Acknowledgement

I would like to thank my advisor Prof. Zachary T. Ball for teaching me everything I know in synthetic skills during my first year and guiding me through the rest of my time here at Rice. He has helped me better myself through constant constructive criticism, both in research and writing. I would also like to thank past and present Ball group members, Dr. Brian Popp, Dr. Alex Zaykov, Dr. Jessica Herron, Vincenzo Russo, Ramya Sambasivan, Cara Bovet, Zhen Chen, Dr. Jane Coughlin, Farrukh Vohidov, Matt Minus, Rob Ferguson, Julian Cooper. They have been a very good support in discussing research, exchanging ideas, and have become great pals in the last few years. Especially, Brian and Ramya’s encouragement in tough times is really valued. I thank my friends Rajkishore Barik and Meenu Adhikari for making Houston a second home. Again, Ramya Sambasivan for being a very good friend and always a willing ear to vent out frustrations associated with the graduate life and Avani Verma for bringing all kinds of fun in life and being so encouraging during my panic moments in thesis writing process. I would like to take this opportunity to express my gratitude and respect for my father Kuru Ram, and mother Amita Kundu, for the confidence and faith they always had in me, my sister Aruna Hajra who always supported me in my endeavors. I would like to thank my fiancée Soumya Sarkar, for his never wavering trust, love and support throughout my Ph.D in our 6 yr long, long- distance relation. Last but not least thanks to everyone who cared!

-Rituparna Kundu Abstract

Developing Dirhodium-Complexes for Protein Inhibition and Modification & Copper-Catalyzed Remote Chlorination of Alkyl-Hydroperoxides

by

Rituparna Kundu

The work describes the development of a new class of protein-inhibitors for protein-protein interactions, based on metallopeptides comprised of a dirhodium metal center. The metal incorporation in the peptide sequence leads to high increase in binding affinity of the inhibitors. The source of this strong affinity is the interaction of histidine on the protein surface with the rhodium center. In addition to this work, rhodium-based small molecule inhibitors for FK-506 binding proteins are investigated. Also, methodology for rhodium-catalyzed modification of proteins containing surface cysteine has been developed where a simple rhodium(II) complex catalyzes cysteine modification with diazo reagents. The reaction is marked by clean cysteine selectivity and mild reaction conditions. The resulting linkage is significantly more stable in human plasma serum, when compared to common maleimide reagents. Apart from this body of work in chemical-biology, the thesis contains the discussion of development of copper-catalyzed remote chlorination of alkyl hydroperoxides. The atom transfer chlorination utilizes simple ammonium chloride salts as the chlorine source and the internal redox process requires no external redox reagents. Table of Content

Chapter 1 Rational Design of Metal-Based Inhibitors for Proteins ...... 1 1.1. Introduction ...... 1 1.2. Rational design of metal-based protein-inhibitors ...... 2 1.2.1. Improving target specificity by fine tuning ligand exchange reactions at metal center...... 2 1.2.2. Improving potency and drug residence time by fine tuning metal and chelator combination .. 4 1.2.3. Kinetically inert metal complexes for highly selective inhibition of protein kinases ...... 7 1.2.4. Affinity enhancement of an inhibitor by combining two weak interacting fragments ...... 12 1.2.5. Inducing selectivity by incorporating a biological recognition sequence in a metal inhibitor . 15 1.2.6. Altering target and action of a drug by incorporating a metal-fragment ...... 16 1.2.7. Other examples ...... 17 1.3. Conclusions ...... 18 1.4. References ...... 18 Chapter 2 Metal catalyzed directed sp3 C–H bond activation ...... 22 2.1. Introduction ...... 22 2.2. Functionalizing sp3 C–H bonds using proximal-directing groups ...... 23 2.2.1. Rh-catalyzed amination of sp3 C–H bonds ...... 24 2.2.2. Pt-mediated directed sp3 C–H activation ...... 26 2.2.3. Pd-catalyzed directed sp3 C–H activation ...... 28 2.3. Conclusions ...... 31 2.4. References ...... 31 Chapter 3 Copper–Catalyzed Remote sp3 C–H Chlorination of Alkyl Hydroperoxide ...... 33 3.1. Introduction ...... 33 3.2. Development of copper-catalyzed C-H chlorination ...... 35 3.2.1. Substrate synthesis ...... 35 3.2.2. Preliminary assumptions ...... 37 3.2.3. Substrate-dependent optimization of copper-catalyzed C-H chlorination...... 39 3.2.4. Substrate-scope of copper-catalyzed C-H chlorination ...... 44 3.3. Conclusions ...... 47 3.4. Experimental ...... 48 3.4.1. General Considerations ...... 48

3.4.2. Synthesis of Alkyl-hydroperoxides ...... 50 3.4.3. Copper-Catalyzed C-H chlorination of Alkyl hydroperoxides:...... 64 3.5. References ...... 76 Chapter 4 Hybrid Organic-Inorganic inhibitors for PDZ Domain ...... 79 4.1. Introduction ...... 79 4.2. Dirhodium‒histidine interaction ...... 81 4.3. Design of metallopeptide based inhibitors ...... 85 4.4. Screening the potency of metalated inhibitors for CALP ...... 91 4.5. Developing dirhodium metallopeptide inhibitors for N1P1 ...... 95 4.6. Conclusions ...... 98 4.7. Experimental ...... 99 4.7.1. General Considerations...... 99 4.7.2. Synthetic Procedures...... 102 4.7.3. Characterization data ...... 103 4.8. References ...... 116 Chapter 5 Designing Dirhodium-Ligated Small Molecule Inhibitors for FK506-Binding Proteins ...... 120 5.1. Introduction ...... 120 5.2. Designing small-molecule metalated inhibitors ...... 121 5.2.1. Synthesis of compound B ...... 122 5.3. Future direction ...... 123 5.4. Experimental ...... 124 5.4.1. General Considerations: ...... 124 5.4.2. Synthesis of inhibitors and precursors ...... 125 5.5. References: ...... 134 Chapter 6 Rhodium Catalyzed Cysteine Modification with Diazo Reagents ...... 135 6.1. Introduction ...... 135 6.2. Dirhodium catalyzed CALP modification ...... 136 6.2.1. Screening modification on CALP ...... 137 6.2.2. Cysteine as the site of modification ...... 139 6.3. Dose-dependent modification of CALP ...... 141 6.4. Effect of neighboring residues on cysteine reactivity ...... 143 6.5. Stability of S-C linkage in biological medium ...... 144

6.6. Cysteine selectivity in a mixture of proteins ...... 146 6.7. Conclusions ...... 148 6.8. Experimental ...... 148 6.8.1. General Considerations...... 148 6.8.2. Modification of CALP (10 M): ...... 150 6.8.3. General Procedure for Trypsin Digestion of Modified CALP and MS/MS Analysis of Modified CALP Peptides: ...... 150 6.8.4. General Procedure for Blocking Cysteine residue on CALP:31...... 151 6.8.5. Control Reactions for Modification of CALP: ...... 151 6.8.6. General Protocol for the Modification of Peptides (25  : ...... 152 6.8.7. General protocol for the Modification of Peptide AQPADRCGGLA using Biotin-Maleimide: 153 6.8.8. General Protocol for Stability Study of S-C Linkages in Human Plasma Serum: ...... 153

6.8.9. Rh2(OAc)4 Catalyzed Modification of CALP in a Mixture of Other Proteins: ...... 154 6.8.10. Biotin-Maleimide Catalyzed Modification of CALP in a Mixture of Other Proteins: ...... 154 6.8.11. Western Blot Analysis: ...... 155 6.8.12. Characterization for AQPADRXCGGLA (X= R, K, A) Peptides:...... 156 6.9. References ...... 158 Appendix A: Selected Spectra for Chapter 3 ...... 161 Appendix B. Selected Spectra for Chapter 5 ...... 220

List of Figures

Figure 1.1 Structures of different ligands for CXCR4 inhibition ...... 6 Figure 1.2 Structure of octahedral metal complexes bound to protein kinase ...... 8 Figure 1.3 Kinetically inert octahedral metal complexes for protein kinases...... 10 Figure 1.4 Chemical formula of [Ir(ppy)2(biq)]PF6, ∆ enantiomer, ref.27 ...... 11 Figure 1.5 Chemical structures of sulfanilamide-Cu-GGH complex ...... 13 Figure 1.6 Ethacrynic acid derivatives as inhibitors for GST P1-1 ...... 14 Figure 1.7 Chemical structures of a) Co(III)-schiif base complex, b) the duplex DNA Ebox, c) Co(III)-Ebox conjugate, ref.41 ...... 15 Figure 1.8 Molecular structures of aspirin and hexacarbonyldicobalt derivative of aspirin ...... 17 Figure 2.1 Rhodium catalyzed C–H insertion of imidoiodobenzene derivative 1.7 ...... 24 Figure 2.2 Rhodium catalyzed C–H amination of primary carbamates.8 ...... 25 Figure 2.3 Regioselective formation of oxathiazinane N,O-acetal (2) and its utility as iminium ion equivalent.9 ...... 26 Figure 2.4 a) Catalytic hydroxylation of L-valine, b) Proposed catalytic cycle for chelate-directed C–H bond functionalization.14 ...... 27 Figure 2.5 Pd-catalyzed oxidation of nonactivated sp3 C–H bonds ...... 28 Figure 2.6 a) Pd-catalyzed iodination of nonactivated sp3 C–H bonds. b) isolated trinuclear palladium alkyl species ...... 29 Figure 2.7 Pd-catalyzed Boc-directed acetoxylation of sp3 C–H bonds ...... 29 Figure 2.8 Pd-catalyzed MPyS-directed acyloxylation of sp3 C–H bonds ...... 30 Figure 2.9 Pd-catalyzed oxime-directed functionalization of sp3 C–H bonds ...... 30 Figure 3.1 A plausible catalytic cycle.1 ...... 38

Figure 4.1 Depiction of E3gH–K3a,eRh2 assembly a) Axial coordination in E3gH–K3a,eRh2 stabilizes the coiled coil. (b) Sequences used in this study. Lower-case grey letters represent positions on a helical- wheel depiction...... 81

Figure 4.2 Selection of thermal denaturation profiles for stoichiometric mixtures of E3H and K3a,eRh2. Vertical lines indicate melting temperature (Tm). See Table 1 for sample conditions and Tm values...... 83 Figure 4.3 Visible absorption spectra of dirhodium-metallopeptide assemblies...... 84 Figure 4.4 Structure of the CAL PDZ domain (orange ribbon) bound to a CFTR peptide (green stick figure)51 All CALP His side chains are shown explicitly (stick figures colored by element; grey = C, blue = N)...... 86 Figure 4.5 Synthesis of a PDZ-binding metallopeptide based on the C-terminal CFTR sequence...... 88 Figure 4.6 HPLC traces and ESI-MS of VQDTRL series ...... 89 Figure 4.7 NMR comparison of (A) VQDTRL peptide B) VQDTRLRh and C) VQDRhTRL...... 90

Figure 4.8 Fluorescence anisotropy displacement isotherms for candidate CALP inhibitors. Ki values are reported in Table 4.2...... 92 Figure 4.9 Metalation improves inhibitor potency ...... 94 Figure 4.10 Structure of N1P1 with a bound peptide ligand ...... 96 Figure 4.11 NMR comparison of (A) EVQSTRL peptide B) EVQSTRLRh and C) ERhVQSTRL ...... 103

Figure 4.12 HPLC traces and ESI-MS data for EVQSTRL series ...... 104 Figure 4.13 HPLC traces and ESI-MS data for QLDVTR series ...... 105 Figure 4.14 HPLC traces and ESI-MS data for EWPTSII series ...... 106 Figure 4.15 HPLC traces and ESI-MS data for EVQSTRI series ...... 107 Figure 4.16 HPLC traces and ESI-MS data for EVQSTRF series ...... 108 Figure 4.17 a) HPLC trace of pure peptide VQDTRL and b) ESI-MS data for isolated peptide VQDTRL. .. 109 Figure 4.18 a) HPLC trace of pure peptide EVQSTRL and b) ESI-MS data for isolated peptide EVQSTRL. 110 Figure 4.19 . a) HPLC trace of pure peptide QLDVTR and b) MS data for isolated peptide QLDVTR...... 111 Figure 4.20 a) HPLC trace of pure peptide EWPTSII and b) MS data for isolated peptide EWPTSII...... 112 Figure 4.21 a) HPLC trace of pure peptide EVQSTRI and b) MS data for isolated peptide EVQSTRI...... 113 Figure 4.22 a) HPLC trace of pure peptide EVQSTRF and b) MS data for isolated peptide EVQSTRF...... 114 Figure 4.23 HSQC spectra of 15N-CALP ...... 115 Figure 5.1 FKBP12 inhibitor selected as design inspiration...... 121 Figure 5.2 Retrosynthetic pathway for designing compound B...... 122 Figure 5.3 Synthesis of metalated inhibitors for FKBP12 ...... 123 Figure 6.1 Modification of CALP using dirhodium metallocarbenes...... 138 Figure 6.2 Control reactions of CALP 16 ...... 139 Figure 6.3 Modified CALP trypsin digest ...... 140 Figure 6.4 Blocking surface cysteines on CALP ...... 141

Figure 6.5 Dose-dependence of CALP modification on biotin-diazo 1 and Rh2(OAc)4 ...... 142 Figure 6.6 Peptide models to study dirhodium metallocarbenoid mediated cysteine modification...... 143 Figure 6.7 Stability study for the linkages in peptide models...... 145 Figure 6.8 Modification of CALP in a mixture of proteins...... 146 Figure 6.9 a) HPLC trace of pure peptide AQPADRCGGLA and b) MALDI-MS of purified peptide AQPADRCGGLA...... 156 Figure 6.10 a) HPLC trace of pure peptide AQPADKCGGLA and b) MALDI-MS of purified peptide AQPADKCGGLA...... 157 Figure 6.11 HPLC trace and ESI-MS of peptide AQPADACGGLA ...... 158

List of Tables

Table 3.1 Optimization of remote C–H functionalization of alkyl-hydroperoxides...... 41 Table 3.2 Optimization for secondary alkyl hydroperoxide (2-octyl hydroperoxide) ...... 43 Table 3.3 Optimizations for tertiary alkyl hydroperoxide (2-methyl, 2-hexyl hydroperoxide) as substrate...... 44 Table 3.4 CuCl catalyzed remote C-H functionalization of alkyl hydroperoxides ...... 45 Table 4.1 Thermal denaturation of metallopeptide coiled coils.[a] ...... 82 Table 4.2 Metallopeptide inhibitors of CALP.[a] ...... 93 Table 4.3 Comparision of binding affinity of metallopeptide inhibitors for N1P1 and CALP.[a]* ...... 97

List of Schemes

+ 6 Scheme 1.1 Proposed two-step process for the reaction of [(R2Im)2Au] with Cys/Sec. ref...... 4 Scheme 2.1 . Directing groups facilitating C–H functionalization.5 ...... 23 Scheme 2.2 Possible metallacycles for Pt–catalyzed remote C–H hydroxylation...... 26 Scheme 3.1 Postulated copper-catalyzed chlorination of alkyl hydroperoxides ...... 34 Scheme 3.2 Synthetic routes of various alkyl hydroperoxides ...... 36 Scheme 3.3 Mechanism of Atom Transfer Radical Polymerization...... 37 Scheme 3.4 Chlorination of a substrate providing mixtures of 1,5 and 1,6 hydrogen abstraction...... 47 Scheme 4.1 Conceptual depiction of a hybrid organic-inorganic inhibitor stabilized by metal–histidine interactions near the interface ...... 80

Abbreviations

Instrumentation:

GC-MS Gas Chromatography-Mass Spectrometry

CD Circular Dichroism

COSY (Two-dimensional NMR) Correlation Spectroscopy

ESI ElectroSpray Ionization

FP Fluorescence Polarization

HPLC High-Performance Liquid Chromatography

NMR Nuclear Magnetic Resonance Spectroscopy

HSQC (Two-dimensional NMR) Heteronuclear Single Quantum Coherence Spectroscopy

MALDI Matrix-Assisted Laser Desorption/Ionization

MS Mass Spectrometry

MS/MS Tandem Mass Spectrometry

UV-vis Ultraviolet–visible Spectroscopy

Materials:

Bipy 2,2'-Bipyridine

DCM Dichloromethane

DMF Dimethylformamide

DMSO Dimethyl Sulfoxide

NHC N-Heterocyclic Carbene

THF Tetrahydrofuran

CHCA α-Cyano-4-hydroxycinnamic Acid

DIEA Diisopropyl Ethyl Amine

MES 2-(N-morpholino)ethanesulfonic Acid

PEG Polyethylene Glycol

TBHA N-tert-butyl Hydroxylamine

TFA Trifluoroacetic Acid

TFE Trifluoroethanol

TIS Triisopropyl Silane

Other:

PDB Protein Data Bank

ET Electron Transfer

rt Room Temperature aq aqueous ee Enantiomeric Excess

Chapter 1 Rational Design of Metal-Based Inhibitors for Proteins

1.1. Introduction

Metal-complexes offer access to a wide range of functional groups owing to their unique properties such as different coordination geometries, tunability of several chemical properties and reactivities (like ligand exchange at the metal center, redox activity etc.), which makes them very desirable for utilization in in unexplored areas of therapeutics.1-4

Since the discovery of cisplatin, a plethora of metalated complexes are used to target DNA for anticancer treatment. The non-specificity of DNA targeting by the metal adversely affects healthy cells as well, causing cytotoxicity.5 Due to this reason, research in medicinal chemistry and chemical biology is increasingly focused on development of drugs with newer mechanistic pathways that target at improved specificity, increased potency and decreased side-effects.

Though traditionally proteins and enzymes have rarely been explored as targets for drugs but owing to their high structural diversity as compared to the conserved DNA structure, proteins can provide unique motifs as targets. Because of the varied functional groups in the proteins, novel metal-based drugs can be designed rationally to target specific proteins and thereby causing least cytotoxicity.5 This article will focus on recent development in the field of metal-

1 based protein inhibitors. Specifically, the approaches towards rational designing of such inhibitors will be emphasized.

1.2. Rational design of metal-based protein-inhibitors

There are numerous proteins that have a regulatory role in signaling pathways, cell proliferation, apoptosis, and onset of a variety of diseases.

Detecting the role of one particular protein for a cellular function and the mechanism of operation is a daunting task for a researcher, and therefore, designing a drug to inhibit a particular protein function becomes even more challenging. Modern chemists and biochemists often get inspired by natural inhibitors, or resort to virtual screening of potential candidates present in literature, and sometimes, use fundamental ideas from the arsenal of chemistry to improve target specificity of a drug. In the current chapter, we have tried to highlight a few examples in recent literature that makes use of these techniques in rational design of metal-based protein-inhibitors.

1.2.1. Improving target specificity by fine tuning ligand exchange reactions at metal center

Berners-Price and Filipovska have shown that lipophilic, cationic Au(I) complexes with fine-tuned ligand-exchange reactions at Au(I) center can selectively target mitochondrial selenoenzymes like TrxR and thereby induce apoptosis in cancer cell lines.6 The thioredoxin (Trx)/thioredoxin reductase (TrxR) regulates cellular redox potential in both cytoplasm and mitochondria. An increase

2 in Trx and TrxR activities are correlated with acceleration in tumor growth and decline of apoptosis,7 whereas the inhibition of TrxR leads to apoptosis in cancer cells.8,9 The mammalian selenoenzyme TrxR is inhibited by Au(I) complexes such as Auranofin and its derivatives9-12 but these complexes also inhibit closely related enzyme glutathione reductase (GR), due to the high thiol reactivity of Au (I) complexes.12 So in spite of exhibiting good activity against growth of tumor cells in vitro,13 the in vivo antitumor activity of Auranofin is limited.14

The authors used Au (I)-N-heterocyclic carbenes (NHCs) complexes,

+ [(R2Im)2Au] to access structurally similar compounds with varying lipophilicity

(log P), as it was suggested in modeling studies15,16 that tuning lipophilicity can increase selectivity of cancer cells over normal cells by increasing the penetration of cations in mitochondria. It was observed that all the Au(I) complexes were selectively toxic to breast cancer cell lines but not to the normal cells and the degree of selectivity correlates with their lipophilicity (log P values). The complex

+ 1(a) [(R2Im)2Au] (R = i-Pr) (Scheme 1.1), having intermediate lipophilicity showed optimum selectivity and cytotoxicity. When MDA-MB-231 breast cancer cell lines were treated with 1a, TrxR activity was inhibited by ~50% whereas no inhibition of glutathione reductase was observed. The improved selectivity was explained by time-dependent 1H NMR studies of the reaction mixture consisting of 1(a) along with cysteine or selenocysteine (Sec). It was established that the reaction proceeds via a two-step process (Scheme 1.1), to successively substitute

+ the two imidazole ligands with Cys/Sec to form either [Au(Cys)2] (5) or

3

+ [Au(Sec)2] (6) and the NHC ligand for Au(I) selenoate was much more labile relative to the thiolate intermediate, thereby increasing the rate constant for the reaction of Au(I) NHC complexes with selenocysteine 20- to 80-fold higher than the reactions with cysteine. This increased rate was explained by the difference in the pKa values of cysteine (pKa = 8.5) and selenocysteine (pKa = 5.2), so at physiological pH 7.2, the selenol is fully ionized, while the thiol is not.

+ Scheme 1.1 Proposed two-step process for the reaction of [(R2Im)2Au] with Cys/Sec. ref.6

1.2.2. Improving potency and drug residence time by fine tuning metal and chelator combination

The chemokine receptor CXCR4, has been identified as a co-receptor protein that helps in the entry of HIV in human cells.17 The xylyl-bicyclam compound AMD3100 (Figure 1.2.a, also known as Mozobil or Plerixafor) was found to selectively target CXCR4 through hydrogen bonding and hydrophobic interactions.18 On complexation of the bicyclam with metals like Zn(II), Ni(II) and

Cu(II), the affinity for CXCR4 increased by factor of 50, 36, and 7 respectively.19

4

The enhanced affinity was attributed to the coordination of an aspartate residue present on the CXCR4 surface to the metal ion.20,21 Hubin, Archibald and coworkers hypothesized that tuning this coordination chemistry, by appropriate pairing of a metal and a rigid chelator, can increase the time an inhibitor remains bound to the receptor (drug residence-time) and the inhibitor’s potency.22

For this purpose, they prepared the copper(II) complex of the highly rigid ethyl-xylyl-bicyclam (Figure 1.1.c). Earlier studies have confirmed that the complex binds to CXCR4 receptors on the cell-surface.23 Experimental assays indicated that both AMD3100 and its copper-bound complex remained active only

4+ till 24 hours, whereas the [Cu2(ethyl-xylyl-bicyclam)] retains activity even at 48 hours. Also, the inhibitor was found to be very potent with EC50 value of 4.3 nM, compared to the published EC50 values24 of AMD3100 (11 nM) and

4+ [Cu2(AMD3100)] (47 nM), in its anti-HIV activity against the R4 strain of HIV.

This fact highlights that the copper(II) complex of the configurationally fixed chelator was able to increase the drug-residence time at CXCR4 receptor and the increased potency directly correlates with the enhanced residence-time.

5

Figure 1.1 Structures of different ligands for CXCR4 inhibition a) Structure of AMD3100. b) Crystal structure of a Ni(II)-cyclam unit with axially coordinated water and Asp101 ((2H9J.pdb generated with PyMOL, Ni green, O red, N blue, C grey). The benzyl linker with the second cyclam unit of AMD3100 could not be located in the electron density map as they are probably disordered. c) Structure of cross-bridged cyclam derivatives: ethyl-xylyl-bicyclam d) folded cis-configuration of ethyl-xylyl-bicyclam. e) ethyl-benzyl-cyclam, f) Crystal structure of [Cu(ethyl-benzyl- cyclam)(OAc)]+, with Cu(II)-centre coordinating to acetate through an equatorial bond (adapted with permission from ref.22).

4+ The greater residence-time of [Cu2(ethyl-xylyl-bicyclam)] relative to

4+ [Cu2(AMD3100)] was explained to be the result of the mode of coordination of the metal ion with the aspartate residue. In case of the square pyramidal complex

6

4+ [Cu2(AMD3100)] , the copper is bound in a stable trans-configuration to the ring

N atoms at all the four equatorial sites in the planar base (Figure 1.1.b), whereas

4+ for the ethyl cross-bridged derivative complex [Cu2(ethyl-xylyl-bicyclam)] , the macrocycle is forced into a folded cis-configuration upon complexation with the metal, where three ring N atoms occupy the equatorial sites, and one ring N atom is axially bound (Figure 1.1.d). When exposed to CXCR4 receptor,

4+ [Cu2(AMD3100)] coordinates with the surface aspartate residue of CXCR4

4+ through a weak axial bond (Figure 1.1.b), whereas [Cu2(ethyl-xylyl-bicyclam)] forms a shorter, stronger equatorial bond with the surface aspartate. This was proved by the X-ray crystallographic studies of acetate complex of the cyclam analogue (Figure 1.1. e, f), to model the interaction of aspartate residue.22 This is a fine example of increasing inhibitor potency by taking advantage of an optimal metal-chelator combination.

1.2.3. Kinetically inert metal complexes for highly selective inhibition of protein kinases

Meggers lab has put forward the idea of using kinetically inert octahedral metal complexes for designing highly selective metal-based inhibitors with increased access to chemical space and increased structural complexity. Inspired by the naturally occurring protein kinase inhibitor staurosporine, they created the

‘octasporine’ design (Figure 1.2.a). This consists of a metal (Ru(II), Ir(III)), placed within the ATP-binding pocket of protein kinase at an optimum position which is

7 neither too exposed to the solvent, nor deeply buried in the hinge-region of protein kinases, and can exert influence for selectivity and potency (Figure 1.2.b).1,25,26

Figure 1.2 Structure of octahedral metal complexes bound to protein kinase a) Staurosporine as an inspiration for the design of Octasporines as protein kinase inhibitor. b) Binding of the octahedral pyridocarbazole metal complex scaffold to the ATP-binding site of a protein kinase, ref.25

By exploiting different octahedral coordination spheres, Meggers and coworkers were able to design highly selective protein kinase inhibitors with sub- nanomolar to nanomolar binding affinities. Apart from several hydrogen-bond

8 interactions and hydrophobic interactions, the role of a monodentate axial ligand was found to be pivotal in determining the selectivity of a metal complex towards one particular protein kinase among 100 or more other protein kinases.25 For example, the Ru-complex OS1, having a carbonyl (CO) axial ligand (Figure 1.3.a) shows an IC50 value of 0.9 nM for the protein kinase GSK3. Within a tested panel of 102 protein kinases, 98 protein kinases were not inhibited, and GSK3 showed the highest inhibition with residual activity of only 6%. Changing the axial ligand to thiocyanate (NCS) (along with some other changes in the peripheral ligands), for Ru-complex OS4 (Figure 1.3.b), the selectivity changes towards the protein kinase DAPK1 with an IC50 value of 2 nM. Again, within a tested panel of 102 protein kinases, OS4 was found to be selective for DAPK1 with a selectivity factor of ~ 4.4. Cocrystal structures of the metalated-inhibitors along with the proteins showed that the axial ligand interacts with the flexible glycine- rich loop of the protein kinase, and the selectivity arises from the shape- complimentarity of the ligand with the glycine-rich loop conformation. The small and linear CO ligand fits the small hydrophobic pocket formed by the glycine-rich loop in GSK3whereas a more ‘stretched-out’ pocket of glycine-rich loop of

DAPK1 can accommodate the larger and bent NCS ligand for optimal interactions

(Figure 1.3.c).25

9

Figure 1.3 Kinetically inert octahedral metal complexes for protein kinases a) Molecular structure of Λ OS1, b) molecular structure of OS4, c) Superimposed structures of GSK3β/OS1 (carbon atoms in gray) and DAPK1/OS4 (carbon atoms in pink) showing relative binding position with the hinge region and glycine-rich loop of protein kinase, ref.25

Another example in this section is the cyclometalated iridium(III) complex

[Ir(ppy)2(biq)]PF6 (Figure 1.4), which was reported as the first metal-based inhibitor of tumor necrosis factor–(TNF–The complex has an extended

10 aromatic biq ligand, that can potentially form favorable interactions with the hydrophobic binding interface of TNF-.

Figure 1.4 Chemical formula of [Ir(ppy)2(biq)]PF6, ∆ enantiomer, ref.27

The mechanism of action of the metal complex with TNF-, was analyzed by studying the interactions of both ∆ and Λ enantiomers of [Ir(ppy)2(biq)]PF6, with the protein using molecular modeling. The authors found that both the enantiomers of [Ir(ppy)2(biq)]PF6 were predicted to interact with the subunits of

TNF-, in the binding pocket via hydrophobic interactions to an equal extent.

Corresponding rhodium-analogues, though predicted to occupy the same binding pocket, was found to interact inefficiently with the subunits, via molecular modeling. The computational studies were verified by carrying out the inhibition assays in cells, and it was observed that rhodium complexes showed 15 % inhibition as opposed to 70 % inhibition by iridium analogues. Also other analogues of the iridium complex Ir(ppy)2(biq)]PF6 , devoid of the biq-ligand such as [Ir2(ppy)4Cl2], [Ir2(bzq)4Cl2], [Ir(ppy)2(H2O)2]OTf, did not show high binding

11 affinities, indicating the role of biq ligand in effective interaction with the TNF- subunits. This report highlights the use of molecular modeling in designing a potent inhibitor based on its shape-complimentarity with the binding pocket of the protein.

Recently, examples of other kinetically inert octahedral metal complexes, including Ir-based complex selectively inhibiting protein kinase and triggering visible-light induced apoptosis in cancer cells28,29 and complexes of Rh30,31 and planar metal complexes of Pt32 as inhibitors for protein kinases have been reported.

1.2.4. Affinity enhancement of an inhibitor by combining two weak interacting fragments

Coupling two molecules that independently bind to two separate, specific sites can significantly enhance enzyme-inhibitor affinity.33 This approach was utilized in the inhibition of the zinc-enzyme carbonic anhydrase, where an active pharmacophore of carbonic anhydrase, sulfanilamide was attached to a Cu(II) bound peptide GGH (Figure 1.5). Carbonic anhydrase contains an amino terminal

Cu/Ni binding motif (ATCUN), which has a high affinity towards Cu/ Ni metal ions under physiological conditions. Therefore, the sulfanilamide-Cu-GGH complex is initially directed by the sulfanilamide fragment to the active site of enzyme, where the ATCUN motif coordinates to the metal through surface exposed histidines. Once bound to the enzyme, the metal mediates oxidation of

12 residues in its close proximity within the active site, and thus inhibits the enzyme.34

Figure 1.5 Chemical structures of sulfanilamide-Cu-GGH complex

Also, previous reports have shown that coupling cupric iminodiacetate

(IDA-Cu2+; designed to bind to solvent exposed histidine residues of carbonic anhydrase I) to benzenesulfonamide results in up to 100-fold increases in binding affinity.35,36

The Ru-ethacrynic acid inhibitor of glutathione transferase Pi class (GST

P1-1) is another example in this category. The over expression of GST P1-1 in tumor cells, after exposure to antitumor drugs, is implicated in detoxification of anticancer agents, making it a target in anticancer therapeutics.37 Ethacrynic acid

(EA) (Figure 1.6.a) is a known inhibitor of GST P1-1 and binds to its hydrophobic site (H-site).38 GST P1-1 also contains two solvent accessible cysteine residues. To utilize the availability of these cysteine residues, the authors looked into ruthenium-derivatives of ethacrynic acid (EA). Ruthenium-complexes interact with soft nucleophilic centers like thiols and are also known to be less toxic than their platinum analogues.39 Because of these advantages, the authors

13 used ruthenium to prepare metalated- ethacrynic acid derivatives (Figure 1.6 b) that can bind to the H-site of GST P1-1 and can also interact with its reactive cysteine residues. The resulting complexes showed an affinity enhancement (3-4 times) as compared to the uncomplexed ethacrynic acid.40 Cocrystal structure of the enzyme along with the ruthenium-complex showed that chloride ligands coordinated to ruthenium are displaced by thiol side-chains of the two solvent exposed cysteine residues of GST P1-1 (Figure 1.6 d).

