The Development of Novel Cysteine Cross-Linkers and Their Application Towards Neurodegenerative Disorders

The Development of Novel Cysteine Cross-linkers and Their Application Towards Neurodegenerative Disorders

by Daniel Patrick Donnelly

B.A. in English Literature, New York University B.S. in Chemistry, Salem State University

A dissertation submitted to

The Faculty of the College of Science of Northeastern University in partial fulfillment of the requirements for the degree of Doctor of Philosophy

April 9th, 2019

Dissertation directed by

Jeffrey N. Agar Associate Professor of Chemistry and Chemical Biology & Pharmaceutical Sciences

Dedication

To my mother, Loretta Yannaco, my father, Robert Jude Donnelly (Jan 26, 1955-Dec 3, 2015), my sister and brother Claire Donnelly and Nicholas Rieber, and my soon-to-be wife, Rebecca Towers for your constant support over the years. I truly would not be where I am today without you. Thank you.

Acknowledgements

I must first acknowledge my advisor, my mentor, my boss, and my friend, Jeffrey N. Agar, for his constant support and guidance throughout my Ph.D. The training, skillset, and knowledge

I have gained under his mentorship is immeasurable. Jeff taught me that good research is thorough research, that publishing in high-impact journals is tedious but worth it, and that scientific problems can be approached from a number of angles. I am a better scientist as a result of his mentorship.

I would like to thank the current and former Agar Lab members: Dr. Catherine M. Rawlins, for being an exceptional colleague, collaborator, and, most importantly, friend; Dr. Jeniffer V.

Quijada, for being a mentor, teacher, and friend throughout my first years at Northeastern

University; Nicholas D. Schmitt, for always being critical of our data and our writing and, consequently, making me a better scientist; Dr. Joseph P. Salisbury, for paving the way for my project and always allowing me to discuss ideas and problems throughout the years; Md Amin

Hossain, for being an exceptional and supportive mentee; Richa Sarin, for being a friend in the lab and always offering your knowledge and constructive feedback; Krishna Aluri, for taking our project to new areas of chemistry; and Jeremy B. Conway for continuously coming back to the

Agar Lab and supporting my research. You all have made my time at Northeastern University that much better.

Thank you to my collaborators, committee members, and mentors: Dr. Jared R. Auclair, for inspiring me to be a better experimentalist and giving me unwavering access to your resources and your knowledge; Dr. Steven A. Lopez, for teaching me a side of chemistry I thought I would never understand and being patient throughout the learning process; Dr. Roman Manestch, for diving headfirst into our project and always supporting my research; Dr. Alexander R. Ivanov, for being a welcoming face in the Barnett Institute and offering resources and support in the completion of my project.

I would also like to thank my collaborators from the Consortium for Top-Down proteomics

(of which there are too many to name). A special thanks goes to Jeremy Wolff of Bruker Daltronics for teaching me the ins and outs of the FT-ICR MS and supporting my research.

Abstract of Dissertation

Neurodegenerative diseases often result from the aggregation of destabilized proteins in the central nervous system caused by genetic mutations or chemical modifications. Mutations in the gene responsible for the expression of Cu/Zn superoxide dismutase, for example, lead to proteins with aberrant conformations and are implicated in 2-7% of all amyotrophic lateral sclerosis (ALS) cases. These mutations destabilize the native homodimer of SOD1 and contribute to its dissociation into toxic monomers prone to self-assemble into higher-order aggregates. The development of pharmacological chaperones, small molecules that assist in the stabilization of nascent proteins, is an accepted therapeutic strategy. Past attempts to stabilize the native dimer of

SOD1 through covalent cross-linking were successful in rescuing enzymatic activity of variant proteins but utilized toxic commercial cross-linkers. In this dissertation, a new class of compounds, cyclic disulfides and cyclic thiosulfinates, is introduced. Cyclic disulfides and cyclic thiosulfinates are the first reported cysteine-specific cross-linkers that target pairs of closely-spaced cysteine residues but avoid the formation of dead-end modification on lone cysteine residues (the reason all other cysteine cross-linkers are toxic in vivo). The chapters presented here explore the cross- linking mechanism of cyclic disulfides and cyclic thiosulfinates, identify the mechanistic driving force of cyclic thiosulfinate cross-linking rate-acceleration, apply this chemistry to ALS disease models (in vitro, in cellulo, and in vivo), and present newly established protocols to analyze intact proteins and protein modifications via mass spectrometry.

Table of Contents

Dedication ...... 2 Acknowledgements ...... 3 Abstract of Dissertation ...... 5 Table of Contents ...... 6 List of Tables ...... 9 List of Figures ...... 10 Abbreviations ...... 15 Chapter 1 ...... 21 Introduction ...... 21 1.1 Proteinopathies and Neurodegeneration ...... 21 1.2 Mechanism of Seeded Protein Aggregation and Neurotoxicity ...... 23 1.3 Amyotrophic Lateral Sclerosis ...... 25 1.4 Aggregation of Superoxide Dismutase (SOD1) and ALS ...... 26 1.5 Protein Stabilization and Chemical/Pharmacological Chaperones ...... 31 1.6 Tafamidis and the stabilization of Transthyretin ...... 33 1.7 Identification of compounds that stabilize the SOD1 dimer ...... 36 1.8 Stabilization of SOD1 with homobifunctional maleimides ...... 39 1.9 Cyclic disulfides...... 41 Chapter 2 ...... 44 Cyclic Thiosulfinates and Cyclic Disulfides Selectively Cross-Link Thiols While Avoiding Modification of Lone Thiols ...... 44 2.0 Statement of Contribution ...... 45 2.1 Abstract ...... 46 2.2 Introduction ...... 47 2.3 Results ad Discussion ...... 51 2.4 Conclusion ...... 55 2.5 Methods...... 56 Chapter 3 ...... 69 Nucleophilic substitution reactions of cyclic thiosulfinates are accelerated by hyperconjugative interactions ...... 69 3.0 Statement of Contribution ...... 70 3.1 Abstract ...... 71 3.2 Introduction ...... 71

3.3 Results and Discussion ...... 73 3.4 Methods...... 86 3.5 Conclusion ...... 87 Chapter 4 ...... 89 New Pharmacological Chaperone Scaffold for the Stabilization of Disease-Associated Variants of Cu/Zn Superoxide Dismutase ...... 89 4.0 Statement of Contribution ...... 90 4.1 Abstract ...... 91 4.2 Introduction ...... 91 4.3 Results ...... 95 4.4 Conclusion ...... 104 4.5 Methods...... 105 Chapter 5 ...... 114 Best Practices and Benchmarks for Mass Spectrometry of Intact Proteins ...... 114 5.0 Statement of Contribution ...... 116 5.1 Abstract ...... 117 5.2 Introduction ...... 117 5.2.1 Origins of Signal Suppression and Signal Spreading ...... 119 5.3 Results ...... 121 5.3.1 Defining the Problems: Signal Suppression by Common Buffer Components ...... 121 5.3.2 The Intact Protein MS (IPMS) Decision Tree for Choosing the Appropriate Experiment ...... 123 5.3.3 Protein Standards and Benchmarks ...... 125 5.3.4 Dilution ...... 126 5.3.5 Ultrafiltration ...... 126 5.3.6 Precipitation ...... 128 5.3.7 Native vs. Denaturing MS...... 130 5.3.8 LC-MS of Intact Proteins ...... 132 5.3.9 Special Methodological Considerations for Intact Antibody Mass Spectrometry ..... 138 5.4 Discussion ...... 138 5.5 Supplementary Protocols ...... 139 5.5.1 Supplementary Protocol 5-1: Dilution ...... 139 5.5.2 Supplementary Protocol 5-2a: MWCO-Ultrafiltration Additional Details ...... 139 5.5.3 Supplementary Protocol 5-2b: Native Membrane Protein Preparation Additional Details ...... 140

5.5.4 Supplementary Protocol 5-3: Protein Precipitation Additional Details ...... 141 5.5.5 Supplementary Protocol 5-4a: Denaturing Mass Spectrometry Additional Details .. 142 5.5.6 Supplementary Protocol 5-4b: Native Mass Spectrometry Additional Details ...... 142 5.5.7 Supplementary Protocol 5a: LC-MS Benchmarks Additional Details ...... 146 5.5.8 Protocol 5b: Denaturing Reversed Phase LC-MS of bacteriorhodopsin from Halobacterium salinarum ...... 153 5.6 Supplementary Notes ...... 157 5.6.1 Supplementary Note 5-1: Protein Standard Mixture ...... 157 5.6.2 Supplementary Note 5-2: Details on Signal Suppression Curves ...... 157 5.6.3 Supplementary Note 5-3: Typical Multicomponent Mixtures of Excipients: ...... 158 5.5.4 Supplementary Note 5-4: Special Considerations for Intact Antibody Mass Spectrometry Additional Details...... 159 5.5.5 Supplementary Note 5-5: S/N Calculations ...... 160 Conclusions and Future Directions ...... 163 Appendix 1: Supplementary Figures for Cyclic Thiosulfinates and Cyclic Disulfides Selectively Cross-Link Thiols While Avoiding Modification of Lone Thiols ...... 165 Appendix 2: Supplementary Figures for Closed-Shell Repulsions Pre-Distort Cyclic Thiosulfinates and Accelerate Nucleophilic Substitution Reactions ...... 215 Appendix 3: Supplementary Figures for New Pharmacological Chaperone Scaffold for the Stabilization of Disease-Associated Variants of Cu/Zn Superoxide Dismutase ...... 279 Appendix 4: Supplementary Figures for Best Practices and Benchmarks for Mass Spectrometry of Intact Proteins ...... 284 References ...... 297

List of Tables

Table 1-1. Table of the most common neurodegenerative diseases, proteins implicated with disease onset/progression, and estimated prevalence in the United States.

Table 3-1. ΔGrxn of the nucleophilic attack of MeS—on (3—10)a and (3—10)b.

∗ Table 3-2. Summary of S1—S2 bond lengths, nO and σSS energies, and the interaction energies ∗ between the nO and σSS orbitals.

Table A3-1: Radius of gyration of each sample calculated in RAW.

List of Figures

Figure 1-1. Chemical Structures of in vivo proteasomal activators.

Figure 1-2. Hypothesized model for SOD1-mediated fALS disease progression.

Figure 1-3. C4F6 positive staining is shown for 4 SALS cases.

Figure 1-4. Mechanism of TTR aggregation and TTR stabilization by tafamidis.

Figure 1-5. Representative compounds identified through in silico screen that inhibit SOD1 aggregation.

Figure 1-6. Representation of SOD1 stabilization by homobifunctional maleimides and corresponding in vitro stabilization.

Figure 1-7. Cyclic disulfides currently in use as drugs.

Figure 2-1. Proposed mechanism of thiol cross-linking using cyclic disulfides (blue) and cyclic thiosulfinates (red).

Figure 2-2. Thiol cross-linking by cyclic thiosulfinates kinetically stabilizes the SOD1 dimer in vitro and in cells.

Scheme 3-1. Nucleophilic substitution towards cyclic thiosulfinate and cyclic disulfides.

Figure 3-1. Transition structures for the reaction of MeS– with cyclic thiosulfinates (3—10)a and cyclic disulfides (3—10)b.

‡ Figure 3-2. Plot of ΔG vs. –ΔGrxn of series (3—10)a and (3—10)b.

Figure 3-3. Activation, distortion, and interaction energies of TS-(3—10)a and TS-(3—10)b.

– Figure 3-4. ΔE⧧ vs. ΔEd⧧ and ΔE⧧ vs. ΔEi of the nucleophilic addition of MeS towards cyclic thiosulfinates and cyclic disulfides.

Figure 3-5. ΔS1—S2 bond length between reactant and transition state of cyclic thiosulfinates and cyclic disulfides vs. calculated distortion energy.

∗ Figure 3-6. Hyperconjugative nO→훔퐒퐒 orbital interaction.

Figure 3-7. Visual representation of MeS– HOMO, and cyclic thiosulfinate and cyclic disulfide

LUMO, and corresponding orbital energies and occupancies.

Figure 4-1. Cyclic thiosulfinates cross-link SOD1 variants via Cys111 residues on adjacent monomers.

Figure 4-2. Cyclic thiosulfinate cross-linking increases thermal stability and proceeds in the cellular environment.

Figure 4-3. Cyclic thiosfulinate cross-links do not perturb native structure.

Figure 4-4. Target engagement in mouse-model confirmed via LC-MS.

Figure 5-1. Common buffer components suppress MS signal.

Figure 5-2. Decision Tree for intact protein sample clean-up, preparation, and analysis.

Figure 5-3. Dilution (P.1), MWCO-ultrafiltration (P.2a), and precipitation (P.3) sample preparation protocols applied to common buffers.

Figure 5-4. Denatured vs. Native ESI-MS of carbonic anhydrase.

Figure 5-5. LC-MS of protein standard mixture prepared following the given SOP and separated on a Dionex UPLC with a Thermo Orbitrap Elite system using PLRP-S stationary phase.

Figure 5-6. LC-MS of bacteriorhodopsin-containing purple membrane of Halobacterium prepared following Protocol 5b and analyzed on an Agilent HPLC system coupled to a Thermo linear ion trap (LTQ) mass spectrometer.

Figure A1-1. Determination of 1,2-dithiane-1-oxide cross-linking half-life.

Figure A1-2. Confirmation of 1,2-dithiane-1-oxide cross-linking site.

Figure A1-3. Complete kinetics of 1,2-dithiane + SOD1 reaction.

Figure A1-4. 1,2-dithiane-1-oxide cross-links WT SOD1 in HeLa cells. Figure A1-5. 1,2-dithiane-1-oxide cell viability assay.

Figure A1-6. The cyclic thiosulfinate, β-lipoic acid cross-links SOD1 in Hep G2 cells, and does so more efficiently than the cyclic disulfide, α-lipoic acid.

Figure A1-7. Mass spectrometry assay of α-lipoic acid vs. β-lipoic acid cross-linking of SOD1.

Figure A1-8. Cross-linking efficiency of 1,2-dithaine-1-oxide in the presence of glutathione.

Figure A1-9. Mass Spectrometric assay of 1,2-dithiane-1-oxide/DTT competition.

Figure A1-10. Coordinates of optimized stationary points using M06-2X/6-31+G(d,p) IEF-

PCMH2O.

Figure A1-11. The energies of the HOMO and LUMOs of MeS–, 1,2-dithiane, and 1,2-dithiane-

1-oxide, respectively.

Figure A1-12. Cross-linking SOD1 using 1,2-dithiepane-1-oxide. 1,2-dithiepane-1-oxide forms rapid and complete cross-link of SOD1 following the same proposed mechanism as 1,2-dithiane-

1-oxide while 1,2-dithiepane does not.

Figure A1-13. 1H NMR of 1,2-dithiane.

Figure A1-14. 13C NMR of 1,2-dithiane.

Figure A1-15. 1H NMR of 1,2-dithiane-1-oxide.

Figure A1-16. 13C NMR of 1,2-dithiane-1-oxide.

Figure A1-17. High resolution mass spectrum of 1,2-dithiane-1-oxide.

Figure A1-18. 1H NMR of 1,2-dithiepane.

Figure A1-19. 13C NMR of 1,2-dithiepane.

Figure A1-20. 1H NMR of 1,2-dithiepane-1-oxide.

Figure A1-21. 13C NMR of 1,2-dithiepane-1-oxide.

Figure A1-22. High resolution mass spectrum of 1,2-dithiepane-1-oxide.

Figure A1-23. Full 1H NMR of β-lipoic acid.

Figure A1-24. Expanded 1H NMR of β-lipoic acid.

Figure A1-25. Full 13C NMR of β-lipoic acid.

Figure A1-26. Expanded 13C NMR of β-lipoic acid #1.

Figure A1-27. Expanded 13C NMR of β-lipoic acid #2.

Figure A1-28. Expanded 13C NMR of β-lipoic acid #3.

Figure A1-29. Expanded 13C NMR of β-lipoic acid #4.

Figure A1-30. High resolution mass spectrum of β-lipoic acid.

Figure A1-31. Structures of two regioisomers of β-lipoic acid.

Figure A2-1. Coordinates of optimized stationary points of reactants (3—10)a and (3—10)b using

M06-2X/6-311++G(d,p) IEF-PCMH2O.

Figure A3-1. 1,2-dithiane-1-oxide cross-links SOD1 at Cys111.

Figure A3-2. Analysis of α-lipoic acid shows presence of oxidized form.

Figure A3-3. 1H NMR of 1,2-dithiolane-1-oxide.

Figure A3-4. 13C NMR of 1,2-dithiolane-1-oxide.

Figure A4-1. Signal suppression curves of common components.

Figure A4-2. Fractionation of human whole-cell lysate prior to top-down mass spectrometry.

Figure A4-3. Antibody Buffer and Gentle Elution Buffer Ablate MS Signal; MWCO-

Ultrafiltration and Precipitation Rescue Signal.

Figure A4-4. Sample Preparation of Protein Mixture following Protocol 3 (MWCO-

Ultrafiltration).

Figure A4-5. Native vs. Denatured MS of AquaporinZ (AqpZ) from E. coli.

Figure A4-6. Denaturing vs. Native Analysis of Carbonic Anhydrase.

Figure A4-7. LC-MS of protein standard mixture run on Waters Acquity-Xevo G2-S QTOF.

Figure A4-8. LC-MS of protein standard mixture run on Waters nanoAcquity coupled to a Bruker

QTOF and a Bruker FT-ICR MS.

Figure A4-9. LC-MS of protein standard mixture run on Dionex UPLC coupled to three different orbitrap mass spectrometers.

Figure A4-10. LC-MS of protein standard mixture run on Dionex UPLC coupled to a Thermo

Orbitrap Fusion Lumos.

Figure A4-11. Capillary Zone Electrophoresis Separation of Protein Mixture.

Figure A4-12. LC-MS of Halobacterium salinarum prepared following Supplemental Protocol 5b.

Figure A4-13. LC-MS of Antibodies prepared following ultrafiltration and precipitation protocol.

Abbreviations

CNS Central nervous system

AD Alzheimer’s Disease

Aβ Amyloid β

PD Parkinson’s Disease

FTLD Frontotemporal lobar degeneration

TDP-43 TAR DNA-binding protein 43

ALS Amyotrophic lateral sclerosis

SOD1 Cu/Zn superoxide dismutase

FUS RNA-binding protein FUS/TLS

C9orf72 Chromosome 9 open reading frame 72

UBQLN2 Ubiquilin-2

HD Huntington’s Disease

ATTR Transthyretin amyloidosis

BACE Beta-secretase

PQC Protein quality control

UPS Ubiquitin-proteasome system

ALP Autophagosome-lysosome pathway sALS Sporadic amyotrophic lateral sclerosis

15 fALS Familial amyotrophic lateral sclerosis

FDA Food and drug administration

ROS Reactive oxygen species

HDX Hydrogen-deuterium exchange

MS Mass spectrometry

WT Wild type

CFTR cystic fibrosis transmembrane conductance regulator

TMAO Trimethylamine N-oxide

AQP2 aquaporin-2 water channel

NDI nephrogenic diabetes

CHO Chinese hamster ovary

DMSO Dimethyl sulfoxide

PAH Phenylalanine hydroxylase

PKU Phenylketonuria

GLA Galactosidase A

GCase β-Glucocerebrosidase

TTR Transthyretin

T4 Thyroxine

CYS Cysteine

TRP Tryptophan

DTME dithiobismaleimidoethane

BMOE bismaleimidoethane

DFT Density functional theory

EC Effective concentration eV Electron volts

Kcal Kilo calorie t1/2 Half life

QM Quantum mechanical h hours

Da Daltons

HepG2 hepatocellular carcinoma cells

DTT Dithiothreitol

ALA α-lipoic acid

BLA β-lipoic acid mL Milliliters

M Molar

NaOH Sodium hydroxide

TLC Thin-layer chromatography

NMR Nuclear magnetic resonance

EtOAc Ethyl acetate

CDCl3 Chloroform-D

NaCl Sodium chloride mp Melting point mCPBA meta-Chloroperoxybenzoic acid

HRMS High-resolution mass spectrometry

HPLC High-Performance Liquid Chromatography

FT-ICR Fourier transform ion cyclotron resonance

MaxEnt Maximum entropy

EDTA Ethylenediaminetetraacetic acid

Tris Tris(hydroxymethyl)aminomethane

LC-MS Liquid chromatography mass spectrometry

FA Formic acid

DMEM Dulbecco's Modified Eagle Medium

CO2 Carbon dioxide

RPM Revolutions per minute

PBS Phosphate buffered saline

SDS Sodium dodecyl sulfate

HRP Horseradish peroxidase

ECL enhanced chemiluminescence

β-ME Beta mercaptoethanol

EMEM Eagle’s minimum essential medium

PCM Polarizable continuum model

IEF integral equation formalism

TS Transition state

MM Molecular mechanics

MeS– Methyl thiolate

Cu Copper

Zn Zinc

Tm Melting point

MTT 3-(4,5-Dimethylthiazol-2-Yl)-2,5-Diphenyltetrazolium Bromide

SDS PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis

LC50 Lethal concentration

SAXS Small angle x-ray scattering

MD Molecular dynamics

DMPK Drug metabolism and pharmacokinetics

HCl Hydrochloric acid

MWCO Molecular weight cut-off

Q-TOF Quadrupole time-of-flight

UPLC Ultra-performance liquid chromatography

DSF Differential scanning fluorimetry

PCR Polymerase chain reaction

MALDI Matrix assisted laser desorption ionization

FBS Fetal bovine serum

BBB Blood brain barrier

MSI Mass spectrometry imagining

YFP Yellow fluorescence protein

CHESS Cornell High Energy Synchrotron Source

RNA ribonucleic acid

PTM Posttranslational modification

ESI Electrospray ionization

FASP Filter aided sample preparation

S/N Signal to noise

NRTDP National Resource for Translational and Developmental Proteomics

CZE Capillary zone electrophoresis

IEX Ion exchange

SEC Size exclusion chromatography

RIPA Radioimmunoprecipitation assay

RP-LC Reversed-phase liquid chromatography

CTDP Consortium for top-down proteomics

CMC Critical micelle concentration

RF Radio frequency

CASI Continuous accumulation of selected ions

ACN Acetonitrile

AGC Automatic gain control

CHAPS Dimethyl[3-( propyl]. azaniumyl}propane-1-sulfonate

CID Collision induced dissociation

AqpZ Aquaporin Z

LSF Least squared fit

Chapter 1

Introduction

1.1 Proteinopathies and Neurodegeneration

Proteinopathies are a group of diseases characterized by the presence of misfolded proteins within a subset of cells. Generally, these misfolded proteins result in the formation of oligomeric protein aggregates that lead to an imbalance of the cellular proteostasis mechanisms and proteotoxicity.1, 2 Often, this disease mechanism presents itself in neurodegenerative disorders and affects important cells within the central nervous system (CNS). The most common proteinopathies present in the human population include: Alzheimer’s Disease (AD), characterized by the presence of misfolded Amyloid β (Aβ) peptides and Tau proteins;3-5 Parkinson’s disease

(PD), characterized by the presence of misfolded α-Synuclein;6, 7 Frontotemporal lobar degeneration (FTLD), characterized by the presence of misfolded TAR DNA-binding protein 43

(TDP-43);8, 9 Amyotrophic lateral sclerosis (ALS), characterized by the presence of misfolded

Cu/Zn superoxide dismutase (SOD1),10, 11 TDP-43,12-14 RNA-binding protein FUS/TLS (FUS),15,

16 chromosome 9 open reading frame 72 (C9orf72),17, 18 or Ubiquilin-2 (UBQLN2);19, 20

Huntington’s disease, characterized by the presence of misfolded huntingtin protein;21, 22 and familial amyloidotic neuropathy, characterized by misfolded transthyretin.23, 24 The initial protein misfolding in all of these diseases can be attributed either to mutations in genes responsible for the translation of these proteins or modifications on already translated proteins due to aging, chemical or radiative exposure, or environmental stress. Although assessments of disease prevalence in the

United States often vary due to difficulty in diagnosis and often similar disease phenotypes, it is

21 estimated that ~1% of the US population is affected by these neurodegenerative disorders at a given time (Table 1-1).

Disease Aggregated protein Estimated Prevalence in U.S.

Alzheimer’s disease (AD)1 Amyloid-β, Tau 6,000,000

Parkinson’s Disease (PD)2 α-Synuclein 1,000,000

Amyotrophic Lateral SOD1; TDP-43; FUS; 30,000 Sclerosis (ALS)3 C9orf72; UBQLN2

Frontotemporal lobar TDP-43 60,000 degeneration (FTLD)4

Huntington’s Disease (HD)5 Huntingtin 30,000

Transthyretin Amyloidosis Transthyretin 5,000 (ATTR)6

Table 1-1. Table of the most common neurodegenerative diseases, proteins implicated with disease onset/progression, and estimated prevalence in the United States. 1Alzheimer’s

Association, 2American Parkinson’s Disease Association, 3Amyotrphic Lateral Sclerosis

Association, 4The Association for Frontotemporal Degeneration, 5Huntington’s Disease Society of America, 6Amyloidosis Foundation.

Despite prevalence, significant government funding, and decades of research, many of these neurological disorders still lack any effective treatment. They are often associated with a prolonged prodromal phase in which neuropathological and neurodegenerative changes precede neurodegenerative symptoms in patients.25 For example, over 75% of the dopaminergic activity in 22

Parkinson’s disease patients is lost before symptoms arise.26 Similarly, many studies have revealed that protofibril plaques of amyloid-β begin to form decades before clinical symptoms arise in

Alzheimer’s patients.27, 28 Drugs designed to inhibit the aggregation of amyloid-β routinely fail.

Most recently, Merck halted the development of the beta-secretase (BACE) 1 inhibitor

Verubecestat because patients involved in clinical trials (>85 years of age) did not receive treatment earlier enough and it failed to alleviate progression/symptoms. In the past two decades, over 100 drugs have met the same fate including Crenezumab (Genetech), Lanabecestat

(AstraZeneca/Eli Lilly and Company), and Solanezumab (Eli Lilly and Company). The disconnect between causality and onset of actual patient symptoms that is common to almost all forms of neurodegeneration remains one of the most challenging aspects of drug discovery. Consequently, new strategies for drug development are needed in the field of neurodegeneration.

1.2 Mechanism of Seeded Protein Aggregation and Neurotoxicity

The proteinopathies discussed above share a common etiology: aberrant protein conformation resulting in aggregation of misfolded protein and fibral formation. In many cases, aggregation of the aberrant protein is restricted to the central nervous system, though the protein is commonly expressed ubiquitously. For example, aggregates in mice expressing human SOD1 are found only in brain and spinal cord tissue and are absent the liver and in skeletal muscle.29 This aggregation is an irreversible stochastic process beginning years prior to the appearance of clinical symptoms and is a constant risk to the patient over their lifetime. The aberrant protein conformation exposes hydrophobic residues which are natively buried within the protein quaternary structure, introducing new “toxic” epitopes prone to interact with other species in the cell. Intermediates in the aggregation process (e.g., soluble oligomers and protofibrils) advance 23 neurodegeneration by altering the membrane permeability, suppressing electron transport, and triggering overproduction of reactive oxygen species.30, 31 Amyloid formation most likely occurs through seeded polymerization similar to that which is seen in prion diseases.32 The intracerebral infusion of brain extracts containing small amounts of Aβ into mice expressing the human Aβ precursor protein initiates the formation of senile plaques in an otherwise healthy mouse.33, 34

Similarly, transgenic mouse models with neuron specific expression of human amyloid precursor protein show an eventual “spreading” or diffusion of Aβ plaques into interconnected regions of the brain.35 Experiments have also been conducted to study the seeding propensity of SOD1 in

ALS models; the transduction of human apo-SOD1 (intrasubunit disulfide reduced) fibrils into cultured mouse neuroblastoma initiates aggregation of stably transfected human SOD1.36

In healthy cells, the protein quality control (PQC) system recognizes and degrades misfolded proteins. Approximately 30% of newly synthesized proteins in regularly-functioning cells are improperly folded and immediately enter this degradation pathway.37 Molecular chaperones such as heat shock proteins recognize non-native protein conformations and work to repair or refold these aberrant species, helping the cell maintain proteostasis.38 If refolding of a protein is not possible, often these chaperones interact with ubiquitin-proteasome system (UPS) or the autophagosome-lysosome pathway (ALP) to degrade improperly folded protein and clear them from the cytoplasm.39, 40 In neurodegeneration models, however, proteostasis is disrupted; the UPS and the ALP mechanisms become overwhelmed and cannot degrade improperly folded proteins fast enough. An obvious approach to reclaim proteostasis is the use of small molecules that upregulate proteasomal activity. A number of proteasome activity-enhancing molecules have been discovered yet only a handful of them can activate the proteasome in vivo:41 Sulforaphane activates the transcription factor Nrf2 and increases proteasomal activity;42 IU1 inhibits ESP14 and

24 increases proteasomal degradation;43 Rolipram activates protein kinase A and reduces levels of aggregated tau (Figure 1-1).44 Still, however, once aggregates are formed, they are often resistant to degradation even if proteasomal activity is increased.

Figure 1-1: Chemical Structures of in vivo proteasomal activators. (a.) Sulforaphane (b.)

Rolipram (c.) IU1

Amyloid formation is the final step in the aggregation cascade. Most commonly, amyloids consist of repetitive parallel (in-register) or anti-parallel (out-of register) β-sheets. These structures are thermodynamically stable, existing in a low energy well on the potential energy surface, which contributes to their resistance to proteasomal degradation.45

1.3 Amyotrophic Lateral Sclerosis

Amyotrophic lateral sclerosis is a fatal neurodegenerative disease that affects motor neurons in the central nervous system. The disease was first described in 1869 by Jean-Martin

Charcot, a French neurologist. It became better known in the United States when the famous New

York Yankee, Lou Gehrig, was suddenly diagnosed with ALS in 1939. Until his disease on-set,

Gehrig was known as the “iron horse” of baseball, playing in 2,130 consecutive games (a record left standing until 1995). On May 2, 1939, however, he voluntarily sat himself from a game after

25 struggling to play well, stating it was “for the good of the team.” Two years later, Gehrig passed away.

The fast progression of ALS and the short life expectancy after diagnosis is common in many cases. On average, patients live 2-5 years after diagnosis. The mean age of disease onset is between 50-65 though some early onset patients can show symptoms as young as 30. Disease symptoms first appear as general muscle weakness and progress to complete paralysis eventually leading to death from respiratory failure. The incidence of ALS is approximately two per 100,000 people worldwide (http://www.alsa.org).

Like most neurodegenerative diseases, ALS can be classified as either familial ALS (fALS) caused by genetic mutations in a disease-related gene leading to expression of variant protein, or sporadic (sALS) which to date has no definitive disease mechanism. Sporadic ALS accounts for

80-90% of cases, with only 10-20% of cases having a known genetic basis. Currently, there is no cure for either fALS or sALS. Riluzole, approved by the FDA in 1995, is a neuroprotective glutamate blocker and extends life by months.46 Edaravone, a free radical and peroxynitrite scavenger, was approved by the FDA in 2018. Studies are ongoing to measure the effectiveness of the drug.47

1.4 Aggregation of Superoxide Dismutase (SOD1) and ALS

The research presented in this thesis focuses on the aggregation of SOD1 and its implications in both familial amyotrophic lateral sclerosis (fALS) and sporadic ALS (sALS).

Genetic mutations in the gene that expresses SOD1 (21q22.11) were first implicated in ALS disease onset and progression in a 1993 publication by Rosen and coworkers.48 Prior to this discovery, no other gene had been associated with the disease. Since this connection was made,

26 over 180 mutations in the SOD1 gene have been linked to fALS accounting to 2-7% of all ALS cases.49

– SOD1 is a homodimeric metalloenzyme that catalyzes the conversion of superoxide (O2 )

50, 51 to oxygen (O2) and hydrogen peroxide (H2O2). Initially, the loss of SOD1 enzymatic activity, which would result in increased levels of reactive oxygen species (ROS) in the cell, was thought to be the basis of variant SOD1 toxicity. ALS mouse models expressing SOD1G93A, however, show comparable SOD1 enzymatic activity. Furthermore, SOD1 knockout mice do not develop motor neuron disease.52 Instead, it is believed that SOD1 fALS results from a toxic gain of function. The leading hypothesis of toxicity is the destabilization of SOD1 homodimer resulting in dimer dissociation and aggregation (Figure 1-2).53-57

Figure 1-2. Reprinted with permission from John Wiley and Sons Inc. from: Schmitt ND, Agar

JN. Parsing disease-relevant protein modifications from epiphenomena: perspective on the structural basis of SOD1-mediated ALS. J. Mass Spectrom. 2017;52(7):480-491. Hypothesized model for SOD1-mediated fALS disease progression. (Top) Mutations at various locations in 27

SOD1, as demonstrated by HDX-MS in Molnar et al. 2009, lead to perturbation of the electrostatic loop (residues 121-142). (Middle) This perturbation exposes a toxic epitope, which enables a gain- of-interaction, possibly between a region of the electrostatic loop (red, rest of loop is pink) of one

SOD1 homodimer and the exposed beta strand edges of beta strands V and VI (green and blue) of another SOD1 dimer, as demonstrated by the Hart group (Elam, J.S., et al. 2003, adapted from pdb entry 1OZU). (Bottom) This leads to aberrant interactions of SOD1 dimers, resulting in greater unnatural quaternary structure, fibrillization, and aggregation.

Through a combination of x-ray crystallography and hydrogen-deuterium exchange mass spectrometry (HDX-MS) experiments, the structural consequences of a series of common genetic mutations in SOD1 were identified.55, 58-61 HDX is an increasingly popular analytical technique used to study changes in protein structure caused by mutations, modifications, or ligand binding.

By subjecting native proteins to incubation with deuterium, all solvent accessible amide hydrogens on the protein backbone are exchanged with deuterium. Following a reaction quench at low pH and digestion by the endoproteinase pepsin, peptide mapping is performed to determine the in- solution dynamics of a protein (i.e., more solvent accessible residues/higher exchange indicates less folded proteins). A study examining the in-solution dynamics of thirteen fALS-associated

SOD1 variants identified a common change in protein structure: the perturbation of the electrostatic loop located between residues 120-140 (Figure 1-2).55 Additionally, metal-deficient variants (e.g., SOD1S134N, SOD1G85R, SOD1D124V, SOD1D125H, and SOD1H46R) contained a destabilized metal-binding loop (residues 60-100). The perturbation of the electrostatic loop introduces a toxic epitope which results in non-native interactions between SOD1 dimers.

Bosco and coworkers developed a monoclonal antibody (C4F6) against fALS SOD1G93A which interacts specifically with the toxic epitope introduced by the destabilization of the electrostatic loop (i.e., stable SOD1WT does not interact with C4F6).53 Many other destabilized variants showed similar affinity towards the C4F6 antibody (e.g., SOD1D90A, SOD1G85R, SOD1A4V, etc.) confirming the perturbed electrostatic loop is common amongst ALS SOD1 variants. In one of the first studies that identified SOD1 as a potential common etiology between fALS and sALS

Bosco et al. showed that oxidized SOD1WT (on either cysteine-111 or tryptophan-32) also interacts with the C4F6 antibody, suggesting chemical modification of SOD1WT can lead to the same destabilization of the native conformation seen in fALS variants. Furthermore, immunohistological studies, using the same destabilized SOD1-specific antibody, identified large

29 aggregates of misfolded SOD1WT in sALS patient tissue (Figure 1-3). These data confirm the destabilization of the electrostatic loop is the main contributing factor to SOD1 ALS.

Figure 1-3. Reprinted by permission from Springer Nature: Springer Nature, Nature

Neuroscience from: Bosco DA, Morfini G, Karabacak NM, et al. Wild-type and mutant SOD1 share an aberrant conformation and a common pathogenic pathway in ALS. Nature Neuroscience.

2010;13(11):1396-403. (a,d,e,f) C4F6 positive staining is shown for 4 SALS cases (SALS1-4).

The positive staining observed for SALS1 (shown in panel a) is lost when C4F6 is excluded from the staining protocol (b) or when the alternative SOD1-mutant specific A9G3 antibody is employed (c). Representative control cases (g-h) and an SOD1-negative FALS case (i) illustrate the lack of positive C4F6 reactivity for such cases. In total, in 4/9 SALS cases exhibited positive

C4F6 staining, whereas 0/17 control cases exhibited positive staining. The perturbation of the electrostatic loop is followed by dimer dissociation. This mechanism is common amongst many proteinopathies. Dimer dissociation results in the demetallation of SOD1 monomers and multimeric assembly of apo-SOD1 monomers to form higher-order aggregates.

1.5 Protein Stabilization and Chemical/Pharmacological Chaperones

Since protein destabilization leading to aggregation is a common mechanism in neurodegenerative disorders, the development of molecules that stabilize native protein structure and prevent aggregation is an accepted therapeutic strategy. Chemical chaperones and osmolytes are low molecular weight compounds (e.g., polyols, amines, and organic solvents) that are commonly used in in vitro assays because of their protein-stabilizing properties. In many cases chemical chaperones have been used to stabilize variant proteins with perturbed structure and rescue activity. The ΔF508 genetic variant of the CFTR protein (cystic fibrosis transmembrane conductance regulator) shows a reduced folding efficiency resulting in misfolded protein, decreased transmembrane activity, and increased degradation. Cells expressing ΔF508 CFTR incubated with high concentrations of glycerol, butyrate, and other chemical chaperones increases the concentration of functional surface ΔF508 CFTR most likely due to enhanced folding and decreased degradation of misfolded protein.62-64 Moreover, injection of ΔF508 CFTR-expressing mice with Trimethylamine N-oxide (TMAO) showed increased forskolin dependent chloride

31 channel activity.65 Similarly, misfolded mutants of aquaporin-2 water channel (AQP2) have been implicated in up to 10% of all nephrogenic diabetes cases (NDI). Incubation of CHO cells expressing the A147T and the T126M variants of AQP2 with 1 M glycerol, 100 mM TMAO, or

280 nM DMSO led to increased concentration of properly folded AQP2.66, 67

Chemical chaperones, however, are only efficient at high concentrations (i.e., 1 M glycerol) and act in a non-specific fashion binding to and stabilizing virtually any protein and limiting their potential use in vivo. The development of pharmacological chaperones, molecules that work at much lower concentrations, offer the same increase in folding efficiency but are specific to a particular target.68 Mutations in the enzyme phenylalanine hydroxylase (PAH) result in the phenylketonuria (PKU), a disease associated with an abnormal phenotype that includes poor skin pigmentations, seizures, developmental delay, and intellectual impairment due to a patient’s inability to metabolize the amino acid phenylalanine.69 Treatment of PKU patients with the PAH cofactor, tetrahydrobiopterin, stabilizes the variant conformation and increases metabolic activity.70

A number of pharmacological chaperones have been in development to treat conformational diseases. Migalastat (tradename Galafold) was developed by Amicus therapeutics as a potential therapy for Fabry disease, a rare genetic disorder caused by mutations in the gene that expresses galactosidase A (GLA).71 Migalastat reversibly binds the active site of variant GLA, stabilizing the functionally deficient enzyme and promoting its trafficking to lysosomes where it is active.72, 73 As of August 2018, Migalastat was approved under accelerated approval by the FDA as a daily 123 mg capsule. Similarly, mutations in the lysosomal protein β-Glucocerebrosidase

(GCase) result in the excessive accumulation of glucosylceramide in the liver, spleen, bone, and bone marrow and lead to Gaucher’s Disease.74 Non-covalent active site inhibitors, including

Zavesca (FDA approved in 2003) and ambroxol, promote mutant GCase stability, increase enzyme activity, and decrease glucosylceramide levels.75-77

1.6 Tafamidis and the stabilization of Transthyretin

Unlike the compounds listed above, which rescue activity of a misfolded protein by decreasing degradation and increasing the proper localization of the protein, pharmacological chaperones have also been used to inhibit the aggregation of aberrant proteins. Tafamidis, developed by Jeffrey Kelly and coworkers at Scripps, is an approved pharmacological chaperone that functions by stabilizing the native tetramer of transthyretin (TTR) and preventing the

78 formation of TTR amyloids. Rate-limiting dissociation of the TTR tetramer at the thyroxine (T4) binding site results in unstable dimers that rapidly dissociate into monomers.79, 80 These monomers have the propensity to form soluble oligomers and amyloid fibrils, leading to the onset of familial amyloidotic polyneuropathy and familial amyloidotic cardiomyopathy (Figure 1-4a).81 Until the early 2000’s, the main therapeutic strategy to ameliorate these proteinopathies was liver transplantation since TTR is mainly synthesized in the liver.82, 83

It was first observed by Sato and coworkers that increased TTR binding of its cofactor,

WT thyroxine (T4), at the destabilized dimer interface increased the thermostability of both TTR and

TTRV30M.84 This finding prompted a thorough screening and structure-based drug design strategy to find T4 analogs that bind tightly and selectively to the T4 binding site at the dimer interface which could potentially stabilize the native tetramer.85-90 The result was the discovery of tafamidis, a compound that stabilizes the weak dimer-dimer interface of many TTR variants through hydrophobic and electrostatic interactions (Figure 1-4b).

In a randomized, placebo-controlled, double-blind trail in patients with TTR familial amyloidotic polyneuropathy, tafamidis failed to achieve the prespecified statistical significance.

Despite this, however, patients treated with tafamidis showed reduction in neurologic deterioration, preservation of nerve fiber function, improved nutritional states, and TTR stabilization.91 Most recently, in a multicenter, international, double-blind, placebo-controlled phase-3 trial for tafamidis in transthyretin amyloid cardiomyopathy patients, all patients that received tafamidis treatment (either with 80 mg or 20 mg for 30 months) showed decreased all- cause mortality and rates of cardiovascular-related hospitalizations.92 These clinical results confirm the potential therapeutic benefits for the development of pharmacological chaperones and

34 kinetic stabilizers for a number of other proteinopathies, including ALS, that result from protein conformational changes.

Figure 1-4. Reprinted with permission from Proceedings of the National Academy of Sciences of the United States of America. Bulawa CE, Connelly S, Devit M, et al. Tafamidis, a potent and selective transthyretin kinetic stabilizer that inhibits the amyloid cascade. Proc Natl Acad Sci U S

A. 2012;109(24):9629-34. (a.) The TTR amyloid cascade. Amyloid formation by TTR requires rate-limiting tetramer dissociation to a pair of folded dimers, which then quickly dissociate into folded monomers. Partial unfolding of the monomers yields the aggregation-prone amyloidogenic intermediate. The amyloidogenic intermediate of TTR (Lower Right) retains much of its native structure (shown in purple), probably with some β-strand dissociation (shown in turquoise). The amyloidogenic intermediate can misassemble to form a variety of aggregate morphologies, including spherical oligomers, amorphous aggregates, and fibrils. (b.)Tafamidis binding to the

TTR tetramer (Upper Left) dramatically slows dissociation, thereby efficiently inhibiting aggregation. Crystal structure of tafamidis bound to TTR. The coordinates are available in the

Protein Data Bank under accession code 3TCT. 3D ribbon diagram depiction of the TTR tetramer with tafamidis bound. The four TTR monomers are individually colored. (c.) Magnified image of tafamidis bound in one of the T4-binding sites. Connolly analytical surface representation

(translucent gray, hydrophobic; translucent purple, polar) depicts the hydrophobicity of the binding site. The 3,5-chloro groups are placed in the HBPs 3 and 3′ making hydrophobic interactions, whereas the carboxylate of tafamidis engages in water-mediated H-bonds with the Lys15/15′ and

Glu54/54′ residues of TTR represented by dotted lines (Lys15′-water H-bond not shown owing to tafamidis orientation)

1.7 Identification of compounds that stabilize the SOD1 dimer

Due to the success Kelly et al. had in the stabilization of TTR, the stabilization of the SOD1 homodimer has been accepted as an attractive therapeutic approach to fALS. Unlike TTR, however, SOD1 has no known natural cofactors. The earliest approach to develop a small molecule pharmacological chaperones that would bind the SOD1 dimer interface (at the

Val7/Val148 interface) and stabilize SOD1 used in silico screening of 1.5 million compounds with hypothesized specificity towards the region adjacent to the dimer interface of SOD1.93, 94 Through this screening, 15 compounds were identified to significantly stabilize SOD1A4V in vitro. However, questions arose as to whether these compounds were actually specific to SOD1 or could potentially target numerous other proteins. Compounds 1 and 2 are representative compounds from this screen. Additional screens were performed to optimize specificity towards SOD1, resulting in compound 3. Screens to optimize the pharmacological properties (i.e., blood-brain penetration, potency, and metabolic stability) identified several other promising compounds that were blood- brain penetrant and potent (nanomolar IC50s) (Compounds 4-7) (Figure 1-5). Mice expressing hSOD1G93A were dosed with compound 5 (R=O) and showed a significant increase in survival

(142.5 days in treated mice vs. 125.7 days in untreated mice). Though these compounds showed efficacy both in vitro and in vivo, the direct mechanism in which they were acting was never reported.

Figure 1-5. Representative compounds identified through in silico screen that inhibit SOD1 aggregation.

In later studies, x-ray crystallography was used to determine the actual binding site of the above molecules and their analogs. Though co-crystallization of these compounds with SOD1 had proved difficult, Hasnain et al. recognized that many of them contained an uracil group. They incubated SOD1L38V with 5’monophospate (UMP) and solved the crystal structure of the resulting complex, discovering that it was not binding at the intended Val7/Val148 dimer interface, but in a groove between the electrostatic and Zn-binding loop.95,96 Furthermore, co-crystallization of methoxy-5-methylaniline and L-methionine bound to SOD1L38V revealed a new binding site near tryptophan 32. Although the determined binding sites can lead to compounds that reduce aberrant

38 self-association of two dimers, they cannot inhibit the dissociation of the SOD1, the first step of aggregation.

1.8 Stabilization of SOD1 with homobifunctional maleimides

While compounds binding at the Val7/Val148 dimer interface resulted in no potential

SOD1 stabilizing drugs, attempts were made to target a different SOD1 dimer-interface motif,

Cys111A and Cys111B (located approximately 9 angstroms apart). In a proof of concept experiment,

Auclair et al. used homobifunctional maleimide cross-linkers (e.g., bismaleimidoethane and

54 dithiobismaleimidoethane) to target the Cys111 pair and covalently stabilize the SOD1 dimer. The covalent bridge formed between the two opposing cysteine residues stabilized disease-associated variants of SOD1 (i.e., SOD1G93A and SOD1G85R) by up to 40 °C (Figure 1-6). Additionally, after incubation with these commercial cross-linkers, the inactive, mainly monomeric SOD1G85R, showed WT-like activity.

Figure 1-6. Reprinted with permission from Proceedings of the National Academy of Sciences of the United States of America. Auclair JR, Boggio KJ, Petsko GA, Ringe D, Agar JN. Strategies for stabilizing superoxide dismutase (SOD1), the protein destabilized in the most common form of familial amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A. 2010;107(50):21394-9. Model of SOD1 after cross-linking DTME with adjacent C111 residues at the dimer interface. (Top, a.)

Free (unreacted) DTME. (b.) Maleimide-chemistry mediated, “cross-linked” reaction product of

SOD1 and DTME, with DTME bridging individual SOD1 monomers. SOD1 Cys111 constituents are shown in red, and the two SOD1 monomer/subunits are labeled “suba” and “subb.” Cysteine111 rotomers are oriented such that the cysteinyl sulfur spacing is 13 Å (the minor rotomer observed in crystal structures). (c.) Thiol-disulfide exchange- and maleimide-chemistry mediated cross- linked reaction product of SOD1 and DTME. Cysteine111 rotomers are oriented such that the cysteinyl sulfur spacing is 9 Å (the major model, including site of cross-linking, was confirmed via LC-FTMS analysis (d-g) Stabilization of fALS-associated SOD1 variants by chemical cross- linking. Stability of mutant SOD1 measured by thermofluorescence assay.

The maleimides used in these studies, however, are toxic with LD50s in mice of 9 mg/kg with renal, hepatic, neurologic and hematologic toxicities as the principal effects.97 Though this strategy for stabilizing SOD1 via its adjacent cysteine111 residues proved successful, identification of molecules that could similarly target this binding pocket, but with better toxicological profiles, was required.

1.9 Cyclic disulfides

Considering the previous attempts to identify molecules that target SOD1 and the success

Auclair et al. had in stabilizing SOD1 variants, we searched for new molecules with the same cross-linking propensity as commercial cross-linkers but with more promising toxicological properties. We rationalized that cyclic disulfides, a group of compounds that have either been used as drugs already or are present in foods we eat, could potentially cross-link pairs of closely-spaced cysteine residues and would have low toxicity in vivo (Figure 1-7). Oltipraz (4-methyl-5-(pyrazin-

2-yl)-3H-1,2-dithiole-3-thione) was previously used in human as an antischistosomal agent and has shown to inhibit carcinogenesis in bladder, colon, breast, stomach, and skin cancer models.98,

99 Sufarlem (5-(4-Methoxyphenyl)-3H-1,2-dithiole-3-thione) is used to treat xerostomia (dry mouth)100, 101 and has been studied as a potential treatment to some cancers.102 Both α-lipoic acid

(ALA) and asparagusic acid are natural products and have been deemed safe for consumption.

ALA is often used in the treatment of diabetic polyneuropathy (at doses up to 2 g/day) and is sold over-the-counter as a universal antioxidant.103-105 Although the mechanism of action of these particular compounds is unknown, and the likelihood that they modify cysteines or pairs of cysteines in vivo is high, their toxicological properties are promising.

Figure 1-7. Cyclic disulfides currently in use as drugs. Oltipraz have been used as an antischistosomal agent. Sufarlem is known to increase production of saliva. Α-lipoic acid reduces symptoms of polyneuropathy.

The subsequent chapters introduce the mechanism in which these compounds cross-link pairs of closely spaced cysteine residues. We modify the most basic cyclic disulfide, 1,2-dithiane,

42 to its thiosulfinate form, 1,2-dithiane-1-oxide, and increase the rate of cross-linking by 104-fold.

We also rationalize the rate-enhancement of the thiol-disulfide exchange step of cross-linking using quantum mechanical calculations. Finally, we apply this cross-linking chemistry to destabilized SOD1 variants and show that: 1. Cross-linking of SOD1 variants is efficient, 2. Cross- link formation increases stability of SOD1 variants, 3. Cross-linking proceeds in cells in the presence of cellular reductants, 4. Cross-linking works in an ALS mouse model. Additionally, we introduce robust intact protein mass spectrometry protocols that can be used in the study of covalent drugs.

Chapter 2

Cyclic Thiosulfinates and Cyclic Disulfides Selectively Cross-Link Thiols While Avoiding Modification of Lone Thiols

Daniel P. Donnelly§,†, Matthew G. Dowgiallo§, Joseph P. Salisbury§,†, Krishna C. Aluri§,†, § §,† § § §,† Suhasini Iyengar , Meenal Chaudhari , Merlit Mathew , Isabella Miele , Jared R. Auclair , Steven A. Lopez§, Roman Manetsch§,‡, Jeffrey N. Agar§,†,‡,* § Department of Chemistry and Chemical Biology, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115, United States † Barnett Institute of Chemical and Biological Analysis, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115, United States ‡ Department of Pharmaceutical Sciences, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115, United States

Reprinted with permission from

Cyclic Thiosulfinates and Cyclic Disulfides Selectively Cross-Link Thiols While Avoiding Modification of Lone Thiols, Journal of the American Chemical Society 2018 140 (24), 7377- 7380 DOI: 10.1021/jacs.8b01136. Copywrite 2018 Journal of American Chemical Society.

2.0 Statement of Contribution

Experimental contributions made to this chapter by Daniel P. Donnelly are as follows: all protein purification, all in vitro assays, all mass spectrometry method development and analysis, all cell culture, and all in cellulo assays. The manuscript was written by Daniel P. Donnelly and Jeffrey

N. Agar with careful review and contributions made by Roman Manetsch, Steven A. Lopez, and

Matthew G. Dowgiallo. All figures were prepared by Daniel P. Donnelly.

2.1 Abstract

This work addresses the need for chemical tools that can selectively form cross-links.

Contemporary thiol-selective cross-linkers, for example, modify all accessible thiols, but only form cross-links between a subset. The resulting terminal “dead-end” modifications of lone thiols are toxic, confound cross-linking-based studies of macromolecular structure, and are an undesired—and currently unavoidable—byproduct in polymer synthesis. Using the thiol pair of

Cu/Zn-superoxide dismutase (SOD1) we demonstrated that cyclic disulfides—including the drug/nutritional supplement lipoic acid—efficiently cross-linked thiol pairs but avoided dead-end modifications. Thiolate-directed nucleophilic attack upon the cyclic disulfide result-ed in thiol- disulfide exchange and ring cleavage. The resulting disulfide-tethered terminal thiolate moiety either directed the reverse reaction, releasing the cyclic disulfide, or participated in oxidative disulfide (cross-link) formation. We hypothesized—and confirmed with density functional theory

(DFT) calculations—that mono-S-oxo derivatives of cyclic disulfides formed a terminal sulfenic acid upon ring cleavage that obviated the previously rate-limiting step, thiol oxidation, and accelerated the new rate-determining step, ring cleavage. Our calculations suggest that the origin of accelerated ring cleavage is improved frontier molecular orbital overlap in the thiolate-disulfide interchange transition. Five to seven-membered cyclic thiosulfinates were synthesized and efficiently cross-linked up to 104-fold faster than their cyclic disulfide precursors; functioned in the presence of biological concentrations of glutathione; and acted as cell-permeable, potent, tolerable, intracellular cross-linkers. This new class of thiol cross-linkers exhibited click-like attributes including, high yields driven by the enthalpies of disulfide and water formation, orthogonality with common functional groups, water-compatibility, and ring strain-dependence.

2.2 Introduction

Thiol-ene reactions are prevalent in applications requiring thiol cross-linking.106 Synthetic applications of thiol-ene cross-linking reactions include: self-healing polymers,107 nanogels,108 thermosetting polymers, hydrogels,109 and dendrimers.110 Prevalent biochemical applications of thiol-ene cross-linking include functionalizing or stabilizing biotherapeutics in vitro,111 and probing high-order protein structure and protein-protein interactions.54 One shortcoming of these thiol-ene cross-linking tools—in fact all current tools—is that they are not cross-linking selective.

