Synthesis of Antibiotic Scaffolds Inspired by the Tan 1057-D Biologically Active Natural Product Through the Utilization of Ullmann Copper(I) and Gold Cyclization Chemistry

As bacterial resistance to drugs continues to rise, the search for new antibiotics is of paramount importance. One key element to finding successful antibiotics is to uncover small molecules that act through novel mechanisms of action in order to decrease the risk of cross-resistance and provide antibiotics with longer clinical lifetimes. Based on previous work conducted in our research group, the TAN-1057 antibiotic family shows promise as new candidates in part due to their inhibition of protein synthesis at the point of tRNA acyl-transfer in the ribosome. This project involves the synthesis of the TAN-1057 D heterocyclic core, as well as the elucidation of TAN- 1057 D interactions in the ribosome. A modular synthetic strategy has been implemented to construct the heterocyclic core of TAN-1057 D. This strategy was chosen in order to streamline analog development via facile functional group additions and control of the heterocyclic ring oxidation state. The core has been synthesized through the utilization of Ullmann copper(I) chemistry, and manipulation of the core is currently in progress to allow for analog development. The installation of functional group handles will allow for a wide variety of analogs to be synthesized, and then undergo biological testing. Upon completion of TAN-1057 D derivatives, we will be able to explore the mechanism of action of the parent compound and gather insight into the structural characteristics of a future drug candidate.

ABSTRACT As bacterial resistance to drugs continues to rise, the search for new antibiotics is of paramount importance. One key element to finding successful antibiotics is to uncover small molecules that act through novel mechanisms of action in order to decrease the risk of cross-resistance and provide antibiotics with longer clinical lifetimes. Based on previous work conducted in our research group, the TAN-1057 antibiotic family shows promise as new candidates in part due to their inhibition of protein synthesis at the point of tRNA acyl-transfer in the ribosome. This project involves the synthesis of the TAN-1057 D heterocyclic core, as well as the elucidation of TAN1057 D interactions in the ribosome. A modular synthetic strategy has been implemented to construct the heterocyclic core of TAN-1057 D. This strategy was chosen in order to streamline analog development via facile functional group additions and control of the heterocyclic ring oxidation state. The core has been synthesized through the utilization of Ullmann copper(I) chemistry, and manipulation of the core is currently in progress to allow for analog development. The installation of functional group handles will allow for a wide variety of analogs to be synthesized, and then undergo biological testing. Upon completion of TAN-1057 D derivatives, we will be able to explore the mechanism of action of the parent compound and gather insight into the structural characteristics of a future drug candidate. ii ACKNOWLEDGMENTS I would like to thank my advisor Prof. Dr. Ryan E. Looper for allowing me to conduct research in his group for the past three and a half years, and for opening my eyes to the beauty of total synthesis. He has been a wonderful mentor and invaluable resource through the entirety of my time conducting research in his lab. I also want to thank Matthew Nelli, a graduate student in the Looper group who has taught me many of the laboratory techniques I know and use today, as well as how to better think like a chemist. His expertise and guidance have been invaluable during my time working on the TAN-inspired core project, due both to the fact that my project is a side project related to his own work, and because he is a brilliant scientist. Thanks are also in order for Cody Bender, M.S., a former member of the Looper Group and the first graduate student I worked with, as well as Chelsea Harmon; working with these two in bacterial culturing and compound isolation helped me gain insight into the connections between biology and synthetic chemistry. I would also like to thank Markus Menke, a graduate student in the Schulz Group at the Technical University of Braunschweig, for pushing me to become a more independent and confident chemist during my summer internship synthesizing farnesol-derived terpenoid frog pheromones. I want to thank the rest of the Looper Group, past and present, for all the support they have given me in science and in life. Last but certainly not least, I want to thank my family for everything they’ve done to help me reach this point in my academic career. My parents continuously encouraged me to explore the world around me and helped foster a sense of curiosity that has directly benefited my scientific endeavors, and my brother has and is always inspiring me to live life to the fullest. I am thankful for and care about each of you more than you know, and I hope I can, in a small way, honor you all through my work and through my life. iii TABLE OF CONTENTS Abstract ii Acknowledgements iii Introduction 1 Retrosynthesis 9 Methods 11 Results and Discussion 19 Conclusion 22 Experimentals 26 References 39 iv Introduction A rise in antimicrobial resistance is leading to a major antibiotic shortage that must be addressed. Despite