Research Collection

Doctoral Thesis

Molecular recognition at the active site of factor Xa

Author(s): Salonen, Laura Maria

Publication Date: 2011

Permanent Link: https://doi.org/10.3929/ethz-a-006882501

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.

ETH Library Diss. ETH No. 19728

Molecular Recognition at the Active Site of Factor Xa

A dissertation submitted to the ETH ZURICH for the degree of DOCTOR OF SCIENCES

Presented by Laura Maria Salonen

M.Sc., University of Turku, Finland Born December 3rd, 1980 Citizen of Finland

Accepted on the recommendation of

Prof. Dr. François Diederich, examiner Prof. Dr. Bernhard Jaun, co-examiner Dr. Wolfgang Haap, co-examiner

Zürich 2011

Vanhemmilleni

Acknowledgements

First, I would like to thank Prof. Dr. François Diederich for giving me the opportunity to work in his group on this exciting project. I really appreciate the freedom, the possibilities, the support, and the encouragement he gave me during my thesis. I especially want to thank him for giving me the chance to write a review: this was an immense learning experience for me, and very rewarding.

I would like to thank Prof. Dr. Bernhard Jaun for accepting to be my co-examiner, and for his help in NMR analysis.

Dr. Wolfgang Haap was always extremely helpful in all factor Xa-related questions. I would also like to thank him for all his support prior to and during this thesis, and for accepting to be my co-examiner.

I would like to thank Dr. David W. Banner for the X-ray cocrystal structures, which contributed greatly to this work. I especially appreciate his enthusiasm during the many helpful discussions on this project.

For enabling the biological measurements performed during this work, I would like to thank Dr. Jacques Himber and Jean-Luc Mary. A special thanks goes to Olivier Kuster for advice and for teaching me how to perform the assays.

I would like to thank David Wechsler and Daniel Zimmerli for their great help in LC- MS-related questions and for the enantiomeric resolutions.

I would like to thank all staff members of the LOC for the valuable services, in particular: the MS service: Dr. Walter Amrein, Dr. Xiangyang Zhang, Louis Bertschi, Oswald Greter, and Rolf Häfliger for measuring the mass spectra; the NMR service: Prof. Dr. Bernhard Jaun, Dr. Marc-Olivier Ebert, Rainer Frankenstein, and Philipp Zumbrunnen for their help and recording the NMR spectra; the Micro-Laboratory for the elemental analyses; Dr. W. Bernd Schweizer and Paul Seiler for X-ray structure analysis; Thomas Mäder for his great help and support in all HPLC-related problems. Many thanks to Irma Näf for her help in all administrative issues, her kindness, and support.

I would like to thank Prof. Dr. Carlo Thilgen for his administrative help and advice on the student projects.

I would like to thank Dr. Bruno Bernet for correcting the experimental part and his advice in all questions related to it.

I would like to thank Pablo Rivera Fuentes for performing the calculations presented in this thesis.

For their patient help in all computer problems I encountered, I would like to thank Dr. Christoph Fäh, Dr. Christian Eberle, Luzi Barandun, Andri Schütz, and Daniel Fankhauser.

The students Christoph Bucher, Mareike Holland, Philip Kaib, Sandra Kienast, Moritz Hunkeler, and Matthias Knecht, who I had the opportunity to work with, I would like to thank for contributions to this work and for creating a nice working atmosphere in the lab.

I would like to thank Paolo Mombelli, Petra Fesser, Dr. Lorenzo Alonso Gómez, Dr. Matthias Vogt, Dr. Nicolas Marion, and Jesse Roose for their great help in correcting this thesis in record time.

I greatly appreciate all the discussions with Leo Hardegger and Paolo on my project, their projects, and any noncovalent interaction.

I would like to thank Nicolas and Paolo for their constant encouragement during coffee breaks, which never failed to make me feel that the goals that I set were reachable.

Working in G320 was a lot of fun! To the great atmosphere contributed Paolo, Julie Geist, Dr. Manuel Ellermann, Dr. Markus Jordan, Dr. Thomas Gottschalk, and Dr. Hai Xu. Thanks for all the people that made my long-ish stay in G327 much more pleasant: Manuel, Dr. Philipp Kohler, Dr. Kara Howes, Michael Harder, Haraldur Gardarsson, Boris Tchitchanov, Dr. Valentina Aureggi, and Julie.

I thank the Spanish–German lunch group, Lorenzo, Dr. Henry Dube, Dr. Martina Zürcher, Petra, Pablo, Kara, and Manuel, for making the lunch breaks a lot of fun, sometimes even educational.

I would like to thank the usual suspects for many great, late evenings!

Many people in the DCC have made the time in the group, both at and after work, so enjoyable! Many thanks to Petra, Paolo, Lorenzo, Nicolas, Pablo, Kara, Dr. Anthoni van Zijl, Henry, Dr. Bernhard Stump, Leo, Julie, Manuel, Michael H., Halli, Markus, Dr. Fabio Silvestri, Jesse, Dr. Tobias Voigt, Veronika Ehmke, Sophie Müller, Benjamin Breiten, Michael Seet, Luzi, Dr. Anna Hirsch, Dr. Anna Vogt, Dr. Simone Hörtner, Martina, Dr. Agnieszka Kraszewska, Ebi, Dr. Peter Jarowski, and Kasper Lincke.

A lot of people from other groups have also contributed to many fun moments in Zürich. Thanks foremost to Dr. Katja Chiesa, Dr. Martina Adams, Prof. Dr. Ryan Gilmour, and Dr. Karolin Geyer.

I would especially like to thank Team Grützmacher, i.e. Matthias, Amos Rosenthal, and Dr. Mónica Trincado Rodríguez, for all the lunches, evenings, excursions all around Zürich, and, most of all, their friendship.

I would like to thank Petra for her never-ending support, for the friendship, and for always being there for me.

Lorenzo, muchas gracias por todo.

Kiitos Äiti, Isä ja Iina tuesta ja kannustuksesta koko väitöskirjan ajan. Ja aina muutenkin. Olette tärkeimmät.

Publications

Laura M. Salonen, Mareike C. Holland, Philip S. J. Kaib, Wolfgang Haap, Jörg Benz, Jean-Luc Mary, Olivier Kuster, W. Bernd Schweizer, David W. Banner, and François Diederich, “Molecular Recognition at the Active Site of Factor Xa: Cation– Interactions, Stacking on Planar Peptide Surfaces, and Replacement of Structural Water”, in preparation.

Laura M. Salonen, Manuel Ellermann, François Diederich, “Aromatic Rings in Chemical and Biological Recognition: Energetics and Structures”, Angew. Chem. Int. Ed. 2011, 50, 4808–4842.

Laura M. Salonen, Christoph Bucher, David W. Banner, Wolfgang Haap, Jean-Luc Mary, Jörg Benz, Olivier Kuster, Paul Seiler, W. Bernd Schweizer, François Diederich, “Cation– Interactions at the Active Site of Factor Xa: Dramatic Enhancement upon Stepwise N-Alkylation of Ammonium Ions”, Angew. Chem. Int. Ed. 2009, 48, 811–814.

Oral and Poster Presentations

"Cation– Interactions in the S4 Pocket of Factor Xa", Laura M. Salonen, David W. Banner, François Diederich, SCS Fall Meeting 2009, Lausanne, Switzerland, 04.09.2009 (Oral Communication).

“Cation– Interactions in the S4 Pocket of Factor Xa: Dramatic Enhancement upon Stepwise N-Alkylation of Ammonium Ions”, Laura M. Salonen, François Diederich, ASMC2009 Kiev, 3rd International Symposium on Advances in Synthetic and Medicinal Chemistry, Kiev, Ukraine, 23.–27.08.2009 (Oral Communication).

“Cation– Interactions at the Active Site of Factor Xa”, Laura M. Salonen, David W. Banner, François Diederich, 2nd Symposium of the SSCI, ETH Hönggerberg, Zurich, Switzerland, 19.11.2009 (Poster Presentation). “Cation– Interactions at the Active Site of Factor Xa: Dramatic Enhancement upon Stepwise N-Alkylation of Ammonium Ions”, Laura M. Salonen, Christoph Bucher, David W. Banner, François Diederich, 1st Symposium of the SSCI, ETH Hönggerberg, Zurich, Switzerland, 28.11.2008 (Poster Presentation).

“The Cation– Interaction in the S4 Pocket of Factor Xa”, Laura M. Salonen, Kaspar Schärer, David W. Banner, François Diederich, SCS Fall Meeting 2008, Zurich, Switzerland, 11.09.2008 (Poster Presentation).

“The Quantification of Cation– Interactions in the S4 Pocket of Factor Xa”, Laura M. Salonen, Christoph Bucher, David W. Banner, François Diederich, EFMC-ISMC 2008 20th International Symposium of Medicinal Chemistry, Vienna, Austria, 31.08.– 04.09.2008 (Poster Presentation).

Table of Contents

Abbreviations...... I Summary...... IV Zusammenfassung...... VII 1. Introduction...... 1 1.1. Molecular Recognition with Biological Systems...... 1 1.2. Cation– Interactions in Chemical and Biological Recognition...... 2 1.3. The Human Blood Coagulation Cascade: Factor Xa and ...... 8 1.4. Current Therapy...... 13 1.5. Factor Xa and the Cation– Interaction...... 15 1.6. Project Goals ...... 19 2. Enhancing Binding Affinity by S1 Needle Replacement...... 21 2.1. Inhibitor Design: Exchanging the Phenylamidinium for a Neutral S1 Needle...... 21 2.2. Synthesis of the Inhibitors Bearing Neutral S1 Needles...... 25 2.3. Enzyme Assays...... 29 2.4. Biological Activities ...... 31 3. The Cation– Interaction...... 35 3.1. Quantification of Cation– Interactions in the S4 Pocket...... 35 3.2. X-Ray Cocrystal Structure of (±)-34 with Factor Xa...... 41 3.3. Identifying the Active Enantiomer of (±)-34 ...... 45 3.4. Chain Length Variation...... 46 3.5. Probing the Size of the S4 Pocket ...... 49 3.6. Methylation Series and Uncharged Control Compounds ...... 53 3.7. Further Aspects of the Cation– Interaction: Towards the Dication Ligand, Other Onium Ions, and Counteranion Effects ...... 63 4. Water Replacement in the S1 Pocket ...... 69 4.1. Water Replacement in Drug Design...... 69 4.2. Water at the Active Site of Factor Xa...... 71 4.3. Halogen– Interactions...... 72 4.4. Ligand Design ...... 75 4.5. Synthesis ...... 75 4.6. Biological Activities ...... 79

5. Stacking on Polar Peptide Surfaces...... 82 5.1. Introduction and Inhibitor Design ...... 82 5.2. Synthesis ...... 85 5.3. Biological Activities ...... 88 5.4. X-Ray Cocrystal Structures and Conformational Analysis ...... 89 6. Towards Water Replacement in the S4 Pocket...... 100 6.1. Conserved Water in the S4 Pocket ...... 100 6.2. Inhibitor Design...... 101 6.3. Synthesis ...... 103 6.4. Biological Activities ...... 105 6.5. X-Ray Cocrystal Structure of (±)-229 with Factor Xa ...... 107 7. Conclusions and Outlook...... 109 8. Experimental ...... 113 8.1. Materials and Methods...... 113 8.2. Synthetic Procedures...... 116 8.3. Enzyme Assays...... 228 8.4. Small-Molecule Crystal Structures...... 230 8.5. X-Ray Cocrystal Structures...... 232 8.6. 1H 1D-NOE Difference Spectra of (±)-78, (±)-79, and (±)-80 ...... 235 9. References...... 236 Abbreviations ______

Abbreviations

Ac acyl ACh acetylcholine

Ac2O acetic anhydride Ala alanine Arg arginine Asn asparagine Asp aspartic acid ATR attenuated total reflection Bis-Tris bis(2-hydroxyethyl)aminotris(hydroxymethyl)methane Bn benzyl Bu butyl Cbz carbobenzyloxy CC column chromatography clogD calculated logaritmic distribution coefficient COMT catechol-O-methyltransferase COSY correlated spectroscopy CSD Cambridge Structural Database Cys cysteine DABCO 1,4-diazabicyclo[2.2.2]octane DFT density functional theory DIAD diisopropyl azodicarboxylate DMAP N,N-dimethyl-4-aminopyridine DMF N,N-dimethylformamide DMP Dess-Martin periodinane DQF-COSY double quantum filtered correlation spectroscopy

EC50 half maximal effective concentration EDTA ethylenediaminetetraacetic acid EGF epidermal growth factor EP electrostatic potential Et ethyl

Et2O diethyl ether

I Abbreviations ______

EtOAc ethyl acetate EtOH ethanol N-FBSI N-fluorobenzenesulfonimide GABA -aminobutyric acid Gla -carboxyglutamic acid Gln glutamine Glu glutamate Gly glycine GPCR G-protein coupled receptor H3 histone 3 HF Hartree–Fock His histidine HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HK high-molecular-weight kininogen HP1 heterochromatin-associated protein 1 HPLC high pressure liquid chromatography HR-EI-MS high resolution electrospray ionization mass spectrometry HR-MALDI-MS high resolution matrix assisted laser desorption/ionization mass spectrometry

IC50 inhibitor concentration at which the normal enzyme activity towards a substrate is inhibited by 50% Ile isoleucine IR infrared ITC isothermal titration calorimetry

Ka association constant

Ki inhibitory constant

Km Michaelis-Menten constant LC-MS liquid chromatography–mass spectrometry LMWH low-molecular-weight Lys lysine Me methyl MeCN acetonitrile MeOH methanol

II Abbreviations ______

MP2 second order Møller–Plesset perturbation theory NAD+ nicotinamide adenine dinucleotide nBu n-butyl NCS N-chlorosuccinimide n.d. not determined Nle norleucine (2-aminohexanoic acid) NMR nuclear magnetic resonance NOE Nuclear Overhauser Effect PCC pyridinium chlorochromate PDB Protein Data Bank PEG poly(ethylene glycol) Phe phenylalanine PK prekallikrein PL phospholipids RT room temperature S substrate concentration SAM S-adenosylmethionine Ser serine SPR surface plasmon resonance TBDMS tert-butyldimethylsilyl tBu tert-butyl THF tetrahydrofuran Thr threonine TLC thin layer chromatography TMA tetramethylammonium Tris tris(hydroxymethyl)aminomethane Trp tryptophan Ts para-toluenesulfonyl Tyr tyrosine UFH unfractionated heparin Val valine VKA

III Summary ______

Summary

Molecular recognition studies are crucial to unravel noncovalent interactions in biological systems. In this thesis, a multidimensional approach was used, comprising structure- based design of enzyme inhibitors and analysis of binding and crystallographic data, to investigate and quantify individual protein–ligand interactions at the molecular level. The in-depth understanding acquired through this approach can ultimately benefit not only drug design, but also other areas of chemical research, such as supramolecular chemistry and catalyst development. Factor Xa, a central of the human blood coagulation cascade, has been shown to be an excellent model system for molecular recognition studies, and thus, was selected as a target for our studies. The used inhibitor scaffold features a tricyclic central core, which orients in an L-shaped manner the quaternary ammonium S4 vector and the chlorothienyl-isoxazolyl S1 needle in their respective pockets.

Cation– interactions O in the S4 pocket Tyr99 Gln192 S4 OH H N N Water replacement O in the S4 pocket H Phe174 HN Cys191 H O Trp215 N O S Stacking interactions Gly216 Ala190 in the S1 pocket S1 Asp189 Water replacement in the S1 pocket OH Tyr228 The cation– interaction in the aromatic S4 pocket, formed by the side chains of Tyr99, Phe174, and Trp215, was quantified by comparing the binding affinities of a

trimethylammonium inhibitor (Ki = 9 nM) and the corresponding tert-butyl inhibitor (Ki =

550 nM). The free enthalpy increment for the cation– interaction originating from the C/N+ single-atom mutation was –G = 2.5 kcal mol–1, corresponding to approximately 0.8 kcal mol–1 per aromatic ring. The X-ray cocrystal structure of the cationic inhibitor bound to factor Xa showed the quaternary ammonium ion residing in the middle of the “aromatic box” of the S4 pocket, and the chlorothienyl moiety interacting with Tyr228 in the S1 pocket.

IV Summary ______

To determine the influence of N-methylation of the terminal amine center on the magnitude of cation– interactions, inhibitors bearing primary, secondary, and tertiary ammonium moieties were prepared. A dramatic effect of the degree of methylation was found: The binding affinity increased proportionally going from the primary (Ki =

9800 nM) to the quaternary ammonium ion (Ki = 9 nM), by a factor of 1000 overall, with an average gain in binding free enthalpy of –G = 1.2–1.8 kcal mol–1 per methylation. The S4 pocket was also found to tolerate larger ammonium ions than trimethylammonium, such N-methylpyrrolidinium and N-methylpiperidinium moieties. The role of the halogen–arene interaction of the thienyl moiety in the S1 pocket was studied by preparing a series of inhibitors bearing differently substituted thienyl residues. The thienyl substituent replaces a conserved water molecule located above Tyr228 in the X-ray crystal structure of factor Xa without a bound ligand. Bromine and chlorine substituents were found to provide inhibitors featuring high binding affinities, whereas incorporating fluorine and iodine resulted in an activity loss by a factor of about five, suggesting that size and polarizability play a significant role in interactions with Tyr228. As compared to the unsubstituted thiophene, chlorine was found to enhance the binding affinity by a factor of 70. To investigate the stacking interactions of the isoxazolyl moiety with the polarizable walls of the S1 pocket, two inhibitors bearing different oxazole rings were prepared, resulting in a significant loss of binding affinity as compared to the isoxazole inhibitor. The loss of binding affinity is proposed to originate from the interplay of different factors, some of which include the slightly displaced position of the Cl atom of the thienyl moiety with respect to Tyr228, some potentially repulsive contacts between the ligand and the active site, and the suboptimal matching of polar contacts between the oxazole rings and the peptide backbone of the S1 pocket, leading to weakened stacking interactions. With the aim of replacing a conserved water molecule located at the back of the S4 pocket, the most active factor Xa inhibitor of the Diederich group to date was obtained

(Ki = 2 nM), featuring a hydroxyethyl pyrrolidinium S4 moiety. In the X-ray cocrystal structure with factor Xa, the inhibitor was found to undergo efficient cation– interactions in the S4 pocket and, instead of water replacement, water-mediated hydrogen bonding, accounting for the enhancement of binding affinity.

V Summary ______

To conclude, the work presented in this thesis shows factor Xa to be an excellent target for molecular recognition studies. Systematic variations of the inhibitor enabled investigations of individual binding interactions. Cation– interactions in the aromatic box of the S4 pocket were studied in detail, and deeper insight into other noncovalent binding interactions at the active site of factor Xa was gained. In the future, variation of the heteroatoms of the biaryl S1 needle could provide detailed information on optimizing stacking interactions with polar peptide surfaces. In addition, incorporating preorganized S4 moieties on the inhibitor, such as spirocyclic ammonium ions, could provide the optimal geometry for water displacement.

VI Zusammenfassung ______

Zusammenfassung

Um nichtkovalente Wechselwirkungen in biologischen Systemen zu entschlüsseln, sind Untersuchungen im Bereich der molekularen Erkennung ausschlaggebend. In der vorliegenden Doktorarbeit wurde ein multidimensionaler Ansatz verwendet, in dem vom strukturbasierten Design von Enzyminhibitoren Gebrauch gemacht wurde. Die Bindungsaffinität dieser Inhibitoren wurde analysiert und deren Bindungsanordnung in Kristallstrukturen untersucht, um individuelle Protein-Ligand-Wechselwirkungen auf molekularer Ebene zu quantifizieren. Dadurch wurde umfangreiches Wissen erlangt, welches nicht nur die Entwicklung von Inhibitoren, sondern auch andere Bereiche chemischer Forschung, wie die supramolekulare Chemie und die Entwicklung neuer Katalysatoren, bereichern kann. Für Studien zur molekularen Erkennung stellt die zentrale Serinprotease der menschlichen Blutgerinnungskaskade Faktor Xa ein exzellentes Modellsystem dar und wurde daher als Gegenstand unserer Forschung auserwählt. Das in dieser Arbeit verwendete Inhibitorsystem besteht aus einem trizyklischen Gerüst, welches den quaternären Ammonium-S4-Vektor und die Chlorthienylisoxazolyl-S1-Nadel zu deren jeweiligen Taschen ausrichtet.

Kation- - Wechselwirkungen in O Tyr99 Gln192 der S4-Tasche S4 OH H N N Wasserverdrängung O in der S4-Tasche H Phe174 HN Cys191 H O Trp215 N O S Stapelwechselwirkungen Gly216 Ala190 in der S1-Tasche S1 Asp189 Wasserverdrängung in der S1-Tasche OH Tyr228 Die Kation--Wechselwirkungen in dem aus den Seitenketten von Tyr99, Phe174 und Trp215 bestehenden Aromaten-Kasten der S4-Tasche wurden durch Vergleich der

Bindungsaffinitäten eines Trimethylammoniuminhibitors (Ki = 9 nM) mit dem + entsprechenden tert-Butylinhibitor (Ki = 550 nM) quantifiziert. Aus der C/N - Einzelatommutation ergab sich ein Gewinn an freier Enthalpie von –G = 2.5 kcal mol–1

VII Zusammenfassung ______

für die Kation--Wechselwirkung, was einem Wert von 0.8 kcal mol–1 pro aromatischen Ring entspricht. Die Cokristallstruktur des an Faktor Xa gebundenen kationischen Inhibitors zeigt, dass das quaternäre Ammoniumion im Zentrum des aromatischen S4-Kasten positioniert ist und der Chlorthienylrest mit Tyr228 aus der S1-Tasche wechselwirkt. Um den Einfluss der N-Methylierung an der endständigen Aminogruppe auf die Größe der Kation--Wechselwirkung zu bestimmen, wurden Inhibitoren mit unterschiedlichem Substitutionsgrad am Aminstickstoff synthetisiert. Es wurde ein substantieller Effekt in Abhängigkeit des Methylierungsgrades gefunden: Die

Bindungsaffinität steigt proportional vom primären (Ki = 9800 nM) zum quaternären

Ammoniumion (Ki = 9 nM) um einen Faktor von 1000, wobei die freie Enthalpie –G im Mittel zwischen 1.2–1.8 kcal mol–1 pro zusätzlicher Methylgruppe ansteigt. Interessanterweise werden in der S4-Tasche auch größere quaternäre Ammoniumionen als Trimethylammonium, wie N-Methylpyrrolidinium und N-Methylpiperidinium, toleriert. Zudem wurde die Bedeutung der Halogen-Aren-Wechselwirkung des Thienylrests in der S1-Tasche anhand einer Inhibitorserie mit unterschiedlich substituierten Thienylresten analysiert. Der erwähnte Thienylsubstituent ersetzt ein über Tyr228 befindliches, strukturelles Wassermolekül, die in der Kristallstruktur von Faktor Xa ohne gebundenen Liganden sich befindet. Dabei konnten mit Brom- und Chlorsubstituenten hohe Bindungsaffinitäten erzielt werden, während Fluor und Iod zu einem Verlust an Aktivität von ca. einem Faktor fünf führten. Dies legt nahe, dass Größe und Polarisierbarkeit maßgeblich die Wechselwirkung mit Tyr228 beeinflussen. Verglichen mit einem unsubstituierten Thiophenrest konnte durch Einführung eines Chlorsubstituenten eine 70fache Erhöhung der Bindungsaffinität erzielt werden. Um die Stapelwechselwirkungen zwischen der Isoxazoleinheit und den polarisierbaren Wänden der S1-Tasche zu analysieren, wurden zwei Inhibitoren mit unterschiedlichen Oxazolringen hergestellt, die im Vergleich zum Isoxazolsystem einen signifikanten Aktivitätverlust zeigten. Die verringerte Bindungsaffinität ist auf verschiedene Faktoren zurückzuführen, unter anderem eingeschlossen: Eine geringfügige Verschiebung des Chlorsubstituenten der Thienyleinheit bezüglich Tyr228, potenziell auftretende Abstossungen zwischen den Heteroatomen des Oxazolringes und der aktiven Tasche, sowie ungünstige polare Kontakte zwischen den Oxazolringen und den Peptidrückgrat der S1-Tasche, was zu schwächeren Stapelwechselwirkungen führt.

VIII Zusammenfassung ______

Mit dem Ziel, das strukturelle Wassermolekül hinter der S4-Tasche zu verdrängen, wurde der aktivste Inhibitor der Diederich-Gruppe (Ki = 2 nM) mit einem Hydroxyethylpyrrolidinium-S4-Substituenten entworfen. Durch Analyse der Cokristallstruktur stellte sich heraus, dass effiziente Kation--Wechselwirkungen in der S4-Tasche eingegangen werden und sich, anstelle der Verdrängung eines Wassermoleküls, eine Wasserstoffbrücke am äusseren Rand der S4-Tasche zu einem Wassermolekül ausbildet, was zur Erhöhung der Bindungsaffinität führt. Zusammenfassend zeigt sich in dieser Doktorarbeit, dass Faktor Xa ein ausgezeichnetes Enzym für Studien über molekulare Erkennung ist. Die systematische Variation des Inhibitors ermöglichte Untersuchungen individueller Bindungswechselwirkungen. Kation--Wechselswirkungen im aromatischen S4-Kasten wurden so im Detail analysiert, und ein tiefergehendes Verständnis der nichtkovalenten Wechselwirkungen im aktiven Zentrum von Faktor Xa wurde erlangt. Eine systematische Variation der Heteroatome des Biaryls der S1-Nadel könnte in Zukunft detailierte Informationen zur Optimierung von Stapelwechselwirkungen mit polaren Peptidoberflächen liefern. Zusätzlich könnte eine stärkere Präorganisation des in der S4-Tasche bindenden Molekülteils, beispielsweise durch Einbau spirozyklischer Ammoniumionen, zu einer Geometrieoptimierung führen, welche die Verdrängung des Wassermoleküls ermöglichen sollte.

IX 1. Introduction ______

1. Introduction

1.1. Molecular Recognition with Biological Systems Noncovalent specific interactions between two or more molecules, e.g. in host–guest and protein–ligand complexes, are referred to as molecular recognition. Detailed studies of the binding phenomena in these complexes are crucial to understand the recognition processes in biological systems. The gained knowledge is essential to enable more accurate predictions of ligand binding, and thus, ultimately a more efficient drug discovery process. Model system studies, enabling the investigation of individual interactions, have been pivotal to gain understanding on how nature recognizes substrates.[1] In addition to synthetic host–guest chemistry, the study of biological systems can be employed as well:[2] modifications of the ligand, i.e. the guest, allow the deciphering of individual interactions at enzyme active sites, and thus, the study of these in a biological context.

Moreover, comparison of the inhibitory constant Ki and the free binding enthalpy G =

–RTlnKi of different ligands allows the quantification of individual intermolecular binding forces. The use of iterative structure-based drug design methods[3] has provided more knowledge on molecular recognition processes in protein–ligand complexes.[4,5] The design cycle begins with computational ligand design at the active site of the target enzyme using the enzyme X-ray crystal structure. After synthesis of the designed compounds, the ligands are subjected to biological assays. If X-ray cocrystals with selected ligands bound at the enzyme active site are obtained, they are analyzed together with the binding data, and the knowledge gained from this analysis can then be applied to the computational design of new ligands. In order to investigate and quantify individual protein–ligand interactions at the molecular level, a multidimensional approach is employed in the Diederich group, comprising structure-based design of enzyme inhibitors and analysis of binding, crystallographic, and physicochemical data, which simultaneously allows the obtainment of highly active inhibitors. The in-depth understanding acquired through this strategy can ultimately be applied not only to drug design, but also to many other areas of chemical research, such as supramolecular chemistry and catalyst development.

1 1. Introduction ______

1.2. Cation– Interactions in Chemical and Biological Recognition Cation– interactions[6] are ubiquitous in nature. Model studies, pioneered by Dougherty,[7-9] have made seminal contributions to understand the way nature exploits these interactions in binding biologically relevant molecules. In addition to model systems, studies on synthetic receptors for onium ion recognition have shown the strength of the cation binding to be proportional to the number of aromatic rings and the free enthalpy contribution to reach values of about 0.5–1 kcal mol–1 per aromatic ring.[10] However, quantification studies in biological systems still remain scarce. Cation– interactions are of particular relevance in many recognition processes, involving neuroreceptors containing binding pockets lined by the side chains of aromatic amino acids Phe, Tyr, and Trp to bind substrates such as nicotine, serotonin (5-hydroxytryptamine), dopamine, acetylcholine (ACh), or GABA (-aminobutyric acid).[11] Such binding sites are also involved e.g. in the binding of ammonium ions upon the uptake to the ammonium transport channel,[12] in the binding of the positively charged 7-methylguanosine ring in human nuclear cap binding complex (Figure 1, left),[13] and in the binding of N,N,N-trimethylated lysine to the aromatic box of the BPTF PHD finger (Figure 1, right).[14] More examples of cation– interactions occurring in biological systems emerge constantly, thus highlighting the importance of elucidating them thoroughly and investigating different aspects of the interaction with model systems.

Figure 1. Left: Binding of the positively charged 7-methylguanosine ring in human nuclear cap-binding complex (PDB code: 1H2T).[13] Right: N,N,N-Trimethylated Lys of histone H3K4me3 in the aromatic pocket of the BPTF PHD finger (PDB code: 2F6J).[14]

In 1981, Kebarle and co-workers reported the potassium cation to bind more strongly to benzene than to water in the gas phase, with enthalpies of H° = –19.2 and

2 1. Introduction ______

–1 + + [15] –17.9 kcal mol for K –benzene and K –H2O, respectively. Calculations showed the major contributions to the binding in the benzene complex to originate from electrostatic and induction forces, with a significant contribution from dispersion as well. For the + Na –benzene complex, calculations showed the cation to sit on the C6 symmetry axis of benzene in the most stable geometry. Dougherty and co-workers observed the cyclophane receptor 1 (Figure 2) to bind adamantyltrimethylammonium iodide (2)[7] and quinolinium guests[9] 3 and 4 with high affinity in water, which was attributed to cation– interactions between the quaternary ammonium ion and the concave aromatic cavity. The finding was supported by the observed loss of binding affinity when the para-xylyl linkers of the macrocycle were replaced by cyclohexyl moieties.

COO–Cs+ N I– +Cs–OOC

O O 2

– N I

3 O O + – N Cs OOC I– COO–Cs+ 4 1

Figure 2. Cyclophane receptor 1 by Dougherty and co-workers was shown to bind the cationic moieties of guests 2–4 in the aromatic cavity.[7,9]

Cation– interactions can be mostly rationalized through the electrostatic potential of the aromatic ring: with their positive charge, cations interact with the negative electrostatic potential of the arene (Figure 3).[16] Although electrostatics play an important role in cation– interactions,[6] calculations have shown induction to contribute significantly to the stabilization as well, in particular in complexes of e.g. Li+, and Na+.[17] However, even though electrostatic contributions are not the sole origin of the interaction, trends can be predicted using only this term.[18]

3 1. Introduction ______

H O

N H

Figure 3. Electrostatic potential surfaces of benzene, phenol, and indole showing the regions of negative (red) and positive (blue) potential (Spartan, HF/3-21*, scale –27 to +21).

With larger organic cations, such as tetramethylammonium (TMA), polarizability is more important than for small inorganic cations.[6] Using the ESP CHELPG (ElectroStatic Potential Charges from Electrostatic Potentials Generalized) method, the atomic partial charges of the TMA cation were calculated as +0.28 for N, –0.30 for C, and +0.16 for H.[19] Although the interaction between an arene and a TMA cation could be described as a strong C–H··· or that of a primary ammonium cation as a N–H··· interaction, the physical origin of the attraction is different.[17] According to calculations, the attraction in both complexes originates from the same forces as those of the alkali metal cation– interaction, thus qualifying the former as cation– interactions. PDB (Protein Data Bank) and CSD (Cambridge Structural Database) searches have been employed to elucidate the occurrences, geometries, and preferences of the cation– interaction. In a survey of the PDB, Gallivan and Dougherty found one energetically significant cation– interaction for every 77 amino acid residues on average,[20] with 26% of Trp side chains taking part in such interactions. This was attributed to an enhanced cation-binding ability of the indole ring in Trp as compared to Tyr or Phe. In the PDB, a clear preference for the cation to reside over the six-membered ring of Trp was found, as has been predicted by calculations and electrostatic potentials.[6] Despite being a relatively uncommon amino acid, Trp seems to appear often at cation– binding sites.[21,22] This can be rationalized by the much more intense negative electrostatic potential over Trp as compared to Tyr or Phe (Figure 3). Tyr is preferred to Phe in the cation– interaction and appears on cation– binding sites nearly as often as

4 1. Introduction ______

Trp and considerably more frequently than Phe. Although the differences in the electrostatic potentials of Tyr and Phe are minor, the ability of Tyr to form hydrogen bonds enhances the negative electrostatic potential on the phenol ring and could help to position the ring properly, if corresponding H-bond acceptors are present in the ligand. Histidine does not appear to be involved in cation– interactions as the  system possibly due to protonation at physiological pH, which would prevent His from interacting with cationic moieties. For the cationic amino acid residues, Arg was found to engage in cation– interactions more often than Lys, possibly due to the ability of Arg to undergo parallel- stacked cation– interactions while remaining hydrogen-bonded to another amino acid side chain.[20] Moreover, Lys was found to interact more often through the C–H moieties of its -carbon than through the ammonium moiety, possibly enabling simultaneous hydrogen-bonding interactions of the cation with the solvent and favorable van der Waals contacts of the C–H moieties and the arene. To analyze the influence of cation– interactions in DNA-binding proteins, 62 X-ray crystal structures were investigated, 73% of which showed cation– contacts. Arg– interactions were found more often than Lys– contacts.[23,24] The side chain of Arg was observed to interact preferentially with Phe and Tyr rather than with Trp. The computed interaction energies varied from –2.35 to –9.94 kcal mol–1, the average being –5.00 and –4.28 kcal mol–1 for Arg and Lys, respectively.[23] A PDB search was carried out to investigate the cation– interactions between adenine and Arg/Lys in ATP-binding proteins.[25] Cation– interactions were found in 59% of the adenylate–protein complexes. In addition to cation– interactions, both positively charged residues were found to interact with adenine through additional interactions, namely hydrogen bonding for Lys and – stacking for Arg. At protein–protein interfaces, cation– interactions between Arg and Trp were found as the most abundant (43%) of the possible cation– pairs in a PDB search.[26] On average, the interaction was calculated to account for 3.3 kcal mol–1 of electrostatic binding free energy, and for over 60% of the cases Arg was found to undergo hydrogen- bonding interactions in addition to the cation– contact. Coplanar geometry was favored (51%) over orthogonal and oblique. The key review by Ma and Dougherty[6] has been followed up by more recent insights into the research progress in the field of cation– interactions,[10] mainly from the

5 1. Introduction ______

point of view of computational research.[17,27] Some studies of energetic quantifications of the cation– interaction with model systems and its use in organic synthesis will be presented in the following. Further examples can be found within this thesis in the appropriate chapters.

Waters and co-workers investigated the recognition of trimethyllysine (LysMe3) with a -hairpin peptide model system and with mutated histone 3 (H3) peptides binding [28] to the HP1 chromodomain. In the hairpin system, the interaction of Trp with LysMe3 is stronger than with its purely aliphatic counterpart, tert-butyl norleucine (tBuNle). The –1 interaction free enthalpies are –1.0 ± 0.1 kcal mol for LysMe3–Trp, and –0.6 ± 0.1 kcal –1 mol for tBuNle–Trp (D2O/acetate-d3 buffer). Thermal denaturation studies revealed the

interaction with LysMe3 to be entropy-driven with a negligible enthalpic contribution, whereas the binding of tBuNle showed unfavorable enthalpy and greatly enhanced

favorable entropy of folding. In the case of the HP1 chromodomain, the LysMe3-bearing

H3 peptide was shown to bind with much higher affinity (Kd = 10 M) than the

tBuNle-bearing peptide (Kd = 310 M) in a fluorescence polarization assay in buffer solution at 288 K. The Dougherty and Lester groups have used an elegant way to probe cation– interactions[29-32] by introducing non-natural aromatic amino acids with increasingly fluorinated aromatic rings to the receptors. Fluorine substitution reduces the electron density of the aromatic rings and cation binding weakens, as evaluated in electrophysiological measurements. This strategy was used to probe and identify [30] [29] cation– interactions at the GABAA and GABAC receptor binding sites, at the dopamine binding site of the D2 receptor of a G-protein coupled receptor (GPCR),[31] and at the nicotinic ACh 42 brain receptor binding site.[32] As an example, mutation of

2Tyr97 in the GABAA receptor to the corresponding increasingly fluorinated Tyr

derivatives caused a 20-fold increase in the EC50 value with each additional fluorine substituent, indicative of weakened cation– interactions in the binding of GABA.[30] The highly preorganized molecular clips and tweezers, introduced by Klärner and co-workers, recognize organic cations.[33] The electron-rich cavities of these aromatic host molecules, such as clip molecule 5 (Figure 4), have been shown to bind Lys and Arg in water,[34] as well as a variety of N-alkylpyridinium guests,[35] such as NAD+,[35,36] and even sulfonium guests, such as S-adenosylmethionine (SAM).[37] The positively charged moiety of the guests was shown to bind in the aromatic cavity by 1H NMR titration

6 1. Introduction ______

studies. Additionally, clip and tweezer molecules were shown to inhibit alcohol dehydrogenase by binding either to NAD+ or to Lys residues on the surface of the enzyme.[38] R

R + R = O P O N(nBu)4 O

5

Figure 4. Clip molecule 5 by Klärner and co-workers.[35]

High-level ab initio calculations by Tsuzuki et al. showed the major contributions to the interaction energy of N-methylpyridinium cations with  systems to stem from electrostatics and induction, thus qualifying the interaction as a cation– interaction.[39] N-Methylpyridinium– interactions have been found to increase the folding stability of oligo(arylene-ethynylene)s in acetonitrile, with the interaction contributing at least 1.8 kcal mol–1 compared to the folding stability of the non-methylated oligomer.[40] However, the relationship between the folding stability and oligomer length is yet to be precisely determined. Additionally, the stabilizing effect increases when electron- donating para-substituents are incorporated into the pyridinium ring, whereas electron- withdrawing substituents have a destabilizing effect.[41] The cation– interaction has been rarely employed in organic synthesis despite its high occurrence in biology and wide application in diverse areas of chemistry. However, there are some interesting reports in the literature, and new ones emerge constantly. In their early studies, Dougherty and co-workers observed rate enhancements of the alkylation of quinoline derivatives to the corresponding quinolinium salts when performing the reaction in the presence of a cyclophane host (Figure 2), which stabilizes the positively charged transition state of the transformation through cation– interactions.[42,43] Yamada has reviewed the use of intramolecular cation– interactions in organic synthesis, focusing mainly on N-substituted pyridinium– interactions.[44] Cation– interactions were found to influence the regiochemical outcome of an intramolecular Schmidt reaction of 2-azidoalkyl ketones in dichloromethane.[45] A typically disfavored pathway becomes preferred due to cation– interactions between the intermediate diazonium cation and an aromatic substituent in -position to the ketone.

7 1. Introduction ______

An increase in selectivity was found when the aromatic ring was substituted with electron-donating groups; the opposite was observed with electron-withdrawing groups. The cation– interaction has been used to influence the regiochemistry of [2+2] photodimerization of styrylpyridinium salts,[46] also within cyclodextrin or cucurbituril hosts.[47] Chiral pyridinium and quinolinium salts with tethered phenyl rings were allylated regio- and enantioselectively due to the cation– interaction between the rings.[48] The N-methylquinolinium ring in one of the starting materials was shown to lie parallel to the phenyl ring at an interplanar distance of 3.4 Å by X-ray crystallography. Pyridinium salts have been used analogously in enantioselective cyclopropanation reactions.[49] Very recently, cation– interactions were reported to enable intramolecular olefin metathesis reactions of compounds such as 6: Macrocyclization products 7 were obtained in 45–90% yield when N-methylquinolinium salt 8 was added as a conformational control element, whereas no product formation was observed in the absence of the salt (Scheme 1).[50] Due to the cation– interaction between the quinolinium and the aryl ring of the starting material, one side of the aromatic ring is shielded, and the substituents are forced into an orientation where the metathesis reaction can occur. In the same manner, the quinolinium control element enabled intramolecular Glaser-Hay coupling reactions with yields around 40%.

PCy3 Cl Ru Cl Ph PCy3 O 10 mol% O

CH2Cl2

MeOOC MeOOC O N O – PF6 678

Scheme 1. Addition of N-methylquinolinium salt 8 as a conformation control element enabled the macrocyclization of 6 via olefin metathesis.[50]

1.3. The Human Blood Coagulation Cascade: Factor Xa and Thrombin Cardiovascular diseases are the leading cause of death in the world according to the World Health Organization, with around 17 million deaths in 2004, 29% of all global

8 1. Introduction ______

cases.[51] Venous thromboembolism, i.e. blood clots in the veins, belongs to this group of diseases. The coagulation cascade is the mechanism enabling the regulation of blood flow, e.g. in the case of vascular injury, which is essential for survival. It is a complicated process and a delicate balance of many components, where defects can lead to serious complications such as the formation of thrombi or bleeding. There are two types of blood coagulation processes: Hemostasis is the ceasing of bleeding of cuts and severed blood vessels, whereas thrombosis occurs with damaged endothelium of blood vessels walls.[52] Common to both mechanisms are three phases: First, a loose and temporary aggregate is formed at the site of the injury. The aggregate is further stabilized by the formation of and ultimately , produced by the blood coagulation cascade. In the final phase, the hemostatic plug or thrombus is dissolved by plasmin. The blood coagulation cascade,[52] leading to the formation of a fibrin clot, can be initiated by two pathways (Scheme 2). A tissue injury switches on the extrinsic pathway. The process, which activates the intrinsic pathway, remains unclear. However, the two pathways are not independent of each other and there are indications of cross-activation. The pathways merge at , and the formation of fibrin occurs through the final common pathway. The intrinsic pathway is initiated by the activation of factor XII by prekallikrein (PK) and high-molecular-weight kininogen (HK; Scheme 2). Factor XIIa, HK, and Ca2+ convert factor XI into its active form XIa, which, together with Ca2+, in turn activates factor IX. This conversion is additionally activated by the VIIa–tissue factor complex of

the extrinsic pathway. Factor X is a central enzyme of the coagulation cascade, since it can be activated to factor Xa by both the extrinsic and the intrinsic pathway. In the extrinsic pathway, factor VIIa together with tissue factor converts factor X into its active form on the surface of activated . In the intrinsic pathway, this activation occurs by factor IXa with cofactor VIIIa, Ca2+, and phospholipids (PL). The two-chain serine protease factor Xa is formed by the cleavage of the Arg52-Ile53 bond of factor X.[53] In the final common pathway, factor Xa in the prothrombinase complex cleaves prothrombin (factor II) at two sites, Arg271-Thr272 and Arg320-Ile321, to produce thrombin (factor IIa).[54] The activation occurs on the surface of activated platelets with the assistance of the prothrombinase complex: prothrombin, factors Xa and Va, Ca2+, and

9 1. Introduction ______platelet anionic phospholipids.[52] The formation of this prothrombinase complex enhances the catalytic activity of factor Xa by a factor 5.[53] Active thrombin converts fibrinogen to fibrin by hydrolyzing Arg-Gly bonds between fibrinopeptides.[52] The release of fibrinopeptides creates fibrin monomers, which aggregate to form a fibrin clot. This fibrin polymer forms the initial, rather weak thrombus by trapping red cells, platelets, and other compounds. Factor XIIIa, also activated by thrombin, strengthens the initial blood clot by cross-linking fibrin molecules.

Intrinsic Pathway PK HK XII XIIa

HK Ca2+

XI XIa Extrinsic Pathway Ca2+ VII IX IXa

Ca2+ VIIa–Tissue factor VIII VIIIa PL

XXaX

Ca2+ VVa PL

Prothrombin Thrombin

Fibrinogen XIII

Fibrin monomer XIIIa

Fibrin polymer

Cross-linked fibrin polymer

Scheme 2. The coagulation cascade.[52] The feedback mechanisms are omitted for clarity. HK = high-molecular-weight kininogen, PK = prekallikrein, PL = phospholipids.

Factor Xa is a vitamin K-dependent serine protease (59 kDa), which is synthesized in the liver and secreted into blood as a zymogen.[53] It consists of two chains, which are linked by a disulfide bridge. The heavy chain comprises 303 amino acids, the light chain 139. The catalytic triad, as in all serine proteases, is located in the heavy chain and consists of Ser195, His57, and Asp102.

10 1. Introduction ______

During catalysis, the substrate, prothrombin, is bound to form the Michaelis complex (Scheme 3).[55,56] The rate-determining step is the nucleophilic attack of Ser195 on the substrate carbonyl of the peptide bond that will be cleaved, forming a tetrahedral intermediate. His57 acts as base enabling the reaction, and Asp102 stabilizes the formed imidazolium cation.

Asp102 Asp102 Asp102 His57 His57 His57 O O O H H H N Ser195 Ser195 N Ser195 O O N O N N N H H O H O O R' N Michaelis Tetrahedral R' R Acyl–enzyme R' R N H O R complex N C intermediate H O intermediate H O

R'NH2

Asp102 His57 Asp102 Asp102 His57 His57 O H Ser195 O O N H H O N Ser195 N Ser195 N O O N N H O H Active Tetrahedral O H O R R enzyme O intermediate O O R H O H O H O

Scheme 3. The catalytic mechanism of factor Xa.[55,56]

In the next step, the peptide bond is cleaved with His57 protonating the formed amine, generating the acyl–enzyme intermediate. Analogous to the previous steps, the active enzyme is regenerated by the deacylation of the acyl–enzyme intermediate and the release of the carboxylate product. His57 deprotonates a water molecule that acts as a nucleophile, and a second tetrahedral intermediate is formed. The serine is protonated by His57, and the active enzyme is regenerated. The active site of factor Xa is located mainly on the surface of the enzyme (Figure 5). The naming of the binding pockets follows the Schechter and Berger nomenclature,[57] where the cleavage point is between S1 and S1’ subsites, P1 and P1’ in the substrate.[58] The numbering increases from the cleavage point with S1, S2, S3 etc. towards the N-terminus and S1’, S2’ etc. towards the C-terminus.

11 1. Introduction ______

Tyr99 H N OH 2 Gln192 O

NH Cys191 O S4 Phe174 HN S

HN Trp215 HN O O OAla190 Gly216 O HN O

O Asp189 OH S1 Tyr228

Figure 5. Central features of the active site of factor Xa.

The S1 subsite is a narrow pocket reaching towards the core of the enzyme (Figure 5). The walls of the S1 pocket are lined by Trp215-Gly216 on one side, and Ala190-Cys191-Gln192 on the other. The negatively charged Asp189 and the side chain of Tyr228 form the bottom of the pocket. The natural substrate prothrombin binds to the S1 pocket through ionic hydrogen bonding of the side chain of Arg with that of Asp189. Factor Xa and thrombin feature alanine in position 190, whereas in trypsin this amino acid is replaced by Ser190, and this change in the volume and electrostatic properties of the S1 pocket can be exploited to gain selectivity.[59] In contrast to other serine proteases, in factor Xa the access to the S2 pocket is blocked by Tyr99. The S4 binding pocket is an aromatic box formed by the side chains of Tyr99, Phe174, and Trp215. This binding site differs greatly from those of other trypsin- like serine proteases,[59] and highly selective inhibitors can be obtained by targeting the S4 pocket. Thrombin (Figure 6), for example, features only one aromatic residue, Trp215, in the D pocket (corresponding to S4 in factor Xa), and thus, selective inhibitors of factor Xa can be obtained by incorporating S4 vectors which interact favorably with the aromatic box.

12 1. Introduction ______

Tyr60A HN Trp60D

HO P

Asn98 HN O O Ser195 NH2 D HN

H Gly193 Trp215 O N O Gly219 Gly216 NH O O

Asp189 OH S1 Tyr228

Figure 6. The active site of thrombin.

1.4. Current Anticoagulant Therapy As mentioned above, thrombosis is a major cause of disease and death in the Western world, and therefore, the use of is extensive, with about 0.7% of the Western population being treated with them.[60] Current anticoagulant treatments include vitamin K antagonists (VKAs) such as or other derivatives, unfractionated heparin (UFH), and low-molecular-weight (LMWHs).[61] These therapies, however, are associated with several significant limitations. Heparin inhibits factor Xa and thrombin indirectly by activating III,[62] a naturally occurring inhibitor of thrombin, factor Xa and many others, which is part of the natural feedback mechanism of the coagulation cascade.[52] Heparin has a narrow therapeutic window, i.e. the difference between the effective dosage and the dosage where side effects start to appear is small.[62] In addition, it has a highly variable dose-response relationship, and some patients suffer from heparin-induced thrombocytopenia, a hypercoagulability state leading to the generation of blood clots, during longer periods of treatment. Although LMWHs, such as the pentasaccharide (9) (Figure 7, left), have a wider therapeutic window and do not trigger

13 1. Introduction ______

thrombocytopenia, they still require parenteral administration, making their use outside of the hospital complicated. O – – – OSO3 COO– OSO3 OSO3 OH O O O –O O OH OH OSO – COO OH O 3 OH O OH O O OMe – – – – NHSO3 OH NHSO3 OSO3 NHSO3 O O fondaparinux (9) warfarin (10)

Figure 7. Left: Fondaparinux (9) (Arixtra®, GlaxoSmithKline), a synthetic indirect factor Xa inhibitor.[62] Right: Warfarin (10),[63] the most commonly used oral anticoagulant.

Warfarin (10) (Figure 7, right), the most commonly used oral anticoagulant,[64] inhibits the synthesis of several enzymes of the coagulation cascade by interfering with the metabolism of vitamin K.[63,65] Despite being orally available, warfarin treatment has numerous limitations, such as delayed onset of action, prolonged anticoagulant effects after discontinuing the treatment, and the need of constant monitoring due to the narrow therapeutic window, variable patient response, and drug–drug and food–drug interactions. Due to the serious disadvantages of heparin-derived and warfarin-based therapies, the development of anticoagulant agents has been of great interest to the pharmaceutical industry, and several reviews have emerged on the subject in general,[60,65-68] and on factor Xa inhibitors in particular.[58,61,64-66,69-76] The advantage of direct small-molecule factor Xa inhibitors is their ability to not only inhibit the enzyme in plasma, but also when it is bound to a fibrin clot.[72] Moreover, factor Xa does not seem to have many roles in addition to its main function, cleaving prothrombin to give thrombin, making it an attractive and specific target for drug development. A search (April 2011) in Thomson Reuters Integrity for factor Xa inhibitors currently in development gave five direct small-molecule inhibitors of factor Xa in Phase III clinical trials or higher (Figure 8): (11)[64,77] (Xarelto®) was launched in 2008 as the first direct small-molecule factor Xa inhibitor for clinical use. (12)[78] (Eliquis®) has been recommended for approval, tosylate (13)[79] has been pre-registered, and both maleate (14)[80] and (15)[81] are in Phase III clinical trials. Most of the structures feature an uncharged S1 vector; only otamixaban has a phenylamidinium moiety binding into the S1 pocket.

14 1. Introduction ______

O O O N N O N N NH2 O N O N S O O HN N O N N HN O HN O SO3H S O N NH H O O 2 Cl Cl rivaroxaban (11, Xarelto) apixaban (12, Eliquis) edoxaban tosylate (13) Bayer HealthCare Bristol-Myers Squibb; Pfizer Daiichi Sankyo launched 2008 recommended approval registered 2011

O N N N

O H OH N HN O O

HN O COOH

O H2N COOH O •HCl NH

darexaban maleate (14) otamixaban (15) Astellas Pharma -Aventis Phase III Phase III

Figure 8. Five direct factor Xa inhibitors 11–15 currently in Phase III clinical trials or higher.[64,77-81]

1.5. Factor Xa and the Cation– Interaction During the course of the extensive molecular recognition project with thrombin in the Diederich group, one goal was to investigate cation– interactions in the D pocket. This led to the synthesis of quaternary ammonium inhibitor (±)-16 (Figure 9), designed to undergo cation– interactions with Trp215.[82] Compound (±)-16 features a phenylamidinium vector to bind to Asp189 at the bottom of the S1 pocket. Uncharged tert-butyl ligand (±)-17 was prepared as a control compound for the cation– interaction.

Surprisingly, ammonium ion (±)-16 was a poor thrombin inhibitor (Ki = 7.76 μM,

Figure 9, left), and control compound (±)-17 was found to bind more strongly (Ki = 0.13 μM). However, reversed activity was found towards factor Xa: Whereas neutral

(±)-17 was a very weak factor Xa inhibitor (Ki = 29 μM), ammonium ion (±)-16 was much

more potent (Ki = 0.28 μM).

15 1. Introduction ______

O H N X N O H

H2N NH HCl

Ki Ki thrombin factor Xa

(±)-16 X = N+Br– 7.76 M 0.28 M (±)-17 X = C 0.13 M 29 M

Figure 9. Left: The binding affinities of (±)-16 and (±)-17 towards thrombin and factor Xa.[82] Right: The X-ray cocrystal structure of (±)-16 bound at the active site of factor Xa (PDB code: 2BOK, resolution 1.64 Å). Distances in Å. Color code: gray

Cenzyme, green Cinhibitor, red O, blue N, yellow S.

The cation– interaction between the ammonium ion and Trp215 in the D pocket of thrombin is presumably not strong enough to compensate for the solvation energy lost during the binding. In the S4 pocket of factor Xa, the aromatic box formed by the three aromatic residues Tyr99, Phe174, and Trp215, however, is able to compensate for the desolvation energy loss. The affinity of factor Xa for (±)-16 is significantly higher than

for the non-cationic (±)-17 (Ki((±)-17)/Ki((±)-16) = 100), corresponding to an increase in binding free enthalpy of –G = 2.8 kcal mol–1 for the cation– interaction, approximately 0.9 kcal mol–1 per aromatic ring. The X-ray cocrystal structure of ammonium ion inhibitor (±)-16 bound at the active site of factor Xa was solved (Figure 9, right, PDB code: 2BOK).[82] In the S1 pocket, the phenylamidinium residue forms a salt bridge with the carboxylate side chain of Asp189 (d(N(H)O-Asp189) = 3.2 and 3.3 Å). In addition, one carbonyl moiety of the imide forms a hydrogen bond with Gly216 (d(Gly216-N(H)O) = 3.1 Å). The quaternary ammonium ion of (±)-16 resides in the middle of the aromatic box of the S4 pocket. The shortest heavy atom distances between the N+ ammonium ion and the aromatic ring C atoms of Tyr99, Phe174, and Trp215 are 4.6, 5.0, and 4.6 Å, respectively, almost ideal for cation– interactions.[6]

16 1. Introduction ______

A survey of the Protein Data Bank (PDB) performed in the Diederich group revealed that basic residues, such as 4-amino-substituted pyridine, pyrrolidines, amidines, and imidazoles, have been often employed as moieties to bind into the S4 pocket of factor Xa.[82] Additionally, a search for similar aromatic boxes binding quaternary ammonium ions revealed the cation always positioned close to the intersection of the normals passing through the centroids of the aromatic rings, at distances 4–5 Å between the N+ center and closest aromatic C atom. Similar results were obtained for tertiary ammonium ions, the binding of which was often accompanied by an additional hydrogen- bonding interaction of the N+–H residue with the protein or a water molecule.

The central tricyclic core of the ligands featured in all target molecules throughout this thesis was designed de novo by Dr. Ulrike Obst-Sander during her PhD thesis on thrombin inhibitors.[83] The tricyclic core was found to give rise to very active ligands with good selectivity against trypsin as compared to bicyclic inhibitors, which was attributed to the high conformational rigidity of this class of compounds.[84,85] Such rigidification of inhibitor scaffolds often correlates with enhanced binding affinity.[86] Consequently, the tricyclic core has been used in all thrombin and factor Xa inhibitors of the Diederich group since.

The key step of the inhibitor synthesis is a 1,3-dipolar cycloaddition of L-proline (18), aryl aldehyde 19, and maleimide 20,[87] which allows the build-up of the central core, featuring four stereogenic centers, in one step (Scheme 4). The first step is the condensation of 18 and aldehyde 19 to give iminiumcarboxylate 21, which subsequently cyclizes to the thermodynamically more stable trans-oxazolidinone 22. The thermally

allowed 1,3-dipolar cycloreversion under concomitant loss of CO2 yields azomethine ylide (E)-23 as the kinetic product, which is in equilibrium with (Z)-23. Because the following reaction with the highly reactive dipolarophile maleimide 20 is fast as compared to the isomerization process, trans-products (±)-24 and (±)-25 are formed in excess. The syn attack of azomethine ylide 23 on maleimide 20 leads to cis-configuration of the bridging C(3a) and C(8b) atoms on the tricycle (Scheme 4).[88] The cycloaddition can go through either endo- or exo-transition states, resulting in a mixture of four possible diastereoisomeric enantiomer pairs: (±)-exo,trans-24, (±)-endo,trans-25, (±)-exo,cis-26, and endo,cis-configured compound (±)-27. The nomenclature exo and endo of the final

17 1. Introduction ______products refers to the orientation of the aryl substituent at C(4) relative to the bicyclic perhydropyrrolo[3,4-c]pyrrole scaffold (Scheme 4); cis and trans refer to the position of the C(4) substituent in respect of the configuration at C(8a). In previous studies, endo,trans-configured inhibitors were shown to bind with higher affinity than exo,trans- configured ones.[89] O + HO2C N Ar H H 18 19

O O N+ N O O Ar Ar 21 22

– CO2

O O

+ R N + N N+ + N R Ar O Ar O 20 (E)-23 (Z)-23 20

N N Ar N N Ar O O Ar O Ar O R N R N R N R N O O O O endo exo exo endo

O O O O H H H H R 8a R R R N 8b N N N 3a 4 N N N N O H O H O H O H Ar Ar Ar Ar (±)-exo,trans-24 (±)-endo,trans-25 (±)-exo,cis-26 (±)-endo,cis-27

Scheme 4. The 1,3-dipolar cycloaddition reaction.

The diastereoisomers formed in the cycloaddition can be separated by column chromatography on silica or normal phase HPLC, and assignment can be performed by 1H NMR spectroscopy due to the characteristic coupling constants of the tricycle protons. Despite the often tedious separation procedures leading to low yields of the isolated

18 1. Introduction ______products, the advantage of this synthetic strategy is the formation of a complex tricyclic structure, orienting the S1 and S4 vector in an ideal L-shaped form, in one step. In addition, the rigidity of the tricyclic core facilitates molecular modeling[90,91] by restricting the degrees of freedom of the molecule, and thus, the binding mode of the inhibitor at the enzyme active site can be predicted with high accuracy. Also, direct comparisons of the binding affinities of molecules with single-atom mutations can be done due to similar binding modes, allowing for quantifications of individual interactions.

1.6. Project Goals The objective of this PhD Thesis was to study the noncovalent interactions involved in the molecular recognition of inhibitors of enzyme factor Xa. The design and synthesis of systematically varied highly active inhibitors would allow the investigation and quantification of different noncovalent interactions involved in the binding of the ligands. The initial aim of this project was to gain insight into cation– interactions in aromatic molecular boxes, namely the S4 pocket of enzyme factor Xa (Figure 10). In the first stage, exchanging the very basic phenylamidine moiety in (±)-16 for a neutral S1 vector was necessary to enhance the binding affinity of the inhibitors, since high activity is crucial to allow for the precise quantification of the cation– interaction. Then, systematic variation of the S4 moiety, by incorporation of both cation as well as neutral residues as comparison, would be performed to probe the interactions in detail.

Cation– interactions O in the S4 pocket Tyr99 Gln192 S4 OH H N N O H Phe174 HN Cys191 H O Trp215 N O

Gly216 Ala190

S1 Asp189 Exchanging the S1 needle OH Tyr228

Figure 10. Initial goals of molecular recognition at the active site of factor Xa.

19 1. Introduction ______

During the course of this thesis, the investigation of interactions involved in the binding of the newly incorporated neutral S1 needle became of interest, and were studied by variation of the S1 vector. Additionally, insight was gained into water replacement at the active site of factor Xa.

20 2. Enhancing Binding Affinity by S1 Needle Replacement ______

2. Enhancing Binding Affinity by S1 Needle Replacement

2.1. Inhibitor Design: Exchanging the Phenylamidinium for a Neutral S1 Needle Early synthetic factor Xa inhibitors have commonly featured a phenylamidinium moiety as the S1 vector,[58,61,70] which interacts with Asp189 through a salt bridge, mimicking the side chains of Arg271 and Arg320 of the natural substrate prothrombin.[53] However, amidine groups have usually disadvantageous pharmacokinetic properties and contribute to poor oral availability due to their high basicity.[58,61,70] Therefore, significant efforts have been made in the pharmaceutical industry to replace the amidinium moiety with an uncharged one,[77,92-98] and residues such as chlorophenyl, chloronaphthyl, chlorobenzothienyl, chlorothienyl, and chloropyridyl have been shown to bind in the S1 pocket through interactions with Tyr228. During studies of cation– interactions at the active site of thrombin, quaternary

ammonium ion (±)-16 was found to be an active factor Xa inhibitor (Ki = 0.28 M),

whereas its uncharged tert-butyl counterpart (±)-17 bound very weakly (Ki = 29 M; Figure 11).[82] The N+/C single-atom exchange allowed, by comparison of the binding affinities of the two inhibitors, the quantification of the cation– interaction in the S4 pocket as –G = 2.8 kcal mol–1, with an average of 0.9 kcal mol–1 per aromatic ring. The X-ray cocrystal structure showed the ammonium ion residing in the center of the aromatic box of the S4 pocket, formed by the side chains of Tyr99, Phe174, and Trp215 (PDB code: 2BOK). To gain a deeper insight into cation– interactions in aromatic molecular boxes in biological systems, further investigations were of interest, e.g. on determining the influence of the size of the cation (Chapter 3.5) or the effect of methylation on the interaction strength (Chapter 3.6). In order to obtain more accurate results, however, enhancing the binding affinity of the inhibitors was necessary. This was envisioned by exchanging the phenylamidinium moiety in (±)-16 for an uncharged S1 vector, which could lead to more favorable physicochemical properties as well, such as enhanced membrane permeability and chemical tractability.

21 2. Enhancing Binding Affinity by S1 Needle Replacement ______

O H N R N O H

H2N NH HCl

+ – (±)-16 R = N Me3Br Ki = 0.28 M (±)-17 R = CMe3 Ki = 29 M

Figure 11. Left: The phenylamidinium vector of the active enantiomer of (±)-16 bound in the S1 pocket of factor Xa as seen in the X-ray cocrystal structure (PDB code: 2BOK, resolution 1.64 Å).[82] Right: The binding affinities of (±)-16 and (±)-17 towards factor

Xa. Distances in Å. Color code: gray Cenzyme, green Cinhibitor, red O, blue N, yellow S.

The cationic phenylamidinium moiety in (±)-16 had been a common feature of the thrombin inhibitors of the Diederich group since the beginning of the project.[83] Efforts to replace the amidine moiety with a less basic group, with the aim of improving the physicochemical profile of the ligands, had already been made during different phases of the thrombin project.[99,100] Residues such as e.g. aniline, benzylamine,

para-chlorophenyl, para-CF3-phenyl, 4-pyridyl, and 3-thienyl were investiagted as replacements, but unfortunately, in all cases the exchange resulted in inactive compounds or considerably weakened binding affinity. At the time of starting this thesis, several X-ray cocrystal structures of factor Xa inhibitors featuring a neutral residue in the S1 pocket had been reported in the [77,92,94-96,98,101] literature. As an example, a very potent factor Xa inhibitor 28 (Ki =

0.07 nM) by Aventis Pharma, which bears a neutral S1 needle, was cocrystallized with factor Xa (Figure 12, PDB code: 2BQ6).[101] The Cl atom of the biaryl needle interacts with Tyr228 at the bottom of the S1 pocket (d(C–Cl···C–O(H)-Tyr228) = 3.4 Å). The heteroatoms of the isoxazole ring are directed towards Ser195 at the back of the S1

pocket, at a distance indicating weak hydrogen bonding (d(Ser195-O(H)···Oisoxazole) = 3.5 Å).

22 2. Enhancing Binding Affinity by S1 Needle Replacement ______

N

N NH

O N

N O

S Cl 28 Ki = 0.07 nM

Figure 12. The X-ray cocrystal structure of 28, featuring a neutral S1 needle, in complex [101] with factor Xa (PDB code: 2BQ6). Distances in Å. Color code: gray Cenzyme, green

Cinhibitor, red O, blue N, yellow S, lime Cl.

Encouraged by the constantly emerging literature reports on successful phenylamidinium replacement in factor Xa inhibitors, various ligands containing neutral S1 vectors were designed using the program Moloc.[90,91] The neutral S1 vectors were incorporated into the tricyclic core, which had been proven to be an excellent central scaffold due to its rigid, preorganized conformation, which allows the introduction of the S1 and the S4 vectors in the L-shaped manner typical for factor Xa inhibitors. Incorporation of uncharged biaryl needles, such as the S1 vector of 28, into the tricyclic core of inhibitor (±)-16 was envisioned to lead to highly active factor Xa inhibitors. Novel inhibitors 29–33 bearing five different S1 needles attached to the tricyclic core were chosen as first target molecules (Figure 13): isoxazolyl-linked chlorothienyl compound 29, which bears the S1 needle of compound 28, thiazolyl- 30, ethynyl- 31, and phenyl-chlorothienyl ligand 32, and 33 featuring a chloroquinolyl needle. Four of these, 29–32, have chlorothienyl as the Cl-bearing moiety with different linkers to the tricyclic core of the inhibitor to change the approach angle of the Cl atom towards Tyr228. In addition to chlorothienyl, chloroquinoline needle 33 also seemed to provide a good fit in the S1 pocket. A focal point in the choice of the S1 vector was also the synthetic access: all potential needles should be readily obtained to enable a rapid development of a new inhibitor scaffold.

23 2. Enhancing Binding Affinity by S1 Needle Replacement ______

S1 needles: O S4 pocket N S N O N N N O S S S S

S1 pocket Cl Cl Cl Cl Cl 2930 31 32 33

Figure 13. The targeted neutral S1 needles 29–33.

Mainly the 3aS,4R,8aS,8bR-configured enantiomers of the endo,trans- diastereoisomeric compounds were modeled at the enzyme active site, corresponding to the configuration of (±)-16 seen in the X-ray cocrystal structure.[82] Enzymes of either crystal structure 2BOK or 2BQ6 were used, depending on which inhibitor the designed ligand resembled more closely. For example, the binding mode of ligand 3aS,4R,8aS,8bR-34 (Figure 14) with the isoxazolyl-chlorothienyl needle incorporated into the tricyclic scaffold resembled in the modeling very closely that of 28 (Figure 12, enzyme PDB code: 2BQ6).[101] The biaryl needle was predicted to bind in the S1 pocket (d(C–Cl···C–OH-Tyr228) = 3.1 Å) and the cation to reside in the center of the S4 pocket similarly to the structure of (±)-16 (Figure 9).

O H N N N O H N O

S

Cl 3aS,4R,8aS,8bR-34

Figure 14. Modeled ligand 3aS,4R,8aS,8bR-34 (magenta) superimposed with inhibitor 28 (green) at the active site of factor Xa of the X-ray crystal structure 2BQ6.[101] Distances

shown to (±)-34 in Å. Color code: gray Cenzyme, green Cinhibitor_28, magenta Cinhibitor_34, red O, blue N, yellow S, lime Cl.

24 2. Enhancing Binding Affinity by S1 Needle Replacement ______

Modeling of phenyl-chlorothienyl ligand 3aS,4R,8aS,8bR-35 (enzyme from 2BOK) predicted the Cl atom to be positioned at 3.3 Å distance of the ipso-C atom of Tyr228 (Figure 15). The quaternary ammonium ions and the tricyclic core of inhibitor 3aS,4R,8aS,8bR-16 and modeled 3aS,4R,8aS,8bR-35 reside at very similar positions within the pocket, indicating the latter to be a good target molecule.

O H N N N O H

S Cl

3aS,4R,8aS,8bR-35

Figure 15. Modeled ligand 3aS,4R,8aS,8bR-35 and the active enantiomer of phenylamidinium inhibitor (±)-16 at the active site of factor Xa of the X-ray crystal

structure 2BOK. Distances in Å. Color code: gray Cenzyme, green Cinhibitor_35, magenta

Cinhibitor_16, red O, blue N, yellow S, lime Cl.

2.2. Synthesis of the Inhibitors Bearing Neutral S1 Needles The key step to form the tricyclic core of the inhibitors involves a 1,3-dipolar cycloaddition reaction (Chapter 1.5). The synthesis of a central component of the reaction, maleimide 36, started by the reaction of maleic anhydride (37) with 3-aminopropyl bromide hydrobromide (38) in the presence of base to give acyclic acid 39 in nearly quantitative yield (Scheme 5).[89] However, when the previously utilized cyclization conditions to form 36, namely oxalyl chloride and DMF, were employed, partial substitution of bromine with chlorine was observed and all attempts to separate the chlorinated and brominated products, or to substitute chlorine with bromine, failed. Therefore, ring closure was carried out by heating 39 and sodium acetate in acetic anhydride,[102] giving 36 in 53% yield.

25 2. Enhancing Binding Affinity by S1 Needle Replacement ______

O Et3N, CH2Cl2, O AcONa, Ac2O, O Br 0 °C  RT OH reflux O + H3N Br H N Br 97% N Br 53% O O O 37 38 39 36

Scheme 5. Synthesis of the maleimide 36.[89,102]

The biaryl aldehydes required for the cycloaddition reaction to form the S1 needles were prepared. The synthesis of the isoxazolyl-chlorothienyl aldehyde 40 followed mainly literature-known procedures (Scheme 6).[101] The condensation of 2-acetyl-5-chlorothiophene (41) with diethyl afforded dioxobutanoate 42 in good yield. Cyclization to the isoxazole-containing ester 43 was performed with hydroxylamine hydrochloride and proceeded almost quantitatively. Finally, ester 43 was [103] reduced to the corresponding alcohol 44 with NaBH4, which was oxidized with pyridinium chlorochromate (PCC) to aldehyde 40 in 77% yield.

. NH2OH HCl, diethyl oxalate, KOtBu, EtOH, reflux, O O O O N toluene, 0 °C  RT, 18.5 h S O 17 h Cl S O Cl S Cl 84% O 98% O

41 42 43

NaBH4, 95% EtOH, RT, 4.5 d

PCC, CH2Cl2, O N O N RT, 18.5 h Cl S Cl S

O 77% OH 40 44

Scheme 6. Synthesis of 40.

The first step in the synthesis of thiazole aldehyde 45 was -bromination of 2-acetyl-5-chlorothiophene (41) to give bromide 46 in 78% yield (Scheme 7).[104] The thiazole ring was formed in a Hantzsch reaction[105] of bromide 46 and ethyl thioxamate, [101] and subsequent reduction of ester 47 with NaBH4 gave alcohol 48 in 96% yield. Oxidation[103] with PCC afforded aldehyde 45.

26 2. Enhancing Binding Affinity by S1 Needle Replacement ______

CuBr2,

CHCl3/EtOAc, ethyl thioxamate, O reflux, 53 h O EtOH, reflux, 5 h S S O Cl S Cl S Cl Br N 78% 60% O 41 46 47

NaBH4, 96% EtOH, 0 °C  RT, 24 h

PCC, CH2Cl2, S S Cl S RT, 16 h Cl S N N O 66% OH 45 48

Scheme 7. Synthesis of 45.

The synthesis of the chlorothienyl propiolaldehyde 49 was carried out in two steps (Scheme 8), beginning with Sonogashira cross-coupling of 2-bromo-5-chlorothiophene (50) and propargyl alcohol (51). Subsequent oxidation of alcohol 52 with Dess-Martin periodinane gave the desired aldehyde 49 in 41% yield.[100]

Dess-Martin

Et3N, [Pd(PPh3)4], OH periodinane, O CuI, THF, reflux, 6 h CH Cl , RT, 10 h S OH S 2 2 S Cl Br + Cl Cl 83% 41% 50 51 52 49

Scheme 8. Synthesis of aldehyde 49.

Aldehyde 53 was readily obtained by a Suzuki cross-coupling reaction of 2-bromo-5-chlorothiophene (50) and 3-formylphenylboronic acid (54) (Scheme 9).[106] The synthesis of the chloroquinoline needle 55 was carried out in one step by oxidation of [107] 7-chloro-2-quinaldine (56) with SeO2 (Scheme 9).

27 2. Enhancing Binding Affinity by S1 Needle Replacement ______

[(PPh3)2PdCl2], K2CO3, O toluene/H2O 7:1, O reflux, 96 h Cl S Br + Cl S HO 68% B OH 50 54 53

SeO2, dioxane, O Cl N reflux, 0.5 h Cl N

63%

56 55

Scheme 9. Synthesis of aldehydes 53 and 55.

The 1,3-dipolar cycloaddition of maleimide 36, L-proline (18) and aldehydes 40, 45, 49, 53, and 55, respectively, gave diastereoisomeric mixtures as products, from which the desired endo,trans-configured compounds (±)-57–(±)-59 were isolated (Scheme 10).

O O O Br MeCN, 80 °C, H H O 21–26 h N N N + + R N R N N HOOC H Ar O H O H O Ar Ar Ar

N O (±)-57 R = Br, 4% (±)-60 R = Br, 10% 36 18 40 Cl S a) a) + – + – (±)-34 R = N Me3Br , 93% (±)-65 R = N Me3Br , 87%

S (±)-61 R = Br, 13% 45 Cl S – a) N + – (±)-66 R = N Me3Br , quant.

(±)-58 R = Br, 25% 49 Cl S a) – + – (±)-64 R = N Me3Br , 68%

(±)-59 R = Br, 33% (±)-62 R = Br, 29% 53 S a) a) Cl + – + – (±)-35 R = N Me3Br , (±)-67 R = N Me3Br , 59% 67%

Cl N (±)-63 R = Br, 27% 55 – a) + – (±)-68 R = N Me3Br , 40%

Scheme 10. Synthesis of final compounds (±)-34, (±)-35, and (±)-64–(±)-68. a) Me3N in EtOH, RT or reflux, 24–144 h.

In addition, exo,trans-configured compounds (±)-60–(±)-63 were obtained. For the thiazolyl- and quinolyl-bearing compounds, the exo,trans-configured diastereoisomers

28 2. Enhancing Binding Affinity by S1 Needle Replacement ______

(±)-61 and (±)-63 were the sole compounds isolated from the mixture due to the large excess of the exo,trans-diastereoisomer formed in the reaction; the desired endo,trans-configured compounds could not be obtained. Single crystals suitable for X-ray diffraction analysis of exo,trans-configured

(±)-60 were grown from a mixture of CH2Cl2 and MeOH (Figure 16). The structure confirmed the relative configuration assigned by NMR spectroscopy (for a discussion on the NMR assignment, see chapter 3.1). In the crystal packing, stacking interactions between the biaryl S1 moieties of the two enantiomers were identified. Closest heavy atom distances between the aromatic rings were around 3.5 Å and 3.6 Å, optimal for arene–arene interactions.[10]

O H N Br N O H N O

S

Cl

(±)-60

Figure 16. Left: Small molecule X-ray crystal structure of (±)-60 confirming the relative configuration assigned by NMR spectroscopy (CCDC-702887). Right: Stacking interactions between the biaryl needles in the crystal packing. Distances in Å. Color code: green C, red O, blue N, yellow S, lime Cl, brown Br.

Finally, the quaternary ammonium ions (±)-34, (±)-35, and (±)-64–(±)-68 were obtained by treatment of the brominated precursors (±)-57–(±)-63 with Me3N in EtOH (Scheme 10).

