., Kalasz, H., Petroianu, G. A., and Tekes, K. (2008) Entry of oximes into the brain: a review, Current medicinal chemistry 15, 743-753. [65] Worek, F., Thiermann, H., and Wille, T. (2016) Oximes in organophosphate poisoning: 60 years of hope and despair, Chemico-biological interactions 259, 93- 98. [66] Snoy, P. J. (2010) Establishing efficacy of human products using animals: the US food and drug administration's "animal rule", Vet Pathol 47, 774-778. [67] FDA. (2015) Product Development Under the Animal Rule Guidance for Industry, (Services, U. S. D. o. H. a. H., Ed.). [68] Langston, J. L., and Myers, T. M. (2016) VX toxicity in the Gottingen minipig, Toxicology letters 264, 12-19. [69] Murray, D. B., Eddleston, M., Thomas, S., Jefferson, R. D., Thompson, A., Dunn, M., Vidler, D. S., Clutton, R. E., and Blain, P. G. (2012) Rapid and complete bioavailability of antidotes for organophosphorus nerve agent and cyanide poisoning in minipigs after intraosseous administration, Annals of emergency medicine 60, 424-430. [70] Saxena, A., Sun, W., Dabisch, P. A., Hulet, S. W., Hastings, N. B., Jakubowski, E. M., Mioduszewski, R. J., and Doctor, B. P. (2011) Pretreatment with human serum butyrylcholinesterase alone prevents cardiac abnormalities, seizures, and death in Gottingen minipigs exposed to sarin vapor, Biochemical pharmacology 82, 1984-1993. [71] Worek, F., and Thiermann, H. (2013) The value of novel oximes for treatment of poisoning by organophosphorus compounds, Pharmacology & therapeutics 139, 249-259. [72] Luo, C., Tong, M., Maxwell, D. M., and Saxena, A. (2008) Comparison of oxime reactivation and aging of nerve agent-inhibited monkey and human acetylcholinesterases, Chemico-biological interactions 175, 261-266. [73] Aurbek, N., Thiermann, H., Szinicz, L., Eyer, P., and Worek, F. (2006) Analysis of inhibition, reactivation and aging kinetics of highly toxic organophosphorus compounds with human and pig acetylcholinesterase, Toxicology 224, 91-99. [74] Worek, F., Aurbek, N., Wetherell, J., Pearce, P., Mann, T., and Thiermann, H. (2008) Inhibition, reactivation and aging kinetics of highly toxic

114

organophosphorus compounds: pig versus minipig acetylcholinesterase, Toxicology 244, 35-41. [75] Worek, F., Reiter, G., Eyer, P., and Szinicz, L. (2002) Reactivation kinetics of acetylcholinesterase from different species inhibited by highly toxic organophosphates, Archives of toxicology 76, 523-529. [76] Nachon, F., Carletti, E., Worek, F., and Masson, P. (2010) Aging mechanism of butyrylcholinesterase inhibited by an N-methyl analogue of tabun: implications of the trigonal-bipyramidal transition state rearrangement for the phosphylation or reactivation of cholinesterases, Chemico-biological interactions 187, 44-48. [77] Dorandeu, F., Mikler, J. R., Thiermann, H., Tenn, C., Davidson, C., Sawyer, T. W., Lallement, G., and Worek, F. (2007) Swine models in the design of more effective medical countermeasures against organophosphorus poisoning, Toxicology 233, 128-144. [78] Bartek, M. J., LaBudde, J. A., and Maibach, H. I. (1972) Skin permeability in vivo: comparison in rat, rabbit, pig and man, J Invest Dermatol 58, 114-123. [79] Simon, G. A., and Maibach, H. I. (2000) The pig as an experimental animal model of percutaneous permeation in man: qualitative and quantitative observations--an overview, Skin Pharmacol Appl Skin Physiol 13, 229-234. [80] Taysse, L., Daulon, S., Delamanche, S., Bellier, B., and Breton, P. (2007) Skin decontamination of mustards and organophosphates: comparative efficiency of RSDL and Fuller's earth in domestic swine, Human & experimental toxicology 26, 135-141. [81] Bjarnason, S., Mikler, J., Hill, I., Tenn, C., Garrett, M., Caddy, N., and Sawyer, T. W. (2008) Comparison of selected skin decontaminant products and regimens against VX in domestic swine, Human & experimental toxicology 27, 253-261. [82] Qiao, G. L., and Riviere, J. E. (1995) Significant effects of application site and occlusion on the pharmacokinetics of cutaneous penetration and biotransformation of parathion in vivo in swine, Journal of pharmaceutical sciences 84, 425-432. [83] Baynes, R. E., Halling, K. B., and Riviere, J. E. (1997) The influence of diethyl-m- toluamide (DEET) on the percutaneous absorption of permethrin and carbaryl, Toxicology and applied pharmacology 144, 332-339. [84] Hulet, S. W., Sommerville, D. R., Crosier, R. B., Dabisch, P. A., Miller, D. B., Benton, B. J., Forster, J. S., Scotto, J. A., Jarvis, J. R., Krauthauser, C., Muse, W. T., Reutter, S. A., Mioduszewski, R. J., and Thomson, S. A. (2006) Comparison of low-level sarin and cyclosarin vapor exposure on pupil size of the Gottingen minipig: effects of exposure concentration and duration, Inhalation toxicology 18, 143-153. [85] Hulet, S. W., Sommerville, D. R., Miller, D. B., Scotto, J. A., Muse, W. T., and Burnett, D. C. (2014) Comparison of sarin and cyclosarin toxicity by subcutaneous, intravenous and inhalation exposure in Gottingen minipigs, Inhalation toxicology 26, 175-184. 115