Figure 1.6 Ethacrynic acid derivatives as inhibitors for GST P1-1 Chemical structures of a) ethacrynic acid, and b) Ru-derivative of ethacrynic acid. c) Interactions between ethacrynic acid (EA) and GSTP1-1, and d) interactions between Ru- derivative of EA and GST P1-1, as seen in X-ray crystallography, ref.40

14

1.2.5. Inducing selectivity by incorporating a biological recognition sequence in a metal inhibitor

An interesting approach to increase selectivity for a particular protein target is to incorporate a recognition sequence in a known inhibitor. Recently Meade group has developed an oligonucleotide adduct of Co(III)-schiff base complex for selective inhibition of Snail family zinc-finger transcription factors.41 Previous work from this group has shown that Co(III)-schiff base complexes can inhibit

DNA bnding of zinc-finger transcription factors.42

Figure 1.7 Chemical structures of a) Co(III)-schiif base complex, b) the duplex DNA Ebox, c) Co(III)-Ebox conjugate, ref.41

The mechanism of inhibition involved irreversible coordination of Co(III) to nitrogen lone pairs of histidine present in zinc-finger proteins, which disrupts

15 the coordination environment of Zn(II). For selective binding to only a subgroup of zinc-finger proteins, the Snail family (known to be responsible for tumor metastasis), they incorporated a modified oligonucleotide called Ebox, containing the recognition sequence ‘CAGGTG’ specific to Snail factors, with the Co(III)- schiff base complex. It was observed that the specificity to inhibit only the Snail family transcriptase increases to 150 fold over unconjugated Co(III)-complex.

Also, controls of just oligonucleotide or Co(III)-schiff base complex by themselves were found to be insufficient for effective inhibition of the target DNA transcriptase.

1.2.6. Altering target and action of a drug by incorporating a metal- fragment

Coupling metals with organic fragments can not only enhance the affinity of an inhibitor for its protein but can also alter the target and action of the inhibitor. Aspirin (acetylsalicylic acid), a universally used non-steroidal anti- inflammatory drug (NSAID), functions through inhibition of cyclooxygenase

(COX) enzymes. On attaching an alkyne-derivative of hexacarbonyldicobalt to aspirin, a compound with anti-proliferative properties against breast cancer cells was formed.43 It was found that the anti-proliferative properties of the new cobalt- aspirin complex results from the alteration in the mode of inhibition. Aspirin preferentially inhibits COX-1 enzyme, whereas the cobalt-aspirin adduct equally inhibits both COX-1 and COX-2 enzymes (COX-2 enzyme was found to be

16 overexpressed in tumor cells). Experimental evidence indicates that the cobalt- aspirin complex acetylates multiple lysine residues in proximity of the active site whereas aspirin acetylates a serine residue. Thus, attachment of a metal-fragment has altered the target and mode of inhibition of the COX inhibitor aspirin.

Figure 1.8 Molecular structures of aspirin and hexacarbonyldicobalt derivative of aspirin

Another interesting example in this category is Tamoxifen, which is used for preventing cell proliferation in hormone-dependent breast-cancer. It’s known to act via inhibition of the hormone estrogen through competitive binding to the estrogen receptor ER-. For inducing cytotoxicity, tamoxifen derivatives were conjugated with redox-active organometallic fragments like ferrocene. When an amine-containing side-chain, was substituted with a ferrocenyl-group, the activity of the resulting complex shifts from antiestrogenicity to cytotoxicity. It was also found that additional phenol substituents and a keto-group adjacent to the ferrocene group were essential in inducing the cytotoxicity to the complex.44

1.2.7. Other examples Other examples in this field includes Au(III)-dithiocarbamate complex that induces apoptosis in highly metastatic cancer cells, by inhibiting proteasomes.45

17

Also, there is a new class of Au(III)-porphyrin complexes that exhibits anticancer properties. It is suggested that these complexes inhibit class I histone deacetylase

(HDAC) and thus effect aberrant signal pathways.46 The mechanisms for these pathways are currently under investigation.

1.3. Conclusions

This chapter illustrates the advantages of incorporating metals in protein inhibitors for increasing their potency and the recent advances in this field. Of particular interest are the strategies involved in optimizing selectivity and potency of an inhibitor and the role played by the incorporated metal in doing so. The modulation of ligand exchange reactions at the metal center to target a specific protein; or combining a specific metal and ligand that increases the inhibitor- potency due to shape-complementarity, or by a stronger mode of coordination to protein leading to increase in drug-residence time, highlights the multiple functionality of a metal center. This field is still under-developed, but very promising and the use of metalated inhibitors has the potential to uncover novel, unprecedented properties that can act as solutions to several problems in medicine and chemical biology.

1.4. References

(1) Meggers, E.: Targeting proteins with metal complexes. Chem. Commun. 2009, 1001-1010. (2) Davies, C. L.; Dux, E. L.; Duhme-Klair, A.-K.: Supramolecular interactions between functional metal complexes and proteins. Dalton Trans. 2009, 10141-10154.

18

(3) Bruijnincx, P. C. A.; Sadler, P. J.: New trends for metal complexes with anticancer activity. Curr. Opin. Chem. Biol. 2008, 12, 197-206. (4) Patra, M.; Gasser, G.: Organometallic Compounds: An Opportunity for Chemical Biology? ChemBioChem 2012, 13, 1232-1252. (5) Lu, L.; Zhu, M.: Metal-based inhibitors of protein tyrosine phosphatases. Anti- cancer Agents in Medicinal Chemistry 2011, 11, 164-71. (6) Hickey, J. L.; Ruhayel, R. A.; Barnard, P. J.; Baker, M. V.; Berners-Price, S. J.; Filipovska, A.: Mitochondria-Targeted Chemotherapeutics: The Rational Design of Gold(I) N- Heterocyclic Carbene Complexes That Are Selectively Toxic to Cancer Cells and Target Protein Selenols in Preference to Thiols. J. Am. Chem. Soc. 2008, 130, 12570-12571. (7) Lincoln, D. T.; Ali Emadi, E. M.; Tonissen, K. F.; Clarke, F. M.: The thioredoxin- thioredoxin reductase system: over-expression in human cancer. Anticancer Res. 2003, 23, 2425-33. (8) K, A.; ES, A.: - Rapid induction of cell death by selenium-compromised thioredoxin reductase 1 but. In - J Biol Chem. 2003 May 2;278(18):15966-72. Epub 2003 Feb 6. (9) C, M.; V, G.; A, F.; G, S.; A, B.; MP, R.: - Inhibition of thioredoxin reductase by auranofin induces apoptosis in. In - Free Radic Biol Med. 2007 Mar 15;42(6):872-81. Epub 2006 Dec 22. (10) Gromer, S.; Urig, S.; Becker, K.: The thioredoxin system--from science to clinic. Med Res Rev. 2004, 24, 40-89. (11) Gromer, S.; Arscott, L. D.; Williams, C. H., Jr.; Schirmer, R. H.; Becker, K.: Human placenta thioredoxin reductase. Isolation of the selenoenzyme, steady state kinetics, and inhibition by therapeutic gold compounds. J Biol Chem. 1998, 273, 20096-101. (12) Urig, S.; Fritz-Wolf, K.; Reau, R.; Herold-Mende, C.; Toth, K.; Davioud-Charvet, E.; Becker, K.: Undressing of phosphine gold(I) complexes as irreversible inhibitors of human disulfide reductases. Angew Chem Int Ed Engl. 2006, 45, 1881-6. (13) Mirabelli, C. K.; Johnson, R. K.; Hill, D. T.; Faucette, L. F.; Girard, G. R.; Kuo, G. Y.; Sung, C. M.; Crooke, S. T.: Correlation of the in vitro cytotoxic and in vivo antitumor activities of gold(I) coordination complexes. J Med Chem. 1986, 29, 218-23. (14) Mirabelli, C. K.; Johnson, R. K.; Sung, C. M.; Faucette, L.; Muirhead, K.; Crooke, S. T.: Evaluation of the in vivo antitumor activity and in vitro cytotoxic properties of auranofin, a coordinated gold compound, in murine tumor models. Cancer Res. 1985, 45, 32-9. (15) Kandela, I. K.; Lee, W.; Indig, G. L.: Effect of the lipophilic/hydrophilic character of cationic triarylmethane dyes on their selective phototoxicity toward tumor cells. Biotech Histochem. 2003, 78, 157-69. (16) S, T.; RW, H.: - A predictive model for the selective accumulation of chemicals in tumor cells. - Eur Biophys J. 2005 Oct;34(7):959-66. Epub 2005 May 14. (17) Liang, X.: CXCR4, Inhibitors and Mechanisms of Action. Chemical Biology & Drug Design 2008, 72, 97-110. (18) De Clercq, E.: The bicyclam AMD3100 story. Nat. Rev. Drug Discovery 2003, 2, 581-587. (19) Gerlach, L. O.; Jakobsen, J. S.; Jensen, K. P.; Rosenkilde, M. R.; Skerlj, R. T.; Ryde, U.; Bridger, G. J.; Schwartz, T. W.: Metal Ion Enhanced Binding of AMD3100 to Asp262 in the CXCR4 Receptor†. Biochemistry 2002, 42, 710-717. (20) Hunter, T. M.; McNae, I. W.; Liang, X.; Bella, J.; Parsons, S.; Walkinshaw, M. D.; Sadler, P. J.: Protein recognition of macrocycles: Binding of anti-HIV metallocyclams to lysozyme. Proc. Nat. Acad. Sci. U.S.A. 2005, 102, 2288-2292.

19

(21) Hunter, T. M.; McNae, I. W.; Simpson, D. P.; Smith, A. M.; Moggach, S.; White, F.; Walkinshaw, M. D.; Parsons, S.; Sadler, P. J.: Configurations of Nickel–Cyclam Antiviral Complexes and Protein Recognition. Chemistry – A European Journal 2007, 13, 40-50. (22) Khan, A.; Nicholson, G.; Greenman, J.; Madden, L.; McRobbie, G.; Pannecouque, C.; De Clercq, E.; Ullom, R.; Maples, D. L.; Maples, R. D.; Silversides, J. D.; Hubin, T. J.; Archibald, S. J.: Binding Optimization through Coordination Chemistry: CXCR4 Chemokine Receptor Antagonists from Ultrarigid Metal Complexes. J. Am. Chem. Soc. 2009, 131, 3416-3417. (23) Lee, B.; Sharron, M.; Montaner, L. J.; Weissman, D.; Doms, R. W.: Quantification of CD4, CCR5, and CXCR4 levels on lymphocyte subsets, dendritic cells, and differentially conditioned monocyte-derived macrophages. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 5215-5220. (24) Este, J. A.; Cabrera, C.; De Clercq, E.; Struyf, S.; Van Damme, J.; Bridger, G.; Skerlj, R. T.; Abrams, M. J.; Henson, G.; Gutierrez, A.; Clotet, B.; Schols, D.: Activity of different bicyclam derivatives against human immunodeficiency virus depends on their interaction with the CXCR4 chemokine receptor. Mol Pharmacol. 1999, 55, 67-73. (25) Feng, L.; Geisselbrecht, Y.; Blanck, S.; Wilbuer, A.; Atilla- o cumen, . . Filippa opoulos, P. r ling, K.; Celik, M. A.; Harms, K.; Maksimoska, J.; Marmorstein, R.; Frenking, G.; Knapp, S.; Essen, L.-O.; Meggers, E.: Structurally Sophisticated Octahedral Metal Complexes as Highly Selective Protein Kinase Inhibitors. J. Am. Chem. Soc. 2011, 133, 5976-5986. (26) Blanck, S.; Maksimoska, J.; Baumeister, J.; Harms, K.; Marmorstein, R.; Meggers, E.: The Art of Filling Protein Pockets Efficiently with Octahedral Metal Complexes. Angew. Chem. Int. Ed. 2012, 51, 5244-5246. (27) Leung, C.-H.; Zhong, H.-J.; Yang, H.; Cheng, Z.; Chan, D. S.-H.; Ma, V. P.-Y.; Abagyan, R.; Wong, C.-Y.; Ma, D.-L.: A Metal-Based Inhibitor of Tumor Necrosis Factor-α. Angew. Chem. Int. Ed. 2012, 51, 9010-9014. (28) Wilbuer, A.; Vlecken, D. H.; Schmitz, D. J.; Kräling, K.; Harms, K.; Bagowski, C. P.; Meggers, E.: Iridium Complex with Antiangiogenic Properties. Angew. Chem. Int. Ed. 2010, 49, 3839-3842. (29) Kastl, A.; Wilbuer, A.; Merkel, A. L.; Feng, L.; Di Fazio, P.; Ocker, M.; Meggers, E.: Dual anticancer activity in a single compound: visible-light-induced apoptosis by an antiangiogenic iridium complex. Chem. Commun. 2012, 48, 1863-1865. (30) Dieckmann, S.; Riedel, R.; Harms, K.; Meggers, E.: Pyridocarbazole-Rhodium(III) Complexes as Protein Kinase Inhibitors. Eur. J. Inorg. Chem. 2012, 2012, 813-821. (31) Leung, C.-H.; Yang, H.; Ma, V. P.-Y.; Chan, D. S.-H.; Zhong, H.-J.; Li, Y.-W.; Fong, W.-F.; Ma, D.-L.: Inhibition of Janus kinase 2 by cyclometalated rhodium complexes. MedChemComm 2012, 3, 696-698. (32) Child, E.; Georgiades, S.; Rose, K.; Stafford, V.; Patel, C.; Steinke, J.; Mann, D.; Vilar, R.: Inhibition of mitogen-activated protein kinase (MAPK) and cyclin-dependent kinase 2 (Cdk2) by platinum(II) phenanthroline complexes. J. Chem. Biol. 2011, 4, 159-165. (33) Jude, K. M.; Banerjee, A. L.; Haldar, M. K.; Manokaran, S.; Roy, B.; Mallik, S.; Srivastava, D. K.; Christianson, D. W.: Ultrahigh Resolution Crystal Structures of Human Carbonic Anhydrases I and II Complexed with “Two-Prong” Inhibitors Reveal the Molecular Basis of High Affinity. J. Am. Chem. Soc. 2006, 128, 3011-3018. (34) Gokhale, N. H.; Bradford, S.; Cowan, J. A.: Catalytic Inactivation of Human Carbonic Anhydrase I by a Metallopeptide−Sulfonamide Conjugate is Mediated by Oxidation of Active Site Residues. J. Am. Chem. Soc. 2008, 130, 2388-2389. (35) Banerjee, A. L.; Swanson, M.; Roy, B. C.; Jia, X.; Haldar, M. K.; Mallik, S.; Srivastava, D. K.: Protein Surface-Assisted Enhancement in the Binding Affinity of an Inhibitor for Recombinant Human Carbonic Anhydrase-II. J. Am. Chem. Soc. 2004, 126, 10875-10883.

20

(36) Roy, B. C.; Banerjee, A. L.; Swanson, M.; Jia, X. G.; Haldar, M. K.; Mallik, S.; Srivastava, D. K.: Two-Prong Inhibitors for Human Carbonic Anhydrase II. J. Am. Chem. Soc. 2004, 126, 13206-13207. (37) Townsend, D. M.; Findlay, V. L.; Tew, K. D.: Glutathione S-transferases as regulators of kinase pathways and anticancer drug targets. Methods Enzymol 2005, 401, 287- 307. (38) Lo, H. W.; Ali-Osman, F.: Genetic polymorphism and function of glutathione S- transferases in tumor drug resistance. Curr Opin Pharmacol. 2007, 7, 367-74. Epub 2007 Aug 6. (39) Hartinger, C. G.; Dyson, P. J.: Bioorganometallic chemistry--from teaching paradigms to medicinal applications. Chem Soc Rev. 2009, 38, 391-401. Epub 2008 Nov 25. (40) Ang, W. H.; Parker, L. J.; De Luca, A.; Juillerat-Jeanneret, L.; Morton, C. J.; Lo Bello, M.; Parker, M. W.; Dyson, P. J.: Rational Design of an Organometallic Glutathione Transferase Inhibitor. Angew. Chem. Int. Ed. 2009, 48, 3854-3857. (41) Harney, A. S.; Lee, J.; Manus, L. M.; Wang, P.; Ballweg, D. M.; LaBonne, C.; Meade, T. J.: Targeted inhibition of Snail family zinc finger transcription factors by oligonucleotide-Co(III) Schiff base conjugate. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 13667- 13672. (42) Louie, A. Y.; Meade, T. J.: A Cobalt Complex that Selectively Disrupts the Structure and Function of Zinc Fingers. Proc. Nat. Acad. Sci. U.S.A. 1998, 95, 6663-6668. (43) Ott, I.; Kircher, B.; Bagowski, C. P.; Vlecken, D. H. W.; Ott, E. B.; Will, J.; Bensdorf, K.; Sheldrick, W. S.; Gust, R.: Modulation of the Biological Properties of Aspirin by Formation of a Bioorganometallic Derivative. Angew. Chem. Int. Ed. 2009, 48, 1160-1163. (44) Nguyen, A.; Top, S.; Pigeon, P.; Vessières, A.; Hillard, E. A.; Plamont, M.-A.; Huché, M.; Rigamonti, C.; Jaouen, G.: Synthesis and Structure–Activity Relationships of Ferrocenyl Tamoxifen Derivatives with Modified Side Chains. Chemistry – A European Journal 2009, 15, 684-696. (45) Milacic, V.; Chen, D.; Ronconi, L.; Landis-Piwowar, K. R.; Fregona, D.; Dou, Q. P.: A Novel Anticancer Gold(III) Dithiocarbamate Compound Inhibits the Activity of a Purified 20S Proteasome and 26S Proteasome in Human Breast Cancer Cell Cultures and Xenografts. Cancer Research 2006, 66, 10478-10486. (46) Chow, K. H.-M.; Sun, R. W.-Y.; Lam, J. B. B.; Li, C. K.-L.; Xu, A.; Ma, D.-L.; Abagyan, R.; Wang, Y.; Che, C.-M.: A Gold(III) Porphyrin Complex with Antitumor Properties Targets the Wnt/β-catenin Pathway. Cancer Research 2010, 70, 329-337.

21

Chapter 2

Metal catalyzed directed sp3 C–H bond activation

2.1. Introduction

Hydrocarbons not only form the basic building blocks of life, but are also the main components of oil and natural gas. But in spite of such an abundance of

C–H bonds in organic molecules, general organic synthetic methods focus on transformations of pre-existing functionality present on the substrate to avail a target molecule. This is mainly because of the inertness of C–H bonds, which arises due to the high homolytic bond strength, low acidity and basicity.1,2 It can be envisioned that if any synthetic method selectively functionalizes the unreactive aliphatic C–H bond, then it can potentially form a powerful and broadly applicable class of organic transformations through new disconnections in the retrosynthetic analysis. A few processes such as hydroxylation, oxidative coupling, and oxidative dehydrogenation of C–H bonds could provide more valuable products such as alcohols, higher alkanes, and alkenes. But though most of the processes are thermodynamically favorable transformations, they are difficult to carry out at high selectivity and reaction rates.1,3,4

This chapter focuses on a subfield of C–H functionalization, involving substrates containing one or more pre-existing functional groups that can be used to chelate a metal catalyst and direct it for selective C–H cleavage. In particular

22 we have tried to emphasize on developments in the area of using directing groups for functionalizing sp3 C–H bonds via transition-metal catalyzed reactions.

2.2. Functionalizing sp3 C–H bonds using proximal-directing groups

Transition metal-catalyzed C–H activation on substrates with a pre-existing functional group is called “further functionalization”.5 The pre-existing functional group of the substrate can be used to coordinate the metal and position it for a C–

H cleavage to install a functional group on an nonactiavted sp3 carbon center

(Scheme 2.1). This strategy proceeds in a site-selective fashion and effectively avoids the extra steps of installation and removal of an external directing group.

This directing group can either be strong σ-donor or π-acceptor groups containing nitrogen, sulfur, phosphorous or weak coordinating groups (ketone, ethers, carboxylic acids etc.).

Scheme 2.1 . Directing groups facilitating C–H functionalization.5

Although various methodologies focusing on metal catalyzed directed sp3

C–H bond activation have been developed over the years with diverse range of substrates, most reports are those for activated sp3 C–H functionalization. This is due to several reasons such as the high bond dissociation energy of nonactivated sp3 C–H bonds that strongly affects the reactivity, the susceptibility of the products

23 towards over-oxidation, and the difficulty of achieving regioselective functionalization in complex molecules. The bond dissociation energy for nonactivated sp3 C–H bonds is generally above 95 kcal/mol in contrast with that of activated sp3 C–H bonds such as allylic (88 kcal/mol), benzylic (90 kcal/mol) or

Cα–H bonds of ethers or amines (92-90 kcal/mol).6 Due to this reason, new methodologies continue to be developed in the field of sp3 C–H bond activation in order to address all the above mentioned issues.

2.2.1. Rh-catalyzed amination of sp3 C–H bonds

In 1983, Breslow reported the rhodium catalyzed C–H insertion of imidoiodo benzene derivative 1.7 The reaction proceeded with high selectivity giving 86 % of the insertion product 2. This reaction involved benzylic C–H bonds, and remains the first report of rhodium catalyzed C–H activation. This reaction is believed to occur through intramolecular metal-nitrene insertion into an isopropyl methine C–H bond, though no mechanistic experimental data was offered. It can be mentioned here that imidoiodo benzene derivatives are labile and can decompose e plosi ely at 1 C.

Figure 2.1 Rhodium catalyzed C–H insertion of imidoiodobenzene derivative 1.7

24

In 2001, Du Bois utilized the above reaction conditions for the intramolecular remote sp3 C–H amination of primary carbamates. The reaction was carried out in the presence of iodosobenzene diacetate, bastilie and catalytic

8 Rh2(OAc)4, to generate in situ imidoiodo benzene derivative. It was observed that the reaction proceeds with retention of stereochemistry indicating direct insertion of nitrene into the methine center. The substrates containing benzylic or tertiary

C–H bonds showed much higher conversions (74-84%) compared to less reactive compounds containg nonactivated C–H bonds (44%). The reaction gives rise to substituted oxazolidinones which can act as precursors for synthesis of 1,2 –amino alcohols.

Figure 2.2 Rhodium catalyzed C–H amination of primary carbamates.8

Du Bois’s strategy of Rh-catalyzed amination of ethereal Cα-H bonds is also very interesting.9 They have used the reactivity of ethereal Cα-H bonds and architecture of a Rh catalyst in combination to direct regioselective amination of substituted sulfamate starting materials. This kind of amination gives rise to unique N,O-acetals (2, Figure 2.3) which display high versatility as iminium ion equivalents. These N,O-acetals can couple smoothly and with high

25 diastereoinduction to allyl silanes, enol ethers, silyl ketene acetals and alkynyl zinc reagents.9,10

Figure 2.3 Regioselective formation of oxathiazinane N,O-acetal (2) and its utility as iminium ion equivalent.9 2.2.2. Pt-mediated directed sp3 C–H activation

Scheme 2.2 Possible metallacycles for Pt–catalyzed remote C–H hydroxylation.

In early 199 ’s, Sen and coworkers reported chelate-assisted remote hydroxylation of C–H bonds in alcohols and acids with Pt-catalyzed reactions.11-13

The order of reactivity was C–H << C–H < C–H < C–H for alcohols

26 and C–H << C–-H < C–H ≥ C–H for the acids. It was observed that the regioselectivity arises due to the tendency of formation of the metallacycle with least ring strain.11

In 2001, Sames reported the directed activation of remote nonactivated sp3

C–H bonds of amino-acids in water.14 Though the product yields of the reaction are moderate, it provides a rare example of catalytic heteroatom-directed functionalization of remote alkyl groups.

Figure 2.4 a) Catalytic hydroxylation of L-valine, b) Proposed catalytic cycle for chelate- directed C–H bond functionalization.14

27

2.2.3. Pd-catalyzed directed sp3 C–H activation

In 2004, Sanford and coworkers reported a highly regio- and chemoselective Pd(II)-catalyzed oxygenation of unactivated sp3 C–H bonds of

15 oxime and pyridine substrates, using PhI(OAc)2 as a stoichiometric oxidant. The regioselectivite oxidation of 1° versus 2° carbon was explained by the strong steric preference for formation of less hindered 1° Pd alkyls, and high reactivity of β

(versus α or γ) C–H bonds was explained by the ease of formation of five- membered palladacycles.

Figure 2.5 Pd-catalyzed oxidation of nonactivated sp3 C–H bonds

Yu group has done an extensive body of work in the field of Pd-catalyzed

C–H activation.5 In 2005, they reported an auxiliary approach for chemoselective and asymmetric room-temperature iodination of α-methyl groups of saturated aliphatic acids.16 They used a hindered σ-chelating auxiliary, such as oxazoline to facilitate the assembly of the pre-transition state for cyclometalation (Figure 2.6 b).

28

Figure 2.6 a) Pd-catalyzed iodination of nonactivated sp3 C–H bonds. b) isolated trinuclear palladium alkyl species

Yu group has also shown weakly coordinating groups like N-tert- butoxycarbonyl (Boc) directing groups to activate and functionalize sp3 C–H bonds via Pd-catalyzed reaction.17

Figure 2.7 Pd-catalyzed Boc-directed acetoxylation of sp3 C–H bonds

29

Recently, Pd-catalyzed acyloxylation of 1°- β sp3 C–H bonds facilitated through a novel directing group S-methyl-S-2-pyridylsulfoximine (MPyS) was reported. The reaction takes place at room temperature, and the directing group can be easily removed and reused.18

Figure 2.8 Pd-catalyzed MPyS-directed acyloxylation of sp3 C–H bonds

Dong and coworkers strategy of using designed oximes as alcohol surrogate and a directing group (DG) for metal catalyzed site-selective functionalization of unactivated β sp3 C–H bonds is also noteworthy.19

Figure 2.9 Pd-catalyzed oxime-directed functionalization of sp3 C–H bonds

30

2.3. Conclusions

Utilizing substrates with directing groups for transition metal catalyzed C–

H activation has high potential for developing synthetically versatile reactions.

Though still underdeveloped, this approach can play an instrumental role in the study of ligand-controlled catalytic C–H functionalization reactions with transition metals in the future.

2.4. References

(1) Shilov, A. . Shul'pin, . B.: Activation of C−H Bonds by Metal Complexes. Chem. Rev. 1997, 97, 2879-2932. (2) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.; Peterson, T. H.: Selective Intermolecular Carbon-Hydrogen Bond Activation by Synthetic Metal Complexes in Homogeneous Solution. Acc. Chem. Res. 1995, 28, 154-162. (3) Crabtree, R. H.: Aspects of Methane Chemistry. Chem. Rev. 1995, 95, 987-1007. (4) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E.: Homogeneous Oxidation of Alkanes by Electrophilic Late Transition Metals. Angew. Chem. Int. Ed. 1998, 37, 2180-2192. (5) Engle, K. M.; Mei, T.-S.; Wasa, M.; Yu, J.-Q.: Weak Coordination as a Powerful Means for Developing Broadly Useful C–H Functionalization Reactions. Acc. Chem. Res. 2012, 45, 788-802. (6) a -Reque o, M. M. P re , P. J.: Coinage Metal Cataly ed C−H Bond Functionalization of Hydrocarbons. Chem. Rev. 2008, 108, 3379-3394. (7) Breslow, R.; Gellman, S. H.: Intramolecular nitrene carbon-hydrogen insertions mediated by transition-metal complexes as nitrogen analogs of cytochrome P-450 reactions. J. Am. Chem. Soc. 1983, 105, 6728-6729. (8) Espino, C. G.; Du Bois, J.: A Rh-Cataly ed C−H Insertion Reaction for the Oxidative Conversion of Carbamates to Oxazolidinones. Angew. Chem. Int. Ed. 2001, 40, 598- 600. (9) Williams Fiori, K.; Fleming, J. J.; Du Bois, J.: Rh-Catalyzed Amination of Ethereal Cα–H Bonds: A Versatile Strategy for the Synthesis of Complex Amines. Angew. Chem. Int. Ed. 2004, 43, 4349-4352. (10) Fleming, J. J.; Fiori, K. W.; Du Bois, J.: Novel Iminium Ion Equivalents Prepared through C−H Oxidation for the Stereocontrolled Synthesis of Functionalized Propargylic Amine Derivatives. J. Am. Chem. Soc. 2003, 125, 2028-2029. (11) Basickes, N.; Sen, A.: Platinum(II) mediated oxidation of remote C H bonds in functionalized organic molecules. Polyhedron 1995, 14, 197-202. (12) Sen, A.; Benvenuto, M. A.; Lin, M.; Hutson, A. C.; Basickes, N.: Activation of Methane and Ethane and Their Selective Oxidation to the Alcohols in Protic Media. J. Am. Chem. Soc. 1994, 116, 998-1003. (13) Sen, A.; Lin, M.; Kao, L. C.; Hutson, A. C.: Carbon-hydrogen activation in aqueous medium. The diverse roles of platinum(II) and metallic platinum in the catalytic and

31 stoichiometric oxidative functionalization of organic substrates including alkanes. J. Am. Chem. Soc. 1992, 114, 6385-6392. (14) Dangel, B. D.; Johnson, J. A.; Sames, D.: Selective Functionalization of Amino Acids in Water: A Synthetic Method via Catalytic C−H Bond Activation. J. Am. Chem. Soc. 2001, 123, 8149-8150. (15) Desai, L. V.; Hull, K. L.; Sanford, M. S.: Palladium-Catalyzed Oxygenation of Unactivated sp3 C−H Bonds. J. Am. Chem. Soc. 2004, 126, 9542-9543. (16) Giri, R.; Chen, X.; Yu, J.-Q.: Palladium-Catalyzed Asymmetric Iodination of ild Conditions. Angew. Chem. Int. Ed. 2005, 44, 2112-2115. (17) Wang, D.-H.; Hao, X.-S.; Wu, D.-F.; Yu, J.-Q.: Palladium-Catalyzed Oxidation of Boc-Protected N-Methylamines with IOAc as the Oxidant: A Boc- irected sp3 C−H Bond Activation. Org. Lett. 2006, 8, 3387-3390. (18) Rit, R. K.; Yadav, M. R.; Sahoo, A. K.: Pd(II)-Catalyzed Primary-C(sp3)–H Acyloxylation at Room Temperature. Org. Lett. 2012, 14, 3724-3727. (19) Ren, Z.; Mo, F.; Dong, G.: Catalytic Functionalization of Unactivated sp3 C–H Bonds via exo-Directing Groups: Synthesis of Chemically Differentiated 1,2-Diols. J. Am. Chem. Soc. 2012, 134, 16991-16994.

32

Chapter 3

Copper–Catalyzed Remote sp3 C–H Chlorination of Alkyl Hydroperoxide

Parts of this chapter have been adapted from:1

Kundu, R.; Ball, Z. T., “Copper-Catalyzed Remote sp3 C−H Chlorination of Alkyl Hydroperoxides.” Org. Lett. 2010, 12 (11), 2460-2463.

3.1. Introduction

Metal-catalyzed reactions involving radical-based C–H-activation remain underdeveloped relative to two-electron C–H-activation processes.2-6 However, the possibility of combining metal- and ligand-induced selectivity with the unique reactivity and functional-group tolerance of radical intermediates makes the development of metal-catalyzed radical processes attractive. This chapter describes the development of a novel methodology involving catalytic remote C–

H chlorination of alkyl hydroperoxides. The reaction is an internal redox reaction requiring no external oxidant and utilizing simple amine hydrochloride salts as the chlorine source.

Metal-mediated atom transfer is an important elementary step that has been developed in catalytic methods for the synthesis of small molecules7-11 and polymers.12-16 Extending this concept to C–H functionalization17-20 requires a

33 catalytically competent means of abstracting an unactivated hydrogen atom to generate the radical intermediate necessary for metal-mediated atom-transfer functionalization. We envisioned that alkyl hydroperoxides could be suitable substrates for this process, since metal-mediated reduction of the hydroperoxide moiety would serve two purposes by enabling intramolecular C–H activation and serving as an internal reoxidant, permitting catalytic turnover after the reductive atom transfer functionalization step (Scheme 3.1).

Scheme 3.1 Postulated copper-catalyzed chlorination of alkyl hydroperoxides

Activation of O–X bonds for remote functionalization has a significant history.21 Lead(IV) reagents react with alcohol O–H bonds to give ether products following distal hydrogen abstraction,22-25 and alkyl nitrites are classical intermediates for photolytic processes, including the functionalization of steroidal methyl groups.26-33 Metal-mediated halogenation of hydroperoxides has been demonstrated with both a stoichiometric metal oxidant and a stoichiometric metal reductant,34-39 but to our knowledge, no metal-catalyzed versions of remote C–H abstraction have been reported prior to our work.

34

3.2. Development of copper-catalyzed C-H chlorination

There are various ways by which alkoxy radicals can be generated.40 For our system, we examined alkyl-hydroperoxides as substrates that can generate a reactive alkoxy radical in the presence of transition metal catalysts. The alkoxy radical formed can potentially perform an intramolecular functionalization at a non-acti ated δ carbon (see Scheme 3.1). The alkyl hydroperoxide substrates can be synthesized from the corresponding alcohols.41-45 They can be purified by silica gel chromatography and are stable for several months in a freezer.