2.3. Enzyme Assays The biological assays for factor Xa,[108] thrombin,[109] and factor IXa were performed at F. Hoffmann-La Roche in Basel, Switzerland, in the laboratory of Dr. Jacques Himber

29 2. Enhancing Binding Affinity by S1 Needle Replacement ______

and Jean-Luc Mary, with the help of Olivier Kuster. The inhibitor solutions were prepared in the laboratory of Dr. Wolfgang Haap. The biological activity against human factor Xa (Enzyme Research Laboratories) was measured at pH 7.8 using chromogenic substrate S-2222 (69, Chromogenix AB, Mölndal, Sweden; Figure 17),[110] which, upon cleavage by factor Xa, releases para-nitroaniline. The thereby induced color change, proportional to proteolytic activity, is followed spectrophotometrically (405 nm) at 25 °C. Cl– H3N NH RO O HN

O O H H H N N N N N H H O O O NO2

S-2222 (69) cleavage R = H (50%) R = CH3 (50%)

Figure 17. S-2222 (69), chromogenic substrate for factor Xa.[110]

The Ki values can be calculated from the measured IC50 values with the following formula:

Ki = IC50/(1+(S/Km));

IC50 = inhibitor concentration at which the normal enzyme activity towards a substrate is [111] inhibited by 50%, S = substrate concentration, Km = Michaelis-Menten constant.

The thrombin assay was performed similarly to that of factor Xa using chromogenic substrate S-2366 (70, Figure 18, left).[109] Human factor IXa  (Enzyme Research Laboratories) activity was measured using chromogenic substrate #299 (71, LOXO, American Diagnostica, Figure 18, right). Further details of the assays are described in Chapter 8.3.

30 2. Enhancing Binding Affinity by S1 Needle Replacement ______

O

Cl– O H3N NH H3N NH O HN HN NH O O O H O O H H N N S N N N N N H H H O O O NO2 NO2

S-2366 (70) cleavage #299 (71) cleavage

Figure 18. Left: S-2366 (70), chromogenic substrate for thrombin. Right: #299 (71), chromogenic substrate for factor IXa.

2.4. Biological Activities Biological activities were measured at F. Hoffmann-La Roche in Basel, Switzerland, as described above (Chapter 2.3). The binding affinities towards factor Xa of the ligands varied drastically upon changing the S1 needle (Figure 19). The isoxazolyl-chlorothienyl needle rendered the inhibitor highly potent: the binding affinity of (±)-34 was in the

single-digit nanomolar range (Ki = 9 nM), resulting in an increase in binding affinity of a

factor of 30 as compared to the phenylamidinium inhibitor (±)-16 (Ki = 280 nM; G = 2.0 kcal mol–1). The exo,trans-configured ligand (±)-65 was a fairly weak factor Xa inhibitor (Ki = 445 nM), proving that endo,trans-configured compounds provide the better fit into the active site of factor Xa, as predicted by modeling. In the absence of crystallographic data it remains unknown which enantiomer of the exo,trans-ligands is preferentially bound. Even the bromine-bearing precursors of (±)-34 and (±)-65, namely

(±)-57 and (±)-60, were found to be weak inhibitors of factor Xa with Ki = 10.6 M and

Ki = 47.1 M, respectively (Figure 20).

31 2. Enhancing Binding Affinity by S1 Needle Replacement ______

O O H H N N N N N N – – Br O H Br O H

O N (±)-34 (±)-65 S Cl Ki(FXa) = 9 nM Ki(FXa) = 445 nM

Ki(Thr) > 35.1 M Ki(Thr) > 35.1 M

S (±)-66 S – Cl N Ki(FXa) > 75.4 M

(±)-64 K (FXa) = 9700 nM Cl S i – Ki(Thr) > 35.1 M

Ki(FIXa) > 72.2 M

(±)-35 (±)-67 K (FXa) = 125 nM K (FXa) = 440 nM Cl S i i Ki(Thr) = 9.8 M

(±)-68 Cl N K (FXa) = 73700 M – i Ki(Thr) > 35.1 M K (FIXa) > 72.2 M i

Figure 19. Binding affinities of (±)-34, (±)-35, and (±)-64–(±)-68.

Ligand (±)-64 bearing the ethynyl-chlorothienyl S1 needle is a weak factor Xa –1 inhibitor (Ki = 9.7 M), with a loss of activity by a factor of 1000 (G = 4.3 kcal mol ) as compared to (±)-34. The Cl atom possibly approaches Tyr228 in a suboptimal angle, and the rigidity of the acetylene linker does not allow for large adjustments in the binding conformation. Relatively high binding affinity towards factor Xa was measured for the

endo,trans-configured ligand (±)-35 with a meta-substituted phenyl ring (Ki = 125 nM). In addition, (±)-35 was the only compound of the series showing activity, although weak,

towards thrombin (Ki = 9.8 M). Its exo,trans-configured counterpart (±)-67 had a

considerably weaker binding affinity (Ki = 440 nM). The exo,trans-configured thiazolyl- bearing ligand (±)-66 was inactive against factor Xa, similarly to quinolinyl ligand (±)-68.

32 2. Enhancing Binding Affinity by S1 Needle Replacement ______

O O H H N N Br N Br N O H O H N N O O

S S

Cl Cl (±)-57 (±)-60

Ki(FXa) = 10.6 M Ki(FXa) = 47.1 M

Ki(Thr) > 35.1 M Ki(Thr) > 35.1 M Ki(FIXa) > 72.2 M

Figure 20. Binding affinities of the brominated precursors (±)-57 and (±)-60 towards factor Xa, thrombin, and factor IXa for (±)-57.

A Relibase[112-114] search was performed in August 2008 to identify orthogonal dipolar C–Cl···C–O(H) contacts with distances between 2.0 and 4.0 Å between bound ligand and Tyr228 at the bottom of the S1 pocket of factor Xa. The search gave 46 hits, with the Cl···C–O(H) distances varying between 2.9 and 3.9 Å, and the Cl···C–O(H) angle adopting values between 79° and 100° (Figure 21). This very high geometrical preference for a specific distance and angle of the Cl atom in regard to Tyr228 can partly explain the drastic differences in the binding affinities of the ligands (±)-34, (±)-35, and (±)-64 (Figure 19). Modeling of isoxazolyl-bearing ligand 3aS,4R,8aS,8bR-34 (magenta, Figure 21) predicted the Cl···C–O(H) distance to be 3.1 Å and the angle to be 89°. For the modeled phenyl-bearing compound 3aS,4R,8aS,8bR-35 (light blue, Figure 21), these values were predicted to be 3.8 Å and 84°, and for the ethynyl-bearing 3aS,4R,8aS,8bR-64 (not shown) 3.3 Å and 91°. Although all angles and distances fall within the range found in the PDB search, modeling predicts the Cl atom of (±)-34, the most active inhibitor

(Ki = 9 nM), to form the closest contact with Tyr228. However, the distance of the Cl atom to Tyr228 alone is unlikely to account for the differences in the binding affinities between ligands (±)-34, (±)-35, and (±)-64, since modeling predicts the Cl atom of (±)-64,

a weak inhibitor (Ki = 9700 nM), to be closer to Tyr228 than that of (±)-35 (Ki = 125 nM). Therefore, it is presumably the rigidity of the ethynyl-chlorothienyl needle, preventing ligand (±)-64 from adjusting optimally at the active site, or the lack of additional favorable interactions as for the phenyl moiety of (±)-35 or the isoxazolyl moiety of (±)-34 in the S1 pocket, which lead to the loss of binding affinity.

33 2. Enhancing Binding Affinity by S1 Needle Replacement ______

Figure 21. Left: Relibase search in the PDB with distances between 2.0 and 4.0 Å for orthogonal dipolar C–Cl···C–O(H) contacts between bound ligand and Tyr228 at the bottom of the S1 pocket of factor Xa. Right: Overlay of some of the search results with modeled ligands 3aS,4R,8aS,8bR-34 and 3aS,4R,8aS,8bR-35. Distance in Å. Color code:

gray Cfactor Xa, magenta Cmodeled_34, light blue Cmodeled_35, yellow Cmodeled_64, red O, blue N, dark yellow S, green Cl.

To summarize, the incorporation of an uncharged S1 moiety to the tricyclic inhibitor scaffold yielded highly active factor Xa inhibitors, with the best isoxazolyl- chlorothienyl-bearing ligand (±)-34 binding with Ki = 9 nM, more active by a factor of 30 as compared to the previous phenylamidinium inhibitor (±)-16 (Ki = 280 nM). All compounds showed remarkable selectivity for factor Xa (Figure 19). Thus, the five- membered isoxazole ring was shown to be the most suitable linker between the tricyclic core and the chlorothienyl moiety, and the chlorothienyl-isoxazolyl needle was the S1 needle of choice to be incorporated into the inhibitor scaffold for further studies at the active site of factor Xa.

34 3. The Cation– Interaction ______

3. The Cation– Interaction

3.1. Quantification of Cation– Interactions in the S4 Pocket Factor Xa is an ideal enzyme for molecular recognition studies due to the highly conserved nature of its active site, as has been shown by X-ray crystal structures. This allows direct comparisons between the binding affinities of different inhibitors and thus investigations of individual interactions. Accordingly, the strength of cation– interactions in the S4 pocket of factor Xa was previously determined by comparing the binding affinities of phenylamidinium-bearing quaternary ammonium ion (±)-16 and tert-butyl inhibitor (±)-17 (Figure 22):[82] Whereas cationic (±)-16 was bound with good

affinity (Ki = 280 nM), a drastic loss of activity was observed for control compound (±)-17

(Ki = 29000 nM).

O H N R N O H

H2N NH HCl

+ – (±)-16 R = N Me3Br Ki = 280 nM (±)-17 R = CMe3 Ki = 29000 nM

Figure 22. The binding affinities of (±)-16 and (±)-17 towards factor Xa, and the quaternary ammonium ion moiety of (±)-16 bound in the S4 pocket of factor Xa as seen in the X-ray cocrystal structure (PDB code: 2BOK, resolution 1.64 Å).[82] Distances in Å.

Color code: gray Cenzyme, green Cinhibitor, red O, blue N.

The difference in activities (Ki((±)-17)/Ki((±)-16) = 100), favoring ammonium ion recognition, amounts to an increment in binding free enthalpy of –G = 2.8 kcal mol–1. This can be attributed to strong cation– interactions of the quaternary ammonium ion in the aromatic box of the S4 pocket formed by the side chains of Tyr99, Phe174, and Trp215, which corresponds to approximately 0.9 kcal mol–1 per aromatic ring. In the X-ray cocrystal structure of (±)-16 and factor Xa, only the 3aS,4R,8aS,8bR-configured enantiomer is seen bound at the active site, with the quaternary ammonium ion located in

35 3. The Cation– Interaction ______

the middle of the S4 pocket at closest heavy atom distances of d(N+···C) = 4.6–5.0 Å (Figure 22). As described in Chapter 2, inhibitors with higher binding affinity and better physicochemical profiles were targeted by exchanging the phenylamidinium moiety in

(±)-16 for a neutral S1 needle. Due to its high inhibition constant (9 nM), the isoxazolyl- chlorothienyl-bearing inhibitor (±)-34 was selected as the model compound, variation of which would allow investigations of different noncovalent interactions at the active site of factor Xa. First, quantification of cation– interactions in the S4 pocket was pursued, similarly to the earlier study with the phenylamidinium inhibitors.[82] In order to quantify the cation– interaction between quaternary ammonium ion (±)-34 and the aromatic box, the preparation of the corresponding uncharged tert-butyl derivative as a control compound was required. The synthetic strategy followed that of the phenylamidinium inhibitor (±)-17. The intermediate maleimide 72 was synthesized according to literature-known procedures (Scheme 11).[89] 3,3-Dimethylbutanol (73) was reacted with para-toluenesulfonyl chloride to give 74 in 84% yield. Compound 74 was then treated with NaCN to afford nitrile 75. Subsequent reduction using LiAlH4 gave amine 76, which was then reacted with maleic anhydride to afford acyclic 77 in quantitative yield. Finally, treatment of 77 with oxalyl chloride followed by Et3N gave maleimide 72 in 68% yield.

TsCl, pyridine, NaCN, Me2SO,

CHCl3, 0 °C, 5 h O 85 °C, 3 h S HO O 84% O 48% N 73 74 75

LiAlH4, Et2O, 47% reflux, 3.5 h 1. oxalyl chloride, DMF,

O CH2Cl2, 0 °C  RT, 48 h maleic anhydride, 2. Et N, CH Cl , RT, 24 h CH Cl , 0 °C, 1 h 3 2 2 O OH 2 2 N HN H2N 68% quant. O O

72 77 76

Scheme 11. Synthesis of tert-butyl maleimide 72.

At the time when the synthesis of the neutral ligands was started, binding affinities of the cationic inhibitors described in Chapter 2 were not known. Hence, control

36 3. The Cation– Interaction ______compounds for all cationic inhibitors bearing neutral S1 needles were targeted (Figure 13). Described herein, however, are those syntheses, which were successfully completed and where the final compounds were subjected to the biological assays.

Maleimide 72 was reacted with L-proline (18) and ethynyl-chlorothienyl needle 49 in a 1,3-dipolar cycloaddition (Scheme 12). Three diastereoisomers were isolated: (±)-endo,trans-78 in 11%, (±)-exo,trans-79 in 14%, and (±)-exo,cis-80 in 6% yield.

O O H H N N N N O H O H

S S O O MeCN, 80 °C, Cl Cl 16 h N ++ (±)-78 (±)-79 HOOC N H 11% 14% O S Cl O H 72 18 49 N N O H

S

Cl (±)-80 6%

Scheme 12. Synthesis of (±)-78–(±)-80.

The assignment of the relative configuration of compounds (±)-78–(±)-80, carried out by Prof. Bernhard Jaun at ETH Zurich, was performed by NMR spectroscopy. The signals of the protons of the tricyclic core were assigned using 1H-1H DQF-COSY spectra; the assignment for ligand (±)-78 is shown in Figure 23 as an example.

37 3. The Cation– Interaction ______O H 8 7 N 8b 8a 3a 4 N 6 O H

(±)-78 S

Cl

1 1 Figure 23. The H- H DQF-COSY spectrum (CDCl3, 400 MHz) of (±)-78, and the assignment of the proton signals of the tricyclic core.

The assignment of the relative configuration was possible because of the different 3J coupling constants of the protons H-C(4), H-C(3a), H-C(8b), and H-C(8a) of the tricyclic core (Figure 24, top, middle). For the desired endo-configured diastereoisomer (±)-78 the coupling constant between H-C(4) and H-C(3a) is large (3J = 8.2 Hz), whereas in the exo-configured ligands (±)-79 and (±)-80 the coupling constant is only about 2.5 Hz. Protons H-C(3a) and H-C(8b) are always cis to each other, resulting in a large coupling constant ( 8.3 Hz). When protons H-C(8a) and H-C(8b) are trans to each other, as in ligands (±)-78 and (±)-80, they have a small coupling constant ( 2.5 Hz). The 1H 1D-NOE difference spectra were also measured for this compound series (Chapter 8.6, Figure 82); however, the values are rather similar for all compounds, and

38 3. The Cation– Interaction ______thus inconclusive, although this was the method of choice for the assigment of the diastereoisomers of the previous phenylamidinium scaffold.[83,89]

2.6 8.4 2.7 O O O 8.3 8.2 9 8.3 H H 9 H H 9 H H 8b N 8b 8a N 8b 8a N 8a R 3a 6 R 3a 6 R 3a 6 4 N 4 N 4 N O H O H O H Ar H H Ar H Ar 2.5 2.4 8.2 (±)-78 (±)-79 (±)-80

Figure 24. The coupling constants (J [Hz]) of the hydrogens of the chiral carbon atoms of 1 ligands (±)-78–(±)-80 (top), as seen in the H NMR spectra (CDCl3, 400 MHz) (bottom).

The variation of the aromatic S1 needle does not seem to have a significant influence on the coupling pattern of the protons of the tricyclic core, and thus, all products synthesized during this thesis were assigned based on this analysis. In addition, the assignment was later confirmed by X-ray crystal structure analysis (see e.g. Chapters 2.2, 3.2, and 3.6).

The reaction of aldehyde 40 with L-proline (18) and 72 gave (±)-endo,trans-81 and (±)-exo,trans-82 in 3% and 9% isolated yield, respectively (Scheme 13).

O O H H N N O N O MeCN, 80 °C, N N O 24 h O H O H N + + N + N N HOOC H O O O S

Cl S S 72 18 40 Cl Cl

(±)-81 (±)-82 3% 9%

Scheme 13. Synthesis of inhibitors (±)-81 and (±)-82.

39 3. The Cation– Interaction ______

As expected, the biological activities of the neutral compounds (±)-78, (±)-79, (±)-81, and (±)-82 are dramatically decreased as compared to the quaternary ammonium counterparts (Figure 25). Both ethynyl-bearing compounds (±)-78 and (±)-79 were inactive against factor Xa, as well as thrombin and factor IXa. This was expected since

quaternary ammonium ion (±)-64 (Ki = 9.7 M, Figure 19), the cationic counterpart of (±)-78, is a poor factor Xa inhibitor despite undergoing cation– interactions in the S4 pocket, and the hydrophobic interactions of the tert-butyl chain in (±)-78 and (±)-79 are much weaker.

O O H H N N N N O H O H

(±)-78 (±)-79 K (FXa) > 75.4 M K (FXa) > 75.4 M Cl S i i Ki(Thr) > 35.1 M Ki(Thr) > 35.1 M

Ki(FIXa) > 72.2 M Ki(FIXa) > 72.2 M

O N (±)-81 (±)-82 S Cl Ki(FXa) = 0.550 M Ki(FXa) = 75.3 M

Ki(Thr) = 17.8 M Ki(Thr) > 35.1 M

Figure 25. The biological activities of the tert-butyl derivatives (±)-78, (±)-79, (±)-81, and (±)-82.

The isoxazolyl-bearing ligand (±)-81, however, is a fairly active factor Xa

inhibitor (Ki = 550 nM), as well as a weak thrombin inhibitor (Ki = 17.8 M, Figure 25). Its exo,trans-configured counterpart (±)-82 is inactive against both factor Xa and thrombin.

Comparing the binding affinity of quaternary ammonium ion (±)-34 (Ki = 9 nM)

with its tert-butyl counterpart (±)-81 (Ki = 550 nM; Ki((±)-81/(±)-34) = 61) allowed the quantification of cation– interactions in the S4 pocket as –G = 2.5 kcal mol–1, approximately 0.8 kcal mol–1 per aromatic ring. This is well in accordance with results previously obtained with the phenylamidinium inhibitor scaffold,[82] and underlines that the cation– interaction really is among the strongest noncovalent binding forces in biological complexation.[6,11]

40 3. The Cation– Interaction ______

3.2. X-Ray Cocrystal Structure of (±)-34 with Factor Xa In order to verify the binding mode of (±)-34 at the active site of factor Xa, crystals of endo,trans-configured (±)-34 in complex with factor Xa were grown at F. Hoffmann-La Roche in Basel, Switzerland. After pre-incubating (±)-34 with short-form factor Xa, [82,115] produced as Arg150Glu mutant, addition of 0.1 M Bis-Tris buffer (bis(2-hydroxyethyl)aminotris(hydroxymethyl)methane; pH 6.5) with 25% polyethylene glycol 3350 using the microbatch method gave crystals grown to full size after a week. The data were collected at the Swiss Light Source at the Paul Scherrer Institute in Switzerland. The X-ray cocrystal structure was solved at 1.25 Å resolution by Dr. David Banner (Figure 26, PDB code: 2JKH). The structure confirmed that, similarly to (±)-16 (PDB code: 2BOK),[82] only the 3aS,4R,8aS,8bR-configured enantiomer of (±)-34 is bound at the active site of factor Xa. The binding conformation of 3aS,4R,8aS,8bR-34 resembles very closely the one predicted by modeling with Moloc (Figure 14).[90,91] In the crystal structure, one carbonyl of the maleimide core is directed towards N–H of Gly216 forming a hydrogen bond (d(N(H)···O=C) = 3.5 Å, Figure 26), albeit a quite weak one compared to that formed by (±)-16 (d(N(H)···O=C) = 3.1 Å).[82] For the tricyclic phenylamidinium inhibitors binding to thrombin, this hydrogen bond was found to contribute –G = 0.8 kcal mol–1.[84]

Figure 26. The X-ray cocrystal structure of the 3aS,4R,8aS,8bR-configured enantiomer of (±)-34 in complex with factor Xa at 1.25 Å resolution (PDB code: 2JKH). Distances in

Å. Color code: gray Cenzyme, green Cinhibitor, red O, blue N, yellow S, lime Cl.

41 3. The Cation– Interaction ______

The tricyclic scaffold directs the chlorothienyl moiety into the S1 pocket where the Cl atom interacts with the side chain of Tyr228 (d(Cl···C–OH) = 3.6 Å, angle(C– Cl···Tyr228-plane) = 66°), and the surrounding amino acids Ala190, Val213, Gly226, and Ile227 (d(Cl···C/N) = 3.5–3.8 Å) (Figure 27). The Cl atom of the S1 vector replaces a structural water molecule that is present in the phenylamidinium structure (Figure 30), as suggested in the literature.[94,116-119]

Figure 27. Chlorothienyl moiety of (±)-34 bound in the S1 pocket of factor Xa (PDB

code: 2JKH). Distances in Å. Color code: gray Cenzyme, green Cinhibitor, red O, blue N, yellow S, lime Cl.

The quaternary ammonium ion binds in the S4 pocket, where it is located in the center of the aromatic box, on the intersection of the normals passing through the centroids of the aromatic rings of the side chains of Trp215, Phe174, and Tyr99 (Figure 28). The distances between the N+ center and the closest aromatic C atom (d(N···C- Trp215) = 4.3 Å; d(N···C-Phe174)= 4.7 Å; d(N···C-Tyr99) = 4.8 Å) correspond well to the values previously determined for (±)-16 bound to factor Xa.[82]

Figure 28. The quaternary ammonium ion of (±)-34 in the S4 pocket of factor Xa (PDB

code: 2JKH). Distances in Å. Color code: gray Cenzyme, green Cinhibitor, red O, blue N.

42 3. The Cation– Interaction ______

The biaryl vector of (±)-34 undergoes very efficient stacking interactions with the surrounding flat and highly polarizable walls of the S1 pocket (shortest heavy atom distances 3.3–3.7 Å; Figure 29), namely the backbone of Gly216 and Trp215 on one side and Gln192-Cys191-Ala190 on the other, which contribute to the high binding affinity of the inhibitor (for a more detailed analysis of the stacking, see Chapter 5 and Figure 55).

Figure 29. The biaryl needle of (±)-34 undergoes efficient stacking interactions with the peptide walls lining the S1 pocket (PDB code: 2JKH). Distance in Å. Color code: gray

Cenzyme, green Cinhibitor, red O, blue N, yellow S, lime Cl.

The enzyme in the X-ray cocrystal structures of both (±)-16 and (±)-34 is very well conserved, underlining the suitability of factor Xa for molecular recognition studies (Figure 30). Yet, the binding mode of (±)-34 differs significantly from that of (±)-16.

43 3. The Cation– Interaction ______

Figure 30. Superimposition of the X-ray cocrystal structures of (±)-16 and (±)-34, showing the 3aS,4R,8aS,8bR-configured enantiomer bound at the active site (PDB code:

2BOK and 2JKH, respectively). Color code: gray Cenzyme_2JKH, purple Cenzyme_2BOK, green

Cinhibitor_34, magenta Cinhibitor_16, red O, blue N, yellow S, lime Cl.

The rigid tricyclic central scaffold undergoes a dramatic repositioning to reach either the side chain of Asp189 with the phenylamidinium residue in the case of (±)-16, or the side chain of Tyr228 with the chlorothienyl residue in the case of (±)-34 (Figure 31). Atoms of the tricyclic scaffold are shifted by up to 6.5 Å in the two structures.

Figure 31. The tricyclic core undergoes a drastic repositioning upon exchanging phenylamidinium ((±)-16, magenta) for the isoxazolyl-chlorothienyl moiety ((±)-34,

green). Distances in Å. Color code: gray Cenzyme_2JKH, green Cinhibitor_34, magenta

Cinhibitor_16, red O, blue N, yellow S, lime Cl.

44 3. The Cation– Interaction ______

Despite the extent of the repositioning, the space occupied by the tricyclic core at the active site is large enough to accommodate both binding modes, since both (±)-16 and (±)-34 are active inhibitors of factor Xa. Importantly, the location of the ammonium ion moiety in the S4 pocket is unchanged despite the shifting of the tricyclic core (Figure 31), underlining the importance of the cation– interaction. The tricycle repositioning is the result of the requirement of the chlorothienyl needle of (±)-34 to interact with Tyr228 on one hand, and the phenylamidinium needle of (±)-16 with Asp189 on the other.

3.3. Identifying the Active Enantiomer of (±)-34 In order to verify which of the (+)- or (–)-enantiomers corresponds to the 3aS,4R,8aS,8bR- configured one bound at the active site of factor Xa in the X-ray cocrystal structure of (±)-34 (PDB code: 2JKH) and to determine the degree of enantioselectivity of the binding process, enantiomerically pure material was required. Due to the low solubility of ammonium ion (±)-34 in organic solvents, the chiral resolution of precursor (±)-57 was performed by HPLC on a chiral stationary phase (Chiralpak AD) at F. Hoffmann- La Roche in Basel, Switzerland, by David Wechsler and Daniel Zimmerli. The enantiomers were subsequently converted into the quaternary ammonium derivatives

(+)-34 and (–)-34 by reaction with Me3N in excellent yield (Scheme 14). The enantiomers were tested for their biological activity towards factor Xa and for selectivity towards thrombin as described in Chapter 2.3. Enantiomer (+)-34 has a much

higher affinity towards factor Xa than (–)-34 (Ki = 5 nM and 271 nM, respectively), and consequently, the absolute configuration of (+)-34 can be unambiguously assigned as 3aS,4R,8aS,8bR. Thus, although (–)-34 is also a relatively active factor Xa inhibitor, on the basis of the loss of binding affinity by a factor of about 50 as compared to (+)-34 and the

similarity of the activities of the racemic ligand (±)-34 (Ki = 9 nM) and (+)-34, it can be concluded that the binding of the ligands to factor Xa is quite enantioselective, and biological activity of the racemic compounds stems mainly from the 3aS,4R,8aS,8bR- configured enantiomers.

45 3. The Cation– Interaction ______

O O H H (+)-57 N (–)-57 N R R = Br R = Br R N Me3N in EtOH, N O O H EtOH, RT, 46 h H N N O O (+)-34, 89% (–)-34, 94% R = N+Me Br– R = N+Me Br– S 3 3 S Ki = 5 nM Ki = 271 nM Cl Cl

Scheme 14. Synthesis and binding affinities towards factor Xa of enantiomerically pure (+)-34 and (–)-34.

3.4. Chain Length Variation The correct positioning of the cation with respect to the arene is crucial for optimal cation– interactions.[6] In a study on a -hairpin peptide, Waters and co-workers showed the hairpin stability to decrease, as a sign of weakened cation– interactions, upon shortening of the chain of the ammonium ion interacting with Trp.[120] In order to investigate the position of the quaternary ammonium ion in the S4 pocket, a series of molecules was targeted with the length of the alkyl spacer connecting the onium ion + Me3N –R and the tricyclic core of C2, C4, and C5, in addition to C3 in (±)-34. This work was part of the student semester project and Master’s Thesis of Christoph Bucher.

First, a series of maleimides bearing C2, C4, and C5 spacers had to be prepared. Brominated amines 83, 84, and 85 for the maleimide synthesis were obtained as HBr salts from the corresponding hydroxyl alkyl amines 86, 87, and 88 by heating to reflux in concentrated HBr (Scheme 15). The amines were then reacted with maleic anhydride, and acids 89 and 90 were obtained in excellent yield. However, the reaction to the acid

bearing the C2 spacer did not give the desired product. Subsequently, acids 89 and 90 were cyclized to the corresponding maleimides 91 and 92 by reaction with oxalyl chloride followed by treatment with Et3N in 75% and 57% yield, respectively.

For C5 maleimide 92, partial chlorination of the desired product occurred during the reaction. However, since only the brominated product reacts in the last synthetic step to form the ammonium ion, this did not interfere with the synthesis.

46 3. The Cation– Interaction ______

1. oxalyl chloride,

DMF, CH2Cl2, maleic anhydride, 0 °C  RT, 2 d

conc. HBr, Et3N, CH2Cl2, 2. Et3N, CH2Cl2, O 0 °C RT, 1.5–2.5 h O OH Br OH reflux, 3–3.5 h Br  HN RT, 0.5 h H N H N n–1 N 2 n–1 3 n–1 Br Br O O n–1 86 n = 2 83 n = 2, 67% 89 n = 4, 96% 91 n = 4, 75% 87 n = 4 84 n = 4, 50% 90 n = 5, 86% 92 n = 5, 57% 88 n = 5 85 n = 5, 39%

Scheme 15. Synthesis of the maleimides 91 and 92.

Due to the failed reaction of 2-aminoethylbromide hydrobromide (83) with maleic anhydride (37), an alternative synthesis route was developed. Maleimide (93) was reacted under basic conditions with methyl chloroformate to give 94 (Scheme 16),[121] which was, without further purification, treated with ethanolamine to give alcohol 95 in 47% overall yield.[122] Alcohol 95 was then converted to the desired bromide 96 in an Appel reaction in 37% yield.

methyl chloroformate, ethanolamine, sat. aq.

N-methylmorpholine, NaHCO3 solution, CBr4, PPh3,

O EtOAc, 0 °C, 30 min O 0 °C, 30 min O CH2Cl2, O  RT, 45 min O  RT, 15 min OH 0 °C  RT, 23 h Br NH N N N O 47% over 2 steps 37% O O O O 93 94 95 96

Scheme 16. The synthesis of C2 maleimide 96.

Maleimides with varying chain length, namely maleimide 96 with a C2 chain, 91

with a C4 chain, and 92 with a C5 chain, were each reacted with L-proline (18) and aldehyde 40 in the 1,3-dipolar cycloaddition (Scheme 17). Unfortunately, despite many attempts, the purification process of the C2 derivative was not successful and the

compound could not be isolated. However, C4 derivative (±)-97 and C5 derivative (±)-98 were both isolated in 2% yield. These were converted further into ammonium ions (±)-99

and (±)-100 in a following reaction with Me3N in 38% and 60% yield, respectively.

47 3. The Cation– Interaction ______

O H N O O MeCN, 80 °C, R N O Br 16–44 h n–1 N N + + O n–1 N H HOOC H S O N O Cl 91 n = 4 18 40 S 92 n = 5 Cl (±)-97 n = 4; R = Br, 2% (±)-98 n = 5; R = Br, 2% Me3N in EtOH,

EtOH, RT, + – (±)-99 n = 4; R = N Me3Br , 38% 48–53 h + – (±)-100 n = 5; R = N Me3Br , 60%

Scheme 17. Synthesis of (±)-99 and (±)-100.

Binding affinities of inhibitors (±)-99 and (±)-100 were measured following the previously described protocols (Chapter 2.3). The complexation of (±)-34 (with a

C3 chain) and (±)-99 (with a C4 chain) is equally efficient (Ki = 9 nM), ligand (±)-100,

however, with a longer C5 spacer is a much weaker binder (Ki = 146 nM) (Figure 32). All compounds show excellent selectivity against thrombin.

Compound R Ki(FXa) Ki(Thr) O H N R N (±)-34 9 nM > 35.1 μM Br N O H N N O (±)-99 9 nM > 35.1 μM

S

Cl N (±)-100 146 nM > 35.1 μM

Figure 32. Binding affinities of the compounds towards factor Xa and thrombin with varying alkyl spacers (±)-34, (±)-99, and (±)-100.

Clearly, factor Xa shows some tolerance for the variation of the chain length. This can be attributed to the space accommodating the tricyclic core, which is large enough to allow for changes in the binding conformation as shown by the comparison of the X-ray cocrystal structures of (±)-16 and (±)-34. For optimal binding, the cation needs to be positioned in the middle of the aromatic box with the N+ center on the intersect of the normals passing through the centers of the aromatic rings, which is possible for (±)-34

48 3. The Cation– Interaction ______

and (±)-99, but only in a strained gauche conformation for (±)-100 (Figure 33). Ligand (±)-100 is a weaker inhibitor than (±)-34 by a factor of 16, amounting to a loss in binding free enthalpy of about 1.6 kcal mol–1, corresponding well with the energy attributed to two gauche conformations ( 1.8 kcal mol–1).[123]

Figure 33. Compound 3aS,4R,8aS,8bR-100 bearing a C5 linker as modeled by [90,91] Moloc. Color code: gray Cenzyme, green Cinhibitor, red O, blue N, yellow S, lime Cl.

This is shown in the modeling for (±)-100 (Figure 33), where the C5 chain needs to adopt a double gauche conformation in order to position the ammonium ion in the middle of the S4 pocket, thus leading to a loss of binding affinity. Thus, the C3 alkyl chain was the linker of choice to be incorporated into ligands designed for further exploration of cation– interactions in the S4 pocket.

3.5. Probing the Size of the S4 Pocket To investigate the volume of the aromatic S4 pocket and the influence of the size of the cationic moiety on the cation– interaction, a series of ligands was designed, which feature quaternary ammonium ion moieties larger than trimethylammonium in (±)-34. Increase in binding affinity could be expected through incorporation of larger cations with further positively polarized C–H moieties. Modeling with Moloc[90,91] predicted target compounds (±)-101–(±)-106 (Scheme 18) to bind to factor Xa in a similar manner to that of (±)-34, as can be seen for ligand (±)-102 bearing a methylpyrrolidinium S4 moiety (Figure 34).

49 3. The Cation– Interaction ______

O H N N N O H N O

S

Cl 3aS,4R,8aS,8bR-102

Figure 34. Compound 3aS,4R,8aS,8bR-102 modeled at the active site (2JKH) of factor

Xa. Distances in Å. Color code: gray Cenzyme, green Cinhibitor, red O, blue N, yellow S, lime Cl.

All compounds were synthesized in a straightforward manner using bromide (±)-57 as common precursor, which was reacted with the corresponding tertiary amines in a substitution reaction to afford the cationic target molecules (±)-101–(±)-106 (Scheme 18). Although reaction times are rather long, ranging from 3 h to 7 days, the products were obtained in moderate to quantitative yields, and were readily isolated through

precipitation from MeOH and Et2O.

50 3. The Cation– Interaction ______

O O H H N N Br N R N O H O H N solvent Br N O + tert. amine O

S S

Cl Cl

(±)-57

Compound RSolvent T Time Yield

(±)-101 N EtOH RT 4 d 81%

(±)-102 N EtOH RT 7 d 71%

(±)-103 N acetone reflux 4.5 d 66%

(±)-104 N EtOH RT 45 h 77%

N (±)-105 N EtOH RT 16 h 41%

(±)-106 N pyridine reflux 3 h quant.

Scheme 18. Synthesis of the cationic products (±)-101–(±)-106.

The biological activities (see Chapter 2.3) of the series show that the aromatic box also accepts larger quaternary onium ions than trimethylammonium (Figure 35).

Ethyldimethylammonium ion (±)-101 (Ki = 8 nM) is as active as the trimethylammonium ion (±)-34. Expanding the interaction surface further by positively polarized C–H residues enhanced the cation– interactions in the S4 pocket, as seen for N-methylpyrrolidinium ion (±)-102 and N-methylpiperidinium ion (±)-103, both binding

with Ki = 5 nM and thus showing the highest binding affinities of the series.

51 3. The Cation– Interaction ______

N Br

(±)-34

Ki(FXa) = 9 nM

Ki(Thr) > 35.1 M clogD = –3.64 logD = –1.46 O H N N N N Br Br Br R N O H (±)-101 (±)-102 (±)-103 N Ki(FXa) = 8 nM Ki(FXa) = 5 nM Ki(FXa) = 5 nM O Ki(Thr) > 35.1 M Ki(Thr) = 29.7 M Ki(Thr) = 16.0 M clogD = –3.73 clogD = –3.31 clogD = –3.03 S

Cl N N N Br N Br Br

(±)-104 (±)-105 (±)-106

Ki(FXa) = 12 nM Ki(FXa) = 50 nM Ki(FXa) = 30 nM

Ki(Thr) > 35.1 M Ki(Thr) > 35.1 M Ki(Thr) > 35.1 M clogD = –3.06 clogD = –3.26 clogD = –2.92

Figure 35. The biological affinities towards factor Xa and thrombin and the clogD values of the cationic products (±)-34, (±)-101–(±)-106. The clogD values were calculated (April 2011) at pH 7.4 with ACD/Labs Software, Release 12.00, Product Version 12.01 (see Chapter 8.1).

A further expansion of the cation in quinuclidinium ion (±)-104, results in a slight decrease in activity (Ki = 12 nM), indicating that this residue is slightly too large to fit comfortably into the S4 pocket. Activity was lost even further with the monoquaternized

1,4-diazabicyclo[2.2.2]octane (DABCO) derivative (±)-105 (Ki = 50 nM). This may be due to the higher desolvation penalty upon incorporation of (±)-105 into the active site as compared to (±)-104. In agreement with this, the calculated logarithmic distribution coefficient (clogD) of DABCO derivative (±)-105 (–3.26) is considerably more negative than that of quinuclidinium ion (±)-104 (–3.06). In addition, the N atom of DABCO in position 4 may engage in some repulsive contacts at the back of the S4 pocket. Indeed, modeling shows the ammonium moiety slightly moved away from the center of the S4 pocket as compared to (±)-34, which could be due to repulsions with the backbone carbonyl of Thr98-Tyr99 peptide bond (Figure 36).

52 3. The Cation– Interaction ______

Figure 36. Left: Ligand 3aS,4R,8aS,8bR-105 modeled at the active site of factor Xa. Right: Superimposition of modeled 3aS,4R,8aS,8bR-105 and crystal (±)-34 (PDB code:

2JKH) bound in the S4 pocket. Color code: gray Cenzyme, magenta Cinhibitor_105, green

Cinhibitor_34, red O, blue N, yellow S, lime Cl.

The flatter N-pyridinium ion (±)-106 has a weaker binding affinity (Ki = 30 nM) than (±)-34, indicating that a bulkier alkyl chain-bearing onium ion moiety is required to obtain inhibitors with high affinity. Tsuzuki et al. calculated slipped parallel as the most favorable geometry for the benzene–N-methylpyridinium complex, with the size of the interaction strongly dependent on the orientation of benzene.[39] Thus, the optimization of the interaction geometry might not be possible in the S4 pocket, leading to weaker activity as compared to (±)-34. To summarize, some binding affinity can be gained through incorporation of larger cations in the S4 pocket, with pyrrolidinium ion (±)-102 and piperidinium ion (±)-103 as most active inhibitors of the series. Bicyclic quinuclidinium and DABCO derivatives seem too large for the pocket as a loss of binding affinity was observed.

3.6. Methylation Series and Uncharged Control Compounds Lysine side chain methylation in histones is an epigenetic modification, which is critical in gene expression. N-Methylated Lys side chains have been shown to bind in aromatic environments through cation– interactions, which raises high interest to understand the effect of N-methylation on the interaction.[14,120,124] Correspondingly, many such studies have emerged in the literature.

53 3. The Cation– Interaction ______

Patel and co-workers showed by NMR, calorimetric, and surface plasmon resonance (SPR) studies that the binding of Lys4 of histone 3 (H3K4) to the aromatic pocket of the plant homeodomain (PHD) finger of bromodomain PHD finger transcription factor (BPTF) is enhanced by increasing the degree of Lys side chain methylation (Figure 37).[14] Dissociation constants were estimated from isothermal titration calorimetry (ITC)

measurements as Kd  2.7 M for H3K4me3 (LysMe3) and Kd  5.0 M for H3K4me2

(LysMe2).

Figure 37. Binding of trimethylated lysine of H3K4me3 in the aromatic pocket of BPTF [14] PHD finger (PDB code: 2F6J). Color code: gray CBPTF PHD finger, green CH3K4me3, red O, blue N.

Waters and co-workers studied the effect of methylation on the cation– interaction in their -hairpin system 107 (Figure 38).[120,125-127] When the hairpin is folded, the side chain of Lys is in proximity to the side chain of Trp. By determining the extent of the folding of the peptide by NMR studies, folding free energies G were determined. Methylation increased the stability of the hairpin by –G = 0.3 (Lys–Trp), –1 0.5 (LysMe–Trp), 0.7 (LysMe2–Trp), and 1.0 (LysMe3–Trp) (± 0.1) kcal mol (T = 298 K), as evaluated with double-mutant cycles in buffer solution, amounting to an enhancement of the cation– interaction of about 0.2–0.3 kcal mol–1 per methylation.[120] With each methylation, the folding entropy became more favorable, consistent with increased interaction sites of the lysine side chain with the aromatic ring. The enthalpic driving force, in contrast, was reduced with N-methylation, which was explained through increasingly distributed positive charge across the N-methyl groups. In a related study, both asymmetric as well as symmetric dimethylation of Arg were shown to stabilize the hairpin by –G = 1.0 ± 0.1 kcal mol–1.[127] Similar binding enhancements upon Lys

54 3. The Cation– Interaction ______

methylation were observed by the same group using a synthetic Dougherty-type cyclophane host with disulfide bridges.[128]

H2N NH2 HN

O O O H H H N N N NH N N N 2 H H H O O O O R = Thr NH 1 O Leu O O NH R2 = Lys O R1 O R2 O H H H LysMe N N N H2N N N LysMe H H 2 O O O LysMe3 tBuNle O NH2 H3N 107

Figure 38. -Hairpin peptide model system 107 employed by Waters and co-workers to study the effect of N-methylation on cation– interactions.[120] Nle = norleucine.

Hof and co-workers observed an increase in the binding affinity of Lys and Arg upon methylation to a p-sulfonatocalix[4]arene host by 1H NMR spectroscopy in water [124] (T = 298 K). LysMe3 was bound 70-fold more strongly than its non-methylated counterpart. ITC studies showed a strong enthalpic driving force and a smaller favorable entropic component for the complexation, with the enthalpic component increasing –1 significantly with each methylation (–H: LysMe  LysMe2: 0.8 kcal mol ; LysMe2 –1  LysMe3: 0.5 kcal mol ). In order to systematically investigate the effect of methylation on the cation– interaction in the S4 pocket of factor Xa, ligands featuring primary, secondary, and tertiary amine S4 moieties were targeted. This work was performed by Christoph Bucher during the student semester project and Master’s Thesis. To obtain primary amine

(±)-108, (±)-57 was treated with NaN3 to give (±)-109 in 96% yield (Scheme 19). Azide (±)-109 was converted into primary amine (±)-108 in good yield using Staudinger reaction conditions. An intramolecular condensation of (±)-108 to a cyclic product, as detected by LC-MS, seems to occur at RT during storage, which was prevented by storing the compound at 278 K. Different conditions were investigated to obtain secondary amine (±)-110.

Treating (±)-57 with MeNH2 in EtOH resulted in a mixture of products, containing e.g.

55 3. The Cation– Interaction ______

the product with ring-opened maleimide. The reaction of (±)-57 with MeNH2·HCl in DMF afforded the chlorinated equivalent of (±)-57 by substitution of bromide. Reactions

performed with MeNH2 in THF seemed to afford the desired product (±)-110, but with very long reaction times. However, reaction time was reduced under basic conditions, and consequently, amine (±)-110 was obtained in 32% yield by treating (±)-57 with

MeNH2 in THF using K2CO3 (Scheme 19). Tertiary amine (±)-111 was obtained by

treatment of (±)-57 with aq. Me2NH in EtOH in 72% yield (Scheme 19). O O O H H H N N N N3 N Br N N N O H O H O H NaN3, H2O, DMF, Me2NH in H2O, N N N RT  45 °C, 7 h EtOH, RT, 23 h O O O 96% 72% S S S

Cl Cl Cl

(±)-109 (±)-57 (±)-111

PPh3, cat. H2O, MeNH2 in THF, K2CO3, THF, RT, 19 h DMF, RT, 56 h 70% 32%

O O H H N N H2N N N N H O H O H N N O O

S S

Cl Cl (±)-108 (±)-110

Scheme 19. The synthesis of primary amine (±)-108, secondary amine (±)-110, and tertiary amine (±)-111 starting from the common precursor (±)-57.

Crystals for small molecule X-ray analysis of (±)-111 were grown from a mixture

of MeOH and Et2O (Figure 39), confirming the endo,trans-configuration deduced from NMR analyses (vide supra). The structure shows the heteroatoms of the biaryl needle residing on the same side, and opposite to the pyrrolidinyl N atom possibly to avoid repulsion with the lone pair, similarly to the conformation seen in the X-ray cocrystal of (±)-34 with factor Xa (Figure 26).

56 3. The Cation– Interaction ______

O H H N N N – Br O H N O

S

Cl

(±)-111

Figure 39. Above: Small molecule X-ray crystal structure of (±)-111 (CCDC-702886). Below: The unit cell of the crystal structure. Hydrogens are omitted for clarity. Color code: green C, red O, blue N, yellow S, lime Cl, brown Br.

The compounds of the methylation series were measured for their binding affinities following the procedures described in Chapter 2.3. Under the conditions of the

assay at pH 7.8, all terminal amines are protonated (for calculated pKa values, see Figure 40). A dramatic dependence of the magnitude of the binding affinities and thus, cation– interactions, on the degree of N-methylation of the terminal amine center was found: the

primary ammonium ion (±)-108 is a poor binder (Ki = 9800 nM, logD = –0.04); its affinity for factor Xa is much lower than that of the tert-butyl derivative (±)-81 (Ki = 550

nM) (Figure 40). Binding affinity increases with each additional methyl group, so that the

secondary ammonium ion (±)-110 binds with modest affinity (Ki = 911 nM). A strong increase in affinity is observed upon further methylation to the tertiary ammonium ion

(±)-111 (Ki = 56 nM, logD = 0.77) and ultimately to the quaternary (±)-34 (Ki = 9 nM, logD = –1.46).

57 3. The Cation– Interaction ______

O H N HN N N H2N R N Br O H N (±)-108 (±)-110 (±)-111 (±)-34 O Ki(FXa) = 9800 nM Ki(FXa) = 911 nM Ki(FXa) = 56 nM Ki(FXa) = 9 nM

Ki(Thr) > 35.1 M Ki(Thr) > 35.1 M Ki(Thr) > 35.1 M Ki(Thr) > 35.1 M S clogD = –2.59 clogD = –2.72 clogD = –1.64 clogD = –3.64 logD = –0.04 logD = n.d. logD = 0.77 logD = –1.46 Cl pKa = 9.7 ± 0.1 pKa = 10.3 ± 0.1 pKa = 9.5 ± 0.3

G / kcal mol–1 1.5 1.8 1.2

Figure 40. Biological activities of the methylation series (±)-108, (±)-110, (±)-111 and (±)-34. The clogD values were calculated (April 2011) at pH 7.4 with ACD/Labs Software, Release 12.00, Product Version 12.01 (see Chapter 8.1). The logD values were

measured at F. Hoffman-La Roche, Basel, Switzerland (see Chapter 8.1). The pKa values of the conjugate acids of the terminal amines were calculated with the software ACD pKa from ACD/Labs. FXa = factor Xa; Thr = thrombin.

In the case of the primary and secondary ammonium ions, desolvation of three ((±)-108) and two ((±)-110) N+–H residues must occur upon insertion into the aromatic box and these desolvation costs are not recovered by the gain in cation– interactions. The much stronger binding of the tertiary ammonium ion (±)-111 is in agreement with the results of a previously reported PDB search:[82] while the three alkyl residues undergo favorable C–H··· interactions in the aromatic box, the N+–H residue presumably reaches towards the bulk water, where it can become solvated by one or more water molecules, which are hydrogen-bonded to the protein. On average, the gain in binding free enthalpy amounts to an increment of –G = 1.2–1.8 kcal mol–1 per methylation. This corresponds to a value of –G = 0.4–0.6 kcal mol–1 per aromatic ring. While increases in lipophilicity upon N-methylation certainly also make a contribution to the enhancement in affinity, the clogD data (Figure 40) clearly show it to be a rather minor factor. The weak binding of the non-methylated ammonium inhibitor (±)-108 suggests that the cation– interaction between primary ammonium ion moieties, such as Lys, and aromatic rings in the absence of a counterion can be discarded. Interestingly, Berry et al. reported a strong solvation dependence of the interaction of the side chain of Lys with [129] that of Trp in an 3Trp model protein. Using double-mutant cycles, solvation effects were studied by shuffling the order of the amino acids. Upon exposing the interaction to

58 3. The Cation– Interaction ______

the solvent, the interaction energy for Trp–Lys was found to decrease by 0.9 kcal mol–1, from –0.73 to +0.15 kcal mol–1, suggesting that solvent-exposed cation– interactions are destabilizing or weak at best. Thus, Lys presumably rather interacts with aromatic rings through C–H··· interactions with its side chain, as proposed by Gallivan and Dougherty,[20] while the primary ammonium turns away from the aromatic ring in order to benefit from solvation.

To investigate the role of the increased volume, and thus more suitable fit of the S4 moiety, in the enhancement of the cation– interaction upon methylation, an uncharged methylation series (±)-112–(±)-116 was designed (Figure 41). The synthetic strategy followed that of tert-butyl inhibitor (±)-81, and began with the preparation of the alkyl-bearing maleimides. Compound (±)-116 was prepared by Sandra Kienast during the Organic Chemistry Laboratory Course 2.

Compound R

(±)-112

O H R (±)-113 N N O H N (±)-114 O

S (±)-115 Cl

(±)-116

Figure 41. The uncharged control compound series (±)-112–(±)-116.

N-Methylmaleimide (117), required for the synthesis of (±)-112, was commercially available. Butyl- and pentyl-bearing maleimides 118 and 119 were obtained, similarly to 72, via reaction of maleic anhydride (37) with the corresponding amines 120 and 121, giving acyclic acids 122 and 123, respectively, in excellent yield (Scheme 20). Cyclization with oxalyl chloride and subsequent treatment with base gave maleimides 118 and 119 in 71% and 82% yield, respectively.

59 3. The Cation– Interaction ______

maleic anhydride (37), 1. oxalyl chloride, DMF, O CH2Cl2, 0 °C  RT, CH2Cl2, 0 °C  RT, 2–7 d O R 0.5–72 h N R 2. Et3N, CH2Cl2, RT, 7–20 h H2N R H N O

OH O

120 R = CH3 122 R = CH3, 89% 118 R = CH3, 71%

121 R = CH2CH3 123 R = CH2CH3, 93% 119 R = CH2CH3, 82%

127 R = CH(CH3)2 124 R = CH(CH3)2, 52%, over 2 steps

Scheme 20. Synthesis of the alkyl-bearing maleimides 118, 119, and 124.

The first synthetic approach to obtain 4-methylpentyl maleimide 124 was via Mitsunobu reaction[130] of maleimide (93) and 4-methyl-1-pentanol (125) (Scheme 21). However, no product formation was observed. Using a Gabriel synthesis strategy, alcohol 125 was first activated with a tosyl group to give 126, and subsequently substituted by phthalimide to afford 127. Both steps proceeded smoothly, but all hydrazinolysis and acidic hydrolysis conditions failed to yield the desired ammonium salt or amine 128 as product. Finally, amine 128 was obtained by reduction of isocapronitrile

(129) with LiAlH4 in low yield. Amine 128 was subsequently converted into the corresponding maleimide 124 in 52% yield over 2 steps following the procedures described for butyl- and pentylmaleimide 118 and 119 (Scheme 20).

pTsCl, Et3N, DMAP, potassium phthalimide, O OH CH2Cl2, RT, 24 h OTs DMF, reflux, 24 h N

82% 73% O 125 126 127

maleimide (93), PPh3, DIAD, . NH2NH2 H2O, neopentyl alcohol, THF, CH2Cl2, MeOH, –78 °C  RT RT or reflux

aq. 6 N HCl, reflux O

N 48% HBr, AcOH, reflux O 124 LiAlH4, Et2O, N RT  reflux, 4.5 h NH2 25% 129 128

Scheme 21. Synthesis of 4-methylpentan-1-amine (128). In the first step of the synthesis of cyclohexyl-bearing maleimide 130, 3-cyclohexylpropanol (131) was reacted with para-toluenesulfonyl chloride to give 132 in

60 3. The Cation– Interaction ______

50% yield (Scheme 22). The activated alcohol was then substituted by phthalimide to afford 133 in 82% yield, and the amine subsequently deprotected using hydrazine monohydrate to give 134. Amine 134 was then reacted with maleic anhydride (37) to 135 in quantitative yield. In the final step, cyclization of the maleimide using oxalyl chloride and subsequent treatment with base gave maleimide 130 in excellent yield. It should be noted, however, that purification of the cyclohexyl-bearing compounds proved to be problematic, and thus crude products were characterized.

pTsCl, pyridine, potassium phthalimide, O CHCl , 0 °C, 5 h DMF, 85 °C, 18 h N OH 3 OTs 50% 82% O

131 132 133

1. oxalyl chloride, NH2NH2•H2O, 43% CH Cl , MeOH, DMF, CH2Cl2, 2 2 0 °C  RT, 15 d maleic anhydride (37), RT, 4 d 2. Et N, CH Cl , O CH Cl , 0 °C O 3 2 2 2 2  RT, 5 d RT, 4 d N NH2 N H 94% HO quant. O O 130 135 134

Scheme 22. Synthesis of the cyclohexyl-bearing maleimide 130.

The synthesis of the uncharged control compounds (±)-112–(±)-116 followed a protocol corresponding to that of the tert-butyl inhibitor (±)-81, with a 1,3-dipolar cycloaddition of the neutral maleimide, L-proline (18), and the biaryl aldehyde 40 (Scheme 23). The endo,trans-configured products (±)-112–(±)-116 were obtained in low yields due to the tedious diastereoisomer separation procedures. The compounds were then tested for their binding affinities according to procedures described in Chapter 2.3. It should be noted that the compounds of this neutral series feature low water-solubility and some precipitation occurred during the binding assays, resulting in large uncertainties in the measured biological activity values. An enormous loss of binding affinity was observed when the S4 vector was omitted: N-methyl ligand (±)-112 binds with Ki = 10.3 M, showing that occupying the S4 pocket is crucial to maintain the high activity of the inhibitors (Figure 42).

61 3. The Cation– Interaction ______

O H N O R N O O N MeCN, 80 °C O N R + O N + H H S N OH O O Cl

18 40 S

Cl Reaction Maleimide R Compound Time Yield

117 (±)-112 3 h 1%

118 (±)-113 42 h 3%

119 (±)-114 19 h 2%

124 (±)-115 18 h 1%

130 (±)-116 19 h 4%

Scheme 23. Synthesis of the control compounds (±)-112–(±)-116.

The n-butyl bearing compound (±)-113 is inactive within the limit of the

biological assay (Ki > 75.4 M). Possibly, the linear butyl chain is either not sufficiently bulky to benefit from hydrophobic interactions in the S4 pocket, or does not reach deep enough into the pocket, which leads to repulsions or suboptimal positioning of the inhibitor at the active site, and ultimately to loss of binding activity. In addition, as mentioned above, precipitation during the assay may also have a large effect. Activity by a factor of 10 as compared to (±)-112 is gained by incorporating an

n-pentyl vector as in (±)-114 (Ki = 1.4 M), presumably resulting from hydrophobic interactions of the alkyl chain. Slight enhancements in affinity are observed when further methyl groups are incorporated, with the 4-methylpentyl-bearing ligand (±)-115 binding

with Ki = 0.75 M, and finally the tert-butyl derivative (±)-81 with Ki = 0.55 M.

Somewhat unexpectedly, cyclohexyl derivative (±)-116 binds very weakly with Ki =

15.1 M. The binding affinity of (±)-116, however, should be regarded rather as a trend than an absolute value: as mentioned above, the purification of the cyclohexyl-bearing

62 3. The Cation– Interaction ______

products was problematic and the biological testing was done with crude (±)-116. In addition, the high hydrophobicity of the compound could cause a large part of it to precipitate from the buffer solution used in the binding assay, thus partially explaining the weakness of (±)-116 as an inhibitor.

Compound R Ki / M

Compound R Ki / M O (±)-112 10.3 H R N (±)-108 H N 9.8 (±)-113 > 75.4 2 N O H (±)-110 N 0.911 (±)-114 1.4 H N O (±)-111 N 0.056 (±)-115 0.75 S (±)-34 N 0.009 (±)-81 0.55 Br Cl (±)-116[a] 15.1

Figure 42. Binding affinities of the cationic and neutral methylation series towards factor Xa. [a]Measured as crude product.

The binding affinity is enhanced by approximately a factor of three when moving from pentyl derivative (±)-114 to tert-butyl derivative (±)-81 in the neutral alkyl-bearing series, showing that some binding affinity can be gained merely through proper filling of the S4 pocket and through increasing hydrophobic interactions (Figure 42). However, these weak interactions are not sufficient to render the inhibitors highly active. Incorporation of ammonium ion moieties capable of undergoing strong cation– interactions results in highly active factor Xa inhibitors, leading to a gain of binding affinity by a factor of 1000 when moving from the primary to the quaternary ammonium ion inhibitor.

3.7. Further Aspects of the Cation– Interaction: Towards the Dication Ligand, Other Onium Ions, and Counteranion Effects

Although monoquaternized DABCO derivative (±)-105 was slightly less active (Ki =

50 nM) than trimethylammonium (±)-34, the additional quaternization of the second tertiary amine was envisioned to lead to further gain in binding affinity. Thus, dication-

63 3. The Cation– Interaction ______

bearing compound (±)-136 was designed. First, 1,4-diazabicyclo[2.2.2]octane (137) was methylated with MeI to give 138 in 62% yield (Scheme 24).

N N 137

MeI, EtOAc, 0 °C  RT, O O H 10 min H N 62% N Br N N N MeCN, O H N O H N 80 °C Br N N + I O N O I S S 138

Cl Cl

(±)-57 (±)-136

Scheme 24. Synthesis of dication (±)-136.

Subsequently, bromide (±)-57 was treated with tertiary amine 138 to give dication (±)-136. The conversion was successful as evaluated by LC-MS analysis; however, dicationic product (±)-136 could not be isolated. The usual procedure to isolate the ammonium ion compounds involves the precipitation of the product from MeOH through

addition of Et2O. In this case, although precipitation occurred, a mixture of the residual methylated DABCO 138 and (±)-136 was obtained. Neither (±)-136 nor 138 are soluble in organic solvents, thus preventing the use of column chromatography. In addition, reversed phase HPLC cannot be used with the tricyclic inhibitors, since partial opening of the maleimide ring has been observed when this method of purification was used. The reaction conditions were not optimized due to time restrictions, but in the future, an excess of bromide (±)-57 could be used in the reaction to entirely consume 138 and allow the isolation of (±)-136.

In order to determine the influence of nature and size of the onium ion, sulfonium and phosphonium ions as S4 moieties were envisioned. Despite many attempts, it was not possible to obtain dimethyl- or pentamethylene sulfonium-bearing counterparts of (±)-34. However, trimethylphosphonium derivatives (±)-endo,trans-139 and (±)-exo,trans-140 were prepared from the corresponding bromine bearing derivatives (±)-57 and (±)-60 by

64 3. The Cation– Interaction ______

treatment with a solution of Me3P in THF in 69% and 26% yield, respectively (Scheme 25, (±)-140 not shown). O H Compound R Ki / nM N R N + – (±)-34 Me3N Br 9 O H + – N (±)-139 Me3P Br 7

O + – (±)-145 Me3N I 18

S

Me3P in THF, Cl THF, RT, 6 d 69% + – (±)-57 R = Br (±)-139 R = Me3P Br NaI, acetone, RT, 43 h quant. (±)-146 R = I Me3N in EtOH, EtOH, RT, 3 d

+ – 90% (±)-145 R = Me3N I

Scheme 25. Synthesis and binding affinities of phosphonium derivative (±)-139 and iodide (±)-145. The exo,trans-configured phosphonium ion (±)-140 was synthesized similarly to (±)-139 using precursor (±)-60.

Phosphonium ion (±)-139 was found to bind to factor Xa with Ki = 7 nM (Scheme 25), in a similar range as (±)-34. The exo,trans-derivative (±)-140 was a much weaker inhibitor (Ki = 1.01 M), also as compared to the corresponding ammonium inhibitor

(±)-65 (Ki = 0.445 M). Both phosphonium compounds were inactive against thrombin. Thus, changing to a softer and slightly larger phosphonium does not significantly affect the binding affinity of the endo,trans-configured ligand. Similar results have been reported in the literature, where the larger tetramethylphosphonium cation (115 Å3) was bound with a similar association constant as the tetramethylammonium guest (105 Å3) in the aromatic cavity of both resorcinarene and pyrogallolarene cavitands.[131,132]

The binding of a cation to a  system, especially in low-polarity environments, is a three-component process, involving the counteranion in addition to the cation and the  system.[10] Thus, a number of investigations with synthetic receptors have addressed specifically the influence of counteranions on cation– interactions.

65 3. The Cation– Interaction ______

In polar, aprotic media ion-pairing interactions are strong, enhancing the importance of the electrostatic interactions between the anion and the cation.[10] Roelens and co-workers showed with their tetraester cyclophane receptor 141 (Figure 43) that loosely ion-paired tetramethylammonium (TMA) salts interact more strongly with the receptor.[133,134] The electrostatic contribution of the counteranion inhibits the cation– interaction, and thus ion pairs with more disperse anions, which are also less soluble, give 1 rise to stronger binding affinity, as measured by H NMR spectroscopy in CDCl3 at 296 K. A fairly good correlation was indeed found between the solubility logS of the ion pair and the standard free energy of binding –G°.[133] In addition, an excellent correlation was found between the electrostatic potential of the ion pair and the experimental free energy of binding.[134]

O O

O O O O

O O O O O O

141 142

Figure 43. Tetraester 141 and tetraether 142 receptors by Roelens and co-workers to investigate counteranion effects on the cation– interaction.[133,134]

Recently, Roelens and co-workers redesigned their cyclophanic receptor: the tetraether host 142 [135] (Figure 43) binds cations in chloroform with higher affinity than the previously synthesized tetraester.[133,134] As in the case of the tetraester, enhanced binding affinity of the salt was observed upon changing the counterion from chloride + – –1 + – –1 (TMA Cl , Ka = 165 M ) to dimethyltrichlorostannate (TMA Me2SnCl3 , Ka = 1004 M ), a softer ion, which confirms the beneficial effect of charge dispersion of the anion on cation binding.[135] The inhibitory effect of the anion towards the cation binding was found to be characteristic of the anion, independent of the host, and can be predicted by calculating the ion-pair electrostatic potential (EP): the higher the EP, the more stable the cation–cyclophane complex. However, in the case of conformationally rigid hosts

66 3. The Cation– Interaction ______

binding tight ion pairs, the results might be affected by steric hindrance in the case of large anions. Hunter et al. studied the role of the counterion in the cation– interaction with their synthetic supramolecular zipper complex 143 (Figure 44).[136] Chemical double- [137] mutant cycles in CDCl3 at 300K showed that the pyridinium– interaction remains constant at –0.6 kcal mol–1 upon changing the counterion. However, the binding constant of the ion pair depends strongly on the nature of the anion, which is due to the anion competing for some other binding site in the complex and not due to a change in the cation– interaction. Calculations have also shown that varying the counteranion has a negligible influence on the energy of the cation– interaction.[138,139]

O X– –1 HN N X G / kcal mol H O N I –0.5 ± 0.2 O PF –0.6 ± 0.1 H 6 BPh –0.6 ± 0.2 N HN 4 O

143

Figure 44. Supramolecular zipper complex 143 by Hunter et al. to study counteranion effects on the cation– interaction.[136]

Hof and co-workers studied the complexation of ammonium cations with host 144 bearing three tryptophan rings by NMR spectroscopy in phosphate buffer at pH 8.0 at 295 K (Figure 45).[140] No significant change in the association constant was observed upon varying the counteranion of the TMA guest (Cl–, I–, AcO–). However, the considerably higher affinity of larger cations such as BuN+Cl– suggested that the hydrophobic effect is more important for binding in this flexible model system than the cation– interaction. This was further corroborated by the lack of binding observed for

those larger cations in CDCl3.

67 3. The Cation– Interaction ______

O–Na+ O N O –1 Guest Kassoc / M O–Na+ + – Me4N Cl 32 ± 2 N Me N+I– 35 ± 6 N 4 + – Me4N AcO 26 ± 8

+Na–O O

144

Figure 45. Model system 144 by Hof and co-workers to study cation– interactions by [140] NMR spectroscopy in phosphate-buffered D2O at pH 8.0 at 295 K.

Although counteranion effects on cation– interactions have been investigated in a number of studies, most results are obtained with model systems in organic solvents. In contrast, data from biological environment remain scarce. To address potential counteranion effects in the binding of the quaternary ammonium ion inhibitors to factor Xa, iodide (±)-145 was prepared by a reaction of (±)-57 with NaI to give (±)-146,

followed by treatment with Me3N (Scheme 25). Counterion exchange of bromide in (±)-34 to chloride was also attempted using Amberlite IRA 68 resin. However, due to the scarcity of the material the purification of the product was very difficult. Additionally, the amount of material was not sufficient to verify if the counterion exchange had actually taken place. Therefore, only iodide (±)-145 was tested for its biological activity. The inhibitory constant of iodide (±)-145 towards factor Xa was determined to be

Ki = 18 nM, in a similar range as that of bromide (±)-34. Thus, in the S4 pocket of factor Xa, exchanging the counteranion seems to have a negligible influence on the strength of the cation– interaction, within the error of the assay in buffer solution at pH 7.8. Clearly, for a more complete evaluation, a series of compounds with different counteranions, e.g. polyatomic anions, should be targeted, which would allow the evaluation of the importance of counteranion effects on the cation– interaction in the S4 pocket of factor Xa.

68 4. Water Replacement in the S1 Pocket ______

4. Water Replacement in the S1 Pocket

4.1. Water Replacement in Drug Design In nature, water molecules are often found at protein–ligand interfaces, e.g. mediating the interactions between polar groups.[141,142] When an inhibitor binds to an enzyme, binding affinity can be gained by bridging the protein with the ligand by hydrogen bonding to a water molecule, or by releasing the water to the bulk solvent.[143] Therefore, water replacement is of enormous interest in structure-based drug design. However, identifying those water molecules, which can be displaced with concomitant gain in binding free enthalpy, remains a challenge.[141,143-145] Although replacement of water results in a gain in entropy, it can give rise to an unfavorable enthalpic term, if the interactions established by the water molecule at the active site are not restored or replaced by more favorable ones of the ligand. Both binding affinity and specificity can be enhanced by the presence of water molecules at protein–ligand interfaces.[142] A water molecule can be incorporated into the interface in many ways: bound mainly to either the protein or the ligand, or bound equally to both, either more centrally or at the periphery of the binding site. An analysis of 392 protein–ligand X-ray cocrystal structures of the PDB revealed that 72% of all ligand- bound water molecules are interfacial, i.e. bound to both the protein and the ligand, whereas 18% are surface water molecules bound only to the ligand.[146] On average, there are 4.6 ligand-bound water molecules per crystal structure. Interfacial water molecules were found to have three polar interactions on average, whereas surface water molecules have only one. Water molecules solvating active sites of proteins are often entropically unfavorable due to the constraints imposed by the protein surface, or energetically unfavorable due to the inability of the water molecule to form all possible hydrogen- bonding interactions.[147] Therefore, free energy can be gained by liberating these water molecules to the bulk water by incorporating a suitable ligand, which interacts favorably with the active site. The replacement of weakly bound water molecules probably requires a small free energy penalty, which can be compensated by the interactions of the ligand moiety filling the gap, resulting in stronger binding of the ligand.[144] Essex and co-workers calculated the binding free energies of 54 water molecules in different protein–ligand complexes, and found the energies to be dependent on the binding

69 4. Water Replacement in the S1 Pocket ______

environment.[148] Strongly bound water molecules were usually located in polar environments forming at least three hydrogen bonds, whereas loosely bound water was found in partially apolar cavities, undergoing less than three hydrogen-bonding interactions. Dunitz estimated the free enthalpy gain resulting from entropy when transferring a water molecule from the protein to the bulk water to range between 0 and 2 kcal mol–1 at room temperature.[149] Water displacement by ligand parts at enzyme active sites has been a central topic in the medicinal chemistry program of the Diederich group in the recent years.[150-152] In the binding of highly active bisubstrate inhibitors of catechol-O-methyltransferase (COMT), the displacement of a ligand-imported water molecule was suggested to lead into a gain in binding free enthalpy of at least 1.8 kcal mol–1 (Figure 46).[150] The

N-methyl group of the adenine moiety of the inhibitor (Ki = 3 nM) was shown by X-ray crystallography (PDB code: 3HVH) to bind in the energetically unfavorable s-trans conformation, and to replace a ligand-bound water molecule seen in the X-ray crystal structure of an inhibitor bearing adenine unsubstituted at the exocyclic N atom. By 1 H NMR spectroscopy in D2O/CD3SO and CF3CD2OD, the s-cis conformation was –1 estimated to be preferred by at least Gs-trans  s-cis  –1.8 kcal mol , which corresponds to the binding free enthalpy that has to be invested when the ligand is bound at the active site of COMT in the unfavorable s-trans conformation. However, the alkylated ligands bind with similar affinities to the non-alkylated one, indicating that the binding free enthalpy gained from water replacement is at least –1.8 kcal mol–1.

HN NH

N N N N

N N N N

s-cis s-trans

G –1.8 kcal mol–1 s-trans  s-cis 

Figure 46. Water replacement at the active site of COMT.[150] Binding free enthalpy by at least –1.8 kcal mol–1 is gained upon complexation of N-alkyladenine, which replaces a ligand-bound water molecule (PDB code: 3HVI, resolution 1.20 Å). Distances in Å.

Color code: gray Cenzyme, green Cinhibitor, red O, blue N, yellow S.

70 4. Water Replacement in the S1 Pocket ______

4.2. Water at the Active Site of Factor Xa In 1993, Padmanabhan et al. published an X-ray crystal structure of factor Xa without a bound ligand at 2.2 Å resolution.[53] At the active site, nine water molecules can be identified (Figure 47), four of which reside within the S4 pocket. In the center of the active site, a water molecule forms a hydrogen bond with Gly216 (d(N(H)···O) = 2.7 Å), and two water molecules are located close to the carbonyl of Gly216. One water molecule sits near the disulfide bridge formed by Cys191 and Cys220 (d(Cys220- S···(H)O) = 2.6 Å), and one is located above the ipso C atom of Tyr228 (d(O···C–O- Tyr228) = 3.2 Å, angle(O···C–O) = 92°).

Figure 47. Water molecules at the active site of factor Xa as seen in the X-ray crystal structure of the enzyme without a bound ligand (resolution 2.2 Å, PDB code: 1HCG).[53] Distances in Å. Color code: gray C, red O, blue N, yellow S.

In order to identify the water molecules that can be displaced by a ligand part with a gain in binding free enthalpy, Friesner and co-workers calculated the hydration sites and the thermodynamic properties of the water molecules at the active site of factor Xa, and compared the hydration map to the locations of the water molecules bound at the active site of factor Xa in the X-ray crystal structure of the free enzyme (PDB code: 1HCG, Figure 47).[147] They identified much more hydration sites than the water molecules seen in the crystal structure, which was attributed to the low resolution of the crystal structure (2.2 Å). The water molecules, which could be displaced with a favorable contribution to binding free enthalpy, reside in the S4 pocket (three water molecules), above the disulfide bridge formed by Cys191 and Cys220, and on top of Tyr228 in the S1 pocket.

71 4. Water Replacement in the S1 Pocket ______

In the X-ray cocrystal structure of (±)-34 bound to factor Xa, the Cl atom of the thiophene is directed towards Tyr228 in the S1 pocket (Figure 27, d(Cl···C–OH) = 3.6 Å, angle(Cl···C–OH) = 82°), and undergoes van der Waals interactions with the surrounding amino acids Ala190, Val213, and Gly226, and Ile227 (d(Cl···C/N) = 3.5–3.8 Å). Previously, in a search of the Protein Data Bank (PDB) by Relibase,[113,114] the Cl position and angle were found to be highly conserved among the X-ray cocrystal structures of factor Xa inhibitors (Chapter 2.4, Figure 21). As mentioned in Chapter 3.2 (Figure 30, Figure 31), when comparing the X-ray cocrystal structures of factor Xa in complex with phenylamidinium ligand (±)-16 and chlorothienyl ligand (±)-34, a water molecule of the phenylamidinium structure resides at a very similar position to that of the Cl atom of the chlorothienyl moiety (Figure 48, left). This water molecule undergoes dipolar interactions with Tyr228 and hydrogen bonding to Ile227 (d(C=O···(H)OH) = 2.9 Å) and to the phenylamidinium S1 needle of the inhibitor (d(N···(H)OH) = 3.1 Å, Figure 48, right).

Figure 48. Left: The chlorothienyl vector of (±)-34 in the S1 pocket of factor Xa (PDB code: 2JKH) superimposed with the water molecule of the phenylamidinium structure (PDB code: 2BOK), showing the similar positions of the water molecule and the Cl atom. Right: The water molecule in the S1 pocket of factor Xa in the X-ray crystal structure of

phenylamidinium inhibitor (±)-16. Distances in Å. Color code: gray Cenzyme, green

Cinhibitor, red O, blue N, yellow S, lime Cl.

4.3. Halogen– Interactions Halogen– interactions, such as that seen in the S1 pocket of factor Xa between (±)-34 and Tyr228, have raised increasingly interest in the recent years, and many searches for

72 4. Water Replacement in the S1 Pocket ______

such contacts in the CSD and the PDB have been reported.[94,153-158] A CSD search performed by Maignan et al. at Aventis Pharma gave 400 potential Cl–aromatic contacts, with 88° as the most common angle between the C–Cl bond and the aromatic plane, and a maximum of occurrences at d(Cl–centroid-Ar) = 4.0 Å.[94] In a PDB search, Imai et al. found 200 hits with 338 interactions for Cl– contacts.[157] After removing geometries where Cl interacts with the hydrogen atoms of the aromatic ring ((Ar-plane···Cl) > 140°), 59 Cl– interactions remained ( < 140°; Figure 49, left). Of these complexes, 21 are interactions with Phe, 15 with Tyr and Trp each, and eight with His. The average distances between the arene centroid and the Cl atom (r1) varied from 3.9 Å (Trp and Tyr) and 4.0 Å (His) to 4.3 Å (Phe). In 57 cases, the Cl atom was a substituent on an aromatic ring. Two dominant geometries were found for the Cl– interaction: face-on (Figure 49, middle) corresponds to the one where the Cl atom approaches the center of the arene; in the edge-on geometry (Figure 49, right) the Cl atom establishes contacts rather with either the arene bonds or the C atoms. The latter geometry was found to be preferred with Phe and His, whereas face-on was prevalent for the more electron-rich Trp and Tyr. Thus, the -electron density of the aromatic ring was deduced to have an influence on the interaction geometry. From the comparison of the calculated interaction energies at HF and MP2 level of theory, the major contributing force to the attraction was concluded to stem from dispersion.[157] In MP2 calculations on the chloroethyne–benzene complex, the potential energy curves showed no preference for the ’ angle in the face-on geometry (Figure 49, left), except for the unstable geometry when ’ = 0°. For the edge-on complex, angles ’ = 10°–20° resulted in destabilization of the complex. The interaction energy for the edge-on chloroethyne–benzene complex in the geometry derived from a crystal structure –1 of a ligand–factor Xa complex was calculated to be Eint = –2.01 kcal mol . Lu et al. found 189 halogen– interactions in 146 protein structures in the PDB, with the mean intermolecular distance increasing in the order Cl < Br < I.[158] The angle between the normal of the arene and Cl···centroid-Ar was found to have a maximum at around 25°.