[86] Hill, S. L., Thomas, S. H., Flecknell, P. A., Thomas, A. A., Morris, C. M., Henderson, D., Dunn, M., and Blain, P. G. (2015) Rapid and equivalent systemic bioavailability of the antidotes HI-6 and dicobalt edetate via the intraosseous and intravenous routes, Emergency medicine journal : EMJ 32, 626-631. [87] Bahar, F. G., Ohura, K., Ogihara, T., and Imai, T. (2012) Species difference of esterase expression and hydrolase activity in plasma, Journal of pharmaceutical sciences 101, 3979-3988. [88] Saxena, A., Belinskaya, T., Schopfer, L. M., and Lockridge, O. (2018) Characterization of butyrylcholinesterase from porcine milk, Arch Biochem Biophys 652, 38-49. [89] Herkert, N. M., Lallement, G., Clarencon, D., Thiermann, H., and Worek, F. (2009) Comparison of the oxime-induced reactivation of rhesus monkey, swine and guinea pig erythrocyte acetylcholinesterase following inhibition by sarin or paraoxon, using a perfusion model for the real-time determination of membrane- bound acetylcholinesterase activity, Toxicology 258, 79-83. [90] Worek, F., Wille, T., Aurbek, N., Eyer, P., and Thiermann, H. (2010) Reactivation of organophosphate-inhibited human, Cynomolgus monkey, swine and guinea pig acetylcholinesterase by MMB-4: a modified kinetic approach, Toxicology and applied pharmacology 249, 231-237. [91] Bartling, A., Worek, F., Szinicz, L., and Thiermann, H. (2007) Enzyme-kinetic investigation of different sarin analogues reacting with human acetylcholinesterase and butyrylcholinesterase, Toxicology 233, 166-172. [92] Aurbek, N., Thiermann, H., Eyer, F., Eyer, P., and Worek, F. (2009) Suitability of human butyrylcholinesterase as therapeutic marker and pseudo catalytic scavenger in organophosphate poisoning: a kinetic analysis, Toxicology 259, 133- 139. [93] Worek, F., Eyer, P., and Thiermann, H. (2012) Determination of acetylcholinesterase activity by the Ellman assay: a versatile tool for in vitro research on medical countermeasures against organophosphate poisoning, Drug testing and analysis 4, 282-291. [94] Brazzolotto, X., Froment, M. T., Gessay, F., Worek, F., Dorandeu, F., and Nachon, F. (2015) Biochemical and structural study of a self-reactivating butyrylcholinesterase after V-type nerve agent inhibition, In 12th International Meeting of Cholinesterases - 6th-Paraoxonase Conference, Elche, Spain. [95] Franjesevic, A. J., Sillart, S. B., Beck, J. M., Vyas, S., Callam, C. S., and Hadad, C. M. (2018) Resurrection and Reactivation of Acetylcholinesterase and Butyrylcholinesterase, Chemistry. [96] Mumford, H., Docx, C. J., Price, M. E., Green, A. C., Tattersall, J. E., and Armstrong, S. J. (2013) Human plasma-derived BuChE as a stoichiometric bioscavenger for treatment of nerve agent poisoning, Chemico-biological interactions 203, 160-166.

116

[97] Lushchekina, S. V., Schopfer, L. M., Grigorenko, B. L., Nemukhin, A. V., Varfolomeev, S. D., Lockridge, O., and Masson, P. (2018) Optimization of Cholinesterase-Based Catalytic Bioscavengers Against Organophosphorus Agents, Frontiers in pharmacology 9, 211. [98] Nachon, F., Brazzolotto, X., Trovaslet, M., and Masson, P. (2013) Progress in the development of enzyme-based nerve agent bioscavengers, Chemico-biological interactions 206, 536-544. [99] Rice, H., Mann, T. M., Armstrong, S. J., Price, M. E., Green, A. C., and Tattersall, J. E. H. (2016) The potential role of bioscavenger in the medical management of nerve-agent poisoned casualties, Chemico-biological interactions 259, 175-181. [100] Dafferner, A. J., Lushchekina, S., Masson, P., Xiao, G., Schopfer, L. M., and Lockridge, O. (2017) Characterization of butyrylcholinesterase in bovine serum, Chemico-biological interactions 266, 17-27. [101] Dorandeu, F., Foquin, A., Briot, R., Delacour, C., Denis, J., Alonso, A., Froment, M. T., Renault, F., Lallement, G., and Masson, P. (2008) An unexpected plasma cholinesterase activity rebound after challenge with a high dose of the nerve agent VX, Toxicology 248, 151-157. [102] Terekhov, S. S., Smirnov, I. V., Stepanova, A. V., Bobik, T. V., Mokrushina, Y. A., Ponomarenko, N. A., Belogurov, A. A., Jr., Rubtsova, M. P., Kartseva, O. V., Gomzikova, M. O., Moskovtsev, A. A., Bukatin, A. S., Dubina, M. V., Kostryukova, E. S., Babenko, V. V., Vakhitova, M. T., Manolov, A. I., Malakhova, M. V., Kornienko, M. A., Tyakht, A. V., Vanyushkina, A. A., Ilina, E. N., Masson, P., Gabibov, A. G., and Altman, S. (2017) Microfluidic droplet platform for ultrahigh-throughput single-cell screening of biodiversity, Proceedings of the National Academy of Sciences of the United States of America 114, 2550-2555. [103] Murphy, A. J. M., Kung, A. L., Swirski, R. A., and Shimke, R. T. (1992) cDNA expression cloning in human cells using the pλDR2 episomal vector system, Methods 4, 111-131. [104] Becke, A. D. (1993) Density‐functional thermochemistry. III. The role of exact exchange, The Journal of Chemical Physics 98, 5648-5652. [105] Stephens, P. J., Devlin, F. J., Chabalowski, C. F., and Frisch, M. J. (1988) Ab Initio calculation of vibrational absorption and circular dichroism spectra using density functional force fields, Journal of Physical Chemistry 98, 11623-11627. [106] Kim, K., and Jordan, K. D. (1994) Comparison of density functional and MP2 calculations on the water monomer and dimer, Journal of Physical Chemistry 98, 10089-10094. [107] Dunning, T. H. (1989) Gaussian basis sets for use in correlated molecular calculations. I. The atoms boron through neon and hydrogen, The Journal of Chemical Physics 90, 1007-1023.