3.2.1. Substrate synthesis

Different substrates were synthesized from commercially available alcohols utilizing protocols obtained from literature based upon their structure.34,41,43,44

Scheme 3.4 represents the different methods employed for the synthesis of various alkyl hydroperoxides. All products were characterized with the help of 1H NMR.

For the synthesis of primary alkyl-hydroperoxides,41 the parent alcohols were mesylated, and then reacted with alkaline hydrogen peroxide, to yield the product. In case of secondary alkyl-hydroperoxides synthesis, initially we used low temperature oxidation of Grignard solution of the corresponding alkyl magnesium chloride.43 An alternative method44 was later employed to provide improved yields. This method comprised of synthesizing the parent hydrazide (a better leaving group than the corresponding mesylate) which on treatment with alkaline hydrogen peroxide formed the desired product. The tertiary alkyl

35 hydroperoxides were synthesized by acid catalyzed reaction of the parent alcohol with hydrogen peroxide.34 For several other hydroperoxides, containing a variety of functional groups, modified procedures were employed (see section 1.4

Experimental, for details).

Scheme 3.2 Synthetic routes of various alkyl hydroperoxides (Synthesis of alkyl hydroperoxides from the corresponding alcohols.)34,41,43,44

36

3.2.2. Preliminary assumptions

For the role of the catalyst, we needed something that performs the function of alkoxy radical generation as well as functionalization of the subsequently produced carbon radical. The catalyst also needs to get regenerated in the course of the reaction. The generation of alkoxy radicals from alkyl hydroperoxide can be carried out by reduction with transition metal salts or by thermal or photochemical decomposition.34,36,37,46

We also looked at the atom-transfer radical polymerization methodology where transition metal catalysts act as halogenation agents in a reversible redox reaction15 (shown in Scheme 3.2). Here a transition metal catalyst abstracts a readily removable group (X) and gets oxidized by one unit, forming a carbon radical which in the presence of a suitable monomer undergoes chain propagation.

The catalyst gets regenerated by introducing the leaving group X into the final product. The regenerated catalyst can promote a new redox process.

Scheme 3.3 Mechanism of Atom Transfer Radical Polymerization.

37

With these considerations in mind, CuICl seemed to be a good starting point for our catalyst optimization. It can act as a one-electron reductant in this process and can cause in situ generation of alkoxy radical. It can proceed with atom transfer functionalization and in the process, can be regenerated back to its original oxidation state to further continue its role as a catalyst. Also, a suitable ligand is pivotal for the formation of a ligated complex with the catalyst, which can facilitate successful interaction of the catalyst with the substrate.

Figure 3.1 A plausible catalytic cycle.1

We hypothesized a mechanism1 where the ligated CuICl catalyst abstracts hydroxyl radical from the substrate alkyl hydroperoxide forming an alkoxy radical

38

(Figure 3.1). In this process the metal catalyst is raised from an oxidation state of

+1 to a higher oxidation state of +2. Alkoxy radicals can act as efficient electrophiles40,47 and can easily abstract hydrogen atom from a non-activated C-H bond. Thus the alkoxy radical formed takes up a hydrogen atom from the δ carbon forming the corresponding carbon centered radical. The 1,5-hydrogen atom abstraction occurs selectively here, because of the favorable formation of a stable

6-membered transition state instead of 5-membered or 7-membered transition state.48

This is very similar to what Barton first reported in 196129 regarding alkyl nitrites. The ligated [CuIIClOH] catalyst transfers a Cl∙ to the δ carbon radical thus formed, giving rise to a chlorinated alcohol as the final product. This mechanism generates a ligated CuIOH. In order to regenerate the catalyst a suitable chlorinating source along with acid is required. The regenerated catalyst can then react with another substrate molecule to yield the product.

3.2.3. Substrate-dependent optimization of copper-catalyzed C-H chlorination

There can be undesirable side reactions29 as the alkoxy radicals generally undergo disproportionation to form the corresponding alcohol and carbonyl compound as shown in equation 3.1; therefore there is a need to optimize the reaction conditions.

2 RCH2-O· → RCH=O + RCH2-OH (3.1)

39

In order to optimize the system, a number of parameters, ranging from solvent system, substrate concentration, water input, acid concentration, type of acid, ligand, catalyst loading, temperature, chlorinating source etc., were screened, and the ideal conditions resulting in an optimum product yield and minimal side- products were identified.

As the mechanism of the reaction suggests the presence of a paramagnetic metal ion like CuII, the progress of the reaction could not be monitored by 1H

NMR. Therefore, HPLC was chosen as the primary tool for monitoring the progress of the reactions. As only the carbonyl side-product is UV sensitive, the refractive index detector (RID) is used for monitoring rest of the components, i.e. the chlorinated alcohol, the side-product alcohol, and the starting material alkyl hydroperoxide. HPLC allows us to observe the formation of new products, estimate their yields and separate them for further analysis. Hence the screening of various conditions can be carried out at ease without the need for work up.

We chose the primary alkyl hydroperoxide 1a as our first test substrate.

Among the potential catalysts, we found that copper (I) complexes with chelating nitrogen ligands are capable catalysts for the formation of chloride 2a from the hydroperoxide 1a, together with the byproducts 3a and 4a. The combination of

CuCl with N,N,N′,N′′,N′′-pentamethyldiethylenetriamine (PMDTA) in the presence of acetic acid and tetrabutylammonium chloride provided 16% of the alkyl chloride product (Table 3.1, entry 3). Since, the reaction pathway was assumed to go via radical mechanism, the undesired side products could result

40 from the intermolecular radical disproportionation as discussed earlier in equation

1. In order to minimize this kind of intermolecular reaction and increase the probability of intramolecular processes like the 1,5- hydrogen abstraction, dilution of the substrate concentration in the solvent medium seemed to be a good option. It was found that on dilution of the reaction mixture from 0.3 M to 0.06 M the yield of chlorinated alcohol increased to 28% (Table 3.1, entry 4), though variation of the solvent was not productive. A brief survey of other sp3 nitrogen ligands did not improve the reaction efficiency (Table 3.1, entries 5–7).

Table 3.1 Optimization of remote C–H functionalization of alkyl-hydroperoxides.

entry ligand chlorine source comments yield (%)a 2 3 4

1 bipyridine Bu4NCl 0.3 M 0 - -

2 - Bu4NCl 0.3 M 0 - -

3 PMDTA Bu4NCl 0.3 M 16 26 18

4 PMDTA Bu4NCl 0.06 M 28 11 24

5 TREN Bu4NCl - 28 22 10

6 Me6TREN Bu4NCl - 16 6 39

7 CYCLAM Bu4NCl - 12 5 5 b b 8 PMDTA Bu4NCl anhydrous 13 6 20 b 9 PMDTA Bu4NCl 1 vol % H2O 24 5 33 i 10 PMDTA Pr2NH·HCl 1 vol % H2O 28 4 29 c d 11 PMDTA NH4Cl 24 h addn 41 8 7

1 vol % H2O

41 a b c Yields determined by HPLC, Anhydrous Bu4NCl was used instead of the hydrate, N-decyl hydroperoxide was added over 24 h, d Isolated yields, e The amount of AcOH was increased from 4 equiv to 10 equiv.

Variable and irreproducible yields led us to examine the effect of water on reaction efficiency. Perhaps surprisingly, rigorous exclusion of water was detrimental to reaction yield (Table 3.1, entry 8). The commercial chlorination source Bu4NCl is quite hygroscopic and always contained some amount of moisture. On using anhydrous Bu4NCl only 13% of the chlorinated-alcohol was obtained. The controlled addition of water provided reproducibility and optimal results were achieved by adding 1 vol % water to the solvent along with the anhydrous chlorinating agent. Also, only a trace amount of product was detected when the reaction was carried out in air. Other chlorination sources were screened and the use of NH4Cl along with syringe pump addition of the hydroperoxide substrate over 24 h improved the yield to 41% (Table 3.1, entry 11), but we have been unable to improve this result with primary hydroperoxide substrates.

Moving to a secondary hydroperoxide 1b, we found that the formation of alcohol (2-octanol) and carbonyl (2-octanone) byproducts was significantly decreased. Screening various types of acid for the reaction did not improve the product yield (Table 3.2, entries 4 and 6-8). However after a brief optimization, we obtained the chloride 2b in 65% yield (Table 3.2, entry 9). The optimal method thus varies depending upon the type of chorination source used. For primary alkyl hydroperoxide substrates, which were highly reactive, NH4Cl was found to give

42 better yield (Table 3.1, entry 11), whereas for secondary alkyl hydroperoxide substrates, diisopropylamine hydrochloride gave the best yields (Table 3.2, entries

4 and 9). A control experiment with substrate 1b in the absence of ligand (Table

3.2, entry 10) exhibited similar rates of starting material conversion but inferior product formation, indicating that the ligand plays an important role in determining the chemoselectivity of the catalytic cycle.

Table 3.2 Optimization for secondary alkyl hydroperoxide (2-octyl hydroperoxide)

entry Acid chlorine source comments yield (%)a 2 3 4

1 AcOH Bu4NCl 0.06 M 51 6 43

2 AcOH Bu4NCl 25 mol% CuCl 43 18 39 b 3 AcOH Bu4NCl 1 vol % H2O 58 5 21 i 4 AcOH Pr2NH·HCl 1 vol % H2O 61 4 19

5 AcOH NH4Cl 1 vol % H2O 50 8 21 i 6 Pivalic acid Pr2NH·HCl 1 vol % H2O 50 8 42 i 7 Formic acid Pr2NH·HCl 1 vol % H2O 18 19 57 i 8 Oxalic acid Pr2NH·HCl 1 vol % H2O 55 5 40 i d 9 AcOH Pr2NH·HCl freeze-pump-thaw 77(65 ) 3 20 i 10 AcOH Pr2NH·HCl No ligand 25 26 35 a b c Yields determined by HPLC, Anhydrous Bu4NCl was used instead of the hydrate, N-decyl hydroperoxide was added over 24 h, d Isolated yields, e The amount of AcOH was increased from 4 equiv to 10 equiv.

43

In the case of the tertiary alkyl hydroperoxide 1c (Table 3.3), the reaction was very sluggish. Even after 10 days, the starting material remained unreacted

(Table 3.3, entry 1). We hypothesized that copper(I) catalyst was being oxidized to an inactive copper(II) species over long reaction times and so examined reductants that might regenerate an active copper(I) species. Addition of a catalytic amount of sodium ascorbate as a reducing agent remarkably improves the rate of the reaction, giving 74% isolated yield of the product 2c within 1 h.

Table 3.3 Optimizations for tertiary alkyl hydroperoxide (2-methyl, 2-hexyl hydroperoxide) as substrate.

entry Reaction time Reducing agent yield (%)a 2 3 4

1 10 d - 75 2 - 2 1 h Hydroquinone 55 14 - 3 17 h Hydroquinone 68 15 - 4 1 h Sodium ascorbate 74 b 10 - a Yields determined by HPLC, b Isolated yields

3.2.4. Substrate-scope of copper-catalyzed C-H chlorination

We examined chlorination of a variety of alkyl hydroperoxide substrates using the standard methods developed above. Surprisingly, reaction efficiency is similar for the functionalization of secondary or tertiary C–H bonds, despite the

44 significant difference in stability of the intermediate radical species (Table 3.4, entries 1 and 4, and entries 2 and 6). For substrates that target benzylic C–H bonds, similar yields are observed despite the increased reactivity of the C–H bond, and the product cyclizes into the corresponding furan under the reaction conditions (Table 3.4, entries 5, 7, 10). Ester groups are also tolerated under the reaction conditions (Table 3.4, entries 8, 9, 11).

Table 3.4 CuCl catalyzed remote C-H functionalization of alkyl hydroperoxides

entry substrate product yield (%)a

1b 1a 41

2c 1b 65

3c 1cd 74

4b 1d 32e

(47)f

5b 1e 58 (1.8: 1)g

45

6c 1f 46e

(68)f

7c 1g 54 g (1:1.8)

8b 1h 40

9c 1i 55

10c 1jd 74

11c 1k 65

12c 1l 33

a b c Isolated yields, NH4Cl used as chorination source, slow addition of substrate over 24 h. i d Pr2NH·HCl used as chlorination source, slow addition of substrate over 1 h. 0.05 equiv sodium ascorbate was added. e Lower isolated yields were observed in these cases due to product volatility . f Yields determined by HPLC, g Ratio, chlorinated product:cyclized product.

Since a new stereocenter is created in the process of C–H activation, the investigation of the effect of a neighboring stereogenic center on diastereoselectivity of the reaction was of interest. To address this issue, a substrate with a methoxy group adjacent to the target C–H bond was designed

46

(Scheme 3.3). In this case, no substantial diastereoselectivity was observed, and the product was isolated as a mixture of diasteromers (1.3:1). In addition to the expected 1,5 H-atom abstraction product 2m, two other products, 6m and 7m, of 1,6 abstraction of a C-H bond on the methoxy group were also formed.

Scheme 3.4 Chlorination of a substrate providing mixtures of 1,5 and 1,6 hydrogen abstraction. a Reaction conditions: 0.1 equiv CuCl, 0.12 equiv PMDTA, 4 equiv AcOH, 1.2 equiv diisopropylamine hydrochloride, 0.06 M hydroperoxide in MeCN, 35 ºC, slow addition of substrate.

3.3. Conclusions

The remote functionalization described here is characterized by operational simplicity. Because the alkyl hydroperoxide substrates are synthesized by substitution reactions with hydrogen peroxide, the inexpensive and environmentally benign hydrogen peroxide is the terminal oxidant for C–H functionalization. We believe this report will serve as a foundation for the

47 development of catalytic stereoselective processes and the further development of metal-catalyzed radical processes triggered by C–H activation.

3.4. Experimental

3.4.1. General Considerations

All reactions were carried out in oven dried glassware unless otherwise mentioned. THF and ether were freshly distilled from sodium benzophenone ketyl under nitrogen. DCM, triethylamine, and diisopropylamine were freshly distilled over calcium hydride. All other solvents were reagent grade and obtained from

Fisher Chemicals. NMR solvents CDCl3, C6D6, CD3CN (Cambridge Isotope

Laboratories) were used as received. Flash Chromatography was performed on 40-

63 μm particle size silica gel. All the alcohols for hydropero ide synthesis were purchased from Acros Organics except 2-octanol (Sigma) and were used as received. The following chemicals were purchased and used as received: 2- undecanone (Acros), 1,2-epoxy octane (Acros), ethyl octanoate (Acros), 6- oxoheptanoic acid (Aldrich), trans-3-octene-2-one (Aldrich-Kosher), MeMgBr

(1.4 M soln in toluene/THF, Acros), n-butyllithium (1.6 M soln in THF, Acros), tert-butyl(chloro)dimethylsilane (Gelest), imidazole (Sigma), triethylamine

i trihydrofluoride (Aldrich), CuCl (Strem), Pr2NH·HCl (TCI America), NH4Cl

(Fisher), PMDTA (Aldrich), methanesulfonyl chloride (Acros), thionyl chloride

(Aldrich), pyridine (Fisher), hydrazine monohydrate (Fisher), hydrogen peroxide

(30 wt % soln in water, and 50 wt % soln in water, Fisher), sulfuric acid (Fisher),

48 hydrochloric acid (Fisher), acetonitrile (Fisher), acetic acid (EM), MgSO4 (Fisher),

NaOH (Fisher), KOH (Fisher).

HPLC: HPLC was performed on a Shimadzu CBM-20A instrument with

Phenomene Jupiter 4 μ Proteo 9 A (25 × 15) and Phenomene Jupiter 4 μ

Proteo 90A (250 × 4.6) columns. Flow rates of 8 mL/min and 1 mL/min were used for the preparatory and analytical columns respectively. A refractive index detector was employed. Volumes of 15 μL of the reaction aliquot were used for sample injection. An isocratic eluent of 55:45 MeCN/H2O mixture was used for separation of different compounds in the aliquot.

GC-MS: GC-MS was conducted on an Agilent 5973N MSD interfaced to an Agilent 6890N GC System equipped with an Rtx-35 MS column (30 m × 0.25 mm in diameter, .1 μm film thickness; Restek, Bellefonte, PA). The sample was prepared with 1-mg/mL concentration of substrate in methanol. The major signals are quoted in m/z with the corresponding ions in parentheses. The methods used start with the injection temperature T0; after holding this temperature for 2 min, the column is heated to temperature T1 (ramp) and this temperature is held for an additional time t:

Method 50_220: T0 = 50 °C, T1 = 220 °C, ramp = 10 °C/min, t = 5 min;

The GC-MS data of alkyl hydroperoxides is not recorded due to decomposition of these compounds into corresponding alcohol and ketone under

GC-MS conditions. It was also observed that benzylic chlorides cyclizes rapidly

49 into the furan analogues under GC-MS conditions, so only the GC-MS data for the cyclized product is recorded for these substrates.

FTIR: IR spectra were recorded using a Nicolet Avatar 320 FT-IR spectrometer on KBr pellets.

NMR: 1H and 13C spectra were recorded on Bruker 5 UltraSield™ (5

MHz) spectrometer or O ford (4 MHz) spectrometer. The chemical shifts (δ) are reported in units of part per million (ppm) relative to TMS internal standard

1 13 (for H) or CDCl3 solvent (for C).

All the compounds reported have >95% purity unless otherwise mentioned.

Compounds 1a,42 1d,36,42 1j,45 2-octyl chloride49 and 5-phenyl-2-pentanone50 were synthesized according to previously published protocols.

3.4.2. Synthesis of Alkyl-hydroperoxides

2-Octyl hydroperoxide (1b)

2-octyl hydroperoxide was prepared by low-temperature oxidation of the corresponding 2-octylmagnesium chloride according to procedure reported for synthesis of analogous compounds.42,43 2-Octyl chloride was synthesized from 2- octanol (6.7 g, 51 mmol) using procedure adapted from the literature.49 A Grignard solution was prepared from magnesium (1.2 g, 50 mmol), 2-octyl chloride (3.6 g,

24 mmol) and distilled ether (12 mL). In another round bottom flask, distilled ether (75 mL) was cooled to -78 ºC. Oxygen was bubbled through this ether

50 solution for 30 min. The Grignard solution was slowly added under the surface of this cooled solution of ether with continuous and simultaneous bubbling of oxygen. Oxygen was bubbled for an additional 30 min after the addition of

Grignard solution. After stopping the oxygen flow, a solution of concentrated sulfuric acid (20 mL) and water (25 mL) was slowly dropped into the stirring solution. The organic phase was separated and washed with water, sodium- bicarbonate (sat. aq soln), and again with water. After drying over magnesium sulfate, excess solvent was removed under vacuum. The product was further purified by flash column chromatography (eluent: 8:1 hexanes/ ethyl acetate), giving 1.6 g (21%) of 2-octyl hydroperoxide over two steps. NMR data is consistent with that reported in literature.51

2-Methyl-2-undecyl hydroperoxide (1c)

Synthesis of 2-methyl-2-undecanol: A solution of methylmagnesium bromide (29.4 mmol) in dry ether (21 mL) was cooled to -78 ºC. To this solution, a mixture of 2-undecanone (5 g, 29.4 mmol) in dry ether (11 mL) was added dropwise while stirring. The reaction mixture was allowed to warm to rt. The progress of the reaction was monitored by thin layer chromatography and all the starting material was consumed in 1 h. The reaction was quenched with 2 M hydrochloric acid and the organic layer was extracted with ether and washed with a saturated solution of sodium bicarbonate, followed by brine. The product was

51 further purified by column chromatography (eluent: 8:1 hexanes/ ethyl actetate),

1 affording 5 g (92%) pure product. H NMR (400 MHz, CDCl3) δ 1.51-1.24 (m,

13 16H), 1.21 (s, 6H), 0.88 (t, 3H, J = 7.0 Hz). C NMR (400 MHz, CDCl3) δ 71.3,

44.2, 32.1, 30.4, 29.9, 29.8, 29.6, 29.4 (2C), 24.6, 22.9, 14.3. IR (thin film): 3362,

-1 2962, 2927, 2854, 1468, 1378, 1152 cm . GC-MS tR 7.9 min, (> 95%) m/z : [M-

+ CH3] calcd for C11H23O : 171.3, found : 171.2.

Synthesis of 2-methyl-2-undecyl hydroperoxide (1c): A solution of 96% sulfuric acid (0.7 mL, 13 mmol) in water (0.5 g, 26 mmol) was prepared and cooled to -10 ºC, hydrogen peroxide (4 mL, 35.42 mmol, 30 wt % aq soln) was added. To this cold mixture, a solution of the tertiary alcohol 2-methyl-2- undecanol (0.6 g, 3.22 mmol) in dioxane (3 mL) was added dropwise. After this, the reaction was heated at 60 ºC for 24 h and the progress of the reaction was monitored by thin layer chromatography. After the starting material was consumed, the reaction was quenched with a saturated solution of potassium carbonate. The organic layer was separated and washed with sodium bicarbonate

(sat. aq soln), followed by brine and water. The product was further purified by flash column chromatography (eluent: 8:1 hexanes/ ethyl actetate), affording 0.4 g

1 (58%) pure product. H NMR (400 MHz, CDCl3) δ 7.17 (br s, 1H), 1.62-1.24 (m,

13 16H), 1.22 (s, 6H), 0.88 (t, 3H, J = 7.0 Hz). C NMR (400 MHz, CDCl3) δ 83.2,

38.7, 32.1, 30.4, 29.84, 29.80, 29.6, 24.2, 24.0 (2C), 22.9, 14.3. IR (thin film):

3403, 2925, 2854, 1467, 1379, 1365 cm-1.

52

4-Phenyl-butyl-hydroperoxide (1e)

Synthesis of 4-phenyl-1-butyl-methanesulfonate: 4-phenyl-1-butyl- methanesulfonate was prepared following the previously reported procedure for synthesis of analogous compounds.41 4-Phenyl-1-butanol (2.0 g, 13 mmol) was placed in a round bottom flask with a stir bar under nitrogen and cooled in an ice bath. Methanesulfonyl chloride (1.5 g, 13 mmol) was added to the flask, followed by pyridine (2.0 g, 26 mmol). The reaction was quenched with 10% hydrochloric acid (20 mL) after 20 min. The organic layer was extracted with ether and washed with sodium bicarbonate (2×15 mL, sat. aq soln), followed by brine and water.

After drying over magnesium sulfate, excess solvent was removed under vacuum giving 3.0 g (>99%) of 4-phenyl-1-butyl-methanesulfonate. The crude product was used directly in the subsequent synthetic step without purification. 1H NMR

(400 MHz, CDCl3) δ 7.33-7.12 (m, 5H), 4.23 (t, 2H, J = 6.2 Hz ), 2.98 (s, 3H),

13 2.66 (t, 2H, J = 7.2 Hz ), 1.86-1.66 (m, 4H). C NMR (500 MHz, CDCl3) δ 141.7,

128.64 (2C), 128.60 (2C), 126.2, 70.0, 37.6, 35.4, 28.8, 27.4. IR (thin film): 3027,

-1 2940, 2862, 1354, 1175, 973, 935, 750, 701 cm . GC-MS tR 13.2 min (> 95%)),

+ m/z : [M] calcd for C11H16O3S : 228.1 found : 228.1

Synthesis of 4-phenyl-butyl hydroperoxide (1e): 4-phenyl-butyl hydroperoxide was prepared from 4-phenyl-1-butyl-methanesulfonate by treatment with an alkaline solution of hydrogen peroxide according to the

53 previously reported procedure for synthesis of analogous compounds.41 A solution of 4-phenyl-1-butyl-methanesulfonate (3.0 g, 13 mmol) in methanol (40 mL) and water (3 mL) was prepared and cooled to 0 ºC. This was followed by the addition of hydrogen peroxide (6 mL, 30 wt% aq soln) and potassium hydroxide (1.6 g,

50% aq soln). The reaction was allowed to warm to rt and stirred for 1 d. The reaction mixture was cooled to 0 ºC and potassium hydroxide (4 g, 50% aq soln) was added. The basic solution was then extracted three times with hexane. The aqueous layer was neutralized with 2 M hydrochloric acid solution and extracted three times with toluene. The organic layer was washed with water and brine.

After drying over magnesium sulfate, excess solvent was removed under vacuum.

The product was further purified by flash column chromatography (eluent: 8:1 hexanes/ ethyl actetate), affording 1.1 g (49%) pure product. 1H NMR (400 MHz,

CDCl3) δ 7.83 (s, 1H), 7.32-7.14 (m, 5H), 4.04 (t, 2H, J = 6.1 Hz ), 2.65 (t, 2H, J =

13 7.1 Hz ), 1.77-1.62 (m, 4H). C NMR (500 MHz, CDCl3) δ 142.3, 128.6 (2C),

128.5 (2C), 126.0, 77.1, 35.8, 27.8, 27.3. IR (thin film): 3399, 3026, 2941, 2861,

1496, 1453, 1376, 748, 699 cm-1.

5-Methyl-2-hexyl hydroperoxide (1f)

Previously reported procedure from literature44 for the synthesis of analogous compounds was modified and employed for the synthesis of 5-methyl-

2-hexyl hydroperoxide. Following the procedure41 described earlier for the

54 synthesis of 4-phenyl-1-butyl-methanesulfonate, 5-methyl-2-hexanol (0.68 g, 5.9 mmol) was converted into 5-methyl-2-hexyl-methanesulfonate (1.0 g, 88%). A reaction mixture of 5-methyl-2-hexyl-mesylate (1.0 g, 5.2 mmol) along with hydrazine monhydrate (8.50 g, 170 mmol) and methanol (8 mL) was prepared and kept in a sealed flask. This was heated at 90 ºC in an oil bath for 1.5 h. The progress of the reaction was monitored by NMR. When all the starting material was consumed, the reaction mixture was extracted with ether (2 × 20 mL) and then washed with potassium hydroxide (50 wt % aq soln), followed by washing with water. The organic phase was dried over magnesium sulfate and excess solvent was removed under vacuum. The crude product was immediately used for the next step without any purification. Partial characterization data: 1H NMR (400 MHz,

CDCl3) δ 2.60 (m, 1H), 1.63-1.06 (m, 5H), 1.10 (d, 3H, J = 6.3 Hz), 0.89 (d, 6H, J

= 6.6 Hz). A solution of the synthesized hydrazine (0.66 g, 5.0 mmol) was prepared in THF (60 mL). To this solution sodium hydroxide (0.75 g, 19.0 mmol) was added, followed by hydrogen peroxide (90 mL, 30% wt aq soln). The reaction mixture was stirred at rt for 3 d while its progress was monitored by thin layer chromatography. The reaction mixture was extracted with dichloromethane, and organic phase was washed with sodium chloride (sat. aq soln), followed by water.

After drying over magnesium sulfate, excess solvent was removed under vacuum.

The product was further purified by column chromatography over silica (eluent:

9:1 hexanes/ ethyl actetate), affording 0.4 g (50%) product over three steps.

Characterization data matches with that reported in literature.43

55

5-Phenyl-2-pentyl hydroperoxide (1g)

Synthesis of 5-phenyl-2-pentanol: Sodium borohydride (131 mg, 3.47 mmol) was added to a solution of 5-phenyl-2-pentanone50 (0.51 g, 3.15 mmol) in dry methanol (5.1 mL) kept in an ice bath under nitrogen. The reaction was stirred for 2 h and quenched with saturated ammonium chloride solution. The organic layer was extracted thrice with ether and dried over magnesium sulfate. The excess solvent was evaporated under vacuum giving 0.51 g (> 99%) of 5-phenyl-

2-pentanol. The crude product was used directly in the subsequent synthetic step without purification. NMR data is consistent with data reported in literature.52

Synthesis of 5-phenyl-2-pentyl hydroperoxide (1g): Following the procedure described earlier for the synthesis of 1f, 5-phenyl-2-pentanol (0.51 g, 31 mmol) was converted to 5-phenyl-2-pentyl-methanesulfonate (0.75 g, > 99%) and used in the next step without further purification. 5-phenyl-2-pentyl- methanesulfonate (0.75 g, 3.09 mmol) was then converted to the corresponding hydrazine, which was used to synthesize 5-phenyl, 2-pentyl hydroperoxide. The reaction mixture was extracted with ether and washed with brine and water. After drying over magnesium sulfate, excess solvent was removed under vacuum. The product was further purified by flash column chromatography (eluent: 9:1 hexanes/ ethyl actetate), affording 0.32 g (57% yield over three steps) of product.

1 H NMR (400 MHz, CDCl3) δ 7.76 (s, 1H), 7.40-7.12 (m, 5H), 4.08 (m, 1H), 2.62

56

(t, 3H, J = 7.4 Hz), 1.79-1.39 (m, 4H), 1.21 (d, 3H, J = 6.0 Hz). 13C NMR (400

MHz, CDCl3) δ 142.4, 128.6 (2C), 128.5 (2C), 126.0, 81.7, 36.0, 33.7, 27.3, 18.3.

IR (thin film): 3393, 2936, 1496, 1453, 1375, 749, 699 cm-1.

Ethyl-2-hydroperoxy-octanoate (1h)

Ethyl-2-hydroperoxy-octanoate was prepared by direct low temperature oxygenation of ester enolate anion using a procedure53 reported earlier for the synthesis of analogous compounds. Anhydrous THF (100 mL) and diisopropylamine (2.8 mL, 20 mmol) were added in a dry flask and cooled to -75

ºC. n-Butyllithium (20 mmol, THF, 1.6 M) and ethyl octanoate (3.45 g, 20 mmol) in THF (50 mL) were added sequentially while the solution was maintained at -75

ºC, then stirred for an additional 15 min to complete enolate anion formation.

HMPA (3.58 mL, 20 mmol) was added, stirring was continued for 15 min, and oxygen was bubbled into the solution for 1 h. The solution was acidified with 2 M hydrochloric acid and extracted with ether. The organic layer was dried over magnesium sulfate and excess solvent was evaporated under vacuum. Further purification was achieved by column chromatography (eluent: 15:1 hexanes/ ethyl

1 actetate), affording 0.63 g (18%) of product. H NMR (400 MHz, CDCl3) δ 9.25

(s, 1H), 4.52 (m, 1H), 4.26 (m, 2H), 1.76 (m, 2H), 1.5-1.23 (m, 11H), 0.88 (t, 3H,

13 J = 7.0 Hz). C NMR (400 MHz, CDCl3) δ 172.7, 83.8, 61.5, 31.7, 30.2, 29.1,

57

25.4, 22.7, 14.4, 14.2. IR (thin film): 3419, 2930, 2860, 1733, 1467, 1378, 1208,

1024 cm-1.

2-Hydroperoxy-octyl-acetate (1i)

Synthesis of 2-hydroperoxy-octanol: 1,2-epoxy-octane (1.0 g, 7.8 mmol) was treated with ethereal hydrogen peroxide (20 mL) in presence of catalytic amount of PMA (commercially available phosphomolybdic acid hydrate, 0.12 g,

0.78 mmol) using previously reported procedure54 for the synthesis of analogous compounds. The reaction mixture was stirred at ambient temperature for 6 h.

Diethylether (20 mL) was added to the reaction mixture followed by water (5 mL).

The phases were separated and the aqueous layer was back-extracted with ethyl acetate (3×20 mL). The combined organic phases were washed in turn with water

(2×10 mL) and brine (2×10 mL) before being dried over anhydrous magnesium sulfate. Filtration and evaporation of excess solvent under vacuum left a crude residue, which was purified by flash column chromatography (eluent:

15:1hexanes/ ethyl actetate), affording 0.57 g (45%) of product. 1H NMR (400

MHz, CDCl3) δ 8.34 (s, 1H), 4. 5 (m, 1H), 3.85 (dd, 1H, J1 = 12.1 Hz, J2 = 2.7

Hz), 3.71(dd, 1H, J1 = 11.8 Hz, J2 = 6.3 Hz), 1.71-1.19 (m, 10H), 0.88 (t, 3H, J =

13 6.9 Hz). C NMR (400 MHz, CDCl3) δ 86.2, 64.2, 31.9, 29.5, 28.8, 25.8, 22.8,

14.3. IR (thin film): 3353, 2955, 2858, 1467, 1040 cm-1.

58

A solution of 2-hydroperoxy-octanol (0.30 g, 1.85 mmol) in dry DCM (19 mL) was prepared and treated with tert-butyl(chloro)dimethylsilane (0.28 g, 1.85 mmol) and imidazole (0.126 mg, 1.85 mmol). The reaction mixture was stirred at rt for 2 h. The reaction mixture was diluted with DCM and washed with potassium hydrogen sulfate (1×, sat. aq soln), followed by water. After drying over magnesium sulfate, excess solvent was evaporated under vacuum. The crude product (0.51 g) obtained from this reaction was used in the subsequent synthetic step without purification.