 Cl Cl Cl 1 2 1 2  r r r r

Cl– interactions face-on edge-on  < 140° r2–r1  0.3 Å r2–r1 > 0.3 Å

Figure 49. Definitions of Cl– interactions by Imai et al.[157]

73 4. Water Replacement in the S1 Pocket ______

Calculations have shown the major contribution to the halogen–arene interaction to originate from dispersion.[159] At high level of theory, Wallnoefer et al. calculated energies of –1.3 kcal mol–1 for the complex of para-cresol with chlorobenzene and –2.3 kcal mol–1 for that with bromobenzene. They found displacements parallel to the plane of the aromatic ring to have a minor influence on the interaction energy, whereas merely small deviations from the optimal distance (3.3 Å) led to significant decreases in the interaction strength. No attractive interaction was found using calculation methods that do not account for dispersion. Since the finding that highly efficient inhibitors can be obtained by engaging in chloroarene–Tyr228 interactions in the S1 pocket of factor Xa, studies of changing the substituent on the arene interacting with Tyr228 have been reported in the literature.[77,97,118,160-163] Matter et al. at Sanofi-Aventis Deutschland GmbH observed large variations in the binding affinity of their factor Xa inhibitors 147a–f when different substituents were incorporated into the phenyl S1 vector (Figure 50).[164] A large gain in binding affinity, –G = 2.5 kcal mol–1, was observed when the unsubstituted phenyl ring in 147a (Ki = 204 nM) was replaced with para-chlorophenyl in 147b or

para-bromophenyl in 147c residues (for both, Ki = 3 nM). Fluorine-bearing inhibitor

147d was bound less strongly (Ki = 63 nM) due to the smaller van der Waals radius as compared to the larger halogens. Methylthienyl ligand 147e was bound in a similar range

(Ki = 56 nM). The inefficiency of cyano-substituted ligand 147f was attributed to repulsive effects in the S1 pocket. The high binding affinity of the chlorine- and bromine-bearing inhibitors 147b and 147c was ascribed to Cl/Br– interactions with Tyr228. In a Relibase+ search, the authors found 78 hits in 52 protein crystal structures for favorable interactions between chlorine and aromatic rings (d(Cl···centroid-Ar) = 2.8–4.2 Å), 24 of which were serine protease complexes. The majority of the complexes showed distances between 3.4 and 3.8 Å. A corresponding search in the CSD revealed 458 hits with 578 occurrences for intermolecular contacts between Cl and arenes. The data indicate some directionality for the interaction: maxima were found for interplanar angles around 0°, corresponding to parallel stacking interactions between the chloroaryl and the second aromatic ring, and 60–90°, corresponding to edge-to-face interactions between the two rings.

74 4. Water Replacement in the S1 Pocket ______

–1 Compound R Ki / nM G / kcal mol N NH a H 204 0 O N b Cl 3 –2.5 O c Br 3 –2.5 HN d F 63 –0.8

e Me 56 –0.8 R 147 f CN 2855 1.6

Figure 50. Matter et al. observed large changes in the binding affinity of factor Xa inhibitors 147a–f upon changing the phenyl substituent.[164]

4.4. Ligand Design In order to gain more insight into the role and characteristics of the thienyl-Cl–arene interaction at the active site of factor Xa, where the Cl atom replaces a structural water molecule in the S1 pocket, a series of inhibitors with systematically varied thienyl substituents was designed: unsubstituted (±)-148, hydroxyl-bearing ligand (±)-149, methoxy-bearing ligand (±)-150, fluorinated (±)-151, brominated (±)-152, iodinated (±)-153, and methyl-bearing ligand (±)-154 (Figure 51). O H (±)-148 X = H N N N (±)-149 X = OH O H (±)-150 X = OMe Br– N (±)-151 X = F O (±)-152 X = Br

S (±)-153 X = I (±)-154 X = Me X

Figure 51. The targeted inhibitors (±)-148–(±)-154 with varying thienyl substituents.

4.5. Synthesis The preparation of the target compounds (±)-148–(±)-152 and (±)-154 began from the corresponding 2,5-substituted acetylthiophenes 155–159. While unsubstituted 155, bromo- 158, and methyl-bearing acetylthiophene 159 are commercially available, methoxy- and fluoro-substituted intermediates 156 and 157 had to be prepared. 2-Acetyl-5-methoxythiophene (156) was prepared from

75 4. Water Replacement in the S1 Pocket ______

[165] 2-methoxythiophene by Friedel-Crafts acylation using AlCl3 and acetyl chloride. Cleavage of the methyl ether group would allow the preparation of the hydroxyl-bearing ligand (±)-149. 2-Acetyl-5-fluorothiophene (157) was obtained starting from 2-acetyl- 5-bromothiophene (158) following a literature procedure (Scheme 26).[166] The ketone was protected using ethylene glycol in a Dean-Stark apparatus giving ketal 160 in 52% yield. Compound 160 was unstable, and cleavage of the protecting group occurred upon storage at RT after a few days. Ketal 160 was lithiated with nBuLi and subsequently treated with N-fluorobenzenesulfonimide to afford the fluorinated ketal, which upon purification on silica underwent deprotection to give fluorothiophene 157 in 22% yield.

1. nBuLi, THF, –78 °C, 65 min ethylene glycol, 2. N-FBSI, THF, –78 °C, 30 min

pTsOH•H2O,  RT, 19 h O O toluene, reflux, 2.2 d O O 3. SiO2 Br S Br S F S 52% 22%

158 160 157

Scheme 26. Synthesis of 2-acetyl-5-fluorothiophene (157). N-FBSI = N-fluorobenzenesulfonimide.

The 2,5-substituted acetylthiophenes 155–159 were converted to the corresponding aldehydes 161–165 following the synthetic route[101] used to obtain chlorothienyl S1 needle 40. First, condensation with diethyl oxalate gave dioxobutanoates 166–170 (Scheme 27), which were then treated with hydroxylamine hydrochloride to afford the isoxazolyl compounds 171–175. Reduction of the ester

moiety using NaBH4 gave alcohols 176–180, which were subsequently treated with PCC to give aldehydes 161–165. All attempts to cleave the methyl ether in ester 172

(48% HBr or Me3SiI with MeOH) or in alcohol 177 (BBr3 or AlCl3) failed. Thus, hydroxylthienyl ligand (±)-149 was discarded as target molecule. Methylthienyl aldehyde 165 was synthesized by Moritz Hunkeler during Organic Chemistry Laboratory Course 2.

76 4. Water Replacement in the S1 Pocket ______

KOtBu, diethyl . oxalate, toluene, NH2OH HCl, 0 °C  RT, EtOH, reflux, O O N O O O 6 h – 7 d S 3–21 h S R O R S R O O 155 R = H 166 R = H 171 R = H, 42% over 2 steps 156 R = OMe 167 R = OMe 172 R = OMe, 40% over 2 steps 157 R = F 168 R = F, 61% 173 R = F, 50% 158 R = Br 169 R = Br, 92% 174 R = Br, 90% 159 R = Me 170 R = Me, 70% 175 R = Me, 72%

NaBH4, EtOH, 0 °C  RT, 1–7 d

PCC, CH2Cl2, O N N RT, 19 h – 6.5 d O S R R S O OH

161 R = H, 78% 176 R = H, 48% 162 R = OMe, 66% 177 R = OMe, 63% 163 R = F, 67% 178 R = F, 56% 164 R = Br, 98% 179 R = Br, 43% 165 R = Me, 79% 180 R = Me, 22%

Scheme 27. Synthesis of the S1 needles 161–165.

Iodothienyl aldehyde 181 was prepared from chlorothienyl alcohol 44 in four steps (Scheme 28). The alcohol was protected with a silyl protecting group to give 182; lithiation with tBuLi and subsequent treatment with iodine gave 183. Deprotection with aq. HCl afforded crude alcohol 184, which was then oxidized to aldehyde 181. ON OX Cl S

tBuMe2SiCl, 44, X = H imidazole, DMF, 1. tBuLi, THF, –78 °C, 1.5 h RT, 4.5 h ON 2. I2, –78 °C  RT, 1 h OTBDMS 93% 182, X = TBDMS I S quant. 183

1 M HCl solution, THF, RT, 41 h quant.

PCC, CH2Cl2, ON ON O RT, 16 h OH I S I S 56% 181 184

Scheme 28. Synthesis of iodothienyl aldehyde 181.

77 4. Water Replacement in the S1 Pocket ______

Aldehydes 161–165 and 181, L-proline (18), and maleimide 36 were reacted in a 1,3-dipolar cycloaddition to give (±)-185–(±)-189, respectively (Scheme 29). Unfortunately, methylthienyl-bearing (±)-190 could not be isolated. Separation of the diastereoisomers to obtain methoxythienyl ligand (±)-186 and iodothienyl ligand (±)-189 was extremely difficult. After numerous attempts to purify the products by column chromatography, the compounds were sent to F. Hoffmann-La Roche, Basel, Switzerland, where they were purified by chiral column chromatography in the laboratory of David Wechsler and Daniel Zimmerli. Instead of a racemic mixture, slightly impure enantiomers of methoxy-bearing ligands (+)-186 and (–)-186 were isolated. For the bromothienyl compound (±)-188, the corresponding exo,trans-configured product (±)-191 was also isolated in 9% yield, and converted into the quaternary ammonium ion (±)-192

by treatment with Me3N in 30% yield (not shown). In a similar manner, compounds (±)-185–(±)-189 were converted into the cationic products (±)-148, (+)- and (–)-150, and (±)-151–(±)-153, respectively. O H N N O R N O Br MeCN, 80 °C, O O H 7–30 h N + O N + N H S O OH O X S 36 18 161 X = H 162 X = OMe X

163 X = F R = Br (±)-185 X = H, 3% 164 X = Br (+)-186 X = OMe, 3% 181 X = I (–)-186 X = OMe, 2% 165 X = Me (±)-187 X = F, 5% (±)-188 X = Br, 1% Me3N in EtOH, (±)-189 X = I, 3% EtOH, RT, (±)-190 X = Me, – 2.5–9 d + – R = Me3N Br (±)-148 X = H, 72% (+)-150 X = OMe, quant. (–)-150 X = OMe, 89% (±)-151 X = F, 73% (±)-152 X = Br, 42% (±)-153 X = I, 51%

Scheme 29. Synthesis of ligands (±)-148, (+)- and (–)-150, and (±)-151–(±)-153.

78 4. Water Replacement in the S1 Pocket ______

4.6. Biological Activities Biological activities of (±)-148, (+)- and (–)-150, and (±)-151–(±)-153 were measured at F. Hoffmann-La Roche following the procedures described in Chapter 2.3. The binding affinities of the inhibitors show that the Cl substituent contributes considerably to the biological activity (Figure 52): When the substituent on the thiophene is omitted as in

(±)-148 (Ki = 612 nM), the binding affinity decreases by a factor of 70, corresponding to a loss of binding free enthalpy of G = 2.6 kcal mol–1. Evidently, preserving the interactions with Tyr228 is essential to obtain highly active inhibitors, in agreement with the results reported by Matter et al.[164] Modeling suggests that also ligand (±)-148 induces replacement of the structural water molecule above Tyr228. However, the unsubstituted thiophene is unable to form favorable interactions with Tyr228, thus resulting in the observed loss in binding affinity.

factor Xa: thrombin: Compound R Ki / nM Ki / M O (±)-148 H 612 n.d. H N [a] (+)-150 OMe 509 > 35.1 N N O H (–)-150[a] OMe 3540 > 35.1 Br N (±)-151 F 48 > 35.1 O

(±)-34 Cl 9 > 35.1 S (±)-152 Br 6 n.d. R (±)-153 I 47 > 35.1

Figure 52. Binding affinities of inhibitors (±)-34, (±)-148, (+)- and (–)-150, and (±)-151– (±)-153 bearing differently substituted thienyl vectors. [a]Binding affinity was measured for slightly impure compound.

The methoxy-substituted ligands (+)-150 and (–)-150 are both weaker factor Xa

inhibitors (Ki = 509 and 3540 nM, respectively) than the corresponding chlorinated ones

(Ki = 5 nM for (+)-34 and 271 nM for (–)-34). The tightness of the S1 pocket presumably prevents the methoxy group from orienting appropriately to form favorable interactions with Tyr228. This could potentially lead to the inhibitor shifting upwards in the S1 pocket, resulting in a loss of binding affinity. Interestingly, Nazaré et al. showed by

X-ray crystallography that an inhibitor (Ki = 89 nM) bearing a methoxyphenyl S1 vector binds to factor Xa similarly to inhibitors bearing a phenylamidinium S1 vector: oriented

79 4. Water Replacement in the S1 Pocket ______

towards Asp189 and undergoing hydrogen-bonding interactions with the conserved water molecule residing above Tyr228.[101] Thus, the possibility of methoxy-substituted ligands (+)-150 and (–)-150 interacting through a different binding mode from that of (+)-34 and (–)-34 cannot be excluded.

In the halogen series, fluorinated ligand (±)-151 (Ki = 48 nM) binds less strongly

than chlorinated (±)-34. Brominated (±)-152 (Ki = 6 nM) binds in a similar range as

(±)-34. The exo,trans-configured bromothienyl ligand (±)-192 binds with Ki = 0.50 M, slightly more strongly than its chlorinated counterpart (±)-65. Increasing the size and polarizability of the substituent further by incorporating iodine in (±)-153 leads to a loss of activity (Ki = 47 nM). The trend of the binding affinities of the halogen series and the analysis of the binding geometry between the chlorothienyl moiety and Tyr228 in the X-ray cocrystal structure of (±)-34 (PDB code: 2JKH) suggests favorable dipolar interactions of the thienyl Cl atom with the phenolic C–O(H).[167] This is additionally supported by the Relibase search for the C–Cl···C–O(H) contacts between inhibitor and Tyr228 in factor Xa (Chapter 2.4, Figure 21), which showed the position and the approach angle of the Cl atom to be highly conserved. Moreover, the structural water molecule in the phenylamidinium crystal structure (Figure 48, PDB code: 2BOK), which is replaced by the Cl atom in the cocrystal stucture of (±)-34, undergoes dipolar interactions with the C–O(H) dipole of the phenol ring of Tyr228 (d(O···C–O(H)) = 3.4 Å, angle(O···C–O(H)) = 98°), in addition to hydrogen bonding to Ile227 (d(C=O···(H)OH) = 2.9 Å) and to the phenylamidinium S1 needle of the inhibitor (d(N···(H)OH) = 3.1 Å). Importantly, a water molecule is located at a similar position also in the X-ray crystal structure of factor Xa without a bound ligand (PDB code: 1HCG),[53] supporting the suggested water replacement. The Cl substituent in (±)-34 displaces this water and restores the dipolar interactions with the phenolic C–O(H) group, while additionally undergoing favorable van der Waals interactions with the surrounding lipophilic amino acid residues, thus resulting in high affinity. Although stronger dipolar interactions are expected for the C–F bond with the C–O(H) moiety of Tyr228, the observed loss of binding affinity for (±)-151 is presumably due to the unfavorable proximity of the F atom to the  surface of Tyr228[168] and the weakened van der Waals interactions with the surrounding amino acids,[169] which overcompensate the favorable dipolar interactions.[167] Matter et al. suggest the weakened binding of fluorinated compounds to be due to the lower

80 4. Water Replacement in the S1 Pocket ______

polarizability and smaller size of fluorine as compared to chlorine, resulting in an unfavorable interaction with the aromatic plane.[164] In the case of iodine-bearing ligand (±)-153, the observed loss in binding affinity could be caused by the weakened dipolar interactions with Tyr228 due to the softer character of iodine as compared to chlorine, and the inhibitor shifting outwards from the S1 pocket due to steric reasons. Although calculations show dispersion to be the main contributing force in the Cl– interaction,[159] and many examples of halogen– interactions have been found in the CSD and PDB,[94,153-158] the geometry found in the (±)-34–factor Xa complex seems to point to a role of favorable dipolar interactions in the binding of this inhibitor system. Obviously, the trend seen for the inhibitor binding affinity is a result of the interplay of different contributions, such as van der Waals contacts, dipolar interactions, and geometrical preferences in the narrow S1 pocket, and thus does not reflect that expected for a pure dipolar interaction. These results underline the importance of the choice of the substituent to replace water and interact with Tyr228. In the S1 pocket of factor Xa, the best substituents are chlorine and bromine, which can be attributed to optimal fit in the S1 pocket and polarizability. In the future, investigating the effect of incorporating substituents on the thiophene, such as hydroxyl or amino groups, which resemble the replaced water molecule more closely, would be of interest.

81 5. Stacking on Polar Peptide Surfaces ______

5. Stacking on Polar Peptide Surfaces

5.1. Introduction and Inhibitor Design The introduction of heteroatoms into aromatic rings has a significant influence on – stacking interactions.[10] Using their macrocyclic receptor (Figure 53), Hamilton and co-workers showed the importance of electrostatic complementarity between partial [170,171] –1 charges. Receptor 193a binds 1-butylthymine (194) with Ka = 570 M in CDCl3 at 298 K as determined by 1H NMR spectroscopy. When the ester substituents were omitted –1 –1 (193b), binding affinity decreased to 290 M , and even further to 140 M with 193c, when the substituents were replaced by butoxy groups. The high association constant in the case of diester receptor 193a was attributed to the favorable alignment of oppositely partially polarized atoms: five complementary contacts are established between the receptor and the guest. The binding geometry was confirmed by X-ray diffraction analysis, where the naphthyl moiety undergoes face-to-face stacking interactions with the thymidine ring at an interplanar distance of 3.5 Å. Interestingly, for the complex of butoxy-substituted receptor 193c with 194, the naphthalene moiety was seen to undergo edge-to-face interactions with the pyridine–thymine plane, as shown by 1H NMR spectroscopy and X-ray crystallography, demonstrating that electrostatic complementarity can lead to face-to-face stacking, whereas in the absence of such contacts edge-to-face interactions are preferred.

O R R R = O N O O a O N O H H H H N N N 194 b O O O 193 c

Figure 53. Macrocyclic receptors 193, which bind 1-butylthymidine (194) through combination of stacking on the naphthyl moiety and hydrogen bonding to the pyridyl moiety. [170,171]

In addition to arene–arene pairs, stacking can also occur between planar hydrogen- bonding arrays, such as those between the side chains of Arg and Glu or Asp, and aromatic rings.[10] Although these polar residues require solvation within the plane, the

82 5. Stacking on Polar Peptide Surfaces ______

arrays are rather apolar orthogonal to the plane, where the highly polarizable bonds enable interactions with aromatic rings. As an example, this was observed in the Diederich group in the X-ray crystal structure of a 2:1 complex of a Rebek-imide receptor and 9-ethyladenine (Figure 54).[172] In the complex, one adenine molecule undergoes stacking interactions with the hydrogen-bonding array of another adenine molecule and the receptor.

Figure 54. The X-ray crystal structure of Rebek-imide receptor and 9-ethyladenine (CCDC-167545).[172] Distances in Å. Some hydrogen atoms omitted for clarity. Color code: gray C, red O, blue N, white H.

Planar peptide backbone lines the walls of the narrow S1 pocket, Ala190-Cys191- Gln192 on one side, and Trp215-Gly216 on the other. In the X-ray cocrystal structure of (±)-34 with factor Xa (PDB code: 2JKH), these polarizable walls were observed to undergo stacking interactions with the isoxazolyl-chlorothienyl needle (Figure 55), with shortest heavy atom distances between 3.3 Å and 3.7 Å.

83 5. Stacking on Polar Peptide Surfaces ______

Figure 55. The biaryl needle of (±)-34 undergoes efficient stacking interactions with the peptide walls lining the S1 pocket (PDB code: 2JKH). Distances in Å. Color code: gray

Cenzyme, green Cinhibitor, red O, blue N, yellow S, lime Cl.

To investigate whether binding affinity is influenced by the position of the heteroatoms on the isoxazole ring of (±)-34, the series (±)-195–(±)-199 of inhibitors was designed by systematically varying the positions of the heteroatoms (Figure 56). Modeling with Moloc predicted the compounds to bind to factor Xa in a similar fashion to (±)-34.

O N N O Ar = Cl S Cl S

(±)-34 (±)-195

Ki = 9 nM O H N N N Cl S Cl S O N N O O H Br Ar (±)-196 (±)-197

O O Cl S Cl S N N

(±)-198 (±)-199

Figure 56. The inhibitors (±)-195–(±)-199 designed to investigate stacking interactions in the S1 pocket.

84 5. Stacking on Polar Peptide Surfaces ______

5.2. Synthesis Parts of the synthesis were performed during Organic Chemistry Laboratory Course 2 of Moritz Hunkeler and Matthias Knecht and student semester projects of Philip Kaib and Mareike Holland. The synthesis of aldehyde 200, required to obtain (±)-195, began from commercially available 5-chloro-2-thiophenecarboxaldehyde (201) (Scheme 30). Reaction with hydroxylamine hydrochloride gave oxime 202 in 76% yield as a mixture of cis- and trans-isomers. The mixture was further converted into the corresponding imidoyl chloride 203. Subsequently, the formation of the isoxazolyl aldehyde was attempted via a reaction of 203 with bromoalkene 204, which was generated in situ by bromination of acrolein (205).[173] However, no formation of aldehyde 200 was observed. Therefore, an alternative method involving the formation of the isoxazole ring in a Cu(I)-catalyzed reaction[174] of imidoyl chloride 203 and propargyl alcohol to give alcohol 206 was employed. No reaction occurred under these conditions, and the synthesis of aldehyde 200 was not pursued further.

NH2OH•OH, NCS, pyridine, pyridine, EtOH, HO THF, CHCl3, HO O RT, 16 h N RT, 3 d N S S Cl Cl S Cl H 76% H 7% Cl 201 202 203

propargyl alcohol, Br2, CH2Cl2, Et3N CuSO , KHCO , O RT O 4 3 sodium ascorbate, Br 205 204 tBuOH, RT

NO NO S Cl Cl S O OH 200 206

Scheme 30. Attempts to synthesize isoxazole-bearing S1 needle 200.

In the first step towards aldehyde 207, required for the synthesis of (±)-196, -brominated acetylthiophene 46 was treated with sodium diformylamide[175] to afford imide 208, which was subsequently deformylated to ammonium salt 209 in excellent yield by treatment with ethanolic HCl (Scheme 31). Compound 209 was then reacted with ethyl chlorooxoacetate to afford the corresponding amide 210.[176] The oxazole ring [177,178] of ester 211 was formed in good yield via a POCl3-triggered cyclization, and the

85 5. Stacking on Polar Peptide Surfaces ______product was converted into alcohol 212, and subsequently into the corresponding aldehyde 207, following a reduction and oxidation sequence using NaBH4 and PCC.

NaN(CHO) , 2 5% HCl in EtOH, O MeCN, RT, 7 h O O S S O Cl Cl RT, 2.5 d Cl S Br N – 52% O NH3Cl 96% 46 208 209

O Et3N, CH2Cl2, O –5 °C, 0.5 h Cl O  RT, 22 h 72%

NaBH4, EtOH, N POCl3, S O N 0 °C  RT, 44 h Cl O reflux, 17 h S S Cl H O Cl OH O N O O 95% 80% O O 212 211 210

PCC, CH2Cl2, RT, 20 h 30%

N Cl S O O

207

Scheme 31. Synthesis of aldehyde 207.

The synthesis of S1 needle 213, required for the preparation of ligand (±)-197, began by converting 5-chlorothiophene-2-carboxylic acid (214) into acid chloride 215 using thionyl chloride (Scheme 32). Compound 215 was reacted further with propargylamine to give amide 216 in 58% yield. The key step of this synthesis, an intramolecular Pd-catalyzed cyclization and oxidation sequence,[179] failed when

1,4-benzoquinone was employed as oxidizing agent. Exchanging this reagent for CuCl2 gave 213 in modest 7% yield. Due to the extremely low yield of the cyclization– oxidation sequence, it did not seem feasible to produce aldehyde 213 through this synthetic route in satisfactory amounts required for the cycloaddition reaction to ultimately afford (±)-197. As an alternative approach, benzonitriles have been reported to undergo a copper-catalyzed reaction with diazopyruvate to give ethyl esters of the desired oxazole.[180]

86 5. Stacking on Polar Peptide Surfaces ______

SOCl2, DMF, propargylamine,

toluene, 80 °C, Et3N, CH2Cl2, O 45 min O RT, 2 h O S S S Cl Cl Cl OH 72% Cl 58% N H 214 215 216

CuCl2, 1,4-benzoquinone, [PdCl2(MeCN)2], [PdCl2(MeCN)2], DMF, 100 °C, 21 h DMF, THF 7%

N S Cl O O 213

Scheme 32. Synthesis of the S1 needle 213.

The reaction sequence towards aldehyde 217 started with the conversion of acid

chloride 215 into amide 218 using 7 N NH3 in MeOH (Scheme 33). Compound 218 was then heated in 1,3-dichloroacetone[181] to form oxazolyl chloride 219 in 63% yield. Only solvent-free conditions provided the desired product. Since direct oxidation only gave aldehyde 217 in poor yield (PCC, Kornblum oxidation,[182] Sommelet method[183]), chloride 219 was hydrolyzed to 220 and subsequently oxidized with PCC to give aldehyde 217 in 54% yield. Due to time constrictions, preparation of the aldehyde required for the synthesis of (±)-199 was not pursued.

7 N NH in MeOH, dichloroacetone, O 3 O O Cl S toluene, RT, 1.5 h Cl S 130 °C, 18 h Cl S Cl Cl NH2 N 92% 63% 215 218 219

1. NaOAc, DMF, 80 °C, 15 h 63% 2. K2CO3, H2O, MeOH, RT, 1 h

PCC, CH2Cl2, O O O RT, 17 h Cl S Cl S OH N N 54% 217 220

Scheme 33. Synthesis of aldehyde 217.

87 5. Stacking on Polar Peptide Surfaces ______

Aldehydes 207 and 217 were reacted with maleimide 36 and L-proline (18), and tricyclic oxazole-bearing compounds (±)-221 and (±)-222 were isolated in 6% and 7% yield, respectively (Scheme 34). Quaternary ammonium ions (±)-196 and (±)-198 were obtained via treatment with Me3N in EtOH. O H N O R N O Br Het MeCN, 80 °C, O 1–3 d H N ++ O N Het H S O OH Cl S N 36 18 207 Het = O Cl

R = Br 217 Het = N N O O N Het = O

(±)-221, 6% (±)-222, 7%

Me3N in EtOH, EtOH, 5–6 d

+ – R = Me3N Br N Het = O N O

(±)-196, 74% (±)-198, quant.

Scheme 34. Synthesis of oxazole-bearing inhibitors (±)-196 and (±)-198.

5.3. Biological Activities Oxazoles (±)-196 and (±)-198 were tested for their biological activity according to previously described protocols (Chapter 2.3). Unexpectedly, they were found to be much

weaker factor Xa inhibitors than (±)-34 (Ki = 9 nM), (±)-196 binding with Ki = 146 nM,

and (±)-198 with Ki = 1620 nM (Figure 57). Such large differences were rather unexpected, since modeling had predicted both oxazole compounds to bind to factor Xa in a similar manner as (±)-34, without any apparent repulsive contacts between the ligand and the active site. Both oxazoles showed excellent selectivity for factor Xa.

88 5. Stacking on Polar Peptide Surfaces ______

Ki / nM Ki / M Compound Het factor Xa thrombin

(±)-34 9 > 35.1 O N H O N N N O H Br (±)-196 146 25.8 Het N O

S

Cl (±)-198 1620 > 35.1 N O

Figure 57. Biological activities of inhibitors (±)-34, (±)-196, and (±)-198 towards factor Xa and thrombin.

5.4. X-Ray Cocrystal Structures and Conformational Analysis To rationalize the considerable binding affinity loss upon exchanging the isoxazole ring in (±)-34 to oxazole rings, X-ray cocrystal structures of both (±)-196 and (±)-198 with factor Xa were solved by Dr. David W. Banner at F. Hoffmann-La Roche (for (±)-196: Figure 58, PDB code: 2Y5G, resolution 1.33 Å; for (±)-198: Figure 59, PDB code: 2Y5H, 1.29 Å resolution). For both compounds, only the 3aS,4R,8aS,8bR-configured enantiomer was bound at the active site. In both structures, the key interactions are maintained: the quaternary ammonium ion is located in the center of the S4 pocket and the chlorothiophene undergoes interactions with Tyr228, analogously to (±)-34 (Figure 55).

89 5. Stacking on Polar Peptide Surfaces ______

Figure 58. The X-ray cocrystal structure of (±)-196 bound at the active site of factor Xa

(PDB code: 2Y5G; resolution 1.33 Å). Distances in Å. Color code: gray Cenzyme,

turquoise Cinhibitor, red O, blue N, yellow S, lime Cl.

Figure 59. The X-ray cocrystal structure of (±)-198 bound at the active site of factor Xa

(PDB code: 2Y5H; resolution 1.29 Å). Distances in Å. Color code: gray Cenzyme, magenta

Cinhibitor, red O, blue N, yellow S, lime Cl.

In addition, the active site is completely conserved in all three structures (Figure 60).

90 5. Stacking on Polar Peptide Surfaces ______

Figure 60. Superimposition of the active sites of factor Xa as seen in the X-ray cocrystal structures of (±)-34 (PDB code: 2JKH; green), (±)-196 (2Y5G; purple), and (±)-198 (2Y5H; pink).

However, the overall binding conformation of the inhibitors is different for (±)-196 and (±)-198 as compared to (±)-34: the oxazole ring has flipped relative to the isoxazole ring of (±)-34, thus inducing the tricyclic core to shift to the back of the active site (Figure 61). The new location of the central core resembles more closely that of the phenylamidinium inhibitor (±)-16 (PDB code: 2BOK).[82] Even though repositioning of the tricycle is not expected to induce major changes in the binding affinity due to the sizable central area of the active site where the tricycle is located, minor effects could occur.

Figure 61. Two superimpositions of the crystal structures of (±)-34, (±)-196, and (±)-198 showing drastic differences in the position of the tricyclic core. Color code: gray

Cenzyme_2JKH, green Cinhibitor_34, turquoise Cinhibitor_196, magenta Cinhibitor_198, red O, blue N, yellow S, lime Cl.

91 5. Stacking on Polar Peptide Surfaces ______

The strongly reduced affinity and the different conformation of bound (±)-196

(Ki = 146 nM) and (±)-198 (Ki = 1620 nM) as compared to (±)-34 (Ki = 9 nM) must originate from the interplay of several factors, which taken alone cannot fully explain the experimental findings. In the crystal structure of (±)-198, a potentially repulsive contact between the O atom of the oxazole ring and the carbonyl O atom of Gly218 is established,

which could account for some of the activity loss (d(Ooxazole···O=C–Gly218) = 3.5 Å, Figure 59). In addition, a repulsive contact is observed between the O atom and the

S atom of Cys220 (d(Ooxazole···S–Cys220) = 3.7 Å). For (±)-196, a possibly favorable

close contact exists between the oxazole C–H and Gly218 (d(C(H)oxazole···O=C–Gly218) = 3.5 Å, Figure 58). Thus, a part of the loss of binding affinity for (±)-198 as compared to (±)-34 could originate from the observed repulsive contacts. To gain more insight into the reasons for the change in binding conformation for the oxazole-bearing compounds (±)-196 and (±)-198, conformational analysis of the endo,trans-configured mimics of 34, 196, and 198 was performed by Pablo Rivera Fuentes. The S4 vector was omitted to enable shorter calculation times. Relevant conformers of 34, 196, and 198 were found by a mixed Monte Carlo Multiple Minimum/Low-Mode Conformational Search method, using the OPLS95 force field as implemented in MacroModel 9.7,[184] and keeping all conformers with energies no higher than 3 kcal mol–1 relative to the global minimum. All conformers found were further optimized using density functional theory (DFT) at the B3LYP/6-311G (d,p) level of theory using Gaussian 09.[185] Harmonic analysis at the same level of theory confirmed that all structures are minima of the potential energy surface. All reported energies are Gibbs free enthalpies (G°), at 1 bar and 298 K, in the gas phase. Three conformers were found for the structure mimicking (±)-34 showing a clear preference for the isoxazole heteroatoms to reside opposite to the lone pair of the pyrrolidinyl N atom (Figure 62). In the lowest-energy conformer, the heteroatoms of the thiophene ring and the isoxazole ring are on opposite sides of the biaryl needle to avoid repulsions of the lone pairs of the isoxazole ring and the pyrrolidinyl N atom of the tricyclic core. The second lowest energy conformer, 1.28 kcal mol–1 higher in energy, resembles closely the one seen at the active site of factor Xa in the X-ray crystal structure (PDB code: 2JKH): the heteroatoms of the aryl rings are on the same side, but opposite to the lone pair of the pyrrolidinyl N atom of the tricyclic core. Turning the biaryl needle

92 5. Stacking on Polar Peptide Surfaces ______

180° results in a conformer, which is 2.85 kcal mol–1 higher in energy than the lowest energy conformer.

Figure 62. Conformational analysis of the structure mimicking (+)-34. All reported energies are Gibbs free enthalpies (G°) at 1 bar and 298 K in the gas phase.

Four relevant conformers were found for the compound mimicking (±)-196 (Figure 63). The lowest energy conformation, also predicted by Moloc to bind at the active site of factor Xa, features the thiophene S atom on the same side as the oxazolyl C–H moiety and the N atom. At almost the same energy level is the conformer with both the S and the O atom on the same side as the pyrrolidinyl N atom. The third conformer, seen in the cocrystal structure of (±)-196 with factor Xa (PDB code: 2Y5H), is higher in energy by 1.06 kcal mol–1. This is probably due to repulsive contacts between the oxazolyl N atom and the lone pair of the pyrrolidinyl N atom. The fourth conformer is 1.81 kcal mol–1 higher in energy than the lowest one, featuring the thiophene S atom and the oxazolyl N atom on the same side as the pyrrolidinyl N atom.

Figure 63. Conformational analysis of the structure mimicking (±)-196. All reported energies are Gibbs free enthalpies (G°) at 1 bar and 298 K in the gas phase.

93 5. Stacking on Polar Peptide Surfaces ______

For the mimic of (±)-198, the lowest energy was found for the conformation seen in the X-ray crystal structure (PDB code: 2Y5H), with the thiophene S atom and the oxazolyl N atom on the same side, directed away from the pyrrolidinyl N atom (Figure 64). This can be rationalized through minimization of repulsive contacts between the oxazolyl and pyrrolidinyl N atoms. The conformer with the thienyl ring rotated by 180° is almost at the same energy level. However, an increase in energy is seen when the oxazolyl and pyrrolidinyl N atoms are on the same side: this conformer is 2.05 kcal mol–1 higher in energy than the lowest energy conformer, which can be attributed to repulsion between the heteroatoms. Interestingly, modeling with Moloc did not show any repulsive intramolecular contacts for this conformer.

Figure 64. Conformational analysis of the structure mimicking (±)-198. All reported energies are Gibbs free enthalpies (G°) at 1 bar and 298 K in the gas phase.

Therefore, only (±)-198, the weakest inhibitor of the series, shows the lowest energy conformation bound at the active site of factor Xa in the X-ray crystal structure. Thus, according to calculations, around 1 kcal mol–1 needs to be invested upon binding of inhibitors (±)-34 and (±)-196 in the less than optimal conformations seen in the crystal structures. A CSD search showed that there is a large variation in the angles of the two bonds departing from the isoxazole as in (±)-34 and from the oxazole rings as in (±)-196 and (±)-198 (Figure 65): the angle in 3,5-disubstituted isoxazoles was found to vary from 153° to 162°, with a predominance around 157° (Figure 65, top left). In contrast, the angles in the two oxazoles are much more narrow. The angle in 2,4-disubstituted oxazoles as in (±)-196 ranges from 127° to 137°, with a maximum of occurrences around 130° (Figure 65, top right), whereas in 2,5-disubstituted oxazoles as in (±)-198, it varies between 132°

94 5. Stacking on Polar Peptide Surfaces ______

and 152°, with a preference for 145–150° (Figure 65, bottom). In the cocrystal structures, the angle for the isoxazole ring in (±)-34 is 149°, 143° for oxazole in (±)-196, and 147° for oxazole in (±)-198.

N N H O O  H 

H N O 

Figure 65. A CSD search for the angle between the two vectors of the isoxazole ring and the two oxazole rings. The search (April 2011) was conducted with ConQuest 1.13 in CSD 5.32 (November 2010).

Although exchanging the isoxazole ring to an oxazole affects the angle in which the chlorothienyl moiety approaches Tyr228 in the S1 pocket, the CSD search shows that there is large variation in the angle between the vectors of the (is)oxazole rings, indicating that the energy cost for adjusting this angle is not large. In order to properly establish the key binding interactions in the aromatic box of the S4 pocket and at the bottom of the S1 pocket, the oxazole ring must flip causing the repositioning of the tricyclic core. However, according to calculations the differences between the lowest energy conformer and the one seen in the crystal structure are 1.3 and 1.1 kcal mol–1 for (±)-34 and (±)-196, respectively, and thus, does not explain the loss in binding affinity of (±)-196 and (±)-198 as compared to (±)-34.

95 5. Stacking on Polar Peptide Surfaces ______

When comparing the three cocrystal structures, the chlorothienyl moiety of (±)-196 and (±)-198 is at a similar distance and angle to Tyr228 as in the X-ray cocrystal structure of (±)-34 (Figure 66). The distance between the Cl atom of (±)-198 and Tyr228 is slightly longer, possibly accounting for some of the loss of binding affinity, as has been shown by calculations,[159] but it is clearly not the sole reason.

(±)-34 (±)-196 (±)-198 S S Ki / nM 91461620

Cl a / Å 4.2 4.0 4.4 Cl

b / Å 3.6 3.6 3.9 a b c / Å 3.8 3.7 4.0  c OH  66° 61° 63° OH

Figure 66. Comparison of the position of the Cl atom relative to Tyr228 and the angle between the C–Cl bond and Tyr228 plane in the X-ray cocrystal structures of (±)-34, (±)-196 and (±)-198.

The van der Waals contacts between the Cl atom and the surrounding amino acids are similar in all three X-ray crystal structures. Additionally, the distances from the thiophene ring to Gly216-Trp215 are in the same range (Figure 55, Figure 68, Figure 69). However, the oxazole rings are twisted by approximately 30° as compared to the isoxazole ring of (±)-34 (Figure 67).

Figure 67. The oxazole rings in the cocrystals of (±)-196 and (±)-198 are twisted by approximately 30° in comparison to the isoxazole ring in (±)-34. Color code: gray

Cenzyme, green Cinhibitor_34, turquoise Cinhibitor_196, magenta Cinhibitor_198, red O, blue N, yellow S, lime Cl.

96 5. Stacking on Polar Peptide Surfaces ______

The isoxazole ring of (±)-34 (Figure 55) establishes closer contacts to Gln192 than (±)-196 (Figure 68) or (±)-198 (Figure 69), with closest heavy atom distances of d((is)oxazole–Gln192) = 3.3, 3.8, and 3.7 Å, respectively.

Figure 68. The stacking of the biaryl needle of (±)-196 bound at the active site of factor

Xa (PDB code: 2Y5G; resolution 1.33 Å). Distances in Å. Color code: gray Cenzyme,

turquoise Cinhibitor, red O, blue N, yellow S, lime Cl.

Figure 69. The stacking of the biaryl needle of (±)-198 bound at the active site of factor

Xa (PDB code: 2Y5H; resolution 1.29 Å). Distances in Å. Color code: gray Cenzyme,

magenta Cinhibitor, red O, blue N, yellow S, lime Cl.

Suboptimal matching of the polar contacts between the oxazole ring and the peptide backbone may contribute considerably to the activity loss. The positions of

97 5. Stacking on Polar Peptide Surfaces ______

+ and – polarized atoms are shifted when replacing the isoxazole with either of the oxazole rings (Figure 70), which affects the stacking interaction between the heterocycle and the peptide surface. Whereas complementary contacts are established for (±)-34, this is not the case for (±)-196 or (±)-198, leading them to be weaker factor Xa inhibitors than (±)-34. However, a systematic study, where heterocycles with differently polarized frameworks are incorporated into the tricyclic core, would be required to validate this hypothesis.

Figure 70. Shifting of the + and – polarized atoms when changing from isoxazole in

(±)-34 to the different oxazoles in (±)-196 and (±)-198. Color code: gray Cenzyme, green

Cinhibitor_34, turquoise Cinhibitor_196, magenta Cinhibitor_198, red O, blue N, yellow S, lime Cl.

In summary, the reason for the loss of binding affinity for (±)-196 and (±)-198 as compared to (±)-34 seems to be a combination of various smaller contributions. Some of these include:

In the X-ray crystal structures, the tricyclic core in (±)-196 and (±)-198 is repositioned to the back of the active site as compared to that of (±)-34. Although major changes in the binding affinity are not expected due to the sizable central area of the active site where the tricycle is located, minor effects could occur.

The distance of the Cl atom of (±)-198 to the C–O(H) ipso-C atom of Tyr228 is slightly too large to fully benefit from the dipolar interactions.

98 5. Stacking on Polar Peptide Surfaces ______

In the structure of (±)-198, a repulsive electrostatic contact between the O atom of the oxazole ring and the carbonyl O atom of Gly218 is established, (d(O···O=C– Gly218) = 3.5 Å, Figure 59). In addition, a repulsive contact exists between the O atom and the S atom of Cys220 (d(O···S–Cys220) = 3.7 Å). For (±)-196, a potentially favorable close contact is observed between the oxazole C–H and Gly218 (d(C(H)···O=C–Gly218) = 3.5 Å, Figure 58).

The oxazole rings in (±)-196 and (±)-198 are twisted by approximately 30° as compared to the isoxazole ring of (±)-34 (Figure 67), leading to weakening of the interactions with the walls of the S1 pocket. Thus, the isoxazole ring of (±)-34 (Figure 55) establishes more attractive contacts with Gln192 than (±)-196 (Figure 68) or (±)-198 (Figure 69), with closest contacts between heavy atoms of d((is)oxazole···Gln192) = 3.3, 3.8, and 3.7 Å, respectively.

The position of positively (+) and negatively (–) polarized atoms in the interacting surfaces is shifted when replacing the isoxazole with either of the oxazole rings (Figure 70), giving rise to less efficient stacking interactions between the heterocycle and the flat polarizable peptide surface.

99 6. Towards Water Replacement in the S4 Pocket ______

6. Towards Water Replacement in the S4 Pocket

6.1. Conserved Water in the S4 Pocket Water molecules seen in X-ray crystal structures at enzyme active sites can be exploited to enhance the binding affinity of a ligand, either by replacing them in the case of loosely bound water or, in the case of tightly bound water, considering them as a part of the active site and binding to the enzyme through water-mediated hydrogen bonds.[142] However, predicting whether water replacement will lead to a gain in binding affinity remains a challenge in drug design,[186] and thus makes such investigations of great interest. When superimposing the X-ray cocrystal structures of (±)-34, (±)-196, and (±)-198 (PDB code: 2JKH, 2Y5G, and 2Y5H, respectively), a conserved chain of five water molecules was identified at the back of the S4 pocket (Figure 71). The water molecules are hydrogen-bonded to each other and the surrounding amino acids Ser173, Ile175, Glu97, and Thr98 (d(O(H)···(H)O/N) = 2.7–2.9 Å). The water molecules closest + to the quaternary ammonium ion are located at distances of d(O(H)···(H3)C–N (CH3)2) = 3.5 and 3.6 Å.

Figure 71. Left: Conserved water molecules behind the S4 pocket as seen in the X-ray cocrystal structures of (±)-34, (±)-196, and (±)-198 with factor Xa (PDB codes: 2JKH, 2Y5G, and 2Y5H, respectively; only enzyme of 2JKH is shown). Right: The hydrogen- bonding contacts of the water molecules and the closest distances to the ammonium ion as

seen in the structure of (±)-34. Distances in Å. Color code: gray Cenzyme_2JKH, green

Cinhibitor_34, turquoise Cinhibitor_196, magenta Cinhibitor_198, red O, blue N.

100 6. Towards Water Replacement in the S4 Pocket ______

In the plethora of publications from the pharmaceutical industry on the development of highly efficient factor Xa inhibitors, no reports were found on replacing the water molecules behind the S4 pocket. Few accounts have emerged describing N-oxide[187] and pyridine[188] moieties as vectors designed to reach behind the S4 pocket. However, as seen in the X-ray cocrystal structure of factor Xa with a highly active

pyridine-N-oxide inhibitor 223 (Ki = 0.4 nM) prepared at Aventis Pharmaceuticals (Figure 72, PDB code: 1KSN),[189] no water displacement takes place, but the negatively charged O atom rather undergoes water-mediated hydrogen-bonding interactions. This is presumably due to the rigidity of the biaryl S4 vector, which does not orient the N-oxide moiety appropriately to allow water replacement.

O O N HN O

O

H2N NH 223 Ki = 0.4 nM

Figure 72. A highly active factor Xa inhibitor 223 by Aventis Pharmaceuticals.[189] In the X-ray structure the pyridine-N-oxide moiety of the inhibitor reaches behind the S4 pocket

(PDB code: 1KSN). Color code: gray Cenzyme, green Cinhibitor, red O, blue N.

6.2. Inhibitor Design The inhibitors were designed to establish if the binding affinity of the ammonium ion inhibitors would be enhanced by water replacement in the S4 pocket of factor Xa. As seen in the X-ray cocrystal structure of (±)-34, some of the water molecules reside in the narrow cleft formed by amino acids Glu97, Thr98, Phe174, and Ile175 (Figure 73). Occupying this cleft was envisioned by replacing a methyl group of the quaternary ammonium moiety of (±)-34 with a H-bond-donating vector, which would displace a water molecule and its interactions with the surrounding amino acid residues.

101 6. Towards Water Replacement in the S4 Pocket ______

Figure 73. The narrow cleft behind the S4 pocket as seen in the X-ray crystal structure of (±)-34 with factor Xa (PDB code: 2JKH). Extending one methyl group of the ammonium moiety was envisioned to replace a water molecule residing in the cleft. Color code: gray

Cenzyme, green Cinhibitor, red O, blue N, yellow S.

Amido, amino, and hydroxyl groups were selected as water-replacing vectors, leading to the design of compounds (±)-224–(±)-229 (Figure 74). Similar moieties have been used for water replacement in studies of tRNA-guanine transglycosylase inhibitors in the Diederich group.[151] O O R = N N N H H2N H2N HO N R N (±)-224 (±)-225 (±)-226 Br O H N O

N HO N N S HO HO Cl (±)-227 (±)-228 (±)-229

Figure 74. Inhibitors (±)-224–(±)-229 designed to replace water at the back of the S4 pocket.

Modeling by Moloc predicted the hydrogen-bond donor moieties of the inhibitors to extend to the back of the aromatic box where the water molecules are located in the crystal structures, as shown for example in the modeling of 3aS,4R,8aS,8bR-226 in Figure 75. The amido group in (±)-224 could act both as H-bond-accepting and donating moiety,

102 6. Towards Water Replacement in the S4 Pocket ______

whereas the amino group of (±)-225 will be protonated under the conditions of the assay at pH 7.8, and can thus donate three hydrogen bonds. Ligands (±)-226–(±)-229 feature a hydroxyl group as a water-replacement vector.

Figure 75. Modeling of 3aS,4R,8aS,8bR-226 at the active site of factor Xa (2JKH) shows the hydroxyethyl chain extending to the back of the S4 pocket where the water molecules are located, and the ammonium ion residing at a similar position as seen in the crystal

structure of (±)-34. Color code: gray Cenzyme_2JKH, green Cinhibitor_34, light blue Cinhibitor_226, red O, blue N.

6.3. Synthesis The inhibitors were readily prepared by reacting intermediate (±)-57 with the corresponding tertiary amines (Scheme 35). 2-(Dimethylamino)acetamide (230), required for the synthesis of (±)-224, was obtained by treating 2-bromoacetamide with aq. Me2NH solution. The reactions of the tertiary amines bearing hydrogen-bond donors with bromide (±)-57 were extremely slow at RT, and at 40 °C the reaction times spanned from 2 to 7 d. Higher temperatures were avoided due to the occasionally observed ring- opening of the maleimide under heating.

103 6. Towards Water Replacement in the S4 Pocket ______

Reaction Compound R Time T Yield

O (±)-224 N 7 d RT  40 °C 71% H2N

O O H H (±)-225 N – – N N H2N Br N R N O H + O H N tert. amine N (±)-226 N 2 d RT  40 °C 73% Br HO O O

S S (±)-227 N – – HO Cl Cl

(±)-57 (±)-228 HO N 4 d RT  40 °C 95%

(±)-229 N 5 d RT  40 °C 12% HO

Scheme 35. Synthesis of (±)-224–(±)-229.

Although the reaction of (±)-57 with the corresponding tertiary amine to give ligands (±)-225 and (±)-227 was successful, purification of the products failed. Compounds (±)-224, (±)-226, and (±)-228 were obtained in good yield, whereas (±)-229 was isolated in very low yield.

Crystals of (±)-228 were grown from a mixture of MeOH and Et2O, and the single crystal diffraction study was performed by Dr. W. Bernd Schweizer at ETH Zurich (Figure 76). In the crystal structure, dimers formed by one enantiomer show halogen- bonding interactions[190-192] between the Cl substituent of the thiophene and the carbonyl of the maleimide core at a distance and angle typical of the interaction (d(Cl···O=C) = 3.08 Å, (C–Cl–O) = 157.2°, (Cl–O–C) = 133.3°). These interactions were recently studied systematically in protein–ligand complexes, in a collaboration of the Diederich group with F. Hoffmann-La Roche, by preparation of human Cathepsin L inhibitors bearing differently para-substituted phenyl rings.[193] The interaction was found to strengthen in the order Cl < Br < I, and to be nonexistent with organofluorine compounds.

104 6. Towards Water Replacement in the S4 Pocket ______

Figure 76. Small molecule crystal structure of (±)-228 showing halogen-bonding interactions between the Cl substituent of the thiophene and the carbonyl of the maleimide core (CCDC-808209). Distance in Å. Color code: green C, red O, blue N, yellow S, lime Cl.

Crystals of compound (±)-229 were grown from THF, but unfortunately poor quality of the single crystals precludes high resolution of the obtained X-ray crystal structure. However, the structure of the compound could be confirmed (not shown).

6.4. Biological Activities Binding affinities of (±)-224, (±)-226, (±)-228, and (±)-229 towards factor Xa were measured as previously described (Chapter 2.3). Amide derivative (±)-224 (Ki = 26 nM) is a weaker inhibitor of factor Xa than (±)-34 (Figure 77). The same holds for two of the three hydroxyl-bearing compounds, namely dimethyl hydroxyethyl derivative (±)-226

(Ki = 14 nM) and hydroxymethyl pyridinium ion (±)-228 (Ki = 76 nM).

105 6. Towards Water Replacement in the S4 Pocket ______

factor Xa: thrombin: factor Xa: thrombin: Compound R Ki / nM Ki / M clogD Compound R Ki / nM Ki / M clogD

O (±)-224 26 > 35.1 –4.64 O N H H N 2 N R N (±)-226 N 14 > 35.1 –4.88 O H (±)-101 8 > 35.1 –3.73 HO N Br N O (±)-228 HO N 76 > 35.1 –3.75 (±)-106 30 > 35.1 –2.92 S N

Cl (±)-229 N 2 13.2 –3.95 (±)-102 5 29.7 –3.31 N HO

Figure 77. Biological activities and clogD values of the compounds of the water replacement series (±)-224, (±)-226, (±)-228, and (±)-229, and their non-hydroxylated counterparts (±)-101, (±)-106, and (±)-102. The clogD values were calculated (April 2011) at pH 7.4 with ACD/Labs Software, Release 12.00, Product Version 12.01 (see Chapter 8.1).

Compared to their non-hydroxylated counterparts (±)-101 (Ki = 8 nM) and (±)-106

(Ki = 30 nM), the hydroxyl-bearing compounds (±)-226 and (±)-228 are weaker factor Xa inhibitors. Certainly, the H-bond-donating groups in (±)-226 and (±)-228 increase the desolvation penalty that needs to be compensated upon binding of the ligand at the active site. This is also evident by comparing the clogD values of the inhibitors: the ones calculated for the hydroxyl-bearing ligands (±)-226 (–4.88) and (±)-228 (–3.75) are more negative than those of the non-hydroxylated compounds (±)-101 (–3.73) and (±)-106 (–2.92). Thus, if water replacement takes place when (±)-226 and (±)-228 are bound at the active site, the gained binding free enthalpy is not sufficient to compensate for the desolvation penalty. On the other hand, N-(2-hydroxyethyl)pyrrolidinium derivative (±)-229 is the

most efficacious factor Xa inhibitor of the Diederich group to date with Ki = 2 nM. This increase in binding affinity, by a factor of about five as compared to (±)-34 and a factor of about two as compared to the pyrrolidinium ion (±)-102, was thought to indicate successful water replacement.

106 6. Towards Water Replacement in the S4 Pocket ______

6.5. X-Ray Cocrystal Structure of (±)-229 with Factor Xa The X-ray cocrystal structure of (±)-229 in complex with factor Xa was solved by Dr. David W. Banner at F. Hoffmann-La Roche (Figure 78, PDB code: 2Y5F, resolution 1.29 Å), which shows the 3aS,4R,8aS,8bR-configured enantiomer bound at the active site. The binding mode resembles closely that of (±)-34 (PDB code: 2JKH), and in both structures the active site is completely conserved (Figure 79).

Figure 78. The X-ray cocrystal structure of (±)-229 with factor Xa (PDB code: 2Y5F,

resolution 1.29 Å). Distances in Å. Color code: gray Cenzyme, peach Cinhibitor, red O, blue N, yellow S, lime Cl.

However, instead of reaching to the back of the S4 pocket to replace a water molecule as designed, the hydroxyethyl chain of (±)-229 is directed towards the bulk water and undergoes water-mediated hydrogen-bonding interactions to the carbonyl of

Glu97 (d(H2O···(H)OCH2-(±)-229) = 2.8 Å; d(Glu97-C=O···(H)OH) = 2.9 Å). Such bridged interactions between two hydrophilic moieties at the periphery of active sites have been suggested to contribute favorably to ligand binding affinity.[194] As compared with the previously prepared N-methylpyrrolidinium compound

(±)-102 (Ki = 5 nM), the hydrogen bonding of (±)-229 (Ki = 2 nM) evidently provides some stabilization to the complex, compensating for the additional desolvation costs from the hydroxyethyl chain and resulting in higher binding affinity for (±)-229 than for (±)-102. As compared to (±)-34, the higher activity of (±)-229 can be partly attributed to the additional positively polarized C–H residues of the pyrrolidinium cation, similarly to

107 6. Towards Water Replacement in the S4 Pocket ______

(±)-102, which expand the surface capable of undergoing cation– interactions, and partly to the water-mediated hydrogen bonding of the hydroxyethyl chain.

Figure 79. Overlay of the X-ray cocrystal structures of (±)-34 and (±)-229 bound in

factor Xa (PDB code: 2JKH and 2Y5F, respectively). Color code: gray Cenzyme_2Y5F,

purple Cenzyme_2JKH, green Cinhibitor_34, peach Cinhibitor_229, red O, blue N, yellow S, lime Cl.

The results show that the hydroxyethyl chain is too flexible for a water- replacement vector, as it can readily flip to benefit from solvation of the bulk water. To achieve successful water displacement, the S4 substituent needs to be more preorganized. More rigid substituents, such as spirocyclic quaternary ammonium ions as in 3aS,4R,8aS,8bR-231 (Figure 80), are envisioned to enable optimal positioning of the H-bond-donating moiety and thus successful displacement of water.

O H N O N N O H HN N O

S

Cl

3aS,4R,8aS,8bR-231

Figure 80. Overlay of (±)-34 and the modeled 3aS,4R,8aS,8bR-231 at the active site of

factor Xa. Color code: gray Cenzyme_2JKH, green Cinhibitor_34, orange Cinhibitor_231, red O, blue N, yellow S, lime Cl.

108 7. Conclusions and Outlook ______

7. Conclusions and Outlook

Factor Xa has been shown to be an excellent model system for molecular recognition studies. The rigid nature of its active site, completely conserved in all four X-ray cocrystal structures obtained during this thesis, allows the investigation and quantification of noncovalent interactions involved in the binding of the inhibitors. In addition, due to the rigidity of the tricyclic core of the ligands, the inhibitors bind in a similar manner, which enables direct comparison between the binding affinities and thus, quantification of key interactions by single-atom mutations. The first goal of this PhD thesis was to enhance the binding affinity of the

previously prepared ammonium ion inhibitor (Ki = 280 nM) by exchanging the phenylamidinium vector for a neutral S1 needle. This was achieved by incorporating a chlorothienyl-isoxazolyl moiety into the tricyclic scaffold, resulting in an enhancement of binding affinity by a factor of 30 (Ki = 9 nM). The X-ray cocrystal structure of the inhibitor with factor Xa solved by Dr. David W. Banner at F. Hoffmann-La Roche showed the chlorothienyl moiety interacting with Tyr228, and the biaryl needle undergoing stacking interactions with the polarizable peptide walls of the S1 pocket.

Cation– interactions O S4 Tyr99 Gln192 OH H N N O Water replacement H Phe174 HN Cys191 H O Trp215 N O

Ala190 Stacking interactions Gly216

S1 Asp189

Water replacement OH Tyr228

With a highly active inhibitor in hand, the project was focused on the systematic exploration of cation– interactions in the aromatic box of the S4 pocket, formed by the side chains of Tyr99, Phe174, and Trp215. Using the inhibitor scaffold bearing the chlorothienyl-isoxazolyl needle, cation– interactions of the trimethylammonium ion in the S4 pocket were quantified by N+/C single-atom mutation as –G = 2.8 kcal mol–1, corresponding to approximately 0.8 kcal mol–1 per aromatic ring. The S4 pocket was also

109 7. Conclusions and Outlook ______

found to tolerate larger ammonium ions than trimethylammonium, such as N-methylpyrrolidinium and N-methylpiperidinium. The effect of methylation on the magnitude of the cation– interaction was found to be dramatic. Moving from the primary ammonium ion inhibitor to the quaternary ammonium one, a stepwise enhancement of binding free enthalpy of 1.2–1.8 kcal mol–1 per methylation was observed. Although the quaternary ammonium ion inhibitor is clearly superior, the tertiary ammonium ion also binds with very high affinity, indicating that such residues could be optimal to bind to aromatic binding sites in general. In addition to the ammonium methylation series, a neutral alkyl-bearing series was prepared. This series showed that although some binding affinity can be gained through proper filling and increasing the hydrophobic interactions in the S4 pocket, these are not responsible for the large activity enhancements in the ammonium ion series, but rather the strong cation– interactions are responsible for the high activity of the ammonium ion inhibitors. The role of the thienyl substituent interacting with Tyr228 in the S1 pocket, and concomitantly replacing a water molecule, was investigated. Bromine and chlorine were found to provide the most active inhibitors, whereas fluoro- and iodothienyl-bearing ligands were bound more weakly, suggesting that size and polarizability play a significant role in replacing the conserved water molecule in the S1 pocket. The positioning of the heteroatoms on the isoxazole ring was found to substantially influence the binding affinity of the inhibitors. As seen in the X-ray cocrystal structures, the oxazole rings adopted a flipped conformation as compared to the isoxazole ring, inducing a repositioning of the tricyclic core of the inhibitors. This is thought to stem from the requirement of a specific approach angle of the chlorothienyl moiety towards Tyr228, as well as from conformational preferences of the inhibitors. The loss of binding affinity is proposed to originate from the interplay of different factors, such as the slightly displaced position of the Cl atom of the thienyl moiety with respect to Tyr228, and some potentially repulsive contacts between the ligand and the active site. More importantly however, the suboptimal matching of polar +···– contacts between the oxazole rings and the polarizable peptide walls of the S1 pocket and the twisting of the oxazole rings by an angle of 30° as compared to the isoxazole ring result in the weakening of the stacking interactions, and thus, in the decrease of binding affinity of the inhibitors, highlighting the importance of complementary contacts when binding on peptide surfaces.

110 7. Conclusions and Outlook ______

Finally, the identification of a chain of conserved water molecules residing at the back of the S4 pocket led to the first attempts to replace these by incorporating H-bond- donating residues onto the quaternary ammonium ion. The inhibitor containing a pyrrolidinium hydroxyethyl S4 moiety was the most active factor Xa inhibitor prepared

during this thesis, and in the Diederich group to date, with Ki = 2 nM. Although the high affinity suggested successful water replacement, the X-ray cocrystal structure showed the chain of water molecules conserved, and the hydroxyethyl chain undergoing water- mediated hydrogen bonding to Glu97 instead. The hydrogen bond seems to account for a gain in binding affinity by at least a factor of two, as compared to the non-hydroxylated counterpart of the pyrrolidinium inhibitor.

The results described above raise new interesting questions, which could be addressed in a follow-up study. First, new insight into cation– interactions could be gained by investigating the thermodynamics of the binding of the cationic and the neutral methylation series. Due to the high cost of the enzyme, performing ITC studies is not feasible. As an altenative, thermodynamic studies could be conducted using surface plasmon resonance: If Kd values are measured at different temperatures, the thermodynamics of binding can be determined using the van’t Hoff equation. The study of counteranion effects on the magnitude of the cation– interaction was initiated: the iodide compound was found only slightly less active than the bromide one, which indicates negligible counterion effects under the conditions of the assay. However, to ascertain this proposition, a series of compounds bearing different counteranions should be prepared. This series should contain chloride, a smaller ion than the already investigated bromide and iodide, polyatomic anions, such as nitrate, and larger counterions, such as tosylate or hexafluorophosphate. To verify the influence of the heteroatom positions on the strength of the stacking interactions between the oxazole linker and the peptide walls of the S1 pocket, a series of inhibitors containing different heterocyclic linkers between the tricyclic core and the chlorothienyl moiety should be prepared (Figure 81, right). Obtaining X-ray cocrystal structures is especially relevant to evaluate the polar contacts between the heterocycles and the peptide walls and to verify the binding conformation of the inhibitor. After completing the oxazole series targeted during this Thesis, the isoxazole linker could be replaced with a heterocycle containing an H-bond-donating vector, such as pyrrole or

111 7. Conclusions and Outlook ______

imidazole, to establish a hydrogen bond with the carbonyl of Gly218 and to enable the evaluation of the repulsive contacts observed for the oxazole ring. Conformational preferences of the inhibitors could be investigated with an oxadiazole series. To enable successful water replacement in the S4 pocket, a more preorganized vector is required to target the cleft behind the S4 pocket (Figure 81, left). Moieties such as spirocyclic ammonium ions are envisioned as suitable vectors. In addition, the pyrrolidinium ring of the inhibitor with the highest binding affinity could be substituted with a second hydroxyalkyl moiety, which would be directed to the cleft while the other hydroxyl group undergoes water-mediated hydrogen-bonding interactions in the S4 pocket.

HO O O NH O N H N N O N R = N N R N HN O H Br N OH Het Het = O O N N NH

S O N N N HN Cl O

Figure 81. The proposed target compounds for water replacement in the S4 pocket (left), and to investigate the stacking interactions in the S1 pocket (right).

Overall, a deeper understanding into the molecular recognition phenomena involved in the binding of ligands to factor Xa was gained in this work. In addition, the results presented herein also raise interesting questions, which could be answered by preparing and studying the proposed target compounds, leading to further insight into the fascinating interplay of different interactions involved in the binding of inhibitors at the active site. The knowledge gained through this work can be applied to other biological systems as well, thus ultimately benefiting the design process of new efficient ligands for drug discovery.

112 8. Experimental ______8. Experimental

8.1. Materials and Methods

Commercially avalable chemicals were used without further purification.

Column chromatography (CC) was carried out with SiO2 60 (particle size 0.040– 0.063 mm, 230–400 mesh; Fluka) and distilled technical solvents.

Dry solvents (DMF, CH2Cl2, MeOH, n-hexane, 1,4-dioxane, and toluene) for reactions were purified by a solvent drying system from LC Technology Solutions

Inc. SP-105 under nitrogen atmosphere (H2O content < 10 ppm as determined by Karl-Fischer titration). THF was freshly distilled from sodium benzophenone ketyl, all other solvents were purchased in p.a. quality.

Thin-layer chromatography (TLC) was conducted on aluminum sheets coated with

SiO2 60 F254 obtained from Macherey-Nagel; visualization with a UV lamp (254 nm).

Melting points (m.p.) were measured on a Büchi B-540 melting point apparatus in open capillaries and are uncorrected.

Optical rotation () was recorded on a Perkin Elmer 241 polarimeter using a 1 dm cell at RT,  = 589 nm (Na D-line). Concentrations are given as mg mL–1.

1H NMR, 13C NMR, 19F NMR, and 31P NMR spectra were measured with Varian Gemini 300, Varian Mercury 300, Bruker ARX 300, Bruker DRX400, Bruker AV400, Bruker DRX500 or on Bruker AV600 spectrometer. Chemical shifts are 1 reported in ppm relative to the signal of Me4Si. Residual solvent signals in the H and 13C NMR spectra were used as an internal reference. Coupling constants (J) are given in Hz. The resonance multiplicity is described as s (singlet), d (doublet), t (triplet), q (quadruplet), quint. (quintet), m (multiplet), and br (broad).

113 8. Experimental ______Infrared spectra (IR) were recorded on a Varian 800 FT-IR spectrometer or on Perkin Elmer Spectrum BX FT-IR system. The spectra were measured in the neat state or as a thin film; absorption bands are reported in wavenumbers (cm–1).

High-resolution (HR) EI-MS spectra were measured with a Waters Micromass AutoSpec Ultima spectrometer. HR ESI spectra were measured with Bruker's maXis ESI/NanoSpray-Qq-TOF instrument. HR FT-MALDI spectra were measured with a Varian IonSpec FT-ICR instrument using 3-hydroxypyridine-2-carboxylic acid (3-HPA) as matrix.

Elemental analyses were performed at the Laboratorium für Organische Chemie, ETH Zürich, with a LECO CHN/900 instrument.

Nomenclature follows the suggestions proposed by the software ACD Names of ACD/Labs.[195] Atom numbering is arbitrary and does not follow the numbering of the IUPAC names.

The clogD values were calculated (April 2011) at pH 7.4 with ACD/Labs Software, Release 12.00, Product Version 12.01. The values obtained using this version differ from the ones obtained in 2009 using an older version of the same software, however, the values follow the same trend. The values calculated in 2009 for the inhibitor series probing the size of the S4 pocket (Figure 35) are –1.19 (±)-34, –1.26 for (±)-101, –0.02 for (±)-103, –0.15 for (±)-104, –0.91 for (±)-105, and –1.90 for (±)-106. For the ammonium ion methylation series, the values are –0.3 for (±)-108,

–0.45 for (±)-110, and +0.74 for (±)-111 (Figure 40). The pKa values of the conjugate acids of the terminal amines were calculated with the software ACD pKa from ACD/Labs.

The logD values were determined at F. Hoffmann-La Roche by high-throughput shake-flask method. The applied methods, called CAMDIS and HTlogD for the determination of the distribution coefficient logD are derived from the conventional 'shake-flask' method. The compound is distributed between water buffered at a specific pH and 1-octanol. The distribution coefficient is then calculated from the difference in concentration in the aqueous phase before and after partitioning and the

114 8. Experimental ______ratio of the two phases. The "one phase-analyzed" experiment starts with a pure

(CH3)2SO solution of the molecule of interest, which is dispensed in aqueous buffer. A part of this solution is then analyzed by measuring the UV absorption. The obtained optical density (reference) is equal to the concentration of the substance before partitioning. An exact amount of 1-octanol is added and the mixture incubated by quiet shaking (2 h). The emulsion is allowed to stand overnight to ascertain reaching of the partition equilibrium. The concentration of the substance in the aqueous phase is then determined again by measuring the UV absorption. This procedure is carried out at four different octanol/water ratios, two with a large volume of octanol for hydrophilic compounds (logD < 1) and two with a low volume of octanol for the lipophilic compounds (logD > 1).

High-performance column chromatography (HPLC) was run on a Merck-Hitachi LaChrom D-Line system equipped with a D-7000 Interface, an L-7100 pump, and an L-7400 UV-detector. Preparative Recycling HPLC was run on a Japan Analytical Industry LC-9101 system with an L-7110 pump and a 3702 UV-detector.

115 8. Experimental ______8.2. Synthetic Procedures

3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-34) O 8 12 9 H 11 8a 7 12 N 8b 3a N 10 N 6 O 4 Br 12 H N 3' O

2' S 1' Cl

4.2 M Me3N in EtOH (0.49 mL, 2.06 mmol, 20 equiv.) was added to (±)-57 (50 mg, 0.10 mmol, 1 equiv.) in EtOH (1.5 mL) and the mixture stirred at RT for 96 h. The mixture was concentrated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (±)-34 (52 mg, 93%) as a white solid. 1 M.p. 154 °C (decomp); H NMR (300 MHz, CD3OD): 1.80–1.91 (m, 2H, H-C(7), H-C(8)), 2.00–2.22 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.82–2.91 (m, 1H, H-C(6)), 2.92–3.02 (m, 1H, H-C(6)), 3.15 (s, 9H, H-C(12)), 3.35–3.39 (m, 2H, H-C(11)), 3.50 (dd, J = 8.1, 1.4, 1H, H-C(8b)), 3.53–3.62 (m, 2H, H-C(9)), 3.72–3.77 (m, 1H, H-C(8a)), 3.86 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.43 (d, J = 8.7, 1H, H-C(4)), 6.63 (s, 1H, H-C(3’)), 7.10 (d, J = 4.0, 1H, H-C(1’)), 7.42 (d, J = 4.0, 1H, H-C(2’)); 13C NMR

(75 MHz, CD3OD): 22.7, 24.1, 30.2, 36.7, 50.6, 50.8, 52.3, 53.6 (t), 62.6, 65.2 (t), 69.5, 101.3, 128.1, 128.97, 129.03, 134.1, 164.8, 165.0, 177.4, 179.9; IR (ATR): 2957, 2872, 1773, 1699, 1614, 1522, 1476, 1432, 1396, 1366, 1352, 1352, 1311, 1235, 1182, 1145, 1086, 1066, 998, 925, 878, 795; HR-MALDI-MS: m/z (%): + 37 + 465.1535 (38, [M–Br] , calcd for C22H28 ClN4O3S : 465.1537), 463.1565 (100, [M– + 35 + Br] , calcd for C22H28 ClN4O3S : 463.1565).

116 8. Experimental ______3-[(3aS,4R,8aS,8bR)-4-[5-(5-Chloro-2-thienyl)-1,2-oxazol-3-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((+)-34)

O 8 12 9 H 11 8a 7 12 N 8b 3a N 10 N 6 O H 4 12 Br N 3' O (+) 2' S 1' Cl To a mixture of (+)-57 (20 mg, 0.04 mmol, 1 equiv.) in EtOH (1 mL) at RT was

added 4.2 M Me3N in EtOH (200 L, 0.83 mmol, 20 equiv.). The mixture was stirred at RT for 46 h and evaporated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (+)-34 (20 mg, 89%) as off-white crystals. 25 1 M.p. 214 °C (decomp); []D : +141.2 (c = 1, CH3OH); H NMR (600 MHz,

CD3OD): 1.80–1.89 (m, 2H, H-C(7), H-C(8)), 2.01–2.09 (m, 2H, H-C(10)), 2.11–2.19 (m, 2H, H-C(7), H-C(8)), 2.87 (ddd, J = 12.7, 8.3, 4.3, 1H, H-C(6)), 2.97 (ddd, J = 12.6, 8.7, 6.9, 1H, H-C(6)), 3.15 (s, 9H, H-C(12)), 3.34–3.36 (m, 2H, H-C(11)), 3.49 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.56 (qt, J = 14.9, 6.5, 2H, H-C(9)), 3.73–3.76 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.43 (d, J = 8.8, 1H, H-C(4)), 6.63 (s, 1H, H-C(3’)), 7.11 (d, J = 4.0, 1H, H-C(1’)), 7.42 (d, J = 4.0, 1H, H-C(2’)); 13 C NMR (150 MHz, CD3OD): 22.8, 24.2, 30.2, 36.7, 50.6, 50.8, 52.4, 53.6 (t), 62.7, 65.3 (t), 69.6, 101.4, 128.1, 129.0, 129.1, 134.2, 164.8, 165.1, 177.4, 179.9; IR (ATR): 3386 (b), 2957, 1772, 1694, 1614, 1524, 1476, 1423, 1402, 1351, 1319, 1283, 1183, 1146, 1067, 1030, 1000, 965, 923, 878, 793, 751, 666, 628; HR-MALDI- + 37 + MS: m/z (%): 465.1536 (42, [M–Br] , calcd for C22H28 ClN4O3S : 465.1537), + 35 + 463.1570 (100, [M–Br] , calcd for C22H28 ClN4O3S : 463.1565).

117 8. Experimental ______3-[(3aR,4S,8aR,8bS)-4-[5-(5-Chloro-2-thienyl)-1,2-oxazol-3-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((–)-34) O 8 12 9 H 11 8a 7 12 N 8b 3a N 10 N 6 O H 4 12 Br N 3' O

(–) 2' S 1' Cl To a mixture of (–)-57 (20 mg, 0.04 mmol, 1 equiv.) in EtOH (1 mL) at RT was

added 4.2 M Me3N in EtOH (200 L, 0.83 mmol, 20 equiv.). The mixture was stirred at RT for 46 h and evaporated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (–)-34 (21 mg, 94%) as off-white crystals. 25 1 M.p. 214 °C (decomp); []D : –137.2 (c = 1, CH3OH); H NMR (600 MHz, CD3OD): 1.79–1.89 (m, 2H, H-C(7), H-C(8)), 2.02–2.09 (m, 2H, H-C(10)), 2.11–2.19 (m, 2H, H-C(7), H-C(8)), 2.87 (ddd, J = 12.7, 8.4, 4.4, 1H, H-C(6)), 2.97 (ddd, J = 12.7, 8.7, 7.0, 1H, H-C(6)), 3.14 (s, 9H, H-C(12)), 3.33–3.36 (m, 2H, H-C(11)), 3.48 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.56 (qt, J = 15.0, 6.5, 2H, H-C(9)), 3.73–3.76 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.42 (d, J = 8.8, 1H, H-C(4)), 6.63 (s, 1H, H-C(3’)), 7.10 (d, J = 4.0, 1H, H-C(1’)), 7.42 (d, J = 3.9, 1H, H-C(2’)); 13C NMR

(150 MHz, CD3OD): 22.8, 24.2, 30.2, 36.7, 50.7, 50.8, 52.4, 53.7 (t), 62.7, 65.3 (t), 69.6, 101.4, 128.1, 129.0, 129.1, 134.2, 164.9, 165.1, 177.5, 179.9; IR (ATR): 3385 (b), 2957, 1772, 1694, 1524, 1476, 1423, 1402, 1351, 1319, 1283, 1246, 1206, 1183, 1145, 1068, 1030, 1000, 965, 923, 898, 878, 794, 750, 665, 627; HR-MALDI-MS: + 37 + m/z (%): 465.1528 (41, [M–Br] , calcd for C22H28 ClN4O3S : 465.1537), 463.1562 + 35 + (100, [M–Br] , calcd for C22H28 ClN4O3S : 463.1565).