117

[108] Baldridge, K., and Klamt, A. (1997) First principles implementation of solvent effects without outlying charge error, The Journal of Chemical Physics 106, 6622-6633. [109] Cossi, M., Barone, V., Cammi, R., and Tomasi, J. (1996) Ab initio study of solvated molecules: A new implementation of the polarizable continuum model, Chemical Physics Letters 255, 327-335. [110] Tomasi, J., Mennucci, B., and Cammi, R. (2005) Quantum mechanical continuum solvation models, Chemical Reviews 105, 2999-3093. [111] Case, D. A., Ben-Shalom, I. Y., Brozell, S. R., Cerutti, D. S., Cheatham, T. E., Cruzeiro III, V.W.D., Darden, T. A., Duke, R. E., Ghoreishi, D., Gilson, M. K., Gohlke, H., Goetz, A. W., Greene, D., Harris, R., Homeyer, N., Huang, Y., Izadi, S., Kovalenko, A., Kurtzman, T., Lee, T. S., LeGrand, S., Li, P., Lin, C., Liu, J., Luchko, T., Luo, R., Mermelstein, D. J., Merz, K. M., Miao, Y., Monard, G., Nguyen, C., Nguyen, H., Omelyan, I., Onufriev, A., Pan, F., Qi, R., Roe, D. R., Roitberg, A., Sagui, C., Schott-Verdugo, S., Shen, J., Simmerling, C. L., Smith, J., Salomon-Ferrer, R., Swails, J., Walker, R. C., Wang, J., Wei, H., Wolf, R. M., Wu, X., Xiao, L., York, D. M., and Kollman, P. A. (2018) Amber 2018. [112] Duan, Y., Wu, C., Chowdhury, S., Lee, M. C., Xiong, G., Zhang, W., Yang, R., Cieplak, P., Luo, R., Lee, T., Caldwell, J., Wang, J., and Kollman, P. (2003) A Point-Charge Force Field for Molecular Mechanics Simulations of Proteins Based on Condensed-Phase Quantum Mechanical Calculations, Journal of Computational Chemistry 24, 1999-2012. [113] Šali, A., and Blundell, T. L. (1993) Comparative Protein Modelling by Satisfaction of Spatial Restraints, Journal of Molecular Biology 234, 779-815. [114] Cui, Q., and Karplus, M. (2003) Catalysis and specificity in enzymes: A study of triosephosphate isomerase and comparison with methyl glyoxal synthase, Advances in Protein Chemistry 66, 315-372. [115] Pearlman, D. A., Case, D. A., Caldwell, J. W., Ross, W. S., Cheatham, T. E., DeBolt, S., Ferguson, D., Seibel, G., and Kollman, P. (1995) AMBER, a package of computer programs for applying molecular mechanics, normal mode analysis, molecular dynamics and free energy calculations to simulate the structural and energetic properties of molecules, Computer Physics Communications 91, 1-41. [116] Dolinsky, T. J., Nielsen, J. E., McCammon, J. A., and Baker, N. A. (2004) PDB2PQR: An automated pipeline for the setup of Poisson-Boltzmann electrostatics calculations, Nucleic acids research 32, 665-667. [117] Pettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G. S., Greenblatt, D. M., Meng, E. C., and Ferrin, T. E. (2004) UCSF Chimera - A visualization system for exploratory research and analysis, Journal of Computational Chemistry 25, 1605- 1612. [118] Wandhammer, M., Carletti, E., Van Der Schans, M., Gillon, E., Nicolet, Y., Masson, P., Goeldner, M., Noort, D., and Nachon, F. (2011) Structural study of

118

the complex stereoselectivity of human butyrylcholinesterase for the neurotoxic V-agents, Journal of Biological Chemistry 286, 16783-16789. [119] Shapovalov, M. V., and Dunbrack, R. L. (2011) A smoothed backbone-dependent rotamer library for proteins derived from adaptive kernel density estimates and regressions, Structure 19, 844-858. [120] Jorgensen, W. L. C., J.; Madura, J. D.; Impey, R. W.; Klein, M. L. (1983) Comparison of simple potential functions for simulating liquid water, The Journal of Chemical Physics 79, 926-935. [121] Roe, D. R., and Cheatham, T. E. (2013) PTRAJ and CPPTRAJ: Software for processing and analysis of molecular dynamics trajectory data, Journal of Chemical Theory and Computation 9, 3084-3095. [122] Munro, N. B., Ambrose, K. R., and Watson, A. P. (1994) Toxicity of the organophosphate chemical warfare agents GA, GB, and VX: implications for public protection, Environmental health perspectives 102, 18-38. [123] Lyagin, I. V., Andrianova, M. S., and Efremenko, E. N. (2016) Extensive hydrolysis of phosphonates as unexpected behaviour of the known His6- organophosphorus hydrolase, Appl Microbiol Biotechnol 100, 5829-5838. [124] Singh, B. K., and Walker, A. (2006) Microbial degradation of organophosphorus compounds, FEMS Microbiol Rev 30, 428-471. [125] Lockridge, O., David, E., Schopfer, L. M., Masson, P., Brazzolotto, X., and Nachon, F. (2018) Purification of recombinant human butyrylcholinesterase on Hupresin(R), Journal of chromatography. B, Analytical technologies in the biomedical and life sciences 1102-1103, 109-115. [126] Rosenberry, T. L., Brazzolotto, X., Macdonald, I. R., Wandhammer, M., Trovaslet- Leroy, M., Darvesh, S., and Nachon, F. (2017) Comparison of the Binding of Reversible Inhibitors to Human Butyrylcholinesterase and Acetylcholinesterase: A Crystallographic, Kinetic and Calorimetric Study, Molecules 22. [127] Perler, F. B., Davis, E. O., Dean, G. E., Gimble, F. S., Jack, W. E., Neff, N., Noren, C. J., Thorner, J., and Belfort, M. (1994) Protein splicing elements: inteins and exteins--a definition of terms and recommended nomenclature, Nucleic acids research 22, 1125-1127. [128] Wood, D. W., Wu, W., Belfort, G., Derbyshire, V., and Belfort, M. (1999) A genetic system yields self-cleaving inteins for bioseparations, Nature biotechnology 17, 889-892. [129] Iwai, H., Zuger, S., Jin, J., and Tam, P. H. (2006) Highly efficient protein trans- splicing by a naturally split DnaE intein from Nostoc punctiforme, FEBS Lett 580, 1853-1858. [130] Zettler, J., Schutz, V., and Mootz, H. D. (2009) The naturally split Npu DnaE intein exhibits an extraordinarily high rate in the protein trans-splicing reaction, FEBS Lett 583, 909-914.

119

[131] Cooper, M. A., Taris, J. E., Shi, C., and Wood, D. W. (2018) A Convenient Split- Intein Tag Method for the Purification of Tagless Target Proteins, Curr Protoc Protein Sci 91, 5 29 21-25 29 23.

