Copyright by Gayle Diane Burstein 2014

The Dissertation Committee for Gayle Diane Burstein Certifies that this is the approved version of the following dissertation:

An Investigation of the Irreversible Inhibition of Human Nω, Nω- Dimethylarginine Dimethylaminohydrolase (DDAH1)

Committee:

Walter Fast, Supervisor

Christian Whitman

Yan Jessie Zhang

Brent Iverson

David Hoffman An Investigation of the Irreversible Inhibition of Human Nω, Nω- Dimethylarginine Dimethylaminohydrolase (DDAH1)

by

Gayle Diane Burstein, B.S.

Dissertation Presented to the Faculty of the Graduate School of The University of Texas at Austin in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

The University of Texas at Austin August 2014

Dedication

For my loving family

Acknowledgements

Thank you to my advisor, Dr. Walter Fast. Dr. Fast was always available to talk through my scientific and personal difficulties. He believed in me even when I didn’t believe in myself and I could not be more thankful for that. Thank you to my friends and colleagues in the Fast lab, past and present, for all your insights, thoughts, and our scientific collaborations. Thank you to Dr. Art Monzigo and the Robertus lab for our collaborative efforts and their X-ray crystallographic work. Thank you to my friends that have helped me and pushed me to pursue my degree, your encouragement is always appreciated. Finally, thank you to Jordan Teitelbaum for pushing me to work hard and being there for me during the difficult times (and Lazer Wolf for his purrs).

v An Investigation of the Irreversible Inhibition of Human Nω, Nω- Dimethylarginine Dimethylaminohydrolase (DDAH1)

Gayle Diane Burstein, Ph.D. The University of Texas at Austin, 2014

Supervisor: Walter Fast

Nitric oxide synthases (NOS) are responsible for the production of nitric oxide (NO), an essential cell-signaling molecule, in mammals. There are three isoforms of NOS with widely different tissue distribution. The overproduction of NO is marked in many human disease states and cancers, however due to the similarities of the enzyme isoforms, targeting NOS for inhibition has proven challenging. Endogenously, the methylated arginines, Nω-monomethyl-L-arginine (NMMA) and asymmetric Nω, Nω- dimethyl-L-arginine (ADMA), inhibit NOS. Nω, Nω-Dimethylarginine dimethylaminohydrolase (DDAH1) metabolizes these methylated arginines and thus relieves NOS inhibition. The role of DDAH1 in the regulation of diseases such as cancer and septic shock is still being elucidated. It is thought that targeting DDAH1 for inhibition rather than NOS may circumvent many of the current problems with the treatment of NO overproduction such as isoform selectivity. My PhD studies focus on the synthesis of a series of irreversible inhibitors of DDAH1, an extensive study of their in vitro mode of inhibition, a comparison of analytical fitting methods, and the viability and efficacy of the inactivators in a human cell line. I also studied a potential endogenous inactivator of DDAH1, nitroxyl (HNO), a one-electron reduction product of NO. vi Table of Contents

List of Tables ...... x

List of Figures ...... xi

List of Schemes ...... xiii

Chapter 1: Background and Significance ...... 1 1.1: The Dual Nature of Nitric Oxide (NO) ...... 1 1.1.1: Nitric Oxide (NO) is an important cell signaling molecule ...... 1 1.1.2: NO overproduction is implicated in many diseases states...... 5 1.2: Dimethylarginine Dimethylaminohydrolase as a Means to Control NO Production ...... 7 1.2.1: Methylated arginines inhibit NOS ...... 7 1.2.2: Dimethylarginine Dimethylaminohydrolase (DDAH) contributes to the regulation of NO production ...... 9 1.2.3: Manipulation of DDAH can alter NO levels ...... 13 1.3 Towards the irreversible inhibition of Dimethylarginine Dimethylamniohydrolase ...... 14 1.4 References ...... 18

Chapter 2: Optimizing Chloroacetamidine Inactivators of DDAH1 and a Comparison of Fitting Methods ...... 26 2.1 Introduction ...... 26 2.2 Materials and Methods ...... 28 General Synthesis ...... 28 DDAH1 Expression and Purification ...... 30 Time-dependent Inactivation of DDAH1 by Irreversible Inhibitors ...31 Partition ratio ...... 33 Crystal Structure ...... 33 Arginase Assay ...... 34

In cell IC50 value determination ...... 34 In Cell Single Point Assay ...... 36

vii 3-(4,5-Dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)- 2-(4- sulfophenyl)-2H-tetrazolium (MTS) Assay ...... 36 2.3 Results and Discussion ...... 37 Kinetic Analysis of DDAH1 Time-dependent Inactivation by Chloroacetamidines...... 39 Cl-NIL acts as a mechanism based inhibitor ...... 49 Cl-NIL forms a covalent bond with the active-site cysteine ...... 50 Cl-NIL is selective for DDAH1 over Mn2+-containing human arginase I (Mn-hArgI) ...... 56 Compounds inhibit DDAH1 within cultured cells ...... 57 2.4 Conclusions ...... 63 2.5 References ...... 63

Chapter 3: Inhibition of Dimethylarginine Dimethylaminohydrolase by Angeli’s Salt ...... 67 3.1 Introduction ...... 67 3.2 Materials and Methods ...... 70 Materials ...... 70 Expression and Purification of DDAH1 ...... 70 Time-dependent Inactivation of DDAH1 by Angeli’s Salt ...... 71 Time-dependent Inactivation by Decomposed Angeli’s Salt ...... 72 Time-dependent Inactivation by Nitrite ...... 72 Time-dependent Inactivation by GSH Incubated Angeli’s Salt ...... 72 Acid Quench of Inactive DDAH1 ...... 72 LC-ESI-MS Analysis ...... 73 In Cell Inhibition of DDAH1 by Angeli’s Salt ...... 74 3.3 Results and Discussion ...... 75 Angeli’s Salt Covalently Inactivates DDAH1 ...... 75 HNO is the Predominant Inactivating Species of DDAH1 by Angeli’s Salt ...... 79 GSH Prevents DDAH1 Inactivation by Angeli’s Salt in vitro...... 82 Angeli’s Salt does not Inactivate DDAH1 in HEK293T Cells...... 84

viii 3.4 Conclusions ...... 86 3.5 References ...... 87

Appendix: Kinetics of Chloroacetamidines ...... 90

Bibliography ...... 108

Vita ...... 120

ix List of Tables

Table 1.1 Representative DDAH inhibitors ...... 16 Table 2.1 Inactivation constants for DDAH1 inactivators ...... 40 Table 2.2 Rate constants obtained during global fitting to the model in scheme 2.1 ...... 46 Table 2.3 Crystallographic Data ...... 51 Table A.1 Rate constants obtained during global fitting to the model in scheme 2.1 ...... 95 Table A.2 Rate constants obtained during global fitting to the model in scheme 2.1 ...... 101 Table A.3 Rate constants obtained during global fitting to the model in scheme 2.1 ...... 107

x

List of Figures

Figure 1.1 Nitric oxide (NO) and cGMP signaling cascade ...... 2 Figure 1.2 Production of NO by NOS ...... 3 Figure 1.3 NO metabolism ...... 5 Figure 1.4 Different NO concentrations activate multiple signaling pathways ...... 7 Figure 1.5 Methylated arginines ...... 8 Figure 1.6 Methylated arginine metabolism ...... 10 Figure 1.7 Substrate interactions with the active site of DDAH1 ...... 12 Figure 1.8 Proposed catalytic mechanism of DDAH1 ...... 13 Figure 2.1 Arginine metabolism and regulation...... 27 Figure 2.2 Substrates and selected inhibitors of DDAH1 ...... 28 Figure 2.3 DDAH1 substrates and synthesized covalent inactivator structures ....38 Figure 2.4 Inactivation of DDAH1 by Cl-NIL (Tsou method) ...... 41 Figure 2.5 Inactivation of DDAH1 by Cl-NIL (Kiz-Wilson and Morrison methods) ...... 43 Figure 2.6 Global fit of DDAH1 inactivation by Cl-NIL to a kinetic model ...... 45 Figure 2.7 FitSpace contour plot for rate constants in the kinetic model ...... 46 Figure 2.8 Partition ratio of Cl-NIL to DDAH1 ...... 49

Figure 2.9 Cl-NIL covalently bound to DDAH1 ...... 52 Figure 2.10 Overlay of DDAH1 bound with L-citrulline and Cl-NIL ...... 54 Figure 2.11 Overlay of DDAH1 bound with Cl-NIL, IPO, and IPnO ...... 55 Figure 2.12. Metabolism of L-arginine and regulation by DDAH1 ...... 56 Figure 2.13 Inhibition of Mn-hArgI by Cl-NIL ...... 57 Figure 2.14. Click-Cl labels DDAH1 active sites within cultured cells ...... 59

xi Figure 2.15 Chloroacetamidines inhibit DDAH1 in cultured HEK293T cells...... 60 Figure 2.16 Concentration dependent inhibition of DDAH1 in HEK293T cells by Cl- NIL...... 61 Figure 2.17 Toxicity of Cl-NIL on HEK293T cells...... 62 Figure 3.1 NO and HNO reactions in biology ...... 68 Figure 3.2 The metabolism of ADMA is regulated by DDAH1 and NO ...... 69 Figure 3.3 Angeli’s salt inactivation of DDAH1 ...... 77 Figure 3.4 ESI-MS of covalently modified DDAH1 ...... 78

- Figure 3.5 Time-dependence of DDAH1 inactivation by NO2 or HNO ...... 80 Figure 3.6 Inactivation of DDAH1 by freshly prepared and decomposed Angeli’s salt ...... 82 Figure 3.7 Inactivation of DDAH1 by Angeli’s salt before and after treatment with GSH ...... 84 Figure 3.8 In cell inhibition of DDAH1 by Angeli’s salt...... 86 Figure A.1 Inactivation of DDAH1 by Cl-NIO (Tsou method) ...... 91 Figure A.2 Inactivation of DDAH1 by Cl-NIO (Kitz-Wilson and Morrison methods) ...... 93 Figure A.3 Inactivation of DDAH1 by Cl-NIO (Kin-Tek Explorer) ...... 95 Figure A.4 Inactivation of DDAH1 by Me-Cl-NIL (Tsou method) ...... 96 Figure A.4 Inactivation of DDAH1 by Me-Cl-NIL (Kitz-Wilson method) ...... 99 Figure A.5 Inactivation of DDAH1 by Me-Cl-NIL (Kin-Tek Explorer) ...... 101 Figure A.6 Inactivation of DDAH1 by Me-Cl-NIO (Tsou method) ...... 103 Figure A.7 Inactivation of DDAH1 by Me-Cl-NI) (Kitz-Wilson method) ...... 105 Figure A.8 Inactivation of DDAH1 by Me-Cl-NIO (Kin-Tek Explorer) ...... 107

xii List of Schemes

Scheme 2.1 Global fitting model for enzyme inactivation ...... 44 Scheme 2.2 Inactivation of DDAH1 by chloroacetamidines ...... 47 Scheme 3.1 Angeli’s Salt decomposes nitroxyl and nitrite ...... 70 Scheme 3.2 Possible adducts formed by Angeli’s salt inactivation of DDAH1 ....79 Scheme 3.3 HNO dimerization ...... 81

xiii Chapter 1: Background and Significance

1.1: THE DUAL NATURE OF NITRIC OXIDE (NO)

1.1.1: Nitric Oxide (NO) is an important cell signaling molecule

Nitric oxide (NO) is a short-lived gaseous radical that functions primarily as a biological messenger (1). NO signaling cascades are involved in the respiratory, nervous, gastrointestinal, genitourinary, and cardiovascular systems (2). The NO radical acts as a signal transducer at low concentrations, first discovered to target the heme-containing protein soluble guanylate cyclase (sGC), where NO binds to heme-FeII as well as many other transition metal containing proteins (3). A conformational change is undergone upon NO binding sGC and confers a 400-fold increase of the conversion of GTP to cGMP (3). cGMP is a secondary messenger responsible for activating protein kinase-G (PKG), phosphodiesterases, and cyclic nucleotide-gated channel (4). PKG is thought to be responsible for smooth muscle cell relaxation and neurotransmission (4,5).

1 Figure 1.1 Nitric oxide (NO) and cGMP signaling cascade sGC catalyzes the transformation of guanosine triphosphate (GTP) into 3’5’-cyclic guanosine monophosphate (cGMP) when it is activated by the binding of NO to its heme moiety. cGMP activates protein kinase G (PKG) and cyclic nucleotide-gated channels which causes smooth muscle relaxation and activates neurotransmission. Figure adapted from (6).

Nitric oxide is synthesized from the oxidation of L-arginine by nitric oxide synthases (NOS, EC 1.14.13.39) and has many other targets besides sGC (5,7). There are three known isoforms of NOS in humans, endothelial NOS (eNOS or nos3), neuronal NOS (nNOS or nos1), and inducible NOS (iNOS or nos2). These isoforms share about 50% amino acid sequence identity (7). Endothelial produced NO is thought to stimulate the relaxation of the endothelial tissue of blood vessels regulating blood flow and pressure (8). In the central nervous system, NO acts as a neurotransmitter modulating responses to other neurotransmitters as a retrograde messenger and it has an important function for long-term potentiation (9,10). NO derived from iNOS takes part in the inflammatory response toward invading pathogens. The body makes use of the reactive nature of the NO radical to attack microorganisms (11). NOS is a P450-like enzyme that acts as a homodimer by first sequestering ZnII to generate binding sites for a heme cofactor, tetrahydrobiopterin (BH4) containing oxygenase domain, as well as flavin adenenine dinucleotide (FAD) and flavin mononucleotide (FMN) which are necessary to transfer electrons provided by NADPH (Figure 1.1) (2,11,12). The oxygenase and reductase domains are connected by a calmodulin (CaM)-binding region (12). When intracellular Ca2+ levels increase, Ca2+ binds to the CaM causing it to bind to NOS and activate NO production in the nNOS and eNOS isoforms (12). iNOS is fully active at basal Ca2+ levels (13).

2 H2N NH2 '()($ '()($ NO

NH +NADPH, O2 !%&$ *'+$ *'+$ !%&$

!"#$ ./ ,- $ !"#$ H2N O + - H3N COO

L-Arginine NH 01%$ 01%$

+ - H3N COO L-Citrulline

Figure 1.2 Production of NO by NOS

Nitric oxide synthases (NOS) dimerize and bind the cofactors tetrahydrobiopterin (BH4), heme, flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD). Once calmodulin (CaM) binds, the active enzyme catalyzes the oxidation of L-arginine to L- citrulline and nitric oxide (NO) and requires molecular oxygen and NADPH as co- substrates. Each NOS dimer coordinates a single zinc (Zn2+) atom. Figure adapted from (2).

Once synthesized, the NO radical can rapidly diffuse through cell membranes and interact with many biomolecules (7). In addition to reacting with sGC as described earlier, NO can directly react with the heme groups of O2 transport proteins and it reacts rapidly with other free radicals such as O2.- or lipid peroxyl radical (14,15). The most well studied NO reaction is the formation of the peroxynitrite (ONOO-) radical from the reaction of NO with superoxide (O2-), which leads to the oxidation of lipids and proteins (often in the form of 3-nitrotyrosine) (14,16). NO is oxidized to form another reactive species nitrogen dioxide (NO2) (17). In the presence of excess NO, NO2 is further converted to the nitrosating agent dinitrogen trioxide (N2O3) (14). N2O3 readily

- + + decomposes to nitrite (NO2 ) and the nitrosium cation (NO ); NO is primary protein

3 nitrosylating and DNA base deaminating species (18). These reactive nitrogen species (RNS) described along with reactive oxygen species (ROS) (superoxide, hydroxyl) are essential in the immune cell response for invading microorganisms, however, overproduction of ROS and RNS can be detrimental to native cellular functions. Nitrosative stresses, such as DNA deamination, nitration and nitrosylation, and oxidative stresses, such as lipid peroxidation and DNA strand breakage, can lead to inflammatory reactions, inhibit mitochondrial respiration, and eventually lead to cell apoptosis due to cellular damage (19). To study these downstream NO products, NO donors and NO-like molecules are used to determine the effects these molecules have in the cell and in vitro (14). In experiments using a donor molecule, nitroxyl (HNO), the one-electron reduction product of NO, has been shown to cause vasodilation independent of NO production and is of interest as a potential endogenous mediator of such effects (20).

4

Figure 1.3 NO metabolism

- NO can react with superoxide anions (O2 ) to form highly reactive peroxynitrite ions (ONOO-­). These ions can cause tyrosine nitrosylation (Tyr) leading to the formation of 3-nitrotyrosine (3-NT). NO can also interact with thiols like glutathione or albumin to produce S-nitrosothiols (RS-NO), or oxidize into nitrite or nitrate. MPO indicates myeloperoxidase; SOD, superoxide dismutase. Figure taken and adapted from (15).

1.1.2: NO overproduction is implicated in many diseases states.

NO dysregulation has been implicated in many disease states including septic shock, cancer, and diabetes (1,4,5). Increased NO production is associated with certain types of neurotoxicity, as shown by L-NMDA (N-methyl-D-aspartate) induced neurotoxicity is reduced by nNOS inhibitors (22). NO production due to iNOS overexpression or activation is observed in models of inflammatory pain, arthritis, bowel disease, asthma, ischemia, septic shock and many cancers (2,23–26). NO overproduction

5 due to eNOS has also been shown in cancers and in diabetes (27). Many NOS inhibitors currently exist and are being studied as a means to control NO production in the disease states mentioned previously. Due to the sequence similarity between the NOS isoforms, it has been challenging to create tissue- and disease-specific inhibitors of these enzymes (28). In many cancers, NO generated by iNOS activates multiple oncogenic signaling cascades whereas NO generated by eNOS or nNOS is still required for the cell to function normally (Figure 1.4) (29). Targeting NO production in certain cancers is thought to be especially beneficial for limiting tumor angiogenesis and growth (30). The nature of NO as an essential cell signaling molecule at lower concentrations, a pro-tumor signaling molecule at elevated concentrations, and a mediator of cell death at very high concentrations makes targeting this molecule in a tissue-specific manner through NOS inhibition very difficult (31). Attempts have been made to selectively inhibit tissue specific isoforms of NOS to control NO production, however many of these inhibitors are not selective for just one isoform of NOS (30). Another route to control NO production in a tissue specific manner would be highly beneficial to this field of research.

6

Figure 1.4 Different NO concentrations activate multiple signaling pathways

Low NO concentrations (< 30 nM) maintain endothelial cells and vascular tone through HIF-1α and sGC signaling. NO levels are elevated during the immune response or in tumor microenvironments and lead to the activation of oncogenic signaling pathways (ERK-1/1, Akt, Src, Ras, and EGFR). Above 500 nM, anti-tumor NO effects are observed through nitrosative stress and the activation of the p53 tumor suppressor. Figure taken and adapted from (29).

1.2: DIMETHYLARGININE DIMETHYLAMINOHYDROLASE AS A MEANS TO CONTROL NO PRODUCTION

1.2.1: Methylated arginines inhibit NOS

Regulation of NOS is of the upmost importance in the cell, as the radical species NO can have many adverse effects if produced above normal concentrations as described above. As mentioned before, intercellular Ca2+ levels and calmodulin (CaM) binding regulate the activity of nNOS and eNOS (11). iNOS expression is induced by bacterial lipopolysaccharides, cytokines, and on the translational level by RNA-binding protein. This isoform is constitutively active because Ca2+ and calmodulin (CaM) are tightly bound unlike eNOS and nNOS (32,33). Another notable mechanism of NOS regulation is 7 ω through methylated arginines (34). N -Methyl-L-arginine (mono-methyl L-arginine, L-

ω ω ω NMMA), N , N -dimethyl-L-arginine (asymmetric dimethyl L-arginine, ADMA), and N ,

ω’ N -dimethyl-L-arginine (symmetric dimethyl L-arginine, SDMA) are all synthesized by they methylation of protein arginines by protein arginine methyltransferases (PRMTs)

(35). PRMT-1 catalyzes the formation of ADMA and L-NMMA and PRMT-2 forms

SDMA and L-NMMA (35). Once these methylated proteins are hydrolyzed, free methylated arginines appear in the cytosol. ADMA and L-NMMA both act as substrate mimics and compete with L-arginine for binding at the active site of NOS (35) but SDMA has no inhibitory activity. ADMA acts as a competitive reversible inhibitor (36) and L- NMMA acts as a mechanism-based inactivator (37) of NOS.

HN NH2 N NH2 HN NH

NH NH NH

H H H COO COO COO H3N H3N H3N L-NMMA ADMA SDMA

Figure 1.5 Methylated arginines

PRMTs post-translationally methylate arginine residues to produce the three types of methylated arginine SDMA, L-NMMA, and ADMA. These molecules are released into the cytosol upon protein hydrolysis where ADMA and L-NMMA inhibit NOS.

In addition to aiding in the physiological regulation of NO production, methylated arginines can have destructive cellular effects when overproduced. Elevated levels of

AMDA are found in hypercholesterolemic patients (38) and are associated with 8 hypertension and immune dysfunction associated with chronic renal failure (39). Endothelial dysfunction caused by hyperhomocysteinimia is marked by elevated ADMA levels (40). Increased ADMA levels are also found in patients with congestive heart failure, diabetes, depression, and stroke (41–44).

1.2.2: Dimethylarginine Dimethylaminohydrolase (DDAH) contributes to the regulation of NO production

Methylated arginines are eliminated from the cell in several ways. Urinary excretion is thought to be almost entirely responsible for the elimination of SDMA and partially for ADMA and L-NMMA (45). Other methods of elimination involve the cationic amino acid transporter system (46). Human alanine-glyoxylate aminotransferase

2 decreases ADMA levels in mice (47). Also ADMA and L-NMMA are eliminated through enzymatic hydrolysis of ADMA and L-NMMA, but not SDMA, as catalyzed by

ω ω N ,N -dimethyl-L-arginine dimethylaminohydrolase (DDAH, E.C. 3.5.3.18) (34,48). There are two isoforms of DDAH expressed in mammals, DDAH1 and DDAH2 (48). DDAH from the bacterium Pseudomonas aeruginosa (PaDDAH) has also been extensively studied and was essential in mechanistic studies (49,50). In mammals, DDAH1 is expressed predominantly in the brain, along with nNOS, and the expression profile of DDAH2 is mainly in endothelial tissues coinciding with eNOS (34). Although DDAH2 is the most widely expressed isoform, DDAH2 does not metabolize methylated arginines (51,52) even though both DDAH1 and DDAH2 have been shown to regulate cellular NO concentrations (53). Estimates show elimination of more than 70% of circulating ADMA is due to metabolism by DDAH1 (54). Figure 1.6 highlights the ADMA/DDAH/NOS/PRMT regulation pathway.

9

Figure 1.6 Methylated arginine metabolism

Protein arginine methyltransferases (PRMTs), specifically methylate protein-incorporated L-arginine residues to generate protein-incorporated L-NMMA (MMA), ADMA, or SDMA. Upon proteolytic cleavage of arginine-methylated proteins, free intracellular L- NMMA, ADMA, or SDMA are generated. Free L-Arg can be metabolized by arginases to L-ornithine and urea, or by nitric oxide synthases (NOS) to NO and L-citrulline. Free methylated arginines can also be released to the extracellular space by cationic amino acid transporters (CAT) to induce distinct biological effects, undergo metabolism, or renal excretion. L-NMMA and ADMA, but not SDMA can be converted to L-citrulline and the corresponding alklyl-amine by dimethylarginine dimethylaminohydrolase (DDAH). L-NMMA and ADMA, but not SDMA, act as potent endogenous inhibitors of NOS enzymes. Finally, NO and L-citrulline can both cause the inhibition of DDAH. Figure taken and adapted from (55).