118 8. Experimental ______3-[(3aSR,4RS,8aSR,8bRS)-4-[3-(5-Chloro-2-thienyl)phenyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-35) O 8 12 9 H 11 8a 7 12 N 8b 3a N 10 N 6 O 4 Br 12 H 3' 6'

S 5' Cl 4' 2' 1'

4.2 M Me3N in EtOH (1.93 mL, 8.10 mmol, 20 equiv.) was added to (±)-59 (200 mg, 0.41 mmol, 1 equiv.) in EtOH (5 mL) and the mixture stirred at RT for 120 h. The mixture was evaporated in vacuo. The residue was dissolved in MeOH. Precipitation

from Et2O gave (±)-35 (150 mg, 67%) as a white solid. 1 M.p. 210 °C (decomp); H NMR (300 MHz, CD3OD): 1.78–2.02 (m, 4H, H-C(7), H- C(8), H-C(10)), 2.06–2.20 (m, 2H, H-C(7), H-C(8)), 2.78 (ddd, J = 12.5, 8.3, 4.2, 1H, H-C(6)), 2.88–2.98 (m, 1H, H-C(6)), 3.05 (s, 9H, H-C(12)), 3.21–3.35 (m, 2H, H- C(11)), 3.46 (t, J = 6.9, 3H, H-C(9), H-C(8b)), 3.71–3.78 (m, 2H, H-C(3a), H-C(8a)), 4.27 (d, J =8.7, 1H, H-C(4)), 6.97 (d, J = 3.9, 1H, H-C(2’)), 7.21 (d, J = 3.9, 1H, H- C(1’)), 7.28 (dt, J = 7.8, 1.5, 1H, H-C(6’)), 7.34 (t, J = 7.5, 1H, H-C(5’)), 7.47 (dt, J = 13 7.5, 1.5, 1H, H-C(4’)), 7.55 (s, 1H, H-C(3’)); C NMR (125 MHz, CD3OD): 22.9, 24.3, 30.5, 36.7, 50.5, 51.9, 52.2, 53.6 (t), 65.3 (t), 69.3, 69.7, 123.9, 125.8, 126.1, 128.8, 129.3, 129.9, 130.0, 134.7, 140.6, 144.5, 177.7, 180.3; IR (ATR): 1768, 1694, 1603, 1482, 1448, 1433, 1398, 1360, 1318, 1228, 1181, 1143, 1093, 1068, 1021, 985, 927, 877, 793, 703; HR-MALDI-MS: m/z (%): 474.1788 (39, [M–Br]+, calcd for 37 + + 35 + C25H31 ClN3O2S : 474.1793), 472.1824 (100, [M–Br] , calcd for C25H31 ClN3O2S : 472.1820).

119 8. Experimental ______1-(3-Bromopropyl)-1H-pyrrole-2,5-dione (36)[89] O Br 3 1 4 N 2 1 O Compound 39 (6.21 g, 26.31 mmol, 1 equiv.) and AcONa (1.10 g, 13.42 mmol,

0.51 equiv.) were heated to reflux in Ac2O (65 mL) for 23 h. The mixture was poured

onto aq. 1 M NaOH solution (600 mL). The aq. phase was extracted with CH2Cl2.

The org. phase was dried over MgSO4, filtrated, and evaporated in vacuo.

Purification by CC (SiO2; CH2Cl2) gave 36 (3.02 g, 53%) as a white solid. 1 Rf = 0.58 (hexane/EtOAc); H NMR (300 MHz, CDCl3): 2.18 (q, J = 6.9, 2H, H-C(3)), 3.37 (t, J = 6.6, 2H, H-C(2)), 3.68 (t, J = 6.6, 2H, H-C(4)), 6.72 (s, 2H, 13 H-C(1)); C NMR (75 MHz, CDCl3): 29.7, 31.6, 36.7, 134.1, 170.4; HR-EI-MS: m/z + 79 + + (%): 216.9730 (19, [M] , calcd for C7H8 BrNO2 : 216.9738), 138.0549 (34, [M–Br] , + + calcd for C7H8NO2 : 138.0555), 110.0239 (100, [M–CH2CH2Br] , calcd for + C5H4NO2 : 110.0242).

(2Z)-4-[(3-Bromopropyl)amino]-4-oxobut-2-enoic Acid (39)[89] O

1 OH 4 2 NH Br 3 5 O

Et3N (10.5 mL, 75.4 mmol, 1.1 equiv.) was added slowly to a mixture of 3-aminopropylbromide hydrobromide (15.0 g, 68.5 mmol, 1 equiv.) and maleic

anhydride (6.72 g, 68.5 mmol, 1 equiv.) in CH2Cl2 (200 mL) at 0 °C. The mixture was stirred at RT for 2.5 h. The mixture was evaporated in vacuo and the residue

dissolved in CH2Cl2. Conc. HBr (1.25 mL) was added and the mixture was washed with aq. 1 M HBr solution. The org. phase was dried over MgSO4, filtrated, and evaporated in vacuo to give 39 (11.6 g, 72%) as a white solid. 1 M.p. 98 °C; H NMR (300 MHz, CD3OD): 2.11 (quint., J = 6.6, 2H, H-C(4)), 3.44 (t, J = 6.7, 2H, H-C(3)), 3.50 (t, J = 6.6, 2H, H-C(5)), 6.23 (d, J = 12.6, 1H, H-C(1)), 13 6.45 (d, J = 12.6, 1H, H-C(2)); C NMR (75 MHz, CD3OD): 31.2, 33.1, 39.4, 133.0, 134.4, 168.0, 168.2; IR (ATR): 3236, 3066, 2977, 2914, 2870, 1908, 1854, 1699,

120 8. Experimental ______1636, 1578, 1498, 1461, 1446, 1407, 1374, 1318, 1292, 1257, 1225, 1214, 1163, 1076, 1034, 956, 894, 849, 796, 757, 738, 661, 619; HR-EI-MS: m/z (%): 156.0655 + + + (10, [M–Br] , calcd for C7H10NO3 : 156.0656), 99.0074 (54, [M–Br(CH2)3NH] , calcd + for C4H3O3 : 99.0082), 55.0208 (27), 54.0157 (22), 30.0428 (100), 27.0340 (30).

5-(5-Chloro-2-thienyl)-3-isoxazolecarbaldehyde (40) ON O Cl S 4 3 1 2 PCC (6.00 g, 27.8 mmol, 1.5 equiv.) was added to a solution of 44 (4.00 g,

18.5 mmol, 1 equiv.) in CH2Cl2 (300 mL) and the mixture was stirred at RT for 18.5

h. The mixture was filtrated and concentrated in vacuo. Purification by CC (SiO2; hexane/CH2Cl2 1:3) gave 40 (3.06 g, 77%) as a yellow solid. 1 Rf = 0.35 (hexane/CH2Cl2 1:3); m.p. 112–113 °C; H NMR (300 MHz, CDCl3): 6.70 (s, 1H, H-C(3)), 6.99 (d, J = 3.9, 1H, H-C(1)), 7.36 (d, J = 3.9, 1H, H-C(2)), 10.15 (s, 13 1H, H-C(4)); C NMR (75 MHz, CDCl3): 96.0, 126.5, 127.4, 127.5, 134.6, 162.5, 165.9, 184.3; IR (ATR): 3390, 3174, 3135, 3110, 3090, 3039, 2866, 2810, 2711, 2625, 2531, 2159, 2050, 1881, 1808, 1749, 1703, 1648, 1587, 1523, 1449, 1417, 1365, 1342, 1273, 1244, 1205, 1170, 1115, 1082, 1052, 1022, 994, 945, 898, 819, 806, 743, 680, 667, 637; HR-EI-MS: m/z (%): 214.9615 (63, [M]+, calcd for 37 + + 35 + C8H4 ClNO2S : 214.9622), 212.9646 (23, [M] , calcd for C8H4 ClNO2S : + 37 + 212.9651), 146.9476 (36, [M–C3H2NO] , calcd for C5H2 ClOS : 146.9485), + 35 + 144.9510 (100, [M–C3H2NO] , calcd for C5H2 ClOS : 144.9515); elemental analysis

calcd for C8H4ClNO2S (212.96): C 44.98, H 1.89, N 6.56; found: C 44.92, H 1.87, N 6.52.

Ethyl 4-(5-Chloro-2-thienyl)-2,4-dioxobutanoate (42)[101] O O Cl S O 3 5 12 O 4 KOtBu (3.46 g, 30.8 mmol, 1 equiv.) was added to 2-acetyl-5-chlorothiophene (41) (4.95 g, 30.8 mmol, 1 equiv.) in toluene (150 mL) at 0 °C. Diethyl oxalate (4.20 mL, 30.8 mmol, 1 equiv.) was added dropwise along with toluene (20 mL). After stirring

121 8. Experimental ______for 18.5 h at RT, the mixture was filtrated and the precipitate was washed with

toluene. The residue was dissolved in EtOAc and the org. phase washed with aq. 1 M

HCl solution. The org. phase was dried over MgSO4, filtrated, and evaporated in vacuo. Compound 42 (6.71 g, 84%) was obtained as an orange solid. 1 Rf = 0.43 (hexane/EtOAc 1:1); m.p. 80–82 °C; H NMR (300 MHz, CDCl3): 1.40 (t, J = 7.2, 3H, H-C(5)), 4.39 (q, J = 7.2, 2H, H-C(4)), 6.83 (s, 1H, H-C(3)), 7.02 (d, 13 J = 4.2, 1H, H-C(1)), 7.64 (d, J = 4.2, 1H, H-C(2)); C NMR (75 MHz, CDCl3): 14.1, 62.7, 98.8, 128.2, 132.1, 140.7, 141.2, 162.0, 164.5, 185.2; IR (ATR): 3099, 2981, 1726, 1626, 1573, 1522, 1403, 1362, 1322, 1245, 1112, 1082, 1018, 927, 901, 864, + 37 + 809, 687, 652; HR-EI-MS: m/z (%): 261.9875 (3, [M] , calcd for C10H9 ClO4S : + 35 + 261.9881), 259.9906 (9, [M] , calcd for C10H9 ClO4S : 259.9905), 188.9585 (36, [M– + 37 + CH3CH2OCO] , calcd for C7H4 ClO2S : 188.9591), 186.9613 (100, [M– + 35 + CH3CH2OCO] , calcd for C7H4 ClO2S : 186.9621), 146.9480 (13, [M– + 37 + CH3CH2OCOCOCH2] , calcd for C5H2 ClOS : 146.9485), 144.9512 (35, [M– + 35 + CH3CH2OCOCOCH2] , calcd for C5H2 ClOS : 144.9515); elemental analysis calcd

for C10H9ClO4S (259.99): C 46.07, H 3.48; found: C 45.97, H 3.44.

Ethyl 5-(5-Chloro-2-thienyl)isoxazole-3-carboxylate (43)[101] O N 5 O Cl S 4 3 O 12 Hydroxylamine hydrochloride (6.60 g, 94.9 mmol, 3.75 equiv.) was added to 42 (6.60 g, 25.3 mmol, 1 equiv.) in EtOH (350 mL) at RT. The mixture was heated to reflux for 17 h. The mixture was evaporated in vacuo and suspended in water. The pH was set to 7 with aq. ammonium hydroxide (25%) and the mixture extracted with

EtOAc. The org. phase was dried over MgSO4, filtrated, and evaporated in vacuo. Compound 43 (6.41 g, 98%) was obtained as a brown solid. 1 Rf = 0.86 (hexane/EtOAc 1:1); m.p. 59–60 °C; H NMR (300 MHz, CDCl3): 1.43 (t, J = 7.2, 3H, H-C(5)), 4.39 (q, J = 7.2, 2H, H-C(4)), 6.73 (s, 1H, H-C(3)), 6.98 (d, 13 J = 3.9, 1H, H-C(1)), 7.34 (d, J = 3.9, 1H, H-C(2)); C NMR (75 MHz, CDCl3): 14.2, 62.4, 99.6, 126.7, 127.1, 127.5, 134.3, 157.0, 159.7, 165.5; IR (ATR): 3140, 3099, 2982, 1726, 1592, 1522, 1480, 1461, 1453, 1414, 1392, 1360, 1339, 1253, 1224, 1193, 1135, 1114, 1070, 1024, 1016, 1000, 929, 900, 866, 829, 806, 772, 674, 668,

122 8. Experimental ______

+ 37 + 618; HR-EI-MS: m/z (%): 158.9881 (17, [M] , calcd for C10H8 ClNO3S : 258.9884), + 35 + 256.9908 (46, [M] , calcd for C10H8 ClNO3S : 256.9908), 146.9488 (15, [M– + 37 + + C5H6NO2] , calcd for C5H2 ClOS : 146.9485), 144.9518 (40, [M–C5H6NO2] , calcd 35 + for C5H2 ClOS : 144.9515), 29.0587 (100).

[5-(5-Chloro-2-thienyl)isoxazol-3-yl]methanol (44)[101] O N OH Cl S 4 3 12

NaBH4 (0.85 g, 22.5 mmol, 1 equiv.) was added to a solution of 43 (5.80 g, 22.5 mmol, 1 equiv.) in EtOH (250 mL) at 0 °C and the mixture stirred at RT for 4.5 d. The mixture was evaporated in vacuo and dissolved in water. The pH was set to 3–4 with conc. HCl. The precipitated product was filtrated and washed with water. Compound 44 (4.60 g, 95%) was obtained as a yellow solid. 1 Rf = 0.50 (hexane/EtOAc 1:1); m.p. 88–89 °C; H NMR (300 MHz, CDCl3): 2.01 (br. s, 1H, OH), 4.78 (s, 2H, H-C(4)), 6.40 (s, 1H, H-C(3)), 6.95 (d, J = 3.9, 1H, H-C(1)), 13 7.28 (d, J = 3.9, 1H, H-C(2)); C NMR (75 MHz, CDCl3): 57.0, 98.0, 126.4, 127.3, 127.6, 133.4, 164.1, 164.2; IR (ATR): 3389, 3093, 2935, 2891, 1717, 1598, 1530, 1423, 1396, 1338, 1244, 1202, 1174, 1244, 1202, 1174, 1072, 1038, 992, 959, 897, 820, 794, 747, 677, 636; HR-EI-MS: m/z (%): 216.9767 (23, [M]+, calcd for 37 + + 35 + C8H6 ClNO2S : 216.9778), 214.9803 (62, [M] , calcd for C8H6 ClNO2S : + 37 + 214.9802), 146.9482 (36, [M–C3H4NO] , calcd for C5H2 ClOS : 146.9485), + 35 + 144.9513 (100, [M–C3H4NO] , calcd for C5H2 ClOS : 144.9515), 31.0343 (43);

elemental analysis calcd for C8H6ClNO2S (214.98): C 44.56, H 2.80, N 6.49; found: C 44.53, H 2.78, N 6.45.

4-(5-Chloro-2-thienyl)-1,3-thiazole-2-carbaldehyde (45)

3 S Cl S N O 1 2 A mixture of compound 48 (278 mg, 1.20 mmol, 1 equiv.) and PCC (389 mg,

1.80 mmol, 1.5 equiv.) in CH2Cl2 (25 mL) was stirred at RT for 16 h. The mixture

123 8. Experimental ______

was filtered and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2) gave 45 (181 mg, 66%) as beige solid. 1 Rf = 0.44 (hexane/EtOAc 4:1); H NMR (300 MHz, CDCl3): 6.93 (d, J = 3.9, 1H, H-C(1)), 7.29 (d, J = 3.9, 1H, H-C(2)), 7.68 (d, J = 1.2, 1H, H-C(3)), 10.01 (d, J = 13 1.2, 1H, CHO); C NMR (75 MHz, CDCl3): 118.3, 124.2, 127.0, 131.2, 135.2, 151.9, 165.5, 183.4; IR (ATR): 3107, 2924, 2853, 1681, 1452, 1406, 1285, 1235, 1175, 1087, 994, 878, 797, 647; HR-EI-MS: m/z (%): 230.9389 (41, [M]+, calculated for 37 + + C8H4 ClNOS2 : 230.9393), 228.9417 (100, [M] , calculated for 35 + + + C8H4 ClNOS2 :228.9423), 173.9364 (26, [M–C2HNO] , calculated for C6H3ClS2 : 173.9365), 138.9676 (29).

2-Bromo-1-(5-chloro-2-thienyl)ethanone (46)[104] O S Cl Br 3 1 2 2-Acetyl-5-chlorothiophene (41) (1.50 g, 9.34 mmol, 1 equiv.) was heated to reflux in

CHCl3/EtOAc (1:1; 30 mL) under N2. CuBr2 (3.96 g, 17.74 mmol, 1.9 equiv) was added in four portions and heating continued for 53 h. The mixture was filtrated and

evaporated in vacuo. Purification by CC (SiO2; hexane/EtOAc 9:1) gave 46 (1.85 g, 78%) as a yellow solid. 1 Rf = 0.34 (hexane/EtOAc 4:1); H NMR (300 MHz, CDCl3): 4.28 (s, 2H, H-C(3)), 7.00 (d, J = 4.1, 1H, H-C(1)), 7.60 (d, J = 4.1, 1H, H-C(2)); 13C NMR (75 MHz,

CDCl3): 29.6, 128.0, 133.3, 139.4, 141.5, 183.7; IR (ATR): 1666, 1390, 1192, 1008, + 79 35 + 791, 524; HR-EI-MS: m/z (%): 237.8846 (5, [M] , calcd for C6H4 Br ClOS : + 35 + 237.8849), 144.9511 (51, [M–CH2Br] , calcd for C5H2 ClOS : 144.9515), 18.0373 (100).

Ethyl 4-(5-chloro-2-thienyl)-1,3-thiazole-2-carboxylate (47)

3 5 S S O Cl 4 N O 1 2 Compound 46 (500 mg, 2.01 mmol, 1 equiv.) and ethyl thioxamate (278 mg,

2.01 mmol, 95%, 1 equiv.) were heated to reflux in dry EtOH (5 mL) under N2 for

124 8. Experimental ______

5 h. The mixture was evaporated in vacuo, diluted with CH2Cl2, washed with sat. aq.

NaHCO3 solution and brine, dried over MgSO4, filtrated, and evaporated in vacuo.

Repeated purification by CC (SiO2; hexane/EtOAc 5:1; hexane/EtOAc 6:1) gave 47 (342 mg, 60%) as a yellow solid. 1 H NMR (300 MHz, CDCl3): 1.46 (t, J = 7.1, 3H, H-C(5)), 4.49 (q, J = 7.1, 2H, H-C(4)), 6.89 (d, J = 3.9, 1H, H-C(1)), 7.30 (d, J = 3.9, 1H, H-C(2)), 7.53 (s, 1H, 13 H-C(3)); C NMR (75 MHz, CDCl3): 14.2, 62.8, 117.3, 124.4, 126.9, 130.9, 135.6, 151.2, 158.4, 159.7; HR-EI-MS: m/z (%): 274.9647 (41, [M]+, calcd for 37 + + 35 + C10H8 ClNO2S2 : 274.9655), 272.9680 (100, [M] , calcd for C10H8 ClNO2S2 : + 37 + 272.9685), 143.9611 (30, [M–C4H5NO2S] , calcd for C6H3 ClS : 143.9614), + 35 + 141.9639 (81, [M–C4H5NO2S] , calcd for C6H3 ClS : 141.9644).

[4-(5-Chloro-2-thienyl)-1,3-thiazol-2-yl]methanol (48)

3 S S Cl 4 N OH 1 2 To a solution of compound 47 (342mg, 1.25 mmol, 1 equiv.) in EtOH (10 mL) at 0 °C

was added NaBH4 (47 mg, 1.25 mmol, 1 equiv.) in portions. The mixture stirred at RT for 24 h, and evaporated in vacuo. The brown solid was taken up in water and the

pH set to 3 with aq. 1 M HCl solution. The precipitate was collected and washed with water to give 48 (278 mg, 96%) as a brown solid. 1 H NMR (300 MHz, CDCl3): 2.71 (s, 1H, OH), 4.97 (s, 2H, H-C(4)), 6.87 (d, J = 3.9, 1H, H-C(1)), 7.18 (d, J = 3.9, 1H, H-C(2)), 7.25 (s, 1H, H-C(3)); 13C NMR (75 MHz,

CDCl3): 62.4, 111.7, 123.3, 127.0, 130.1, 136.8, 149.1, 171.4; IR (ATR): 3124, 1696, 1471, 1417, 1339, 1164, 1062, 967, 705; HR-EI-MS: m/z (%): 232.9544 (41, [M]+, 37 + + 35 + calcd for C8H6 ClNOS2 : 232.9550), 230.9574 (100, [M] , calcd for C8H6 ClNOS2 : + 35 + 230.9579), 201.9547 (45), 173.9355 (39, [M–C3H3NO] , calcd for C6H3 ClS2 : 173.9365), 138.9666 (44).

125 8. Experimental ______3-(5-Chlorothiophen-2-yl)propiolaldehyde (49) O

Cl S

1 2

Dess-Martin periodinane (15% in CH2Cl2, 9.78 mL, 4.71 mmol, 1.5 equiv.) was added to 52 (540 mg, 3.14 mmol, 1 equiv.) in anhydrous CH2Cl2 (20 mL), the mixture was stirred at RT overnight. The excess reactant was destroyed with MeOH, and the

mixture was diluted with water and extracted with EtOAc and CH2Cl2. The org.

phase was dried over MgSO4, filtrated, and evaporated in vacuo. Purification by CC

(SiO2; hexane/CH2Cl2 1:9) gave 49 (220 mg, 41%) as a dark brown solid. 1 Rf = 0.61 (hexane/EtOAc 1:1); m.p. 105 °C (decomp); H NMR (300 MHz, CDCl3): 6.92 (d, J = 4.1, 1H, H-C(1)), 7.33 (d, J = 3.7, 1H, H-C(2)), 9.38 (s, 1H, CHO); 13 C NMR (75 MHz, CDCl3): 87.9, 93.2, 127.3, 136.7, 137.2, 163.2, 175.6; IR (ATR): 2174, 1642, 1412, 1063, 928, 802, 691; HR-EI-MS: m/z (%): 171.9551 (40, [M]+, 37 + + 35 + calcd for C7H5 ClOS : 171.9564), 169.9582 (100, [M] , calcd for C7H5 ClOS : 169.9593), 143.9604 (31), 141.9635 (79), 106.9952 (23), 105.9876 (26), 63.0230 (29), 18.0454 (66).

3-(5-Chloro-2-thienyl)-2-propyn-1-ol (52) OH

Cl S

1 2 2-Bromo-5-chlorothiophene (50) (100 mg, 0.51 mmol, 1 equiv.), propargyl alcohol

(0.80 mL, 1.38 mmol, 2.7 equiv.), Et3N (0.19 mL, 1.38 mmol, 2.7 equiv.), [Pd(PPh3)4] (62 mg, 0.05 mmol, 0.1 equiv.), and CuI (5 mg, 0.03 mmol, 0.05 equiv.) were heated

to reflux in anhydrous THF (5 mL) under N2 for 6 h. The mixture was diluted with

aq. 2 M HCl solution and extracted with EtOAc. The org. phase was washed with aq.

5% NaHCO3 solution, dried over Na2SO4, filtrated, and evaporated in vacuo.

Purification by CC (SiO2; CH2Cl2) gave 52 (72 mg, 83%) as a brown oil. 1 Rf = 0.46 (hexane/EtOAc 1:1); H NMR (300 MHz, CDCl3): 1.77 (br. s, 1H, OH),

4.49 (s, 2H, CH2), 6.78 (d, J = 3.9, 1H, H-C(1)), 6.98 (d, J = 3.9, 1H, H-C(2)); 13 C NMR (75 MHz, CDCl3): 51.8, 78.4, 91.6, 121.3, 126.3, 131.0, 132.0; HR-EI-MS:

126 8. Experimental ______

+ 35 + m/z (%): 171.9743 (28, [M] , calcd for C7H5 ClOS : 171.9744), 137.0057 (28, [M– + + Cl] , calcd for C7H5OS : 137.0061), 109.0105 (33), 28.0216 (61), 18.0431 (100).

3-(5-Chloro-2-thienyl)benzaldehyde (53)

5 4 Cl S 6 1 2 3 7 O 2-Bromo-5-chlorothiophene (50) (278 L, 2.53 mmol, 1 equiv.),

3-formylphenylboronic acid (54) (456 mg, 3.04 mmol, 1.2 equiv.), and K2CO3

(700 mg, 5.06 mmol, 2 equiv.) were degassed in H2O (3 mL) and toluene (22 mL).

[(PPh3)2PdCl2] (89 mg, 0.13 mmol, 0.05 equiv.) was added and the mixture heated to

reflux for 96 h. The mixture was extracted with CH2Cl2. The org. phase was dried

over MgSO4, filtrated, and evaporated in vacuo. Purification by CC (SiO2; hexane/EtOAc 9:1) gave 53 (385 mg, 68%) as a yellow solid.

1 Rf = 0.45 (hexane/EtOAc 1:1); m.p. 59 °C; H NMR (300 MHz, CDCl3): 6.94 (d, J = 3.9, 1H, H-C(1)), 7.17 (d, J = 3.9, 1H, H-C(2)), 7.56 (t, J = 7.7, 1H, H-C(5)), 7.76 (ddd, J = 7.8, 2.0, 1.2, 1H, H-C(6)), 7.80 (dt, J = 7.6, 1.3, 1H, H-C(4)), 8.01 (t, J = 13 1.8, 1H, H-C(3)), 10.05 (s, 1H, H-C(7)); C NMR (75 MHz, CDCl3): 123.4, 126.1, 127.4, 129.1, 129.8, 131.2, 134.7, 137.0, 141.4, 191.8, 1 signal not visible; IR (ATR): 2848, 2747, 1683, 1672, 1447, 1255, 1152, 997, 866, 778, 653, 647; HR-EI-MS: m/z + 37 + + (%): 223.9873 (37, [M] , calcd for C11H7 ClOS : 223.9877), 221.9902 (100, [M] , 35 + calcd for C11H7 ClOS : 221.9901), 158.0174 (33).

7-Chloro-2-quinolinecarbaldehyde (55)[196]

1 6 N Cl O

2 5 3 4

To SeO2 (1.87 g, 16.89 mmol, 6 equiv.) in dioxane (75 mL) at 60 °C was added 7-chloro-2-methylquinoline (56) (0.50 g, 2.82 mmol, 1 equiv.). The mixture was heated to reflux for 30 min, filtered warm, evaporated to a small volume, and filtered again. The solid was suspended in CH2Cl2. The org. phase was washed with water,

127 8. Experimental ______

dried over MgSO4, filtrated, and evaporated in vacuo. Repeated purification by CC

(SiO2; CH2Cl2; hexane/CH2Cl2 1:3) gave 55 (0.34 g, 63%) as a white solid.

[197] 1 Rf = 0.60 (hexane/EtOAc 1:1); m.p. 160 °C (Lit. : 159 °C); H NMR (300 MHz,

CDCl3): 7.65 (dd, J = 8.8, 2.1, 1H, H-C(5)), 7.86 (d, J = 8.7, 1H, H-C(4)), 8.03 (d, J = 8.4, 1H, H-C(2)), 8.26 (d, J = 2.1, 1H, H-C(6)), 8.31 (d, J = 8.4, 1H, H-C(3)), 10.21 13 (d, J = 0.9, 1H, H-C(1)); C NMR (75 MHz, CDCl3): 117.5, 128.3, 129.0, 129.2, 130.1, 136.4, 137.2, 148.1, 153.1, 193.1; IR (ATR): 1693, 1434, 1236, 1063, 791.8, + 37 + 751.6; HR-EI-MS: m/z (%): 192.1063 (6, [M] , calcd for C10H6 ClNO : 192.1071), + 35 + 191.0133 (49, [M] , calcd for C10H6 ClNO : 191.0132), 163.0175 (100), 128.0486

(34); elemental analysis calcd (%) for C10H6ClNO (191.0): C 62.68, H 3.16, N 7.31; found: C 62.59, H 3.33, N 7.16.

(3aSR,4RS,8aSR,8bRS)-2-(3-Bromopropyl)-4-[5-(5-chloro-2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-57) O 8 9 H 11 8a N 8b 7 3a Br 10 N 6 O H 4

N 3' O

2' S 1' Cl A suspension of 36 (3.05 g, 14.0 mmol, 1 equiv.), L-proline (18) (1.70 g, 14.7 mmol, 1.1 equiv.), and 40 (3.14 g, 14.7 mmol, 1.1 equiv.) in MeCN (30 mL) was heated to reflux for 24 h, leading to a mixture of products, and concentrated in vacuo. Repeated

purification by CC (SiO2; CH2Cl2/Et2O 9:1; CH2Cl2/Et2O 95:5; pentane/EtOAc 1:1; pentane/EtOAc 1:3) followed by purification by HPLC (Merck LiChrospher® Si 60; 25025 mm, 5m; EtOAc) gave (±)-57 (267 mg, 4%) as a white solid. Enantiomers were separated by HPLC (105 mg) with a Chiralpak AD column (heptane/isopropanol 3:2; 35 mL min–1), giving (+)-6 (42 mg) and (–)-6 (40 mg) as white solids. 25 Rf = 0.28 (pentane/EtOAc 1:1); m.p. 142 °C; []D : +158.3 ((+)-57, c = 1, CHCl3); 25 1 []D : –163.1 ((–)-57, c = 1, CHCl3); H NMR (400 MHz, CDCl3): 1.68 (dtd, J =

128 8. Experimental ______12.6, 10.2, 7.9, 1H, H-C(7) or H-C(8)), 1.82 (dqd, J = 13.0, 8.6, 4.3, 1H, H-C(7) or H- C(8)), 2.01–2.23 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.75 (ddd, J = 12.6, 8.3, 4.4, 1H, H-C(6)), 2.90 (ddd, J = 12.6, 9.0, 7.3, 1H, H-C(6)), 3.31–3.38 (m, 3H, H-C(8b), H- C(11)), 3.60–3.64 (m, 3H, H-C(3a), H-(9)), 3.77 (ddd, J = 9.3, 7.6, 1.4, 1H, H-C(8a)), 4.27 (d, J = 8.4, 1H, H-C(4)), 6.26 (s, 1H, H-C(3’)), 6.93 (d, J = 4.2, 1H, H-C(1’)), 13 7.26 (d, J = 4.4, 3H, H-C(2’)); C NMR (100 MHz, CDCl3): 23.6, 29.6, 29.9, 30.9, 38.0, 49.4, 49.6, 50.8, 60.9, 68.4, 99.0, 126.5, 127.3, 128.0, 133.3, 163.6, 163.9, 175.1, 177.8; IR (ATR): 3119, 2958, 2889, 1770, 1699, 1613, 1480, 1433, 1397, 1322, 1236, 1155, 1077, 1031, 999, 919, 870, 821, 798, 738, 666; HR-MALDI-MS: + 81 35 + m/z (%): 486.0066 (100, [M+H] , calcd for C19H20 Br ClN3O3S : 486.0069), + 79 35 + 484.0086 (76, [M+H] , calcd for C19H20 Br ClN3O3S : 484.0092); elemental

analysis calcd (%) for C19H19BrClN3O3S (484.8): C 47.07, H 3.95, N 8.67; found: C 47.16, H 4.18, N 8.80.

(3aSR,4RS,8aSR,8bRS)-2-(3-Bromopropyl)-4-[(5-Chloro-2- thienyl)ethynyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-58) O 8 9 H 11 7 N 8b 8a Br 10 3a N 6 O H 4

S 2'

1' Cl A solution of 36 (1.73 g, 7.98 mmol, 1 equiv.), L-proline (18) (0.96 g, 8.38 mmol, 1.1 equiv.), and 49 (1.42 g, 8.38 mmol, 1.1 equiv.) was heated to reflux for 26 h, leading to a mixture of products, and concentrated in vacuo. Repeated purification by

CC (SiO2; hexane/EtOAc 2:1  EtOAc; CH2Cl2/EtOAc 95:5) gave (±)-58 (82 mg, 25%) as a brown solid. 1 Rf = 0.40 (CH2Cl2/EtOAc 4:1); m.p. 203 °C (decomp); H NMR (300 MHz, CDCl3): 1.60–2.26 (m, 6H, H-C(7), H-C(8), H-C(10)), 2.91–3.08 (m, 2H, H-C(6)), 3.21 (dd, J = 8.4, 2.4, 1H, H-C(8b)), 3.35 (t, J = 6.8, 2H, H-C(11)), 3.57 (dd, J = 8.3, 8.3, 1H, H-C(3a)), 3.67 (t, J = 6.9, 2H, H-C(9)), 3.77 (td, J = 7.8, 2.2, 1H, H-C(8a)), 4.09 (d, J

129 8. Experimental ______= 8.3, 1H, H-C(4)), 6.76 (d, J = 3.9, 1H, H-C(1’)), 6.99 (d, J = 3.9, 1H, H-C(2’)); 13 C NMR (75 MHz, CDCl3): 23.6, 29.9, 30.0, 31.0, 38.1, 49.3, 50.5, 51.3, 56.9, 67.8, 79.2, 89.2, 126.2, 131.0, 132.2, 140.7, 174.6, 177.5; IR (ATR): 1779, 1699, 1433, 1398, 1355, 1233, 1189, 1108, 1059, 1000, 799; HR-EI-MS: m/z (%): 441.9932 (13, + 81 37 + + [M] , calcd for C18H18 Br ClN2O2S : 441.9931), 439.9957 (10, [M] , calcd for 79 35 + C18H18 Br ClN2O2S : 439.9956), 223.0217 (41), 18.0464 (100).

(3aSR,4RS,8aSR,8bRS)-2-(3-Bromopropyl)-4-[3-(5-chloro-2- thienyl)phenyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-59) and (3aSR,4SR,8aRS,8bRS)-2-(3-Bromopropyl)-4-[3-(5-chloro-2- thienyl)phenyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-62) O O 8 8 9 H 9 H 11 8a 11 8a N 8b 7 N 8b 7 Br 10 3a N Br 10 3a N 6 6 O H 4 O H 4 3' 6' 3' 6'

S 5' S 5' Cl 4' Cl 4' 2' 2' 1' 1'

(±)-59 (±)-62 A suspension of 36 (1.06 g, 4.89 mmol, 1 equiv.), L-proline (18) (0.59 g, 5.13 mmol, 1.1 equiv.), and 53 (1.14 g, 5.13 mmol, 1.1 equiv.) in MeCN (50 mL) was heated to reflux for 24 h, leading to a mixture of products, and concentrated in vacuo. Repeated

purification by CC (SiO2; hexane/EtOAc 3:2; hexane/EtOAc 1:1; hexane/EtOAc 1:1;

CH2Cl2/Et2O 95:5;) gave (±)-59 (789 mg, 33%) as light brown foam and (±)-62 (700 mg, 29%) as a yellow solid. 1 (±)-59: Rf = 0.24 (pentane/EtOAc 1:1); H NMR (300 MHz, CDCl3): 1.66–1.85 (m, 2H, H-C(7), H-C(8)), 1.96–2.19 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.74 (ddd, J = 15.2, 6.7, 3.8, 1H, H-C(6)), 2.87–2.97 (m, 1H, H-C(6)), 3.23–3.31 (m, 3H, H-C(8b), H-C(11)), 3.47–3.56 (m, 3H, H-C(3a), H-C(9)), 3.80 (dd, J = 9.9, 7.4, 1H, H-C(8a)), 4.13 (d, J = 8.7, 1H, H-C(4)), 6.87 (d, J = 3.9, 1H, H-C(2’)), 7.07 (d, J = 3.9, 1H, H- C(1’)), 7.23 (partly hidden under solvent peak, 1H, H-C(6’)), 7.35 (t, J = 7.6, 1H, H- C(5’)), 7.44 (dt, J = 7.7, 1.4, 1H, H-C(4’)), 7.48 (br. s, 1H, H-C(3’)); 13C NMR

130 8. Experimental ______

(75 MHz, CDCl3): 23.4, 29.6, 30.9, 37.6, 49.1, 50.5, 50.9, 67.9, 68.6, 122.4, 124.9, 125.1, 127.1, 127.6, 128.9, 129.1, 133.6, 139.0, 142.9, 175.3, 178.2, 1 signal not visible; IR (ATR): 2953, 1696, 1233, 1065, 877, 785, 692; HR-MALDI-MS: m/z (%): + 81 37 + 497.0292 (29, [M+H] , calcd for C22H23 Br ClN2O2S : 497.0299), 495.0315 (100, + 81 35 + + [M+H] , calcd for C22H23 Br ClN2O2S : 495.0324), 493.0358 (52, [M+H] , calcd 79 37 + for C22H23 Br ClN2O2S : 493.0347). 1 (±)-62: Rf = 0.21 (pentane/EtOAc 1:1); m.p. 112 °C; H NMR (300 MHz, CDCl3): 1.57–2.10 (m, 4H, H-C(7), H-C(8)), 2.17 (quintet, J = 7.0, 2H, H-C(10)), 2.61–2.70 (m, 1H, H-C(6)), 3.03 (ddd, J = 11.6, 8.8, 5.6, 1H, H-C(6)), 3.34–3.41 (m, 3H, H- C(3a), H-C(11)), 3.58–3.69 (m, 3H, H-C(8b), H-C(9)), 3.93 (td, J = 8.8, 6.9, 1H, H- C(8a)), 4.11 (d, J = 6.2, 1H, H-C(4)), 6.89 (d, J = 3.9, 1H, H-C(2’)), 7.11 (d, J = 3.9, 1H, H-C(1’)), 7.33–7.44 (m, 3H, H-C(4’), H-C(5’), H-C(6’)), 7.66 (t, J = 1.7, 1H, H- 13 C(3’)); C NMR (75 MHz, CDCl3): 24.7, 26.8, 29.8, 30.8, 37.9, 47.9. 52.2, 55.5, 66.4, 69.5, 122.6, 124.2, 124.9, 126.6, 127.2, 129.4, 129.5, 134.2, 142.9, 143.1, 177.0, 178.0; IR (ATR): 2972, 2901, 1773, 1694, 1602, 1531, 1489, 1434, 1396, 1373, 1344, 1233, 1187, 1146, 1065, 897, 786, 730, 697, 659, 624; HR-MALDI-MS: m/z (%): + 81 37 + 497.0346 (33, [M+H] , calcd for C22H23 Br ClN2O2S : 497.0299), 495.0378 (100, + 81 35 + + [M+H] , calcd for C22H23 Br ClN2O2S : 495.0324), 493.0356 (52, [M+H] , calcd 79 37 + for C22H23 Br ClN2O2S : 493.0347).

(3aSR,4SR,8aRS,8bRS)-2-(3-Bromopropyl)-4-[5-(5-chloro-2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-60) O 8 9 H 11 8a N 8b 7 3a Br 10 N 6 O H 4

N 3' O

2' S 1' Cl A suspension of 36 (905 mg, 4.17 mmol, 1 equiv.), L-proline (18) (504 mg, 4.38 mmol, 1.1 equiv.), and 40 (933 mg, 4.38 mmol, 1.1 equiv.) in MeCN (15 mL) was heated to reflux for 22 h, leading to a mixture of products, and concentrated in

131 8. Experimental ______vacuo. Repeated purification by CC (SiO2; CH2Cl2/Et2O 95:5; CH2Cl2/Et2O 9:1; hexane/EtOAc 2:1) gave (±)-60 (207 mg, 10%) as a light brown solid. 1 Rf = 0.25 (pentane/EtOAc 1:1); m.p. 120 °C; H NMR (500 MHz, CDCl3): 1.75–1.86 (m, 1H, H-C(7)), 1.86–1.92 (m, 1H, H-C(7)), 1.92–2.01 (m, 1H, H-C(8)), 2.05–2.12 (m, 1H, H-C(8)), 2.16 (quint., J = 7.1, 2H, H-C(10)), 2.56–2.62 (m, 1H, H-C(6)), 3.14–3.19 (m, 1H, H-C(6)), 3.38 (t, J = 6.7, 2H, H-C(11)), 3.50 (dd, J = 9.0, 9.0, 1H, H-C(8b)), 3.64 (td, J = 7.2, 0.9, 2H, H-C(9)), 3.82–3.92 (m, 2H, H-C(3a), H-C(8a)), 4.58 (d, J = 3.3, 1H, H-C(4)), 6.42 (s, 1H, H-C(3’)), 6.94 (d, J = 4.0, 1H, H-C(1’)), 13 7.27 (d, J = 3.9, 1H, H-C(2’)); C NMR (125 MHz, CDCl3): 24.7, 25.8, 29.6, 30.3, 37.9, 48.5, 52.2, 52.7, 62.5, 66.4, 98.3, 126.4, 127.3, 127.5, 133.4, 164.4, 165.1, 177.1, 177.9; IR (ATR): 3059, 2919, 2872, 1771, 1693, 1601, 1531, 1466, 1432, 1399, 1373, 1344, 1323, 1237, 1200, 1162, 1145, 1111, 1060, 1031, 1007, 963, 930, 897, 829, 809, 786, 730, 698, 662, 637, 624; HR-MALDI-MS: m/z (%): 488.0034 (30, + 81 37 + + [M+H] , calcd for C19H20 Br ClN3O3S : 488.0043), 486.0062 (4, [M+H] , calcd for 81 35 + C19H20 Br ClN3O3S : 486.0069); elemental analysis calcd (%) for

C19H19BrClN3O3S (483.0): C 47.07, H 3.95, N 8.67; found: C 47.31, H 4.09, N 8.50.

(3aSR,4SR,8aRS,8bRS)-2-(3-Bromopropyl)-4-[4-(5-chloro-2-thienyl)-1,3-thiazol- 2-yl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-61) O 8 9 H 11 8a N 8b 7 3a Br 10 N 6 O H 4 N S

S 3' Cl 2' 1' A mixture of 36 (99 mg, 0.45 mmol, 1 equiv.), L-proline (18) (55 mg, 0.48 mmol, 1.05 equiv.), and 45 (110 mg, 0.48 mmol, 1.05 equiv.) was heated to reflux in MeCN

(5 mL) for 21 h. The mixture was evaporated in vacuo. Purification by CC (SiO2; hexane/EtOAc 2:1) gave (±)-61 (30 mg, 13%) a light brown solid. 1 Rf = 0.63 (pentane/EtOAc 1:1); H NMR (300 MHz, CDCl3): 1.65–1.77 (m, 1H, H-C(7)), 1.84–2.00 (m, 1H, H-C(7)), 2.02–2.20 (m, 4H, H-C(8), H-C(10)), 2.56 (dt, J = 9.9, 8.1, 1H, H-C(6)), 3.27–3.33 (m, 1H, H-C(6)), 3.35–3.41 (m, 3H, H-C(8b), H-

132 8. Experimental ______C(11)), 3.65 (t, J = 6.9, 2H, H-C(9)), 3.95 (ddd, J = 9.6, 7.2, 4.5, 1H, H-C(8a)), 4.20 (dd, J = 8.7, 2.1, 1H, H-C(3a)), 4.86 (d, J = 1.8, 1H, H-C(4)), 6.86 (d, J = 3.9, 1H, H- C(1’)), 7.17 (d, J = 3.9, 1H, H-C(2’)), 7.22 (s, 1H, H-C(3’)); 13C NMR (75 MHz,

CDCl3): 25.2, 26.5, 29.7, 30.5, 38.0, 49.1, 53.7, 54.0, 66.7, 68.1, 112.6, 122.8, 126.7, 129.7, 136.9, 149.2, 174.4, 177.0, 177.9; IR (ATR): 2922, 2853, 1771, 1696, 1612, 1441, 1397, 1343, 1235, 1154, 1069, 1031, 995, 918, 870, 797, 738, 653; HR- + 81 37 + MALDI-MS: m/z (%): 503.9812 (33, [M+H] , calcd for C19H20 Br ClN3O2S2 : + 81 35 + 503.9813), 501.9845 (100, [M+H] , calcd for C19H20 Br ClN3O2S2 : 501.9840), + 79 35 + 499.9877 (47, [M+H] , calcd for C19H20 Br ClN3O2S2 : 499.9863).

(3aSR,4SR,8aRS,8bRS)-2-(3-Bromopropyl)-4-(7-chloro-2- quinolinyl)hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-63) O 8 9 H 11 8a N 8b 7 Br 10 3a N 6 O H 4 N 5'

1' 4'

Cl 3' 2' A suspension of 36 (389 mg, 1.79 mmol, 1 equiv.), L-proline (18) (217 mg, 1.88 mmol, 1.1 equiv.), and 55[196] (480 mg, 1.88 mmol, 1.1 equiv.) in MeCN (10 mL) was heated to reflux for 24 h, leading to a mixture of products, and concentrated in

vacuo. Repeated purification by CC (SiO2; CH2Cl2; CH2Cl2/EtOAc 95:5; hexane/EtOAc 1:1) gave (±)-63 (224 mg, 27%) as a light brown solid. 1 Rf = 0.20 (pentane/EtOAc 1:1); m.p. 180 °C (decomp); H NMR (400 MHz, CDCl3): 1.74–1.84 (m, 1H, H-C(7)), 1.86–1.95 (m, 1H, H-C(7)), 2.04–2.10 (m, 2H, H-C(8)), 2.15–2.21 (m, 2H, H-C(10)), 2.60–2.66 (m, 1H, H-C(6)), 3.21–3.29 (m, 1H, H-C(6)), 3.40 (t, J = 6.7, 2H, H-C(11)), 3.52 (dd, J = 9.0, 9.0, 1H, H-C(8b)), 3.66 (t, J = 7.3, 2H, H-C(9)), 3.85–3.90 (m, 1H, H-C(8a)), 4.39 (dd, J = 8.6, 2.9, 1H, H-C(3a)), 4.77 (d, J = 2.9, 1H, H-C(4)), 7.48 (dd, J = 8.7, 2.1, 1H, H-C(2’)), 7.66 (d, J = 8.5, 1H, H- C(5’)), 7.74 (d, J = 8.7, 1H, H-C(3’)), 8.06 (d, J = 2.1, 1H, H-C(1’)), 8.13 (d, J = 8.5, 13 1H, H-C(4’)); C NMR (100 MHz, CDCl3): 24.8, 26.0, 29.7, 30.5, 37.9, 49.3, 52.0, 53.7, 66.4, 71.6, 120.5, 125.8, 127.5, 128.5, 128.7, 135.3, 136.7, 147.8, 161.5, 177.8,

133 8. Experimental ______179.4; IR (ATR): 2956, 1690, 1601, 1364, 1357, 1175, 1075, 1066, 946, 834, 813; + 81 37 + HR-EI-MS: m/z (%): 460.0423 (8, [M–H] , calcd for C21H20 Br ClN3O2 : 460.0427), 287.0220 (39), 177.0343 (100), 108.0749 (23).

(3aSR,4RS,8aSR,8bRS)-4-[(5-Chloro-2-thienyl)ethynyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-64) O H 8 12 9 11 8b 7 12 N 8a N 10 3a N 6 O H 4 12 Br

S 2'

1' Cl

A solution of (±)-58 (30 mg, 0.07 mmol, 1 equiv.) in 4.2 M Me3N in EtOH (1 mL, 4.20 mmol, 62 equiv.) was heated to reflux for 48 h. The mixture was evaporated in

vacuo. The crude product was dissolved in MeOH and precipitated from Et2O to give (±)-64 (23 mg, 68%) as a brown solid. 1 H NMR (300 MHz, CD3OD): 1.66–1.93 (m, 2H, H-C(7), H-C(8)), 2.04–2.18 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.92–3.01 (m, 1H, H-C(6)), 3.07 (s, 9H, H-C(12)), 3.10– 3.19 (m, 1H, H-C(6)), 3.24–3.34 (m, 3H, H-C(8b), H-C(11)), 3.59–3.64 (m, 3H, H- C(8a), H-C(9)), 3.69 (dd, J = 8.3, 8.3, 1H, H-C(3a)), 4.17 (d, J = 8.5, 1H, H-C(4)), 6.91 (d, J = 4.0, 1H, H-C(1’)), 7.07 (d, J = 3.9, 1H, H-C(2’)); 13C NMR (75 MHz,

CD3OD): 22.8, 23.9, 30.2, 36.9, 50.1, 50.7, 51.9, 53.5 (t), 57.3, 69.1, 90.8, 122.5, 127.8, 128.7, 133.3, 176.7, 179.4, 2 signals not visible; IR (ATR): 1700, 1405, 660; + 37 + HR-MALDI-MS: m/z (%): 422.1471 (33, [M–Br] , calcd for C21H27 ClN3O2S : + 35 + 422.1478), 420.1503 (100, [M–Br] , calcd for C21H27 ClN3O2S : 420.1507).

134 8. Experimental ______3-[(3aSR,4SR,8aRS,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-65) O 8 9 H 12 11 8a 8b 7 12 N N 10 3a N 6 12 O H 4 Br N 3' O

2' S 1' Cl

4.2 M Me3N in EtOH (0.98 mL, 4.12 mmol, 20 equiv.) was added to (±)-60 (100 mg, 0.21 mmol, 1 equiv.) in EtOH (2.5 mL) and the mixture stirred at RT for 120 h. The mixture was evaporated in vacuo. The residue was dissolved in MeOH. Precipitation

from Et2O gave (±)-65 (98 mg, 87%) as a light pink solid. 1 M.p. 204 °C (decomp); H NMR (300 MHz, CD3OD): 1.79–2.15 (m, 6H, H-C(7), H- C(8), H-C(10)), 2.68 (dt, J = 11.2, 7.1, 1H, H-C(6)), 3.02–3.14 (m, 10H, H-C(6), H- C(12)), 3.38–3.44 (m, 2H, H-C(11)), 3.60 (t, J = 6.7, 2H, H-C(9)), 3.70 (dd, J = 9.0, 9.0, 1H, H-C(8b)), 3.83–3.90 (m, 2H, H-C(3a), H-C(8a)), 4.45 (d, J = 4.7, 1H, H- C(4)), 6.76 (s, 1H, H-C(3’)), 7.10 (d, J = 4.0, 1H, H-C(1’)), 7.44 (d, J = 4.0, 1H, H- 13 C(2’)); C NMR (75 MHz, CD3OD): 22.4, 25.2, 26.8, 36.4, 52.7, 53.3, 53.5, 62.8, 64.9, 67.2, 99.2, 127.7, 128.4, 128.5, 131.3, 133.7, 165.1, 166.4, 178.1, 178.7; IR (ATR): 2962, 1691, 1421, 1181, 1062, 784, 648; HR-MALDI-MS: m/z (%): 465.1535 + 37 + + (38, [M–Br] , calcd for C22H28 ClN4O3S : 465.1537), 463.1568 (100, [M–Br] , calcd 35 + + for C22H28 ClN4O3S : 463.1565), 404.0822 (69, [M–N(CH3)3] , calcd for 35 + C19H19 ClN3O3S : 404.0836), 335.0247 (50); elemental analysis calcd (%) for

C22H28BrClN4O3S (463.2): C 48.58, H 5.19, N 10.30; found: C 48.42, H 5.31, N 9.92.

135 8. Experimental ______3-[(3aSR,4SR,8aRS,8bRS)-4-[4-(5-Chloro-2-thienyl)-1,3-thiazol-2-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)yl-N,N,N-trimethyl-1- propanaminium Bromide ((±)-66)

O H 8 12 9 11 8a N 8b 7 12 3a N 10 N 6 O 4 12 H N S

S 3' Cl 2' 1'

A solution of (±)-61 (30 mg, 0.06 mmol, 1 equiv.) in 4.2 M Me3N in EtOH (285 L, 1.20 mmol, 20 equiv.) in EtOH (1 mL) was stirred at RT for 96 h. The mixture was evaporated in vacuo. The crude product was dissolved in MeOH and precipitated from Et2O to give (±)-66 (35 mg, quant.) as a light brown solid. 1 M.p. 188 °C (decomp); H NMR (400 MHz, CD3OD): 1.70–1.80 (m, 1H, H-C(7)), 1.83–1.94 (m, 1H, H-C(7)), 1.97–2.15 (m, 4H, H-C(8), H-C(10)), 2.63 (dt, J = 10.4, 7.7, 1H, H-C(6)), 3.14 (s, 9H, H-C(12)), 3.21–3.29 (m, 1H, H-C(6)), 3.42 (m, 2H, H-C(11)), 3.53 (dd, J = 9.0, 9.0, 1H, H-C(8b)), 3.59–3.63 (m, 2H, H-C(9)), 3.95–4.00 (m, 1H, H-C(8a)), 4.16 (dd, J = 8.6, 2.8, 1H, H-C(3a)), 4.80 (d, J = 2.8, 1H, H-C(4)), 6.94 (d, J = 3.9, 1H, H-C(1’)), 7.29 (d, J = 3.9, 1H, H-C(2’)), 7.60 (s, 1H, H-C(3’)); 13 C NMR (100 MHz, CD3OD): 22.7, 26.0, 27.5, 36.8, 50.3, 53.7 (t), 54.4, 55.8, 65.4 (t), 67.8, 69.2, 114.5, 124.3, 128.3, 130.5, 138.7, 150.6, 176.6, 179.0, 179.9; + 37 + HR-MALDI-MS: m/z (%): 481.1304 (100, [M–Br] , calcd for C22H28 ClN4O2S2 : + 35 + 481.1307), 479.1336 (100, [M–Br] , calcd for C22H28 ClN4O2S2 : 479.1337).

136 8. Experimental ______3-[(3aSR,4SR,8aRS,8bRS)-4-[3-(5-Chloro-2-thienyl)phenyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-67) O 8 12 9 H 11 8a 8b 7 12 N 3a N 10 N 6 O 4 Br 12 H 3' 6'

S 5' Cl 4' 2' 1'

4.2 M Me3N in EtOH (1.12 mL, 4.70 mmol, 20 equiv.) was added to (±)-62 (116 mg, 0.24 mmol, 1 equiv.) in EtOH (3 mL) and the mixture stirred at RT for 144 h. The mixture was evaporated in vacuo. The residue was dissolved in MeOH. Precipitation

from Et2O gave (±)-67 (76 mg, 59%) as a yellow solid. 1 M.p. 219 °C (decomp); H NMR (500 MHz, CD3OD): 1.63–1.71 (m, 1H, H-C(8)), 1.79–1.88 (m, 1H, H-C(7)), 1.95–2.17 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.68–2.74 (m, 1H, H-C(6)), 2.93–2.99 (m, 1H, H-C(6)), 3.14 (s, 9H, H-C(12)), 3.42 (t, J = 8.7, 2H, H-C(11)), 3.52 (dd, J = 9.0, 7.3, 1H, H-C(3a)), 3.62 (t, J = 6.8, 2H, H-C(9)), 3.81 (dd, J = 9.1, 9.1, 1H, H-C(8b)), 3.88–3.93 (m, 1H, H-C(8a)), 4.08 (d, J = 7.2, 1H, H- C(4)), 6.97 (d, J = 3.9, 1H, H-C(2’)), 7.24 (d, J = 3.9, 1H, H-C(1’)), 7.40 (t, J = 7.6, 1H, H-C(5’)), 7.44–7.46 (m, 1H, H-C(6’)), 7.50 (dt, J = 7.7, 1.4, 1H, H-C(4’)), 7.71 13 (t, J = 1.8, 1H, H-C(3’)); C NMR (125 MHz, CD3OD): 22.9, 25.3, 27.8, 36.7, 52.4, 53.7 (t), 56.5, 65.4 (t), 67.4, 70.4, 124.0, 125.6, 125.8, 128.2, 128.7, 130.0, 130.5, 135.3, 144.1, 144.3, 178.7, 179.5, 1 signal not visible; IR (ATR): 2148, 1686, 1419, 1173, 924, 783; HR-MALDI-MS: m/z (%): 474.1776 (38, [M–Br]+, calcd for 37 + + 35 + C25H31 ClN3O2S : 474.1793), 472.1817 (100, [M–Br] , calcd for C25H31 ClN3O2S :

472.1820); elemental analysis calcd (%) for C25H31BrClN3O2S (551.1): C 54.30, H 5.65, N 7.60; found: C 54.02, H 5.76, N 7.39.

137 8. Experimental ______3-[(3aSR,4SR,8aRS,8bRS)-4-(7-Chloro-2-quinolinyl)-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-68) O 8 12 9 H 11 8a 8b 7 12 N 3a N 10 N 6 O 4 12 Br H N 5'

1' 4'

Cl 3' 2'

A solution of 4.2 M Me3N in EtOH (1.0 mL, 4.20 mmol, 98 equiv.) and (±)-63 (20 mg, 0.04 mmol, 1 equiv.) were heated to reflux for 24 h and concentrated in

vacuo. The residue was dissolved in MeOH. Precipitation from Et2O gave (±)-68 (9 mg, 40%) as a light brown solid. 1 M.p. 161 °C (decomp); H NMR (300 MHz, CD3OD): 1.85–2.17 (m, 6H, H-C(7), H- C(8), H-C(10)), 2.63–2.72 (m, 1H, H-C(6)), 3.07–3.12 (m, 1H, H-C(6)), 3.14 (s, 9H, H-C(12)), 3.39–3.49 (m, 2H, H-C(11)), 3.62 (t, J = 6.9, 2H, H-C(9)), 3.73 (dd, J = 9.0, 9.0, 1H, H-C(8b)), 3.90–3.98 (m, 1H, H-C(8a)), 4.20 (dd, J = 8.8, 4.5, 1H, H- C(3a)), 4.61 (d, J = 4.6, 1H, H-C(4)), 7.57 (dd, J = 8.7, 2.1, 1H, H-C(2’)), 7.77 (d, J = 8.5, 1H, H-C(5’)), 7.93 (d, J = 8.8, 1H, H-C(3’)), 8.03 (d, J = 1.7, 1H, H-C(1’)), 8.35 13 (d, J = 8.6, 1H, H-C(4’)); C NMR (75 MHz, CD3OD): 22.5, 25.2, 27.0, 36.4, 53.3 (t), 53.5, 53.9, 65.0 (t), 67.3, 72.2, 121.4, 127.1, 128.12, 128.15, 130.1, 136.2, 138.1, 148.4, 162.9, 178.6, 179.7, 1 signal not visible; IR (ATR): 2943, 2863, 1772, 1699, 1614, 1599, 1497, 1438, 1394, 1360, 1317, 1291, 1244, 1179, 1144, 1108, 1067, 940, 870, 843, 773; HR-MALDI-MS: m/z (%): 443.2027 (33, [M–Br]+, calcd for 37 + + 35 + C24H30 ClN4O2 : 443.2025), 441.2059 (100, [M–Br] , calcd for C24H30 ClN4O2 : 441.2052.

138 8. Experimental ______1-(4,4-Dimethylpentyl)-1H-pyrrole-2,5-dione (72)[89] O 2 4 5 N 1 3 5 1 O 5 Compound 76 (594 mg, 4.51 mmol, 1 equiv.) was added to a solution of maleic

anhydride (442 mg, 4.51 mmol, 1 equiv.) in CH2Cl2 (9 mL) at 0 °C. The mixture was stirred at RT for 2 h. The mixture was cooled down to 0 °C, DMF (4 L) and oxalyl chloride (423 L, 4.96 mmol, 1.1 equiv.) were added dropwise, and the mixture was stirred at RT for 48 h. The mixture was concentrated in vacuo, and the oily residue

was dissolved in CH2Cl2 (5 mL). Et3N (689 L, 4.96 mmol, 1.1 equiv.) was added dropwise and the mixture stirred at RT for 24 h. The mixture was diluted with

CH2Cl2, washed with aq. 1 M HCl solution, dried over MgSO4, filtrated, and

concentrated in vacuo. Repeated purification by CC (SiO2; hexane/EtOAc 1:1;

CH2Cl2) gave 72 (597 mg, 68%) as a white solid. [89] 1 Rf = 0.42 (CH2Cl2); m.p. 66 °C (Lit. : 62 °C); H NMR (300 MHz, CDCl3): 0.85 (s, 9H, H-C(5)), 1.12–1.18 (m, 2H, H-C(4)), 1.49–1.60 (m, 2H, H-C(3)), 3.48 (t, J = 7.5, 13 2H, H-C(2)), 6.68 (s, 2H, H-C(1)); C NMR (75 MHz, CDCl3): 23.9, 29.3, 30.1, 38.6, 40.8, 134.0, 170.9; IR (ATR): 2946, 1696, 1364, 1131, 834, 692; HR-EI-MS: + + m/z (%): 195.1255 (9, [M] , calcd for C11H17NO2 : 195.1254), 110.0238 (16, [M– + + + (CH3)3CCH2CH2] , calcd for C5H4NO2 : 110.0242), 57.0691 (26, [C(CH3)] , calcd for + C4H9 : 57.0704), 18.0147 (100), 17.0082 (29).

3,3-Dimethylbutyl 4-Methylbenzenesulfonate (74)[89]

1 1

1 3 O 4 5 2 O S 6 O 4 5 3,3-Dimethylbutanol (73) (4.74 mL, 39.2 mmol, 1 equiv.) and p-toluenesulfonyl

chloride (7.09 g, 37.2 mmol, 0.95 equiv.) were dissolved in dry CHCl3 (17 mL) and cooled to 0 °C. Pyridine (5.38 mL, 66.6 mmol, 1.7 equiv.) was added dropwise keeping the temperature under 5 °C and stirring continued at 0 °C for 5 h. The mixture was poured onto ice (32 g) and conc. HCl (10.3 mL). The aq. phase was extracted with Et2O. The org. phase was washed with brine, dried over MgSO4,

139 8. Experimental ______

filtrated, and evaporated in vacuo. Purification by CC (SiO; hexane/CH2Cl21:1) gave 74 (8.40 g, 84%) as a colorless oil. 1 H NMR (300 MHz, CDCl3): 0.87 (s, 9H, H-C(1)), 1.58 (t, J = 7.2, 2H, H-C(2)), 2.45 (s, 3H, H-C(6)), 4.09 (t, J = 7.5, 2H, H-C(3)), 7.35 (d, J = 7.8, 2H, H-C(5)), 7.79 (d, J = 8.4, 2H, H-C(4)); IR (ATR): 3352, 2956, 1697, 1601, 1357, 1175, 1075, 946, 834, + + 813; HR-EI-MS: m/z (%): 256.1129 (5, [M] , calcd for C13H20O3S : 256.1128), + + 173.0255 (24), 172.0193 (90), 91.0540 (46, [C7H7] , calcd for C7H7 : 91.0548), + + 69.0695 (100), 57.0713 (30, [C(CH3)3] , calcd for C4H9 : 57.0704), 56.0639 (29),

41.0454 (25); elemental analysis calcd (%) for C13H20O3S (256.1): C 60.91, H 7.86; found: C 60.97, H 7.76.

4,4-Dimethylpentanenitrile (75)[89]

N 2 3

1 3 3

NaCN (1.98 g, 40.40 mmol, 1.2 equiv) in Me2SO (20 mL) was heated to 90 °C. The oil bath was removed and 74 was added dropwise. The mixture was stirred at 85 °C for 3 h, cooled to RT, and treated with water (43 mL). The aq. phase was extracted

with Et2O. The org. phase was washed with brine, dried over MgSO4, filtrated, and evaporated in vacuo. The crude product was purified by distillation to give 75 (3.74 g, quant.) as colorless oil. 1 H NMR (300 MHz, CDCl3): 0.93 (s, 9H, H-C(3)), 1.62 (t, J = 8.4, 2H, H-C(2)), 2.28 13 (t, J = 8.4, 2H, H-C(1)); C NMR (75 MHz, CDCl3): 12.8, 28.7, 30.4, 39.2, 120.5.

4,4-Dimethylpentylamine (76)[89]

1 3 4 H N 2 2 4 4

Compound 75 (3.74 g, 33.7 mmol, 1 equiv.) in Et2O (12 mL) was added dropwise to a

solution of LiAlH4 (1.92 g, 50.5 mmol, 1.5 equiv.) in Et2O (35 mL) under N2 at RT.

The mixture was heated to reflux for 3.5 h. Excess LiAlH4 was destroyed with sat. aq.

NH4Cl solution. Et2O was added to the suspension and decanted. The org. phase was dried over MgSO4, filtrated, and evaporated in vacuo. Purification by distillation gave 76 (1.84 g, 47%) as a colorless oil.

140 8. Experimental ______

1 H NMR (300 MHz, CDCl3): 0.87 (s, 9H, H-C(4)), 1.13 (s, 2H, NH2), 1.14–1.20 (m, 2H, H-C(3)), 1.35–1.45 (m, 2H, H-C(2)), 2.65 (t, J = 7.2, 2H, H-C(1)); 13C NMR

(75 MHz, CDCl3): 29.1, 29.5, 30.2, 41.3, 43.2; HR-EI-MS: m/z (%): 115.1356 (7, + + + [M] , calcd for C7H17N : 115.1356), 83.0843 (21), 30.0340 (100, [CH2NH2] , calcd + for CH2NH2 : 30.0344).

(2Z)-4-[(4,4-Dimethylpentyl)amino]-4-oxo-2-butenoic Acid (77)[89] O OH 5 6 HN 3 1 4 6 2 O 6 Compound 35 (500 mg, 4.34 mmol, 1 equiv.) was added dropwise to a solution of maleic anhydride (425 mg, 4.34 mmol, 1 equiv.) in CH2Cl2 (9 mL) at 0 °C and stirred for 1 h. The mixture was evaporated in vacuo to give 77 (923 mg, quant.) as a white solid. 1 M.p. 106 °C; H NMR (300 MHz, CD3OD): 0.90 (s, 9H, H-C(6)), 1.21–1.27 (m, 2H, H-C(5)), 1.50–1.60 (m, 2H, H-C(4)), 3.26 (t, J = 6.9, 2H, H-C(3)), 6.24 (d, J = 12.6, 13 1H, H-C(1)), 6.43 (d, J = 12.6, 1H, H-C(2)); C NMR (75 MHz, CD3OD): 25.2, 29.7, 30.9, 41.8, 42.3, 133.9, 134.0, 167.6, 168.1; IR (ATR): 3295, 2950, 1705, 1495, 1364, 1214, 1049, 861, 748, 630; HR-EI-MS: m/z (%): 213.1359 (2, [M]+, calcd for + + + C11H19NO3 : 213.1359), 156.0653 (23, [M–C(CH3)] , calcd for C7H10NO3 : 156.0661), 129.0419 (45), 99.0110 (44), 83.0858 (39), 30.0313 (100); elemental analysis calcd (%) for C7H17N (213.1): C 61.95, H 8.98, N 6.57; found: C 61.65, H 9.17, N 6.66.

141 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-4-[(5-Chloro-2-thienyl)ethynyl]-2-(4,4- dimethylpentyl)hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-78), (3aSR,4SR,8aRS,8bRS)-4-[(5-Chloro-2-thienyl)ethynyl]-2-(4,4- dimethylpentyl)hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-79), and (3aSR,4SR,8aSR,8bRS)-4-[(5-Chloro-2-thienyl)ethynyl]-2-(4,4- dimethylpentyl)hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-80) O O O H 8 H 8 H 8 12 9 12 9 12 9 11 8a 7 11 8a 7 11 8a 7 N 8b N 8b N 8b 12 3a 12 12 3a 10 N 10 3a N 10 N 6 6 6 O 4 O 4 O 4 12 H 12 H 12 H

S 2' S 2' S 2'

Cl 1' Cl 1' Cl 1' (±)-78 (±)-79 (±)-80 A suspension of 72 (150 mg, 0.77 mmol, 1 equiv.), L-proline (18) (93 mg, 0.80 mmol, 1.1 equiv.), and 49 (137 mg, 0.80 mmol, 1.1 equiv.) in MeCN (15 mL) was heated to reflux for 16 h, leading to a mixture of products, and concentrated in vacuo. Repeated

purification by CC (SiO2; hexane/EtOAc 4:1) gave (±)-78 (35 mg, 11%), (±)-79 (46 mg, 14%), and (±)-80 (18 mg, 6%) as a yellow solid. 1 (±)-78: Rf = 0.35 (pentane/EtOAc 1:1); m.p. 108 °C; H NMR (400 MHz, CDCl3): 0.82 (s, 9H, H-C(12)), 1.14–1.19 (m, 2H, H-C(11)), 1.48–1.54 (m, 2H, H-C(10)), 1.62–1.71 (m, 1H, H-C(8)), 1.81–1.92 (m, 1H, H-C(7)), 1.97–2.08 (m, 1H, H-C(7)), 2.16–2.24 (m, 1H, H-C(8)), 2.92–3.06 (m, 2H, H-C(6)), 3.18 (dd, J = 8.2, 2.6, 1H, H- C(8b)), 3.46 (t, J = 7.6, 2H, H-C(9)), 3.56 (dd, J = 8.3, 8.3, 1H, H-C(3a)), 3.78 (td, J = 7.6, 2.4, 1H, H-C(8a)), 4.09 (d, J = 8.2, 1H, H-C(4)), 6.74 (d, J = 3.9, 1H, H-C(1’)), 13 6.96 (d, J = 3.9, 1H, H-C(2’)); C NMR (100 MHz, CDCl3): 23.1, 23.5, 29.3, 29.9, 30.1, 40.0, 40.9, 49.2, 50.4, 51.2, 56.8, 67.6, 78.9, 89.3, 121.3, 126.1, 130.8, 132.0, 174.8, 177.7; IR (ATR): 2955, 1773, 1696, 1528, 1435, 1403, 1358, 1342, 1295, 1216, 1194, 1160, 1092, 1060, 1012, 951, 879, 808, 685; HR-MALDI-MS: m/z (%): + 35 + 421.1529 (35), 419.1546 (100, [M+H] , calcd for C22H28 ClN2O2S : 419.1555). 1 (±)-79: Rf = 0.50 (pentane/EtOAc 1:1); H NMR (400 MHz, CDCl3): 0.85 (s, 9H, H-C(12)), 1.13–1.18 (m, 2H, H-C(11)), 1.47–1.56 (m, 2H, H-C(10)), 1.81–2.04 (m, 4H, H-C(7), H-C(8)), 2.57–2.64 (m, 1H, H-C(6)), 2.85–2.92 (m, 1H, H-C(6)), 3.36

142 8. Experimental ______(dd, J = 8.4, 8.4, 1H, H-C(8b)), 3.44 (td, J = 7.0, 1.3, 2H, H-C(9)), 3.59 (dd, J = 8.3, 2.5, 1H, H-C(3a)), 3.74 (q, J = 8.4, 6.6, 1H, H-C(8a)), 4.28 (d, J = 2.5, 1H, H-C(4)), 6.77 (d, J = 3.9, 1H, H-C(1’)), 6.96 (dd, J = 3.9, 0.2, 1H, H-C(2’)); 13C NMR (100

MHz, CDCl3): 22.7, 24.6, 25.2, 29.3, 30.1, 39.9, 40.9, 46.3, 48.9, 55.6, 55.7, 66.9, 77.2, 91.2, 121.3, 126.2, 130.7, 131.9, 176.7, 177.3; IR (ATR): 2953, 2867, 1774, 1697, 1603, 1528, 1435, 1397, 1362, 1349, 1293, 1273, 1234, 1187, 1152, 1115, 1066, 1019, 998, 7790, 735, 697, 651; HR-MALDI-MS: m/z (%): 421.1527 (37), + 35 + 419.1549 (100, [M+H] , calcd for C22H28 ClN2O2S : 419.1555). 1 (±)-80: Rf = 0.53 (pentane/EtOAc 1:1); H NMR (400 MHz, CDCl3): 0.85 (s, 9H, H-C(12)), 1.13–1.17 (m, 2H, H-C(11)), 1.50–1.56 (m, 2H, H-C(10)), 1.75–1.91 (m, 2H, H-C(7), H-C(8)), 2.07–2.19 (m, 2H, H-C(7), H-C(8)), 2.95–3.02 (m, 1H, H-C(6)), 3.11–3.17 (m, 1H, H-C(6)), 3.29 (dd, J = 8.2, 2.7, 1H, H-C(8b)), 3.46 (t, J = 7.4, 2H, H-C(9)), 3.52–3.59 (m, 2H, H-C(3a), H-C(8a)), 4.45 (d, J = 2.4, 1H, H-C(4)), 6.78 (d, 13 J = 3.9, 1H, H-C(1’)), 6.94 (d, J = 3.9, 1H, H-C(2’)); C NMR (100 MHz, CDCl3): 23.0, 25.9, 29.3, 30.1, 31.9, 39.8, 40.8, 48.8, 48.9, 53.1, 57.0, 69.9, 78.2, 92.9, 121.3, 126.3, 130.9, 131.5, 176.7, 177.9; IR (ATR): 2953, 1778, 1703, 1549, 1399, 1363, 1218, 1153, 1067, 1001, 857, 813, 736, 676; HR-MALDI-MS: m/z (%): 421.1523 + 35 + (36), 419.1552 (100, [M+H] , calcd for C22H28 ClN2O2S : 419.1555).

143 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-2-(4,4- dimethylpentyl)hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-81) and (3aSR,4SR,8aRS,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-2- (4,4-dimethylpentyl)hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-82) O O 8 8 12 9 H 12 9 H 11 7 11 7 12 N 8b 8a 12 N 8b 8a 3a 3a 10 N 10 N 6 6 O 4 O 4 12 H 12 H N 3' N 3' O O

2' 2' S S 1' 1' Cl Cl (±)-81 (±)-82 Compound 72 (500 mg, 2.56 mmol, 1 equiv.), L-proline (18) (310 mg, 2.69 mmol, 1.1 equiv.), and 40 (573 mg, 2.69 mmol, 1 equiv.) in MeCN (6 mL) were heated to reflux for 24 h, leading to a mixture of products, and concentrated in vacuo. Repeated purification by CC (SiO2; CH2Cl2/Et2O 9:1  8:2; hexane/EtOAc 2:1; CH2Cl2/Et2O

95:5) gave (±)-81 (31 mg, 3%) as a brown solid. Repeated purification by CC (SiO2;

CH2Cl2/Et2O 9:1  8:2; hexane/EtOAc 1:1) gave (±)-82 (111 mg, 9%) as a brown solid. 1 (±)-81: Rf = 0.37 (CH2Cl2/EtOAc 9:1); m.p. 61–63 °C; H NMR (300 MHz, CDCl3): 0.85 (s, 9H, H-C(12)), 1.12–1.18 (m, 2H, H-C(11)), 1.44–1.55 (m, 2H, H-C(10)), 1.60–1.86 (m, 2H, H-C(7), H-C(8)), 1.99–2.22 (m, 2H, H-C(7), H-C(8)), 2.74 (ddd, J = 12.7, 8.2, 4.3, 1H, H-C(6)), 2.88 (ddd, J = 12.7, 8.9, 7.3, 1H, H-C(6)), 3.30 (dd, J = 8.1, 1.3, 1H, H-C(8b)), 3.45–3.40 (m, 2H, H-C(9)), 3.59 (dd, J = 8.2, 8.2, 1H, H- C(3a)), 3.75 (t, J = 8.5, 1H, H-C(8a)), 4.25 (d, J = 8.3, 1H, H-C(4)), 6.25 (s, 1H, H- C(3’)), 6.91 (d, J = 4.0, 1H, H-C(1’)), 7.22 (d, J = 4.0, 1H, H-C(2’)); 13C NMR (75

MHz, CDCl3): 22.9, 24.9, 26.0, 29.4, 30.3, 40.0, 41.1, 48.6, 52.4, 52.9, 62.6, 66.6, 98.5, 126.5, 127.4, 127.8, 133.5, 164.5, 165.4, 177.4, 178.2; IR (ATR): 2951, 2866, 1773, 1695, 1601, 1474, 1423, 1400, 1347, 1217, 1157, 1085, 1067, 1026, 998, 921, 900, 874, 794; HR-MALDI-MS: m/z (%): 464.1582 (42, [M+H]+, calcd for

144 8. Experimental ______

37 + + 35 + C23H29 ClN3O3S : 464.1585), 462.1603 (100, [M+H] , calcd for C23H29 ClN3O3S : 462.1613). 1 (±)-82: Rf = 0.22 (pentane/EtOAc 1:1); m.p. 99 °C; H NMR (300 MHz, CDCl3): 0.86 (s, 9H, H-C(12)), 1.15–1.21 (m, 2H, H-C(11)), 1.48–1.59 (m, 2H, H-C(10)), 1.75–2.11 (m, 4H, H-C(7), H-C(8)), 2.58–2.66 (m, 1H, H-C(6)), 3.16 (ddd, J = 10.5, 8.1, 5.2, 1H, H-C(6)), 3.42–3.51 (m, 3H, H-C(8b), H-C(9)), 3.83–3.90 (m, 2H, H- C(3a), H-C(8a)), 4.57 (d, J = 3.3, 1H, H-C(4)), 6.42 (s, 1H, H-C(3’)), 6.94 (d, J = 4.0, 13 1H, H-C(1’)), 7.26 (d, J = 3.9, 1H, H-C(2’)); C NMR (75 MHz, CDCl3): 22.9, 24.9, 26.0, 29.4, 30.3, 40.0, 41.1, 48.6, 52.4, 52.9, 62.6, 66.6, 98.5, 126.5, 127.4, 127.8, 133.5, 164.5, 165.4, 177.4, 178.2; IR (ATR): 3116, 2948, 1693, 1601, 1370, 1357, 1154, 798; HR-MALDI-MS: m/z (%): 464.1587 (38, [M+H]+, calcd for 37 + + 35 + C23H29 ClN3O3S : 464.1585), 462.1617 (100, [M+H] , calcd for C23H29 ClN3O3S :

462.1613); elemental analysis calcd (%) for C23H28ClN3O3S (461.2): C 59.79, H 6.11, N 9.10; found: C 59.66, H 5.98, N 8.89.