120

Appendix A. Plasmid Maps

121

Figure 21. pT7CFE1-BChE (WT)-His6

122

Figure 22. pET- BChE-His6

123

Figure 23. pTT- BChE-His6

124

Figure 24. pTT-BChE-His6 (without restriction enzymes)

125

HN Figure 25. pTT-IK-NpuC -mBChE-His6

126

Figure 26. pRcCMV-BChE-His6

127

Figure 27. pRcCMV-BChE-His6 (without restriction enzymes)

128

Figure 28. CGT8288-BChE (WT - Tobacco)-His6

129

Figure 29. AKK1448-BChE (WT BChE - Tobacco)-His6

130

Figure 30. AKK1472-BChE (WT BChE - Tobacco)-His6

131

Figure 31. pOpt_CCA_gLuc_Hyg[7646]

132

Figure 32. pOpt_CCA_gLuc_Paro[7647]

133

Figure 33. pOpt_CCA_BChE (without Intron)_Hyg

134

Figure 34. pOpt_CCA_BChE (without Intron)_Hyg (with restriction enzymes)

135

Figure 35. pOpt_CCA_BChE (with Intron)_Hyg

136

Figure 36. pOpt_CCA_BChE (with Intron)_Hyg (with restriction enzymes)

137

Appendix B: Supplemental Information for Human-Porcine Hybrid BChE

Figure 37. Sequence alignment comparison of human BChE vs. BChE from various species.

AfGrn 42 MHSKVTIICIRFLFWFLLLCMLIGKSHTEDDIVIATKNGKVRGMNLTVLGGTVTAFLGIP 101 Human 1 ...... I...... F...... 60 Rhes 1 MHSKVTIICIRLLFWFLLLCMLIGKSHTEDDIVIATKNGKVRGMNLTVLGGTVTAFLGIP 60 Human 1 ...... F...... I...... F...... 60 Cyno 1 MDSKVTIICIRLLFWFLLLCMLIGKSHTEDDIVIATKNGKVRGMNLTVLGGTVTAFLGIP 60 Human 1 .H...... F...... I...... F...... 60 Pig 1 MQRKGTIMYIRYFLWFLLLWMLVGKSYAEEDIIVTTKNGKVRGMNLPVLGGTVTAFLGIP 60 Human 1 .HS.V..IC..FLF.....C..I...HT.D...IA...... T.F...... 60

AfGrn 102 YAQPPLGRLRFKKPQSLTKWSDVWNATKYANSCYQNIDQSFPGFHGSEMWNPNTDLSEDC 161 Human 61 ...... I...... C...... 120 Rhes 61 YAQPPLGRLRFKKPQSLTKWSDIWNATKYANSCYQNIDQSFPGFHGSEMWNPNTDLSEDC 120 Human 61 ...... C...... 120 Cyno 61 YAQPPLGRLRFKKPQSLTKWSDIWNATKYANSCYQNIDQSFPGFHGSEMWNPNTDLSEDC 120 Human 61 ...... C...... 120 Pig 61 YAQPPLGRLRFKKPQSMTKWPDIWNATKYANSCYQNTDQSFPGFVGSEMWNPNTELSEDC 120 Human 61 ...... L...S...... C..I...... H...... D..... 120

AfGrn 162 LYLNVWIPAPKPKNATVMIWIYGGGFQTGTSSLHVYDGKFLARVERVIVVSMNYRVGALG 221 Human 121 ...... L...... 180 Rhes 121 LYLNVWIPAPKPKNATVMIWIYGGGFQTGTSSLHVYDGKFLARVERVIVVSMNYRVGALG 180 Human 121 ...... L...... 180 Cyno 121 LYLNVWIPAPKPKNATVMIWIYGGGFQTGTSSLHVYDGKFLARVERVIVVSMNYRVGALG 180 Human 121 ...... L...... 180 Pig 121 LYLNVWIPAPKPKNATVMIWIYGGGFQTGTSSLHVYDGKFLSRVERVIVVSMNYRVGALG 180 Human 121 ...... L...... A...... 180

AfGrn 222 FLALPGNPEAPGNMGLFDQQLALQWVQKNIAAFGGNPKSVTLFGESAGAASVSLHLLSPG 281

138

Human 181 ...... 240 Rhes 181 FLALPGNPEAPGNMGLFDQQLALQWVQKNIAAFGGNPKSVTLFGESAGAASVSLHLLSPG 240 Human 181 ...... 240 Cyno 181 FLALPGNPEAPGNMGLFDQQLALQWVQKNIAAFGGNPKSVTLFGESAGAASVSLHLLSPG 240 Human 181 ...... 240 Pig 181 FLALPGNPEAPGNMGLFDQQLALQWVQKNIAAFGGNPKSVTLFGESAGAVSVSLHLLSPR 240 Human 181 ...... A...... G 240

AfGrn 282 SHSLFTRAILQSGSSNAPWAVTSLYEARNRTLTLAKLTGCSRDNETEIVKCLRNKDPHEI 341 Human 241 ...... F...... N...... E.....I...... Q.. 300 Rhesu 241 SHPLFTRAILQSGSSNAPWAVTSLYEARNRTLTLAKLTGCSRDNETEIVKCLRNKDPHEI 300 Human 241 ..S...... F...... N...... E.....I...... Q.. 300 Cyno 241 SHSLFTRAILQSGSSNAPWAVTSLYEARNRTLTLAKLTGCSRDNETEIVKCLRNKDPHEI 300 Human 241 ...... F...... N...... E.....I...... Q.. 300 Pig 241 SHPLFARAILQSGSSNAPWAVTSLYEARNRTLTLAKFIGCSRENETEIIKCLRNKDPQEI 300 Human 241 ..S..T...... F...... N...LT...... 300

AfGrn 342 LLNEAFVVPYGTLLSVNFGPTMDGDFLTDMPDILLELGQFKKTQILVGVNKDEGTAFLVY 401 Human 301 ...... P...... V...... 360 Rhes 301 LLNEAFVVPYGTLLSVNFGPTMDGDFLTEMPDILLELGQFKKTQILVGVNKDEGTAFLVY 360 Human 301 ...... P...... V...... D...... 360 Cyno 301 LLNEAFVVPYGTLLSVNFGPTMDGDFLTEMPDILLELGQFKKTQILVGVNKDEGTAFLVY 360 Human 301 ...... P...... V...... D...... 360 Pig 301 LQNEVFVVPNHMLLSVNFGPTVDGDFLTDLPDTLLQLGQFKKTQILVGVNKDEGTAFLVY 360 Human 301 .L..A....YGTP...... M..I..E...... 360