DDAH1 is a member of the pentein superfamily of proteins that share an α/β propeller fold and modify guanidines through a hydrolase mechanism, a dihydrolase mechanism, and an amidinotransferase reaction (49). DDAH1 uses a hydrolase

10 mechanism to convert its substrates (ADMA or L-NMMA) to L-citrulline and the corresponding alkylamine (56,57). In the proposed mechanism, Glu78 makes a monodentate interaction with the unsubstituted Nω of the methylated substrate, while Asp79 makes a bidentate interaction with the ω- and δ-nitrogens of the substrate (Figure

1.7). Upon substrate binding, the pKa of the active site cysteine, Cys274, is depressed and this thiolate nucleophile is activated (56). The active site Cys274 is then free to attack the guanidine carbon to form a tetrahedral intermediate. His173 will then donate a proton to the substituted ω-nitrogen, which aids in bond cleavage from the guanidine carbon to form a planar S-alkyl thiouronium intermediate following the loss of an alkyl amine (50,56). His173 then acts as a catalytic base to deprotonate a water molecule to attack the guanidine carbon of the thiourionium intermediate to again form a tetrahedral intermediate (49). This adduct collapses to release the product, L-citrulline and restores the enzyme to its original state. The proposed mechanism is shown in figure 1.8.

11 H173 E78 O

O

HN NH

N NH2

O S NH

C274 O D79

O

H3N

O ADMA

Figure 1.7 Substrate interactions with the active site of DDAH1

Asp79 forms a bidentate interaction with the ω- and δ-nitrogens of the substrate while Glu78 forms a monodentate interaction with the ω-nitrogen to assist in substrate placement.

12 H173 H173

N N NH2 NH2 HN HN NH H173 S NH NH S NH C274 +ADMA C274 -H+ O O H3N H3N SH HN NH O O C274 H173

(CH3)2NH NH2 S N NH NH C274 -L-citrulline O H173 H N 3 O

-(CH ) NH O 3 2 +H 0 HN 2 H2N NH H173 NH H173 SH C274 O NH2 H N N 2 S NH H3N O OH S HN NH NH HOH NH C274 C274 O O H3N H N O 3 O

Figure 1.8 Proposed catalytic mechanism of DDAH1

Upon substrate binding, Cys274 attacks the guanidine carbon to form a tetrahedral intermediate. His173 acts as an active site acid and donates a proton to facilitate the leaving alkylamine group. His173 then acts as a base to activate a catalytic water which forms the second tetrahedral intermediate. This intermediate collapses to release L- citrulline and restore the enzyme to its resting state.

1.2.3: Manipulation of DDAH can alter NO levels

As previously mentioned, developing isoform selective NOS inhibitors to control NO production has proven to be a challenge, therefore alternative efforts are being made to control NO production through one of its regulation pathways, namely the DDAH enzymes (30,58–62). Studies have shown that the pharmacological inhibition of DDAH1 can cause a rise in the levels of methylated arginines that is sufficient to block NO

13 production (63) and overexpression of DDAH1 reduces plasma ADMA and enhances angiogenesis in transgenic mice (64). Down-regulation of DDAH1 by siRNA resulted in a reduced amount of bioavailable NO in bovine aortic endothelial cells (53). In addition, DDAH1 has been found to be upregulated in over 80 % of various melanoma strains as compared to normal melanocytes (61) and the overexpression of DDAH1 in C6 gliomas results in increased vascularization and enhanced tumor growth (65,66). The fact that DDAH1 is expressed in vascular tissue, but not significantly in immune tissues make it a potential target for tissue-specific regulation of NO in cases when total NO reduction is not desirable, such as septic shock (48). Septic shock is a frequently fatal, NO-mediated condition, where NO is produced as a response to bacterial infection (67–69). This NO then activates macrophages and immune tissue, but also dilates blood vessels. Dilated blood vessels can cause excessive hypotension that can lead to multiple organ failure and death. Attempts to use NO synthase inhibitors to relieve septic shock induced hypotension were unsuccessful. Treatment of septic shock with selective DDAH1 inhibitors, however, could retain the beneficial aspects of reducing NO in vascular tissues, but avoid the detrimental effects of inhibiting NO production in immune tissue (30). Plasma ADMA is lowered in children with septic shock, perhaps due to increased DDAH1 activity (70). These results make it apparent that DDAH1 could be a viable target for isoform selective reduction of NO.

1.3 TOWARDS THE IRREVERSIBLE INHIBITION OF DIMETHYLARGININE DIMETHYLAMNIOHYDROLASE

Over the past decade, many DDAH inhibitors have been studied. Both reversible and irreversible inhibitors of DDAH are of interest and high-throughput screening has been recently used to discover new scaffolds for both types of compounds (71–73). 14 Irreversible modifiers of proteins have the advantage of increased biochemical efficiency as the covalent bonds created allow for an increased duration of action (74). Additionally, increasing the concentration of an enzyme’s substrate could potentially overwhelm reversible inhibition. If the enzyme were covalently inhibited, increasing the concentration of substrate would not be able to overcome the effects of the inhibitor. To achieve selectivity, inhibitors must inhibit DDAH1 but not the NOS isoforms or arginase, which metabolizes L-arginine to L-ornithine and urea (Figure 1.6) (61). Table 1.1 is a representative list of selected DDAH inhibitors. S-2-Amino-4-(3- methyl-guanidino) butanoic acid was first used to help identify a functional role for DDAH1 in the regulation of NO production (63). Leiper and colleagues have found that

ω N -(2-methyoxyethyl)-L-arginine caused an increase in circulating ADMA concentrations in rodents (30). Our lab has found that increasing the chain length of non-hydrolysable L-

5 5 NMMA mimics like N -(1-iminopropyl)-L-ornithine and N -(1-iminopentyl)-L-ornithine increase potency (58). 4-Hydroxy-2-nonenal is a reactive small molecule that is a product of free radical oxidation of polyunsaturated fatty acids and that can modify DDAH1 at endogenous concentrations (43). Ebselen, a compound which has been studied in the treatment of the same pathological states in which nitric oxide overproduction is involved (e.g. septic shock, ischemic stroke) was found through a high- throughput screen as a potent irreversible inhibitory of DDAH1 (73,75). In addition to ebselen, the fragment 2-hydroxymethyl-4-chloropyridine was discovered as a novel inhibitor of PaDDAH and its scaffold, the 4-halopyridines, are being used to create new cell permeable probes for DDAH active sites (76,77). Our lab discovered that 2- chloroacetamidine irreversibly inactivates PaDDAH (78). Using this small warhead, Wang et al. built onto it to create a potent, bioavailable inactivator of DDAH1, N5-(1- imino-2- chloroethyl)-L-ornithine (61). 15

Inhibitor Name Structure IC50 Ki, KI, kinact or Type 2nd order rate Constant O

Reversible S-2-amino-4-(3- H H 250 μM N.A. N N O methyl-guanidino) (DDAH1) NH2 NH3 butanoic acid (63)

NH2 O Reversible Nω-(2- 22 μM N.A. O N N O methyoxyethyl)- H H (DDAH1) NH3 L-arginine (30)

- H2N COO Reversible N5-(1- N.A. 52 μM NH iminopropyl)-L- NH3 (DDAH1) H ornithine (58)

- H2N COO 5 Reversible N -(1- NH N.A. 7.5 μM

NH3 iminopentyl)-L- H (DDAH1)

ornithine (60) O Irreversible 4-hydroxy-2- 50 μM N.A.

nonenal (79) (DDAH1) OH O −1 −1 Irreversible Ebselen (75) N.A. 44,000 M s

N (DDAH1) Se

Table 1.1 Representative DDAH inhibitors

16

Cl -1 -1 Irreversible 2-hydroxymethyl- N.A. 0.65 M s

4-chloropyridine (PaDDAH)

(76) N

OH NH -1 -1 Irreversible 2- 2 N.A. 6.4 M s

chloroacetamidine Cl (PaDDAH)

NH3 (78) NH 2 -1 -1 Irreversible N5-(1-imino-2- N.A. 4000 M s Cl COO- N chloroethyl)-L- H (DDAH1) H NH3 ornithine (61)

Table 1.1 Representative DDAH inhibitors

Inhibitors were studied either in human DDAH1 (DDAH1) or DDAH from P. aeruginosa DDAH (PaDDAH), as indicated in parentheses. Note that list is not comprehensive.

In addition to these inhibitors, DDAH1 is also inhibited endogenously by its

2+ reaction product L-citrulline (62), by Zn (80,81), by L-homocysteine (40), and by S- nitroso-L-homocysteine (57), among other compounds. DDAH1 is reversibly S- nitrosylated by NO donors and after cytokine-induced expression of iNOS, which both cause accumulation of ADMA and thus the inhibition of eNOS and nNOS, which is indicative of a feedback loop (82). Recently, activity based protein profiling probes have been developed by our lab for DDAH1 which allows the visualization of the active enzyme in cells and serves as a method to screen inhibitors for bioavailability (60,61,83). Although much progress has been made towards creating effective and viable DDAH1 inhibitors and probes, there is still progress to be made. DDAH1’s role in cancer biology is still being elucidated and selective activity probes can be very useful to further this 17 endeavor. My PhD studies at the University of Texas at Austin have focused on furthering this field. In the following chapters, I describe my studies of new covalent

5 inactivators of DDAH1 based on the inactivator N -(1-imino-2-chloroethyl)-L-ornithine to create more potent and selective inhibitors and my studies on the irreversible modification of DDAH1 by Angeli’s salt, a compound that donates nitroxyl (HNO), a one-electron reduction product of NO.

1.4 REFERENCES 1. Iyer AKV, Azad N, Wang L, Rojanasakul Y. Role of S-nitrosylation in apoptosis resistance and carcinogenesis. Nitric Oxide Biol. Chem. Off. J. Nitric Oxide Soc. 2008 Sep;19(2):146–51.

2. Vallance P, Leiper J. Blocking NO synthesis: how, where and why? Nat. Rev. Drug Discov. 2002 Dec;1(12):939–50.

3. Irvine JC, Favaloro JL, Kemp-Harper BK. NO- activates soluble guanylate cyclase and Kv channels to vasodilate resistance arteries. Hypertension. 2003 Jun;41(6):1301–7.

4. Mocellin S, Bronte V, Nitti D. Nitric oxide, a double edged sword in cancer biology: Searching for therapeutic opportunities. Med. Res. Rev. 2007 May 1;27(3):317–52.

5. Martínez-Ruiz A, Cadenas S, Lamas S. Nitric oxide signaling: Classical, less classical, and nonclassical mechanisms. Free Radic. Biol. Med. 2011 Jul 1;51(1):17– 29.

6. Calabrese V, Mancuso C, Calvani M, Rizzarelli E, Butterfield DA, Giuffrida Stella AM. Nitric oxide in the central nervous system: neuroprotection versus neurotoxicity. Nat. Rev. Neurosci. 2007 Oct;8(10):766–75.

7. Stuehr D., Griffith O. Mammalian Nitric Oxide Synthases. Adv. Enzym. Relat. Areas Mol. Biol. 1993;(65):287–346.

8. Shaul PW. Regulation of endothelial nitric oxide synthase: location, location, location. Annu. Rev. Physiol. 2002;64:749–74.

9. Susswein AJ, Katzoff A, Miller N, Hurwitz I. Nitric oxide and memory. Neurosci Rev J Bringing Neurobiol Neurol Psychiatry. 2004 Apr;10(2):153–62. 18 10. Boehning D, Snyder SH. Novel neural modulators. Annu. Rev. Neurosci. 2003;26:105–31.

11. Knipp M. How to control NO production in cells: Nω,Nω-dimethyl-L-arginine dimethylaminohydrolase as a novel drug target. Chembiochem Eur J Chem Biol. 2006 Jun;7(6):879–89.

12. Bretscher LE, Li H, Poulos TL, Griffith OW. Structural characterization and kinetics of nitric-oxide synthase inhibition by novel N5-(iminoalkyl)- and N5-(iminoalkenyl)- ornithines. J. Biol. Chem. 2003 Nov 21;278(47):46789–97.

13. Ekmekcioglu S, Ellerhorst J, Smid CM, Prieto VG, Munsell M, Buzaid AC, et al. Inducible nitric oxide synthase and nitrotyrosine in human metastatic melanoma tumors correlate with poor survival. Clin. Cancer Res. 2000;6(12):4768–75.

14. Flores-Santana W, Switzer C, Ridnour LA, Basudhar D, Mancardi D, Donzelli S, et al. Comparing the chemical biology of NO and HNO. Arch. Pharm. Res. 2009 Aug;32(8):1139–53.

15. Serrano C, Valero A, Picado C. Nasal Nitric Oxide. Arch. Bronconeumol. Engl. Ed. 2004 May;40(5):222–30.

16. Druhan LJ, Forbes SP, Pope AJ, Chen CA, Zweier JL, Cardounel AJ. Regulation of eNOS-derived superoxide by endogenous methylarginines. Biochemistry. 2008;47(27):7256–63.

17. Wink DA, Miranda KM, Katori T, Mancardi D, Thomas DD, Ridnour L, et al. Orthogonal properties of the redox siblings nitroxyl and nitric oxide in the cardiovascular system: a novel redox paradigm. Am. J. Physiol. Heart Circ. Physiol. 2003 Dec;285(6):H2264–2276.

18. Foster MW, McMahon TJ, Stamler JS. S-nitrosylation in health and disease. Trends Mol. Med. 2003 Apr;9(4):160–8.

19. Davis KL, Martin E, Turko IV, Murad F. Novel effects of nitric oxide. Annu. Rev. Pharmacol. Toxicol. 2001;41:203–36.

20. Chin KY, Qin C, Cao N, Kemp-Harper BK, Woodman OL, Ritchie RH. The concomitant coronary vasodilator and positive inotropic actions of the nitroxyl donor Angeli’s salt in the intact rat heart: contribution of soluble guanylyl cyclase- dependent and -independent mechanisms. Br. J. Pharmacol. 2014 Apr;171(7):1722– 34.

19 21. Wang J-Z, Zhu W-D, Xu Z-X, Du W-T, Zhang H-Y, Sun X-W, et al. Pin1, endothelial nitric oxide synthase, and amyloid-β form a feedback signaling loop involved in the pathogenesis of Alzheimer’s disease, hypertension, and cerebral amyloid angiopathy. Med Hypotheses. 2014 Feb;82(2):145–50.

22. Izumi Y, Benz AM, Clifford DB, Zorumski CF. Nitric oxide inhibitors attenuate N- methyl-D-aspartate excitotoxicity in rat hippocampal slices. Neurosci Lett. 1992 Feb 3;135(2):227–30.

23. Huang Z, Huang PL, Panahian N, Dalkara T, Fishman MC, Moskowitz MA. Effects of cerebral ischemia in mice deficient in neuronal nitric oxide synthase. Science. 1994 Sep 23;265(5180):1883–5.

24. Luo ZD, Cizkova D. The role of nitric oxide in nociception. Curr. Rev. Pain. 2000;4(6):459–66.

25. Kubes P, McCafferty DM. Nitric oxide and intestinal inflammation. Am J Med. 2000 Aug 1;109(2):150–8.

26. Silkoff PE, Robbins RA, Gaston B, Lundberg JO, Townley RG. Endogenous nitric oxide in allergic airway disease. J. Allergy Clin. Immunol. 2000 Mar;105(3):438–48.

27. Pannirselvam M, Anderson TJ, Triggle CR. Endothelial cell dysfunction in type I and II diabetes: The cellular basis for dysfunction. Drug Dev Res. 2003 Jan 1;58(1):28– 41.

28. Vittecek, J., Lojek A., Valacchi G.,Kubabla, L., et al. Arginine-Based Inhibitors of Nitric Oxide Synthase: Therapeutic Potential and Challenges. Mediators Inflamm. 2012 Sep 4;2012:e318087.

29. Switzer CH, Glynn SA, Ridnour LA, Cheng RY-S, Vitek MP, Ambs S, et al. Nitric oxide and protein phosphatase 2A provide novel therapeutic opportunities in ER- negative breast cancer. Trends Pharmacol Sci. 2011 Nov;32(11):644–51.

30. Leiper J, Nandi M. The therapeutic potential of targeting endogenous inhibitors of nitric oxide synthesis. Nat. Rev. Drug Discov. 2011 Apr;10(4):277–91.

31. Engin AB. Dual function of nitric oxide in carcinogenesis, reappraisal. Curr. Drug Metab. 2011 Nov;12(9):891–9.

32. Bogdan C. Nitric oxide and the immune response. Nat. Immunol. 2001 Oct;2(10):907–16.

33. Förstermann U, Sessa WC. Nitric oxide synthases: regulation and function. Eur. Heart J. 2012 Apr;33(7):829–837, 837a–837d. 20 34. Leiper JM. The DDAH-ADMA-NOS pathway. Ther Drug Monit. 2005;27(6):744–6.

35. Bedford MT, Clarke SG. Protein Arginine Methylation in Mammals: Who, What, and Why. Mol. Cell. 2009 Jan 16;33(1):1–13.

36. Tsikas D, Böger RH, Sandmann J, Bode-Böger SM, Frölich JC. Endogenous nitric oxide synthase inhibitors are responsible for the L-arginine paradox. FEBS Lett. 2000 Jul 28;478(1-2):1–3.

37. Olken NM, Osawa Y, Marletta MA. Characterization of the inactivation of nitric oxide synthase by NG-methyl-L-arginine: evidence for heme loss. Biochemistry. 1994 Dec 13;33(49):14784–91.

38. Böger RH, Bode-Böger SM, Szuba A, Tsao PS, Chan JR, Tangphao O, et al. Asymmetric dimethylarginine (ADMA): a novel risk factor for endothelial dysfunction: its role in hypercholesterolemia. Circulation. 1998 Nov 3;98(18):1842– 7.

39. Vallance P, Leone A, Calver A, Collier J, Moncada S. Accumulation of an endogenous inhibitor of nitric oxide synthesis in chronic renal failure. Lancet. 1992 Mar 7;339(8793):572–5.

40. Stühlinger MC, Oka RK, Graf EE, Schmölzer I, Upson BM, Kapoor O, et al. Endothelial dysfunction induced by hyperhomocyst(e)inemia: role of asymmetric dimethylarginine. Circulation. 2003 Aug 26;108(8):933–8.

41. Usui M, Matsuoka H, Miyazaki H, Ueda S, Okuda S, Imaizumi T. Increased endogenous nitric oxide synthase inhibitor in patients with congestive heart failure. Life Sci. 1998;62(26):2425–30.

42. Yoo JH, Lee SC. Elevated levels of plasma homocyst(e)ine and asymmetric dimethylarginine in elderly patients with stroke. Atherosclerosis. 2001 Oct;158(2):425–30.

43. Selley ML. Increased (E)-4-hydroxy-2-nonenal and asymmetric dimethylarginine concentrations and decreased nitric oxide concentrations in the plasma of patients with major depression. J. Affect. Disord. 2004 Jun;80(2-3):249–56.

44. Altinova AE, Arslan M, Sepici-Dincel A, Akturk M, Altan N, Toruner FB. Uncomplicated type 1 diabetes is associated with increased asymmetric dimethylarginine concentrations. J. Clin. Endocrinol. Metab. 2007 May;92(5):1881– 5.

21 45. Siroen MPC, van der Sijp JRM, Teerlink T, van Schaik C, Nijveldt RJ, van Leeuwen PAM. The human liver clears both asymmetric and symmetric dimethylarginine. Hepatol Baltim Md. 2005 Mar;41(3):559–65.

46. Vallance P, Leiper J. Cardiovascular biology of the asymmetric dimethylarginine:dimethylarginine dimethylaminohydrolase pathway. Arterioscler Thromb Vasc Biol. 2004 Jun;24(6):1023–30.

47. Rodionov RN, Murry DJ, Vaulman SF, Stevens JW, Lentz SR. Human alanine- glyoxylate aminotransferase 2 lowers asymmetric dimethylarginine and protects from inhibition of nitric oxide production. J. Biol. Chem. 2010 Feb 19;285(8):5385–91.

48. Leiper JM, Santa Maria J, Chubb A, MacAllister RJ, Charles IG, Whitley GS, et al. Identification of two human dimethylarginine dimethylaminohydrolases with distinct tissue distributions and homology with microbial arginine deiminases. Biochem J. 1999 Oct 1;343 Pt 1:209–14.

49. Linsky T, Fast W. Mechanistic similarity and diversity among the guanidine- modifying members of the pentein superfamily. Biochim. Biophys. Acta. 2010 Oct;1804(10):1943–53.

50. Linsky TW, Monzingo AF, Stone EM, Robertus JD, Fast W. Promiscuous partitioning of a covalent intermediate common in the pentein superfamily. Chem. Biol. 2008 May;15(5):467–75.

51. Palm F, Onozato ML, Luo Z, Wilcox CS. Dimethylarginine dimethylaminohydrolase (DDAH): expression, regulation, and function in the cardiovascular and renal systems. Am J Physiol Heart Circ Physiol. 2007 Dec;293(6):H3227–3245.

52. Altmann KS, Havemeyer A, Beitz E, Clement B. Dimethylarginine- dimethylaminohydrolase-2 (DDAH-2) does not metabolize methylarginines. Chembiochem 2012 Nov 26;13(17):2599–604.

53. Pope AJ, Karrupiah K, Kearns PN, Xia Y, Cardounel AJ. Role of dimethylarginine dimethylaminohydrolases in the regulation of endothelial nitric oxide production. J. Biol. Chem. 2009 Dec 18;284(51):35338–47.

54. Pope AJ, Karuppiah K, Cardounel AJ. Role of the PRMT–DDAH–ADMA axis in the regulation of endothelial nitric oxide production. Pharmacol. Res. 2009 Dec;60(6):461–5.

55. Zakrzewicz D, Eickelberg O. From arginine methylation to ADMA: A novel mechanism with therapeutic potential in chronic lung diseases. BMC Pulm Med. 2009 Jan 29;9(1):5.

22 56. Stone EM, Costello A, Tierney D, Fast W. Substrate-assisted cysteine deprotonation in the mechanism of dimethylarginase (DDAH) from Pseudomonas aeruginosa. Biochemistry. 2006;17:5618–30.

57. Hong L, Fast W. Inhibition of human dimethylarginine dimethylaminohydrolase-1 by S-nitroso-L-homocysteine and hydrogen peroxide. Analysis, quantification, and implications for hyperhomocysteinemia. J. Biol. Chem. 2007 Nov 30;282(48):34684– 92.

58. Wang Y MA. Developing dual and specific inhibitors of dimethylarginine dimethylaminohydrolase-1 and nitric oxide synthase: Toward a targeted polypharmacology to control nitric oxide. Biochemistry. 2009;48:8624–35.

59. Nandi M, Kelly P, Torondel B, Wang Z, Starr A, Ma Y, et al. Genetic and pharmacological inhibition of dimethylarginine dimethylaminohydrolase 1 is protective in endotoxic shock. Arterioscler Thromb Vasc Biol. 2012 Nov;32(11):2589–97.

60. Lluis M, Wang Y, Monzingo AF, Fast W, Robertus JD. Characterization of C-alkyl amidines as bioavailable covalent reversible inhibitors of human DDAH-1. ChemMedChem. 6(1):81–8.

61. Wang Y, Hu S, Gabisi AM Jr, Er JAV, Pope A, Burstein G, et al. Developing an Irreversible Inhibitor of Human DDAH-1, an Enzyme Upregulated in Melanoma. ChemMedChem. 2014 Feb 26;

62. Kotthaus J, Schade D, Kotthaus J, Clement B. Designing modulators of dimethylarginine dimethylaminohydrolase (DDAH): a focus on selectivity over arginase. J. Enzyme. Inhib. Med. Chem. 27(1):24–8.

63. MacAllister RJ, Parry H, Kimoto M, Ogawa T, Russell RJ, Hodson H, et al. Regulation of nitric oxide synthesis by dimethylarginine dimethylaminohydrolase. Br. J. Pharmacol. 1996 Dec;119(8):1533–40.

64. Jacobi J, Sydow K, von Degenfeld G, Zhang Y, Dayoub H, Wang B, et al. Overexpression of dimethylarginine dimethylaminohydrolase reduces tissue asymmetric dimethylarginine levels and enhances angiogenesis. Circulation. 2005 Mar 22;111(11):1431–8.

65. Kostourou V, Robinson SP, Whitley GSJ, Griffiths JR. Effects of overexpression of dimethylarginine dimethylaminohydrolase on tumor angiogenesis assessed by susceptibility magnetic resonance imaging. Cancer Res. 2003 Aug 15;63(16):4960–6.

23 66. Kostourou V, Troy H, Murray JF, Cullis ER, Whitley GSJ, Griffiths JR, et al. Overexpression of dimethylarginine dimethylaminohydrolase enhances tumor hypoxia: an insight into the relationship of hypoxia and angiogenesis in vivo. Neoplasia N Y N. 2004 Aug;6(4):401–11.

67. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit. Care Med. 2001 Jul;29(7):1303–10.

68. Russel JA. The current management of septic shock. Minerva Med. 2008 Oct;99(5):431–58.

69. Fortin CF, McDonald PP, Fülöp T, Lesur O. Sepsis, leukocytes, and nitric oxide (NO): an intricate affair. Shock Augusta Ga. 2010 Apr;33(4):344–52.

70. Weiss SL, Haymond S, Ralay Ranaivo H, Wang D, De Jesus VR, Chace DH, et al. Evaluation of asymmetric dimethylarginine, arginine, and carnitine metabolism in pediatric sepsis. Pediatr Crit Care Med J. 2012 Jul;13(4):e210–218.

71. Linsky T, Wang Y, Fast W. Screening for dimethylarginine dimethylaminohydrolase inhibitors reveals ebselen as a bioavailable inactivator. ACS Med Chem Lett. 2011;2(8):592–6.

72. Ghebremariam YT, Erlanson DA, Yamada K, Cooke JP. Development of a dimethylarginine dimethylaminohydrolase (DDAH) assay for high-throughput chemical screening. J. Biomol. Screen. 2012 Jun;17(5):651–61.

73. Linsky T, Fast W. A continuous, fluorescent, high-throughput assay for human dimethylarginine dimethylaminohydrolase-1. J. Biomol. Screen. 2011 Oct;16(9):1089–97.

74. Kalgutkar AS, Dalvie DK. Drug discovery for a new generation of covalent drugs. Expert. Opin. Drug. Discov. 2012 Jul;7(7):561–81.

75. Linsky T, Wang Y, Fast W. Screening for dimethylarginine dimethylaminohydrolase inhibitors reveals ebselen as a bioavailable inactivator. ACS Med Chem Lett. 2011;2(8):592–6.

76. Johnson CM, Monzingo AF, Ke Z, Yoon D-W, Linsky TW, Guo H, et al. On the mechanism of dimethylarginine dimethylaminohydrolase inactivation by 4- halopyridines. J. Am. Chem. Soc. 2011 Jul 20;133(28):10951–9.

24 77. Johnson CM, Linsky TW, Yoon D-W, Person MD, Fast W. Discovery of halopyridines as quiescent affinity labels: inactivation of dimethylarginine dimethylaminohydrolase. J. Am. Chem. Soc. 2011 Feb 9;133(5):1553–62.

78. Stone EM, Schaller TH, Bianchi H, Person MD, Fast W. Inactivation of two diverse enzymes in the amidinotransferase superfamily by 2-chloroacetamidine: dimethylargininase and peptidylarginine deiminase. Biochemistry. 2005;44(42):13744–52.

79. Forbes SP, Druhan LJ, Guzman JE, Parinandi N, Zhang L, Green-Church KB, et al. Mechanism of 4-HNE mediated inhibition of hDDAH-1: implications in NO regulation. Biochemistry. 2008 Feb 12;47(6):1819–26.

80. Knipp M, Braun O, Gehrig PM, Sack R, Vasák M. Zn(II)-free dimethylargininase-1 (DDAH-1) is inhibited upon specific Cys-S-nitrosylation. J. Biol. Chem. 2003 Jan 31;278(5):3410–6.

81. Knipp M, Charnock JM, Garner CD, Vašák M. Structural and Functional Characterization of the Zn(II) Site in Dimethylargininase-1 (DDAH-1) from Bovine Brain Zn(II) RELEASE ACTIVATES DDAH-1. J. Biol. Chem. 2001 Nov 2;276(44):40449–56.

82. Leiper J, Murray-Rust J, McDonald N, Vallance P. S-nitrosylation of dimethylarginine dimethylaminohydrolase regulates enzyme activity: further interactions between nitric oxide synthase and dimethylarginine dimethylaminohydrolase. Proc. Natl. Acad. Sci. U S A. 2002 Oct 15;99(21):13527– 32.

83. Wang Y, Hu S, Fast W. A click chemistry mediated in vivo activity probe for dimethylarginine dimethylaminohydrolase. J. Am .Chem. Soc. 2009;131(42):15096– 7.

25 Chapter 2: Optimizing Chloroacetamidine Inactivators of DDAH1 and a Comparison of Fitting Methods

2.1 INTRODUCTION

Nitric oxide (NO) is an essential cell-signaling molecule involved in blood pressure regulation, blood vessel formation, pain modulation and cytotoxicity against bacteria (1). NO also plays an essential role in tumoregenesis at low, but elevated from normal, cellular levels (nM) and the upregulation of Nitric Oxide Synthases (NOS) is found in many types of tumors that are correlated with poor patient prognosis (2). Conversely, high (μM) levels of NO are associated with tumor cell killing (3). Targeted inhibition of NOS has shown promise in some human melanomas, however increased potency is needed and alternative methods of NO control are desired to avoid off-target cell killing (4).

Nitric oxide synthases catalyze the oxidation of L-arginine to L-citrulline and NO in an NADPH- and O2-dependent reaction (5). Asymmetric dimethylarginine (ADMA) and monomethylarginine (L-NMMA) are released into the cytoplasm by hydrolysis of post-translationally methylated protein arginine residues (1). L-NMMA and ADMA are endogenous inhibitors of NOS (6). ADMA acts as a competitive inhibitors of NOS while

L-NMMA acts as a mechanism-based inactivator (7). NO production is further controlled by the metabolism of these methylated arginines. Dimethylarginine dimethylaminohydrolase (DDAH1) catalyzes the hydrolytic conversion of ADMA or L- NMMA into citrulline and dimethylamine or monomethylamine respectively and relieves their inhibitory effects on NOS (Figure 2.1) (8).

26 DDAH1

Figure 2.1 Arginine metabolism and regulation.

DDAH1 upregulation results a decrease in tumor ADMA and increased NO production along with increased tumor growth and angiogenesis (9). Several studies propose DDAH1 as a good therapeutic target for NO regulation in tumor cells (1,9). Preliminary studies from a collaboration between our lab and the Ekmekcioglu lab at the MD Anderson Cancer Center have shown that a DDAH1 specific inhibitor, N5-(1-imino-

2-chloroethyl)-L-ornithine (Cl-NIO), decreases NO production in a melanoma cell line (10). We previously studied the mechanism of DDAH from Pseudomonas aeruginosa inactivation by 2-chloroacetamidine (CAA) and have linked the active site warhead with

L-norvaline to yield a potent selective inactivator of DDAH1, Cl-NIO (KI = 1.3 ± 0.6 μM,

-1 kinact = 0.34 ± 0.07 min ) which mimics one of the substrates of DDAH1, L-NMMA (10) (Figure 2.2 shows the structures of these compounds). We expected at least an additive effect on the KI, from linking the CAA (KI = 300 ± 20 μM) and the L-norvaline (Ki = 700

± 30 μM) fragments based on the combined ΔG of binding (-9.08 kcal/mol) with an

27 additional benefit of a smaller entropic penalty for the binding of one molecule instead of two (11,12). However, Cl-NIO has a ΔG of binding of -8.00 kcal/mol, less than expected. Cl Cl

N NH2 HN NH2 NH2 NH2

NH NH NH NH2

H H H H

- - - - H3N COO H3N COO H3N COO H3N COO

ADMA L-NMMA Cl-NIO CAA L-Norvaline

Figure 2.2 Substrates and selected inhibitors of DDAH1

Shown are the two substrates of DDAH1, ADMA and L-NMMA, the covalent irreversible inhibitor of DDAH1, Cl-NIO, and the two fragment sized inhibitors that led the synthesis of Cl-NIO, CAA and L-norvaline. Here, I synthesized three derivatives of this inactivator to try to improve the potency of Cl-NIO. The inactivation of DDAH1 by these inactivators is studied in detail using different methods to analyze the kinetic parameters. A crystal structure gives insight into the binding of one of the inactivators, and the compounds are found to be cell-permeable active-site directed inhibitors of DDAH1.

2.2 MATERIALS AND METHODS

General Synthesis

2-Chloro-ethanimidic acid, ethyl ester: Chloroacetornitrile (2.55 mL, 40.3 mmol) and anhydrous ethanol (2.6 mL, 22.4 mmol) were added to 22 mL of anhydrous diethyl ether

o and cooled to 0 C. The solution was bubbled through with HCl(g) until a white precipitate formed. The precipitate was washed with cold diethyl ether followed by cold hexanes to

28 1 yield 2-chloroethanimidic acid, ethyl ester (1.8 g, 78 %). H-NMR (300 MHz, D2O): 4.66 (2 H, overlapped with reference peak), 4.16 (2 H, s), 1.16 (3 H, d, J = 7.2 Hz).

2-Chloro-propanimidic acid, ethyl ester: Chloropropionitrile (3.5 mL, 39.56 mmol) and anhydrous ethanol (2.6 mL, 22.4 mmol) were added to 20 mL of anhydrous diethyl ether

o and cooled to 0 C. The solution was bubbled through with HCl(g) until a white precipitate formed. The precipitate was washed with cold diethyl ether followed by cold hexanes to

1 yield 2-chloropropanimidic acid, ethyl esther. (1.4 g, 46 %) H-NMR (400 MHz, D2O):

4.10 (1 H, q, J = 7.6 Hz), 3.19 (2 H,m), 1.48 (3 H, dd, J1 = 6.8 Hz, J2 = 22.8 Hz), 1.07- 1.15 (3 H, m).

α α N -(Tert-butoxycarbonyl)-L-ornithine or N -(t-butoxycarbonyl)-L-lysine (1 mmol) was dissolved in 5 mL water at 0 oC followed by the dropwise addition of 2.5 M NaOH to bring the pH to 10. 1 or 2 was added in portions (2.5 mmol), holding the reaction at pH 10 by addition of NaOH. After the completion of addition, the reaction was stirred at 0 oC for 2 h and at room temperature for 1 h. The reaction was neutralized by the dropwise addition of 1 M HCl and stirred at 4 oC for 1.5 days. The solvents were removed by reduced pressure rotary evaporation and purified using the Teledyne/Isco CombiFlash

Rf200 RediSep Rf 26 g C18 Reverse Phase column (0.1 % TFA/H2O to 0.1 % TFA/MeOH linear gradient) at the Texas Institute for Drug and Diagnostic Development at the University of Texas at Austin. The solvents were removed by reduced pressure rotary evaporation and the remaining residue was treated with 4 M HCl in 1,4-dioxane for 2 h to remove the t-Boc group. The solvent was again removed to yield the desired products.

29 6 1 N -(1-Imino-2-chloroethyl)-L-lysine (Cl-NIL) H-NMR (400 MHz, D2O): 4.26 (2 H, s), 3.87 (1 H, t, J = 6 Hz), 3.22 (2 H, t, J = 7.2 Hz), 1.80-1.84 (2 H, m), 1.57-1.60 (2 H, m),

13 1.38-1.47 (2H, m). C-NMR(D2O): 171.58, 162.19, 52.24, 41.64, 38.61, 28.75, 25.51,

+ 21.0. ESI (m/z) M + H calcd for C8H17ClN3O2, 222.10038; found 222.10029.

6 1 N -(1-Imino-2-chloroisopropyl)-L-lysine (Me-Cl-NIL) H-NMR (400 MHz, D2O): 4.68 (1 H, overlapped with reference peak), 3.831 (1 H, t, J = 6.2 Hz), 3.188 (2 H, t, J = 7.2 Hz),

13 1.25-1.91(9 H, m). C-NMR(D2O): 172.54, 166.21, 53.07, 51.63, 42.06, 29.38, 26.0,

+ 22.04, 21.53. ESI (m/z) M + H calcd for C9H19ClN3O2, 236.11603; found 236.11632.

5 1 N -(1-Imino-2-chloroisopropyl)-L-ornithine (Me-Cl-NIO) H-NMR (400 MHz, D2O): 4.70 (1 H, overlapped with reference peak), 3.95 (1 H, t, J = 6.4 Hz), 3.241 (2 H, t, J =

13 6.8 Hz), 1.63-1.94 (6 H, m), 1.61 (3 H, d, J = 7.2 Hz). C-NMR(D2O): 172.44, 166.73,

+ 53.06, 51.89, 41.95, 27.32, 22.63, 22.04. ESI (m/z) M + H calcd for C8H17ClN3O2, 222.10031; found 222.10023.

The broad specificity probe N-but-3-ynyl-2-chloroacetamidine was synthesized as previously described (12). Unless specified otherwise, all other chemicals were from the Sigma Aldrich Chemical Co. (St. Louis, MO).

DDAH1 Expression and Purification

Wild-type human DDAH1 bearing an N-terminal His6 affinity tag was expressed using an expression plasmid with a reengineeried N-terminus (pET28a-hDDAH-1re) to avoid N-terminal nonenzymatic gluconylation as described earlier (11). Protein was purified according to the procedure published earlier (11) with the following

30 modifications: Elution Buffer (25 mL) (50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole, 15% glycerol pH 8.0) was used to elute protein from the Ni-NTA affinity resin

(8mL, Qiagen). The eluent was diluted into Buffer A (50 mM NaH2PO4, 3.0 M NaCl, 15 % glycerol, pH 7.0) to a final NaCl concentration of 1.8 M and loaded onto a 1.5 × 18 cm phenyl-sepharose column (GE Healthcare, Piscataway, NJ). Proteins were eluted by a stepwise gradient using the described Buffer A but with an NaCl concentration of 1.8 M, followed by 1.6 M, followed by 1.3 M, followed by 1 M. As gauged by SDS-PAGE, fractions containing DDAH1 were pooled and concentrated using an Amicon Centrifugal Filter (Millipore, Billerica, MA) with a 10 kDa molecular weight cutoff. The final protein was dialyzed overnight in Dialysis Buffer 1 (2 mM 1,10-phenathroline, 100 mM KCl, 10 % glycerol, pH 7.3) followed by 2 × 4 h dialyses in Chelex-100 (BioRad) treated Dialysis

Buffer 2 (50 mM KH2PO4, 100 mM KCl, 10 % glycerol, pH 7.3), flash frozen and stored in aliquots at -80 oC. All buffers were made with deionized water.

Time-dependent Inactivation of DDAH1 by Irreversible Inhibitors

Inactivation kinetics of DDAH1 were derived from the nonlinear progress curves of incubations with purified His6-DDAH1 and inactivators (Cl-NIO, Cl-NIL, Me-Cl-NIO, or Me-Cl-NIL) in competition with substrate, ADMA (KM = 42 μM). Briefly 10 mM or 5 mM ADMA was incubated with varying concentrations of inactivators (0-1.5 mM), in reaction buffer (100 mM KH2PO4, 100 mM KCl, 2 mM EDTA, 1 % Tween-20, pH 7.3) in a 96-well plate format. The reaction was initiated upon the addition of 1.2 μM DDAH1 and quenched with 10 μL TFA at various time points (0-60 min). L-Citrulline was detected by a color developing reagent which detects ureido groups (200 μL, COLDER) (13). The apparent inactivation rates (A) were determined by the method of Tsou, and replots of 1/A led to the determination of kinact and Ki values as described elsewhere (14).

31 A second method was used to determine the inactivation parameters. Purified

His6-DDAH1 was incubated with inactivators (Cl-NIL, Cl-NIO, Me-Cl-NIL, or Me-Cl- NIO) (0-64 μM) in reaction buffer. To test for time-dependent loss of activity, aliquots were removed from those incubations at time points (0-60 min) and diluted two-fold into an assay solution containing a large excess (300 μM) of the alternative substrate, S- methyl-L-thiocitrulline (SMTC). The remaining enzyme activity was assayed by detecting methanethiol release upon substrate hydrolysis using 7-diethylamino-3-(4'- maleimidylphenyl)-4-methylcoumarin (CPM), as described elsewhere (15). The percent remaining activity was fit to a single exponential to determine the observed rate, kobs. The plot of the observed rate versus inhibitor concentration was then fit to the Morrison equation for tight binding to estimate the apparent Ki using equation 2.1 as described, if the enzyme and inhibitor concentration were similar. If the inhibitor concentration was well above that of the enzyme, the observed rates were fit to equation 2.2 to determine kinact and KI by a non-linear version of that described by Kitz and Wilson (16).

! ! + ! + !! − ! + ! + !!) − 4[!!][!]) ! = ! ! 2[!]

Equation 2.1: where V0 is the reaction rate observed in the absence of inhibitor, [I] is the

inhibitor concentration and KI is the apparent constant for reversible competitive inhibition.

(!!"#$% ! ) !!"# = (!! + [!])

Equation 2.2: where [I] is the inhibitor concentration, kinact is the inactivation rate

constant, and KI is the constant for the reversible component of inhibition

32 A third method, described below, was also used to fit the inactivation of DDAH1 by the inactivators. The remaining enzyme activity (%) was fit globally using KinTek software (KinTek Corporation)

Partition ratio

To determine the partition ratio of Cl-NIL to DDAH1, purified DDAH1 was incubated with varying concentration of Cl-NIL (0-400 nM) for one hour and then subjected to rapid dilution into excess substrate (300 μM SMTC). The remaining fractional activity was determined by the CPM assay (15) and the ratio of [Cl-NIL] to [DDAH1] was plotted versus the remaining fractional activity.

Crystal Structure

Prior to crystallization set-up, protein was incubated with Cl-NIL in the following mixture: 12 mg/mL human DDAH-1 in 100 mM KCl, 100 mM potassium phosphate, pH 7.3, 10 % glycerol, 2 mM dithiothreitol, 0.7 mM Cl-NIL. The Macromolecular Structure Core facility then grew co-crystals of the protein-compound complex were grown at 4 ˚C using the sitting drop method from the protein-compound mixture and 20 % PEG 3350, 0.2 M NaCl. Prior to data collection, a crystal was transferred briefly to a drop of 20 % PEG 3350, 0.2 M NaCl, 0.1 M HEPES at pH 7.2, 10 % propylene glycol for cryoprotection. The crystal was mounted in a cryoloop (Hampton Research, Laguna Niguel, CA), flash frozen in liquid nitrogen, and mounted in the cold stream on the goniostat. X-ray diffraction data were collected from the co-crystal at 100 oK at the Advanced Light Source (ALS) beamline 5.0.3 at the Lawrence Berkeley National Laboratory. Diffraction images were processed and data reduced using HKL2000 (17).

The structure of the complex was solved by molecular replacement with MOLREP (18) 33 using the structure of human DDAH1 (PDB accession code 3I2E) (11) as the search model. Model building was carried out using Coot (19). Refinement of models was done using PHENIX (20). There were several rounds of refinement followed by manual rebuilding of the model. To facilitate manual rebuilding, a difference map and a 2Fo-Fc map, σA-weighted to eliminate bias from the model (21), were prepared. 5 % of the diffraction data were set aside throughout refinement for cross-validation (22). MolProbity (23) were used to determine areas of poor geometry and to make Ramachandran plots.

Arginase Assay

To determine if manganese-loaded human arginase-I (Mn-hArgI) is inhibited by Cl-NIL, Mn-hArgI (2.5 μM) was incubated with varying concentrations of Cl-NIL (46

μM − 12 mM) in the presence of L-arginine (1.5 mM), MnSO4 (10 μM) in HEPES buffer (100 mM) at pH 7.4 for 30 min at 37 °C. The reaction was quenched with 10 μL of trichloroacetic acid (6 M) and the formation of urea was determined by the COLDER assay as described above. The remaining fractional enzyme activity vs. the inhibitor

h concentration were plotted and fit using the equation 1/(1+(IC50/[Inhibitor]) ), with h as the Hill coefficient, in KaliedaGraph and the Ki value was calculated using the method of Cheng and Prusoff (24) assuming competitive inhibition.

In cell IC50 value determination

To determine apparent “in cell” IC50 values for inhibition of DDAH1, a competitive labeling strategy was used. Cultured HEK293T cells were seeded in two 6- well polystyrene plates using complete growth medium containing DMEM with 10 % FBS (Invitrogen, Carlsbad, CA) and grown to 80 % confluency. The pEF6a-DDAH-1 plasmid was transiently transfected into HEK293T cells using Lipofectamine 2000 34 (Invitrogen, Carlsbad, CA) (12). After 24 h, spent medium was removed and the cells were each washed with 1 mL of 0.5 mL PBS at pH 7.2 (Invitrogen, Carlsbad, CA). Stock solutions of Cl-NIL (10 mM) were diluted into complete growth medium in the cultures (1000 μL each well) to give final concentrations of 0.078, 0.156, 0.313, 0.625, 1.25, 2.5, 5, 10, 20, 40, 80, and 0 μM. The resulting cultures were subsequently incubated for 15 min at 37 ºC in an atmosphere of 5 % CO2 before addition of the activity probe (110 μM), followed by an additional 10 min incubation. After treatment, cells were washed two times with PBS (1 mL) to remove the media, and harvested in 1 mL PBS followed by centrifugation at 14000 g for 5 min at 4 ºC. Cell pellets were stored at – 80 ºC.

Frozen cell pellets were lysed and labeled with biotin-PEO3-azide as described earlier (12). Briefly, the cell pellets were resuspended in PB buffer (10 mM sodium / potassium phosphate buffer at pH 8.0 containing the complete mini EDTA-free protease cocktail inhibitor (Roche, Indianapolis, IN) and Triton X-100 (1%)) and then were subjected to three cycles of vortexing, freezing in liquid N2, and thawing at room temperature. The cells were then centrifuged at 14000 g for 5 min at 4 ºC to remove insoluble cell debris. The supernatant was removed and total protein concentration was determined using a Bradford assay. Biotin-PEO3-azide was appended to alkyne labeled proteins using the Cu (I) catalyzed 1,3 dipolar cycloaddition reaction as follows: 25 μg total protein in 50 μL PB buffer was mixed with 0.5 μL of biotin-PEO3-azide (2.5 mM), 0.5 μL CuSO4 (50 mM), 0.5 μL tris(2-carboxyethyl)phosphine (TCEP, 50 mM) and 1.65 μL tris[(1-benzyl-1H-1,2,3-trazol-4-yl)methyl]amine (TBTA, 1.5 mM), vortexed and incubated at 25° C for 1 h. Reactions were quenched by addition of 2 × SDS loading buffer and heat inactivated at 100 °C for 10 min. Samples were directly used for SDS- PAGE or stored at -80 °C before analysis. Two-color Western blot detection was used to detect the expression levels of DDAH1 and the response to the biotin tag as described 35 previously (12). Images were scanned using an Odyssey Infrared Imaging System (Li- Cor Biosciences, Lincoln, NE) at the Core DNA Facility (University of Texas, Austin). Fluorescence intensities for both 680 nm and 800 nm channels were integrated. Fits for the concentration dependence of the normalized I800/I680 are determined using the

h equation: I800 (or I800/I680) = 1/(1+(IC50/[Inhibitor]) ), with h as the Hill coefficient.

In Cell Single Point Assay

To evaluate the efficacy of the compounds as compared to one another, cells were grown, seeded, and transfected as described above before adding 10 μM of Cl-NIL, Cl- NIO, Me-Cl-NIL, or Me-Cl-NIO to separate wells. The compounds were allowed to incubate at 37 ºC in an atmosphere of 5 % CO2 for 15 min before the addition of the activity probe (110 μM), followed by an additional 10 min incubation. After treatment, cells were washed two times with PBS (1 mL) and harvested in 1 mL PBS followed by centrifugation at 14000 g for 5 min at 4 ºC. Cell pellets were stored at – 80 ºC. The frozen cell pellets were thawed, lysed, and labeled with biotin-PEO3-azide before analysis by two-color Western Blot detection as described above.

3-(4,5-Dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)- 2-(4-sulfophenyl)-2H- tetrazolium (MTS) Assay

Cultured HEK293T cells were seeded in a 96-well polystyrene plate using complete growth medium containing DMEM with 10 % FBS (Invitrogen Carlsbad, CA). Post-seeding (24h), 1000, 333.3, 111.1, 37, 12.3, 4.1, 1.27, and 0 μM of Cl-NIL were diluted in growth medium added to the wells in triplicate and incubated for 72 h at 37 oC in an atmosphere of 5 % CO2. MTS reagent (20 μL) (Promega) was added to each well

o and then incubated for an additional 2 h at 37 C in an atmosphere of 5 % CO2 before the absorbance was read at 490 nm.