A solution of the crude product silyl peroxy ether (0.51 g) from the previous step was prepared in DCM (20 mL) and treated with acetic anhydride

(0.19 mL) in presence of catalytic amount of DMAP. The reaction was stirred at rt for 1h. The reaction mixture was extracted with DCM and the organic layer was washed with potassium hydrogen sulphate (1× 10 mL, sat. aq soln), sodium bicarbonate (1× 10 mL, sat. aq soln) and sodium chloride (1× 10 mL, sat. aq soln).

After drying over magnesium sulfate, excess solvent was evaporated under vacuum. The crude product (0.57 g) obtained from this reaction was used in the subsequent synthetic step without purification.

Synthesis of 2-hydroperoxy-octyl-acetate (1i): The resulting acetate (0.57 g) was taken in THF (9 mL) and the solution was cooled to 0 ºC. To this solution, triethylamine trihydrofluoride (0.58 mL) was added. The reaction was stirred at rt for 1 h and quenched with sodium bicarbonate (sat. aq soln). The organic layer was extracted with ether and washed with potassium hydrogen sulfate (sat. aq

59 soln) and brine. After drying over magnesium sulfate, excess solvent was evaporated under vacuum. Further purification was achieved by flash column chromatography (eluent: 15:1 hexanes/ ethyl acetate), giving 0.106 g (28%) of

1 pure 2-hydroperoxyoctyl acetate 1i over three steps. H NMR (400 MHz, CDCl3) δ

9.32(s, 1H), 4.52 (dd, 1H, J1= 12.5, J2= 2.7 Hz), 4.11 (dd, 1H, J1 = 12.6 Hz, J2 =

4.8 Hz), 3.99 (m, 1H), 2.15 (s, 3H), 1.68-1.24 (m, 10H), 0.89 (t, 3H, J = 7.3 Hz).

13 C NMR (400 MHz, CDCl3) δ 173.3, 83.5, 63.0, 31.9, 29.4, 28.8, 25.8, 22.8, 21.2,

14.3. IR (thin film): 3380, 2956, 2929, 1741, 1721, 1460, 1384, 1240, 1042 cm-1.

Ethyl-6-methyl-6-hydroperoxy-heptanoate (1k)

Synthesis of ethyl-6-oxo-heptanoate: Thionyl chloride (3.3 g, 28 mmol) was added dropwise to a solution of 6-oxo-heptanoic acid (2.0 g, 14 mmol) in ethanol (93 mL). The reaction was complete within 1 h, after which excess solvent was evaporated under vacuum and the organic residue was extracted with ether and washed with sodium bicarbonate. The organic phase was dried over anhydrous magnesium sulfate. After filtration and evaporation of excess solvent under vacuum, 2.3 g (95%) of product was obtained and was used in the subsequent

1 synthetic step without further purification. H NMR (400 MHz, CDCl3) δ 4.13 (q,

2H, J = 7.2 Hz), 2.46 (t, 2H, J = 6.8 Hz), 2.32 (t, 2H, J = 6.8 Hz), 2.15 (s, 3H),

13 1.7-1.5 (m, 4H), 1.26 (t, 3H, J = 6.9 Hz). C NMR (400 MHz, CDCl3) δ 208.9,

173.6, 60.5, 43.4, 34.2, 30.0, 24.5, 23.3, 14.4. IR (thin film): 2941, 1734, 1373,

60

-1 + 1181 cm . GC-MS. tR 8.4 min, (> 95%) m/z : [M] calcd for C9H16O3: 172.2, found : 172.1

Synthesis of ethyl-6-methyl-6-hydroxy-heptanoate: Following the procedure described for the synthesis of 2-methyl-2-undecanol, ethyl-6-oxo- heptanoate (1 g, 6 mmol) was treated with methylmagnesium bromide (6 mmol) to give 0.9 g (82%) of the product. The product was used in the next synthetic step

1 without further purification. H NMR (500 MHz, CDCl3) δ 4.13 (q, 2H, J = 7.3

Hz), 2.32 (t, 2H, J = 7.5 Hz), 1.64 (m, 2H), 1.52-1.34 (m, 5H), 1.26 (t, 3H, J = 7.3

13 Hz), 1.21 (s, 6H). C NMR (500 MHz, CDCl3) δ 173.9, 71. , 6 .4, 43.7, 34.5,

29.4 (2C), 25.6, 24.1, 14.4. IR (thin film): 3434, 2969, 2939, 1735, 1374, 1246,

-1 + 1182, 1121 cm . GC-MS. tR 8.5 min, (> 95%) m/z : [M-CH3] calcd for C9H17O3:

173.2, found : 173.2.

Synthesis of ethyl-6-methyl-6-hydroperoxy-heptanoate (1k): The resulting tertiary alcohol ethyl-6-methyl-6-hydroxy-heptanoate (0.9 g, 4.7 mmol) was treated with 85% phosphoric acid (2.4 mL) and hydrogen peroxide (3.6 mL,

50 wt % aq soln) at 0 ºC. After the addition, the reaction mixture was warmed to

35 ºC. The reaction was complete within 2 h and was quenched with saturated solution of potassium carbonate. The organic layer was separated and washed with a saturated solution of sodium bicarbonate, followed by brine and water. The product was further purified by flash column chromatography (eluent: 7:1

1 hexanes/ ethyl acetate), giving 0.72 g (75%) product. H NMR (400 MHz, CDCl3)

δ 7.51 (s, 1H), 4.14 (q, 2H, J = 7.2 Hz), 2.33 (t, 2H, J = 7.6 Hz), 1.70-1.55 (m,

61

6H), 1.5-1.3 (m, 2H), 1.26 (t, 3H, J = 7.0 Hz), 1.21 (s, 6H). 13C NMR (400 MHz,

CDCl3) δ 174.1, 82.7, 6 .5, 37.7, 34.2, 25.4, 24.1 (2C), 23.2, 14.4. IR (thin film):

3416, 2982, 2942, 2869, 1733, 1467, 1373, 1252, 1187, 1124, 1030 cm-1.

4-Hydroperoxy heptane (1l)

A solution of 4-heptanone (1.0 g, 8.8 mmol) in ethanol (18 mL) was treated with tosylhydrazine (1.8 g, 9.6 mmol). The reaction was complete within 1 h, after which excess solvent was evaporated under vacuum giving 2.8 g (> 95%) of the corresponding tosylhydrazone. A solution of this compound (2.8 g, 9.9 mmol) in

THF (150 mL) was then treated with borane dimethylsulfide (5.6 mL, 2 M in

THF) at 0 ºC. The progress of the reaction was monitored by NMR. The reaction was quenched with methanol and excess solvent was evaporated under vacuum.

The crude product was immediately used for the next step without any purification. A solution of the synthesized hydrazine (3.43 g, 12.0 mmol) was prepared in THF (125 mL). To this solution sodium hydroxide (0.96 g, 24.0 mmol) was added, followed by hydrogen peroxide (190 mL, 30% wt aq soln). The reaction mixture was stirred at rt for 3 d while its progress was monitored by thin layer chromatography. The reaction mixture was extracted with dichloromethane, and organic phase was washed with sodium chloride (sat. aq soln), followed by water. After drying over magnesium sulfate, excess solvent was removed under vacuum. The product was further purified by column chromatography over silica

62

(eluent: 9:1 hexanes/ ethyl actetate), affording 0.5 g (45%) product over three

43 steps. Characterization data matches with that reported in literature.

4-Methoxy-2-methyl-2-octyl hydroperoxide (1m)

Synthesis of 4-methoxy-2-octanone: 4-Methoxy-2-octanone was synthesized according to a procedure55 reported earlier for the synthesis of analogous compounds. A solution of 3-octen-2-one (5.0 g, 40 mmol) in methanol

(40 mL) was prepared and subjected to three cycles of freeze-pump-thaw.

Catalytic amount of triethylphosphine (2 mmol) was added to the solution. The reaction was stirred overnight and filtered through a pad of silica and concentrated under vacuum. The product was further purified by flash column chromatography

(eluent: 9:1/ 6:1 hexanes/ ethyl acetate gradient) to give 3.34 g (53%) product and

1 was used for the subsequent synthetic step. H NMR (400 MHz, CDCl3) δ 3.67

(m, 1H), 3.32 (s, 3H), 2.67 (dd, 1H, J1 = 16.1 Hz, J2 = 7.7 Hz), 2.46 (dd, 1H, J1 =

15.9 Hz, J2 = 4.9 Hz), 2.18 (s, 3H), 1.56-1.19 (m, 6H), 0.90 (t, 3H, J = 6.9 Hz).

NMR data is consistent with data reported in literature.56

Synthesis of 4-methoxy-2-methyl-2-octanol: Following the procedure described for the synthesis of 2-methyl-2-undecanol, 4-methoxy-2-octanone (2.0 g, 13 mmol) was treated with methylmagnesium bromide (19 mmol) to give 2.1 g

(94%) of the product, which was used in the next synthetic step without further

1 purification. H NMR (400 MHz, CDCl3) δ 4.04 (s, 1H), 3.57 (m, 1H), 3.36 (s,

63

3H), 1.78-1.26 (m, 8H), 1.21 (s, 3H), 1.25 (s, 3H), 0.92 (t, 3H, J = 7.1 Hz). 13C

NMR (400 MHz, CDCl3) δ 79.6, 70.4, 55.8, 45.9, 32.7, 31.3, 28.4, 26.9, 23.1,

14.2. IR (thin film): 3446, 2965, 2933, 2874, 1464, 1394, 1377, 1201, 1147, 1084,

-1 + 913, 743, 668 cm . GC-MS. tR 6.5 min, (> 95%) m/z : [M-CH3] calcd for

C9H19O2: 159.2, found : 159.2

Synthesis of 4-methoxy-2-methyl-2-octyl hydroperoxide (1m): 4-

Methoxy-2-methyl-2-octanol (1.0 g, 5.7 mmol) was treated with 85% phosphoric acid (2.9 mL) and hydrogen peroxide (4.3 mL, 50 wt % aq soln) at 0 ºC. After the addition, the reaction mixture was heated at 50 ºC for 12 h and was worked up as mentioned earlier. Further purification was achieved by flash column chromatography (eluent: 6:1 hexanes/ ethyl acetate), giving 0.78 g (77%) pure

1 product. H NMR (400 MHz, CDCl3) δ 9.96 (s, 1H), 3.44 (m, 1H), 3.41 (s, 3H),

2.04 (m, 1H), 1.64-1.17 (m, 10H), 1.13 (s, 3H), 0.92 (t, 3H, J = 7.2 Hz). 13C NMR

(400 MHz, CDCl3) δ 80.9, 78.4, 56.4, 42.5, 32.7, 26.7, 26.2, 23.8, 22.8, 14.0. IR

(thin film): 3310, 2983, 2874, 2828, 1468, 1379, 1364, 1201, 1079 cm-1.

3.4.3. Copper-Catalyzed C-H chlorination of Alkyl hydroperoxides:

3.4.3.1. Representative procedure for copper-catalyzed remote C-H chlorination for primary alkyl hydroperoxides (conversion of 1a to 2a)

4-Chloro-decanol (2a):

64

NH4Cl (36.6 mg, 1.2 equiv) and CuCl (5.64 mg, 0.1 equiv) were weighed in a vial. A magnetic stir-bar was placed in the vial, which was then closed with a septum and placed under nitrogen. Degassed acetonitrile (5 mL) was added to the mi ture, followed by water (95 μL), PMDTA (14 μL, .12 equi ), and acetic acid

(13 μL, 4 equi ). The colorless or faintly blue solution was warmed to 35 ºC.

Meanwhile, the substrate n-decyl hydroperoxide (100 mg, 0.57 mmol, 1 equiv) was diluted with acetonitrile (4.5 mL) and was added to the reaction mixture with the help of a syringe pump at the flow rate of 0.14 mL/h. The color of the reaction mixture changes from colorless (or pale yellow/ blue) to bright blue with addition of substrate. The progress of the reaction was monitored using HPLC. After 1 d, the reaction mixture was filtered through silica, the silica pad was washed with ether multiple times, and excess solvent was evaporated under vacuum. The residue was further purified by flash column chromatography over silica (eluent:

9:1 hexanes/ ethyl acetate) to give 47 mg (41%) of product. 1H NMR (400 MHz,

CDCl3) δ 3.93 (m, 1H), 3.69 (t, 2H, J = 6.1 Hz), 2.16-1.34 (m, 8H), 1.93-1.22 (m,

13 14H), 0.89 (t, 3H, J = 7.1 Hz). C NMR (500 MHz, CDCl3) δ 64.2, 62.6, 38.8,

35.0, 31.9, 29.9, 29.0, 26.7, 22.8, 14.3. IR (thin film): 3344, 2930, 2858, 1457,

1377, 1058 cm-1. NMR data is consistent with data reported in literature.57 GC-MS

+ tR 9.7 min, (87%), m/z : [M- C2H6O] calcd for C10H21OCl : 146.2, found : 146.1.

+ Peak m/z : 146.1 was found along with another peak m/z: [(M+2)-C2H6O] : 148.1 in the intensity ratio of 3:1 exhibiting characteristic isotopic pattern for 35Cl and

37Cl and thus confirming the presence of chlorine in the compound.

65

3.4.3.2. Representative procedure for copper-catalyzed remote C-H chlorination for secondary alkyl hydroperoxides (conversion of 1b to 2b)

5-Chloro-2-octanol (2b):

Diisopropylamine hydrochloride (94.15 mg, 1.2 equiv) and CuCl (5.64 mg,

0.1 equiv) were weighed in a vial. A magnetic stir-bar was placed in the vial, which was then closed with a septum and placed under nitrogen. Meanwhile, a solution of acetic acid (325 μL, 1 equi ) and water (95 μL) was prepared and subjected to three cycles of freeze-pump-thaw. To the mixture of diisopropylamine hydrochloride and CuCl, degassed acetonitrile (5 mL), PMDTA

(14 μL, .12 equi ), and acetic acid-water mi ture (42 μL) was added. The colorless or faintly blue solution was warmed to 35 ºC. Meanwhile the substrate 2- octyl hydroperoxide (83 mg, 0.57 mmol, 1 equiv) was diluted with acetonitrile

(4.5 mL) and was added to the reaction mixture with the help of a syringe pump at the flow rate of 4.5 mL/h. The color of the reaction mixture changes from colorless (or pale yellow/ blue) to bright blue with addition of substrate. The progress of the reaction was monitored using HPLC. After 1 d, the reaction mixture was filtered through silica, the silica pad was washed with ether multiple times and excess solvent was evaporated under vacuum. The residue was further purified by flash column chromatography (eluent: 9:1 hexanes/ ethyl acetate) to

1 give 61 mg (65%) of product. H NMR (400 MHz, CDCl3) δ 3.94 (m, 1H), 3.83

66

(m, 1H), 1.32-2.07 (m, 9H), 1.22 (d, 3H, J = 6.2 Hz), 0.93 (t, 3H, J = 7.3 Hz). 13C

NMR (400 MHz, CDCl3) δ 67.9, 67.4, 64.1, 63.7, 40.7, 40.6, 36.2, 35.7, 34.9,

34.4, 23.70, 23.67, 19.7, 19.6, 13.5 (2C). IR (thin film): 3373, 2962, 2933, 2874,

1459, 1377, 1114, 954, 667, 608 cm-1. The diastereomeric mixture was not

+ 35 resolved by GC-MS. tR 6.8 min, (> 95%), m/z : [M-CH3] calcd for C8H17O Cl :

+ 37 149.2, found : 149.1, m/z : [(M+2)-CH3] calcd for C8H17O Cl : 151.2, found :

151.1

3.4.3.3. Representative procedure for copper-catalyzed remote C-H chlorination for tertiary alkyl hydroperoxides (conversion of 1c to 2c)

5-Chloro-2-methyl-2-undecanol (2c):

Diisopropylamine hydrochloride (94.15 mg, 1.2 equiv), sodium ascorbate

(5.64 mg, 0.05 equiv) and CuCl (5.64 mg, 0.1 equiv) were weighed in a vial. A magnetic stir-bar was placed in the vial, which was then closed with a septum and placed under nitrogen. Degassed acetonitrile (5 mL) was added to the mixture, followed by water (95 μL), PMDTA (14 μL, .12 equi ), and acetic acid (13 μL,

4 equiv). The colorless or faintly blue solution was warmed to 35 ºC. Meanwhile the substrate 2-methyl-2-undecanol hydroperoxide (115 mg, 0.57 mmol, 1 equiv) was diluted with acetonitrile (4.5 mL) and was added to the reaction mixture with the help of a syringe pump at the flow rate of 4.5 mL/h. The color of the reaction

67 mixture changes from colorless (or pale yellow/ blue) to bright blue with addition of substrate. The progress of the reaction was monitored using HPLC. After 1 d, the reaction mixture was filtered through silica, the silica pad was washed with ether, and excess solvent was evaporated under vacuum. The residue was further purified by flash column chromatography (eluent: 9:1 hexanes/ ethyl acetate) to

1 give 93 mg (74%) of product. H NMR (500 MHz, CDCl3) δ 3.90 (m, 1H), 1.95-

1.25 (m, 14H), 1.24 (s, 3H), 1.23 (s, 3H), 0.88 (t, 3H, J = 7.3 Hz). 13C NMR (500

MHz, CDCl3) δ 7 .6, 64.7, 4 .4, 38.6, 33.2, 31.7, 29.7, 29.1, 28.8, 26.4, 22.6,

14.0. IR (thin film): 3374, 2957, 2930, 2858, 1467, 1378, 1206, 1140, 930, 907,

-1 + 35 725 cm . GC-MS tR 10.1 min, (93%), m/z : [M-CH3] calcd for C11H22O Cl :

+ 37 205.3, found : 205.2. m/z : [(M+2)-CH3] calcd for C11H22O Cl: 207.2, found :

207.1

4-Chloro-4-methyl-pentanol (2d):

Following the procedure described for the synthesis of 2a, 4-methyl-pentyl- hydroperoxide 1b (67.4 mg, 0.57 mmol) was converted into the desired product

(25 mg, 32%). Chromatography eluent: 8:1 hexanes/ ethyl acetate. 1H NMR (400

13 MHz, CDCl3) δ 3.69 (t, 2H, J = 6.1 Hz), 1.92-1.70 (m, 4H), 1.60 (s, 6H). C

NMR (500 MHz, CDCl3) δ 71. , 63.1, 42.4, 32.7 (2C), 28.7. NMR data was found to be consistent with data reported in literature.57 IR (thin film): 3352, 2953, 2931,

-1 2875, 2360, 1456, 1387, 1370, 1117, 1060, 1023, 668 cm . GC-MS tR 5.5 min,

68

+ 35 (91%), m/z : [M-C3H8O] calcd for C3H5 Cl : 77.2, found 77.1. m/z : [(M+2)-

+ 37 C3H8O] calcd for C3H5 Cl: 79.2, found : 79.1

4-Chloro-4-phenyl-1-butanol (2e) and 2-phenyltetrahydrofuran (5e)

Following the procedure described for the synthesis of 2a, 4-phenyl-butyl hydroperoxide 1e (94.7 mg, 0.57 mmol) was converted into a mixture of 4-chloro-

4-phenyl-1-butanol 2e and 2-phenyltetrahydrofuran 5e in a 1.8:1 ratio. For characterization purposes, the individual components were obtained through careful chromatography on the mixture, affording samples of each compound contaminated with small amount of the other product: 2e (39 mg, 37%, and 92% pure), 5e (18 mg, 21%, and 94% pure). The mixture of these compounds was separated using flash column chromatography over silica (eluent: 8:1 hexanes/ ethyl acetate).

Characterization data for 2e: 1H NMR (500 MHz, CDCl3) δ 7.43-7.27

(m, 5H), 4.90 (dd, 1 H, J1 = 8.2 Hz, J2= 6.2 Hz), 3.68 (dt, 2H, J1 = 6.4 Hz, J2 = 1.9

Hz), 2.27-2.10 (m, 2H), 1.85-1.54 (m, 2H), 1.40 (s, 1H). 13C NMR (500 MHz,

CDCl3) δ 141.9, 128.9(2H), 128.5, 127.1 (2H), 63.8, 62.4, 36.7, 3 .4. IR (thin film): 3352, 2948, 2875, 1454, 1057, 1030, 754, 697 cm-1. 2c cyclizes partially to

+ 3c under GC-MS conditions. GC-MS tR 10.8 min, (70%), m/z : [M] calcd for

35 + 37 C10H13O Cl : 184.2, found : 184.1. m/z : [(M+2)] calcd for C10H13O Cl: 207.2, found : 207.1

69

Characterization data for 5e: 1H NMR (500 MHz, CDCl3) δ 7.39-7.16

(m, 5H), 4.89 (t, 1H, J = 7.2 Hz), 4.10 (m, 1H), 3.94 (m, 1H), 2.32 (m, 1H), 2.08-

1.94 (m, 2H), 1.81 (m, 1H). 13C NMR (5 MHz, CDCl3) δ 143.7, 128.5 (2H),

127.3, 125.8 (2C), 80.9, 68.9, 34.8, 26.2. NMR data was found to be consistent with data reported in literature.13 IR (thin film): 2969, 2868, 1452, 1080, 1060,

-1 + 754, 699 cm . GC-MS tR 8.0 min, (> 95%), m/z: [M] calcd for C10H12O : 148.2, found : 148.1

5-Chloro-5-methyl-2-hexanol (2f):

Following the procedure described for the synthesis of 2b, 5-methyl-2- hexyl hydroperoxide 1f (120 mg, 0.91 mmol) was converted into 5-chloro-5- methyl-2-hexanol 2f (63 mg, 46%). Chromatography eluent: 8:1 hexanes/ ethyl

1 acetate. H NMR (400 MHz, CDCl3) δ 3.80 (m, 1H), 1.95-1.60 (m, 4H), 1.57 (s,

13 3H), 1.56 (s, 3H), 1.23 (d, 3H, J = 3.6 Hz). C NMR (400 MHz, CDCl3) δ 70.9,

68.1, 42.1, 34.7, 32.7, 32.3, 23.7. IR (thin film): 3356, 2940, 2929, 1458, 1389,

-1 + 1371, 1119, 1077 cm . GC-MS tR 5.9 min, (88%), m/z : [M-CH5O] calcd for

35 + 37 C6H10 Cl : 117.2, found : 117.1. m/z : [(M+2)-CH5O] calcd for C6H10 Cl: 119.2, found : 119.1

70

5-Chloro-5-phenyl-2-pentanol (2g) and 2-methyl-5-phenyltetrahydrofuran (5g)

5-Phenyl-2-pentyl hydroperoxide 1g (95.8 mg, 0.48 mmol) was subjected to the same reaction conditions described for the synthesis of 2b, giving 5-chloro-

5-phenyl-2-pentanol 2g (20 mg, 19%) and 2-phenyltetrahydrofuran 5g (30 mg,

35%). Chromatography eluent: 10:1 hexanes/ ethyl acetate.

1 Characterization data for 2g: H NMR (400 MHz, CDCl3) δ 7.43-7.14

(m, 5H), 4.89 (m, 1H), 3.83 (m, 1H), 2.34-2.02 (m, 2H), 1.73-1.39 (m, 2H), 1.21-

13 1.18 (2 sets of d, 3H). C NMR (400 MHz, CDCl3) δ 141.9 (2C), 128.8 (4C),

128.5 (2C), 127.14 (2C), 127.11 (2C), 67.9, 67.6, 64.2, 63.9, 36.8, 36.63, 36.59,

36.4, 23.95, 23.91. IR (thin film): 3371, 2966, 2929, 2870, 1495, 1454, 1375,

1130, 1086, 1013, 957, 937, 758, 798 cm-1. The compound cyclizes completely, to the corresponding furan 3f under GC-MS conditions.

1 Characterization data for 5g: H NMR (400 MHz, CDCl3) δ 7.37-7.20

(m, 10 H), 5.04 (m, 1H), 4.88 (m, 1H), 4.35 (m, 1H), 4.17 (m, 1H), 2.47-1.56 (m,

8H), 1.37(d, 3H, J = 6.1 Hz), 1.32 (d, 3H, J = 5.8 Hz). 13C NMR (400 MHz,

CDCl3) δ 144.2, 143.7, 128.5 (2C), 128.46 (2C), 127.3, 127.2, 126.1 (2C), 125.8

(2C), 81.3, 80.5, 76.2, 76.1, 35.8, 34.9, 34.5, 33.3, 21.8, 21.6. IR (thin film): 2968,

-1 1085, 1069, 753, 699 cm . The diastereomeric mixture was resolved in GC-MS. tR

+ 8.3 min (43%), 8.4 min (50%), m/z : [M] calcd for C11H14O : 162.2, found : 162.1.

71

Ethyl-5-chloro-2-hydroxy-octanoate (2h)

The substrate ethyl-2-hydroperoxy-octanoate was found to be very reactive and hence the reaction conditions for the synthesis of 2a were employed here.

Ethyl-2-hydroperoxy-octanoate (116 mg, 0.57 mmol) was converted to the desired product (51 mg, 40%). Chromatography eluent: 15:1 hexanes/ ethyl acetate. 1H

NMR (500 MHz, CDCl3) δ 4.26 (m, 2H), 4.19 (m, 1H), 3.94 (m, 1H), 2.16-1.34

(m, 8H), 1.31 (t, 3H, J = 7.2 Hz), 0.93 (t, 3H, J = 7.1 Hz). 13C NMR (500 MHz,

CDCl3) δ = 175.29, 175.25, 70.4, 69.9, 63.7, 63.4, 62.1 (2C), 40.9, 40.8, 34.2,

33.6, 31.7, 31.4, 19.91, 19.86, 14.4 (2C), 13.8 (2C). IR (thin film): 3466, 2960,

2934, 2874, 1734, 1466, 1369, 1261, 1215, 1096, 1023 cm-1. The diastereomeric mixture decomposes under GC-MS conditions.

5-Chloro-1,2-octanediol-l-acetate (2i)

Following the procedure described for the synthesis of 2b, 2-hydroperoxy- octyl-acetate 1i (104 mg, 0.47 mmol) was converted into 5-chloro-1,2-octanediol-

1-acetate 2i (62 mg, 56%). Chromatography eluent: 9:1 hexanes/ ethyl acetate. 1H

NMR (500 MHz, CDCl3) δ 4.16 (m, 1H), 4. -3.8 (m, 3H), 2.11 (s, 3H), 1.93-1.38

13 (m, 8H), 0.93 (t, 3H, J = 7.34 Hz). C NMR (500 MHz, CDCl3) δ 171.4 (2C),

70.2, 69.5, 68.99, 68.94, 64.1, 63.5, 41.0, 40.8, 34.9, 34.2, 21.1 (2C), 19.98, 19.91,

72

13.8 (2C). IR (thin film): 3458, 2959, 2874, 1741, 1458, 1369, 1241, 1042 cm-1.

The diastereomeric mixture decomposes under GC-MS conditions.

5-Chloro-2-methyl -5-phenyl-2-pentanol (2j) and 2,2-dimethyl-5- phenyltetrahydrofuran (5j)

5-phenyl-2-methyl-2-pentyl hydroperoxide 1j (110.7 mg, 0.57 mmol) was subjected to same reaction conditions as described for the synthesis of 2c. The reaction was allowed to run for 6 d and 2, 2-dimethyl-5-phenyltetrahydrofuran 5j

(74 mg, 74%) was obtained as the only product. Chromatography eluent: 9:1

1 hexanes/ ethyl acetate. H NMR (400 MHz, CDCl3) δ 7.41- 7.19 (m, 5H), 4.98 (m,

1H), 2.38-2.24 (m, 1H), 1.97-1.82 (m, 3H), 1.39 (s, 3H), 1.35 (s, 3H). 13C NMR

(400 MHz, CDCl3) δ 143.8, 128.5 (2C), 127.3, 126 (2C), 81.5, 80.7, 39.3, 35.9,

29.3, 28.6. NMR data was found to be consistent with data reported in literature.58

IR (thin film): 2969, 2869, 1451, 1379, 1365, 1143, 1080, 1048, 1029, 858, 726,

-1 + 699 cm . GC-MS tR 8.5 min, (> 95%), m/z : [M] calcd for C12H17OCl : 176.14, found : 176.2

Ethyl-3-chloro-6-methyl-6-hydroxy-heptanoate (2k)

Following the procedure described for the synthesis of 2c, ethyl-6-methyl-

6-hydroperoxy-heptanoate 1k (116 mg, 0.57 mmol) was converted into desired

73 product 2k (80 mg, 65%) after a reaction time of 2 d. However, no sodium ascorbate was added to this reaction. Chromatography eluent: 8:1 hexanes/ ethyl

1 acetate. H NMR (500 MHz, CDCl3) δ 4.32 (m, 1H), 4.19 (q, 2H), 2.77 (m, 2H),

2.00-1.31 (m, 5H), 1.28 (t, 3H, J = 7.2 Hz), 1.24 (s, 3H), 1.23 (s, 3H). 13C NMR

(500 MHz, CDCl3) δ 170.4, 70.7, 61.1, 58.7, 43.9, 40.4, 33.1, 29.9, 29.4, 14.4. IR

(thin film): 3434, 2972, 2933, 1738, 1378, 1369, 1254, 1220, 1160, 1130, 1028

-1 + 35 cm . GC-MS tR 9.9 min, (93%), m/z : [M-CH3] calcd for C9H16O3 Cl : 207.2,

+ 37 found : 207.1. m/z : [(M+2)-CH3] calcd for C9H16O3 Cl: 209.2, found : 209.1

1-Chloro-4-heptanol (2l)

Following the procedure described for the synthesis of 2b, 4-hydroperoxy- heptane 1l (66 mg, 0.50 mmol) was converted into 1-chloro-4-heptanol 2l (14 mg,

33%). Chromatography eluent: 9:1 hexanes/ ethyl acetate. 1H NMR (500 MHz,

CDCl3) δ 3.65 (m, 1H), 3.59 (m, 2H), 2.11-1.78 (m, 2H), 1.72-1.24 (m, 6H), 0.94

13 (t, 3H, J = 6.7 Hz). C NMR (400 MHz, CDCl3) δ 71.3, 45.5, 4 .1, 34.8, 29.1,

19.0, 14.3. IR (thin film): 3365, 2958, 2931, 2872, 1459, 1308, 1125, 1083, 999,

-1 + 35 652 cm . GC-MS tR 6.9 min, (>95%), m/z : [M-C3H7] calcd for C4H8O Cl :

+ 37 107.2, found : 107.1. m/z : [(M+2)-C3H7] calcd for C4H8O Cl: 109.2, found :

109.1

74

5-Chloro-4-methoxy-2-methyl-2-octanol (2m), (2-hydroxy-2-methyloctan-4- yloxy)methyl acetate (6m), and 6-butyl-4,4-dimethyl-1,3-dioxane (7m)

Following the procedure described for the synthesis of 2c, 4-methoxy-2- methyl-2-octyl hydroperoxide 1m (108 mg, 0.57 mmol) was converted into a mixture of 2m (60 mg, 51%), 6m (18 mg, 14%) and 7m (19 mg, 19%).

Chromatography eluent: 20:1 hexanes/ ethyl acetate with 0.1 vol % triethylamine.