AfGrn 402 GAPGFSKDNNSIITRNEFQEGLKIFFPGVSEFGKESILFHYTDWVDDQRPENYREALDDV 461 Human 361 ...... K...... G.. 420 Rhes 361 GAPGFSKDNDSIITRNEFQEGLKIFFPGVSEFGKESILFHYTDWVDDQRPENYREALDDV 420 Human 361 ...... N.....K...... G.. 420 Cyno 361 GAPGFSKDNDSIITRNEFQEGLKIFFPGVSEFGKESILFHYTDWVDDQRPENYREALDDV 420 Human 361 ...... N.....K...... G.. 420 Pig 361 GAPGFSKDNNSIITRKEFEEGLKIFFPGVSEFGKESILFHYMDWTDDQRAENYRDALDDV 420 Human 361 ...... Q...... T..V....P....E..G.. 420

AfGrn 462 VGDYNIICPALEFTKKFSEWGNNAFFYYFEHRSSKLPWPEWMGVMHGYEIEFVFGLPLER 521 Human 421 .....F...... 480 Rhes 421 VGDYNIICPALEFTKKFSEWGNNAFFYYFEHRSSKLPWPEWMGVMHGYEIEFVFGLPLER 480 Human 421 .....F...... 480 139

Cyno 421 VGDYNIICPALEFTKKFSEWGNNAFFYYFEHRSSKLPWPEWMGVMHGYEIEFVFGLPLER 480 Human 421 .....F...... 480 Pig 421 VGDYDIICPALEFTKKFSEMGNNAFFYYFEHRSSKLPWPEWMGVMHGYEIEFVFGLPLER 480 Human 421 ....NF...... W...... 480

AfGrn 522 RVNYTKAEEILSRSIVKRWANFAKYGNPNGTHNNSTKWPVFKSTEQKYLTLNTESSRILT 581 Human 481 .D...... E.Q....S...... T..M. 540 Rhes 481 RVNYTKAEEILSRSIVKRWANFAKYGNPNGTHNNSTKWPVFKSTEQKYLTLNTESSRILT 540 Human 481 .D...... E.Q....S...... T..M. 540 Cyno 481 RVNYTKAEEILSRSIVKRWANFAKYGNPNGTHNNSTKWPVFKSTEQKYLTLNTESSRILT 540 Human 481 .D...... E.Q....S...... T..M. 540 Pig 481 RANYTKAEEILSRSIMKRWANFAKYGNPNGTQNNSTRWPVFKSNEQKYLTLNAESPRVYT 540 Human 481 .D...... V...... E...... S...... T...... T..T.IM. 540

AfGrn 582 KLRAQQCRFWTSFFPKVLEMTGNIDEAEWEWKAGFHRWSNYMMDWKNQFNDYTSKKESCV 641 Human 541 ...... N...... 600 Rhes 541 KLRAQQCRFWTSFFPKVLEMTGNIDEAEWEWKAGFHRWSNYMMDWKNQFNDYTSKKESCV 600 Human 541 ...... N...... 600 Cyno 541 KLRAQQCRFWTSFFPKVLEMTGNIDEAEWEWKAGFHRWSNYMMDWKNQFNDYTSKKESCV 600 Human 541 ...... N...... 600 Pig 541 KLRAQQCRFWTLFFPKVLEMTGNIDEAEREWKAGFHRWNNYMMDWKNQFNDYTSKKESCA 600 Human 541 ...... S...... W...... V 600

AfGrn 642 GL 643 Human 601 .. 602 Rhes 601 GL 602 Human 601 .. 602 Cyno 601 GL 602 Human 601 .. 602 Pig 601 DL 602 Human 601 G. 602

140

Figure 38. 48-Hr Aging Experiment Results

R e c o m b in a n t H u m a n W T B C h E v s G B

R e c o m b in a n t H u m a n (W T ) 1 0 0

l B C h E , 0 µ M 2 P A M C l

n R e c o m b in a n t H u m a n (W T )

o t

i 7 5 B C h E , 5 0 µ M 2 P A M C l

n 5 0

c 2 5

% n

U A g in g R a te (t1 /2 ): 2 7 .3 H rs

S p o n ta n e o u s R e a c tiv a tio n R a te (t1 /2 ): 9 .0 H rs 0 0 1 2 2 4 3 6 4 8 T im e (H rs ) n = 6 E r ro r R e p re s e n ts  S E M

R e c o m b in a n t H u m a n - P ig A c y l B C h E v s G B

R e c o m b in a n t H u m a n (P ig A c y l) 1 0 0

l B C h E , 0 µ M 2 P A M C l

n R e c o m b in a n t H u m a n (P ig A c y l)

o t

i 7 5

C B C h E , 5 0 µ M 2 P A M C l

n 5 0

l h

l S p o n ta n e o u s R e a c tiv a tio n R a te (t ): 2 .4 H rs C

a 1 /2

c 2 5 R e a c tiv a tio n R a te w ith O x im e (t ): 3 .4 H rs

% 1 /2

n U

0 0 8 1 6 2 4 3 2 4 0 4 8 T im e (H rs ) n = 6 E r ro r R e p re s e n ts  S E M

141

Table 9. Measured Concentrations of Sarin and IMPA using LC-MS/MS

[Sarin] Sample ID Time (hours) [IMPA] (ng/mL) (ng/mL) Human (WT) BChE 0 ND ND Human (WT) BChE 1 ND 0.17 Human (WT) BChE 2 ND 0.20 Human (WT) BChE 4 ND 0.19 Human (WT) BChE 6 ND 0.21 Human (WT) BChE 22 ND 0.26 Human (Hybrid) BChE 0 ND 0.31 Human (Hybrid) BChE 1 ND 0.44 Human (Hybrid) BChE 2 ND 0.45 Human (Hybrid) BChE 4 ND 0.57 Human (Hybrid) BChE 6 ND 0.66 Human (Hybrid) BChE 22 ND 0.68 ND: Not Detected