These tools will form terminal “dead-end” modifications unless two functional groups happen to be within their reach.112 Dead-end modifications are toxic in vivo; in particular the modification of essential catalytic cysteines (e.g. phosphatases and cysteine [Cys] proteases) and the creation of

“non-self” epitopes that increase the risk of an adverse immune response.113 This inherent toxicity, and poor cell permeability, have stymied in vivo cross-linking. To enable the in vivo use of the cross-linking applications described above, our objective is a chemical tool with improved selectivity for cross-linking thiols (i.e., higher cross-linking efficiency).

In addition, we introduce in vivo thiol cross-linking as a strategy for pharmacological protein stabilization, and a long-sought, non-inhibitory alternative to stabilization with substrate analogues.114 A number of diseases, including familial Amyotrophic Lateral Sclerosis (fALS), are associated with loss of quaternary structure and protein destabilization (seen with Cu/Zn- superoxide dismutase (SOD1) mutations). Multimer stabilization—exemplified by the substrate/cargo analogue, transthyretin-stabilizing drug tafamidis115—is a therapeutic strategy in these diseases.

We used thiol-ene cross-linkers in a proof-of-concept study to demonstrate that cross- linking the thiol pair (Cys 111A + B, 8 Å apart) on adjacent subunits of SOD1 could stabilize fALS-

SOD1 variants by up to 40 °C.54 This approach also rescued the enzymatic activity of inherently inactive fALS SOD1 variants.54 Through a computational screen of human protein structures, 20 additional multimeric proteins with quaternary structures that could be stabilized by intersubunit cysteine cross-linking were discovered.

We surveyed drugs to identify mechanisms for selective thiol binding that can be tolerated in vivo. One recent approach to drug design is to attach a soft, sometimes finely “tuned” Cys- selective electrophile113 to a high-affinity binder.116, 117 Unfortunately, as is often the case, the lack of high-affinity SOD1 binders ruled out this structure-based approach. The other mechanism used by thiolate-selective drugs, disulfide bond formation, is the most prevalent and mature

(disulfiram/Antabuse treatments began in 1948).118 Inactive prodrugs are transformed into thiols, which, after spontaneous oxidation, form long-lived disulfide bonds between the drug metabolite’s sulfenic acid and a target protein’s cysteine thiolate. Some drugs form disulfides with enzyme active site Cys (e.g. disulfiram119) and others with allosteric Cys (e.g. omeprazole/Prilosec,120 prasugrel/Effiant,121 etc.). Unfortunately, the obvious strategy of binding two of these drugs to create a bifunctional cross-linker would not result in a tool that could avoid dead-end modifications.

Having ruled out cross-linkers composed of existing thiolate-selective warheads, we sought molecules with the ability to minimize dead-end Cys modifications as a starting point for cross-linkers. Cyclic disulfides are the only thiolate-selective scaffold we knew of that can form transient bonds (i.e., can avoid dead-end modifications without the aid of other molecules). Cyclic disulfide chemistry was extensively characterized in a series of publications by the Whitesides’ group.122-124 These studies demonstrated the high effective concentration (EC – i.e., the entropically-driven propensity to remain oxidized and cyclic; specifically the Keq between a dithiol

48 forming a cyclic disulfide and a dialkyl disulfide forming two thiols) of cyclic disulfides results in transient binding to lone thiols. Moreover, the Keq of cyclic disulfide binding to lone thiols (i.e.,

122, 123 “Keq dead-end”) is highly ring strain-dependent, varying over three orders of magnitude.

Cyclic disulfide-tethered drug cargos can even be transported across the cell membrane via reversible binding to a transferrin receptor Cys.125, 126 Cyclic disulfides can be tolerated at doses up to 5 g/day/person and have an LD50 in the range of ethanol, fructose, and sodium chloride.

We reasoned that cyclic disulfides could cross-link thiol pairs while minimizing dead-end

Cys modifications. A reversible SN2-type attack of a Cys thiolate upon a cyclic disulfide would result in thiolate-disulfide interchange concomitant to ring opening to form a terminal thiolate. If this terminal thiolate was within binding distance of a sulfenate (i.e., oxidized Cys), a cross-link could form by their condensation to a disulfide bond (Figure 2-1, Mechanism II).127 Otherwise, the cyclic disulfide would be released by the reverse (thiolate-disulfide interchange) reaction

(Figure 2-1, Mechanism I). Furthermore, if mono-S-oxo cyclic disulfides (cyclic thiosulfinates) were used instead, thiolate oxidation, the slowest step of the cross-linking reaction sequence, would not be required (Figure 2-1, Mechanism III). Instead, thiolate-disulfide interchange with a cyclic thiosulfinate would lead directly to a disulfide-bound terminal sulfenic acid, which would rapidly form a cross-link by condensing with the second, nearby thiolate, releasing water.127

Thiolate-disulfide interchange proceeds through a linear trisulfide-like intermediate comprised of nucleophilic-(Sn), center-(Sc), and leaving group-(Sl) sulfurs with Brønstead coefficients (β) of

~0.5, -0.3, and -0.7, respectively.123, 128 The Brønstead coefficients of -0.7 and -0.3 for the leaving group and central sulfur, respectively, and our quantum mechanical calculations (Figure 2-1,

49 bottom panel) imply that the rate of thiolate-disulfide interchange is highly sensitive and inversely proportional to the pKa of Sl.

Figure 2-1. Proposed mechanism of thiol cross-linking using cyclic disulfides (blue) and cyclic thiosulfinates (red). (Top) Formation of the first disulfide bond is reversible. Dead-end modification is minimized by entropically favorable ring closure (I). Cross-linking proceeds through condensation of CysA, and a sulfenic acid derived from either rate limiting S-oxidation of thiolateB (II) or a cyclic thiosulfinate (III). (Bottom) A potential energy surface for the non- enzymatic reaction according to the mechanism proposed in the top panel. The free energy values are computed using M06-2X/6-311+G(d,p) IEF-PCMH2O//M06-2X/6-31+G(d,p) IEF-PCMH2O.

#The dotted blue line indicates an activation barrier derived from the zeroth-order half-life kinetics 50 of the oxidation of 5. Details on energy value of structures 3a and 3b are given in SI Section II,

kcal mol-1 experimentally-derived enthalpy from reported values of heat of formation of -132װ .O water and experimentally determined bond energy of dimethyl disulfide.25-26 ◊The activation barrier for TS(3a→4) is 0.5 kcal mol-1 higher than TS(3b→4) (11.5 kcal mol-1) (Figure A1-10).

Transition structures for the nucleophilic addition of MeS– to 1 and 5 are shown. The bond lengths

–1 and energy values are reported in Å and kcal mol , respectively.

2.3 Results ad Discussion

We used DFT calculations to understand the origins of the nearly 110-fold reactivity increase of 1 towards lone thiolates over 5 and the observed reversibility. We performed a conformational search on the starting materials, transition structures, and intermediates (full computational details are provided in the 2.5.13). We represented a Cys thiolate as methyl thiolate to reduce the conformational search space and computation time. The 2.9 kcal mol–1 lower activation free energy of TS(1→2b) vs. TS(5→2a) is due to more favorable frontier molecular orbital interactions in the transition state. The σ* orbital (Figure A1-11) of 1 is 0.21 eV lower in energy than that of 5, thus lowering the energy of TS(1→2b). The ring-opening step for 1 and 5 are endergonic (∆G = 4.6 and 5.4 kcal mol–1, respectively) and reversible, consistent with experiments. The formation of the cross-linked product and water is thermodynamically favored, over 130 kcal mol–1 lower in free energy than the reactants (1 and 5).129, 130 Upon ring-opening of

1,2-dithiane, 2a is slowly oxidized (t1/2 = 10 days) to 3a, which is rate-determining and affords the final cross-linked product. The QM results highlight two major implications for the low pKa sulfenic acid group. First, the nucleophilic attack on the non-oxo-thiosulfinate S, which releases sulfenic acid, is >10-fold faster and therefore more likely than the attack on the more electrophilic

51 sulfinyl sulfur. Second, in addition to eliminating the need for rate-limiting thiol oxidation, thiosulfinates, through the sulfenate intermediate generated, also increase the rate of thiolate- disulfide interchange.

To test the cross-linking activity of cyclic disulfides and cyclic thiosulfinates, 1,2-dithiane and 1,2-dithiane-1-oxide were synthesized and incubated with SOD1, a homodimeric protein containing a solvent accessible thiol pair (Cys111A + B, 8 Å apart) on adjacent subunits. The reaction was monitored using a mass spectrometry (MS) assay that uses a combination of increased voltage within the region of hypersonic gas expansion and brief treatment with 10% formic acid to create exclusively monomeric SOD1 or covalently cross-linked SOD1 dimer. Formic acid also quenches the reaction, transforming any reactive thiolates to unreactive thiols. As a result, if no reaction occurs, only apo SOD1 monomer is detected (Figure 2-2, top). Consistent with our hypothesized mechanism: 1) 1,2-dithiane-1-oxide, but not 1,2-dithiane, resulted in rapid and complete dithiolate cross-linking (half-life ~2-3 min) of SOD1 (Figures 2-2, middle and A1-1); 2) no binding to single Cys residues (SOD1 has free Cys111 and Cys6) was observed in any sample with either compound; 3) No cross-linking was observed without the loss of the oxygen from the S-oxo of

1,2-dithiane-1-oxide; 4) No cross-linking was observed when incubating 1,2-dithiane-1-oxide with

C111S SOD1 (Figure A1-2). Given sufficient time for thiolates to be oxidized to sulfenic acid

(which occurs on the order of days-weeks), even 1,2-dithiane was expected to cross-link SOD1.

After 72 h of incubation with 1,2-dithiane, 11% of SOD1 had formed the expected covalent dimer

(Figure A1-3). Comparable results were observed from the incubation of SOD1 with 1,2- dithiepane and 1,2-dithiepane-1-oxide (Figure A1-12).

Figure 2-2. Thiol cross-linking by cyclic thiosulfinates kinetically stabilizes the SOD1 dimer in vitro and in cells. Representative raw (Top left) and deconvoluted (Top right) mass spectra used for calculating cross-linking rates (Middle). The 31,808 Da molecular mass of the cross-linked dimer (D) supports the mechanism proposed in Figure 2-1 (e.g. two SOD1 monomers [2 x 15,844

Da (M)] + 1,2-dithiane-1-oxide [136 Da] – oxygen [16 Da]). After 10 min of incubation 1,2- dithiane-1-oxide and 1,2-dithiane cross-linked 95% and 0% of SOD1, respectively. Consistent with the rate of cyclic disulfide cross-linkers being limited by thiol oxidation 1,2-dithiane cross-

53 links only 11% of SOD1 after three days (Figure A1-3). (Bottom) Western blot of SOD1 from

HepG2 cells incubated with various concentrations of compounds for 30 min. EC50s for 1,2- dithiane-1-oxide cross-linking were 1-5 µM in HepG2 and HELA (Figure A1-4) cells.

To demonstrate the utility of 1,2-dithaine-1-oxide as a cell penetrating, dithiol pair cross- linker, the cross-linking reaction was examined both in two widely used human cell lines (Hep G2 and HeLa), and with purified SOD1 in the presence of competing reduced glutathione or DTT.

Hep G2 and HeLa cells both contain approximately 5 mM glutathione.131 Hep G2 cells (Figure

2-2, bottom) and HeLa cells (Figure A1-4) incubated with various concentrations of 1,2-dithaine-

1-oxide for 30 min showed an EC50 of ~5 µM in western blots, confirming that cellular conditions do not prohibit cross-linking. Cell viability was not affected by 1,2-dithiane-1-oxide concentrations that were 50-fold higher than the EC50, and the LC50 of 1,2-dithiane-1-oxide was approximately 200-fold greater than its EC50 (Figure A1-5). Consistent with the cellular studies, cross-linking of purified SOD proceeded to completion in the presence of 10:1 ratio of glutathione:1,2-dithiane-1-oxide (Figure A1-8), and even in the presence of equimolar concentrations of the reducing agent dithiothreitol (DTT) (Figure A1-9). The rate of cross-linking was decreased in the presence of competing reductants, presumably due to reversible thiolate- disulfide interchange between reductants and 1,2-dithiane-1-oxide (no glutathionyl or DTT adducts with 1,2-dithiane-1-oxide or SOD1 were observed). These results confirm the utility of these cross-linkers even in presence of modest amounts of additional reducing agents.

Cyclic disulfides132 and their derivitives (e.g., dithiolene thiones)133 have been used therapeutically and many of their targets are known. However, the binding mechanism of these drugs, including that of the nutritional supplement and diabetic complication treatment, α-lipoic

54 acid (ALA), have not been characterized.134 To broaden the applicability of cyclic disulfide mediated cross-linking and explore a potential MOA, ALA was purchased and β-lipoic acid (BLA) was synthesized and assayed as above. Compared to ALA, BLA cross-linked SOD1 in cells and cross-linked 30% more SOD1 in vitro (Figure A1-6 and A1-7). Notably, the terminal carboxylic acid on ALA and BLA presents an opportunity for functionalization.135

2.4 Conclusion

Critical features of cyclic thiosulfinate reactivity include: 1) toxic binding to single thiolates is reversible through thiolate-disulfide interchange but thiol pair cross-linking is not; 2) the leaving group is expended upon cross-linking; 3) cross-linking proceeds in water and is driven by the considerable bond enthalpies of S-S bond and water formation (-132 kcal mol-1), resulting in high yields; 4) Reactive functional groups, including carboxylates, amines, and disulfides, are avoided; 5) Cyclic disulfide S-S bond strength and reactivity has a strain-dependence that greatly exceeds that of rings composed of period-two elements;122 and 6) competing reductants are tolerated and cross-linking can occurs in cells. In summary, cyclic disulfide reactivity, including reversible binding to lone thiols, is predictable and highly tunable. Cyclic thiosulfinate cross- linkers have potential as: 1) A less toxic alternative to Cys specific di-ene cross-linkers and phenylarsine oxide cross-linkers, which can both react with monothiols,136 2) Probes for proteinaceous Cys-dithiolates, which perform essential in vivo functions and often serve as metal and metallocofactor ligands, 137-139 3) Inter-functional group distance measurement tools, 4)

Biocompatible templates for higher order structures in polymer synthesis, and 5) Cellular thiol pair cross-linkers.

2.5 Methods

2.5.1 Synthesis of 1,2-dithiane

Preparation of 1,2-dithiane was adapted from a known literature procedure.140 Silica gel

(40-60 µm particle size, 60 Å pore size, 41 g) was added to a round bottom flask and distilled water (102 mL) was added slowly with rigorous stirring until a uniform suspension had formed.

Dichloromethane (200 mL) and 1,4-butanedithiol (2.00 g, 16.4 mmol, 1 equiv) were added to the suspension while stirring. A solution of Br2 (2.88 g, 18 mmol, 1.10 equiv) in dichloromethane (16 mL) was added dropwise to the off-white suspension while stirring vigorously. The reaction mixture was stirred for 5 minutes, and reaction completion was confirmed by TLC analysis. The reaction mixture was filtered over celite into a flask containing a stirred solution of 1.25 M NaOH

(12 mL). The colorless organic phase was removed, washed with distilled water (3 x 50 mL) and dried over sodium sulfate. The solvent was removed under reduced pressure and the crude product crystallized from hexanes at –20 °C to yield 1,2-dithiane as a white crystalline solid (1.63 g, 83%).

1 13 Rf = 0.80 (5:1 hexanes:EtOAc); H NMR (400 MHz, CDCl3, δ): 1.97 (bs, 4H), 2.85 (bs, 4H); C

140 NMR (100 MHz, CDCl3, δ): 27.9, 33.5. mp 29 - 32 °C (lit. 28-30 °C).

2.5.2 Synthesis of 1,2-dithiane-1-oxide

Preparation of 1,2-dithiane-1-oxide was adapted from a known literature procedure.140 A solution of sodium periodate (843 mg, 3.94 mmol, 1.10 equiv) in water (64 mL) was added dropwise to a stirred solution of 1,2-dithiane (431 mg, 3.58 mmol, 1 equiv) in methanol (193 mL) at 0 °C. The reaction mixture was stirred 16 h, and reaction completion was confirmed by TLC analysis. The white slurry warmed to room temperature, filtered over celite, and the filtrate was concentrated under reduced pressure. The remaining solution was diluted with chloroform (40 mL) and transferred to a separatory funnel. A small amount of solid NaCl was added and the aqueous layer was extracted with CHCl3 (3 x 40 mL). The combined organic layer was dried over sodium sulfate, the solvent removed under reduced pressure, and purified by flash column chromatography on silica gel with 2% MeOH / CH2Cl2 yielded 1,2-dithiane-1-oxide as a colorless solid (214 mg,

1 44%). Rf = 0.33 (2% MeOH / CH2Cl2); H NMR (400 MHz, CDCl3, δ): 1.85-1.88 (m, 1H), 1.96-

2.06 (dtt, J = 13.7, 12.7, 3.0 Hz, 1H), 2.11-2.15 (m, 1H), 2.62-2.71 (m, 2H), 3.04 - 3.12 (dt, J =

13.1, 3.0 Hz, 1H), 3.18-3.23 (td, J = 13.3, 3.6 Hz, 1H), 3.62-3.70 (ddd, J = 14.5, 12.0, 2.5 Hz, 1H);

13 + C NMR (100 MHz, CDCl3, δ): 15.3, 23.5, 25.8, 51.9. HRMS-ESI (m / z): [M + H] calcd for

140 C4H8OS2, 137.00893 Da; found 137.00893 Da. mp 83-86 °C (lit. 85 °C).

2.5.3 Synthesis of 1,2-dithiepane

Preparation of 1,2-dithiepane was adapted from a known literature procedure.140 Silica gel

(40-60 µm particle size, 60 Å pore size, 46 g) was added to a round bottom flask and distilled water (23 mL) was added slowly with rigorous stirring until a uniform suspension had formed.

Dichloromethane (230 mL) and 1,5-pentanedithiol (3.00 g, 22.01 mmol, 1 equiv) were added to the suspension while stirring. A solution of Br2 (3.87 g, 24.22 mmol, 1.10 equiv) in dichloromethane (23 mL) was added dropwise to the off-white suspension while stirring vigorously. The reaction mixture was stirred for 5 minutes, and reaction completion was confirmed by TLC analysis. The reaction mixture was filtered over celite into a flask containing a stirred solution of 1.25 M NaOH (110 mL). The colorless organic phase was removed, washed with distilled water (3 x 70 mL) and dried over sodium sulfate. The solvent was removed under reduced pressure and the crude product purified by flash column chromatography on silica gel with 20:1

Hex / EtOAc yielded 1,2- dithiepane as a clear liquid (2.83 g, 96%). Rf = 0.82 (9:1

1 hexanes:EtOAc); H NMR (400 MHz, CDCl3, δ): 1.75-1.79 (m, 2H), 2.00-2.05 (m, 4H), 2.82-2.84

13 (t, J = 6.3, 4H); C NMR (100 MHz, CDCl3, δ): 26.0, 30.0, 39.2.

2.5.4 Synthesis of 1,2-dithiepane-1-oxide

Preparation of 1,2-dithiepane was adapted from a known literature procedure for the oxidation of disulfides.141 mCPBA (73%, 214 mg, 0.909 mmol, 1.07 equiv) was added at 0 °C to a solution of 1,2-dithiepane (114 mg, 0.849 mmol, 1 equiv) in anhydrous dichloromethane (5.5 mL). The solution was stirred in an ice bath for 1 h, then sodium carbonate (1 g) was added and stirred for 30 min at 0 °C. The solution was filtered over a celite pad and magnesium sulfate, solvent was removed under reduced pressure and the crude product purified by flash column chromatography on silica gel with 5:1 Hex / EtOAc yielded 1,2- dithiepane-1-oxide as a clear

1 liquid (97 mg, 76%). Rf = 0.19 (5:1 hexanes:EtOAc); H NMR (400 MHz, CDCl3, δ): 1.73-1.81

(m, 1H), 1.91-2.02 (m, 4H), 2.09-2.16 (m, 1H), 2.77-2.86 (m, 2H), 3.36-3.41 (dd, J = 6.5, 13.8 Hz,

13 1H), 3.50-3.57 (m, 1H); C NMR (100 MHz, CDCl3, δ): 18.02, 22.31, 26.77, 28.46, 60.56.

+ HRMS-ESI (m / z): [M + H] calcd for C5H10OS2, 151.024584 Da; found 151.02498 Da.

2.5.5 Synthesis of β-lipoic acid

Preparation of β-lipoic acid was adapted from a known literature procedure.142 Aqueous hydrogen peroxide (0.841 mL, 35% in H2O, 9.77 mmol, 2 equiv) was added to a solution of DL- thioctic acid (α -lipoic acid, 1.01 g, 4.89 mmol, 1 equiv) in acetone (2.5 mL) and allowed to stir

59 for 24 h. Solvent was removed under reduced pressure, diluted with dichloromethane (25 mL) and then added to brine (50 mL). The aqueous layer was extracted with dichloromethane (3 x 25 mL) and dried over sodium sulfate. The solvent was removed under reduced pressure and purified by flash column chromatography on silica gel with 3% MeOH / CH2Cl2, 0.1% AcOH yielding beta-

1 lipoic acid as a colorless oil (418 mg, 39%). Rf = 0.46 (8% MeOH / CH2Cl2, 0.1% AcOH). H-

NMR spectrum and 13C-NMR data matched those reported by others142, 143 for a mixture of all four stereo- and regioisomers: the 1H-NMR and 13C-NMR chemical shifts have been exactly observed as reported by 2D techniques performed by Müller et al.143 HRMS-ESI (m / z): [M + H]+ calcd for

C4H14O3S2, 223.04571 Da; found 223.04626 Da. The β-lipoic acid samples used is a mixture of two regioisomers (Figure A1-31). Furthermore, each regioisomer contains 2 stereogenic centers and hence is comprised of pairs of enantiomers ((R,R) and (S,S)) and ((S,R) and (R,S)). Therefore, the β-lipoic acid mixture consists of a total of 8 isomers ((1R,3R), (1S,3R), (1R,3S), (1S,3S),

(1R,5R), (1S,5R), (1R,5S), (1S,5S)). We have not been able to separate these 8 isomers nor have we determined the stereo or regiospecificity of this cross-linking reaction.

2.5.6 Kinetics of 1,2-dithiane-1-oxide and 1,2-dithiane cross-linking

Kinetic studies of the cross-linking efficiency of 1,2-dithiane compared to 1,2-dithiane-1- oxide were performed in tandem. 50 µM human WT SOD1 in 10 mM ammonium acetate pH 7.4 was incubated with 1000 µM 1,2-dithiane or 1,2-dithiane-1-oxide (20X excess) at 37 °C. Human

WT SOD1 was expressed in yeast as previously reported. Both 1,2-dithiane and 1,2-dithiane-1- oxide were dissolved in 100% HPLC grade methanol to a stock concentration of 25 mM. Dilution in HPLC-Grade water to 1 mM cyclic disulfide was performed prior to experimentation. As a negative control, 50 µM human WT SOD1 was incubated at 37 °C with 4% HPLC-grade methanol.

At every time point, 1 µL of each sample was removed from their respective reaction vial for analysis. Samples were briefly ( ̴30 seconds) incubated at room temperature with 10% formic acid to remove metals from SOD1 and quench cross-linking reaction prior to mass spectrometry analysis. Samples were then diluted to 1 µM SOD1 in 50:50 acetonitrile:water, 0.1% FA and analyzed by direct infusion into a Bruker SolariX XR FT-ICR mass spectrometer. The percent dimer formation was calculated by comparing the relative dimer (31,808 Da) and monomer

(15,844 Da) MaxEnt deconvoluted peak heights (Dimer/(Dimer+Monomer)). Identical methods were used in the kinetics of α-lipoic acid and β-lipoic acid. Kinetics of 1,2-dithiepane vs. 1,2- dithiepane-1-oxide were performed at a 1:1 ratio of SOD1 dimer:cross-linker.

2.5.7 Expression and Purification of hSOD1 (WT and C111S)

Expression and purification of WT SOD1 and C111S SOD1 was carried out as previously published.54, 144, 145 Briefly, human SOD1 cDNA cloned into the yeast expression vector YEp-351 was transformed into EGy118ΔSOD1 yeast and grown at 30 °C for 36-48 hours. Cultures were pelleted, lysed using 0.5 mm glass beads and a blender, and subjected to a 60% ammonium sulfate cut on ice. After ammonium sulfate precipitation, the sample was pelleted and the supernatant was diluted with ~0.19 volumes buffer (50 mM sodium phosphate, 150 mM sodium chloride, 0.1 M

EDTA, 0.25 mM DTT, pH 7.0) to a final concentration of 2.0 M ammonium sulfate. The sample was then purified using a phenyl-Sepharose hydrophobic interaction chromatography column with a 300 mL linearly decreasing salt gradient from a high salt buffer (2.0 M ammonium sulfate, 50 mM sodium phosphate, 150 mM sodium chloride, 0.1 M EDTA, 0.25 mM DTT, pH 7.0) to a low salt buffer (50 mM sodium phosphate, 150 mM sodium chloride, 0.1 M EDTA, 0.25 mM DTT, pH

7.0). Samples containing SOD1 typically elute between 1.6-1.1 M ammonium sulfate, which was

61 confirmed using gel electrophoresis. SOD1 containing fractions were pooled and exchanged to a

10 mM Tris (pH 8.0) buffer. The protein was then loaded onto a Mono Q 10/100 anion exchange chromatography column and eluted using a 200 mL linearly increasing salt gradient from a low salt buffer (10 mM Tris, pH 8.0) to a high salt buffer (10 mM Tris, pH 8.0, 1 M sodium chloride).

The gradient is run from 0-30% 10 mM Tris, pH 8.0, 1 M sodium chloride and SOD1 eluted between 5-12% 10 mM Tris, pH 8.0, 1 M sodium chloride. SOD1 fractions from anion exchange were confirmed using gel electrophoresis. Final SOD1 containing fractions were pooled together, washed three times with 10 mM ammonium bicarbonate in a Millipore amicon centrifugal filter, and buffer exchanged into 10 mM ammonium acetate pH 7.4. Protein was stored at -80 °C until use.

2.5.8 Localization of cross-link using point mutated C111S SOD1

1,2-dithiane-1-oxide was initially dissolved in 100% DMSO. Recombinant human WT

SOD1 and C111S SOD1 purified from yeast (10 µM) were incubated overnight at room temperature with 1,2-dithiane-1-oxide at 100 µM in 0.1% DMSO after dilution. Aliquots from each of the SOD1 incubations were individually diluted ten-fold with H2O containing 0.1% formic acid and analyzed using reversed phase C18 LC-MS on a Bruker HCT Ultra ion trap. The resulting data were processed using DataAnalysis 3.4 (Bruker Daltonics). Mass spectra were averaged across the retention times corresponding to when SOD1 eluted and were deconvoluted to determine the molecular weight of the uncharged species. Note, the acidic conditions employed during liquid chromatography and the relatively harsh electrospray ionization process resulted in loss of native metals and native dimer dissociation.

2.5.9 In vitro glutathione competition assay

Purified WT SOD1 was diluted to 10 µM in 10 mM ammonium acetate, pH 7.4. Protein was incubated with freshly prepared 100 µM 1,2-dithiane-1-oxide in 5% methanol and 1000µM reduced glutathione (in water) at 37 °C for given time periods. Samples were extracted, briefly (10 s) with 10% formic acid to and analyzed at 0 min, 1 min, 10 min, 100 min, and 1000 mins. Control samples included: 1) 10 µM SOD1 2) 10 µM SOD1 + 100 µM 1,2-dithiane-1-oxide 3) 10 µM

SOD1 + 1000 µM reduced glutathione. Samples were briefly ( ̴30 seconds) incubated at room temperature with 10% formic acid to remove metals from SOD1 prior to mass spectrometry analysis. Samples were then diluted to 1 µM SOD1 in 50:50 acetonitrile:water, 0.1% FA and analyzed by direct infusion into a Bruker Solarix XR FT-ICR mass spectrometer. The percent dimer formation was calculated by comparing the relative dimer (31,808 Da) and monomer

(15,844 Da) MaxEnt deconvoluted peak heights (Dimer/(Dimer+Monomer)).

2.5.10 In vitro DTT competition assay

Purified WT SOD1 was diluted to 50 µM in 10 mM ammonium acetate, pH 7.4. 2 µL of

50 µM SOD1 solution was incubated with 1 µL of 2 mM 1,2-dithiane-1-oxide and 1 µL of various concentrations of DTT. DTT solutions were made in 100% HPLC grade H2O. DTT solution concentrations were 1 mM, 2 mM and 4 mM. Samples were left to incubate at 37 °C for 24 h.

Samples were then diluted to 1 µM SOD1 in 50:50 acetonitrile:water, 0.1% FA and analyzed by direct infusion into a Bruker Solarix XR FT-ICR mass spectrometer

2.5.11 Cell culture and 1,2-dithiane-1-oxide/1,2-dithiane dosing and α-lipoic acid/β-lipoic acid in Hep-G2 cells

Hep G2 cells were cultures in DMEM with 10% fetal bovine serum and penicillin/streptomycin in 96 well Costar® Corning CellBIND plate with 5% CO2 at 37 °C. The cells were cultured to monolayer confluency. 20 mM, 2 mM, 200 µM, 20 µM, and 2 µM stocks of

1,2-dithiane and 1,2-dithiane-1-oxide were prepared in 100% DMSO. Stocks were diluted twenty- fold in 1X PBS (final DMSO 5%). Cells were treated with 200 uL of compound and incubated for

30 minutes at 37 °C 5%CO2. Cells were washed with 1X PBS and 20 µL of 6x nonreducing sample buffer was added to each well. The 96 well plate was heated to 90 °C for 10 minutes. Samplers were spun at 14000 RPM for 5 minutes in a Beckman Coulter Microfuge®18 centrifuge. Samples were run using a Bio Rad Mini-Protean electrophoresis chamber on 12% precast TGX polyacrylamide gels at 150V. After separation was complete, gels were extracted from the cassettes and incubated in an in-house transfer buffer for 10 minutes are 90 °C to ensure thorough transfer of protein bands and thorough binding of antibody (25 mM Tris, 192 mM glycine, 10 mM 2- mercaptoethanol, 0.1% SDS). Transfer was performed with a Bio Rad Trans-blot turbo transfer system using a trans-clot turbo transfer pack. Membranes were dried, blocked with 5% milk for 2 h at room temperature. Membranes were then probed with primary antibody, anti-SOD1 antibody

SOD100 overnight at 4 °C. Membranes are incubated with HRP-labeled secondary antibodies and visualized using ECL Western Blotting Substrate and imaged using the ChemiDoc MP.

2.5.12 Cell culture and 1,2-dithaine-1-oxide dosing assay in HeLa Cells

HeLa cells were cultured in DMEM with 10% fetal bovine serum and penicillin/streptomycin in 24 well Costar® Corning CellBIND plate with 5% CO2 at 37 °C. The cells were cultured to monolayer confluency. A 100 mM concentration of 1,2-Dithiane-1-oxide was made by dissolving in DMSO. The 100 mM solution diluted to various concentrations with

1X PBS solution. 1 mL of solutions was added to the HeLa cells in the 24 well plate. 1,2-Dithiane-

1-oxide was allowed to interact with the cells for 1 h. Cells were washed 2X with 1 mL PBS. After removing the second PBS wash, the cells were then treated with 6X non-reducing SDS sample buffer and collected. The collected cell lysate was divided into two parts. One half was treated with 5% β-ME, to remove bound cyclic disulfides, as negative controls. The other half was treated with equal amount of milliQ water. All the cell lysate samples were then heated at 80 ℃ for 10 minutes in an Eppendorf Thermomixer. The samples were spun at 14000 rpm for 5 minutes in

Beckman Coulter Microfuge centrifuge. HeLa cell extracts were separated on 12% SDS- polyacrylamide Tris·HCl gels run at 80 V at room temperature. Proteins were transferred to nitrocellulose membranes for Western blotting. Monomer and covalently-linked dimer SOD1 were detected using the rabbit polyclonal anti-SOD1 antibody SOD100. Membranes are incubated with

HRP-labeled secondary antibodies and visualized using ECL Western Blotting Substrate and imaged using the ChemiDoc MP.

2.5.13 Cell culture and 1,2-dithiane-1-oxide cell viability assay

Human immortalized neuroblastoma cells (SH-SY5Y) were purchased from ATCC

(Manassas, Virginia, USA). Cells were grown in Eagle’s Minimum Essential Medium (EMEM) supplemented with 10% fetal bovine serum at 37 °C and 5% CO2 to >70% confluency. Cells were treated with 200 µL of various concentrations of 1,2-dithiane-1-oxide in 2% DMSO (250

µM, 500 µM, 750 µM, and 1000 µM) and allowed to incubate for 24 h in a 96 well plate

(Corning Inc., Corning, New York, USA). Following incubation with 1,2-dithaine-1-oxide, 22

µL of PrestoBlue® was added to each well and allowed to incubate for 10 min at 37 °C.

Excitation/emission was measured at 560/590 (in nm) using an Infinite® 200 plate reader

(Tecan, Mannedorf, Switzerland). Cell viability was quantified relative to untreated cells.

2.5.14 Computational methods

All computations were carried out with Gaussian 09, Rev E.01.146 The reported stationary points were optimized using the M06-2X147 density functional at the DFT level of theory. We utilized the Gaussian 09 ultrafine integration grid and the polarizable continuum model (PCM) using the integral equation formalism variant (IEF-PCM)148-150 using parameters for water. Each stationary point was subjected to a vibrational analysis and all transition structures have exactly one negative frequency, while all other stationary points have all frequencies greater than 0 indicating local minima. We then performed a single point energy calculation on the 6-31+G(d,p)- optimized geometries at the DFT level of theory with ultrafine grid and PCM model as mentioned earlier. Single point energy were performed using the with 6-311+G(d,p) basis set. The total energy was calculated by adding the free energy correction and the single point energy which can be seen from the Figure A1-10 below. ESIgen was used to organize geometries, energies, and frequencies for the stationary points utilized in Figure A1-10.151 The parameters for the transition structure conformational search were identical for the minima, but we constrained the forming S–S bond at

2.4 Angstroms for the cyclic disulfide ring opening step (TS5->2a and TS1->2b) and 2.3

Angstroms for the thiolate/sulfenic acid condensation reaction (TS(3a,b->4)). Each constrained transition state search produced 10 structures within 12 kcal/mol (according to OPLS2005) which were subjected to a transition state search. The reported structures in Figure 2-1 are the lowest energy transition structures for each transformation.

2.5.15 QM calculations are consistent with empirical studies.

Using transition state theory (Eyring equation) we extrapolated rates from the calculated transition states (TS(1→2b) and TS(5→2a)), and found these were consistent with the myriad of published small molecule and protein experimental data presented here.152-155 In particular, the

QM results provide mechanistic insight into how decreasing the pKa of the leaving group sulfur

(via sulfenic acid anion instead of a sulfur anion) increased the overall reaction rate by two separate mechanisms (i.e. in addition to eliminating the rate determining thiol oxidation, cyclic thiosulfinates decrease the barrier of the new rate limiting step, thiolate-disulfide interchange).

Specifically, we performed a series of detailed (multi-concentration) kinetics experiments following the protocol of Singh et al. and determined the overall second order rate constant of cyclic thiosulfinate-mediated cross-linking product formation to be 1.5 x 104 M-1 min-1 which, under our experimental conditions, extrapolates to a predicted half-life of 2.7 min for cross-linking

SOD1 with 1,2-dithiane-1-oxide.153 Notably this value is consistent with half-life fitted from

Figure 2-1 (2.2 min) and is consistent with the absolute rate extrapolated from the calculated value of 13.1 kcal/mol for the thiol-disulfide intermediate 2b. being the rate determining step. The thiol- disulfide exchange reaction of oxidized DTT (cyclic disulfide) and reduced glutathione, which provides the closest analogy to the 1,2-dithiane data presented here, has an extrapolated half-life of >200 mins (rate constant of 1.8 x 102 M-1 min-1).152 In fact, the rates we observed and calculated for cyclic thiosulfinates were faster than any ever reported for cyclic disulfides (to our knowledge, the fastest reported rate of thiol-disulfide interchange is the reaction of Papain-S-SCH3 with DTT which gives a rate constant of 3.3 x 103 M-1 min-1 and an extrapolated half-life of 12 min).154 These combined results demonstrate that thiosulfinates speed the overall rate both by eliminating the rate determining-S oxidation, and then by decreasing the new rate limiting step, thiolate-disulfide interchange. We note that although the reported condensation reaction rates of thiolates and

67 sulfenic acids vary considerably, our experimentally determined rate constants are near the median of reported values, and are also consistent with the 11.0 kcal/mol transition state TS(3b→4) reported in Figure 2-1. For example, Gupta et al. reported the rate constants of disulfide formation through sulfenic acid/thiolate condensation range from >106 M-1 min-1 (Cys-SOH + Cys) to 1.3 x

103 M-1 min-1 (HSA-SOH + Cys), which extrapolate to half-lives ranging from milliseconds to 30 mins, respectively, under our experimental conditions.155

Thapa et al. highlighted the difficulty of computing pKa’s of thiols without explicit solvation, and experimental rates of thiol-disulfide exchange generally decrease by three orders of magnitude in protic compared to aprotic solvents, and by as much as nine orders of magnitude in water compared to the gas phase.156, 157 Given these obstacles, and that our protein includes both protic and aprotic environments, we do not report continuum model-based calculations for intermediates 3a and 3b. We do not have the experimental data required to estimate the pKa of this intermediate, which is therefore listed as the arbitrary value of 0 kcal/mol.158 Note, however, this value is consistent with our experimentally determined rates, which imply that the energy of intermediate 3a,3b cannot be higher than c.a. 3 kcal/mol, as well as the lack of observation of mono-thiol intermediates, which implies that its energy probably isn’t lower than c.a. -3 kcal/mol.

Chapter 3

Nucleophilic substitution reactions of cyclic thiosulfinates are accelerated by hyperconjugative interactions

Daniel P. Donnelly1,2, Jeffrey N. Agar1,2, Steven A. Lopez*1

1Department of Chemistry and Chemical Biology, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115, United States

2Barnett Institute of Chemical and Biological Analysis, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115, United States

This chapter was written with the intent to publish. It is currently under review at Chemical Science.

3.0 Statement of Contribution

Experimental contributions made to this chapter by Daniel P. Donnelly are as follows: all calculations and data analysis. The manuscript was written by Daniel P. Donnelly and Steven A.

Lopez with careful review and contributions made by Jeffrey N. Agar. All figures were prepared by Daniel P. Donnelly.

3.1 Abstract

Cyclic thiosulfinates are a class of biocompatible molecules, currently expanding our in vivo toolkit. Agar and co-workers have shown that they are capable of efficient cross-linking reactions. While strain energy has been shown to promote the nucleophilic substitution reactions of cyclic disulfides, the reactivities of cyclic thiosulfinate nucleophilic substitution is unexplored.

We used density functional theory calculations [M06-2X/6-311++G(d,p)] to determine the activation and reaction free energies for the reactions of 3—10-membered cyclic thiosulfinates and cyclic disulfides with methyl thiolate. The nucleophilic substitution reaction of cyclic thiosulfinates was found to be strain-promoted, similar to the strain-promoted nucleophilic substitution reactions of cyclic disulfides. The origin of the nearly 100-fold rate enhancement of cyclic thiosulfinates over cyclic disulfides was understood using the distortion/interaction model and natural bond order analysis. The cyclic thiosulfinates benefit from a hyperconjugative

∗ ∗ interaction between an oxygen lone pair and the σSS orbital (nO → σSS). This interaction generally lengthens the reactant S1—S2 bond, which pre-distorts cyclic thiosulfinates to resemble their corresponding transition structures. The inductive effect of the oxygen in cyclic thiosulfinates

∗ lowers the σSS orbital energies relative to cyclic disulfides and results in more stabilizing transition state frontier molecular orbital interactions with methyl thiolate.

3.2 Introduction

Cyclic disulfides are a privileged class of molecules that have long played important roles in energy metabolism and the modulation of cellular redox status.134, 159-163 Recently, they have been used for in-vivo applications in biochemistry and biomaterials as building blocks for self-

71 healing, biocompatible polymers, and hydrogels.164, 165 They can also serve as vehicles to shuttle large molecules and apoptosis-inducing substrates through cell membranes via the transferrin receptor.126, 166-168 In many cases, α-lipoic acid, a cyclic disulfide, has been used in a variety of functional assemblies at the gold surface.169-171 Cyclic disulfides are one of the first known cross- linking-specific molecules; Agar and co-workers showed that they could selectively cross-link cysteine pairs while reversibly modifying lone cysteines in vivo.172 Our groups recently introduced a six-membered cyclic thiosulfinate (1,2-dithiane-1-oxide) capable of cross-linking free cysteine pairs up to 104-fold faster than a six-membered cyclic disulfide (1,2-dithiane), by circumventing the rate-determining oxidation step.172 A covalent cross-link is efficiently formed between the sulfenic acid intermediate and a second thiolate. Whitesides and coworkers augmented their experiments with MM2173 calculations to show that the reactivities of cyclic disulfides towards biological thiolate-based nucleophiles were strain-promoted.122, 124 Bachrach and co-workers used

DFT calculations to locate transition structures for the nucleophilic substitution reactions of a model thiolate to a series of cyclic disulfides.174 This computational study builds on the results of

Whitesides and Bachrach to determine the origin of the increased nucleophilic substitution reactivities of 3—10-membered cyclic thiosulfinates relative to cyclic disulfides (Scheme 3-1). A rigorous conformational search was employed to identify the global minima of reactants, ring- opened intermediates, and the lowest-energy transition structures. The DFT calculations are used to predict the reactivities of 3—10-membered cyclic thiosulfinates (3—10)a towards a model thiolate (methyl thiolate) by locating transition structures and disulfide-exchange intermediates.

‡ The corresponding activation free energies and reaction energies (ΔG and ΔGrxn, respectively) were compared to those of an analogous series of cyclic disulfides to understand why cyclic thiosulfinates are more reactive than cyclic disulfides towards thiolates.

Scheme 3-1. Nucleophilic substitution towards cyclic thiosulfinate and cyclic disulfides. (a.)

Structures of 3—10-membered cyclic thiosulfinates (blue) and cyclic disulfides (red). (b.)

Mechanism of thiol-disulfide exchange between nucleophilic methyl thiolate (MeS—) and a cyclic thiosulfinate (blue) or a cyclic disulfide (red)

3.3 Results and Discussion

Ring strain is released upon nucleophilic addition of MeS– ; as such, we defined strain

– energy as –ΔGrxn. We computed reaction energies (ΔGrxn) for the nucleophilic addition of MeS to (3—10)a and (3—10)b to assess the reversibilities of the nucleophilic substitution reactions.

These energies are summarized in Table 3-1.

For the most strained reactants, (3—4)a and (3—4)b, the reaction energies for the nucleophilic addition of MeS– range from –11.9 to –11.7 and –16.9 to –16.7 kcal mol–1, respectively. These reactions are exergonic because ring strain is released upon ring-opening. (5—

7)a and (5—7)b lead to endergonic reaction energies ranging from 1.2 to 4.4 kcal mol–1 and 1.2 to

5.1 kcal mol–1, respectively. The larger cyclic structures, (8—10)a and (8—10)b have reaction free

73 energies that range from –2.3 to –4.6 kcal mol-1 and +0.7 to –1.5 kcal mol–1 , respectively. The reaction energies of these series follow a similar trend to cycloalkanes in which (3—4)a and (3—

4)b are significantly strained, (5—7)a and (5—7)b are relatively unstrained, and (8—10)a and

174-176 (8—10)b are moderately strained. The longer S1—S2 bond in the thiacycles relieves some strain compared the corresponding cycloalkanes (e.g., 1,2-dithiolane vs. cyclopentane).

— Table 3-1. ΔGrxn of the nucleophilic attack of MeS on (3—10)a and (3—10)b. Computed using

M06-2X/6-311++G(d,p) IEF-PMCH2O.

We assessed the reactivities of the cyclic thiosulfinates and cyclic disulfides towards methyl thiolate by locating transition structures and computing their corresponding activation free energies and enthalpies (Figure 3-1). The transition structures shown in Figure 1 generally have a

– nearly linear MeS —S1—S2 angle; the transition states range from exactly synchronous to asynchronous. The breaking S1—S2 bonds of TS-(3—10)a and TS-(3—10)b range from 2.26-2.48

Å and 2.26-2.50 Å, respectively. The S—S1 distance in TS-(3—10)a and TS-(3—10)b ranges from 2.42-2.72 Å and 2.39-2.79 Å, respectively. TS-10b is exactly synchronous (2.46 Å), while

TS-5b is the most asynchronous (2.56 and 2.38 Å). The C—C and C—S σ bonds of (3—4)a and

(3—4)b are well-described by Walsh orbitals due to the nearly 60° and 90° bonding angles, respectively. As such, incipient nucleophiles will interact with bent S1—S2 σ* orbitals, which results in the non-linear transition state geometries of TS-(3—4)a and TS-(3—4)b. The activation

74 free energies of the smallest rings TS-(3—4)a and TS-(3—4)b are the lowest (2.3-4.4 kcal mol–

1). The low activation energies of (3—4)a and (3—4)b are consistent with the established strain- promoted reactions of cyclic disulfides. The activation free of energies of TS-(5—10)a are generally higher and range from 10.9 to 13.1 kcal mol–1. The activation free energies of TS-(6—

10)a are substantially lower than those of TS-(6—10)b; ΔΔG‡ range from –2.5 to –7.4 kcal mol–

1, which corresponds to a 102-105–fold rate enhancement for cyclic thiosulfinates relative to cyclic disulfides.

Figure 3-1. Transition structures for the reaction of MeS– with cyclic thiosulfinates (3—10)a and cyclic disulfides (3—10)b. The ΔG⧧ and ΔH⧧ (in brackets) were computed with M06-2X/6-

311++G(d,p) IEF-PCMH2O and provided for each transition structure. The bond lengths and energies are reported in Å and kcal mol–1, respectively. *We were unable to locate transition states for (3—4)a and (3—4)b using B3LYP-D3BJ, M06-2X, or MP2 methods. Reported structures are

76 energies for a constrained transition structure featuring one negative frequency corresponding to the substitution reaction using M06-2X/6-311++G(d,p).

‡ Figure 3-2. Plot of ΔG vs. –ΔGrxn of series (3—10)a and (3—10)b. The linear equation for (3—

‡ ‡ 10)a is ΔG = –0.53(–ΔGrxn) + 11.33. The linear equation for (3—10)b is ΔG = –0.66(–ΔGrxn) +

14.71. Computed using M06-2X/6-311++G(d,p) IEF-PCMH2O.

Are these reactions strain promoted?

Nucleophilic substitution reactions of cyclic disulfides are often described as strain- promoted.122, 174, 177 We assessed the role of strain energy on the reactivities of (3—10)a and (3—

10)b by plotting ΔG⧧ against ΔGrxn (Figure 3-2). Figure 3-2 shows a linear correlation between

2 ΔG⧧ and –ΔGrxn for the cyclic disulfide and cyclic thiosulfinate reactions (R =0.81 and 0.80, respectively). This suggests that strain energy controls the reactivities for a broad set of cyclic

77 disulfides and establishes that the reactivities of cyclic thiosulfinates are also controlled by strain energy. The activation free energies of cyclic thiosulfinates are generally lower than those of cyclic disulfides; the y–intercept values are 11.3 and 14.7 kcal mol–1, respectively.

Distortion/Interaction Model

To understand the origin of generally lower activation barriers and strain energies of cyclic thiosulfinates relative to the cyclic disulfides, we turned to the distortion/interaction model.178-180

The distortion/interaction model dissects activation barriers for bimolecular reactions into two terms: distortion and interaction energy [ΔE⧧= ΔEd⧧ + ΔEi⧧]. Distortion energy (ΔEd⧧) is the energy required to deform reactants from their equilibrium structures to their distorted transition structure geometries without allowing them to interact. Interaction energy (ΔEi⧧) results from the difference in ΔE⧧ and ΔEi⧧ and has been attributed to favorable intermolecular electrostatic, dispersion, and charge transfer interactions. The distortion/interaction model has been used to explain the reactivities and selectivities of pericyclic181-184 and organometallic185-187 reactions. The computed distortion and interaction energies of TS-(3—10)a and TS-(3—10)b are summarized in Figure 3-

Figure 3-3. Activation, distortion, and interaction energies of TS-(3—10)a and TS-(3—10)b.

Computed using M06-2X/6-311++G(d,p) IEF-PCMH2O. The bond lengths and energies are

79 reported in Å and kcal mol–1, respectively. *Transition structures are constrained and have one negative frequency connecting reactants to product using M06-2X/6-311++G(d,p).

The distortion energies of cyclic thiosulfinates (3—10)a range from 3.7 to 14.3 kcal mol–1 and the distortion energies of cyclic disulfides (3—10)b range from 4.3 to 20.6 kcal mol–1. We plotted activation energies against distortion energies for the reactions of cyclic thiosulfinates

(blue) and cyclic disulfides (red) in Figure 3-4.

– Figure 3-4. (a.) ΔE⧧ vs. ΔEd⧧ of the nucleophilic addition of MeS towards cyclic thiosulfinates

(blue) and cyclic disulfides (red). The linear equation for (3—10)a is ΔE⧧ = 0.84ΔEd⧧ – 7.40. The

linear equation for (3—10)b is ΔE⧧ = 0.96ΔEd⧧ – 10.12. (b.) ΔE⧧ vs. ΔEi⧧ of the nucleophilic addition of MeS– towards cyclic thiosulfinates (blue, R2=0.001) and cyclic disulfides (red,

R2=0.05). Computed using M06-2X/6-311++G(d,p) IEF-PCMH2O. The energies are reported in kcal mol-1.

These plots show that there is a linear relationship between ΔE⧧ and ΔEd⧧ for cyclic disulfides (R2 = 0.88) and cyclic thiosulfinates (R2 = 0.84). This suggests that the reactivities are controlled by distortion energy. The interaction energies of cyclic thiosulfinates (3—10)a and cyclic disulfides (3—10)b range from –6.1 to –11.2 kcal mol–1 and –6.9 to –14.4 kcal mol–1,

2 respectively. There is no correlation between ΔE⧧ and ΔEi⧧ (R =0.001 for cyclic thiosulfinates and

R2=0.05 for cyclic disulfides), which implies that the interaction energies do not influence reactivities. We hypothesized that the strain energy would manifest itself as a structural pre- distortion of the reactants, an effect that results in distortion-accelerated reactions.188 To this end, we analyzed (3—10)a and (3—10)b in their equilibrium and distorted transition state geometries; strained cyclic thiosulfinates and disulfides require less distortion to achieve their transition state geometries. This is demonstrated in Figure 3-5, where we show the relationship between distortion energy and the difference in S1—S2 bond lengths in the reactant and transition state (ΔS1—S2).

Figure 3-5. ΔS1—S2 bond length between reactant and transition state of cyclic thiosulfinates

(blue) and cyclic disulfides (red) vs. calculated distortion energy. The combined linear equation is

ΔEd⧧ = 60.55(ΔS1—S2) – 6.39.

The reactions with the lowest activation energies resulted from reactants with the longest

(pre-distorted) S1—S2 bonds at equilibrium. The linear relationship between ΔEd⧧ and ΔS1—S2

2 (R = 0.97) confirms that the S1—S2 pre-distortion of cyclic thiosulfinates results in lower activation energies.

We then scrutinized the geometric and electronic structures of the cyclic thiosulfinates to understand why they are more pre-distorted than the cyclic disulfides. One of the oxygen lone pair orbitals adjacent to the S1—S2 bond is ideally positioned for a hyperconjugative interaction with

∗ the 훔퐒퐒 orbital, via the general anomeric effect. There is a rich literature on this effect from the

189-196 ∗ experimental and theoretical communities. Figure 6 illustrates the possible nO→훔퐒퐒 orbital interaction.

∗ Figure 3-6. (a.) Hyperconjugative nO→훔퐒퐒 orbital interaction. (b.) Computed LUMO of 6a

We quantified this effect with natural bond order (NBO)197 calculations and second order perturbation theory analysis on the optimized structures of the cyclic thiosulfinates. Table 3-2

∗ shows the hyperconjugative nO→σSS interaction energies, and the effect on S1—S2 bond lengths.

∗ Table 3-2. Summary of S1—S2 bond lengths, nO and σSS energies, and the interaction energies

∗ 1 2 ∗ between the nO and σSS orbitals. S1—S2 bond lengths are reported in Å and nO→σSS energies are reported in kcal mol–1.

∗ S1—S2 bond distances, energies for the nO and σSS orbitals participating in the

∗ hyperconjugative interaction, and the energies of the corresponding nO→σSS interactions are given in Table 2. The σ framework of 3a and 4a have relatively high-lying σ orbitals because of the increased p-character associated with the so-called banana bonds.198 As such, 3a and 4a benefit

∗ ∗ from smaller energy gaps between the nO and σSS orbitals, which results in nO→σSS interaction energies of –47.3 and –40.3 kcal mol-1, respectively. (5—10)a have smaller, but similar, orbital

-1 interaction energies (–33.9 to –36.9 kcal mol ), because of the larger energy gap between the nO

∗ ∗ and σSS orbitals and linear 훔퐒퐒 orbitals.

∗ We then compared the σSS orbital energies of cyclic thiosulfinates to those of cyclic

∗ disulfides to quantify the extent in which the nO→σSS hyperconjugative interaction contributes to

∗ nucleophilic substitution rate-enhancement. The HOMO energy of methyl thiolate, the σSS orbital

∗ (LUMO) energies of (3—10)a and (3—10)b, and the occupancies of the 훔퐒퐒 and nO orbitals are shown in Figure 3-7.

Figure 3-7. (a.) Visual representation of MeS– HOMO and corresponding orbital energy. (b.)

∗ Visual representation of σSS orbitals of (3—4)a and (3—4)b and the corresponding orbital energies

∗ and occupancies. (c.) Visual representation of σSS orbitals of (5—10)a and (5—10)b and the corresponding orbital energies and occupancies. Computed M06-2X/6-311++G(d,p) IEF-

PMCH2O. Energies are reported in eV.