1 Characterization data for (2m): H NMR (400 MHz, CDCl3) δ 4.21-4.14

(m, 1H), 4.13-4.07 (m, 1H), 3.79-3.71 (m, 1H), 3.69-3.63 (m, 1H), 3.47 (s, 3H),

3.34 (s, 1H), 3.18 (s, 1H), 2.00-1.30 (m, 12H), 1.26 (s, 6H), 1.24 (s, 6H), 0.99-0.92

13 (m, 6H). C NMR (400 MHz, CDCl3) δ 82.3, 82.1, 70.2, 70.0, 63.1, 60.9, 57.5,

57.1, 41.8, 40.6, 36.3, 33.7, 31.10, 31.06, 29.0, 28.9, 20.5, 20.4, 13.8 (2C). IR (thin film): 3449, 2935, 2934, 2875, 1466, 1379, 1201, 1151, 1093 cm-1. The diastereomeric mixture was resolved in GC-MS. tR 8.2 min, (28%), 8.4 min,

+ 35 (65%), m/z : [M-CH3] calcd for C9H18O2 Cl : 193.2, found : 193.1, m/z : [(M+2)-

+ 37 CH3] calcd for C9H18O2 Cl: 195.2, found : 195.1

1 Characterization data for (6m): H NMR (500 MHz, C6D6) δ 5.11 (s,

2H), 3.86 (m, 1H), 1.69 (s, 3H), 1.43-1.08 (m, 8H), 1.19 (s, 3H), 0.85 (t, 3H, J =

13 6.9 Hz). C NMR (500 MHz, C6D6) δ 169.7, 86.3, 76.8, 69.4, 46.5, 34.5, 31.0,

28.8, 26.8, 22.9, 20.3, 14.0. IR (thin film): 3458, 2964, 2934, 2873, 1750, 1468,

75

-1 1367, 1228, 1153, 1103, 1011, 937 cm . GC-MS tR 9.8 min, (93%), m/z : [M-

+ CH2OAc] calcd for C9H19O2 : 159.2 found : 159.1

1 Characterization data for (7m): H NMR (500 MHz, CDCl3) δ 4.90 (dd,

2H, J1 = 25.0 Hz, J2 = 6.9 Hz), 3.66 (m, 1H), 1.57-1.31 (m, 8 H), 1.30 (s, 3H), 1.26

13 (s, 3H), 0.91 (t, 3H, J =7.2 Hz). C NMR (500 MHz, CDCl3) δ= 88.1, 72.9, 71.5,

42.7, 36.1, 31.9, 27.4, 22.9, 21.8, 14.3. IR (thin film): 2958, 2930, 2859, 1177,

-1 + 1088, 1039 cm . GC-MS tR 6.4 min, (91%), m/z : [M] calcd for C10H20O2 : 172.2 found : 172.1

3.5. References

(1) Kundu, R.; Ball, Z. T.: Copper-Cataly ed Remote sp3 C−H Chlorination of Al yl Hydroperoxides. Org. Lett. 2010, 12, 2460-2463. (2) Labinger, J. A.; Bercaw, J. E.: Understanding and exploiting C-H bond activation. Nature 2002, 417, 507-514. (3) Godula, K.; Sames, D.: C-H Bond Functionalization in Complex Organic Synthesis. Science 2006, 312, 67-72. (4) Bergman, R. G.: Organometallic chemistry: C-H activation. Nature 2007, 446, 391-393. (5) Goldman A. S.; Goldberg K. I.: Organometallic C-H Bond Activation: An Introduction. In Activation and Functionalization of C-H Bonds; American Chemical Society, 2004; Vol. 885; pp 1-43. (6) a -Reque o, M. M. P re , P. J.: Coinage Metal Cataly ed C−H Bond Functionalization of Hydrocarbons. Chem. Rev. 2008, 108, 3379-3394. (7) Fukuyama, T.; Nishitani, S.; Inouye, T.; Morimoto, K.; Ryu, I.: Org. Lett. 2006, 8, 1383–1386. . (8) Roncaglia, F.; Stevens, C. V.; Ghelfi, F.; Van der Steen, M.; Pattarozzi, M.; De Buyck, L.: Tetrahedron 2009, 65, 1481-1487. (9) Tenn, W. J.; Conley, B. L.; Hövelmann, C. H.; Ahlquist, M.; Nielsen, R. J.; Ess, D. H.; Oxgaard, J.; Bischof, S. M.; Goddard, W. A.; Periana, R. A.: J. Am. Chem. Soc. 2009, 131, 2466– 2468. (10) Nevárez, Z.; Woerpel, K. A.: Org. Lett. 2007, 9, 3773–3776. (11) Yang, D.; Yan, Y.; Zheng, B.; Gao, Q.; Zhu, N.: Org. Lett. 2006, 8, 5757–5760. (12) Matyjaszewski, K.: Chem. Eur. J. 1999, 5, 3095-3102. (13) Matyjaszewski, K.; Tsarevsky, N. V.: Nature Chem. 2009 1, 276-288. (14) Matyjaszewski, K.; Xia, J.: Chem. Rev. 2001, 101, 2921-2990.

76

(15) Wang, J.; Matyjaszewski, K.: J. Am. Chem. Soc. 1995, 117, 5614–5615. (16) Percec, V.; Guliashvili, T.; Ladislaw, J. S.; Wistrand, A.; Stjerndahl, A.; Sienkowska, M. J.; Monteiro, M. J.; Sahoo, S.: J. Am. Chem. Soc. 2006, 128, 14156-14165. (17) Crabtree, R. H.: Chem. Rev. 1995, 95, 987-1007. (18) Shilov, A. . Shul’pin, . B.: Chem. Rev. 1997, 97, 2879-2932. (19) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E.: Angew. Chem. Int. Ed. 1998, 37, 2181- 2192. (20) Labinger, J. A.; Bercaw, J. E.: Nature 2002, 417, 507-514. (21) Zhu, H.; Wickenden, J. G.; Campbell, N. E.; Leung, J. C. T.; Johnson, K. M.; Sammis, G. M.: Org. Lett. 2009, 11, 2019-2022. (22) Bowers, A.; Denot, E.; Ibanez, L. C.; Cabezas, M. E.; Ringold, H. J. J.: Org. Chem. 1962, 27, 1862-1867. (23) Mihailović, M. L. Če ović, Ž. Andre ević, V. Matić, R. Jeremić, .: Tetrahedron 1968, 24, 4947–4961. (24) Brun, P.; Pally, M.; Waegell, B.: Tetrahedron Lett. 1970, 331–334. (25) Brun, P.; Waegell, B.: Bull. Soc. Chim. Fr. 1972, 1825–1832. (26) Barton, D. H. R.: Pure Appl. Chem. 1968, 16, 1. (27) Barton, D. H. R.; Akhtar, M. J.: J. Am. Chem. Soc. 1962, 84, 1496-1497. (28) Barton, D. H. R.; Akhtar, M. J.: J. Am. Chem. Soc. 1964, 86, 1528-1536. (29) Barton, D. H. R.; Beaton, J. M.; Geller, L. E.; Pechet, M. M.: J. Am. Chem. Soc. 1961, 83, 4076-4083. (30) Barton, D. H. R.; Budhiraja, R. P.; McGhie, J. F.: J. Chem Soc., Part C 1969, 336- 338. (31) Batten, P. L.; Bentley, T. J.; Boar, R. B.; Draper, R. W.; McGhie, J. F.; Barton, D. H. R.: J. Chem. Soc., Perkin Trans. 1972, 1, 739-748. (32) Stotter, P. L.; Hill, K. A.; Friedman, M. D.: Heterocycles 1987, 25, 259. (33) Suginome, H.; Nakayama, Y.; Senboku, H.: J. Chem. Soc., Perkin Trans. 1 1992, 1837-1842 (34) Acott, B.; Beckwith, A. L. J.: Aust. J. Chem. 1964, 17, 1342-1353. (35) Če ović, Ž. Čvet ović, M.: Tetrahedron Lett. 1982, 23, 3791-3794. (36) Če ović, Ž. imitri ević, L. o ić, . Srnić, T.: Tetrahedron 1979, 35, 2021- 2026. (37) Če ović, Ž. reen, M. M.: J. Am. Chem. Soc. 1974, 96, 3000-3002. (38) Če ović, Ž. Ili ev, .: Tetrahedron Lett. 1988, 29, 1441-1444. (39) Petrović, . Če ović, Ž.: Tetrahedron 1999, 55, 1377-1390. (40) Će ović, Ž.: Reactions of δ-carbon Radicals Generated by 1,5-Hydrogen Transfer to Alkoxyl radicals. Tetrahedron 2003, 59, 8073-8090. (41) Williams, H. R.; Mosher, H. S.: J. Am. Chem. Soc. 1954, 76, 2984-2987. (42) Walling, C.; Buckler, S. A.: J. Am. Chem. Soc. 1955, 77, 6032-6038. (43) Jinsheng, L.; Pritzkow, W.; Voerckel, V.: J. Prakt. Chem. 1994, 336, 43-52. (44) Casteel, D. A.; Jung, K.: J. Chem.Soc. Perkin Trans. 1 1991, 2597-2598. (45) Kropf, H.; von Wallis, H.: Synthesis 1981, 8, 633-635. (46) Beckwith, A. L. J.; Hay, B. P.: Generation of Alkoxy radicals From N- alkoxypyridinethiones. J. Am. Chem. Soc. 1988, 110, 4415-4416. (47) Backwith, A. L. J. K.; Ingold, U.: Free Radical Rearrangements in Rearragements in Gound and Excited States, de Mayo, P., Ed.; Academic: New York 1980, 1, 161–309. (48) Robertson, J.; Pillai, J.; Lush, R. K.: Radical Translocation Reactions in Synthesis Chem. Soc. Rev. 2001, 30, 94-103.

77

(49) Cason, J.; Correia, J. S.: J. Org. Chem. 1961, 26, 3645-3649. (50) Montgomery, F. C.; Saunders, W. H.: J. Org. Chem. 1976, 41, 2368-2372. (51) Dussault, P. H.; Zope, U. R.; Westermeyer, T. A.: J. Org. Chem. 1994, 59, 8267- 8268. (52) Kropp, P. J.; Breton, G. W.; Craig, S. L.; Crawford, S. D.; Durland, W. F. J.; Jones, J. E. I.; Raleigh, S. J.: J. Org. Chem. 1995, 60, 4146-4152. (53) Konen, D. A.; Silbert, L. S.; Pfeffer, P. E.: J. Org. Chem. 1975, 40, 3253-3258. (54) Li, Y.; Hao, H.; Wu, Y.: Org. Lett. 2009, 11, 2691-2694. (55) Stewart, I. C.; Bergman, R. G.; Toste, F. D.: J. Am. Chem. Soc., 2003, 125, 8696- 8697. (56) Ramachary, D. B.; Mondal, R.: Tetrahedron Lett. 2006, 47, 7689-7693. (57) Kapustina, N. I.; Lisitsyn, A. V.; Nikishin, G. I.: Izv. Akad. Nauk SSSR, Ser. Khim. 1989, 1, 98-105. (58) Loh, T. P.; Hu, Q. Y.; Ma, L. T.: J. Am. Chem. Soc. 2001, 123, 2450-2451.

78

Chapter 4 Hybrid Organic-Inorganic inhibitors for PDZ Domain

Parts of this chapter have been adapted from:1

Kundu, R.; Cushing, P. R.; Popp, B. V.; Zhao, Y.; Madden, D. R.; Ball, Z. T., “Hybrid Organic–Inorganic Inhibitors of a PDZ Interaction that Regulates the Endocytic Fate of CFTR.” Angew. Chem. Int. Ed. 2012, 51 (29), 7217-7220.

4.1. Introduction

Protein-protein interactions (PPIs) play key roles in cellular processes and consequently in human disease, but often lack a compact, high-affinity pocket accessible to traditional ligand-discovery approaches. However, combining multiple weak interactions can be combined to yield polyvalent2-4 ligands that more closely mimics the complexity of natural macromolecular interactions and thus pro ides enhanced potency and specificity for such “undruggable” PPI targets.5 Initial studies validated the fragment-based ligand design approach6 for enzyme inhibition using tethering approaches,7,8 and the ideas have been extended to protein-protein interactions.9,10 In part, this approach involves increasing the potency of a weak ligand by adding groups that make “peripheral” contacts near the primary binding site. In this chapter, we have described the extension of the ideas of fragment-based PPI inhibition to hybrid structures utilizing cooperative organic-inorganic binding to the surface of a target protein (Scheme 4.1). In

79 particular, we describe the engineering of a stable rhodium(II) metallopeptide that displaces representative peptide ligands from the PDZ domain of the cystic fibrosis transmembrane conductance regulator (CFTR)-associated ligand (CAL).

PDZ domains are a family of peptide-binding PPI modules named for the first three members: PSD‒95, Dlg, and ZO‒1.

Scheme 4.1 Conceptual depiction of a hybrid organic-inorganic inhibitor stabilized by metal–histidine interactions near the interface

In an aqueous environment, metal-ligand interactions offer potentially dramatic increases in stability compared to non-covalent organic interactions, which are typically weak ( < 1 kcal mol ‒1). However, taking advantage of this possibility is challenging. Recruiting endogenous metal ions can stabilize a protein-inhibitor interface,11-14 but the low in vivo availability of transition-metal ions is a major limitation.15 Stable metal-based protein inhibitors have a significant history,16-18 often with substitution-inert metal centers.19,20 Although metal-ligand- conjugate therapeutics have been proposed,21 few stable inhibitors that display cooperative organic-inorganic binding have been reported, such as DNA-metal conjugates that serve as irreversible enzyme inhibitors.22,23

80

To successfully bind a protein target, a discrete organic-inorganic complex must contain a stable organic-metal linkage while allowing ligand exchange at the metal center in order to bind targeted side chains. Fortunately, di-metal

“pinwheel” structures, such as rhodium(II) tetracarbo ylate, ha e well differentiated ligand environments containing both kinetically inert, equatorial κ2- carboxylate ligands and kinetically labile axial ligand sites (e.g., Figure 4.1.a),24 with a demonstrated capability to bind biologically relevant thiol and imidazole compounds in a reversible manner.25,26 The interactions of rhodium(II) complexes with nucleic acids are well studied,27-30 and a few reports have described rhodium(II) interactions with proteins.31,32 Finally, rhodium(II) complexes can have low toxicity toward mammalian cells33 or living animals.34

Figure 4.1 Depiction of E3gH–K3a,eRh2 assembly a) Axial coordination in E3gH–K3a,eRh2 stabilizes the coiled coil. (b) Sequences used in this study. Lower-case grey letters represent positions on a helical-wheel depiction.

4.2. Dirhodium‒histidine interaction

(Thermal denaturation studies and UV–absorption studies in this section were performed by Dr. Brian V. Popp.)

81

From the previous work in our lab, we hypothesized that rhodium(II) centers could form stabilizing secondary contacts with a protein surface while examining the coiled-coil assembly of rhodium(II) metallopeptides with histidine- containing peptides (Figure 4.1).35,36 Based on established coiled-coil models,37 a rhodium(II) center linked to a coil at positions a and e of a heptad repeat sequence

(abcdefg, Table 4.1) would be proximal to position g of the complementary peptide, E3gX. It was found that coordination of appropriate side chains at position g strongly stabilizes the coiled coil. For example, thermal denaturation of a mixture of E3gH and K3a,eRh2 revealed a high melting temperature (Tm = 66.1 °C;

Table 4.1, entry 3, and Figure 4.2), in contrast to simple E3/K3 dimers37 and to control experiments with a mixture of K3a,eRh2 and a non-coordinating phenylalanine analogue E3gF (Tm = 39.5 °C; Table 4.1, entry 1). This coiled-coil stabilization reflects a specific interaction of the rhodium center, in this case with the histidine imidazole group.

Table 4.1 Thermal denaturation of metallopeptide coiled coils.[a] Entry sequence E3X Tm (°C)

1 E3gF 39.5

2 E3gE 50.2

3 E3gH 66.1 [b] 4 E3gH + 50 mM imidazole 46.0

5 E3gM >75 [c] 6 E3gM 70.4

7 E3gC 33.5

82

8 E3cH 47.0

[a] Standard condns: equimolar mixtures of E3X and K3a,eRh2 in aq buffer (pH 5.9–6.2) were monitored by CD (222 nm) from −5 to 95 °C at 1°C/min. All peptides 1 µM unless otherwise noted. [b] CD monitored at 225 nm. [c] Peptide concentration 33 µM.

When the histidine was moved away from the interface, to position c, a

drop in Tm to 47.0 °C was observed (Table 4.1, entry 8, and Figure 4.2). The

addition of large concentrations of imidazole, either before or after coiled-coil

assembly, also led to a significant drop in melting temperature to 46 °C (Table 4.1,

entry 4, and Figure 4.2), thereby providing evidence for a reversible metal-ligand

interaction.

Figure 4.2 Selection of thermal denaturation profiles for stoichiometric mixtures of E3H and K3a,eRh2. Vertical lines indicate melting temperature (Tm). See Table 1 for sample conditions and Tm values.

83

Finally, upon assembly with the E3gH coil, the metallopeptide K3a,eRh2 exhibits a blue shift of the UV-vis absorption peak at 587 nm to 567 nm (Figure

4.3), consistent with a rhodium(II) tetracarboxylate containing axial nitrogen or sulfur ligands.25,38,39

Figure 4.3 Visible absorption spectra of dirhodium-metallopeptide assemblies. (A) K3a,eRh2 [black trace], (B) K3a,eRh2–E3gH [red trace], (C) K3a,eRh2–E3gH [blue trace], (D) K3g,dRh2–E3gH [purple trace]. Sample preparation: 1:1 mixture of metallopeptide and E3-peptide with (A–C) aq buffer at pH 6.2 and (D) aq KOH at pH~7.0.

Other Lewis basic side chains also facilitate stabilization. Coiled-coil assemblies with either glutamate (E3gE) or methionine (E3gM) peptides also exhibit elevated Tm values (50.2 °C and >70 °C, respectively), consistent with carboxylate–rhodium or stronger thioether–rhodium interactions (Table 1, entries

3, 5–7). So far as we are aware, Tm values of 65–70 °C represent the most stable intermolecular coiled coils yet reported for such a short peptide (21 amino acids),40 similar to stabilities achieved with covalent crosslinking.41-43 Insertion of cysteine

84 at the same position, on the other hand, leads to a coiled coil with decreased stability (entry 8). Preliminary modeling indicates the cysteine side chain is too short to reach the rhodium center without disrupting the coiled coil.

4.3. Design of metallopeptide based inhibitors

To extend the concept of organic-inorganic cooperativity to the discovery of potent PPI inhibitors, we examined interactions between the CAL PDZ domain

(CALP) and the cystic fibrosis transmembrane conductance regulator (CFTR). The

C-terminal region of CFTR interacts with several proteins (e.g. CAL, NHERF1,

NHERF2) through PDZ interactions that affect CFTR function in diverse ways.44-

47 Despite its potential value as a target, inhibiting CALP is distinctly difficult due to its broad specificity and comparatively low baseline affinity.48 Recently, our collaborators at Dartmouth combined a screen of inverted peptide arrays with in vitro fluorescence polarization measurements to identify selective CALP

48-50 inhibitors. However, the potency of these inhibitors remains modest, with Ki ≥

1.3 M.

The CAL structure positions several histidine residues near the peptide- binding site in its PDZ domain, making it an attractive target for a hybrid organic– inorganic approach to inhibitor design (Figure 4.4).51

85

Figure 4.4 Structure of the CAL PDZ domain (orange ribbon) bound to a CFTR peptide (green stick figure)51 All CALP His side chains are shown explicitly (stick figures colored by element; grey = C, blue = N).

To test the potential contributions of rhodium-based interactions to CALP inhibitor affinity, we adapted known methods33,52 to prepare metallopeptides based on sequences known to interact with CALP.

Dirhodium metallopeptides can be syntheisized by direct metalation of the free peptide via carboxylate containing residues. However, for PDZ binding, it was necessary to synthesize metallopeptides with a free C-terminal carboxylate.

PDZ domains recognize C-terminal sequences, and affinity is greatly reduced for peptides with C-terminal amides. This requirement creates synthetic difficulties, because the metalation of carboxylate side chains with reactive dirhodium complexes is a kinetic phenomenon that is largely non-selective for poly- carboxylate-containing sequences.33,52 We considered various methods to

86 selectively protect specific carboxylate groups, but found it most convenient to directly metalate a peptide with both C-terminal and aspartate side-chain carboxylates and to separate the resulting mixture of metalated products (Error! eference source not found.4.5). We started from the known VQDTRL sequence.

Using established protocols,33,52 metalation of the fully deprotected peptide provided a roughly statistical mixture of the three possible products, which were separated using preparative HPLC to afford the desired metallopeptide,

VQDRhTRL. Other metallopeptides were produced similarly. After metalation and subsequent purification of the complexes for a series of peptides, a trend was established, where it was found that the side-chain metalated peptide is less polar than the terminal-metalated peptide and always elutes after the terminal-metalated peptide (Figure.4.6.a).

87

Figure 4.5 Synthesis of a PDZ-binding metallopeptide based on the C-terminal CFTR sequence. The desired side-chain metalated peptide containing free C-terminal carboxylate is highlighted. Yields are given for purified, isolated material. Other metallopeptides were synthesized similarly.

88

Figure 4.6 HPLC traces and ESI-MS of VQDTRL series a) HPLC traces of (A) unpurified metalation mixture of peptide VQDTRL (B) purified terminal-metalated peptide VQDTRLRh (C) purified sidechain-matalated peptide VQDRhTRL b) ESI- MS data of purified sidechain-matalated peptide for VQDRhTRL.

89

As both of the mono-metalated peptides are identical in mass, they could not be distinguished via mass spectrometry. So NMR was employed to distinguish between the two isomers (Figure 4.7).

Figure 4.7 NMR comparison of (A) VQDTRL peptide B) VQDTRLRh and C) VQDRhTRL. irhodium attachment results in changes to one leucine Hδ methyl (from 0.87 ppm to 0.65 ppm, Δδ1H ca. –0.22 ppm) in the C-terminal bound product VQDTRLRh (A vs B) and to the aspartate Hβ methylene (from 2.81 ppm to 2.69 ppm, Δδ1H ca. –0.12 ppm) in the side-chain-bound metallopeptide VQDRhTRL (A vs C).

90

The site of metalation was determined by 1H NMR spectroscopy on the basis of chemical shifts of protons proximal to the bound dirhodium. Dirhodium attachment resulted in discernible changes to one leucine Hδ methyl (from .87 ppm to .65 ppm ,Δδ1H ca. –0.22 ppm) in the C-terminal bound product,

VQDTRLRh, and to the aspartate Hβ methylene (from 2.81 ppm to 2.69 ppm, Δδ1H ca. –0.12 ppm) in the side-chain-bound metallopeptide, VQDRhTRL (Figure 4.7).

4.4. Screening the potency of metalated inhibitors for CALP

(The fluorescence polarization measurements and pull down assay was performed in collaboration with Dr. Patrick R. Cushing, Yu Zhao and Prof. Dear

R. Madden from Dartmouth Medical School.)

We measured inhibitor equilibrium dissociation constants (Ki), using fluorescence anisotropy to observe the displacement of a fluorescent reporter peptide (Figure 4.8 and Table 4.2). The CFTR sequence, VQDTRL, derived from the natural CALP-binding sequence of CFTR, has weak affinity for CALP. Direct incorporation of a rhodium(II) center at the aspartate side-chain carboxylate of the natural sequence (VQDRhTRL) resulted in a modest decrease in the apparent inhibitory constant (Ki), but only to a value comparable to that observed with a simple rhodium complex, Rh2(OAc)4 (Figure 4.8). The apparent affinity of

VQDRhTRL was also comparable to that of a metallopeptide derived from a non- binding, scrambled control sequence, QLDRhVTR. Together, these data suggest

91 that the effects seen with rhodium(II) addition at the P-3 site (P0 = C-terminal residue) of VQDTRL are not specific.

Figure 4.8 Fluorescence anisotropy displacement isotherms for candidate CALP inhibitors. Ki values are reported in Table 4.2.

Based on structural analysis of the CALP domain,51 we designed three metallopeptides ( Table 4.2, entries 6, 8, 10; ERhWPTSII, ERhVQSTRL,

ERhVQSTRI) having a site of rhodium attachment at the P‒6 position, because structural analysis of the CALP domain51 indicated that the P‒6 position should be proximal to the His301 residue in the target (Figure 4.4). All three new metallopeptides exhibited significant and reproducible affinity enhancement in relative to the non-metalated controls (Table 4.2, entries 5-10). When a sequence

(EWPTSII) carrying a glutamate side chain at the P-6 position, metalation increased binding affinity from 65 to 1.7 µM, an approximately 40-fold enhancement (Table 4.2, entries 5–6). An even larger enhancement was seen with

92

the sequence EVQSTRL, which contains the dominant C-terminal tetrapeptide

identified by array screens.53 In this case, the metalated peptide ERhVQSTRL

bound with the highest affinity yet reported for CALP (0.56 µM), a nearly 75-fold

change relative to the parent peptide (Table 4.2, entries 7–8). Recently, we

designed a new sequence EVQSTRI with an isoleucine residue instead of leucine

at P0 position, upon metalation an even higher binding affinity is observed, from

171 to 0.45 µM, an approximately 380 fold enhancement (Table 4.2, entries 9-10).

HSQC footprinting revealed in presence of either the peptide or metallopeptide

(see section 1.6 Experimentals) were well-dispersed, thus confirming that each

binds in the canonical PDZ binding site, with localized differences.

Table 4.2 Metallopeptide inhibitors of CALP.[a] entry peptide sequence CALP Ki (μM) H301 Ki (μM) 1 1 VQDTRL 320 ± 43 - Rh 2 1-Rh VQD TRL 6.3 ± 0.7 [b] -

3 2 QLDVTR >500 - Rh 4 2-Rh QLD VTR 10.0 ± 0.9 [b] -

5 3 EWPTSII 65 ± 5 - Rh 6 3-Rh E WPTSII 1.7 ± 0.2 [c] -

7 4 EVQSTRL 42 ± 5 89 ± 3 Rh 8 4-Rh E VQSTRL 0.56 ± 0.08[c] 11.3 ± 0.7

9* 5 EVQSTRI 171 ± 7 375 ± 67 Rh 10* 5-Rh E VQSTRI 0.45 ± 0.14[c] 12.3 ± 0.1

[a] Inhibitor equilibrium dissociation (Ki) constants were determined for cognate peptide/metallopeptide pairs. [b] All rhodium(II) complexes, including Rh2(OAc)4 (Ki =13 ± 5 μM) exhibit nonspecific inhibition, establishing an upper bound for these measurements. [c] Ki value significantly different from Rh2(OAc)4 (p<0.05, n=3). * unpublished data.

93

A CALP-H301A mutant was prepared to ascertain the role of His301 in metallopeptide affinity (Table 4.2, entries 7-10). This mutant binds the parent

EVQSTRL with Ki 88.9 µM, only slightly (~2-fold) weaker than the wild-type protein. However, the mutant binds the metallopeptide with a Ki of 11.3 µM, a

>25-fold loss of affinity relative to wild-type, consistent with the predicted

His301–rhodium ligation (Table 4.2, entries 7-8). The same trend is observed when the mutant binds to the sequence EVQSTRI with Ki of 375 µM, 2 fold loss of affinity with respect to the wild-type protein. Upon binding to the metallopeptide, a Ki of 12.3 µM was observed, a >25-fold loss of affinity relative to wild-type, consistent with the earlier data (Table 4.2, entries 9-10).

Figure 4.9 Metalation improves inhibitor potency (a) Western blot of native CAL captured from epithelial lysates, 50 in the presence of increasing concentrations of EVQSTRL peptide with (+Rh) or without (-Rh) rhodium. (b) Quantification reveals dose-dependent inhibition of CAL pulldown (PD) by the metalated peptide (n=3).

94

To provide an independent demonstration of rhodium-based affinity enhancement and to establish the efficacy of rhodium metallopeptides in a more complex environment, a pulldown inhibition assay was performed using epithelial cell lysate (Figure 4.9). Relative to the non-metalated control, the metallopeptide

ERhVQSTRL exhibits improved inhibition, demonstrating that the affinity gains carry over to a more physiological environment.

4.5. Developing dirhodium metallopeptide inhibitors for N1P1

(The fluorescence polarization measurements and pull down assay was performed in collaboration with Dr. Patrick R. Cushing, Yu Zhao and Prof. Dear

R. Madden from Dartmouth Medical School. This data is unpublished.)

NHERF1 (Na+/H+ Exchanger Regulator Factor 1) is a scaffolding protein that interacts with transmembrane receptors and transporters, coupling them to effectors and cytoskeletal proteins.54,55 It contains two PDZ (PSD-95, Dlg, ZO-1) domains that bind to the C-termini of its partners, interactions that can be mimicked by short linear peptides. The first PDZ domain of NHERF1 (N1P1) is a well-studied example, and reported structural work indicates the presence of two histidine residues near the ligand-binding groove, making N1P1 a good test for the development of dirhodium metallopeptide ligands.

95

Figure 4.10 Structure of N1P1 with a bound peptide ligand (based on PDB ID: 1GQ4).56 Histidine side chains proximal to the binding pocket have been highlighted in purple.

While developing inhibitors for CALP, we investigated N1P1 as a target too. The parent peptide VQDTRL had a very good binding affinity of Ki = 0.70

µM to begin with. Upon metalation we observed a modest 5 fold increment in binding affinity to 0.14 µM (Table 4.3, entries 1, 2). A further enhancement of binding affinity was observed with the hit sequence for CALP inhibition,

EVQSTRL. With the metalated inhibitor ERhVQSTRL, approximately >20 fold enhancement was observed (Table 4.3, entries 3, 4). However, the affinity of the inhibitors for N1P1, though approximately 30 fold higher than that for CALP, was still not enough for a selective inhibition of either of the proteins in presence of the other.

96

Table 4.3 Comparision of binding affinity of metallopeptide inhibitors for N1P1 and CALP.[a]*

entry peptide sequence N1P1 Ki (μM) CALP Ki (μM) 1 1 VQDTRL 0.70 320 ± 43 Rh 2 1-Rh VQD TRL 0.14 6.3 ± 0.7 [b]

3 4 EVQSTRL 0.71 ± 0.04 42 ± 5 Rh 4 4-Rh E VQSTRL 0.032 ± 0.005[c] 0.56 ± 0.08[c]

5 3 EWPTSII >500 65 ± 5 Rh 6 3-Rh E WPTSII >50 1.7 ± 0.2 [c]

7 6 EVQSTRF 1.9 ± 0.1 >1,000 Rh 8 6-Rh E VQSTRF 0.15 ± 0.05 >15

[a] Inhibitor equilibrium dissociation (Ki) constants were determined for cognate peptide/metallopeptide pairs. [b] All rhodium(II) complexes, including Rh2(OAc)4 (Ki =13 ± 5 μM) exhibit nonspecific inhibition, establishing an upper bound for these measurements. [c] Ki value significantly different from Rh2(OAc)4 (p<0.05, n=3). * unpublished data.

For attaining selective inhibition for either N1P1 or CALP, two sequences

EWPTSII and EVQSTRF were synthesized. For the sequence EWPTSII, upon

metalation, a high binding affinity of Ki = 1.7 µM was observed with CALP, but

the binding diminished considerably with N1P1 (Table 4.3, entries 5-6). The

sequence EVQSTRF, having a phenylalanine residue at P0 position, showed

almost no binding with CALP, but a good binding affinity towards N1P1. Upon

metalation, though binding improves with CALP but only to modest values.

97

However for N1P1, metalated peptide inhibitor shows a 13 fold enhancement in binding affinity (Table 4.3, entries 7-8). Further studies to footprint the key interactions for binding of the metalated inhibitors to N1P1 are undergoing in our lab.

4.6. Conclusions

Rh The metallopeptide E VQSTRL (Ki = 0.56 µM) is the first reported inhibitor with sub-micromolar affinity for the CAL PDZ domain, and is significantly shorter than decameric single-micromolar alternatives.50 The

Rh metallopeptide E VQSTRI (Ki = 0.45µM) also shows inhibition with an even better affinity for the CAL PDZ domain. Also, the metallopeptides ERhWPTSII and ERhVQSTRF have the potential to selectively inhibit CALP and N1P1 respectively. The ability to quickly generate a potent inhibitor for a recalcitrant protein target demonstrates that combining organic inhibitors with coordination chemistry is a viable strategy to inhibit protein–protein interactions. In addition, comparative binding studies with the CALP-H301A mutant and with P-3 metalated peptides indicate that rhodium-mediated affinity enhancement is specific. This capability bodes well for the development of inhibitors not only for PDZ domains, but for other structural classes of protein regulators as well.

98

4.7. Experimental

4.7.1. General Considerations.

Solvents and reagents were purchased from Fisher Scientific and used as received. Millipore ultra-purified water (18 MΩ) was used in all cases.

Peptide synthesis: All peptides were synthesized with an Advanced

ChemTech APEX 396 Automated Multipeptide Synthesizer using standard solid- phase Fmoc protocols. Rink amide MBHA resin (AAPPTEC) or preloaded Wang resin (AAPPTEC) was used to afford C-terminal amides or carboxylates, respectively. The peptides were acetylated at the N-terminus prior to cleavage from the resin. The purification was accomplished by reverse-phase HPLC with gradients of water-acetonitrile containing 0.1% trifluoroacetic acid, and peptides were isolated by lyophilization. Analysis and purity assessment was attained by mass spectrometry and analytical HPLC.