4-Bromobutan-1-amine Hydrobromide (84)[198]

1 3 Br H3N 2 4 Br 4-Aminobutan-1-ol (87) (1.40 mL, 15.1 mmol) in conc. HBr (10 mL) was heated to reflux for 3 h and stirred at RT overnight. HBr was removed under vacuum at 60 °C. The product was recrystallized from EtOAc, filtrated, and washed with cold EtOAc. Compound 84 (1.75 g, 50%) was obtained as a silver solid. [198] 1 M.p. 158–159 °C (Lit. : 146–147 °C); H NMR (300 MHz, CD3OD): 1.76–1.89 (m, 2H, H-C(2)), 1.89–2.00 (m, 2H, H-C(3)), 2.97 (t, J = 7.5, 2H, H-C(1)), 3.50 (t, 13 J = 6.0, 2H, H-C(4)); C NMR (75 MHz, CD3OD): 27.4, 30.7, 33.3, 40.1; IR (ATR): 2992, 2513, 2004, 1596, 1479, 1464, 1440, 1390, 1366, 1347, 1330, 1290, 1256, 1201, 1170, 1128, 1081, 1051, 1028, 969, 935, 902, 868, 788, 735; HR-ESI-MS: m/z + 81 + (%): 154.0049 (87, [M–Br] , calcd for C4H11 BrN : 154.0049), 152.0069 (83, [M– + 79 + + Br] , calcd for C4H11 BrN : 152.0069), 136.9783 (96, [M–NH3Br] , calcd for 81 + + 79 + C4H8 Br : 136.9789), 134.9804 (100, [M–NH3Br] , calcd for C4H8 Br : 134.9809);

elemental analysis calcd for C4H11Br2N (233.0): C 20.62, H 4.76, N 6.01; found: C 20.90, H 4.59, N 5.97.

145 8. Experimental ______5-Bromopentan-1-amine Hydrobromide (85)[199]

1 3 5 H N Br 2 2 4 Br 5-Aminopentan-1-ol (88) (4.0 mL, 38.2 mmol, 1 equiv.) in conc. HBr (15 mL) was heated to reflux for 3.5 h and stirred at RT overnight. HBr was removed under vacuum at 60 °C. The product was recrystallized from EtOAc, filtrated, and washed with cold EtOAc. Compound 85 (3.66 g, 39%) was obtained as a white solid. [199] 1 M.p. 110–112 °C (Lit. : 136–138 °C); H NMR (300 MHz, CD3OD): 1.48–1.60 (m, 2H, H-C(3)), 1.62–1.74 (m, 2H, H-C(4)), 1.85–1.95 (m, 2H, H-C(2)), 2.93 (t, 13 J = 7.2, 2H, H-C(1)), 3.50 (t, J = 6.6, 2H, H-C(5)); C NMR (75 MHz, CD3OD): 26.1, 27.8, 33.4, 33.9, 40.7; IR (ATR): 2906, 2630, 2361, 2342, 1913, 1734, 1571, 1486, 1466, 1448, 1399, 1368, 1330, 1270, 1207, 1148, 1062, 1028, 998, 937, 910, + 81 + 811, 730, 650; HR-ESI-MS: m/z (%): 168.0205 (97, [M–Br] , calcd for C5H13 BrN : + 79 + 168.0205), 166.0226 (91, [M–Br] , calcd for C5H13 BrN : 166.0226).

(Z)-4-(4-Bromobutylamino)-4-oxobut-2-enoic Acid (89) HO

1 O H 4 6 N 2 Br 3 5 O

Et3N (0.20 mL, 1.42 mmol, 1.1 equiv.) was slowly added to a mixture of 84 (300 mg,

1.29 mmol, 1 equiv.) and maleic anhydride (127 mg, 1.29 mmol, 1 equiv.) in CH2Cl2 (4 mL) at 0 °C. The mixture was stirred for 90 min, evaporated in vacuo, and

dissolved in CH2Cl2. Conc. HBr (25 μL) was added and the mixture washed with aq.

1 M HCl solution. The org. phase was dried over MgSO4, filtrated, and evaporated in vacuo. Compound 89 (309 mg, 96%) was obtained as a yellow solid. 1 M.p. 69–71 °C; H NMR (300 MHz, CD3OD): 1.66–1.78 (m, 2H, H-C(4)), 1.85–1.96 (m, 2H, H-C(5)), 3.32 (t, J = 6.8, 2H, H-C(3)), 3.48 (t, J = 6.8, 2H, H-C(6)), 6.23 (d, 13 J = 12.5, 1H, H-C(1)), 6.43 (d, J = 12.5, 1H, H-C(2)); C NMR (75 MHz, CD3OD): 28.6, 31.2, 33.8, 40.0, 133.5, 134.2, 167.9, 168.1; IR (ATR): 3232, 3058, 2928, 2884, 1703, 1629, 1578, 1518, 1458, 1383, 1301, 1244, 1217, 1067, 1037, 963, 897, 742, + 79 + 629; HR-EI-MS: m/z (%): 248.9994 (1, [M] , calcd for C8H12 BrNO3 : 248.9995),

146 8. Experimental ______

+ + 110.0226 (31), 99.0074 (54, [M–Br(CH2)4NH] , calcd for C4H3O3 : 99.0082), 30.0336 (100).

(Z)-4-(5-Bromopentylamino)-4-oxobut-2-enoic Acid (90) HO

1 O H 4 6 2 N Br 3 5 7 O

Et3N (620 μL, 4.45 mmol, 1.1 equiv.) was slowly added to a mixture of 85 (1.00 g,

4.05 mmol, 1 equiv.) and maleic anhydride (397 mg, 4.05 mmol, 1 equiv.) in CH2Cl2 (25 mL) at 0 °C. The mixture was stirred for 2.5 h, evaporated in vacuo, and

dissolved in CH2Cl2. Conc. HBr (120 μL) was added and the mixture washed with

aq. 1 M HCl solution. The org. phase was dried over MgSO4, filtrated, and evaporated in vacuo. Compound 90 (920 mg, 86%) was obtained as a white solid. 1 M.p. 107–109 °C; H NMR (300 MHz, CD3OD): 1.43–1.68 (m, 4H, H-C(4), H-C(6)), 1.88 (quint., J = 6.9, 2H, H-C(5)), 3.30 (t, J = 6.9, 2H, H-C(3)), 3.45 (t, J = 6.6, 2H, H-C(7)), 6.24 (d, J = 12.9, 1H, H-C(1)), 6.43 (d, J = 12.9, 1H, H-C(2)); 13C NMR

(75 MHz, CDCl3): 26.6, 29.1, 33.6, 34.1, 40.7, 133,7, 134.1, 167.8, 168.6; IR (ATR): 3232, 3058, 2928, 2884, 1703, 1629, 1578, 1518, 1458, 1383, 1301, 1244, 1217, 1067, 1037, 963, 897, 742, 629; HR-EI-MS: m/z (%): 263.0151 (1, [M]+, calcd for 79 + + + C9H14 BrNO3 : 263.0152), 99.0067 (25, [M–Br(CH2)5NH] , calcd for C4H3O3 : 99.0082), 30.0532 (100).

1-(4-Bromobutyl)-1H-pyrrole-2,5-dione (91) O 2 4 Br N 1 3 5 1 O

Compound 89 (750 mg, 3.00 mmol, 1.0 equiv.) was dissolved in CH2Cl2 (4 mL) and cooled to 0 °C. DMF (37 L, 0.48 mmol, 0.2 equiv.) was added to the solution at 0 °C. Oxalyl chloride (270 L, 3.15 mmol, 1.1 equiv.) was added dropwise over 30 min and the mixture stirred at RT for 44 h. The mixture was concentrated in vacuo

and the residue dissolved in CH2Cl2. Et3N (450 L, 3.30 mmol, 1.1 equiv.) was added dropwise and the mixture stirred at RT for 30 min. The solution was washed with aq.

147 8. Experimental ______

1 M HCl solution and the org. phase concentrated in vacuo. Purification by CC (SiO2; hexane/EtOAc 1:1) gave 91 (540 mg, 75%) as a yellow solid. 1 Rf = 0.54 (hexane/EtOAc 1:1); m.p. 30 °C; H NMR (300 MHz, CDCl3): 1.70–1.91 (m, 4H, H-C(3), H-C(4)), 3.42 (t, J = 6.3, 2H, H-C(2)), 3.56 (t, J = 6.6, 2H, H-C(5)), 13 6.70 (s, 2H, H-C(1)); C NMR (75 MHz, CDCl3): 27.2, 29.8, 32.7, 36.9, 134.2, 170.8; IR (ATR): 3734, 3629, 3448, 3076, 2953, 1687, 1587, 1446, 1404, 1362, 1333, 1287, 1252, 1216, 1180, 1126, 1042, 988, 974, 954, 839, 816, 767, 744, 710, 691, + 81 + 668, 645; HR-EI-MS: m/z (%): 232.9875 (9, [M] , calcd for C8H10 BrNO2 : + 79 + 232.9874), 230.9888 (8, [M] , calcd for C8H10 BrNO2 : 230.9889), 152.0701 (32, + + + [M–Br] , calcd for C8H10NO2 : 152.0712), 110.0233 (100, [M–Br(CH2)3] , calcd for + C5H4NO2 : 110.0242); elemental analysis calcd for C8H10BrNO2 (232.1): C 41.40, H 4.34, N 6.04; found: C 41.57, H 4.33, N 5.91.

1-(5-Bromopentyl)-1H-pyrrole-2,5-dione (92) Br O 5 3 6 1 4 N 2 1 O DMF (100 μL, 1.29 mmol, 0.17 equiv.) was added to a solution of 90 (2.00 g,

7.57 mmol, 1.0 equiv.) in CH2Cl2 (10 mL) at 0 °C. Oxalyl chloride (683 μL, 7.95 mmol, 1.1 equiv.) was added over 30 min and the mixture stirred at RT for 49 h. The

mixture was evaporated in vacuo and dissolved in CH2Cl2. Et3N (1.20 mL) was added dropwise and the mixture stirred at RT for 30 min. The solution was washed

with aq. 1 M HCl solution. The org. phase was dried over MgSO4, filtrated, and evaporated in vacuo. Purification by CC (SiO2; hexane/EtOAc 3:1) afforded 92 (1.02 g, 57%) as yellow oil. 1 Rf = 0.30 (hexane/EtOAc 3:1); H NMR (300 MHz, CDCl3): 1.40–1.49 (m, 2H, H-C(4)), 1.57–1.68 (m, 2H, H-C(3)), 1.88 (quint., J = 7.2, 2H, H-C(5)), 3.39 (t, J = 6.9, 2H, H-C(6)), 3.53 (t, J = 7.2, 2H, H-C(2)), 6.70 (s, 2H, H-C(1)); 13C NMR

(75 MHz, CDCl3): 25.3, 27.7, 32.1, 33.4, 37.5, 134.1, 170.8; IR (ATR): 3459, 3099, 2940, 2863, 1698, 1568, 1441, 1407, 1368, 1246, 1211, 1130, 1093, 1043, 995, 893, + 81 + 826, 694, 633; HR-EI-MS: m/z (%): 247.0030 (22, [M] , calcd for C9H12 BrNO2 : + 79 + 247.0031), 245.0047 (22, [M] , calcd for C9H12 BrNO2 : 245.0046), 166.0845 (28,

148 8. Experimental ______

+ + + [M–Br] , calcd for C9H12NO2 : 166.0868), 110.0236 (100, [M–Br(CH2)4] , calcd for + C5H4NO2 : 110.0242), 82.0132 (23).

1-(2-Hydroxyethyl)-1H-pyrrole-2,5-dione (95) O 3 1 OH N 2 1 O Methyl chloroformate (4.39 mL, 56.7 mmol, 1.1 equiv.) was added to a solution of maleimide (5.00 g, 51.5 mmol, 1 equiv.) and N-methylmorpholine (6.23 mL, 56.7 mmol, 1.1 equiv.) in EtOAc (250 mL) at 0 °C. The mixture was stirred for 30 min at 0 °C and 45 min at RT, filtrated, and the precipitate washed with EtOAc. The

organic phase was dried over MgSO4, filtrated, and evaporated to dryness. Compound 94[121] was obtained as a brown solid and was reacted to 95 without further purification. Compound 94 (7.00 g, 45.1 mmol, 1 equiv.) was added to a

solution of ethanolamine (3.00 mL, 49.6 mmol, 1.1 equiv.) in sat. aq. NaHCO3 solution (210 mL) at 0 °C. The mixture was stirred for 30 min at 0 °C and 15 min at

RT, and extracted with CH2Cl2. The organic phase was dried over MgSO4, filtrated,

and evaporated to dryness. Purification by CC (SiO2; CH2Cl2/EtOAc 2:1) afforded 95 (3.02 g, 47%) as colorless crystals. [200] 1 Rf = 0.45 (CH2Cl2/EtOAc 2:1); m.p. 72 °C (Lit. : 72–73 °C); H NMR (400 MHz,

CDCl3): 2.18 (t, J = 5.6, 1H, OH), 3.68–3.74 (m, 2H, H-C(4)), 3.74–3.80 (m, 2H, H- 13 C(3)), 6.73 (s, 2H, H-C(1)); C NMR (100 MHz, CDCl3): 40.7, 60.8, 134.2, 171.1; IR (ATR): 3254, 3113, 2960, 2932, 1780, 1764, 1707, 1527, 1441, 1403, 1386, 1362, 1321, 1245, 1156, 1092, 1066, 1051, 971, 930, 852, 825, 775, 714, 692; HR-EI-MS: + + m/z (%): 141.0421 (2, [M] , calcd for C6H7NO3 : 141.0426), 110.0235 (100, [M– + + CH2OH] , calcd for C5H4NO2 : 110.0242), 98.0235 (98), 82.0165 (57).

149 8. Experimental ______1-(2-Bromoethyl)-1H-pyrrole-2,5-dione (96) O 3 1 Br N 2 1 O

PPh3 (5.62 g, 21.4 mmol, 1.1 equiv.) was slowly added to a solution of 95 (2.75 g,

19.5 mmol, 1 equiv.) and CBr4 (7.11 g, 21.4 mmol, 1.1 equiv.) in CH2Cl2 (100 mL) at

0 °C under N2. The mixture was stirred at RT for 23 h and evaporated to dryness.

Repeated purification by CC (SiO2; pentane/CH2Cl2 1:1  CH2Cl2, pentane/EtOAc 4:1) afforded 96 (1.40 g, 37%) as colorless crystals. 1 Rf = 0.27 (pentane/EtOAc 4:1); m.p. = 64 °C; H NMR (300 MHz, CDCl3): 3.52 (t, J = 6.6, 2H, H-C(3)), 3.94 (t, J = 6.6, 2H, H-C(2)), 6.74 (s, 2H, H-C(1)); 13C NMR

(100 MHz, CDCl3): 28.1, 39.1, 134.2, 170.1; IR (ATR): 3094, 2924, 1770, 1686, 1582, 1439, 1432, 1407, 1388, 1372, 1322, 1254, 1202, 1138, 1121, 1060, 1044, 976, 962, 920, 874, 828, 718, 693; HR-EI-MS: m/z (%): 204.9551 (17, [M]+, calcd for 81 + + 79 + C6H6 BrNO2 : 204.9561), 202.9576 (17, [M] , calcd for C6H6 BrNO2 : 202.9582), + + 110.0220 (100, [M–CH2Br] , calcd for C5H4NO2 : 110.0242).

(3aSR,4RS,8aSR,8bRS)-2-4-Bromobutyl)-4-[5-(5-chloro-2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-97) O 8 9 H 11 8a N 8b 7 Br 10 3a 12 N 6 O H 4

N 3' O

2' S 1' Cl A suspension of 91 (561 mg, 2.62 mmol, 1.1 equiv.), L-proline (18) (302 mg, 2.62 mmol, 1.1 equiv.), and 40 (580 mg, 2.50 mmol, 1 equiv.) in MeCN (4 mL) was heated to reflux for 44 h, leading to a mixture of products, and concentrated in vacuo.

Repeated purification by CC (SiO2; CH2Cl2/Et2O 95:5  9:1; pentane/EtOAc 2:1) gave (±)-97 (25 mg, 2%) as a white solid.

150 8. Experimental ______

1 Rf = 0.66 (pentane/EtOAc 1:1); m.p. 109 °C; H NMR (300 MHz, CDCl3): 1.59–1.91 (m, 6H, H-C(7), H-C(8), H-C(10), H-C(11)), 1.98–2.26 (m, 2H, H-C(7), H-C(8)), 2.70–2.80 (m, 1H, H-C(6)), 2.84–2.97 (m, 1H, H-C(6)), 3.31 (dd, J = 6.8, 1.4, 1H, H- C(8b)), 3.41 (t, J = 6.4, 2H, H-C(12)), 3.51 (t, J = 7.0, 2H, H-C(9)), 3.61 (dd, J = 8.3, 8.3, 1H, H-C(3a)), 3.72–3.80 (m, 1H, H-C(8a)), 4.27 (d, J = 8.3, 1H, H-C(4)), 6.28 (s, 1H, H-C(3’)), 6.93 (d, J = 4.0, 1H, H-C(1’)), 7.26 (d, J = 4.0, 1H, H-C(2’)); 13C NMR

(75 MHz, CDCl3): 23.4, 26.4, 29.8, 32.8, 38.0, 49.2, 49.5, 50.7, 60.7, 68.3, 99.0, 126.3, 127.2, 127.9, 133.2, 163.6, 163.7, 175.1, 177.8, 1 signal not visible; IR (ATR): 3124, 2947, 2873, 1768, 1693, 1600, 1532, 1477, 1433, 1401, 1372, 1346, 1308, 1286, 1273, 1245, 1207, 1181, 1148, 1068, 1038, 999, 954, 898 879, 864, 845, 820, 807, 743, 653, 639; HR-MALDI-MS: m/z (%): 502.0199 (26, [M+H]+, calcd for 81 37 + + C20H22 Br ClN3O3S : 502.0200), 500.0217 (100, [M+H] , calcd for 81 35 + + C20H22 Br ClN3O3S : 500.0226), 498.0244 (64, [M+H] , calcd for 79 35 + C20H22 Br ClN3O3S : 498.0248).

(3aSR,4RS,8aSR,8bRS)-2-(5-Bromopentyl)-4-[5-(5-chloro-2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-98) O 8 9 H 11 8a 13 N 8b 7 10 3a Br 12 N 6 O H 4

N 3' O

2' S 1' Cl A suspension of 92 (1.00 g, 4.06 mmol, 1 equiv.), L-proline (18) (491 mg, 4.27 mmol, 1.1 equiv.), and 40 (912 mg, 4.27 mmol, 1.1 equiv.) in MeCN (15 mL) was heated to reflux for 16 h, leading to a mixture of products, and concentrated in vacuo. Repeated

purification by CC (SiO2; CH2Cl2/Et2O 98:2  1:2; hexane/CH2Cl2/EtOAc 4:2:4) gave (±)-98 (45 mg, 2%) as yellow oil. 1 Rf = 0.32 (hexane/EtOAc 1:1); H NMR (300 MHz, CDCl3): 1.35–1.93 (m, 8H, H- C(7), H-C(8), H-C(10), H-C(11), H-C(12)), 1.98–2.25 (m, 2H, H-C(7), H-C(8)), 2.69–2.80 (m, 1H, H-C(6)), 2.83–2.95 (m, 1H, H-C(6)), 3.30 (dd, J = 8.1, 1.2, 1H, H-

151 8. Experimental ______C(8b)), 3.38 (t, J = 6.6, 2H, H-C(13)), 3.48 (t, J = 7.2, 2H, H-C(9)), 3.61 (dd, J = 8.4, 8.4, 1H, H-C(3a)), 3.72–3.82 (m, 1H, H-C(8a)), 4.26 (d, J = 8.4, 1H, H-C(4)), 6.26 (s, 1H, H-C(3’)), 6.93 (d, J = 4.2, 1H, H-C(1’)), 7.25 (d, J = 4.2, 1H, H-C(2’)); 13C NMR

(75 MHz, CDCl3): 23.4, 25.2, 26.8, 29.8, 32.1, 33.4, 38.7, 49.2, 49.4, 50.7, 60.7, 68.2, 99.0, 126.2, 127.2, 127.9, 133.1, 163.6, 175.1, 177.8, 1 signal not visible; IR (ATR): 2941, 1774, 1697, 1601, 1523, 1475, 1432, 1400, 1349, 1242, 1210, 1149, 1068, 1028, 998, 900, 875, 797, 729, 629; HR-MALDI-MS: m/z (%): 516.0342 (27, + 81 37 + + [M+H] , calcd for C21H24 Br ClN3O3S : 516.0357), 514.0364 (100, [M+H] , calcd 81 35 + + for C21H24 Br ClN3O3S : 514.0383), 512.0395 (64, [M+H] , calcd for 79 35 + C21H24 Br ClN3O3S : 512.0405).

4-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- butanaminium Bromide ((±)-99) O 8 13 9 H 11 8a 13 N 8b 7 N 10 3a 12 N 6 13 O H 4 Br N 3' O

2' S 1' Cl To a solution of (±)-97 (19 mg, 0.04 mmol, 1 equiv.) in EtOH (0.4 mL) at RT was added 4.2 M Me3N in EtOH (180 L, 0.76 mmol, 20 equiv.). The mixture was stirred 48 h and concentrated in vacuo. The residue was dissolved in MeOH, precipitated

from Et2O, filtrated, and washed with Et2O. Compound (±)-99 (8 mg, 38%) was obtained as a white solid. 1 M.p. 120 °C (decomp); H NMR (300 MHz, CD3OD): 1.53–1.91 (m, 6H, H-C(7), H- C(8), H-C(10), H-C(11)), 2.03–2.21 (m, 2H, H-C(7), H-C(8)), 2.78–3.01 (m, 2H, H- C(6)), 3.12 (s, 9H, H-C(13)), 3.28–3.47 (m, 2H, H-C(12)), 3.46 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.48–3.55 (m, 2H, H-C(9)), 3.68–3.77 (m, 1H, H-C(8a)), 3.82 (dd, J = 8.3, 8.3, 1H, H-C(3a)), 4.41 (d, J = 8.5, 1H, H-C(4)), 6.60 (s, 1H, H-C(3’)), 7.10 (d, 13 J = 4.0, 1H, H-C(1’)), 7.42 (d, J = 4.0, 1H, H-C(2’)); C NMR (75 MHz, CD3OD):

152 8. Experimental ______21.2, 24.3, 25.3, 30.4, 38.8, 50.6, 50.8, 52.4, 53.67, 53.70, 53.74, 62.7, 67.2, 69.6, 101.2, 128.1, 129.1, 134.1, 164.8, 164.9, 177.6, 180.2, 1 signal not visible; IR (ATR): 3397, 2924, 1771, 1693, 1600, 1531, 1476, 1422, 1404, 1344, 1289, 1239, 1208, 1179, 1145, 1068, 1029, 998, 969, 900, 797, 664, 625; HR-MALDI-MS: m/z (%): + 35 + 477.1728 (73, [M–Br] , calcd for C23H30 ClN4O3S : 477.1722).

5-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- pentanaminium Bromide ((±)-100) O 8 9 H 14 11 8a 13 N 8b 7 14 10 3a N 12 N 6 O 4 Br H 14 N 3' O

2' S 1' Cl

4.2 M Me3N in EtOH (981 L, 4.12 mmol, 20 equiv.) was added to a solution of (±)-98 (30 mg, 0.06 mmol, 1 equiv.) in EtOH (0.5 mL) at RT. The mixture was stirred for 53 h and concentrated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (±)-100 (20 mg, 60%) as a white solid. 1 M.p. 135–137 °C; H NMR (300 MHz, CD3OD): 1.27–1.40 (m, 2H, H-C(11)), 1.55– 1.66 (m, 2H, H-C(10)), 1.72–1.92 (m, 4H, H-C(7), H-C(8), H-C(12)), 2.03–2.22 (m, 2H, H-C(7), H-C(8)), 2.75–2.87 (m, 1H, H-C(6)), 2.89–3.01 (m, 1H, H-C(6)), 3.11 (s, 9H, H-C(14)), 3.25–3.30 (m, 2H, H-C(13)), 3.42–3.50 (m, 3H, H-C(8b), H-C(9)), 3.73 (t, J = 8.7, 1H, H-C(8a)), 3.80 (dd, J = 8.4, 8.4, 1H, H-C(3a)), 4.40 (d, J = 8.7, 1H, H-C(4)), 6.55 (s, 1H, H-C(3’)), 7.10 (d, J = 4.2, 1H, H-C(1’)), 7.42 (d, J = 4.2, 13 1H, H-C(2’)); C NMR (75 MHz, CD3OD): 20.4, 21.5, 25.1, 27.8, 36.3, 48.0, 49.5, 50.8 (t), 59.7, 64.9 (m), 66.8, 98.2, 125.2, 126.28, 126.31, 131.2, 162.0, 162.2, 174.7, 177.4, 2 signals not visible; IR (ATR): 3384, 2945, 1767, 1691, 1613, 1478, 1436, 1419, 1406, 1380, 1346, 1317, 1222, 1177, 1150, 1091, 1074, 1025, 999, 976, 911, 876, 841, 796, 735, 666, 618; HR-MALDI-MS: m/z (%): 493.1846 (36, [M–Br]+, 37 + + calcd for C24H32 ClN4O3S : 493.1851), 491.1880 (95, [M–Br] , calcd for

153 8. Experimental ______

35 + + C24H32 ClN4O3S : 491.1878), 306.2172 (100, [M–C7H4BrClNOS] , calcd for + C17H28N3O2 : 306.2176).

3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N-ethyl-N,N-dimethyl-1- propanaminium Bromide ((±)-101) O 8 12 9 H 11 8a 13 N 8b 7 3a N 10 N 14 6 O 4 Br 12 H N 3' O

2' S 1' Cl N,N-Dimethylethanamine (90 L, 0.83 mmol, 20 equiv.) was added to a solution of (±)-57 (20 mg, 0.04 mmol, 1 equiv.) in EtOH (0.35 mL) at RT. The mixture was stirred for 4 d and treated with further N,N-dimethylethanamine (90 L, 0.83 mmol, 20 equiv.). The mixture was stirred for 4 d, evaporated in vacuo, and the residue was dissolved in MeOH. Precipitation from Et2O gave (±)-101 (19 mg, 81%) as a white solid. 1 M.p. 170 °C (decomp); H NMR (300 MHz, CD3OD): 1.30–1.40 (m, 3H, H-C(14)), 1.74–1.93 (m, 2H, H-C(7), H-C(8)), 1.93–2.23 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.80–3.04 (m, 2H, H-C(6)), 3.07 (s, 6H, H-C(12)), 3.25–3.34 (m, 2H, H-C(11)), 3.40 (q, J = 7.3, 2H, H-C(13)), 3.48 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.51–3.60 (m, 2H, H-C(9)), 3.69–3.77 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.4, 8.4, 1H, H-C(3a)), 4.42 (d, J = 8.7, 1H, H-C(4)), 6.63 (s, 1H, H-C(3’)), 7.10 (d, J = 4.0, 1H, H-C(1’)), 7.42 (d, 13 J = 4.0, 1H, H-C(2’)); C NMR (100 MHz, CD3OD): 8.5, 22.3, 24.2, 30.2, 36.8, 50.56, 50.63, 50.8, 52.3, 61.2, 62.5, 62.7, 69.6, 101.3, 128.1, 129.0, 129.1, 134.2, 164.8, 165.1, 177.5, 180.0; IR (ATR): 3387, 2950, 1770, 1694, 1600, 1530, 1475, 1426, 1400, 1350, 1315, 1233, 1205, 1177, 1145, 1087, 1065, 1020, 999, 923, 900, 802, 742, 662, 635; HR-MALDI-MS: m/z (%): 479.1689 (36, [M–Br]+, calcd for 37 + + 35 + C23H30 ClN4O3S : 479.1694), 477.1723 (100, [M–Br] , calcd for C23H30 ClN4O3S : 477.1722).

154 8. Experimental ______1-{3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-1,2-oxazol-3-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]propyl}-1- methylpyrrolidinium Bromide ((±)-102) O 8 14 13 9 H 11 8a 7 N 8b 14 N 10 3a N 6 13 O 4 12 H Br N 3' O

2' S 1' Cl To a solution of (±)-57 (20 mg, 0.04 mmol, 1 equiv.) in EtOH (1 mL) at RT was added N-methylpyrrolidine (90 L, 0.83 mmol, 20 equiv.). The mixture was stirred at RT for 7 d and evaporated in vacuo. The residue was dissolved in MeOH, and precipitation from Et2O gave (±)-102 (17 mg, 71%) as a white solid. 1 M.p. 197 °C (decomp); H NMR (600 MHz, CDCl3): 1.80–1.90 (m, 2H, H-C(7), H-C(8)), 2.03–2.10 (m, 2H, H-C(10)), 2.11–2.19 (m, 2H, H-C(7), H-C(8)), 2.21–2.28 (m, 4H, H-C(14)), 2.87 (ddd, J = 12.7, 8.3, 4.3, 1H, H-C(6)), 2.96 (ddd, J = 12.6, 8.8, 6.9, 1H, H-C(6)), 3.07 (s, 3H, H-C(12)), 3.34–3.39 (m, 2H, H-C(11)), 3.49 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.51–3.61 (m, 6H, H-C(9), H-C(13)), 3.73–3.76 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.42 (d, J = 8.8, 1H, H-C(4)), 6.63 (s, 1H, H-C(3’)), 7.11 (d, J = 4.0, 1H, H-C(1’)), 7.42 (d, J = 3.9, 1H, H-C(2’)); 13C NMR

(150 MHz, CDCl3): 22.6, 23.5, 24.1, 30.2, 36.8, 50.6, 50.7, 52.3 62.6, 63.1, 65.6, 65.7, 69.6, 101.3, 128.1, 128.98, 129.04, 134.1, 164.8, 165.0, 177.4, 179.9; IR (ATR): 2937, 2880, 1770, 1694, 1600, 1536, 1496, 1479, 1460, 1424, 1400, 1342, 1312, 1280, 1230, 1205, 1151, 1085, 1057, 1018, 1002, 988, 918, 898, 880, 869, 802, 786, + 37 + 741; HR-MALDI-MS: m/z (%): 491.1687 (37, [M–Br] , calcd for C24H30 ClN4O3S : + 35 + 491.1695), 489.1716 (100, [M–Br] , calcd for C24H30 ClN4O3S : 489.1722),

235.0713 (23); elemental analysis calcd (%) for C24H30BrClN4O3S (569.9): C 50.58, H 5.31, N 9.83; found: C 50.44, H 5.27, N 9.71.

155 8. Experimental ______1-{3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]propyl}-1- methylpiperidinium Bromide ((±)-103) O 8 12 9 H 11 8a 13 N 8b 7 3a 14 N 10 N 6 13 O H 4 15 14 Br N 3' O

2' S 1' Cl 1-Methylpiperidine (25 L, 0.21 mmol, 5 equiv.) was added to a solution of (±)-57 (20 mg, 0.04 mmol, 1 equiv.) in acetone (2.5 mL) at RT. The mixture was heated to reflux for 4.5 d, cooled to RT, and evaporated in vacuo. The residue was dissolved in

MeOH. Precipitation from Et2O gave (±)-103 (16 mg, 66%) as a yellow solid. 1 M.p. 145 °C (decomp); H NMR (300 MHz, CD3OD): 1.59–2.22 (m, 12H, H-C(7), H-C(8), H-C(10), H-C(14), H-C(15)), 2.77–3.04 (m, 2H, H-C(6)), 3.06 (s, 3H, H- C(12)), 3.31–3.43 (m, 6H, H-C(11), H-C(13)), 3.49 (dd, J = 8.7, 1.5, 1H, H-C(8b)), 3.51–3.62 (m, 2H, H-C(9)), 3.69–3.78 (m, 1H, H-C(8a)), 3.85 (dd, J = 8.4, 8.4, 1H, H-C(3a)), 4.42 (d, J = 8.7, 1H, H-C(4)), 6.63 (s, 1H, H-C(3’)), 7.10 (d, J = 4.0, 1H, H- 13 C(1’)), 7.42 (d, J = 4.0, 1H, H-C(2’)); C NMR (75 MHz, CD3OD): 21.0, 21.6, 22.2, 24.2, 30.2, 36.9, 50.6, 50.8, 52.3, 62.4, 62.5, 62.6, 69.6, 101.3, 128.1, 129.0, 129.1, 134.1, 164.8, 165.0, 177.5, 180.0, 1 signal not visible; IR (ATR): 3386, 2944, 1770, 1694, 1600, 1476, 1427, 1398, 1346, 1312, 1188, 1085, 1066, 999, 916, 900, 871, 802, 636; HR-MALDI-MS: m/z (%): 505.1851 (38, [M–Br]+, calcd for 37 + + 35 + C25H32 ClN4O3S : 505.1852), 503.1881 (100, [M–Br] , calcd for C25H32 ClN4O3S : + + 503.1878), 318.2175 (21, [M–C7H3ClNOS] , calcd for C18H28N3O2 : 318.2176).

156 8. Experimental ______1-{3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]propyl}-1- azoniabicyclo[2.2.2]octane Bromide ((±)-104) O 8 9 H 11 8a 12 N 8b 7 3a 13 13 N 10 N 6 12 12 O H 4 14 13 Br N 3' O

2' S 1' Cl Quinuclidine (92 mg, 0.83 mmol, 20 equiv.) was added to a solution of (±)-57 (20 mg, 0.04 mmol, 1 equiv.) in EtOH (0.4 mL) at RT. The mixture was stirred for 45 h and concentrated in vacuo. The residue was dissolved in MeOH, and

precipitation from Et2O gave (±)-104 (19 mg, 77%) as a yellow solid. 1 M.p. 130 °C (decomp); H NMR (300 MHz, CD3OD): 1.76–1.92 (m, 2H, H-C(7), H-C(8)), 1.92–2.07 (m, 8H, H-C(10), H-C(13)), 2.07–2.22 (m, 3H, H-C(7), H-C(8), H-C(14)), 2.80–2.91 (m, 1H, H-C(6)), 2.91–3.02 (m, 1H, H-C(6)), 3.10–3.19 (m, 2H, H-C(11)), 3.43 (t, J = 7.9, 6H, H-C(12)), 3.46–3.62 (m, 3H, H-C(8b), H-C(9)), 3.70– 3.79 (m, 1H, H-C(8a)), 3.85 (dd, J = 8.4, 8.4, 1H, H-C(3a)), 4.43 (d, J = 8.8, 1H, H- C(4)), 6.63 (s, 1H, H-C(3’)), 7.11 (d, J = 4.0, 1H, H-C(1’)), 7.43 (d, J = 4.0, 1H, H- 13 C(2’)); C NMR (125 MHz, CD3OD): 20.9, 21.9, 24.2, 24.9, 30.3, 37.0, 50.76, 50.79, 52.4, 55.8, 62.6, 63.0, 69.5, 101.3, 128.2, 129.0, 129.1, 134.2, 164.9, 165.0, 177.5, 179.9; IR (ATR): 3385, 2951, 2881, 1772, 1694, 1600, 1470, 1422, 1402, 1348, 1315, 1215, 1180, 1068, 999, 966, 900, 799, 668, 626; HR-MALDI-MS: m/z (%): 517.1872 + 37 + + (45, [M–Br] , calcd for C26H32 ClN4O3S : 517.1852), 515.1886 (100, [M–Br] , calcd 35 + for C26H32 ClN4O3S : 515.1878).

157 8. Experimental ______1-{3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]propyl}-4-aza-1- azoniabicyclo[2.2.2]octane Bromide ((±)-105) O 8 9 H 11 8a 12 N 8b 7 3a 13 13 N 10 N 6 N 12 12 O H 4 13 Br N 3' O

2' S 1' Cl 1,4-Diazabicyclo[2.2.2]octane (93 L, 0.83 mmol, 20 equiv.) was added to a solution of (±)-57 (20 mg, 0.04 mmol, 1 equiv.) in EtOH (0.35 mL) at RT. The mixture was stirred for 16 h and concentrated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (±)-105 (10 mg, 41%) as a white solid. 1 M.p. 160–161 °C; H NMR (300 MHz, CD3OD): 1.74–1.92 (m, 2H, H-C(7), H-C(8)), 1.98–2.09 (m, 2H, H-C(10)), 2.09–2.22 (m, 2H, H-C(7), H-C(8)), 2.82–2.91 (m, 1H, H-C(6)), 2.93–3.02 (m, 1H, H-C(6)), 3.16–3.29 (m, 8H, H-C(11), H-C(12)), 3.33– 3.41 (m, 6H, H-C(13)), 3.36 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.49–3.61 (m, 2H, H-C(9)), 3.71–3.79 (m, 1H, H-C(8a)), 3.85 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.43 (d, J = 8.7, 1H, H-C(4)), 6.63 (s, 1H, H-C(3’)), 7.11 (d, J = 4.0, 1H, H-C(1’)), 7.43 (d, 13 J = 4.0, 1H, H-C(2’)); C NMR (75 MHz, CD3OD): 21.6, 24.2, 30.2, 36.8, 46.1, 50.7, 50.8, 52.3, 53.6 (m), 62.6, 63.1 (m), 69.5, 101.3, 128.1, 129.0, 129.1, 134.1, 164.9, 165.0, 177.4, 180.0; IR (ATR): 3382, 2956, 1772, 1693, 1600, 1471, 1403, 1353, 1312, 1242, 1199, 1176, 1056, 998, 900, 841, 795; HR-MALDI-MS: m/z (%): + 37 + 518.1794 (37, [M–Br] , calcd for C25H31 ClN5O3S : 518.1804), 516.1828 (100, [M– + 35 + + Br] , calcd for C25H31 ClN5O3S : 516.1831), 331.2127 (43, [M–C7H4BrClNOS] , + calcd for C18H27N4O2 : 331.2129).

158 8. Experimental ______1-{3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]propyl}pyridinium Bromide ((±)-106) O 8 9 H 11 8a 12 N 8b 7 3a 13 N 10 N 6 12 O H 4 14 13 Br N 3' O

2' S 1' Cl A solution of (±)-57 (20 mg, 0.04 mmol, 1 equiv.) in pyridine (500 L, 6.19 mmol, 150 equiv.) was heated to reflux for 3 h. The mixture was concentrated in vacuo, and dissolved in CH2Cl2. Precipitation from Et2O gave (±)-106 (23 mg, quant.) as a brown solid. 1 M.p. 67 °C (decomp); H NMR (300 MHz, CD3OD): 1.78–1.93 (m, 2H, H-C(7), H-C(8)), 2.09–2.32 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.83–2.91 (m, 1H, H-C(6)), 3.04–2.95 (m, 1H, H-C(6)), 3.50–3.57 (m, 3H, H-C(8b), H-C(9)), 3.78 (t, J = 8.2, 1H, H-C(8a)), 3.88 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.46 (d, J = 8.7, 1H, H-C(4)), 4.64 (t, J = 7.5, 2H, H-C(11)), 6.63 (s, 1H, H-C(3’)), 7.09 (d, J = 4.0, 1H, H-C(1’)), 7.40 (d, J = 4.0, 1H, H-C(2’)), 8.13 (dd, J = 7.6, 6.8, 2H, H-C(13)), 8.61 (tt, J = 7.8, 1.1, 1H, 13 H-C(14)), 9.04 (d, J = 5.5, 2H, H-C(12)); C NMR (75 MHz, CD3OD): 24.2, 30.3, 30.6, 36.3, 50.8, 52.4, 60.4, 62.6, 69.5 (m), 101.15, 101.21, 128.1, 128.9, 129.0, 129.5, 134.1, 146.2, 147.1, 164.9, 165.0, 177.5, 180.0; IR (ATR): 2955, 2879, 1773, 1692, 1633, 1597, 1487, 1473, 1420, 1400, 1343, 1305, 1282, 1213, 1174, 1128, 1082, 1060, 997, 900, 873, 798, 772; HR-MALDI-MS: m/z (%): 485.1229 (31, [M– + 37 + + Br] , calcd for C24H24 ClN4O3S : 485.1228), 483.1257 (100, [M–Br] , calcd for 35 + + C24H24 ClN4O3S : 483.1258), 298.1548 (23, [M–C7H4BrClNOS] , calcd for + C17H20N3O2 : 298.1550).

159 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-2-(3-Aminopropyl)-4-[5-(5-chloro-2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-108) O 8 9 H 11 8a N 8b 7 3a H2N 10 N 6 O H 4

N 3' O

2' S 1' Cl

PPh3 (59 mg, 0.22 mmol, 2 equiv.) was added to a solution of (±)-109 (50 mg,

0.11 mmol, 1 equiv.) in H2O (45 L) and THF (1.2 mL) at RT. The mixture was stirred for 19 h and concentrated in vacuo. The product was dissolved in MeOH and precipitated from Et2O. Purification by CC (SiO2; CH2Cl2/MeOH/NH3 94:5:1) gave (±)-108 (33 mg, 70%) as a yellow solid. 1 M.p. 129 °C; H NMR (300 MHz, CD3OD): 1.65 (quint., J = 6.8, 2H, H-C(10)), 1.72– 1.90 (m, 2H, H-C(7), H-C(8)), 2.01–2.22 (m, 2H, H-C(7), H-C(8)), 2.58 (t, J = 6.8, 2H, H-C(11)), 2.72–2.82 (m, 1H, H-C(6)), 2.87–2.99 (m, 1H, H-C(6)), 3.43 (dd, J = 8.2, 1.7, 1H, H-C(8b)), 3.49 (t, J = 6.8, 2H, H-C(9)), 3.68–3.76 (m, 1H, H-C(8a)), 3.77 (dd, J = 8.3, 8.3, 1H, H-C(3a)), 4.36 (d, J = 8.5, 1H, H-C(4)), 6.51 (s, 1H, H- C(3’)), 7.08 (d, J = 4.0, 1H, H-C(1’)), 7.41 (d, J = 4.0, 1H, H-C(2’)); 13C NMR

(75 MHz, CDCl3): 23.5, 29.9, 31.5, 36.6, 39.1, 49.3, 49.6, 50.8, 60.8, 68.4, 99.1, 126.4, 127.3, 128.0, 133.2, 163.71, 163.74, 175.4, 178.1; IR (ATR): 3357, 3122, 2936, 1766, 1690, 1600, 1531, 1475, 1431, 1398, 1368, 1344, 1310, 1269, 1209, 1168, 1154, 1112, 1069, 1034, 999, 955, 922, 899, 866, 846, 818, 741, 666, 638; HR- + 37 + MALDI-MS: m/z (%): 423.1076 (36, [M+H] , calcd for C19H22 ClN4O3S : + 35 + 423.1067), 421.1097 (100, [M+H] , calcd for C19H22 ClN4O3S : 421.1096).

160 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-2-(3-Azidopropyl)-4-[5-(5-chloro-2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-109) O 8 9 H 11 8a N 8b 7 3a N3 10 N 6 O H 4

N 3' O

2' S 1' Cl

NaN3 (18 mg, 0.28 mmol, 2.25 equiv.) in H2O (70 L) was added to a solution of (±)-57 (60 mg, 0.12 mmol, 1 equiv.) in DMF (0.2 mL) at RT. The mixture was stirred at 45 °C for 7 h, poured onto ice water, and extracted with EtOAc. The org. phase was washed with brine, dried over Na2SO4, filtrated, and concentrated in vacuo. Compound (±)-109 (53 mg, 96%) was obtained as a light yellow solid. 1 M.p. 95 °C; H NMR (300 MHz, CDCl3): 1.59–1.88 (m, 4H, H-C(7), H-C(8), H-C(10)), 1.96–2.23 (m, 2H, H-C(7), H-C(8)), 2.68–2.79 (m, 1H, H-C(6)), 2.82–2.91 (m, 1H, H-C(6)), 3.24–3.36 (m, 3H, H-C(8b), H-C(11)), 3.54 (t, J = 6.9, 2H, H-C(9)), 3.62 (dd, J = 8.2, 8.2, 1H, H-C(3a)), 3.71–3.80 (m, 1H, H-C(8a)), 4.25 (d, J = 8.3, 1H, H-C(4)), 6.26 (s, 1H, H-C(3’)), 6.91 (d, J = 4.0, 1H, H-C(1’)), 7.23 (d, J = 4.0, 1H, 13 H-C(2’)); C NMR (75 MHz, CDCl3): 23.4, 27.1, 29.8, 36.6, 49.0, 49.3, 49.4, 50.7, 60.7, 68.2, 98.9, 126.3, 127.2, 127.8, 133.2, 163.5, 163.7, 175.1, 177.7; IR (ATR): 3462, 2929, 2880, 2097, 1774, 1701, 1666, 1601, 1475, 1434, 1402, 1386, 1347, 1314, 1255, 1173, 1090, 1066, 1030, 999, 922, 900, 877, 803, 659, 624; HR-MALDI- + 37 + MS: m/z (%): 449.0972 (40, [M+H] , calcd for C19H20 ClN6O3S : 449.0972), + 35 + 447.1001 (100, [M+H] , calcd for C19H20 ClN6O3S : 447.1001), 179.0820 (33).

161 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-2-[3- (methylamino)propyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)- dione ((±)-110) O 8 9 H 11 8a 7 12 N 8b 3a N 10 N H 6 O H 4

N 3' O

2' S 1' Cl

To a solution of (±)-57 (60 mg, 0.12 mmol, 1 equiv.) and K2CO3 (51 mg, 0.37 mmol,

3 equiv.) in DMF (15 ml) at RT was added 2 M MeNH2 in THF (1.24 mL, 2.48 mmol, 20 equiv.). The mixture was stirred for 23 h and concentrated in vacuo. Purification by CC (SiO2; CH2Cl2/MeOH/NH3 94:5:1) gave (±)-110 (17 mg, 32%) as a white solid. 1 M.p. 109–110 °C; H NMR (300 MHz, CDCl3): 1.65–1.92 (m, 4H, H-C(7), H-C(8), H-C(10)), 1.99–2.23 (m, 2H, H-C(7), H-C(8)), 2.39 (s, 3H, H-C(12)), 2.55 (t, J = 6.8, 2H, H-C(11)), 2.69–2.94 (m, 2H, H-C(6)), 3.31 (dd, J = 8.1, 1.4, 1H, H-C(8b)), 3.55 (t, J = 7.0, 2H, H-C(9)), 3.60 (dd, J = 8.2, 8.2, 1H, H-C(3a)), 3.71–3.79 (m, 1H, H- C(8a)), 4.25 (d, J = 8.3, 1H, H-C(4)), 6.27 (s, 1H, H-C(3’)), 6.93 (d, J = 4.0, 1H, H- 13 C(1’)), 7.26 (d, J = 4.0, 1H, H-C(2’)); C NMR (75 MHz, CDCl3): 23.4, 27.7, 29.8, 36.4, 37.0, 48.9, 49.2, 49.5, 50.6, 60.7, 68.3, 99.1, 126.3, 127.2, 127.9, 133.1, 163.6, 163.7, 175.3, 177.9; IR (ATR): 3124, 2935, 2893, 2792, 1768, 1690, 1601, 1531, 1474, 1431, 1400, 1342, 1311, 1287, 1245, 1179, 1158, 1117, 1066, 1036, 998, 898, 866, 846, 817, 808, 738, 664, 640; HR-MALDI-MS: m/z (%): 437.1221 (37, [M+H]+, 37 + + calcd for C20H24 ClN4O3S : 437.1223), 435.1248 (100, [M+H] , calcd for 35 + C20H24 ClN4O3S : 435.1252).

162 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-2-[3- (dimethylamino)propyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)- dione ((±)-111) O 8 9 H 11 8a 7 12 N 8b N 10 3a N 6 O 4 12 H N 3' O

2' S 1' Cl To a solution of (±)-57 (15 mg, 0.03 mmol, 1 equiv.) in EtOH (0.3 mL) at RT was added 7.9 M Me2NH in H2O (78 L, 0.62 mmol, 20 equiv.). The mixture was stirred for 56 h and concentrated in vacuo. The residue was dissolved in MeOH and product

precipitated from Et2O. Purification by CC (SiO2; CH2Cl2/MeOH/NH3 94:5:1) gave (±)-111 (10 mg, 72%) as a white solid. 1 M.p. 114–115 °C; H NMR (300 MHz, CD3OD): 1.65–1.93 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.01–2.25 (m, 2H, H-C(7), H-C(8)), 2.29 (s, 6H, H-C(12)), 2.41 (t, J = 7.6, 2H, H-C(11)), 2.73–2.85 (m, 1H, H-C(6)), 2.87–3.01 (m, 1H, H-C(6)), 3.40–3.51 (m, 3H, H-C(8b), H-C(9)), 3.69–3.83 (m, 2H, H-C(3a), H-C(8a)), 4.38 (d, J = 8.6, 1H, H-C(4)), 6.54 (s, 1H, H-C(3’)), 7.09 (d, J = 4.0, 1H, H-C(1’)), 7.42 (d, J = 4.0, 1H, 13 H-C(2’)); C NMR (75 MHz, CD3OD): 24.3, 26.1, 30.5, 37.9, 45.1, 50.78, 50.83, 52.1, 57.6, 62.2, 69.5, 100.8, 128.0, 129.0, 129.1, 134.0, 164.8, 165.1, 177.4, 180.0; IR (ATR): 3126, 2946, 2820, 2771, 1767, 1693, 1601, 1531, 1472, 1431, 1403, 1387, 1347, 1308, 1288, 1253, 1205, 1177, 1152, 1117, 1080, 1063, 1039, 1020, 999, 965, 899, 870, 846, 805, 754, 737, 665, 636; HR-MALDI-MS: m/z (%): 451.1377 (33, + 37 + + [M+H] , calcd for C21H26 ClN4O3S : 451.1380), 449.1406 (100, [M+H] , calcd for 35 + C21H26 ClN4O3S : 449.1409).

163 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-2- methylhexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-112) O H 8 9 8a N 8b 7 3a N 6 O H 4

N 3' O

2' S 1' Cl A suspension of N-methylmaleimide (117) (600 mg, 5.40 mmol, 1.0 equiv.), L-proline (18) (653 mg, 5.67 mmol, 1.1 equiv.), and 40 (1208 mg, 5.67 mmol, 1.1 equiv.) in MeCN (7 mL) was heated to reflux for 3 h, leading to a mixture of products.

Repeated purification by CC (SiO2; CH2Cl2/Et2O 9:1  4:1; pentane/EtOAc 2:1) and HPLC (Merck LiChrospher® Si 60; 25025 mm, 5 m; EtOAc) gave (±)-112 (18 mg, 1%) as a light brown solid. 1 Rf = 0.22 (CH2Cl2/Et2O 4:1); m.p. 175 °C (decomp); H NMR (400 MHz, CD2Cl2): 1.63–1.78 (m, 2H, H-C(7), H-C(8)), 1.99–2.18 (m, 2H, H-C(7), H-C(8)), 2.72 (ddd, J = 12.7, 8.3, 4.3, 1H, H-C(6)), 2.84 (ddd, J = 12.6, 9.0, 7.4, 1H, H-C(6)), 2.92 (s, 3H, H-C(9)), 3.31 (dd, J = 8.0, 1.1, 1H, H-C(8b)), 3.59 (dd, J = 8.2, 8.2, 1H, H-C(3a)), 3.67–3.71 (m, 1H, H-C(8a)), 4.23 (d, J = 8.4, 1H, H-C(4)), 6.29 (s, 1H, H-C(3’)), 6.97 (d, J = 4.0, 1H, H-C(1’)), 7.26 (d, J = 4.0, 1H, H-C(2’)); 13C NMR (100 MHz,

CD2Cl2): 23.8, 25.3, 30.0, 49.6, 50.0, 50.9, 61.0, 68.6, 99.6, 126.6, 127.7, 128.4, 133.3, 163.8, 164.4, 175.6, 178.2; IR (ATR): 2923, 2856, 1690, 1595, 1530, 1465, 1433, 1381, 1282, 1209, 1133, 1059, 1010, 973, 895, 808, 738, 644; HR-MALDI- + 37 + MS: m/z (%): 380.0653 (37, [M+H] , calcd for C17H17 ClN3O3S : 380.0644), + 35 + 378.0677 (100, [M+H] , calcd for C17H17 ClN3O3S : 378.0674).

164 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-2-Butyl-4-[5-(5-chloro-2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-113) O 8 9 H 11 8a N 8b 7 3a 12 10 N 6 O H 4

N 3' O

2' S 1' Cl A suspension of 118 (631 mg, 4.12 mmol, 1 equiv.), L-proline (18) (498 mg, 4.33 mmol, 1.1 equiv.), and 40 (921 mg, 4.33 mmol, 1.1 equiv.) in MeCN (8 mL) was heated to reflux for 42 h, leading to a mixture of products. Repeated purification by

CC (SiO2; CH2Cl2/EtOAc 3:1; CH2Cl2/EtOAc 5:1; pentane/EtOAc 2:1) followed by purification by HPLC (Merck LiChrospher® Si 60; 25025 mm, 5 m; EtOAc) gave (±)-113 (50 mg, 3%) as a white solid. 1 Rf = 0.44 (EtOAc); m.p. 159 °C; H NMR (400 MHz, CDCl3): 0.92 (t, J = 7.3, 3H, H-C(12)), 1.30 (sextet, J = 15.3, 7.7, 2H, H-C(11)), 1.49–1.56 (m, 2H, H-C(10)), 1.66 (dtd, J = 12.6, 10.2, 7.9, 1H, H-C(8)), 1.81 (dqd, J = 12.9, 8.6, 4.3, 1H, H-C(7)), 2.01–2.10 (m, 1H, H-C(7)), 2.12–2.22 (m, 1H, H-C(8)), 2.75 (ddd, J = 12.6, 8.3, 4.4, 1H, H-C(6)), 2.89 (ddd, J = 12.6, 9.0, 7.3, 1H, H-C(6)), 3.30 (dd, J = 8.1, 1.4, 1H, H- C(8b)), 3.47 (t, J = 7.4, 2H, H-C(9)), 3.59 (dd, J = 8.3, 8.3, 1H, H-C(3a)), 3.75–3.79 (m, 1H, H-C(8a)), 4.26 (d, J = 8.4, 1H, H-C(4)), 6.24 (s, 1H, H-C(3’)), 6.93 (d, J = 13 4.0, 1H, H-C(1’)), 7.24 (d, J = 3.9, 1H, H-C(2’)); C NMR (100 MHz, CDCl3): 13.8, 20.1, 23.6, 29.9, 30.0, 39.1, 49.4, 49.5, 50.8, 60.9, 68.4, 99.2, 126.3, 127.3, 128.1, 133.2, 163.7, 163.8, 175.3, 177.9; IR (ATR): 2954, 2932, 2872, 1767, 1691, 1600, 1473, 1432, 1402, 1351, 1249, 1189, 1142, 1073, 1000, 899, 803, 737, 640; HR- + 37 + MALDI-MS: m/z (%): 422.1120 (40, [M+H] , calcd for C20H23 ClN3O3S : + 35 + 422.1114), 420.1145 (100, [M+H] , calcd for C20H23 ClN3O3S : 420.1143), + + 235.1439 (20, [M–C7H3ClNOS] , calcd for C13H19N2O2 : 235.1447).

165 8. Experimental ______(3aRS,4RS,8aSR,8bSR)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-2- pentylhexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-114) O 8 9 H 11 8a 7 13 N 8b 10 3a 12 N 6 O H 4

N 3' O

2' S 1' Cl Compound 119 (2.57 g, 15.4 mmol, 1 equiv.), L-proline (18) (1.86 g, 16.1 mmol, 1.1 equiv.), and 40 (3.44 g, 16.1 mmol, 1.1 equiv.) in MeCN (25 mL) were heated to reflux for 19 h, leading to a mixture of products, and concentrated in vacuo. Repeated purification by CC (SiO2; pentane/CH2Cl2 1:1  CH2Cl2  CH2Cl2/Et2O 300:1 

Et2O; CH2Cl2/Et2O 240:1) gave (±)-114 (112 mg, 2%) as a light brown solid. 1 Rf = 0.30 (pentane/EtOAc 1:1); m.p. 139 °C; H NMR (400 MHz, CDCl3): 0.88 (t, J = 7.1, 3H, H-C(13)), 1.22–1.37 (m, 4H, H-C(11), H-C(12))), 1.50–1.58 (m, 2H, H-C(10)), 1.67 (dtd, J = 12.5, 10.2, 7.9, 1H, H-C(8)), 1.81 (dqd, J = 12.9, 8.6, 4.3, 1H, H-C(7)), 2.01–2.12 (m, 1H, H-C(7)), 2.12–2.22 (m, 1H, H-C(8)), 2.75 (ddd, J = 12.7, 8.3, 4.4, 1H, H-C(6)), 2.89 (ddd, J = 12.7, 8.9, 7.3, 1H, H-C(6)), 3.30 (dd, J = 8.1, 1.4, 1H, H-C(8b)), 3.45 (t, J = 7.5, 2H, H-C(9)), 3.59 (dd, J = 8.2, 8.2, 1H, H-C(3a)), 3.74–3.79 (m, 1H, H-C(8a)), 4.26 (d, J = 8.4, 1H, H-C(4)), 6.24 (s, 1H, H-C(3’)), 6.93 (d, J = 4.0, 1H, H-C(1’)), 7.24 (d, J = 4.0, 1H, H-C(2’)); 13C NMR (100 MHz,

CDCl3): 14.1, 22.4, 23.6, 27.5, 29.0, 30.0, 39.3, 49.4, 49.5, 50.8, 60.9, 68.4, 99.2, 126.3, 127.3, 128.1, 133.2, 163.7, 163.8, 175.3, 177.9; IR (ATR): 2926, 1767, 1691, 1469, 1431, 1404, 1347, 1308, 1286, 1242, 1189, 1142, 1071, 999, 962, 899, 865, 820, 803, 726, 666, 640, 626, 600; HR-MALDI-MS: m/z (%): 436.1286 (47, [M+H]+, 37 + + calcd for C21H24 ClN3O3S : 436.1271), 434.1300 (100, [M+H] , calcd for 35 + C21H24 ClN3O3S : 434.1300); elemental analysis calcd (%) for C21H24ClN3O3S (434.0): C 58.12, H 5.57, N 9.68; found: C 58.09, H 5.47, N 9.50.

166 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-1,2-oxazol-3-yl]-2-(4- methylpentyl)hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-115) O 8 9 H 11 8a 7 13 12 N 8b 3a 10 N 6 O 4 13 H N 3' O

2' S 1' Cl A mixture of maleimide 124 (398 mg, 2.20 mmol, 1 equiv.), L-proline (18) (266 mg, 2.31 mmol, 1.1 equiv.), and 40 (493 mg, 2.31 mmol, 1.1 equiv.) was heated to reflux for 18 h, leading to a mixture of products, and concentrated in vacuo. Repeated

purification by CC (SiO2; CH2Cl2  CH2Cl2/Et2O 97:3  5:1; CH2Cl2/EtOAc 9:1;

CH2Cl2/Et2O 97:3) gave (±)-115 (13 mg, 1%) as yellow crystals. 1 Rf = 0.40 (CH2Cl2/EtOAc 4:1); m.p. 106–109 °C; H NMR (400 MHz, CDCl3): 0.86 (dd, J = 6.6, 2.1, 6H, H-C(13)), 1.13–1.19 (m, 2H, H-C(11)), 1.46–1.57 (m, 3H, H- C(10), H-C(12)), 1.66 (dtd, J = 12.6, 10.2, 7.9, 1H, H-C(8)), 1.75–1.86 (m, 1H, H- C(7)), 2.01–2.11 (m, 1H, H-C(7)), 2.13–2.21 (m, 1H, H-C(8)), 2.75 (ddd, J = 12.7, 8.3, 4.4, 1H, H-C(6)), 2.88 (ddd, J = 12.6, 9.1, 7.3, 1H, H-C(6)), 3.30 (dd, J = 8.1, 1.4, 1H, H-C(8b)), 3.44 (t, J = 7.5, 2H, H-C(9)), 3.59 (dd, J = 8.2, 8.2, 1H, H-C(3a)), 3.74–3.78 (m, 1H, H-C(8a)), 4.25 (d, J = 8.3, 1H, H-C(4)), 6.24 (s, 1H, H-C(3’)), 6.92 (d, J = 4.0, 1H, H-C(2’)), 7.23 (d, J = 3.9, 1H, H-C(1’)); 13C NMR (100 MHz,

CDCl3): 22.6, 22.7, 23.6, 25.7, 27.8, 29.9, 35.9, 39.5, 49.3, 49.5, 50.8, 60.8, 68.4, 99.1, 126.3, 127.3, 128.1, 133.2, 163.7, 163.8, 175.3, 177.9; IR (ATR): 2962, 2930, 2906, 1767, 1689, 1605, 1533, 1477, 1430, 1402, 1371, 1348, 1303, 1244, 1210, 1186, 1146, 1082, 1066, 1000, 900, 871, 808, 752, 729, 704, 668, 639, 629, 611; HR- + 37 + MALDI-MS: m/z (%): 450.1428 (32, [M+H] , calcd for C22H27 ClN3O3S : + 35 + 450.1428), 448.1459 (100, [M+H] , calcd for C22H27 ClN3O3S : 448.1456).

167 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-1,2-oxazol-3-yl]-2-(3- cyclohexylpropyl)hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-116) O 8 9 H 11 8a 7 13 12 N 8b 3a 14 10 N 6 13 O H 4 15 14 N 3' O

2' S 1' Cl A mixture of 130 (420 mg, 1.90 mmol, 1 equiv.), L-proline (18) (229 mg, 1.99 mmol, 1.1 equiv.), and 40 (424 mg, 1.99 mmol, 1.1 equiv.) in MeCN (6 mL) was heated to reflux for 19 h, leading to a mixture of products, and concentrated in vacuo. Repeated

purification by CC (SiO2; CH2Cl2/Et2O 98:2; CH2Cl2/EtOAc 24:1; pentane/EtOAc 2:1) gave (±)-116 (40 mg, 4%) as a brown solid. 1 Rf = 0.45 (pentane/EtOAc 1:1); m.p. 79–81 °C; H NMR (400 MHz, CDCl3): 0.79– 0.91 (m, 2H, H-C(13)), 1.07–1.22 (m, 6H, H-C(11), H-C(12), H-C(15)), 1.49–1.58 (m, 6H, H-C(10)), 1.60–1.71 (m, 5H, H-C(8), H-C(10), H-C(13), H-C(14)), 1.76–1.87 (m, 1H, H-C(7)), 2.01–2.10 (m, 1H, H-C(7)), 2.12–2.22 (m, 1H, H-C(8)), 2.75 (ddd, J = 12.7, 8.4, 4.3, 1H, H-C(6)), 2.85–2.92 (m, 1H, H-C(6)), 3.30 (dd, J = 8.1, 1.3, 1H, H-C(8b)), 3.43 (t, J = 7.6, 2H, H-C(9)), 3.59 (dd, J = 8.3, 8.3, 1H, H-C(3a)), 3.74– 3.79 (m, 1H, H-C(8a)), 4.26 (d, J = 8.5, 1H, H-C(4)), 6.24 (s, 1H, H-C(3’)), 6.93 (d, J = 3.9, 1H, H-C(1’)), 7.24 (d, J = 3.9, 1H, H-C(2’)), some excess proton signals in the 13 aliphatic region; C NMR (150 MHz, CDCl3): 23.6, 25.3, 26.4, 26.7, 29.9, 33.42, 33.45, 34.5, 37.5, 39.6, 49.4, 49.5, 50.8, 60.8, 68.4, 99.2, 126.3, 127.3, 128.1, 133.2, 163.7, 163.8, 175.3, 177.9, 1 signal in excess; IR (ATR): 3115, 2922, 2851, 1772, 1693, 1605, 1526, 1477, 1433, 1402, 1346, 1244, 1210, 1164, 1144, 1076, 1032, 998, 923, 900, 792, 744, 667, 638; HR-MALDI-MS: m/z (%): 490.1744 (44, [M+H]+, 37 + + calcd for C25H31 ClN3O3S : 490.1742), 488.1764 (100, [M+H] , calcd for 35 + + C25H31 ClN3O3S : 488.1796), 303.2062 (22, [M–C7H3ClNOS] , calcd for + C18H27N2O2 : 303.2073).

168 8. Experimental ______1-Butyl-1H-pyrrole-2,5-dione (118)[201,202]

5 O 3 1 4 N 1 2 O

A solution of 122 (1.00 g, 5.84 mmol, 1 equiv.) in CH2Cl2 (10 mL) was cooled to 0 °C. DMF (6 L, 0.08 mmol, 0.01 equiv.) and oxalyl chloride (552 L, 6.43 mmol, 1.1 equiv.) were added dropwise, and the mixture stirred at RT for 43 h. The mixture

was concentrated in vacuo, and the residue was dissolved in CH2Cl2 (7 mL). Et3N (895 L, 6.43 mmol, 1.1 equiv.) was added dropwise, and the mixture stirred at RT for 7 h. The mixture was diluted with CH2Cl2, washed with aq. 1 M HCl solution,

dried over MgSO4, filtrated, and concentrated in vacuo. Purification by CC (SiO2;

CH2Cl2) gave 118 (631 mg, 71%) as a yellow oil. 1 Rf = 0.35 (CH2Cl2); H NMR (300 MHz, CDCl3): 0.93 (t, J = 7.3, 3H, H-C(5)), 1.31 (dq, J = 15.1, 7.5, 2H, H-C(4)), 1.52–1.62 (m, 2H, H-C(3)), 3.52 (t, J = 7.2, 2H, 13 H-C(2)), 6.68 (s, 2H, H-C(1)); C NMR (100 MHz, CDCl3): 13.7, 20.1, 30.7, 37.8, + + 134.2, 171.0; HR-EI-MS: m/z (%): 153.0783 (35, [M] , calcd for C8H11NO2 : + + 153.0785), 110.0229 (100, [M–CH3CH2CH2] , calcd for C5H4NO2 : 110.0242).

1-Pentyl-1H-pyrrole-2,5-dione (119)

5 O 6 3 1 4 N 2 1 O

Compound 123 (3.50 g, 18.9 mmol, 1 equiv.) was dissolved in CH2Cl2 (45 mL) and cooled to 0 °C. DMF (17 L) and oxalyl chloride (1.79 mL, 20.8 mmol, 1.1 equiv.) were added dropwise. The mixture was stirred at RT. After 7 d, the mixture was

evaporated in vacuo. The dark red oily residue was dissolved in CH2Cl2 (20 mL), and

Et3N (2.89, 20.8 mmol, 1.1 equiv.) was added. The mixture was stirred at RT for 20

h, diluted with CH2Cl2, washed with aq. 1 M HCl solution, dried over MgSO4, filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2) gave 119 (2.58 g, 82%) as a light yellow oil.

169 8. Experimental ______

1 Rf = 0.48 (CH2Cl2); H NMR (300 MHz, CDCl3): 0.88 (t, J = 7.0, 3H, H-(6)), 1.19– 1.39 (m, 4H, H-C(4), H-C(5)), 1.53–1.63 (m, 2H, H-C(3)), 3.50 (t, J = 7.3, 2H, 13 H-C(2)), 6.67 (s, 2H, H-C(1)); C NMR (100 MHz, CDCl3): 14.0, 22.3, 28.4, 29.0, 38.1, 134.2, 171.0; IR (ATR): 3100, 2933, 2862, 1770, 1697, 1587, 1442, 1407, 1365, 1336, 1263, 1239, 1173, 1159, 1113, 1013, 951, 826, 728, 694, 636, 622; HR-EI-MS: + + m/z (%): 167.0942 (32, [M] , calcd for C9H13NO2 : 167.0941), 110.0242 (100, [M– + + CH3(CH2)3] , calcd for C5H4NO2 : 110.0242); 82.0201 (32); elemental analysis calcd

(%) for C9H13NO2 (167.3): C 64.65, H 7.84, N 8.38; found: C 64.59, H 7.82, N 8.37.

(2Z)-4-(Butylamino)-4-oxo-2-butenoic Acid (122)[203,204] O OH3 5 HN 1 4 6

2 O 1-Butylamine (4.00 mL, 25.7 mmol, 1 equiv.) was added to a solution of maleic

anhydride (2.52 g, 25.7 mmol, 1 equiv.) in CH2Cl2 (50 mL) at 0 °C. The mixture was

stirred at RT for 72 h, evaporated in vacuo, and diluted with CH2Cl2. The org. phase

was washed with aq. 1 M HCl solution, dried over MgSO4, filtrated, and evaporated in vacuo, giving 122 (3.92 g, 89%) as a white solid. [204] 1 M.p. 82–83 °C (Lit. : 85 °C); H NMR (300 MHz, CD3OD): 0.95 (t, J = 7.3, 3H, H-C(6)), 1.38 (dq, J = 15.1, 7.4, 2H, H-C(5)), 1.50–1.60 (m, 2H, H-C(4)), 3.24–3.36 (m, 2H, H-C(3)), 6.24 (d, J = 12.7, 1H, H-C(1)), 6.43 (d, J = 12.7, 1H, H-C(2); + + HR-EI-MS: m/z (%): 171.0891 (2, [M] , calcd for C8H13NO3 : 171.0890), 99.0082 + + (33, [M–CH3(CH2)3NH] , calcd for C4H3O3 : 99.0082), 30.0467 (100).

(2Z)-4-Oxo-4-(pentylamino)-2-butenoic Acid (123) O OH3 5 7 HN 1 4 6

2 O n-Pentylamine (2.37 mL, 20.4 mmol, 1 equiv.) was added dropwise to maleic

anhydride (2.00 g, 20.4 mmol, 1 equiv.) in CH2Cl2 (35 mL) at 0 °C. The mixture was

stirred at RT for 30 min, washed with aq. 1 M HCl solution, dried over MgSO4, filtrated, and evaporated in vacuo to give 123 (3.53 g, 93%) as a white solid.

170 8. Experimental ______

[205] 1 M.p. 72 °C (Lit. : 72 °C); H NMR (300 MHz, CDCl3): 0.91 (t, J = 6.9, 3H, H- C(7)), 1.32–1.36 (m, 4H, H-C(5), H-C(6)), 1.56–1.66 (m, 2H, H-C(4)), 3.38 (td, J = 7.2, 6.0, 2H, H-C(3)), 6.32 (d, J = 12.8, 1H, H-C(1)), 6.40 (d, J = 12.8, 1H, H-C(2)), 13 7.40 (br. s, 1H, NH); C NMR (100 MHz, CDCl3): 14.0, 22.4, 28.6, 29.1, 40.8, 131.7, 136.3, 165.7, 166.3; IR (ATR): 3238, 3068, 2956, 2931, 2858, 1915, 1779, 1702, 1626, 1577, 1513, 1454, 1401, 1372, 1320, 1292, 1216, 1130, 1056, 1040, 959, 893, 845, 799, 728, 667, 627; HR-ESI-MS: m/z (%): 184.0979 (78, [M–H]+, calcd for + + + C9H14NO3 : 184.0979), 140.1078 (100, [M–COOH] , calcd for C8H14NO : 140.1075);

elemental analysis calcd (%) for C9H15NO3 (185.2): C 58.36, H 8.16, N 7.56; found: C 58.29, H 8.17, N 7.47.

1-(4-Methylpentyl)-1H-pyrrole-2,5-dione (124)

6 5 O 6 3 1 4 N 2 1 O

To maleic anhydride (402 mg, 4.10 mmol, 1 equiv.) in CH2Cl2 (6 mL), 128 (415 mg,

4.10 mmol, 1 equiv.) in CH2Cl2 (2 mL) was added. The mixture was stirred at RT for 3 h, then cooled to 0 °C. DMF (14 L) and oxalyl chloride (390 L, 4.51 mmol, 1.1 equiv.) were added dropwise. The mixture was stirred at RT for 22 h and evaporated in vacuo. The oily residue was dissolved in CH2Cl2 (4 mL) and Et3N (630 L, 4.51

mmol, 1.1 equiv.) was added. After 3 h at RT, the mixture was diluted with CH2Cl2, washed with aq. 1 M HCl solution, dried over MgSO4, filtrated, and evaporated in

vacuo. Purification by CC (SiO2; CH2Cl2) gave 124 (398 mg, 54%) as a white solid. 1 Rf = 0.50 (CH2Cl2); m.p. 40 °C; H NMR (300 MHz, CDCl3): 0.85 (d, J = 6.6, 6H, H-C(6)), 1.11–1.18 (m, 2H, H-C(4)), 1.47–1.61 (m, 3H, H-C(3), H-C(5)), 3.48 (t, J = 13 7.4, 2H, H-C(2)), 6.67 (s, 2H, H-C(1)); C NMR (75 MHz, CDCl3): 22.6, 26.6, 27.8, 35.9, 38.3, 134.2, 171.0; IR (ATR): 3363 (Br.), 2956, 1763, 1700, 1685, 1609, 1459, 1436, 1402, 1343, 1308, 1249, 1178, 1143, 1106, 1050, 1031, 996, 941, 903, 817, 781, 760, 740, 653, 631; HR-EI-MS: m/z (%): 181.1098 (45, [M]+, calcd for + + C10H15NO2 : 181.1098), 110.0236 (100, [M–CH2CH2CH(CH3)2] , calcd for + C5H4NO2 : 110.0242), 99.0321 (49); elemental analysis calcd (%) for C10H15NO2 (181.2): C 66.27, H 8.34, N 7.73; found: C 66.25, H 8.31, N 7.71.

171 8. Experimental ______4-Methylpentyl 4-Methylbenzenesulfonate (126)

1 3 5 2 O O 6 4 S 7 1 O 6 7 8

4-Methyl-1-pentanol (125) (1.0 g, 9.79 mmol, 1 equiv.) was dissolved in CH2Cl2

(14 mL) at RT under N2. TsCl (2.1 g, 10.7 mmol, 1.1 equiv.) and a catalytic amount of DMAP were added. The mixture was stirred for 1 d at RT. The solution was

diluted with CH2Cl2, washed with sat. aq. NaHCO3 solution and water, and dried over

Na2SO4. The solvents were evaporated in vacuo to give 126 (2.0 g, 82%) as yellow oil. 1 Rf = 0.59 (CH2Cl2); H NMR (300 MHz, CDCl3): 0.83 (d, J = 6.6, 6H, H-C(1)), 1.13– 1.20 (m, 2H, H-C(3)), 1.41–1.52 (m, 1H, H-C(2)), 1.58–1.68 (m, 2H, H-C(4)), 2.45 (s, 3H, H-C(8)), 4.00 (t, J = 6.6, 2H, H-C(5)), 7.34 (d, J = 8.0, 2H, H-C(7)), 7.77– 13 7.81 (m, 2H, H-C(6)); C NMR (100 MHz, CDCl3): 21.8, 22.5, 26.9, 27.7, 34.5, 71.1, 128.0, 130.0, 133.6, 144.7; IR (ATR): 3673, 2956, 2870, 2193, 2165, 1715, 1653, 1598, 1467, 1356, 1211, 1188, 1173, 1120, 1097, 1038, 1019, 965, 913, 833,

814, 790, 736, 706, 663, 630; elemental analysis calcd for C13H20O3S (256.37): C 60.91, H 7.86; found: C 60.76, H 7.84.