∗ ∗ The σSS energies of cyclic thiosulfinates range from –0.54 to 0.53 eV; the σSS energies of

∗ cyclic disulfides range from 0.74 to 2.37 eV. The σSS orbitals of cyclic thiosulfinates are relatively low-lying because of the adjacent oxygen that is inductively electron withdrawing. The electron density of the sulfoxide oxygen disfavors nucleophilic attack of thiolates at S1 because of substantial closed-shell repulsions with the incipient thiolate lone pair orbitals. (3—4)a and (3—

∗ 4)b feature bent σSS orbitals because of the small C—S—S bond angles in the three- and four- membered rings (54° and 78°, respectively). The nO orbitals of cyclic thiosulfinates have reduced occupancies (1.77-1.81e) from the ideal value of 2.00e due to the hyperconjugative interaction;

∗ the σSS orbitals of cyclic thiosulfinates have increased occupancies (0.17-0.20e) from the ideal

∗ value of 0.00e. The large stabilizing nO→σSS energies corroborate Cyclic disulfides have a

∗ significantly lower occupancy of the σSS orbitals ranging from 0.00-0.03e. The generally lower

∗ ∗ σSS orbital energies of cyclic thiosulfinates, resulting from the nO→σSS interaction, lead to stronger frontier molecular orbital interactions with the MeS– lone pair orbitals in the transition state. These more favorable interactions contribute to the general rate-enhancement of nucleophilic substitution towards cyclic thiosulfinates.

3.4 Methods

Initial conformational searches of all structures studied were performed within Maestro

11199 using low-mode sampling with a maximum atom deviation cutoff of 0.5 Å within the

OPLS3e force field in the dielectric constant of water (휺=78).200 Each conformational search produced as maximum of 10 low energy conformations (energy cutoff=100 kJ/mol). Each of these

OPLSe-minimized structures were subjected to geometrical optimization using the hybrid density functional M06-2X/6-311++G(d,p) and the polarizable continuum model using the integral equation formalism variant (IEF-PCM) with the parameters for water.148, 150 Each stationary point was subjected to vibrational analysis from which exactly one negative frequency was identified for transition structures and an absence of negative frequencies for the minima. All DFT calculations were performed using the Gaussian 16201 program. All stationary points for (5—10)a and (5—10)b were optimized using M06-2X/6-311++G(d,p). Transition state scans of (3—4)a and (3—4)b were performed using the coupled cluster method (CCSD).202 The distance between methyl thiolate and the cyclic disulfide/thiosulfinate was scanned between 2.3—2.8 Å. 9 steps at

0.04 Å per step were taken. Each result from this scan was optimized by constraining the forming

– MeS —S2 bond distance using the coupled cluster singles doubles method (CCSD) with the 6-

31+G(d,p) basis set. Single point energy calculations were performed on the constrained transition structures with the same method and basis set [M06-2X/6-311++G(d,p)] as the unconstrained transition states to evaluate activation free energies. NBO analysis and second order perturbation theory analysis on the optimized cyclic thiosulfate reactants was performed to measure the

∗ interaction between the oxygen lone pair and the σSS orbital and the corresponding orbital energies

86 and occupancies. All chemical structures were prepared using CylView.203 The supporting information was prepared using ESIgen.204

3.5 Conclusion

We used DFT calculations to predict the reactivities of a series of 3—10-membered cyclic thiosulfinates towards methyl thiolate for the first time. Similar to previous reports of cyclic disulfide reactivity, the reaction of thiolates towards cyclic thiosulfinates is strain-promoted. Our calculations suggest that the rate of nucleophilic substitution reactions of 6–10-membered cyclic thiosulfinates will be 102-105-fold faster than 6–10-membered cyclic disulfides. Our calculations

∗ show that (3—4)a and (3—4)b have bent σSS orbitals that contribute to their significantly higher strain-dependence and lower activation barriers through increased p-character. The S1—S2 bonds in cyclic thiosulfinates (6—10)a are pre-distorted towards their transition structures and require less distortion energy (ΔEd⧧) to deform reactants from their equilibrium geometries relative to corresponding cyclic disulfides (6—10)b. This results in generally lower activation barriers. A

∗ ∗ hyperconjugative interaction between the oxygen lone pair and the σSS orbitals (nO→σSS) is responsible for the pre-distortion of cyclic thiosulfinates and was verified by decreased

∗ occupancies of nO orbitals and increased occupancies of σSS orbitals. The activation barriers are

∗ further lowered because the σSS orbital energies are decreased by an inductive effect of the adjacent oxygen, which improves transition state frontier molecular orbital interactions. This effect is not

∗ observed in cyclic disulfides which have higher energy σSS orbitals. These theoretical insights have begun to guide our development of new cross-linking tools that avoid toxic dead-end modifications and increase reaction rates in vitro and in vivo. We predict that cyclic thiosulfinate

7a will make the best cross-linking scaffold. Its relatively low strain energy results in a reversible

87 nucleophilic substitution reaction, which will prevent off-target (dead-end) modification of cysteine residues. Additionally, the nucleophilic substitution towards 7a is 7.4 kcal mol–1 lower in activation energy than 7b, resulting in a 105-fold increase in reaction rate

Chapter 4

New Pharmacological Chaperone Scaffold for the Stabilization of Disease- Associated Variants of Cu/Zn Superoxide Dismutase

Daniel P. Donnelly†1,2, Jenifer N. Winters†1, Jeremy B. Conway1, Md Amin Hossain1,2, Matthew G. Dowgiallo1, Nicholas D. Schmitt1,2, Catherine M. Rawlins1,2, Joseph P. Salisbury1, Jared R. Auclair1,2, Roman Manetsch1,3, Lee Makowski1,4, Mary Jo Ondrechen1, Jeffrey N. Agar *1,2,3

†Co-first authors *Corresponding Author 1Department of Chemistry and Chemical Biology, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115, United States 2Barnett Institute of Chemical and Biological Analysis, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115, United States 3Department of Pharmaceutical Sciences, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115, United States 4Department of Bioengineering, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115, United States

This chapter was written with the intent to publish. It is currently submitted to Science Advances.

4.0 Statement of Contribution

Experimental contributions made to this chapter by Daniel P. Donnelly are as follows: all protein purification, all in vitro assays, all mass spectrometry method development and analysis (except

β-lipoic acid experiments performed by Nicholas D. Schmitt, all cell culture, and all in cellulo assays with help from (with help from Md Amin Hossian). Small angle scattering data was collected and analyzed by Jenifer N. Winters. The manuscript was written by Daniel P. Donnelly and Jeffrey N. Agar with careful review and contributions made by Jenifer N. Winters, Jeremy B.

Conway, Md Amin Hossain, Matthew G. Dowgiallo, Nicholas D. Schmitt, Catherine M. Rawlins,

Joseph P. Salisbury, Jared R. Auclair, Roman Manetsch, Lee Makowski, and Mary Jo Ondrechen.

All figures expect Figure 4-3 were prepared by Daniel P. Donnelly.

4.1 Abstract

The destabilization of the Cu/Zn-superoxide dismutase homodimer by genetic mutations or oxidative modifications is known to cause amyotrophic lateral sclerosis (ALS). The initial destabilization of the native non-covalent dimer results in dissociation into monomers that gain a toxic function and form higher-order aggregates within motor neurons. We apply the newly reported cyclic thiosulfinate cross-linking reaction to covalently tether the SOD1 dimer and prevent its toxic dissociation. Cyclic thiosulfinate cross-links stabilize SOD1WT (wildtype) and two of the most prevalent variants SOD1G93A and SOD1A4V. We characterize the stabilization of these proteins in vitro, in cellulo, and in vivo using mass spectrometry, differential scanning fluorimetry, western blotting, and small-angle X-ray scattering and define their therapeutic window using traditional toxicity assays. Our results show that cyclic thiosulfinates are promising pharmacological chaperones for the treatment of ALS.

4.2 Introduction

The destabilization of proteins by mutations or post-translational modifications (PTMs) can promote toxic protein aggregation in a number of diseases.205 Stabilizing the native conformation of these altered proteins with small molecules may prevent these diseases or slow their onset.206-208 For example, tafamidis is a pharmacological chaperone for transthyretin amyloidosis and transthyretin cardiomyopathy that stabilizes the native transthyretin tetramer and prevents its dissociation into toxic monomers prone to aggregation.209-213 Similar native dimer dissociation of the protein Cu/Zn-superoxide dismutase (SOD1) has been implicated in the onset of amyotrophic lateral sclerosis (ALS). Over 180 genetic mutations in SOD1 have been linked to

91 familial ALS (fALS) disease progression48, 145, 214-217 and oxidative modifications on aggregated

SOD1WT have been observed in a subset of sporadic ALS (sALS) cases.218-221 The loss of SOD1’s thermodynamic stability of the protein and its propensity for aggregation are not epiphenomena – they are major risk factors in more aggressive fALS progression.222-224 The success of transthyretin stabilization with tafamidis and a paucity of effective treatments for ALS led others to attempt the stabilization of the SOD1 dimer. In silico screening for molecules with non-covalent interactions at the SOD1 dimer interface was attempted and resulted in the discovery of in vitro SOD1 stabilizers.94, 225 Subsequent x-ray co-crystallization studies with these molecules, however, showed that these chaperones did not bind the intended site at the dimer-interface but instead bound to the β-barrel region associated with aggregation and fibril formation, and ultimately failed to increase thermal stability and prevent aggregation.226

We previously had sought an alternative approach to stabilize the SOD1 native dimer. We postulated that the pair of free cysteines on adjacent monomers, Cys111A and Cys111B (within 9-13

Å of each other, depending upon their residue orientation), could be cross-linked to prevent dimer dissociation and presumably also prevent aggregate formation (Figure 4-1). Thiolate is one of the few reactive amino acid functional groups that is residue-specific. This, combined with the unique reactivity (nucleophilicity and polarizability) of thiolate, allows cysteine residues to be targeted with high specificity. This is evidenced by the numerous drugs that form covalent bonds to cysteine residues, including Antabuse (disulfiram),227-229 the blockbuster proton pump inhibitors omeprazole (Prilosec) and its single enantiomer esomeprazole (Nexium),230 and second generation kinase inhibitors including Afatinib.231, 232 In proof-of-concept experiments, we demonstrated that

G93A G85R tethering the Cys111 cysteine pair on the disease variants SOD1 and SOD1 could thermodynamically stabilize native structure by unprecedented amounts (as much as 40 °C) and

rescue superoxide dismutatse activity.54 The bifunctional maleimides used in these studies,

Figure 4-1. Cyclic thiosulfinates cross-link SOD1 variants via Cys111 residues on adjacent monomers. (a.) Mass spectra of uncross-linked and cross-linked SOD1WT (black), SOD1A4V

(blue), and SOD1G93A (teal) confirming complete covalent cross-link formation. Deconvoluted masses of cross-links samples confirm the proposed mechanism of cross-linking (correspond to variant dimer mass + cross-linker mass – oxygen). Unreacted SOD1 appears at its monomeric mass due to the denaturing solvents used during analysis (50% acetonitrile, 0.1% formic acid) and an increased voltage applied within the region of hypersonic gas expansion. The experimental

93 nominal average masses of uncross-linked SODWT, SOD1A4V, and SOD1G93A, were 15844 Da,

15872 Da, and 15858 Da, respectively. Cross-linked SOD1 is easily distinguishable due to the presence of 2x charge states (from two monomers) and a mass shift corresponding to the mass of the specific cross-linker used minus the oxygen lost during the sulfenic acid/thiolate condensation

36 step. (b.) Crystal structure of SOD1 (PDB: 1SPD) highlighting opposing Cys111 residues on monomer A (Blue) and monomer B (wheat) with representation of thiosulfinate cross-links. (c.)

Mechanism of cyclic thiosulfinate cross-linking.

however, are toxic97 because they target any free cysteine, many of which serve essential catalytic and redox roles in the biological milieu. Recent efforts to stabilize SOD1 employ a pharmacological chaperone, ebselen, which covalently binds each Cys111 (in 2:1 ligand-to-dimer stoichiometry) residue and prevents dimer dissociation through π- π stacking of the adjacent molecules. Lone cysteine modification and its subsequent toxicity, however, are not prevented by this approach.233 Initial reports confirm ebselen-SOD1 conjugate is readily reduced by glutathione in the cytosol yet off-target cysteine modifications that are less easily reduced could prompt an immune response.

In a recent publication, we introduced a new cross-linking tool, cyclic thiosulfinates, which selectively cross-link closely spaced thiol pairs while avoiding “dead-end” modification of lone cysteines.172, 234 Cyclic thiosulfinate cross-linking proceeds through an initial thiol-disulfide exchange between a cysteine thiolate and a cyclic thiosulfinate, leading to ring cleavage and a sulfenic acid intermediate. In the presence of a second cysteine thiolate, the sulfenic acid intermediate undergoes rapid condensation to form a cross-link.

We hypothesize that this chemistry can be used to stabilize the native structure of disease variants SOD1A4V and SOD1G93A and prevent aggregation; in fact, we invented cyclic thiosulfinate cross-linkers for this purpose. In this manuscript, we confirm cyclic thiosulfinate cross-link formation of SOD1WT, SOD1A4V, and SOD1G93A in vitro; test in cellulo toxicity and target engagement; qualify changes in quaternary structure resulting from cross-link formation; and confirm target engagement in human blood samples and in vivo in an ALS mouse model.

4.3 Results

4.3.1 Cyclic thiosulfinates efficiently cross-link SOD1 variants.

Native SOD1WT is a noncovalent homodimer at physiological conditions, but ALS- associated variants increase the propensity of this dimer to dissociate into monomers. We therefore targeted two closely spaced cysteine residues (9-13 Å, depending on rotameric state) on adjacent subunits of SOD1; we join them with cross-linking specific cyclic thiosulfinates to covalently tether the native dimer. We synthesized five, six, and seven-membered cyclic thiosulfinates (1,2- dithiolane-1-oxide, 1,2-dithiane-1-oxide, and 1,2-dithiepane-1-oxide, respectively) and incubated each with purified SOD1WT, SOD1A4V, and SOD1G93A in vitro to assess cross-linking efficiency.

Covalent dimer formation was monitored by direct infusion, intact mass spectrometry (Figure 4-

1a). These cross-links form a thermodynamically irreversible covalent dimer (Figure 4-1b).

Complete cross-link formation on SOD1WT, SOD1A4V, and SOD1G93A by 1,2-dithiolane-1-oxide,

1,2-dithiane-1-oxide, and 1,2-dithiepane-1-oxide occurred rapidly in vitro and depends upon the protein:cross-linker stoichiometry (Figure 4-1a). Monomeric SOD1 a with cyclic thiosulfinate

“dead-end” modification was not observed in any sample. We confirmed cross-link formation at

Cys111 by endoproteinase digest and peptide mass fingerprinting analysis (Figure A3-1).

Figure 4-2. Cyclic thiosulfinate cross-linking increases thermal stability and proceeds in the cellular environment. (a.) Differential scanning fluorimetry (DSF) confirms increases in thermal stability by as much as 15 °C in vitro. Incubation with 1,2-dithiolane-1-oxide, 1,2-dithiane-1-

WT oxide, or 1,2-dithiepane-1-oxide shifts the Tm of: apo-SOD1 from 76.0 °C to 88.3 °C, 90.2 °C,

WT and 89.2 °C, respectively, and the Tm of holo-SOD1 shift beyond 100 °C and out of range of

A4V accurate Tm quantification; SOD1 from 63.2 to 77.1 °C, 79.3 °C, and 79.7 °C, respectively; apo-SOD1G93A from 70.6 °C to 83.2 °C, 85.0 °C, and 84.4 °C, respectively, and holo SOD1G93A from 84.7 °C to 95.9 °C, 97.4 °C, and 95.5 °C, respectively. Derivative plots of SODWT + 1,2- dithiepane-1-oxide indicate the presence of some unreacted protein. Standard errors of triplicate analysis are all less than 0.3 °C (b.) Cross-linking proceeds in HEP G2 cells with an EC50 of ~5

µM (1X PBS) or 10 µM (complete growth media). (c.) MTT assays confirm low toxicity of 1,2- dithiolane-1-oxide and 1,2-dithiane-1-oxide (50x and 100x EC50) and higher toxicity for 1,2-

96 dithiepane-oxide (10x EC50). Results are shown as percentage of viable cells comparing to control.

All samples were assessed in triplicate spanning the cytotoxicity dosing range. Control vehicle wells were treated with 1% DMSO (dimethyl sulfoxide) and 500 µM of chlorpromazine was used as a positive control.

4.3.2 Cross-links increase thermal stability of SOD1.

In a previous study, we identified a common toxic conformation, a destabilized electrostatic loop, shared by fourteen fALS SOD1 variants using hydrogen-deuterium exchange

(HDX), as well as perturbation near the dimer interface.55 Perturbation of the electrostatic loop leads to an exposed epitope responsible for aggregation through non-native SOD1-SOD1 interactions.235 Destabilized SOD1 conformations decrease the thermal stability of fALS variants compared to unmodified SOD1WT.144, 236 We sought to quantify changes in thermal stability resulting from cross-link formation. Purified SOD1WT, SOD1A4V, and SOD1G93A were incubated with and without tenfold concentration of 1,2-dithiolane-1-oxide, 1,2-dithiane-1-oxide, or 1,2- dithiepane-1-oxide in vitro at 37 °C. Thermally-induced denaturation of SOD1 variants was measured using a previously published technique, differential scanning fluorimetry (DSF), whereby samples were incubated with SYPRO® orange dye that fluoresces when bound to hydrophobic residues (Figure 4-2a).54, 237 DSF provides temperatures of unfolding, not reversible

Tms, but in keeping with common practice, these values are referred to as Tms throughout this manuscript. As-purified (i.e., uncross-linked) SOD1WT has two melting points at 75.9 °C and 87.6

°C, associated with its apo and holo forms. Incubation of as-purified SOD1WT with 1,2-dithiolane-

WT 1-oxide, 1,2-dithiane-1-oxide, or 1,2-dithiepane-1-oxide increased the Tm of apo and holo SOD1 by a range of c.a. 12-14 °C and >13 °C, respectively (Figure 4-2). As-purified SOD1A4V has a

97 single Tm at 63.2 °C, and cross-link formation by 1,2-dithiolane-1-oxide, 1,2-dithiane-1-oxide, or

G93A 1,2-dithiepane-1-oxide increased the Tm by 14-16 °C. Uncross-linked SOD1 also has two distinct apo and holo melting points at 70.6 °C and 84.7 °C. Cross-link formation by 1,2-dithiolane-

G93A 1-oxide shifts apo and holo SOD1 Tms by a range of 13-15 °C and 11-13 °C, respectively

(Figure 4-2). In summary, each of the 5, 6, and 7-membered cyclic thiosulfinates stabilized both

WT apo- and holo ALS-associated variants to at-or-above the Tms of SOD1 . This degree of stabilization suggests the tethering of SOD1 subunits could viably prevent destabilization and aggregation.

4.3.3 Cyclic thiosulfinates cross-link SOD1 in cellulo.

To verify cyclic thiosulfinates’ ability to cross the cellular membrane and maintain their activity towards SOD1 within the cellular environment, Hep G2 cells were incubated with increasing concentrations of cyclic thiosulfinates both in 1x PBS and complete growth media

(Figure 4-2b). SOD1 cross-link formation was monitored using SDS PAGE and western blotting.

1,2-dithiolane-1-oxide, 1,2-dithiane-1-oxide, and 1,2-dithiepane-1-oxide successfully cross-linked

WT SOD1 in cells in 1x PBS with an observed EC50 of c.a. 5 µM (Figure 4-2b). The EC50s of cellular cross-linking increased to c.a. 10 µM when incubated in complete growth media indicating minor association with serum proteins. These cell-based studies illustrate an important characteristic of cyclic thiosulfinates – their high effective concentration (EC) compared to monothiols (e.g., glutathione) allows them to pass through the cell without being reduced. By analogy to the EC of cyclic disulfides with similar ring strain determined in seminal small molecule studies,122 we estimate the EC of cyclic thiosulfinates to be at least 10,000 times that of glutathione,

98 and consistent with our cellular studies, not reduced (inactivated) by physiological concentrations of cellular reductants. In previous studies we confirmed that the ternary complex is not reduced

(uncross-linked) by glutathione at cellular concentrations.172

4.3.4 Cyclic thiosulfinates have promising toxicological properties.

The cytotoxicities of 1,2-dithiolane-1-oxide, 1,2-dithiane-1-oxide, and 1,2-dithiepane-1- oxide were measured in Hep G2 using a standard MTT assay (Figure 4-2c). We hypothesized these molecules would have low toxicity due to their resemblance to the widely used dietary supplement α-lipoic acid and the naturally occurring asparagusic acid. We have observed oxidative modification of α-lipoic acid to its thiosulfinate form within dietary supplements (Figure A3-2).

Both 1,2-dithiolane-1-oxide and 1,2-dithiane-1-oxide had relatively high LC50s (~250 µM and

~500 µM, respectively), and 1,2-dithiepane-1-oxide had an LC50 of ~60 µM. Taken together with their EC50, these compounds offer potential therapeutic windows of 50x, 100x, and 10x of their

EC50s, respectively. Extension of ring size beyond six-membered thiosulfinates should begin to favor the forward reaction of thiol-disulfide exchange creating more dead-end modifications and could account for the observed toxicity of 1,2-dithiepane-1-oxide. Although no dead-end modifications of cysteines were observed on SOD1 in vitro, other modifications on essential cysteines in the cellular environment could result in increased toxicity.

Figure 4-3. Cyclic thiosulfinate cross-links do not perturb native structure. Semi-log plots for

SOD1WT, SOD1A4V, and SOD1G93A with and without cyclic thiosulfinate cross-link (a, c, e).

Significant differences are seen between SOD1A4V and SOD1G93A with cross-link compared to without. Kratky plots of SOD1WT with or without cross-link show that all samples are properly folded (b). SOD1G93A and SOD1A4V without cross-links are completely unfolded while with cross- links maintain their structure (d, f). Three-dimensional envelopes of SOD1WT with and without compound. Every sample has an elongated shape that is almost fibril-like even though these samples had different radii of gyration (g). The sample with the largest radius of gyration can fit seven SOD1 subunits within the envelope. Three-dimensional envelopes generated from SAXS data for both SOD1A4V and SOD1G93A variants. Envelopes of SOD1G93A with each compound; 1,2- dithiolane-1-oxide shows slight fluctuations with the SOD1 structure (h). While 1,2-dithiane-1- oxide may cross-link SOD1, it may also cause some structural changes given that the model structure has a poor fit. 1,2-dithiepane-1-oxide fits within the envelope with some flexibility.

Envelopes of SOD1A4V with each compound; the envelope with 1,2-dithiolane-1-oxide fits well

100 with the model structure. However, 1,2-dithiane-1-oxide and 1,2-dithiepane-1-oxide either cause

SOD1 to undergo some structural changes or restructure the dimer interface (i).

4.3.5 Cyclic thiosulfinate cross-linking of SOD1 variants reinstates native conformations

We performed Small Angle X-Ray Scattering (SAXS) experiments on SOD1WT, SOD1A4V, and SOD1G93A with and without cyclic thiosulfinate cross-links to measure potential changes due to cross-link formation. The radius of gyration was calculated for each sample from the SAXS experimental data (Table A3-1). For reference, the SOD1WT crystal structure,238 SOD1G93A and

SOD1A4V homology models were submitted to the WAXSIS server239, 240 to calculate the radius of gyration for a static crystal structure. The WAXSIS server calculates the radius of gyration directly from these models based on explicit-solvent all-atom molecular dynamics (MD) simulations done in YASARA.241, 242 Calculations resulted in a radius of gyration ~21 Å for all SOD1 dimer structures. Experimental results show a range of radii of gyration from the predicted ~21 Å to 2-

3x that of the predicted (Table A3-1).

Figure 4-3 shows scattering data in six separate semi-log plots from SOD1WT, SOD1G93A, and SOD1A4V with and without cyclic thiosulfinate cross-links (Figure 4a, c, e). The scattering patterns for SOD1WT vary most notably at small angles (q ~.05) based on the level of aggregation.

At larger angles (q > .15), there are differences between each pattern (Figure 4-3a). Kratky plots confirmed that the samples were properly folded for a globular protein. Data for SOD1WT cross- linked by 1,2-dithiepane-1-oxide (red), however, show it is larger than the other complexes, consistent with the calculated radius of gyration (Figure 4-3b). The scattering plots for SOD1G93A highlight distinct changes upon addition of cross-linkers (Figure 4-3c). Scattering patterns for cross-linked SOD1G93A are similar except at small angles where there is some deviation due to

101 different levels of aggregation. The Kratky plots of these samples confirm that SOD1G93A without compound was completely unfolded whereas once cross-linked SOD1G93A retained its native-like structure (Figure 4-3d). Similarly, the scattering plot of SOD1A4V without compound (black) suggests the protein was completely unfolded (also verified by the Kratky plot (Figure 4-3f)).

Cross-linked SOD1A4V had scattering patterns consistent with that of a well-folded protein, and the patterns were similar to one another except at smaller angles where the protein cross-linked by

1,2-dithiane-1-oxide and 1,2-dithiepane-1-oxide exhibit some aggregation due to experimental concentrations higher than physiological concentrations.

Figure 4-3g shows three-dimensional shape reconstructions of SOD1WT with and without cross-links. Analysis of SAXS data from the SOD1WT samples produced three-dimensional envelopes that are elongated, fitting more than one dimer into the envelope (Figure 4-3g). These envelopes are similar to other three-dimensional reconstructions that have been reported on aggregates of SOD1.243 Attempts to reconstruct the SOD1G93A and SOD1A4V variants without cross-links did not converge. This behavior is common for unfolded proteins. Models of SOD1G93A cross-linked with both 1,2-dithiolane-1-oxide and 1,2-dithiepane-1-oxide fit relatively well inside the three-dimensional envelopes (Figure 4-3h). Additional unoccupied volume can be seen on either side of the envelope of the 1,2-dithiolane-1-oxide and 1,2-dithiepane-1-oxide cross-linked

SOD1G93A. This volume may account for some flexibility within the linker region of the crystal structure (residues 49-54) based on the region’s high B-factor in the reported crystal structure.244

The 1,2-dithiane-1-oxide linked species does not fit well into the envelope and has additional unoccupied space suggesting that this cross-linked structure may undergo some structural rearrangement. The model of SOD1A4V cross-linked by 1,2-dithiolane-1-oxide fits well inside the envelope with no additional unoccupied volume (Figure 4-3i). The model of 1,2-dithiane-1-oxide

102 cross-linked SOD1A4V suggests some structural changes based on the additional unoccupied volumes on the sides of the reconstruction. By contrast, the envelope of SOD1A4V cross-linked by

1,2-dithiepane-1-oxide has unoccupied volume on the top-left and top-right of the reconstruction.

If the subunits were rotated a few degrees, however, the structure may fit well inside the envelope

(Figure 4-3i). Therefore, this compound may restructure the dimer interface of SOD1.

4.3.6 Cyclic thiosulfinates cross-link SOD1G93A in vivo.

To confirm that target engagement is possible in vivo, hemizygous mice expressing human

SOD1G93A (“fast-line” B6SJL.SOD1-G93A)245, 246 were bred and dosed with 1,2-dithiolane-1- oxide, 1,2-dithiane-1-oxide, and 1,2-dithiepane-1-oxide via lateral tail vein injection. Mice were euthanized twenty minutes post-injection and blood was collected by cardiac puncture. We developed a modified “mass-spectrometry-friendly” isolation procedure that employed the hemoglobin precipitation of Fridovich and McCord, and SOD1G93A was purified from blood and analyzed by intact LC- MS.50 Chromatograms and corresponding mass spectra of monomeric and cyclic thiosulfinate mediated dimer SOD1G93A are shown in Figure 4-4. Western blot analysis of the same blood samples of the treated mice confirmed the presence of c.a. 50% of covalent

SOD1G93A dimer.

103

Figure 4-4. Target engagement in mouse-model confirmed via LC-MS. (a.) SOD1G93A purified from hemizygous mice post cyclic thiosulfinate dosing confirm activity in vivo via intact LC-MS.

(b.) Western blot analysis using SOD1-specific antibodies supplement LC-MS results and confirm

~ 50% of SOD1G93A from blood is cross-linked. Lanes 1, 3, and 5 contain SOD1G93A purified from the same mouse not dosed with cyclic thiosulfinate compound. The red D and M highlight the location of dimer and monomer of SOD1G93A.

4.4 Conclusion

In previous studies we developed a viable strategy for stabilizing ALS-associated SOD1 variants by covalently cross-linking the Cys111 pair at the dimer interface with toxic maleimides.

This strategy not only prevented dissociation of the native dimer, but rescued wildtype-like activity. These molecules were toxic and surely modified countless essential cysteine residues in cellulo and in vivo. To preserve the stabilization offered by cross-linking, while mitigating its toxic

104 effects, we developed the first cross-linking molecules that would mitigate dead-end modifications, the major off-target modification of cross-linking molecules. These cross-linkers function effectively in vitro against two of the most common ALS-associated SOD1 variants, in

G93A cellulo at promisingly low EC50s, and cross-linked SOD1 in a mouse model. Solution scattering confirmed that cross-link formation does not significantly modify native quaternary structure, suggesting immune response due to conformational changes is mitigated. The results presented here confirm a promising new strategy for ALS therapy development.

We do, however, recognize that these studies are only the first step in therapy development.

Proper DMPK studies are required to define toxicity and metabolism profiles of cyclic thiosulfinates and are underway. Here we use the simplest form of cyclic thiosulfinates which, although they avoid modifying many lone cysteines, could cross-link any closely spaced cysteine pair in the biological milieu. A computational search of the protein database identified at least 10 other proteins that could be potential sites of off-target cross-link formation. Through future library development by medicinal chemistry, we hope to develop protein-selective cyclic thiosulfinates, which has the potential to further reduce their toxicity and increase their activity. Due to the lack of effective treatments for ALS, these molecules show promise as a starting point for future therapy development.

4.5 Methods

4.5.1 Synthesis of Cyclic Thiosulfinates

1,2-dithiane-1-oxide and 1,2-dithiepane-1-oxide were prepared as previously reported.172

Preparation of 1,2-dithiolane-1-oxide was adapted from a known literature procedure.247 Ceric

105 ammonium nitrate (10.3 g, 21.9 mmol, 4 equiv.) was added portion-wise over 30 mins to a solution of 2,2-dimethyl-1,3-dithiane4 (0.813 mg, 5.48 mmol, 1 equiv.) in MeCN:H2O (4:1, 91 mL) and allowed to stir for 1 h. The reaction mixture was added to water (100 mL), diluted with dichloromethane (100 mL) and extracted with dichloromethane (3 x 50 mL). The combined organic layer was washed with water (4 x 50 mL) and subsequently dried over sodium sulfate. The solvent was removed under reduced pressure and purified by flash column chromatography on silica gel with 4% EtOH / EtOAc yielded 1,2-dithiolane-1-oxide as a light-yellow oil (298 mg,

44%). Rf = 0.36 (4% EtOH / EtOAc). 1H NMR (400 MHz, CDCl3, δ): 2.87-2.93 (m, 3H), 3.26-

3.31 (m, 1H), 3.60-3.63 (m, 1H), 3.72-3.76 (m, 1H); 13C NMR (100 MHz, CDCl3, δ): 29.6, 38.0,

63.1.

4.5.2 Expression and Purification of SODWT, SOD1A4V, and SOD1G93A

Expression and purification of SOD1 variants was achieved as previously published.54, 144,

145 Human SOD1 cDNA cloned into the yeast expression vector YEp-351 was transformed into

EGy118ΔSOD1 yeast and grown at 30 °C for 36-48 h. Cultures were pelleted and stored at –80

°C until purification. To start purification, pellets were thawed on ice, suspended in lysis buffer

(200 mM Tris HCl, 0.1 mM EDTA, 50 mM NaCl, 1 Pierce Protease Inhibitor/50mL) and lysed using 0.5 mm glass beads in a blender. Lysate was centrifuged, and supernatant was collected and subjected to a 60% ammonium sulfate cut on ice. Protein precipitate was again pelleted, and supernatant was collected and diluted with ~0.19 volumes of buffer (50 mM sodium phosphate,

150 mM sodium chloride, 0.1 M EDTA, 0.25 mM DTT, pH 7.0) to a final concentration of 2.0 M ammonium sulfate. The sample was then purified using a phenyl-Sepharose hydrophobic interaction chromatography column with a 300 mL linearly decreasing salt gradient from 100%

106

Buffer B (2.0 M ammonium sulfate, 50 mM sodium phosphate, 150 mM sodium chloride, 0.1 M

EDTA, 0.25 mM DTT, pH 7.0) to 100% Buffer A (50 mM sodium phosphate, 150 mM sodium chloride, 0.1 M EDTA, 0.25 mM DTT, pH 7.0). Fractions containing SOD1 typically elute between 1.6 and 1.1 M ammonium sulfate, which was confirmed using SDS PAGE. Fraction containing SOD1 were pooled and buffer exchanged into 10 mM Tris (pH 8.0). The sample was then loaded onto a Mono Q 10/100 anion exchange chromatography column and eluted using a

200 mL linearly increasing salt gradient from a low salt buffer (10 mM Tris, pH 8.0) to a high salt buffer (10 mM Tris, pH 8.0, 1 M sodium chloride). The gradient is run from 0-30% 10 mM Tris, pH 8.0, 1 M sodium chloride and SOD1 eluted between 5-12% 10 mM Tris, pH 8.0, 1 M sodium chloride. SOD1 fractions from anion exchange were confirmed using gel electrophoresis. Final

SOD1-containing fractions were pooled together, washed three times with 10 mM ammonium bicarbonate in a Millipore amicon centrifugal filter (MWCO 10000 Da), and buffer exchanged into 10 mM ammonium acetate pH 7.4. Proteins were flash frozen and stored at -80 °C until use.

4.5.3 Confirmation of Cross-link Formation

SOD1WT, SOD1A4V, and SOD1G93A stock solutions were diluted to 63 µM in 10 mM ammonium acetate, pH 7.4. 1,2-dithiolane-1-oxide, 1,2-dithiane-1-oxide, and 1,2-dithiepane-1- oxide were prepared at a concentration of 315 µM (fivefold protein concentration) in 10 mM ammonium acetate, 2% methanol. Protein and cross-linker samples were combined in equal volumes (final concentration 37 µM SOD1, 185 µM cross-linkers, 1% MeOH) and incubated at

37 °C for 5 hours. Complete cross-linking was confirmed by mass spectrometry on a 9.4T Bruker

SolariX XR as previously described.172 Prior to infusion, samples were briefly (c.a. 30 s) treated with 10% formic acid and diluted to 1 µM in 50:50 acetonitrile:water, 0.1% formic acid. During

107 analysis, 32 scans were acquired in positive mode and averaged. Funnel 1 and skimmer 1 were kept at 150 V and 20 V, respectively, and funnel RF amplitude was held at 60.0 Vpp. In vivo cross- linking was confirmed using an H-Class Acquity UPLC system coupled to a Xevo G2-S Q-TOF mass spectrometer (Waters Corp, Milford, MA). The columns used was an Acquity UPLC Protein

BEH C4 (300 Å pore size, 1.7 µm particle size, 100 mm bed length, 2.1 mm ID x 100 mm) column

(Waters Corp, Milford, MA).

4.5.4 Differential Scanning Fluorimetry

The effect of cross-link formation on the thermal stability of SOD1 was determined by differential scanning fluorimetry as previously described.54 SOD1A4V, SOD1G93A, and SOD1WT

(20 μM in 10 mM ammonium acetate, pH 7.4) were incubated with tenfold molar excess 1,2- dithiolane-1-oxide, 1,2-dithiane-1-oxide, and 1,2-dithiepane-1-oxide for 5 hours at 37 °C in a 96- well reaction plate (MicroAmp Fast Optical, Applied Biosystems, Life Technologies Corporation,

Carlsbad, California, USA). After incubation, SYPRO Orange (Life Technologies Corporation,

Carlsbad, California, USA), an environmentally sensitive fluorescent dye which is quenched in an aqueous environment but becomes unquenched once it binds hydrophobic residues, was added to the reaction mixture to a final concentration of 20X, and the plate was spun down to eliminate bubbles. The reaction plate was monitored using a real time PCR machine (Applied Biosystems,

Life Technologies Corporation, Carlsbad, California, USA) with a temperature gradient ranging from 25 to 99.9 °C. Samples containing buffer without protein mixed with either cross-linkers dissolved in DMSO (5% final DMSO conc.) or 5% DMSO alone were analyzed; background due to these reaction components was negligible. All samples were run in triplicate. Tms were

108 quantified by averaging the negative first derivative of relative fluorescence of all three runs and identifying the local minima.

4.5.5 Proteolytic Digestion and MALDI-TOF-MS Peptide Analysis

Samples of SOD1A4V with or without tenfold molar excess of 1,2-dithiane-1-oxide in 10 mM Tris HCl, pH 7.4 were incubated for 4 hours at 37 °C. After incubation, samples were alkylated with iodoacetamide (100 mM for 30 minutes), heated to 75 °C for 20 minutes, and then treated with two volumes of Poroszyme immobilized trypsin (Applied Biosystems, Life

Technologies Corporation, Carlsbad, CA, USA) at 37 °C for 15 minutes, mixing every few minutes to keep beads suspended. Beads were removed by centrifugation and digest reactions were analyzed using a microflex MALDI-TOF mass spectrometer (Bruker Daltonics, Billerica,

Massachusetts, USA) in reflectron mode in the 2-5 kDa range and linear mode in the 4-20 kDa range. Spectra were calibrated using Peptide and Protein I Calibrant (Bruker Daltonics, Billerica,

Massachusetts, USA). Matrix only and trypsin digest reaction mixture without SOD1 spectra were acquired as negative controls. Spectra were analyzed in flexAnalysis and BioTools 3.2 (Bruker

Daltonics, Billerica, Massachusetts, USA). Peptide mass fingerprinting was performed using

MASCOT (Matrix Science, Boston, MA, USA) using trypsin as the enzyme with up to 5 missed cleavages, 100 ppm mass tolerance, and cysteine carbamidomethylation as a variable modification.

4.5.6 Hep G2 cell culture

The Hep G2 human hepatocarcinoma cell line (American Type Culture Collection, ATCC,

Manassas, VA) was cultured using EMEM (Eagle’s minimal essential medium, ATCC) supplemented with 10% FBS (fetal bovine serum, ATCC) and 1% penicillin streptomycin (10,000

109 units/mL penicillin and 10,000 µg/mL streptomycin, Fisher scientific, Hampton, NH). The cells were grown, and sub cultured around 72 hours at 37°C, 5% CO2 under controlled humidity. The cells growth and morphology were inspected using an inverted Zeiss microscope.

4.5.7 Cell viability assay/Cell cytotoxicity assay

Using manufacturer’s instructions, Hep G2 cells were cultured for 24 hours with a density

5 of 2 x 10 cells/mL in a 96-well culture plate at 37 °C, 5% CO2 under controlled humidity. The cells were treated with cyclic thiosulfinates (1,2-dithiolane, 1,2-dithiane, 1,2-dithiepane) ranging with 1, 15, 10, 20, 40, 60, 80, 100, 120, 150, 250, 400, 600, 1000 µM for 24 hours. Cell cytotoxicity was performed via MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay at

570 nm absorbance using BioTek synergy H1 (Vermont, USA) plate reader. The MTT stock was made at 5 mg/mL using 0.9% sodium chloride solution which was purchased from Sigma Aldrich

(St. Louis, MO). Twenty-four hours after dosing the plate, the cultured medium was aspirated and replaced with the equal volume of MTT solution (tenfold dilution) which was diluted using the

EMEM complete media. The plate was incubated for 4 hours (37 °C, 5% CO2). The MTT solution was replaced by the equal amount of acidified isopropanol (0.3% hydrochloric acid (v/v)) the formazan was dissolved by shaking the plate gently for 30 minutes. The absorbance was measured at 570 nm for both the samples and acidified isopropanol for background calculation.

4.5.8 Preparation of SOD1 Samples for SAXS Analysis

SOD1WT, SOD1A4V, SOD1G93A was prepared at 3 mg/mL (approx. 190 µM) in HEPES buffer (115 mM NaCl, 1.2 mM CaCl2, 1.2 mM MgCl2, 2.4 mM K2HPO4, 20 mM HEPES, pH 7.4).

Stock solutions of 1,2-dithiolane-1-oxide, 1,2-dithiane-1-oxide, and 1,2-dithiepane-1-oxide were 110 freshly made at 10 mM in HPLC grade methanol and diluted to 946 µM in HPLC grade water

(fivefold concentration of protein). 100 µL of 3 mg/mL protein was mixed with 110 µL of 946

µM. Control sample contained 5% MeOH (final conc. 2.5%). Samples were then incubated at 37

°C for 6 hours to ensure complete cross-linking. After incubation, excess compound was buffer exchanged out of each sample using a 10 kDa MWCO ultrafiltration device. 320 µL is diluted to

15 mL into HEPES buffer, spun down to approximately 500 µL, and resuspended in another 15 mL of HEPES buffer. After the final spin, samples are removed and concentrated in a smaller ultrafiltration device and brought to approximately 70 µL final volume (4.7 µM). Samples were flash frozen and stored at –80 °C.

4.5.9 Mouse Dosing and SOD1 Isolation

Mice hemizygous for SODG93A (B6SJL-Tg(SOD1*G93A)1Gur/J, JAX stock #002726 ) were purchased from The Jackson Laboratory. Mice were bred to express YPF in neurons in anticipation of BBB-penetration assays and MALDI MSI (Imaging) analysis. 1,2-dithiolane, 1,2- dithiane, and 1,2-dithiepane were prepared from 100 mM stock solutions (100% DMSO) and diluted to 5 mM in 1x PBS (final 5% DMSO). Syringes were prepared with 150 µL of cyclic thiosulfinate solution and 30-gauge needles. Solutions were injected slowly (~10 min) via the lateral tail vein and mice. 20 min post injection, mice were euthanized by CO2 exposure. Blood was collected by cardiac puncture. Brains, livers, and kidneys were collected and immediately left to freeze on liquid nitrogen. Blood cells were washed three times with an acid citrate dextrose solution (0.48% citric acid, 1.32% sodium citrate, 1.47% glucose) and centrifuged at 2000 rpm at

4 °C for 5 minutes. After the first wash, plasma was collected and stored flash frozen in liquid nitrogen. Following the third wash, the supernatant was removed and blood cells were lysed by

111 the addition of 10 mM ammonium acetate to a total of 500 µL of hemolysate. To the hemolysate,

0.15 equivalents of cold chloroform and 0.25 equivalents of cold ethanol was added. Samples were vortexed for 15 mins at 4 °C and centrifuged at 3000 RPM for 10 minutes. The supernatant was collected and stored for LC-MS analysis. Prior to LC-MS analysis, samples were acidified to 8% formic acid.

4.5.10 Small-angle x-ray solution scattering (SAXS)

SAXS data were collected at the G1 beamline at Cornell High Energy Synchrotron Source

(CHESS). For each sample, two measurements were taken, one of the protein with buffer and one with buffer by itself. Solution scattering data were captured every second for 10 frames. The 10 frames of both buffer and protein were then averaged and the buffer was subtracted out to get the scattering for the protein. Samples were run in a 96-well plate and held at 4 °C continuously. Data collection was in the scattering angle (q) range of 0.008 to 0.71 Å-1 and processed using the software, RAW.248

4.5.11 SAXS Data Analysis and Reconstruction of Molecular Envelopes

Programs within the ATSAS suite249 were used to determine the three-dimensional molecular envelopes for SOD1WT, SOD1A4V, and SOD1G93A with and without cyclic thiosulfinate cross-linkers using the x-ray solution scattering data. The GNOM program250 was used to evaluate the pair distribution plot using an indirect Fourier transform. The Dmax, or maximum intra-particle distance, for SOD1WT was much higher (~150 Å) than other variants due to the amount of aggregation, shown as a sharp spike at very low scattering angles (q). Both SOD1G93A and

A4V SOD1 with compound had Dmaxs ranging between 70-85 Å whereas these proteins without

112 compound had a much higher Dmax of over 200 Å due to protein unfolding. The GASBOR program251 was used to generate three-dimensional ab initio models of connected beads to fit the

GNOM data, with the number of beads set approximately to the total number of amino acids in the

SOD1 constructs. In order to assess the uniqueness of these solutions, 10 bead models were generated without any symmetry applied, then compared and averaged. Figures were produced using CHIMERA252 followed by superimposition of envelopes.

113

Chapter 5 Best Practices and Benchmarks for Mass Spectrometry of Intact Proteins

Daniel P. Donnelly1†, Catherine M. Rawlins1†, Caroline J. DeHart2, Luca Fornelli2, Luis F. Schachner2, Ziqing Lin3, Jennifer L. Lippens4, Krishna C. Aluri1,14, Richa Sarin1,15, Bifan Chen3, Carter Lantz13, Wonhyeuk Jung13, Kendall R. Johnson1, Antonius Koller1, Jeremy J. Wolff5, Iain D. G. Campuzano4, Jared R. Auclair6, Alexander R. Ivanov1, Julian P. Whitelegge7, Ljiljana Paša-Tolić8, Julia Chamot-Rooke9, Paul O. Danis10, Lloyd M. Smith11, Yury O. Tsybin12, Joseph A. Loo13, Ying Ge3, Neil L. Kelleher2, Jeffrey N. Agar1*

1Barnett Institute of Chemical and Biological Analysis and Departments of Chemistry & Chemical Biology and Pharmaceutical Sciences, Northeastern University, Boston, Massachusetts, USA 2Departments of Chemistry and Molecular Biosciences and the Proteomics Center of Excellence, Northwestern University, Evanston, Illinois, USA 3Department of Cell and Regenerative Biology, Department of Chemistry, Human Proteomics Program, University of Wisconsin-Madison, Madison, Wisconsin, USA 4Discovery Analytical Sciences, Amgen, Thousand Oaks, California, USA 5Bruker Daltonics, Billerica, Massachusetts, USA 6Biopharmaceutical Analysis Training Laboratory, Northeastern University, Burlington, Massachusetts, USA 7 The Pasarow Mass Spectrometry Laboratory, The Jane and Terry Semel Institute for Neuroscience and Human Behavior, David Geffen School of Medicine, UCLA, Los Angeles, California, USA. 8Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, Richland, Washington, USA 9Mass Spectrometry for Biology Unit, Institut Pasteur, USR 2000, CNRS, Paris, France 10Eastwoods Consulting, Boylston, Massachusetts, USA 11Department of Chemistry, University of Wisconsin, Genome Center of Wisconsin, University of Wisconsin-Madison, Madison, Wisconsin, USA 12Spectroswiss, 1015 Lausanne, Switzerland 13Department of Chemistry and Biochemistry, Department of Biological Chemistry, and UCLA/DOE Institute of Genomics and Proteomics, University of California-Los Angeles, Los Angeles, California, USA 14Alnylam Pharmaceuticals, Cambridge, Massachusetts, USA 15Biogen, Cambridge, Massachusetts, USA

†These authors contributed equally to the manuscript *Corresponding Author

114

This chapter was written with the intent to publish. It is currently under review at Nature Methods.

115

5.0 Statement of Contribution

Experimental contributions made to this chapter by Daniel P. Donnelly are as follows: Mass spectrometry method development on Bruker SolariX and Waters Xevo G2S, method development of denaturing analysis of membrane proteins, optimization and analysis of protein preparation techniques. The manuscript was written by Daniel P. Donnelly, Catherine M. Rawlins, and Jeffrey

N. Agar. Figures were made by Daniel P. Donnelly, Catherine M. Rawlins, and Jeffrey N. Agar.

116

5.1 Abstract

One gene can give rise to many functionally distinct proteoforms, each of which has a characteristic molecular mass. Many laboratories, however, forgo the challenging mass analysis of intact proteins and instead resort to established methods for proteolytic digests. Here, we review the driving forces of protein signal suppression, provide a decision tree that guides researchers to robust protocols for mass analysis of intact proteins (antibodies, membrane proteins, etc.) from mixtures of varying complexity, and present cross-platform analytical benchmarks to allow users to gauge their proficiency.

5.2 Introduction

Mutations, polymorphisms, RNA processing, and post-translational modifications (PTMs) such as acetylation, methylation, and phosphorylation can lead to a single gene producing many functionally distinct “proteoforms.”253 The distinct proteoforms that arise from these phenomena can have different effects on important biological processes including gene regulation, cell signaling, and protein activity; consequently, the ability to characterize these species is essential to understanding the biological response to disease. The identity of a proteoform can often be inferred254 from an accurate experimentally determined intact mass.255 The sensitivity of intact mass-based proteoform ID can be increased by determining the relative abundance of a particular amino acid by using isotopic labeling; by using mass similarities to cluster proteoforms into gene families; and by reducing the search space using sample-specific search databases.254 Localizing

PTMs, and in some cases definitive proteoform ID, requires tandem mass spectrometry (MSn) analysis. Measurement of intact protein mass and MSn have been coined “top-down” mass

117 spectrometry,256-260 and have their origins in Fenn and colleagues’ discovery that large biomolecules could be ionized261 and fragmented262-264 using electrospray ionization (ESI) MS.

One advantage of ESI over the alternative “soft” ionization method, Matrix Assisted Laser

Desorption-Ionization (MALDI), is that ESI imparts more charge per protein. This enables the mass determination of large biomolecules using mass analyzers with moderate m/z upper limits

(e.g., m/z ≤ 4000), which happen to offer the highest resolving power. Higher charge per molecular mass also facilitates gas-phase fragmentation and, therefore, the characterization of primary sequence and PTMs by MSn.265, 266 The superior fragmentation and ability to interface with liquid chromatography (LC) systems are the reasons ESI—the focus of this study—is used for most top- down MS characterization. Projects requiring rapid MS analysis,267 the ability to analyze hundreds of proteins in a single spectrum, protein imaging capabilities, or less signal suppression by common protein buffer components268 may be better suited for MALDI-MS. The in-gel digestion method,269 multi-dimensional chromatography techniques,270 and Filter Aided Sample Preparation

(FASP)271, 272 protocols were designed to remove signal suppressing components from peptide samples, and have been employed in tens of thousands of LC-ESI-MS/MS proteomics studies.273

These “clean up” techniques are applicable to “bottom-up” MS, namely MS preceded by a endoproteinase digestion step, but unfortunately, are not compatible with intact mass measurement.

Top-down approaches provide additional layers of information: detecting modifications that are removed or scrambled274 during peptide sample preparation (e.g., S-thiolation); elucidating functional relationships (e.g., “cross-talk”) between PTMs on the same protein molecule; characterizing drug-target interactions; observing important modifications on biopharmaceuticals; and identifying and quantifying distinct proteoforms that would have been convoluted by

118 endoproteinase digestion.275-279 In addition, sample preparation for intact protein MS comprises fewer steps and does not require chemical modification (e.g., reduction and alkylation), thereby reducing the number of experimental artifacts.280 Current top-down sample cleanup methods (e.g., protein precipitation281 or molecular weight cut off (MWCO)-ultrafiltration) are not applicable to all sample types or downstream MS analyses. The demand for robust, generally applicable methods for intact protein MS is the most common request made to members of the Consortium for Top-Down Proteomics282, 283 (http://topdownproteomics.org/), and addressing this unmet need is, therefore, the goal of this study.

We begin by quantifying the signal suppression associated with common buffer components and biotherapeutic excipients. This provides the rationale for most failed intact MS measurements, and in addition, a path to designing MS-compatible buffers. Next, a decision tree is presented, which is based upon sample composition and experimental goals, and guides users to a best-practices protocol and corresponding benchmark data. In summary, we present here the first step towards a universal guide for practitioners at all levels of expertise to acquire high-quality intact protein mass spectra by ESI-MS.

5.2.1 Origins of Signal Suppression and Signal Spreading

Biological, biochemical, and biotherapeutic sample preparations usually contain numerous interfering substances (e.g., salts, detergents, chaotropes, and buffers) that lead to signal suppression during ESI-MS analysis. To provide a theoretical context for this work, we will describe the two major drivers of the quality of intact protein (positive ion) ESI-MS, and how these are affected by interfering substances. The first driver of quality is the formation of desolvated protein ions, which can be understood in terms of a few critical steps during the ESI process.284-286

119

Interfering substances generally affect the ESI process after the formation of nanodroplets at the

Rayleigh charge limit. Two salient, often opposing, processes that occur within these nanodroplets are the partitioning of net charge towards the droplet surface and the minimization of solvation energy. Polar species such as salts and native proteins partition towards the droplet interior to optimize solvation energy; their ionization, therefore, requires evaporation of solvent molecules.266

Hydrophobic species such as detergent monomers and unfolded proteins migrate to the droplet surface to optimize solvation energy and, in a faster process that requires less energy, evaporate or are ejected. Many of the techniques presented here for reducing signal suppression can be rationalized within the framework above. For example: organic solvents that decrease surface tension should promote the ionization of both polar and non-polar analytes; detergents partition to the surface where they can outcompete analytes for a limited number of protons; organic solvents and acids that unfold proteins should promote ejection-based ionization; native MS (nMS) requires greater desolvation energy and is more sensitive to polar contaminants.

The second driver of the quality of intact protein MS is signal spreading (i.e., the distribution of the signal from a single proteoform across multiple channels), which increases with protein mass. Each channel has its own respective noise; consequently, the cumulative noise increases proportionally to the number of channels. The ESI process promotes signal spreading, via adduct formation, by increasing the concentrations of interfering substances and proteins.

Heavy isotopes and charge states further distribute signal intensity across multiple channels; the former can be mitigated by isotope depletion.287 This work concerns experimental techniques that minimize signal spreading (i.e., increase signal-to-noise ratio, or S/N), including employing nMS to reduce the number of charge states, and the use of volatile salts (e.g., ammonium acetate) or purification to minimize the effects of alkaline salts.