HPLC: HPLC was performed on a Shimadzu CBM-20A instrument with

Phenomene Jupiter 4μ Proteo 9 A (25 × 15 mm preparati e) and Phenomenex

Jupiter 4μ Proteo 9 A (25 × 4.6 mm analytical) columns. Flow rates of 8 mL/min and 1 mL/min were used for preparative and analytical columns, respectively.

Analytical and preparative HPLC were performed with gradient of acetonitrile in water. Both solvents contained 0.1% trifluoroacetic acid (TFA) unless otherwise noted. Data was collected using UV-vis absorption at 220 nm and 300 nm.

99

Mass Spectrometry: MALDI-MS was performed on a Bruker Daltonics

Autoflex MALDI-TOF/TOF mass spectrometer with CHCA matrix (10 mg/mL,

Thermo Scientific Pierce). ESI-MS was performed on a Bruker Daltonics microTOF instrument. Data analysis was performed on mMass program.57,58

Protein reagents: The expression vector for the CALP-H301A mutant was prepared by PCR mutagenesis59 using the WT vector60 as a template. Mutagenesis was verified by DNA sequencing. Expression and purification of the WT and mutant CAL PDZ domains were performed as described.60 The preparation of 15N- labeled protein for NMR analysis followed published protocols.50 This work was performed by Madden research group, Dartmouth medical school.

Protein binding studies: Fluorescence anisotropy inhibition binding assays were performed using standard procedures.49 Briefly, wells were prepared containing 1.8 M (WT) or 5.5 M (H301A) CAL PDZ protein, 30 nM fluorescein-labeled iCAL36 reporter (Tufts University Core Facility), and varying concentrations of inhibitor peptides (VQDTRL and EVQSTRL, with and without rhodium side-chain modification). Following equilibration, fluorescence anisotropy values were determined using a Tecan Infinite M1000 plate reader

(n=3). Inhibitor equilibrium dissociation (Ki) constants were estimated as

50 described. DMSO and Rh2(OAc)4 were used as negative controls for unlabeled and labeled peptides, respectively. This work was performed in collaboration with

Madden research group, Dartmouth medical school.

100

Capture inhibition assays. To determine the ability of inhibitor peptides to displace the interactions of full-length CAL in the presence of the epithelial-cell proteome, a capture inhibition assay was developed. Briefly, using published methods50 a biotin-conjugated peptide “bait” sequence (BT-iCAL36) was immobilized on streptavidin beads and incubated with clarified lysates of

CFBE41o- cells to capture full-length CAL in the presence of cellular proteins.

For the inhibition assay used here, capture was performed in the presence of a dilution series of metalated or non-metalated EVQSTRL inhibitor peptides. Beads were washed and eluted, and bound proteins were separated by SDS-PAGE and immunoblotted using an -CAL antibody. Band intensities were separately quantified and averaged (n=3). This work was performed by Madden research group, Dartmouth medical school.

NMR: 1H spectra were recorded on Bruker 5 UltraShield™ (5 MHz) spectrometer (for EVQSTRL system) and on Oxford (400 MHz) spectrometer (for

VQDTRL system). The chemical shifts (δ) are reported in units of part per million

(ppm) relative to solvent peak.

HSQC footprinting: 25 M 15N-labeled CAL PDZ protein was also subjected to 1H, 15N heteronuclear single quantum correlation spectroscopy

(HSQC) analysis, as described,50 except that tris(2-carboxyethyl)phosphine was omitted from the final dialysis buffers. Spectra were measured either in the presence of 125 M EVQSTRL, 45 M Rh-EVQSTRL, or no peptide, each at a

101 final concentration of 2.5% (v/v) DMSO. HSQC backbone crosspeaks were assigned by comparison with the previously assigned CALP apo spectrum. This work was performed by Madden research group, Dartmouth medical school.

4.7.2. Synthetic Procedures.

General procedure for synthesis of CFTR-derived metallopeptides.

Example synthesis of VQDRhTRL: The peptide VQDTRL (15.0 mg, 19.4

μmol) and Rh2(tfa)(OAc)3 (1 .1 mg, 2 .4 μmol) were charged into a 1-dram vial equipped with a stir bar. A solution of MES buffer (2-(N- morpholino)ethanesulfonic acid, 19.4 mL, 0.1 M aq soln, pH 4.9) was added. The reaction was heated at 50 °C overnight, after which all reactants were consumed based on HPLC analysis. The resulting mixture of dirhodium-peptide complexes were purified by direct injection of the reaction mixture onto a preparative HPLC column. The metallopeptide bound through the C-terminal carboxylate eluted first, followed by the desired aspartate-bound metallopeptide, which was isolated by lyophilization to afford a fluffy light blue powder (8.35 mg, 46 % yield). Smaller amounts of the later-eluting bis-dirhodium metallopeptide could also be isolated.

Analysis and purity assessment was attained by ESI–MS and analytical HPLC.

102

4.7.3. Characterization data

4.7.3.1. NMR analysis for determining position of dihodium ligation with peptide EVQSTRL

Figure 4.11 NMR comparison of (A) EVQSTRL peptide B) EVQSTRLRh and C) ERhVQSTRL irhodium attachment resulted in upfield shift of one leucine Hδ methyl (from ~ 0.9 ppm to 0.75 ppm , Δδ1H ca. –0.15 ppm) in the C-terminal bound product EVQSTRLRh (A vs B), and of the glutamate Hδ methylene from 2.45 ppm to 2.2 ppm, Δδ1H ca. –0.15 ppm, in the side-chain-bound metallopeptide, ERhVQSTRL (A vs C).

103

4.7.3.2. HPLC analysis and ESI-MS characterization for the rhodium complexation of peptides

Figure 4.12 HPLC traces and ESI-MS data for EVQSTRL series a) HPLC traces of (A) unpurified metalation mixture of peptide EVQSTRL (B) purified terminal-metalated peptide EVQSTRLRh (C) purified sidechain-matalated peptide ERhVQSTRL b) ESI- MS data of purified sidechain-matalated peptide for ERhVQSTRL.

104

Figure 4.13 HPLC traces and ESI-MS data for QLDVTR series a) HPLC traces of (A) unpurified metalation mixture of peptide QLDVTR (B) purified terminal-metalated peptide QLDVTRRh (C) purified sidechain-matalated peptide QLDRhVTR b) ESI- MS data of purified sidechain-matalated peptide for QLDRhVTR.

105

Figure 4.14 HPLC traces and ESI-MS data for EWPTSII series a) HPLC traces of (A) unpurified metalation mixture of peptide EWPTSII (B) purified terminal-metalated peptide EWPTSIIRh (C) purified sidechain-metalated peptide ERhWPTSII b) ESI- MS data of purified sidechain-metalated peptide for ERhWPTSII.

106

Figure 4.15 HPLC traces and ESI-MS data for EVQSTRI series a) HPLC traces of (A) unpurified metalation mixture of peptide EVQSTRI (B) purified sidechain-metalated peptide ERhVQSTRI b) ESI- MS data of purified sidechain-metalated peptide for ERhVQSTRI.

107

Figure 4.16 HPLC traces and ESI-MS data for EVQSTRF series a) HPLC traces of (A) unpurified metalation mixture of peptide EVQSTRF,(only side-chain metalated peptide and the bis-metalated peptide was formed in this reaction.) (B) purified sidechain-metalated peptide ERhVQSTRF, b) ESI- MS data of purified sidechain-metalated peptide for ERhVQSTRF.

108

4.7.3.3. MALDI-MS characterization and purity analysis by HPLC for synthesized peptides a)

b)

Figure 4.17 a) HPLC trace of pure peptide VQDTRL and b) ESI-MS data for isolated peptide VQDTRL.

109

a)

b)

Figure 4.18 a) HPLC trace of pure peptide EVQSTRL and b) ESI-MS data for isolated peptide EVQSTRL.

110

a)

SPD-20A Ch1-220nm RK3-142-2-pur-001 RK3-142-2-pur-001.dat [Group] Name

1 2 3 4 5 6 7 8 9 10 11 12 13 Minutes

b)

Figure 4.19 . a) HPLC trace of pure peptide QLDVTR and b) MS data for isolated peptide QLDVTR.

111

a)

SPD-20A Ch2-220nm RK-142-1-pure trace-RK-analytical-20m.met RK-142-1-pure trace-RK-analytical-20m.met

9 10 11 12 13 14 15 16 17 18 19 20 Minutes

b)

Figure 4.20 a) HPLC trace of pure peptide EWPTSII and b) MS data for isolated peptide EWPTSII.

112

a)

SPD-20A Ch1-220nm

RK-4-3-EVQSTRI-purif-009-.dat

7 8 9 10 11 12 13 14 15 16 17 Minutes b)

Figure 4.21 a) HPLC trace of pure peptide EVQSTRI and b) MS data for isolated peptide EVQSTRI.

113

a)

SPD-20A Ch1-220nm

JC-1-21-EVQSTRF-purif-002-20-50_20m.met

5 6 7 8 9 10 11 12 13 14 Minutes b)

Figure 4.22 a) HPLC trace of pure peptide EVQSTRF and b) MS data for isolated peptide EVQSTRF.

114

4.7.3.4. HSQC analysis of proteins in presence and absence of metallopeptides

Figure 4.23 HSQC spectra of 15N-CALP (top) HSQC spectra of 15N-CALP determined separately (red) and in the presence of 125- M EVQSTRL peptide (magenta) in 2.5% (v/v) DMSO. (bottom) HSQC spectra of 15N-CALP determined separately (red) and in the presence of 125-M ERhVQSTRL metallopeptide (green) in 2.5% (v/v) DMSO.

115

4.8. References

(1) Kundu, R.; Cushing, P. R.; Popp, B. V.; Zhao, Y.; Madden, D. R.; Ball, Z. T.: Hybrid Organic–Inorganic Inhibitors of a PDZ Interaction that Regulates the Endocytic Fate of CFTR. Angew. Chem. Int. Ed. 2012, 51, 7217-7220. (2) Mammen, M.; Choi, S.-K.; Whitesides, G. M.: Polyvalent Interactions in Biological Systems: Implications for Design and Use of Multivalent Ligands and Inhibitors. Angew. Chem. Int. Ed. 1998, 37, 2754-2794. (3) Kiessling, L. L.; Strong, L. E.; Gestwicki, J. E.: Principles for multivalent ligand design. In Annu. Rep. Med. Chem., 2000; Vol. 35; pp 321-330. (4) Gestwicki, J. E.; Cairo, C. W.; Strong, L. E.; Oetjen, K. A.; Kiessling, L. L.: Influencing receptor-ligand binding mechanisms with multivalent ligand architecture. J. Am. Chem. Soc. 2002, 124, 14922-14933. (5) Verdine, G. L.: Drugging "undruggable" targets. Mol. Cancer Ther. 2007, 6, 3623S-3623S. (6) Hajduk, P. J.; Greer, J.: A decade of fragment-based drug design: strategic advances and lessons learned. Nat. Rev. Drug Discov. 2007, 6, 211-219. (7) DeLano, W. L.: Unraveling hot spots in binding interfaces: progress and challenges. Curr. Opin. in Struct. Biol. 2002, 12, 14-20. (8) Erlanson, D. A.; Wells, J. A.; Braisted, A. C.: Tethering: fragment-based drug discovery. Annu. Rev. Biophys. Biomol. Struct. 2004, 33, 199-223.

(9) Hu, X.; Sun, J.; Wang, H.-G.; Manetsch, R.: Bcl-XL-Templated Assembly of Its Own Protein–Protein Interaction Modulator from Fragments Decorated with Thio Acids and Sulfonyl Azides. J. Am. Chem. Soc. 2008, 130, 13820-13821. (10) Kulkarni, S. S.; Hu, X.; Doi, K.; Wang, H.-G.; Manetsch, R.: Screening of Protein– Protein Interaction Modulators via Sulfo-Click Kinetic Target-Guided Synthesis. ACS Chem. Biol. 2011, 6, 724–732. (11) Salgado, E. N.; Ambroggio, X. I.; Brodin, J. D.; Lewis, R. A.; Kuhlman, B.; Tezcan, F. A.: Metal templated design of protein interfaces. Proc. Nat. Acad. Sci. U.S.A. 2010, 107, 1827- 1832. (12) Salgado, E. N.; Radford, R. J.; Tezcan, F. A.: Metal-Directed Protein Self- Assembly. Acc. Chem. Res. 2010, 43, 661-672. (13) Katz, B. A.; Clark, J. M.; Finer-Moore, J. S.; Jenkins, T. E.; Johnson, C. R.; Ross, M. J.; Luong, C.; Moore, W. R.; Stroud, R. M.: Design of potent selective zinc-mediated serine protease inhibitors. Nature 1998, 391, 608-612. (14) Iyaguchi, D.; Kawano, S.; Takada, K.; Toyota, E.: Structural basis for the design of novel Schiff base metal chelate inhibitors of trypsin. Biorg. Med. Chem. 2010, 18, 2076-2080. (15) Schiffmann, R.; Heine, A.; Klebe, G.; Klein, C. D. P.: Metal Ions as Cofactors for the Binding of Inhibitors to Methionine Aminopeptidase: A Critical View of the Relevance of In Vitro Metalloenzyme Assays. Angew. Chem. Int. Ed. 2005, 44, 3620-3623. (16) Takeuchi, T.; Böttcher, A.; Quezada, C. M.; Meade, T. J.; Gray, H. B.: Inhibition of thermolysin and human α-thrombin by cobalt(III) Schiff base complexes. Bioorg. Med. Chem. 1999, 7, 815-819. (17) Venkatraman, S.; Wu, W.; Prongay, A.; Girijavallabhan, V.; George Njoroge, F.: Potent inhibitors of HCV-NS3 protease derived from boronic acids. Bioorg. Med. Chem. Lett. 2009, 19, 180-3.

116

(18) LeBeau, A. M.; Singh, P.; Isaacs, J. T.; Denmeade, S. R.: Potent and Selective Peptidyl Boronic Acid Inhibitors of the Serine Protease Prostate-Specific Antigen. Chem. Biol. 2008, 15, 665-674. (19) Wilbuer, A.; Vlecken, D. H.; Schmitz, D. J.; Kräling, K.; Harms, K.; Bagowski, C. P.; Meggers, E.: Iridium Complex with Antiangiogenic Properties. Angew. Chem. Int. Ed. 2010, 49, 3839-3842. (20) Xie, P.; Williams, D. S.; Atilla-Gokcumen, G. E.; Milk, L.; Xiao, M.; Smalley, K. S. M.; Herlyn, M.; Meggers, E.; Marmorstein, R.: Structure-Based Design of an Organoruthenium Phosphatidyl-inositol-3-kinase Inhibitor Reveals a Switch Governing Lipid Kinase Potency and Selectivity. ACS Chem. Biol. 2008, 3, 305-316. (21) Hocharoen, L.; Cowan, J. A.: Metallotherapeutics: Novel strategies in drug design. Chem. Eur. J. 2009, 15, 8670-8676. (22) Harney, A. S.; Lee, J.; Manus, L. M.; Wang, P.; Ballweg, D. M.; LaBonne, C.; Meade, T. J.: Targeted inhibition of Snail family zinc finger transcription factors by oligonucleotide-Co(III) Schiff base conjugate. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 13667- 13672. (23) Hurtado, R. R.; Harney, A. S.; Heffern, M. C.; Holbrook, R. J.; Holmgren, R. A.; Meade, T. J.: Specific Inhibition of the Transcription Factor Ci by a Cobalt(III) Schiff Base–DNA Conjugate. Molecular Pharmaceutics 2012, 9, 325-333. (24) Chifotides, H. T.; Dunbar, K. R.: Rhodium Compounds. In Multiple Bonds Between Metal Atoms; 3rd ed.; Cotton, F. A., Murillo, C., Walton, R. A., Eds.; Springer: New York, 2005; pp 465-589. (25) Jakimowicz, P.; Ostropolska, L.; Pruchnik, F. P.: Interaction of [Rh2(O2CCH3)4 (H2O)2] and [Rh2(O2CCH(OH)Ph)2 (phen)2(H2O)2] (O2C-CH(OH)Ph)2 with sulfhydryl compounds and ceruloplasmin. Met.-Based Drugs 2000, 7, 201-209. (26) Das, K.; Simmons, E. L.; Bear, J. L.: Thermodynamics and kinetics of some tetra- .mu.-carboxylato-dirhodium(II) adduct formation reactions. Inorg. Chem. 1977, 16, 1268-1271. (27) Chifotides, H. T.; Dunbar, K. R.: Interactions of metal-metal-bonded antitumor active complexes with DNA fragments and DNA. Acc. Chem. Res. 2005, 38, 146-156. (28) Angeles-Boza, A. M.; Chifotides, H. T.; Aguirre, J. D.; Chouai, A.; Fu, P. K. L.; Dunbar, K. R.; Turro, C.: Dirhodium(II,II) complexes: Molecular characteristics that affect in vitro activity. J. Med. Chem. 2006, 49, 6841-6847. (29) Rainen, L.; Howard, R. A.; Kimball, A. P.; Bear, J. L.: Complexes of rhodium(II) carboxylates with adenosine 5'-monophosphates, 5'-diphosphates, and 5'-triphosphates. Inorg. Chem. 1975, 14, 2752-2754. (30) Kang, M.; Chouai, A.; Chifotides, H. T.; Dunbar, K. R.: 2D NMR spectroscopic evidence for unprecedented interactions of cis-[Rh-2(dap)-(mu-O2CCH3)(2)(eta(1)- O2CCH3)(CH3OH)](O2CCH3) with a DNA oligonucleotide: Combination of intercalative and coordinative binding. Angew. Chem. Int. Ed. 2006, 45, 6148-6151. (31) Lee, S. H.; Chao, D. L.; Bear, J. L.; Kimball, A. P.: Inhibition of deamination of arabinosylcytosine(NSC-63878) by rhodium(II) acetate. Cancer Chemother. Rep., Part 1 1975, 59, 661-663. (32) Howard, R. A.; Spring, T. G.; Bear, J. L.: Interaction of rhodium(II) carboxylates with enzymes. Cancer Res. 1976, 36, 4402-4405. (33) Zaykov, A. N.; Popp, B. V.; Ball, Z. T.: Helix Induction by Dirhodium: Access to Biocompatible Metallopeptides with Defined Secondary Structure. Chem. Eur. J. 2010, 16, 6651- 6659.

117

(34) Howard, R. A.; Sherwood, E.; Erck, A.; Kimball, A. P.; Bear, J. L.: Hydrophobicity of several rhodium(II) carboxylates correlated with their biologic activity. J. Med. Chem. 1977, 20, 943-946. (35) Popp, B. V.; Ball, Z. T.: Structure-Selective Modification of Aromatic Side Chains with Dirhodium Metallopeptide Catalysts. J. Am. Chem. Soc. 2010, 132, 6660-6662. (36) Popp, B. V.; Ball, Z. T.: Proximity-driven metallopeptide catalysis: Remarkable side-chain scope enables modification of the Fos bZip domain. Chem. Sci. 2011, 2, 690-695. (37) Litowski, J. R.; Hodges, R. S.: Designing heterodimeric two-stranded α-helical coiled-coils - ffects of hydrophobicity and α-helical propensity on protein folding, stability, and specificity. J. Biol. Chem. 2002, 277, 37272-37279. (38) Dennis, A. M.; Howard, R. A.; Bear, J. L.: THE REACTIVITY OF TETRA-MU- ACETATODIRHODIUM(II) WITH SELECTED DIPEPTIDES AND TRIPEPTIDES, SUBSTITUTED PYRIDINES AND IMIDAZOLE LIGANDS. Inorg. Chim. Acta. 1982, 66, L31-L34. (39) Das, K.; Bear, J. L.: COMPLEXATION OF TETRA-MU-CARBOXYLATO- DIRHODIUM(II) WITH IMIDAZOLE. Inorg. Chem. 1976, 15, 2093-2095. (40) Woolfson, D. N.; Parry, D. A. D.; Squire, J. M.: The Design of Coiled-Coil Structures and Assemblies. In Adv. Protein Chem.; Academic Press, 2005; Vol. 70; pp 79-112. (41) Antonsson, P.; Kammerer, R. A.; Schulthess, T.; Hänisch, G.; Engel, J.: Stabilization of the -Helical Coiled-coil Domain in Laminin by C-terminal Disulfide Bonds. J. Mol. Biol. 1995, 250, 74-79. (42) Hoshino, M.; Yumoto, N.; Yoshikawa, S.; Goto, Y.: Design and characterization of the anion-sensitive coiled-coil peptide. Protein Sci. 1997, 6, 1396-404. (43) Taylor, C. M.; Keating, A. E.: Orientation and Oligomerization Specificity of the Bcr Coiled-Coil Oligomeri ation omain†. Biochemistry 2005, 44, 16246-16256. (44) Cheng, J.; Moyer, B. D.; Milewski, M.; Loffing, J.; Ikeda, M.; Mickle, J. E.; Cutting, G. R.; Li, M.; Stanton, B. A.; Guggino, W. B.: A Golgi-associated PDZ Domain Protein Modulates Cystic Fibrosis Transmembrane Regulator Plasma Membrane Expression. J. Biol. Chem. 2002, 277, 3520-3529. (45) Moyer, B. D.; Denton, J.; Karlson, K. H.; Reynolds, D.; Wang, S.; Mickle, J. E.; Milewski, M.; Cutting, G. R.; Guggino, W. B.; Li, M.; Stanton, B. A.: A PDZ-interacting domain in CFTR is an apical membrane polarization signal. J. Clin. Invest. 1999, 104, 1353-1361. (46) Estell, K.; Braunstein, G.; Tucker, T.; Varga, K.; Collawn, J. F.; Schwiebert, L. M.: Plasma membrane CFTR regulates RANTES expression via its C-terminal PDZ-interacting motif. Mol. Cell. Biol. 2003, 23, 594-606. (47) Milewski, M. I.; Mickle, J. E.; Forrest, J. K.; Stafford, D. M.; Moyer, B. D.; Cheng, J.; Guggino, W. B.; Stanton, B. A.; Cutting, G. R.: A PDZ-binding motif is essential but not sufficient to localize the C terminus of CFTR to the apical membrane. J. Cell Sci. 2001, 114, 719- 726. (48) Cushing, P. R.; Fellows, A.; Villone, D.; Boisguerin, P.; Madden, D. R.: The relative binding affinities of PDZ partners for CFTR: a biochemical basis for efficient endocytic recycling. Biochemistry 2008, 47, 10084-98. (49) Vouilleme, L.; Cushing, P. R.; Volkmer, R.; Madden, D. R.; Boisguerin, P.: Engineering Peptide Inhibitors to Overcome PDZ Binding Promiscuity. Angew. Chem. Int. Ed. 2010, 49, 9912–9916. (50) Cushing, P. R.; Vouilleme, L.; Pellegrini, M.; Boisguerin, P.; Madden, D. R.: A Stabilizing Influence: CAL PDZ Inhibition Extends the Half-Life of ΔF508-CFTR. Angew. Chem. Int. Ed. 2010, 49, 9907-9911.

118

(51) Piserchio, A.; Fellows, A.; Madden, D. R.; Mierke, D. F.: Association of the cystic fibrosis transmembrane regulator with CAL: structural features and molecular dynamics. Biochemistry 2005, 44, 16158-66. (52) Zaykov, A. N.; MacKenzie, K. R.; Ball, Z. T.: Controlling Peptide Structure with Coordination Chemistry: Robust and Reversible Peptide-Dirhodium Ligation. Chem. Eur. J. 2009, 15, 8961-8965. (53) Roberts, K. E.; Cushing, P. R.; Boisguerin, P.; Madden, D. R.; Donald, B. R.: Computational design of a PDZ domain peptide inhibitor that rescues CFTR activity. PLoS Comp. Biol. 2012, 8, e1002477. (54) Shenolikar, S.; Voltz, J. W.; Cunningham, R.; Weinman, E. J.: Regulation of Ion Transport by the NHERF Family of PDZ Proteins. Physiology 2004, 19, 362-369. (55) Li, J.; Dai, Z.; Jana, D.; Callaway, D. J. E.; Bu, Z.: Ezrin Controls the Macromolecular Complexes Formed between an Adapter Protein Na+/H+ Exchanger Regulatory Factor and the Cystic Fibrosis Transmembrane Conductance Regulator. J. Biol. Chem. 2005, 280, 37634-37643. (56) Karthikeyan, S.; Leung, T.; Ladias, J. A. A.: Structural Determinants of the Na+/H+Exchanger Regulatory Factor Interaction with the β2Adrenergic and Platelet-derived Growth Factor Receptors. J. Biol. Chem. 2002, 277, 18973-18978. (57) Strohalm, M. Hassman, M. ošata, B. od če , M.: mMass data miner: an open source alternative for mass spectrometric data analysis. Rapid Commun. Mass Spectrom. 2008, 22, 905-908. (58) Strohalm, M. avan, . ov , P. Voln , M. Havl če , V. r.: mMass 3: A Cross- Platform Software Environment for Precise Analysis of Mass Spectrometric Data. Anal. Chem. 2010, 82, 4648-4651. (59) Fisher, C. L.; Pei, G. K.: BioTechniques 1997, 23, 570. (60) Amacher, J. F.; Cushing, P. R.; Weiner, J. A.; Madden, D. R.: Crystallization and preliminary diffraction analysis of the CAL PDZ domain in complex with a selective peptide inhibitor. Acta Crystallographica Section F 2011, 67, 600-603.

119

Chapter 5 Designing Dirhodium-Ligated Small Molecule Inhibitors for FK506-Binding Proteins

5.1. Introduction

FK-506 binding proteins (FKBP) are a subset of the immunophilin family of proteins, and are characterized by the presence of at least one peptidyl prolyl isomerase (PPIase) domain. These proteins are present almost in all tissues, in several eukaryotes from yeasts to humans. Several isoforms of FKBP play a regulatory role in diverse cellular pathways effecting tumorigenesis, neurodegenerative conditions, psychiatric disorders etc. and therefore, present interesting targets for drug development.1,2

In chapter 4, we have introduced our strategy of developing potent inhibitors for undruggable protein targets by incorporating a metal in the inhibitor, and thereby introducing stronger covalent interactions with the protein surface.

For this approach, we metalated a terminal carboxylate peptide that could target the protein active site via the C-terminus carboxylate moiety.

In order to extend the scope of our metalation strategy towards small- molecule inhibitors, the protein FKBP 12 appeared to be an apt choice as a target.

It is a well-studied protein and various FKBP 12 inhibitors have already been described in the literature.

120

5.2. Designing small-molecule metalated inhibitors

For designing the first generation of metalated inhibitors for FKBP 12, we looked at various reported inhibitors.3,4 Of all these, FKBP inhibitor A was selected as the inhibitor for our design inspiration. This was due to its average affinity for FKBP 12, as an inhibitor with an already high affinity in nano or subnanomolar range would not facilitate the detection of any increment in potency after metalation.

Figure 5.1 FKBP12 inhibitor selected as design inspiration.

In order to incorporate a dirhodium center in the core of compound A, a carboxylate functionality has to be installed, as the metalation process involves ligating the dirhodium to the carboxylate moeity.5,6 For this purpose, we designed compound B (Figure 5.2). The retrosynthesis to attain compound B is shown

(Figure 5.2), which involves the coupling of the modified L-pipecolic acid with the S or R versions of the alcohol methyl 5-hydroxy-5-phenylpentanoate containing masked carboxylate functionality to yield the corresponding (S, S) or

(S, R) diasteromers (Figure 5.2).

121

Figure 5.2 Retrosynthetic pathway for designing compound B.

5.2.1. Synthesis of compound B

(Undergraduate researcher Julian Cooper helped in several synthetic steps involved in this section.)

The compound B was synthesized using the procedure detailed below

(Figure 5.3). Commercially available L-pipecolic acid was esterified with EtOH in presence of thionyl chloride; the resulting ethyl-ester of L-pipecolic acid was then treated with benzylsulfonyl chloride in presence of the base triethylamine to yield the corresponding benzylsulfonylpiperidine adduct. The ethyl ester was then hydrolyzed in presence of LiOH·H2O to the corresponding acid, which was then subjected to DCC-coupling with S/R versions of methyl-5-hydroxy-5-

122 phenylpentanoate to give the corresponding diasteromers (S,S) or (S,R). (For simplification, the stereochemistry is not indicated in Figure 5.3). Each diasteromer was then subjected to metalation using Rh2(OAc)3(tfa) in trifluoroethanol to give the metalated inhibitors. Detailed experimental procedure for each step is described in the experimental section.

Figure 5.3 Synthesis of metalated inhibitors for FKBP12

5.3. Future direction

The metalated inhibitors along with the non-metalated version compound

B, was assayed to check its activity against FKBP 12, by postdoctoral fellow

123

researcher Dr. Jane Coughlin. The results are currently being analyzed to ascertain

if the incorporation of metal results in an increment in the FKBP12 affinity of the

inhibitor.

5.4. Experimental

5.4.1. General Considerations:

All reactions were carried out in oven dried glassware unless otherwise

mentioned. Sure/sealed acetonitrile (Aldrich) was used for DCC coupling.

Trifluroethanol (TFE), triethylamine, diisopropylethylamine, ethanol,

dimethylformamide (DMF) and tetrahydrofuran (THF) were reagent grade and

obtained from Fisher Chemicals. NMR solvents CDCl3 (Cambridge Isotope

Laboratories) was used as received. Flash Chromatography was performed on 40-

63 μm particle size silica gel. The following chemicals were purchased and used

as received: L-pipecolic acid (TCI America), 5-oxo-5-phenylpentanoic acid

(Aldrich), (R)-(+)-2-methyl-CBS-oxazaborolidine solution (1M in toluene)

(Aldrich), (S)-(+)-2-methyl-CBS-oxazaborolidine solution (1M in toluene)

(Aldrich), thionyl chloride (Aldrich), phenylmethanesulfonyl chloride (Aldrich),

LiOH·.H2O (Fisher), DCC (Acros), DMAP (Aldrich), DIEA (Fisher), acetonitrile

(Fisher), MgSO4 (Fisher).

HPLC was performed on a Shimadzu CBM-20A instrument with

Phenomene Jupiter 4 μ Proteo 9 A (25 × 15) and Phenomene Jupiter 4 μ

Proteo 90A (250 × 4.6) columns. Flow rates of 8 mL/min and 1 mL/min were used

124 for the preparatory and analytical columns respectively. Analytical and preparative

HPLC were performed with gradient of acetonitrile in water. Both solvents contained 0.1% trifluoroacetic acid (TFA) unless otherwise noted. Two wavelengths — 220 nm and 300 nm — were used to allow for independent analysis of peptides and dirhodium complexes. ESI-MS was performed on Bruker

Daltonics micrOTOF instrument. IR spectra were recorded using a Nicolet Avatar

320 FT-IR spectrometer, and KBr pellets. 1H and 13C spectra were recorded on

Bruker 5 UltraSield™ (5 MHz) spectrometer or O ford (4 MHz) spectrometer. The chemical shifts (δ) are reported in units of part per million

1 13 (ppm) relative to TMS internal standard (for H) or CDCl3 solvent (for C).

Compounds 5.1, 5.2, 5.3 were synthesized according to previously published protocol.7

5.4.2. Synthesis of inhibitors and precursors

(S)-(S)-5-methoxy-5-oxo-1-phenylpentyl 1-(benzylsulfonyl) piperidine-2- carboxylate (5.4-S)

Compound 5.4-S was prepared by DCC-coupling of compound 5.3 and (S)- methyl 5-hydroxy-5-phenylpentanoate.

125

Synthesis of (S)-methyl 5-hydroxy-5-phenylpentanoate: The commercially available keto-acid, 5-oxo-5-phenylpentanoic acid was methylated to form the keto-ester, methyl 5-oxo-5-phenylpentanoate. The keto-ester was then reduced to the corresponding S-alcohol, (S)-methyl 5-hydroxy-5-phenylpentanoate using (S)-(+)-2-methyl-CBS-oxazaborolidine catalyst along with borane dimethyl sulfide using published protocol.8 The crude product was analyzed by 1H NMR,9 chiral HPLC (96% ee) and was used directly in the subsequent synthetic step without purification.