142

Table 10. Measured Concentrations of sarin over time using GC-HRMS

Sample ID Time (hours) [Sarin] (ng/mL)

10 mM MOPS (-) Control 0 1.61 10 mM MOPS (-) Control 2 0.84 10 mM MOPS (-) Control 5 0.60 Hybrid-BChE 0 1.57 Hybrid -BChE 2 0.97 Hybrid -BChE 5 0.42

143

B-1: Supplemental Information for Molecular Dynamics Simulations

Molecular Dynamics analysis of the native enzyme before and after mutation Molecular dynamics (MD) simulations were carried out for WT and hybrid forms of native BChE, that included each possible protonation states for the His283 in the hybrid form of the enzyme. All MD simulations were performed with the AMBER 16 molecular dynamics package 111 with the ff03 force field. The protein preparation was performed for the WT and hybrid forms of the enzyme with 1P0I human native form of BChE (Ref. 42) as previously eluded to in the computational methods.

Clustering analysis for the native enzyme A clustering protocol using AMBER’s CPPTRAJ module was used in order to decrease the amount of structures for analysis in the native forms of the enzyme. A “representative” structure was chosen for each cluster that is color-coded by the relative populations of each cluster.

144

Figure 39. Cluster analysis by RMSD for the native forms of BChE color coded by relative populations of each cluster where blue is most populated, red is 2nd most populated, green is 3rd most populated, yellow is 4th most populated, pink is 5th most populated, and black is any cluster more than 5th.

145

Root mean square fluctuation analysis for the native enzyme The root mean square fluctuation (RMSF) was calculated using AMBER’s CPPTRAJ module for each residue in order to evaluate the flexibility of certain residues for the native enzyme. In order to compare the flexibility of the mutation site to the rest of the protein it is color-coded.

Figure 40. RMSF analysis for the native forms of BChE where the site of the mutation is color-coded red.

146

Distance analysis for the active site of the native enzyme To observe the important interactions in the active site of the WT and hybrid forms of the native enzyme, AMBER’s PYTRAJ module was used in order to extract distances. Heat map plots were made for the WT and the HID, HIE, and HIP protonation states of the G283H mutation were made. The heat maps display the distances as a function of color over time. The interatomic distances measured are summarized below: Interatomic distances: • Oxyanion hole: Average distance from hydrogens from the backbone amides of Gly116, Gly117, and Ala199 to the O of Ser198. • SerH − HisN: Distance from the H of Ser198 to the -nitrogen of the catalytic His438. • HisH − Glu325: Distance from the -hydrogen of the catalytic His438 to the catalytic Glu325 oxyanion. • HisH − Glu441: Distance from the -hydrogen of the catalytic His438 to the non-catalytic Glu441 oxyanion, this is an interaction that is often observed in the inhibited form of the enzyme. • SerH − Glu197: Distance from the H of Ser198 to the adjacent Glu197 oxyanion. • Gly115H − Glu197: Distance from the hydrogen from the backbone amide of Gly115 to the Glu325 oxyanion. • Leu286 − HisN: Distance from the closest carbon of the methyl group on Leu286 in the acyl-binding pocket to the  - Nitrogen of the catalytic His438. • Phe329 − HisN: Distance from the centroid of the benzyl group of Phe329 in the acyl- binding pocket to the  - Nitrogen of the catalytic His438.

147

Figure 41. Active site distances native forms of BChE color-coded by the different interactions in the active site. Distances are displayed as heat maps and are shown as a function of color over time.

148

Distance analysis for the mutation site/acyl loop of the native enzyme To observe the important interactions in the acyl loop and the site of the mutation of the WT and hybrid forms of the native enzyme, AMBER’s PYTRAJ module was used in order to extract distances. Heat map plots were made for the WT and the HID, HIE, and HIP protonation states of the G283H mutation were made. The heat maps display the distances as a function of color over time. The top 6 interatomic distances are shared between the WT and hybrid forms of the enzyme Shared interatomic distances: • Tyr332 − Pro285back/Leu285back: Distance from the hydrogen of the alcohol group of Tyr332 to the backbone of Pro285 (WT) or Leu285 (hybrid) backbone. • Tyr332 − Ser79: Distance from the Distance from the hydrogen of the alcohol group of Tyr332 to the oxygen of the alcohol group of Ser79 located on the -loop. • Tyr332 − Gly75back: Distance from the hydrogen of the alcohol group of Tyr332 to the backbone of Gly75 located on the -loop. • Tyr332 − Gln71back: Distance from the hydrogen of the alcohol group of Tyr332 to the backbone of Gln71 located on the -loop. • Tyr332 − Phe329Cen: Distance from the hydrogen of the alcohol group of Tyr332 to the centroid of the benzyl group of Phe329 in the acyl-binding pocket. • Tyr332 − HisN: Distance from the hydrogen of the alcohol group of Tyr332 to the - nitrogen of the catalytic His438 located in the active site. WT interatomic distances: • Tyr282 − Val288back: Distance from the hydrogen of the alcohol group of Tyr282 to the backbone of Val288 located in the acyl-binding pocket. • Tyr282 − Tyr332: Distance from the hydrogen of the alcohol group of Tyr282 to the hydrogen of the alcohol group of Tyr332. • Tyr282 − Tyr332back: Distance from the hydrogen of the alcohol group of Tyr282 to the backbone of Tyr332. • Tyr282 − Ile356back: Distance from the hydrogen of the alcohol group of Tyr282 to the backbone of Ile356. Hybrid interatomic distances: • His283 − Tyr332: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the oxygen of the alcohol group of Tyr322. • His283 − Ser72back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Ser72 located on the -loop. • His283 − Pro230back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Pro230 located on the bottom of the acyl- binding pocket. • His283 − Asn282back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Asn282 located above the acyl-binding pocket. • His283 − Pro281back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Pro281 located above the acyl-binding pocket.

149

• His283 − Val280back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Val280 located above the acyl-binding pocket.

Figure 42. Distances of the acyl loop and the site of the mutation for the native forms of BChE to evaluate different interactions near the mutation. Distances are displayed as heat maps and are shown as a function of color over time.