120

5.3 Results

5.3.1 Defining the Problems: Signal Suppression by Common Buffer Components

Using the intact protein standard mixture (ubiquitin, myoglobin, trypsinogen, and carbonic anhydrase) established by the National Resource for Translational and Developmental Proteomics

(NRTDP),288 common buffer components (Figure 5-1c) were evaluated in order to quantify the concentration required for 50% signal suppression during direct infusion ESI. By analogy to IC50

(half-maximal inhibitory concentration) nomenclature, we termed this metric SC50 (half-maximal suppression concentration) (Figure 5-1 and Figure A4-1). At their typical concentrations, all common buffer additives significantly suppressed ESI signal. Consistent with the mechanisms of

ESI ionization described above, detergents produced the highest degree of signal suppression, less volatile salts (e.g., metallic) produced intermediate levels, and volatile components the lowest suppression. Additional details of the experimental parameters employed here are provided in the

Supplementary Note 5-1 and 5-2 . The SC50 values given in Figure 5-1 allow users to design

MS-compatible buffers. In addition, the SC50 and buffer composition serve as the entry point into the decision tree outlined below, leading users to the appropriate protocol. While the trends in SC50 values reported here should generally be consistent across MS platforms, parameter-dependent variations in the reported values are likely (in particular flow rate, voltages, temperatures, and pressures that affect ionization and desolvation). Here, for example, we calculate SC50 obtained by direct infusion using a standard microflow ESI source (ca. µl/min), but nano-ESI (<µl/min) is less affected by salts due to the order of magnitude decrease in initial droplet size.289, 290

121

Figure 5-1. Common buffer components suppress MS signal. a) MgCl2 reduces signal (and signal- to-noise ratio, S/N) in a concentration-dependent manner, b) Fit of experimental data to determine the concentration of MgCl2 required for 50% signal suppression (SC50, indicated by black arrow), c) Table of common buffer components and the concentration threshold for 50% SC50

(Experimental data curves and their fits are shown in Figure A4-1) and the calculations in the

122

Supplementary Note 5-2. Mass spectrometry compatible detergents are discussed in protocol 2b

. *Signal suppression by detergents is less pronounced above their critical micellar concentration

(CMC) (described in protocol 2b). The antibody formulation buffer included 10 mM arginine, 10 mM tris-HCl, 10 mM histidine, 10 mM KH2PO4, 10 mM citric acid, pH 5.5.

5.3.2 The Intact Protein MS (IPMS) Decision Tree for Choosing the Appropriate Experiment

The IPMS Decision Tree (Figure 5-2) directs practitioners to a protocol or a combination of protocols based upon buffer composition, the number of proteins in the sample, and whether native or denaturing conditions are to be employed. Consider, for example, a purified protein in phosphate buffered saline (PBS). Based upon the 1.5 mM SC50 exhibited by NaCl (Figure 5-1c) and the 137 mM NaCl present in PBS, a protein sample in PBS requires a 91-fold dilution in order to achieve 50% of the potential MS signal. Therefore, if the concentration of protein is greater than

90 µM, and salt adducts will not impede data analysis, the sample can be diluted following

Protocol 1 (P.1). Otherwise, sample cleanup by ultrafiltration using spin cartridges with a MWCO- membrane is recommended following Protocol 2 (P.2). Interest in certain PTMs (e.g., metallation) or protein complex quaternary structure would then dictate the use of native MS methods following Protocol 4b (P.4b); otherwise the denaturing MS Protocol 4a (P.4a) is recommended. The objective of this decision tree is to provide a proven workflow for any sample, not to rule out alternative methods. For example, depending upon sample stability, user expertise, and available resources, precipitation (P.3), size exclusion “spin cartridges”, or LC (P.5) could be suitable alternatives to MWCO-ultrafiltration. All protocols and benchmarks referenced by the

123 decision tree and alternative methods are summarized below and further detailed in Supplementary

Methods and Notes.

124

Figure 5-2: Decision Tree for intact protein sample clean-up, preparation, and analysis. The red dashed line, for example, denotes the decision path for the native MS analysis of a membrane protein. *LC can also be applied at this stage in the decision tree. †Minimally complex protein samples prepared by P.3 can be analyzed via denaturing direct infusion (P.4a) if desired. Protocols

P.1 through P.5 are included in the main text and in Supplementary Methods. Refer to Figure A4-

2 for a recommended example of 2D separation. ‡Other viable alternative separation techniques include Capillary Zone Electrophoresis (CZE), Ion Exchange (IEX), and Size Exclusion

Chromatography (SEC).

5.3.3 Protein Standards and Benchmarks

To promote standardization and allow users to benchmark their own data using readily available proteins, representative results are provided for each protocol using the following commercially available standards: 1) the NRTDP intact protein standard mixture (See

Supplementary Notes for preparation instructions), 2) NIST monoclonal antibody reference material 8671 (“NISTmAb”), containing humanized IgG1ĸ in 12.5 mM L-histidine, 12.5 mM L- histidine HCl (pH 6.0), and 3) Sigma bacteriorhodopsin from Halobacterium salinarum (B0184).

Benchmarks for mass accuracy depend upon the instrumentation platform and have been reviewed.255, 291-296 Rules of thumb include requiring 10 ppm accuracy for modern Fourier transform MS and 20 ppm accuracy for modern quadrupole-time-of-flight (QTOF) MS. We suggest the use of Proforma Notation297 for standardized proteoform nomenclature, and note that the PeptideMass tool (https://web.expasy.org/peptide_mass/) can be used to calculate the mass of a given sequence, or of proteoforms contained within the Uniprot database.

125

5.3.4 Dilution

Protocol 1: Sample Preparation by Dilution of Interfering Substances

Consistent with the mechanisms of ESI and signal spreading detailed above, common buffer components render proteins undetectable (Figure 5-3). Minimally complex, concentrated protein solutions can often be analyzed by direct infusion, following dilution to approximately 1

µM final protein concentration in the appropriate sample buffer. Consider using this protocol if dilution can decrease the concentration of a given interfering substance below its SC50 value

(Figure 5-1, Supplementary Protocol 5-1). Assuming a practical upper limit of approximately

10 mM protein concentration, this protocol is potentially applicable to any of the components listed in Figure 5-1. As detailed above, however, nMS utilizes an ESI process that is more sensitive to many interferents, including salts. Consequently, dilution is less likely to adequately improve nMS. See Protocol 4 for methods to dilute native proteins into whichever solution will be used to introduce samples to the MS. Mass spectra obtained by this method have the lowest S/N of any of the protocols in this work and may contain adducts.

5.3.5 Ultrafiltration

Protocol 2: Sample Preparation Using MWCO-Ultrafiltration

We recommend remediating non-volatile salt adducts by employing buffer exchange into a solution of volatile salts. The MWCO of the ultrafiltration device should not exceed half the molecular mass of any given protein in a sample to prevent possible sample loss. No particular pH is optimal for all proteins, but pH extremes should be avoided, as should pH that is equivalent to a protein’s pI, where protein solubility is at a local minimum.298 We utilize ammonium acetate 126 throughout these protocols due to its volatility and ability to act as a stabilizing background electrolyte during ESI.299 Ammonium acetate provides maximal buffering around pH 4.75

(acetate) and 9.25 (ammonium) and results in a neutral pH upon dissolving in water (~pH 6.5-7).

Prior to the addition of protein sample, MWCO-ultrafiltration devices should be rinsed with the appropriate buffer. Additional details for this method can be found in Supplementary Protocol

5-2.

Protocol 2a, Soluble Proteins

Based upon the protein masses in the NRTDP intact protein standard, a MWCO of 3 kDa was used according to manufacturer’s instructions. The protein preparation is subjected to three

(1:20 dilution) buffer exchanges into 10 mM ammonium acetate (pH 6.5) using a MWCO- ultrafiltration device, followed by an additional three exchanges into 2.5 mM ammonium acetate

(pH 6.5) (Figure 5-3, Figures A4-3-4, Supplementary Protocol 5-2a). Denaturing and non- denaturing samples can then be diluted and introduced to the MS as described below in Protocol

4a.

Protocol 2b, Native membrane proteins

Membrane proteins are estimated to account for 23% of the total human proteome and represent approximately 60% of targets for currently approved drugs.300, 301 The mass analysis of native, intact membrane proteins can further provide key information regarding stoichiometry, ligand binding, and lipid association. A typical analysis of a membrane protein complex requires either size-exclusion chromatography (SEC) or MWCO-ultrafiltration to remove alkali salt adducts while maintaining the detergent used to solubilize the protein (Supplementary Protocol

5-2b).302 This differs fundamentally from the MWCO-ultrafiltration employed during filter-aided

127 sample preparation (FASP) to improve the bottom-up proteomics analysis of membrane proteins, which removes detergents.272, 303 For users interested in native membrane proteins we recommend the protocols of Robinson and coworkers (10.1038/nprot.2013.024).302 Their protocols are based upon comprehensive optimization and include a complete list of non-ionic detergents compatible with mass spectrometry and detailed sample preparation considerations. Successful application of

Robinson and coworker’s protocols is demonstrated for the native tetramer of Aquaporin Z (AqpZ) from E. coli (Figure A4-5a).

5.3.6 Precipitation

Protocol 3: Sample Preparation Using Protein Precipitation

Common precipitation protocols use organic solvents to agglomerate proteins while leaving small molecules, including salts and detergents, solubilized. Whereas MWCO- ultrafiltration using Protocol 2a does not rescue protein signal from a preparation containing harsh surfactants (e.g., SDS and Triton), precipitation of proteins following Protocol 3 does (Figure 5-

3, Supplementary Protocol 5-3). A volume ratio of 1:1:4:3 of aqueous protein sample:chloroform:methanol:water is recommended to precipitate proteins.281 The supernatant is removed by aspiration and the precipitated pellet can be further washed with one more addition and removal of methanol. Pellets are resolubilized for 15 minutes at -20 °C using a small volume of 80% (v/v) formic acid (ca., 25% of the starting volume), and are then diluted to the starting sample volume with HPLC grade water or a solution of volatile salts (e.g., ammonium acetate).304

As an alternative method, acetone precipitation has the distinct advantage of leaving many proteins folded. This method, however, has been shown to modify proteins with +98 Da adducts;305 requires

128 longer incubation at -20 °C (at least 1 h); requires that all steps be performed at or below 0 °C to maximize resolubilization; and can be compromised by detergents.

Figure 5-3: Dilution (P.1), MWCO-ultrafiltration (P.2a), and precipitation (P.3) sample preparation protocols applied to common buffers. Protein standard mixture in PBS (a) and detergent-containing RIPA Buffer (b). In the buffer containing harsh detergents protein signal is only attained using precipitation. (c) NISTmAb in 12.5 mM L-histidine, 12.5 mM L-histidine HCl

(pH 6.0). Mass spectra were obtained using an ion cyclcotron resonance MS (FT-ICR) (Bruker

129

Daltronics SolariX 9.4T MS) using denaturing direct infusion (P.4a). See Figure A4-3 for additional results with “Gentle Elution” immunoaffinity elution buffer and a second antibody buffer.

5.3.7 Native vs. Denaturing MS

Protocol 4a: Denaturing Direct-Infusion MS

Denaturing direct-infusion ESI (P.4a) mass spectra can usually be obtained by introducing samples to the MS in a mixture of 49.95% HPLC grade acetonitrile, 49.95% HPLC grade water, and 0.1% formic acid (v/v). A 60:35:5 ratio of HPLC grade methanol:water:acetic acid may be used as an alternative and, in some cases, can improve S/N. As described above, the use of these organic solvents and acids results in efficient ionization from a droplet’s surface, often allowing

MS analysis to be performed using instrumentation parameters typically employed for peptides. A more detailed description of instrument parameters for the Bruker SolariX FT-ICR MS used during denaturing direct infusion studies is found in Supplementary Protocol 5-4a.

Protocol 4b: Native Direct-Infusion MS

Although these protocols may not necessarily produce folded “native” ions that match exactly to their in-solution structures, they can be used to achieve accurate mass measurements of native structures and complexes.306 Consequently, Native direct-infusion MS can provide unique structural information, including the characterization of labile PTMs, metal-binding sites, non- covalent interactions with small molecules, and protein tertiary and quaternary structure.

Detergent-free samples can be infused directly in aqueous 2.5 mM ammonium acetate, the same 130 solution used in the final stage of Protocol 2a (concentrations of ammonium acetate up to 500 mM can even be employed). Figure 5-4 compares mass spectra of carbonic anhydrase in denatured and native states, with the intensity of the base peak in the native sample being ~2-fold higher than that of the denatured sample. This comparison was repeated by four additional labs on six different instruments to illustrate the possible range of relative intensities (Figure A4-6, Supplementary

Protocol 5-4b). Membrane protein complexes with MS compatible detergents can be infused directly from the final solution described in Protocol 2b.302 In order to observe native membrane proteins, removal of the detergent ions from the protein-micelle complex by increasing collisional activation is required. This may be achieved through an increase in collision voltage applied to the source or the collision cell (typically 50-200 V), but could require additional critical parameters that are described in detail by Robinson and coworkers, and in part in Supplementary Protocol

5-4b.302, 307, 308

131

Figure 5-4: Denatured vs. Native ESI-MS of carbonic anhydrase. Intensity is scaled to demonstrate the difference between denaturing MS (left) and native MS (right). These spectra were collected on the same instrument using the same concentration (10 µM). The native MS will have lower and fewer charge states and thus will have higher intensity and appear at a higher m/z. The inset includes the most abundant charge state and the S/N.

5.3.8 LC-MS of Intact Proteins

Protocol 5: Intact Protein Analysis Using LC-MS

Ionization suppression by excipients and by other proteins generally makes the analysis of multiple proteins and proteoforms by direct infusion intractable. For example, many “high purity” proteins (as judged by SDS-PAGE) contain numerous proteoforms that cannot be reliably detected and quantitatively assessed without upfront separation.309, 310 Liquid phase separation approaches, including liquid chromatography (LC) (e.g., reversed-phase (RP), size-exclusion, ion exchange, chromatofocusing) and capillary electrophoresis techniques (e.g., capillary zone electrophoresis, capillary isoelectric focusing) can remove excipients and provide the resolving power for deep characterization of proteins. As directed in the decision tree (Figure 5-2), separation of particularly complex samples (>100 proteins) requires an additional dimension of separation prior to LC-MS.

Figure A4-2 shows the use of GELFrEE separation prior to LC, which fractionates samples based on protein MWs, and has resulted in the largest number of characterized proteoforms.311

132

Protocol 5a. LC-MS of soluble proteins.

RP-LC is recommended for all samples containing >5 unique proteins, but is also a viable option for samples with fewer proteins, provided they do not contain high salt concentrations (>1

M) or harsh detergents. The recommended reversed-phase LC protocol is described in its entirety in Supplementary Protocol 5-5a and by following this link

(http://nrtdp.northwestern.edu/protocols/).

133

Figure 5-5: LC-MS of protein standard mixture prepared following the given SOP and separated on a Dionex UPLC with a Thermo Orbitrap Elite system using PLRP-S stationary phase. The final concentrations of each protein loaded onto the column were; 0.14 pmol ubiquitin, 0.49 pmol

134 trypsinogen, 1.09 pmol myoglobin, and 0.64 pmol carbonic anhydrase (top). Summary of S/N values calculated for each protein on all instrumentation platforms using the given SOP (bottom) including Dionex Ultimate 3000-Thermo Orbitrap Elite, Waters Acquity-Xevo G2-S QTOF,

Waters nanoAcquity-Bruker impact II QTOF, Waters nanoAcquity-Bruker SolariX FT-ICR,

Dionex Ultimate 3000-Thermo Fusion Lumos, and Dionex Ultimate 3000-Thermo QE-HF. As described in detail below, S/N calculations differ per manufacturer and do not reflect absolute performance.

Figure 5-5 demonstrates that sufficient intact MS signal was attained, and four unique chromatographic peaks were observed, using Protocol 5a with a PLRP-S stationary phase (1000

Å pore size, 5 µm particle size) on a Dionex UPLC coupled to a Thermo Orbitrap Elite.

Benchmarks for this SOP, as well as for additional data acquired using Monolithic and C4 stationary phases, are provided for six widely used platforms (Waters Xevo G2-S QTOF: Figure

A4-7; Bruker Impact II QTOF and Bruker SolariX FT-ICR: Figure A4-8; Thermo Orbitrap Elite,

Thermo Orbitrap Fusion Lumos, and Thermo Orbitrap QE-HF: Figures A4-9 & A4-10.

Experimental and data analysis methods are described in detail in the Supplementary Protocol

5-5a. To allow users to compare their performance with that of experienced operators using instruments that are operating within specifications, we report S/N for the platforms used here

(Figure 5-5). However, instrument vendors employ proprietary, non-standardized techniques to preprocess data, display data, and determine S/N, and as a result, our data cannot be used for a cross-platform comparison. As an example of a viable alternative method, notably that is better suited for proteoforms with similar mass and RP-LC retention (e.g., deamidation), we provide a separation of the same protein mix using CZE (Figure A4-11).

135

Protocol 5b. Intact Membrane Protein LC-MS.

Denaturing LC-MS of intact membrane proteins is not straightforward due to their inherent hydrophobicity.312, 313 Whitelegge et al. provided the earliest example of denaturing LC-MS of membrane proteins using high concentrations of mobile phase additives and, for the first time, demonstrated that ESI of membrane proteins could achieve the 0.01% mass accuracy benchmark established for ESI of soluble proteins.312 For thorough reviews of the current state of membrane

Figure 5-6: LC-MS of bacteriorhodopsin-containing purple membrane of Halobacterium prepared following Protocol 5b and analyzed on an Agilent HPLC system coupled to a Thermo linear ion trap (LTQ) mass spectrometer. Proteins were separated using an Agilent PLRP-S 300Å, 136

2.1 x 150 mm, 3 µm. Figure A4-12 demonstrates this analysis on four additional instrumentation platforms.

protein analysis via LC-MS314, 315 and the corresponding protocols, readers are directed to the following publications.314, 316, 317 Current denaturing LC-MS methods for membrane proteins utilize either size-exclusion318, 319 or reversed-phase separation. Here, due to observed ease of implementation across a variety of MS platforms, we suggest analysis via reversed-phase LC-MS using a polystyrene-divinyl benzene co-polymer stationary phase (PLRP-S, 300 Å, Agilent). Use of long chain bonded stationary phases such as C8 and C18 is not recommended, as membrane proteins will likely be retained on the column. We solubilize enriched bacteriorhodopsin from

Halobacterium salinarum (Sigma B0184) in 88% formic acid to separate the protein from lipid contaminants. To avoid risk of formic acid adduction (+28 Da), samples are immediately injected onto the column and solvent exchanged to much lower acid concentrations (0.1%). In the case of membrane protein preparations containing high enough concentrations of lipid contaminants to confound analysis or damage the column, we recommend precipitation following Protocol 3 prior to analysis. Proteins are eluted using an increasing gradient of 49.95% acetonitrile, 49.95% isopropanol, 0.1% formic acid. Figure 5-6 shows the analysis of denatured bacteriorhodopsin of

Halobacterium following this protocol. While elution efficiency for some integral proteins may fall well below 100%, PLRP-S columns can be regenerated with 90% formic acid injections. This protocol was performed by five labs on five different instrument platforms (Figure A4-12,

Supplementary Protocol 5-5b). An example of an alternative LC-MS method using a more common stationary phase (ZORBAX RRHD 300SB-C3) is provided for Aquaporin Z in Figure

A4-5b.

137

5.3.9 Special Methodological Considerations for Intact Antibody Mass Spectrometry

With the increasing development of biotherapeutics and biosimilars in the pharmaceutical industry, and an increasingly stringent route to regulatory approvals, there is a growing need for intact Ab MS. We demonstrate that every protocol presented here can be applied to the analysis of intact antibodies (Figure 5-3, Figure A4-13). However, as antibodies are relatively large and signal spreading increases in proportion to protein size, we recommend against the use of Protocol

1 (dilution) for any regulatory filing.

5.4 Discussion

The IPMS Decision Tree (Figure 5-2) guides practitioners of all levels towards broadly applicable methods to obtain high-quality intact mass spectra from any protein sample. The protocols described here have been scrutinized and optimized by over ten expert intact protein MS labs, and successfully applied by laboratories without experience in intact protein MS. It is our hope that these protocols will enable any research group to adopt intact protein mass analysis. The accurate mass measurement of an intact protein is the sine qua non of top-down mass spectrometry, which can characterize how proteoforms interact and identify prevalent PTMs that are lost in other analyses. High-throughput top-down analysis of whole proteomes has proven successful in the unambiguous identification of hundreds of proteins and proteoforms from a single biological sample,320 and revealed prevalent yet previously uncharacterized, biologically relevant modifications.321 Quantitative top-down proteomics has been utilized to identify disease-relevant differences in protein levels, an encouraging step forward in the field of proteomics-based personalized medicine.322 Additionally, utilizing native proteomics following the top-down

138 workflow we can observe previously unknown protein-protein interactions, protein-ligand binding, protein-cofactor association, and protein-complex stoichiometry, allowing us to assess their relationship to important biological pathways.323 We believe starting with intact mass analysis, utilizing these intact protein MS protocols coupled to top-down MS analysis, and by identifying proteoforms rather than proteins, new insights into the human proteome can be achieved. It is also our hope that these protocols serve as a starting point for users to push, even further, the current limits of high molecular weight mass spectrometry. The CTDP hopes that this work will enable users at any level to contribute to the ongoing development of top-down MS and proteomics.

5.5 Supplementary Protocols

5.5.1 Supplementary Protocol 5-1: Dilution

In order to sufficiently concentrate protein to allow for dilution out of the typical sample buffers, the protein standard mixture was first precipitated following Protocol 3. Pelleted proteins were then resuspended in one tenth of the original volume of sample, to concentrate them to a 10X mixture. The protein mixture in 1X PBS (Figure 3a) was diluted 1:50 in 49.95% HPLC grade acetonitrile, 49.95% HPLC grade water, and 0.1% formic acid (v/v). The 10X concentrated sample was diluted 1:500 in 49.95% HPLC grade acetonitrile, 49.95% HPLC grade water, and 0.1% formic acid (v/v). This same procedure was repeated with the RIPA buffer sample (Figure 3b).

5.5.2 Supplementary Protocol 5-2a: MWCO-Ultrafiltration Additional Details

Typical buffer samples were transferred to an Amicon® Ultra 0.5 mL Centrifugal Filter with a regenerated cellulose membrane and a 3 kDa MWCO (Millipore, Billerica, Massachusetts,

139

USA). Samples were diluted with 450 µL of 10 mM ammonium acetate, pH 6.5, and centrifuged at 14000 RPM for 15 min three times (concentrating to 25-50 µL during each centrifugation).

Using the same filter devices, samples were exchanged into 2.5 mM ammonium acetate, pH 6.5, by centrifuging at 14000 RPM for 15 minutes three times. Following the last spin, samples were diluted 50-fold their initial volume in 49.95% HPLC grade acetonitrile, 49.95% HPLC grade water, and 0.1% formic acid (v/v). Sample preparation by size exclusion “spin cartridges” can often be a suitable alternative for this protocol.324

5.5.3 Supplementary Protocol 5-2b: Native Membrane Protein Preparation Additional Details

AquaporinZ (AqpZ) from E. coli was provided in 150 mM NaCl, 20 mM Tris-HCl (pH =

8.0), 5% glycerol and 40 mM octyl glucoside. The initial stages of detergent exchange were performed by diluting 150 µg of protein (100 µL) into an equal volume of running buffer containing 2x CMC tetraethylene glycol monooctyl ether (C8E4) at RT. This solution was then further buffer and detergent exchanged using FPLC-SEC (GE Healthcare, Sephadex 200, 5/150

GL column). If preparation of membrane proteins for native analysis is performed using MWCO- ultrafiltration instead, we recommend against the use of the same MWCO-ultrafiltration devices used in denaturing experiments (MWCO 3 kDa). Most detergent micelles will not pass through these membranes and are, therefore, concentrated rather than removed. Even after dilution of the sample to below the CMC of a given detergent, the dissociation of non-ionic micelles to monomers occurs slowly enough that equilibrium may not be reached during ultrafiltration. A MWCO membrane with a pore size that exceeds the mass of the empty micelles is recommended.325, 326 A

50 kDa cutoff membrane, for example, is sufficient to retain protein-micelle complexes of the MS- compatible detergent C8E4, but able to remove excess free detergent. We also recommend against

140 using either SEC or MWCO-ultrafiltration for the outright removal of detergents. Removing detergents often results in the loss of native structure, loss of protein solubility, and protein aggregates that clog filters. If outright detergent removal is desired, LC-MS, preceded by precipitation using Protocol 3 is recommended. If proteins can’t be re-solubilized in formic acid following precipitation, LC-MS can be preceded by exchange into MS compatible detergents using the protocol for native MS analysis.

5.5.4 Supplementary Protocol 5-3: Protein Precipitation Additional Details

All samples were volume normalized prior to precipitation to ensure protein concentration consistency. Each sample was subjected to chloroform/methanol precipitation following a previously published protocol.327 100 µL aliquots of each sample were combined with 400 µL of methanol, 100 µL of chloroform, and 300 µL of HPLC grade H2O followed by gentle vortexing to ensure complete mixing. Samples were centrifuged for 15 minutes at 14000 RPM and 4 °C in an Eppendorf minispin centrifuge (Hauppauge, New York, USA). The top layer of each sample was aspirated carefully to not disturb the precipitated protein wafer located within the organic/aqueous interface. An additional 400 µL of methanol was added to each sample followed by gentle vortexing and 15 minutes of centrifugation. Solvent was completely removed by aspiration and the pellet was dried briefly. To ensure complete removal of interfering substances, an additional 400 µL of methanol was added to each pellet, followed by vortexing, centrifugation, and aspiration. Following a resolubilization protocol published by Doucette et al.328, the resulting pellets were chilled at -20 °C for 15 minutes and incubated in 25 µL of 80% formic acid at -20 °C for two minutes. The samples were then mixed and incubated at -20 °C for an additional 15 minutes. 75 µL of cold HPLC grade water was then added to each sample, followed by a final 1:50

141 dilution in 49.95% HPLC grade acetonitrile, 49.95% HPLC grade water, and 0.1% formic acid

(v/v).

5.5.5 Supplementary Protocol 5-4a: Denaturing Mass Spectrometry Additional Details

All experiments with signal suppression curves and Protocols 1-3 were performed on a 9.4

T SolariX FT-ICR MS (Bruker Daltonics, Billerica, MA) in positive mode with ESI. The m/z range was 154 – 4500 m/z, with a 0.1 second ion accumulation, and 41.5 second accumulation time (32 scans). The time of flight (TOF) was set at 1.7 ms with 25% Sweep Excitation Power and at 2 M size resolution. The flow rate was kept at 2 µL/min with the Dry Gas Temperature at 180 °C and the ESI Capillary voltage at 4.5 kV. The RF Amplitude in the funnel was set to 200 Vpp with 150

V for Funnel 1 and 40 V for Skimmer 1. The RF Frequency for the Collision Cell was 2 MHz with the RF Amplitude set to 1300 Vpp. The RF Frequency for the Transfer Optics was 4 MHz with the

RF Amplitude set to 250 Vpp. Although isotopic resolution is not observed here for intact antibodies, it can be achieved with further instrument optimization of the Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometer. Previous studies have combined the use of

Continuous Accumulation of Selected Ions (CASI), broadband phase correction, and high field magnets to isotopically resolve intact antibodies.329-331 Note that nomenclature (e.g., CAD vs. CID vs. HCD) and optimal parameters will vary across different instrumentation platforms.

5.5.6 Supplementary Protocol 5-4b: Native Mass Spectrometry Additional Details

Native MS has been shown to reduce the signal spreading that is often observed in denaturing MS by producing fewer charge states.332-334 However, in another study, a 20-fold decrease in S/N was observed in mass spectra obtained from native samples compared to those

142 obtained from denatured samples.335 Desolvation can be enhanced by increasing collisional activation energy in the source atmosphere/vacuum interface region or collision cell, increasing gas pressure or capillary temperature, and employing nano-ESI. Moreover, small concentrations of supercharging additives (e.g., meta-nitrobenzyl alcohol) that preserve the native conditions of the sample can be used to increase the average charge state of an ionized protein and lower the m/z of large proteins into a range compatible with the instrumentation, thereby increasing ion signal in most instruments.336 Increasing desolvation energy increases the likelihood of bond fragmentation

– in particular the loss of labile modifications and non-covalent adducts.

The major application of nMS, for which there are guides337 and online tutorials

(http://msr.dom.wustl.edu/tutorial-native-mass-spectrometry/) available, has been in the targeted study of individual (or simple mixtures of) proteins and protein complexes.338 Carbonic anhydrase samples analyzed by native mass spectrometry (nMS) were buffer exchanged into 150 mM ammonium acetate, pH 7.0, following the MWCO-ultrafiltration procedure described in Protocol

2 & Supplemental Protocol 5-2, with Amicon® Ultra 0.5 mL 10-100 kDa MWCO spin filters and a nominal final protein concentration of 10-20 µM. All denaturing vs. native comparative analyses were run using this same sample prep. Ammonium acetate at a concentration of 50-200 mM is the buffer predominantly used in nMS due to its high volatility, low adduction behavior, and non-denaturing properties.339 By using non-denaturing buffers, non-covalent interactions are preserved and proteins can be maintained in a folded configuration.339

Native Analysis Method 1: Bruker 9.4 T SolariX FT-ICR MS

Data were acquired using the Bruker nano-ESI source in positive mode. The capillary was kept at 1200 V with a drying gas temperature of 150 °C. The following source parameters were used: Skimmer 1 at 165 V, Funnel RF amplitude at 250 Vpp, Funnel 1 at 200 V. Time of flight

143 was set to 2.2 ms and sweep excitation was set to 27%. 50 mass spectra were averaged at 4 MW

(Figure 5-4).

Native Analysis Method 2: Thermo Fisher Scientific Q Exactive HF MS

A customized Thermo Fisher Scientific Q Exactive HF Mass Spectrometer with Extended

Mass Range was utilized for nMS analysis of Carbonic Anhydrase in Figure A4-6a. 340 The nMS platform employs direct infusion of sample into a native electrospray ionization (nESI) source held at +2 kV. The instrument was set in EMR MS mode for intact protein mass analysis. The samples were analyzed with the following acquisition parameters: capillary temperature, 330 °C; FT resolution, 15,000 (at 200 m/z); S-lens RF level, 50; scan range, 350−10000 m/z; desolvation, in- source CID 1-195 V; polarity, positive; microscans, 10; AGC target, 3 × 106; maximum injection time, 50 ms; averaging, 0; trapping gas pressure setting, 1-4. Mass spectrometric data were analyzed using Thermo Fisher Xcalibur 4.0.

Native Analysis Method 3: Bruker 15 T SolariX FT-ICR MS

Data were acquired using the Bruker Nano-ESI source in positive mode. The capillary was kept at 1100 V with a drying gas temperature of 150 °C. The following source parameters were used: Skimmer 1 at 120 V, Funnel RF amplitude at 250 Vpp, Funnel 1 at 150 V. Time of flight was set to 2.0 ms and sweep excitation was set to 27%. 50 mass spectra were averaged at 1 MW

(Figure A4-6b).

Native Analysis Method 4: Bruker 12 T SolariX FT-ICR MS

10 µM of carbonic anhydrase in 150 mM ammonium acetate pH=7 was directly infused via a TriVersa NanoMate (Advion Bioscience, Ithaca, NY) nano-ESI source with a spray voltage of 1.8 kV. 50 scans were acquired at 512k words at 202.76-3000 m/z with 0.05 s accumulation.

The following settings were used: Funnel 1 150 V; Skimmer 1 120 V, Funnel RF amplitude 300

144

Vpp; Octopole frequency 2 MHz, Octopole RF Amplitude 600 Vpp; Collision cell frequency 1.4

MHz; Collision RF amplitude 2000 Vpp; Transfer Optics time of light 1.5 ms; Transfer Optics frequency 2 MHz; Transfer Optics RF Amplitude 450 Vpp; Sweep excitation power 40%; Gas control 40% (Figure A4-6c).

Native Analysis Method 5: Bruker maXis II ETD Q-TOF

10 µM of carbonic anhydrase in 150 mM ammonium acetate pH=7 was directly infused at

3 µL/min via an ESI source. The following settings were used: Mass range 200-6000 m/z at 1 Hz acquisition rate; Capillary voltage 3800 V; Nebulizer 1.5 Bar; Dry gas 5 l/min; Dry temp 150 ˚C; isCID energy 150 eV; Multipole RF 600 Vpp; Collision cell transfer time 110 µs and pre pulse storage 45 µs. 120 scans were averaged (Figure A4-6d).

Native Analysis Method 6: Bruker 15 T SolariX FT-ICR MS

Data were acquired using the Bruker nano-ESI source in positive mode. The capillary was kept at 600 V. The following source parameters were used: Skimmer 1 at 25 V, Funnel 1 at 100

V. Time of flight was set to 1.5 ms. 50 mass spectra were averaged (Figure A4-6e).

Native Analysis Method 7: Waters SynaptG2Si

120 scans were averages during acquisition. Capillary voltage was kept at 1.6 kV, source temperature at 80 °C, Sample cone at 150 V, Desolvation Temperature at 150 °C, Nanoflow Gas

Pressure at 0.3 Bar, and Nebulizer Gas Flow at 6.5 Bar (Figure A4-6f).

Native Analysis Method 8: AqpZ

The nMS of membrane protein AqpZ (E. coli) was acquired on a Waters Synapt G1 QTOF.

AqpZ solution was infused into the Q-TOF MS by nanoESI (Figure A4-5a). Typical instrument voltages and pressures were: capillary voltage 0.5-1 kV; sample cone 40 V, extraction cone 1V; trap collision energy 4.0 to 110V (range to determine optimal protein ejection voltage) and source

145 backing pressure of 6.0 mbar. The instrument was externally calibrated using a 50 µg/µL CsI solution, over the m/z range 500-20,000. In the case of multimeric membrane proteins, care must be taken to balance the removal of detergent ions from the protein-micelle complex and the dissociation of the multimeric complex through increased collisional activation.341 Too little voltage can render protein charge states unobservable, while too high a voltage can produce lower multimeric species resulting from subunit ejection.342

5.5.7 Supplementary Protocol 5a: LC-MS Benchmarks Additional Details

All participating laboratories used the same protein standard mixture and the same SOP which was provided by the National Resource for Translational and Developmental Proteomics

(NRTDP) at Northwestern University (http://nrtdp.northwestern.edu/protocols/). All labs used a

PLRP-S column; however, as the PLRP-S resin is becoming less commercially available in bulk, some labs also performed the same SOP with their column of choice.

Method 1: Waters UPLC-QTOF

The protein standard mixture was analyzed on an H-Class Acquity UPLC system coupled to a Xevo G2-S Q-TOF mass spectrometer (Waters Corp, Milford, MA). The columns used were an Agilent PLRP-S, 1000 Å pore size, 5 µm particle size, 50 mm bed length, 4.6 mm ID column

(Agilent Technologies, Santa Clara, CA) and an Acquity UPLC Protein BEH C4 (300 Å pore size,

1.7 µm particle size, 100 mm bed length, 2.1 mm ID x 100 mm) column (Waters Corp, Milford,

MA). The flow rate for the C4 column was 0.2 mL/min while the flow rate for the PLRP-S column was 0.8 mL/min. Solvent A: 95% HPLC Grade H2O with 0.2% formic acid (Fisher Scientific,

Hampton, NH), 5% HPLC Grade Acetonitrile (Fisher Scientific, Hampton, NH), and for Solvent

146

B: 5% HPLC Grade H2O with 0.2% formic acid, 95% HPLC Grade Acetonitrile. The following gradient was used for both columns:

Time (min) % B 0.0 5.0 10.0 5.0 12.0 15.0 37.0 55.0 40.0 95.0 43.0 95.0 45.0 5.0 60.0 5.0

The capillary voltage was set to 3 kV and the sample cone voltage was set at 40 V. The source temperature was kept at 150 °C and desolvation temperature 350 °C with a gas flow of 800 L/h.

The method ran in the Sensitivity Analyzer mode with a 500 – 4000 m/z mass range. For the PLRP-

S column the scan time was 0.5 s, and for the C4 column the scan time was 1 s. Figure A4-7 shows the results from one of three replicates of the concentrated sample, 1:10 diluted sample, and the

1:100 diluted sample separated with the PLRP-S and C4 stationary phases. For each method, 2.5

µL was injected.

A total ion chromatogram is given in the top panel of each figure, with each chromatographic peak numbered in order of elution. The corresponding mass spectrum for each of the four proteins is presented in four panels below each chromatogram. The undiluted sample

(Figure A4-7a) contained: 14 pmol ubiquitin, 49 pmol trypsinogen, 109 pmol myoglobin, and 64 pmol carbonic anhydrase. The 1:10 diluted sample (Figure A4-7b) contained: 1.4 pmol ubiquitin,

4.9 pmol trypsinogen, 10.9 pmol myoglobin, and 6.4 pmol carbonic anhydrase. The 1:100 diluted sample (Figure A4-7c) contained: 0.14 pmol ubiquitin, 0.49 pmol trypsinogen, 1.09 pmol myoglobin, and 0.64 pmol carbonic anhydrase. Unlike the methods that used capillary columns

147

(Methods 2-6), the 1:100 dilution showed no protein signal with the PLRP-S (Figure A4-7c top panel). The 1:10 diluted sample produced comparable results within the S/N range of the C4 column (Figure A4-7b), but contained broader peaks than expected. The undiluted sample (Figure

A4-7a) showed ideal chromatographic peaks for both stationary phases.

Method 2: Waters UPLC – Bruker QTOF

The protein standard mixture was prepared as described above except for bovine carbonic anhydrase (PS-121-1) which was purchased from Protea Biosciences (Morgantown, WV). 2 mg/mL stocks of each protein were made in HPLC Grade H2O (Fisher Scientific, Hampton, NH).

This standard mixture was divided into 2 µL aliquots and stored in – 80 °C. One aliquot was diluted to 200 µL and run in triplicate on a nanoAcquity UPLC system (Waters Corp, Milford, MA) coupled to an impact II QTOF (Bruker Daltonics) with a self-packed Agilent PLRP-S (1000 Å pore size, 5 µm particle size), 20 cm bed length, 500 µm ID capillary column. For the LC, Solvent

A: 95% HPLC Grade H2O, 5% HPLC Grade Acetonitrile, 0.2% MS‐grade formic acid (Fisher

Scientific, Hampton, NH) and Solvent B: 5% HPLC Grade H2O, 95% HPLC Grade Acetonitrile

(Fisher Scientific, Hampton, NH), 0.2% MS‐grade formic acid were used with the following gradient:

Time (min) % B 0.0 5.0 5.0 5.0 42.0 60.0 44.0 95.0 46.0 95.0 47.0 5.0 60.0 5.0

148

For each replicate, 5 µL was injected at a flow rate of 12 µL/min with the mass range at 500 –

2500 m/z for a final injected amount of 0.14 pmol ubiquitin, 0.49 pmol trypsinogen, 1.09 pmol myoglobin, and 0.64 pmol carbonic anhydrase. The Collision Cell RF was set to 2500 Vpp, the

Capillary voltage at 4500 V, the Funnel RF at 400 Vpp, the Collision cell energy at 8 V, the

Quadrupole energy at 4 V, and the in-source CID at 40 V. The source gas temperature was set to

220 °C with a 4 L/min flow rate. The results are displayed in Figure A4-8a with the S/N calculated for each protein.

Method 3: Waters UPLC – Bruker FTICR

All sample preparation, column, and LC parameters were identical to those described above in Method 2. In this method, the nanoAdvance HPLC (Bruker Daltonics, Billerica, MA) was coupled to a solariX XR 12T FTICR (Bruker Daltonics, Billerica, MA) with a capillary column of 15 cm bed length, 250 µm ID, running at 4 µL/min. The mass range was 500 – 2500 m/z with a 0.08 s ion accumulation with 1.4 s per scan. The TOF was set at 1.00 ms with 30%

Sweep Excitation Power and at 1M transient length with Full Sine processing. The dry gas temperature was set to 220 °C with a 4 L/min flow rate and the ESI Capillary voltage at 4.5 kV.

The RF Amplitude in the funnel was set to 150 Vpp and 40 V for in-source CID. The RF Frequency for the Transfer Optics was 4 MHz with 350 Vpp RF Amplitude and the Collision Cell was 2 MHz with the RF Amplitude set to 2000 Vpp. The results are displayed in Figure A4-8b with the S/N calculated for each protein.

Method 4: Dionex UPLC – Thermo Orbitrap Elite

The intact protein standard mixture was prepared as described above and divided into 2.5

µL aliquots for storage at -80 °C. Prior to analysis, aliquots of standard were diluted 1:240 in 600

µL Solvent A (95% Optima grade water, 5% Optima grade acetonitrile, and 0.2% MS-grade formic

149 acid; all Fisher Scientific, Hampton, NH), for a final injected amount of 0.14 pmol ubiquitin, 0.49 pmol trypsinogen, 1.09 pmol myoglobin, and 0.64 pmol carbonic anhydrase. Ultra-high performance liquid chromatography (UPLC) was performed on-line with the MS using a Dionex

Ultimate 3000 (Thermo Fisher Scientific, San Jose, CA). Trap (150 µm ID x 2 cm) and analytical

(75 µm ID x 15 cm) columns were packed in-house with PLRP-S resin (5 µm, 1000 Å) (Agilent

Technologies, Santa Clara, CA) and enclosed within a column oven compartment maintained at

45 °C. Monolithic PepSwift trap (150 µm ID x 0.5 cm) and ProSwift RP-4H analytical (100 µm

ID x 50 cm) columns were purchased from Thermo Fisher Scientific, conditioned according to the manufacturer’s protocol, and enclosed within a column oven compartment maintained at 35 °C. 6

µL injections were concentrated and desalted on the trap column in Solvent A for 3 minutes at 10

µL/min (Monolithic) or 10 minutes at 3 µL/min (PLRP-S). Intact proteins were then further resolved prior to MS analysis using the gradients of Solvent A and Solvent B (5% Optima grade water, 95% Optima grade acetonitrile, and 0.2% MS-grade formic acid) shown in the tables below:

Monolithic: 1 µL/min flow rate

Time (min) % B 0.0 5.0 3.0 5.0 33.0 50.0 35.0 95.0 38.0 95.0 39.0 5.0 53.0 5.0

PLRP-S: 0.3 µL/min flow rate

Time (min) % B 0.0 5.0 10 5.0 12.0 15.0 37.0 55.0 40.0 95.0

150

43.0 95.0 45.0 5.0 60.0 5.0

Chromatographically resolved proteins were introduced into the MS using a custom nano-

ESI source containing a high-voltage union (Thermo Fisher Scientific) coupled to the end of the analytical column, through which a 1.9-2.2 kV potential was applied, and connected to a 15 µm

ID nanospray emitter (New Objective, Woburn, MA) self-packed with 2 mm of PLRP-S resin.

Intact protein (MS1) spectra were acquired over an m/z range of 500-2000 at an FT resolution of

120,000 (at 400 m/z) on an LTQ-Velos Orbitrap Elite mass spectrometer (Thermo Fisher

Scientific) operating in “protein mode” under control of Xcalibur (Thermo Fisher Scientific). MS1 method details included 4 microscans, an AGC target of 1 x 106 ions, a maximum injection time of 1s, and an additional 15 V applied within the source region to facilitate desolvation. Source region parameters also included a transfer capillary temperature of 320 °C and an S-lens RF amplitude of 50%. Mass spectrometric data were analyzed using Xcalibur 4.0 (Thermo Fisher

Scientific). The resulting chromatogram, intact protein spectra, and respective S/N calculation from a representative replicate injection are shown in Figure A4-9a.

Method 5: Dionex UPLC – Thermo Orbitrap Fusion Lumos

All sample preparation, PLRP-S column, and LC parameters were identical to those described above in Method 4. Intact protein (MS1) spectra were acquired over an m/z range of

400-2000 at an FT resolution of 120,000 (at 200 m/z) on an Orbitrap Fusion Lumos mass spectrometer (Thermo Fisher Scientific) operating in “intact protein mode” under control of

Xcalibur. MS1 method details included 4 microscans, an AGC target of 2 x 105 ions, a maximum ion injection time of 100 ms, and an additional 15V applied within the source region to facilitate desolvation. Source region parameters also included a transfer capillary temperature of 320 °C and

151 an ion funnel RF amplitude of 30%. Mass spectrometric data were analyzed using Xcalibur 4.0

(Thermo Fisher Scientific). The resulting chromatogram, intact protein spectra, and respective S/N calculations from a representative injection are shown in Figure A4-9b.

Method 6: Dionex UPLC – Thermo QE-HF

All sample preparation, PLRP-S column, and LC parameters were identical to those described above in Method 4. Intact protein (MS1) spectra were acquired over an m/z range of

500-2000 at an FT resolution of 120,000 (at 200 m/z) on a QE-HF mass spectrometer (Thermo

Fisher Scientific) operating in “intact protein mode” under control of Xcalibur. MS1 method details included 4 microscans, an AGC target of 1 x 106 ions, a maximum ion injection time of 50 ms, and an additional 15V applied within the source region to facilitate desolvation. Source region parameters also included a transfer capillary temperature of 320 °C and an S-lens RF amplitude of

50%. Mass spectrometric data were analyzed using Xcalibur 4.0 (Thermo Fisher Scientific). The resulting chromatogram, intact protein spectra, and respective S/N calculations from a representative injection are shown in Figure A4-9c.

Method 7: Dionex UltimMate 300 RSLCnano System – Thermo Fusion Lumos

The Protein standard mixture was prepared as described in the given SOP by diluting 14 pmol ubiquitin, 49 pmol trypsinogen, 109 pmol myoglobin, and 64 pmol carbonic anhydrase in

100 uL of 95% ACN, 5% Water, 0.2% formic acid and injecting 1 µL for analysis. Samples were separated on an in-house pulled 100 µm ID capillary with integrated spray tip packed with PLRP-

S resin (1000 Å pore size, 5 µm particle size) to 15 cm. Sample was directly loaded onto this column by the autosampler at 0.3 ul/min for 30 min and the same gradient as described in Method

4 was applied. The EASY-spray source (Thermo Fisher Scientific) was used in conjunction with a Jailbreak Column Heater (Phoenix S&T) to heat to 50 °C and position the column to the front of

152 the mass spectrometer. Intact protein (MS1) spectra were acquired over an m/z range of 500-2000 at an FT resolution of 75,000 (at 400 m/z) on a Thermo Orbitrap Fusion Lumos (Thermo Fisher

Scientific) operating in “protein mode” under control of Xcalibur (Thermo Fisher Scientific). MS1 method details included 4 microscans, an AGC target of 1 x 106 ions, a maximum injection time of 50ms, and an additional 15V applied within the source region to facilitate desolvation. Source region parameters also included a transfer capillary temperature of 320 °C and an S-lens RF amplitude of 60%. Mass spectrometric data were analyzed using Xcalibur 4.0 (Thermo Fisher

Scientific). The resulting chromatogram, intact protein spectra, and respective S/N calculation from a representative replicate injection are shown in Figure A4-10.

5.5.8 Protocol 5b: Denaturing Reversed Phase LC-MS of bacteriorhodopsin from Halobacterium salinarum

Method 1: Agilent 1200 HPLC – Thermo LTQ (Figure 5-6)

Bacteriorhodopsin from Halobacterium salinarum was purchased through Sigma-Aldrich

(B0184). 1 mg of protein was suspended in 100 µL of 1% CHAPS in HPLC-grade water to a final concentration of 10 mg/mL. Membrane protein solution was aliquoted into 25 µg aliquots and stored at 4 °C until use (<5 days) or -80 °C for extended storage. Prior to analysis, 1 aliquot of protein (25 µg) was mixed with 10 µL of 88% formic acid (ACS Grade). Upon addition of acid, the protein solution changed from purple to orange, and was immediately injected onto the column

(approximately 1-3 minutes depending on plumbing). The idea was to minimize exposure to formic acid (less than 2 mins) as it is known to modify proteins at this concentration. 3 µL (224 pmol) of protein is injected onto a PLRP-S (Agilent) reversed-phase column (300 Å pore size, 3

µm bead size). Solvent A was water + 0.1% formic acid and solvent B was 49.95% HPLC grade

153 acetonitrile, 49.95% HPLC grade isopropanol, and 0.1% formic acid. We used an increasing gradient of solvent B during separation (specified below) and a flow rate of 0.120 mL/min:

Time (min) % Solvent B 0 5 5 5 10 30 50 80 55 99 65 99 66 5 75 5

Method 2: Waters nanoAcquity – Bruker SolariX FT-ICR (Figure A4-12a)

25 µg of bacteriorhodopsin from Halobacterium salinarum was suspended in 2.5 µL of

HPLC grade water to a final concentration of 10 mg/mL. Water was used instead of 1% CHAPS due the FT-ICR’s sensitivity to CHAPS, which overpowered protein signal. 110 µL of 88% formic acid was added to the protein solution, turning it from a deep purple to an orange immediately. 3

µL (22.4 pmol) of sample was loaded onto a self-packed PLRP-S capillary column (7 cm length,

100 µm I.D., 300 Å pore size, 3 µm bead size). The same gradient given above was run at 2 µL/min.

Mass analysis was performed on a 9.4 T SolariX FT-ICR-MS (Bruker Daltonics, Billerica, MA) in the positive mode with ESI. The m/z range was 154 – 5000 m/z, with a 0.15 second ion accumulation, The time of flight (TOF) was set at 1.4 ms with 18% Sweep Excitation Power and at 1M size resolution. Dry Gas Temperature was kept at 180 °C and the ESI Capillary voltage at

4.5 kV. The RF Amplitude in the funnel was set to 200 Vpp with 150 V for Funnel 1 and 40 V for

Skimmer 1. The RF Frequency for the Collision Cell was 2 MHz with the RF Amplitude set to

1300 Vpp. The RF Frequency for the Transfer Optics was 4 MHz with the RF Amplitude set to 250

Vpp.

154

Method 3: H-Class Acquity UPLC - Xevo G2-S Q-TOF (Figure A4-12b)

Bacteriorhodopsin from Halobacterium salinarum was prepared following the same procedure as described in Method 1, above. 3 uL (224 pmol) of protein was loaded onto an PLRP-

S column (3 µm, 300 Å, 2.1x50 mm) (Agilent Technologies, Santa Clara, CA). Samples were run at 200 µL/min. Solvent A was 100% HPLC Grade H2O with 0.1% formic acid and Solvent B was

49.95% HPLC Grade Acetonitrile, 49.95% Isopropanol, 0.1% formic acid. The same linear gradient applied in B was used. The capillary voltage was set to 3 kV and the sample cone voltage was set at 40 V. The source temperature was kept at 150 °C and desolvation temperature at 350

°C with a gas flow of 800 L/h. The method ran in the Sensitivity Analyzer mode with a 500 – 4000 m/z mass range. Scan time was set to 1 s.

Method 4: Thermo Scientific Vanquish – Thermo Q Exactive (Figure A4-12c)

Bacteriorhodopsin from Halobacterium salinarum was prepared following the same procedure as described in Method 1 above. 3 uL (224 pmol) of protein was loaded onto an

PLRP-S column (3 µm, 300 Å, 2.1x50 mm) (Agilent Technologies, Santa Clara, CA). The following linear gradient was used at a flow rate of 400 µL/min.

Time (min) % B 0.0 5.0 5 5.0 80 100

The mass spectrometer was run in positive ion mode at 140,000 resolution, scanning a range m/z from 850-3000 m/z. Mass spectra were collected after 40 minutes to ensure CHAPS was directed towards waste. Capillary temperature was kept at 263 °C and S-Lens RF level was set at 50. In-source CID was set at 5.0 eV.

155

Method 5: Agilent 1290 – Thermo Exactive Plus (Figure A4-12d)

Bacteriorhodopsin from Halobacterium salinarum was prepared following the same procedure as described in Method 1 above. 3 uL 224 (pmol) of protein was loaded onto an

PLRP-S column (3 µm, 300 Å, 2.1x50 mm) (Agilent Technologies, Santa Clara, CA). The following linear gradient was run at a flow rate of 0.3 mL/min:

Time (min) % B 0.0 5.0 5 5.0 80 100

The mass spectrometer was run in positive ion mode at 140,000 resolution, scanning a range m/z from 900-3500 m/z. Capillary temperature was kept at 265 °C and S-Lens RF level was set at 60. In-source CID was set at 20.0 eV.

Denaturing LC-MS of Membrane Protein AqpZ

LC-MS of AqpZ (E. coli) was performed using an Acquity UPLC, XevoQ-ToF (Waters

Corporation, MS Technologies Center) and an Agilent ZORBAX RRHD 300SB-C3, 2.1 x 50 mm,

1.8 µm, 300 Å column. Using mobile phases H2O containing trifluoroacetic acid, 0.1% v/v, formic acid, 0.1% v/v, as mobile phase A and 90% n-propanol containing trifluoroacetic acid, 0.1% v/v, formic acid, 0.1% v/v, as mobile phase B. UPLC analysis was achieved using a 5 minute UPLC method (including re-equilibration) operating at a flow rate of 400 µL/min and a column temperature of 65 °C. AqpZ analyzed, as received from purification, under the above UPLC-MS conditions yielded spectra displayed in Figure A4-5b. Contrary to nMS, denaturing LC-MS requires no further sample preparation (FPLC detergent exchange and buffer exchange) prior to analysis. The instrument was externally calibrated using a 50 µg/µL CsI solution over m/z range

156 of 1000-4500. Real-time lock mass correction was performed on the leucine encephalin dimer (m/z

1111.5464).

5.6 Supplementary Notes

5.6.1 Supplementary Note 5-1: Protein Standard Mixture

Bovine ubiquitin (U6253), bovine trypsinogen (T1143), equine myoglobin (M5696), and bovine carbonic anhydrase (C2624) were all purchased from Sigma Aldrich (St. Louis, MO). Stock solutions at 2 mg/mL were made of each solution in HPLC Grade H2O (Fisher Scientific,

Hampton, NH) and the standard was prepared as follows:

Volume Molecular Weight Stock Concentration Protein (µL) (g/mol) (pmol/µL) Carbonic Anhydrase 40 29030 25.7 Myoglobin 40 16951 43.9 Trypsinogen 25 23981 19.6 Ubiquitin 2.5 8560 5.5 Total 107.5

This standard mixture was used for the signal suppression curves and for sample clean-up experiments with denaturing mass spectrometry methods. It should be noted that superoxide dismutase (SOD1) can sometimes appear in purchased stocks of bovine carbonic anhydrase.