36 2.3 RESULTS AND DISCUSSION

The strategy of inhibiting DDAH1 to lead to the accumulation of ADMA and the subsequent inhibition of NO production could allow for promising new drugs and circumvent the challenge of NOS isoform specific inhibition (1). We hypothesized that varying two moieties of Cl-NIO could lead to improved potency. First, we analyzed the warhead portion of the molecule compared to the natural substrate of DDAH1, ADMA.

ADMA contains two omega methyl groups compared to one of L-NMMA. ADMA is thought to be the more important of the two molecules because L-NMMA circulating concentrations are much lower than ADMA (11). Additionally, DDAH1 is thought to be responsible for over 90% of ADMA metabolism and the KM for ADMA is lower than that of L-NMMA (25,26). Therefore we installed an extra methyl group in the inactivator to mimic ADMA and potentially increase the potency of the inhibitor. Next, we analyzed the amino acid portion of Cl-NIO and the crystal structure of DDAH1 complexed with L- citrulline (PDB 2JAI) (27). We hypothesized that lengthening the linker between the amino acid portion of the molecule and the warhead would result in a molecule with more rotational degrees of freedom and perhaps could allow better positioning of the warhead in the active site of DDAH1. Using these hypotheses, we synthesized three

6 6 derivatives of Cl-NIO: N -(1-imino-2-chloroethyl)-L-lysine (Cl-NIL), N -(1-imino-2-

5 chloroisopropyl)-L-lysine (Me-Cl-NIL) and N -(1-imino-2-chloroisopropyl)-L-ornithine (Me-Cl-NIO) and studied their inhibitory effect on DDAH1 (Figure 2.3 shows the structures compared with their respective substrate mimics). Herein, I find that the compounds synthesized are cell permeable and show potent inhibition of DDAH1. The crystal structure obtained for DDAH1 inactivation by Cl-NIL gives further insight into the mechanism of inhibition.

37 Cl Cl

HN NH2 NH2 NH2

NH NH NH

H H H COO COO H3N H3N H3N

L-NMMA Cl-NIO COO Cl-NIL

Cl Cl

N NH2 NH2 NH2

NH NH NH

H H H COO COO H3N H3N H3N COO ADMA Me-Cl-NIO Me-Cl-NIL

Figure 2.3 DDAH1 substrates and synthesized covalent inactivator structures

Shown are the structures of the two DDAH1 substrates, ADMA and L-NMMA as well as the first generation inhibitor Cl-NIO and the three inactivators I synthesized, Cl-NIL which mimics the head group of L-NMMA, as well as Me-Cl-NIO and Me-Cl-NIL which mimic the head group of AMDA.

38 Kinetic Analysis of DDAH1 Time-dependent Inactivation by Chloroacetamidines.

Schwartz et al have discussed the importance of understanding drug potency by being able to separate molecular determinates of inhibition (non-covalent binding affinity and chemical reactivity) and the difficulty of doing so by conventional methods (28,29). Herein, I compare conventional fitting analyses with globally fitting the data using KinTek Explorer Software. Fitting data by conventional methods to measure the rates of product formation is often followed by subsequent analysis of the observed inactivation rates which often results in fitting redundant information or simplifying assumptions (25).

Globally fitting data to a model by computer simulation has the advantage of using all aspects of the data such as rates and amplitudes of reactions without the need to make extra simplifying assumptions (30). We previously used the method of Tsou to analyze the first generation inhibitor Cl-NIO using excess substrate ADMA (10). Briefly, excess substrate and varying inhibitor concentrations are mixed prior to the addition of the enzyme. Product formation is monitored and an apparent inactivation rate is determined and the inverse of this apparent rate is then plotted versus inhibitor concentration to determine kinact and KI values. Here, we used Tsou’s method to analyze the time-dependent inactivation of DDAH1 by three derivatives of Cl-NIO (Cl-NIL, Me-Cl-NIO, and Me-Cl-NIL). The kinetics of Cl-NIL will be discussed here in the main text, but the information for Cl- NIO, Me-Cl-NIO and Me-Cl-NIL is included in table 2.1 as well as in the appendix of this dissertation. Figure 2.4 shows the inactivation and Tsou plots of DDAH1 by Cl-NIL,

-1 which lead to the determination of a kinact of 0.3 ± 0.1 min and a KI of 0.10 ± 0.06 μM.

This KI is a 10-fold improvement when compared to Cl-NIO, which we also determined using the Tsou method.

39 I next used a different type of experiment to determine kinact and KI. I conducted a jump-dilution assay and analyzed the results in three different ways.

Table 2.1 Inactivation constants for DDAH1 inactivators

The inhibition constants for four DDAH1 inhibitors (Cl-NIO, Cl-NIL, Me-Cl-NIO, Me- Cl-NIL) are determined using three or four different methods depending on the potency of the inhibitor.

40 y = m1*(1-exp(-m2*m0))+m3 Value Error A m1 6502.2 85000 m2 0.00019678 0.0025826 80 m3 0.56379 0.59022 Chisq 13.498 NA 70 R 0.99742 NA

60 y = m1*(1-exp(-m2*m0))+m3 Value Error 50 m1 9.5847 1.054 m2 0.10258 0.026792 40 m3 -0.186 1.2048 Chisq 3.4462 NA M Citrulline

µ 30 R 0.99723 NA y = m1*(1-exp(-m2*m0))+m3 20 Value Error m1 5.2007 1.6464 10 m2 0.17133 0.060355 m3 -1.9115 1.688 0 Chisq 6.0861 NA 0 10 20 30 40 50 60 70 y = m1*(1-exp(-m2*m0))+m3R 0.89031 NA Time (min) Value Error m1 15.702 1.1939 m2 0.12446 0.02217 m3 -1.3853 1.3254 B y = m1*(1-exp(-m2*m0))+m3 y = m1*(1-exp(-m2*m0))+m3 y = m1*(1-exp(-m2*m0))+m3 Chisq 2.0235 NA Value Error Value Error Value Error R 0.99828 NA m1 38.316 1.1947 m1 15.944 1.8342 m1 26.907 0.82328 320m2 0.056407 0.0068311 m2 0.079643 0.018067 m2 0.052985 0.008146 m3 -1.3095 0.55029 m3 1.4622 2.0176 m3 0.53467 1.104 Chisq 18.024 NA Chisq 1.1105 NA 280 Chisq 0.12701 NA R 0.99609 NA R 0.99529 NA R 0.99995 NA 240

200

1/A/ min 160

120

80

40

0 5 10 15 20 25 30 35 40 Cl-NIL (µ!)

Figure 2.4 Inactivation of DDAH1 by Cl-NIL (Tsou method)

(A) Time- and concentrationy =- dependent(10042*m2 + 42*M0)/(42*m... inactivation of purified DDAH-1 is observed Value Error when assayed in the presence ofm1 competing 0.31238 substrate0.12442 and 0 (¡), 5 (o), 8 (u), 12 (‚), 15 (+), 18.75 (r), 37.5 (l) µM ofm2 Cl -NIL.0.10367 Solid0.059696 lines are fits as described in Materials and Chisq 6.6311 NA Methods. (B) The inverse of theR apparent0.69826 inactivationNA rates (A) are replotted as described -1 in Materials and Methods to derive KI (0.10 ± 0.06 µM) and kinact (0.3 ± 0.1 min ) values for Cl-NIL inactivation of DDAH-1. 41 To confirm the findings of the Tsou method, a jump dilution assay was performed using the alternative substrate S-methyl-L-thiocitrulline (SMTC), and clearly show time- and concentration-dependent inactivation by each inactivator. Figure 2.5 shows two of the methods used to fit the observed inactivation rates (kobs) versus inhibitor concentration. Because several of the lower Cl-NIL concentrations (12.5 to 50 nM) were below or very near enzyme concentration, they were omitted in the direct fitting by Kitz-

Wilson method. To determine the reversible component of binding (KI) for the entire inhibitor concentration range, the observed inactivation rates (kobs) were fitted to the Morrison equation for tight-binding inhibitors. As seen in table 2.1, when fitted using the

-1 Kitz-Wilson method, a kinact of 0.25 ± 0.04 min and a KI of 0.2 ± 0.1 μM are observed.

These values are within error of the values achieved by the Tsou method. The KI determined by the Morrison equation is 0.3 ± 0.1 μM which agrees again with both the Kitz-Wilson and the Tsou methods. In a subsequent analysis, we fit the exponential loss of activity (Figure 2.5 A) using a global fitting software (KinTek Explorer) to minimize the need to manipulate the data for analysis.

42 A y = m3*(2.718^(-m2*M0)) Value Error y = m3*(2.718^(-m2*M0))y = m3*(2.718^(-m2*M0)) 100 m2 0.025669Value 0.00071506Error y = m3*(2.718^(-m2*M0))Value Error m2m3 0.04799796.271 0.000593330.9055 m2 Value-2.1722e-15 Error 0 Chisqm3 92.38697.149 0.88149NA m2 0.0072045m3 0.00036822100 1.3545 0 Chisqy = m3*(2.718^(-m2*M0))R 411.180.97413 NANA 0.02 80 m3 Chisq95.2543.7788e-230.50476 NA R 0.96417Value ErrorNA 0.04 m2 Chisq0.07648 R63.6730.0010188 1 NA NA 0.08 R 0.93962 NA 0.1 m3y = m3*(2.718^(-m2*M0))95.614 0.95892 0.2 60 Chisq 323.72Value ErrorNA 0.3 m2R 0.975760.33773 0.02351NA 0.4 m3 99.772 0.99938 Chisq 50.21 NA 40 R 0.99589y = m3*(2.718^(-m2*M0))NA

% Remaining Activity Value Error m2 0.54991 0.042505 20 m3 99.988 0.99999 Chisq 9.5522 NA R 0.99942 NA y = m3*(2.718^(-m2*M0)) 0 Value Error 0 10 20 30 40 50 60 70 m2 0.39892 0.025494 Time (min) m3 99.505 0.9988 Chisq 202.1 NA B R 0.98938 NA 0.25

0.2

y = (m2*M0)/(m1+M0)

) 0.15 Value Error -1 m1 211.48 120.83

(min m2 0.24614 0.04256

obs Chisq 0.0032083 NA k 0.1 R 0.91183 NA

y = (((m2+M0+m1)-((m2+M0+m1)... 0.05 Value Error m1 250.81 101.83 m2 0.50501 0.034804 Chisq 0.0040882 NA 0 R 0.96341 NA 0 500 1000 1500 2000 Cl-NIL (nM)

Figure 2.5 Inactivation of DDAH1 by Cl-NIL (Kiz-Wilson and Morrison methods)

(A) Exponential fits to the observed inactivation by different concentrations of Cl-NIL: 0 (¡), 12.5 (o), 50(u), 100 (‚), 200 (+), 400 (r), 800 (l), 1600 (n) nM in the presence of 300 μM SMTC. (B) Concentration dependence of the kobs values for inactivation, fit to either equation 1 (black line, open circles) or to equation 2 (red line, red crosses).

43 As mentioned earlier, global fitting eliminates the need for redundant calculations and some simplifying assumptions (31). A kinetic model was constructed as shown in scheme 2.1 where enzyme (E) and inhibitor (I) first form a reversible complex (EI) which is then inactivated (EX). The inactivation of EI to EX is assumed to be irreversible.

-1 -1 Assuming rapid binding of Cl-NIL, I fixed k+1 to the diffusion limit of 18,000 μM min . Figure 2.7 shows the fitting of the data using this model.

k1 k2 E+I EI EX k -1

Scheme 2.1 Global fitting model for enzyme inactivation

The data fit to this model decently well (Figure 2.6, χ2 = 736.23, χ2/DoF = 13.6)

-1 -1 and rate constants were extracted. A k-1 of 6000 ± 4000 min and a k2 of 0.5 ± 0.2 min were obtained where k2 corresponds to the rate of inactivation kinact and k-1/k+1 correspond

-1 -1 to Ki (0.4 ± 0.2 μM, when k+1 is set at 18,000 μM min ). To obtain an estimate of whether the information content of the data is sufficient to define these constants, FitSpace (30) calculations with a χ2 threshold limit of 2 were performed. The results demonstrate that k-1 and k2 vary as a pair and have a lower limit but not an upper limit (Figure 2.7), suggesting that the data contains sufficient information to extract the fit parameters in the model and respective confidence estimates (Table 2.2). The best-fit values were used, as they were the values in which we had confidence, and correspond to our values obtained from the other methods of fitting.

44

Figure 2.6 Global fit of DDAH1 inactivation by Cl-NIL to a kinetic model

Global fitting of the time-dependent inactiviation of purified DDAH1 to the model in scheme 2.1 by varying concentrations Cl-NIL : 0 (red), 20 (green), 40 (blue), 70 (yellow), - 100 (cyan), 200 (pink), 300 (forrest green), 400 (purple) nM. k+1 was set to 18,000 μM 1min-1

45

Figure 2.7 FitSpace contour plot for rate constants in the kinetic model

The FitSpace Confidence contour is shown for the values of k-1 and k+2 . The linearity shows that the two parameters vary with respect to each other and that there is a defined lower boundary but the large upper boundary indicates that it is not well defined.

Table 2.2 Rate constants obtained during global fitting to the model in scheme 2.1

*Upper and lower bounds were calculated with a χ2 threshold of 1.1 at the boundaries.

46 The kinetic mechanism supported by this data is consistent with the proposed chemical mechanism in which DDAH1 first, and quickly, forms a reversible complex with Cl-NIL, followed by the attack of the active-site cysteine residue and the irreversible loss of chloride and the resulting inactivation of DDAH1 as shown in scheme 2.2, most likely, path b. Path b is the more likely mechanism as it mimics the formation of the tetrahedral intermediate of DDAH1’s native mechanism (32) and there is evidence of path b shown for the inactivation of a related guandine-modifying enzyme: protein arginine deiminase (PAD) (33) by haloacetamidines.

Cl DDAH a.

S a. NH2 NH2 -Cl- S DDAH b. HN NH R R

b.

Cl b. NH2 NH2 - S S -Cl

DDAH NH DDAH NH

R R

Scheme 2.2 Inactivation of DDAH1 by chloroacetamidines

The reaction can proceed in one of two ways, pathway (a) describes an SN2 type mechanism while pathway (b) mimics the enzyme’s attack of the substrate. Figure adapted from (34). 47 As shown in table 2.1, the kinetic parameters from all four methods are consistent. It is evident that the Tsou method results in large fitting or very small fitting errors when compared with the other three methods. Using traditional methods of fitting such as the non-linearized version of the Kitz-Wilson is only valid when the inhibitor concentration is well above the concentration of the enzyme and the Morrison equation only allows the determination of the reversible component of binding in our case. The global fitting method is consistent with a two-step inactivation mechanism which consists of the fast formation of an initial reversible complex, followed by a slow inactivation of DDAH1 rather than a one-step inactivation as a one-step model did not fit this data well (not shown). Although some assumptions must be made in the global fitting method, comparing global fitting with at least one other analysis method gives more confidence in the values obtained and gives more insight into how many steps are needed in the enzyme’s inactivation as different models can be evaluated. Overall, the data show that Cl-NIL inactivates DDAH1 at about the same rate as

Cl-NIO, however the binding appears to be much more favorable with a KI of 0.4 ± 0.2

μM compared with Cl-NIO which has a KI of 1.8 ± 0.8 μM (from global fits). This 5-fold decrease in KI is likely due to the superior position in the active site allowed by the additional methylene in the backbone of the molecule. Likely, this is caused by a hydrophobic effect (35). The additional methylene causes the inhibitor to seek the more hydrophobic interior of the protein. Me-Cl-NIO, on the other hand, loses some potency due to the addition of the methyl group to the warhead portion of the molecule with a KI of 15 ± 2 μM. Me-Cl-NIL again loses even more potency with a KI of 30 ± 10 μM. The active site of DDAH1 can likely accommodate either additional length in the backbone of the linker portion of the molecule or additional bulk in the warhead portion but cannot

48 accommodate both as highlighted by the 75-fold difference in KI between Cl-NIL and Me-Cl-NIL.

Cl-NIL acts as a mechanism based inhibitor

Our lab has previously described the inactivation of DDAH1 and P. aeruginosa DDAH by chloroacetamidines and molecules containing the chloroacetamidine group as a warhead (10,12,34). To confirm that these inactivators react with DDAH1 in a 1:1 ratio, the partition ratio was determined for Cl-NIL and DDAH1 as see in figure 2.8. Figure 2.8 confirms that indeed, Cl-NIL reacts with DDAH1 in a 1:1 manner further confirming that Cl-NIL and chloroacetamidine derivatives tested here selective modifiers of DDAH1 and do not target multiple residues for modification. 1

0.8

0.6

Fractional Activty Fractional 0.4

0.2

0 0 1 2 3 4 5 6 7 [Cl-NIL]/[E]

Figure 2.8 Partition ratio of Cl-NIL to DDAH1

49 Partition ratio of Cl-NIL to DDAH1. DDAH1 (60 nM) was incubated with Cl-NIL (20, 40, 70, 100, 200, 400 nM) until reaction completion. Remaining activity was determined and plotted against the ratio of [Cl-NIL] to [DDAH1] to determine a partition ratio of 0.98 ± 0.02.

Cl-NIL forms a covalent bond with the active-site cysteine

Previous studies by our lab have shown by mass spectrometry that chloracetamidine (CAA) forms a covalent bond with the active-site cysteine of P. aeruginosa DDAH (34). We wanted to be able to visualize this covalent bond via x-ray crystallography to gain insight into how Cl-NIL interacts with the active site of DDAH1. We obtained a crystal structure of DDAH1 covalently bound to Cl-NIL at 1.9 Å. Table 2.3 contains the crystallographic data for this structure and figure 2.9 shows Cl-NIL covalently attached to the active site cysteine, Cys274.

50

Table 2.3 Crystallographic Data

51

Figure 2.9 Cl-NIL covalently bound to DDAH1

The crystal structure of DDAH1 with Cl-NIL shows density consistent with a covalent bond between Cl-NIL and active site Cys-274.

Cys274, the active site cysteine, forms a covalent bond with Cl-NIL following the loss of the chloride. The remainder of the protein structure is mostly unchanged following inactivation by Cl-NIL (see figure 2.10). The Cl-NIL bound structure of DDAH1 aligns very well with the structure of DDAH1 complexed with its product, L-citrulline (RMSD = 0.302) (27). The density from the methylene carbon of the warhead component occupies a space near Glu78 where one of the Nω-methyl groups of the substrate ADMA would likely occupy; this may be caused by the length gained from the addition of a methylene in the linker portion of the molecule as compared with Cl-NIO. The carboxyl portion of the inhibitor forms hydrogen bonds with Arg144 as in the L-citrulline bound structure (27). This indicates that the extra methylene allows the warhead portion of the

52 inhibitor to push itself further into the binding pocket of DDAH1, which may explain why the KI for Cl-NIL is 5 fold better than for Cl-NIO, which has the same number of carbons in the backbone chain as the substrate. Leung et al have found that often times, there is an affinity boost of 10-fold or more occurring with the addition of a methyl group on a compound (35). The improvement in affinity often arises from coupling the conformational gain of the methyl group with the burial of it in a hydrophobic region of the protein (35). The addition of the methylene in the chain showed a ΔΔG of -1.6 kcal/mol. The contribution either comes from Cl-NIL being pushed up further into the area that would accommodate the substrate, ADMA, or from they hydrophobic effect of the additional bulk in the backbone causing the inhibitor to seek the hydrophobic interior of the protein. Although there is a significant gain in the potency for Cl-NIL as compared to Cl-NIO, presumably there is also a negative entropic effect due to increasing linker portion of the molecule (36). His173 appears to be in a different conformation than it is in the crystal structure of DDAH1 bound with the product L-citrulline (PDB 2JAI) (27). The inactivation of DDAH1 by chloroacetamidines does not require assistance from His173 and appears to be in the same position as His173 from the apo-enzyme structure (PDB 3I2E) or His173 has moved to accommodate the larger warhead portion of Cl-NIL (11). The crystal structure also sheds light on the differences in activity between Me-

Cl-NIL and Me-Cl-NIO. Me-Cl-NIL and Me-Cl-NIO have a KI of 30 ± 10 μM and 15 ± 2

-1 -1 μM respectively, with similar kinact of 0.6 ± 0.2 min and 0.3 ± 0.2 min . By examining the structure of Cl-NIL covalently bound to DDAH1, it is apparent that Cl-NIL takes up the space likely used by the Nω-methyl groups of ADMA in the active site pocket, Me-Cl- NIL probably adds too much bulk to this area which excludes it from the active site hence the 75-fold increase in KI. Me-Cl-NIO on the other hand, mimics Cl-NIO more closely yet still has a larger KI. In this case, the extra methyl groups to mimic ADMA are again 53 unfavorable but less so than Me-Cl-NIL. Likely, DDAH1 can accommodate a longer chain or some extra bulk in the warhead region, but not both.

Figure 2.10 Overlay of DDAH1 bound with L-citrulline and Cl-NIL

DDAH1 bound with L-citrulline is shown in cyan (27) and with Cl-NIL in pink

I then compared the structure of Cl-NIL bound to DDAH1 with previously

5 5 studied reversible-covalent inhibitors, N -(1-iminoproply)-L-ornithine (IPO) and N -(1- iminopentyl)-L-ornithine (IPnO), which both inhibit DDAH1 with a KI of 52 μM and 7.5 μM respectively (11,37) and mimic the substrates of DDAH1, much like Cl-NIL and the other chloroacetamidines. Figure 2.11 shows the overlay of the three structures. One of the most striking differences is the movement of the active site Cys275 in the structure of DDAH1 with Cl-NIL compared to the other two structures. Perhaps the movement of the

54 warheard portion of Cl-NIL due to the lysine chain rather than the ornithine chain as with the other two molecules is part of the reason the KI for Cl-NIL is so much more potent (0.4 ± 0.2 μM) as compared with IPO, IPnO, and even Cl-NIO. IPnO is more potent that IPO perhaps due to the interactions of the longer methylene chain that points out towards the active site His173 (37). Cl-NIL’s longer chain backbone chain points out a bit in this same direction, perhaps aiding in the increased potency for this compound. One caveat of this comparison is that IPO and IPnO are covalent reversible inhibitors and their crystal structures are of that covalent reversible complex while that of Cl-NIL is a covalent irreversible product complex.

Figure 2.11 Overlay of DDAH1 bound with Cl-NIL, IPO, and IPnO

DDAH1 bound with Cl-NIL is shown in pink, with IPO is shown in green (11), and with IPnO is shown in purple (37).

55 Cl-NIL is selective for DDAH1 over Mn2+-containing human arginase I (Mn-hArgI)

2+ Mn -containing human arginase I (Mn-hArgI) catalyzes the conversion of L- arginine to L-ornithine and urea. Cross-inhibition of Mn-hArgI would be unfavorable because an increase in circulating L-arginine could potentially overwhelm NOS inhibition by ADMA (Figure 2.12). To address this possibility, Cl-NIL was tested as a potential inhibitor of Mn-hArgI. The inhibition of manganese-containing human arginase I by Cl-

NIL was fit to an IC50 value of 20 ± 2 mM and a Hill coefficient of 0.91 ± 0.02. Under these experimental conditions (substrate L-arginine concentration of 1.5 mM, KM = 2.3 mM), a KI value of 12 mM was calculated using the method of Cheng and Prussoff, assuming competitive inhibition (Figure 2.13) (24). This makes Cl-NIL more selective for DDAH1 than for Mn-hArgI by approximately 1x105 fold. Cl-NIO also shows selectivity for DDAH1 over Mn-hArgI but not by as much (>500-fold) (10). Because the selectivity for DDAH1 over Mn-hArgI is so marked, we predict that Cl-NIL will not affect cellular levels of L-arginine due to Mn-hArgI inhibition.

Figure 2.12. Metabolism of L-arginine and regulation by DDAH1 56 1.2

1

0.8

0.6

Fractional Activity 0.4

0.2

0 10 100 1000 104 105 [Cl-NIL] µM

Figure 2.13 Inhibition of Mn-hArgI by Cl-NIL

The activity of purified Mn2+ -containing human arginase is inhibited at millimolar concentrations of Cl-NIL. Concentration-dependent inhibition in the presence of L- arginine (1.5mM, KM = 2.3 mM) is fit by IC50 = 20 ± 2 mM and a Hill coefficient of 0.91 ± 0.02, which give a calculated KI value of 12 mM, assuming competitive inhibition.