the need for new antibiotics, no novel antibiotic classes have been developed since the 2000’s [Fig. 1].1 When developing an antibiotic, many challenges must be overcome. Factors such as low economic appeal, difficulties in identifying promising candidates, and a lull in fully characterizing compounds with novel mechanisms of action all contribute to the lack of discovery of novel antibiotics.2 Because of the multiple challenges related to antibiotic development, many private Figure 1: Timeline of antibiotic class development. Numerous classes of antibiotics have been studied and utilized by the general public. However, the most recent class, the lipopeptides, appeared in 2000. No new classes have been fully developed after that point.1 corporations prefer to devote time and resources to developing other drugs, and researchers in academia are often concerned just as much with uncovering chemical insight as they are with contributing to the medicinal chemistry field.3 It can take years, even decades after a promising natural product is identified for the compound to become utilized as an antibiotic. Therefore, the time to start developing novel antibiotics is now. While there are several classes of antibiotics that target the cell wall and its biogenesis, (e.g. penicillins, cephalosporins, carbapenems and other classes like the vancomycins), current antibiotics that act intracellularly are confined to inhibiting just three targets. For example, 1 tetracyclines, macrolides, aminoglycosides, inhibit protein synthesis by attacking the ribosome.4 Quinolones inhibit transcription and the sulfonamides inhibit nucleic acid synthesis, as seen in Figure 2 designed by Elongavan et. al.4 While these categories are very general, Figure 2: Common intracellular antibiotic target areas. As seen in the figure, common use antibiotics often attack one of three areas: cell wall synthesis, nucleic acid synthesis, and protein synthesis processes. The specific area a compound targets as well as it’s mechanism of action help determine which class a compound belongs to.4 the fact that there are only three main areas where antibiotics affect cells is an important factor to consider when searching for antibiotics with a new mechanism of action. Bacteria can easily adapt against repeated inhibition. The ribosome, cell wall, and nucleic acid syntheses processes are all validated targets. However, new mechanisms of action are always needed due to the speed at which bacteria develop resistance to current antibiotics.5 Because bacteria are learning to combat current antibiotics, it is crucial to develop new classes of antibiotics that cells via novel mechanisms of action. While there are already multiple compounds that target elements of the ribosome, there are still many stages of the translation process that are only inhibited by antibiotics with low efficacy and result in severe side effects, or are targeted by compounds currently resisted by certain bacterial strains. The ribosome is an integral organelle for translation of RNA. To start 2 the translation process, the 30S and 50S subunits must come together, along with the RNA template [Fig. 3].6 Then, a complementary tRNA enters via the A site. The tRNA moves into the P site where a peptide bond is formed between the first tRNA and the next tRNA in the sequence. This process continues, forming a chain that grows out of the P site, and empty tRNA carriers exit via the E site post amino acid transfer. Once Figure 3: RNA translation process. Once the ribosomal subunits bind and mRNA is present, the translation process can begin. Multiple factors and involved in the success or failure of protein production. A number of antibiotics attack this process; however, the efficacy and side effects of such antibiotics can vary drastically. 6 translation of the nascent peptide is complete, the 30S and 50S subunits of the ribosome split apart and will be utilized in future translation operations. Because translation involves a lot of moving parts, there are multiple aspects of the process that can be utilized as antibiotic targets. One aspect of translation not targeted by antibiotics currently on the market is inhibition of peptide bond formation via interruption of the Watson Crick base paired between the template and tRNA through binding at the P site. The P site is where the growing peptide chain is held throughout the translation process.7 At this time, there are some antibiotic classes that block the exit of the P site, for example, the lincosamides and streptogramins, effectively halting translation by preventing the nascent peptide chain from leaving the ribosome.8,9 However, the Looper Research Group is interested in compounds that inhibit the peptidyl transferase center 3 (PTC) through P site binding [Fig. 4]. Essentially, the inhibition occurs when the compound of interest clogs the P site and halts formation of the peptide chain by interrupting the Watson Crick base pairing between incoming tRNA and the template. While some natural products interrupt the PTC at the A site, no antibiotics on the market target via the P site. There are a few such natural products currently under investigation within the Looper Research Group, for example, amicetin and the TAN 1057 family, and the compound blasticidin Figure 4: Crystal structure of amicetin bound at the P site in the ribosome. No antibiotics currently on the