Further details regarding the preparation of this protein standard mixture can be found at http://nrtdp.northwestern.edu/protocols/.

5.6.2 Supplementary Note 5-2: Details on Signal Suppression Curves

157

Each interfering substance (listed in Figure 5-1c) was added directly to the concentrated protein standard mixture and then diluted 1:10 in succession. The samples were then diluted 1:40 in 49.95% HPLC grade acetonitrile, 49.95% HPLC grade water, and 0.1% formic acid (v/v), followed by direct infusion. The clean standard mixture, diluted to the same concentration, was used as a positive control. Three mass spectra were collected from each sample and the S/N was calculated and averaged for the three replicates. Additional samples were run between each data point to calculate the S/N threshold and the equation for the curve. S/N was normalized so that the positive control represented zero loss of S/N. The threshold where 50% of S/N was lost and the least squared fit (LSF) were determined using the sum-of-squares Solver function in Microsoft

Excel described below:

[퐶] 푦%푠:푛 푙표푠푡 = [퐶] + 푦1/2

In a similar vein as Michaelis-Menten kinetics, the signal is normalized to 1 (ymax) and

SC50 is, in a sense, the Km; the concentration of [C] at y% = 0.5 will determine the signal suppression threshold for each buffer component or interfering substance.

5.6.3 Supplementary Note 5-3: Typical Multicomponent Mixtures of Excipients:

Four “typical buffer” samples made up of commonly used buffers were analyzed for their effects on protein signal. These samples included: 1) 1X PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4); 2) 50% Thermo RIPA buffer (25mM Tris-HCl pH 7.6, 150mM

NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS); 3) Antibody buffer (10 mM arginine, 10 mM tris-HCl, 10 mM histidine, 10 mM KH2PO4, 10 mM citric acid, pH 5.5); and 4) 50% Thermo

Fisher Gentle Elution Buffer (containing molar amounts of salts). Figure 5-3a-b shows protein signal in 1X PBS and 50% RIPA, respectively. Supplementary Figure 5-3a-b shows protein signal in antibody buffer and 50% Thermo Fisher Gentle Elution Buffer, respectively.

158

5.5.4 Supplementary Note 5-4: Special Considerations for Intact Antibody Mass Spectrometry Additional Details

Humanized IgG1к (RM8671) was purchased through the National Institute of Standards and Technology (Gaithersburg, MD) The antibody was resuspended in HPLC Grade H2O to a concentration of 67 µM and stored in 20 µL aliquots at -80 °C. Figure 5-3c spectra were collected on a 9.4 T solariX FTICR MS (Bruker Daltonics, Billerica, MA) with ESI. Each sample was cleaned using Protocols 1-3. The mass range was 610 – 6000 m/z with a 1 min 37 s accumulation time (400 scans) and 0.030 s ion accumulation. The time of flight (TOF) was set at 1.500 ms with

20% Sweep Excitation Power and at 64k transient length with Sine2 processing. The flow rate was kept at 2 µL/min with the Dry Gas Temperature at 180 °C and the ESI Capillary voltage at 4.5 kV.

The RF Amplitude in the funnel was set to 200 Vpp with 150 V for Funnel 1 and 120 V for Skimmer

1. The RF Frequency for the Transfer Optics and the Collision Cell was 2 MHz with the RF

Amplitude set to 400 and 1300 Vpp respectively.

Figure A4-13 spectra were collected on a H-Class Acquity UPLC system coupled to a

Xevo G2-S Q-ToF mass spectrometer (Waters Corp, Milford, MA) using a Acquity UPLC Protein

BEH C4 (300 Å pore size, 1.7 µm particle size, 2.1 mm ID x 100 mm) column (Waters Corp,

Milford, MA). The solvents were, A: 95% H2O, 5% acetonitrile, 0.2% formic acid and B: 5% H20,

95% acetonitrile, 0.2% formic acid, and were used with the following gradient:

Time (min.) % B 0 5 10 5 12 15 37 55 40 95 43 95

159

45 5 60 5

The flow rate was 200 µL/min with the capillary voltage set to 3 kV and the sample cone voltage set at 40 V. The source temperature was kept at 150 °C and desolvation temperature 350 °C with a gas flow of 800 L/h. The method ran in the Sensitivity Analyzer mode with a 500 – 4000 m/z mass range at 1.00 s/scan time. All analyses and processing were performed using Waters UNIFI

1.7.1 software.

5.5.5 Supplementary Note 5-5: S/N Calculations

We are not aware of a program with a facile graphical user interface that can preprocess intact protein MS data from multiple vendors. For information on what we consider to be the most accurate method for S/N determination, we invite readers to refer to the work of Tsybin, et al.343

Vendors do not provide preprocessing options that would have allowed us to standardize S/N calculations in this study. In addition, not all vendors offer acceptable deisotoping and deconvolution options for intact proteins, and when they do these options require additional licenses. To avoid subjecting users to the tedious, albeit comprehensive, manual analysis of all charge states, S/N was calculated here by summing the S/N of the top five most abundant charge states for each protein as indicated below. As such, the reported S/N values represent a reproducible subset of the S/N summed from a greater number of charge states during automated deisotoping and deconvolution.

Waters XevoG2: The Waters UNIFI software did not have the capability to calculate protein S/N, therefore, S/N calculations from Figure A4-7 had to be calculated manually. Noise level was

160 estimated from a 10-20 m/z segment, adjacent to the most abundant charge state, containing no discernable molecular ion signals. From this segment, a range of intensity that contained ~97% of signal was estimated. Assuming this contained only normally distributed noise, the range would equal four times the value of “N,” i.e. N=σ. Signal “S” was defined as the maximum intensity observed within a given charge state. The total S/N across for the five charge states was then summed to give the S/N ratios reported here, and relative standard deviation (RSD) was determined using the three replicates.

Waters Synapt: For Figure 5-6, the S/N for both spectra were calculated by hand. The baseline of the noise was determined as the range of intensity containing no molecular ion signals. Assuming this contained only normally distributed noise, the range would equal two times the value of “N,” i.e. N=σ. Signal “S” was defined as the maximum intensity observed within a given charge state.

The values represented are the sum of the S/N calculations for the top 5 charge states.

Bruker instruments: For Figure A4-8, S/N was calculated using DataAnalysis 4.3 (Bruker

Daltonics) via the SNAPII (Sophisticated Numerical Annotation Procedure) algorithm with a S/N threshold of 3 (defined as the height of the mass peak above its baseline relative to one standard deviation of the noise) and a quality factor threshold of 0.9. The total S/N across for the five charge states was then summed to give the S/N ratios reported here, and relative standard deviation (RSD) was determined using the three replicates.

Thermo instruments: For Figures 5-5, Figure A4-6a, and Figure A4-9, the S/N ratios for each protein were calculated manually. For the convenience of users with the necessary licenses, this

161 method was designed to give results that are comparable to automated deconvolution and deisotoping using the Thermo Xtract or Thrash algorithms. Briefly, the five most abundant charge states of each electrosprayed protein were selected (after averaging all MS1 scans across the respective chromatographic peak). For each charge state peak, the five most abundant isotopomers

푆 푁퐿−퐵 were considered, and S/N calculated for each isotopomer by the following relation: = . 푁 푁−퐵

Here, the signal level is the NL (normalization level) for a given peak, N is empirically determined

(and related to the electronic noise level) quantity, and B is the baseline level. The NL, N, and B quantities are returned by Xcalibur 4.0 (Thermo Fisher Scientific). The S/N value for each charge state was calculated by averaging the five single isotopomer S/N values, while the final S/N reported for each protein comprised the sum of the five charge state S/N values from a single analysis.

162

Conclusions and Future Directions

This dissertation demonstrates the never-reported cross-linking reaction of cyclic disulfides and cyclic thiosulfinates. The presented work, however, is the just the first step in the development of new pharmacological chaperones. Pre-clinical studies examining the in vivo efficiency and toxicity of lead compounds are currently underway on our basic cyclic thiosulfinate scaffold 1,2-dithiane-1-oxide. Here, we target SOD1 and the most prevalent disease-associated variants of SOD1 and show efficient cross-linking in vitro, in cellulo, and in vivo with the most basic cyclic thiosulfinate scaffolds. Future directions of this work involve: 1. The targeting of different proteins with similar cysteine pair motifs, and 2. the modification of our basic scaffold,

1,2-dithiane-1-oxide (and other sized cyclic thiosulfinates), to more specific pharmacological chaperones.

A thorough search of the protein database identified dozens of other potential targets of the cyclic thiosulfinate cross-linking scaffold. We have already shown the ability to cross-link a number of these potential targets including DJ-1, a protein with implications in Parkinson’s disease, caspase-3, a protein that controls cell apoptosis and whose inactivation through cross-link formation could inhibit cell death (applicable to all neurodegenerative diseases), and ISCU, a protein with implications in certain lung cancers. These targets represent just a small percentage of potential targets. Others involve additional programmed cell death proteins [Caspases 6 and 7

(cysteine distance of ~9 Å)]; multiple cancers [nuclear estrogen receptor, ERβ (cysteine distance of 12 Å)]; tumor suppression344 breast cancer345 and infertility346 [microtubule associated protein,

MAST2 (cysteine distance of 10 Å)]; and neurologic disease [S100 calcium-binding protein B,

S100B (cysteine distance of 8 Å)].347

163

The second future direction is to modify these pharmacological chaperone scaffolds to add specificity for a particular protein target (i.e., SOD1 over DJ-1). We are currently working to synthesize a cyclic thiosulfinate with a clickable group on the 4 or 5 carbon meta to the sulfonyl group. This scaffold will allow us to rapidly screen hundreds of functional groups and measure their specificity towards a particular target using a multi-protein LC-MS assay we have developed.

Findings from these experiments could likely result in lead compounds to take

164

Appendix 1: Supplementary Figures for Cyclic Thiosulfinates and Cyclic Disulfides Selectively Cross-Link Thiols While Avoiding Modification of Lone Thiols

Figure A1-1. Determination of 1,2-dithiane-1-oxide cross-linking half-life. We performed a series of multi-concentration kinetics experiments following the protocol of Singh et al.153 and determined the overall second order rate constant of cyclic thiosulfinate-mediated cross-linking product formation to be 1.5 x 104 M-1 min-1 which, under our experimental conditions, extrapolates to a predicted half-life of 3 min for cross-linking SOD1 with 1,2-dithiane-1-oxide. Shown here are

165 representative results from one concentration, 50 µM SOD1 with 20X excess 1,2-dithiane-1-oxide

(1 mM).

Figure A1-2. Confirmation of 1,2-dithiane-1-oxide cross-linking site. LC-MS analysis of WT and

Cys111Ser SOD1 incubated with 1,2-dithiane-1-oxide. SOD1 variant Cys111Ser shows no covalent dimer formation confirming the location of 1,2-dithiane-1-oxide cross-link at the Cys111 pair at

SOD1’s dimer interface.

166

Figure A1-3. Complete kinetics of 1,2-dithiane + SOD1 reaction. Incubation of 50 μM SOD1 with

20X excess 1,2-dithiane shows slow cross-link formation (compared to cyclic thiosulfinates) at the expected dimer mass of 31,808 Da, consistent with the mechanism proposed in Figure 2-1.

167

Figure A1-4. 1,2-dithiane-1-oxide cross-links WT SOD1 in HeLa cells. Western blots of SOD1 extracted from HeLa cells treated with increasing concentrations of cyclic thiosulfinate 1,2- dithiane-1-oxide show an approximately 5 μM EC50 for 1,2-dithiane-1-oxide cross-linking of

SOD1.

168

Figure A1-5. 1,2-dithiane-1-oxide cell viability assay. SH-SY5Y cells incubated with various concentrations of 1,2-dithiane-1-oxide show cell viability is equivalent to untreated cells at cross- linker concentrations 50-fold higher than measured EC50, and that the LC50 is c.a. 200-fold that of the EC50.

Figure A1-6. The cyclic thiosulfinate, β-lipoic acid cross-links SOD1 in Hep G2 cells, and does so more efficiently than than the cyclic disulfide, α-lipoic acid. Western blots of SOD1 extracted from Hep G2 treated with increasing concentrations of lipoic acids (α and β) shows the 1-oxide

169 form increases in vivo cross-linking (as measured by formation of the SOD1 dimer, red arrow, following denaturing SDS-PAGE).

Figure A1-7. Mass spectrometry assay of α-lipoic acid vs. β-lipoic acid cross-linking of SOD1.

Both rate and extent of covalent cross-linking of SOD1 is increased by oxidation of α-lipoic acid

170 to β-lipoic acid. One of the two possible regioisomers present in our β-lipoic acid preparation is illustrated.

Figure A1-8. Cross-linking efficiency of 1,2-dithaine-1-oxide in the presence of glutathione.

Incubation of SOD1 (A) 10x 1,2-dithiane-1-oxide, (B) 10x 1,2-dithiane-1-oxide and 100x reduced glutathione shows that even in the presence of excess glutathione, 1,2-dithiane-1-oxide can efficiently cross-link SOD1. In the presence of glutathione, the rate of cross-linking was decreased, but the cross-linking yield (100%) was not affected.

171

Figure A1-9. Mass Spectrometric assay of 1,2-dithiane-1-oxide/DTT competition. SOD1 samples incubated for 24 h at various concentrations of DTT confirm 1,2-dithiane-1-oxide’s ability to cross-link the thiol pair of SOD1 at 1:1=cross-linker:DTT.

172

Compound 1

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C4H8OS2 | | Number of Basis Functions | 181 | | Electronic Energy (Eh) | -1028.6548354 | | Sum of electronic and zero-point Energies (Eh) | -1028.534059 | | Sum of electronic and thermal Energies (Eh) | -1028.526785 | | Sum of electronic and enthalpy Energies (Eh) | -1028.525841 | | Sum of electronic and thermal Free Energies (Eh) | -1028.565863 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 36 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C 1.964067 0.386173 0.061675 C 1.353054 -0.898299 0.610541 C -0.231359 1.396534 -0.797182 C 0.999089 1.573666 0.085218 S -0.009582 -1.553320 -0.418605 S -1.351967 0.076432 -0.212381 O -1.535930 0.352361 1.265379 H 2.324573 0.213376 -0.959070 H 2.836395 0.630528 0.677363 H 2.083248 -1.710653 0.639190 H 0.966389 -0.753549 1.623887 H 0.019221 1.148538 -1.832705 H -0.856871 2.293663 -0.790831 H 1.518114 2.473246 -0.259486 H 0.672055 1.767733 1.112887 ```

__Frequencies__ (Top 10 out of 39)

``` 1. 107.9031 cm-1 (Symmetry: A) 2. 189.7377 cm-1 (Symmetry: A)

173

3. 242.0999 cm-1 (Symmetry: A) 4. 305.0132 cm-1 (Symmetry: A) 5. 336.8457 cm-1 (Symmetry: A) 6. 360.1050 cm-1 (Symmetry: A) 7. 428.0133 cm-1 (Symmetry: A) 8. 449.3701 cm-1 (Symmetry: A) 9. 523.4418 cm-1 (Symmetry: A) 10. 645.3749 cm-1 (Symmetry: A) ``` ***

Compound 1_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C4H8OS2 | | Number of Basis Functions | 218 | | Electronic Energy (Eh) | -1028.76778046 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 36 |

***

Compound 2a

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11S3(1-) |

174

| Number of Basis Functions | 219 | | Electronic Energy (Eh) | -1391.67088725 | | Sum of electronic and zero-point Energies (Eh) | -1391.516024 | | Sum of electronic and thermal Energies (Eh) | -1391.504764 | | Sum of electronic and enthalpy Energies (Eh) | -1391.50382 | | Sum of electronic and thermal Free Energies (Eh) | -1391.555589 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 45 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C 1.868328 0.230491 0.369685 C 2.974277 -0.117094 -0.625283 C -0.554395 0.866479 0.699928 C 0.525903 0.511061 -0.311874 S 4.600250 -0.467144 0.162123 S -2.197607 1.195946 -0.033888 S -2.788869 -0.668164 -0.696856 C -3.531943 -1.390301 0.799220 H 2.174683 1.107440 0.955437 H 1.752845 -0.597342 1.081673 H 2.656034 -0.988168 -1.211260 H 3.081661 0.713841 -1.333405 H -0.310463 1.798428 1.222408 H -0.664727 0.084048 1.457573 H 0.210875 -0.372187 -0.881830 H 0.633920 1.335426 -1.027665 H -4.371383 -0.783183 1.138593 H -2.788231 -1.489274 1.590731 H -3.888615 -2.383048 0.515632 ```

__Frequencies__ (Top 10 out of 51)

``` 1. 33.8684 cm-1 (Symmetry: A) 2. 55.3741 cm-1 (Symmetry: A) 3. 71.5328 cm-1 (Symmetry: A) 4. 102.8667 cm-1 (Symmetry: A) 5. 126.3400 cm-1 (Symmetry: A) 6. 141.8635 cm-1 (Symmetry: A) 7. 171.4083 cm-1 (Symmetry: A) 8. 187.2660 cm-1 (Symmetry: A) 9. 256.1444 cm-1 (Symmetry: A) 10. 279.9002 cm-1 (Symmetry: A) ``` ***

175

Compound 2a_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11S3(1-) | | Number of Basis Functions | 266 | | Electronic Energy (Eh) | -1391.80213748 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 45 |

***

Compound 2b

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11OS3(1-) | | Number of Basis Functions | 238 | | Electronic Energy (Eh) | -1466.83135417 | | Sum of electronic and zero-point Energies (Eh) | -1466.673775 | | Sum of electronic and thermal Energies (Eh) | -1466.661196 |

176

| Sum of electronic and enthalpy Energies (Eh) | -1466.660251 | | Sum of electronic and thermal Free Energies (Eh) | -1466.715684 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 49 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C -1.474148 0.030694 0.299098 C -2.522510 -0.542333 -0.649633 C 0.894742 0.872463 0.559272 C -0.135245 0.287536 -0.396767 H -2.249781 -1.557947 -0.964726 H -2.591240 0.080132 -1.551907 H -1.858059 0.969050 0.718438 H -1.330763 -0.661296 1.138964 H 1.049487 0.224506 1.427678 H 0.570819 1.851197 0.931026 H -0.279268 0.978877 -1.236058 H 0.250288 -0.651017 -0.814945 S -4.186114 -0.664937 0.080001 S 2.520193 1.213798 -0.207669 S 3.266474 -0.676374 -0.575100 C 4.070348 -1.087378 1.005276 H 4.850796 -0.361626 1.234579 H 3.338585 -1.128624 1.812768 H 4.516770 -2.075641 0.873885 O -4.609450 0.875838 0.346387 ```

__Frequencies__ (Top 10 out of 54)

``` 1. 27.8930 cm-1 (Symmetry: A) 2. 36.8692 cm-1 (Symmetry: A) 3. 56.4270 cm-1 (Symmetry: A) 4. 83.0357 cm-1 (Symmetry: A) 5. 107.9356 cm-1 (Symmetry: A) 6. 125.2356 cm-1 (Symmetry: A) 7. 145.6633 cm-1 (Symmetry: A) 8. 169.2902 cm-1 (Symmetry: A) 9. 190.2284 cm-1 (Symmetry: A) 10. 246.7086 cm-1 (Symmetry: A) ``` ***

Compound 2b_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value |

177

|:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11OS3(1-) | | Number of Basis Functions | 288 | | Electronic Energy (Eh) | -1466.98332086 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 49 | ***

Compound 3b

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C5H12OS3 | | Number of Basis Functions | 243 | | Electronic Energy (Eh) | -1467.31975395 | | Sum of electronic and zero-point Energies (Eh) | -1467.149337 | | Sum of electronic and thermal Energies (Eh) | -1467.136401 | | Sum of electronic and enthalpy Energies (Eh) | -1467.135457 | | Sum of electronic and thermal Free Energies (Eh) | -1467.191632 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 49 |

__Molecular Geometry in Cartesian Coordinates__

178

```xyz C 1.432734 0.101575 0.268062 C 2.465610 -0.475808 -0.692738 C -0.940378 0.919179 0.528991 C 0.089067 0.333566 -0.427596 S 4.089895 -0.767645 0.047507 S -2.579199 1.205662 -0.229124 S -3.281169 -0.709663 -0.549379 C -4.051374 -1.111153 1.050080 H 1.801786 1.052520 0.671927 H 1.303194 -0.582925 1.114671 H 2.156506 -1.464310 -1.054237 H 2.601833 0.169302 -1.566837 H -0.633092 1.912465 0.874637 H -1.070237 0.287047 1.412830 H -0.285934 -0.616907 -0.826529 H 0.220348 1.013688 -1.277432 H -4.847865 -0.402307 1.276890 H -3.307701 -1.117927 1.847626 H -4.472687 -2.113300 0.943430 O 4.571455 0.817685 0.335312 H 4.315816 1.063371 1.235669 ```

__Frequencies__ (Top 10 out of 57)

``` 1. 20.7858 cm-1 (Symmetry: A) 2. 38.0627 cm-1 (Symmetry: A) 3. 59.1774 cm-1 (Symmetry: A) 4. 89.2028 cm-1 (Symmetry: A) 5. 106.1720 cm-1 (Symmetry: A) 6. 119.2715 cm-1 (Symmetry: A) 7. 148.3095 cm-1 (Symmetry: A) 8. 170.3174 cm-1 (Symmetry: A) 9. 187.7952 cm-1 (Symmetry: A) 10. 243.8667 cm-1 (Symmetry: A) ``` ***

Compound 3b_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C5H12OS3 | | Number of Basis Functions | 294 | | Electronic Energy (Eh) | -1467.47177084 |

179

| Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 49 |

***

Compound 5

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C4H8S2 | | Number of Basis Functions | 162 | | Electronic Energy (Eh) | -953.493944195 | | Sum of electronic and zero-point Energies (Eh) | -953.376889 | | Sum of electronic and thermal Energies (Eh) | -953.370664 | | Sum of electronic and enthalpy Energies (Eh) | -953.369719 | | Sum of electronic and thermal Free Energies (Eh) | -953.407259 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 32 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C 1.658266 0.751255 0.152766 C 0.465934 1.514096 -0.421972 C 0.465947 -1.514093 0.421972 C 1.658272 -0.751242 -0.152766 S -1.121858 0.990013 0.312126 S -1.121848 -0.990023 -0.312127 H 1.691146 0.906561 1.237998 H 2.571032 1.192674 -0.262781

180

H 0.542614 2.581440 -0.196009 H 0.399572 1.394269 -1.507782 H 0.399586 -1.394265 1.507783 H 0.542635 -2.581436 0.196011 H 2.571043 -1.192652 0.262781 H 1.691155 -0.906548 -1.237999 ```

__Frequencies__ (Top 10 out of 36)

``` 1. 179.6633 cm-1 (Symmetry: A) 2. 229.6942 cm-1 (Symmetry: A) 3. 292.6208 cm-1 (Symmetry: A) 4. 349.8603 cm-1 (Symmetry: A) 5. 368.3507 cm-1 (Symmetry: A) 6. 490.6125 cm-1 (Symmetry: A) 7. 512.7474 cm-1 (Symmetry: A) 8. 672.6474 cm-1 (Symmetry: A) 9. 680.7006 cm-1 (Symmetry: A) 10. 803.9016 cm-1 (Symmetry: A) ``` ***

Compound 5_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C4H8S2 | | Number of Basis Functions | 196 | | Electronic Energy (Eh) | -953.586078338 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 32 |

***

181

H2O

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | H2O | | Number of Basis Functions | 29 | | Electronic Energy (Eh) | -76.4034163646 | | Sum of electronic and zero-point Energies (Eh) | -76.381923 | | Sum of electronic and thermal Energies (Eh) | -76.379087 | | Sum of electronic and enthalpy Energies (Eh) | -76.378143 | | Sum of electronic and thermal Free Energies (Eh) | -76.39957 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 5 |

__Molecular Geometry in Cartesian Coordinates__

```xyz O 0.000000 0.000000 0.117054 H -0.000000 0.765770 -0.468215 H -0.000000 -0.765770 -0.468215 ```

__Frequencies__ (Top 10 out of 3)

``` 1. 1600.5560 cm-1 (Symmetry: A1) 2. 3864.1091 cm-1 (Symmetry: A1) 3. 3969.7889 cm-1 (Symmetry: B2) ``` *** H2O_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 |

182

| Stoichiometry | H2O | | Number of Basis Functions | 34 | | Electronic Energy (Eh) | -76.4288811325 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 5 |

***

+ H3O

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 1 | | Multiplicity | 1 | | Stoichiometry | H3O(1+) | | Number of Basis Functions | 34 | | Electronic Energy (Eh) | -76.7941179852 | | Sum of electronic and zero-point Energies (Eh) | -76.758967 | | Sum of electronic and thermal Energies (Eh) | -76.756083 | | Sum of electronic and enthalpy Energies (Eh) | -76.755138 | | Sum of electronic and thermal Free Energies (Eh) | -76.778096 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 5 |

__Molecular Geometry in Cartesian Coordinates__

```xyz O 1.621991 -1.619069 -0.219138 H 2.548446 -1.706535 0.083874 H 1.229983 -0.775059 0.083799

183

H 1.082338 -2.382095 0.071465 ```

__Frequencies__ (Top 10 out of 6)

``` 1. 933.8456 cm-1 (Symmetry: A) 2. 1672.7220 cm-1 (Symmetry: A) 3. 1683.4338 cm-1 (Symmetry: A) 4. 3641.8467 cm-1 (Symmetry: A) 5. 3743.3818 cm-1 (Symmetry: A) 6. 3754.1724 cm-1 (Symmetry: A) ``` *** + H3O _Single Point Energies

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 1 | | Multiplicity | 1 | | Stoichiometry | H3O(1+) | | Number of Basis Functions | 40 | | Electronic Energy (Eh) | -76.8177981822 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 5 |

***

OH–

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 |

184

| Multiplicity | 1 | | Stoichiometry | HO(1-) | | Number of Basis Functions | 24 | | Electronic Energy (Eh) | -75.892480289 | | Sum of electronic and zero-point Energies (Eh) | -75.883653 | | Sum of electronic and thermal Energies (Eh) | -75.881292 | | Sum of electronic and enthalpy Energies (Eh) | -75.880348 | | Sum of electronic and thermal Free Energies (Eh) | -75.899905 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 5 |

__Molecular Geometry in Cartesian Coordinates__

```xyz O -0.000000 0.000000 0.107168 H 0.000000 -0.000000 -0.857345 ```

__Frequencies__ (Top 10 out of 1)

``` 1. 3874.7480 cm-1 (Symmetry: SG) ``` ***

OH–_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | HO(1-) | | Number of Basis Functions | 28 | | Electronic Energy (Eh) | -75.9179929945 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A |

185

| Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 5 |

***

TS(1→2b)

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11OS3(1-) | | Number of Basis Functions | 238 | | Electronic Energy (Eh) | -1466.82067268 | | Sum of electronic and zero-point Energies (Eh) | -1466.662876 | | Sum of electronic and thermal Energies (Eh) | -1466.65143 | | Sum of electronic and enthalpy Energies (Eh) | -1466.650486 | | Sum of electronic and thermal Free Energies (Eh) | -1466.701512 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 49 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C 0.696031 1.897188 0.360284 C -0.158468 0.731086 0.854477 C 2.455195 0.433074 -0.833541 C 2.173638 1.543840 0.173743 H 0.370118 0.179595 1.640200 H -1.110755 1.083512 1.252153 H 0.272871 2.289822 -0.573257 H 0.631022 2.702376 1.102504 H 3.534740 0.299948 -0.959592 H 2.017764 0.647179 -1.815306 H 2.592225 1.243334 1.141597 H 2.723153 2.436781 -0.145723

186

S -0.585046 -0.480507 -0.461008 S 1.767435 -1.184817 -0.333093 O 2.206107 -1.370478 1.157804 C -3.541884 -0.524741 0.867448 H -3.370414 -1.603631 0.839123 H -4.616742 -0.343524 0.944703 H -3.060391 -0.117532 1.761475 S -2.893609 0.269280 -0.642447 ```

__Frequencies__ (Top 10 out of 54)

``` 1. -175.5524 cm-1 (Symmetry: A) * 2. 50.5474 cm-1 (Symmetry: A) 3. 60.0109 cm-1 (Symmetry: A) 4. 100.4443 cm-1 (Symmetry: A) 5. 116.3230 cm-1 (Symmetry: A) 6. 146.4042 cm-1 (Symmetry: A) 7. 148.8266 cm-1 (Symmetry: A) 8. 188.2539 cm-1 (Symmetry: A) 9. 224.8842 cm-1 (Symmetry: A) 10. 269.0928 cm-1 (Symmetry: A) ``` ***

TS(1→2b)_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11OS3(1-) | | Number of Basis Functions | 288 | | Electronic Energy (Eh) | -1466.97327716 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 49 |

187

***

TS(3b→4)

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C6H15OS4(1-) | | Number of Basis Functions | 300 | | Electronic Energy (Eh) | -1905.4865875 | | Sum of electronic and zero-point Energies (Eh) | -1905.279523 | | Sum of electronic and thermal Energies (Eh) | -1905.262528 | | Sum of electronic and enthalpy Energies (Eh) | -1905.261584 | | Sum of electronic and thermal Free Energies (Eh) | -1905.327996 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 62 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C 0.518876 -0.814017 -0.392966 C 1.659869 -0.628662 0.597299 C -1.977266 -1.052567 -0.685594 C -0.840819 -0.859626 0.308514 S 3.303893 -0.565599 -0.203152 S -3.640090 -1.087453 0.075208 S -3.901406 0.861989 0.703759 C -4.551007 1.666398 -0.794213 H 0.683165 -1.741991 -0.953951 H 0.532783 0.012427 -1.115146 H 1.524757 0.293627 1.169792 H 1.714334 -1.471467 1.288155 H -1.898413 -2.023331 -1.187696 H -1.965946 -0.278873 -1.459685 H -0.999542 0.074618 0.861736 H -0.858906 -1.676334 1.040401 H -5.483200 1.194462 -1.105141 H -3.818008 1.632105 -1.600944 H -4.740029 2.707958 -0.524782 O 3.327188 -2.657343 0.260388 H 4.048057 -2.980657 -0.294469

188

S 3.150699 1.650513 -0.677121 C 3.652321 2.379965 0.916274 H 4.652746 2.041822 1.195340 H 3.659771 3.468358 0.820536 H 2.949555 2.105863 1.707757 ```

__Frequencies__ (Top 10 out of 72)

``` 1. -166.9159 cm-1 (Symmetry: A) * 2. 17.4466 cm-1 (Symmetry: A) 3. 22.8543 cm-1 (Symmetry: A) 4. 47.1952 cm-1 (Symmetry: A) 5. 54.8378 cm-1 (Symmetry: A) 6. 71.7013 cm-1 (Symmetry: A) 7. 106.8647 cm-1 (Symmetry: A) 8. 118.8890 cm-1 (Symmetry: A) 9. 124.3354 cm-1 (Symmetry: A) 10. 136.7940 cm-1 (Symmetry: A) ``` ***

TS(3b→4)_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C6H15OS4(1-) | | Number of Basis Functions | 364 | | Electronic Energy (Eh) | -1905.6808201 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 62 |

***

189

TS(5→2a)

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11S3(1-) | | Number of Basis Functions | 219 | | Electronic Energy (Eh) | -1391.65573095 | | Sum of electronic and zero-point Energies (Eh) | -1391.501262 | | Sum of electronic and thermal Energies (Eh) | -1391.49097 | | Sum of electronic and enthalpy Energies (Eh) | -1391.490025 | | Sum of electronic and thermal Free Energies (Eh) | -1391.538353 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 45 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C 1.021773 1.781901 0.012509 C -0.003094 0.895871 0.718564 C 2.706021 -0.078032 -0.574676 C 2.469260 1.286532 0.068696 H 0.350772 0.610340 1.712702 H -0.947495 1.431815 0.823085 H 0.712609 1.918582 -1.032250 H 0.980678 2.772219 0.483944 H 3.778496 -0.296076 -0.592616 H 2.357153 -0.063247 -1.614567 H 2.795289 1.245182 1.116234 H 3.106663 2.025697 -0.434209 S -0.412084 -0.659390 -0.187317 S 1.868467 -1.447750 0.303077 C -3.438575 -0.093617 0.853066 H -3.364487 -1.129005 1.194735 H -4.495176 0.174151 0.774132 H -2.974937 0.556195 1.601355 S -2.633375 0.107028 -0.772104 ```

190

__Frequencies__ (Top 10 out of 51)

``` 1. -198.0829 cm-1 (Symmetry: A) * 2. 52.6577 cm-1 (Symmetry: A) 3. 62.8482 cm-1 (Symmetry: A) 4. 115.7516 cm-1 (Symmetry: A) 5. 129.6520 cm-1 (Symmetry: A) 6. 159.0236 cm-1 (Symmetry: A) 7. 211.5483 cm-1 (Symmetry: A) 8. 248.1296 cm-1 (Symmetry: A) 9. 263.4596 cm-1 (Symmetry: A) 10. 277.8272 cm-1 (Symmetry: A) ``` ***

TS(5→2a)_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11S3(1-) | | Number of Basis Functions | 266 | | Electronic Energy (Eh) | -1391.78736939 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 45 |

***

MeS–

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:|

191

| Charge | -1 | | Multiplicity | 1 | | Stoichiometry | CH3S(1-) | | Number of Basis Functions | 57 | | Electronic Energy (Eh) | -438.170119418 | | Sum of electronic and zero-point Energies (Eh) | -438.133136 | | Sum of electronic and thermal Energies (Eh) | -438.13009 | | Sum of electronic and enthalpy Energies (Eh) | -438.129146 | | Sum of electronic and thermal Free Energies (Eh) | -438.15674 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 13 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 3.815429 0.963899 0.840898 C 4.237882 2.128317 -0.522788 H 3.502012 2.077639 -1.330756 H 5.217466 1.899019 -0.952500 H 4.267051 3.162297 -0.166624 ```

__Frequencies__ (Top 10 out of 9)

``` 1. 717.5827 cm-1 (Symmetry: A) 2. 951.4018 cm-1 (Symmetry: A) 3. 953.9670 cm-1 (Symmetry: A) 4. 1349.5968 cm-1 (Symmetry: A) 5. 1483.6661 cm-1 (Symmetry: A) 6. 1484.8735 cm-1 (Symmetry: A) 7. 3049.0206 cm-1 (Symmetry: A) 8. 3121.1911 cm-1 (Symmetry: A) 9. 3122.3645 cm-1 (Symmetry: A) ``` ***

MeS–_Single Point Energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 |

192

| Multiplicity | 1 | | Stoichiometry | CH3S(1-) | | Number of Basis Functions | 70 | | Electronic Energy (Eh) | -438.209484579 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 13 |

*** TS(3a→4) Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C6H15OS4(1-) | | Number of Basis Functions | 300 | | Electronic Energy (Eh) | -1905.48540546 | | Sum of electronic and zero-point Energies (Eh) | -1905.27846 | | Sum of electronic and thermal Energies (Eh) | -1905.261393 | | Sum of electronic and enthalpy Energies (Eh) | -1905.260449 | | Sum of electronic and thermal Free Energies (Eh) | -1905.327586 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 62 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C 0.279355 -1.284340 -0.453823 C 1.365397 -1.559750 0.581439

193

C -2.191194 -0.960126 -0.863100 C -1.111016 -1.204276 0.181239 S 3.055097 -1.587547 -0.114868 S -3.880851 -0.825885 -0.175166 S -3.820643 0.966187 0.848656 C -4.205286 2.170586 -0.460646 H 0.297278 -2.076308 -1.213269 H 0.506139 -0.341723 -0.967438 H 1.327732 -0.801742 1.373789 H 1.192982 -2.531385 1.057609 H -2.258561 -1.801766 -1.561539 H -1.982897 -0.060092 -1.450117 H -1.130666 -0.392318 0.918860 H -1.332812 -2.134715 0.718340 H -5.186457 1.965920 -0.889362 H -3.439040 2.156366 -1.236511 H -4.212550 3.151985 0.018856 O 3.870387 2.721116 -0.572290 H 3.762747 2.880741 -1.518525 S 3.451633 0.652403 -0.409369 C 3.944564 0.867040 1.334840 H 4.891704 0.360849 1.528239 H 4.048847 1.938326 1.495958 H 3.177755 0.459603 1.995699 ```

__Frequencies__ (Top 10 out of 72)

``` 1. -176.4482 cm-1 (Symmetry: A) * 2. 12.3702 cm-1 (Symmetry: A) 3. 18.8335 cm-1 (Symmetry: A) 4. 48.8225 cm-1 (Symmetry: A) 5. 56.1643 cm-1 (Symmetry: A) 6. 74.6258 cm-1 (Symmetry: A) 7. 93.8656 cm-1 (Symmetry: A) 8. 106.0095 cm-1 (Symmetry: A) 9. 110.1965 cm-1 (Symmetry: A) 10. 129.3395 cm-1 (Symmetry: A) ``` ***

TS(3a→4)_single point energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C6H15OS4(1-) |

194

| Number of Basis Functions | 364 | | Electronic Energy (Eh) | -1905.67946076 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 62 |

MeSOH

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | CH4OS | | Number of Basis Functions | 81 | | Electronic Energy (Eh) | -513.819362371 | | Sum of electronic and zero-point Energies (Eh) | -513.766713 | | Sum of electronic and thermal Energies (Eh) | -513.762248 | | Sum of electronic and enthalpy Energies (Eh) | -513.761304 | | Sum of electronic and thermal Free Energies (Eh) | -513.793215 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 17 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C -1.372941 0.444353 -0.000754 S 0.086793 -0.612793 0.012490 H -2.235573 -0.227663 0.006270 H -1.396000 1.047918 -0.909226 O 1.292449 0.550281 -0.118570

195

H 1.546681 0.836314 0.770667 H -1.405744 1.079755 0.885528 ```

__Frequencies__ (Top 10 out of 15)

``` 1. 191.7965 cm-1 (Symmetry: A) 2. 298.0856 cm-1 (Symmetry: A) 3. 372.6649 cm-1 (Symmetry: A) 4. 725.2424 cm-1 (Symmetry: A) 5. 793.4169 cm-1 (Symmetry: A) 6. 982.5542 cm-1 (Symmetry: A) 7. 987.0715 cm-1 (Symmetry: A) 8. 1191.2447 cm-1 (Symmetry: A) 9. 1357.3622 cm-1 (Symmetry: A) 10. 1448.4966 cm-1 (Symmetry: A) ``` ***

MeSOH_Single_point_energy

Run with Gaussian 09revisionE.01.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | CH4OS | | Number of Basis Functions | 98 | | Electronic Energy (Eh) | -513.87926544 | | Sum of electronic and zero-point Energies (Eh) | N/A | | Sum of electronic and thermal Energies (Eh) | N/A | | Sum of electronic and enthalpy Energies (Eh) | N/A | | Sum of electronic and thermal Free Energies (Eh) | N/A | | Number of Imaginary Frequencies | N/A | | Mean of alpha and beta Electrons | 17 |

***

Figure A1-10. Coordinates of optimized stationary points using M06-2X/6-31+G(d,p) IEF-

PCMH2O.

196

Figure A1-11. The energies of the HOMO and LUMOs of MeS–, 1,2-dithiane, and 1,2-dithiane-

1-oxide, respectively. Computed using M06-2X/6-311++G(d,p) IEF-PCMH2O// M06-2X/6-

31+G(d,p).

197

Figure A1-12. Cross-linking SOD1 using 1,2-dithiepane-1-oxide. 1,2-dithiepane-1-oxide forms rapid and complete cross-link of SOD1 following the same proposed mechanism as 1,2-dithiane-

1-oxide while 1,2-dithiepane does not. The observed covalent dimer (D) appeared at 31,824 Da

(two SOD1 monomers [2 x 15,844 Da] + 1,2-dithiepane-1-oxide [152 Da] – oxygen [16 Da]).

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

Appendix 2: Supplementary Figures for Closed-Shell Repulsions Pre-Distort Cyclic Thiosulfinates and Accelerate Nucleophilic Substitution Reactions

Supplementary Figure A2-1. Coordinates of optimized stationary points of reactants (3—10)a and (3—10)b using M06-2X/6-311++G(d,p) IEF-PCMH2O.

Methyl_Thiolate

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | CH3S(1-) | | Number of Basis Functions | 73 | | Electronic Energy (Eh) | -438.209663176 | | Sum of electronic and zero-point Energies (Eh) | -438.17273 | | Sum of electronic and thermal Energies (Eh) | -438.169681 | | Sum of electronic and enthalpy Energies (Eh) | -438.168737 |

215

| Sum of electronic and thermal Free Energies (Eh) | -438.195294 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 13 |

__Molecular Geometry in Cartesian Coordinates__

```xyz C -0.000000 0.000000 -1.130787 H -0.000000 1.017917 -1.525484 H 0.881542 -0.508958 -1.525484 S 0.000000 0.000000 0.710073 H -0.881542 -0.508958 -1.525484 ```

__Frequencies__ (Top 10 out of 9)

``` 1. 710.5292 cm-1 (Symmetry: ?A) 2. 953.9094 cm-1 (Symmetry: ?A) 3. 957.4586 cm-1 (Symmetry: ?A) 4. 1347.5282 cm-1 (Symmetry: ?A) 5. 1479.6620 cm-1 (Symmetry: ?A) 6. 1481.2514 cm-1 (Symmetry: ?A) 7. 3047.4960 cm-1 (Symmetry: A1) 8. 3116.8720 cm-1 (Symmetry: E) 9. 3117.2872 cm-1 (Symmetry: E) ``` *** 3a_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | CH2OS2 | | Number of Basis Functions | 118 | | Electronic Energy (Eh) | -910.82939738 | | Sum of electronic and zero-point Energies (Eh) | -910.796407 | | Sum of electronic and thermal Energies (Eh) | -910.792152 | | Sum of electronic and enthalpy Energies (Eh) | -910.791207 | | Sum of electronic and thermal Free Energies (Eh) | -910.824074 |

216

| Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 24 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -0.697029 0.014477 -0.482443 C 0.524109 1.143479 0.201047 S 1.281392 -0.489006 0.083583 O -1.706161 -0.330171 0.556929 H 0.255186 1.473649 1.198529 H 0.899634 1.899313 -0.478490 ```

__Frequencies__ (Top 10 out of 12)

``` 1. 281.1701 cm-1 (Symmetry: A) 2. 361.0781 cm-1 (Symmetry: A) 3. 457.5815 cm-1 (Symmetry: A) 4. 599.7161 cm-1 (Symmetry: A) 5. 851.1966 cm-1 (Symmetry: A) 6. 926.3478 cm-1 (Symmetry: A) 7. 973.2585 cm-1 (Symmetry: A) 8. 1078.8131 cm-1 (Symmetry: A) 9. 1104.7904 cm-1 (Symmetry: A) 10. 1441.6020 cm-1 (Symmetry: A) ``` *** 3b_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | CH2S2 | | Number of Basis Functions | 96 | | Electronic Energy (Eh) | -835.642981205 | | Sum of electronic and zero-point Energies (Eh) | -835.613489 | | Sum of electronic and thermal Energies (Eh) | -835.610123 | | Sum of electronic and enthalpy Energies (Eh) | -835.609179 | | Sum of electronic and thermal Free Energies (Eh) | -835.639689 |

217

| Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 20 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 1.048985 -0.319729 -0.000002 C 0.000272 1.135841 -0.000001 S -1.049193 -0.319383 0.000001 H 0.000812 1.705396 -0.920385 H 0.000877 1.705358 0.920408 ```

__Frequencies__ (Top 10 out of 9)

``` 1. 517.4562 cm-1 (Symmetry: A) 2. 621.3439 cm-1 (Symmetry: A) 3. 885.1527 cm-1 (Symmetry: A) 4. 954.4319 cm-1 (Symmetry: A) 5. 976.6701 cm-1 (Symmetry: A) 6. 1091.5116 cm-1 (Symmetry: A) 7. 1466.6299 cm-1 (Symmetry: A) 8. 3163.3494 cm-1 (Symmetry: A) 9. 3269.0906 cm-1 (Symmetry: A) ``` *** 4a_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C2H4OS2 | | Number of Basis Functions | 154 | | Electronic Energy (Eh) | -950.134114061 | | Sum of electronic and zero-point Energies (Eh) | -950.071859 | | Sum of electronic and thermal Energies (Eh) | -950.066522 | | Sum of electronic and enthalpy Energies (Eh) | -950.065577 | | Sum of electronic and thermal Free Energies (Eh) | -950.101209 | | Number of Imaginary Frequencies | 0 |

218

| Mean of alpha and beta Electrons | 28 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -0.908539 -0.153788 -0.452626 C 0.095676 1.386595 -0.302780 C 1.272123 0.823468 0.483601 S 1.071627 -0.930374 -0.044131 O -1.758881 -0.235500 0.779846 H -0.496324 2.163033 0.179520 H 0.375030 1.682599 -1.313583 H 1.133552 0.902004 1.558928 H 2.242584 1.222568 0.199541 ```

__Frequencies__ (Top 10 out of 21)

``` 1. 131.0597 cm-1 (Symmetry: A) 2. 269.1548 cm-1 (Symmetry: A) 3. 326.6100 cm-1 (Symmetry: A) 4. 445.8414 cm-1 (Symmetry: A) 5. 500.7762 cm-1 (Symmetry: A) 6. 661.3069 cm-1 (Symmetry: A) 7. 722.2829 cm-1 (Symmetry: A) 8. 830.3837 cm-1 (Symmetry: A) 9. 955.5185 cm-1 (Symmetry: A) 10. 1008.8409 cm-1 (Symmetry: A) ``` *** 4b_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C2H4S2 | | Number of Basis Functions | 132 | | Electronic Energy (Eh) | -874.944257737 | | Sum of electronic and zero-point Energies (Eh) | -874.885459 | | Sum of electronic and thermal Energies (Eh) | -874.881144 | | Sum of electronic and enthalpy Energies (Eh) | -874.8802 | | Sum of electronic and thermal Free Energies (Eh) | -874.913265 |

219

| Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 24 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -1.065536 -0.628113 0.098579 C -0.724146 1.151847 -0.242721 C 0.724142 1.151849 0.242721 S 1.065538 -0.628110 -0.098578 H -1.399529 1.814332 0.293974 H -0.788312 1.324364 -1.316056 H 0.788307 1.324367 1.316056 H 1.399522 1.814337 -0.293975 ```

__Frequencies__ (Top 10 out of 18)

``` 1. 203.2265 cm-1 (Symmetry: A) 2. 449.4764 cm-1 (Symmetry: A) 3. 500.8979 cm-1 (Symmetry: A) 4. 688.9446 cm-1 (Symmetry: A) 5. 722.4384 cm-1 (Symmetry: A) 6. 862.4175 cm-1 (Symmetry: A) 7. 959.5563 cm-1 (Symmetry: A) 8. 1001.3543 cm-1 (Symmetry: A) 9. 1103.3037 cm-1 (Symmetry: A) 10. 1203.5006 cm-1 (Symmetry: A) ``` *** 5a_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C3H6OS2 | | Number of Basis Functions | 190 | | Electronic Energy (Eh) | -989.463257447 | | Sum of electronic and zero-point Energies (Eh) | -989.371538 | | Sum of electronic and thermal Energies (Eh) | -989.365319 | | Sum of electronic and enthalpy Energies (Eh) | -989.364375 |

220

| Sum of electronic and thermal Free Energies (Eh) | -989.402099 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 32 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -1.118444 -0.204409 -0.378651 C -0.171055 1.331011 -0.653462 C 0.927916 1.376635 0.394215 C 1.791309 0.131295 0.235773 S 0.686109 -1.308666 -0.078074 O -1.818753 -0.007892 0.940355 H 0.226733 1.282645 -1.669147 H -0.897022 2.139243 -0.570288 H 0.480844 1.409207 1.389676 H 1.537628 2.272399 0.262396 H 2.369455 -0.083973 1.132163 H 2.460724 0.219173 -0.619194 ```

__Frequencies__ (Top 10 out of 30)

``` 1. 111.0170 cm-1 (Symmetry: A) 2. 234.5875 cm-1 (Symmetry: A) 3. 277.5966 cm-1 (Symmetry: A) 4. 349.2926 cm-1 (Symmetry: A) 5. 438.2524 cm-1 (Symmetry: A) 6. 461.6862 cm-1 (Symmetry: A) 7. 531.6345 cm-1 (Symmetry: A) 8. 646.0792 cm-1 (Symmetry: A) 9. 678.9954 cm-1 (Symmetry: A) 10. 861.3629 cm-1 (Symmetry: A) ``` *** 5b_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C3H6S2 | | Number of Basis Functions | 168 | | Electronic Energy (Eh) | -914.274851982 |

221

| Sum of electronic and zero-point Energies (Eh) | -914.187119 | | Sum of electronic and thermal Energies (Eh) | -914.181634 | | Sum of electronic and enthalpy Energies (Eh) | -914.18069 | | Sum of electronic and thermal Free Energies (Eh) | -914.217647 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 28 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -0.794440 -1.145606 0.033144 C -1.450254 0.574913 0.104834 C -0.358070 1.543236 -0.332327 C 0.947180 1.139178 0.337753 S 1.259188 -0.601180 -0.106884 H -2.319754 0.618340 -0.549154 H -1.756897 0.769422 1.131581 H -0.634582 2.559582 -0.040990 H -0.239081 1.513190 -1.416886 H 0.881933 1.232637 1.422428 H 1.799263 1.711459 -0.028699 ```

__Frequencies__ (Top 10 out of 27)

``` 1. 27.9747 cm-1 (Symmetry: A) 2. 300.6580 cm-1 (Symmetry: A) 3. 389.6149 cm-1 (Symmetry: A) 4. 473.4817 cm-1 (Symmetry: A) 5. 517.9033 cm-1 (Symmetry: A) 6. 672.5491 cm-1 (Symmetry: A) 7. 693.6922 cm-1 (Symmetry: A) 8. 853.5899 cm-1 (Symmetry: A) 9. 894.9561 cm-1 (Symmetry: A) 10. 925.3759 cm-1 (Symmetry: A) ``` *** 6a_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C4H8OS2 |

222

| Number of Basis Functions | 226 | | Electronic Energy (Eh) | -1028.76825423 | | Sum of electronic and zero-point Energies (Eh) | -1028.647575 | | Sum of electronic and thermal Energies (Eh) | -1028.64031 | | Sum of electronic and enthalpy Energies (Eh) | -1028.639366 | | Sum of electronic and thermal Free Energies (Eh) | -1028.679333 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 36 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -0.003174 -1.557261 -0.422972 S -1.354357 0.077652 -0.206397 C -0.231419 1.387405 -0.807450 C 0.995449 1.574763 0.075053 C 1.959603 0.387741 0.066585 C 1.341855 -0.889988 0.619955 O -1.522116 0.356887 1.268299 H -0.855018 2.283295 -0.817520 H 0.022043 1.117163 -1.834152 H 1.514705 2.468510 -0.278028 H 0.667466 1.779466 1.098044 H 2.825145 0.636721 0.686593 H 2.326934 0.207137 -0.947970 H 2.069235 -1.700697 0.670965 H 0.933972 -0.732471 1.620726 ```

__Frequencies__ (Top 10 out of 39)

``` 1. 112.5043 cm-1 (Symmetry: A) 2. 194.7829 cm-1 (Symmetry: A) 3. 243.4699 cm-1 (Symmetry: A) 4. 304.8492 cm-1 (Symmetry: A) 5. 338.0509 cm-1 (Symmetry: A) 6. 359.4040 cm-1 (Symmetry: A) 7. 427.8570 cm-1 (Symmetry: A) 8. 445.8754 cm-1 (Symmetry: A) 9. 524.8572 cm-1 (Symmetry: A) 10. 634.6410 cm-1 (Symmetry: A) ``` *** 6b_Reactant

Run with Gaussian 16revisionA.03.

223

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C4H8S2 | | Number of Basis Functions | 204 | | Electronic Energy (Eh) | -953.586417106 | | Sum of electronic and zero-point Energies (Eh) | -953.469429 | | Sum of electronic and thermal Energies (Eh) | -953.463207 | | Sum of electronic and enthalpy Energies (Eh) | -953.462263 | | Sum of electronic and thermal Free Energies (Eh) | -953.499787 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 32 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 1.122019 0.991363 0.315756 S 1.118852 -0.994616 -0.315754 C -0.466961 -1.507778 0.427439 C -1.659119 -0.749097 -0.149176 C -1.656574 0.753807 0.149154 C -0.462201 1.509185 -0.427493 H -0.393912 -1.374164 1.508784 H -0.544181 -2.575210 0.212722 H -1.693884 -0.909540 -1.231320 H -2.570048 -1.185778 0.270275 H -2.566060 1.193581 -0.270152 H -1.690995 0.914410 1.231302 H -0.389377 1.375232 -1.508748 H -0.536357 2.576808 -0.212443 ```

__Frequencies__ (Top 10 out of 36)

``` 1. 181.6062 cm-1 (Symmetry: A) 2. 231.8222 cm-1 (Symmetry: A) 3. 293.4102 cm-1 (Symmetry: A) 4. 349.9327 cm-1 (Symmetry: A) 5. 370.1472 cm-1 (Symmetry: A) 6. 491.1524 cm-1 (Symmetry: A) 7. 508.6505 cm-1 (Symmetry: A) 8. 667.2621 cm-1 (Symmetry: A) 9. 676.1891 cm-1 (Symmetry: A)

224

10. 803.6331 cm-1 (Symmetry: A) ``` *** 7a_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C5H10OS2 | | Number of Basis Functions | 262 | | Electronic Energy (Eh) | -1068.06759181 | | Sum of electronic and zero-point Energies (Eh) | -1067.917889 | | Sum of electronic and thermal Energies (Eh) | -1067.909519 | | Sum of electronic and enthalpy Energies (Eh) | -1067.908574 | | Sum of electronic and thermal Free Energies (Eh) | -1067.950925 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 40 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -1.583325 0.080075 -0.178060 C -0.604704 1.566874 -0.564854 C 0.702452 1.737841 0.203435 C 1.881818 0.921981 -0.353932 C 2.180520 -0.388474 0.379040 C 0.965214 -1.259222 0.684843 S -0.201680 -1.478674 -0.699179 O -1.699000 0.083640 1.328210 H -1.308256 2.369914 -0.330068 H -0.450283 1.546509 -1.645870 H 0.943530 2.801657 0.146738 H 0.534131 1.524125 1.264041 H 1.698744 0.711006 -1.411443 H 2.786340 1.532945 -0.317535 H 2.657124 -0.167232 1.340105 H 2.904113 -0.960905 -0.206592 H 0.369603 -0.855793 1.509462 H 1.265230 -2.267775 0.970112 ```