150

Distance analysis for Trp82 of the native enzyme To visualize the stability and proximity of the omega loop to the active site of the native enzyme, AMBER’s PYTRAJ module was used in order to extract distances as well as an angle. Important distances from the  - Nitrogen of the catalytic His438 to the -carbon (C), the centroid of the amino ring (Cen1), and the centroid of the benzyl ring (Cen2) of Trp82. A summary of the distances and the angle measured are shown below: Interatomic distances: • C: Distance from the -nitrogen of the catalytic His438 located in the active site to the - Carbon of Trp82. • Cen1: Distance from the -nitrogen of the catalytic His438 located in the active site to the centroid of the amino ring of Trp82. • Cen2: Distance from the -nitrogen of the catalytic His438 located in the active site to the centroid of the benzyl ring of Trp82. Interatomic angle: • HisN − Cen1 − Cen2: Angle formed by the -nitrogen of the catalytic His438 located in the active site to the centroid of the amino ring of Trp82 to the centroid of the benzyl ring of Trp82.

151

Figure 43. Distances and angle of Trp82 in the -loop of the native form as compared to the position and orientation of the -nitrogen of the catalytic His438 located in the active site.

152

Figure 44. Illustration of “acyl-pocket encroachment” in the HID form of the native enzyme.

153

Figure 45. Illustration of active site “climbing” in the HIE form of the native enzyme

Molecular Dynamics analysis of the inhibited enzyme before and after mutation Molecular dynamics (MD) simulations were carried out for WT and hybrid forms of inhibited BChE, that included each possible protonation states for the His283 in the hybrid form of the enzyme. All MD simulations were performed with the AMBER 16 molecular dynamics package with the ff03 force field. The protein preparation was performed for the WT and hybrid forms of the enzyme with 2XQJ and 2XQK that are the (R)- and (S)- inhibited stereoisomers of VX, respectively, as previously eluded to in the computational methods.

Clustering analysis for the inhibited enzyme A clustering protocol using AMBER’s CPPTRAJ module was used in order to decrease the amount of structures for analysis in the inhibited forms of the enzyme. A “representative” structure was chosen for each cluster that is color-coded by the relative populations of each cluster.

154

Figure 46. Cluster analysis by RMSD for the 2XQJ inhibited forms of BChE color coded by relative populations of each cluster where blue is most populated, red is 2nd most populated, green is 3rd most populated, yellow is 4th most populated, pink is 5th most populated, and black is any cluster more than 5th.

155

Figure 47. Cluster analysis by RMSD for the 2XQK inhibited forms of BChE color coded by relative populations of each cluster where blue is most populated, red is 2nd most populated, green is 3rd most populated, yellow is 4th most populated, pink is 5th most populated, and black is any cluster more than 5th.

156

Root mean square fluctuation analysis for the inhibited enzyme The root mean square fluctuation (RMSF) was calculated using AMBER’s CPPTRAJ module for each residue in order to evaluate the flexibility of certain residues for the inhibited enzyme. In order to compare the flexibility of the mutation site to the rest of the protein it is color- coded.

Figure 48. RMSF analysis for the 2XQJ inhibited forms of BChE where the site of the mutation is color-coded red.

157

Figure 49. RMSF analysis for the 2XQK inhibited forms of BChE where the site of the mutation is color-coded red.

158

Distance analysis for the active site of the inhibited enzyme To observe the important interactions in the active site of the WT and hybrid forms of the inhibited enzyme, AMBER’s PYTRAJ module was used in order to extract distances. Heat map plots were made for the WT and the HID, HIE, and HIP protonation states of the G283H mutation for 2XQJ and 2XQK were made. The heat maps display the distances as a function of color over time. The interatomic distances measured are summarized below: Interatomic distances: • Oxyanion hole: Average distance from hydrogens from the backbone amides of Gly116, Gly117, and Ala199 to the O1 of phosphylated Ser198 (Sarin198) which is the oxygen that is located closest to these backbone amides. • HisH − O1: Distance from the -hydrogen of the catalytic His438 to the O1 of phosphylated Ser198 (Sarin198) which is the oxygen that is located in the oxyanion hole. • HisH − O2: Distance from the -hydrogen of the catalytic His438 to the O2 of phosphylated Ser198 (Sarin198) which is the oxygen connected to the isopropyl leaving group. • HisH − O: Distance from the -hydrogen of the catalytic His438 to the O1 of phosphylated Ser198 (Sarin198) which is the oxygen connected to the protein. • HisH − Glu325: Distance from the -hydrogen of the catalytic His438 to the catalytic Glu325 oxyanion. • HisH − Glu441: Distance from the -hydrogen of the catalytic His438 to the non-catalytic Glu441 oxyanion, this is an interaction that is often observed in the inhibited form of the enzyme. • HisH − Glu197: Distance from the -hydrogen of the catalytic His438 to the Glu197 oxyanion. • Gly115H − Glu197: Distance from the hydrogen from the backbone amide of Gly115 to the Glu325 oxyanion. • Leu286 − HisN: Distance from the closest carbon of the methyl group on Leu286 in the acyl-binding pocket to the  - Nitrogen of the catalytic His438. • Phe329 − HisN: Distance from the centroid of the benzyl group of Phe329 in the acyl- binding pocket to the  - Nitrogen of the catalytic His438.

159

Figure 50. Active site distances for the 2XQJ inhibited forms of BChE color-coded by the different interactions in the active site. Distances are displayed as heat maps and are shown as a function of color over time.

160

Figure 51. Active site distances for the 2XQK inhibited forms of BChE color-coded by the different interactions in the active site. Distances are displayed as heat maps and are shown as a function of color over time.

161

Distance analysis for the mutation site/acyl loop of the inhibited enzyme To observe the important interactions in the acyl loop and the site of the mutation of the WT and hybrid forms of the inhibited enzyme, AMBER’s PYTRAJ module was used in order to extract distances. Heat map plots were made for the WT and the HID, HIE, and HIP protonation states of the G283H mutation for 2XQJ and 2XQK were made. Heat maps display the distances as a function of color over time. The top 6 interatomic distances are shared between the WT and hybrid forms of the enzyme

Shared interatomic distances: • Tyr332 − Pro285back/Leu285back: Distance from the hydrogen of the alcohol group of Tyr332 to the backbone of Pro285 (WT) or Leu285 (hybrid) backbone. • Tyr332 − Ser79: Distance from the Distance from the hydrogen of the alcohol group of Tyr332 to the oxygen of the alcohol group of Ser79 located on the -loop. • Tyr332 − Gly75back: Distance from the hydrogen of the alcohol group of Tyr332 to the backbone of Gly75 located on the -loop. • Tyr332 − Gln71back: Distance from the hydrogen of the alcohol group of Tyr332 to the backbone of Gln71 located on the -loop. • Tyr332 − Phe329Cen: Distance from the hydrogen of the alcohol group of Tyr332 to the centroid of the benzyl group of Phe329 in the acyl-binding pocket. • Tyr332 − HisN: Distance from the hydrogen of the alcohol group of Tyr332 to the - nitrogen of the catalytic His438 located in the active site.