market utilize the mechanism of action displayed by the P-site inhibitors studied in the Looper Research Group.10 has been studied by others groups but is utilized in comparing selectivity differences between compounds that bind in the P site.10 While the compounds have some shared structural motifs, there are noticeable differences in the selectivity, toxicity, and potency of each compound.11 However, all three are valuable to study due to the fact that they inhibit transcription through a novel mechanism of action. Both blasticidin and amicetin inhibit the P site of the ribosome by utilizing a number of intermolecular interactions. As seen in Figure 5, there are three main intermolecular interactions at play: H-bonding, cation-, and Watson-Crick. The strength of the binding interactions varies for the two molecules; for example, BlaS has three hydrogen bonding interactions near the cation-pi site where Ami has only one. While BlaS and Ami bind in the same location and take advantage of similar interactions, the variety in the strength of those interactions has important implications. 4 very basic guanidine moiety that, when protonated under physiological conditions, affords a strong electro-static interaction on the eastern fragment. These qualities contribute to blasticidin’s nonselectivity and cytotoxicity. Because the TAN-1057 Figure 6: Crystal structures of the prokaryotic (left) and eukaryotic (right) P sites within their respective ribosomes. The prokaryotic P site displays two subunits (L16 and L27), whereas the eukaryotic hosts the much larger L10 and a truncated L29. 12 compounds have structural similarities to Ami and BlaS, and due to the fact that it is known that TAN-1057 A binds in the P site as well, the Looper Group is interested in learning more about the mechanism of action of the TAN compounds. The TAN 1057 family consists of four compounds, where the respective pairs differ by a stereocenter [Fig. 7].14 TAN A and B boast a functionalized six membered ring and guanidine with an alkyl chain separating the two, where TAN C and D have a large sevenmembered core with tails on either side of the molecule. It is known that TAN C can interconvert to TAN B in basic solutions, and TAN D will convert to an epimeric mixture of Figure 7: Scanning Electron Microscopy image of Flexibacter sp. overlaid with TAN 1057 A-D structures. Note that TAN A/B and TAN C/D differ only by a stereocenter. A/B in various conditions when heated.15 TAN A and D display promising minimum inhibitory 6 concentration (MIC) values [Table 1]14. Due to their similarities to BlaS and Ami, the TAN 1057 compounds are of Table 1: Table displaying MIC values for TAN D against three gram positive bacteria strains. great interest to the Looper Group. TAN D is especially intriguing due to a number of factors. First, TAN A and B have been studied and synthesized in the past, and although TAN C and D have been synthesized, little is known about their biological attributes.16 Second, both TAN C and D contain a N-carbamoyl guanidine moiety not often seen in nucleoside antibiotics or even antibiotics in general. Additionally, TAN C is more easily converted into TAN B than TAN D is converted to TAN A/B; therefore, synthesizing TAN D would allow for further investigation into the stability of the molecule. Another important fact is that TAN D’s MIC values are more promising than TAN B/C’s values14. When taking all the background into consideration, two main questions rise to the forefront: (a) is TAN D responsible for the MIC activity shown, or is the activity the result of conversion to TAN A and (b) how exactly does TAN D inhibit translation, ie does it have the same mechanism of action as TAN A, and if so, how does it bind within the ribosome without a canonical Watson Crick base pair? When examining the characteristics of the TAN D molecule, it becomes clear that further investigation into its structure and mechanism of action would prove beneficial to the scientific community. 7 Although TAN D is an undeniably promising molecule regarding P site inhibition, there are also some concerns regarding cytotoxicity. It is useful to compare the structures of the three molecules (Ami, BlaS, and TAN D) in order to see why cytotoxicity might be an issue with TAN D [Fig. 8]. When comparing critical functional groups for substrate binding, it becomes clear that a major difference between Ami and both BlaS and TAN D is that Ami employs electrostatic interactions on the left side of the molecule, where this interaction is missing entirely in BlaS and TAN D.17 Keeping analog Figure 8: Regional interaction comparisons between Ami, BlaS, and TAN D. Ami is the only natural product to take advantage of electrostatic interactions within the P site, which may contribute to its heightened selectivity. Both BlaS and TAN D contain a strongly basic guanidine moiety that contributed strongly to interactions in the cation π region. Of importance is the presence of the Ncarbamoyl guanidine moiety in TAN D; this unique functional group is rarely seen in drug-like compounds. development in mind, it would be beneficial to install functional groups on the left tail of TAN D that would also utilize these interactions to see how the electrostatic region affects efficacy and selectivity. For the Watson-Crick region, BlaS and Ami employ pyrimidone moieties while TAN D displays the unique