Compounds inhibit DDAH1 within cultured cells

Activity based protein profiling probes can be very useful in confirming that a compound targets its protein of choice in a cellular setting. To this end, we used Click-Cl (12) to determine if the four compounds are able to inactivate episomally expressed myc- tagged DDAH1 in HEK293T cells. We have previously shown that Click-Cl is able to label episomally expressed myc-tagged DDAH1 in HEK293T cells and that we can use this compound to probe for “in cell” potency of active site inhibitors of DDAH1 (12). Briefly, myc-tagged DDAH1 is transiently transfected into HEK293T cells, 24 h after transfection Click-Cl is administered to intact cells and incubated for 10 min. After a 57 series of washings and cell lysis to obtain soluble cellular proteins, a click-chemistry mediated reaction is performed to append a biotin-PEO3-azide to any protein that was tagged by the alkyne-containing probe, Click-Cl. A two-color western blot is then performed, with DDAH1 visualized by an antibody for the genetically encoded myc tag (red), and DDAH1 that has been labeled by Click-Cl visualized by an antibody for the biotin tag (green), figure 2.14 is a visual representation of this process. Using this system, I tested the effects of four inactivators on DDAH1 within HEK293T cells (Figure 2.15). Briefly, prior to addition of Click-Cl but after transfection with myc-tagged DDAH1, 10 μM of each compound was added to a different well of cultured HEK293T cells and incubated for 15 min prior to the addition of Click-Cl for an additional 10 min incubation. As described in the materials and methods section, the red fluorescence and green fluorescence intensities were integrated using the imaging system to determine the percent inhibition of myc-tagged DDAH1 by the inactivators. Figure 2.15 shows that both Cl-NIL and Cl-NIO block almost all Click-Cl labeling while Me-Cl-NIO and Me- Cl-NIL are only able to partially block Click-Cl labeling during the same incubation time period. These results are consistent with the kinetic data obtained with purified DDAH1 and the inactivators.

58

Figure 2.14. Click-Cl labels DDAH1 active sites within cultured cells

Click-Cl labels episomally expressed myc-tagged DDAH1 (red E). After a brief incubation period, the cells are washed, lysed, and incubated with biotin-PEO3-azide using a click-chemistry reaction to label any protein that was tagged by Click-Cl. A two- color western blot is used to visualize DDAH1 by an antibody for the genetically encoded myc tag (red) and DDAH1 that has been labeled by Click-Cl is visualized by an antibody for the biotin tag (green).

59 A !"#$%&$'$(")## *+,!-.## *+,!-/# 01,*+,!-.# 01,*+,!-/# !"#2)"'1#

120

B 100

80

60

40 % Activity by % Activity Click-Cl Probe

20

0 Click-Cl Cl-NIL Cl-NIO Me-Cl-NIL Me-Cl-NIO No Click-Cl only

Figure 2.15 Chloroacetamidines inhibit DDAH1 in cultured HEK293T cells.

(A) Two-color western blot of in cell inactivation of myc-tag episomally expressed DDAH1 (red) and a response to a biotin-tagged activity probe (green) by the four compounds. 10 μM of each compound was added to separate wells for 15 min before the addition of the activity probe (110 μM) for an additional 10 min incubation. (B) Percent inhibition of DDAH1 in cultured HEK293T cells by the four chloroacetamidines using Click-Cl as a positive control.

To experimentally determine the “in cell” potency of Cl-NIL, we applied varying concentrations of the compound to HEK293T cells expressing DDAH1 followed by the

Click-Cl probe. We found the apparent “in cell” IC50 to be 10 ± 3 μM (Figure 2.16).

Apparent IC50 values are dependent on incubation time because Cl-NIL is a time- 60 dependent inactivator, yet this data clearly shows that Cl-NIL is able to label DDAH1 in this cellular condition (29). The apparent “in cell” IC50 value of Cl-NIL, which was determined under the same conditions as that of Cl-NIO, is comparable with that of Cl-

NIO (6.6 ± 0.2 μM) yet the in vitro KI value of Cl-NIL is 5 fold less that that of Cl-NIO.

A

B 1

0.8

0.6

0.4 Fractional Fluorescence Fractional

0.2

0 0.01 0.1 1 10 100 [Cl-NIL] µ!

Figure 2.16 Concentration dependent inhibition of DDAH1 in HEK293T cells by Cl-NIL.

(A) Two-color western blot of in cell inactivation of varying concentrations of Cl-NIL : 0, 0.078, 0.156, 0.313, 0.625, 1.25, 2.5, 5, 10, 20, 40, 80 μM. (B) Normalized fluorescence values for the biotin-derived signal are fit to give an apparent “in cell” IC50 value for DDAH1 inhibition of 10 ± 2 μM and a Hill coefficient of 0.8 ± 0.2. 61 Cl-NIL likely uses the same cationic amino acid transporter to enter the cell as ADMA (38). We speculate that either the cationic amino acid transporter is saturated hence the similar “in cell” IC50 values of Cl-NIL and Cl-NIO. Some other protein in the cell may also sequester Cl-NIL. To confirm that Cl-NIL is not toxic to cells under the same conditions for DDAH1 inhibition, we preformed a MTS toxicity assay. Figure 2.17 shows that while Cl- NIL does adversely affect the viability of cultured HEK293T cells, the adverse effects occur at a much higher concentration and alonger incubation time (72 h) than required for DDAH1 inhibition in the same cells. These findings indicate that Cl-NIL can be used as a potent bioavailable DDAH1 inhibitor, at least under these conditions.

120

100

80

60

40 % Viable Cells Remaining

20

0 1 10 100 1000 [Cl-NIL] µM

Figure 2.17 Toxicity of Cl-NIL on HEK293T cells.

62 Toxicity of Cl-NIL to the HEK293T cell line after a 72 h incubation was addressed using the MTS assay (see Materials and Methods) to determine an IC50 of 125 ± 6 μM for cell killing.

2.4 CONCLUSIONS

The compounds I synthesized covalently inactivate DDAH1 in vitro and all show some potency towards DDAH1 in cultured HEK293T cells. To the best of our knowledge, Cl-NIL is the most potent DDAH1 inhibitor to date, with a KI of 400 ± 200

4 -1 -1 nM and a kinact/KI of 2.4 × 10 M s and is selective for DDAH1 over Mn-hArgI by 1 × 105 fold. We obtained a crystal structure of DDAH1 covalently bound to Cl-NIL at the active site cysteine following the loss of the chloride. I used multiple kinetic analyses to confirm the inactivation kinetics that gives more confidence in the obtained values and can hopefully shed light on how to analyze similar, potent time-dependent enzyme inactivators. Future studies might benefit from a focus on restricting the rotation in the linker portion of the molecule to avoid paying an entropic penalty for freezing out rotation upon binding.

2.5 REFERENCES 1. Leiper J, Nandi M. The therapeutic potential of targeting endogenous inhibitors of nitric oxide synthesis. Nat. Rev. Drug Discov. 2011; (4):277–91.

2. Ekmekcioglu S, Ellerhorst JA, Prieto VG, Johnson MM, Broemeling LD, Grimm EA. Tumor iNOS predicts poor survival for stage III melanoma patients. Int. J. Cancer. 2006;119(4):861–6.

3. Fukumura D, Kashiwagi S, Jain RK. The role of nitric oxide in tumour progression. Nat. Rev. Cancer. 2006;6(7):521–34.

4. Sikora AG, Gelbard A, Davies MA, Sano D, Ekmekcioglu S, Kwon J, et al. Targeted inhibition of inducible nitric oxide synthase inhibits growth of human melanoma in vivo and synergizes with chemotherapy. J. Am. Assoc. Cancer Res. 2010; 16(6):1834–44. 63 5. Stuehr D., Griffith O. Mammalian Nitric Oxide Synthases. Adv. Enzym. Relat. Areas Mol. Biol. 1993;(65):287–346.

6. Leiper JM. The DDAH-ADMA-NOS pathway. Ther. Drug Monit. 2005;27(6):744–6.

7. Druhan LJ, Forbes SP, Pope AJ, Chen CA, Zweier JL, Cardounel AJ. Regulation of eNOS-derived superoxide by endogenous methylarginines. Biochemistry 2008;47(27):7256–63.

8. Pope AJ, Karuppiah K, Cardounel AJ. Role of the PRMT–DDAH–ADMA axis in the regulation of endothelial nitric oxide production. Pharmacol. Res. 2009 (6):461–5.

9. Kostourou V, Robinson SP, Cartwright JE, Whitley GSJ. Dimethylarginine dimethylaminohydrolase I enhances tumour growth and angiogenesis. Br. J. Cancer. 2002 Sep 9;87(6):673–80.

10. Wang Y, Hu S, Gabisi AM Jr, Er JAV, Pope A, Burstein G, et al. Developing an Irreversible Inhibitor of Human DDAH-1, an Enzyme Upregulated in Melanoma. ChemMedChem. 2014 Feb 26;

11. Wang Y MA. Developing dual and specific inhibitors of dimethylarginine dimethylaminohydrolase-1 and nitric oxide synthase: Toward a targeted polypharmacology to control nitric oxide. Biochemistry. 2009;48:8624–35.

12. Wang Y, Hu S, Fast W. A click chemistry mediated in vivo activity probe for dimethylarginine dimethylaminohydrolase. J. Am. Chem. Soc. 2009;131(42):15096– 7.

13. Knipp M, Vasak M. A colorimetric 96-well microtiter plate assay for the determination of enzymatically formed citrulline. Anal. Biochem. 2000;286(2):257– 64.

14. Tsou CL. Kinetics of substrate reaction during irreversible modification of enzyme activity. Adv. Enzym. Relat. Areas Mol. Biol. 1988;61:381–436.

15. Linsky T, Fast W. A continuous, fluorescent, high-throughput assay for human dimethylarginine dimethylaminohydrolase-1. J. Biomol. Screen. 2011 Oct;16(9):1089–97.

16. Morrison JF. Kinetics of the reversible inhibition of enzyme-catalysed reactions by tight-binding inhibitors. Biochim. Biophys. Acta. 1969;185(2):269–86.

17. Otwinowski Z, Minor W. [20] Processing of X-ray diffraction data collected in oscillation mode. In: Charles W. Carter J, editor. Methods in Enzymology. Academic Press; 1997 [cited 2014 Apr 14]. p. 307–26. 64 18. Vagin A TA. MOLREP: an automated program for molecular replacement. J. Appl. Cryst. 1997;(30):1022–5.

19. Emsley P LB. Features and development of Coot. Acta. Cryst. 2010;D66:486–501.

20. Adams PD AP. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta. Cryst. 2010;D66:213–21.

21. Read R. Improved Fourier coefficients for maps using phases from partial structures with errors. Acta. Cryst. 1986;Sect A 42:140–9.

22. Brunger A. Assessment of phase accuracy by cross validation: the free R value. Methods and applications. Acta. Cryst. 1993;D49:24–36.

23. Davis IW L-FA. MolProbity: all-atom contacts and structure validation for proteins and nucleic acids. Nucleic Acids Res. 2007;35:W375–83.

24. Cheng Y, Prusoff WH. Relationship between the inhibition constant (K1) and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction. Biochem. Pharmacol. 1973;22(23):3099–108.

25. Bełtowski J, Kedra A. Asymmetric dimethylarginine (ADMA) as a target for pharmacotherapy. Pharmacol. Rep. PR. 2006 Apr;58(2):159–78.

26. Forbes SP, Druhan LJ, Guzman JE, Parinandi N, Zhang L, Green-Church KB, et al. Mechanism of 4-HNE mediated inhibition of hDDAH-1: implications in no regulation. Biochemistry 2008; 47(6):1819–26.

27. Leiper J, Nandi M, Torondel B, Murray-Rust J, Malaki M, O’Hara B, et al. Disruption of methylarginine metabolism impairs vascular homeostasis. Nat. Med. 2007; (2):198–203.

28. Schwartz PA, Kuzmic P, Solowiej J, Bergqvist S, Bolanos B, Almaden C, et al. Covalent EGFR inhibitor analysis reveals importance of reversible interactions to potency and mechanisms of drug resistance. Proc. Natl. Acad. Sci. U S A. 2014; 111(1):173–8.

29. Copeland RA. Evaluation of enzyme inhibitors in drug discovery: a guide for medicinal chemists and pharmacologists. Hoboken, NJ: Wiley-Interscience; 2005.

30. Johnson KA. Fitting enzyme kinetic data with KinTek Global Kinetic Explorer. Methods. Enzymol. 2009;467:601–26.

31. Johnson KA. A century of enzyme kinetic analysis, 1913 to 2013. FEBS Lett. 2013;587(17):2753–66. 65 32. Stone EM, Costello A, Tierney D, Fast W. Substrate-assisted cysteine deprotonation in the mechanism of dimethylarginase (DDAH) from Pseudomonas aeruginosa. Biochemistry. 2006;17:5618–30.

33. Knuckley B, Causey CP, Pellechia PJ, Cook PF, Thompson PR. Haloacetamidine- based inactivators of protein arginine deiminase 4 (PAD4): evidence that general acid catalysis promotes efficient inactivation. Chembiochem 2010; 11(2):161–5.

34. Stone EM, Schaller TH, Bianchi H, Person MD, Fast W. Inactivation of two diverse enzymes in the amidinotransferase superfamily by 2-chloroacetamidine: dimethylargininase and peptidylarginine deiminase. Biochemistry. 2005;44(42):13744–52.

35. Leung C, Leung S, Tirado-Rives J, Jorgensen W. Methyl Effects on Protein-Ligand Binding. J. Med. Chem. 2012; 55:4489–500.

36. Lee Y, Marletta MA, Martasek P, Roman LJ, Masters BS, Silverman RB. Conformationally-restricted arginine analogues as alternative substrates and inhibitors of nitric oxide synthases. Bioorg. Med. Chem. 1999; (6):1097–104.

37. Lluis M, Wang Y, Monzingo AF, Fast W, Robertus JD. Characterization of C-alkyl amidines as bioavailable covalent reversible inhibitors of human DDAH-1. ChemMedChem. 6(1):81–8.

38. Vasta V, Meacci E, Farnararo M, Bruni P. Identification of a specific transport system for L-arginine in human platelets. Biochem. Biophys. Res. Commun. 1995; 206(3):878–84.

66

Chapter 3: Inhibition of Dimethylarginine Dimethylaminohydrolase by Angeli’s Salt

3.1 INTRODUCTION

Nitric oxide (NO) plays multiple physiological and pathological functions in biology, from regulating cardiovascular processes to aiding with the immune response to regulating brain function (1–4). Researchers have investigated the oxidation products of nitric oxide to further study how this tiny molecule can possibly affect so many processes. Some of the well-known oxidation products are peroxynitrite, nitrogen dioxide and nitrogen trioxide. One of the least studied and most intriguing molecules is nitroxyl

(HNO, pKa = 11.4) (5), the product of one-electron reduction of NO (6). Recent discoveries that HNO may act as an endogenous mediator of different biological effects suggest that regulation of HNO pathways may have promising pharmaceutical applications (7,8). NO and it’s redox sibling, HNO, often target similar proteins in vitro and in vivo but may result in the formation of different products through distinct reaction mechanisms (7,9). NO reacts with protein thiols, metal complexes, hemes and radicals (10). HNO chemistry differs from NO chemistry in that it prefers to react with oxidized heme and metal complexes as well as thiols but does not readily react directly with, or generate radicals (7). HNO can directly modify thiols, but NO must use autoxidation to react with thiols (9). Metals and sulfhydryls react most readily with HNO. HNO reacts with metal complexes at a similar rate with NO (11). The next most reactive targets for HNO are thiols and have recently been proposed as the most relevant biological targets for HNO (9). Thiols react with HNO to form an N-hydroxy-sulfenamide (RSN(H)OH) 67 (12) and in the case of glutathione (GSH) the predominant reaction is a rearrangement to sulfonamide (RS(O)NH2). The reaction rate of HNO with small molecule thiols is typically slower than with protein thiols (7). Thus the environment in which HNO is produced and reacts must have a profound effect on which thiols become modified. A summary of the chemical reactions that HNO and NO may undergo in biological conditions is shown in Figure 3.1. O

S R NH2 N O + H O 2 2 R'S-NO + RSH III - H2O HbFe + NO3

R'SH 2 HNO

RS NH2+ OH -O R'SH RSH II RS-NO + NO - + H+ Mn-1 -NO+ HNO 2

I n-1 + HbFe + Cu M + H RSH +H n -. OH M O2 + .NO ONOO- RS N -H NO- CuII OH H .NO - . . R'SH RS + NO O2 R

N O ONOO- OH NH2OH + RSSR' 2 2 +H+ .NO II + .NO HbFe -O , H NO2 2 R ONOONO OH- + N2O O . III H2NO + H2O2 + HbFe N2O2

. + R H2NO +H - 2 NO2

- - NO2 + N2O NO2 N2 + 2H2O H O N O 2 - - + 2 4 NO2 + NO3 + 2H

Figure 3.1 NO and HNO reactions in biology

Figure taken and adapted from (7) The formation of endogenous HNO in mammalian cells has not yet been proven, however many of the precursors and pathways that would lead to HNO formation have

G been found in vivo (6). HNO may be derived from the oxidation of N -hydroxy-L- arginine, an intermediate of nitric oxide synthases enzymatic conversion of L-arginine to NO (13) or possibly from the direct reduction of NO via ubiquionol, cytochrome,

68 MnSOD, and xanthine oxidase (6,8). Another possibility is from the decomposition of nitrosothiols (14) or heme protein mediated peroxidation of hydroxylamine (15). These reactions make it plausible for HNO to potentially regulate signals in biological systems.

Nitric oxide is produced by nitric oxide synthases (NOS) which convert L- arginine to nitric oxide and citrulline in an NADPH dependent oxidation (16). NOS enzymes are catalytically active as homodimers that use flavin adenine dinucleotide, flavin mononucleotide, heme, tetrahydrobioptern, calmodulin, Zn2+ and Ca2+ as cofactors. This reaction is inhibited by endogenously produced Nω-methylated arginines which are in turn metabolized by the enzyme human dimethylarginine dimethylaminohydrolase (DDAH1) (17). DDAH1 catalyzes the conversion of asymmetric dimethylarginine

(ADMA) and monomethylarginine (L-NMMA) to L-citrulline and the corresponding methylamine (18). Nitric oxide produced by NOS can directly reversibly inhibit DDAH1 by the S-nitrosation of the active site cysteine residue which results in the accumulation of ADMA and thus feedback inhibition of NOS (19). Figure 3.2 highlights this regulation process.

H2N NH2 H2N O O2, NADPH NO Synthase NO + NH NH R R Arginine Citrulline

DDAH1

N NH2

NH2

NH Dimethylamine R

Asymmetric dimethylarginine + - R = (CH2)3CHNH3 COO

Figure 3.2 The metabolism of ADMA is regulated by DDAH1 and NO 69 Given the similarities of HNO and NO biology, we set out to determine whether DDAH1, which uses an active site cysteine and is regulated by NO, could also be regulated by HNO. HNO dimerizes with itself to form hyponitrous acid and then decomposes to form nitrous oxide and water. As such, it is impossible to store HNO, so HNO is typically studied using donor molecules to generate HNO in situ (6). To study the effects of HNO on DDAH1, I used the well-known HNO donor molecule Angeli’s salt

(Na2N2O3) to conduct the experiments. Angeli’s salt is stable at basic pH (pH > 10) and

- decomposes to nitroxyl (HNO) and nitrite (NO2 ) at physiological pH or below, as shown is scheme 3.1, and is the most common HNO donor molecule used in biological studies.

pH (4-8) + 2- + - H + N2O3 HNO NO2 aqueous conditions

Scheme 3.1 Angeli’s Salt decomposes nitroxyl and nitrite

3.2 MATERIALS AND METHODS

Materials

Unless otherwise noted, all materials were purchased from Sigma Aldrich Chemical Company (Saint Louis, MO).

Expression and Purification of DDAH1

Recombinant wild-type human DDAH1 bearing an N-terminal His6 affinity tag was expressed using an expression plasmid with a reengineered N-terminus (pET28a- hDDAH-1re) in BL21 (DE3) Escherichia coli cells to avoid N-terminal nonenzymatic gluconylation as described earlier (20). Protein was purified according to the procedure as described earlier (21) with the following modifications: 25 mL of Elution Buffer (50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole, 15% glycerol, pH 8.0) was used to

70 elute protein from an Ni-NTA affinity resin (8 mL, Qiagen). The eluent was diluted into

Buffer A (50 mM NaH2PO4, 3.0 M NaCl, 15 % glycerol, pH 7.0) to a final NaCl concentration of 1.8 M and loaded onto a 1.5 × 18 cm phenyl-sepharose column (GE Healthcare, Piscataway, NJ). Proteins were eluted by a stepwise gradient from the described phosphate Buffer A but with an NaCl concentration of 1.8 M, followed by 1.6 M, followed by 1.3 M, followed by 1 M. As gauged by SDS-page, fractions containing DDAH1 were pooled and concentrated using an Amicon Centrifugal Filter (Millipore, Billerica) with a 10 kDa molecular weight cutoff. The final protein was dialyzed overnight in Dialysis Buffer 1 (2 mM 1,10-phenathroline, 100 mM KCl, 10 % glycerol, pH 7.3) followed by 2 × 4 h dialyses in Dialysis Buffer 2 (50 mM KH2PO4, 100 mM KCl, 10 % glycerol, pH 7.3), flash frozen and stored in aliquots at -80 oC.

Time-dependent Inactivation of DDAH1 by Angeli’s Salt

Purified His6-DDAH1 (2 μM) was incubated with 1 % Tween-20, 20 mM EDTA and 200 μM of Angeli’s salt (stock solution kept at 0.5 M in 10 mM NaOH) in Phosphate

Buffer (100 mM KH2PO4, 100 mM KCl, pH 7.3). To test for time-dependent loss of activity, aliquots were removed from the incubation at time points (0-80 min) and diluted fifty-fold into an assay solution containing an excess (2 mM) of substrate, asymmetric dimethylarginine (ADMA). Remaining enzyme activity was assayed by detecting product formation (L-citrulline), after an incubation time of 40 min and quenching with 10 μL 6 N trifluoroacetic acid, using a color developing reagent (COLDER) assay (22). Angeli’s salt decomposes to HNO at a non-linear rate (6) hence an attempt to characterize a concentration dependent rate constant for the inactivation of DDAH1 by HNO using Angeli’s salt as a donor species was not made. Using one concentration of Angeli’s salt

71 inactivation data was fit directly to a single exponential to obtain an observed rate (kobs) for the inactivation of DDAH1.

Time-dependent Inactivation by Decomposed Angeli’s Salt

Prior to an incubation like that described above for Angeli’s salt and DDAH1, Angeli’s salt was allowed to decompose overnight at 37 oC in Phosphate Buffer in the absence of DDAH1. A time-dependent loss of DDAH1 activity experiment was conducted as described above using 200 μM of decomposed Angeli’s salt and a positive control of freshly diluted Angeli’s salt at 200 μM.

Time-dependent Inactivation by Nitrite

- Angeli’s salt releases nitroxyl (HNO) and nitrite (NO2 ) in a 1:1 ratio (6). To determine whether nitrite or nitroxyl is the inactivating species, a time-dependent loss of activity experiment was conducted with sodium nitrite (200 μM) as well as fresh Angeli’s salt (200 μM) as a positive control, as described above.

Time-dependent Inactivation by GSH Incubated Angeli’s Salt

HNO is known to react with thiols (8). To test this possibility, Angeli’s salt (400 μM) was incubated with 5 mM glutathione (GSH) in Phosphate Buffer for 1 h prior to a time-dependent loss of DDAH1 activity assay as described above (200 μM final concentration of Angeli’s salt) with freshly prepared Angeli’s salt as a positive control.

Acid Quench of Inactive DDAH1

To characterize a putative covalent DDAH1 adduct found by nitroxyl treatment, DDAH1 was incubated with excess Angeli’s salt for 2 h prior to acid quench. Typically, 69 μM protein was mixed with 400 μM of Angeli’s salt in DDAH1 Assay Buffer (0.1 M

KH2PO4, 1 mM EDTA, pH 7.3) in 100 μL total volume. Reactions mixtures were

72 incubated for 2 h at 25 °C, followed by quenching with 1 M trifluoroacetic acid to a final concentration of approximately 80 mM. Samples were directly frozen at -80 °C and subjected to Liquid Chromatography Electrospray Ionization Mass Spectrometry (LC- ESI-MS) analysis.