225

__Frequencies__ (Top 10 out of 48)

``` 1. 118.8111 cm-1 (Symmetry: A) 2. 169.3134 cm-1 (Symmetry: A) 3. 211.8459 cm-1 (Symmetry: A) 4. 226.1384 cm-1 (Symmetry: A) 5. 260.1081 cm-1 (Symmetry: A) 6. 321.8407 cm-1 (Symmetry: A) 7. 342.4044 cm-1 (Symmetry: A) 8. 394.5000 cm-1 (Symmetry: A) 9. 426.5671 cm-1 (Symmetry: A) 10. 494.4263 cm-1 (Symmetry: A) ``` *** 7b_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C5H10S2 | | Number of Basis Functions | 240 | | Electronic Energy (Eh) | -992.887077713 | | Sum of electronic and zero-point Energies (Eh) | -992.741301 | | Sum of electronic and thermal Energies (Eh) | -992.733819 | | Sum of electronic and enthalpy Energies (Eh) | -992.732875 | | Sum of electronic and thermal Free Energies (Eh) | -992.773288 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 36 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -0.301573 -1.466776 0.559343 C 1.266325 -1.263625 -0.362844 C 2.032026 0.020002 -0.069701 C 1.303546 1.334701 -0.381932 C 0.192989 1.704261 0.618775 C -1.236837 1.448040 0.154244 S -1.551573 -0.226844 -0.527781 H 1.863353 -2.133789 -0.079577 H 1.022518 -1.348604 -1.423259

226

H 2.950679 -0.020963 -0.664562 H 2.340040 0.024712 0.980725 H 2.054461 2.127766 -0.372964 H 0.899865 1.304523 -1.399880 H 0.359642 1.177079 1.561356 H 0.247201 2.771017 0.854340 H -1.939167 1.631082 0.967371 H -1.496564 2.104827 -0.679803 ```

__Frequencies__ (Top 10 out of 45)

``` 1. 130.0263 cm-1 (Symmetry: A) 2. 182.4703 cm-1 (Symmetry: A) 3. 220.2901 cm-1 (Symmetry: A) 4. 262.3870 cm-1 (Symmetry: A) 5. 278.8449 cm-1 (Symmetry: A) 6. 346.8667 cm-1 (Symmetry: A) 7. 457.6359 cm-1 (Symmetry: A) 8. 469.8397 cm-1 (Symmetry: A) 9. 513.8968 cm-1 (Symmetry: A) 10. 651.3520 cm-1 (Symmetry: A) ``` *** 8a_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C6H12OS2 | | Number of Basis Functions | 298 | | Electronic Energy (Eh) | -1107.36736296 | | Sum of electronic and zero-point Energies (Eh) | -1107.188603 | | Sum of electronic and thermal Energies (Eh) | -1107.179109 | | Sum of electronic and enthalpy Energies (Eh) | -1107.178165 | | Sum of electronic and thermal Free Energies (Eh) | -1107.223028 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 44 |

__Molecular Geometry in Cartesian Coordinates__

227

```xyz S -1.745677 0.286387 -0.287685 C -0.488794 1.571639 -0.650162 C 0.567126 1.767915 0.441575 C 1.989010 1.376833 0.022467 C 2.184478 -0.027304 -0.551229 C 1.944690 -1.196937 0.404453 C 0.497238 -1.492234 0.808457 S -0.685745 -1.544313 -0.585077 O -2.006330 0.418023 1.194177 H -1.114837 2.458567 -0.768090 H -0.075116 1.322380 -1.628866 H 0.258638 1.245595 1.349102 H 0.574851 2.825251 0.713928 H 2.644400 1.500948 0.890892 H 2.331869 2.095836 -0.728538 H 1.565883 -0.162391 -1.444409 H 3.219501 -0.097231 -0.897061 H 2.513336 -1.038655 1.328157 H 2.350907 -2.100061 -0.058852 H 0.087565 -0.787527 1.534532 H 0.433907 -2.479559 1.266623 ```

__Frequencies__ (Top 10 out of 57)

``` 1. 104.5385 cm-1 (Symmetry: A) 2. 139.1307 cm-1 (Symmetry: A) 3. 185.9257 cm-1 (Symmetry: A) 4. 193.8756 cm-1 (Symmetry: A) 5. 229.3645 cm-1 (Symmetry: A) 6. 288.6110 cm-1 (Symmetry: A) 7. 316.3881 cm-1 (Symmetry: A) 8. 339.7632 cm-1 (Symmetry: A) 9. 356.9698 cm-1 (Symmetry: A) 10. 415.7173 cm-1 (Symmetry: A) ``` *** 8b_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C6H12S2 | | Number of Basis Functions | 276 | | Electronic Energy (Eh) | -1032.18848046 |

228

| Sum of electronic and zero-point Energies (Eh) | -1032.013568 | | Sum of electronic and thermal Energies (Eh) | -1032.005061 | | Sum of electronic and enthalpy Energies (Eh) | -1032.004117 | | Sum of electronic and thermal Free Energies (Eh) | -1032.046681 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 40 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -0.961776 -1.445563 -0.394620 C -1.862865 -0.169977 0.565598 C -1.898460 1.221470 -0.072625 C -0.544181 1.754274 -0.544537 C 0.544313 1.754251 0.544531 C 1.898552 1.221338 0.072630 C 1.862847 -0.170104 -0.565598 S 0.961670 -1.445628 0.394621 H -1.424744 -0.163226 1.563502 H -2.876621 -0.563985 0.661398 H -2.330414 1.902479 0.669757 H -2.588105 1.203757 -0.921045 H -0.223796 1.151286 -1.397544 H -0.683814 2.765788 -0.933474 H 0.684018 2.765766 0.933440 H 0.223887 1.151310 1.397555 H 2.588190 1.203569 0.921054 H 2.330564 1.902312 -0.669750 H 2.876572 -0.564189 -0.661408 H 1.424717 -0.163316 -1.563498 ```

__Frequencies__ (Top 10 out of 54)

``` 1. 155.3572 cm-1 (Symmetry: A) 2. 175.0785 cm-1 (Symmetry: A) 3. 194.2543 cm-1 (Symmetry: A) 4. 196.6446 cm-1 (Symmetry: A) 5. 269.8353 cm-1 (Symmetry: A) 6. 288.4424 cm-1 (Symmetry: A) 7. 353.1223 cm-1 (Symmetry: A) 8. 354.8649 cm-1 (Symmetry: A) 9. 463.0481 cm-1 (Symmetry: A) 10. 481.1488 cm-1 (Symmetry: A) ``` *** 9a_Reactant

Run with Gaussian 16revisionA.03.

229

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C7H14OS2 | | Number of Basis Functions | 334 | | Electronic Energy (Eh) | -1146.66826856 | | Sum of electronic and zero-point Energies (Eh) | -1146.460733 | | Sum of electronic and thermal Energies (Eh) | -1146.450038 | | Sum of electronic and enthalpy Energies (Eh) | -1146.449094 | | Sum of electronic and thermal Free Energies (Eh) | -1146.496473 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 48 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 1.474502 -0.205271 -0.538855 C 1.028763 -1.721912 0.387142 C -0.446408 -1.822586 0.755779 C -1.445393 1.408886 -0.765261 C -0.300845 2.207932 -0.142227 S 0.851170 1.230244 0.903731 C -2.316215 0.638056 0.244160 C -1.411407 -1.693706 -0.441612 C -2.634893 -0.806512 -0.170379 O 2.984843 -0.229192 -0.585290 H 1.683899 -1.733976 1.260547 H 1.335420 -2.516943 -0.297577 H -0.665286 -1.069016 1.512752 H -0.583433 -2.789931 1.245187 H -2.057527 2.121528 -1.325352 H -1.027082 0.734496 -1.515207 H -0.680784 2.954389 0.559532 H 0.280299 2.725877 -0.904662 H -3.258952 1.170200 0.391868 H -1.823532 0.634661 1.219728 H -1.760231 -2.686786 -0.734160 H -0.876887 -1.312785 -1.316469 H -3.242870 -1.266114 0.615872 H -3.254139 -0.792578 -1.073370 ```

__Frequencies__ (Top 10 out of 66)

230

``` 1. 108.0713 cm-1 (Symmetry: A) 2. 140.2451 cm-1 (Symmetry: A) 3. 146.5121 cm-1 (Symmetry: A) 4. 167.7579 cm-1 (Symmetry: A) 5. 234.2927 cm-1 (Symmetry: A) 6. 257.7856 cm-1 (Symmetry: A) 7. 262.6891 cm-1 (Symmetry: A) 8. 285.9254 cm-1 (Symmetry: A) 9. 312.2245 cm-1 (Symmetry: A) 10. 333.7029 cm-1 (Symmetry: A) ``` *** 9b_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C7H14S2 | | Number of Basis Functions | 312 | | Electronic Energy (Eh) | -1071.49052658 | | Sum of electronic and zero-point Energies (Eh) | -1071.286773 | | Sum of electronic and thermal Energies (Eh) | -1071.277036 | | Sum of electronic and enthalpy Energies (Eh) | -1071.276092 | | Sum of electronic and thermal Free Energies (Eh) | -1071.32146 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 44 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -1.594583 0.781227 0.686665 C -0.627073 2.100196 -0.156323 C 0.743332 1.697573 -0.683038 C 0.744965 -1.696734 0.683195 C -0.625044 -2.100655 0.156454 S -1.593628 -0.782655 -0.686817 C 1.728436 -1.226154 -0.400266 C 1.727078 1.227412 0.400372 C 2.558914 0.001139 -0.000130 H -0.545538 2.863096 0.623671 H -1.248212 2.500953 -0.957245

231

H 1.150083 2.563888 -1.213247 H 0.597828 0.926528 -1.441790 H 1.152372 -2.562508 1.213783 H 0.598739 -0.925499 1.441625 H -1.245891 -2.501807 0.957405 H -0.542788 -2.863623 -0.623399 H 1.173814 -1.006492 -1.316303 H 2.408928 -2.042747 -0.655490 H 1.172521 1.006770 1.316220 H 2.406707 2.044570 0.656088 H 3.214870 -0.263816 0.836088 H 3.214296 0.266872 -0.836553 ```

__Frequencies__ (Top 10 out of 63)

``` 1. 122.7262 cm-1 (Symmetry: A) 2. 144.8838 cm-1 (Symmetry: A) 3. 145.4642 cm-1 (Symmetry: A) 4. 167.0791 cm-1 (Symmetry: A) 5. 249.8331 cm-1 (Symmetry: A) 6. 262.4500 cm-1 (Symmetry: A) 7. 286.3590 cm-1 (Symmetry: A) 8. 317.2203 cm-1 (Symmetry: A) 9. 355.2039 cm-1 (Symmetry: A) 10. 399.4279 cm-1 (Symmetry: A) ``` *** 10a_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C8H16OS2 | | Number of Basis Functions | 370 | | Electronic Energy (Eh) | -1185.97367104 | | Sum of electronic and zero-point Energies (Eh) | -1185.737491 | | Sum of electronic and thermal Energies (Eh) | -1185.725512 | | Sum of electronic and enthalpy Energies (Eh) | -1185.724568 | | Sum of electronic and thermal Free Energies (Eh) | -1185.774973 | | Number of Imaginary Frequencies | 0 |

232

| Mean of alpha and beta Electrons | 52 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -1.578471 0.167425 -0.550594 C -1.370536 1.749588 0.347958 C 0.066463 2.076976 0.729203 C 1.049279 2.091664 -0.458402 C 1.938293 -0.929647 -0.751133 C 1.459515 -2.271358 -0.171860 C -0.058303 -2.403345 -0.060728 S -0.876796 -1.129177 0.983916 C 2.411885 0.054232 0.328338 C 2.431537 1.521719 -0.113372 O -3.073667 0.024344 -0.696684 H -1.781112 2.470978 -0.363288 H -2.037594 1.690209 1.210391 H 0.391396 1.370874 1.493730 H 0.051492 3.057826 1.211414 H 1.163910 3.117039 -0.818440 H 0.632437 1.534307 -1.301870 H 2.757371 -1.098814 -1.455289 H 1.127071 -0.495018 -1.343453 H 1.912913 -2.436979 0.810303 H 1.787738 -3.099885 -0.806659 H -0.526636 -2.362270 -1.046508 H -0.335236 -3.355121 0.393439 H 1.781874 -0.057011 1.213852 H 3.418760 -0.237114 0.644022 H 2.876175 2.125754 0.684943 H 3.084240 1.629523 -0.986289 ```

__Frequencies__ (Top 10 out of 75)

``` 1. 77.3679 cm-1 (Symmetry: A) 2. 104.2078 cm-1 (Symmetry: A) 3. 129.5450 cm-1 (Symmetry: A) 4. 155.4657 cm-1 (Symmetry: A) 5. 195.8908 cm-1 (Symmetry: A) 6. 218.9618 cm-1 (Symmetry: A) 7. 245.1579 cm-1 (Symmetry: A) 8. 272.5575 cm-1 (Symmetry: A) 9. 281.6402 cm-1 (Symmetry: A) 10. 299.3065 cm-1 (Symmetry: A) ``` *** 10b_Reactant

Run with Gaussian 16revisionA.03.

233

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C8H16S2 | | Number of Basis Functions | 348 | | Electronic Energy (Eh) | -1110.79310363 | | Sum of electronic and zero-point Energies (Eh) | -1110.560655 | | Sum of electronic and thermal Energies (Eh) | -1110.549605 | | Sum of electronic and enthalpy Energies (Eh) | -1110.548661 | | Sum of electronic and thermal Free Energies (Eh) | -1110.597147 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 48 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 1.906687 -0.780468 0.689798 C 0.944689 -2.086856 -0.178015 C -0.428510 -1.670812 -0.678030 C -1.515799 -1.541530 0.393174 C -1.515858 1.541493 -0.393169 C -0.428582 1.670812 0.678043 C 0.944612 2.086878 0.178034 S 1.906621 0.780519 -0.689815 C -2.741168 0.760300 0.114576 C -2.741136 -0.760386 -0.114580 H 1.564675 -2.443657 -1.000093 H 0.881387 -2.883982 0.568534 H -0.754005 -2.416347 -1.411655 H -0.318468 -0.740341 -1.236200 H -1.824176 -2.550115 0.684908 H -1.114960 -1.080158 1.300097 H -1.824271 2.550067 -0.684902 H -1.114997 1.080138 -1.300090 H -0.318524 0.740349 1.236222 H -0.754100 2.416346 1.411658 H 1.564596 2.443670 1.000117 H 0.881298 2.884018 -0.568499 H -2.851484 0.956044 1.188286 H -3.644000 1.158110 -0.357607 H -3.643956 -1.158232 0.357594 H -2.851436 -0.956133 -1.188292 ```

234

__Frequencies__ (Top 10 out of 72)

``` 1. 89.1305 cm-1 (Symmetry: A) 2. 99.5591 cm-1 (Symmetry: A) 3. 126.2113 cm-1 (Symmetry: A) 4. 177.8482 cm-1 (Symmetry: A) 5. 208.2965 cm-1 (Symmetry: A) 6. 238.7385 cm-1 (Symmetry: A) 7. 261.7031 cm-1 (Symmetry: A) 8. 268.7878 cm-1 (Symmetry: A) 9. 290.9594 cm-1 (Symmetry: A) 10. 301.0285 cm-1 (Symmetry: A) ``` ***

3a_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C2H5OS3(1-) | | Number of Basis Functions | 191 | | Electronic Energy (Eh) | -1349.07448786 | | Sum of electronic and zero-point Energies (Eh) | -1349.003117 | | Sum of electronic and thermal Energies (Eh) | -1348.994473 | | Sum of electronic and enthalpy Energies (Eh) | -1348.993528 | | Sum of electronic and thermal Free Energies (Eh) | -1349.038314 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 37 |

__Molecular Geometry in Cartesian Coordinates__

235

```xyz S -2.402657 0.572171 0.098779 C -0.688139 0.588164 -0.453738 S 0.347313 -0.441353 0.665333 O -2.953669 -0.874662 -0.318591 S 2.145439 -0.616758 -0.363438 C 2.969735 0.948346 0.059368 H -0.335874 1.620183 -0.425104 H -0.608933 0.185781 -1.464773 H 2.398819 1.795090 -0.317348 H 3.098157 1.025951 1.137197 H 3.946076 0.926272 -0.425804 ```

__Frequencies__ (Top 10 out of 27)

``` 1. 48.6967 cm-1 (Symmetry: A) 2. 67.5728 cm-1 (Symmetry: A) 3. 113.9681 cm-1 (Symmetry: A) 4. 138.8533 cm-1 (Symmetry: A) 5. 183.0317 cm-1 (Symmetry: A) 6. 218.0819 cm-1 (Symmetry: A) 7. 261.1068 cm-1 (Symmetry: A) 8. 341.6703 cm-1 (Symmetry: A) 9. 498.3396 cm-1 (Symmetry: A) 10. 684.3605 cm-1 (Symmetry: A) ``` *** 3b_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C2H5S3(1-) | | Number of Basis Functions | 169 | | Electronic Energy (Eh) | -1273.89740381 | | Sum of electronic and zero-point Energies (Eh) | -1273.828595 | | Sum of electronic and thermal Energies (Eh) | -1273.821256 | | Sum of electronic and enthalpy Energies (Eh) | -1273.820312 | | Sum of electronic and thermal Free Energies (Eh) | -1273.861537 | | Number of Imaginary Frequencies | 0 |

236

| Mean of alpha and beta Electrons | 33 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 2.776835 0.228587 -0.065032 C 1.094066 0.342604 0.600225 S -0.003382 -0.672442 -0.469967 S -1.884408 -0.445625 0.398914 C -2.459285 1.106427 -0.356556 H 1.018416 -0.055266 1.611650 H 0.715768 1.364670 0.588397 H -3.455466 1.298885 0.042962 H -1.798533 1.929161 -0.089445 H -2.513587 1.000028 -1.438226 ```

__Frequencies__ (Top 10 out of 24)

``` 1. 73.8157 cm-1 (Symmetry: A) 2. 100.9252 cm-1 (Symmetry: A) 3. 136.0659 cm-1 (Symmetry: A) 4. 183.6063 cm-1 (Symmetry: A) 5. 224.0376 cm-1 (Symmetry: A) 6. 271.6557 cm-1 (Symmetry: A) 7. 490.0227 cm-1 (Symmetry: A) 8. 698.4814 cm-1 (Symmetry: A) 9. 716.6399 cm-1 (Symmetry: A) 10. 747.6444 cm-1 (Symmetry: A) ``` *** 4a_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C3H7OS3(1-) | | Number of Basis Functions | 227 | | Electronic Energy (Eh) | -1388.37856381 | | Sum of electronic and zero-point Energies (Eh) | -1388.278216 | | Sum of electronic and thermal Energies (Eh) | -1388.268412 | | Sum of electronic and enthalpy Energies (Eh) | -1388.267468 |

237

| Sum of electronic and thermal Free Energies (Eh) | -1388.315193 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 41 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -2.762062 -0.770304 -0.056688 C -1.145459 -0.171146 -0.633016 C -0.358190 0.470039 0.494914 S 1.205406 1.258352 -0.035971 O -3.518838 0.580591 0.408004 S 2.336258 -0.374040 -0.633962 C 3.028001 -0.936602 0.950308 H -0.615402 -1.039012 -1.034967 H -1.302168 0.550359 -1.440163 H -0.137063 -0.253799 1.281256 H -0.934814 1.288395 0.933314 H 3.653851 -1.800243 0.723231 H 3.634369 -0.151022 1.396556 H 2.232194 -1.237284 1.629458 ```

__Frequencies__ (Top 10 out of 36)

``` 1. 49.2465 cm-1 (Symmetry: A) 2. 63.2115 cm-1 (Symmetry: A) 3. 79.1492 cm-1 (Symmetry: A) 4. 117.9007 cm-1 (Symmetry: A) 5. 177.0455 cm-1 (Symmetry: A) 6. 188.4146 cm-1 (Symmetry: A) 7. 224.2425 cm-1 (Symmetry: A) 8. 263.9557 cm-1 (Symmetry: A) 9. 314.1532 cm-1 (Symmetry: A) 10. 374.6532 cm-1 (Symmetry: A) ``` *** 4b_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C3H7S3(1-) | | Number of Basis Functions | 205 |

238

| Electronic Energy (Eh) | -1313.19822735 | | Sum of electronic and zero-point Energies (Eh) | -1313.100563 | | Sum of electronic and thermal Energies (Eh) | -1313.092039 | | Sum of electronic and enthalpy Energies (Eh) | -1313.091095 | | Sum of electronic and thermal Free Energies (Eh) | -1313.13548 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 37 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -3.182433 -0.559828 0.055467 C -1.611713 0.201972 -0.515501 C -0.681317 0.499000 0.648440 S 0.918150 1.248545 0.147572 S 1.895130 -0.358003 -0.728584 C 2.635471 -1.164014 0.723109 H -1.816392 1.133470 -1.048452 H -1.103784 -0.469753 -1.211135 H -0.469532 -0.402867 1.225488 H -1.123493 1.236515 1.322863 H 3.322852 -0.483918 1.222119 H 1.860251 -1.495826 1.411641 H 3.181898 -2.030785 0.349920 ```

__Frequencies__ (Top 10 out of 33)

``` 1. 56.7985 cm-1 (Symmetry: A) 2. 86.7607 cm-1 (Symmetry: A) 3. 101.4925 cm-1 (Symmetry: A) 4. 173.1689 cm-1 (Symmetry: A) 5. 196.4587 cm-1 (Symmetry: A) 6. 241.3492 cm-1 (Symmetry: A) 7. 269.6668 cm-1 (Symmetry: A) 8. 333.4416 cm-1 (Symmetry: A) 9. 505.4995 cm-1 (Symmetry: A) 10. 689.7437 cm-1 (Symmetry: A) ``` *** 5a_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:|

239

| Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C4H9OS3(1-) | | Number of Basis Functions | 263 | | Electronic Energy (Eh) | -1427.67913295 | | Sum of electronic and zero-point Energies (Eh) | -1427.550319 | | Sum of electronic and thermal Energies (Eh) | -1427.538988 | | Sum of electronic and enthalpy Energies (Eh) | -1427.538044 | | Sum of electronic and thermal Free Energies (Eh) | -1427.592097 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 45 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -3.153489 -0.523330 0.400608 C -1.727293 0.501126 -0.085248 C -0.616785 0.401001 0.952761 C 0.589547 1.276183 0.634844 S 1.514071 0.783615 -0.866053 O -4.190738 -0.342419 -0.829429 S 2.365134 -1.011734 -0.276571 C 3.855783 -0.438866 0.592676 H -2.061001 1.539014 -0.186440 H -1.377799 0.156144 -1.061751 H -0.289418 -0.638172 1.057401 H -1.001194 0.710507 1.930758 H 1.293514 1.295251 1.467997 H 0.285024 2.306164 0.429339 H 4.491585 0.127528 -0.084678 H 3.586221 0.164131 1.457990 H 4.380021 -1.334712 0.926875 ```

__Frequencies__ (Top 10 out of 45)

``` 1. 5.4341 cm-1 (Symmetry: A) 2. 29.6709 cm-1 (Symmetry: A) 3. 56.0951 cm-1 (Symmetry: A) 4. 90.1877 cm-1 (Symmetry: A) 5. 116.4181 cm-1 (Symmetry: A) 6. 156.0177 cm-1 (Symmetry: A) 7. 179.0582 cm-1 (Symmetry: A) 8. 190.3076 cm-1 (Symmetry: A) 9. 256.9552 cm-1 (Symmetry: A)

240

10. 303.1657 cm-1 (Symmetry: A) ``` *** 5b_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C4H9S3(1-) | | Number of Basis Functions | 241 | | Electronic Energy (Eh) | -1352.49836192 | | Sum of electronic and zero-point Energies (Eh) | -1352.372247 | | Sum of electronic and thermal Energies (Eh) | -1352.362271 | | Sum of electronic and enthalpy Energies (Eh) | -1352.361327 | | Sum of electronic and thermal Free Energies (Eh) | -1352.411014 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 41 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -3.419128 -0.882703 -0.308123 C -2.707589 0.437746 0.756203 C -1.431924 1.083080 0.206197 C -0.272821 0.097047 0.137104 S 1.217298 0.968328 -0.467651 S 2.610279 -0.562961 -0.556562 C 3.203569 -0.624665 1.160961 H -3.451151 1.226159 0.893358 H -2.490986 0.030452 1.748761 H -1.152590 1.924073 0.852333 H -1.631085 1.486555 -0.791776 H -0.498065 -0.717778 -0.549099 H -0.058995 -0.312615 1.126896 H 3.962544 -1.406885 1.195902 H 2.390619 -0.880682 1.837958 H 3.647119 0.328850 1.440248 ```

__Frequencies__ (Top 10 out of 42)

```

241

1. 9.7218 cm-1 (Symmetry: A) 2. 56.2627 cm-1 (Symmetry: A) 3. 74.8331 cm-1 (Symmetry: A) 4. 121.1486 cm-1 (Symmetry: A) 5. 152.1403 cm-1 (Symmetry: A) 6. 181.5714 cm-1 (Symmetry: A) 7. 230.9692 cm-1 (Symmetry: A) 8. 255.6603 cm-1 (Symmetry: A) 9. 295.2209 cm-1 (Symmetry: A) 10. 416.2393 cm-1 (Symmetry: A) ``` *** 6a_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11OS3(1-) | | Number of Basis Functions | 299 | | Electronic Energy (Eh) | -1466.98296679 | | Sum of electronic and zero-point Energies (Eh) | -1466.825447 | | Sum of electronic and thermal Energies (Eh) | -1466.812842 | | Sum of electronic and enthalpy Energies (Eh) | -1466.811898 | | Sum of electronic and thermal Free Energies (Eh) | -1466.867597 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 49 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 2.017382 0.561516 -0.985460 S -3.773548 -0.104364 -0.537671 C -2.383130 -0.071894 0.639296 C -1.175273 0.635424 0.036147 C 0.009071 0.674416 0.999900 C 1.227019 1.396492 0.439358 O -4.946507 -0.873589 0.272590 S 2.829694 -1.126810 -0.099438 C 4.401879 -0.479768 0.544961 H -2.131058 -1.104144 0.902310 H -2.715950 0.439554 1.548072 H -1.449647 1.660819 -0.237206

242

H -0.877584 0.129063 -0.887350 H 0.294167 -0.342600 1.288304 H -0.287925 1.189138 1.920218 H 0.953683 2.380236 0.046904 H 1.990840 1.542810 1.204215 H 4.220871 0.296957 1.285849 H 4.908015 -1.320378 1.020692 H 5.012802 -0.096247 -0.269617 ```

__Frequencies__ (Top 10 out of 54)

``` 1. 24.3712 cm-1 (Symmetry: A) 2. 41.2416 cm-1 (Symmetry: A) 3. 49.0159 cm-1 (Symmetry: A) 4. 81.3338 cm-1 (Symmetry: A) 5. 96.1229 cm-1 (Symmetry: A) 6. 105.8812 cm-1 (Symmetry: A) 7. 152.3208 cm-1 (Symmetry: A) 8. 162.3434 cm-1 (Symmetry: A) 9. 193.5904 cm-1 (Symmetry: A) 10. 248.6267 cm-1 (Symmetry: A) ``` *** 6b_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11S3(1-) | | Number of Basis Functions | 277 | | Electronic Energy (Eh) | -1391.80218709 | | Sum of electronic and zero-point Energies (Eh) | -1391.647356 | | Sum of electronic and thermal Energies (Eh) | -1391.636267 | | Sum of electronic and enthalpy Energies (Eh) | -1391.635323 | | Sum of electronic and thermal Free Energies (Eh) | -1391.687008 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 45 |

__Molecular Geometry in Cartesian Coordinates__

243

```xyz S 4.070420 -0.550114 0.314734 S -1.712490 1.148013 -0.130010 C -0.463813 0.614001 1.100312 C 0.729446 -0.130329 0.518504 C 1.620642 0.731650 -0.368522 C 2.824476 -0.021847 -0.936303 S -2.539925 -0.659255 -0.717174 C -3.772756 -0.937959 0.589942 H -0.149391 1.549547 1.572481 H -0.980780 0.018215 1.854054 H 0.373628 -1.001433 -0.043906 H 1.323534 -0.517021 1.351477 H 1.972985 1.597720 0.204429 H 1.024222 1.121342 -1.202256 H 2.460530 -0.902516 -1.475853 H 3.317855 0.616541 -1.673045 H -4.272572 -1.876119 0.346535 H -4.498496 -0.127346 0.601627 H -3.287570 -1.030321 1.560052 ```

__Frequencies__ (Top 10 out of 51)

``` 1. 30.8690 cm-1 (Symmetry: A) 2. 34.5233 cm-1 (Symmetry: A) 3. 54.9304 cm-1 (Symmetry: A) 4. 102.2761 cm-1 (Symmetry: A) 5. 135.4149 cm-1 (Symmetry: A) 6. 163.3771 cm-1 (Symmetry: A) 7. 195.0853 cm-1 (Symmetry: A) 8. 247.5786 cm-1 (Symmetry: A) 9. 255.3627 cm-1 (Symmetry: A) 10. 301.9241 cm-1 (Symmetry: A) ``` *** 7a_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C6H13OS3(1-) | | Number of Basis Functions | 335 | | Electronic Energy (Eh) | -1506.28743208 |

244

| Sum of electronic and zero-point Energies (Eh) | -1506.101034 | | Sum of electronic and thermal Energies (Eh) | -1506.087302 | | Sum of electronic and enthalpy Energies (Eh) | -1506.086357 | | Sum of electronic and thermal Free Energies (Eh) | -1506.144352 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 53 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -4.117024 0.291203 -0.398047 C -2.600143 -0.250003 0.453758 C -1.483416 -0.540500 -0.541785 C -0.198792 -1.005281 0.143459 C 0.918227 -1.288598 -0.856596 C 2.236516 -1.715260 -0.225608 S 2.997242 -0.483643 0.896830 O -5.161537 0.563780 0.810563 S 3.289299 1.151560 -0.342218 C 1.769764 2.119575 -0.083144 H -2.297193 0.540208 1.148122 H -2.834712 -1.145183 1.038202 H -1.814177 -1.306191 -1.252639 H -1.274537 0.362037 -1.128701 H -0.400313 -1.911186 0.726822 H 0.126873 -0.242505 0.858502 H 1.092540 -0.412049 -1.490572 H 0.604412 -2.091047 -1.533340 H 2.972998 -1.966312 -0.989112 H 2.100630 -2.589316 0.418278 H 0.902503 1.612402 -0.502014 H 1.923531 3.066074 -0.602593 H 1.624553 2.307310 0.978982 ```

__Frequencies__ (Top 10 out of 63)

``` 1. 20.5013 cm-1 (Symmetry: A) 2. 49.7737 cm-1 (Symmetry: A) 3. 58.2384 cm-1 (Symmetry: A) 4. 78.7160 cm-1 (Symmetry: A) 5. 100.7071 cm-1 (Symmetry: A) 6. 112.8763 cm-1 (Symmetry: A) 7. 124.6027 cm-1 (Symmetry: A) 8. 159.7509 cm-1 (Symmetry: A) 9. 183.0915 cm-1 (Symmetry: A) 10. 207.3232 cm-1 (Symmetry: A) ``` *** 7b_Product

245

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C6H13S3(1-) | | Number of Basis Functions | 313 | | Electronic Energy (Eh) | -1431.10553117 | | Sum of electronic and zero-point Energies (Eh) | -1430.922107 | | Sum of electronic and thermal Energies (Eh) | -1430.909629 | | Sum of electronic and enthalpy Energies (Eh) | -1430.908685 | | Sum of electronic and thermal Free Energies (Eh) | -1430.963696 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 49 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -2.548460 -1.224760 -0.276514 C -0.909065 -0.888626 0.458475 C 0.077461 -0.231436 -0.494190 C 1.419198 0.048287 0.182030 C 2.422997 0.704977 -0.761130 C 3.764368 1.037403 -0.106089 S 4.763790 -0.425831 0.398313 S -3.330974 0.683398 -0.486922 C -4.027619 0.972804 1.167133 H -0.555294 -1.874539 0.772795 H -1.056599 -0.286213 1.357648 H 0.227990 -0.882434 -1.361560 H -0.351204 0.705452 -0.864900 H 1.260874 0.702595 1.048815 H 1.842861 -0.885093 0.564903 H 1.982844 1.633087 -1.150106 H 2.595463 0.050296 -1.623302 H 3.577251 1.665468 0.771264 H 4.353103 1.638315 -0.803551 H -4.790670 0.229726 1.389502 H -3.242861 0.952162 1.921284 H -4.477505 1.965809 1.141803 ```

__Frequencies__ (Top 10 out of 60)

246

``` 1. 31.6487 cm-1 (Symmetry: A) 2. 33.9339 cm-1 (Symmetry: A) 3. 57.4893 cm-1 (Symmetry: A) 4. 91.0928 cm-1 (Symmetry: A) 5. 116.0681 cm-1 (Symmetry: A) 6. 118.9348 cm-1 (Symmetry: A) 7. 166.3499 cm-1 (Symmetry: A) 8. 171.9884 cm-1 (Symmetry: A) 9. 204.6876 cm-1 (Symmetry: A) 10. 254.6644 cm-1 (Symmetry: A) ``` *** 8a_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C7H15OS3(1-) | | Number of Basis Functions | 371 | | Electronic Energy (Eh) | -1545.59017696 | | Sum of electronic and zero-point Energies (Eh) | -1545.375524 | | Sum of electronic and thermal Energies (Eh) | -1545.360211 | | Sum of electronic and enthalpy Energies (Eh) | -1545.359267 | | Sum of electronic and thermal Free Energies (Eh) | -1545.421977 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 57 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -5.161393 -0.211100 -0.590520 C -3.712330 -0.202843 0.514173 C -2.469318 0.305973 -0.206677 C -1.241037 0.339071 0.702533 C 0.008354 0.852956 -0.008629 C 1.224769 0.902809 0.911328 C 2.483220 1.435718 0.240060 S 3.156800 0.357516 -1.077634 O -6.360353 -0.763904 0.349498 S 3.853491 -1.264706 0.007820

247

C 5.493749 -0.678443 0.529137 H -3.553155 -1.222349 0.879666 H -3.942149 0.434339 1.374148 H -2.658191 1.313650 -0.594133 H -2.260783 -0.331347 -1.073515 H -1.451661 0.973434 1.571540 H -1.049025 -0.668226 1.090332 H -0.185762 1.857635 -0.403486 H 0.225600 0.213357 -0.870487 H 1.429138 -0.092426 1.320443 H 1.005090 1.551421 1.766872 H 3.279798 1.610506 0.964495 H 2.284262 2.381186 -0.272840 H 5.942050 -1.490434 1.102572 H 6.109086 -0.459808 -0.341142 H 5.401707 0.201491 1.163342 ```

__Frequencies__ (Top 10 out of 72)

``` 1. 19.3105 cm-1 (Symmetry: A) 2. 33.6616 cm-1 (Symmetry: A) 3. 45.1472 cm-1 (Symmetry: A) 4. 62.8575 cm-1 (Symmetry: A) 5. 66.6797 cm-1 (Symmetry: A) 6. 82.5897 cm-1 (Symmetry: A) 7. 108.0032 cm-1 (Symmetry: A) 8. 129.4748 cm-1 (Symmetry: A) 9. 143.1579 cm-1 (Symmetry: A) 10. 162.0832 cm-1 (Symmetry: A) ``` *** 8b_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C7H15S3(1-) | | Number of Basis Functions | 349 | | Electronic Energy (Eh) | -1470.4088502 | | Sum of electronic and zero-point Energies (Eh) | -1470.196886 | | Sum of electronic and thermal Energies (Eh) | -1470.183047 | | Sum of electronic and enthalpy Energies (Eh) | -1470.182103 |

248

| Sum of electronic and thermal Free Energies (Eh) | -1470.240784 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 53 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 4.816171 1.423425 -0.215256 C 4.609658 -0.399561 -0.038440 C 3.308874 -0.836928 0.637134 C 2.055953 -0.561749 -0.189002 C 0.775206 -1.023713 0.502638 C -0.475671 -0.750004 -0.331945 C -1.740070 -1.216550 0.372471 S -3.277800 -0.895743 -0.561640 S -3.431419 1.168694 -0.452273 C -4.252983 1.385546 1.154866 H 5.451865 -0.790979 0.537502 H 4.661394 -0.864784 -1.028456 H 3.222727 -0.336844 1.609119 H 3.364320 -1.915245 0.839524 H 2.147672 -1.068719 -1.158267 H 1.991236 0.509887 -0.400432 H 0.841291 -2.096960 0.718460 H 0.683655 -0.516421 1.470533 H -0.551232 0.323230 -0.534174 H -0.394328 -1.256089 -1.299262 H -1.839037 -0.758490 1.359134 H -1.737337 -2.301777 0.507940 H -4.387355 2.459349 1.288961 H -5.222631 0.891951 1.151544 H -3.629271 0.997624 1.958253 ```

__Frequencies__ (Top 10 out of 69)

``` 1. 19.7217 cm-1 (Symmetry: A) 2. 39.9138 cm-1 (Symmetry: A) 3. 52.8828 cm-1 (Symmetry: A) 4. 79.7781 cm-1 (Symmetry: A) 5. 90.6647 cm-1 (Symmetry: A) 6. 106.8085 cm-1 (Symmetry: A) 7. 136.5199 cm-1 (Symmetry: A) 8. 157.2468 cm-1 (Symmetry: A) 9. 173.5441 cm-1 (Symmetry: A) 10. 191.1347 cm-1 (Symmetry: A) ``` *** 9a_Product

Run with Gaussian 16revisionA.03.

249

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C8H17OS3(1-) | | Number of Basis Functions | 407 | | Electronic Energy (Eh) | -1584.8937175 | | Sum of electronic and zero-point Energies (Eh) | -1584.650517 | | Sum of electronic and thermal Energies (Eh) | -1584.633767 | | Sum of electronic and enthalpy Energies (Eh) | -1584.632823 | | Sum of electronic and thermal Free Energies (Eh) | -1584.699022 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 61 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 6.015254 0.501134 -0.518010 C 4.600536 -0.071001 0.477596 C 3.303171 0.017353 -0.318010 C -1.735735 -0.654470 -0.303944 C -2.947924 -1.066931 0.516283 S -4.531123 -0.957980 -0.389847 C -0.443160 -0.740589 0.506833 C 2.081759 -0.408242 0.496392 C 0.783905 -0.330477 -0.304043 O 7.286894 0.323843 0.470828 S -4.788374 1.091262 -0.568639 C -5.570699 1.496513 1.021584 H 4.792047 -1.103527 0.786667 H 4.540691 0.548387 1.378234 H 3.379217 -0.612697 -1.211450 H 3.159795 1.045718 -0.669334 H -1.874788 0.371233 -0.660410 H -1.662912 -1.296755 -1.187345 H -3.035045 -0.470568 1.427220 H -2.884861 -2.118268 0.810745 H -0.309865 -1.764635 0.875018 H -0.524822 -0.097933 1.391258 H 1.998696 0.228282 1.385109 H 2.224553 -1.433360 0.857988 H 0.864737 -0.973136 -1.188741 H 0.647245 0.692813 -0.673638 H -4.904700 1.254101 1.847808 H -5.753930 2.571401 1.009127

250

H -6.514448 0.964621 1.122904 ```

__Frequencies__ (Top 10 out of 81)

``` 1. 20.4944 cm-1 (Symmetry: A) 2. 36.4071 cm-1 (Symmetry: A) 3. 41.9245 cm-1 (Symmetry: A) 4. 47.3515 cm-1 (Symmetry: A) 5. 69.6060 cm-1 (Symmetry: A) 6. 74.9370 cm-1 (Symmetry: A) 7. 100.2665 cm-1 (Symmetry: A) 8. 113.3858 cm-1 (Symmetry: A) 9. 127.0768 cm-1 (Symmetry: A) 10. 147.9526 cm-1 (Symmetry: A) ``` *** 9b_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C8H17S3(1-) | | Number of Basis Functions | 385 | | Electronic Energy (Eh) | -1509.71233777 | | Sum of electronic and zero-point Energies (Eh) | -1509.471795 | | Sum of electronic and thermal Energies (Eh) | -1509.456598 | | Sum of electronic and enthalpy Energies (Eh) | -1509.455654 | | Sum of electronic and thermal Free Energies (Eh) | -1509.517747 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 57 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 3.813256 -1.136263 -0.358948 C 2.172463 -1.063035 0.442404 C 1.082554 -0.459100 -0.429354 C -3.822777 0.869689 -0.698114 C -5.172103 0.980354 0.013596 S -5.975434 -0.631664 0.404416

251

C -2.712107 0.275084 0.162538 C -0.258593 -0.393167 0.300505 C -1.370209 0.200570 -0.562037 S 4.352620 0.864246 -0.426811 C 5.068906 1.101612 1.227187 H 2.277620 -0.519212 1.383800 H 1.950777 -2.106425 0.683591 H 0.979617 -1.054825 -1.341948 H 1.382267 0.549012 -0.733421 H -3.514907 1.872642 -1.023982 H -3.944320 0.265693 -1.605164 H -5.851409 1.564674 -0.611919 H -5.032107 1.545932 0.940854 H -2.600306 0.880864 1.071318 H -3.008283 -0.726711 0.488436 H -0.147184 0.206569 1.211590 H -0.549078 -1.400014 0.623027 H -1.074524 1.204395 -0.890904 H -1.482802 -0.402719 -1.471107 H 5.397183 2.140448 1.273727 H 4.319625 0.925853 1.997054 H 5.921960 0.440682 1.366193 ```

__Frequencies__ (Top 10 out of 78)

``` 1. 22.4756 cm-1 (Symmetry: A) 2. 29.1806 cm-1 (Symmetry: A) 3. 49.5016 cm-1 (Symmetry: A) 4. 64.2626 cm-1 (Symmetry: A) 5. 78.7083 cm-1 (Symmetry: A) 6. 83.3491 cm-1 (Symmetry: A) 7. 130.0298 cm-1 (Symmetry: A) 8. 140.3116 cm-1 (Symmetry: A) 9. 144.6659 cm-1 (Symmetry: A) 10. 176.0329 cm-1 (Symmetry: A) ``` *** 10a_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C9H19OS3(1-) | | Number of Basis Functions | 443 | | Electronic Energy (Eh) | -1624.19685346 |

252

| Sum of electronic and zero-point Energies (Eh) | -1623.925064 | | Sum of electronic and thermal Energies (Eh) | -1623.90717 | | Sum of electronic and enthalpy Energies (Eh) | -1623.906226 | | Sum of electronic and thermal Free Energies (Eh) | -1623.975604 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 65 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -5.547008 1.160959 0.015212 C -5.297416 -0.629611 -0.234263 C -3.902690 -0.951828 -0.763053 C -2.771455 -0.529959 0.174797 C 1.097556 -1.106477 0.167643 C 2.228928 -0.688137 1.102158 C 3.596806 -1.209208 0.682842 S 4.233158 -0.499931 -0.880547 C -0.265508 -0.597054 0.629506 C -1.404851 -1.023626 -0.292792 O -7.136963 1.275584 0.310617 S 4.647393 1.453921 -0.327038 C 6.302743 1.280901 0.405043 H -5.472845 -1.147432 0.714977 H -6.062218 -0.956244 -0.944321 H -3.838844 -2.033195 -0.930607 H -3.762297 -0.474623 -1.739711 H -2.751005 0.561784 0.263202 H -2.975830 -0.918555 1.180495 H 1.300774 -0.736310 -0.842818 H 1.072307 -2.201087 0.101551 H 2.024363 -1.066322 2.110179 H 2.268543 0.403561 1.181960 H 3.563809 -2.285019 0.487628 H 4.346929 -1.033122 1.454951 H -0.465762 -0.962963 1.643856 H -0.240868 0.497447 0.693572 H -1.421320 -2.118040 -0.364854 H -1.210833 -0.649159 -1.305412 H 7.002367 0.896715 -0.334398 H 6.269505 0.628051 1.275534 H 6.607562 2.280646 0.715978 ```

__Frequencies__ (Top 10 out of 90)

``` 1. 17.0446 cm-1 (Symmetry: A) 2. 22.9816 cm-1 (Symmetry: A) 3. 38.5886 cm-1 (Symmetry: A) 4. 47.8199 cm-1 (Symmetry: A)

253

5. 61.9716 cm-1 (Symmetry: A) 6. 75.2526 cm-1 (Symmetry: A) 7. 81.4318 cm-1 (Symmetry: A) 8. 88.8408 cm-1 (Symmetry: A) 9. 117.1148 cm-1 (Symmetry: A) 10. 143.2411 cm-1 (Symmetry: A) ``` *** 10b_Product

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C9H19S3(1-) | | Number of Basis Functions | 421 | | Electronic Energy (Eh) | -1549.0158641 | | Sum of electronic and zero-point Energies (Eh) | -1548.746522 | | Sum of electronic and thermal Energies (Eh) | -1548.730191 | | Sum of electronic and enthalpy Energies (Eh) | -1548.729247 | | Sum of electronic and thermal Free Energies (Eh) | -1548.794938 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 61 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -2.671946 -1.120379 0.245807 C -2.425982 0.309409 1.362054 C -2.065201 1.620194 0.671338 C -0.859151 1.548422 -0.264985 C 2.905103 0.562580 0.135922 C 4.124999 0.664738 -0.775029 C 5.432006 0.201905 -0.129590 S 5.526874 -1.603307 0.230176 C 1.631660 1.100616 -0.512456 C 0.407251 1.007084 0.395273 S -4.195452 -0.470069 -0.999145 C -5.666394 -0.770018 0.026895 H -1.625166 -0.029431 2.024239 H -3.325232 0.427443 1.968130 H -1.870191 2.352238 1.463344 H -2.937253 1.981429 0.119874

254

H -0.661700 2.555404 -0.647325 H -1.105217 0.928907 -1.134255 H 3.102494 1.119509 1.061430 H 2.756676 -0.482177 0.426022 H 3.948648 0.083391 -1.687854 H 4.242002 1.712308 -1.084896 H 5.577889 0.761475 0.800453 H 6.261091 0.468572 -0.789829 H 1.435099 0.547488 -1.439489 H 1.784779 2.146987 -0.804240 H 0.259660 -0.040913 0.681290 H 0.598334 1.560013 1.324085 H -5.634618 -0.159769 0.927793 H -5.738488 -1.825404 0.281777 H -6.526161 -0.476956 -0.576500 ```

__Frequencies__ (Top 10 out of 87)

``` 1. 16.4352 cm-1 (Symmetry: A) 2. 19.5902 cm-1 (Symmetry: A) 3. 29.1608 cm-1 (Symmetry: A) 4. 51.3798 cm-1 (Symmetry: A) 5. 65.6468 cm-1 (Symmetry: A) 6. 81.6052 cm-1 (Symmetry: A) 7. 105.3403 cm-1 (Symmetry: A) 8. 129.4932 cm-1 (Symmetry: A) 9. 147.7672 cm-1 (Symmetry: A) 10. 162.8538 cm-1 (Symmetry: A) ``` ***

255

TS–3a (Constrained Optimization, M06-2X/6-311++G(d,p))

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C2H5OS3(1-) | | Number of Basis Functions | 191 | | Electronic Energy (Eh) | -1349.04764193 | | Sum of electronic and zero-point Energies (Eh) | -1348.977288 | | Sum of electronic and thermal Energies (Eh) | -1348.96886 | | Sum of electronic and enthalpy Energies (Eh) | -1348.967916 | | Sum of electronic and thermal Free Energies (Eh) | -1349.012752 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 37 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -2.233212 -0.269267 0.091731 C -0.727342 0.035259 1.055998 S -0.074712 -0.657033 -0.447712 O -2.732963 1.070448 -0.446820 H -0.574454 1.101371 1.197464 H -0.617797 -0.566066 1.953184 C 2.532822 1.343897 -0.056470 H 1.867179 1.741925 0.711827 H 3.526733 1.763576 0.105091 S 2.598967 -0.483469 0.021029 H 2.172456 1.676981 -1.030952 ```

256

__Frequencies__ (Top 10 out of 27)

``` 1. -147.2093 cm-1 (Symmetry: A) * 2. 47.6565 cm-1 (Symmetry: A) 3. 71.1399 cm-1 (Symmetry: A) 4. 86.8871 cm-1 (Symmetry: A) 5. 111.6082 cm-1 (Symmetry: A) 6. 155.3225 cm-1 (Symmetry: A) 7. 219.3794 cm-1 (Symmetry: A) 8. 317.0194 cm-1 (Symmetry: A) 9. 361.6671 cm-1 (Symmetry: A) 10. 629.6306 cm-1 (Symmetry: A) ``` ***

TS–3b (Constrained Optimization, M06-2X/6-311++G(d,p))

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C2H5S3(1-) | | Number of Basis Functions | 169 | | Electronic Energy (Eh) | -1273.86401613 | | Sum of electronic and zero-point Energies (Eh) | -1273.79692 | | Sum of electronic and thermal Energies (Eh) | -1273.789451 | | Sum of electronic and enthalpy Energies (Eh) | -1273.788507 | | Sum of electronic and thermal Free Energies (Eh) | -1273.831214 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 33 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 2.613341 0.151315 -0.142204 C 1.154511 0.523109 0.858310 S 0.484670 -0.563266 -0.375126 H 0.812841 1.550923 0.798490 H 1.173426 0.116306 1.863509 C -2.504461 1.100174 -0.408638 H -3.526893 1.427621 -0.213212 H -1.822710 1.822206 0.044001

257

H -2.342821 1.104269 -1.487430 S -2.235144 -0.573113 0.285869 ```

__Frequencies__ (Top 10 out of 24)

``` 1. -181.9612 cm-1 (Symmetry: A) * 2. 42.5536 cm-1 (Symmetry: A) 3. 77.5611 cm-1 (Symmetry: A) 4. 83.4860 cm-1 (Symmetry: A) 5. 124.8252 cm-1 (Symmetry: A) 6. 173.1423 cm-1 (Symmetry: A) 7. 286.2374 cm-1 (Symmetry: A) 8. 663.5008 cm-1 (Symmetry: A) 9. 725.6141 cm-1 (Symmetry: A) 10. 873.8017 cm-1 (Symmetry: A) ``` ***

TS–4a (Constrained Optimization, M06-2X/6-311++G(d,p))

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C3H7OS3(1-) | | Number of Basis Functions | 227 | | Electronic Energy (Eh) | -1388.35318226 | | Sum of electronic and zero-point Energies (Eh) | -1388.25345 | | Sum of electronic and thermal Energies (Eh) | -1388.244127 | | Sum of electronic and enthalpy Energies (Eh) | -1388.243183 | | Sum of electronic and thermal Free Energies (Eh) | -1388.289539 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 41 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 0.049285 -0.630670 -0.340675 C -0.140262 0.841843 0.751987 C -1.424040 1.424638 0.177609 S -2.196658 -0.072475 -0.529966

258

H 0.724369 1.495967 0.671479 H -0.253016 0.510175 1.782613 H -1.224468 2.102574 -0.653192 H -2.102574 1.891839 0.890851 C 3.093955 0.887084 -0.453146 H 3.104406 1.700231 0.274701 H 4.087732 0.811950 -0.896223 S 2.623086 -0.699696 0.327445 H 2.385324 1.139813 -1.245759 O -2.938889 -0.766058 0.625996 ```

__Frequencies__ (Top 10 out of 36)

``` 1. -126.9990 cm-1 (Symmetry: A) * 2. 58.9859 cm-1 (Symmetry: A) 3. 74.1594 cm-1 (Symmetry: A) 4. 106.3819 cm-1 (Symmetry: A) 5. 131.7314 cm-1 (Symmetry: A) 6. 151.6912 cm-1 (Symmetry: A) 7. 164.1087 cm-1 (Symmetry: A) 8. 243.1360 cm-1 (Symmetry: A) 9. 299.3939 cm-1 (Symmetry: A) 10. 336.0194 cm-1 (Symmetry: A) ``` ***

TS–4b (Constrained Optimization, M06-2X/6-311++G(d,p))

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C3H7S3(1-) | | Number of Basis Functions | 205 | | Electronic Energy (Eh) | -1313.16651405 | | Sum of electronic and zero-point Energies (Eh) | -1313.070113 | | Sum of electronic and thermal Energies (Eh) | -1313.061829 | | Sum of electronic and enthalpy Energies (Eh) | -1313.060884 | | Sum of electronic and thermal Free Energies (Eh) | -1313.104882 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 37 |

259

__Molecular Geometry in Cartesian Coordinates__

```xyz S 2.513636 -0.412027 0.394544 C 0.475103 1.039683 -0.241062 S 0.197187 -0.767852 -0.012663 H -0.175226 1.425994 -1.024159 H 0.262768 1.544127 0.702546 C -2.699454 0.698105 0.909298 H -2.296620 1.695723 0.727582 H -3.773350 0.785851 1.080291 H -2.236982 0.295082 1.812019 S -2.387289 -0.405640 -0.513905 C 1.963531 1.056463 -0.555485 H 2.137215 0.887008 -1.618769 H 2.470563 1.969015 -0.243633 ```

__Frequencies__ (Top 10 out of 33)

``` 1. -185.3353 cm-1 (Symmetry: A) * 2. 48.7141 cm-1 (Symmetry: A) 3. 81.5742 cm-1 (Symmetry: A) 4. 129.8703 cm-1 (Symmetry: A) 5. 154.4063 cm-1 (Symmetry: A) 6. 174.9251 cm-1 (Symmetry: A) 7. 213.5210 cm-1 (Symmetry: A) 8. 250.8046 cm-1 (Symmetry: A) 9. 446.9031 cm-1 (Symmetry: A) 10. 709.3480 cm-1 (Symmetry: A) ``` ***