guanidinium-carbamide functional group. This is a definite benefit of the TAN D molecule over the other two Figure 9: Scaffold design inspired by the TAN D natural product. The synthetic routes utilized in this project host varying bridge elements and allow for a variety of amine/diamine fragments to be installed. Overall, the design facilitates analog development. compounds regarding expansion of scientific knowledge. Finally, the cation-pi region reveals the 8 main concern about the selectivity of TAN D. Both BlaS and TAN D have guanidinium functional groups on their right tails, where as Ami only has a substituted amine. Recall from Figure 5 that Ami has only one hydrogen bonding interaction in the addition to the cation-pi binding, where BlaS has three. Likely the presence of the guanidinium moiety is contributing to the cytotoxicity of BlaS; therefore, it would be beneficial to develop TAN D analogs that do not have the guanidinium functional group. In summary, the naturally-occurring TAN D compound shows promise, but will need to be tailored for use as an antibiotic [Figure 9]. In order to investigate the antimicrobial effectiveness of the TAN D compound and its analogs, the molecule must first be synthesized. When approaching TAN D, the main goal was to determine how to synthesize the core. This project explores two approaches to the core, ultimately resulting in two TAN D inspired cores at various oxidation stages. The first approach hinges upon an Ullmann copper(I) cyclization, while the second depends on gold catalysis.18,19 The retrosynthetic approaches, synthetic methods, and future directions will all be presented in this thesis. Retrosynthesis A basic retrosynthetic scheme is useful for demonstrating the manipulability of the synthetic route undertaken in the project. As shown in Scheme 1, the ultimate goal of developing Core A Scheme 1: Retrosynthetic scheme for Core A. Ultimately, a variety of analogs with different amine/diamine groups on the tails will be synthesized from a general core. The core will have alcohols that can be differentiated (aryl vs alkyl), and will be derived from an iodoallylic aldehyde and serinamide pair. To generate the iodoallylic aldehyde, iodophenyl methanol will undergo a number of transformations. 9 the cyclic core is to allow for multiple TAN D inspire analogs to be synthesized. The final product will have any number of amines and diamines attached so that the binding properties can be studied and the potency and cytotoxicity associated with varying functional groups can be assessed. Before developing analogs, the core of the molecule must first be synthesized. The desired cyclized product, ie Core A, has two protecting groups on the alcohols. These protecting groups can be varied to allow for selective deprotection, or two of the same groups can be used to facilitate global deprotection and subsequent selective oxidation of the free alcohols. Regarding formation of the cyclized product, using an amino acid derivative provides a large amount of synthetic flexibility. Any number of carboxamides can be utilized in the place of the serinamide, affording numerous analogs from a simple starting material change. However, the underlying goal in utilizing an amino acid is to set the stereocenter from the starting materials rather than having to set it synthetically. Additionally, the serinamide was chosen to allow for facile analog development once the core has been deprotected. Flexibility within the synthetic plan is also evident in the generation of the iodoaldehyde; although the aldehyde for the model substrate is derived from an aryl iodide, couplings between serinamide and alkyl iodo enals should be facile. This provides yet another opportunity for manipulation of the core and tails. Retrosynthetic analysis of Core B helps emphasize a core manipulability similar to that seen in the development of Core A [Scheme 2]. While the final product of Core B does not afford as much freedom regarding analog development as Core A due to the lack of free alcohol moieties, the synthetic route for Core B still takes advantage of the use of a modified amino acid as a source of variety and facile insertion of a stereocenter. Additionally, the starting material used to synthesize the yne- acid can be easily adjusted to allow for the development of a number of analogs with unique side chains. Of importance to note, however, is that Core B was not 10 specifically designed with analog development in mind; the main goal of the Core B project was as a methods project highlighting a gold cyclization. Therefore, the relative reduction in freedom was not seen as a detriment. Core B Scheme 2: Retrosynthetic scheme for Core B. While analogs can ultimately be synthesized from this core as well, the main focus of the chemistry is on showcasing the gold cyclization methodology. Core B will be derived from a valine hydroxamate and an yne acid, which will be synthesized via manipulation of 4-pentyn-1-ol. Methods: To arrive at Core A, an aryl iodoketone and a protected serinamide had to be synthesized [Scheme 3]. The synthesis of the aryl iodoketone is shown in Scheme 2. Protection of iodophenyl methanol afforded aromatic silane 2 in 94% yield. Compound 2 was then coupled with propargyl alcohol via a Sonogashira coupling, which resulted in the generation of the alkynol 3 in 78% yield. The next step in the synthesis consisted of a reductive iodination utilizing Red-Al® to reduce the alkyne followed by quenching of the organo-aluminum complex with n-iodosuccinimide. Ensuring anhydrous conditions is crucial to the success of the iodination; if the reaction is exposed to a proton source, the allylic alcohol is produced. To produce the aryl iodo enal, the iodo allylic alcohol is exposed to manganese (IV) dioxide overnight to give the aldehyde 5 in 95% yield. 