2-(4-Methylpentyl)-1H-isoindole-1,3(2H)-dione (127)

7 6 O 7 1 4 2 6 N 5 3 1 O Potassium phthalimide (2.18 g, 11.7 mmol, 1.5 equiv.) was added to 126 (2.00 g,

7.80 mmol, 1 equiv.) in DMF (39 mL) at RT under N2. The mixture was heated to reflux for 24 h. Water (40 mL) was added, and the pH was set to 2 with conc. HCl. The aq. phase was extracted with EtOAc. The combined org. phases were dried over

Na2SO4, and the solvents were evaporated in vacuo. The product was purified by CC

(SiO2; pentane/EtOAc 8:1  2:1  EtOAc) giving 127 (1.32 g, 73%) as yellow oil. 1 Rf = 0.75 (CH2Cl2); H NMR (400 MHz, CDCl3): 0.87 (m, 6H, H-C(1)), 1.20–1.25 (m, 2H, H-C(3)), 1.54–1.71 (m, 3H, H-C(2), H-C(4)), 3.66 (t, J = 7.2, 2H, H-C(5)),

172 8. Experimental ______7.68–7.72 (m, 2H, H-C(7)), 7.81–7.86 (m, 2H, H-C(6)); 13C NMR (100 MHz,

CDCl3): 22.6, 26.7, 27.9, 36.1, 38.4, 123.3, 132.4, 134.0, 168.6; IR (ATR): 2953, 2869, 2165, 1773, 1703, 1677, 1616, 1466, 1437, 1394, 1368, 1334, 1240, 1187, 1172, 1062, 995, 957, 880, 794, 718, 692, 670, 659, 620; HR-EI-MS: m/z (%): + + 231.1255 (49, [M] , calcd for C14H17NO2 : 231.1254), 160.0380 (100, [M– + + (CH3)2CH(CH2)2] , calcd for C9H6NO2 : 160.0399).

4-Methyl-1-pentanamine (128)

5

2 4 H2N 1 3 5

To a suspension of LiAlH4 (0.94 g, 24.7 mmol, 1.5 equiv.) in dry Et2O (16 mL) was

added isocapronitrile (129) (2.9 mL, 16.5 mmol, 1 equiv.) in Et2O (6 mL). The mixture was heated to reflux. After 4.5 h, the mixture was cooled down and treated

with sat. aq. NH4Cl solution. Et2O was added to the suspension and decanted. The

org. phase was dried over MgSO4, filtrated, and evaporated in vacuo to give 128 (415 mg, 25%) as colorless oil. 1 H NMR (300 MHz, CDCl3): 0.86 (s, 3H, H-C(5)), 0.88 (s, 3H, H-C(5)), 1.22–1.16

(m, 2H, H-C(3)), 1.32 (s, 2H, NH2), 1.46–1.37 (m, 2H, H-C(2)), 1.52 (td, J = 12.4, 13 5.8, 1H, H-C(4)), 2.65 (t, J = 7.1, 2H, H-(1)); C NMR (75 MHz, CDCl3): 22.9, 28.2, 32.0, 36.4, 42.8.

1-(3-Cyclohexylpropyl)-1H-pyrrole-2,5-dione (130)

7 1 O 6 8 1 3 5 N 7 2 4 6 O

Compound 135 (491 mg, 2.05 mmol, 1 equiv.) was dissolved in CH2Cl2 (5 mL). DMF (1.9 L) and oxalyl chloride (194 L, 2.26 mmol, 1.1 equiv.) were added dropwise to the solution at 0 °C. The mixture was stirred for 14.5 d and evaporated in vacuo. The oily residue was dissolved in CH2Cl2 (2.3 mL), Et3N (314 L, 2.26 mmol, 1.1 equiv.) was added dropwise, and the mixture was stirred for 5 d. The mixture was diluted with CH2Cl2, and washed with aq. 1 M HCl solution (4 x 90 mL).

173 8. Experimental ______

The org. phase was dried over Na2SO4, filtrated, and evaporated in vacuo to give compound 130 (427 mg, 94%) as a dark brown solid. 1 Rf = 0.46 (CH2Cl2); m.p. 62–71°C; H NMR (400 MHz, CDCl3): 0.84–0.91 (m, 2H), 1.14–1.26 (m, 6H), 1.54–1.70 (m, 7H, H-C(3)), 3.50 (t, J = 7.2, 2H, H-C(2)), 6.68 (s, 13 2H, H-C(1)); C NMR (100 MHz, CDCl3): 26.1, 26.5, 26.8, 33.4, 34.5, 37.4, 38.4, 134.2, 171.0; IR (ATR): 3093, 2920, 2850, 1696, 1439, 1410, 1374, 1338, 1309, 1260, 1221, 1164, 1124, 1028, 974, 890, 842, 790, 734, 692, 648, 622; HR-EI-MS: + + m/z (%): 221.1413 (73, [M] , calcd for C13H19NO2 : 221.1411), 122.1090 (81), + + 111.0310 (64, [M–C5H4NO2] ), 110.0232 (71, [M–CH2CH2C6H11] ), 81.0692 (62), 55.0570 (100).

3-Cyclohexylpropyl 4-Methylbenzenesulfonate (132)[206]

1 2 3 9 O 8 10 2 S 5 3 O 7 O 9 4 6 8 A solution of 3-cyclohexyl-1-propanol (4.4 mL, 28.1 mmol, 1.0 equiv.) and

p-toluene-4-sulfonylchloride (5.1 g, 26.7 mmol, 1.0 equiv.) in CHCl3 (12.2 mL) was cooled to 0 °C. Pyridine (3.9 mL, 47.8 mmol, 1.7 equiv.) was added dropwise to the mixture keeping the temperature under 5 °C. After 5 h at 0 °C, the mixture was poured onto ice (23 g) and conc. HCl (7.40 mL) was added. The aq. phase was

extracted with Et2O. The combined org. phases were washed with brine, dried over

Mg2SO4, and evaporated in vacuo. Purification by CC (SiO2; pentane/CH2Cl2 1:1) gave compound 132 (4.2 g, 50%) as a colorless oil. 1 Rf = 0.44 (pentane/EtOAc 1:1); H NMR (300 MHz, CDCl3): 0.75–0.88 (m, 2H), 1.08–1.27 (m, 6H), 1.59–1.68 (m, 7H), 2.45 (s, 3H, H-C(1)), 4.00 (t, J = 6.6, 2H, H- C(4)), 7.32–7.36 (m, 2H, H-C(2)), 7.77–7.81 (m, 2H, H-C(3)); 13C NMR (100 MHz,

CDCl3): 21.7, 26.4, 26.7, 31.1, 33.0, 33.3, 37.2, 71.2, 128.0, 129.9, 133.5, 144.7; IR (ATR): 2922, 2850, 1599, 1514, 1449, 1358, 1307, 1291, 1256, 1212, 1188, 1174, 1098, 1037, 970, 952, 926, 880, 828, 813, 778, 733, 705, 663, 630.

174 8. Experimental ______2-(3-Cyclohexylpropyl)-1H-isoindole-1,3(2H)-dione (133)[207]

1 2 O 8 1 7 9 4 6 2 N 8 3 5 7 O To a solution of potassium phthalimide (3.9 g, 13.9 mmol, 1.0 equiv.) and DMF (50 mL), compound 132 (4.1 g, 13.9 mmol, 1.0 equiv.) was added. The mixture was stirred at 85 °C for 18 h, and diluted with water (50 mL) and EtOAc (50 mL). The pH was set to 2–3 using conc. HCl. The aq. phase was extracted with EtOAc (3 x 50

mL). The combined org. phases were dried over Na2SO4. The mixture was evaporated in vacuo and purified by CC (SiO2; pentane/CH2Cl2 3:1) to give compound 133 (3.1 g, 82%) as a light yellow solid. [207] 1 Rf = 0.25 (pentane/CH2Cl2 3:1); m.p. 49–50 °C (Lit. : 51 °C); H NMR (300 MHz,

CDCl3): 0.83–0.90 (m, 2H), 1.08–1.25 (m, 6H), 1.60–1.72 (m, 7H), 3.65 (t, J = 7.5, 13 2H, H-C(3)), 7.67–7.74 (m, 2H), 7.82–7.88 (m, 2H); C NMR (100 MHz, CDCl3): 26.1, 26.5, 26.8, 33.4, 34.6, 37.5, 38.5, 123.3, 132.4, 133.9, 168.6; IR (ATR): 3058, 2921, 2849, 1773, 1711, 1615, 1466, 1438, 1396, 1361, 1188, 1071, 1004, 868, 793, + + 719, 625; HR-EI-MS: m/z (%): 271.1564 (85, [M] , calcd for C17H21NO2 : 271.1567), + + 160.0402 (100, [M–CH2CH2C6H11] , calcd for C9H6NO2 : 160.0399).

(3-Cyclohexylpropyl)amine (134)[207]

6 5 7 2 4 H2N 6 1 3 5

Compound 133 (1.9 g, 7.1 mmol, 1 equiv.) was dissolved in CH2Cl2 (77 mL) and MeOH (77 mL). Hydrazine monohydrate (3.5 mL, 71.9 mmol, 10 equiv.) was added, the mixture stirred for 4 d at RT, and evaporated in vacuo. The mixture was taken up

in aq. 1 M NaOH solution and extracted with CH2Cl2 (5 x 250 mL). The combined

org. phases were dried over Na2SO4, filtrated, and evaporated in vacuo. Compound 134 (0.4 g, 43%) was obtained as an orange-brown oil. 1 H NMR (300 MHz, CDCl3): 0.86–0.93 (m, 2H), 1.15–1.28 (m, 6H), 1.39–1.49 (m, 4H), 1.62–1.72 (m, 5H), 2.66 (t, J = 7.2, 2H, H-C(1)); HR-EI-MS: m/z (%): 141.1508 + + (6, [M] , calcd for C9H19N : 141.1512), 81.0718 (10), 29.9490 (100).

175 8. Experimental ______(2Z)-4-[(3-Cyclohexylpropyl)amino]-4-oxo-2-butenoic Acid (135) O 8 1 OH 7 9 H 4 6 2 N 8 3 5 7 O Amine 134 (430 mg, 3.05 mmol, 1 equiv.) was added dropwise to a solution of maleic

anhydride (298 mg, 3.05 mmol, 1 equiv.) in CH2Cl2 (7.7 mL) at 0 °C. The mixture was stirred for 4 d at RT. The mixture was evaporated in vacuo to give crude 135 (858 mg, quant.) as a light brown solid. 1 M.p. 92–99 °C; H NMR (300 MHz, CDCl3): 0.85–0.93 (m, 2H), 1.15–1.28 (m, 6H), 1.54–1.71 (m, 7H, H-C(4)), 3.34 (q, J = 7.2, 2H, H-C(3)), 6.20 (d, J = 12.9, 1H, H-C(1)), 6.34 (d, J = 12.6, 1H, H-C(2), 6.93 (br. s, 1H, NH); IR (ATR): 3714, 3239, 3037, 2918, 2849, 1851, 1780, 1700, 1633, 1586, 1463, 1373, 1289, 1240, 1190, 1174, 1151, 1069, 1018, 955, 887, 848, 765, 744, 696, 636, 623; HR-EI-MS: m/z (%): + + + 221.1410 (6, [M–H2O] , calcd for C13H19NO2 : 221.1411), 156.0648 (7, [M–C6H11] ), 129.0412 (4).

1-Methyl-4-aza-1-azoniabicyclo[2.2.2]octane Iodide (138)[208] N N 1 3 2 I 1,4-Diaza[2.2.2]bicyclooctane (137) (200 mg, 1.78 mmol, 1 equiv.) was dissolved in

EtOAc (3.3 mL) under N2 and cooled down to 0 °C. MeI (111 L, 1.78 mmol, 1 equiv.) was added dropwise and the mixture stirred at RT for 10 min. The precipitated product was filtrated and washed with EtOAc, giving 138 (282 mg, 62%) as a white solid. [208] 1 M.p. = 211 °C (Lit. : 213 °C); H NMR (300 MHz, CD3OD): 3.06 (s, 3H, H-C(3)), 3.20 (br. t, J = 7.5, 6H), 3.38 (br. t, J = 7.5, 6H); HR-EI-MS: m/z (%): 127.1231 (80, + + [M–I] , calcd for C7H15N2 : 127.1230), 42.0446 (100).

176 8. Experimental ______{3-[(3aRS,4RS,8aSR,8bSR)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3-dioxo- octahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]propyl}- (trimethyl)phosphonium Bromide ((±)-139) O 8 12 9 H 11 8a 7 12 N 8b 3a P 10 N 6 O H 4 12 Br N 3' O

2' S 1' Cl

A mixture of 1 M Me3P in THF (1.24 mL, 1.24 mmol, 20 equiv.) and (±)-57 (30 mg,

0.06 mmol, 1 equiv.) was stirred at RT under N2 for 16 h, treated with additional 1 M

Me3P in THF (1 mL), and stirred for 5 d. The precipitate was filtered and washed

with THF. The residue was dissolved in MeOH. Precipitation from Et2O gave (±)-139 (24 mg, 69%) as a white solid. 1 M.p. 217 °C (decomp); H NMR (400 MHz, CD3OD): 1.77–1.90 (m, 4H, H-C(7), H- C(8), H-C(10)), 1.88 (d, J = 14.5, 9H, H-C(12)), 2.10–2.25 (m, 4H, H-C(7), H-C(8), H-C(11)), 2.86 (ddd, J = 12.4, 8.1, 4.4, 1H, H-C(6)), 2.97 (ddd, J = 12.7, 8.6, 6.9, 1H, H-C(6)), 3.49 (dd, J = 8.2, 1.5, 1H, H-C(8b)), 3.51–3.63 (m, 2H, H-C(9)), 3.72–3.77 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.43 (d, J = 8.8, 1H, H-C(4)), 6.62 (s, 1H, H-C(3’)), 7.11 (d, J = 4.0, 1H, H-C(1’)), 7.42 (d, J = 4.0, 1H, H-C(2’)); 13 1 2 1 C NMR (100 MHz, CD3OD): 7.7 (d, JCP = 55.1), 20.9 (d, JCP = 3.7), 21.6 (d, JCP 3 = 54.1), 24.1, 30.2, 39.8 (d, JCP = 18.4), 50.65, 50.71, 52.4, 62.7, 69.6, 101.3, 128.1, 31 128.98, 129.02, 134.1, 164.9, 165.0, 177.4, 180.0; P NMR (172 MHz, CD3OD): 28.0; IR (ATR): 2956, 1769, 1694, 1604, 1527, 1475, 1424, 1398, 1370, 1336, 1306, 1280, 1245, 1203, 1159, 1103, 1063, 1029, 990, 973, 901, 873, 798, 768, 756, 727, 682, 667, 654, 623; HR-MALDI-MS: m/z (%): 482.1238 (39, [M–Br]+, calcd for 37 + + C22H28 ClN3O3PS : 482.1248), 480.1272 (100, [M–Br] , calcd for 35 + C22H28 ClN3O3PS : 480.1272).

177 8. Experimental ______{3-[(3aRS,4SR,8aRS,8bSR)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3-dioxo- octahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]propyl}- (trimethyl)phosphonium Bromide ((±)-140) O 8 12 9 H 11 8a 7 12 N 8b 3a P 10 N 6 O 4 12 H Br N 3' O

2' S 1' Cl

A mixture of (±)-60 (20 mg, 0.04 mmol, 1 equiv.) and 1 M Me3P in THF (50 L,

0.05 mmol, 1.2 equiv.) in THF (1.5 mL) under N2 was stirred at 50 °C for 24 h, and

treated with additional 1 M Me3P in THF (500 L, 0.5 mmol, 12 equiv.). The mixture

was stirred at 50 °C 96 h, further 1 M Me3P in THF (700 L, 0.7 mmol, 17 equiv.) was added, and the mixture stirred for 77 h. The precipitated product was filtered and washed with THF. Compound (±)-140 (6 mg, 26%) was obtained as a white solid. 1 M.p. 120 °C (decomp); H NMR (400 MHz, CD3OD): 1.88 (d, J = 14.5, 9H, H-C(12)), 1.84–1.90 (m, 5H, H-C(7), H-C(8), H-C(10)), 1.97–2.06 (m, 1H, H-C(8)), 2.22–2.29 (m, 2H, H-C(11)), 2.66–2.72 (m, 1H, H-C(6)), 3.05–3.11 (m, 1H, H-C(6)), 3.62 (td, J = 6.9, 0.9, 2H, H-C(9)), 3.69 (dd, J = 9.0, 9.0, 1H, H-C(8b)), 3.85–3.91 (m, 2H, H-C(3a), H-C(8a)), 4.46 (d, J = 4.9, 1H, H-C(4)), 6.76 (s, 1H, H-C(3’)), 7.10 (d, 13 J = 4.0, 1H, H-C(1’)), 7.44 (d, J = 4.0, 1H, H-C(2’)); C NMR (100 MHz, CD3OD): 1 2 1 3 7.8 (d, JCP = 55.2), 20.9 (d, JCP = 3.3), 21.9 (d, JCP = 53.9), 25.5, 27.1, 39.9 (d, JCP = 18.7), 49.8, 53.2, 53.9, 63.3, 67.6, 99.6, 128.2, 128.95, 129.02, 134.3, 165.7, 167.0, 31 178.7, 179.3; P NMR (161 MHz, CD3OD): 28.0; IR (ATR): 2966, 2895, 1771, 1694, 1604, 1472, 1399, 1354, 1299, 1243, 1205, 1155, 1063, 1036, 980, 898, 782, 739, 670; HR-MALDI-MS: m/z (%): 482.1234 (37, [M–Br]+, calcd for 37 + + C22H28 ClN3O3PS : 482.1248), 480.1265 (100, [M–Br] , calcd for 35 + C22H28 ClN3O3PS : 480.1272).

178 8. Experimental ______3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-1,2-oxazol-3-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Iodide ((±)-145) O 8 12 9 H 11 8a 7 12 N 8b N 10 3a N 6 O 4 12 H I N 3' O

2' S 1' Cl

A mixture of 4.2 M Me3N in EtOH (90 L, 0.38 mmol, 20 equiv.) and (±)-146 (10 mg, 0.02 mmol, 1 equiv.) in EtOH (1 mL) was stirred at RT for 3 d. The mixture was concentrated in vacuo. The residue was dissolved in MeOH. Precipitation from

Et2O gave (±)-145 (10 mg, 90%) as an off-white solid. 1 M.p. 140 °C (decomp); H NMR (600 MHz, CD3OD): 1.81–1.88 (m, 2H, H-C(7), H-C(8)), 2.02–2.08 (m, 2H, H-C(10)), 2.11–2.18 (m, 2H, H-C(7), H-C(8)), 2.87 (ddd, J = 12.7, 8.3, 4.4, 1H, H-C(6)), 2.95–2.99 (m, 1H, H-C(6)), 3.15 (s, 9H, H-C(12)), 3.33–3.36 (m, 2H, H-C(11)), 3.49 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.51–3.61 (m, 2H, H-C(9)), 3.73–3.76 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.43 (d, J = 8.8, 1H, H-C(4)), 6.63 (s, 1H, H-C(3’)), 7.11 (d, J = 4.0, 1H, H-C(1’)), 7.42 (d, J 13 = 4.0, 1H, H-C(2’)); C NMR (150 MHz, CDCl3): 22.8, 24.2, 30.2, 36.7, 50.6, 50.8, 52.4, 53.6 (t), 62.7, 65.3 (t), 69.6, 101.4, 128.1, 129.0, 129.1, 134.2, 164.8, 165.1, 177.5, 179.9; IR (ATR): 3419 (br.), 2951, 1773, 1693, 1603, 1528, 1474, 1402, 1349, 1317, 1241, 1181, 1143, 1066, 1031, 998, 963, 926, 899, 874, 796, 667, 627; HR- + 37 + MALDI-MS: m/z (%): 465.1562 (39, [M–I] , calcd for C22H28 ClN4O3S : 465.1537), + 35 + 463.1562 (100, [M–I] , calcd for C22H28 ClN4O3S : 463.1565).

179 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-1,2-oxazol-3-yl]-2-(3- iodopropyl)hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-146) O 8 9 H 11 8a 7 N 8b I 10 3a N 6 O H 4

N 3' O

2' S 1' Cl To a mixture of (±)-57 (20 mg, 0.041 mmol, 1 equiv.) in acetone (0.5 mL) at RT, NaI (11 mg, 0.074 mmol, 1.8 equiv.) was added. The mixture was stirred at RT for 21 h, additional NaI (9 mg, 0.060 mmol, 1.4 equiv.) was added, and the mixture was stirred for 22 h, and evaporated in vacuo. The residue was dissolved in EtOAc, washed with

aq. 0.5 M Na2S2O3 solution and water, dried over MgSO4, filtrated, and evaporated in vacuo to give crude (±)-146 (22 mg, quant.) as a white solid. 1 Rf = 0.20 (pentane/EtOAc 1:1); m.p. 131 °C; H NMR (400 MHz, CDCl3): 1.65–1.73 (m, 1H, H-C(7) or H-C(8)), 1.77–1.88 (m, 1H, H-C(7) or H-C(8)), 2.01–2.14 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.75 (ddd, J = 12.6, 8.3, 4.4, 1H, H-C(6)), 2.90 (ddd, J = 12.6, 9.1, 7.3, 1H, H-C(6)), 3.09 (td, J = 7.3, 0.5, 2H, H-C(11)), 3.32 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.55 (t, J = 6.9, 2H, H-C(9)), 3.60–3.66 (m, 2H, H-C(3a)), 3.76 (ddd, J = 9.5, 7.6, 1.6, 1H, H-C(8a)), 4.27 (d, J = 8.3, 1H, H-C(4)), 6.25 (s, 1H, H-C(3’)), 6.93 (d, J = 4.0, 1H, H-C(1’)), 7.28 (d, J = 4.0, 1H, H-C(2’)); 13C NMR (100 MHz,

CDCl3): 0.8, 23.5, 29.8, 31.8, 39.9, 49.3, 49.5, 50.8, 60.8, 68.3, 99.0, 126.5, 127.3, 127.9, 133.3, 163.6, 163.9, 175.1, 177.8; IR (ATR): 3135, 2923, 2899, 2861, 1765, 1691, 1599, 1528, 1473, 1433, 1422, 1398, 1358, 1347, 1312, 1281, 1262, 1243, 1208, 1177, 1145, 1119, 1075, 1064, 1001, 972, 952, 921, 898, 865, 852, 806, 739, 706, 666, 648; HR-MALDI-MS: m/z (%): 533.9921 (41, [M+H]+, calcd for 37 + + 35 + C19H20 ClIN3O3S : 531.9925), 531.9961 (100, [M+H] , calcd for C19H20 ClIN3O3S : + + 531.9953), 347.0248 (30, [M–C7H3ClNOS] , calcd for C12H16IN2O2 : 347.0256).

180 8. Experimental ______3-[(3aSR,4RS,8aSR,8bRS)-1,3-Dioxo-4-[5-(2-thienyl)-3- isoxazolyl]octahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-148) O 8 12 9 H 11 8a 7 12 N 8b N 10 3a N 6 O 4 12 H Br N 4' O

3' S 2' 1'

A micture of 4.2 M Me3N in EtOH (170 L, 0.71 mmol, 20 equiv.) and crude (±)-185 (16 mg, 0.04 mmol, 1 equiv.) in EtOH (1 mL) was stirred at RT for 72 h. The mixture was evaporated in vacuo. The residue was dissolved in MeOH. Precipitation from Et2O gave (±)-148) (13 mg, 72%) as a white solid. 1 M.p. 221 °C (decomp); H NMR (400 MHz, CD3OD): 1.79–1.91 (m, 2H, H-C(7), H-C(8)), 2.06 (m, 2H, H-C(10)), 2.12–2.20 (m, 2H, H-C(7), H-C(8)), 2.88 (ddd, J = 12.5, 8.2, 4.3, 1H, H-C(6)), 2.95–3.03 (m, 1H, H-C(6)), 3.15 (s, 9H, H-C(12)), 3.34– 3.38 (m, 2H, H-C(11)), 3.49 (dd, J = 8.1, 1.6, 1H, H-C(8b)), 3.51–3.63 (m, 2H, H-C(9)), 3.74–3.78 (m, 1H, H-C(8a)), 3.86 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.44 (d, J = 8.8, 1H, H-C(4)), 6.62 (s, 1H, H-C(4’)), 7.19 (dd, J = 5.0, 3.7, 1H, H-C(2’)), 7.59 (dd, J = 3.7, 1.1, 1H, H-C(3’)), 7.65 (dd, J = 5.0, 1.1, 1H, H-C(1’)); 13C NMR (100

MHz, CD3OD): 22.8, 24.2, 30.2, 36.7, 50.8 (C(3a), C(8b)), 52.4, 53.6 (t), 62.7, 65.3 (t), 69.5, 101.0, 128.5, 129.4, 129.7, 130.1, 164.7, 166.4, 177.4, 179.9; IR (ATR): 2965, 2884, 1768, 1698, 1601, 1474, 1400, 1357, 1318, 1181, 1145, 1063, 967, 923, 718, 643; HR-MALDI-MS: m/z (%): 429.1951 (100, [M–Br]+, calcd for + C22H29N4O3S : 429.1955).

181 8. Experimental ______3-[(3aS,4R,8aS,8bR)-4-[5-(5-Methoxy-2-thienyl)-1,2-oxazol-3-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((+)-150) O 8 12 9 H 11 8a 7 12 N 8b N 10 3a N 6 O H 4 12 Br N 3' O

2' S 1' O

To a mixture of (+)-186 (7.0 mg, 0.02 mmol, 1 equiv.) in EtOH (0.8 mL) at RT, 4.2 M

Me3N in EtOH (90 L, 0.38 mmol, 26 equiv.) was added. The mixture was stirred at RT for 4 d and evaporated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (+)-150 (7.9 mg, quant.) as a light brown solid. 25 1 M.p. 174 °C (decomp); []D : +73.4 (c = 0.3, CH3OH); H NMR (600 MHz,

CD3OD): 1.80–1.89 (m, 2H, H-C(7), H-C(8)), 2.00–2.09 (m, 2H, H-C(10)), 2.10–2.19 (m, 2H, H-C(7), H-C(8)), 2.87 (ddd, J = 12.6, 8.3, 4.3, 1H, H-C(6)), 2.97 (ddd, J = 12.6, 8.7, 6.9, 1H, H-C(6)), 3.14 (s, 9H, H-C(12)), 3.35 (dd, J = 9.5, 7.6, 2H, H-C(11)), 3.48 (dd, J = 8.2, 1.6, 1H, H-C(8b)), 3.51–3.61 (m, 2H, H-C(9)), 3.74–3.77 (m, 1H, H-C(8a)), 3.83 (dd, J = 8.6, 8.6, 1H, H-C(3a)), 3.96 (s, 3H, OMe), 4.40 (d, J = 8.9, 1H, H-C(4)), 6.36 (d, J = 4.1, 1H, H-C(1’)), 6.42 (s, 1H, H-C(3’)), 7.24 (d, 13 J = 4.1, 1H, H-C(2’)); C NMR (150 MHz, CD3OD): 22.8, 24.1, 30.2, 36.7, 50.7, 50.8, 52.4, 53.6 (t), 61.1, 62.7, 65.3 (t), 69.5, 99.2, 106.0, 116.0, 127.3, 164.5, 166.6, 170.8, 177.4, 179.9; IR (ATR): 2967, 1769, 1697, 1614, 1537, 1494, 1422, 1397, 1365, 1318, 1207, 1179, 1142, 1061, 1015, 979, 919, 898, 869, 822, 772, 725; HR- + + ESI-MS: m/z (%): 459.2061 (100, [M–Br] , calcd for C23H31N4O4S : 459.2061).

182 8. Experimental ______3-[(3aR,4S,8aR,8bS)-4-[5-(5-Methoxy-2-thienyl)-1,2-oxazol-3-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((–)-150) O 8 12 9 H 11 8a 7 12 N 8b N 10 3a N 6 O H 4 12 Br N 3' O

2' S 1' O

To a mixture of (–)-186 (6.3 mg, 0.01 mmol, 1 equiv.) in EtOH (0.8 mL) at RT, 4.2 M

Me3N in EtOH (90 L, 0.38 mmol, 29 equiv.) was added. The mixture was stirred at RT for 4 d and evaporated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (–)-150 (6.3 mg, 89%) as a light brown solid. 25 1 M.p. 181 °C (decomp); []D : –114.9 (c = 0.4, CH3OH); H NMR (600 MHz,

CD3OD): 1.80–1.90 (m, 2H, H-C(7), H-C(8)), 2.01–2.07 (m, 2H, H-C(10)), 2.11–2.17 (m, 2H, H-C(7), H-C(8)), 2.87 (ddd, J = 12.6, 8.3, 4.4, 1H, H-C(6)), 2.97 (ddd, J = 12.6, 8.6, 6.8, 1H, H-C(6)), 3.14 (s, 9H, H-C(12)), 3.34–3.36 (m, 2H, H-C(11)), 3.48 (dd, J = 8.3, 1.7, 1H, H-C(8b)), 3.51–3.61 (m, 2H, H-C(9)), 3.74–3.77 (m, 1H, H-C(8a)), 3.83 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 3.96 (s, 3H, OMe), 4.40 (d, J = 8.9, 1H, H-C(4)), 6.36 (d, J = 4.1, 1H, H-C(1’)), 6.42 (s, 1H, H-C(3’)), 7.24 (d, J = 4.1, 13 1H, H-C(2’)); C NMR (150 MHz, CD3OD): 22.8, 24.1, 30.2, 36.7, 50.7, 50.8, 52.4, 53.6 (t), 61.1, 62.7, 65.3 (t), 69.5, 99.2, 106.0, 116.0, 127.3, 164.5, 166.6, 170.8, 177.4, 179.9; IR (ATR): 2955, 1768, 1698, 1614, 1537, 1495, 1422, 1396, 1366, 1318, 1207, 1178, 1142, 1061, 1015, 978, 918, 868, 823, 806, 770, 725; HR-ESI-MS: + + m/z (%): 459.2063 (100, [M–Br] , calcd for C23H31N4O4S : 459.2061), 338.3418 (48).

183 8. Experimental ______3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Fluoro-2-thienyl)-1,2-oxazol-3-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-151) O 8 12 9 H 11 8a 7 12 N 8b 3a N 10 N 6 O 4 12 H Br N 3' O

2' S 1' F To a mixture of (±)-187 (10 mg, 0.02 mmol, 1 equiv.) in EtOH (1 mL) at RT, 4.2 M

Me3N in EtOH (102 L, 0.43 mmol, 20 equiv.) was added. The mixture was stirred at RT for 2.5 d and evaporated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (±)-151 (8 mg, 73%) as a light yellow solid. 1 M.p. 196 °C (decomp); H NMR (600 MHz, CD3OD): 1.81–1.90 (m, 2H, H-C(7), H-C(8)), 2.01–2.10 (m, 2H, H-C(10)), 2.10–2.19 (m, 2H, H-C(7), H-C(8)), 2.87 (ddd, J = 12.7, 8.3, 4.3, 1H, H-C(6)), 2.97 (ddd, J = 12.6, 8.7, 6.9, 1H, H-C(6)), 3.15 (s, 9H, H-C(12)), 3.37–3.33 (m, 2H, H-C(11)), 3.48 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.56 (dtd, J = 24.5, 12.8, 5.6, 2H, H-C(9)), 3.73–3.76 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.5, 3 8.5, 1H, H-C(3a)), 4.42 (d, J = 8.8, 1H, H-C(4)), 6.59 (s, 1H, H-C(3’)), 6.72 (dd, JHH 4 3 3 13 = 4.2, JHF = 1.9, 1H, H-C(2’)), 7.28 (t, JHH = 3.9, JHF = 3.9, 1H, H-C(1’)); C NMR

(150 MHz, CD3OD): 22.7, 24.1, 30.2, 36.7, 50.6, 50.7, 52.3, 53.6 (t), 62.6, 65.2 (t), 6 2 3 69.5, 100.8 (d, JCF = 2.5), 110.3 (d, JCF = 11.3), 119.2 (d, JCF = 4.3), 125.6 (d, 4 1 19 JCF = 4.1), 164.7, 165.5, 168.6 (d, JCF = 293.0), 177.4, 179.9; F NMR (282 MHz,

CD3OD): –127.8; IR (ATR): 3010, 2951, 1773, 1696, 1616, 1551, 1494, 1438, 1397, 1351, 1318, 1180, 1144, 1087, 1063, 1012, 968, 929, 900, 869, 793, 733, 713, 668, + + 640; HR-MALDI-MS: m/z (%): 447.1867 (100, [M–Br] , calcd for C22H28FN4O3S : 447.1861).

184 8. Experimental ______3-[(3aRS,4RS,8aSR,8bSR)-4-[5-(5-Bromo-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-152) O 8 12 9 H 11 8a 7 12 N 8b 3a N 10 N 6 O H 4 12 Br N 3' O

2' S 1' Br To a solution of (±)-188 (19 mg, 0.04 mmol, 1 equiv.) in EtOH (1.0 mL) at RT, 4.2 M

Me3N in EtOH (171 L, 0.72 mmol, 20 equiv.) was added. The mixture was stirred at RT for 9 d. The mixture was evaporated in vacuo, dissolved in MeOH, and

precipitation from Et2O gave (±)-152 (9 mg, 43%) as an off-white solid. 1 M.p. 202 °C (decomp); H NMR (600 MHz, CD3OD): 1.79–1.91 (m, 2H, H-C(7), H-C(8)), 2.01–2.09 (m, 2H, H-C(10)), 2.12–2.19 (m, 2H, H-C(7), H-C(8)), 2.87 (ddd, J = 12.7, 8.3, 4.3, 1H, H-C(6)), 2.97 (ddd, J = 12.7, 8.7, 7.0, 1H, H-C(6)), 3.14 (s, 9H, H-C(12)), 3.35 (dd, J = 10.4, 6.7, 2H, H-C(11)), 3.49 (dd, J = 8.3, 1.6, 1H, H-C(8b)), 3.51–3.61 (m, 2H, H-C(9)), 3.73–3.76 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.43 (d, J = 8.8, 1H, H-C(4)), 6.64 (s, 1H, H-C(3’)), 7.23 (d, J = 4.0, 1H, 13 H-C(1’)), 7.39 (d, J = 3.9, 1H, H-C(2’)); C NMR (150 MHz, CD3OD): 22.8, 24.2, 30.2, 36.7, 50.6, 50.8, 52.3, 53.6 (t), 62.7, 65.3 (t), 69.6, 101.4, 116.7, 129.0, 131.8, 132.8, 164.8, 165.1, 177.5, 179.9; IR (ATR): 2945, 1766, 1700, 1696, 1616, 1491, 1477, 1433, 1416, 1394, 1363, 1348, 1316, 1203, 1179, 1142, 1062, 1014, 968, 918, 868, 795, 736, 670, 655, 637; HR-MALDI-MS: m/z (%): 509.1039 (100, [M–Br]+, 81 + + calcd for C22H28 BrN4O3S : 509.1040), 507.1067 (4, [M–Br] , calcd for 79 + C22H28 BrN4O3S : 507.1060).

185 8. Experimental ______3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Iodo-2-thienyl)-1,2-oxazol-3-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-153) O 8 12 9 H 11 8a 7 12 N 8b 3a N 10 N 6 O H 4 12 Br N 3' O

2' S 1' I To a mixture of (±)-189 (9.4 mg, 0.02 mmol, 1 equiv.) in EtOH (0.8 mL) at RT, 4.2 M

Me3N in EtOH (90 L, 0.38 mmol, 23 equiv.) was added. The mixture was stirred at RT for 4 d and evaporated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (±)-153 (5.3 mg, 51%) as a light brown solid. 1 M.p. 214 °C (decomp); H NMR (600 MHz, CD3OD): 1.81–1.89 (m, 2H, H-C(7), H-C(8)), 2.02–2.09 (m, 2H, H-C(10)), 2.10–2.19 (m, 2H, H-C(7), H-C(8)), 2.87 (ddd, J = 12.7, 8.3, 4.3, 1H, H-C(6)), 2.97 (ddd, J = 12.6, 8.7, 6.9, 1H, H-C(6)), 3.14 (s, 9H, H-C(12)), 3.33–3.36 (m, 2H, H-C(11)), 3.48 (dd, J = 8.2, 1.5, 1H, H-C(8b)), 3.51– 3.61 (m, 2H, H-(9)), 3.73–3.76 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.5, 8.5, 1H, H- C(3a)), 4.43 (d, J = 8.8, 1H, H-C(4)), 6.62 (s, 1H, H-C(3’)), 7.26 (d, J = 3.8, 1H, H- 13 (1’)), 7.37 (d, J = 3.8, 1H, H-(2’)); C NMR (150 MHz, CD3OD): 22.7, 24.1, 30.2, 36.7, 50.6, 50.7, 52.3, 53.6 (t), 62.7, 65.3 (t), 69.5, 79.1, 101.3, 129.9, 136.0, 139.6, 164.7, 165.0, 177.4, 179.9; IR (ATR): 2964, 1767, 1694, 1615, 1471, 1430, 1397, 1351, 1317, 1179, 1141, 1064, 1013, 965, 921, 869, 820; HR-ESI-MS: m/z (%): + + 556.0960 (26), 555.0928 (100, [M–Br] , calcd for C22H28IN4O3S : 555.0921).

186 8. Experimental ______1-(5-Methoxy-2-thienyl)ethanone (156)[209] O O S 1 4 23

To AlCl3 (852 mg, 6.40 mmol, 1.5 equiv.) in CH2Cl2 (30 mL) at 0 °C under N2, acetyl chloride (310 L, 4.40 mmol, 1 equiv.) was added, and the mixture stirred at 0 °C for 30 min. This mixture was added dropwise to a solution of 2-methoxythiophene (440

L, 4.40 mmol, 1 equiv.) in CH2Cl2 (25 mL) at 0 °C under N2, and stirring was continued for 1 h at 0 °C. The reaction was quenched with water (0.5 mL). The org. phase was washed with sat. aq. NH4Cl solution, dried over MgSO4, filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2) gave 156 (234 mg, 34%) as a yellow oil. 1 Rf = 0.13 (CH2Cl2); H NMR (300 MHz, CDCl3): 2.46 (s, 3H, H-C(1)), 3.96 (s, 3H, H-C(4)), 6.25 (d, J = 4.3, 1H, H-C(2)), 7.44 (d, J = 4.3, 1H, H-C(3)); 13C NMR

(100 MHz, CDCl3): 25.5, 60.5, 106.1, 131.0, 133.1, 174.7, 190.1; IR (ATR): 3128, 2960, 2930, 1767, 1691, 1645, 1603, 1533, 1473, 1403, 1371, 1347, 1290, 1245, 1211, 1187, 1146, 1081, 1066, 1031, 1000, 927, 900, 871, 807, 751, 730, 704, 667, + + 638, 608; HR-EI-MS: m/z (%): 156.0240 (67, [M] , calcd for C7H8O2S : 156.0240), + + 140.9998 (100, [M–CH3] , calcd for C6H5O2S : 141.0010), 43.0219 (38).

1-(5-Fluoro-2-thienyl)ethanone (157)[210] O F S 3 12

To a solution of 160 (3.20 g, 12.9 mmol, 1 equiv.) in THF (32 mL) under N2 at

–78 °C was added 1.6 M nBuLi in hexanes (10.5 mL, 16.7 mmol, 1.3 equiv.) dropwise over 25 min. The mixture was stirred at –78 °C for 45 min. A solution of N-fluorobenzenesulfonimide (5.27 g, 16.7 mmol, 1.3 equiv.) in THF (14 mL) was added dropwise over 30 min. The mixture was slowly warmed to RT overnight.

After 19 h the mixture was poured onto cold sat. aq. NH4Cl solution. The aq. phase was extracted with Et2O. The org. phase was washed with water and brine, dried over

MgSO4, filtrated, and evaporated in vacuo. Repeated purification by CC (SiO2;

187 8. Experimental ______

pentane  pentane/CH2Cl2 3:2  CH2Cl2; CH2Cl2) gave 157 (540 mg, 22%) as a light brown oil. 1 Rf = 0.44 (CH2Cl2); H NMR (400 MHz, CDCl3): 2.48 (s, 3H, H-C(3)), 6.54 (dd, 3 4 3 3 JHH = 4.2, JHF = 1.3, 1H, H-C(2)), 7.38 (dd, JHH = 3.9, JHF = 3.9, 1H, H-C(1)); 13 2 3 C NMR (100 MHz, CDCl3): 25.6, 109.7 (d, JCF = 12.4), 130.6 (d, JCF = 4.8), 134.0 4 1 5 19 (d, JCF = 1.4), 172.2 (d, JCF = 299.3), 190.5 (d, JCF = 2.4); F NMR (376 MHz,

CDCl3): –117.2; IR (ATR): 3080, 1655, 1555, 1456, 1360, 1340, 1276, 1197, 1059, 1024, 963, 928, 791, 718, 640, 604; HR-EI-MS: m/z (%): 144.0040 (8, [M]+, calcd for + + + C6H5FOS : 144.0040), 128.9803 (13, [M–CH3] , calcd for C5H2FOS : 128.9805), + + 57.0152 (3, [FC(CH)2] , calcd for C3H2F ; 57.0141), 18.0101 (100); elemental

analysis calcd (%) for C6H5FOS (144.2): C 49.99, H 3.50; found: C 50.23, H 3.63.

2-(5-Bromo-2-thienyl)-2-methyl-1,3-dioxolane (160)

3 3

O O Br S 4 12 A mixture of 2-acetyl-5-bromothiophene (5.0 g, 24.4 mmol, 1 equiv.), ethylene glycol (5.4 mL, 97.3 mmol, 4 equiv.), and p-toluenesulfonic acid monohydrate (0.23 g,

1.2 mmol, 0.1 equiv.) in toluene (138 mL) was heated to reflux under N2 for 2.2 d using a Dean-Stark trap. The mixture was cooled to RT, and EtOAc (320 mL) and

water (160 mL) were added. The organic layer was washed with sat. aq. NaHCO3

solution and brine, dried over MgSO4, and evaporated in vacuo. Purification by CC

(SiO2; pentane/CH2Cl2 2:3) gave 160 (3.2 g, 52%) as a white solid. [211] 1 Rf = 0.59 (CH2Cl2); m.p. 87 °C (Lit. : 62–63 °C); H NMR (400 MHz, CDCl3): 1.73 (s, 3H, H-C(4)), 3.99 (m, 4H, H-C(3)), 6.79 (d, J = 4.0, 1H, H-C(2)), 6.89 (d, J = 13 4.0, 1H, H-C(1)); C NMR (100 MHz, CDCl3): 27.5, 65.1, 107.0, 111.9, 124.5, 129.8, 149.1; IR (ATR): 1789, 1644, 1521, 1407, 1353, 1316, 1271, 1213, 1086, 1036, 980, 926, 883, 804, 739, 681; HR-EI-MS: m/z (%): 249.9481 (4, [M]+, calcd for 81 + + 79 + C8H9 BrO2S : 249.9486), 247.9501 (5, [M] , calcd for C8H9 BrO2S : 247.9502), + 79 + 232.9261 (31, [M–CH3] , calcd for C7H6 BrO2S ; 232.9266), 188.9003 (12), 18.0676

(100); elemental analysis calcd (%) for C8H9BrO2S (248.0): C 38.57, H 3.64; found: C 38.64, H 3.62.

188 8. Experimental ______5-(2-Thienyl)-3-isoxazolecarbaldehyde (161) ON S O 1 4 H 2 3 5 PCC (1.55 g, 7.2 mmol, 1.5 equiv.) was added to a solution of 176 (0.87 g, 4.8 mmol,

1 equiv.) in CH2Cl2 (300 mL) at RT. The mixture was stirred at RT for 72 h, filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2) gave 161 (670 mg, 78%) as a white solid. 1 Rf = 0.51 (CH2Cl2); m.p. 65 °C; H NMR (300 MHz, CDCl3): 6.74 (s, 1H, H-C(4)), 7.16 (dd, J = 5.0, 3.7, 1H, H-C(2)), 7.52 (dd, J = 5.0, 1.2, 1H, H-C(1)), 7.59 (dd, J = 13 3.7, 1.2, 1H, H-C(3)), 10.16 (s, 1H, H-C(5)); C NMR (100 MHz, CDCl3): 96.1, 128.18, 128.21, 128.5, 129.3, 162.7, 167.2, 184.6; IR (ATR): 3114, 2848, 1703, 1586, 1451, 1410, 1357, 1253, 1179, 1050, 1010, 943, 908, 854, 816, 748, 714; HR-EI-MS: + + m/z (%): 179.0034 (50, [M] , calcd for C8H5NO2S : 179.0036), 110.9899 (100, [M– + + C3H2NO] , calcd for C5H3OS : 110.9905); elemental analysis calcd (%) for

C8H5NO2S (179.0): C 53.62, H 2.81, N 7.82; found: C 53.60, H 2.92, N 7.85.

5-(5-Methoxy-2-thienyl)-1,2-oxazole-3-carbaldehyde (162) O N O O S 5 1 4 23

To a solution of 177 (47 mg, 0.22 mmol, 1 equiv.) in CH2Cl2 (3 mL) at RT was added PCC (72 mg, 0.33 mmol, 1.5 equiv.). The mixture was stirred at RT for 4 d, filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2) gave 162 (31 mg, 66%) as a light yellow solid. 1 Rf = 0.40 (CH2Cl2); m.p. 104 °C; H NMR (300 MHz, CDCl3): 3.97 (s, 3H, H-C(1)), 6.26 (d, J = 4.1, 1H, H-C(2)), 6.55 (s, 1H, H-C(4)), 7.26 (d, J = 4.1, 1H, H-C(3)), 13 10.13 (s, 1H, H-C(5)); C NMR (100 MHz, CDCl3): 60.7, 94.1, 105.3, 114.2, 127.0, 162.6, 167.4, 170.1, 184.9; IR (ATR): 3138, 2924, 1750, 1707, 1590, 1534, 1479, 1455, 1424, 1416, 1355, 1258, 1238, 1217, 1180, 1148, 1066, 1051, 1014, 1003, 985, 926, 898, 808, 789, 773, 753, 683; HR-EI-MS: m/z (%): 209.0141 (100, [M]+, calcd + + + for C9H7NO3S : 209.0142), 141.0008 (57, [M–C3H2NO] , calcd for C6H5O2S : 141.0010).

189 8. Experimental ______5-(5-Fluoro-2-thienyl)-1,2-oxazole-3-carbaldehyde (163) O N F S 4 3 O 1 2

To a mixture of 178 (140 mg, 0.70 mmol, 1 equiv.) in CH2Cl2 (10 mL) at RT was added PCC (227 mg, 1.05 mmol, 1.5 equiv.). The mixture was stirred at RT for 46 h,

filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2) gave 163 (93 mg, 67%) as a white solid. 1 3 Rf = 0.44 (CH2Cl2); m.p. 82 °C; H NMR (300 MHz, CDCl3): 6.58 (dd, JHH = 4.2, 4 3 3 JHF = 1.6, 1H, H-C(2)), 6.66 (s, 1H, H-C(3)), 7.23 (dd, JHH = 3.9, JHF = 3.9, 1H, 13 6 H-C(1)), 10.14 (s, 1H, H-C(4)); C NMR (100 MHz, CDCl3): 95.6 (d, JCF = 2.1), 2 3 4 109.2 (d, JCF = 11.1), 117.0 (d, JCF = 4.7), 125.2 (d, JCF = 4.3), 162.6, 166.5, 168.2 1 19 (d, JCF = 296.3), 184.5; F NMR (376 MHz, CDCl3): –124.4; IR (ATR): 3379, 3120, 2900, 1736, 1712, 1595, 1547, 1479, 1450, 1372, 1350, 1274, 1248, 1204, 1130, 1042, 1016, 938, 900, 814, 792, 752, 715, 682, 638; HR-EI-MS: m/z (%): 196.9941 + + + (73, [M] , calcd for C8H4FNO2S : 196.9942), 128.9807 (100, [M–CHCNCHO] , calcd + + + for C5H2FOS : 128.9810), 100.9859 (10, [M–COCHCNCHO] , calcd for C4H2FS :

100.9861); elemental analysis calcd (%) for C8H4FNO2S (197.2): C 48.73, H 2.04, N 7.10; found: C 48.66, H 2.22, N 7.06.

5-(5-Bromo-2-thienyl)-3-isoxazolecarbaldehyde (164) ON O Br S 3 4 1 2 PCC (1.63 g, 7.55 mmol, 1.5 equiv.) was added to a solution of 179 (1.31 g,

5.04 mmol, 1 equiv.) in CH2Cl2 (80 mL). The mixture was stirred at RT for 19 h.

The mixture was filtrated and concentrated in vacuo. Purification by CC (SiO2;

CH2Cl2) gave 164 (1.28 g, 98%) as a light yellow solid. 1 Rf = 0.49 (CH2Cl2); m.p. 117 °C; H NMR (400 MHz, CDCl3): 6.70 (s, 1H, H-C(3)), 7.13 (d, J = 4.0, 1H, H-C(1)), 7.33 (d, J = 3.9, 1H, H-C(2)), 10.15 (s, 1H, H-C(4)); 13 C NMR (100 MHz, CDCl3): 96.2, 117.2, 128.3, 129.5, 131.4, 162.7, 165.9, 184.4; IR (ATR): 3134, 3105, 3087, 3033, 2874, 1808, 1706, 1585, 1521, 1451, 1412, 1364, 1339, 1242, 1203, 1171, 1074, 975, 944, 898, 818, 803, 747, 732, 680, 661; HR-EI-

190 8. Experimental ______

+ 81 + MS: m/z (%): 258.9120 (68, [M] , calcd for C8H4 BrNO2S : 258.9126), 256.9143 + 79 + + (65, [M] , calcd for C8H4 BrNO2S : 256.9141), 190.8985 (100, [C5H2BrOS] , calcd 81 + + 79 + for C5H2 BrOS : 190.8989), 188.9004 (96, [C5H2BrOS] , calcd for C5H2 BrOS :

188.9010); elemental analysis calcd (%) for C8H4BrNO2S (258.1): C 37.23, H 1.56, N 5.43; found: C 37.27, H 1.47, N 5.28.

5-(5-Methyl-2-thienyl)-3-isoxazolecarbaldehyde (165) O N 1 S 5

4 O 2 3 PCC (0.75 g, 3.46 mmol, 1.5 equiv.) was added to a solution of 180 (0.45 g,

2.30 mmol, 1 equiv.) in CH2Cl2 (34 mL) at RT. The mixture was stirred at RT for

6.5 d, filtrated, washed with CH2Cl2, and evaporated in vacuo. Purification by CC

(SiO2; pentane/CH2Cl2 1:4) gave 165 (0.35 g, 79%) as a white solid. 1 Rf = 0.58 (CH2Cl2); m.p. 108 °C; H NMR (300 MHz, CDCl3): 2.55 (d, J = 0.9, 3H, H-C(1)), 6.64 (s, 1H, H-C(4)), 6.81 (d, J = 3.6, 1H, H-C(2)), 7.38 (d, J = 3.6, 1H, 13 H-C(3)), 10.15 (s, 1H, H-C(5)); C NMR (75 MHz, CDCl3): 15.6, 95.2, 125.8, 126.8, 128.4, 144.8, 162.6, 167.3, 184.8; IR (ATR): 3135, 2920, 2358, 2342, 1701, 1592, 1531, 1472, 1446, 1434, 1380, 1366, 1253, 1203, 1183, 1160, 1066, 1006, 938, 900, 798, 752, 742, 685, 668; HR-EI-MS: m/z (%): 193.0192 (68, [M]+, calcd for + + + C9H7NO2S : 193.0192), 125.0049 (100, [M–CHCNCHO] , calcd for C6H5OS :

125.0056); elemental analysis calcd for C9H7NO2S (193.2): C 55.94, H 3.65, N 7.25; found: C 55.64, H 3.87, N 7.11.

Ethyl 4-(5-Methoxy-2-thienyl)-2,4-dioxobutanoate (167) O O O S O 1 4 6 23 O 5

To acetylthiophene 156 (500 mg, 3.20 mmol, 1 equiv.) in toluene (15 mL) under N2 at 0 °C, KOtBu (359 mg, 3.20 mmol, 1 equiv.) was added in portions. The mixture was stirred at 0 °C for 30 min, then diethyl oxalate (165 L, 3.84 mmol, 1.2 equiv.) was added. The mixture was stirred at RT for 4 d. The precipitated product was filtrated

and dissolved in EtOAc. The org. phase was washed with aq. 1 M HCl solution, dried

191 8. Experimental ______

over MgSO4, filtrated, and evaporated in vacuo to give crude 167 (374 mg, 46%) as a dark brown solid. 1 H NMR (300 MHz, CDCl3): 1.41 (t, J = 7.1, 3H, H-C(6)), 4.01 (s, 3H, H-C(1)), 4.39 (q, J = 7.1, 2H, H-C(5)), 6.33 (d, J = 4.4, 1H, H-C(2)), 6.82 (s, 1H, H-C(4)), 7.62 (d, + + J = 4.4, 1H, H-C(3)); HR-EI-MS: m/z (%): 256.0399 (18, [M] , calcd for C11H12O5S : + + 256.0400), 183.0113 (100, [M–COOCH2CH3] , calcd for C8H7O3S : 183.0116), + + 141.0004 (65, [M–CH2COCOOCH2CH3] , calcd for C6H5O2S : 141.0010).

Ethyl 4-(5-Fluoro-2-thienyl)-2,4-dioxobutanoate (168) O O F S O 5 3 4 12 O A solution of compound 157 (624 mg, 4.33 mmol, 1 equiv.) in toluene (17 mL) under

N2 was cooled to 0 °C, and KOtBu (486 mg, 4.33 mmol, 1 equiv.) was added in portions. Diethyl oxalate (220 L, 5.19 mmol, 1.2 equiv.) was added and the mixture stirred at RT for 26 h. The mixture was filtrated and the residue was dissolved in

EtOAc. The org. phase was washed with aq. 1 M HCl solution, dried over MgSO4, filtrated, and evaporated in vacuo to give 168 (650 mg, 61%) as a brown oil. Crude 168 was used in the next step without further purification. 1 H NMR (300 MHz, CDCl3): 1.40 (t, J = 7.1, 3H, H-C(5)), 4.39 (q, J = 7.1, 2H, 3 4 H-C(4)), 6.62 (dd, JHH = 4.4, JHF = 1.3, 1H, H-C(2), 6.85 (s, 1H, H-C(3)), 7.57 (dd, 3 3 19 JHH = 4.0, JHF = 4.0, 1H, H-C(1)); F NMR (376 MHz, CDCl3): –113.9.

Ethyl 4-(5-Bromo-2-thienyl)-2,4-dioxobutanoate (169) O O Br S O 3 5 1 2 O 4 KOtBu (1.64 g, 14.6 mmol, 1 equiv.) was added to 2-acetyl-5-bromothiophene

(3.00 g, 14.6 mmol, 1 equiv.) in toluene (75 mL) under N2 at 0 °C in four portions. After 10 min, diethyl oxalate (2.38 mL, 17.6 mmol, 1.2 equiv.) was added dropwise. After 6 h at RT, the reaction was complete and the mixture was

filtrated. The residue was dissolved in EtOAc and washed with aq. 1 M HCl

solution. The org. phase was dried over Na2SO4, filtrated, and evaporated in vacuo,

192 8. Experimental ______

affording crude 169 (4.07 g, 92%) as an orange solid. Crude 169 was used in the next step without further purification. 1 H NMR (300 MHz, CDCl3): 1.41 (t, J = 7.4, 3H, H-C(5)), 4.40 (q, J = 7.1, 2H, H-C(4)), 6.83 (s, 1H, H-C(3)), 7.16 (d, J = 4.1, 1H, H-C(1)), 7.59 (d, J = 4.1, 1H, H-C(2)); IR (ATR): 3090, 2978, 1726, 1626, 1570, 1519, 1468, 1429, 1398, 1361, 1317, 1262, 1245, 1112, 1076, 1030, 1004, 969, 929, 900, 865, 843, 826, 808, 769, + 81 + 684, 648; HR-EI-MS: m/z (%): 305.9390 (10, [M] , calcd for C10H9 BrO4S : + 79 + 305.9384), 303.9402 (10, [M] , calcd for C10H9 BrO4S : 303.9405), 232.9085 (100, + 81 + [M–CH3CH2OCO] , calcd for C7H4 BrO2S : 232.9095), 230.9113 (98, [M– + 79 + + CH3CH2OCO] , calcd for C7H4 BrO2S : 230.9115), 190.8987 (37, [M–C5H7O3] , 81 + + calcd for C5H2 BrOS : 190.8989), 188.9010 (37, [M–C5H7O3] , calcd for 79 + C5H2 BrOS : 188.9010); elemental analysis calcd (%) for C10H9BrO4S (305.2): C 39.36, H 2.97; found: C 39.41, H 3.02.

Ethyl 4-(5-Methyl-2-thienyl)-2,4-dioxobutanoate (170) O O 1 S O 4 6 2 3 O 5 KOtBu (2.1 g, 18.7 mmol, 0.87 equiv.) was added in portions to 2-acetyl-

5-methylthiophene (2.7 mL, 21.4 mmol, 1 equiv.) in toluene (40 mL) under N2 at 0 °C. After 10 min, diethyl oxalate (3.0 mL, 22.4 mmol, 1.1 equiv.) was added dropwise. The mixture was stirred at RT for 13 h. The mixture was filtrated, the precipitate washed with toluene, and dissolved in EtOAc (250 mL). The org. phase

was washed with aq. 1 M HCl solution, dried over Na2SO4, and evaporated in vacuo to give crude 170 (3.6 g, 70%) as an orange solid. The crude product was used in the next step without further purification. 1 Rf = 0.50 (pentane/EtOAc 1:1); m.p. 31 °C; H NMR (300 MHz, CDCl3): 1.40 (t, J = 7.2, 3H, H-C(6)), 2.57 (s, 3H, H-C(1)), 4.39 (q, J = 7.2, 2H, H-C(5)), 6.86 (s, 2H, H-C(4)), 6.87 (dd, J = 3.9, 0.9, 1H, H-C(2)), 7.67 (d, J = 3.9, 1H, H-C(3)); HR-EI- + + MS: m/z (%): 240.0451 (9, [M] , calcd for C11H12O4S : 240.0451), 167.0158 (100, + + [M–COOCH2CH3] , calcd for C8H7O2S : 167.0161), 125.0060 (41, [M– + + CH2COCOOCH2CH3] , calcd for C6H5OS : 125.0056).

193 8. Experimental ______Ethyl 5-(2-Thienyl)-3-isoxazolecarboxylate (171) ON 6 S O 1 5 4 O 2 3 KOtBu (2.67 g, 23.8 mmol, 1 equiv.) was added to a solution of 2-acetylthiophene (2.59 mL, 23.8 mmol, 1 equiv.) in toluene (75 mL) at 0 °C. Diethyl oxalate (3.87 mL, 30.8 mmol, 1 equiv.) was added dropwise. After stirring for 7 d at RT, the mixture

was filtrated. The residue was dissolved in EtOAc and washed with aq. 1 M HCl

solution. The org. phase was dried over Na2SO4, filtrated, and evaporated in vacuo. The crude dioxobutanoate 166 was used in the next step without further purification. A solution of 166 (2.52 g, 11.1 mmol, 1 equiv.) in EtOH (125 mL) was treated with hydroxylamine hydrochloride (2.90 g, 41.8 mmol, 3.75 equiv.). The mixture was heated to reflux for 3 h, evaporated in vacuo, and diluted with water. The pH was set to 7 with aq. ammonium hydroxide (25%). The precipitate was filtrated and dissolved

in CH2Cl2. The aqueous phase was extracted with CH2Cl2. The combined org.

phases were dried over MgSO4, filtrated, and evaporated in vacuo, giving 171 (2.24 g, 42%) as a brown solid. [212] 1 Rf = 0.63 (pentane/EtOAc 1:1); m.p. 46 °C (Lit. : 53–54 °C); H NMR (400 MHz,

CDCl3): 1.44 (t, J = 7.1, 3H, H-C(6)), 4.47 (q, J = 7.1, 2H, H-C(5)), 6.78 (s, 1H, H-C(4)), 7.15 (dd, J = 5.0, 3.7, 1H, H-C(2)), 7.50 (dd, J = 5.0, 1.2, 1H, H-C(1)), 7.57 13 (dd, J = 3.6, 1.1, 1H, H-C(3)); C NMR (100 MHz, CDCl3): 14.3, 62.4, 99.7, 127.9, 128.38, 128.45, 129.0, 157.1, 160.0, 166.8; IR (ATR): 3177, 3141, 3113, 3085, 2987, 2928, 1725, 1591, 1516, 1453, 1412, 1361, 1266, 1247, 1208, 1143, 1114, 1079, 1016, 994, 920, 907, 862, 848, 830, 779, 727, 682, 644, 618; HR-EI-MS: m/z (%): + + + 223.0300 (32, [M] , calcd for C10H9NO3S : 223.0298), 110.9897 (49, [M–C5H6NO2] , + calcd for C5H3OS : 110.9905), 29.0678 (100); elemental analysis calcd (%) for

C10H9NO3S (223.0): C 53.80, H 4.06, N 6.27; found: C 53.88, H 4.24, N 6.18.

194 8. Experimental ______Ethyl 5-(5-Methoxy-2-thienyl)-1,2-oxazole-3-carboxylate (172) O N 6 O O S 5 1 4 O 23 To crude 167 (374 mg, 1.46 mmol, 1 equiv.) in EtOH (15 mL), hydroxylamine hydrochloride (406 mg, 5.84 mmol, 4 equiv.) was added and the mixture was heated to reflux. After 21 h, the mixture was evaporated in vacuo, the residue suspended in water, and the pH set to 7 with aq. ammonium hydroxide (25%). The aq. phase was extracted with CH2Cl2. The org. phase was washed with water, dried over MgSO4, filtrated, and evaporated in vacuo to give 172 (327 mg, 40% over 2 steps) as a dark brown solid. 1 Rf = 0.38 (CH2Cl2); m.p. 54 °C; H NMR (300 MHz, CDCl3): 1.42 (t, J = 7.2, 3H, H-C(6)), 3.96 (s, 3H, H-C(1)), 4.45 (q, J = 7.2, 2H, H-C(5)), 6.25 (d, J = 4.1, 1H, H-C(2)), 6.59 (s, 1H, H-C(4)), 7.23 (d, J = 4.1, 1H, H-C(3)); 13C NMR (100 MHz,

CDCl3): 14.3, 60.7, 62.3, 97.9, 105.2, 114.5, 126.6, 156.9, 160.1, 167.0, 169.9; IR (ATR): 2941, 1719, 1594, 1537, 1489, 1450, 1423, 1390, 1362, 1248, 1213, 1133, 1061, 1016, 996, 979, 913, 898, 884, 869, 827, 785, 769, 744, 725, 678, 615; HR-EI- + + MS: m/z (%): 253.0405 (100, [M] , calcd for C11H11NO4S : 253.0404), 140.9999 (39, + + [M–C5H6NO2] , calcd for C6H5O2S : 141.0010), 29.0627 (90).

Ethyl 5-(5-Fluoro-2-thienyl)-3-isoxazolecarboxylate (173) O N O F S 3 O 4 1 2

5 A mixture of crude 168 (650 mg, 1.66 mmol, 1 equiv.) and hydroxylamine hydrochloride (694 mg, 9.98 mmol, 3.75 equiv.) in EtOH (35 mL) was heated to reflux for 16 h. The mixture was evaporated in vacuo. The black residue was diluted with water, and the pH set to 7 with aq. ammonium hydroxide (25%). The aq. phase

was extracted with CH2Cl2. The org. phase was dried over MgSO4, filtrated, and

evaporated in vacuo. Purification by CC (SiO2; pentane/CH2Cl2 5:2  CH2Cl2) gave 173 (318 mg, 50%) as a brown solid.

195 8. Experimental ______

1 Rf = 0.50 (CH2Cl2); m.p. 55–56 °C; H NMR (400 MHz, CDCl3): 1.43 (t, J = 7.1, 3H, 3 4 H-C(5)), 4.46 (q, J = 7.1, 2H, H-C(4)), 6.56 (dd, JHH = 4.2, JHF = 1.6, 1H, H-C(2)), 3 3 13 6.68 (s, 1H, H-C(3)), 7.20 (t, JHH = 3.9, JHF = 3.9, 1H, H-C(1)); C NMR 6 2 (100 MHz, CDCl3): 14.3, 62.5, 99.2 (d, JCF = 2.2), 109.1 (d, JCF = 11.1), 117.2 (d, 3 4 1 JCF = 4.7), 124.9 (d, JCF = 4.3), 157.1, 159.8, 166.1, 168.0 (d, JCF = 295.7); 19 F NMR (376 MHz, CDCl3): –124.9; IR (ATR): 3132, 3072, 2990, 1723, 1597, 1548, 1480, 1448, 1436, 1389, 1366, 1301, 1260, 1239, 1206, 1171, 1134, 1094, 1046, 1018, 984, 926, 899, 856, 834, 811, 779, 734, 712, 677, 622; HR-EI-MS: m/z + + (%): 241.0206 (36, [M] , calcd for C10H8FNO3S : 241.0204), 195.9859 (8, [M– + + OCH2CH3] , calcd for C8H3FNO2S : 195.9863), 128.9806 (39, [M– + + + CHNCH2OOCH2CH3] , calcd for C5H2FOS : 128.9810), 29.0420 (100, [CH2CH3] , + calcd for C2H5 : 29.0391); elemental analysis calcd (%) for C10H8FNO3S (241.2): C 49.79, H 3.34, N 5.81; found: C 49.77, H 3.39, N 5.71.

Ethyl 5-(5-Bromo-2-thienyl)-3-isoxazolecarboxylate (174) ON 5 O Br S 4 3 O 1 2 Crude 169 (4.07 g, 13.3 mmol, 1 equiv.) was dissolved in EtOH (200 mL), hydroxylamine hydrochloride (3.48 g, 50.0 mmol, 3.75 equiv.) was added, and the mixture was heated to reflux for 18 h. The mixture was evaporated in vacuo and the residue taken up in water. The mixture was neutralized to pH 7 with aq. ammonium

hydroxide (25%). The precipitate was collected and dissolved in CH2Cl2, dried over

MgSO4, filtrated, and evaporated in vacuo, affording 174 (3.60 g, 90%) as a brown oil. 1 Rf = 0.74 (pentane/EtOAc 1:1); H NMR (300 MHz, CDCl3): 1.43 (t, J = 7.1, 3H, H-C(5)), 4.46 (q, J = 7.1, 2H, H-C(4)), 6.73 (s, 1H, H-C(3)), 7.11 (d, J = 4.0, 1H, 13 H-C(1)), 7.30 (d, J = 4.0, 1H, H-C(2)); C NMR (75 MHz, CDCl3): 14.3, 62.5, 99.8, 116.8, 128.1, 129.7, 131.3, 157.1, 159.8, 165.6; IR (ATR): 3136, 2988, 1726, 1592, 1454, 1425, 1314, 1250, 1224, 1135, 1018, 973, 931, 899, 863, 832, 798, 774, 679, + 81 + 660, 617; HR-EI-MS: m/z (%): 302.9379 (26, [M] , calcd for C10H8 BrNO3S : + 79 + 302.9388), 300.9405 (26, [M] , calcd for C10H8 BrNO3S : 300.9403), 190.8990 (21, + 81 + + [M–C5H6NO2] , calcd for C5H2 BrOS : 190.8989), 188.9012 (21, [M–C5H6NO2] ,

196 8. Experimental ______

79 + calcd for C5H2 BrOS : 188.9010), 29.0622 (100); elemental analysis calcd (%) for

C10H8BrNO3S (302.1): C 39.75, H 2.67, N 4.64; found: C 39.82, H 2.75, N 4.71.

Ethyl 5-(5-Methyl-2-thienyl)-3-isoxazolecarboxylate (175) O N 1 S O

4 O 5 2 3

6 To a solution of 170 (3.6 g, 14.98 mmol, 1 equiv.) in anh. EtOH (47 mL) at RT, hydroxylamine hydrochloride (3.9 g, 56.28 mmol, 3.75 equiv.) was added. The mixture was heated to reflux for 17 h. The solvents were evaporated in vacuo, and water (80 mL) was added. The pH was set to 9 with aq. ammonium hydroxide (25%), and the aq. phase was extracted with EtOAc (3 x 60 mL). The org. phase was dried

over MgSO4, filtrated, and evaporated in vacuo to give 175 (2.54 g, 72%) as a brown solid. [212] 1 Rf = 0.71 (pentane/EtOAc 1:1); m.p. 35 °C (Lit. : 45–46 °C); H NMR (300 MHz,

CDCl3): 1.42 (t, J = 7.2, 3H, H-C(6)), 2.53 (s, 3H, H-C(1)), 4.45 (q, J = 7.2, 2H, H-C(5)), 6.67 (s, 1H, H-C(4)), 6.78 (dd, J = 3.6, 1.2, 1H, H-C(2)), 7.35 (d, J = 3.6, 13 1H, H-C(3)); C NMR (100 MHz, CDCl3): 14.3, 15.5, 62.3, 98.9, 126.0, 126.7, 128.0, 144.4, 156.9, 160.0, 166.9; IR (ATR): 3142, 3089, 2985, 2920, 1725, 1590, 1534, 1476, 1435, 1390, 1368, 1299, 1262, 1245, 1209, 1167, 1133, 1094, 1062, 1013, 983, 910, 898, 852, 833, 825, 809, 780, 743, 676, 623; HR-EI-MS: m/z (%): + + 237.0456 (64, [M] , calcd for C11H11NO3S : 237.0455), 125.0011 (71, [M– + + C5H6NO2] , calcd for C6H5OS : 125.0061), 29.0723 (100); elemental analysis calcd

for C11H11NO3S (237.1): C 55.68, H 4.67, N 5.90; found: C 55.46, H 4.83, N 6.14.

5-(2-Thienyl)-3-isoxazolecarbaldehyde (176) ON S OH 1 5 4 2 3

NaBH4 (0.38 g, 10 mmol, 1 equiv.) was added to a solution of 171 (2.24 g, 10 mmol, 1 equiv.) in EtOH (100 mL) at 0 °C. The mixture was stirred at RT for 7 d,

evaporated in vacuo, and diluted with water. The pH was set to 3 with aq. 1 M HCl

197 8. Experimental ______

solution. The aq. phase was extracted with CH2Cl2. The org. phase was dried over

MgSO4, filtrated, and evaporated to dryness. Purification by CC (SiO2;

pentane/CH2Cl2 1:1  CH2Cl2  CH2Cl2/Et2O 9:1) gave 176 (880 mg, 48%) as a brown solid. 1 Rf = 0.28 (pentane/EtOAc 1:1); m.p. 39 °C; H NMR (300 MHz, CDCl3): 4.02 (s, 1H, OH), 4.73 (s, 2H, H-C(5)), 6.42 (s, 1H, H-C(4)), 7.02–7.05 (m, 1H, H-C(2)), 7.36– 13 7.41 (m, 2H, H-C(1), H-C(3)); C NMR (100 MHz, CDCl3): 56.5, 98.2, 127.2, 128.0, 128.1, 129.0, 164.5, 165.2; IR (ATR): 3315 (br.), 3145, 3105, 2913, 2865, 1602, 1465, 1415, 1221, 1060, 1000, 970, 925, 851, 800, 719, 680; HR-EI-MS: m/z (%): + + + 181.0190 (57, [M] , calcd for C8H7NO2S : 181.0192), 110.9895 (100, [M–C3H4NO] , + calcd for C5H3OS : 110.9905), 39.0370 (22), 31.0417 (33); elemental analysis calcd

(%) for C8H7NO2S (181.0): C 53.02, H 3.89, N 7.73; found: C 53.29, H 3.92, N 7.63.

[5-(5-Methoxy-2-thienyl)-1,2-oxazol-3-yl]methanol (177) O N OH O S 5 1 4 23

To 172 (325 mg, 1.28 mmol, 1 equiv.) in EtOH (20 mL) at 0 °C, NaBH4 (49 mg, 1.28 mmol, 1 equiv.) was added. The mixture was stirred at RT for 23 h, evaporated in

vacuo, and the residue taken up in water. The pH was set to 3 with aq. 1 M HCl

solution, and the aq. phase was extracted with CH2Cl2. The org. phase was dried over

MgSO4, filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2/EtOAc 3:1) gave 177 (170 mg, 63%) as a light brown solid. 1 Rf = 0.31 (CH2Cl2/EtOAc 3:1); m.p. 53 °C; H NMR (400 MHz, CDCl3): 2.01 (t, J = 5.9, 1H, OH), 3.95 (s, 3H, H-C(1)), 4.76 (d, J = 5.5, 2H, H-C(5)), 6.23 (d, J = 4.1, 1H, H-C(2)), 6.26 (s, 1H, H-C(4)), 7.17 (d, J = 4.1, 1H, H-C(3)); 13C NMR (100 MHz,

CDCl3): 57.3, 60.6, 96.2, 105.0, 115.4, 125.8, 164.0, 165.7, 169.2; IR (ATR): 3446 (br.), 2939, 1613, 1538, 1499, 1453, 1423, 1332, 1232, 1206, 1157, 1072, 1051, 1016, 980, 921, 897, 770, 745, 730, 708, 684; HR-EI-MS: m/z (%): 211.0299 (100, [M]+, + + + calcd for C9H9NO3S : 211.0298), 141.006 (56, [M–C3H4NO] , calcd for C6H5O2S :

141.0010), 31.0427 (29); elemental analysis calcd (%) for C9H9NO3S (211.2): C 51.17, H 4.29, N 6.63; found: C 51.42, H 4.48, N 6.58.

198 8. Experimental ______[5-(5-Fluoro-2-thienyl)-1,2-oxazol-3-yl]methanol (178) O N F S 4 3 OH 1 2 Compound 173 (300 mg, 1.24 mmol, 1 equiv.) was dissolved in EtOH (20 mL) and

cooled to 0 °C. NaBH4 was added in portions and the mixture stirred at RT for 5 d. The mixture was evaporated in vacuo, diluted with water, and the pH was set to 2

with aq. 1 M HCl solution. The solid was filtered, dissolved in CH2Cl2, dried over

MgSO4, filtrated, and evaporated in vacuo. Purification by CC (SiO2;

CH2Cl2/EtOAc 3:2) gave 178 (140 mg, 56%) as a white solid. 1 Rf = 0.54 (CH2Cl2/EtOAc 1:1); m.p. 56 °C; H NMR (300 MHz, CDCl3): 2.28 (t, J = 3 6.1, 1H, OH), 4.77 (d, J = 5.9, 2H, H-C(4)), 6.36 (s, 1H, H-C(3)), 6.53 (dd, JHH = 4 3 3 13 4.2, JHF = 1.7, 1H, H-C(2)), 7.13 (dd, JHH = 3.9, JHF = 3.9, 1H, H-C(1)); C NMR 6 2 3 (100 MHz, CDCl3): 57.1, 97.6 (d, JCF = 2.5), 108.8 (d, JCF = 11.0), 118.0 (d, JCF = 4 1 19 4.4), 124.1 (d, JCF = 3.8), 164.2, 164.8, 167.6 (d, JCF = 294.9); F NMR (376 MHz,

CDCl3): –126.1; IR (ATR): 3276 (br.), 3115, 2933, 2649, 1745, 1603, 1557, 1497, 1455, 1431, 1351, 1252, 1201, 1028, 997, 963, 899, 875, 820, 791, 742, 716, 676, + + 638; HR-ESI-MS: m/z (%): 200.0176 (100, [M+H] , calcd for C8H7FNO2S : 200.0177), 185.1287 (45).