TS–5a

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C4H9OS3(1-) | | Number of Basis Functions | 263 | | Electronic Energy (Eh) | -1427.66986652 | | Sum of electronic and zero-point Energies (Eh) | -1427.541148 | | Sum of electronic and thermal Energies (Eh) | -1427.530778 |

260

| Sum of electronic and enthalpy Energies (Eh) | -1427.529834 | | Sum of electronic and thermal Free Energies (Eh) | -1427.578566 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 45 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -2.046690 -0.769728 -0.378073 C -2.315783 1.021487 -0.269595 C -1.221953 1.597289 0.612862 C 0.142014 1.342978 -0.009621 S 0.381469 -0.425487 -0.517106 O -2.356247 -1.308806 1.054149 H -2.255634 1.412204 -1.289553 H -3.319739 1.189081 0.123654 H -1.279858 1.131667 1.600189 H -1.356641 2.675244 0.742413 H 0.935744 1.573832 0.697103 H 0.282866 1.955631 -0.901379 S 2.807678 -0.086014 -0.572654 C 3.083226 -0.290493 1.217930 H 2.746317 -1.274136 1.546872 H 4.147014 -0.192939 1.437496 H 2.545554 0.471962 1.785880 ```

__Frequencies__ (Top 10 out of 45)

``` 1. -185.4293 cm-1 (Symmetry: A) * 2. 52.3649 cm-1 (Symmetry: A) 3. 60.6178 cm-1 (Symmetry: A) 4. 105.4183 cm-1 (Symmetry: A) 5. 124.2395 cm-1 (Symmetry: A) 6. 148.4073 cm-1 (Symmetry: A) 7. 187.0192 cm-1 (Symmetry: A) 8. 198.6764 cm-1 (Symmetry: A) 9. 247.0361 cm-1 (Symmetry: A) 10. 294.9943 cm-1 (Symmetry: A) ``` ***

TS–5b

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 |

261

| Multiplicity | 1 | | Stoichiometry | C4H9S3(1-) | | Number of Basis Functions | 241 | | Electronic Energy (Eh) | -1352.4839797 | | Sum of electronic and zero-point Energies (Eh) | -1352.358616 | | Sum of electronic and thermal Energies (Eh) | -1352.34925 | | Sum of electronic and enthalpy Energies (Eh) | -1352.348306 | | Sum of electronic and thermal Free Energies (Eh) | -1352.394952 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 41 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 0.123220 -0.771624 -0.062631 C -0.274788 0.864241 -0.803026 C -1.346553 1.573077 0.010875 C -2.632815 0.757008 -0.000437 S -2.212466 -0.997962 0.321154 H 0.654941 1.427326 -0.819830 H -0.612320 0.707091 -1.830077 H -1.528846 2.566103 -0.412885 H -0.992386 1.703584 1.037324 H -3.117585 0.831349 -0.975369 H -3.331111 1.106225 0.760791 S 2.635346 -0.413299 -0.424423 C 2.842362 0.743074 0.974830 H 3.878161 1.080038 1.036087 H 2.204684 1.621825 0.849298 H 2.577619 0.258231 1.915617 ```

__Frequencies__ (Top 10 out of 42)

``` 1. -172.9265 cm-1 (Symmetry: A) * 2. 46.1997 cm-1 (Symmetry: A) 3. 64.6655 cm-1 (Symmetry: A) 4. 107.6796 cm-1 (Symmetry: A) 5. 115.5643 cm-1 (Symmetry: A) 6. 166.1406 cm-1 (Symmetry: A) 7. 192.6900 cm-1 (Symmetry: A) 8. 247.6802 cm-1 (Symmetry: A) 9. 309.3249 cm-1 (Symmetry: A) 10. 352.7081 cm-1 (Symmetry: A) ``` ***

262

TS–6a

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11OS3(1-) | | Number of Basis Functions | 299 | | Electronic Energy (Eh) | -1466.97427923 | | Sum of electronic and zero-point Energies (Eh) | -1466.816172 | | Sum of electronic and thermal Energies (Eh) | -1466.805003 | | Sum of electronic and enthalpy Energies (Eh) | -1466.804059 | | Sum of electronic and thermal Free Energies (Eh) | -1466.853808 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 49 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 0.534432 -0.878094 -0.141467 S -1.923343 -0.985448 -0.444601 C -2.118866 0.736999 -1.016951 C -1.756957 1.759956 0.052765 C -0.284966 1.758545 0.468196 C 0.183937 0.442326 1.088645 O -2.628538 -1.029629 0.953155 H -3.159319 0.858615 -1.327611 H -1.479445 0.824171 -1.900042 H -2.026604 2.752435 -0.320111 H -2.378432 1.570826 0.933350 H -0.140362 2.548428 1.213165 H 0.348767 2.007632 -0.388689 H 1.110864 0.574437 1.643699 H -0.579122 0.051293 1.766799 S 2.933971 -0.708560 0.098110 C 3.146905 0.995800 -0.512708 H 4.209701 1.202932 -0.642870 H 2.643215 1.121360 -1.472852 H 2.737761 1.716794 0.197581 ```

__Frequencies__ (Top 10 out of 54)

263

``` 1. -172.9773 cm-1 (Symmetry: A) * 2. 64.7493 cm-1 (Symmetry: A) 3. 76.2458 cm-1 (Symmetry: A) 4. 128.1021 cm-1 (Symmetry: A) 5. 147.3807 cm-1 (Symmetry: A) 6. 166.8054 cm-1 (Symmetry: A) 7. 178.3716 cm-1 (Symmetry: A) 8. 196.9073 cm-1 (Symmetry: A) 9. 238.3199 cm-1 (Symmetry: A) 10. 259.9207 cm-1 (Symmetry: A) ``` ***

TS–6b

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C5H11S3(1-) | | Number of Basis Functions | 277 | | Electronic Energy (Eh) | -1391.78841963 | | Sum of electronic and zero-point Energies (Eh) | -1391.633823 | | Sum of electronic and thermal Energies (Eh) | -1391.623683 | | Sum of electronic and enthalpy Energies (Eh) | -1391.622739 | | Sum of electronic and thermal Free Energies (Eh) | -1391.670279 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 45 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 2.066795 -1.308671 -0.065290 S -0.374716 -0.936611 0.084899 C -0.018530 0.538843 1.133142 C 0.649897 1.692423 0.387055 C 2.133161 1.495040 0.068634 C 2.431695 0.306947 -0.840958 H 0.605714 0.202529 1.961886 H -0.983768 0.852087 1.527985 H 0.097568 1.887722 -0.538069

264

H 0.548186 2.587782 1.010606 H 2.506003 2.406423 -0.412396 H 2.691903 1.376518 1.003491 H 1.844984 0.389834 -1.761275 H 3.487044 0.311007 -1.121621 S -2.753877 -0.536767 0.095535 C -2.773263 1.036315 -0.825050 H -3.799420 1.283732 -1.099448 H -2.371387 1.848230 -0.215842 H -2.175820 0.949512 -1.734569 ```

__Frequencies__ (Top 10 out of 51)

``` 1. -198.8287 cm-1 (Symmetry: A) * 2. 67.6196 cm-1 (Symmetry: A) 3. 74.3616 cm-1 (Symmetry: A) 4. 130.4167 cm-1 (Symmetry: A) 5. 161.8667 cm-1 (Symmetry: A) 6. 176.8240 cm-1 (Symmetry: A) 7. 210.2201 cm-1 (Symmetry: A) 8. 231.6656 cm-1 (Symmetry: A) 9. 257.4465 cm-1 (Symmetry: A) 10. 289.1762 cm-1 (Symmetry: A) ``` ***

TS–7a

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C6H13OS3(1-) | | Number of Basis Functions | 335 | | Electronic Energy (Eh) | -1506.2765748 | | Sum of electronic and zero-point Energies (Eh) | -1506.08957 | | Sum of electronic and thermal Energies (Eh) | -1506.077191 | | Sum of electronic and enthalpy Energies (Eh) | -1506.076247 | | Sum of electronic and thermal Free Energies (Eh) | -1506.128855 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 53 |

265

__Molecular Geometry in Cartesian Coordinates__

```xyz S -1.859668 -1.154428 -0.502262 C -2.223929 0.524592 -1.113304 C -1.927376 1.673833 -0.149986 C -0.444859 2.079984 -0.050356 C 0.297067 1.534529 1.172097 C 0.158059 0.033534 1.391892 S 0.531468 -0.976711 -0.087891 O -2.624666 -1.245157 0.853871 H -3.288899 0.503562 -1.359914 H -1.661665 0.626689 -2.045951 H -2.509994 2.530027 -0.500824 H -2.327595 1.420309 0.837997 H 0.076136 1.760694 -0.958137 H -0.373803 3.170540 -0.021999 H -0.080664 2.030776 2.074220 H 1.358410 1.788431 1.092143 H -0.860068 -0.234578 1.688462 H 0.835548 -0.302995 2.174992 S 2.975487 -0.658013 0.189379 C 3.134410 0.604415 -1.119959 H 2.701809 1.554699 -0.802797 H 4.188541 0.759722 -1.353431 H 2.622752 0.274499 -2.025657 ```

__Frequencies__ (Top 10 out of 63)

``` 1. -169.2664 cm-1 (Symmetry: A) * 2. 49.4183 cm-1 (Symmetry: A) 3. 77.2824 cm-1 (Symmetry: A) 4. 111.7186 cm-1 (Symmetry: A) 5. 119.8444 cm-1 (Symmetry: A) 6. 138.6142 cm-1 (Symmetry: A) 7. 161.4785 cm-1 (Symmetry: A) 8. 182.0178 cm-1 (Symmetry: A) 9. 188.6954 cm-1 (Symmetry: A) 10. 236.7629 cm-1 (Symmetry: A) ``` ***

TS–7b

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 |

266

| Stoichiometry | C6H13S3(1-) | | Number of Basis Functions | 313 | | Electronic Energy (Eh) | -1431.08442835 | | Sum of electronic and zero-point Energies (Eh) | -1430.901155 | | Sum of electronic and thermal Energies (Eh) | -1430.889713 | | Sum of electronic and enthalpy Energies (Eh) | -1430.888769 | | Sum of electronic and thermal Free Energies (Eh) | -1430.939458 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 49 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 1.809883 -1.536993 0.077260 C 2.225198 -0.292692 -1.200591 C 2.465308 1.122470 -0.681928 C 1.263675 1.793442 -0.004022 C 0.910965 1.222754 1.382502 C -0.371427 0.396453 1.452353 S -0.609984 -0.929622 0.172565 H 3.118042 -0.642024 -1.724850 H 1.404180 -0.279389 -1.924513 H 2.772282 1.739050 -1.535420 H 3.308827 1.113973 0.017646 H 1.493570 2.857307 0.099264 H 0.391651 1.728614 -0.665189 H 1.749681 0.639690 1.762340 H 0.758193 2.046881 2.088435 H -0.462813 -0.071247 2.434059 H -1.229310 1.052736 1.316775 S -2.934371 -0.415152 -0.058894 C -2.736047 1.061169 -1.106976 H -2.151942 0.818927 -1.996069 H -3.717779 1.422578 -1.415241 H -2.229068 1.859600 -0.560173 ```

__Frequencies__ (Top 10 out of 60)

``` 1. -207.4044 cm-1 (Symmetry: A) * 2. 40.6747 cm-1 (Symmetry: A) 3. 69.0615 cm-1 (Symmetry: A) 4. 124.2846 cm-1 (Symmetry: A) 5. 145.4820 cm-1 (Symmetry: A) 6. 159.6536 cm-1 (Symmetry: A) 7. 197.3744 cm-1 (Symmetry: A) 8. 209.9390 cm-1 (Symmetry: A)

267

9. 223.4529 cm-1 (Symmetry: A) 10. 235.5937 cm-1 (Symmetry: A) ``` ***

TS–8a

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C7H15OS3(1-) | | Number of Basis Functions | 371 | | Electronic Energy (Eh) | -1545.57554356 | | Sum of electronic and zero-point Energies (Eh) | -1545.359876 | | Sum of electronic and thermal Energies (Eh) | -1545.346172 | | Sum of electronic and enthalpy Energies (Eh) | -1545.345228 | | Sum of electronic and thermal Free Energies (Eh) | -1545.401245 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 57 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -1.201355 -1.746536 -0.387323 C -2.559881 -0.569757 -0.760123 C -2.860759 0.436523 0.356712 C -2.482016 1.884374 0.027545 C -1.033126 2.144256 -0.390064 C 0.034682 1.912167 0.681971 C 0.415478 0.462657 0.984659 S 0.871587 -0.516477 -0.497275 O -1.432894 -2.146766 1.100443 H -3.424331 -1.208234 -0.953441 H -2.285851 -0.089967 -1.703810 H -2.384382 0.101153 1.280776 H -3.934241 0.411365 0.561171 H -2.715403 2.507665 0.898185 H -3.134149 2.230691 -0.782053 H -0.782573 1.551725 -1.275702 H -0.967039 3.190788 -0.702630 H -0.286519 2.376233 1.623220

268

H 0.949816 2.426031 0.374332 H -0.374289 -0.099330 1.484794 H 1.287987 0.446242 1.637987 S 3.134133 0.533975 -0.545223 C 3.896784 -0.654773 0.611436 H 4.949179 -0.406408 0.755079 H 3.399855 -0.621476 1.583291 H 3.828280 -1.670418 0.219589 ```

__Frequencies__ (Top 10 out of 72)

``` 1. -179.9281 cm-1 (Symmetry: A) * 2. 44.6575 cm-1 (Symmetry: A) 3. 48.6534 cm-1 (Symmetry: A) 4. 99.4035 cm-1 (Symmetry: A) 5. 106.6649 cm-1 (Symmetry: A) 6. 120.5401 cm-1 (Symmetry: A) 7. 127.0503 cm-1 (Symmetry: A) 8. 145.8202 cm-1 (Symmetry: A) 9. 193.5953 cm-1 (Symmetry: A) 10. 199.6678 cm-1 (Symmetry: A) ``` ***

TS–8b

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C7H15S3(1-) | | Number of Basis Functions | 349 | | Electronic Energy (Eh) | -1470.39216415 | | Sum of electronic and zero-point Energies (Eh) | -1470.179661 | | Sum of electronic and thermal Energies (Eh) | -1470.167242 | | Sum of electronic and enthalpy Energies (Eh) | -1470.166298 | | Sum of electronic and thermal Free Energies (Eh) | -1470.218923 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 53 |

__Molecular Geometry in Cartesian Coordinates__

269

```xyz S 0.697925 -1.028941 -0.322126 C 0.428552 0.321839 -1.550584 C 0.309041 1.722109 -0.948017 C -0.717219 1.847014 0.179882 C -2.151209 1.480742 -0.246648 C -2.906402 0.606451 0.757366 C -2.205946 -0.706790 1.127889 S -1.626764 -1.713710 -0.286705 H -0.470420 0.038232 -2.093874 H 1.270101 0.285216 -2.240764 H 0.045678 2.405430 -1.765408 H 1.293982 2.030013 -0.585288 H -0.391506 1.202483 1.000086 H -0.692644 2.869066 0.568702 H -2.729411 2.392068 -0.424984 H -2.126727 0.946521 -1.199105 H -3.895197 0.376079 0.347425 H -3.074265 1.167335 1.685718 H -2.889757 -1.323083 1.715461 H -1.332140 -0.517204 1.757127 S 3.046188 -0.301388 -0.014106 C 2.750051 0.522466 1.586676 H 2.193540 1.451797 1.451874 H 2.176551 -0.132485 2.245265 H 3.703424 0.750177 2.065361 ```

__Frequencies__ (Top 10 out of 69)

``` 1. -209.8984 cm-1 (Symmetry: A) * 2. 43.6417 cm-1 (Symmetry: A) 3. 77.5194 cm-1 (Symmetry: A) 4. 110.7186 cm-1 (Symmetry: A) 5. 129.8310 cm-1 (Symmetry: A) 6. 143.1309 cm-1 (Symmetry: A) 7. 170.5801 cm-1 (Symmetry: A) 8. 191.9914 cm-1 (Symmetry: A) 9. 217.9300 cm-1 (Symmetry: A) 10. 225.1364 cm-1 (Symmetry: A) ``` ***

TS–9a

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 |

270

| Stoichiometry | C8H17OS3(1-) | | Number of Basis Functions | 407 | | Electronic Energy (Eh) | -1584.87501531 | | Sum of electronic and zero-point Energies (Eh) | -1584.63049 | | Sum of electronic and thermal Energies (Eh) | -1584.615586 | | Sum of electronic and enthalpy Energies (Eh) | -1584.614642 | | Sum of electronic and thermal Free Energies (Eh) | -1584.67308 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 61 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -1.015147 -1.830067 0.085800 C -2.230652 -1.073556 -1.095364 C -3.180721 -0.042604 -0.479078 C 0.383867 1.719399 1.247980 C 0.907652 0.286125 1.282262 S 0.989553 -0.579750 -0.332104 C -1.119018 1.809089 0.955575 C -2.884188 1.414528 -0.855006 C -1.458417 1.863529 -0.538759 O -1.564176 -1.591319 1.515847 H -1.630304 -0.667714 -1.913984 H -2.787382 -1.928346 -1.481271 H -3.170497 -0.170607 0.606833 H -4.197047 -0.281644 -0.801915 H 0.587914 2.146958 2.236009 H 0.960795 2.306962 0.526014 H 0.260939 -0.324527 1.919809 H 1.921264 0.252191 1.679283 H -1.542546 2.692547 1.444038 H -1.596961 0.938790 1.416328 H -3.598964 2.058072 -0.329285 H -3.071755 1.549205 -1.925818 H -1.308689 2.885092 -0.904436 H -0.756890 1.237958 -1.093847 S 3.222263 0.575669 -0.859780 C 4.237875 -0.635823 0.056556 H 3.947177 -0.667584 1.109200 H 4.119408 -1.635312 -0.364141 H 5.291847 -0.359231 0.002769 ```

__Frequencies__ (Top 10 out of 81)

``` 1. -155.7149 cm-1 (Symmetry: A) *

271

2. 49.7966 cm-1 (Symmetry: A) 3. 58.6524 cm-1 (Symmetry: A) 4. 72.0721 cm-1 (Symmetry: A) 5. 96.4392 cm-1 (Symmetry: A) 6. 112.0144 cm-1 (Symmetry: A) 7. 127.2042 cm-1 (Symmetry: A) 8. 139.1976 cm-1 (Symmetry: A) 9. 177.2006 cm-1 (Symmetry: A) 10. 193.3493 cm-1 (Symmetry: A) ``` ***

TS–9b

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C8H17S3(1-) | | Number of Basis Functions | 385 | | Electronic Energy (Eh) | -1509.68763285 | | Sum of electronic and zero-point Energies (Eh) | -1509.446582 | | Sum of electronic and thermal Energies (Eh) | -1509.432763 | | Sum of electronic and enthalpy Energies (Eh) | -1509.431818 | | Sum of electronic and thermal Free Energies (Eh) | -1509.487641 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 57 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -1.065607 -1.894287 0.819368 C -2.132544 -1.762024 -0.676173 C -2.302561 -0.350013 -1.243188 C -0.456357 1.385895 1.350026 C 0.827140 0.556593 1.315210 S 0.951249 -0.703923 -0.040286 C -0.830840 2.074493 0.028450 C -2.968699 0.645762 -0.278941 C -2.337732 2.047520 -0.275789 H -3.106733 -2.171446 -0.392537 H -1.723139 -2.413413 -1.451698 H -2.892444 -0.426601 -2.163638

272

H -1.316853 0.009280 -1.540906 H -0.327966 2.134501 2.139889 H -1.262113 0.730486 1.675104 H 0.945556 0.022474 2.258960 H 1.697624 1.195187 1.170822 H -0.289412 1.594267 -0.789832 H -0.493998 3.115453 0.040431 H -2.933858 0.230395 0.730959 H -4.030183 0.744105 -0.526241 H -2.865721 2.664792 0.460253 H -2.504043 2.520718 -1.250362 S 2.885539 0.354037 -1.039809 C 4.132324 -0.147848 0.192834 H 4.167845 -1.234538 0.277985 H 5.114455 0.214198 -0.114025 H 3.897708 0.274644 1.171897 ```

__Frequencies__ (Top 10 out of 78)

``` 1. -207.4766 cm-1 (Symmetry: A) * 2. 51.5361 cm-1 (Symmetry: A) 3. 61.3006 cm-1 (Symmetry: A) 4. 86.8857 cm-1 (Symmetry: A) 5. 114.6180 cm-1 (Symmetry: A) 6. 136.5495 cm-1 (Symmetry: A) 7. 140.9816 cm-1 (Symmetry: A) 8. 159.8054 cm-1 (Symmetry: A) 9. 190.1707 cm-1 (Symmetry: A) 10. 203.4153 cm-1 (Symmetry: A) ``` ***

TS–10a

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C9H19OS3(1-) | | Number of Basis Functions | 443 | | Electronic Energy (Eh) | -1624.18113886 | | Sum of electronic and zero-point Energies (Eh) | -1623.907889 | | Sum of electronic and thermal Energies (Eh) | -1623.89175 | | Sum of electronic and enthalpy Energies (Eh) | -1623.890806 |

273

| Sum of electronic and thermal Free Energies (Eh) | -1623.951918 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 65 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S -0.898201 1.713500 0.710722 C -1.805050 1.822434 -0.871752 C -2.237547 0.470077 -1.432932 C -3.093574 -0.382523 -0.476529 C -0.947142 -1.945266 1.220664 C 0.576416 -1.892015 1.428087 C 1.160971 -0.481942 1.424049 S 0.989583 0.438243 -0.167969 C -1.353579 -2.265756 -0.223860 C -2.797850 -1.885890 -0.574004 O -0.373284 3.170706 0.926146 H -2.658835 2.476852 -0.675140 H -1.127747 2.337433 -1.559688 H -1.338213 -0.075020 -1.719817 H -2.792261 0.653721 -2.357821 H -4.153474 -0.212207 -0.685507 H -2.942624 -0.052495 0.554313 H -1.388752 -2.696056 1.883113 H -1.370949 -0.985957 1.532473 H 1.084275 -2.490878 0.664985 H 0.838018 -2.333252 2.395694 H 0.679791 0.125809 2.192486 H 2.228144 -0.510613 1.639690 H -0.661581 -1.763290 -0.903030 H -1.215504 -3.338781 -0.394888 H -3.016197 -2.227131 -1.592348 H -3.484629 -2.425265 0.088122 S 2.953163 -0.712105 -1.074495 C 4.198928 0.420931 -0.373025 H 4.028911 1.440371 -0.721174 H 4.160451 0.411263 0.718146 H 5.195283 0.105347 -0.685099 ```

__Frequencies__ (Top 10 out of 90)

``` 1. -187.9722 cm-1 (Symmetry: A) * 2. 47.6070 cm-1 (Symmetry: A) 3. 52.0603 cm-1 (Symmetry: A) 4. 81.6039 cm-1 (Symmetry: A) 5. 91.8693 cm-1 (Symmetry: A) 6. 108.1707 cm-1 (Symmetry: A) 7. 111.7646 cm-1 (Symmetry: A) 8. 127.5650 cm-1 (Symmetry: A) 9. 146.1626 cm-1 (Symmetry: A) 10. 179.6252 cm-1 (Symmetry: A)

274

``` ***

TS–10b

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | -1 | | Multiplicity | 1 | | Stoichiometry | C9H19S3(1-) | | Number of Basis Functions | 421 | | Electronic Energy (Eh) | -1548.99240144 | | Sum of electronic and zero-point Energies (Eh) | -1548.722665 | | Sum of electronic and thermal Energies (Eh) | -1548.70755 | | Sum of electronic and enthalpy Energies (Eh) | -1548.706606 | | Sum of electronic and thermal Free Energies (Eh) | -1548.765562 | | Number of Imaginary Frequencies | 1 | | Mean of alpha and beta Electrons | 61 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 0.527426 2.215441 0.827071 C 1.609976 2.203433 -0.660268 C 2.026053 0.817075 -1.147176 C 3.075398 0.090140 -0.299587 C 0.734988 -2.037842 0.244155 C 0.385949 -1.092894 1.396575 C -0.984703 -0.434024 1.346930 S -1.238613 0.745044 -0.058290 C 2.241612 -2.355099 0.205586 C 3.139013 -1.415565 -0.620554 H 1.081527 2.722282 -1.463198 H 2.490407 2.804112 -0.413104 H 2.425502 0.916394 -2.164379 H 1.126724 0.208130 -1.234591 H 4.052417 0.542203 -0.500533 H 2.886601 0.248443 0.765665 H 0.170371 -2.967885 0.369440 H 0.409092 -1.614887 -0.710476 H 1.129272 -0.301949 1.471889 H 0.443884 -1.658628 2.335114 H -1.154074 0.132921 2.263192

275

H -1.777869 -1.174326 1.247214 H 2.612993 -2.378480 1.237915 H 2.388019 -3.365435 -0.188717 H 4.167751 -1.768153 -0.497525 H 2.901371 -1.547920 -1.683659 S -2.972582 -0.658245 -1.084660 C -4.326020 -0.278935 0.078561 H -4.070500 -0.607894 1.088034 H -4.522292 0.793831 0.099317 H -5.234479 -0.796337 -0.232864 ```

__Frequencies__ (Top 10 out of 87)

``` 1. -210.6581 cm-1 (Symmetry: A) * 2. 46.0430 cm-1 (Symmetry: A) 3. 51.6250 cm-1 (Symmetry: A) 4. 75.9915 cm-1 (Symmetry: A) 5. 100.5276 cm-1 (Symmetry: A) 6. 116.6576 cm-1 (Symmetry: A) 7. 122.3534 cm-1 (Symmetry: A) 8. 130.4428 cm-1 (Symmetry: A) 9. 178.3091 cm-1 (Symmetry: A) 10. 212.2311 cm-1 (Symmetry: A) ``` ***

6c_Reactant

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 | | Stoichiometry | C4H8S3 | | Number of Basis Functions | 234 | | Electronic Energy (Eh) | -1351.75429984 | | Sum of electronic and zero-point Energies (Eh) | -1351.635218 | | Sum of electronic and thermal Energies (Eh) | -1351.627538 | | Sum of electronic and enthalpy Energies (Eh) | -1351.626594 |

276

| Sum of electronic and thermal Free Energies (Eh) | -1351.667827 | | Number of Imaginary Frequencies | 0 | | Mean of alpha and beta Electrons | 40 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 0.626413 -1.576360 -0.351300 S -0.940747 -0.215649 -0.779352 C 0.084892 1.275485 -1.096684 C 0.946479 1.710062 0.077669 C 2.026441 0.703803 0.477794 C 1.465317 -0.628055 0.960014 H -0.646044 2.038163 -1.366217 H 0.676310 1.010652 -1.975112 H 1.419220 2.650605 -0.215054 H 0.302631 1.930521 0.933445 H 2.614206 1.132322 1.293708 H 2.709653 0.530270 -0.358668 H 2.256000 -1.299498 1.297975 H 0.755353 -0.495038 1.780234 S -2.012297 0.175399 0.886711 ```

__Frequencies__ (Top 10 out of 39)

``` 1. 115.1874 cm-1 (Symmetry: A) 2. 173.6178 cm-1 (Symmetry: A) 3. 219.2121 cm-1 (Symmetry: A) 4. 257.0682 cm-1 (Symmetry: A) 5. 323.2467 cm-1 (Symmetry: A) 6. 344.5467 cm-1 (Symmetry: A) 7. 386.1748 cm-1 (Symmetry: A) 8. 434.9928 cm-1 (Symmetry: A) 9. 490.4004 cm-1 (Symmetry: A) 10. 517.5559 cm-1 (Symmetry: A) ``` ***

6c*_Reactant

__Requested operations__

Run with Gaussian 16revisionA.03.

| Datum | Value | |:------|------:| | Charge | 0 | | Multiplicity | 1 |

277

| Stoichiometry | C4H8S3 | | Number of Basis Functions | 234 | | Electronic Energy (Eh) | -1351.54612168 | | Sum of electronic and zero-point Energies (Eh) | -1351.426279 | | Sum of electronic and thermal Energies (Eh) | -1351.419764 | | Sum of electronic and enthalpy Energies (Eh) | -1351.41882 | | Sum of electronic and thermal Free Energies (Eh) | -1351.457724 | | Number of Imaginary Frequencies | 2 | | Mean of alpha and beta Electrons | 40 |

__Molecular Geometry in Cartesian Coordinates__

```xyz S 0.642805 -1.576595 -0.448497 S -1.156361 -0.143339 -0.523867 C -0.033721 1.278018 -1.048766 C 0.914899 1.715260 0.053383 C 2.033318 0.717937 0.364510 C 1.524219 -0.640251 0.840596 H -0.742626 2.070374 -1.301167 H 0.476026 0.934315 -1.948049 H 1.350921 2.672264 -0.244248 H 0.341063 1.920937 0.965689 H 2.665927 1.141187 1.150408 H 2.663200 0.576965 -0.519072 H 2.345143 -1.280348 1.165659 H 0.855655 -0.521606 1.702378 S -1.773170 0.098692 0.832993 ```

__Frequencies__ (Top 10 out of 39)

``` 1. -487.9592 cm-1 (Symmetry: A) * 2. -454.2941 cm-1 (Symmetry: A) * 3. 202.3063 cm-1 (Symmetry: A) 4. 250.0734 cm-1 (Symmetry: A) 5. 259.1456 cm-1 (Symmetry: A) 6. 287.4842 cm-1 (Symmetry: A) 7. 341.8365 cm-1 (Symmetry: A) 8. 369.5550 cm-1 (Symmetry: A) 9. 497.1117 cm-1 (Symmetry: A) 10. 540.8622 cm-1 (Symmetry: A) ``` ***

278

Appendix 3: Supplementary Figures for New Pharmacological Chaperone Scaffold for the Stabilization of Disease-Associated Variants of Cu/Zn Superoxide Dismutase

Variant Radius of gyration (Å) WT control 47.2 WT + 1,2 Dithiolane-1-oxide 46.1 WT + 1,2 Dithiane-1-oxide 25.2 WT + 1,2 Dithiepane-1-oxide 62.3 G93A control 70 G93A + 1,2 Dithiolane-1-oxide 24.4 G93A + 1,2 Dithiane-1-oxide 23.2 G93A + 1,2 Dithiepane-1-oxide 22.1 A4V control ----- A4V + 1,2 Dithiolane-1-oxide 21.7 A4V + 1,2 Dithiane-1-oxide 22.9 A4V + 1,2 Dithiepane-1-oxide 22.9

Supplementary Table A3-1: Radius of gyration of each sample calculated in RAW.248

279

Figure A3-1. 1,2-dithiane-1-oxide cross-links SOD1 at Cys111. a. Trypsin digest and MALDI-

TOF-MS analysis of SOD1A4V gave a combined sequence coverage of 94%, including peaks corresponding to the Cys57-Cys146 disulfide linked peptides (blue) and the Cys111- containing peptide (purple) linked by 1,2-dithiane-1-oxide. N-terminal acetylation is shown in blue text, cysteines are shown in red text, A4V mutation is shown underlined. b. Examining the higher mass range revealed multiply charged forms of SOD1A4V including the 1,2-dithiane-1-oxide linked dimer, as well as two different peaks corresponding to Cys111-linked peptides.

280

Figure A3-2. Analysis of α-lipoic acid shows presence of oxidized form. Bottom spectrum shows analysis of β-lipoic acid synthesized in-house from α-lipoic acid. Top spectrum shows analysis of commercially acquired α -lipoic acid. Analysis was performed on a mass spectrometer with a reducing source to avoid artifactual oxidation (Bruker SolariX XR FT-ICR-MS).

281

Supplementary Figure A3-3. 1H NMR of 1,2-dithiolane-1-oxide

282

Supplementary Figure A3-4. 13C NMR of 1,2-dithiolane-1-oxide

283

Appendix 4: Supplementary Figures for Best Practices and Benchmarks for Mass Spectrometry of Intact Proteins

Figure A4-1. Signal suppression curves of common components. These components are outlined in Figure 1c. The x-axis represents an increasing concentration of interfering substance [C] and the y-axis represents the fraction of signal lost. Each spectrum was collected in triplicate. S/N was calculated as described in the Online Methods. Standard deviation of each data point was calculated and used to produce error bars. The least-squares fitting (LSF) calculation is included to show the quality of fit to the equation.

284

Figure A4-2. Fractionation of human whole-cell lysate prior to top-down mass spectrometry.

Human colorectal cancer cells were lysed and constituent proteins quantified by the methods described by Anderson et. al.1 Aliquots of lysate containing 400 µg were precipitated in acetone, resuspended in 1% SDS containing 50 mM DTT, and resolved on 8% T (a) or 10% T (b) gel-eluted liquid fraction entrapment electrophoresis (GELFrEE) cartridges following the respective manufacturer’s protocols (GELFrEE 8100 Fractionation System, Expedeon, Inc.). Upon collection of MW-based fractions, 10 µL aliquots were resolved by SDS-PAGE and visualized by AgNO3 stain2 to gauge protein content and quality of resolution. (a,b) Note that the MW ranges of f1

(purple box) and subsequent fractions differ depending on the GELFrEE cartridge selected. While 285

8% cartridges (a) are recommended for quantitative high-throughput top-down MS applications or analysis of higher-MW proteins, 10% (b) cartridges provide superior resolution in the 5-30 kDa

MW range for qualitative high-throughput applications.

Figure A4-3. Antibody Buffer and Gentle Elution Buffer Ablate MS Signal; MWCO-

Ultrafiltration and Precipitation Rescue Signal. Buffers included (a) Thermo Gentle Elution Buffer

(containing molar salt concentration), and (b) Antibody buffer (10 mM Arginine, 10 mM Tris HCl,

10 mM histidine, 10 mM potassium phosphate, 10 mM citric acid, pH 5.5). All the above spectra were obtained using a Bruker SolariX FT-ICR mass spectrometer, 9.4T.

286

Figure A4-4. Sample Preparation of Protein Mixture following Protocol 3 (MWCO-

Ultrafiltration). These samples were analyzed by direct infusion on a (a) Waters Xevo G2-S QTOF and a (b) Thermo Q Exactive Plus.

287

Figure A4-5. Native vs. Denatured MS of AquaporinZ (AqpZ) from E. coli. Native spectrum was acquired on a Waters Synapt G1 Q-TOF with nanoESI via direct infusion while denatured spectra was acquired on a Waters Synapt G1 Q-TOF via nanoESI-LC-MS. The native sample (a) and the denatured sample (b) was acquired at a concentration of 10 µM. *Denotes the most abundant charge state. 24268.7 Da is the deconvoluted mass of the unmodified AqpZ monomer. Formylated

AqpZ was also detected with a mass of 24,296.4 Da. ‡Five native tetramer masses were observed corresponding to five unique combinations of formylated and unformylated monomers.

288

Figure A4-6. Denaturing vs. Native Analysis of Carbonic Anhydrase. Both denaturing and native analysis of carbonic anhydrase was run on a (a.) Thermo Q Exactive HF MS (b.) Bruker 15T

SolariX FT-ICR MS, (c.) Bruker 12T SolariX FT-ICR MS , (d.) Bruker maXis II ETD Q-TOF,

(e.) Bruker 15T SolariX FT-ICR MS, (f.) Waters Synapt G2Si MS.

289

Figure A4-7. LC-MS of protein standard mixture run on Waters Acquity-Xevo G2-S QTOF.

Samples were prepared following the given SOP and separated using a PLRP-S (top panel of a, b, and c) or a C4 (bottom panel of a, b, and c) stationary phase. (a.) The final concentrations of each protein loaded onto the column were; 14 pmol ubiquitin, 49 pmol trypsinogen, 109 pmol myoglobin, and 64 pmol carbonic anhydrase. (b.) The final concentrations of each protein loaded onto the column were; 1.4 pmol ubiquitin, 4.9 pmol trypsinogen, 10.9 pmol myoglobin, and 6.4 pmol carbonic anhydrase. (c.) The final concentrations of each protein loaded onto the column

290 were; 0.14 pmol ubiquitin, 0.49 pmol trypsinogen, 1.09 pmol myoglobin, and 0.64 pmol carbonic anhydrase.

Figure A4-8. LC-MS of protein standard mixture run on Waters nanoAcquity coupled to a Bruker

QTOF and a Bruker FT-ICR MS. Samples were prepared following the given SOP and separated using a PLRP-S on a Waters nanoAcquity coupled to (a.) a Bruker impact II QTOF and (b.) a

Bruker SolariX FT-ICR MS.

291

Figure A4-9. LC-MS of protein standard mixture run on Dionex UPLC coupled to three different orbitrap mass spectrometers. Samples were prepared following the given SOP and separated S on a Dionex UPLC coupled to (a.) a Thermo Orbitrap Elite (monolithic stationary phase (b.) Thermo

Orbitrap Fusion Lumos (PLRP-S stationary phase) (c.) Thermo Orbitrap QE-HF (PLRP-S stationary phase).

292

Figure A4-10. LC-MS of protein standard mixture run on Dionex UPLC coupled to a Thermo

Orbitrap Fusion Lumos. Samples were prepared following the given SOP on a Dionex UltiMate

3000 RSLCNano System with a Thermo Fusion Lumos using PLRP-S stationary phase and analyzed on a Thermo Fusion Lumos. The final concentrations of each protein loaded onto the column were; 0.14 pmol ubiquitin, 0.49 pmol trypsinogen, 1.09 pmol myoglobin, and 0.64 pmol carbonic anhydrase.

293

Figure A4-11. Capillary Zone Electrophoresis Separation of Protein Mixture. Separated using a separated using a prototype CESI-8000 Plus (AB SCIEX) used with a Neutral OptiMS cartridge.

294

Figure A4-12. LC-MS of Halobacterium salinarum prepared following Supplemental Protocol 5b.

Proteins were separated using a PLRP-S stationary phase (300 Å pore size, 3 µm bead size) and analyzed on a (a.) Waters nanoAcquity interfaced with a Bruker SolariX FT-ICR MS (b.) Waters

H-Class Acquity UPLC interfaced with a Waters Xevo G2-S QTOF (c.) Thermo Scientific

Vanquish interfaced with a Thermo Orbitrap Q Exactive (d.) Agilent 1290 interfaced with a

Thermo Orbitrap Exactive Plus.

Figure A4-13.

295

296

References

1. Sweeney, P. et al. Protein misfolding in neurodegenerative diseases: implications and strategies. Translational Neurodegeneration 6, 6 (2017). 2. Lim, J. & Yue, Z. Neuronal aggregates: formation, clearance, and spreading. Developmental cell 32, 491-501 (2015). 3. Murphy, M.P. & LeVine, H., 3rd Alzheimer's disease and the amyloid-beta peptide. Journal of Alzheimer's disease : JAD 19, 311-323 (2010). 4. Roychaudhuri, R., Yang, M., Hoshi, M.M. & Teplow, D.B. Amyloid β-Protein Assembly and Alzheimer Disease. Journal of Biological Chemistry 284, 4749-4753 (2009). 5. Laurent, C., Buée, L. & Blum, D. Tau and neuroinflammation: What impact for Alzheimer's Disease and Tauopathies? Biomedical Journal 41, 21-33 (2018). 6. Breydo, L., Wu, J.W. & Uversky, V.N. Alpha-synuclein misfolding and Parkinson's disease. Biochimica et biophysica acta 1822, 261-285 (2012). 7. Uversky, V.N. Alpha-synuclein misfolding and neurodegenerative diseases. Current protein & peptide science 9, 507-540 (2008). 8. Porta, S. et al. Patient-derived frontotemporal lobar degeneration brain extracts induce formation and spreading of TDP-43 pathology in vivo. Nature Communications 9, 4220 (2018). 9. Lee, E.B. et al. Expansion of the classification of FTLD-TDP: distinct pathology associated with rapidly progressive frontotemporal degeneration. Acta neuropathologica 134, 65-78 (2017). 10. Rotunno, M. & Bosco, D. An emerging role for misfolded wild-type SOD1 in sporadic ALS pathogenesis. Frontiers in Cellular Neuroscience 7 (2013). 11. Valentine, J.S. & Hart, P.J. Misfolded CuZnSOD and amyotrophic lateral sclerosis. Proceedings of the National Academy of Sciences of the United States of America 100, 3617-3622 (2003). 12. Pokrishevsky, E. et al. Aberrant Localization of FUS and TDP43 Is Associated with Misfolding of SOD1 in Amyotrophic Lateral Sclerosis. PLOS ONE 7, e35050 (2012). 13. Tan, C.-F. et al. TDP-43 immunoreactivity in neuronal inclusions in familial amyotrophic lateral sclerosis with or without SOD1 gene mutation. Acta neuropathologica 113, 535- 542 (2007). 14. Arai, T. et al. TDP-43 is a component of ubiquitin-positive tau-negative inclusions in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Biochemical and Biophysical Research Communications 351, 602-611 (2006). 15. Sharma, A. et al. ALS-associated mutant FUS induces selective motor neuron degeneration through toxic gain of function. Nature Communications 7, 10465 (2016). 16. Patel, A. et al. A Liquid-to-Solid Phase Transition of the ALS Protein FUS Accelerated by Disease Mutation. Cell 162, 1066-1077 (2015). 17. Balendra, R. & Isaacs, A.M. C9orf72-mediated ALS and FTD: multiple pathways to disease. Nature Reviews Neurology 14, 544-558 (2018). 18. Lee, Y.-B. et al. C9orf72 poly GA RAN-translated protein plays a key role in amyotrophic lateral sclerosis via aggregation and toxicity. Human Molecular Genetics 26, 4765-4777 (2017).

297

19. Deng, H.-X. et al. Mutations in UBQLN2 cause dominant X-linked juvenile and adult- onset ALS and ALS/dementia. Nature 477, 211 (2011). 20. Kim, S.H. et al. Mutation-dependent aggregation and toxicity in a Drosophila model for UBQLN2-associated ALS. Hum Mol Genet 27, 322-337 (2018). 21. Arrasate, M. & Finkbeiner, S. Protein aggregates in Huntington's disease. Experimental neurology 238, 1-11 (2012). 22. Hoffner, G. & Djian, P. Protein aggregation in Huntington’s disease. Biochimie 84, 273- 278 (2002). 23. Saelices, L. et al. Uncovering the Mechanism of Aggregation of Human Transthyretin. Journal of Biological Chemistry (2015). 24. Hurshman, A.R., White, J.T., Powers, E.T. & Kelly, J.W. Transthyretin Aggregation under Partially Denaturing Conditions Is a Downhill Polymerization. Biochemistry 43, 7365-7381 (2004). 25. Golde, T.E., Borchelt, D.R., Giasson, B.I. & Lewis, J. Thinking laterally about neurodegenerative proteinopathies. The Journal of clinical investigation 123, 1847-1855 (2013). 26. Damier, P., Hirsch, E.C., Agid, Y. & Graybiel, A.M. The substantia nigra of the human brain. II. Patterns of loss of dopamine-containing neurons in Parkinson's disease. Brain : a journal of neurology 122 ( Pt 8), 1437-1448 (1999). 27. Braak, H. & Braak, E. Frequency of stages of Alzheimer-related lesions in different age categories. Neurobiology of aging 18, 351-357 (1997). 28. Nordberg, A. Amyloid plaque imaging in vivo: current achievement and future prospects. European Journal of Nuclear Medicine and Molecular Imaging 35, 46-50 (2008). 29. Bergh, J. et al. Expression of wild-type human superoxide dismutase-1 in mice causes amyotrophic lateral sclerosis. Human Molecular Genetics 22, 51-60 (2012). 30. Stefanova, N. et al. Oxidative stress in transgenic mice with oligodendroglial alpha- synuclein overexpression replicates the characteristic neuropathology of multiple system atrophy. The American journal of pathology 166, 869-876 (2005). 31. Tanaka, Y. et al. Inducible expression of mutant alpha-synuclein decreases proteasome activity and increases sensitivity to mitochondria-dependent apoptosis. Hum Mol Genet 10, 919-926 (2001). 32. Walker, L.C., Diamond, M.I., Duff, K.E. & Hyman, B.T. Mechanisms of protein seeding in neurodegenerative diseases. JAMA neurology 70, 304-310 (2013). 33. Jucker, M. & Walker, L.C. Pathogenic protein seeding in Alzheimer disease and other neurodegenerative disorders. Annals of neurology 70, 532-540 (2011). 34. Kane, M.D. et al. Evidence for seeding of beta -amyloid by intracerebral infusion of Alzheimer brain extracts in beta -amyloid precursor protein-transgenic mice. The Journal of neuroscience : the official journal of the Society for Neuroscience 20, 3606-3611 (2000). 35. Ronnback, A. et al. Amyloid neuropathology in the single Arctic APP transgenic model affects interconnected brain regions. Neurobiology of aging 33, 831.e811-839 (2012). 36. Furukawa, Y., Kaneko, K., Watanabe, S., Yamanaka, K. & Nukina, N. Intracellular seeded aggregation of mutant Cu,Zn-superoxide dismutase associated with amyotrophic lateral sclerosis. FEBS letters 587, 2500-2505 (2013). 37. Schubert, U. et al. Rapid degradation of a large fraction of newly synthesized proteins by proteasomes. Nature 404, 770-774 (2000).

298

38. Hartl, F.U., Bracher, A. & Hayer-Hartl, M. Molecular chaperones in protein folding and proteostasis. Nature 475, 324-332 (2011). 39. Ciechanover, A., Orian, A. & Schwartz, A.L. Ubiquitin-mediated proteolysis: biological regulation via destruction. BioEssays : news and reviews in molecular, cellular and developmental biology 22, 442-451 (2000). 40. Kundu, M. & Thompson, C.B. Autophagy: basic principles and relevance to disease. Annual review of pathology 3, 427-455 (2008). 41. Leestemaker, Y. et al. Proteasome Activation by Small Molecules. Cell chemical biology 24, 725-736.e727 (2017). 42. Liu, Y. et al. Sulforaphane enhances proteasomal and autophagic activities in mice and is a potential therapeutic reagent for Huntington's disease. Journal of Neurochemistry 129, 539-547 (2014). 43. Lee, B.-H. et al. Enhancement of proteasome activity by a small-molecule inhibitor of USP14. Nature 467, 179 (2010). 44. Myeku, N. et al. Tau-driven 26S proteasome impairment and cognitive dysfunction can be prevented early in disease by activating cAMP-PKA signaling. Nature medicine 22, 46-53 (2016). 45. Knauer, M.F., Soreghan, B., Burdick, D., Kosmoski, J. & Glabe, C.G. Intracellular accumulation and resistance to degradation of the Alzheimer amyloid A4/beta protein. Proc Natl Acad Sci U S A 89, 7437-7441 (1992). 46. Dharmadasa, T. & Kiernan, M.C. Riluzole, disease stage and survival in ALS. The Lancet Neurology 17, 385-386 (2018). 47. Mora, J.S. Edaravone for treatment of early-stage ALS. The Lancet Neurology 16, 772 (2017). 48. Rosen, D.R. et al. Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature 362, 59-62 (1993). 49. Schmitt, N.D. & Agar, J.N. Parsing disease-relevant protein modifications from epiphenomena: perspective on the structural basis of SOD1-mediated ALS. Journal of mass spectrometry : JMS 52, 480-491 (2017). 50. McCord, J.M. & Fridovich, I. Superoxide Dismutase: An Enzymic Function for Erythrocuprein (Hemocuprein) J. Biol. Chem. 244, 6049-6055 (1969). 51. Fridovich, I. Superoxide Dismutases. Annual Review of Biochemistry 44, 147-159 (1975). 52. Reaume, A.G. et al. Motor neurons in Cu/Zn superoxide dismutase-deficient mice develop normally but exhibit enhanced cell death after axonal injury. Nature Genetics 13, 43-47 (1996). 53. Bosco, D.A. et al. Wild-type and mutant SOD1 share an aberrant conformation and a common pathogenic pathway in ALS. Nature neuroscience 13, 1396-1403 (2010). 54. Auclair, J.R., Boggio, K.J., Petsko, G.A., Ringe, D. & Agar, J.N. Strategies for stabilizing superoxide dismutase (SOD1), the protein destabilized in the most common form of familial amyotrophic lateral sclerosis. Proc. Natl. Acad. Sci. U.S.A. 107, 21394-21399 (2010). 55. Molnar, K.S. et al. A common property of amyotrophic lateral sclerosis-associated variants: destabilization of the copper/zinc superoxide dismutase electrostatic loop. J. Biol. Chem. 284, 30965-30973 (2009).

299

56. Shaw, B.F. et al. Local unfolding in a destabilized, pathogenic variant of superoxide dismutase 1 observed with H/D exchange and mass spectrometry. The Journal of biological chemistry 281, 18167-18176 (2006). 57. Taylor, D.M. et al. Tryptophan 32 potentiates aggregation and cytotoxicity of a copper/zinc superoxide dismutase mutant associated with familial amyotrophic lateral sclerosis. The Journal of biological chemistry 282, 16329-16335 (2007). 58. Cao, X. et al. Structures of the G85R variant of SOD1 in familial amyotrophic lateral sclerosis. The Journal of biological chemistry 283, 16169-16177 (2008). 59. Galaleldeen, A. et al. Structural and biophysical properties of metal-free pathogenic SOD1 mutants A4V and G93A. Archives of biochemistry and biophysics 492, 40-47 (2009). 60. Rakhit, R. & Chakrabartty, A. Structure, folding, and misfolding of Cu,Zn superoxide dismutase in amyotrophic lateral sclerosis. Biochimica et biophysica acta 1762, 1025- 1037 (2006). 61. Schmidlin, T., Kennedy, B.K. & Daggett, V. Structural changes to monomeric CuZn superoxide dismutase caused by the familial amyotrophic lateral sclerosis-associated mutation A4V. Biophysical journal 97, 1709-1718 (2009). 62. Sato, S., Ward, C.L., Krouse, M.E., Wine, J.J. & Kopito, R.R. Glycerol reverses the misfolding phenotype of the most common cystic fibrosis mutation. The Journal of biological chemistry 271, 635-638 (1996). 63. Gelman, M.S. & Kopito, R.R. Rescuing protein conformation: prospects for pharmacological therapy in cystic fibrosis. The Journal of clinical investigation 110, 1591-1597 (2002). 64. Heda, G.D. & Marino, C.R. Surface expression of the cystic fibrosis transmembrane conductance regulator mutant DeltaF508 is markedly upregulated by combination treatment with sodium butyrate and low temperature. Biochem Biophys Res Commun 271, 659-664 (2000). 65. Fischer, H. et al. Partial restoration of defective chloride conductance in DeltaF508 CF mice by trimethylamine oxide. American journal of physiology. Lung cellular and molecular physiology 281, L52-57 (2001). 66. Tamarappoo, B.K., Yang, B. & Verkman, A.S. Misfolding of Mutant Aquaporin-2 Water Channels in Nephrogenic Diabetes Insipidus. Journal of Biological Chemistry 274, 34825-34831 (1999). 67. Tamarappoo, B. & S Verkman, A. Defective trafficking of AQP2 water channels in Nephrogenic Diabetes Insipidus and correction by chemical chaperones, Vol. 101. (1998). 68. S Rajan, R. et al. Chemical and Pharmacological Chaperones: Application for Recombinant Protein Production and Protein Folding Diseases, Vol. 18. (2011). 69. Williams, R., Mamotte, C. & R Burnett, J. Phenylketonuria: An Inborn Error of Phenylalanine Metabolism, Vol. 29. (2008). 70. Pey, A.L. et al. Mechanisms underlying responsiveness to tetrahydrobiopterin in mild phenylketonuria mutations. Human mutation 24, 388-399 (2004). 71. Zarate, Y.A. & Hopkin, R.J. Fabry's disease. The Lancet 372, 1427-1435 (2008). 72. Germain, D. et al. Treatment of Fabry’s Disease with the Pharmacologic Chaperone Migalastat, Vol. 375. (2016).

300

73. Yam, G.H.-F., Zuber, C. & Roth, J. A synthetic chaperone corrects the trafficking defect and disease phenotype in a protein misfolding disorder. The FASEB Journal 19, 12-18 (2005). 74. Rosenbloom, B.E. & Weinreb, N.J. Gaucher disease: a comprehensive review. Critical reviews in oncogenesis 18, 163-175 (2013). 75. Joosten, A. et al. A systematic investigation of iminosugar click clusters as pharmacological chaperones for the treatment of Gaucher disease. Chembiochem : a European journal of chemical biology 15, 309-319 (2014). 76. Yilmazer, B. et al. Investigation of novel pharmacological chaperones for Gaucher Disease. Journal of molecular graphics & modelling 76, 364-378 (2017). 77. Benito, J.M., Garcia Fernandez, J.M. & Ortiz Mellet, C. Pharmacological chaperone therapy for Gaucher disease: a patent review. Expert opinion on therapeutic patents 21, 885-903 (2011). 78. Bulawa, C.E. et al. Tafamidis, a potent and selective transthyretin kinetic stabilizer that inhibits the amyloid cascade. Proceedings of the National Academy of Sciences of the United States of America 109, 9629-9634 (2012). 79. Foss, T.R., Wiseman, R.L. & Kelly, J.W. The pathway by which the tetrameric protein transthyretin dissociates. Biochemistry 44, 15525-15533 (2005). 80. Colon, W. & Kelly, J.W. Partial denaturation of transthyretin is sufficient for amyloid fibril formation in vitro. Biochemistry 31, 8654-8660 (1992). 81. Damas, A.M. & Saraiva, M.J. Review: TTR Amyloidosis—Structural Features Leading to Protein Aggregation and Their Implications on Therapeutic Strategies. Journal of Structural Biology 130, 290-299 (2000). 82. Holmgren, G. et al. Clinical improvement and amyloid regression after liver transplantation in hereditary transthyretin amyloidosis. Lancet (London, England) 341, 1113-1116 (1993). 83. Herlenius, G., Wilczek, H.E., Larsson, M. & Ericzon, B.G. Ten years of international experience with liver transplantation for familial amyloidotic polyneuropathy: results from the Familial Amyloidotic Polyneuropathy World Transplant Registry. Transplantation 77, 64-71 (2004). 84. Sato, T. et al. Chromium(III) ion and thyroxine cooperate to stabilize the transthyretin tetramer and suppress in vitro amyloid fibril formation. FEBS letters 580, 491-496 (2006). 85. Baures, P.W., Peterson, S.A. & Kelly, J.W. Discovering transthyretin amyloid fibril inhibitors by limited screening. Bioorganic & medicinal chemistry 6, 1389-1401 (1998). 86. Alhamadsheh, M.M. et al. Potent kinetic stabilizers that prevent transthyretin-mediated cardiomyocyte proteotoxicity. Science translational medicine 3, 97ra81 (2011). 87. Choi, S. & Kelly, J.W. A competition assay to identify amyloidogenesis inhibitors by monitoring the fluorescence emitted by the covalent attachment of a stilbene derivative to transthyretin. Bioorganic & medicinal chemistry 19, 1505-1514 (2011). 88. Reixach, N., Adamski-Werner, S.L., Kelly, J.W., Koziol, J. & Buxbaum, J.N. Cell based screening of inhibitors of transthyretin aggregation. Biochem Biophys Res Commun 348, 889-897 (2006). 89. Klabunde, T. et al. Rational design of potent human transthyretin amyloid disease inhibitors. Nature structural biology 7, 312-321 (2000).