LC-ESI-MS Analysis

LC-ESI-MS analysis was performed by Dr. Stony Lo in the Analytical Core, College of Pharmacy, UT-Austin. An electrospray ion trap mass spectrometer (LCQ, Finnigan MAT, San Jose, CA) coupled on-line with a microbore HPLC (Magic 2002, Michrom BioResources, Auburn, CA) was used to acquire spectra of proteins. The protein sample was loaded through a protein desalting trap (Michrom BioResource) and desalted by 4 × 50 μL of mobile phase A (acetonitrile: water: acetic acid: trifluoroacetic aicd, 2:98:0.1:0.02) then injected into microbore HPLC. Protein was eluted with a 0.5 × 50 mm PLRP-S column (8 μm particle diameter, 4000 Å pore size; Michrom Bioresources, Auburn, CA) with mobile phase A and B (acetonitrile: water: acetic acid: trifluoroacetic acid, 90:10:0.09:0.02). The gradient used to elute protein was from 5 % to 95 % of mobile phase B in 10 min followed by 95 % B for 5 min at a flow rate of 20 μL/min. Some of the protein samples were eluted with 50 % B (40 μL/min) into LCQ without a column. Automated acquisition of full scan MS spectra was executed by Finnigan ExcaliburTM software. The settings for the ESI were as follows: spray voltage, 4.5 kV; nitrogen sheath gas and auxiliary gas flow rates, 60 and 5 psi, respectively; capillary temperature, 200 °C; capillary voltage, 22 V; tube lens offset, 40 V. The electron multiplier was set at -860 V; the scan time setting was performed with 50 msec of maximum injection time for full scan. The full scan range for MS was 150-2000Da. The acquired convoluted protein spectra from LCQ were deconvoluted by the Finnigan-

73 MAT BIOWORKS software to afford the MH+ m/z value(s) of the protein sample.

In Cell Inhibition of DDAH1 by Angeli’s Salt A competitive labeling strategy was used to determine if HNO donated by

Angeli’s salt could label DDAH1 in HEK293T cells. Cultured HEK293T cells were seeded in two 6-well polystyrene plates using complete growth medium containing

DMEM with 10 % FBS (Invitrogen, Carlsbad, CA) and grown to 80 % confluency. The pEF6a-hDDAH-1 plasmid (23) was transiently transfected into HEK293T cells using

Lipofectamine 2000 (Invitrogen, Carlsbad, CA). After 24 h, spent medium was removed and cell was washed with 1 mL of 0.5 mL phosphate buffered saline (PBS) at pH 7.2

(Invitrogen, Carlsbad, CA). Stock solutions of Angeli’s salt (10 mM) were diluted into complete growth medium in the cultures (1000 μL each well) to give final concentrations of 1000, 500, 250, 50, 0 μM. The resulting cultures were subsequently incubated for 15

min at 37 ºC in an atmosphere of 5% CO2 before addition of 110 μM of an alkyne labeled

DDAH1 activity based protein profiling probe (Click-Cl), described elsewhere (23), followed by an additional 10 min incubation. After treatment, cells were washed two times with PBS (1 mL) and harvested in 1 mL PBS followed by centrifugation at 14000 g for 5 min at 4 ºC. Cell pellets were stored at – 80 ºC.

Frozen cell pellets were lysed and labeled with biotin-PEO3-azide using click chemistry (23). Briefly, the cell pellets were resuspended in PB buffer (10 mM sodium / potassium phosphate buffer at pH 8.0 containing the complete mini EDTA-free protease cocktail inhibitor (Roche, Indianapolis, IN) and Triton X-100 (1% v/v)) and then were subjected to three cycles of vortexing, freezing in liquid N2, and thawing at room 74 temperature. The cells were then centrifuged at 14000 g for 5 min at 4 ºC to remove insoluble cell debris. The supernatant was removed and total protein concentration was determined using a Bradford assay. Biotin-PEO3-azide was appended to alkyne-labeled proteins using the Cu (I) catalyzed 1,3 dipolar cycloaddition reaction as follows: 25 μg total protein in 50 μL PB buffer was mixed with 0.5 μL of biotin-PEO3-azide (2.5 mM), 0.5 μL CuSO4 (50 mM), 0.5 μL tris(2-carboxyethyl)phosphine (TCEP, 50 mM) and 1.65 μL tris[(1-benzyl-1H-1,2,3-trazol-4-yl)methyl]amine (TBTA, 1.5 mM), vortexed and incubated at 25° C for 1 h. Reactions were quenched by addition of 2 × SDS loading buffer and heat denatured at 100° C for 10 min. Samples were directly used for SDS- PAGE or stored at -80° C before analysis. Two-color Western blot detection was used to detect the expression levels of DDAH1 and the response to the biotin tag as described previously (23). Images were scanned using an Odyssey Infrared Imaging System (Li- Cor Biosciences, Lincoln, NE) at the Core DNA Facility (University of Texas, Austin). Fluorescence intensities for both 680 nm and 800 nm channels were integrated to determine percent inhibition.

3.3 RESULTS AND DISCUSSION

Angeli’s Salt Covalently Inactivates DDAH1

Angeli’s salt is a known HNO donor molecule that releases HNO with a first- order rate constant of 6.8 x 10-4 s-1 at 25 oC (6). To test if one of the species donated by Angeli’s salt inactivates DDAH1, I performed a time-dependent inactivation study of Angeli’s salt with DDAH1. One of the species donated by Angeli’s salt irreversibly inactivates DDAH1 in a time-dependent manner as shown in Figure 3.3. The concentration dependence of DDAH1 inactivation by Angeli’s salt was not determined

- due to complications arising from the non-linear relationship between HNO and NO2

75 release and Angeli’s salt concentration (24). However, an observed rate of 0.0017 ± 0.0005 s-1 was found upon fitting the data to a single exponential which indicates that

- HNO or NO2 reacts with DDAH1 upon its release from Angeli’s salt. One explanation for time-dependent irreversible inactivation of DDAH1 is the formation of a covalent adduct, so an incubation of DDAH1 with Angeli’s salt was characterized by ESI-MS. The prominent adduct formed from HNO reacting with a thiol is a sulfinamide or a sulfenamide (12), which would both correspond to a + 32 Da adduct. Unmodified DDAH1 was first used as a positive control with an expected Mw = 33,303 Da. The ESI-

MS analysis for unmodified DDAH1 showed two major peaks of Mwexptl = 31,405 and

33,303 Da (31,405 corresponds to wild-type DDAH1 that has lost its N-terminal His6 tag, presumably due to contaminating proteases, whereas 33,303 corresponds to expected wild-type DDAH1 that has retained the N-terminal tag) (Figure 3.4B). DDAH1 was also incubated with Angeli’s salt for 2 h (as described in the Materials and Methods) followed by quenching, buffer exchange, and ESI-MS analysis to yield Mwexptl = 31,439 and 33,341 Da (Figure 3.4A). These modifications correspond to a + 34 and + 38 (± 10) Da gain, respectively, to the treated full length DDAH1 sequence, which is within error to the expected + 32 Da adduct from a thiol reacting with HNO (14). An attempt to characterize which amino acid was modified was unsuccessful because the covalent modification was lost during peptide digestion (data not shown). Possible adducts that account for the increase in mass are shown in Scheme 3.2.

76 100 y = m1*(2.718^(m2*M0)) Value Error m1 76.666 9.432 80 m2 -0.097119 0.033763 Chisq 1208.9 NA R 0.91769 NA

60

% Remaining Activity % Remaining 40

20

0 0 20 40 60 80 100

Time (min)

Figure 3.3 Angeli’s salt inactivation of DDAH1

The time dependent inactivation of DDAH1 by Angeli’s salt was carried out as followed: 2 μM of DDAH1 was incubated with 200 μM of Angeli’s salt in phosphate buffer. Aliquots were diluted 50-fold into excess substrate (ADMA, 2 mM) at 1, 3, 5, 12, 20, 40, 60 and 80 min. The concentration of L-citrulline was determined by the COLDER assay and the percent remaining activity of DDAH1 was plotted against time, showing time- dependent inactivation by Angeli’s salt.

77 A

B

Figure 3.4 ESI-MS of covalently modified DDAH1

(A) ESI-MS spectra of modified DDAH1, the taller peak corresponds to a Mw = 31,439 Da while the shorter corresponds to Mw = 33,341 Da. (B) ESI-MS spectra of unmodified DDAH1, the taller peak corresponds to a Mw = 31,405 Da while the shorter corresponds to a Mw = 33,303 Da.

78

Scheme 3.2 Possible adducts formed by Angeli’s salt inactivation of DDAH1

Circled are the proposed adducts formed during covalent modification of DDAH1 by a HNO. Adduct 1 (N-hydroxysulfenamide) and 2 (sulfinamide) each correspond to a +32

Da addition (addition MWcalc = 32 Da, MWexptl= 34 Da and 38 Da)

HNO is the Predominant Inactivating Species of DDAH1 by Angeli’s Salt

- Angeli’s salt releases one equivalent of HNO plus one equivalent of NO2 upon decomposition at physiological pH (25). HNO is thiol reactive and predicted to be the active species of Angeli’s salt (24,26). To rule out nitrite as the molecule responsible for DDAH1 inactivation, sodium nitrite was tested along side of Angeli’s salt. Figure 3.5

- shows that NO2 does not inhibit DDAH1 in a time-dependent manner as Angeli’s salt does, indicating that HNO as the molecule responsible for DDAH1 inactivation. Here we determined the same observed rate of 0.0017 ± 0.0005 s-1 for Angeli’s salt but decay of activity was not observed for sodium nitrite.

79 100

80

60

40 % Remaining Activity % Remaining

20

0 0 20 40 60 80 100

Time (min)

- Figure 3.5 Time-dependence of DDAH1 inactivation by NO2 or HNO

The time-dependent inactivation of DDAH1 by HNO released by Angeli’s salt (black - circles) compared with NO2 (black squares) was carried out by the same procedure as described for the time-dependent inactivation of DDAH1 by Angeli’s salt. 200 μM of sodium nitrite or 200 μM of Angeli’s salt were incubated with 2 μM of DDAH1 followed by a 50-fold dilution into excess substrate (ADMA 2 mM) at 1, 3, 5, 12, 20, 40, 60 and 80 min. The percent remaining enzyme activity was calculated from product concentration determined by the COLDER assay. HNO spontaneously dimerizes to give hyponitrous acid that then decomposes to nitrous oxide and water (Scheme 3.3). Nitroxyl donors are used to circumvent this self- dimerization. However, care must be taken to confirm that the active species from the donor molecule is actually nitroxyl. I addressed this by first testing nitrite (Figure 3.6) and then I allowed Angeli’s salt to decompose at physiological pH (7.3) overnight at 37 oC to address whether any of the downstream products from HNO decomposition 80 modifies DDAH1. Nitrous oxide is likely the predominating species remaining in a decomposed solution of Angeli’s salt (6). Figure 3.5 shows the time-dependent inactivation of DDAH1 by the decomposed Angeli’s salt compared with fresh Angeli’s salt. The data is shown in a bar graph for clarity. About 40% DDAH1 activity remains with decomposed Angeli’s salt compared to complete inactivation by freshly prepared

Angeli’s salt. Likely, either hyponitrous acid, N2O, peroxynitrite (which can form upon HNO oxidation) (6), another unidentified molecule from Angeli’s salt decomposition, or unreacted HNO, has inactivated DDAH1. The inactivating species completely reacted with DDAH1 upon the first dilution attempt, as further time-dependence is not seen in the plot (Figure 3.5). Given that fresh Angeli’s salt completely inactivates DDAH1 and the adduct formed as seen in the mass spectrometry analysis is consistent with HNO reacting with a thiol, it is likely that inactivation of DDAH1 by fresh Angeli’s salt is from HNO due to the fact that the dimerization process and then subsequent breakdown to nitrous oxide takes more than the allotted time during the enzyme inactivation experiment (27).

HNO + HNO ⇌ HONNOH → N2O + H2O

Scheme 3.3 HNO dimerization

81 100

80

60

40 % Remaining Actiivity % Remaining

20

0 0 1 10 30 60 Time (min)

Figure 3.6 Inactivation of DDAH1 by freshly prepared and decomposed Angeli’s salt

The time-dependent inactivation of DDAH1 by HNO released by freshly diluted Angeli’s salt (black bars) compared with Angeli’s salt that had been allowed to incubate at pH 7.3 at 37 oC overnight (white bars) was carried out by the same procedure as described for the time-dependent inactivation of DDAH1 by Angeli’s salt. 200 μM of decomposed or 200 μM of Angeli’s salt were incubated with 2 μM of DDAH1 followed by a 50-fold dilution into excess substrate (ADMA 2 mM) at 1, 3, 5, 12, 20, 40, 60 and 80 min. The percent remaining enzyme activity was calculated from product formation determined by the COLDER assay.

GSH Prevents DDAH1 Inactivation by Angeli’s Salt in vitro.

HNO is known to react with thiols in biological systems (5). Glutathione (GSH) is often used to test whether certain compounds would be available to thiol-containing proteins in a biological setting. Angeli’s salt was pre-incubated with glutathione (GSH) prior to a time-dependent inactivation assay of DDAH1 to determine if HNO would react 82 with GSH over DDAH1. Figure 3.7 shows that pre-incubation of Angeli’s salt with GSH does protect DDAH1 from inactivation by HNO. This is not an unexpected given that HNO is known to be thiol reactive. However, several studies have shown that HNO reacts preferentially with some protein thiols rather than with smaller thiols like GSH (6,7,14) and HNO is thought to possibly be locally produced by an intermediate of NOS under some conditions (11), so it is plausible that although there is much more GSH in a cellular environment than DDAH1, HNO may still preferentially react with DDAH1 rather than GSH under some circumstances. To attempt to address this issue, I preformed an assay to determine if HNO produced by Angeli’s salt would inactivate DDAH1 in HEK293T cells.

83 100

80

60

40 % Remaining Activity % Remaining

20

0 0 20 40 60 80 100

Time (min)

Figure 3.7 Inactivation of DDAH1 by Angeli’s salt before and after treatment with GSH

The time-dependent inactivation of DDAH1 by HNO released by Angeli’s salt (black circles) compared with Angeli’s salt pre-inubated for 1 h with GSH (black squares) was carried out by the same procedure as described for the time-dependent inactivation of DDAH1 by Angeli’s salt. 200 μM of GSH reacted Angeli’s salt or 200 μM of Angeli’s salt were incubated with 2 μM of DDAH1 followed by a 50-fold dilution into excess substrate (ADMA 2 mM) at 1, 3, 5, 12, 20, 40, 60 and 80 min. The percent remaining enzyme activity was calculated from product formation determined by the COLDER assay.

Angeli’s Salt does not Inactivate DDAH1 in HEK293T Cells.

We previously established a system using an activity based protein profiling probe (Click-Cl) to test for active DDAH1 in HEK 293T cells (23). Briefly, myc-tagged DDAH1 is transiently transfected into HEK293T cells. After transfection (24h), Click-Cl is administered to intact cells and incubated for 10 min. After a series of washings and 84 cell lysis to obtain soluble cellular proteins, a click-chemistry mediated reaction is performed to append a biotin-PEO3-azide to any protein that was tagged by the alkyne- containing probe, Click-Cl. A two-color western blot is then performed, with DDAH1 visualized by an antibody for the genetically encoded myc tag (red), and DDAH1 that has been labeled by Click-Cl visualized by an antibody for the biotin tag (green). Using this system, I tested the effect of Angeli’s salt on the available active sites of transiently transfected DDAH1 within HEK293T cells (Figure 3.8). Briefly, prior to addition of Click-Cl but after transfection with myc-tagged DDAH1, varying concentrations of Angeli’s salt were added to cultured HEK293T cells and incubated for 15 min at 37 oC. As described in the materials and methods section, the red fluorescence and green fluorescence intensities were integrated using the imaging system to determine the percent inhibition of myc-tagged DDAH1 by Angeli’s salt. No appreciable inhibition was seen by this method, indicating that HNO donation by Angeli’s salt in this specific cellular context does not lead to DDAH1 inactivation. By introducing Angeli’s salt into cells in such a manner as done in these experiments, GSH or other more abundant thiol- containing proteins could have interacted with the HNO generated from Angeli’s salt thus showing no apparent inhibition of DDAH1. Another limiting factor for this experiment could be that HNO forms an adduct with DDAH1 somewhere other than the active site and the activity based probe Click-Cl will only hit active sites of DDAH1. There are six cysteines in the DDAH1 sequence (including the active site cysteine), any which could have reacted with HNO to form the adduct seen by the mass spectrometry analysis, however this is less likely as these cysteines are buried in the protein (Figure 3.4).

85 A

120 B 100

80

60 % Remaining Activity % Remaining 40

20

0 0 50 100 250 500 [Angeli's Salt] (µ!)

Figure 3.8 In cell inhibition of DDAH1 by Angeli’s salt.

(A)Two-color Western Blot reflect the presence of myc (red) and biotin (green) tags after labeling of overexpressed DDAH1 in intact HEK293T cells in the presence of up to 500 μM of Angeli’s salt. (B) Bar graph representration of the percent remaining enzyme acitivty in cell vs. Angeli’s salt concentration. The percent remaining activity was determined as described in the materials and methods section by normalizing the red to green fluorescence intensities and comparing with a positive control (no Angeli’s salt addition).

3.4 CONCLUSIONS

In conclusion, HNO donated by Angeli’s salt can irreversibly inactivate DDAH1 and form one of two possible covalent modifications (Scheme 3.2) but based on previous 86 studies (14) the likely adduct formed is the sulfinamide, which has been shown to be the predominant adduct formed when both GSH and glyceraldehyde-3-phosphate dehydrogenase are reacted with Angeli’s salt (27). DDAH1 is inactivated by HNO and

- not by NO2 , the other released molecule of Angeli’s salt. A possible HNO dimerization product, perhaps nitrous oxide, also inhibits DDAH1. Nitrous oxide can be produced in a biological context (7) hence its effect on DDAH1 would make an interesting study. A pre-incubation of Angeli’s salt with GSH prevents DDAH1 inactivation and the cellular assay did not show inhibition of overexpressed DDAH1 in HEK 293T cells, indicating that in this context, DDAH1 is not the primary target of nitroxyl. However, HNO has been shown to be more reactive towards some protein thiols rather than smaller ones like GSH by several orders of magnitude (9) and is thought to be locally produced as a byproduct of the nitric oxide synthase reaction (5). These factors should be considered as well as the limitations of introducing a donor molecule such as Angeli’s salt into cultured cells as this experiment may not mimic the exact biological situation.

3.5 REFERENCES 1. Augustyniak A, Skolimowski J, Błaszczyk A. Cytotoxicity of nitroxyl (HNO/NO−) against normal and cancer human cells. Chem. Biol. Interact. 2013; 206(2):262–71.

2. Fukumura D, Kashiwagi S, Jain RK. The role of nitric oxide in tumour progression. Nat. Rev. Cancer. 2006;6(7):521–34. z

3. Huang Z, Huang PL, Panahian N, Dalkara T, Fishman MC, Moskowitz MA. Effects of cerebral ischemia in mice deficient in neuronal nitric oxide synthase. Science. 1994; 265(5180):1883–5.

4. Bogdan C. Nitric oxide and the immune response. Nat. Immunol. 2001; (10):907–16.

5. Fukuto JM, Jackson MI, Kaludercic N, Paolocci N. Examining nitroxyl in biological systems. Methods Enzymol. 2008;440:411–31.

87 6. Flores-Santana W, Switzer C, Ridnour LA, Basudhar D, Mancardi D, Donzelli S, et al. Comparing the chemical biology of NO and HNO. Arch. Pharm. Res. 2009; (8):1139–53.

7. Paolocci N, Wink DA. The shy Angeli and his elusive creature: the HNO route to vasodilation. Am. J. Physiol. Heart Circ. Physiol. 2009; 296(5):H1217–1220.

8. Flores-Santana W, Salmon DJ, Donzelli S, Switzer CH, Basudhar D, Ridnour L, et al. The specificity of nitroxyl chemistry is unique among nitrogen oxides in biological systems. Antioxid. Redox. Signal. 2011;14(9):1659–74.

9. Stuehr D., Griffith O. Mammalian Nitric Oxide Synthases. Adv. Enzym. Relat. Areas Mol. Biol. 1993;(65):287–346.

10. Paolocci N, Jackson MI, Lopez BE, Miranda K, Tocchetti CG, Wink DA, et al. The pharmacology of nitroxyl (HNO) and its therapeutic potential: not just the Janus face of NO. Pharmacol. Ther. 2007; 3(2):442–58.

11. Doyle MP, Mahapatro SN, Broene RD, Guy JK. Oxidation and reduction of hemoproteins by trioxodinitrate(II). The role of nitrosyl hydride and nitrite. J. Am. Chem. Soc. 1988; 110(2):593–9.

12. Fukuto JM, Chiang K, Hszieh R, Wong P, Chaudhuri G. The pharmacological activity of nitroxyl: a potent vasodilator with activity similar to nitric oxide and/or endothelium-derived relaxing factor. J. Pharmacol. Exp. Ther. 1992;263:546–51.

13. Wong PS, Hyun J, Fukuto JM, Shirota FN, DeMaster EG, Shoeman DW, et al. Reaction between S-nitrosothiols and thiols: generation of nitroxyl (HNO) and subsequent chemistry. Biochemistry 1998; 37(16):5362–71.

14. Donzelli S, Espey MG, Flores-Santana W, Switzer CH, Yeh GC, Huang J, et al. Generation of nitroxyl by heme protein-mediated peroxidation of hydroxylamine but not N-hydroxy-L-arginine. Free Radic. Biol. Med. 2008; 45(5):578–84.

15. Shaul PW. Regulation of endothelial nitric oxide synthase: location, location, location. Annu. Rev. Physiol. 2002; 64:749–74.

16. Leiper JM. The DDAH-ADMA-NOS pathway. Ther Drug Monit. 2005;27(6):744–6.

17. Knipp M, Charnock JM, Garner CD, Vasák M. Structural and functional characterization of the Zn(II) site in dimethylargininase-1 (DDAH-1) from bovine brain. Zn(II) release activates DDAH-1. J. Biol. Chem. 2001; 276(44):40449–56.

88 18. Vallance P, Leiper J. Cardiovascular biology of the asymmetric dimethylarginine:dimethylarginine dimethylaminohydrolase pathway. Arterioscler. Thromb. Vasc. Biol. 2004; (6):1023–30.

19. Wang Y MA. Developing dual and specific inhibitors of dimethylarginine dimethylaminohydrolase-1 and nitric oxide synthase: Toward a targeted polypharmacology to control nitric oxide. Biochemistry. 2009; 48 :8624–35.

20. Hong L, Fast W. Inhibition of human dimethylarginine dimethylaminohydrolase-1 by S-nitroso-L-homocysteine and hydrogen peroxide. Analysis, quantification, and implications for hyperhomocysteinemia. J. Biol. Chem. 2007; 282(48):34684–92.

21. Knipp M, Vasak M. A colorimetric 96-well microtiter plate assay for the determination of enzymatically formed citrulline. Anal. Biochem. 2000;286(2):257– 64.

22. Wang Y, Hu S, Fast W. A click chemistry mediated in vivo activity probe for dimethylarginine dimethylaminohydrolase. J. Am. Chem. Soc. 2009;131(42):15096– 7.

23. Amatore C, Arbault S, Ducrocq C, Hu S, Tapsoba I. Angeli’s salt (Na2N2O3) is a precursor of HNO and NO: a voltammetric study of the reactive intermediates released by Angeli’s salt decomposition. ChemMedChem. 2007; (6):898–903.

24. Miranda KM, Dutton AS, Ridnour LA, Foreman CA, Ford E, Paolocci N, et al. Mechanism of aerobic decomposition of Angeli’s salt (sodium trioxodinitrate) at physiological pH. J. Am. Chem. Soc. 2005; 127(2):722–31.