[5-(5-Bromo-2-thienyl)-3-isoxazolyl]methanol (179) ON OH Br S 4 3 1 2

NaBH4 (0.45 g, 11.8 mmol, 1 equiv.) was added in portions to a solution of 174 (3.60 g, 11.8 mmol, 1 equiv.) in EtOH (200 mL) at 0 °C, and the mixture stirred at RT for 72 h. The mixture was evaporated in vacuo and dissolved in water. The pH was set to 3–4 with conc. HCl. The precipitated product was filtrated, dissolved in

CH2Cl2, filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2 

CH2Cl2/EtOAc 4:1  EtOAc) gave 179 (1.31 g, 43%) as a yellow solid. 1 Rf = 0.48 (pentane/EtOAc 1:1); m.p. 92 °C; H NMR (400 MHz, CDCl3): 2.14 (br. s, 1H, OH), 4.78 (s, 2H, H-C(4)), 6.41 (d, J = 0.2, 1H, H-C(3)), 7.08 (d, J = 3.9, 1H, 13 H-C(1)), 7.24 (dd, J = 4.0, 0.2, 1H, H-C(2)); C NMR (100 MHz, CDCl3): 57.2, 98.3,

199 8. Experimental ______116.0, 127.4, 130.7, 131.1, 164.25, 164.29; IR (ATR): 3403, 3091, 2930, 2887, 1598, 1528, 1475, 1417, 1394, 1335, 1242, 1201, 1174, 1039, 993, 971, 894, 819, 794, 746, + 81 + 677, 634; HR-EI-MS: m/z (%): 260.9277 (80, [M] , calcd for C8H6 BrNO2S : + 79 + 260.9282), 258.9300 (80, [M] , calcd for C8H6 BrNO2S : 258.9298), 190.8983 (100, + 81 + + [C5H2BrOS] , calcd for C5H2 BrOS : 190.8989), 188.9005 (100, [C5H2BrOS] , calcd 79 + for C5H2 BrOS : 188.9010); elemental analysis calcd (%) for C8H6BrNO2S (260.1): C 36.94, H 2.32, N 5.38; found: C 37.18, H 2.49, N 5.24.

[5-(5-Methyl-2-thienyl)isoxazol-3-yl]methanol (180) O N 1 S 5

4 OH 2 3

NaBH4 (0.41 g, 10.7 mmol, 1 equiv.) was added to 175 (2.54 g, 10.7 mmol, 1 equiv.) in anhydrous EtOH (30 mL) at 0 °C. The solution was stirred at RT for 4.5 d. The solvents were evaporated in vacuo, and the residue was taken up in water. The pH

was set to 3 using aq. 1 M HCl solution, giving a white precipitate. The precipitate

was filtered, washed with water, dissolved in CH2Cl2, and dried over MgSO4. The

solvents were evaporated in vacuo and the product purified by CC (SiO2; pentane/EtOAc 15:1  3:1  EtOAc), giving 180 (0.45 g, 22%) as a white solid. 1 Rf = 0.42 (pentane/EtOAc 1:1); m.p. 95 °C; H NMR (300 MHz, CDCl3): 2.54 (s, 3H, H-C(1)), 4.78 (s, 2H, H-C(5)), 6.35 (s, 1H, H-C(4)), 6.78 (dd, J = 3.6, 1.2, 1H, 13 H-C(2)), 7.31 (d, J = 3.7, 1H, H-C(3)); C NMR (100 MHz, CDCl3): 15.5, 57.3, 97.3, 126.6, 126.9, 127.4, 143.6, 164.0, 165.7; IR (ATR): 3377, 3125, 3089, 2950, 2889, 1781, 1598, 1538, 1485, 1461, 1442, 1403, 1254, 1206, 1161, 1037, 992, 962, 898, 823, 792, 748, 681, 642; HR-EI-MS: m/z (%): 195.0349 (54, [M]+, calcd for + + + C9H9NO2S : 195.0349), 125.0048 (100, [M–C3H4NO] , calcd for C6H5OS :

125.0061); elemental analysis calcd for C9H9NO2S (195.0): C 55.37, H 4.65, N 7.17; found: C 55.46, H 4.69, N 7.09.

200 8. Experimental ______5-(5-Iodo-2-thienyl)-1,2-oxazole-3-carbaldehyde (181) O N O I S 4 3 12

To a solution of 184 (449 mg, 1.46 mmol, 1 equiv.) in CH2Cl2 (22 mL) at RT, PCC (544 mg, 2.52 mmol, 1.7 equiv.) was added. The mixture was stirred at RT for 16 h,

filtrated, and evaporated in vacuo. Purification by CC (SiO2; pentane/EtOAc 48:2) and HPLC (Dr. Maisch Reprosil Chiral NR; 250 x 20 mm; 9 mL min–1, hexane/EtOAc 98:2  92:8 in 30 min) gave 181 (250 mg, 56%) as a white solid. 1 Rf = 0.50 (CH2Cl2); m.p. 70 °C; H NMR (300 MHz, CDCl3): 6.71 (s, 1H, H-C(3)), 7.23 (d, J = 3.9, 1H, H-C(1)), 7.31 (d, J = 3.9, 1H, H-C(2)), 10.15 (s, 1H, H-C(4)); 13 C NMR (100 MHz, CDCl3): 78.7, 96.4, 129.4, 133.9, 138.3, 162.6, 184.5, 1 signal not visible; IR (ATR): 2885, 1596, 1524, 1477, 1438, 1408, 1352, 1330, 1240, 1225, 1202, 1184, 1105, 1025, 1001, 954, 936, 903, 802, 783, 731, 694; HR-EI-MS: m/z + + (%): 304.9000 (100, [M] , calcd for C8H4INO2S : 304.9002), 236.8871 (80, [M– + + C3H2NO] , calcd for C5H2IOS : 236.8871); elemental analysis calcd (%) for

C8H4INO2S (305.1): C 31.49, H 1.32, N 4.59; found: C 31.34, H 1.49, N 4.65.

5-(5-Chloro-2-thienyl)-3-({[dimethyl(2-methyl-2-propanyl)silyl]oxy}methyl)-1,2- oxazole (182)

6 5 6 ON S O Cl Si 6 4 3 1 2 5 To 44 (100 mg, 0.46 mmol, 1 equiv.) in dry DMF (1 mL), TBDMSCl (77 mg, 0.51 mmol, 1.1 equiv.) and imidazole (63 mg, 0.93 mmol, 2 equiv.) were added. The mixture was stirred at RT for 4.5 h and subsequently diluted with water and EtOAc.

The aq. phase was extracted with EtOAc. The org. phase was dried over MgSO4, filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2) gave 182 (142 mg, 93%) as colorless oil. 1 Rf = 0.53 (CH2Cl2); H NMR (300 MHz, CDCl3): 0.13 (s, 6H, H-C(5)), 0.94 (s, 9H, H-C(6)), 4.77 (s, 2H, H-C(4)), 6.38 (s, 1H, H-C(3)), 6.94 (d, J = 4.0, 1H, H-C(1)), 13 7.26 (d, J = 3.9, 1H, H-C(2)); C NMR (100 MHz, CDCl3): –5.2, 18.5, 26.0, 57.5,

201 8. Experimental ______98.5, 126.3, 127.3, 128.0, 133.2, 163.8, 164.9; IR (ATR): 3099, 2934, 2882, 2859, 1597, 1533, 1477, 1436, 1422, 1374, 1340, 1251, 1206, 1179, 1092, 1029, 1013, 1001, 940, 896, 834, 809, 777, 724, 681, 673, 662, 635; HR-EI-MS: m/z (%): + 35 + 271.9962 (100, [M–C4H9] , calcd for C10H11 ClNO2SSi : 271.9963), 75.0255 (25); elemental analysis calcd (%) for C14H20ClNO2SSi (329.9): C 50.97, H 6.11, N 4.25; found: C 51.00, H 6.13, N 4.52.

3-({[Dimethyl(2-methyl-2-propanyl)silyl]oxy}methyl)-5-(5-iodo-2-thienyl)-1,2- oxazole (183)

6 5 6 ON S O I Si 6 4 3 1 2 5 To a solution of 182 (100 mg, 0.30 mmol, 1 equiv.) in THF (10 mL) at –78 °C under

N2 was added 1.9 M tBuLi in pentane (176 L, 0.33 mmol, 1.1 equiv.). The mixture was stirred at –78 °C for 1.5 h, and I2 (85 mg, 0.33 mmol, 1.1 equiv.) was added. Stirring was continued at –78 °C for 1 h, then the mixture was slowly warmed to RT,

and treated with water. The aq. phase was extracted with Et2O. The org. phase was washed with aq. 0.5 M Na2S2O3 solution and water, dried over MgSO4, filtrated, and evaporated in vacuo to give 183 (128 mg, quant.) as a brown oil. 1 Rf = 0.39 (pentane/CH2Cl2 1:1); H NMR (300 MHz, CDCl3): 0.13 (s, 6H, H-C(5)), 0.94 (s, 9H, H-C(6)), 4.77 (s, 2H, H-C(4)), 6.39 (s, 1H, H-C(3)), 7.15 (d, J = 3.8, 1H, 13 H-C(1)), 7.26 (d, J = 3.8, 1H, H-C(2)); C NMR (100 MHz, CDCl3): –5.2, 18.5, 26.0, 57.5, 77.0, 98.8, 128.2, 135.4, 138.0, 163.6, 164.9; IR (ATR): 3096, 2932, 2883, 2858, 1595, 1528, 1471, 1437, 1411, 1274, 1334, 1251, 1208, 1178, 1092, 1014, 1003, 949, 892, 835, 807, 779, 726, 680, 654, 633; HR-EI-MS: m/z (%): 405.9792 + + + (3, [M–CH3] , calcd for C13H17INO2SSi : 405.9789), 363.9317 (100, [M–C4H9] , + calcd for C10H11INO2SSi : 363.9319).

202 8. Experimental ______[5-(5-Iodo-2-thienyl)-1,2-oxazol-3-yl]methanol (184) ON OH I S 4 3 1 2 To a solution of 183 (400 mg, 0.45 mmol, 1 equiv.) in THF (38 mL), aq. 1 M HCl solution (4.7 mL) was added and the mixture was stirred at RT. After 41 h, the

mixture was diluted with water and neutralized with solid NaHCO3. The aq. phase was extracted with EtOAc. The org. phase was washed with brine, dried over

MgSO4, filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2/EtOAc 4:1  EtOAc) gave crude 184 (532 mg, quant.) as a white solid. 1 Rf = 0.28 (CH2Cl2/EtOAc 4:1); m.p. 72 °C; H NMR (300 MHz, CDCl3): 1.96 (t, J = 6.1, 1H, OH), 4.79 (d, J = 6.0, 2H, H-C(4)), 6.42 (s, 1H, H-C(3)), 7.16 (d, J = 3.9, 1H, 13 H-C(1)), 7.27 (d, J = 3.9, 1H, H-C(2)); C NMR (100 MHz, CDCl3): 57.2, 77.4, 98.4, 128.5, 135.0, 138.1, 164.2, 1 signal not visible; IR (ATR): 3291 (br.), 3114, 2925, 2853, 1719, 1600, 1526, 1471, 1407, 1333, 1239, 1208, 1182, 1065, 1040, 997, 963, 951, 897, 821, 798, 747, 678, 663, 631; HR-EI-MS: m/z (%): 306.9160 (100, [M]+, + + + calcd for C8H6INO2S : 306.9159), 236.8867 (80, [M–C3H4NO] , calcd for C5H2IOS : 236.8871).

(3aSR,4RS,8aSR,8bRS)-2-(3-Bromopropyl)-4-[5-(2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-185) O 8 9 H 11 8a N 8b 7 Br 10 3a N 6 O H 4

N 4' O

3' S 2' 1' A suspension of 36 (760 mg, 3.50 mmol, 1.0 equiv.), L-proline (18) (423 mg, 3.68 mmol, 1.1 equiv.), and 161 (660 mg, 3.68 mmol, 1.1 equiv.) in MeCN (3 mL) was heated to reflux for 7 h, leading to a mixture of products, and was concentrated in

vacuo. Repeated purification by CC (SiO2; CH2Cl2/Et2O 95:5  9:1; pentane/EtOAc

203 8. Experimental ______2:1) and HPLC (Merck LiChrospher® Si 60; 25025 mm, 5 m; EtOAc) gave slightly impure (±)-185 (42 mg, 3%) as a yellow solid. Crude (±)-185 was used for the next step without further purification. 1 Rf = 0.32 (CH2Cl2/Et2O 4:1); H NMR (400 MHz, CDCl3): 1.64–1.85 (m, 2H, H-C(7), H-C(8)), 2.04–2.23 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.77 (ddd, J = 12.3, 8.0, 4.3, 1H, H-C(6)), 2.87–2.95 (m, 1H, H-C(6)), 3.31–3.38 (m, 3H, H-C(8b), H-C(11)), 3.60–3.66 (m, 3H, H-C(3a), H-C(9)), 3.78 (t, J = 8.7, 1H, H-C(8a)), 4.29 (d, J = 8.7, 1H, H-C(4)), 6.31 (s, 1H, H-C(4’)), 7.11 (dd, J = 5.0, 3.7, 1H, H-C(2’)), 7.43 (dd, J = 5.0, 1.1, 1H, H-C(1’)), 7.51 (dd, J = 3.7, 1.0, 1H, H-C(3’)); IR (ATR): 3112, 2954, 2878, 1774, 1696, 1601, 1469, 1436, 1399, 1352, 1322, 1236, 1186, 1155, 1089, 1065, 1020, 908, 875, 849, 804, 710, 626; HR-MALDI-MS: m/z (%): 452.0460 (100, + 81 + + [M+H] , calcd for C19H21 BrN3O3S : 452.0461), 450.0472 (93, [M+H] , calcd for 79 + C19H21 BrN3O3S : 450.0482).

(3aS,4R,8aS,8bR)-2-(3-Bromopropyl)-4-[5-(5-methoxy-2-thienyl)-1,2-oxazol-3- yl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((+)-186) and (3aR,4S,8aR,8bS)-2-(3-Bromopropyl)-4-[5-(5-methoxy-2-thienyl)-1,2-oxazol-3- yl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((–)-186) O O 8 8 9 H 9 H 11 8a 11 8a N 8b 7 N 8b 7 3a 3a Br 10 N Br 10 N 6 6 O H 4 O H 4

N 3' N 3' O O

2' 2' S S 1' 1' O O

(+)-186 (–)-186 A mixture of 36 (176 mg, 0.81 mmol, 1 equiv.), L-proline (18) (98 mg, 0.85 mmol, 1.1 equiv.), and 162 (178 mg, 0.85 mmol, 1.1 equiv.) in MeCN (4 mL) was heated to reflux for 26 h, leading to a mixture of products, and concentrated in vacuo. Repeated

purification by CC (SiO2; CH2Cl2/Et2O 4:1; CH2Cl2/Et2O 9:1) and HPLC (Reprosil

204 8. Experimental ______Chiral-NR, 250 x 4.6 mm, 7 m; 1 mL min–1, 40% EtOH in heptane) gave (+)-186 (10 mg, 3%) and (–)-186 (9 mg, 2%) as a light brown solid. 25 (+)-186: Rf = 0.33 (CH2Cl2/Et2O 4:1); []D : +120.6 (c = 0.5, CH3OH); m.p. 103 °C; 1 H NMR (300 MHz, CDCl3): 1.63–1.74 (m, 1H, H-C(8)), 1.75–1.88 (m, 1H, H-C(7)), 2.00–2.22 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.76 (ddd, J = 12.6, 8.2, 4.3, 1H, H- C(6)), 2.85–2.95 (m, 1H, H-C(6)), 3.30–3.37 (m, 3H, H-C(8b), H-C(11)), 3.59–3.65 (m, 3H, H-C(3a), H-C(9)), 3.77 (br. t, J = 8.4, 1H, H-C(8a)), 3.94 (s, 3H, OMe), 4.26 (d, J = 8.4, 1H, H-C(4)), 6.12 (s, 1H, H-C(3’)), 6.21 (d, J = 4.1, 1H, H-C(1’)), 7.16 (d, J = 4.1, 1H, H-C(2’)); IR (ATR): 2934, 1774, 1698, 1613, 1536, 1495, 1445, 1420, 1398, 1357, 1280, 1229, 1206, 1150, 1095, 1056, 1017, 1005, 987, 918, 880, 815, 785, 729; HR-MALDI-MS: m/z (%): 482.0570 (98, [M+H]+, calcd for 81 + + 79 + C20H23 BrN3O4S : 482.0567), 480.0582 (100, [M+H] , calcd for C20H23 BrN3O4S : 480.0587). 25 (–)-186: Rf = 0.33 (CH2Cl2/Et2O 4:1); []D : –176.3 (c = 0.4, CH3OH); m.p. 112 °C; 1 H NMR (400 MHz, CDCl3): 1.63–1.73 (m, 1H, H-C(8)), 1.75–1.87 (m, 1H, H-C(7)), 2.00–2.22 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.72–2.81 (m, 1H, H-C(6)), 2.84–2.95 (m, 1H, H-C(6)), 3.30–3.37 (m, 3H, H-C(8b), H-C(11)), 3.59–3.65 (m, 3H, H-C(3a), H-C(9)), 3.74–3.82 (m, 1H, H-C(8a)), 3.94 (s, 3H, OMe), 4.26 (d, J = 8.5, 1H, H- C(4)), 6.12 (s, 1H, H-C(3’)), 6.21 (d, J = 4.1, 1H, H-C(1’)), 7.16 (d, J = 3.8, 1H, H- C(2’)); IR (ATR): 2955, 1774, 1698, 1613, 1536, 1496, 1460, 1445, 1421, 1398, 1357, 1340, 1312, 1280, 1235, 1206, 1150, 1095, 1056, 1018, 1005, 987, 950, 918, 897, 880, 864, 814, 785, 757, 729; HR-MALDI-MS: m/z (%): 482.0576 (100, + 81 + + [M+H] , calcd for C20H23 BrN3O4S : 482.0567), 480.0589 (98, [M+H] , calcd for 79 + C20H23 BrN3O4S : 480.0587).

205 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-2-(3-Bromopropyl)-4-[5-(5-fluoro-2-thienyl)-1,2-oxazol- 3-yl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-187) O 8 9 H 11 8a N 8b 7 3a Br 10 N 6 O H 4

N 3' O

2' S 1' F A suspension of maleimide 36 (140 mg, 0.64 mmol, 1.0 equiv.), L-proline (18) (78 mg, 0.67 mmol, 1.1 equiv.), and 163 (133 mg, 0.67 mmol, 1.1 equiv.) in MeCN (2 mL) was heated to reflux for 30 h, leading to a mixture of products, and concentrated in vacuo. Repeated purification by CC (SiO2; CH2Cl2/Et2O 9:1;

CH2Cl2/Et2O 6:1; CH2Cl2/Et2O 6:1; CH2Cl2/Et2O 9:1) and HPLC (Regis Spherical –1 Kromasil; SiO2, 100 Å, 5 m; 9 mL min ; CH2Cl2/EtOAc 3:7  2.5:7.5, 25 min) gave (±)-187 (15 mg, 5%) as a light brown solid. 1 Rf = 0.26 (CH2Cl2/Et2O 8:1); m.p. 138–140 °C; H NMR (300 MHz, CDCl3): 1.64– 1.74 (m, 1H, H-C(8)), 1.83 (dtd, J = 17.0, 8.6, 4.2, 1H, H-C(7)), 2.00–2.24 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.75 (ddd, J = 12.6, 8.2, 4.4, 1H, H-C(6)), 2.85–2.95 (m, 1H, H-C(6)), 3.29–3.39 (m, 3H, H-C(8b), H-C(11)), 3.59–3.65 (m, 3H, H-C(3a), H-C(9)), 3.74–3.79 (m, 1H, H-C(8a)), 4.27 (d, J = 8.4, 1H, H-C(4)), 6.22 (s, 1H, 3 4 3 3 H-C(3’)), 6.52 (dd, JHH = 4.2, JHF = 1.6, 1H, H-C(2’)), 7.13 (t, JHH = 3.9, JHF = 3.9, 13 1H, H-C(1’)); C NMR (100 MHz, CDCl3): 23.5, 29.7, 29.9, 30.9, 38.0, 49.4, 49.5, 6 2 3 50.8, 60.8, 68.3, 98.4 (d, JCF = 1.8), 108.8 (d, JCF = 11.0), 118.2 (d, JCF = 4.2), 4 1 19 124.0 (d, JCF = 3.9,), 163.5, 164.3, 167.4 (d, JCF = 294.7), 175.1, 177.8; F NMR

(282 MHz, CDCl3): –125.9; IR (ATR): 2964, 1699, 1615, 1552, 1504, 1443, 1402, 1372, 1336, 1241, 1197, 1069, 921, 872, 790, 715, 668, 630; HR-MALDI-MS: m/z + 81 + (%): 470.0384 (100, [M+H] , calcd for C19H20 BrFN3O3S : 470.0367), 468.0387 (86, + 79 + [M+H] , calcd for C19H20 BrFN3O3S : 468.0391).

206 8. Experimental ______(3aRS,4RS,8aSR,8bSR)-2-(3-Bromopropyl)-4-[5-(5-bromo-2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-188) and (3aRS,4SR,8aRS,8bSR)-2-(3-Bromopropyl)-4-[5-(5-bromo-2-thienyl)-3- isoxazolyl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-191) O O 8 8 9 H 9 H 11 8a 11 8a N 8b 7 N 8b 7 3a 3a Br 10 N Br 10 N 6 6 O H 4 O H 4

N 3' N 3' O O

2' 2' S S 1' 1' Br Br

(±)-188 (±)-191 Compound 36 (755 mg, 3.46 mmol, 1 equiv.), L-proline (18) (419 mg, 3.64 mmol, 1.1 equiv.), and 164 (938 mg, 3.64 mmol, 1.1 equiv.) in MeCN (9 mL) were heated to reflux for 18 h, leading to a mixture of products, and concentrated in vacuo.

Purification by CC (SiO2; CH2Cl2/EtOAc 19:1  EtOAc) and recycling HPLC ® (Merck LiChrospher Si 60; 25025 mm, 5m; CH2Cl2/EtOAc 3:7, 3 turns) gave (±)-188 (24 mg, 1%) and (±)-191 (166 mg, 9%) as a yellow solid. 1 (±)-188: Rf = 0.27 (pentane/EtOAc 1:1); m.p. 201 °C (decomp); H NMR (400 MHz,

CDCl3): 1.62–1.73 (m, 1H, H-C(8)), 1.76–1.88 (m, 1H, H-C(7)), 2.01–2.23 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.76 (ddd, J = 12.8, 8.4, 4.3, 1H, H-C(6)), 2.90 (ddd, J = 12.6, 8.9, 7.4, 1H, H-C(6)), 3.31–3.37 (m, 3H, H-C(8b), H-C(11)), 3.59–3.65 (m, 3H, H-C(3a), H-C(9)), 3.75–3.79 (m, 1H, H-C(8a)), 4.28 (d, J = 8.4, 1H, H-C(4)), 6.27 (s, 1H, H-C(3’)), 7.07 (d, J = 3.9, 1H, H-C(1’)), 7.24 (d, J = 3.9, 1H, H-C(2’)); 13C NMR

(100 MHz, CDCl3): 23.4, 29.5, 29.8, 30.7, 37.9, 49.3, 49.4, 50.7, 60.7, 68.2, 99.0, 126.1, 127.2, 130.7, 130.9, 163.5, 163.7, 175.0, 177.6; IR (ATR): 2958, 1769, 1696, 1615, 1478, 1435, 1396, 1348, 1317, 1227, 1180, 1143, 1065, 1016, 970, 919, 870, 793, 638; HR-MALDI-MS: m/z (%): 531.9526 (51), 529.9562 (100, [M+H]+, calcd 79 80 + for C19H20 Br BrN3O3S : 529.9566), 527.9300 (45). 1 (±)-191: Rf = 0.31 (pentane/EtOAc 2:3); m.p. 118 °C; H NMR (400 MHz, CDCl3): 1.75–2.11 (m, 4H, H-C(7), H-C(8)), 2.15 (quintet, J = 7.0, 2H, H-C(10)), 2.56–2.62

207 8. Experimental ______(m, 1H, H-C(6)), 3.17 (ddd, J = 10.5, 8.3, 5.0, 1H, H-C(6)), 3.38 (t, J = 6.7, 2H, H- C(11)), 3.50 (dd, J = 8.9, 8.9, 1H, H-C(8b)), 3.64 (t, J = 7.1, 2H, H-C(9)), 3.83–3.91 (m, 2H, H-C(3a), H-C(8a)), 4.58 (d, J = 3.3, 1H, H-C(4)), 6.42 (s, 1H, H-C(3’)), 7.08 (d, J = 3.9, 1H, H-C(1’)), 7.23 (d, J = 3.9, 1H, H-C(2’)); 13C NMR (100 MHz,

CDCl3): 24.9, 26.0, 29.7, 30.5, 38.1, 48.7, 52.4, 52.9, 62.7, 66.6, 98.6, 116.0, 127.4, 130.7, 131.1, 164.5, 165.3, 177.2, 178.0; IR (ATR): 3058, 2924, 2868, 1767, 1694, 1597, 1527, 1468, 1451, 1442, 1431, 1398, 1343, 1321, 1243, 1200, 1188, 1144, 1109, 1061, 1030, 973, 961, 929, 898, 855, 828, 807, 785, 728, 698, 668, 651, 636, + 80 + 622; HR-MALDI-MS: m/z (%): 531.9526 (53, [M+H] , calcd for C19H20 Br2N3O3S : + 79 80 + 531.9546), 529.9563 (100, [M+H] , calcd for C19H20 Br BrN3O3S : 529.9566),

527.9198 (44); elemental analysis calcd (%) for C19H19Br2N3O3S (529.2): C 43.12, H 3.62, N 7.94; found: C 43.4, H 3.67, N 7.81.

(3aSR,4RS,8aSR,8bRS)-2-(3-Bromopropyl)-4-[5-(5-iodo-2-thienyl)-1,2-oxazol-3- yl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione (189) O 8 9 H 11 8a 7 N 8b Br 10 3a N 6 O H 4

N 3' O

2' S 1' I A mixture of 36 (170 mg, 0.78 mmol, 1 equiv.), L-proline (18) (95 mg, 0.82 mmol, 1.1 equiv.), and 181 (251 mg, 0.82 mmol, 1.1 equiv.) in MeCN (2 mL) was heated to reflux for 21 h, leading to a mixture of products, and concentrated in vacuo. Repeated

purification by CC (SiO2; CH2Cl2; CH2Cl2/Et2O 49:1  45:5  3:2 1:1) and HPLC (Reprosil Chiral-NR; 250 x 4.6 mm, 7 m; 1 mL min–1, 40% EtOH in heptane) gave crude 189 (13 mg, 3%) as what 1 Rf = 0.43 (CH2Cl2/Et2O 4:1); m.p. 124 °C (decomp); H NMR (400 MHz, CDCl3): 1.63–1.73 (m, 1H, H-C(8)), 1.77–1.87 (m, 1H, H-C(7)), 2.01–2.22 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.75 (ddd, J = 12.6, 8.3, 4.3, 1H, H-C(6)), 2.90 (ddd, J = 12.6, 9.1, 7.3, 1H, H-C(6)), 3.31–3.38 (m, 3H, H-C(8b), H-C(11)), 3.60–3.65 (m, 3H, H-C(3a),

208 8. Experimental ______H-C(9)), 3.74–3.79 (m, 1H, H-C(8a)), 4.27 (d, J = 8.4, 1H, H-C(4)), 6.27 (s, 1H, H-C(3’)), 7.15 (d, J = 3.8, 1H, H-C(1’)), 7.25 (d, J = 3.9, H-C(2’)); 13C NMR (100

MHz, CDCl3): 23.5, 29.7, 29.9, 30.8, 38.0, 49.4, 49.5, 50.8, 60.8, 68.3, 99.2, 128.4, 135.3, 138.0, 163.6, 163.7, 175.1, 177.8, 1 signal not visible; IR (ATR): 2934, 2852, 1771, 1697, 1607, 1536, 1495, 1445, 1422, 1399, 1356, 1280, 1236, 1206, 1151, 1095, 1057, 1005, 987, 949, 917, 897, 880, 813, 785, 757, 729; HR-MALDI-MS: m/z + 81 + (%): 577.9383 (83, [M+H] , calcd for C19H20 BrIN3O3S : 577.9428), 575.9438 (75, + 79 + + [M+H] , calcd for C19H20 BrIN3O3S : 575.9448) 450.0438 (100, [M–I] , calcd for 81 + C19H19 BrN3O3S : 450.0310).

3-[(3aRS,4SR,8aRS,8bSR)-4-[5-(5-Bromo-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-192) O 8 12 9 H 11 8a 7 12 N 8b 3a N 10 N 6 O H 4 12 Br N 3' O

2' S 1' Br

To a solution of (±)-191 (30 mg, 0.06 mmol, 1 equiv.) in EtOH (1 mL), 4.2 M Me3N in EtOH (270 L, 1.13 mmol, 20 equiv.) was added. The mixture was stirred at RT for 12 d and evaporated in vacuo. The residue was dissolved in MeOH. Precipitation from Et2O gave (±)-192 (10 mg, 30%) as a light brown solid. 1 M.p. 203 °C (decomp); H NMR (500 MHz, CDCl3): 1.82–2.14 (m, 6H, H-C(7), H-C(8), H-C(10)), 2.67–2.72 (m, 1H, H-C(6)), 3.05–3.10 (m, 1H, H-C(6)), 3.14 (s, 9H, H-C(12)), 3.39–3.45 (m, 2H, H-C(11)), 3.56–3.65 (m, 2H, H-C(9)), 3.71 (t, J = 9.0, 1H, H-C(8b)), 3.86–3.90 (m, 2H, H-C(3a), H-C(8a)), 4.46 (d, J = 4.9, 1H, H-C(4)), 6.77 (s, 1H, H-C(3’)), 7.22 (d, J = 4.0, 1H, H-C(1’)), 7.41 (d, J = 3.9, 1H, 13 H-C(2’)); C NMR (125 MHz, CDCl3): 22.7, 25.5, 27.1, 36.8, 49.4, 53.1, 53.7 (t), 53.9, 63.2, 65.3 (t), 67.6, 99.6, 116.8, 129.0, 131.8, 132.7, 165.7, 166.9, 178.6, 179.3; IR (ATR): 3007, 2966, 2874, 1778, 1690, 1599, 1474, 1454, 1406, 1393, 1377, 1354,

209 8. Experimental ______1299, 1236, 1182, 1141, 1113, 1070, 1037, 1020, 969, 954, 915, 898, 806, 740, 667, 648, 625; HR-MALDI-MS: m/z (%): 509.1030 (100, [M–Br]+, calcd for 81 + + 79 + C22H28 BrN4O3S : 509.1040), 507.1059 (99, [M–Br] , calcd for C22H28 BrN4O3S :

507.1060); elemental analysis calcd (%) for C22H28Br2N4O3S (588.4): C 44.91, H 4.80, N 9.52; found: C 45.11, H 4.91, N 9.25.

3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-1,3-oxazol-2-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N,N-trimethyl-1- propanaminium Bromide ((±)-196) O 8 12 9 H 11 8a 7 12 N 8b N 10 3a N 6 O 4 12 H Br N O

3' 2' S 1' Cl To a solution of (±)-221 (17 mg, 0.04 mmol, 1 equiv.) in EtOH (2 mL) at RT, 4.2 M

Me3N in EtOH (170 L, 0.70 mmol, 20 equiv.) was added. The mixture was stirred at RT for 5 d and evaporated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (±)-196 (14 mg, 74%) as a yellow solid. 1 M.p. 223 °C (decomp); H NMR (600 MHz, CDCl3): 1.77–1.89 (m, 2H, H-C(7), H-C(8)), 2.01–2.08 (m, 2H, H-C(10)), 2.10–2.19 (m, 2H, H-C(7), H-C(8)), 2.87 (ddd, J = 12.6, 8.2, 4.4, 1H, H-C(6)), 2.97–3.01 (m, 1H, H-C(6)), 3.13 (s, 9H, H-C(12)), 3.33–3.37 (m, 2H, H-C(11)), 3.49 (dd, J = 8.2, 1.5, 1H, H-C(8b)), 3.54–3.59 (m, 2H, H-C(9)), 3.75 (ddd, J = 9.2, 7.3, 1.6, 1H, H-C(8a)), 3.91 (dd, J = 8.5, 8.5, 1H, H- C(3a)), 4.47 (d, J = 8.8, 1H, H-C(4)), 7.03 (d, J = 4.0, 1H, H-C(1’)), 7.22 (d, J = 3.9, 13 1H, H-C(2’)), 7.30 (s, 1H, H-C(3’)); C NMR (150 MHz, CDCl3): 22.7, 24.2, 30.4, 36.8, 50.5, 50.7, 52.4, 53.6 (t), 63.6, 65.2 (t), 69.6, 122.9, 125.4, 128.6, 129.5, 131.6, 147.9, 162.6, 177.5, 179.8; IR (ATR): 2953, 1773, 1704, 1562, 1478, 1434, 1404, 1357, 1321, 1245, 1226, 1205, 1183, 1146, 1127, 1087, 1068, 1031, 960, 929, 884, 818, 777, 748, 668, 629; HR-MALDI-MS: m/z (%): 465.1531 (41, [M–Br]+, calcd for

210 8. Experimental ______

37 + + 35 + C22H28 ClN4O3S : 465.1537), 463.1567 (100, [M–Br] , calcd for C22H28 ClN4O3S : 463.1565).

N,N,N-Trimethyl-3-[(3aRS,4RS,8aSR,8bSR)-4-[2-(5-Chloro-2-thienyl)-1,3- oxazol-4-yl]-1,3-dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-1- propanaminium Bromide ((±)-198) O 8 12 9 H 11 8a 7 12 N 8b N 10 3a N 6 O 4 12 H 3' N Br O

2' S 1' Cl

To (±)-222 (12 mg, 0.02 mmol, 1 equiv.) in EtOH (0.25 mL) at RT, 4.2 M Me3N in EtOH (0.12 mL, 0.5 mmol, 50 equiv.) was added. The mixture was stirred for 6 d at RT and the solvent evaporated in vacuo. The residue was dissolved in MeOH.

Precipitation from Et2O gave (±)-198 (20 mg, quant.) as a white solid. 1 M.p. 180 °C (decomp); H NMR (600 MHz, CDCl3): 1.81–1.87 (m, 2H, H-C(7), H-C(8)), 1.99–2.08 (m, 2H, H-C(10)), 2.09–2.18 (m, 2H, H-C(7), H-C(8)), 2.85–2.90 (m, 1H, H-C(6)), 2.94–2.98 (m, 1H, H-C(6)), 3.09 (s, 9H, H-C(12)), 3.33–3.36 (m, 2H, H-C(11)), 3.45–3.49 (m, 2H, H-C(8b), H-C(9)), 3.56 (m, 1H, H-C(9)), 3.79–3.82 (m, 2H, H-C(3a), H-C(8a)), 4.28 (d, J = 8.9, 1H, H-C(4)), 7.09 (d, J = 4.0, 1H, H- C(1’)), 7.51 (d, J = 4.0, 1H, H-C(2’)), 7.79 (d, J = 0.7, 1H, H-C(3’)); 13C NMR (150

MHz, CDCl3): 22.8, 24.3, 30.6, 36.8, 50.6, 51.1, 52.4, 53.6 (t), 62.6, 65.4 (t), 69.1, 128.9, 129.0, 129.4, 134.4, 137.9, 141.5, 158.0, 177.6, 180.1; IR (ATR): 3383 (br.), 2959, 1774, 1694, 1593, 1478, 1431, 1403, 1351, 1319, 1241, 1182, 1066, 1026, 982, 929, 906, 877, 803, 725, 663; HR-MALDI-MS: m/z (%): 465.1526 (33, [M–Br]+, 37 + + calcd for C22H28 ClN4O3S : 465.1541), 463.1564 (100, [M–Br] , calcd for 35 + C22H28 ClN4O3S : 463.1565).

211 8. Experimental ______5-Chloro-2-thiophenecarbaldehyde Oxime (202) HO N Cl S H 1 2 5-Chloro-2-thiophenecarboxaldehyde (3.00 g, 20.5 mmol, 1 equiv.) and hydroxylamine hydrochloride (2.13 g, 30.7 mmol, 1.5 equiv.) were dissolved in EtOH (30 mL) at RT. Pyridine (1.5 mL, 18.6 mmol, 0.9 equiv.) was added, and the mixture was heated to reflux for 16 h. The solvents were evaporated in vacuo. The residue was taken up in water and stirred at 0 °C for 30 min. The solution was filtered,

washed with water, and evaporated in vacuo. Purification by CC (SiO2;

CH2Cl2/EtOAc 1:1) gave 202 (2.51 g, 76%) as a white solid, as a mixture of trans and cis isomers. 1 Rf = 0.51 (pentane/EtOAc 1:1); m.p. 106 °C; H NMR (300 MHz, CDCl3): 1.78 (s, 0.4H, OH), 6.94 (d, J = 4.0, 1H, H-C(1)), 7.15 (d, J = 4.0, 1H, H-C(2)), 7.61 (s, 1H, 13 H-C(3)), 9.35 (s, 1H, 0.6H, OH); C NMR (100 MHz, CDCl3): 125.6, 129.5, 131.0, + 37 + 137.0, 141.3; HR-ESI-TOF: m/z (%): 161.9775 (90, [M] , calcd for C5H4NOS Cl :

161.9775), 149.9530 (100); elemental analysis calcd for C5H4NOSCl (161.98): C 37.16, H 2.49, N 8.67; found: C 37.10, H 2.61, N 8.56.

5-Chloro-N-hydroxythiophene-2-carboximidoyl Chloride (203) HO N Cl S Cl 1 2

Compound 202 (0.88 g, 5.41 mmol, 1 equiv.) was dissolved in CHCl3 (72 mL). THF (3.6 mL), N-chlorosuccinimide (0.72 g, 5.41 mmol, 1 equiv.), and pyridine (54 μL, 0.66 mmol, 0.1 equiv.) were added to the solution at RT. The mixture was stirred at RT for 3 d. The solvents were evaporated in vacuo and the product was purified by

CC (SiO2; pentane/EtOAc 12:1) to give 203 (0.07 g, 7%) as a pale white solid. 1 Rf = 0.32 (pentane/EtOAc 12:1); m.p. 124 °C; H NMR (400 MHz, CDCl3): 6.88 (d, J = 4.0, 1H, H-C(1)), 7.31 (d, J = 4.0, 2H, H-C(2)), 7.68 (s, 1H, OH); 13C NMR (100

MHz, CDCl3): 126.6, 129.1, 133.5, 133.9, 134.3; IR (ATR): 3270, 3091, 3010, 2864, 2286, 1986, 1776, 1659, 1619, 1534, 1467, 1443, 1417, 1321, 1242, 1219, 1171,

212 8. Experimental ______1069, 1012, 987, 916, 888, 867, 804, 734, 670, 665, 645; HR-ESI-MS: m/z (%): + 37 + + 198.9236 (8, [M] , calcd for C5H3 Cl2NOS : 198.9253), 196.9275 (41, [M] , calcd for 35 37 + + 35 + C5H3 Cl ClNOS : 196.9253), 194.9305 (62, [M] , calcd for C5H3 Cl2NOS : + 37 + 194.9307), 160.9523 (34, [M–HCl] , calcd for C5H2 ClNOS : 160.9516), 158.9550 + 35 + + (85, [M–HCl] , calcd for C5H2 ClNOS : 158.9546), 144.9568 (38, [M–HClO] , calcd 37 + + 35 + for C5H2 ClNS : 144.9567), 142.9599 (100, [M–HClO] , calcd for C5H2 ClNS : 142.9596).

5-(5-Chloro-2-thienyl)-1,3-oxazole-2-carbaldehyde (207)

3 N O Cl S O 4 12

To crude 212 (0.9 g, 4.0 mmol, 1 equiv.) in CH2Cl2 (60 mL), PCC (1.3 g, 6.0 mmol, 1.5 equiv.) was added in portions. The mixture was stirred at RT for 20 h, filtrated

through a layer of SiO2, and evaporated in vacuo. Purification by CC (SiO2; pentane/CH2Cl2 1:4) gave 207 (0.3 g, 30%) as a light yellow solid. 1 Rf = 0.35 (CH2Cl2); m.p. 8687 °C; H NMR (400 MHz, CDCl3): 6.95 (d, J = 4.0, 1H, H-C(1)), 7.29 (d, J = 3.6, 1H, H-C(2)), 7.41 (s, 1H, H-C(3)), 9.71 (s, 1H, 13 H-C(4)); C NMR (100 MHz, CDCl3): 124.5, 126.6, 126.7, 127.6, 133.7, 149.3, 156.9, 176.6; IR (ATR): 3117, 3067, 2850, 1683, 1584, 1520, 1480, 1424, 1367, 1254, 1117, 1030, 1004, 969, 867, 798, 773, 671; HR-EI-MS: m/z (%): 212.9646 + 35 + (100, [M] , calcd for C8H4 ClNO2S : 212.9646), 144.9508 (64), 129.9637 (33), 94.9955 (30).

[2-(5-Chloro-2-thienyl)-2-oxoethyl]formylformamide (208) O O 4 S Cl N O 3 4 12 A mixture of 46 (6.5 g, 27.3 mmol, 1 equiv.) and sodium diformylamide (3.0 g,

31.7 mmol, 1.2 equiv.) in MeCN (40 mL) under N2 was stirred at RT for 7 h and then heated to 70 °C for 10 min. The hot mixture was filtered and the solid was washed with hot MeCN. The combined filtrates were evaporated in vacuo. Purification by

CC (SiO2; CH2Cl2) gave 208 (3.3 g, 52%) as a white solid.

213 8. Experimental ______

1 Rf = 0.09 (CH2Cl2); m.p. 103 °C; H NMR (400 MHz, CDCl3): 4.95 (s, 2H, H-C(3)), 7.01 (d, J = 4.4, 1H, H-C(1)), 7.60 (d, J = 4.0, 1H, H-C(2)), 9.01 (s, 2H, H-C(4)); 13 C NMR (150 MHz, CDCl3): 44.0, 127.9, 132.1, 139.1, 141.1, 163.0, 181.9; IR (ATR): 3096, 3079, 2930, 1658, 1412, 1347, 1299, 1235, 1140, 1015, 973, 949,

893, 823, 786, 742; elemental analysis calcd (%) for C8H6ClO3SN (231.7): C 41.48, H 2.61, N 6.05; found: C 41.46, H 2.63, N 5.93.

2-Amino-1-(5-chloro-2-thienyl)ethanone Hydrochloride (209) O Cl S – NH3Cl 3 12 A mixture of 208 (2.4 g, 10.3 mmol, 1 equiv.) and 5% ethanolic HCl (1.4 mL) was stirred at RT for 2.5 d. The mixture was evaporated in vacuo to give 209 (2.1 g, 96%) as a white solid. 1 M.p. 195 °C (decomp); H NMR (300 MHz, CD3OD): 4.49 (s, 2H, H-C(3)), 7.20 (d, 13 J = 4.2, 1H, H-C(1)), 7.84 (d, J = 4.2, 1H, H-C(2)); C NMR (100 MHz, CD3OD): 45.5, 129.9, 135.4, 139.9, 142.4, 185.4; IR (ATR): 3066, 3008, 2951 (br.), 2826, 2754, 2182, 2158, 1650, 1521, 1412, 1393, 1362, 1321, 1255, 1154, 1009, 903, 862, + 37 + 819; HR-ESI-TOF: m/z (%): 177.9901 (42, [M–Cl] , calcd for C6H7 ClOSN : + 35 + 177.9907), 175.9931 (100, [M–Cl] , calcd for C6H7 ClOSN : 175.9931).

Ethyl [2-(5-Chloro-2-thienyl)-2-oxoethyl]amino(oxo)acetate (210)

O O S H Cl N 4 3 O 5 12 O

To 209 (6.5 g, 30.7 mmol, 1 equiv.) in CH2Cl2 (100 mL) at –5 °C under N2, ethyl chlorooxoacetate (3.7 mL, 33.7 mmol, 1.1 equiv.) was added. The mixture was

stirred at 5 °C for 30 min. Et3N (13.2 mL, 95.0 mmol, 3.1 equiv.) was added keeping the temperature below 0 °C. The mixture was stirred at RT for 22 h and diluted with CH2Cl2. The org. phase was washed with water, dried over MgSO4, filtrated, and evaporated in vacuo to give 210 (8.5 g, 72%) as an orange solid. 1 Rf = 0.59 (CH2Cl2/EtOAc 1:4); m.p. 91 °C; H NMR (300 MHz, CDCl3): 1.41 (t, J = 7.2, 3H, H-C(5)), 4.39 (q, J = 7.2, 2H, H-C(4)), 4.69 (d, J = 5.1, 2H, H-C(3)), 7.01 (d,

214 8. Experimental ______J = 4.2, 1H, H-C(1)), 7.61 (d, J = 4.2, 1H, H-C(2)), 7.91 (br. s, 1H, NH); 13C NMR

(100 MHz, CDCl3): 45.8, 62.4, 63.3, 127.9, 132.2, 139.1, 141.3, 156.6, 159.7, 184.8; IR (ATR): 2855, 2046, 1665, 1458, 1409, 1244, 1013, 862, 812; HR-EI-MS: m/z (%): + 35 + 275.0016 (6, [M] , calcd for C10H10 ClNO4S : 275.0019), 201.9733 (7, [M– + 35 + COOCH2CH3] , calcd for C7H5 ClNO2S : 201.9730), 144.9492 (100, [M– + 35 + CH2NHCOCOOCH2CH3] , calcd for C5H2 ClOS : 144.9509); elemental analysis

calcd (%) for C10H10ClNO4S (275.0): C 43.56, H 3.66, N 5.08; found: C 43.36, H 3.78, N 5.23.

Ethyl 5-(5-Chloro-2-thienyl)-1,3-oxazole-2-carboxylate (211)

3 N 5 O Cl S O 4 O 12

A solution of 210 (1.2 g, 4.3 mmol, 1 equiv.) in POCl3 (11.1 mL, 120.6 mmol, 28.7 equiv.) was heated to reflux for 17 h. The mixture was cooled to RT, carefully poured onto ice, and extracted with EtOAc. The org. phase was neutralized with sat.

aq. NaHCO3 solution and washed with water and brine. The org. phase was dried over

MgSO4 and evaporated in vacuo. Purification by CC (SiO2; pentane/CH2Cl2 1:4) gave 211 (0.9 g, 80%) as a light yellow solid. 1 Rf = 0.37 (CH2Cl2); m.p. 6971 °C; H NMR (400 MHz, CDCl3): 1.45 (t, J = 7.1, 3H, H-C(5)), 4.49 (q, J = 7.1, 2H, H-C(4)), 6.94 (d, J = 4.0, 1H, H-C(1)), 7.25 (d, J = 4.4, 13 1H, H-C(2)), 7.32 (s, 1H, H-C(3)); C NMR (100 MHz, CDCl3): 14.1, 62.7, 123.5, 125.7, 126.8, 127.2, 132.0, 148.6, 151.2, 155.3; IR (ATR): 3068, 2964, 1721, 1497, 1377, 1327, 1247, 1210, 1173, 1015, 904, 849, 804, 773, 642; HR-EI-MS: m/z (%): + 35 + 256.9910 (100, [M] , calcd for C10H8 ClNO3S : 256.9908), 144.9522 (49, [M– + 35 + CHNCOCOOCH2CH3] , calcd for C5H2 ClOS : 144.9515); 29.0702 (63); elemental

analysis calcd (%) for C10H8ClNO3S (257.0): C 46.61, H 3.13, N 5.44; found: C 46.89, H 3.15, N 5.25.

215 8. Experimental ______[5-(5-Chloro-2-thienyl)-1,3-oxazol-2-yl]methanol (212)

3 N OH Cl S O 4 12 To a solution of compound 211 (1.1 g, 4.2 mmol, 1 equiv.) in EtOH (30 mL) at 0 °C,

NaBH4 (0.16 g, 4.2 mmol, 1 equiv.) was added in portions. The mixture was stirred at 0 °C for 10 min, at RT for 44 h, and evaporated in vacuo. The off-white solid was

taken up in water, and the pH was set to 3 with aq. 1 M HCl solution. The

precipitated solid was filtered, dissolved in EtOAc/CH2Cl2, dried over MgSO4, filtrated, and evaporated in vacuo to give crude 212 (0.9 g, 80%) as a yellow solid. Product was used in the next step without further purification. + 35 + HR-EI-MS: m/z (%): 214.9801 (100, [M] , calcd for C8H6 ClNO2S : 214.9803), 144.9501 (46), 70.0288 (81).

2-(5-Chloro-2-thienyl)-1,3-oxazole-5-carbaldehyde (213)

3 N Cl S O O 4 1 2

A solution of CuCl2 (222 mg, 1.65 mmol, 3.3 equiv.) and [PdCl2(MeCN)2] (7 mg,

0.03 mmol, 0.05 equiv.) in DMF (10 mL) was stirred at RT under N2 for 30 min. Compound 216 (100 mg, 0.50 mmol, 1 equiv.) was dissolved in DMF (7 mL) and added to the mixture. The mixture was stirred at 100 °C for 21 h, cooled to RT, and

diluted with brine. The mixture was extracted with Et2O. The org. phase was washed with brine, dried over Na2SO4, filtrated, and evaporated in vacuo. Purification by CC

(SiO2; CH2Cl2) gave 213 (8 mg, 7%) as a yellow solid. 1 Rf = 0.28 (CH2Cl2); m.p. = 112 °C; H NMR (300 MHz, CDCl3): 7.02 (d, J = 4.1, 1H, H-C(1)), 7.68 (d, J = 4.1, 1H, H-C(2)), 7.88 (s, 1H, H-C(3)), 9.77 (s, 1H, H-C(4)); 13 C NMR (100 MHz, CDCl3): 126.8, 128.1, 130.5, 137.0, 139.2, 149.4, 160.4, 175.9; IR (ATR): 3071, 2924, 2853, 1727, 1678, 1662, 1576, 1552, 1480, 1417, 1382, 1327, 1263, 1232, 1216, 1148, 1088, 1036, 1000, 971, 921, 818, 805, 782, 731, 672, 620; + + HR-EI-MS: m/z (%): 212.9648 (100, [M] , calcd for C8H4ClNO2S : 212.9646), + + 183.9616 (80, [M–CHO] , calcd for C7H3ClNOS : 183.9618), 155.9669 (83, [M– + + OCCHO] , calcd for C6H3ClNS : 155.9669).

216 8. Experimental ______5-Chloro-2-thiophenecarbonyl Chloride (215) O Cl S Cl 12 To 5-chlorothiophene-2-carboxylic acid (1.0 g, 6.15 mmol, 1 equiv.) in toluene

(50 mL) at RT under N2, SOCl2 (1.80 mL, 24.6 mmol, 4 equiv.) and DMF (0.2 mL, 2.6 mmol, 0.4 equiv.) were added. The mixture was stirred for 45 min at 80 °C and

evaporated in vacuo. Purification by CC (SiO2; pentane) gave 215 (801 mg, 72%) as a white solid. 1 Rf = 0.81 (pentane/EtOAc 1:1); H NMR (300 MHz, CDCl3): 7.04 (d, J = 4.2, 1H, H-C(1)), 7.80 (d, J = 4.2, 1H, H-C(2)); HR-EI-MS: m/z (%): 181.9155 (8, [M]+, calcd 37 35 + + 35 + for C5H2 Cl ClOS : 181.9174), 179.9199 (13, [M] , calcd for C5H2 Cl2OS : + 37 + 179.9198), 146.9457 (100, [M–Cl] , calcd for C5H2 ClOS : 146.9485), 144.9510 + 35 + + (100, [M–Cl] , calcd for C5H2 ClOS : 144.9515), 74.9801 (10, [M–C5H3ClNOS] , 37 + + 35 + calcd for C3H2 Cl : 74.9816), 72.9839 (36, [M–C5H3ClNOS] , calcd for C3H2 Cl : 72.9845).

5-Chloro-N-2-propyn-1-yl-2-thiophenecarboxamide (216) O S Cl 3 N H 1 2 4

Compound 215 (2.71 g, 15.0 mmol, 1 equiv.) in CH2Cl2 (20 mL) was cooled to 0 °C.

Et3N (1.16 mL, 6.9 mmol, 1 equiv.) and propargylamine (0.96 mL, 14.8 mmol,

1 equiv.) in CH2Cl2 (60 mL) were added and the mixture was stirred at RT for 2 h. The mixture was washed with 5% HCl solution (100 mL), the organic phase dried

over MgSO4, filtrated, and evaporated in vacuo. Purification by CC (SiO2; CH2Cl2) gave 216 (1.72 g, 58%) as a white solid. 1 Rf = 0.28 (CH2Cl2); m.p. = 141 °C; H NMR (300 MHz, CDCl3): 2.29 (t, J = 2.6, 1H, H-C(4)), 4.21 (dd, J = 5.3, 2.6, 2H, H-C(3)), 6.03 (br. s, 1H, NH), 6.91 (d, J = 4.0, 13 1H, H-C(1)), 7.28 (d, J = 4.0, 1H, H-C(2)); C NMR (75 MHz, CDCl3): 29.9, 72.4, 79.1, 127.1, 127.7, 136.0, 136.6, 160.6; IR (ATR): 3291, 3079, 1617, 1546, 1518, 1416, 1351, 1327, 1306, 1270, 1257, 1227, 1144, 1072, 1043, 1004, 957, 915, 905, 804, 790, 745, 667; HR-EI-MS: m/z (%): 201.9906 (36, [M+H]+, calcd for

217 8. Experimental ______

37 + + 35 + C8H7 ClNOS : 201.9907), 199.9935 (100, [M+H] , calcd for C8H7 ClNOS :

199.9931); elemental analysis calcd (%) for C8H6ClNOS (199.7): C 48.13, H 3.03, N 7.02; found: C 48.05, H 3.03, N 6.91.

2-(5-Chloro-2-thienyl)-1,3-oxazole-4-carbaldehyde (217)

3 O O Cl S N 4 12

To a solution of 220 (5.90 g, 27.4 mmol, 1 equiv.) in CH2Cl2 (270 mL) under N2 at RT, PCC (10.02 g, 46.5 mmol, 1.7 equiv.) was added. The mixture was stirred for

17 h at RT, filtered, and the solvent was removed in vacuo. Purification by CC (SiO2; pentane/CH2Cl2 3:1  CH2Cl2) gave 217 (3.18 g, 54%) as a light yellow solid. 1 Rf = 0.80 (pentane/EtOAc 2:1); m.p. 125–130 °C; H NMR (600 MHz, CDCl3): 6.98 (d, J = 4.0, 1H, H-C(1)), 7.56 (d, J = 4.0, 1H, H-C(2)), 8.24 (s, 1H, H-C(3)), 9.97 (s, 13 1H, H-C(4)); C NMR (150 MHz, CDCl3): 126.4, 127.2, 128.5, 134.9, 141.4, 143.4, 157.6, 183.4; IR (ATR): 3125, 3081, 2852, 1682, 1606, 1592, 1553, 1519, 1505, 1453, 1424, 1388, 1349, 1334, 1319, 1305, 1265, 1206, 1114, 1073, 1021, 989, 975, 912, 899, 856, 787, 722, 668, 626; HR-EI-MS: m/z (%): 214.9610 (37, [M]+, calcd for 37 + + 35 + C8H4 ClNO2S : 214.9622), 212.9644 (100, [M] , calcd for C8H4 ClNO2S :

212.9646); elemental analysis calcd (%) for C8H4ClNO2S (213.6): C 44.98, H 1.89, N 6.56; found: C 44.72, H 1.95, N 6.62.

5-Chloro-2-thiophenecarboxamide (218)[213] O Cl S NH2 12

To 215 (510 mg, 2.82 mmol, 1 equiv.) in toluene (5 mL) at 0 °C under N2, 7 N NH3 in

CH3OH (14.08 mL, 98.60 mmol, 35 equiv.) was added dropwise. The mixture was stirred for 1.5 h at RT. Toluene was added until a white solid precipitated, the mixture was filtrated, and the solvent was removed in vacuo. Purification by CC

(SiO2, pentane/EtOAc 1:1) gave 218 (417 mg, 92%) as white crystals. [214] 1 Rf = 0.36 (pentane/EtOAc 1:1); m.p. 177 °C (Lit. : 166–168 °C); H NMR

(500 MHz, CD3OD): 7.01 (d, J = 4.1, 1H, H-C(1)), 7.53 (d, J = 4.1, 1H, H-C(2));

218 8. Experimental ______

13 C NMR (125 MHz, CD3OD): 128.7, 130.2, 136.6, 139.2, 165.4; IR (ATR): 3378, 3129, 1648, 1605, 1425, 1389, 1221, 1117, 1060, 993, 776, 706, 661; HR-ESI-MS: + 37 + m/z (%): 163.9741 (30, [M] , calcd for C5H5 ClNOS : 163.9737), 161.9775 (100, + 35 + [M] , calcd for C5H5 ClNOS : 161.9771).

4-(Chloromethyl)-2-(5-chloro-2-thienyl)-1,3-oxazole (219)

3 O Cl Cl S N 4 12 Compound 218 (0.38 g, 2.35 mmol, 1 equiv.) and dichloroacetone (1.15 g,

9.03 mmol, 3.84 equiv.) under N2 were heated to 130 °C. After stirring for 18 h, water was added and the mixture was extracted with CH2Cl2. The org. phase was

washed with aq. 2 N NaOH solution. The combined organic layers were dried over

MgSO4, filtrated, and evaporated in vacuo to give 219 (347 mg, 63%) as a white solid. 1 Rf = 0.52 (pentane/CH2Cl2 5:1); m.p. 89 °C; H NMR (600 MHz, CD2Cl2): 4.53 (d, J = 0.9, 2H, H-C(4)), 6.97 (d, J = 4.0, 1H, H-C(1)), 7.46 (d, J = 4.0, 1H, H-C(2)), 7.66 13 (t, J = 0.9, 1H, H-C(3)); C NMR (150 MHz, CD2Cl2): 37.3, 127.80, 127.83, 128.4, 133.9, 136.4, 139.2, 157.5; IR (ATR): 3130, 1594, 1502, 1429, 1346, 1271, 1214, 1153, 1107, 1073, 1024, 994, 982, 894, 812, 786, 742, 715, 693, 665; HR-EI-MS: m/z + 37 + + (%): 236.9410 (10, [M] , calcd for C8H5 Cl2NOS : 236.9410), 234.9436 (51, [M] , 37 35 + + calcd for C8H5 Cl ClNOS : 234.9434), 232.9464 (76, [M] , calcd for 35 + + 37 + C8H5 Cl2NOS : 232.9464), 199.9744 (35, [M–Cl] , calcd for C8H5 ClNOS : + 35 + 199.9751), 197.9772 (100, [M–Cl] , calcd for C8H5 ClNOS : 197.9780), 144.9668 + 37 + + (36, M–C3H3ClO] , calcd for C5H2 ClNS : 144.9567), 142.9698 (90, M–C3H3ClO] , 35 + calcd for C5H2 ClNS : 142.9698); elemental analysis calcd (%) for C8H5Cl2NOS (233.0): C 41.04, H 2.15, N 5.98; found: C 40.82, H 2.28, N 5.95.

219 8. Experimental ______[2-(5-Chloro-2-thienyl)-1,3-oxazol-4-yl]methanol (220)

3 O OH Cl S N 4 12 To 219 (10.35 g, 44 mmol, 1 equiv.) in DMF (200 mL) at RT, NaOAc (14.51 g, 177 mmol, 4 equiv.) was added. The mixture was stirred at 80 °C for 15 h, diluted with water (300 mL), and extracted with EtOAc (3 x 300 mL). The org. phase was washed

with brine (500 mL), dried over MgSO4, and evaporated in vacuo. To a solution of

the crude acetate (11.34 g, 44 mmol, 1 equiv.) in MeOH (250 mL) at RT, H2O (50

mL) and K2CO3 (30.41 g, 220 mmol, 5 equiv.) were added. The mixture was stirred for 1 h at RT, diluted with water (250 mL), and extracted with EtOAc (2 x 300 mL).

The combined organic layers were washed with brine (250 mL), dried over MgSO4, and evaporated in vacuo. Purification by CC (SiO2; pentane/EtOAc 1:2) gave 220 (5.97 g, 62%) as orange crystals. 1 Rf = 0.23 (pentane/EtOAc 1:1); m.p. 123 °C; H NMR (600 MHz, CDCl3): 2.23 (t, J = 6.1, 1H, OH), 4.64 (dd, J = 6.1, 1.0, 2H, H-C(4)), 6.93 (d, J = 4.0, 1H, H-C(1)), 7.44 13 (d, J = 4.0, 1H, H-C(2)), 7.57 (t, J = 1.0, 1H, H-C(3)); C NMR (150 MHz, CDCl3): 57.1, 127.35, 127.36, 128.3, 133.7, 134.8, 141.8, 157.3; IR (ATR): 3255, 3110, 2932, 2914, 2865, 1681, 1591, 1504, 1427, 1365, 1348, 1321, 1291, 1217, 1204, 1117, 1095, 1052, 989, 951, 883, 801, 760, 712, 671, 652, 627; HR-EI-MS: m/z (%): + 37 + + 216.9774 (36, [M] , calcd for C8H6 ClNO2S : 216.9778), 214.9806 (100, [M] , calcd 35 + for C8H6 ClNO2S : 214.9803); elemental analysis calcd (%) for C8H6ClNO2S (215.7): C 44.56, H 2.80, N 6.49; found: C 44.30, H 2.87, N 6.64.

220 8. Experimental ______(3aSR,4RS,8aSR,8bRS)-2-(3-Bromopropyl)-4-[5-(5-chloro-2-thienyl)-1,3-oxazol- 2-yl]hexahydrodipyrrolo[1,2-a:3’,4’-c] pyrrole-1,3(2H,4H)-dione ((±)-221) O 8 9 H 11 8a N 8b 7 3a Br 10 N 6 O H 4 N O

3' 2' S 1' Cl A suspension of maleimide 36 (287 mg, 1.32 mmol, 1.0 equiv.), L-proline (18) (159 mg, 1.38 mmol, 1.1 equiv.), and 207 (295 mg, 1.38 mmol, 1.1 equiv.) in MeCN (4 mL) was heated to reflux for 22 h, leading to a mixture of products, and concentrated in vacuo. Repeated purification by CC (SiO2; CH2Cl2/EtOAc 1:1 

EtOAc; CH2Cl2/EtOAc 2:1; Et2O/EtOAc 4:1), prep. TLC (Silicycle, Ultra Pure Silica

Gel 60 F254, 1000 m; CH2Cl2/Et2O 1:1), and HPLC (Regis Spherical Kromasil; –1 SiO2, 100 Å, 5 m; 9 mL min ; CH2Cl2/EtOAc 3:7  2:8, 40 min) gave (±)-221 (36 mg, 6%) as a light brown solid. 1 Rf = 0.18 (CH2Cl2/EtOAc 4:1); m.p. 110 °C (decomp); H NMR (300 MHz, CDCl3): 1.63–1.76 (m, 1H, H-C(8)), 1.79–1.93 (m, 1H, H-C(7)), 2.00–2.27 (m, 4H, H-C(7), H-C(8), H-C(10)), 2.81 (ddd, J = 12.5, 7.9, 4.9, 1H, H-C(6)), 3.02–3.11 (m, 1H, H-C(6)), 3.29–3.37 (m, 3H, H-C(8b), H-C(11)), 3.59 (t, J = 6.9, 2H, H-C(9)), 3.74 (dd, J = 8.6, 8.6, 1H, H-C(3a)), 3.88 (ddd, J = 8.7, 7.2, 1.8, 1H, H-C(8a)), 4.38 (d, J = 8.6, 1H, H-C(4)), 6.88 (d, J = 3.9, 1H, H-C(1’)), 7.04 (d, J = 3.9, 1H, H-C(2’)), 7.11 13 (s, 1H, H-C(3’)); C NMR (100 MHz, CDCl3): 23.7, 30.0, 30.1, 30.8, 38.1, 49.4, 49.9, 51.8, 63.0, 68.1, 122.2, 123.9, 127.0, 128.3, 130.8, 146.5, 160.5, 175.3, 177.6; IR (ATR): 2957, 1772, 1692, 1559, 1508, 1434, 1403, 1349, 1234, 1156, 1096, 1028, 998, 904, 877, 792, 732, 632; HR-MALDI-MS: m/z (%): 487.0126 (20), 486.0079 + 81 35 + (100, [M+H] , calcd for C19H20 Br ClN3O3S : 486.0079), 485.0157 (11), 484.0087 + 79 35 + (72, [M+H] , calcd for C19H20 Br ClN3O3S : 484.0092).

221 8. Experimental ______(3aRS,4RS,8aSR,8bSR)-2-(3-Bromopropyl)-4-[2-(5-chloro-2-thienyl)-1,3-oxazol- 4-yl]hexahydrodipyrrolo[1,2-a:3',4'-c]pyrrole-1,3(2H,4H)-dione ((±)-222) O 8 9 H 11 8a N 8b 7 Br 10 3a N 6 O H 4 3' N O

2' S 1' Cl A suspension of 36 (1.49 g, 6.85 mmol, 1 equiv.), L-proline (18) (0.87 g, 7.54 mmol, 1.1 equiv.), and 217 (1.61 g, 7.54 mmol, 1.1 equiv.), in MeCN (20 mL) was heated to reflux for 3 d, leading to a mixture of products, and concentrated in vacuo. Repeated

purification by CC (SiO2; CH2Cl2/EtOAc 3:1; CH2Cl2/Et2O 2:1; CH2Cl2/Et2O 3:1) –1 and HPLC (Regis Spherical Kromasil; SiO2, 100 Å, 5 m; 9 mL min ;

CH2Cl2/EtOAc 3:7  1:9, 40 min) gave (±)-222 (108 mg, 7%) as yellow oil. 1 Rf = 0.23 (CH2Cl2/Et2O 2:1); H NMR (400 MHz, CDCl3): 1.63–1.73 (m, 1H, H-C(8)), 1.76–1.87 (m, 1H, H-C(7)), 1.98–2.03 (m, 1H, H-C(7)), 2.05–2.12 (m, 2H, H-C(10)), 2.13–2.22 (m, 1H, H-C(8)), 2.82 (ddd, J = 12.5, 8.0, 4.8, 1H, H-C(6)), 2.97 (ddd, J = 12.2, 8.9, 6.8, 1H, H-C(6)), 3.27 (dd, J = 8.0, 2.0, 1H, H-C(8b)), 3.30 (t, J = 6.8, 2H, H-C(11)), 3.56 (t, J = 6.9, 2H, H-C(9)), 3.68 (dd, J = 8.4, 8.4, 1H, H-C(3a)), 3.81–3.86 (m, 1H, H-C(8a)), 4.17 (d, J = 8.5, 1H, H-C(4)), 6.91 (d, J = 4.0, 1H, H- C(1’)), 7.41 (d, J = 4.0, 1H, H-C(2’)), 7.48 (s, 1H, H-C(3’)); 13C NMR (100 MHz,

CDCl3): 23.7, 29.8, 29.9, 30.9, 37.9, 49.3, 49.8, 51.6, 62.3, 68.0, 127.27, 127.30, 128.4, 133.5, 135.7, 140.8, 157.0, 175.4, 178.0; IR (ATR): 2921, 2851, 1779, 1703, 1698, 1694, 1682, 1673, 1651, 1599, 1557, 1538, 1505, 1428, 1402, 1373, 1258, 1177, 1109, 1027, 900, 797, 719, 667; HR-MALDI-MS: m/z (%): 488.0046 (28, + 81 37 + + [M+H] , calcd for C19H20 Br ClN3O3S : 488.0057), 486.0074 (100, [M+H] , calcd 81 35 + 79 37 + for C19H20 Br ClN3O3S or C19H20 Br ClN3O3S : 486.0077), 484.0094 (64, + 79 35 + [M+H] , calcd for C19H20 Br ClN3O3S : 484.0092).

222 8. Experimental ______N-(2-Amino-2-oxoethyl)-3-[(3aRS,4RS,8aSR,8bSR)-4-[5-(5-chloro-2-thienyl)-3- isoxazolyl]-1,3-dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N,N- dimethyl-1-propanaminium Bromide ((±)-224) O 8 12 9 H 11 8a 7 13 N 8b H2N N 10 3a N 6 O 4 O 12 H Br N 3' O

2' S 1' Cl To a solution of (±)-57 (23 mg, 0.05 mmol, 1 equiv.) in EtOH (1.5 mL) at RT, 230 (63 mg, 0.62 mmol, 13 equiv.) was added. The mixture was stirred at RT for 3 d and at 40 °C for 4 d. The mixture was evaporated in vacuo, dissolved in MeOH, and

precipitation from Et2O gave (±)-224 (20 mg, 71%) as an off-white solid. 1 M.p. 207 °C (decomp.); H NMR (600 MHz, CDCl3): 1.80–1.88 (m, 2H, H-C(7), H-C(8)), 2.02–2.09 (m, 2H, H-C(10)), 2.11–2.19 (m, 2H, H-C(7), H-C(8)), 2.86 (ddd, 1H, J = 12.7, 8.4, 4.3, H-C(6)), 2.96 (ddd, J = 12.6, 8.6, 7.0, 1H, H-C(6)), 3.28 (s, 3H, H-C(12)), 3.30 (s, 3H, H-C(12)), 3.49 (dd, J = 8.1, 1.5, 1H, H-C(8b)), 3.52–3.63 (m, 4H, H-C(9), H-C(11)), 3.74 (br. t, J = 7.7, 1H, H-C(8a)), 3.84 (dd, J = 8.4, 8.4, 1H, H-C(3a)), 4.07 (s, 2H, H-C(13)), 4.41 (d, J = 8.7, 1H, H-C(4)), 6.64 (s, 1H, H-C(3’)), 7.11 (d, J = 4.0, 1H, H-C(1’)), 7.43 (d, J = 4.0, 1H, H-C(2’)); 13C NMR (150 MHz,

CDCl3): 22.5, 24.2, 30.3, 36.6, 50.6, 50.8, 52.2, 52.5, 52.7, 62.5, 62.7, 64.3, 69.6, 101.2, 128.1, 129.0, 129.1, 134.2, 164.9, 165.1, 166.9, 177.5, 180.0; IR (ATR): 3219, 3106, 2928, 1774, 1705, 1595, 1532, 1470, 1441, 1426, 1396, 1356, 1324, 1303, 1242, 1218, 1204, 1164, 1080, 1063, 1026, 997, 952, 907, 898, 871, 824, 789, 737, 667, 638, 609; HR-MALDI-MS: m/z (%): 506.1631 (100, [M–Br]+, calcd for 35 + C23H29 ClN5O4S : 506.1623).

223 8. Experimental ______3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]-N-(2-hydroxyethyl)-N,N- dimethyl-1-propanaminium Bromide ((±)-226) O 8 12 9 H 11 8a 13 N 8b 7 3a HO N 10 N 14 6 O H 4 12 Br N 3' O

2' S 1' Cl To a solution of (±)-57 (30 mg, 0.06 mmol, 1 equiv.) in EtOH (1 mL) at RT, N,N-dimethylethanolamine (125 L, 1.24 mmol, 20 equiv.) was added, the mixture stirred at RT for 24 h, and at 40 °C for further 24 h. The mixture was evaporated in vacuo and the residue dissolved in MeOH. Precipitation from Et2O gave (±)-226 (26 mg, 73%) as a pink solid. 1 M.p. 180 °C (decomp); H NMR (600 MHz, CD3OD): 1.79–1.89 (m, 2H, H-C(7), H-C(8)), 2.00–2.07 (m, 1H, H-C(10)), 2.09–2.19 (m, 3H, H-C(7), H-C(8), H-C(10)), 2.87 (ddd, J = 12.7, 8.3, 4.3, 1H, H-C(6)), 2.96 (ddd, J = 12.7, 8.8, 7.0, 1H, H-C(6)), 3.16 (s, 3H, H-C(12), 3.20 (s, 3H, H-C(12), 3.34–3.61 (m, 7H, H-C(8b), H-C(9), H-C(11), H-C(13)) 3.70–3.73 (m, 1H, H-C(8a)), 3.84 (dd, J = 8.4, 8.4, 1H, H-C(3a)), 3.90–3.95 (m, 1H, H-C(14)), 4.12 (ddq, J = 14.0, 7.2, 2.4, 1H, H-C(14)), 4.44 (d, J = 8.8, 1H, H-C(4)), 6.66 (s, 1H, H-C(3’)), 7.11 (d, J = 4.0, 1H, H-C(1’)), 7.44 (d, J = 13 3.8, 1H, H-C(2’)); C NMR (150 MHz, CD3OD): 22.5, 24.1, 30.2, 36.6, 50.4, 50.7, 52.2, 52.5 (t), 52.6 (t), 56.8, 62.7, 63.9, 66.3, 69.5, 101.4, 128.1, 128.96, 129.02, 134.2, 164.7, 165.1, 177.4, 179.8; IR (ATR): 3283, 3008, 2955, 2882, 1768, 1696, 1600, 1530, 1476, 1428, 1400, 1351, 1313, 1233, 1188, 1156, 1067, 1000, 963, 923, 895, 802, 742, 667, 632; HR-MALDI-MS: m/z (%): 495.1650 (37, [M–Br]+, calcd for 37 + + 35 + C23H30 ClN4O4S : 495.1644), 493.1678 (100, [M–Br] , calcd for C23H30 ClN4O4S :

493.1671); elemental analysis calcd (%) for C23H30BrClN4O4S (572.1): C 48.13, H 5.27, N 9.76; found C 47.84, H 5.32, N 9.58.

224 8. Experimental ______1-{3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-3-isoxazolyl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]propyl}-3- (hydroxymethyl)pyridinium Bromide ((±)-228) O 8 9 H 11 8a 15 N 8b 7 3a 14 N 10 N 6 12 O H 4 13 Br N 3' 16 HO O

2' S 1' Cl To a solution of (±)-57 (30 mg, 0.06 mmol, 1 equiv.) in EtOH (1 mL) at RT, 3-(hydroxymethyl)pyridine (120 L, 1.24 mmol, 20 equiv.) was added, and the mixture stirred at RT for 29 h and further 67 h at 40 °C. The mixture was evaporated in vacuo. The residue was dissolved in MeOH. Precipitation from Et2O gave (±)-228 (35 mg, 95%) as yellow crystals. 1 M.p. 188 °C (decomp); H NMR (400 MHz, CD3OD): 1.80–1.94 (m, 2H, H-C(7), H-C(8)), 2.10–2.21 (m, 2H, H-C(7), H-C(8)), 2.23–2.30 (m, 2H, H-C(10)), 2.87 (ddd, J = 12.5, 8.2, 4.3, 1H, H-C(6)), 3.00 (ddd, J = 12.5, 8.6, 6.9, 1H, H-C(6)), 3.46–3.61 (m, 3H, H-C(8b), H-C(9)), 3.79 (ddd, J = 9.1, 7.3, 1.6, 1H, H-C(8a)), 3.88 (dd, J = 8.5, 8.5, 1H, H-C(3a)), 4.45 (d, J = 8.7, 1H, H-C(4)), 4.63 (td, J = 7.6, 1.2, 2H, H- C(11)), 4.84 (s, 2H, H-C(16)), 6.62 (s, 1H, H-C(3’)), 7.09 (d, J = 4.0, 1H, H-C(1’)), 7.40 (d, J = 4.0, 1H, H-C(2’)), 8.07 (dd, J = 8.0, 6.1, 1H, H-C(14)), 8.55 (d, J = 8.1, 1H, H-C(13)), 8.92 (d, J = 6.1, 1H, H-C(15)), 8.99 (s, 1H, H-C(12)); 13C NMR (100

MHz, CD3OD): 24.2, 30.4, 30.6, 36.4, 50.8, 52.4, 60.4, 61.1, 62.6, 69.5, 101.2, 128.1, 128.9, 129.0, 134.1, 143.9, 144.5, 144.6, 145.5, 164.9, 165.0, 177.5, 180.0, 2 signals not visible; IR (ATR): 3303, 3021, 2950, 2868, 1769, 1699, 1612, 1474, 1427, 1397, 1344, 1297, 1207, 1173, 1065, 998, 920, 878, 796, 687; HR-MALDI-MS: m/z (%): + 37 + 515.1337 (43, [M–Br] , calcd for C25H26 ClN4O4S : 515.1332), 513.1359 (100, [M– + 35 + Br] , calcd for C25H26 ClN4O4S : 513.1358); elemental analysis calcd (%) for

C25H26BrClN4O4S (592.1): C 50.56, H 4.41, N 9.43; found: C 50.83, H 4.63, N 9.29.