301

90. Connelly, S., Choi, S., Johnson, S.M., Kelly, J.W. & Wilson, I.A. Structure-based design of kinetic stabilizers that ameliorate the transthyretin amyloidoses. Current opinion in structural biology 20, 54-62 (2010). 91. Coelho, T. et al. Tafamidis for transthyretin familial amyloid polyneuropathy: a randomized, controlled trial. Neurology 79, 785-792 (2012). 92. Maurer, M.S. et al. Tafamidis Treatment for Patients with Transthyretin Amyloid Cardiomyopathy. New England Journal of Medicine 379, 1007-1016 (2018). 93. Ray, S.S., Nowak, R.J., Brown, R.H., Jr. & Lansbury, P.T., Jr. Small-molecule-mediated stabilization of familial amyotrophic lateral sclerosis-linked superoxide dismutase mutants against unfolding and aggregation. Proceedings of the National Academy of Sciences of the United States of America 102, 3639-3644 (2005). 94. Nowak, R.J., Cuny, G.D., Choi, S., Lansbury, P.T. & Ray, S.S. Improving binding specificity of pharmacological chaperones that target mutant superoxide dismutase-1 linked to familial amyotrophic lateral sclerosis using computational methods. J. Med. Chem. 53, 2709-2718 (2010). 95. Antonyuk, S., Strange, R.W. & Hasnain, S.S. Structural Discovery of Small Molecule Binding Sites in Cu−Zn Human Superoxide Dismutase Familial Amyotrophic Lateral Sclerosis Mutants Provides Insights for Lead Optimization. Journal of Medicinal Chemistry 53, 1402-1406 (2010). 96. Wright, G.S.A., Antonyuk, S.V., Kershaw, N.M., Strange, R.W. & Samar Hasnain, S. Ligand binding and aggregation of pathogenic SOD1. Nature Communications 4, 1758 (2013). 97. Cooney, D.A. et al. Maleimide—Biochemical, pharmacologic and toxicologic studies. Biochem. Pharmacol. 27, 151-166 (1978). 98. Benson, A.B., 3rd Oltipraz: a laboratory and clinical review. Journal of cellular biochemistry. Supplement 17f, 278-291 (1993). 99. Zhang, Y. & Munday, R. Dithiolethiones for cancer chemoprevention: where do we stand? Molecular Cancer Therapeutics 7, 3470 (2008). 100. Schiodt, M., Oxholm, P. & Jacobsen, A. Treatment of xerostomia in patients with primary Sjogren's syndrome with sulfarlem. Scandinavian journal of rheumatology. Supplement 61, 250-252 (1986). 101. Hamada, T. et al. Treatment of xerostomia with the bile secretion-stimulating drug anethole trithione: a clinical trial. The American journal of the medical sciences 318, 146- 151 (1999). 102. Lubet, R.A. et al. Chemopreventive efficacy of anethole trithione, N-acetyl-L-cysteine, miconazole and phenethylisothiocyanate in the DMBA-induced rat mammary cancer model. International Journal of Cancer 72, 95-101 (1997). 103. Vallianou, N., Evangelopoulos, A. & Koutalas, P. Alpha-lipoic Acid and diabetic neuropathy. The review of diabetic studies : RDS 6, 230-236 (2009). 104. Ziegler, D. et al. Oral Treatment With α-Lipoic Acid Improves Symptomatic Diabetic Polyneuropathy. The SYDNEY 2 trial 29, 2365-2370 (2006). 105. Packer, L., Witt, E.H. & Tritschler, H.J. alpha-Lipoic acid as a biological antioxidant. Free radical biology & medicine 19, 227-250 (1995). 106. Hoyle, C.E. & Bowman, C.N. Thiol-ene click chemistry. Angew. Chem., Int. Ed. 49, 1540-1573 (2010).

302

107. Kuhl, N. et al. Increased stability in self-healing polymer networks based on reversible Michael addition reactions. J. Appl. Polym. Sci. 134, 1-8 (2017). 108. Lockhart, J.N., Beezer, D.B., Stevens, D.M., Spears, B.R. & Harth, E. One-pot polyglycidol nanogels via liposome master templates for dual drug delivery. J. Control. Release 244, 366-374 (2016). 109. Wang, J.Q., Zhang, F.J., Tsang, W.P., Wan, C. & Wu, C. Fabrication of injectable high strength hydrogel based on 4-arm star PEG for cartilage tissue engineering. Biomaterials 120, 11-21 (2017). 110. Killops, K.L., Campos, L.M. & Hawker, C.J. Robust, efficient, and orthogonal synthesis of dendrimers via thiol-ene "click" chemistry. J. Am. Chem. Soc. 130, 5062-5064 (2008). 111. Marculescu, C. et al. Aryloxymaleimides for cysteine modification, disulfide bridging and the dual functionalization of disulfide bonds. Chem. Commun. 50, 7139-7142 (2014). 112. Konermann, L., Vahidi, S. & Sowole, M.A. Mass Spectrometry Methods for Studying Structure and Dynamics of Biological Macromolecules. Anal. Chem. 86, 213-232 (2014). 113. Baillie, T.A. Targeted Covalent Inhibitors for Drug Design. Angew. Chem., Int. Ed. 55, 13408-13421 (2016). 114. Valenzano, K.J. et al. Identification and Characterization of Pharmacological Chaperones to Correct Enzyme Deficiencies in Lysosomal Storage Disorders. Assay Drug Dev. Technol. 9, 213-235 (2011). 115. Narita, A. et al. Ambroxol chaperone therapy for neuronopathic Gaucher disease: A pilot study. Ann. Clin. Transl. Neurol. 3, 200-215 (2016). 116. Liu, Q. et al. Developing irreversible inhibitors of the protein kinase cysteinome. Chem. Biol. 20, 146-159 (2013). 117. Nelson, V., Ziehr, J., Agulnik, M. & Johnson, M. Afatinib: emerging next-generation tyrosine kinase inhibitor for NSCLC. Onco Targets Ther. 6, 135-143 (2013). 118. Hald, J. & Jacobsen, E. A drug sensitizing the organism to ethyl alcohol. Lancet (London, England) 2, 1001-1004 (1948). 119. Lindberg, P., Nordberg, P., Alminger, T., Brandstrom, A. & Wallmark, B. THE MECHANISM OF ACTION OF THE GASTRIC-ACID SECRETION INHIBITOR OMEPRAZOLE. J. Med. Chem. 29, 1327-1329 (1986). 120. Shin, J.M., Besancon, M., Simon, A. & Sachs, G. THE SITE OF ACTION OF PANTOPRAZOLE IN THE GASTRIC H+/K+-ATPASE. Biochim. Biophys. Acta 1148, 223-233 (1993). 121. Ferri, N., Corsini, A. & Bellosta, S. Pharmacology of the New P2Y(12) Receptor Inhibitors: Insights on Pharmacokinetic and Pharmacodynamic Properties. Drugs 73, 1681-1709 (2013). 122. Burns, J.A. & Whitesides, G.M. Predicting the stability of cyclic disulfides by molecular modeling: effective concentrations in thiol-disulfide interchange and the design of strongly reducing dithiols. J. Am. Chem. Soc. 112, 6296-6303 (1990). 123. Szajewski, R.P. & Whitesides, G.M. Rate constants and equilibrium constants for thiol- disulfide interchange reactions involving oxidized glutathione. J. Am. Chem. Soc. 102, 2011-2026 (1980). 124. Singh, R. & Whitesides, G.M. Degenerate intermolecular thiolate-disulfide interchange involving cyclic five-membered disulfides is faster by .apprx.103 than that involving six- or seven-membered disulfides. J. Am. Chem. Soc. 112, 6304-6309 (1990).

303

125. Abegg, D. et al. Strained Cyclic Disulfides Enable Cellular Uptake by Reacting with the Transferrin Receptor. J. Am. Chem. Soc. 139, 231-238 (2017). 126. Zong, L. et al. Epidithiodiketopiperazines: Strain-Promoted Thiol-Mediated Cellular Uptake at the Highest Tension. ACS Cent. Sci. 3, 449-453 (2017). 127. Gupta, V. & Carroll, K.S. Profiling the reactivity of cyclic C-nucleophiles towards electrophilic sulfur in cysteine sulfenic acid †Electronic supplementary information (ESI) available. See DOI: 10.1039/c5sc02569a Click here for additional data file. Click here for additional data file. Chem. Sci. 7, 400-415 (2016). 128. Shaked, Z., Szajewski, R.P. & Whitesides, G.M. Rates of thiol-disulfide interchange reactions involving proteins and kinetic measurements of thiol pKa values. Biochemistry 19, 4156-4166 (1980). 129. Cox, J.D., Wagman, D.D. & Medvedev, V.A. CODATA key values for thermodynamics. (Hemisphere Pub. Corp., New York :; 1989). 130. Nicovich, J.M., Kreutter, K.D., Van Dijk, C.A. & Wine, P.H. Temperature-dependent kinetics studies of the reactions bromine atom(2P3/2) + hydrogen sulfide .tautm. mercapto + hydrogen bromide and bromine atom(2P3/2) + methanethiol .tautm. methylthiol + hydrogen bromide. Heats of formation of mercapto and methylthio radicals. J. Phys. Chem. 96, 2518-2528 (1992). 131. Jiang, X. et al. Quantitative Imaging of Glutathione in Live Cells Using a Reversible Reaction-Based Ratiometric Fluorescent Probe. ACS Chem. Biol. 10, 864-874 (2015). 132. Sun, H. et al. Urinary exosomes as a novel biomarker for evaluation of alpha-lipoic acid's protective effect in early diabetic nephropathy. J. Clin. Lab. Anal. 31, 1-7 (2017). 133. Nare, B., Smith, J.M. & Prichard, R.K. Mechanisms of inactivation of Schistosoma mansoni and mammalian glutathione S-transferase activity by the antischistosomal drug oltipraz. Biochem. Pharmacol. 43, 1345-1351 (1992). 134. Shay, K.P., Moreau, R.F., Smith, E.J., Smith, A.R. & Hagen, T.M. Alpha-lipoic acid as a dietary supplement: Molecular mechanisms and therapeutic potential. Biochim. Biophys. Acta 1790, 1149-1160 (2009). 135. Gasparini, G., Sargsyan, G., Bang, E.K., Sakai, N. & Matile, S. Ring Tension Applied to Thiol‐Mediated Cellular Uptake. Angew. Chem., Int. Ed. 54, 7328-7331 (2015). 136. Schmidt, A.-C., Koppelt, J., Neustadt, M. & Otto, M. Mass spectrometric evidence for different complexes of peptides and proteins with arsenic(III), arsenic(V), copper(II), and zinc(II) species. Rapid Commun. Mass Spectrom. 21, 153-163 (2007). 137. Alon, A. et al. The dynamic disulphide relay of quiescin sulphydryl oxidase. Nature 488, 414-418 (2012). 138. Owen, G.R. et al. Human CD4 Metastability Is a Function of the Allosteric Disulfide Bond in Domain 2. Biochemistry 55, 2227-2237 (2016). 139. Gutle, D.D., Roret, T., Hecker, A., Reski, R. & Jacquot, J.P. Dithiol disulphide exchange in redox regulation of chloroplast enzymes in response to evolutionary and structural constraints. Plant Sci. 255, 1-11 (2017). 140. Ananikov, V.P., Gayduk, K.A., Beletskaya, I.P., Khrustalev, V.N. & Antipin, M.Y. Catalyst Leaching as an Efficient Tool for Constructing New Catalytic Reactions: Application to the Synthesis of Cyclic Vinyl Sulfides and Vinyl Selenides. Eur. J. Inorg. Chem. 2009, 1149-1161 (2009). 141. Fong, J. et al. Disulfide Bond-Containing Ajoene Analogues As Novel Quorum Sensing Inhibitors of Pseudomonas aeruginosa. J. Med. Chem. 60, 215-227 (2017).

304

142. Saito, I. & Fukui, S. Studies on the oxidation products of lipoic acid. J. Vitaminol. 13, 115-121 (1967). 143. Müller, A., Knaack, M. & Olbrich, A. NMR Spectroscopic Characterization of Isomeric S‐Oxides Derived from α‐Lipoic Acid. Magn. Reson. Chem. 35, 111-114 (1998). 144. Hayward, L.J. et al. Decreased metallation and activity in subsets of mutant superoxide dismutases associated with familial amyotrophic lateral sclerosis. J. Biol. Chem. 277, 15923-15931 (2002). 145. Doucette, P.A. et al. Dissociation of Human Copper-Zinc Superoxide Dismutase Dimers Using Chaotrope and Reductant: Insights into the Molecular Basis for Dimer Stability. J. Biol. Chem. 279, 54558-54566 (2004). 146. Frisch, M.J. et al. (Wallingford CT; 2009). 147. Zhao, Y. & Truhlar, D.G. The M06 suite of density functionals for main group thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements: two new functionals and systematic testing of four M06-class functionals and 12 other functionals. Theor. Chem. Acc. 120, 215-241 (2008). 148. Pascual-ahuir, J.L., Silla, E. & Tuñon, I. GEPOL: An improved description of molecular surfaces. III. A new algorithm for the computation of a solvent-excluding surface. J. Comput. Chem. 15, 1127-1138 (1994). 149. Miertus,̃ S. & Tomasi, J. Approximate evaluations of the electrostatic free energy and internal energy changes in solution processes. Chem. Phys. 65, 239-245 (1982). 150. Miertuš, S., Scrocco, E. & Tomasi, J. Electrostatic interaction of a solute with a continuum. A direct utilizaion of AB initio molecular potentials for the prevision of solvent effects. Chem. Phys. 55, 117-129 (1981). 151. Rodríguez-Guerra Pedregal, J., Gómez-Orellana, P. & Maréchal, J.-D. ESIgen: Electronic Supporting Information Generator for Computational Chemistry Publications. J. Chem. Inf. Model. 58, 561-564 (2018). 152. Rothwarf, D.M. & Scheraga, H.A. Equilibrium and kinetic constants for the thiol- disulfide interchange reaction between glutathione and dithiothreitol. Proc. Natl. Acad. Sci. U.S.A. 89, 7944-7948 (1992). 153. Singh, J. et al. Structure-Based Design of a Potent, Selective, and Irreversible Inhibitor of the Catalytic Domain of the erbB Receptor Subfamily of Protein Tyrosine Kinases. J. Med. Chem. 40, 1130-1135 (1997). 154. Singh, R. & Whitesides George, M. Thiol—disulfide interchange. (John Wiley & Sons Ltd, 2010). 155. Gupta, V. & Carroll, K.S. Sulfenic acid chemistry, detection and cellular lifetime(). Biochim. Biophys. Acta 1840, 847-875 (2014). 156. Thapa, B. & Schlegel, H.B. Density Functional Theory Calculation of pKa’s of Thiols in Aqueous Solution Using Explicit Water Molecules and the Polarizable Continuum Model. J. Phys. Chem. A 120, 5726-5735 (2016). 157. Singh, R. & Whitesides, G.M. Comparisons of rate constants for thiolate-disulfide interchange in water and in polar aprotic solvents using dynamic proton NMR line shape analysis. J. Am. Chem. Soc. 112, 1190-1197 (1990). 158. Ullmann, G.M. Relations between Protonation Constants and Titration Curves in Polyprotic Acids: A Critical View. J. Phys. Chem. B 107, 1263-1271 (2003).

305

159. Reed, L.J., DeBusk, B.G., Gunsalus, I.C. & Hornberger, C.S. Crystalline α - Lipoic Acid: A Catalytic Agent Associated with Pyruvate Dehydrogenase. Science 114, 93-94 (1951). 160. Solmonson, A. & DeBerardinis, R.J. Lipoic acid metabolism and mitochondrial redox regulation. J. Biol. Chem. 293, 7522-7530 (2018). 161. Rosenberg, H.R. & Culik, R. Effect of α-lipoic acid on vitamin C and vitamin E deficiencies. Arch. Biochem. Biophys. 80, 86-93 (1959). 162. Ou, P., Tritschler, H.J. & Wolff, S.P. Thioctic (lipoic) acid: a therapeutic metal-chelating antioxidant? Biochem. Pharmacol. 50, 123-126 (1995). 163. Feldman, E.L. Oxidative stress and diabetic neuropathy: a new understanding of an old problem. J. Clin. Invest. 111, 431-433 (2003). 164. Ishida, H., Kisanuki, A. & Endo, K. Ring-Opening Polymerization of Aromatic 6- Membered Cyclic Disulfide and Characterization of the Polymer. Polym. J. 41, 110 (2008). 165. Yu, H., Wang, Y., Yang, H., Peng, K. & Zhang, X. Injectable self-healing hydrogels formed via thiol/disulfide exchange of thiol functionalized F127 and dithiolane modified PEG. J. Mater. Chem. B 5, 4121-4127 (2017). 166. Abegg, D. et al. Strained Cyclic Disulfides Enable Cellular Uptake by Reacting with the Transferrin Receptor. J. Am. Chem. Soc. 139, 231-238 (2017). 167. Morelli, P. & Matile, S. Sidechain Engineering in Cell-Penetrating Poly(disulfide)s. Helv. Chim. Acta 100, e1600370 (2017). 168. Chuard, N. et al. Strain-Promoted Thiol-Mediated Cellular Uptake of Giant Substrates: Liposomes and Polymersomes. Angew. Chem. Int. Ed. Engl. 56, 2947-2950 (2017). 169. Dong, Y., Abaci, S., Shannon, C. & Bozack, M.J. Self-Assembly and Electrochemical Desorption of Thioctic Acid Monolayers on Gold Surfaces. Langmuir 19, 8922-8926 (2003). 170. Roux, S. et al. Synthesis, Characterization of Dihydrolipoic Acid Capped Gold Nanoparticles, and Functionalization by the Electroluminescent Luminol. Langmuir 21, 2526-2536 (2005). 171. Abad, J.M., Mertens, S.F.L., Pita, M., Fernández, V.M. & Schiffrin, D.J. Functionalization of Thioctic Acid-Capped Gold Nanoparticles for Specific Immobilization of Histidine-Tagged Proteins. J. Am. Chem. Soc. 127, 5689-5694 (2005). 172. Donnelly, D.P. et al. Cyclic Thiosulfinates and Cyclic Disulfides Selectively Cross-Link Thiols While Avoiding Modification of Lone Thiols. J. Am. Chem. Soc. 140, 7377-7380 (2018). 173. Allinger, N.L. Conformational analysis. 130. MM2. A hydrocarbon force field utilizing V1 and V2 torsional terms. J. Am. Chem. Soc. 99, 8127-8134 (1977). 174. Bachrach, S.M., Woody, J.T. & Mulhearn, D.C. Effect of ring strain on the thiolate- disulfide exchange. A computational study. J. Org. Chem. 67, 8983-8990 (2002). 175. Hanson, P., Hendrickx, R.A.A.J. & Lindsay Smith, J.R. A kinetic investigation, supported by theoretical calculations, of steric and ring strain effects on the oxidation of sulfides and sulfoxides by dimethyldioxirane in acetone. New J. Chem. 34, 65-84 (2010). 176. Boatz, J.A., Gordon, M.S. & Hilderbrandt, R.L. Structure and bonding in cycloalkanes and monosilacycloalkanes. J. Am. Chem. Soc. 110, 352-358 (1988). 177. Dumont, É., Loos, P.-F. & Assfeld, X. Effect of ring strain on disulfide electron attachment. Chem. Phys. Lett. 458, 276-280 (2008).

306

178. Bickelhaupt, F.M. & Houk, K.N. Analyzing Reaction Rates with the Distortion/Interaction-Activation Strain Model. Angew. Chem., Int. Ed. 56, 10070-10086 (2017). 179. Ess, D.H. & Houk, K.N. Theory of 1,3-Dipolar Cycloadditions: Distortion/Interaction and Frontier Molecular Orbital Models. J. Am. Chem. Soc. 130, 10187-10198 (2008). 180. van Zeist, W.-J. & Bickelhaupt, F.M. The activation strain model of chemical reactivity. Org. Biomol. Chem. 8, 3118-3127 (2010). 181. Levandowski, B.J., Hamlin, T.A., Bickelhaupt, F.M. & Houk, K.N. Role of Orbital Interactions and Activation Strain (Distortion Energies) on Reactivities in the Normal and Inverse Electron-Demand Cycloadditions of Strained and Unstrained Cycloalkenes. J. Org. Chem. 82, 8668-8675 (2017). 182. Cheong, P.H.Y. et al. Indolyne and Aryne Distortions and Nucleophilic Regioselectivites. J. Am. Chem. Soc. 132, 1267-1269 (2010). 183. Yu, P. et al. Distortion-Controlled Reactivity and Molecular Dynamics of Dehydro- Diels–Alder Reactions. J. Am. Chem. Soc. 138, 8247-8252 (2016). 184. Liu, F., Paton, R.S., Kim, S., Liang, Y. & Houk, K.N. Diels–Alder Reactivities of Strained and Unstrained Cycloalkenes with Normal and Inverse-Electron-Demand Dienes: Activation Barriers and Distortion/Interaction Analysis. J. Am. Chem. Soc. 135, 15642-15649 (2013). 185. Green, A.G., Liu, P., Merlic, C.A. & Houk, K.N. Distortion/Interaction Analysis Reveals the Origins of Selectivities in Iridium-Catalyzed C–H Borylation of Substituted Arenes and 5-Membered Heterocycles. J. Am. Chem. Soc. 136, 4575-4583 (2014). 186. Lopez, S.A., Pourati, M., Gais, H.-J. & Houk, K.N. How Torsional Effects Cause Attack at Sterically Crowded Concave Faces of Bicyclic Alkenes. J. Org. Chem. 79, 8304-8312 (2014). 187. Bura, T. et al. Direct heteroarylation polymerization: guidelines for defect-free conjugated polymers. Chem. Sci. 8, 3913-3925 (2017). 188. Lopez, S.A. & Houk, K.N. Alkene Distortion Energies and Torsional Effects Control Reactivities, and Stereoselectivities of Azide Cycloadditions to Norbornene and Substituted Norbornenes. J. Org. Chem. 78, 1778-1783 (2013). 189. Alabugin, I.V., Manoharan, M. & Zeidan, T.A. Homoanomeric Effects in Six-Membered Heterocycles. J. Am. Chem. Soc. 125, 14014-14031 (2003). 190. Alabugin, I.V. Stereoelectronic Interactions in Cyclohexane, 1,3-Dioxane, 1,3-Oxathiane, and 1,3-Dithiane: W-Effect, σC-X ↔ σ*C-H Interactions, Anomeric EffectWhat Is Really Important? J. Org. Chem. 65, 3910-3919 (2000). 191. Freitas, M.P. The anomeric effect on the basis of natural bond orbital analysis. Org. Biomol. Chem. 11, 2885-2890 (2013). 192. Cramer, C.J., Truhlar, D.G. & French, A.D. Exo-anomeric effects on energies and geometries of different conformations of glucose and related systems in the gas phase and aqueous solution. Carbohydr. Res. 298, 1-14 (1997). 193. Mo, Y., Jiao, H. & von Rague Schleyer, P. Hyperconjugation effect in substituted methyl boranes: an orbital deletion procedure analysis. J. Org. Chem. 69, 3493-3499 (2004). 194. Alabugin, I.V. Stereoelectronic Interactions in Cyclohexane, 1,3-Dioxane, 1,3-Oxathiane, and 1,3-Dithiane: W-Effect, σC-X ↔ σ*C-H Interactions, Anomeric EffectWhat Is Really Important? The Journal of Organic Chemistry 65, 3910-3919 (2000).

307

195. Juaristi, E., dos Passos Gomes, G., Terent’ev, A.O., Notario, R. & Alabugin, I.V. Stereoelectronic Interactions as a Probe for the Existence of the Intramolecular α-Effect. J. Am. Chem. Soc. 139, 10799-10813 (2017). 196. Gomes, G.D.P., Vil, V., Terent'ev, A. & Alabugin, I.V. Stereoelectronic source of the anomalous stability of bis-peroxides. Chem. Sci. 6, 6783-6791 (2015). 197. Glendening, E.D., Landis, C.R. & Weinhold, F. NBO 6.0: Natural bond orbital analysis program. J. Comput. Chem. 34, 1429-1437 (2013). 198. Wiberg, K.B. Bent Bonds in Organic Compounds. Acc. Chem. Res. 29, 229-234 (1996). 199. . (ed. S.R. 2018-3) (Maestro, Schrödinger, LLC, New York, NY; 2018). 200. Harder, E. et al. OPLS3: A Force Field Providing Broad Coverage of Drug-like Small Molecules and Proteins. J. Chem. Theory Comput. 12, 281-296 (2016). 201. Frisch, M.J. et al. (Wallingford, CT; 2016). 202. Purvis, G.D. & Bartlett, R.J. A full coupled‐cluster singles and doubles model: The inclusion of disconnected triples. J. Chem. Phys. 76, 1910-1918 (1982). 203. Legault, C. CYLview User Manual. (2012). 204. Rodriguez-Guerra Pedregal, J., Gomez-Orellana, P. & Marechal, J.D. ESIgen: Electronic Supporting Information Generator for Computational Chemistry Publications. J. Chem. Inf. Model 58, 561-564 (2018). 205. Gregersen, N., Bross, P., Vang, S. & Christensen, J.H. Protein Misfolding and Human Disease. Annu. Rev. Genomics Hum. Genet. 7, 103-124 (2006). 206. Arakawa, T., Ejima, D., Kita, Y. & Tsumoto, K. Small molecule pharmacological chaperones: From thermodynamic stabilization to pharmaceutical drugs. BBA Proteins Proteom. 1764, 1677-1687 (2006). 207. Morello, J.-P., Bouvier, M., Petäjä-Repo, U.E. & Bichet, D.G. Pharmacological chaperones: a new twist on receptor folding. Trends Pharmacol. Sci. 21, 466-469 (2000). 208. Bernier, V., Lagacé, M., Bichet, D.G. & Bouvier, M. Pharmacological chaperones: potential treatment for conformational diseases. Trends Endocrinol. Metab. 15, 222-228 (2004). 209. Quintas, A., Saraiva, M.J. & Brito, R.M. The tetrameric protein transthyretin dissociates to a non-native monomer in solution. A novel model for amyloidogenesis. J. Biol. Chem. 274, 32943-32949 (1999). 210. Miroy, G.J. et al. Inhibiting transthyretin amyloid fibril formation via protein stabilization. Proc. Natl. Acad. Sci. U. S. A. 93, 15051-15056 (1996). 211. Petrassi, H.M., Klabunde, T., Sacchettini, J. & Kelly, J.W. Structure-Based Design of N- Phenyl Phenoxazine Transthyretin Amyloid Fibril Inhibitors. J. Am. Chem. Soc. 122, 2178-2192 (2000). 212. Waddington Cruz, M. & Benson, M.D. A Review of Tafamidis for the Treatment of Transthyretin-Related Amyloidosis. Neurology and Therapy 4, 61-79 (2015). 213. Maurer, M.S. et al. Tafamidis Treatment for Patients with Transthyretin Amyloid Cardiomyopathy. N. Engl. J. Med. 379, 1007-1016 (2018). 214. Gurney, M.E., Liu, R., Althaus, J.S., Hall, E.D. & Becker, D.A. Mutant Cu,Zn superoxide dismutase in motor neuron disease. Age 21, 85-89 (1998). 215. Cleveland, D.W. & Rothstein, J.D. From Charcot to Lou Gehrig: deciphering selective motor neuron death in ALS. Nat. Rev. Neurosci. 2, 806-819 (2001). 216. Valentine, J.S. & Hart, P.J. Misfolded CuZnSOD and amyotrophic lateral sclerosis. Proc. Natl. Acad. Sci. U. S. A. 100, 3617-3622 (2003).

308

217. Rakhit, R. et al. Monomeric Cu,Zn-superoxide Dismutase Is a Common Misfolding Intermediate in the Oxidation Models of Sporadic and Familial Amyotrophic Lateral Sclerosis. J. Biol. Chem. 279, 15499-15504 (2004). 218. Ezzi, S.A., Urushitani, M. & Julien, J.-P. Wild-type superoxide dismutase acquires binding and toxic properties of ALS-linked mutant forms through oxidation. J. Neurochem. 102, 170-178 (2007). 219. Grad, L.I. et al. Intermolecular transmission of superoxide dismutase 1 misfolding in living cells. Proc. Natl. Acad. Sci. U.S.A. 108, 16398-16403 (2011). 220. Bosco, D.A. et al. Wild-type and mutant SOD1 share an aberrant conformation and a common pathogenic pathway in ALS. Nature neuroscience 13, 1396-1403 (2010). 221. Schmitt, N.D. & Agar, J.N. Parsing disease-relevant protein modifications from epiphenomena: perspective on the structural basis of SOD1-mediated ALS. Journal of mass spectrometry : JMS 52, 480-491 (2017). 222. Wang, Q., Johnson, J.L., Agar, N.Y. & Agar, J.N. Protein aggregation and protein instability govern familial amyotrophic lateral sclerosis patient survival. PLoS Biol. 6, e170 (2008). 223. Hough, M.A. et al. Dimer destabilization in superoxide dismutase may result in disease- causing properties: Structures of motor neuron disease mutants. Proc. Natl. Acad. Sci. U.S.A. 101, 5976-5981 (2004). 224. Vassall, K.A., Stathopulos, P.B., Rumfeldt, J.A.O., Lepock, J.R. & Meiering, E.M. Equilibrium Thermodynamic Analysis of Amyotrophic Lateral Sclerosis-Associated Mutant Apo Cu,Zn Superoxide Dismutases. Biochemistry 45, 7366-7379 (2006). 225. Ray, S.S., Nowak, R.J., Brown, R.H. & Lansbury, P.T. Small-molecule-mediated stabilization of familial amyotrophic lateral sclerosis-linked superoxide dismutase mutants against unfolding and aggregation. Proc. Natl. Acad. Sci. U.S.A. 102, 3639-3644 (2005). 226. Wright, G.S.A., Antonyuk, S.V., Kershaw, N.M., Strange, R.W. & Samar Hasnain, S. Ligand binding and aggregation of pathogenic SOD1. Nat. Commun. 4, 1758 (2013). 227. Adams, H.S. & Ludwig, M. (Google Patents, 1930). 228. Bell, R.G. & Smith, H.W. Preliminary report on clinical trials of antabuse. Can. Med. Assoc. J. 60, 286-288 (1949). 229. Graham, W.D. In vitro inhibition of liver aldehyde dehydrogenase by tetraethylthiuram disulphide. The Journal of pharmacy and pharmacology 3, 160-168 (1951). 230. Olbe, L., Carlsson, E. & Lindberg, P. A proton-pump inhibitor expedition: the case histories of omeprazole and esomeprazole. Nat. Rev. Drug Discov. 2, 132-139 (2003). 231. Li, D. et al. BIBW2992, an irreversible EGFR/HER2 inhibitor highly effective in preclinical lung cancer models. Oncogene 27, 4702-4711 (2008). 232. Nelson, V., Ziehr, J., Agulnik, M. & Johnson, M. Afatinib: emerging next-generation tyrosine kinase inhibitor for NSCLC. Onco. Targets Ther. 6, 135-143 (2013). 233. Capper, M.J. et al. The cysteine-reactive small molecule ebselen facilitates effective SOD1 maturation. Nat. Commun. 9, 1693 (2018). 234. Deng, H.X. et al. Amyotrophic lateral sclerosis and structural defects in Cu,Zn superoxide dismutase. Science (New York, N.Y.) 261, 1047-1051 (1993). 235. Elam, J.S. et al. Amyloid-like filaments and water-filled nanotubes formed by SOD1 mutant proteins linked to familial ALS. Nat. Struct. Biol. 10, 461-467 (2003).

309

236. Stathopulos, P.B. et al. Cu/Zn superoxide dismutase mutants associated with amyotrophic lateral sclerosis show enhanced formation of aggregates Proc. Natl. Acad. Sci. U. S. A. 100, 7021-7026 (2003). 237. Niesen, F.H., Berglund, H. & Vedadi, M. The use of differential scanning fluorimetry to detect ligand interactions that promote protein stability. Nat. Protoc. 2, 2212-2221 (2007). 238. DiDonato, M. et al. ALS mutants of human superoxide dismutase form fibrous aggregates via framework destabilization. J. Mol. Biol. 332, 601-615 (2003). 239. Chen, P.C. & Hub, J.S. Validating solution ensembles from molecular dynamics simulation by wide-angle X-ray scattering data. Biophys. J. 107, 435-447 (2014). 240. Knight, C.J. & Hub, J.S. WAXSiS: a web server for the calculation of SAXS/WAXS curves based on explicit-solvent molecular dynamics. Nucleic Acids Res. 43, W225-230 (2015). 241. Krieger, E. & Vriend, G. YASARA View—molecular graphics for all devices—from smartphones to workstations. Bioinformatics 30, 2981-2982 (2014). 242. Krieger, E. & Vriend, G. New ways to boost molecular dynamics simulations. J. Comput. Chem. 36, 996-1007 (2015). 243. Pratt, A.J. et al. Aggregation propensities of superoxide dismutase G93 hotspot mutants mirror ALS clinical phenotypes. Proc. Natl. Acad. Sci. U. S. A. 111, E4568-E4576 (2014). 244. Galaleldeen, A. et al. Structural and biophysical properties of metal-free pathogenic SOD1 mutants A4V and G93A. Arch. Biochem. Biophys. 492, 40-47 (2009). 245. Tu, P.H. et al. Transgenic mice carrying a human mutant superoxide dismutase transgene develop neuronal cytoskeletal pathology resembling human amyotrophic lateral sclerosis lesions. Proc. Natl. Acad. Sci. U. S. A. 93, 3155-3160 (1996). 246. Gurney, M.E. et al. Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation. Science (New York, N.Y.) 264, 1772-1775 (1994). 247. Chabaud, B., Labaudinière, R. & Christol, H. Removal of Thioacetal Protecting Groups by Cerium (IV) Ammonium Nitrate - Hydrolysis of 1,3-Dithiannes AU - Cristau, Henri- Jean. Synth. Commun. 11, 423-428 (1981). 248. Hopkins, J.B., Gillilan, R.E. & Skou, S. BioXTAS RAW: improvements to a free open- source program for small-angle X-ray scattering data reduction and analysis. J. Appl. Crystallogr. 50, 1545-1553 (2017). 249. Franke, D. et al. ATSAS 2.8: a comprehensive data analysis suite for small-angle scattering from macromolecular solutions. J. Appl. Crystallogr. 50, 1212-1225 (2017). 250. Svergun, D. Determination of the regularization parameter in indirect-transform methods using perceptual criteria. J. Appl. Crystallogr. 25, 495-503 (1992). 251. Svergun, D.I., Petoukhov, M.V. & Koch, M.H. Determination of domain structure of proteins from X-ray solution scattering. Biophys. J. 80, 2946-2953 (2001). 252. Pettersen, E.F. et al. UCSF Chimera--a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605-1612 (2004). 253. Smith, L.M. & Kelleher, N.L. Proteoform: a single term describing protein complexity. Nat Meth 10, 186-187 (2013). 254. Shortreed, M.R. et al. Elucidating Proteoform Families from Proteoform Intact-Mass and Lysine-Count Measurements. Journal of Proteome Research 15, 1213-1221 (2016).

310

255. Mann, M. & Kelleher, N.L. Precision proteomics: The case for high resolution and high mass accuracy. Proceedings of the National Academy of Sciences 105, 18132 (2008). 256. Shaw, J.B. et al. Complete Protein Characterization Using Top-Down Mass Spectrometry and Ultraviolet Photodissociation. J Am Chem Soc 135, 12646-12651 (2013). 257. Ge, Y. et al. Top down characterization of larger proteins (45 kDa) by electron capture dissociation mass spectrometry. J Am Chem Soc 124, 672-678 (2002). 258. Siuti, N. & Kelleher, N.L. Decoding protein modifications using top-down mass spectrometry. Nat Methods 4, 817-821 (2007). 259. Ge, Y., Rybakova, I.N., Xu, Q.G. & Moss, R.L. Top-down high-resolution mass spectrometry of cardiac myosin binding protein C revealed that truncation alters protein phosphorylation state. P Natl Acad Sci USA 106, 12658-12663 (2009). 260. Toby, T.K., Fornelli, L. & Kelleher, N.L. Progress in Top-Down Proteomics and the Analysis of Proteoforms. Annual review of analytical chemistry 9, 499-519 (2016). 261. Fenn, J.B., Mann, M., Meng, C.K., Wong, S.F. & Whitehouse, C.M. Electrospray ionization for mass spectrometry of large biomolecules. Science 246, 64-71 (1989). 262. Loo, J.A., Edmonds, C.G. & Smith, R.D. Tandem mass spectrometry of very large molecules: serum albumin sequence information from multiply charged ions formed by electrospray ionization. Analytical Chemistry 63, 2488-2499 (1991). 263. Reid, G.E. & McLuckey, S.A. ‘Top down’ protein characterization via tandem mass spectrometry. Journal of Mass Spectrometry 37, 663-675 (2002). 264. Alexandrov, M.L. et al. Extraction of ions from solutions under atmospheric pressure as a method for mass spectrometric analysis of bioorganic compounds. Rapid Communications in Mass Spectrometry 22, 267-270 (2008). 265. Kelleher, N.L. et al. Top down versus bottom up protein characterization by tandem high- resolution mass spectrometry. J Am Chem Soc 121, 806-812 (1999). 266. Konermann, L., Ahadi, E., Rodriguez, A.D. & Vahidi, S. (ACS Publications, 2012). 267. Lohnes, K. et al. Combining High-throughput MALDI-TOF Mass Spectrometry and Isoelectric Focusing Gel Electrophoresis for Virtual 2DGel-based Proteomics. Methods (San Diego, Calif.) 104, 163-169 (2016). 268. Fenselau, C. & Demirev, P.A. Characterization of intact microorganisms by MALDI mass spectrometry. Mass Spectrometry Reviews 20, 157-171 (2001). 269. Shevchenko, A., Wilm, M., Vorm, O. & Mann, M. Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal Chem 68, 850-858 (1996). 270. Washburn, M.P., Wolters, D. & Yates, J.R., 3rd Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nature biotechnology 19, 242-247 (2001). 271. Manza, L.L., Stamer, S.L., Ham, A.J., Codreanu, S.G. & Liebler, D.C. Sample preparation and digestion for proteomic analyses using spin filters. Proteomics 5, 1742- 1745 (2005). 272. Wisniewski, J.R., Zougman, A., Nagaraj, N. & Mann, M. Universal sample preparation method for proteome analysis. Nat Meth 6, 359-362 (2009). 273. Aebersold, R. & Mann, M. Mass spectrometry-based proteomics. Nature 422, 198-207 (2003). 274. Auclair, J.R. et al. Artifacts to avoid while taking advantage of top-down mass spectrometry based detection of protein S-thiolation. Proteomics 14, 1152-1157 (2014).

311

275. Auclair, J.R. et al. Post-Translational Modification by Cysteine Protects Cu/Zn- Superoxide Dismutase from Oxidative Damage. Biochemistry 52, 6137-6144 (2013). 276. Ayaz-Guner, S., Zhang, J., Li, L., Walker, J.W. & Ge, Y. In Vivo Phosphorylation Site Mapping in Mouse Cardiac Troponin I by High Resolution Top-Down Electron Capture Dissociation Mass Spectrometry: Ser22/23 Are the Only Sites Basally Phosphorylated. Biochemistry 48, 8161-8170 (2009). 277. Martinez-Garcia, E. et al. The MMSET histone methyl transferase switches global histone methylation and alters gene expression in t(4;14) multiple myeloma cells. Blood 117, 211-220 (2011). 278. Wu, D., Struwe, W.B., Harvey, D.J., Ferguson, M.A.J. & Robinson, C.V. N-glycan microheterogeneity regulates interactions of plasma proteins. Proceedings of the National Academy of Sciences 115, 8763 (2018). 279. Doll, S. & Burlingame, A.L. Mass Spectrometry-Based Detection and Assignment of Protein Posttranslational Modifications. ACS Chemical Biology 10, 63-71 (2015). 280. Chait, B.T. Mass Spectrometry: Bottom-Up or Top-Down? Science 314, 65 (2006). 281. Wessel, D. & Flügge, U.I. A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Analytical Biochemistry 138, 141-143 (1984). 282. Fornelli, L. et al. Top-down proteomics: Where we are, where we are going? Journal of proteomics (2017). 283. Dang, X. et al. The first pilot project of the consortium for top-down proteomics: a status report. Proteomics 14, 1130-1140 (2014). 284. Cech, N.B. & Enke, C.G. Practical implications of some recent studies in electrospray ionization fundamentals. Mass Spectrom Rev 20, 362-387 (2001). 285. Konermann, L., Ahadi, E., Rodriguez, A.D. & Vahidi, S. Unraveling the Mechanism of Electrospray Ionization. Analytical Chemistry 85, 2-9 (2013). 286. Metwally, H., McAllister, R.G. & Konermann, L. Exploring the Mechanism of Salt- Induced Signal Suppression in Protein Electrospray Mass Spectrometry Using Experiments and Molecular Dynamics Simulations. Analytical Chemistry 87, 2434-2442 (2015). 287. Compton, P.D., Zamdborg, L., Thomas, P.M. & Kelleher, N.L. On the Scalability and Requirements of Whole Protein Mass Spectrometry. Analytical Chemistry 83, 6868-6874 (2011). 288. DeHart, C.J. & Fornelli, L. (National Resource for Translational and Developmental Proteomics 2017). 289. Karas, M., Bahr, U. & Dülcks, T. Nano-electrospray ionization mass spectrometry: addressing analytical problems beyond routine. Fresenius' Journal of Analytical Chemistry 366, 669-676 (2000). 290. Juraschek, R., Dulcks, T. & Karas, M. Nanoelectrospray--more than just a minimized- flow electrospray ionization source. J Am Soc Mass Spectrom 10, 300-308 (1999). 291. Evershed, R.P., Robertson, D.H., Beynon, R.J. & Green, B.N. Application of electrospray ionization mass spectrometry with maximum-entropy analysis to allelic 'fingerprinting' of major urinary proteins. Rapid communications in mass spectrometry : RCM 7, 882-886 (1993). 292. Ferrige, A.G. et al. Disentangling electrospray spectra with maximum entropy. Rapid Communications in Mass Spectrometry 6, 707-711 (1992).

312

293. Maquin, F., Schoot, B.M., Devaux, P.G. & Green, B.N. Molecular weight determination of recombinant interleukin 2 and interferon gamma by electrospray ionization mass spectroscopy. Rapid communications in mass spectrometry : RCM 5, 299-302 (1991). 294. Loo, J.A., Udseth, H.R. & Smith, R.D. Peptide and protein analysis by electrospray ionization-mass spectrometry and capillary electrophoresis-mass spectrometry. Analytical Biochemistry 179, 404-412 (1989). 295. Morris, H.R. et al. High sensitivity collisionally-activated decomposition tandem mass spectrometry on a novel quadrupole/orthogonal-acceleration time-of-flight mass spectrometer. Rapid communications in mass spectrometry : RCM 10, 889-896 (1996). 296. Kelleher, N.L., Senko, M.W., Little, D.P., O'Connor, P.B. & McLafferty, F.W. Complete large-molecule high-resolution mass spectra from 50-femtomole microvolume injection. Journal of the American Society for Mass Spectrometry 6, 220-221 (1995). 297. LeDuc, R.D., Schwammle, V., Shortreed, M.R. & Cesnik, A.J. ProForma: A Standard Proteoform Notation. 17, 1321-1325 (2018). 298. Shaw, K.L., Grimsley, G.R., Yakovlev, G.I., Makarov, A.A. & Pace, C.N. The effect of net charge on the solubility, activity, and stability of ribonuclease Sa. Protein Science : A Publication of the Protein Society 10, 1206-1215 (2001). 299. Konermann, L. Addressing a Common Misconception: Ammonium Acetate as Neutral pH “Buffer” for Native Electrospray Mass Spectrometry. Journal of The American Society for Mass Spectrometry 28, 1827-1835 (2017). 300. Uhlén, M. et al. Tissue-based map of the human proteome. Science 347 (2015). 301. Overington, J.P., Al-Lazikani, B. & Hopkins, A.L. How many drug targets are there? Nature Reviews Drug Discovery 5, 993 (2006). 302. Laganowsky, A., Reading, E., Hopper, J.T.S. & Robinson, C.V. Mass Spectrometry of Intact Membrane Protein Complexes. Nature protocols 8, 639-651 (2013). 303. Erde, J., Loo, R.R.O. & Loo, J.A. Enhanced FASP (eFASP) to Increase Proteome Coverage and Sample Recovery for Quantitative Proteomic Experiments. Journal of Proteome Research 13, 1885-1895 (2014). 304. Doucette, A.A., Vieira, D.B., Orton, D.J. & Wall, M.J. Resolubilization of Precipitated Intact Membrane Proteins with Cold Formic Acid for Analysis by Mass Spectrometry. Journal of Proteome Research 13, 6001-6012 (2014). 305. Güray, M.Z., Zheng, S. & Doucette, A.A. Mass Spectrometry of Intact Proteins Reveals +98 u Chemical Artifacts Following Precipitation in Acetone. Journal of Proteome Research 16, 889-897 (2017). 306. Leney, A.C. & Heck, A.J.R. Native Mass Spectrometry: What is in the Name? Journal of the American Society for Mass Spectrometry 28, 5-13 (2017). 307. Lössl, P., Snijder, J. & Heck, A.J.R. Boundaries of Mass Resolution in Native Mass Spectrometry. Journal of The American Society for Mass Spectrometry 25, 906-917 (2014). 308. Gault, J. et al. High-resolution mass spectrometry of small molecules bound to membrane proteins. Nature methods 13, 333-336 (2016). 309. Belov, A.M. et al. Analysis of Proteins, Protein Complexes, and Organellar Proteomes Using Sheathless Capillary Zone Electrophoresis - Native Mass Spectrometry. Journal of The American Society for Mass Spectrometry 28, 2614-2634 (2017).

313

310. Belov, A.M. et al. Complementary middle-down and intact monoclonal antibody proteoform characterization by capillary zone electrophoresis - mass spectrometry. 39, 2069-2082 (2018). 311. Tran, J.C. & Doucette, A.A. Multiplexed size separation of intact proteins in solution phase for mass spectrometry. Anal Chem 81, 6201-6209 (2009). 312. Whitelegge, J.P., Gundersen, C.B. & Faull, K.F. Electrospray-ionization mass spectrometry of intact intrinsic membrane proteins. Protein Science : A Publication of the Protein Society 7, 1423-1430 (1998). 313. le Coutre, J. et al. Proteomics on Full-Length Membrane Proteins Using Mass Spectrometry. Biochemistry 39, 4237-4242 (2000). 314. Whitelegge, J.P. in HPLC of Peptides and Proteins: Methods and Protocols. (ed. M.-I. Aguilar) 323-339 (Springer New York, Totowa, NJ; 2004). 315. Souda, P., Ryan, C.M., Cramer, W.A. & Whitelegge, J. Profiling of integral membrane proteins and their post translational modifications using high-resolution mass spectrometry. Methods (San Diego, Calif.) 55, 330-336 (2011). 316. Souda, P., Ryan, C.M., Cramer, W.A. & Whitelegge, J. Profiling of integral membrane proteins and their post translational modifications using high-resolution mass spectrometry. Methods (San Diego, Calif.) 55, 330-336 (2011). 317. Whitelegge, J.P. Integral Membrane Proteins and Bilayer Proteomics. Analytical chemistry 85, 2558-2568 (2013). 318. Howery, A.E. et al. A designed inhibitor of a CLC antiporter blocks function through a unique binding mode. Chemistry & biology 19, 1460-1470 (2012). 319. Whitelegge, J.P. et al. Toward the bilayer proteome, electrospray ionization-mass spectrometry of large, intact transmembrane proteins. Proc Natl Acad Sci U S A 96, 10695-10698 (1999). 320. Cleland, T.P. et al. High-Throughput Analysis of Intact Human Proteins Using UVPD and HCD on an Orbitrap Mass Spectrometer. Journal of Proteome Research 16, 2072- 2079 (2017). 321. Ansong, C. et al. Top-down proteomics reveals a unique protein S-thiolation switch in Salmonella Typhimurium in response to infection-like conditions. Proc Natl Acad Sci U S A 110, 10153-10158 (2013). 322. Ntai, I. et al. Applying Label-Free Quantitation to Top Down Proteomics. Analytical Chemistry 86, 4961-4968 (2014). 323. Skinner, O.S. et al. Top-down characterization of endogenous protein complexes with native proteomics. Nature Chemical Biology 14, 36 (2017). 324. Porath, J. & Flodin, P. Gel filtration: a method for desalting and group separation. Nature 183, 1657-1659 (1959). 325. Laganowsky, A., Reading, E., Hopper, J.T.S. & Robinson, C.V. Mass Spectrometry of Intact Membrane Protein Complexes. Nat. Protoc. 8, 639-651 (2013). 326. Laganowsky, A. et al. Membrane proteins bind lipids selectively to modulate their structure and function. Nature 510, 172-175 (2014). 327. Wessel, D. & Flügge, U.I. A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal. Biochem. 138, 141-143 (1984). 328. Doucette, A.A., Vieira, D.B., Orton, D.J. & Wall, M.J. Resolubilization of precipitated intact membrane proteins with cold formic acid for analysis by mass spectrometry. J. Proteome Res. 13, 6001-6012 (2014).

314

329. Valeja, S.G. et al. Unit Mass Baseline Resolution for an Intact 148 kDa Therapeutic Monoclonal Antibody by FT-ICR Mass Spectrometry. Anal. Chem. 83, 8391-8395 (2011). 330. Li, H., Wolff, J.J., Van Orden, S.L. & Loo, J.A. Native Top-Down Electrospray Ionization-Mass Spectrometry of 158 kDa Protein Complex by High-Resolution Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. Anal. Chem. 86, 317-320 (2014). 331. Shaw, J.B. & Brodbelt, J.S. Extending the isotopically resolved mass range of Orbitrap mass spectrometers. Anal. Chem. 85, 8313-8318 (2013). 332. de la Mora, J.F. Electrospray ionization of large multiply charged species proceeds via Dole's charged residue mechanism. Anal. Chim. Acta 406, 93-104 (2000). 333. Compton, P.D., Zamdborg, L., Thomas, P.M. & Kelleher, N.L. On the Scalability and Requirements of Whole Protein Mass Spectrometry. Anal. Chem. 83, 6868-6874 (2011). 334. Belov, A.M. et al. Analysis of Proteins, Protein Complexes, and Organellar Proteomes Using Sheathless Capillary Zone Electrophoresis - Native Mass Spectrometry. J. Am. Soc. Mass Spectrom. 28, 2614-2634 (2017). 335. Konermann, L., Ahadi, E., Rodriguez, A.D. & Vahidi, S. Unraveling the Mechanism of Electrospray Ionization. Anal. Chem. 85, 2-9 (2013). 336. Iavarone, A.T. & Williams, E.R. Mechanism of charging and supercharging molecules in electrospray ionization. J. Am. Chem. Soc. 125, 2319-2327 (2003). 337. Kirshenbaum, N., Michaelevski, I. & Sharon, M. Analyzing large protein complexes by structural mass spectrometry. J Vis Exp. (2010). 338. Heck, A.J.R. Native mass spectrometry: a bridge between interactomics and structural biology. Nat. Methods 5, 927-933 (2008). 339. Loo, J.A. Studying noncovalent protein complexes by electrospray ionization mass spectrometry. Mass Spectrom Rev. 16, 1-23 (1997). 340. Belov, M.E. et al. From protein complexes to subunit backbone fragments: a multi-stage approach to native mass spectrometry. Anal. Chem. 85, 11163-11173 (2013). 341. Campuzano, I.D.G. et al. Native MS Analysis of Bacteriorhodopsin and an Empty Nanodisc by Orthogonal Acceleration Time-of-Flight, Orbitrap and Ion Cyclotron Resonance. Anal. Chem. 88, 12427-12436 (2016). 342. Lippens, J.L. et al. Fourier Transform-Ion Cyclotron Resonance Mass Spectrometry as a Platform for Characterizing Multimeric Membrane Protein Complexes. J. Am. Soc. Mass Spectrom. 29, 183-193 (2017). 343. Zhurov, K.O., Kozhinov, A.N., Fornelli, L. & Tsybin, Y.O. Distinguishing analyte from noise components in mass spectra of complex samples: where to cut the noise? Anal. Chem. 86, 3308-3316 (2014). 344. Valiente, M. et al. Binding of PTEN to Specific PDZ Domains Contributes to PTEN Protein Stability and Phosphorylation by Microtubule-associated Serine/Threonine Kinases. J. Biol. Chem. 280, 28936-28943 (2005). 345. Robinson, D.R. et al., Vol. 17 1646-1651 (Nat. Med., Nat. Med.; 2011). 346. Huang, N. et al. A Screen for Genomic Disorders of Infertility Identifies MAST2 Duplications Associated with Nonobstructive Azoospermia in Humans. Biol. Reprod. 93, 61 (2015).

315

347. Chong, Z.Z., Changyaleket, B., Xu, H., Dull, R.O. & Schwartz, D.E. Identifying S100B as a Biomarker and a Therapeutic Target For Brain Injury and Multiple Diseases. Curr. Med. Chem. 23, 1571-1596 (2016).

316