WT interatomic distances: • Tyr282 − Val288back: Distance from the hydrogen of the alcohol group of Tyr282 to the backbone of Val288 located in the acyl-binding pocket. • Tyr282 − Tyr332: Distance from the hydrogen of the alcohol group of Tyr282 to the hydrogen of the alcohol group of Tyr332. • Tyr282 − Tyr332back: Distance from the hydrogen of the alcohol group of Tyr282 to the backbone of Tyr332. • Tyr282 − Ile356back: Distance from the hydrogen of the alcohol group of Tyr282 to the backbone of Ile356.

162

Hybrid interatomic distances: • His283 − Tyr332: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the oxygen of the alcohol group of Tyr322. • His283 − Ser72back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Ser72 located on the -loop. • His283 − Pro230back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Pro230 located on the bottom of the acyl- binding pocket. • His283 − Asn282back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Asn282 located above the acyl-binding pocket. • His283 − Pro281back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Pro281 located above the acyl-binding pocket. • His283 − Val280back: Distance from the -(HID) or -(HIE) hydrogen or the closest of the two (HIP) of His283 to the backbone of Val280 located above the acyl-binding pocket.

Figure 52. Distances of the acyl loop and the site of the mutation for the 2XQJ inhibited forms of BChE to evaluate different interactions near the mutation. Distances are displayed as heat maps and are shown as a function of color over time.

163

Figure 53. Distances of the acyl loop and the site of the mutation for the 2XQK inhibited forms of BChE to evaluate different interactions near the mutation. Distances are displayed as heat maps and are shown as a function of color over time.

164

Distance analysis for Trp82 of the inhibited enzyme To visualize the stability and proximity of the omega loop to the active site of the inhibited enzyme, AMBER’s PYTRAJ module was used in order to extract distances as well as an angle. Distance plots were made for the WT and the HID, HIE, and HIP protonation states of the G283H mutation for 2XQJ and 2XQK were made. Important distances from the  - Nitrogen of the catalytic His438 to the -carbon (C), the centroid of the amino ring (Cen1), and the centroid of the benzyl ring (Cen2) of Trp82. A summary of the distances and the angle measured are shown below: Interatomic distances: • C: Distance from the -nitrogen of the catalytic His438 located in the active site to the - Carbon of Trp82. • Cen1: Distance from the -nitrogen of the catalytic His438 located in the active site to the centroid of the amino ring of Trp82. • Cen2: Distance from the -nitrogen of the catalytic His438 located in the active site to the centroid of the benzyl ring of Trp82. Interatomic angle: • HisN − Cen1 − Cen2: Angle formed by the -nitrogen of the catalytic His438 located in the active site to the centroid of the amino ring of Trp82 to the centroid of the benzyl ring of Trp82.

165

Figure 54. Distances and angle of Trp82 in the -loop of the 2XQJ inhibited form as compared to the position and orientation of the -nitrogen of the catalytic His438 located in the active site.

166

Figure 55. Distances and angle of Trp82 in the -loop of the 2XQK inhibited form as compared to the position and orientation of the -nitrogen of the catalytic His438 located in the active site.

167

Distance analysis for spontaneous hydrolysis of the inhibited enzyme To observe the important water interactions in the WT and hybrid forms of the inhibited enzyme, AMBER’s PYTRAJ module was used in order to extract distances. Heat map plots were made for the WT and the HID, HIE, and HIP protonation states of the G283H mutation for 2XQJ and 2XQK were made. The heat maps display the distances as a function of color over time. The interatomic distances measured are summarized below: Interatomic distances: • phosph − wat1: Distance from the phosphorus of the inhibited Ser198 to the oxygen of the closest water to the phosphorous atom of the inhibited Ser198. • phosph − wat2: Distance from the phosphorus of the inhibited Ser198 to the oxygen of the 2nd closest water to the phosphorous atom of the inhibited Ser198. • HisN − wat1: Distance from the -nitrogen of the catalytic His438 to the center of mass of the hydrogens of the closest water to the phosphorus atom of the inhibited Ser198. • HisN − wat2: Distance from the -nitrogen of the catalytic His438 to the center of mass of the hydrogens of the 2nd closest water to the phosphorus atom of the inhibited Ser198. • HisH − Glu197: Distance from the -hydrogen of the catalytic His438 to the Glu197 oxyanion. • Glu197 − wat1: Distance from the Glu197 oxyanion to the center of mass of the hydrogens of the closest water to the phosphorus atom of the inhibited Ser198. • Glu197 − wat2: Distance from the Glu197 oxyanion to the center of mass of the hydrogens of the closest water to the phosphorus atom of the inhibited Ser198.

168

Figure 56. Water interaction distances for the 2XQJ inhibited forms of BChE color-coded by the different interactions in the active site. Distances are displayed as heat maps and are shown as a function of color over time.

169

Figure 57. Water interaction distances for the 2XQK inhibited forms of BChE color-coded by the different interactions in the active site. Distances are displayed as heat maps and are shown as a function of color over time.

170

Figure 58. Example of an “active” water for Pathway 2 that sits in between the phosphylated serine (SIB) and Glu197. This is taken from the most populated cluster of the HIE form of 2XQK inhibition.

171

Figure 59. Example of an “active” water for Pathways 1 and 2 that sits in between the phosphylated serine (SIB) and Glu197. This is taken from the third most populated cluster of the HIE form of 2XQK.

172

Figure 60. Illustration of the water pocket formed in between Glu197 and phosphylated Ser198 (Sarin198) for the top 3 clusters of the HIE form of 2XQK. Each of the individual clusters is color coded by population: blue, most populated; red, next most populated; and green, lowest population.

173