Synthesis of (S)-(S)-5-methoxy-5–oxo-1-phenylpentyl 1-(benzylsulfonyl) piperidine-2-carboxylate (5.4): The compound 5.3 was coupled with (S)-methyl

5-hydroxy-5-phenylpentanoate using the following protocol. To a dry flask with a stir bar, under nitrogen, containing the compound 5.3 (123 mg, 468 moles),

DMAP (3 mg, 24.6 moles) dissolved in 0.1 mL of MeCN was added, followed by the addition of (S)-methyl 5-hydroxy-5-phenylpentanoate (190 mg, 913

moles) dissolved in 1.7 mL of MeCN. This solution was cooled to 0 °C, and then

DCC (188 mg, 912 moles) dissolved in 0.9 mL MeCN was added to it. The reaction was warmed up to room temperature and stirred overnight. The reaction mixture was diluted with excess MeCN and the precipitating DCC urea was filtered off. The filtered solution was reduced under pressure and purified by silica gel column (16 % EtOAc in Hexanes). The product was further subjected to HPLC purification and afforded 104 mg (47 %) of pure product. 1H NMR (500 MHz,

126

CDCl3) δ 7.52-7.27 (m, 10H), δ 5.80 (m, 1H), δ 4.56 (m, 1H), δ 4.13 (m, 2H), δ

3.65 (s, 3H), δ 3.44 (m, 1H), δ 3.14 (m, 1H), δ 2.33 (t, 2H, J = 7.35 Hz), δ 2.26-

13 1.14 (m, 10H). C NMR (500 MHz, CDCl3) δ 173.9, δ 171.1, δ 139.8, δ 131.1

(2C), δ 129.3, δ 128.9 (2C), δ 128.7 (3C), δ 128.5, δ 126.7 (2C), δ 58.8, δ 56.3, δ

51.9, δ 43.6, δ 35.8, δ 33.7, δ 28. , δ 25.2, δ 21.1, δ 2 .5. IR (thin film): 2946,

-1 + 2861, 1736, 1453, 1336, 1150, 1129, 967, 699, 539 cm . ESI-MS. m/z: [M-Na] calcd for C25H31NO6SNa: 496.2, found : 496.2

(S)-(R)-5-methoxy-5-oxo-1-phenylpentyl 1-(benzylsulfonyl) piperidine-2- carboxylate (5.4-R)

Compound 5.4-R was prepared by DCC-coupling of compound 5.3 and

(R)-methyl 5-hydroxy-5-phenylpentanoate in a procedure similar to that of the synthesis of 5.4-S. The chiral alcohol (R)-methyl 5-hydroxy-5-phenylpentanoate was synthesized by reducing methyl 5-oxo-5-phenylpentanoate using (R)-(+)-2- methyl-CBS-oxazaborolidine solution (1M in toluene) catalyst and borane dimethyl sulfide complex, as described earlier for the S-alcohol. The product was analyzed by 1H NMR and chiral HPLC (95% ee) and was used for the DCC coupling without further purification. To a dry flask with a stir bar, under nitrogen,

127 containing the compound 5.3 (136 mg, 517 moles), DMAP (3 mg, 24.6 moles) dissolved in 0.1 mL of MeCN was added, followed by the addition of (S)-methyl

5-hydroxy-5-phenylpentanoate (193 mg, 928 moles) dissolved in 1.7 mL of

MeCN. This solution was cooled to 0 °C, and then DCC (190 mg, 922 moles) dissolved in 0.9 mL MeCN was added to it. The reaction was warmed up to room temperature and stirred overnight. The reaction mixture was diluted with excess

MeCN and the precipitating DCC urea was filtered off. The filtered solution was reduced under pressure and purified by silica gel column (16 % EtOAc in

Hexanes). The product was further subjected to HPLC purification and afforded

1 66 mg (27 %) of pure product. H NMR (400 MHz, CDCl3) δ 7.55-7.25 (m, 10H),

δ 5.77 (m, 1H), δ 4.6 (m, 1H), δ 4.25 (s, 2H), δ 3.65 (s, 3H), δ 3.40 (m, 1H), δ 3.1

(m, 1H), δ 2.3 (t, 2H, J = 7.42 Hz), δ 2.2-0.9 (m, 10H). 13C NMR (400 MHz,

CDCl3) δ 173.8, δ 171.1, δ 14 .2, δ 131.2 (2C), δ 129.5, δ 128.82 (2C), δ 128.75

(2C), δ 128.67, δ 128.51, δ 126.6 (2C), δ 58.9, δ 56.1, δ 51.8, δ 43.5, δ 35.8, δ

33.7, δ 28. , δ 25.1, δ 21.1, δ 2 .3. IR (thin film): 3 33, 2948, 2861, 2341, 1732,

1496, 1455, 1337, 1200, 1151, 1130, 1110, 968, 734, 700 cm-1. ESI-MS. m/z: [M-

+ Na] calcd for C25H31NO6SNa: 496.2, found : 496.2

128

(S)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl)oxy)-5-phenylpentanoic acid (5.5-S)

The substrate (5.4-S) (S)-(S)-5-methoxy-5-oxo-1-phenylpentyl 1-

(benzylsulfonyl) piperidine-2-carboxylate (36 mg, 76 moles) and LiOH·H2O (3.2 mg, 76 μmol) were charged into a 4 mL vial equipped with a stir bar, at 0 °C. To this THF/H2O (2:1) (1.09 mL, 0.07 M) was added, and the reaction was warmed upto 40 °C, and stirred for 3 h, after which the reaction appeared to be complete based on TLC analysis. The resulting reaction mixture of was quenched with 2N

HCl, and then extracted with DCM. The organic extract was washed with brine, dried and concentrated. The concentrated crude product was dissolved in acetonitrile and purified by direct injection onto a preparative HPLC column to yield 19.5 mg (56 %) of the pure product. Analysis and purity assessment was attained by 1H NMR, 13C NMR, ESI–MS and analytical HPLC. 1H NMR (400

MHz, CDCl3) δ 7.46-7.27 (m, 10H), δ 6.4 (br s), δ 5.81 (m, 1H), δ 4.54 (m, 1H), δ

4.14 (m, 2H), δ 3.44 (m, 1H), δ 3.14 (m, 1H), δ 2.4 (t, 2H, J = 7.43 Hz), δ 2.3-1.1

13 (m, 10H). C NMR (400 MHz, CDCl3) δ 178.9, δ 171.1, δ 139.7, δ 131.1 (2C), δ

129

129.3, δ 128.9 (2C), δ 128.7 (2C), δ 128.6 (2C), δ 126.7 (2C), δ 58.9, δ 56.3, δ

43.7, δ 35.7, δ 33.5, δ 28. , δ 25.2, δ 2 .8, δ 2 .5. IR (thin film): 2944, 2360,

2340, 1734, 1708, 1455, 1336, 1200, 1150, 1129, 968, 699, 668, 538 cm-1. ESI-

+ MS. ESI-MS. m/z: [M-Na] calcd for C24H29NO6SNa: 482.2, found : 482.2.

R)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl)oxy)-5-phenylpentanoic acid (5.5 R)

The substrate (5.4-R) (S)-(R)-5-methoxy-5-oxo-1-phenylpentyl 1-

(benzylsulfonyl) piperidine-2-carboxylate (46 mg, 97 moles) and LiOH·H2O (4.4 mg, 97 μmol) were charged into a 4 mL vial equipped with a stir bar, at 0 °C. To this THF/H2O (2:1) (1.51 mL, 0.07 M) was added, and the reaction was warmed upto 4 °C, and stirred for 6 h, after which the reaction appeared to be complete based on TLC analysis. The resulting reaction mixture of was quenched with 2N

HCl, and then extracted with DCM. The organic extract was washed with brine, dried and concentrated. The concentrated crude product was dissolved in acetonitrile and purified by direct injection onto a preparative HPLC column to yield 20 mg (57 %) of the pure product. Analysis and purity assessment was

130 attained by 1H NMR, 13C NMR, ESI–MS and analytical HPLC. 1H NMR (500

MHz, CDCl3) δ 7.51-7.27 (m, 10H), δ 6.8 (br s, 2H), δ 5.78 (m, 1H), δ 4.58 (m,

1H), δ 4.26 (s, 2H), δ 3.41 (m, 1H), δ 3.11 (m, 1H), δ 2.39 (m, 2H), δ 2.25-0.87

13 (m, 10H). C NMR (500 MHz, CDCl3) δ 178.8, δ 171. , δ 14 .1, δ 131.2 (2C), δ

129.4, δ 128.82 (2C),δ 128.75 (2C), δ 128.7 , δ 128.46, δ 126.7, 126.5 (2C), δ

58.9, δ 56.1, δ 43.5, δ 35.7, δ 33.6, δ 28. , δ 25.1, δ 2 .8, δ 2 .2. IR (thin film):

3033, 2944, 2861, 2360, 2340, 1735, 1707, 1455, 1336, 1200, 1150, 1129, 968,

-1 + 699, 538 cm . ESI-MS. m/z: [M-Na] calcd for C24H29NO6SNa: 482.2, found :

482.2.

Dirhodium ligation of (S)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl) oxy) -5-phenylpentanoic acid (5.6-S)

The substrate (S)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl)oxy)-5- phenylpentanoic acid (8.6 mg, 19 moles) and Rh2(tfa)(OAc)3 (9.2 mg, 19 μmol) were charged into a 20 mL scintillation vial equipped with a stir bar. To this TFE

(trifluroethanol, 14 mL, 1.4 mM) was added, followed by the addition of DIEA

131

(diiisopropylethylamine. 8 L, 43 moles). The reaction was heated at 50 °C overnight, after which all reactants were consumed based on HPLC analysis. The resulting reaction mixture of was purified by direct injection onto a preparative

HPLC column to yield 6.9 mg (44 %) of the pure metalated product. Analysis and purity assessment was attained by 1H NMR, 13C NMR, ESI–MS and analytical

HPLC. The metalated complex was axially ligated to water and MeCN molecules

1 as observed in NMR. H NMR (400 MHz, CDCl3) δ 7.43-7.27 (m, 10H), δ 5.75

(m, 1H), δ 4.52 (m, 1H), δ 4.19 (m, 2H), δ 3.44 (m, 1H), δ 3.12 (m, 1H), δ 2.79 (br s, 6 H), δ 2.58 (s, 1H), δ 2.55 (s, 1H), δ 2.26 (m, 2H), δ 2.15 (m, 1H), δ 2. 1 (m,

1H), δ 1.99 (s, 3H), δ 1.91 (s, 6H), δ 1.88-1.01 (m, 10H). 13C NMR (400 MHz,

CDCl3) δ 193.8, δ 192.1, δ 192. (2C), δ 171.2, δ 139.8, δ 131.1 (3C), δ 129.3, δ

128.8 (5C), δ 128.5 , δ 126.8 (2C), δ 59.1, δ 56.4, δ 43.8, δ 36.7, δ 35.3, δ 27.9, δ

25.1, δ 23.9, δ 23.8 (2C), δ 21.6, δ 2 .5. IR (thin film): 2360, 2341, 1734, 1637,

-1 + 1582, 1417, 1337, 1200, 699, 668 cm . ESI-MS. m/z: [M-Na] calcd for

C30H37NO12SRh2Na: 864.2, found : 864.2.

Dirhodium ligation of (R)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl) oxy)-5-phenylpentanoic acid (5.6-R)

132

The substrate (R)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl)oxy)-5- phenylpentanoic acid (12 mg, 26 moles) and Rh2(tfa)(OAc)3 (12.8 mg, 26 μmol) were charged into a 20 mL scintillation vial equipped with a stir bar. To this TFE

(trifluroethanol, 15 mL, 1.7 mM) was added, followed by the addition of DIEA

(diiisopropylethylamine. 12 L, 65 moles). The reaction was heated at 50 °C overnight, after which all reactants were consumed based on HPLC analysis. The resulting reaction mixture of was purified by direct injection onto a preparative

HPLC column to yield 12.6 mg (57 %) of the pure metalated product. Analysis and purity assessment was attained by 1H NMR, 13C NMR, and ESI–MS.The metalated complex was axially ligated to water and MeCN molecules as observed

1 in NMR. H NMR (400 MHz, CDCl3) δ 7.61-7.20 (m, 10H), δ 5.75 (m, 1H), δ

4.68 (m, 1H), δ 4.36 (s, 2H), δ 3.51 (br s, 6H), δ 3.41 (m, 1H), δ 3.14 (m, 1H), δ

3.10 (m, 1H), δ 2.5 (s, 3H), δ 1.96 (s, 3H), δ 1.91 (s, 6H), δ 1.82-0.75 (m, 10H).

13 C NMR (500 MHz, CDCl3) δ 194.2, δ 192.7, δ 192.5 (2C), δ 171. , δ 14 .2, δ

131.3 (2C), δ 129.5, δ 128.7 (6C), δ 128.3 , δ 126.5 (2C), δ 115.8, δ 59. , δ 56. , δ

43.6, δ 36.9, δ 35.7, δ 28.1, δ 25.2, δ 24. , δ 23.9 (2C), δ 21.8, δ 2 .3, δ 3.58. IR

(thin film): 3249, 2943, 2360, 2341, 1734, 1695, 1636, 1585, 1558, 1417, 1336,

133

-1 + 1200, 1151, 1129, 968, 734, 699, 668, 539 cm . ESI-MS m/z: [M-Na] calcd for

C30H37NO12SRh2Na: 864.2, found : 864.2.

5.5. References:

(1) Blackburn, E. A.; Walkinshaw, M. D.: Targeting FKBP isoforms with small- molecule ligands. Curr. Opin. Pharmacol. 2011, 11, 365-371. (2) Solassol, J.; Mange, A.; Maudelonde, T.: FKBP family proteins as promising new biomarkers for cancer. Curr. Opin. Pharmacol. 2011, 11, 320-325. (3) Holt, D. A.; Konialian-Beck, A. L.; Oh, H.-J.; Yen, H.-K.; Rozamus, L. W.; Krog, A. J.; Erhard, K. F.; Ortiz, E.; Levy, M. A.; Brandt, M.; Bossard, M. J.; Luengo, J. I.: Structure-activity studies of synthetic FKBP ligands as peptidyl-prolyl isomerase inhibitors. Bioorganic & Medicinal Chemistry Letters 1994, 4, 315-320. (4) Yamashita, D. S.; Oh, H.-J.; Yen, H.-K.; Bossard, M. J.; Brandt, M.; Levy, M. A.; Newman-Tarr, T.; Badger, A.; Luengo, J. I.; Holt, D. A.: Design, synthesis and evaluation of dual domain FKBP ligands. Bioorganic & Medicinal Chemistry Letters 1994, 4, 325-328. (5) Zaykov, A. N.; MacKenzie, K. R.; Ball, Z. T.: Controlling Peptide Structure with Coordination Chemistry: Robust and Reversible Peptide-Dirhodium Ligation. Chem. Eur. J. 2009, 15, 8961-8965. (6) Zaykov, A. N.; Popp, B. V.; Ball, Z. T.: Helix Induction by Dirhodium: Access to Biocompatible Metallopeptides with Defined Secondary Structure. Chem. Eur. J. 2010, 16, 6651- 6659. (7) Juli, C.; Sippel, M.; Jager, J.; Thiele, A.; Weiwad, M.; Schweimer, K.; Rosch, P.; Steinert, M.; Sotriffer, C. A.; Holzgrabe, U.: Pipecolic acid derivatives as small-molecule inhibitors of the Legionella MIP protein. J Med Chem. 2011, 54, 277-83. Epub 2010 Dec 13. (8) Corey, E. J.; Helal, C. J.: Reduction of Carbonyl Compounds with Chiral Oxazaborolidine Catalysts: A New Paradigm for Enantioselective Catalysis and a Powerful New Synthetic Method. Angew. Chem. Int. Ed. 1998, 37, 1986-2012. (9) Kim, J.; De Castro, K. A.; Lim, M.; Rhee, H.: Reduction of aromatic and aliphatic keto esters using sodium borohydride/MeOH at room temperature: a thorough investigation. Tetrahedron 2010, 66, 3995-4001.

134

Chapter 6 Rhodium Catalyzed Cysteine Modification with Diazo Reagents

Parts of this chapter have been adapted from:1

undu, R. Ball, Z. T., “Rhodium cataly ed cysteine modification with dia o reagents.” Chem. Commun., 2012, DOI:10.1039/c2cc37323h.

6.1. Introduction

Chemical protein modification plays a key role in chemical biology, biomaterials research, the development of protein-based therapeutics, and other areas of modern research. Modification reactions have demanding requirements for chemoselectivity, pH tolerance, and linkage stability. Research in this area continues because no single method meets these challenges in all settings.2 In this chapter, a rhodium-catalyzed cysteine modification is described which produces serum-stable adducts.

Cysteine is arguably one of the most important targets for modification, owing to its reactivity and relative scarcity. A variety of reaction manifolds have been used to modify cysteine:3 alkylation, oxidative elimination(and subsequent thiol conjugate addition)4,5 and desulfurization of allylic disulfides/ selenosulfides after allylic rearrangement of thioethers.6,7 Many of these methods are robust and

135 useful, but drawbacks remain, for example, alkylation with maleimides is broadly useful, yet the resulting adducts are often unstable in biological environments due to reversible thiol exchange and other side reactions.8,9 Efforts to address maleimide reversibility have been made, but chemoselectivity issues with maleimide have been noted as well.10 Iodoacetimides are the other major class of cysteine-selective reagents, but they can require elevated pH and exhibit promiscuous reactivity with other side chains,11 a limitation ameliorated with chloroacetimides at a significant cost of reaction rate.12

Metal catalysis for protein modification is an important and relatively new tool for protein chemists.13 Metal catalysis can offer new opportunities and benefits, including low reactant concentrations and the ability to tune reactivity with molecular inputs.14 Given its value as a target, cysteine is surprisingly little- studied in metal-catalyzed methods. Reductive desulfurization to alanine is commonly employed in native chemical ligation strategies.15 Ruthenium16 and rhodium17 catalysis has been used to provide cysteine conjugates, but only on residues that have been first modified in a separate step to provide S-allyl cysteine.5

6.2. Dirhodium catalyzed CALP modification

In studies of protein modification, our lab has made extensive use of

“targeting” or “pro imity” strategies to deli er a rhodium catalyst to specific amino acids.18-21 In most cases, no modification occurs with simple rhodium

136 catalysts such as Rh2(OAc)4, in solution. During our research, we found that cysteine is an exception: surface-exposed cysteine thiols do react with diazo reagents in the presence of Rh2(OAc)4, affording alkylated cysteine products via

S–H insertion of a metallocarbene intermediate. (The other exception is tryptophan, which is also modified under these conditions when in a solvent- accessible location.22,23 We have never observed modification of the other sulphur- containing amino-acid, methionine, despite efforts to do so.19,24) The cysteine reactivity is consistent with a variety of reports of diazo insertion into S–H bonds, catalyzed by rhodium(II).25-28 This process has turned out to be a simple way to introduce functional handles that are stable in serum.

6.2.1. Screening modification on CALP

In studies of rhodium-catalyzed reactions on cystic fibrosis transmembrane conductance regulator-associated ligand (CAL) and its PDZ* domain,

24,29,30 CALP, Rh2(OAc)4 afforded >98% modification of the protein (Figure 6.1).

* PDZ domains are a family of peptide-binding PPI modules named for the first three members: PSD‒95, Dlg, and ZO‒1).

137

CALP (+1 mod)

9895

) ) a.u

CALP (+0 mod)

Intensity ( Intensity 9333

8500 9000 9500 10000 10500 Mass (m/z)

Figure 6.1 Modification of CALP using dirhodium metallocarbenes. a) 10 M solution of CALP was treated with biotinylated diazo compound 1 and Rh2(OAc)4 for 5 h. b) The sample was analyzed by MALDI-TOF and singly modified CALP was identified in the spectrum.

Negligible modification of CALP was observed in the absence of either catalyst Rh2(OAc)4 or biotin-diazo reagent 1 (Figure 6.2).

138

9333

(+0 mod)

) ) a.u

Intensity ( Intensity 9333 (+0 mod)

8500 9000 9500 10000 10500 11000 11500 Mass (m/z)

Figure 6.2 Control reactions of CALP

a) Control reaction in absence of the catalyst Rh2(OAc)4. b) Control reaction in absence of diazo reagent 1.

6.2.2. Cysteine as the site of modification

To elucidate the site of modification on CALP, trypsin digestion was performed on the modified protein. A digest fragment, Cys44−Arg62, contained a single modification with biotin–diazo (m/z = 2441) was observed (Figure 6.3 a).

MS/MS analysis of the Cys44−Arg62 segment led to the observation of several daughter ions (Figure 6.3 b), including the b3 ion with modification and the internal ions (2-8) and (2-12) without modification, indicating that modification occurs predominantly at Cys44.

139

2022 2441

) )

a.u Intensity ( Intensity 1755 1879 2921 2066 2576

1500 1800 2100 2400 2700 3000

Mass (m/z)

*

int (2-8) * b3 * * *

Figure 6.3 Modified CALP trypsin digest a) Modified CALP trypsin digest. Native peptide fragment containing C44, expected m/z = 1879; singly-modified, expected m/z = 2441. (m/z = 2022 corresponds to native peptide fragment V6-K24 + K+ adduct). b) MS/MS analysis of the singly modified peptide fragment (m/z = 2441), observed in trypsin digest of modified CALP. * The b, y and internal- ions are indicated in red, blue and green respectively.

140

To further corroborate that cysteine is the site of modification, the protein

CALP was treated with Na2SO3 to reversibly block the cysteine residue as an S- sulfate.31 The blocked protein, on being subjected to reaction conditions, shows minimal modification (Figure 6.4).

9412

CALP (cysteine blocked)

) ) a.u 9895 9333 unblocked CALP unblocked CALP

Intensity ( Intensity (+1 mod) (+0 mod)

8500 9000 9500 10000 10500 11000 Mass (m/z)

Figure 6.4 Blocking surface cysteines on CALP The protein was treated with Na2SO3 to oxidize the cysteine residue, the blocked protein on being subjected to reaction conditions shows minimal modification.

6.3. Dose-dependent modification of CALP

To determine the dependence of CALP modification on the amount of catalyst Rh2(OAc)4 and on the biotin diazo 1, the concentrations of these reagents were varied separately and the percent modification of CALP was determined. It was observed that CALP modification was dependent on amounts of both of these reagents (Figure 6.5). The percent conversions were determined relative to the unreacted protein by using MALDI-TOF. These results were further confirmed (in case of Rh2(OAc)4) by correlating the % conversions obtained in MALD-MS with biotin-specific western blot (Figure 6.5. c, d, e).

141

Figure 6.5 Dose-dependence of CALP modification on biotin-diazo 1 and Rh2(OAc)4 a) modified CALP with varying concentrations of biotin diazo 1 determined by MALDI- MS, b) % conversions of CALP modification relative to diazo concentrations, corresponding to Figure (a).

c) modified CALP with varying concentrations of Rh2(OAc)4 determined by MALDI-MS, d) biotin-specific western blot of CALP modification reactions with varying concentration of Rh2(OAc)4, corresponding to Figure (c). e) Comparison of % conversion of CALP modification relative to Rh2(OAc)4 concentration, determined by MALDI-MS and biotin- specific western blot.

142

6.4. Effect of neighboring residues on cysteine reactivity

1660 X= R (+1 mod)

1099 (+0 mod)

X= K 1633

) ) (+1 mod) a.u

1072

(+0 mod) Intensity ( Intensity

X= A 1614 (+1 mod + K+)

1200 1400 1600 1800 2000 Mass (m/z)

Figure 6.6 Peptide models to study dirhodium metallocarbenoid mediated cysteine modification. No difference in reactivity was observed based on the adjacent residue to cysteine.

143

The significant reactivity of the CALP cysteine was somewhat surprising, given that previous work by us and by others did not uncover this reactivity. We synthesized a series of peptides derived from the CALP sequence

39QPADRCGGL47 to assess if the local sequence, including the presence of a neighboring positively-charged (arginine) residue, might lead to unique reactivity.

For this purpose, we varied the adjacent residue from arginine (R) to another positively-charged residue lysine (K) and a neutral residue like alanine (A). Under standard reaction conditions, clean conversion to >95% biotinylated peptide was observed for all three peptide examined (Figure 6.6). Removing the neighboring positive charge (R43A) had no appreciable effect, indicating that cysteine reactivity is likely to be general, at least for surface-exposed thiol groups.

6.5. Stability of S-C linkage in biological medium

The stability of the linkage formed in cysteine functionalization is a concern of any method. Disulfide linkages, for examples, are readily formed but undergo rapid reduction or disulfide exchange under biological conditions.

Maleimides, despite their frequent use, afford rather unstable cysteine–maleimide linkage.8 MS-based analysis of bioconjugates in serum effectively addresses stability questions.9

144

1660 1099 0 min (+1 mod) (+0 mod)

1h Intensity (a.u) (a.u) Intensity 1434

1200 1400 1600 1800 2000 Mass (m/z) 1099 1550 (+0 mod) (+1 mod) 0 min

1099 1324 1h

Intensity (a.u) (a.u) Intensity (+0 mod)

1000 1200 1400 1600 1800 2000 Mass (m/z)

Figure 6.7 Stability study for the linkages in peptide models. Peptide-biotin conjugates were incubated at 37 °C in human plasma serum and the stability of the conjugate assessed by MALDI-MS. a) Rh2(OAc)4 catalyzed modification and b) maleimide modification.

To assess these stability issues with products of diazo modification, the peptide AQPADRCGGLA was modified with a commercially available biotin- maleimide to afford a peptide–biotin conjugate. In human serum, this conjugate had a half-life of only a few minutes and was completely consumed within an hour, consistent with previous reports8,9 (Figure 6.7 b). The corresponding

145 peptide–biotin conjugate from rhodium-catalyzed modification persisted for hours in serum at 37 °C. (Figure 6.7 a). As expected from the structure of the diazo- insertion products, the biotinylation products of rhodium-catalyzed modification are significantly more stable in human plasma and provide a sturdier bond than maleimide-based bonds.

6.6. Cysteine selectivity in a mixture of proteins

Figure 6.8 Modification of CALP in a mixture of proteins. Each reaction contained 140 g/mL of each of five proteins at 4 ᵒC for 5 h (a) Rh2(OAc)4 t – catalyzed reaction: Rh2(OAc)4 (15 M), diazo 1 (1.5 mM), BuNHOH.HCl (10 mM), pH 6.2, (b) Biotin-maleimide mediated modification: biotin-maleimide (250 M), pH 7.5 (see section 1.8 Experimental, for details).

To check the scope and selectivity of the reaction, we examined cysteine modification within a small mixture of proteins. A mixture of proteins (CALP, triose phosphate isomerise (TIM), myoglobin, -lactalbumin, and aprotinin) that included proteins with accessible, reduced cysteine thiols (TIM, CALP) and those with only buried or oxidized cysteine residues that were expected to serve as

146 negative controls (myoglobin, -lactalbumin, aprotinin)32 were subjected to the standard modification conditions. Western blot analysis indicated TIM and CALP were the primary proteins modified (Fig. 6.8 a). This reactivity profile was similar to that observed with a commercial biotin-maleimide reagent (Fig. 5b).

A brief discussion regarding chemoselectivity is warranted.

Metallocarbenes intermediates derived from rhodium(II) complexes catalyze modification of tryptophan as well as cysteine.22,23 The tryptophan modification reaction was originally discovered at acidic pH (≤4), and we, like those authors,23 have confirmed that cysteine is unreactive under those conditions. Later, tert-butyl hydroxylamine hydrochloride was found to be a unique buffer that allowed tryptophan modification at more neutral pH, albeit with some loss in reaction rate,22 and others have reported tryptophan modifications in the presence of cysteine residues.33 We employ those tert-butyl hydroxylamine hydrochloride conditions to achieve cysteine modification. Thus, it is possible to select for tryptophan reactivity on the basis of reaction pH, and the presence of free cysteine thiols should be considered when choosing reaction conditions. The reversible S- sulfate formation demonstrated here represents another solution to this potential difficulty. Alternatively, the presence of a solvent-accessible tryptophan residue may limit the utility of this cysteine modification protocol. However, we have not found modification of tryptophan to be a problem in practice. Tryptophan is a rare amino acid that is typically buried in hydrophobic pockets; its modification often requires denaturing conditions.22 For example, both myoglobin (PDB ID: 3LR7)

147 and α-lactalbumin (PDB ID: 1HFZ) contain (buried) tryptophan residues, yet neither is modified here (Figure 6.8).

6.7. Conclusions

We describe a new cysteine modification reaction for metallocarbene intermediates and identify new chemoselectivity considerations for the use of rhodium-catalyzed methods for protein modification with diazo reagents. Acting directly on natural cysteine residues, this reaction provides products similar to those from the Kirmse–Doyle reaction,17 but does so in a direct, 1-step process. Product profiles among several substrate proteins indicate that the metal-catalyzed process may have different selectivity considerations based on local steric and electronic factors, making the method described here of interest when selective modification of proteins with multiple cysteine residues is required. The serum stability of the product linkage should make diazo reagents a practical alternative for applications in cellular environments.

6.8. Experimental

6.8.1. General Considerations.

Solvents and reagents were purchased from Fisher Scientific (unless mentioned otherwise) and used as received. Millipore ultra-purified water (18

MΩ) was used in all cases.

Peptide synthesis. All peptides were manually synthesized using standard solid-phase Fmoc protocols.34,35 Rink amide MBHA resin (AAPPTEC) to afford

148

C-terminal amides. The peptides were acetylated at the N-terminus prior to cleavage from the resin. The purification was accomplished by reverse-phase

HPLC with gradients of water-acetonitrile containing 0.1% trifluoroacetic acid, and peptides were isolated by lyophilization. Analysis and purity assessment was attained by mass spectrometry and analytical HPLC.

HPLC. HPLC was performed on a Shimadzu CBM-20A instrument with

Phenomene Jupiter 4μ Proteo 9 A (25 × 15 mm preparati e) and Phenomenex

Jupiter 4μ Proteo 9 A (25 × 4.6 mm analytical) columns. Flow rates of 8 mL/min and 1 mL/min were used for preparative and analytical columns, respectively.

Analytical and preparative HPLC were performed with gradient of acetonitrile in water. Both solvents contained 0.1% trifluoroacetic acid (TFA) unless otherwise noted. Data was collected using UV-vis absorption at 220 nm and 300 nm.

Mass Spectrometry. MALDI-TOF MS and MS/MS analyses were performed on a Bruker Daltonics Autoflex MALDI-TOF/TOF mass spectrometer with α-cyano-4-hydroxycinnamic acid matrix (CHCA, Thermo Scientific, 10 mg/mL) for peptide samples and 2,4-dihydroxyacetophenone (DHAP, Fluka, 10 mg/mL) for protein samples. ESI-MS was performed on a Bruker Daltonics microTOF instrument. Data analysis was performed on mMass program.36,37

149

Synthesis of known compounds. The protein CALP38 and biotin diazo20

reagent 1 were prepared and purified according to published procedures.

6.8.2. Modification of CALP (10 M):

To a 250 L microcentrifuge tube, 68.7 L of millipore water was added, along with 10 L aqueous solution of tert-butyl hydroxylamine hydrochloride

(100 mM, pH 6.2). To this 10 L of CALP (100 M) was added, followed by the addition of 1 L aqueous solution of Rh2(OAc)4 (1 mM). The reaction mixture was vortexed briefly and 1.7 L t-butanol solution of the biotin-diazo 1 (60 mM) was added in the end. The reaction mixture was centrifuged and then incubated on a laboratory low speed shaker at 4 ᵒC for 5 h. The reaction mixture was quenched using 20 L of MeCN and then analyzed using MALDI-TOF. The rhodium catalyst can be removed along with other small molecules like biotin diazo by desalting the reaction mixture using a μC18 ZipTip® pipet tip.

6.8.3. General Procedure for Trypsin Digestion of Modified CALP and MS/MS Analysis of Modified CALP Peptides:

100 L of the CALP (20 M) modification reaction mixture was centrifuged and then heated at 65 ᵒC in a water bath for 30 min. This was diluted with 100 L aqueous solution of NH4HCO3 (80 mM, pH 7.8) followed by the addition of 1.1 L of trypsin solution (Promega, 100 g reconstituted in 100 L of

50 mM acetic acid). The digest was incubated at 37 ᵒC for 6 h. The crude digest

150 was desalted using μC18 ZipTip® pipet tip and then analyzed by MALDI-TOF.

After analyzing the digestion data using mMass program,36,37 MS/MS was performed on the singly modified peptide fragment m/z = 2441 using MALDI-

TOF.

6.8.4. General Procedure for Blocking Cysteine residue on CALP:31

In a 2 mL centrifuge tube, 3 mg of lyophilized CALP was taken and dissolved in 1.5 ml Na2SO3 buffer* and incubated at 25 ᵒC for 1 h keeping the tube open. The blocked protein solution is then desalted by dialyzing it in PDZ buffer* (pH 7.4).

*Na2SO3 buffer recipe– 0.05 M Na2SO3, 0.2 mM cysteine, 6 M guanidine hydrochloride, 0.1 M Tris chloride (pH 8.4).