25. Chin KY, Qin C, Cao N, Kemp-Harper BK, Woodman OL, Ritchie RH. The concomitant coronary vasodilator and positive inotropic actions of the nitroxyl donor Angeli’s salt in the intact rat heart: contribution of soluble guanylyl cyclase- dependent and -independent mechanisms. Br. J. Pharmacol. 2014; 1(7):1722–34.

26. Lopez BE, Wink DA, Fukuto JM. The inhibition of glyceraldehyde-3-phosphate dehydrogenase by nitroxyl (HNO). Arch. Biochem. Biophys. 2007; 465(2):430–6.

27. Switzer CH, Flores-Santana W, Mancardi D, Donzelli S, Basudhar D, Ridnour LA, et al. The emergence of nitroxyl (HNO) as a pharmacological agent. Biochim. Biophys. Acta. 2009 ;1787(7):835–40.

89

Appendix: Kinetics of Chloroacetamidines

y = m1*(1-exp(-m2*M0))+m3 A Value Error m1 0.53617 2.5742 120 m2 0.36211 1.5222 m3 3.0135 2.683 y = m1*(1-exp(-m2*M0))+m3 Chisq 0.68813 NA Value Error 100 R 0.62373 NA m1 2.3127 3.0979 m2 0.35776 0.46975 m3 2.4014 3.157 80 Chisq 1.6567 NA y = m1*(1-exp(-m2*M0))+m3 R 0.97824 NA Value Error m1 2.9474 0.73695 60 M Citrulline m2 0.19638 0.041795 y = m1*(1-exp(-m2*M0))+m3 µ m3 3.0928600 0.74557 Value Error Chisq 7.8539300 NA m1 8.0084 0.83486 40 200 R 0.93299100 NA m2 0.16089 0.017582 y = m1*(1-exp(-m2*M0))+m375 m3 2.2148 0.8443 37.5 Value ErrorChisq 0.66529 NA 20 m1 18.7510.867 1.4951 0 R 0.99903 NA m2 0.13531 0.028689 m3 -0.75359 1.5603 0 y = m1*(1-exp(-m2*M0))+m3Chisq 0.60485 NA 0 10 20 30 40 50 60 70 ValueR 0.99944Error NA Time (min) m1 22.645 0.79268 m2 0.075164 0.0055235 B m3 0.98093 0.90589 y = m1*(1-exp(-m2*M0))+m3Chisq 2.253 NA RValue0.99917 Error NA 2000 m1 437.72 183.07 m2y = m1*(1-exp(-m2*M0))+m30.0043828 0.0019868 m3 -2.2539Value 0.45178Error Chisqm1 7.609121.6 65.106NA Rm2 0.0104810.999 0.0064538NA 1500 m3 1.1051 0.64194 Chisq 3.5127 NA R 0.99598 NA 1/A/min

1000

500

0 100 200 300 400 500 600 700 800 Cl-NIO (µ!) 90

y = (5042*m2 + 42*M0)/(42*m1... Value Error m1 0.399 0.15307 m2 1.3267 0.6988 Chisq 0.63097 NA R 0.95661 NA Figure A.1 Inactivation of DDAH1 by Cl-NIO (Tsou method)

Inactivation of purified DDAH-1 by C-NIO. (A) Time- and concentration-dependent inactivation of purified DDAH-1 is observed when assayed in the presence of competing substrate and 0 (n), 18.75 (l), 37.5 (r), 75(‚), 100 (u), 200 (+), 300(¢) µM of Cl- NIO. Solid lines are fits as described in Materials and Methods. (B) The inverse of the apparent inactivation rates (A) are replotted as described in Chapter 2 to derive KI (1.3 ± -1 0.7 µM) and kinact (0.4 ± 0.2 min ) values for Cl-NIO inactivation of DDAH-1.

91 0 0.03 0.06 0.12 A 0.24 0.48 1 100 2 y = m1*(2.718^(m2*M0)) y = m1*(2.718^(m2*M0))y = m1*(2.718^(m2*M0)) Value Error ValueValue Errorm1Error 100 6.7179e-12 m1 102.22 2.3457m2 9.7267e-16 0 80 m1 95.631y = m1*(2.718^(m2*M0))1.7471 y = m1*(2.718^(m2*M0))m2 m2-0.014612 -0.0057953 0.00077231Value0.00077286Chisq 3.3847e-23Error NA ChisqChisqValue43.493m188.769 Error98.268NA RNA1.6142 1 NA m1 R94.124R 0.99287m20.95287 3.1168-0.030691NA 0.0010825NA 60 m2 -0.06902 Chisq0.005306531.465 NA Chisq 94.367 R NA0.99772 NA y = m1*(2.718^(m2*M0)) R 0.99488 NA Value Error 40 m1 94.918 4.0516 m2 -0.14893 0.019986

% Remaining Activity y = m1*(2.718^(m2*M0)) Chisq 125.72 ValueNA Error R 0.99343 NA 20 m1 99.859 3.1632 m2 -0.46392 0.043012 y = m1*(2.718^(m2*M0))Chisq 60.056 NA R 0.9966 NA 0 Value Error 0 10 20 30 40 50 60 70 m1 99.998 2.2513 Time (min) m2 -0.78454 0.055236 Chisq 30.41 NA R 0.99818 NA B

0.5

ClNIOQuadratic

B 0.4

0.3 y = (((m2+M0+m1)-((m2+M0+m1)... )

-1 Value Error

(min m1 0.51409 0.048824

obs m2 0.56097 0.011428 k 0.2 Chisq 93.616 NA R 0.98926 NA

0.1

0 0 1 2 3 4 5 Cl-NIO (µ!)

92

C 0.5 B

ClNIO [I]>[E] only 0.4

0.3 ) -1 (min obs k 0.2

y = (m2*M0)/(m1+M0) Value Error 0.1 m1 1.2013 0.074559 m2 0.38805 0.014934 Chisq 107.24 NA 0 R 0.98223 NA 0 1 2 3 4 5 Cl-NIO (µ!)

Figure A.2 Inactivation of DDAH1 by Cl-NIO (Kitz-Wilson and Morrison methods)

(A) Exponential fits to the observed inactivation by different concentrations of Cl-NIO: 0 (¡), 0.6 (o), 0.12 (u), 0.24 (‚), 0.48 (+), 1 (r), 2 (l), 4 (n) μM in the presence of

300 μM SMTC. (B) Concentration dependence of the kobs values for inactivation, fit to equation 2.1 (Morrison equation) (C) Concentration dependence of the kobs values for inactivation for inactivator values well above enzyme concentration, fit to equation 2.2 (Kitz-Wilson).

93 A

B

94 Figure A.3 Inactivation of DDAH1 by Cl-NIO (Kin-Tek Explorer)

(A) Global fitting of the time-dependent inactiviation of purified DDAH-1 by varying concentrations Cl-NIO : 0 (red), 30 (green), 60 (blue), 120 (yellow), 240 (cyan), 480 -1 -1 (pink), 1,000 (forrest green), 2,000 (purple) nM. k+1 was set to 18,000 μM min and k-2 was set to 0 and the fitting was allowed to proceed as described in Chapter 2. (B) The

FitSpace Confidence contour is shown for the values of k-1 and k+2. The linearity shows that the two parameters vary with respect to each other and that there is a defined lower boundary but the large upper boundary indicates that it is not well defined.

Table A.1 Rate constants obtained during global fitting to the model in scheme 2.1

*Upper and lower bounds were calculated with a χ2 threshold of 1.1 at the boundaries.

95 A y = m1*(1-exp(-m2*M0))+m3 Value Error y = m1*(1-exp(-m2*M0))+m3m1 9.281 0.69649 80 y = m1*(1-exp(-m2*M0))+m3m2 Value0.13131 Error0.018968 m1m3 Value13.22-0.3865 0.21333Error0.70732 70 y =m1 m1*(1-exp(-m2*M0))+m3m2Chisq 0.08317519.0630.363140.00684510.1529 NA m2m3 0.070635R Value-0.40260.999830.00243710.27414Error NA Chisq 1.466 NA 60 m1m3 -0.674126.964 0.662980.17238 Chisqy m2= m1*(1-exp(-m2*M0))+m3 R0.0456678.87170.99986 0.0038907NA NA m3R 0.154540.99971Value 0.60835ErrorNA 50 Chisqm1 35.2542.311 0.95296NA m2R 0.0326340.99944 0.0021859NA 40 m3 0.65207 0.2593 Chisq 3.7084 NA M Citrulline µ 30 R 0.99987 NA y = m1*(1-exp(-m2*M0))+m3 Value Error 20 m1 54.523 2.1144 m2 0.021953 0.00161 10 m3 0.12133 0.16644 Chisq 0.80769 NA 2000 0 R 0.999971500 NA 0 10 20 30 40 50 60 70 1000 750 Time (min) 500 375 0 B 1.2 104 B y = m1*(1-exp(-m2*M0))+m3 Value Error m1 252.47 87.878 1.1 104 m2 0.0051913 0.0020498 m3 -0.47385 0.36983 Chisq 10.782 NA R 0.99916 NA 1 104

y = (5042*m2 + 42*M0)/(42*m1...

1/A/min Value Error 9000 m1 0.51077 0.15914 m2 34.362 12.616 Chisq 1.6931e+06 NA R 0.84883 NA 8000

7000 0 200 400 600 800 1000 1200 1400 1600

MeClNIL (µ!)

Figure A.4 Inactivation of DDAH1 by Me-Cl-NIL (Tsou method) 96 (A) Time- and concentration-dependent inactivation of purified DDAH-1 is observed when assayed in the presence of competing substrate and 0 (l), 187.5 (r), 375 (+), 500(‚), 750 (u), 1000 (o), 1500(¢) µM of Me-Cl-NIL. Solid lines are fits as described in Materials and Methods of Chapter 2. (B) The inverse of the apparent inactivation rates (A) are replotted as described in Materials and Methods of Chapter 2 to -1 derive KI (30 ± 10 µM) and kinact (0.5 ± 0.2 min ) values for Me-Cl-NIL inactivation of DDAH-1

97 A

100

80

y = 100*(2.718^(m2*M0)) 60 Value Error m2 2.0727e-16 0 Chisq 3.9457e-24 NAy = 100*(2.718^(m2*M0)) y = 100*(2.718^(m2*M0)) R y 1= 100*(2.718^(m2*M0))NA Value Error 40 y = 100*(2.718^(m2*M0))

% Remaining Activity % Remaining Value Error Valuem2Value -0.0088658Error Error0.00044546 m2 -0.071123 0.0039724 m2 m2-0.16042 Chisq-0.036630.0174010.001913747.28 NA y = 100*(2.718^(m2*M0)) Chisq 97.455R Chisq0.97392 NA68.452 NA NA yChisq = 100*(2.718^(m2*M0))115.69 NA Value Error 20 R 0.9883 R 0.99293NA NA R 0.98274Value ErrorNA m2 -0.01929 0.0009005 m2 -0.45232 0.036315 Chisq 78.367 NA y = 100*(2.718^(m2*M0)) R 0.98421 NA Chisq 52.267 ValueNA Error 0 R 0.97953m2 -0.84429NA 0.026213 0 10 20 30 40 50 60 70 Chisq 5.6271 NA Time (min) R 0.99109 NA B

0.3

0.25

0.2 ) -1

(min 0.15 obs k

0.1

0.05 y = (((m2+M0+m1)-((m2+M0+m1)... Value Error m1 25.41 15.745 0 m2 0.66437 0.1696 0 5 10 15 20 25 30 35 Chisq 0.79568 NA R 0.99793 NA MeClNIL (µ!)

98 Figure A.4 Inactivation of DDAH1 by Me-Cl-NIL (Kitz-Wilson method)

(A) Exponential fits to the observed inactivation by different concentrations of Me-Cl- NIL: 0 (¡), 0.5 (o), 1 (u), 2 (‚), 4 (+), 8 (r), 16 (l), 32 (n) μM in the presence of

300 μM SMTC. (B) Concentration dependence of the kobs values for inactivation, fit to equation 2.2 (Kitz-Wilson)

99 A

B

100 Figure A.5 Inactivation of DDAH1 by Me-Cl-NIL (Kin-Tek Explorer)

(A) Global fitting of the time-dependent inactiviation of purified DDAH-1 by varying concentrations Me-Cl-NIL : 0 (red), 0.5 (green), 1 (blue), 2(yellow), 4 (cyan), 8 (pink), -1 -1 16 (forrest green), 32 (purple) μM. k+1 was set to 18,000 μM min and k-2 was set to 0 and the fitting was allowed to proceed as described in Chapter 2. (B) The FitSpace

Confidence contour is shown for the values of k-1 and k+2. The linearity shows that the two parameters vary with respect to each other and that there is a defined lower boundary but the large upper boundary indicates that it is not well defined.

Table A.2 Rate constants obtained during global fitting to the model in scheme 2.1

*Upper and lower bounds were calculated with a χ2 threshold of 1.1 at the boundaries

101 A 2000 1500 750 120 500 y = m1*(1-exp(-m2*M0))+m3 375 187.5 Value Error 0 m1 5.5842 1.9036 100 m2 0.21347 0.07223 m3 2.0137 1.9769 y = m1*(1-exp(-m2*M0))+m3 y = m1*(1-exp(-m2*M0))+m3Chisq 0.094261 NA Value Error 80 Value R Error0.99909 NA m1 11.628 0.45557 m1 19.275 0.80928 m2 0.14082 0.014935 m2 0.086537 0.0086794

Citrulline m3 0.61124 0.50952 60 m3 1.7855 0.94306 ! Chisq 1.308 NA µ Chisq 0.55568 NA R 0.99978 NA R 0.99976 NA 40 y = m1*(1-exp(-m2*M0))+m3 y = m1*(1-exp(-m2*M0))+m3 Value Error Value Error Data 619 m1 44.198 11.437 m1 29.188 0.88877 20 m2 0.043143 0.021273 y = m1*(1-exp(-m2*M0))+m3m2 0.057983y = m1*(1-exp(-m2*M0))+m30.0052906 m3 0.80138 1.5454 m3 Value1.8574 Value0.94299Error Error m1 77.374 2.6533 Chisq 0.25997 NA 0 m1 Chisq 289.551.6822 55.721NA m2 0.02098 0.0012395 R 0.99789 NA 0 10 20 30 40 50 60 70 m2 0.0073018R 0.999220.0016727NA m3 0.62467m3 2.28950.53877 0.20481 Time (min) Chisq 17.591Chisq 16.992NA NA R 0.999R 0.99934NA NA B MeClNIO Tsou Plot 1.2 104

1.1 104

1 104

9000 1/A/min

8000 y = (5042*m2 + 42*M0)/(42*m1... Value Error 7000 m1 0.92163 0.44155 m2 60.502 30.707 Chisq 2.8279 NA 6000 R 0.77886 NA 0 500 1000 1500 2000 2500 MeClNIO (µ!)

102 Figure A.6 Inactivation of DDAH1 by Me-Cl-NIO (Tsou method)

(A) Time- and concentration-dependent inactivation of purified DDAH-1 is observed when assayed in the presence of competing substrate and 0 (l), 187.5 (r), 375 (+), 500(‚), 750 (u), 1000 (o), 1500(¢) µM of Me-Cl-NIO. Solid lines are fits as described in Chapter 2. (B) The inverse of the apparent inactivation rates (A) are replotted -1 as described in Chapter 2 to derive KI (60 ± 30 µM) and kinact (0.9 ± 0.4 min ) values for Me-Cl-NIO inactivation of DDAH-1.

103 A

100

y = 100*(2.718^(m2*M0)) Value Error 80 y = 100*(2.718^(m2*M0)) y = 100*(2.718^(m2*M0))m2 -0.0099701 0.0010326 Value Error ValueChisq Error229.97 NA y = 100*(2.718^(m2*M0)) m2 2.0727e-16 0 m2 -0.025508 0.0012945R 0.94932 NA y = 100*(2.718^(m2*M0))Chisq 98.813Value ChisqErrorNA 3.9457e-24 NA 60 y = 100*(2.718^(m2*M0)) Valuem2R -0.0369340.98699Error 0.0012163NAR 1 NA Value Error m2 -0.07327Chisqy = 100*(2.718^(m2*M0))0.002411138.584 NA m2 -0.16266 0.0094579 Chisq 23.111R 0.99679ValueNA Error NA Chisq 32.813 NA 40 R 0.99798m2 -0.2421NA0.013206 R 0.99608 NA % Remaining Activity % Remaining Chisq 22.303 NA R 0.99613 NA

20 y = 100*(2.718^(m2*M0)) Value Error m2 -0.42034 0.020269 0 Chisq 18.631 NA 0 10 20 30 40 50 60 70 R 0.99374 NA

Time (min)

B

0.25 C

MeClNIO Kobs regular 0.2

0.15 ) -1 (min obs k 0.1 y = (m2*M0)/(m1+M0) Value Error m1 9.3151 2.3667 0.05 m2 0.2899 0.029881 Chisq 0.0012461 NA R 0.98623 NA 0 0 5 10 15 20 25 30 35 MeClNIO (µ!)

104 Figure A.7 Inactivation of DDAH1 by Me-Cl-NI) (Kitz-Wilson method)

(A) Exponential fits to the observed inactivation by different concentrations of Me-Cl- NIO: 0 (¡), 0.5 (o), 1 (u), 2 (‚), 4 (+), 8 (r), 16 (l), 32 (n) μM in the presence of

300 μM SMTC. (B) Concentration dependence of the kobs values for inactivation, fit to equation 2.2 (Kitz-Wilson).

105 A

B

106 Figure A.8 Inactivation of DDAH1 by Me-Cl-NIO (Kin-Tek Explorer)

(A) Global fitting of the time-dependent inactiviation of purified DDAH-1 by varying concentrations Me-Cl-NIO : 0 (red), 0.5 (green), 1 (blue), 2(yellow), 4 (cyan), 8 (pink), -1 -1 16 (forrest green), 32 (purple) μM. k+1 was set to 18,000 μM min and k-2 was set to 0 and the fitting was allowed to proceed as described in Chapter 2. (B) The FitSpace

Confidence contour is shown for the values of k-1 and k+2. The linearity shows that the two parameters vary with respect to each other and that there is a defined lower boundary but and upper boundary in this case.

Table A.3 Rate constants obtained during global fitting to the model in scheme 2.1

*Upper and lower bounds were calculated with a χ2 threshold of 1.1 at the boundaries

107 Bibliography

Adams PD, A. P. (2010). PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Cryst., D66, 213–221.

Altinova, A. E., Arslan, M., Sepici-Dincel, A., Akturk, M., Altan, N., & Toruner, F. B. (2007). Uncomplicated type 1 diabetes is associated with increased asymmetric dimethylarginine concentrations. The Journal of Clinical Endocrinology and Metabolism, 92(5), 1881–1885.

Altmann, K. S., Havemeyer, A., Beitz, E., & Clement, B. (2012). Dimethylarginine- dimethylaminohydrolase-2 (DDAH-2) does not metabolize methylarginines. Chembiochem 13(17), 2599–2604.

Amatore, C., Arbault, S., Ducrocq, C., Hu, S., & Tapsoba, I. (2007). Angeli’s salt (Na2N2O3) is a precursor of HNO and NO: a voltammetric study of the reactive intermediates released by Angeli’s salt decomposition. ChemMedChem, 2(6), 898–903.

Angus, D. C., Linde-Zwirble, W. T., Lidicker, J., Clermont, G., Carcillo, J., & Pinsky, M. R. (2001). Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Critical Care Medicine, 29(7), 1303–1310.

Augustyniak, A., Skolimowski, J., & Błaszczyk, A. (2013). Cytotoxicity of nitroxyl (HNO/NO−) against normal and cancer human cells. Chemico-Biological Interactions, 206(2), 262–271.

Bedford, M. T., & Clarke, S. G. (2009). Protein Arginine Methylation in Mammals: Who, What, and Why. Molecular Cell, 33(1), 1–13.

Bełtowski, J., & Kedra, A. (2006). Asymmetric dimethylarginine (ADMA) as a target for pharmacotherapy. Pharmacological Reports: PR, 58(2), 159–178.

Boehning, D., & Snyder, S. H. (2003). Novel neural modulators. Annual Review of Neuroscience, 26, 105–131.

Bogdan, C. (2001). Nitric oxide and the immune response. Nature Immunology, 2(10), 907–916.

108 Böger, R. H., Bode-Böger, S. M., Szuba, A., Tsao, P. S., Chan, J. R., Tangphao, O., … Cooke, J. P. (1998). Asymmetric dimethylarginine (ADMA): a novel risk factor for endothelial dysfunction: its role in hypercholesterolemia. Circulation, 98(18), 1842–1847.

Bretscher, L. E., Li, H., Poulos, T. L., & Griffith, O. W. (2003). Structural characterization and kinetics of nitric-oxide synthase inhibition by novel N5- (iminoalkyl)- and N5-(iminoalkenyl)-ornithines. The Journal of Biological Chemistry, 278(47), 46789–46797.

Brunger, A. (1993). Assessment of phase accuracy by cross validation: the free R value. Methods and applications. Acta Cryst, D49, 24–36.

Calabrese, V., Mancuso, C., Calvani, M., Rizzarelli, E., Butterfield, D. A., & Giuffrida Stella, A. M. (2007). Nitric oxide in the central nervous system: neuroprotection versus neurotoxicity. Nat Rev Neuroscience, 8(10), 766–775.

Cheng, Y., & Prusoff, W. H. (1973). Relationship between the inhibition constant (K1) and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction. Biochem Pharmacol, 22(23), 3099–108.

Chin, K. Y., Qin, C., Cao, N., Kemp-Harper, B. K., Woodman, O. L., & Ritchie, R. H. (2014). The concomitant coronary vasodilator and positive inotropic actions of the nitroxyl donor Angeli’s salt in the intact rat heart: contribution of soluble guanylyl cyclase-dependent and -independent mechanisms. British Journal of Pharmacology, 171(7), 1722–1734.

Copeland, R. A. (2005). Evaluation of enzyme inhibitors in drug discovery: a guide for medicinal chemists and pharmacologists. Hoboken, NJ: Wiley-Interscience.

Davis IW, L.-F. A. (2007). MolProbity: all-atom contacts and structure validation for proteins and nucleic acids. Nucleic Acids Res, 35, W375–83.

Davis, K. L., Martin, E., Turko, I. V., & Murad, F. (2001). Novel effects of nitric oxide. Annual Review of Pharmacology and Toxicology, 41, 203–236.

Donzelli, S., Espey, M. G., Flores-Santana, W., Switzer, C. H., Yeh, G. C., Huang, J., … Wink, D. A. (2008). Generation of nitroxyl by heme protein-mediated peroxidation of hydroxylamine but not N-hydroxy-L-arginine. Free Radical Biology & Medicine, 45(5), 578–584.

109 Doyle, M. P., Mahapatro, S. N., Broene, R. D., & Guy, J. K. (1988). Oxidation and reduction of hemoproteins by trioxodinitrate(II). The role of nitrosyl hydride and nitrite. Journal of the American Chemical Society, 110(2), 593–599.

Druhan, L. J., Forbes, S. P., Pope, A. J., Chen, C. A., Zweier, J. L., & Cardounel, A. J. (2008). Regulation of eNOS-derived superoxide by endogenous methylarginines. Biochemistry, 47(27), 7256–63.

Ekmekcioglu, S., Ellerhorst, J. A., Prieto, V. G., Johnson, M. M., Broemeling, L. D., & Grimm, E. A. (2006). Tumor iNOS predicts poor survival for stage III melanoma patients. Int J Cancer, 119(4), 861–6.

Ekmekcioglu, S., Ellerhorst, J., Smid, C. M., Prieto, V. G., Munsell, M., Buzaid, A. C., & Grimm, E. A. (2000). Inducible nitric oxide synthase and nitrotyrosine in human metastatic melanoma tumors correlate with poor survival. Clin Cancer Res, 6(12), 4768–75.

Emsley P, L. B. (2010). Features and development of Coot. Acta Cryst, D66, 486–501.

Engin, A. B. (2011). Dual function of nitric oxide in carcinogenesis, reappraisal. Current Drug Metabolism, 12(9), 891–899.

Flores-Santana, W., Salmon, D. J., Donzelli, S., Switzer, C. H., Basudhar, D., Ridnour, L.Wink, D. A. (2011). The specificity of nitroxyl chemistry is unique among nitrogen oxides in biological systems. Antioxidants & Redox Signaling, 14(9), 1659–1674.