225 8. Experimental ______1-{3-[(3aSR,4RS,8aSR,8bRS)-4-[5-(5-Chloro-2-thienyl)-1,2-oxazol-3-yl]-1,3- dioxooctahydrodipyrrolo[1,2-a:3',4'-c]pyrrol-2(3H)-yl]propyl}-1-(2- hydroxyethyl)pyrrolidinium Bromide ((±)-229) HO O 8 14 9 H 15 11 8a N 8b 7 12 3a N 10 N 13 6 12 O H 4 13 Br N 3' O

2' S 1' Cl To a solution of (±)-57 (20 mg, 0.041 mmol, 1 equiv.) in EtOH (1 mL) at RT, N-(2-hydroxyethyl)pyrrolidine (97 L, 0.825 mmol, 20 equiv.) was added. The mixture was stirred at RT for 17 h, at 40 °C for 4 d, and evaporated to dryness. The

residue was dissolved in MeOH. Precipitation from Et2O gave (±)-229 (3 mg, 12%) as colorless crystals. 1 M.p. 182 °C; H NMR (600 MHz, CDCl3): 1.79–1.90 (m, 2H, H-C(7), H-C(8)), 1.96– 2.03 (m, 1H, H-C(13)), 2.12–2.25 (m, 5H, H-C(7), H-C(8), H-C(10), H-C(13)), 2.88 (ddd, J = 12.7, 8.3, 4.3, 1H, H-C(6)), 2.95 (ddd, J = 12.8, 8.9, 7.1, 1H, H-C(6)), 3.34– 3.76 (m, 9 H, H-C(8a), H-C(8b), H-C(9), H-C(11), H-C(12), H-C(14)), 3.82 (dd, J = 8.4, 8.4, 1H, H-C(3a)), 3.87–3.90 (m, 1H, H-C(15)), 4.14–4.17 (m, 1H, H-C(15)), 4.43 (d, J = 8.7, 1H, H-C(4)), 6.68 (s, 1H, H-C(3’)), 7.12 (d, J = 4.0, 1H, H-C(1’)), 7.44 (d, J = 4.0, 1H, H-C(2’)), protons hidden under solvent peak; 13C NMR (150

MHz, CDCl3): 22.5, 22.6, 23.2, 24.1, 30.0, 36.6, 50.1, 50.7, 52.0, 57.3, 59.7, 61.9, 62.7, 64.6, 65.0, 69.5, 101.5, 128.2, 128.9, 129.1, 134.2, 164.5, 165.2, 177.4, 179.7; IR (ATR): 3246, 2950, 1764, 1700, 1696, 1607, 1518, 1472, 1430, 1418, 1399, 1366, 1232, 1182,1151, 1089, 1060, 1026, 995, 953, 922, 868, 811, 789, 671; HR-MALDI- + 37 + MS: m/z (%): 521.1791 (41, [M–Br] , calcd for C25H32 ClN4O4S : 521.1801), + 35 + 519.1827 (100, [M–Br] , calcd for C25H32 ClN4O4S : 519.1827), 334.2123 (18, [M– + + C7H4BrClNOS] , calcd for C18H28N3O3 : 334.2125).

226 8. Experimental ______2-(Dimethylamino)acetamide (230)

2 O N H N 2 1 2

To a solution of 7.9 M Me2NH in H2O (1.84 mL, 14.5 mmol, 20 equiv.) and EtOH (1.0 mL) at 0 °C was added 2-bromoacetamide (100 mg, 0.7 mmol, 1 equiv.) in EtOH (0.6 mL). The mixture was stirred 1.5 h at 0 °C and extracted with EtOAc. The org. phase was dried over Na2SO4, filtrated, and evaporated in vacuo. Compound 230 (18 mg, 24%) was obtained as a yellow solid. [215] 1 M.p. 91 °C (Lit. : 95–96 °C); H NMR (300 MHz, CD3OD): 2.35 (s, 6H, H-C(2)), 3.03 (s, 2H, H-C(1)).

227 8. Experimental ______8.3. Enzyme Assays

All biological assays were conducted at F. Hoffmann-La Roche in Basel, Switzerland, in the laboratory of Dr. Jacques Himber and Jean-Luc Mary, with the guidance of Olivier Kuster.

8.3.1. Factor Xa[108]

A 10 mM stock solution of the inhibitors was prepared in Me2SO, which was diluted

stepwise 1:10 in 96-well microtiter plates using Me2SO, and each well subsequently

1:40 using HNPT buffer (Hepes 100 mM, NaCl 140 mM, PEG 6000 0.1%, Tween 80 0.02%) at pH 7.8. The dilution series were made in duplicate for each inhibitor, which allowed the measurement of six inhibitors per plate at seven different concentrations and one blank. Factor Xa activity was measured spectrophotometrically in microtiter plates in a final volume of 250 L. Inhibition of human factor Xa was tested at an enzyme concentration of 3 nM. First, the enzyme in HNPT buffer was added to the aqueous inhibitor solutions, and the plates were incubated at RT for 10 min. The chromogenic

substrate S-2222 (Chromogenix AB, Mölndal, Sweden, Km = 613 M, final

concentration of 200 M) solution was added, and the reaction followed spectrophotometrically at 405 nm for 5 min at 298 K. The velocity of the reaction was determined by the autoreader from the slope of the linear regression fit to 7 time points (1 min). The initial velocity for each inhibitor concentration was determined by the slope of at least four time points in the linear phase by a linear regression fit (mOD/min2). Apparent dissociation constants [111] Ki were calculated based on the IC50 and the respective Km.

Given Ki values are averages of 1–4 duplicate measurements. The chosen dataset show a good match of the duplicate value measured spectrophotometrically for the velocity of the reaction. As examples, the Ki values with the errors (standard

deviation) for the measurements are Ki = 5 ± 2 nM for (+)-34, Ki = 26 ± 16 nM for

(±)-224, Ki = 146 ± 63 nM for (±)-196, and Ki = 1.4 ± 0.7 M for (±)-114.

8.3.2. Thrombin[109] Biological activity against human thrombin was measured analogously to that of

factor Xa at an enzyme concentration of 1 nM, using chromogenic substrate S-2366

228 8. Experimental ______

(Km = 108 M, final concentration 200 M). The dilution series of the inhibitors were prepared as those used for the measurement of factor Xa activity.

8.3.3. Factor IXa Factor IXa activity was measured spectrophotometrically, following procedures from F. Hoffmann-La Roche, in microtiter plates in a final volume of 125 L in HNTP buffer containing 20% ethylene glycol. Inhibition of human factor IXa was tested at an enzyme concentration of 10 nM, using chromogenic substrate #299 (LOXO,

American diagnostica) at a final concentration of 0.5 mM. The dilution series were prepared analogous to those used for the assays of factor Xa and thrombin using HNTP buffer containing 22% ethylene glycol. The inhibitor was incubated with the enzyme for 4 min at RT, the substrate was added, and the activity measured spectrophotometrically as described for factor Xa.

229 8. Experimental ______8.4. Small-Molecule Crystal Structures

Copies of the data can be obtained free of charge on application to Cambridge Crystallographic Data Centre (CCDC), 12 Union Road, Cambridge CB2 1EZ, UK (fax: (+44)1223-336-033; e-mail: [email protected]).

X-Ray Analysis of Compound (±)-60 and (±)-110 Data has been measured on a Bruker Kappa-CCD diffractometer with MoKa radiation, l = 0.71073 Å, graphite monochromator at 223 K. The structures were solved by direct methods (SIR97).[216] Non-H-atoms were refined anisotropically (over two positions disordered C-atoms of methanol solvent in (±)-110 isotropic) with SHELXL-97.[217] Calculated H-atom positions were included in the structure factor calculation.

Compound (±)-60 monoclinic P21/c a = 21.767(1), b = 7.619(1), c = 12.362(2) Å, b = 102.39(2)°, V = 2002.4(5) Å3 from 21483 reflections; Z = 4. Final R(F) = 0.037, wR(F2) = 0.092 for 254 parameters, 0 restraints, and 3546 reflections with I > 2s(I) and 3.16 < q < 27.46°. Deposition No. CCDC-702887.

Compound (±)-110 monoclinic P21/n a = 6.5917(2), b = 11.9351(4), c = 35.420(1) Å, b = 91.804(2)°, V = 2785.2(2) Å3 from 8932 reflections; Z = 4. Final R(F) = 0.094, wR(F2) = 0.1437 for 303 parameters, 2 restraints, and 2933 reflections with I > 2s(I) and 2.91 < q < 27.49°. Deposition No. CCDC-702886.

X-Ray Analysis of Compound (±)-228

Compound (±)-228, C25H26BrClN4O4S, Mr = 593.933. A crystal of size 0.18x0.15x0.08 mm was measured at 223 K on a Bruker-Nonius Kappa-CCD with –3 MoKa radiation,  = 0.71073 Å. Monoclinic space group C2/c, rcalcd = 1.542g cm , Z = 8, a = 11.2205(2) Å, b = 10.5089(2) Å, c = 43.6149(10) Å,  = 95.9384(9)°, V = 5115.3(2) Å3, μ = 1.833 mm–1. Intensities of 8612 reflections were measured and reduced to 5107 independent ones (Rint = 0.08). Absorption corrections were applied with multi-scan method (Blessing).[218] The structures was solved by direct methods (SHELXS-97)[219] and refined by full-matrix least-squares analysis (SHELXL-97),[219] non-hydrogen atoms with anisotropic displacement parameters and H-atoms isotropically. Final R(F) = 0.061, wR(F2) = 0.159 for 429 parameters and 3537

230 8. Experimental ______reflections with I > 2s(I), corresponding R-values for all 5107 reflections are 0.097 and 0.183. Deposition No. CCDC-808209.

231 8. Experimental ______8.5. X-Ray Cocrystal Structures

Protein Expression and Purification[82] The expression and purification of factor Xa were performed at F. Hoffmann- La Roche. Recombinant factor X was expressed using the system described by Hopfner et al.[115] The construct encodes for a short-form factor X without the N-terminal Gla and epidermal growth factor 1 (EGF1) domains. The protein contains an Ile-Asp-Gly-Arg recognition sequence at the end of the activation peptide allowing activation to factor Xa by autocatalysis. To prevent autolysis, an Arg150Glu mutation was introduced (chymotrypsinogen numbering). At 310 K, the protein expressed in E. coli is in the insoluble fraction. Isolated, purified, and glutathione-

modified inclusion bodies were dissolved in 6 M guanidinium hydrochloride (pH 4.5) and 20 mM EDTA, and rapidly diluted (1:100) into 50 mM Tris/HCl (pH 8.0), 500 mM

arginine, 1 mM EDTA, 20 mM CaCl2, and 0.5 mM cysteine, and then incubated at 277 K for 2 d. The refolding solution was concentrated using a hollow fiber cartrige

(A/G Technology, Needham, MA, USA) and dialyzed against 20 mM Tris/HCl

(pH 8.0) and 50 mM NaCl at 277 K. Precipitated protein was removed by centrifugation. The supernatant was loaded on a Q-Sepharose column equilibrated

with 20 mM Tris/HCl (pH 8.0) and 50mM NaCl, and eluted with a linear NaCl gradient from 50 mM to 500 mM. Fractions containing renatured factor X were further purified by heparin affinity chromatography using the same buffer system. To prevent premature self-activation of the enzyme, both purification steps were performed in the presence of 1 mM benzamidine. The autocatalytic activation of factor X to Xa was enabled by an overnight dialysis at 277 K, using a benzamidine-

free buffer, 50 mM Tris/HCl (pH 8.0), 100 mM NaCl, and 5 mM CaCl2. The excised activation peptide was separated from the active enzyme by gel filtration on a Superdex-75 column in the same buffer.

The crystal structure of (±)-34 and factor Xa (PDB code: 2JKH) Short-form factor Xa was produced as previously described as the Arg150Glu mutant.[82,115] Compound (±)-34 was pre-incubated with the protein and crystals

grown by addition of 0.1 M Bis-Tris pH 6.5, 25% PEG3350 in microbatch mode. Crystals appeared after a week and grew to full size within two weeks. Before data collection, crystals were transferred to crystallization buffer supplemented with 20%

232 8. Experimental ______

Glycerol and flash-frozen in liquid N2. Diffraction data were measured at 100 K and wavelength 1.0 Å on beam line X10SA at the Swiss Light Source using a Mar225 CCD detector. Data from 360 frames of 0.5° were processed to 1.25 Å resolution [220] using XDS and SADABS (Bruker-AXS). The space group is P212121 with unit cell dimensions a = 48.91 Å, b = 74.12 Å, c = 77.18 Å. 75,840 unique reflections were measured with redundancy 6.98. The merging R factor on intensities was 4.4% (60.1% in the outermost shell, 1.34–1.25 Å), with completeness 97.0% (91.3%) and I/ 18.21 (2.43). Data reduction used the CCP4 package.[221] The structure was solved by molecular replacement using 2bok.pdb as model. Model building with Moloc[90,91] and Coot[222] and refinement with Refmac5[223] gave final overall crystallographic R factors of 19.5% (working) and 22.1% (free), with values in the outer shell (1.283–1.25 Å) of 31.2% and 32.5%, respectively, for 2349 atoms, including one Ca2+ ion, one Na+ ion, and 410 water molecules.

The crystal structure of (±)-196 and factor Xa (PDB code: 2Y5G) Data were collected on beam line X10SA (PXII) at the Swiss Light Source (SLS) at

wavelength 1.0 Å using a PILATUS pixel detector. The space group is P212121, with a = 48.89 Å, b = 74.64 Å, c = 77.1 Å. For 64894 unique reflections to 1.33 Å resolution the merging R-factor on intensities was 3.53%. The final R-values, after addition of hydrogen atoms at riding positions and the refinement of anisotropic temperature factors, were 14.7% (all data) and 18.5% (5% R-free).

The crystal structure of (±)-198 and factor Xa (PDB code: 2Y5H) Data were collected on beam line X10SA (PXII) at the Swiss Light Source (SLS) at

wavelength 1.0 Å using a PILATUS pixel detector. The space group is P212121, with a = 48.95 Å, b = 74.3 Å, c = 77.1 Å. For 71402 unique reflections to 1.29 Å resolution the merging R-factor on intensities was 5.85%. The final R-values, after addition of hydrogen atoms at riding positions and the refinement of anisotropic temperature factors, were 14.8% (all data) and 19.0% (5% R-free).

The crystal structure of (±)-229 and factor Xa (PDB code: 2Y5F) Data were collected on beam line X10SA (PXII) at the Swiss Light Source (SLS) at

wavelength 1.0 Å using a PILATUS pixel detector. The space group is P212121, with a = 48.94 Å, b = 74.3 Å, c = 77.2 Å. For 71523 unique reflections to 1.29 Å

233 8. Experimental ______resolution the merging R factor on intensities was 5.1%. The final R-values, after addition of hydrogen atoms at riding positions and the refinement of anisotropic temperature factors, were 14.0% (all data) and 17.5% (5% R-free).

234 8. Experimental ______8.6. 1H 1D-NOE Difference Spectra of (±)-78, (±)-79, and (±)-80

The 1H 1D-NOE difference spectra of (±)-78, (±)-79, and (±)-80 were measured in

CDCl3 at 500 MHz (Figure 82). Protons H-C(4), H-C(3a), H-C(8b), and H-C(8a) were irradiated. O O O H H 8 H H 8 H H 8 7 7 8b 7 N 8b 8a N 8b 8a N 8a 3a 3a 3a 4 N 6 4 N 6 4 N 6 O H O H O H H H H

S S S

Cl Cl Cl (±)-78 (±)-79 (±)-80

Irradiated NOE: Irradiated NOE: Irradiated NOE: signal: signal: signal: H-C(4)  H-C(3a), HH-C(6), H-C(4)  H-C(3a), HH-C(6) H-C(4)  H-C(3a), HH-C(6) HH-C(7), HH-C(8) H-C(3a)  H-C(4), H-C(8b) H-C(3a)  H-C(4), H-C(8b) H-C(3a) and H-C(8a)  H-C(8b), HH-C(8) H-C(8a)  H-C(8b), HH-C(8) H-C(8a)  H-C(4), H-C(8b), HH-C(6) HH-C(8) H-C(8b)  H-C(3a), H-C(8a), H-C(8b)  H-C(3a), H-C(8a) H-C(8b)  H-C(3a), H-C(8a), HH-C(8) HH-C(8)

Figure 82. The irradiated signals and the NOE peaks of the 1H 1D-NOE difference

spectra of compounds (±)-78, (±)-79, and (±)-80 (CDCl3, 500 MHz).

235 9. References ______9. References

[1] E. T. Kool, M. L. Waters, Nat. Chem. Biol. 2007, 3, 70–73. The Model Student: What Chemical Model Systems Can Teach Us about Biology. [2] F. Hof, F. Diederich, Chem. Commun. 2004, 477–480. Medicinal Chemistry in Academia: Molecular Recognition with Biological Receptors. [3] K. Appelt, R. J. Bacquet, C. A. Bartlett, C. L. J. Booth, S. T. Freer, M. A. M. Fuhry, M. R. Gehring, S. M. Herrmann, E. F. Howland, C. A. Janson, T. R. Jones, C.-C. Kan, V. Kathardekar, K. K. Lewis, G. P. Marzoni, D. A. Matthews, C. Mohr, E. W. Moomaw, C. A. Morse, S. J. Oatley, R. C. Ogden, M. R. Reddy, S. H. Reich, W. S. Schoettlin, W. W. Smith, M. D. Varney, J. E. Villafranca, R. W. Ward, S. Webber, S. E. Webber, K. M. Welsh, J. White, J. Med. Chem. 1991, 34, 1925–1934. Design of Enzyme Inhibitors Using Iterative Protein Crystallographic Analysis. [4] R. E. Babine, S. L. Bender, Chem. Rev. 1997, 97, 1359–1472. Molecular Recognition of Protein–Ligand Complexes: Applications to Drug Design. [5] C. Bissantz, B. Kuhn, M. Stahl, J. Med. Chem. 2010, 53, 5061–5084. A Medicinal Chemist's Guide to Molecular Interactions. [6] J. C. Ma, D. A. Dougherty, Chem. Rev. 1997, 97, 1303–1324. The Cation– Interaction. [7] T. J. Shepodd, M. A. Petti, D. A. Dougherty, J. Am. Chem. Soc. 1986, 108, 6085–6087. Tight, Oriented Binding of an Aliphatic Guest by a New Class of Water-Soluble Molecules with Hydrophobic Binding Sites. [8] M. A. Petti, T. J. Shepodd, R. E. Barrans Jr., D. A. Dougherty, J. Am. Chem. Soc. 1988, 110, 6825–6840. "Hydrophobic" Binding of Water-Soluble Guests by High-Symmetry, Chiral Hosts. An Electron-Rich Receptor Site with a General Affinity for Quaternary Ammonium Compounds and Electron- Deficient  Systems. [9] T. J. Shepodd, M. A. Petti, D. A. Dougherty, J. Am. Chem. Soc. 1988, 110, 1983–1985. Molecular Recognition in Aqueous Media: Donor–Acceptor and Ion–Dipole Interactions Produce Tight Binding for Highly Soluble Guests.

236 9. References ______[10] E. A. Meyer, R. K. Castellano, F. Diederich, Angew. Chem. 2003, 115, 1244– 1287; Angew. Chem. Int. Ed. 2003, 42, 1210–1250. Interactions with Aromatic Rings in Chemical and Biological Recognition. [11] N. Zacharias, D. A. Dougherty, Trends Pharmacol. Sci. 2002, 23, 281–287. Cation– Interactions in Ligand Recognition and Catalysis. [12] S. Khademi, J. O'Connell III, J. Remis, Y. Robles-Colmenares, L. J. W. Miercke, R. M. Stroud, Science 2004, 305, 1587–1594. Mechanism of Ammonia Transport by Amt/MEP/Rh: Structure of AmtB at 1.35 Å. [13] R. Worch, M. Jankowska-Anyszka, A. Niedzwiecka, J. Stepinski, C. Mazza, E. Darzynkiewicz, S. Cusack, R. Stolarski, J. Mol. Biol. 2009, 385, 618–627. Diverse Role of Three Tyrosines in Binding of the RNA 5' Cap to the Human Nuclear Cap Binding Complex. [14] H. Li, S. Ilin, W. Wang, E. M. Duncan, J. Wysocka, C. D. Allis, D. J. Patel, Nature 2006, 442, 91–95. Molecular Basis for Site-Specific Read-Out of Histone H3K4me3 by the BPTF PHD Finger of NURF. [15] J. Sunner, K. Nishizawa, P. Kebarle, J. Phys. Chem. 1981, 85, 1814–1820. Ion–Solvent Molecule Interactions in the Gas Phase. The Potassium Ion and Benzene. [16] D. A. Dougherty, J. Nutr. 2007, 137, 1504S–1508S. Cation- Interactions Involving Aromatic Amino Acids. [17] S. Tsuzuki, Struct. Bond. 2005, 115, 149–193. Interactions with Aromatic Rings. [18] S. Mecozzi, A. P. West Jr., D. A. Dougherty, J. Am. Chem. Soc. 1996, 118, 2307–2308. Cation– Interactions in Simple Aromatics: Electrostatics Provide a Predictive Tool. [19] C. Felder, H.-L. Jiang, W.-L. Zhu, K.-X. Chen, I. Silman, S. A. Botti, J. L. Sussman, J. Phys. Chem. A 2001, 105, 1326–1333. Quantum/Classical Mechanical Comparison of Cation– Interactions between Tetramethylammonium and Benzene. [20] J. P. Gallivan, D. A. Dougherty, Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 9459–9464. Cation- Interactions in Structural Biology. [21] D. A. Dougherty, Science 1996, 271, 163–168. Cation- Interactions in Chemistry and Biology: A New View of Benzene, Phe, Tyr, and Trp.

237 9. References ______[22] S. Mecozzi, A. P. West Jr., D. A. Dougherty, Proc. Natl. Acad. Sci. U. S. A. 1996, 93, 10566–10571. Cation- Interactions in Aromatics of Biological and Medicinal Interest: Electrostatic Potential Surfaces as a Useful Qualitative Guide. [23] M. M. Gromiha, C. Santhosh, S. Ahmad, Int. J. Biol. Macromol. 2004, 34, 203–211. Structural Analysis of Cation– Interactions in DNA Binding Proteins. [24] M. M. Gromiha, C. Santhosh, M. Suwa, Polymer 2004, 45, 633–639. Influence of Cation– Interactions in Protein–DNA Complexes. [25] L. Mao, Y. Wang, Y. Liu, X. Hu, J. Mol. Biol. 2004, 336, 787–807. Molecular Determinants for ATP-binding in Proteins: A Data Mining and Quantum Chemical Analysis. [26] P. B. Crowley, A. Golovin, Proteins: Struct. Funct. Bioinf. 2005, 59, 231–239. Cation– Interactions in Protein–Protein Interfaces. [27] J. G. Cheng, X. M. Luo, X. H. Yan, Z. Li, Y. Tang, H. L. Jiang, W. L. Zhu, Sci. China, Ser. B: Chem. 2008, 51, 709–717. Research Progress in Cation- Interactions. [28] R. M. Hughes, K. R. Wiggins, S. Khorasanizadeh, M. L. Waters, Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 11184–11188. Recognition of Trimethyllysine by a Chromodomain Is Not Driven by the Hydrophobic Effect. [29] S. C. R. Lummis, D. L. Beene, N. J. Harrison, H. A. Lester, D. A. Dougherty, Chem. Biol. 2005, 12, 993–997. A Cation- Binding Interaction with a

Tyrosine in the Binding Site of the GABAC Receptor. [30] C. L. Padgett, A. P. Hanek, H. A. Lester, D. A. Dougherty, S. C. R. Lummis, J. Neurosci. 2007, 27, 886–892. Unnatural Amino Acid Mutagenesis of the

GABAA Receptor Binding Site Residues Reveals a Novel Cation–

Interaction between GABA and 2Tyr97. [31] M. M. Torrice, K. S. Bower, H. A. Lester, D. A. Dougherty, Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 11919–11924. Probing the Role of the Cation– Interaction in the Binding Sites of GPCRs Using Unnatural Amino Acids. [32] X. Xiu, N. L. Puskar, J. A. P. Shanata, H. A. Lester, D. A. Dougherty, Nature 2009, 458, 534–538. Nicotine Binding to Brain Receptors Requires a Strong Cation- Interaction.

238 9. References ______[33] F.-G. Klärner, B. Kahlert, Acc. Chem. Res. 2003, 36, 919–932. Molecular Tweezers and Clips as Synthetic Receptors. Molecular Recognition and Dynamics in Receptor–Substrate Complexes. [34] M. Fokkens, T. Schrader, F.-G. Klärner, J. Am. Chem. Soc. 2005, 127, 14415– 14421. A Molecular Tweezer for Lysine and Arginine. [35] M. Fokkens, C. Jasper, T. Schrader, F. Koziol, C. Ochsenfeld, J. Polkowska, M. Lobert, B. Kahlert, F.-G. Klärner, Chem. Eur. J. 2005, 11, 477–494. Selective Complexation of N-Alkylpyridinium Salts: Binding of NAD+ in Water. [36] J. Polkowska, F. Bastkowski, T. Schrader, F.-G. Klärner, J. Zienau, F. Koziol, C. Ochsenfeld, J. Phys. Org. Chem. 2009, 22, 779–790. A Combined Experimental and Theoretical Study of the pH-Dependent Binding Mode of NAD+ by Water-Soluble Molecular Clips. [37] T. Schrader, M. Fokkens, F.-G. Klärner, J. Polkowska, F. Bastkowski, J. Org. Chem. 2005, 70, 10227–10237. Inclusion of Thiamine Diphosphate and S- Adenosylmethionine at Their Chemically Active Sites. [38] P. Talbiersky, F. Bastkowski, F.-G. Klärner, T. Schrader, J. Am. Chem. Soc. 2008, 130, 9824–9828. Molecular Clip and Tweezer Introduce New Mechanisms of Enzyme Inhibition. [39] S. Tsuzuki, M. Mikami, S. Yamada, J. Am. Chem. Soc. 2007, 129, 8656–8662. Origin of Attraction, Magnitude, and Directionality of Interactions in Benzene Complexes with Pyridinium Cations. [40] J. M. Heemstra, J. S. Moore, Chem. Commun. 2004, 1480–1481. Helix Atabilization through Pyridinium– Interactions. [41] J. M. Heemstra, J. S. Moore, Org. Lett. 2004, 6, 659–662. Pyridine- Containing m-Phenylene Ethynylene Oligomers Having Tunable Basicities. [42] D. A. Stauffer, R. E. Barrans Jr., D. A. Dougherty, Angew. Chem. 1990, 102, 953–956; Angew. Chem. Int. Ed. 1990, 29, 915–918. Biomimetic Catalysis of

an SN2 Reaction Resulting from a Novel Form of Transition-State Stabilization. [43] A. McCurdy, L. Jimenez, D. A. Stauffer, D. A. Dougherty, J. Am. Chem. Soc.

1992, 114, 10314–10321. Biomimetic Catalysis of SN2 Reactions through Cation– Interactions. The Role of Polarizability in Catalysis.

239 9. References ______[44] S. Yamada, Org. Biomol. Chem. 2007, 5, 2903–2912. Intramolecular Cation–  Interaction in Organic Synthesis. [45] M. Szostak, L. Yao, J. Aubé, J. Org. Chem. 2010, 75, 1235–1243. Synthesis of Medium-Bridged Twisted Lactams via Cation– Control of the Regiochemistry of the Intramolecular Schmidt Reaction. [46] S. Yamada, N. Uematsu, K. Yamashita, J. Am. Chem. Soc. 2007, 129, 12100– 12101. Role of Cation– Interactions in the Photodimerization of trans-4- Styrylpyridines. [47] R. Kaliappan, M. V. S. N. Maddipatla, L. S. Kaanumalle, V. Ramamurthy, Photochem. Photobiol. Sci. 2007, 6, 737–740. Crystal Engineering Principles Applied to Solution Photochemistry: Controlling the Photodimerization of Stilbazolium Salts within -Cyclodextrin and Cucurbit[8]uril in Water. [48] S. Yamada, M. Inoue, Org. Lett. 2007, 9, 1477–1480. Regio- and Stereoselective Addition of Allylmetal Reagents to Pyridinium– and Quinolinium– Complexes. [49] S. Yamada, J. Yamamoto, E. Ohta, Tetrahedron Lett. 2007, 48, 855–858. Enantioselective Cyclopropanation Reaction Using a Conformationally Fixed Pyridinium Ylide through a Cation– Interaction. [50] P. Bolduc, A. Jacques, S. K. Collins, J. Am. Chem. Soc. 2010, 132, 12790– 12791. Efficient Macrocyclization Achieved via Conformational Control Using Intermolecular Noncovalent -Cation/Arene Interactions [51] http://www.who.int/mediacentre/factsheets/fs317/en/index.html (10.04.2011). [52] R. K. Murray, D. K. Granner, P. A. Mayes, V. W. Rodwell, Harper's Illustrated Biochemistry, (Eds.: J. Foltin, J. Ransom and J. M. Oransky), McGraw-Hill, New York, 2003. [53] K. Padmanabhan, K. P. Padmanabhan, A. Tulinsky, C. H. Park, W. Bode, R. Huber, D. T. Blankenship, A. D. Cardin, W. Kisiel, J. Mol. Biol. 1993, 232, 947–966. Structure of Human Des(1-45) Factor Xa at 2.2 Å Resolution. [54] E. W. Davie, K. Fujikawa, W. Kisiel, Biochemistry 1991, 30, 10363–10370. The Coagulation Cascade: Initiation, Maintenance, and Regulation. [55] L. Hedstrom, Chem. Rev. 2002, 102, 4501–4523. Serine Protease Mechanism and Specificity. [56] T. D. H. Bugg, Introduction to Enzyme and Coenzyme Chemistry, Blackwell, Oxford, 2004, pp. 84–89.

240 9. References ______[57] I. Schechter, A. Berger, Biochem. Biophys. Res. Commun. 1967, 27, 157–162. On the Size of the Active Site in Proteases. I. Papain. [58] Y.-K. Lee, M. R. Player, Med. Res. Rev. 2011, 31, 202–283. Developments in Factor Xa Inhibitors for the Treatment of Thromboembolic Disorders. [59] S. Maignan, V. Mikol, Curr. Top. Med. Chem. 2001, 1, 161–174. The Use of 3D Structural Data in the Design of Specific Factor Xa Inhibitors. [60] D. Gustafsson, R. Bylund, T. Antonsson, I. Nilsson, J.-E. Nyström, U. Eriksson, U. Bredberg, A.-C. Teger-Nilsson, Nat. Rev. Drug Discovery 2004, 3, 649–659. A New Oral Anticoagulant: The 50-Year Challenge. [61] D. J. P. Pinto, J. M. Smallheer, D. L. Cheney, R. M. Knabb, R. R. Wexler, J. Med. Chem. 2010, 53, 6243–6274. Factor Xa Inhibitors: Next-Generation Agents. [62] M. de Kort, R. C. Buijsman, C. A. A. van Boeckel, Drug Discov. Today 2005, 10, 769–779. Synthetic Heparin Derivatives as New Anticoagulant Drugs. [63] J. Hirsh, J. E. Dalen, D. R. Anderson, L. Poller, H. Bussey, J. Ansell, D. Deykin, Chest 2001, 119, 8S–21S. Oral Anticoagulants: Mechanism of Action, Clinical Effectiveness, and Optimal Therapeutic Range. [64] E. Perzborn, S. Roehrig, A. Straub, D. Kubitza, F. Misselwitz, Nat. Rev. Drug Discovery 2011, 10, 61–75. The Discovery and Development of Rivaroxaban, an Oral, Direct Factor Xa Inhibitor. [65] S. Haas, J. Thromb. 2008, 25, 52–60. New Oral Xa and IIa Inhibitors: Updates on Clinical Trial Results. [66] J. P. Vacca, Curr. Opin. Chem. Biol. 2000, 4, 394–400. New Advances in the Discovery of Thrombin and Factor Xa Inhibitors. [67] L.-A. Linkins, J. I. Weitz, Annu. Rev. Med. 2005, 56, 63–77. New Anticoagulant Therapy. [68] R. M. Scarborough, A. Pandey, X. Zhang, Annu. Rep. Med. Chem. 2005, 40, 85–101. Small Molecule Anticoagulant/Antithrombotic Agents. [69] R. J. Leadley Jr., Curr. Top. Med. Chem. 2001, 1, 151–159. Coagulation Factor Xa Inhibition: Biological Background and Rationale. [70] H. W. Pauls, W. R. Ewing, Curr. Top. Med. Chem. 2001, 1, 83–100. The Design of Competitive, Small-molecule Inhibitors of Coagulation Factor Xa. [71] M. L. Quan, R. R. Wexler, Curr. Top. Med. Chem. 2001, 1, 137–149. The Design and Synthesis of Noncovalent Factor Xa Inhibitors.

241 9. References ______[72] B. Kaiser, Cell. Mol. Life Sci. 2002, 59, 189–192. Factor Xa – A Promising Target for Drug Development. [73] M. Ieko, T. Tarumi, T. Nakabayashi, M. Yoshida, S. Naito, T. Koike, Front. Biosci. 2006, 11, 232–248. Factor Xa Inhibitors: New Anti-thrombotic Agents and Their Characteristics. [74] A. Kasani, R. Subedi, M. Stier, D. D. Holsworth, S. N. Maiti, Heterocycles 2007, 73, 47–85. Cardiovascular Agents: Renin Inhibitors and Factor Xa Inhibitors. [75] A. G. G. Turpie, Arterioscler. Thromb. Vasc. Biol. 2007, 27, 1238–1247. Oral, Direct Factor Xa Inhibitors in Development for the Prevention and Treatment of Thromboembolic Diseases. [76] A. Straub, S. Roehrig, A. Hillisch, Angew. Chem. 2011, 123, 4670–4686; Angew. Chem. Int. Ed. 2011, 50, 4574–4590. Oral, Direct Thrombin and Factor Xa Inhibitors: The Replacement for Warfarin, Leeches, and Pig Intestines? [77] S. Roehrig, A. Straub, J. Pohlmann, T. Lampe, J. Pernerstorfer, K.-H. Schlemmer, P. Reinemer, E. Perzborn, J. Med. Chem. 2005, 48, 5900–5908. Discovery of the Novel Antithrombotic Agent 5-Chloro-N-({(5S)-2-oxo-3[4- (3-oxomorpholin-4-yl)phenyl]-1,3-oxazolidin-5-yl}methyl)thiophene-2- carboxamide (BAY 59-7939): An Oral, Direct Factor Xa Inhibitor. [78] D. J. P. Pinto, M. J. Orwat, S. Koch, K. A. Rossi, R. S. Alexander, A. Smallwood, P. C. Wong, A. R. Rendina, J. M. Luettgen, R. M. Knabb, K. He, B. Xin, R. R. Wexler, P. Y. S. Lam, J. Med. Chem. 2007, 50, 5339–5356. Discovery of 1-(4-Methoxyphenyl)-7-oxo-6-(4-(2-oxopiperidin-1-yl)phenyl)- 4,5,6,7-tetrahydro-1H-pyrazolo[3,4-c]pyridine-3-carboxamide (Apixaban, BMS-562247), a Highly Potent, Selective, Efficacious, and Orally Bioavailable Inhibitor of Blood Coagulation Factor Xa. [79] T. Furugohri, K. Isobe, Y. Honda, C. Kamisato-Matsumoto, N. Sugiyama, T. Nagahara, Y. Morishima, T. Shibano, J. Thromb. Haemost. 2008, 6, 1542– 1549. DU-176b, a Potent and Orally Active Factor Xa Inhibitor: in vitro and in vivo Pharmacological Profiles. [80] S. Oba, T. Yasuji, T. Hakomori, K. Sako, US 2010/0144711, 2010. Pharmaceutical Composition for Oral Administration.

242 9. References ______[81] V. Chu, K. Brown, D. Colussi, J. Gao, J. Bostwick, C. Kasiewski, R. Bentley, S. Morgan, K. Guertin, H. W. Pauls, Y. Gong, A. Zulli, M. H. Perrone, C. T. Dunwiddie, R. J. Leadley, Thromb. Res. 2001, 103, 309–324. Pharmacological Characterization of a Novel Factor Xa Inhibitor, FXV673. [82] K. Schärer, M. Morgenthaler, R. Paulini, U. Obst-Sander, D. W. Banner, D. Schlatter, J. Benz, M. Stihle, F. Diederich, Angew. Chem. 2005, 117, 4474– 4479; Angew. Chem. Int. Ed. 2005, 44, 4400–4404. Quantification of Cation–  Interactions in Protein–Ligand Complexes: Crystal-Structure Analysis of Factor Xa Bound to a Quaternary Ammonium Ion Ligand. [83] U. Obst, Dissertation, ETH, No. 12037, 1997. De novo-Design und Synthese neuartiger, nichtpeptidischer Thrombin-Inhibitoren. [84] U. Obst, D. W. Banner, L. Weber, F. Diederich, Chem. Biol. 1997, 4, 287– 295. Molecular Recognition at the Thrombin Active Site: Structure-based Design and Synthesis of Potent and Selective Thrombin Inhibitors and the X- Ray Crystal Structures of Two Thrombin-Inhibitor Complexes. [85] U. Obst, P. Betschmann, C. Lerner, P. Seiler, F. Diederich, V. Gramlich, L. Weber, D. W. Banner, P. Schönholzer, Helv. Chim. Acta 2000, 83, 855–909. Synthesis of Novel Nonpeptidic Thrombin Inhibitors. [86] H.-J. Böhm, G. Klebe, Angew. Chem. 1996, 108, 2750–2778; Angew. Chem. Int. Ed. Engl. 1996, 35, 2589–2614. What Can We Learn from Molecular Recognition in Protein–Ligand Complexes for the Design of New Drugs? [87] R. Huisgen, J. Org. Chem. 1976, 41, 403–419. The Concerted Nature of 1,3- Dipolar Cycloadditions and the Question of Diradical Intermediates. [88] K. V. Gothelf, K. A. Jørgensen, Chem. Rev. 1998, 98, 863–909. Asymmetric 1,3-Dipolar Cycloaddition Reactions. [89] K. Schärer, Dissertation, ETH, No. 15820, 2005. Strukturbasiertes Design und Synthese von nichtpeptidischen Inhibitoren der Serinprotease Thrombin. [90] P. R. Gerber, K. Müller, J. Comput.-Aided Mol. Des. 1995, 9, 251–268. MAB, a Generally Applicable Molecular Force Field for Structure Modelling in Medicinal Chemistry. [91] Gerber Molecular Design, http://www.moloc.ch. [92] M. Adler, M. J. Kochanny, B. Ye, G. Rumennik, D. R. Light, S. Biancalana, M. Whitlow, Biochemistry 2002, 41, 15514–15523. Crystal Structures of Two Potent Nonamidine Inhibitors Bound to Factor Xa.

243 9. References ______[93] Y. M. Choi-Sledeski, R. Kearney, G. Poli, H. Pauls, C. Gardner, Y. Gong, M. Becker, R. Davis, A. Spada, G. Liang, V. Chu, K. Brown, D. Collussi, R. Leadley Jr., S. Rebello, P. Moxey, S. Morgan, R. Bentley, C. Kasiewski, S. Maignan, J.-P. Guilloteau, V. Mikol, J. Med. Chem. 2003, 46, 681–684. Discovery of an Orally Efficacious Inhibitor of Coagulation Factor Xa which Incorporates a Neutral P1 Ligand. [94] S. Maignan, J.-P. Guilloteau, Y. M. Choi-Sledski, M. R. Becker, W. R. Ewing, H. W. Pauls, A. P. Spada, V. Mikol, J. Med. Chem. 2003, 46, 685–690. Molecular Structures of Human Factor Xa Complexed with Ketopiperazine Inhibitors: Preference for a Neutral Group in the S1 Pocket. [95] W. W. K. R. Mederski, D. Dorsch, S. Anzali, J. Gleitz, B. Cezanne, C. Tsaklakidis, Bioorg. Med. Chem. Lett. 2004, 14, 3763–3769. Halothiophene Benzimidazoles as P1 Surrogates of Inhibitors of Blood Coagulation Factor Xa. [96] M. Nazaré, M. Essrich, D. W. Will, H. Matter, K. Ritter, M. Urmann, A. Bauer, H. Schreuder, A. Dudda, J. Czech, M. Lorenz, V. Laux, V. Wehner, Bioorg. Med. Chem. Lett. 2004, 14, 4191–4195. Factor Xa Inhibitors Based on a 2-Carboxyindole Scaffold: SAR of Neutral P1 Substituents. [97] J. B. Franciskovich, J. J. Masters, J. M. Tinsley, T. J. Craft, L. L. Froelich, D. S. Gifford-Moore, V. J. Klimkowski, J. K. Smallwood, G. F. Smith, T. Smith, R. R. Towner, L. C. Weir, M. R. Wiley, Bioorg. Med. Chem. Lett. 2005, 15, 4838–4841. Investigation of Factor Xa Inhibitors Containing Non-Amidine S1 Elements. [98] S. Komoriya, S. Kobayashi, K. Osanai, T. Yoshino, T. Nagata, N. Haginoya, Y. Nakamoto, A. Mochizuki, T. Nagahara, M. Suzuki, T. Shimada, K. Watanabe, Y. Isobe, T. Furugoori, Bioorg. Med. Chem. 2006, 14, 1309–1330. Design, Synthesis, and Biological Activity of Novel Factor Xa Inhibitors: Improving Metabolic Stability by S1 and S4 Ligand Modification. [99] P. Betschmann, Dissertation, ETH, No. 13890, 2000. Strukturbasiertes Design und Synthese nichtpeptidischer Thrombin-Inhibitoren. [100] E. Schweizer, Dissertation, ETH, No. 16531, 2006. Studien zur Hemmung der Dimerisierung des Enzyms HIV-Reverse-Transkriptase (HIV-RT) und zu der Molekularen Erkennung im aktiven Zentrum von Thrombin.

244 9. References ______[101] M. Nazaré, D. W. Will, H. Matter, H. Schreuder, K. Ritter, M. Urmann, M. Essrich, A. Bauer, M. Wagner, J. Czech, M. Lorenz, V. Laux, V. Wehner, J. Med. Chem. 2005, 48, 4511–4525. Probing the Subpockets of Factor Xa Reveals Two Binding Modes for Inhibitors Based on a 2-Carboxyindole Scaffold: A Study Combining Structure-Activity Relationship and X-Ray Crystallography. [102] S. Muramatsu, K. Kinbara, H. Taguchi, N. Ishii, T. Aida, J. Am. Chem. Soc. 2006, 128, 3764–3769. Semibiological Molecular Machine with an Implemented "AND" Logic Gate for Regulation of Protein Folding. [103] A. K. Roy, S. Batra, Synthesis 2003, 2325–2330. Facile Baylis–Hillman Reaction of Substituted 3-Isoxazolecarbaldehydes: The Impact of a Proximal Heteroatom within a Heterocycle on the Acceleration of the Reaction. [104] A. Foroumadi, M. Oboudiat, S. Emami, A. Karimollah, L. Saghaee, M. H. Moshafi, A. Shafiee, Bioorg. Med. Chem. 2006, 14, 3421–3427. Synthesis and Antibacterial Activity of N-[2-[5-(Methylthio)thiophen-2-yl]-2-oxoethyl] and N-[2-[5-(Methylthio)thiophen-2-yl]-2-(oxyimino)ethyl]piperazinyl- quinolone Derivatives. [105] N. Ikemoto, J. Liu, K. M. J. Brands, J. M. McNamara, P. J. Reider, Tetrahedron 2003, 59, 1317–1325. Practical Routes to the Triarylsulfonyl

Chloride Intermediate of a 3 Adrenergic Receptor Agonist. [106] F. C. Krebs, H. Spanggaard, J. Org. Chem. 2002, 67, 7185–7192. An Exceptional Red Shift of Emission Maxima upon Fluorine Substitution. [107] X. Bu, L. W. Deady, G. J. Finlay, B. C. Baguley, W. A. Denny, J. Med. Chem. 2001, 44, 2004–2014. Synthesis and Cytotoxic Activity of 7-Oxo-7H- dibenz[f,ij]isoquinoline and 7-Oxo-7H-benzo[e]perimidine Derivatives. [108] L. Anselm, K. Groebke Zbinden, W. Haap, J. Himber, C. M. Stahl, S. Thomi, US 2005/0215599, 2005. Pyrrolidine-3,4-dicarboxamide Derivatives. [109] K. Hilpert, J. Ackermann, D. W. Banner, A. Gast, K. Gubernator, P. Hadváry, L. Labler, K. Müller, G. Schmid, T. B. Tschopp, H. van de Waterbeemd, J. Med. Chem. 1994, 37, 3889–3901. Design and Synthesis of Potent and Highly Selective Thrombin Inhibitors. [110] L. Aurell, P. Friberger, G. Karlsson, G. Claeson, Thromb. Res. 1977, 11, 595– 609. A New Sensitive and Highly Specific Chromogenic Peptide Substrate for Factor Xa.

245 9. References ______[111] Y. Cheng, W. H. Prusoff, Biochem. Pharmacol. 1973, 22, 3099–3108.

Relationship between Inhibition Constant (KI) and Concentration of Inhibitor

Which Causes 50 Per Cent Inhibition (I50) of an Enzymatic Reaction. [112] Relibase 2.2.2. (August 2008). [113] M. Hendlich, A. Bergner, J. Günther, G. Klebe, J. Mol. Biol. 2003, 326, 607– 620. Relibase: Design and Development of a Database for Comprehensive Analysis of Protein–Ligand Interactions. [114] J. Günther, A. Bergner, M. Hendlich, G. Klebe, J. Mol. Biol. 2003, 326, 621– 636. Utilising Structural Knowledge in Drug Design Strategies: Applications Using Relibase. [115] K.-P. Hopfner, H. Brandstetter, A. Karcher, E. Kopetzki, R. Huber, R. A. Engh, W. Bode, EMBO J. 1997, 16, 6626–6635. Converting Blood Coagulation Factor IXa into Factor Xa: Dramatic Increase in Amidolytic Activity Identifies Important Active Site Determinants. [116] J. T. Kohrt, C. F. Bigge, J. W. Bryant, A. Casimiro-Garcia, L. Chi, W. L. Cody, T. Dahring, D. A. Dudley, K. J. Filipski, S. Haarer, R. Heemstra, N. Janiczek, L. Narasimhan, J. McClanahan, J. T. Peterson, V. Sahasrabudhe, R. Schaum, C. A. Van Huis, K. M. Welch, E. Zhang, R. J. Leadley, J. J. Edmunds, Chem. Biol. Drug Des. 2007, 70, 100–112. The Discovery of (2R,4R)-N-(4-Chlorophenyl)-N-(2-fluoro-4-(2-oxopyridin-1(2H)-yl)phenyl)-4- methoxypyrrolidine-1,2-dicarboxamide (PD 0348292), an Orally Efficacious Factor Xa Inhibitor. [117] J. X. Qiao, C.-H. Chang, D. L. Cheney, P. E. Morin, G. Z. Wang, S. R. King, T. C. Wang, A. R. Rendina, J. M. Luettgen, R. M. Knabb, R. R. Wexler, P. Y. S. Lam, Bioorg. Med. Chem. Lett. 2007, 17, 4419–4427. SAR and X-Ray Structures of Enantiopure 1,2-cis-(1R,2S)-Cyclopentyldiamine and Cyclohexyldiamine Derivatives as Inhibitors of Coagulation Factor Xa. [118] Y. Shi, D. Sitkoff, J. Zhang, H. E. Klei, K. Kish, E. C.-K. Liu, K. S. Hartl, S. M. Seiler, M. Chang, C. Huang, S. Youssef, T. E. Steinbacher, W. A. Schumacher, N. Grazier, A. Pudzianowski, A. Apedo, L. Discenza, J. Yanchunas Jr., P. D. Stein, K. S. Atwal, J. Med. Chem. 2008, 51, 7541–7551. Design, Structure–Activity Relationships, X-ray Crystal Structure, and Energetic Contributions of a Critical P1 Pharmacophore: 3-Chloroindole-7-yl- Based Factor Xa Inhibitors.

246 9. References ______[119] K. Groebke Zbinden, L. Anselm, D. W. Banner, J. Benz, F. Blasco, G. Décoret, J. Himber, B. Kuhn, N. Panday, F. Ricklin, P. Risch, D. Schlatter, M. Stahl, S. Thomi, R. Unger, W. Haap, Eur. J. Med. Chem. 2009, 44, 2787– 2795. Design of Novel Aminopyrrolidine Factor Xa Inhibitors from a Screening Hit. [120] R. M. Hughes, M. L. Benshoff, M. L. Waters, Chem. Eur. J. 2007, 13, 5753– 5764. Effects of Chain Length and N-Methylation on a Cation– Interaction in a -Hairpin Peptide. [121] O. Keller, J. Rudinger, Helv. Chim. Acta 1975, 58, 531–541. Preparation and Some Properties of Maleimido Acids and Maleoyl Derivatives of Peptides. [122] K. A. Keller, J. Guo, S. Punna, M. G. Finn, Tetrahedron Lett. 2005, 46, 1181– 1184. A Thermally-Cleavable Linker for Solid-Phase Synthesis. [123] K. B. Wiberg, Angew. Chem. 1986, 98, 312–322; Angew. Chem. Int. Ed. Engl. 1986, 25, 312–322. The Concept of Strain in Organic Chemistry. [124] C. S. Beshara, C. E. Jones, K. D. Daze, B. J. Lilgert, F. Hof, ChemBioChem 2010, 11, 63–66. A Simple Calixarene Recognizes Post-translationally Methylated Lysine. [125] R. M. Hughes, M. L. Waters, J. Am. Chem. Soc. 2005, 127, 6518–6519. Influence of N-Methylation on a Cation– Interaction Produces a Remarkably Stable -Hairpin Peptide. [126] A. J. Riemen, M. L. Waters, Biochemistry 2009, 48, 1525–1531. Design of Highly Stabilized -Hairpin Peptides through Cation– Interactions of Lysine and N-Methyllysine with an Aromatic Pocket. [127] R. M. Hughes, M. L. Waters, J. Am. Chem. Soc. 2006, 128, 12735–12742. Arginine Methylation in a -Hairpin Peptide: Implications for Arg–

Interactions, Cp°, and the Cold Denatured State. [128] L. A. Ingerman, M. E. Cuellar, M. L. Waters, Chem. Commun. 2010, 46, 1839–1841. A Small Molecule Receptor That Selectively Recognizes Trimethyl Lysine in a Histone Peptide with Native Protein-Like Affinity. [129] C. Tommos, B. W. Berry, M. M. Elvekrog, J. Am. Chem. Soc. 2007, 129, 5308–5309. Environmental Modulation of Protein Cation– Interactions. [130] M. A. Walker, J. Org. Chem. 1995, 60, 5352–5355. A High Yielding Synthesis of N-Alkyl Maleimides Using a Novel Modification of the Mitsunobu Reaction.

247 9. References ______[131] A. Shivanyuk, J. C. Friese, S. Döring, J. Rebek Jr., J. Org. Chem. 2003, 68, 6489–6496. Solvent-Stabilized Molecular Capsules. [132] N. K. Beyeh, A. Valkonen, K. Rissanen, Supramol. Chem. 2009, 21, 142–148. Encapsulation of Tetramethylphosphonium Cations. [133] S. Bartoli, S. Roelens, J. Am. Chem. Soc. 1999, 121, 11908–11909. Electrostatic Attraction of Counterion Dominates the Cation– Interaction of Acetylcholine and Tetramethylammonium with Aromatics in Chloroform. [134] S. Bartoli, S. Roelens, J. Am. Chem. Soc. 2002, 124, 8307–8315. Binding of Acetylcholine and Tetramethylammonium to a Cyclophane Receptor: Anion's Contribution to the Cation– Interaction. [135] P. Sarri, F. Venturi, F. Cuda, S. Roelens, J. Org. Chem. 2004, 69, 3654–3661. Binding of Acetylcholine and Tetramethylammonium to Flexible Cyclophane Receptors: Improving on Binding Ability by Optimizing Host's Geometry. [136] C. A. Hunter, C. M. R. Low, C. Rotger, J. G. Vinter, C. Zonta, Chem. Commun. 2003, 834–835. The Role of the Counteranion in the Cation- Interaction. [137] S. L. Cockroft, C. A. Hunter, Chem. Soc. Rev. 2007, 36, 172–188. Chemical Double-Mutant Cycles: Dissecting Non-Covalent Interactions. [138] M. Albertí, A. Aguilar, F. Pirani, J. Phys. Chem. A 2009, 113, 14741–14748. Cation–-Anion Interaction in Alkali Ion–Benzene–Halogen Ion Clusters. [139] D. Kim, E. C. Lee, K. S. Kim, P. Tarakeshwar, J. Phys. Chem. A 2007, 111, 7980–7986. Cation––Anion Interaction: A Theoretical Investigation of the Role of Induction Energies. [140] A. L. Whiting, N. M. Neufeld, F. Hof, Tetrahedron Lett. 2009, 50, 7035– 7037. A Tryptophan-Analog Host Whose Interactions with Ammonium Ions in Water Are Dominated by the Hydrophobic Effect. [141] Z. Li, T. Lazaridis, Phys. Chem. Chem. Phys. 2007, 9, 573–581. Water at Biomolecular Binding Interfaces. [142] J. E. Ladbury, Chem. Biol. 1996, 3, 973–980. Just Add Water! The Effect of Water on the Specificity of Protein–Ligand Binding Sites and Its Potential Application to Drug Design [143] S. B. A. de Beer, N. P. E. Vermeulen, C. Oostenbrink, Curr. Top. Med. Chem. 2010, 10, 55–66. The Role of Water Molecules in Computational Drug Design.

248 9. References ______[144] S. E. Wong, F. C. Lightstone, Expert Opin. Drug Discov. 2011, 6, 65–74. Accounting for Water Molecules in Drug Design. [145] A. Amadasi, J. A. Surface, F. Spyrakis, P. Cozzini, A. Mozzarelli, G. E. Kellogg, J. Med. Chem. 2008, 51, 1063–1067. Robust Classification of "Relevant" Water Molecules in Putative Protein Binding sites. [146] Y. Lu, R. Wang, C.-Y. Yang, S. Wang, J. Chem. Inf. Model. 2007, 47, 668– 675. Analysis of Ligand-Bound Water Molecules in High-Resolution Crystal Structures of Protein–Ligand Complexes. [147] R. Abel, T. Young, R. Farid, B. J. Berne, R. A. Friesner, J. Am. Chem. Soc. 2008, 130, 2817–2831. Role of the Active-Site Solvent in the Thermodynamics of Factor Xa Ligand Binding. [148] C. Barillari, J. Taylor, R. Viner, J. W. Essex, J. Am. Chem. Soc. 2007, 129, 2577–2587. Classification of Water Molecules in Protein Binding Sites. [149] J. D. Dunitz, Science 1994, 264, 670. The Entropic Cost of Bound Water in Crystals and Biomolecules. [150] M. Ellermann, R. Jakob-Roetne, C. Lerner, E. Borroni, D. Schlatter, D. Roth, A. Ehler, M. G. Rudolph, F. Diederich, Angew. Chem. 2009, 121, 9256–9260; Angew. Chem. Int. Ed. 2009, 48, 9092–9096. Molecular Recognition at the Active Site of Catechol-O-Methyltransferase: Energetically Favorable Replacement of a Water Molecule Imported by a Bisubstrate Inhibitor. [151] P. C. Kohler, T. Ritschel, W. B. Schweizer, G. Klebe, F. Diederich, Chem. Eur. J. 2009, 15, 10809–10817. High-Affinity Inhibitors of tRNA-Guanine Transglycosylase Replacing the Function of a Structural Water Cluster. [152] T. Ritschel, P. C. Kohler, G. Neudert, A. Heine, F. Diederich, G. Klebe, ChemMedChem 2009, 4, 2012–2023. How to Replace the Residual Solvation Shell of Polar Active Site Residues to Achieve Nanomolar Inhibition of tRNA-Guanine Transglycosylase. [153] M. D. Prasanna, T. N. Guru Row, Cryst. Eng. 2000, 3, 135–154. C– halogen··· Interactions and Their Influence on Molecular Conformation and Crystal Packing: A Database Study. [154] I. Saraogi, V. G. Vijay, S. Das, K. Sekar, T. N. Guru Row, Cryst. Eng. 2003, 6, 69–77. C–halogen··· Interactions in Proteins: A Database Study. [155] D. Swierczynski, R. Luboradzki, G. Dolgonos, J. Lipkowski, H.-J. Schneider, Eur. J. Org. Chem. 2005, 1172–1177. Non-Ccovalent Interactions of Organic

249 9. References ______Halogen Compounds with Aromatic Systems – Analyses of Crystal Structure Data. [156] Y. N. Imai, Y. Inoue, Y. Yamamoto, J. Med. Chem. 2007, 50, 1189–1196. Propensities of Polar and Aromatic Amino Acids in Noncanonical Interactions: Nonbonded Contacts Analysis of Protein–Ligand Complexes in Crystal Structures. [157] Y. N. Imai, Y. Inoue, I. Nakanishi, K. Kitaura, Protein Sci. 2008, 17, 1129– 1137. Cl– Interactions in Protein–Ligand Complexes. [158] Y. Lu, Y. Wang, W. Zhu, Phys. Chem. Chem. Phys. 2010, 12, 4543–4551. Nonbonding Interactions of Organic Halogens in Biological Systems: Implications for Drug Discovery and Biomolecular Design. [159] H. G. Wallnoefer, T. Fox, K. R. Liedl, C. S. Tautermann, Phys. Chem. Chem. Phys. 2010, 12, 14941–14949. Dispersion Dominated Halogen– Interactions: Energies and Locations of Minima. [160] Z. J. Jia, Y. Wu, W. Huang, E. Goldman, P. Zhang, J. Woolfrey, P. Wong, B. Huang, U. Sinha, G. Park, A. Reed, R. M. Scarborough, B.-Y. Zhu, Bioorg. Med. Chem. Lett. 2002, 12, 1651–1655. Design, Synthesis and Biological

Activity of Novel Non-Amidine Factor Xa Inhibitors. Part 1: P1 Structure– Activity Relationships of the Substituted 1-(2-Naphthyl)-1H-pyrazole-5- carboxylamides. [161] C. A. Van Huis, C. F. Bigge, A. Casimiro-Garcia, W. L. Cody, D. A. Dudley, K. J. Filipski, R. J. Heemstra, J. T. Kohrt, L. S. Narasimhan, R. P. Schaum, E. Zhang, J. W. Bryant, S. Haarer, N. Janiczek, R. J. Leadley Jr., T. McClanahan, J. T. Peterson, K. M. Welch, J. J. Edmunds, Chem. Biol. Drug Des. 2007, 69, 444–450. Structure-based Drug Design of Pyrrolidine-1,2-dicarboxamides as a Novel Series of Orally Bioavailable Factor Xa Inhibitors. [162] J. M. Smallheer, S. Wang, M. L. Laws, S. Nakajima, Z. Hu, W. Han, I. Jacobson, J. M. Luettgen, K. A. Rossi, A. R. Rendina, R. M. Knabb, R. R. Wexler, P. Y. S. Lam, M. L. Quan, Bioorg. Med. Chem. Lett. 2008, 18, 2428– 2433. Sulfonamidolactam Inhibitors of Coagulation Factor Xa. [163] K. Yoshikawa, A. Yokomizo, H. Naito, N. Haginoya, S. Kobayashi, T. Yoshino, T. Nagata, A. Mochizuki, K. Osanai, K. Watanabe, H. Kanno, T. Ohta, Bioorg. Med. Chem. 2009, 17, 8206–8220. Design, Synthesis, and SAR

250 9. References ______of cis-1,2-Diaminocyclohexane Derivatives as Potent Factor Xa Inhibitors. Part I: Exploration of 5–6 Fused Rings as Alternative S1 Moieties. [164] H. Matter, M. Nazaré, S. Güssregen, D. W. Will, H. Schreuder, A. Bauer, M. Urmann, K. Ritter, M. Wagner, V. Wehner, Angew. Chem. 2009, 121, 2955– 2960; Angew. Chem. Int. Ed. 2009, 48, 2911–2916. Evidence for C–Cl/C– Br··· Interactions as an Important Contribution to Protein–Ligand Binding Affinity. [165] J. Jiang, J. Zhang, N. Vo, H. Vo, S. Chen, WO 2007/087429, 2007. Phenyl and Pyridyl Compounds for Inflammation and Immune-Related Uses. [166] G. Lai, T. Guo, Synth. Commun. 2008, 38, 72–76. A Convenient and Efficient Preparation of 2-Acetyl-4,5-difluorothiophene. [167] R. Paulini, K. Müller, F. Diederich, Angew. Chem. 2005, 117, 1820–1839; Angew. Chem. Int. Ed. 2005, 44, 1788–1805. Orthogonal Multipolar Interactions in Structural Chemistry and Biology. [168] K. Müller, C. Faeh, F. Diederich, Science 2007, 317, 1881–1886. Fluorine in Pharmaceuticals: Looking Beyond Intuition. [169] B. E. Smart, J. Fluorine Chem. 2001, 109, 3–11. Fluorine Substituent Effects (on Bioactivity). [170] A. D. Hamilton, D. Van Engen, J. Am. Chem. Soc. 1987, 109, 5035–5036. Induced Fit in Synthetic Receptors: Base Recognition by a "Molecular Hinge". [171] A. V. Muehldorf, D. Van Engen, J. C. Warner, A. D. Hamilton, J. Am. Chem. Soc. 1988, 110, 6561–6562. Aromatic–Aromatic Interactions in Molecular Recognition: A Family of Artificial Receptors for Thymine That Shows Both Face-to-Face and Edge-to-Face Orientations. [172] R. K. Castellano, V. Gramlich, F. Diederich, Chem. Eur. J. 2002, 8, 118–129. Rebek Imides and Their Adenine Complexes: Preferences for Hoogsteen Binding in the Solid State and in Solution. [173] J. Xu, A. T. Hamme II, Synlett 2008, 919–923. Efficient Access to Isoxazoles from Alkenes. [174] F. Himo, T. Lovell, R. Hilgraf, V. V. Rostovtsev, L. Noodleman, K. B. Sharpless, V. V. Fokin, J. Am. Chem. Soc. 2005, 127, 210–216. Copper(I)- Catalyzed Synthesis of Azoles. DFT Study Predicts Unprecedented Reactivity and Intermediates.

251 9. References ______[175] Y. Han, H. Hu, Synthesis 1990, 122–124. A Convenient Synthesis of Primary Amines Using Sodium Diformylamide as a Modified Gabriel Reagent. [176] J. A. Pfefferkorn, M. L. Greene, R. A. Nugent, R. J. Gross, M. A. Mitchell, B. C. Finzel, M. S. Harris, P. A. Wells, J. A. Shelly, R. A. Anstadt, R. E. Kilkuskie, L. A. Kopta, F. J. Schwende, Bioorg. Med. Chem. Lett. 2005, 15, 2481–2486. Inhibitors of HCV NS5B Polymerase. Part 1: Evaluation of the Southern Region of (2Z)-2-(Benzoylamino)-3-(5-phenyl-2-furyl)acrylic Acid. [177] T. Moriya, S. Takabe, S. Maeda, K. Matsumoto, K. Takashima, T. Mori, S. Takeyama, J. Med. Chem. 1986, 29, 333–341. Synthesis and Hypolipidemic Activities of 5-Thienyl-4-oxazoleacetic Acid Derivatives. [178] R. L. White Jr., F. L. Wessels, T. J. Schwan, K. O. Ellis, J. Med. Chem. 1987, 30, 263–266. 1-[[[5-(Substituted Phenyl)-2-oxazolyl]methylene]amino]-2,4- imidazolidinediones, a New Class of Skeletal Muscle Relaxants. [179] E. M. Beccalli, E. Borsini, G. Broggini, G. Palmisano, S. Sottocornola, J. Org. Chem. 2008, 73, 4746–4749. Intramolecular Pd(II)-Catalyzed Cyclization of Propargylamides: Straightforward Synthesis of 5-Oxazolecarbaldehydes. [180] M. E. Alonso, P. Jano, J. Heterocycl. Chem. 1980, 17, 721–725. The Syntheses of Ethoxycarbonyl-1,3-dioxoles and Oxazoles from the Copper Catalyzed Thermolysis of Ethyl Diazopyruvate in the Presence of Ketones, Aldehydes and Nitriles. [181] G. Campiani, S. Butini, F. Trotta, C. Fattorusso, B. Catalanotti, F. Aiello, S. Gemma, V. Nacci, E. Novellino, J. A. Stark, A. Cagnotto, E. Fumagalli, F. Carnovali, L. Cervo, T. Mennini, J. Med. Chem. 2003, 46, 3822–3839.

Synthesis and Pharmacological Evaluation of Potent and Highly Selective D3 Receptor Ligands: Inhibition of Cocaine-Seeking Behavior and the Role of

Dopamine D3/D2 Receptors. [182] N. Kornblum, W. J. Jones, G. J. Anderson, J. Am. Chem. Soc. 1959, 81, 4113– 4114. A New and Selective Method of Oxidation. The Conversion of Alkyl Halides and Alkyl Tosylates to Aldehydes. [183] J. P. Verge, P. Roffey, J. Med. Chem. 1975, 18, 794–797. Antiprotozoal Thiazoles. 2-(5-Nitro-2-thienyl)thiazoles. [184] MacroModel, version 9.7, Schrödinger, LLC, New York, NY, 2009. [185] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.

252 9. References ______Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009. [186] P. Ball, Chem. Rev. 2008, 108, 74–108. Water as an Active Constituent in Cell Biology. [187] N. Haginoya, S. Kobayashi, S. Komoriya, T. Yoshino, T. Nagata, Y. Hirokawa, T. Nagahara, Bioorg. Med. Chem. 2004, 12, 5579–5586. Design, Synthesis, and Biological Activity of Non-Amidine Factor Xa Inhibitors Containing Pyridine N-Oxide and 2-Carbamoylthiazole Units. [188] Y. Shi, D. Sitkoff, J. Zhang, W. Han, Z. Hu, P. D. Stein, Y. Wang, L. J. Kennedy, S. P. O'Connor, S. Ahmad, E. C.-K. Liu, S. M. Seller, P. Y. S. Lam, J. A. Robl, J. E. Macor, K. S. Atwal, R. Zahler, Bioorg. Med. Chem. Lett. 2007, 17, 5952–5958. Amino(methyl) Pyrrolidines as Novel Scaffolds for Factor Xa Inhibitors. [189] K. R. Guertin, C. J. Gardner, S. I. Klein, A. L. Zulli, M. Czekaj, Y. Gong, A. P. Spada, D. L. Cheney, S. Maignan, J.-P. Guilloteau, K. D. Brown, D. J. Colussi, V. Chu, C. L. Heran, S. R. Morgan, R. G. Bentley, C. T. Dunwiddie, R. J. Leadley, H. W. Pauls, Bioorg. Med. Chem. Lett. 2002, 12, 1671–1674. Optimization of the ß-Aminoester Class of Factor Xa inhibitors. Part 2: Identification of FXV673 as a Potent and Selective Inhibitor with Excellent In Vivo Anticoagulant Activity. [190] G. R. Desiraju, Angew. Chem. 1995, 107, 2541–2558; Angew. Chem. Int. Ed. Engl. 1995, 34, 2311–2327. Supramolecular Synthons in Crystal Engineering–A New Organic Synthesis.

253 9. References ______[191] P. Metrangolo, H. Neukirch, T. Pilati, G. Resnati, Acc. Chem. Res. 2005, 38, 386–395. Halogen Bonding Based Recognition Processes: A World Parallel to Hydrogen Bonding. [192] P. Metrangolo, F. Meyer, T. Pilati, G. Resnati, G. Terraneo, Angew. Chem. 2008, 120, 6206–6220; Angew. Chem. Int. Ed. 2008, 47, 6114–6127. Halogen Bonding in Supramolecular Chemistry. [193] L. A. Hardegger, B. Kuhn, B. Spinnler, L. Anselm, R. Ecabert, M. Stihle, B. Gsell, R. Thoma, J. Diez, J. Benz, J.-M. Plancher, G. Hartmann, D. W. Banner, W. Haap, F. Diederich, Angew. Chem. 2011, 123, 329–334; Angew. Chem. Int. Ed. 2011, 50, 314–318. Systematic Investigation of Halogen Bonding in Protein–Ligand Interactions. [194] H. Wang, A. Ben-Naim, J. Med. Chem. 1996, 39, 1531–1539. A Possible Involvement of Solvent-Induced Interactions in Drug Design. [195] ACD/Name, version 6.0, Advanced Chemistry Development, Inc., Toronto (Canada), 2002. [196] C. Barbier, A. Joissains, A. Commerçon, J.-F. Riou, F. Huet, Heterocycles 2000, 53, 37–48. Preparation of Lavendamycin Analogues. [197] W. Mathes, W. Sauermilch, Chem. Ber. 1957, 90, 758–761. Über einige Substituierte Aldehyde der Chinolin- und Pyridinreihe. [198] D. G. Doherty, R. Shapira, W. T. Burnett Jr., J. Am. Chem. Soc. 1957, 79, 5667–5671. Synthesis of Aminoalkylisothiuronium Salts and Their Conversion to Mercaptoalkylguanidines and Thiazolines. [199] R. F. Brown, G. H. Schmid, J. Org. Chem. 1962, 27, 1288–1294. Synthesis of Some Substituted 5-Bromopentylamine Hydrobromides. [200] G. Deng, Y. Chen, Macromolecules 2004, 37, 18–26. A Novel Way to Synthesize Star Polymers in One Pot by ATRP of N-[2-(2- Bromoisobutyryloxy)ethyl]maleimide and Styrene. [201] R. C. Clevenger, K. D. Turnbull, Synth. Commun. 2000, 30, 1379–1388. Synthesis on N-Alkylated Maleimides. [202] B. Pal, P. K. Pradhan, P. Jaisankar, V. S. Giri, Synthesis 2003, 1549–1552. First Triphenylphosphine Promoted Reduction of Maleimides to Succinimides. [203] T. F. G. Henn, M. C. Garnett, S. R. Chhabra, B. W. Bycroft, R. W. Baldwin, J. Med. Chem. 1993, 36, 1570–1579. Synthesis of 2'-Deoxyuridine and 5-

254 9. References ______Fluoro-2'-Deoxyuridine Derivatives and Evaluation in Antibody Targeting Studies. [204] N. B. Mehta, A. P. Phillips, F. F. Lui, R. E. Brooks, J. Org. Chem. 1960, 25, 1012–1015. Maleamic and Citraconamic Acids, Methyl Esters, and Imides. [205] J. R. Heitz, C. D. Anderson, B. M. Anderson, Arch. Biochem. Biophys. 1968, 127, 627–636. Inactivation of Yeast Alcohol Dehydrogenase by N- Alkylmaleimides. [206] K. B. Becker, A. F. Boschung, C. A. Grob, Helv. Chim. Acta 1973, 56, 2733– 2747. Nucleophilic Reactions at Tertiary Carbon. Part 2. -and -Routes to 8- Hydrindanyl Cation. [207] A. Buschauer, A. Friese-Kimmel, G. Baumann, W. Schunack, Eur. J. Med.

Chem. 1992, 27, 321–330. Synthesis and Histamine H2 Agonistic Activity of Arpromidine Analogs: Replacement of the Pheniramine-like Moiety by Non- Heterocyclic Groups. [208] J.-Y. Kazock, M. Taggougui, B. Carré, P. Willmann, D. Lemordant, Synthesis 2007, 3776–3778. Simple and Efficient Synthesis of N-Quaternary Salts of Quinuclidinium Derivatives. [209] J. Sicé, J. Am. Chem. Soc. 1953, 75, 3697–3700. Preparation and Reactions of 2-Methoxythiophene. [210] R. D. Schuetz, G. P. Nilles, J. Org. Chem. 1971, 36, 2188–2190. Synthesis and Nuclear Magnetic Resonance Investigation of Some Fluorothiophenes. [211] J. L. Casarrubio, S. Conde, C. Corral, J. Lissavetzky, J. Heterocycl. Chem. 1983, 20, 1557–1560. On the Syntheses of Thiophene Analogs of Practolol and "Reversed" Practolol. [212] T. S. Gardner, J. Lee, E. Wenis, J. Org. Chem. 1961, 26, 1514–1518. Synthesis of 5-Substituted 3-Isoxazolecarboxylic Acid Hydrazides and Derivatives. [213] C. M. Beaton, N. B. Chapman, K. Clarke, J. M. Willis, J. Chem. Soc. Perkin Trans. 1 1976, 2355–2363. Some Derivatives of 2- and 3-Phenylthiophen. [214] G. Kühnhanss, H. Reinhardt, J. Teubel, J. Prakt. Chem. 1956, 3, 137–145. Beitrag zur Carboxylierung des Thiophens mit Harnstoffchlorid und Oxalylchlorid. [215] R. A. Turner, J. Am. Chem. Soc. 1946, 68, 1607–1608. - Dimethylaminoethylamine and Dimethylaminoacetonitrile.

255 9. References ______[216] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna, J. Appl. Crystallogr. 1999, 32, 115–119. SIR97: A New Tool for Crystal Structure Determination and Refinement. [217] G. M. Scheldrick (1997). SHELXL97. Program for the Refinement of Crystal Structures. University of Göttingen, Germany. [218] R. H. Blessing, Acta Crystallogr. A 1995, 51, 33–38. An Empirical Correction for Absorption Anisotropy. [219] G. M. Sheldrick, Acta Crystallogr. A 2008, 64, 112–122. A short history of SHELX. [220] W. Kabsch, J. Appl. Crystallogr. 1993, 26, 795–800. Automatic Processing of Rotation Diffraction Data from Crystals of Initially Unknown Symmetry and Cell Constants. [221] Collaborative Computational Project, Number 4, Acta Cryst. 1994, D50, 760– 763. [222] P. Emsley, K. Cowtan, Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004, 60, 2126–2132. Coot: Model-building Tools for Molecular Graphics. [223] G. N. Murshudov, A. A. Vagin, E. J. Dodson, Acta Crystallogr., Sect. D: Biol. Crystallogr. 1997, 53, 240–255. Refinement of Macromolecular Structures by the Maximum-Likelihood Method.

256