PDZ buffer recipe – 5% glycerol, 25 mM Na3PO4, 150 mM NaCl, 0.02%

NaN3, 0.1 mM TCEP (pH 7.4).

6.8.5. Control Reactions for Modification of CALP:

In absence of Rh2(OAc)4: To a 250 L microcentrifuge tube, 69.7 L of millipore water was added, along with 10 L aqueous solution of tert-butyl hydroxylamine hydrochloride (100 mM, pH 6.2). To this 10 L of CALP (100

M) was added. The reaction mixture was vortexed briefly and 1.7 L t-butanol solution of the biotin-diazo 1 (60 mM) was added in the end. The reaction mixture was centrifuged and then incubated on a laboratory low speed shaker at 4 ᵒC for 5

151 h. The reaction mixture was quenched using 20 L of MeCN and then analyzed using MALDI-TOF.

In absence of diazo 1: To a 250 L microcentrifuge tube, 70.4 L of millipore water was added, along with 10 L aqueous solution of tert-butyl hydroxylamine hydrochloride (100 mM, pH 6.2). To this 10 L of CALP (100

M) was added, followed by the addition of 1 L aqueous solution of Rh2(OAc)4

(1 mM). The reaction mixture was centrifuged and then incubated on a laboratory low speed shaker at 4 ᵒC for 5 h. The reaction mixture was quenched using 20 L of MeCN and then analyzed using MALDI-TOF.

6.8.6. General Protocol for the Modification of Peptides (25  :

To a 250 L microcentrifuge tube, 85.3 L of millipore water was added, along with 10 L aqueous solution of tert-butyl hydroxylamine hydrochloride

(100 mM, pH 6.2). To this 2.5 L of peptide (1 mM) was added, followed by the addition of 0.25 L aqueous solution of Rh2(OAc)4 (1 mM). The reaction mixture was vortexed briefly and 2 L t-butanol solution of the biotin-diazo 1 (60 mM) was added in the end. The reaction mixture was centrifuged and then incubated on a laboratory low speed shaker at 4 ᵒC for 5 h. The reaction mixture was quenched using 20 L of MeCN and then analyzed using MALDI-TOF.

152

6.8.7. General protocol for the Modification of Peptide AQPADRCGGLA using Biotin-Maleimide:

To a 500 L centrifuge tube, 314 L of millipore water was added, along with 36.4 L of buffer*. To this 9.1 L of AQPADRCGGLA (1 mM) was added, followed by the addition of 4.55 L of Biotin-maleimide (Sigma Aldrich, 1 mM solution in DMF). The reaction mixture was centrifuged and then incubated on a laboratory low speed shaker at 4 ᵒC for 5 h. The reaction mixture was analyzed using MALDI-TOF. This was lyophilized and used in the stability study in human plasma serum. *Buffer recipe – 100 mM Na3PO4, 5 mM EDTA, pH 7.5.

6.8.8. General Protocol for Stability Study of S-C Linkages in Human Plasma Serum:

To a 500 L centrifuge tube containing the lyophilized modified peptide,

200 L of human plasma serum (Fischer Scientific) was added, along with 2 L of

PMSF protease inhibitor (phenylmethanesulfonylfluoride, 100 mM solution in isopropanol). The reaction mixture was centrifuged and then incubated at 37 ᵒC.

At different time intervals, 10 L aliquots of the reaction mixture were taken out and quenched with 60 L of MeCN. This was centrifuged and the solution was analyzed using MALDI-TOF.

153

6.8.9. Rh2(OAc)4 Catalyzed Modification of CALP in a Mixture of Other Proteins:

To a 250 L microcentrifuge tube, 67.95 L of millipore water was added, along with 10 L aqueous solution of tert-butyl hydroxylamine hydrochloride

(100 mM, pH 6.2). To this 15 L of CALP (100 M) was added, followed by the addition of 3 L of BioRad polypeptide standard†. The reaction mixture was vortexed briefly and 1.5 L aqueous solution of Rh2(OAc)4 (1 mM) was added to it. Finally 2.55 L of biotin-diazo 1 (60 mM solution in t-butanol) was added. The reaction mixture was centrifuged and then incubated on a laboratory low speed shaker at 4 ᵒC for 5 h. The reaction mixture was quenched using 20 L of MeCN and then analyzed using Western blot.

6.8.10. Biotin-Maleimide Catalyzed Modification of CALP in a Mixture of Other Proteins:

To a 250 L microcentrifuge tube, 89 L of buffer* was added, along with

3 L of CALP (500 M) and 3 L of BioRad polypeptide standard†. The reaction mixture was vortexed briefly and 5 L of biotin-maleimide (Sigma Aldrich, 5 mM solution in DMF) was added to it. The reaction mixture was centrifuged and then incubated on a laboratory low speed shaker at 4 ᵒC for 5 h. The reaction mixture was quenched using 20 L of MeCN and then analyzed using Western blot.

*Buffer recipe – 100 mM Na3PO4, 5 mM EDTA, pH 7.5.

154

† 3 L of Biorad polypeptide standard contains 13.5 g of the proteins aprotinin,-lactalbumin, myoglobin and triose-phosphate isomerase (TIM).

6.8.11. Western Blot Analysis:

A crude reaction sample (0.50 L) was analyzed directly by SDS-PAGE

(10% Tris-Tricine gel, Invitrogen). Proteins were then transferred to a nitrocellulose membrane (GE Healthcare), which was subsequently blocked in 8% milk (8% milk in 100 mL TBS-T*) for 2 h. The membrane was incubated with avidin–horseradish peroxidase conjugate (blotting grade, Bio-Rad) solution (1% in

TBS-T) for 1 h, followed by washing with TBS-T (3 × 5 min) and TBS (1×15 min). Blots were developed with chemiluminescent substrate (RapidStep ECL

Reagent, Calbiochem) and images recorded with a Fujifilm LAS-4000 instrument.

An estimate of the extent of biotinylation was obtained by comparison to control lanes of authentic HRP-biotin standard (Invitrogen) and quantified using Multi

Gauge Version 3.0.

*TBS recipe: 2.4 g tris base, 29.22 g NaCl, was dissolved in 900 mL water, and adjusted to pH 7.5 and then the volume was made up to 1 L.

TBS-T recipe: 1 mL of Tween 20 in 500 mL of TBS solution.

155

6.8.12. Characterization for AQPADRXCGGLA (X= R, K, A) Peptides:

SPD-20A Ch2-220nm

RK-4-18-test run-purity check-RK-analytical-20m.met-1mM-15ul.dat

0 1 2 3 4 5 6 7 8 9 10 11 Minutes

1099

) )

a.u Intensity ( Intensity

900 1000 1100 1200 1300 1400 Mass (m/z)

Figure 6.9 a) HPLC trace of pure peptide AQPADRCGGLA and b) MALDI-MS of purified peptide AQPADRCGGLA.

156

SPD-20A Ch2-220nm

RK-4-35-1-KCGGLA-purity check-1 mM soln

1 2 3 4 5 6 7 8 9 10 Minutes

1071

) )

a.u Intensity ( Intensity

250 500 750 1000 1250 1500 1750 2000 2500 Mass (m/z)

Figure 6.10 a) HPLC trace of pure peptide AQPADKCGGLA and b) MALDI-MS of purified peptide AQPADKCGGLA.

157

SPD-20A Ch2-220nm RK-4-35-3-ACGGLA-purity check-1 mM soln. RK-4-35-3-ACGGLA-purity check-1 mM soln

1 2 3 4 5 6 7 8 9 10 Minutes

1015

) )

a.u Intensity ( Intensity

500 750 1000 1250 1750 Mass (m/z)

Figure 6.11 HPLC trace and ESI-MS of peptide AQPADACGGLA

6.9. References

(1) Kundu, R.; Ball, Z. T.: Rhodium-catalyzed cysteine modification with diazo reagents. Chem. Commun. 2012, DOI:10.1039/c2cc37323h. (2) Baslé, E.; Joubert, N.; Pucheault, M.: Protein Chemical Modification on Endogenous Amino Acids. Chem. Biol. 2010, 17, 213-227. (3) Chalker, J. M.; Bernardes, G. J. L.; Lin, Y. A.; Davis, B. G.: Chemical Modification of Proteins at Cysteine: Opportunities in Chemistry and Biology. Chem. Asian J. 2009, 4, 630-640. (4) Holmes, T. J.; Lawton, R. G.: Cysteine modification and cleavage of proteins with 2-methyl-N1-benzenesulfonyl-N4-bromoacetylquinonediimide. J. Am. Chem. Soc. 1977, 99, 1984-1986.

158

(5) Bernardes, G. J. L.; Chalker, J. M.; Errey, J. C.; Davis, B. G.: Facile Conversion of Cysteine and Alkyl Cysteines to Dehydroalanine on Protein Surfaces: Versatile and Switchable Access to Functionalized Proteins. J. Am. Chem. Soc. 2008, 130, 5052-5053. (6) Crich, . rishnamurthy, V. Hutton, T. .: Allylic Selenosulfide Rearrangement: A Method for Chemical Ligation to Cysteine and Other Thiols. J. Am. Chem. Soc. 2006, 128, 2544- 2545. (7) Crich, D.; Krishnamurthy, V.; Brebion, F.; Karatholuvhu, M.; Subramanian, V.; Hutton, T. .: echalcogenative Allylic Selenosulfide and isulfide Rearrangements: Complementary Methods for the Formation of Allylic Sulfides in the Absence of Electrophiles. Scope, Limitations, and Application to the Functionalization of Unprotected Peptides in Aqueous Media. J. Am. Chem. Soc. 2007, 129, 10282-10294. (8) Lewis, M. R.; Shively, J. E.: Maleimidocysteineamido-DOTA erivatives: ew Reagents for Radiometal Chelate Conjugation to Antibody Sulfhydryl Groups Undergo pH- Dependent Cleavage Reactions. Bioconjugate Chem. 1998, 9, 72-86. (9) Shen, B.-Q.; Xu, K.; Liu, L.; Raab, H.; Bhakta, S.; Kenrick, M.; Parsons-Reponte, K. L.; Tien, J.; Yu, S.-F.; Mai, E.; Li, D.; Tibbitts, J.; Baudys, J.; Saad, O. M.; Scales, S. J.; McDonald, P. J.; Hass, P. E.; Eigenbrot, C.; Nguyen, T.; Solis, W. A.; Fuji, R. N.; Flagella, K. M.; Patel, D.; Spencer, S. D.; Khawli, L. A.; Ebens, A.; Wong, W. L.; Vandlen, R.; Kaur, S.; Sliwkowski, M. X.; Scheller, R. H.; Polakis, P.; Junutula, J. R.: Conjugation site modulates the in vivo stability and therapeutic activity of antibody-drug conjugates. Nat. Biotechnol. 2012, 30, 184-189. (10) Stephanopoulos, N.; Francis, M. B.: Choosing an effective protein bioconjugation strategy. Nat. Chem. Biol. 2011, 7, 876-884. (11) Nielsen, M. L.; Vermeulen, M.; Bonaldi, T.; Cox, J.; Moroder, L.; Mann, M.: Iodoacetamide-induced artifact mimics ubiquitination in mass spectrometry. Nat Meth 2008, 5, 459-460. (12) Dahl, K. H.; McKinley-McKee, J. S.: The reactivity of affinity labels: A kinetic study of the reaction of alkyl halides with thiolate anions—a model reaction for protein alkylation. Bioorg. Chem. 1981, 10, 329-341. (13) Antos, J. M.; Francis, M. B.: Transition metal catalyzed methods for site-selective protein modification. Curr. Opin. Chem. Biol. 2006, 10, 253-262. (14) Popp, B. V.; Chen, Z.; Ball, Z. T.: Sequence-specific inhibition of a designed metallopeptide catalyst. Chem. Commun. 2012, 48, 7492-7494. (15) Yan, L. Z.; Dawson, P. E.: Synthesis of Peptides and Proteins without Cysteine Residues by Native Chemical Ligation Combined with Desulfurization. J. Am. Chem. Soc. 2001, 123, 526-533. (16) Lin, Y. A.; Chalker, J. M.; Floyd, N.; Bernardes, G. a. J. L.; Davis, B. G.: Allyl Sulfides Are Privileged Substrates in Aqueous Cross-Metathesis: Application to Site-Selective Protein Modification. J. Am. Chem. Soc. 2008, 130, 9642-9643. (17) Crich, D.; Zou, Y.; Brebion, F.: Sigmatropic Rearrangements as Tools for Amino Acid and Peptide Modification: Application of the Allylic Sulfur Ylide Rearrangement to the Preparation of Neoglycoconjugates and Other Conjugates. J. Org. Chem. 2006, 71, 9172-9177. (18) Chen, Z.; Vohidov, F.; Coughlin, J. M.; Stagg, L. J.; Arold, S. T.; Ladbury, J. E.; Ball, Z. T.: Catalytic Protein Modification with Dirhodium Metallopeptides: Specificity in Designed and Natural Systems. J. Am. Chem. Soc. 2012, 134, 10138-10145. (19) Popp, B. V.; Ball, Z. T.: Proximity-driven metallopeptide catalysis: Remarkable side-chain scope enables modification of the Fos bZip domain. Chem. Sci. 2011, 2, 690-695. (20) Chen, Z.; Popp, B. V.; Bovet, C. L.; Ball, Z. T.: Site-Specific Protein Modification with a Dirhodium Metallopeptide Catalyst. ACS Chem. Biol. 2011, 6, 920-925.

159

(21) Popp, B. V.; Ball, Z. T.: Structure-selective modification of aromatic side chains with dirhodium metallopeptide catalysts. J. Am. Chem. Soc. 2010, 132, 6660-6662. (22) Antos, J. M.; McFarland, J. M.; Iavarone, A. T.; Francis, M. B.: Chemoselective Tryptophan Labeling with Rhodium Carbenoids at Mild pH. J. Am. Chem. Soc. 2009, 131, 6301- 6308. (23) Antos, J. M.; Francis, M. B.: Selective tryptophan modification with rhodium carbenoids in aqueous solution. J. Am. Chem. Soc. 2004, 126, 10256-10257. (24) Kundu, R.; Cushing, P. R.; Popp, B. V.; Zhao, Y.; Madden, D. R.; Ball, Z. T.: Hybrid Organic–Inorganic Inhibitors of a PDZ Interaction that Regulates the Endocytic Fate of CFTR. Angew. Chem. Int. Ed. 2012, 51, 7217-7220. (25) Anthony Mc ervey, M. Ratananu ul, P.: Regiospecific synthesis of α- (phenylthio) etones via rhodium(II) acetate catalysed addition of thiophenol to α-diazoketones. Tetrahedron Lett. 1982, 23, 2509-2512. (26) Moody, C. J.; Taylor, R. J.: Rhodium carbenoid mediated cyclisations. Use of ethyl lithiodia oacetate in the preparation of ω-hydroxy-,-mercapto-, and -boc-amino-α-diazo-β- keto esters. Tetrahedron Lett. 1987, 28, 5351-5352. (27) Paulissen, R.; Hayez, E.; Hubert, A. J.; Teyssie, P.: Transition metal catalysed reactions of diazocompounds - part III a one-step synthesis of substituted furanes and esters. Tetrahedron Lett. 1974, 15, 607-608. (28) Moyer, M. P.; Feldman, P. L.; Rapoport, H.: Intramolecular nitrogen-hydrogen, oxygen-hydrogen and sulfur-hydrogen insertion reactions. Synthesis of heterocycles from .alpha.-diazo .beta.-keto esters. J. Org. Chem. 1985, 50, 5223-5230. (29) Cushing, P. R.; Vouilleme, L.; Pellegrini, M.; Boisguerin, P.; Madden, D. R.: A Stabilizing Influence: CAL PDZ Inhibition Extends the Half-Life of ΔF508-CFTR. Angew. Chem. Int. Ed. 2010, 49, 9907-9911. (30) Vouilleme, L.; Cushing, P. R.; Volkmer, R.; Madden, D. R.; Boisguerin, P.: Engineering Peptide Inhibitors To Overcome PDZ Binding Promiscuity. Angew. Chem. Int. Ed. 2010, 49, 9912-9916. (31) Crankshaw, M. W.; Grant, G. A.: Modification of cysteine. Curr. Protoc. Protein Sci. / editorial board, John E. Coligan [et al.] 2001, Chapter 15, 15.01.09. (32) Protein structures: PDB IDs: 1B0C, 1HFZ, 3LR7, 1R2R, 2DC2. (33) Bao, Z. Wang, S. Shi, W. ong, S. Ma, H.: Selective Modification of Trp19 in β- Lactoglobulin by a New Diazo Fluorescence Probe. J. Proteome Res. 2007, 6, 3835-3841. (34) Wellings, D. A.; Atherton, E.: In Solid-Phase Peptide Synthesis 1997, 289, 44-67. (35) Wellings, D. A.; Atherton, E.: Standard Fmoc protocols. In Methods Enzymol 1997, 289, 44-67. (36) Strohalm, M. Hassman, M. ošata, B. od če , M.: mMass data miner: an open source alternative for mass spectrometric data analysis. Rapid Commun. Mass Spectrom. 2008, 22, 905-908. (37) Strohalm, M. avan, . ov , P. Voln , M. Havl če , V. r.: mMass 3: A Cross- Platform Software Environment for Precise Analysis of Mass Spectrometric Data. Anal. Chem. 2010, 82, 4648-4651. (38) Amacher, J. F.; Cushing, P. R.; Weiner, J. A.; Madden, D. R.: Crystallization and preliminary diffraction analysis of the CAL PDZ domain in complex with a selective peptide inhibitor. Acta Crystallographica Section F 2011, 67, 600-603.

160

Appendix A: Selected Spectra for Chapter 3

161

2-Methyl-2-undecyl hydroperoxide (1c)

9 8

0.77 7 6 5 4 3 2

2.07 14.30

5.80 1 3.00 0 ppm

162

180 160 140 120 100

80 83.19 77.55 77.23 76.91 60

40 38.73 32.11 30.40 29.85 29.80 29.55

20 24.22 24.03 22.90 14.34 ppm

163

4-Phenyl-1-butyl-methanesulfonate

9 8

5.41 7 6 5

1.85 4 3 2.94 2.00 2

3.88 1 0 ppm

164

180 160 140 141.74

128.65 128.61 126.22 120 100 80 77.55 77.24 76.92 70.05 60 40

37.59 35.39

28.83 27.40 20 ppm

165

4-Phenylbutyl hydroperoxide (1e)

10 9 8 0.92

2.56

2.94 7 6

0.03 5

0.07 2.00 4 0.02 0.04

0.03 3 0.04 2.10

0.08 2 4.08 1.10 0.10 1 0 ppm

166

200 180

77.6

77.4 77.49 77.2 160 77.23

77.11 77.0 76.98 ppm

140 142.30

128.61 128.53

120 126.02 100

80 77.49 77.23 77.11 76.98 60 40

35.79

27.85 27.33 20 ppm

167

5-Phenyl, 2-pentyl hydroperoxide (1g)

9 8

0.81

2.12

3.01 7 6 5

0.99 4 3

2.00 2

3.07 1.05 3.07 1 0 ppm

168

180 160

140 142.40

128.59 128.51 125.98 120 100

80 81.68 77.55 77.23 76.91 60 40

36.03 33.70

27.31 20

18.34 ppm

169

2-Hydroperoxy-ethyloctanoate (1h)

0.66 9 8 7 6 5

0.72 1.91 4

0.17 3 2

2.05 10.89

0.45 1 3.00 0 ppm

170

180

172.65 160 140 120 100

83.75 80 77.48 77.23 76.97

60 61.54 40 31.70 30.21 29.11

20 25.39 22.72 14.43 14.24 ppm

171

2-Hydroperoxy-octanol

9

0.55 8 7 6 5

0.77 4 0.73 0.68 3 2

10.69 1 3.00 0 ppm

172

173

2-Hydroperoxy-octyl-acetate (1i) 10

0.92 9 8 7 6 5

0.99

1.02 4 0.99 3

3.00 2

1.18 9.73 1 3.24 0 ppm

174

180

173.31 160 140 120 100

83.51 80 77.49 77.24 76.98

60 63.02 40 31.86 29.35 28.75 25.79 20 22.78 21.19 14.27 ppm

175

Ethyl-6-oxo-heptanoate

9 8 7 6 5

2.00 4 3

2.22 1.95

3.26 2

4.43

3.05 1 0 ppm

176

208.91 200 180

173.63 160 140 120 100 80 77.55 77.24 76.92 60 60.48

40 43.42

34.21 30.04 24.53 20 23.31

14.38 0 ppm

177

Ethyl-6-methyl-6-hydroxy-heptanoate

10 9 8 7 6 5

2.00 4 3

2.04 2

2.10 5.08 3.00 5.84 1 0 ppm

178

200 180

173.94 160 140 120 100 80 77.49 77.23 76.98 71.00 60 60.41

40 43.65

34.47 29.39 25.61

20 24.06

14.42 ppm

179

Ethyl-6-methyl-6-hydroxy-heptanoate (1k)

8.0 7.5 0.90 7.0 6.5 6.0 5.5 5.0 4.5

2.00 4.0 3.5 3.0 2.5

2.05 2.0

5.48 1.5 2.28 3.15 5.98 1.0 0.5 ppm

180

180

173.94 160 140 120 100

82.54 80 77.34 77.02 76.70 60 60.36 40 37.54 34.05 25.25

20 23.92 23.05 14.25 ppm 0.01

181

4-Methoxy-2-methyl-2-octanol

8 7 6 5

0.85 4

0.95 2.83 3 2

7.45 6.50 1 3.00 0 ppm

182

160 140 120 100

79.55 80 77.55 77.23 76.92 70.39 60

55.77

45.87 40

32.66 31.30 28.43 26.90 20 23.12 14.24 ppm

183

4-Methoxy-2-methyl-2-octyl hydroperoxide (1m)

10 0.95 9 8 7 6 5 4

4.05 3

1.07 2

11.38

3.00 1 3.17 ppm

184

160 140 120 100

80.92

80 78.42 77.23 76.98 76.73 60

56.42

40 42.54

32.66 26.70 26.16

20 23.77 22.84 14.02

ppm -0.04

185

4-Chloro-decanol (2a):

9 8 7 6 5 4 0.97 2.00 3

0.04 2

16.52 1 3.00 0 ppm

186

170 160 150 140 130 120 110 100 90 80 77.49 77.23 76.98 70

64.21

60 62.59 50 40

38.83 34.98 31.90 30 29.86 29.03 26.65

20 22.79

14.27 10 ppm

187

5-Chloro-2-octanol (2b): 7.0 6.5 6.0 5.5 5.0 4.5 4.0 0.93 1.03 3.5 3.0 2.5 2.0

9.83 1.5

3.34 1.0 3.00 0.5 ppm

188

170 160 150 140 130 120 110 100 90 80 77.42 77.16 76.91 70 68.08 67.57 64.25

60 63.82 50

40.83 40 40.72 36.37 35.88 35.06 30 34.52

23.87 23.84 20 19.84 19.79

13.69 ppm

189

10 5-Chloro-2-methyl, 2-undecanol (2c): 9

8 7 6 5 4 0.88 3

0.05 2

15.45 5.57

0.60 1 3.00 0 ppm

190

180 160 140 120 100 80 77.53 77.28 77.02 70.92

64.98 60

40.71 40 38.86 33.49 31.96 29.96 29.37 29.09 20 26.72 22.85 14.32 ppm

191

4-Chloro-4-methyl-pentanol (2d):

9 8 7 6 5 4

1.99 3 2

4.14 6.00 1 0 ppm

192

200 180 160 140 120 100 80 77.49 77.23 76.98 70.96

60 63.06

40 42.42

32.69 28.70 20

0.22 ppm

193

4-Chloro-4-phenyl butanol (2e)

10 9 8

5.22 7 6 5 1.00 4

1.94 3

2.02 2 1.14 1.13 1.38

0.34 1 0.29 0 ppm

194

180 160 140 141.90

128.87 128.52 127.13 120 100 80 77.49 77.23 76.98

63.78 60 62.42 40

36.67

30.41 20 ppm 0.22

195

2-Phenyltetrahydrofuran (5e)

9 8

4.37

0.91 7 6 5 1.00

1.03 4 1.02 3

1.05

2.19 2 1.05

0.34 1 0 ppm

196

180 160

140 143.67

128.50 127.33

120 125.84 100

80 80.89 77.49 77.23 76.98 68.89 60 40

34.83

26.25 20 ppm

197

5-Chloro-5-methyl, 2-hexanol (2f):

9 8 7 6 5 4

1.00 3

1.03 2 3.12 5.66 1.31 0.32

0.97 1 3.10 ppm

198

160 140 120 100

80 77.34 77.22 77.02 76.70 70.87 68.14 60

40 42.05

34.72 32.70 32.29

23.74 20

0.01 ppm

199

5-Chloro-5-phenyl-2-pentanol (2g)

9 8

5.21

0.34 7 6 5 1.00 4 1.08 0.02 3

0.15

2.23 2

4.15 0.66 3.08 0.20 1 0.28 0 ppm

200

160 140 141.75

128.67 128.42 128.30 126.95 120 126.92 100 80 77.36 77.04 76.72 67.75 67.43 63.96 60 63.67 40 36.66 36.45 36.40 36.18

23.76

20 23.72 0 ppm

201

2-Methyl-5-phenyltetrahydrofuran (5g)

9 8

7.69

1.95 7 6

1.02 5 0.94

1.04

0.99 4 3

2.26

2.23 2 2.11 2.01 3.00 3.17 1 0 ppm

202

200 180 160

144.22 140 143.73 128.50 128.46 127.34 127.24 120 126.06 125.78 100

81.25 80 80.47 76.19 76.13 60 40 35.82 34.86 34.50 33.31

20 21.77 21.55 ppm

203

5-Chloro-2-hydroxy-ethyloctanoate (2h)

8 7 6 5

2.08

1.28 4 1.04 3

0.48

0.78 2

9.42

4.05 1 3.00 0 ppm

204

180

175.29 175.25 160 140 120 100 80

70.37 69.88 63.70

60 63.39 62.12

40 40.89 40.79 34.21 33.62 31.68 31.37

20 19.92 19.86 14.43 13.76 ppm

205

5-Chloro-1, 2-octanediol-l-acetate (2i)

8 7 6 5

1.00 4 3.09 3

3.36 2 1.23 5.41 2.87 0.23 1 3.00 0 ppm

206

180

171.45 160 140 120 100

80 77.52 77.27 77.01 70.15 69.47 68.99 68.94 60 64.15 63.47

40 40.99 40.84 34.86 34.18 30.79 30.15

20 21.13 19.98 19.91 13.80 ppm

207

2, 2-Dimethyl-5-phenyltetrahydrofuran (5j)

9 8

5.64 7 0.08 6 5 0.98 4 0.03 0.10 3

0.11 1.16 0.15

2.98 2 0.21 0.19 2.71

0.38 1 3.00 0.47 0 ppm

208

180 160

140 143.85

128.46 127.30 120 126.02 100

81.54

80 80.69 77.55 77.24 76.93 60

40 39.28 35.87 29.24 28.62 20 0 ppm

209

3-Chloro-6-methyl-6-hydroxy-ethylheptanoate (2k)

8 7 6 5

1.00

2.09 4 3

1.93

0.95 2 2.26 1.09 0.92 3.35 6.72 1 0.78 0 ppm

210

180

170.39 160 140 120 100

80 77.52 77.27 77.01 70.68

60 61.15 58.66

43.95 40 40.40

33.06 29.88 29.37 20

14.40 ppm

211

1-Chloro-4-heptanol (2l)

9 8 7 6 5 4 0.93 1.73 3

0.96 2 0.95 1.13 5.60 1 2.88 0 ppm

212

213

5-Chloro-4-methoxy-2-methyl-2-octanol (2m)

10 9 8 7 6 5

1.21

0.79 4 1.02 1.23 5.96

1.96 3 0.74 2

14.90 12.96

0.75 1 6.00 0 ppm

214

180

171.45 160 140 120 100

80 77.52 77.27 77.01 70.15 69.47 68.99 68.94 60 64.15 63.47

40 40.99 40.84 34.86 34.18 30.79 30.15

20 21.13 19.98 19.91 13.80 ppm

215

(2-Hydroxy-2-methyloctan-4-yloxy)methyl acetate (6m):

(NMR data in C6D6)

9 8 7 6

1.73 5 4 0.92

0.20 0.09 3 0.07 2

3.80 5.22 7.08 3.71 1 3.00

0.46 0 ppm

216

180

169.68 160 140 120 100

86.35 80

76.79

69.41 60

46.51 40

34.47 31.00 28.76 26.80

20 22.91 20.32

13.96 ppm

217

6-Butyl-4,4-dimethyl-1,3-dioxane (7m)

9 8 7 6 5 1.71 4

0.82 3 2

6.71 3.56

2.85 1 3.00 0 ppm

218

200 180 160 140 120 100

88.10 80

72.86 71.47 60

40 42.73

36.08 31.85 27.43 22.90 20 21.83

14.25 ppm 0.21

219

Appendix B. Selected Spectra for Chapter 5

220

(S)-(S)-5-methoxy-5-oxo-1-phenylpentyl 1-(benzylsulfonyl) piperidine-2- carboxylate (5.4-S)

8 7 6 5 4 3 2 1 0 ppm

0.98 0.98 0.10 2.39 3.00 1.00 0.98 1.99 0.99 1.02 1.02 5.02 1.06 0.99

10.21

173.91 171.07 139.78 131.14 129.35 128.86 128.70 128.52 126.70 58.84 56.28 51.86 43.62 35.77 33.66 28.03 25.19 21.06 20.51

200 180 160 140 120 100 80 60 40 20 ppm

221

(S)-(R)-5-methoxy-5-oxo-1-phenylpentyl 1-(benzylsulfonyl) piperidine-2- carboxylate (5.4-R)

8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm

1.00 0.99 1.74 0.40 3.00 1.03 1.01 2.06 1.04 1.15 1.11 6.95 0.96

10.24

173.76 170.99 140.05 131.20 131.16 129.51 128.87 128.82 128.76 128.68 128.46 126.72 126.60 58.91 56.13 51.83 43.49 35.80 33.71 28.02 25.14 21.12 20.54 20.32

180 160 140 120 100 80 60 40 20 0 -20 ppm

222

(S)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl)oxy)-5-phenylpentanoic acid (5.5-S)

9 8 7 6 5 4 3 2 1 0 ppm

1.14 2.71 1.00 1.83 1.00 1.01 2.02 1.00 1.25 1.05 5.10 1.14 1.05

10.09

187.19 178.91 171.13 139.68 131.13 129.30 128.91 128.74 128.59 126.70 116.61 58.92 56.34 43.68 35.66 33.51 27.99 25.15 20.76 20.50

180 160 140 120 100 80 60 40 20 ppm 223

(R)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl)oxy)-5-phenylpentanoic acid (5.5 R)

8 7 6 5 4 3 2 1 0 ppm

2.01 1.00 1.00 1.69 0.34 1.00 0.99 2.07 1.07 1.15 1.02 6.36 0.82

10.32

178.80 171.01 139.97 131.17 129.36 128.82 128.75 128.70 128.46 126.67 126.51 58.85 56.10 43.47 35.70 33.56 27.95 25.05 20.82 20.23

200 180 160 140 120 100 80 60 40 20 ppm

224

Dirhodium ligation of (S)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl) oxy) -5-phenylpentanoic acid (5.6-S)

10 9 8 7 6 5 4 3 2 1 0 ppm

0.91 0.96 2.05 1.04 1.21 6.05 0.66 0.71 0.19 2.00 1.18 0.96 2.46 0.74 4.88 10.26

10.19

193.83 192.13 191.97 171.16 139.80 131.10 129.35 128.81 128.47 126.75 59.09 56.38 43.82 36.70 35.30 27.89 25.08 23.95 23.80 21.64 20.45

180 160 140 120 100 80 60 40 20 0 ppm

225

Dirhodium ligation of (R)-5-(((S)-1-(benzylsulfonyl)piperidine-2-carbonyl) oxy)-5-phenylpentanoic acid (5.6-R)

8 7 6 5 4 3 2 1 0 ppm

2.04 7.06 1.06 0.20 0.96 0.09 0.97 6.13 2.11 0.21 0.40 1.14 1.13 0.37 2.33 2.02 1.18 2.95 5.90 8.44

1.00

194.2 192.7 192.5 171.0 140.2 131.3 129.5 128.7 128.3 126.5 115.8 59.0 56.0 43.5 36.9 35.7 28.1 25.2 24.0 23.9 21.8 20.3 3.6

200 180 160 140 120 100 80 60 40 20 ppm

226