Flores-Santana, W., Switzer, C., Ridnour, L. A., Basudhar, D., Mancardi, D., Donzelli, S. Wink, D. A. (2009). Comparing the chemical biology of NO and HNO. Archives of Pharmacal Research, 32(8), 1139–1153.

Forbes, S. P., Druhan, L. J., Guzman, J. E., Parinandi, N., Zhang, L., Green-Church, K. B., & Cardounel, A. J. (2008). Mechanism of 4-HNE mediated inhibition of hDDAH-1: implications in NO regulation. Biochemistry, 47(6), 1819–1826.

Förstermann, U., & Sessa, W. C. (2012). Nitric oxide synthases: regulation and function. European Heart Journal, 33(7), 829–837, 837a–837d.

Fortin, C. F., McDonald, P. P., Fülöp, T., & Lesur, O. (2010). Sepsis, leukocytes, and nitric oxide (NO): an intricate affair. Shock (Augusta, Ga.), 33(4), 344–352.

Foster, M. W., McMahon, T. J., & Stamler, J. S. (2003). S-nitrosylation in health and disease. Trends in Molecular Medicine, 9(4), 160–168. 110

Fukumura, D., Kashiwagi, S., & Jain, R. K. (2006). The role of nitric oxide in tumour progression. Nat Rev Cancer, 6(7), 521–34.

Fukuto, J. M., Chiang, K., Hszieh, R., Wong, P., & Chaudhuri, G. (1992). The pharmacological activity of nitroxyl: a potent vasodilator with activity similar to nitric oxide and/or endothelium-derived relaxing factor. J Pharmacol Exp Ther, 263, 546–551.

Fukuto, J. M., Jackson, M. I., Kaludercic, N., & Paolocci, N. (2008). Examining nitroxyl in biological systems. Methods in Enzymology, 440, 411–431.

Ghebremariam, Y. T., Erlanson, D. A., Yamada, K., & Cooke, J. P. (2012). Development of a dimethylarginine dimethylaminohydrolase (DDAH) assay for high- throughput chemical screening. Journal of Biomolecular Screening, 17(5), 651–

Hong, L., & Fast, W. (2007). Inhibition of human dimethylarginine dimethylaminohydrolase-1 by S-nitroso-L-homocysteine and hydrogen peroxide. Analysis, quantification, and implications for hyperhomocysteinemia. The Journal of Biological Chemistry, 282(48), 34684–34692.

Huang, Z., Huang, P. L., Panahian, N., Dalkara, T., Fishman, M. C., & Moskowitz, M. A. (1994). Effects of cerebral ischemia in mice deficient in neuronal nitric oxide synthase. Science (New York, N.Y.), 265(5180), 1883–1885.

Irvine, J. C., Favaloro, J. L., & Kemp-Harper, B. K. (2003). NO- activates soluble guanylate cyclase and Kv channels to vasodilate resistance arteries. Hypertension, 41(6), 1301–1307.

Iyer, A. K. V., Azad, N., Wang, L., & Rojanasakul, Y. (2008). Role of S-nitrosylation in apoptosis resistance and carcinogenesis. Nitric Oxide: Biology and Chemistry / Official Journal of the Nitric Oxide Society, 19(2), 146–151.

Izumi, Y., Benz, A. M., Clifford, D. B., & Zorumski, C. F. (1992). Nitric oxide inhibitors attenuate N-methyl-D-aspartate excitotoxicity in rat hippocampal slices. Neuroscience Letters, 135(2), 227–230.

Jacobi, J., Sydow, K., von Degenfeld, G., Zhang, Y., Dayoub, H., Wang, B., … Cooke, J. P. (2005). Overexpression of dimethylarginine dimethylaminohydrolase reduces tissue asymmetric dimethylarginine levels and enhances angiogenesis. Circulation, 111(11), 1431–1438.

111 Johnson, C. M., Linsky, T. W., Yoon, D.-W., Person, M. D., & Fast, W. (2011). Discovery of halopyridines as quiescent affinity labels: inactivation of dimethylarginine dimethylaminohydrolase. Journal of the American Chemical Society, 133(5), 1553–1562.

Johnson, C. M., Monzingo, A. F., Ke, Z., Yoon, D.-W., Linsky, T. W., Guo, H., … Fast, W. (2011). On the mechanism of dimethylarginine dimethylaminohydrolase inactivation by 4-halopyridines. Journal of the American Chemical Society, 133(28), 10951–10959.

Johnson, K. A. (2009). Fitting enzyme kinetic data with KinTek Global Kinetic Explorer. Methods in Enzymology, 467, 601–626.

Johnson, K. A. (2013). A century of enzyme kinetic analysis, 1913 to 2013. FEBS Letters, 587(17), 2753–2766.

Kalgutkar, A. S., & Dalvie, D. K. (2012). Drug discovery for a new generation of covalent drugs. Expert Opinion on Drug Discovery, 7(7), 561–581.

Knipp, M. (2006). How to control NO production in cells: N(omega),N(omega)- dimethyl-L-arginine dimethylaminohydrolase as a novel drug target. Chembiochem 7(6), 879–889.

Knipp, M., Braun, O., Gehrig, P. M., Sack, R., & Vasák, M. (2003). Zn(II)-free dimethylargininase-1 (DDAH-1) is inhibited upon specific Cys-S-nitrosylation. The Journal of Biological Chemistry, 278(5), 3410–3416.

Knipp, M., Charnock, J. M., Garner, C. D., & Vasák, M. (2001). Structural and functional characterization of the Zn(II) site in dimethylargininase-1 (DDAH-1) from bovine brain. Zn(II) release activates DDAH-1. The Journal of Biological Chemistry, 276(44), 40449–40456.

Knipp, M., Charnock, J. M., Garner, C. D., & Vašák, M. (2001). Structural and Functional Characterization of the Zn(II) Site in Dimethylargininase-1 (DDAH-1) from Bovine Brain Zn(II) RELEASE ACTIVATES DDAH-1. Journal of Biological Chemistry, 276(44), 40449–40456.

Knipp, M., & Vasak, M. (2000). A colorimetric 96-well microtiter plate assay for the determination of enzymatically formed citrulline. Anal Biochem, 286(2), 257–64.

Knuckley, B., Causey, C. P., Pellechia, P. J., Cook, P. F., & Thompson, P. R. (2010). Haloacetamidine-based inactivators of protein arginine deiminase 4 (PAD4):

112 evidence that general acid catalysis promotes efficient inactivation. Chembiochem 11(2), 161–165.

Kostourou, V., Robinson, S. P., Cartwright, J. E., & Whitley, G. S. J. (2002). Dimethylarginine dimethylaminohydrolase I enhances tumour growth and angiogenesis. British Journal of Cancer, 87(6), 673–680.

Kostourou, V., Robinson, S. P., Whitley, G. S. J., & Griffiths, J. R. (2003). Effects of overexpression of dimethylarginine dimethylaminohydrolase on tumor angiogenesis assessed by susceptibility magnetic resonance imaging. Cancer Research, 63(16), 4960–4966.

Kostourou, V., Troy, H., Murray, J. F., Cullis, E. R., Whitley, G. S. J., Griffiths, J. R., & Robinson, S. P. (2004). Overexpression of dimethylarginine dimethylaminohydrolase enhances tumor hypoxia: an insight into the relationship of hypoxia and angiogenesis in vivo. Neoplasia (New York, N.Y.), 6(4), 401–411.

Kotthaus, J., Schade, D., Kotthaus, J., & Clement, B. (n.d.). Designing modulators of dimethylarginine dimethylaminohydrolase (DDAH): a focus on selectivity over arginase. J Enzyme Inhib Med Chem, 27(1), 24–8.

Kubes, P., & McCafferty, D. M. (2000). Nitric oxide and intestinal inflammation. The American Journal of Medicine, 109(2), 150–158.

Lee, Y., Marletta, M. A., Martasek, P., Roman, L. J., Masters, B. S., & Silverman, R. B. (1999). Conformationally-restricted arginine analogues as alternative substrates and inhibitors of nitric oxide synthases. Bioorganic & Medicinal Chemistry, 7(6), 1097–1104.

Leiper, J. M. (2005). The DDAH-ADMA-NOS pathway. Ther Drug Monit, 27(6), 744–6.

Leiper, J. M., Santa Maria, J., Chubb, A., MacAllister, R. J., Charles, I. G., Whitley, G. S., & Vallance, P. (1999). Identification of two human dimethylarginine dimethylaminohydrolases with distinct tissue distributions and homology with microbial arginine deiminases. The Biochemical Journal, 343 Pt 1, 209–214.

Leiper, J., Murray-Rust, J., McDonald, N., & Vallance, P. (2002). S-nitrosylation of dimethylarginine dimethylaminohydrolase regulates enzyme activity: further interactions between nitric oxide synthase and dimethylarginine dimethylaminohydrolase. Proceedings of the National Academy of Sciences of the United States of America, 99(21), 13527–13532.

113 Leiper, J., & Nandi, M. (2011). The therapeutic potential of targeting endogenous inhibitors of nitric oxide synthesis. Nature Reviews. Drug Discovery, 10(4), 277– 291.

Leiper, J., Nandi, M., Torondel, B., Murray-Rust, J., Malaki, M., O’Hara, B., … Vallance, P. (2007). Disruption of methylarginine metabolism impairs vascular homeostasis. Nature Medicine, 13(2), 198–203.

Leung, C., Leung, S., Tirado-Rives, J., & Jorgensen, W. (2012). Methyl Effects on Protein-Ligand Binding. Journal of Medicinal Chemistry, 55, 4489–4500.

Linsky, T., & Fast, W. (2010). Mechanistic similarity and diversity among the guanidine- modifying members of the pentein superfamily. Biochimica et Biophysica Acta, 1804(10), 1943–1953.

Linsky, T., & Fast, W. (2011). A continuous, fluorescent, high-throughput assay for human dimethylarginine dimethylaminohydrolase-1. Journal of Biomolecular Screening, 16(9), 1089–1097.

Linsky, T. W., Monzingo, A. F., Stone, E. M., Robertus, J. D., & Fast, W. (2008). Promiscuous partitioning of a covalent intermediate common in the pentein superfamily. Chemistry & Biology, 15(5), 467–475.

Linsky, T., Wang, Y., & Fast, W. (2011a). Screening for dimethylarginine dimethylaminohydrolase inhibitors reveals ebselen as a bioavailable inactivator. ACS Medicinal Chemistry Letters, 2(8), 592–596.

Linsky, T., Wang, Y., & Fast, W. (2011b). Screening for dimethylarginine dimethylaminohydrolase inhibitors reveals ebselen as a bioavailable inactivator. ACS Medicinal Chemistry Letters, 2(8), 592–596.

Lluis, M., Wang, Y., Monzingo, A. F., Fast, W., & Robertus, J. D. (n.d.). Characterization of C-alkyl amidines as bioavailable covalent reversible inhibitors of human DDAH-1. ChemMedChem, 6(1), 81–8.

Lopez, B. E., Wink, D. A., & Fukuto, J. M. (2007). The inhibition of glyceraldehyde-3- phosphate dehydrogenase by nitroxyl (HNO). Archives of Biochemistry and Biophysics, 465(2), 430–436.

Luo, Z. D., & Cizkova, D. (2000). The role of nitric oxide in nociception. Current Review of Pain, 4(6), 459–466.

114 MacAllister, R. J., Parry, H., Kimoto, M., Ogawa, T., Russell, R. J., Hodson, H., … Vallance, P. (1996). Regulation of nitric oxide synthesis by dimethylarginine dimethylaminohydrolase. British Journal of Pharmacology, 119(8), 1533–1540.

Martínez-Ruiz, A., Cadenas, S., & Lamas, S. (2011). Nitric oxide signaling: Classical, less classical, and nonclassical mechanisms. Free Radical Biology and Medicine, 51(1), 17–29.

Miranda, K. M., Dutton, A. S., Ridnour, L. A., Foreman, C. A., Ford, E., Paolocci, N., … Wink, D. A. (2005). Mechanism of aerobic decomposition of Angeli’s salt (sodium trioxodinitrate) at physiological pH. Journal of the American Chemical Society, 127(2), 722–731.

Mocellin, S., Bronte, V., & Nitti, D. (2007). Nitric oxide, a double edged sword in cancer biology: Searching for therapeutic opportunities. Medicinal Research Reviews, 27(3), 317–352.

Morrison, J. F. (1969). Kinetics of the reversible inhibition of enzyme-catalysed reactions by tight-binding inhibitors. Biochimica et Biophysica Acta, 185(2), 269–286.

Nandi, M., Kelly, P., Torondel, B., Wang, Z., Starr, A., Ma, Y., … Leiper, J. (2012). Genetic and pharmacological inhibition of dimethylarginine dimethylaminohydrolase 1 is protective in endotoxic shock. Arteriosclerosis, Thrombosis, and Vascular Biology, 32(11), 2589–2597.

Olken, N. M., Osawa, Y., & Marletta, M. A. (1994). Characterization of the inactivation of nitric oxide synthase by NG-methyl-L-arginine: evidence for heme loss. Biochemistry, 33(49), 14784–14791.

Otwinowski, Z., & Minor, W. (1997). [20] Processing of X-ray diffraction data collected in oscillation mode. In J. Charles W. Carter (Ed.), Methods in Enzymology (Vol. Volume 276, pp. 307–326). Academic Press.

Palm, F., Onozato, M. L., Luo, Z., & Wilcox, C. S. (2007). Dimethylarginine dimethylaminohydrolase (DDAH): expression, regulation, and function in the cardiovascular and renal systems. American Journal of Physiology. Heart and Circulatory Physiology, 293(6), H3227–3245.

Pannirselvam, M., Anderson, T. J., & Triggle, C. R. (2003). Endothelial cell dysfunction in type I and II diabetes: The cellular basis for dysfunction. Drug Development Research, 58(1), 28–41.

115 Paolocci, N., Jackson, M. I., Lopez, B. E., Miranda, K., Tocchetti, C. G., Wink, D. A., … Fukuto, J. M. (2007). The pharmacology of nitroxyl (HNO) and its therapeutic potential: not just the Janus face of NO. Pharmacology & Therapeutics, 113(2), 442–458.

Paolocci, N., & Wink, D. A. (2009). The shy Angeli and his elusive creature: the HNO route to vasodilation. American Journal of Physiology. Heart and Circulatory Physiology, 296(5), H1217–1220.

Pope, A. J., Karrupiah, K., Kearns, P. N., Xia, Y., & Cardounel, A. J. (2009). Role of dimethylarginine dimethylaminohydrolases in the regulation of endothelial nitric oxide production. The Journal of Biological Chemistry, 284(51), 35338–35347.

Pope, A. J., Karuppiah, K., & Cardounel, A. J. (2009). Role of the PRMT–DDAH– ADMA axis in the regulation of endothelial nitric oxide production. Pharmacological Research, 60(6), 461–465.

Read, R. (1986). Improved Fourier coefficients for maps using phases from partial structures with errors. Acta Cryst, Sect A 42, 140–149.

Rodionov, R. N., Murry, D. J., Vaulman, S. F., Stevens, J. W., & Lentz, S. R. (2010). Human alanine-glyoxylate aminotransferase 2 lowers asymmetric dimethylarginine and protects from inhibition of nitric oxide production. The Journal of Biological Chemistry, 285(8), 5385–5391.

Russel, J. A. (2008). The current management of septic shock. Minerva Medica, 99(5), 431–458.

Schwartz, P. A., Kuzmic, P., Solowiej, J., Bergqvist, S., Bolanos, B., Almaden, C., … Murray, B. W. (2014). Covalent EGFR inhibitor analysis reveals importance of reversible interactions to potency and mechanisms of drug resistance. Proceedings of the National Academy of Sciences of the United States of America, 111(1), 173–178.

Selley, M. L. (2004). Increased (E)-4-hydroxy-2-nonenal and asymmetric dimethylarginine concentrations and decreased nitric oxide concentrations in the plasma of patients with major depression. Journal of Affective Disorders, 80(2-3), 249–256.

Serrano, C., Valero, A., & Picado, C. (2004). Nasal Nitric Oxide. Archivos de Bronconeumología ((English Edition)), 40(5), 222–230.

116 Shaul, P. W. (2002). Regulation of endothelial nitric oxide synthase: location, location, location. Annual Review of Physiology, 64, 749–774.

Sikora, A. G., Gelbard, A., Davies, M. A., Sano, D., Ekmekcioglu, S., Kwon, J., … Overwijk, W. W. (2010). Targeted inhibition of inducible nitric oxide synthase inhibits growth of human melanoma in vivo and synergizes with chemotherapy. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research, 16(6), 1834–1844

Silkoff, P. E., Robbins, R. A., Gaston, B., Lundberg, J. O., & Townley, R. G. (2000). Endogenous nitric oxide in allergic airway disease. The Journal of Allergy and Clinical Immunology, 105(3), 438–448.

Siroen, M. P. C., van der Sijp, J. R. M., Teerlink, T., van Schaik, C., Nijveldt, R. J., & van Leeuwen, P. A. M. (2005). The human liver clears both asymmetric and symmetric dimethylarginine. Hepatology (Baltimore, Md.), 41(3), 559–565.

Stone, E. M., Costello, A., Tierney, D., & Fast, W. (2006). Substrate-assisted cysteine deprotonation in the mechanism of dimethylarginase (DDAH) from Pseudomonas aeruginosa. Biochemistry, 17, 5618–30.

Stone, E. M., Schaller, T. H., Bianchi, H., Person, M. D., & Fast, W. (2005). Inactivation of two diverse enzymes in the amidinotransferase superfamily by 2- chloroacetamidine: dimethylargininase and peptidylarginine deiminase. Biochemistry, 44(42), 13744–52.

Stuehr, D. ., & Griffith, O. . (1993). Mammalian Nitric Oxide Synthases. Adv Enzymol Relat Areas Mol Biol, (65), 287–346.

Stühlinger, M. C., Oka, R. K., Graf, E. E., Schmölzer, I., Upson, B. M., Kapoor, O., … Cooke, J. P. (2003). Endothelial dysfunction induced by hyperhomocyst(e)inemia: role of asymmetric dimethylarginine. Circulation, 108(8), 933–938.

Susswein, A. J., Katzoff, A., Miller, N., & Hurwitz, I. (2004). Nitric oxide and memory. The Neuroscientist: A Review Journal Bringing Neurobiology, Neurology and Psychiatry, 10(2), 153–162.

Switzer, C. H., Flores-Santana, W., Mancardi, D., Donzelli, S., Basudhar, D., Ridnour, L. A., … Wink, D. A. (2009). The emergence of nitroxyl (HNO) as a pharmacological agent. Biochimica et Biophysica Acta, 1787(7), 835–840.

Switzer, C. H., Glynn, S. A., Ridnour, L. A., Cheng, R. Y.-S., Vitek, M. P., Ambs, S., & Wink, D. A. (2011). Nitric oxide and protein phosphatase 2A provide novel 117 therapeutic opportunities in ER-negative breast cancer. Trends in Pharmacological Sciences, 32(11), 644–651.

Tsikas, D., Böger, R. H., Sandmann, J., Bode-Böger, S. M., & Frölich, J. C. (2000). Endogenous nitric oxide synthase inhibitors are responsible for the L-arginine paradox. FEBS Letters, 478(1-2), 1–3.

Tsou, C. L. (1988). Kinetics of substrate reaction during irreversible modification of enzyme activity. Adv Enzymol Relat Areas Mol Biol, 61, 381–436.

Usui, M., Matsuoka, H., Miyazaki, H., Ueda, S., Okuda, S., & Imaizumi, T. (1998). Increased endogenous nitric oxide synthase inhibitor in patients with congestive heart failure. Life Sciences, 62(26), 2425–2430.

Vittecek, J., Lojek, A., Valacchi, G., Kubala, L. (2012). Arginine-Based Inhibitors of Nitric Oxide Synthase: Therapeutic Potential and Challenges. Mediators of Inflammation, 2012, e318087.

Vagin A, T. A. (1997). MOLREP: an automated program for molecular replacement. J. Appl. Cryst., (30), 1022–1025.

Vallance, P., & Leiper, J. (2002). Blocking NO synthesis: how, where and why? Nature Reviews. Drug Discovery, 1(12), 939–950.

Vallance, P., & Leiper, J. (2004). Cardiovascular biology of the asymmetric dimethylarginine:dimethylarginine dimethylaminohydrolase pathway. Arteriosclerosis, Thrombosis, and Vascular Biology, 24(6), 1023–1030.

Vallance, P., Leone, A., Calver, A., Collier, J., & Moncada, S. (1992). Accumulation of an endogenous inhibitor of nitric oxide synthesis in chronic renal failure. Lancet, 339(8793), 572–575.

Vasta, V., Meacci, E., Farnararo, M., & Bruni, P. (1995). Identification of a specific transport system for L-arginine in human platelets. Biochemical and Biophysical Research Communications, 206(3), 878–884.

Wang, J.-Z., Zhu, W.-D., Xu, Z.-X., Du, W.-T., Zhang, H.-Y., Sun, X.-W., & Wang, X.- H. (2014). Pin1, endothelial nitric oxide synthase, and amyloid-β form a feedback signaling loop involved in the pathogenesis of Alzheimer’s disease, hypertension, and cerebral amyloid angiopathy. Medical Hypotheses, 82(2), 145–150.

118 Wang Y, M. A. (2009). Developing dual and specific inhibitors of dimethylarginine dimethylaminohydrolase-1 and nitric oxide synthase: Toward a targeted polypharmacology to control nitric oxide. Biochemistry, 48, 8624–8635.

Wang, Y., Hu, S., & Fast, W. (2009). A click chemistry mediated in vivo activity probe for dimethylarginine dimethylaminohydrolase. J Am Chem Soc, 131(42), 15096– 7.

Wang, Y., Hu, S., Gabisi, A. M., Jr, Er, J. A. V., Pope, A., Burstein, G., … Fast, W. (2014). Developing an Irreversible Inhibitor of Human DDAH-1, an Enzyme Upregulated in Melanoma. ChemMedChem.

Weiss, S. L., Haymond, S., Ralay Ranaivo, H., Wang, D., De Jesus, V. R., Chace, D. H., & Wainwright, M. S. (2012). Evaluation of asymmetric dimethylarginine, arginine, and carnitine metabolism in pediatric sepsis. Pediatric Critical Care Medicine: A Journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies, 13(4), e210–218.

Wink, D. A., Miranda, K. M., Katori, T., Mancardi, D., Thomas, D. D., Ridnour, L., … Paolocci, N. (2003). Orthogonal properties of the redox siblings nitroxyl and nitric oxide in the cardiovascular system: a novel redox paradigm. American Journal of Physiology. Heart and Circulatory Physiology, 285(6), H2264–2276.

Wong, P. S., Hyun, J., Fukuto, J. M., Shirota, F. N., DeMaster, E. G., Shoeman, D. W., & Nagasawa, H. T. (1998). Reaction between S-nitrosothiols and thiols: generation of nitroxyl (HNO) and subsequent chemistry. Biochemistry, 37(16), 5362–5371.

Yoo, J. H., & Lee, S. C. (2001). Elevated levels of plasma homocyst(e)ine and asymmetric dimethylarginine in elderly patients with stroke. Atherosclerosis, 158(2), 425–430.

Zakrzewicz, D., & Eickelberg, O. (2009). From arginine methylation to ADMA: A novel mechanism with therapeutic potential in chronic lung diseases. BMC Pulmonary Medicine, 9(1), 5.

119 Vita

Gayle Burstein was born in San Diego, CA to Sam and Cheryl, two scientists who passed their love of research onto their daughter. She attended Florida Atlantic University (FAU) and received a B.S. in 2007 in Chemistry. At FAU, she studied matrix metalloproteinases and solid-phase peptide synthesis with Dr. Gregg Fields. After college, Gayle took a year to improve her French by teaching at Lyceé General Louis Pasteur in Neuilly-Sur-Seine near Paris, France. While France had its allures, Gayle was nonetheless drawn back to science and joined the Department of Biochemistry at the University of Texas at Austin in 2008. She eventually joined the lab of Dr. Walter Fast in 2009 to pursue her Ph.D.

Permanent email address: gburstei at gmail dot com This dissertation was typed by Gayle Diane Burstein.

120