11 3 2 1 5 4 Scheme 3: Route from iodophenyl methanol to the iodo allylic alcohol. Generation of the O-TBS serinamide was completed by a facile protection of the commericially available L-serinamide HCl salt (6) to produce serinamide 7 in 92% yield, as shown in Figure 10. Treatment of a mixture of 5 and 7 with sodium cyanoborohydride 6 7 followed by addition of formaldehyde afforded the Figure 10: Route from iodophenyl methanol to the iodo allylic l h l reductive amination product 8 in 66% yield [Scheme 4]. The cyclized product 9 was generated in 44% yield by adding a solution of 8 in ethanol to copper (I) iodide and caesium carbonate under anhydrous conditions. Initial cyclization attempts made at elevated temperatures (70 °C) resulted in the degradation of the desired product as well as starting materials. However, while the room 12 7 8 5 9 Scheme 4: Route from the two subunits (iodo allylic ketone and serinamide) to the cyclized core. temperature cyclization resulted in generation of desired product, some starting material remained. Therefore it is necessary to optimize the reaction further. The first step towards core functionalization consisted of treatment of the cyclized product 9 with HF pyridine to produce globally-deprotected heterocycle 10 [Fig. 11]. However, this reaction was unsuccessful. Multiple conditions were screened in an effort to deprotect the cyclized product. Unfortunately, none of the conditions screened resulted in desired product 10; only degradation products were observed. Therefore, an attempt will be made to reduce the alkene moiety within the seven-membered ring in the hopes that removal of the conjugation would facilitate deprotection of the alcohols by removing suspected detrimental functionality via suppression of starting material/product degradation. 13 10 9 Figure 11: Deprotection reaction conditions. As of the writing of this thesis, no conditions had been found to successfully deprotect the core without leading to degradation. An attempt will be made to dehydrate the alkene before further deprotection efforts are undertaken. While optimizing the reactions to reach Core A, work was initiated to synthesize Core B. Initially, the synthetic plan involved generating an yne-acid and carboxamide pair for use in a peptide coupling preceding the cyclization step. Therefore, initial work focused on development of a straightforward yne-acid protocol. The synthesis of the yne-acid 13 is shown is Scheme 5. Protection of 4-pentynol (11) resulted in alkynyl ether 12 in 90% yield. Subsequent treatment of 12 with n-butyllithium and carbon dioxide gas afforded the desired carboxylation product, yne-acid 13, in 83% yield. Ensuring the benzylether solution was properly saturated with carbon dioxide proved crucial to the success of the carboxylation, as well as maintaining temperatures below 0 °C for the entirety of the reaction. 14 Additional hydroxamates were synthesized as well [Fig. 13]. Following the same methods outlined in Scheme 6, compounds 28 and 29 were synthesized in good yield; 73 % yield each [Fig. 13]. Regrettably, following peptide coupling with yne-acid 13, the resulting coupled products were seen in the crude NMR, but were unable to be purified. Yne-acid 13 and valine hydroxamate 27 were utilized in a peptide coupling to form the coupled product 30 in 73% yield. Purification of 30 proved to be quite challenging; however, once a pure sample of 30 was obtained, the material was pushed forward through the gold cyclization with various gold catalysts to result in what was initially believed to be the desired cyclized product 31 [Scheme 9]. However, later analysis revealed that this product was likely the undesired O-cyclization product 32. 27 30 31 13 32 Scheme 9: Route from the Valine hydroxamate and the yne-acid to the cyclized product(s). While initial Carbon NMR analysis suggested the desired product had been synthesized as the major product, it was later revealed that O cyclization was occurring instead. Alternative cyclization products were also observed (see Figure 15). While the hydroxamates did appear to aid in promotion of the cyclization, use of the HCl salts proved to be impractical. Therefore, efforts were made to synthesize carboxylbenzyl (Cbz) protected hydroxamates instead. Cbz-L-valine was converted to the hydroxamate through the same procedure utilized to make hydroxamates 27, 28, and 29, and the resulting salt proved to be much more manageable due to a lower viscosity. 18 Matthew Nelli, a graduate student in the Looper Group, who synthesized a cyclopropane derivative to obtain a solid, which allowed him to acquire a crystal structure of the compound. While he has isolated small amounts of the desired N connectivity product 31, the equilibrium of the reaction still favors the O cyclized product 32 [Fig. 15]. Additionally, the 6-exo product 34 has also been observed. The presence of these byproducts, in addition to the purification issues and difficulty in obtaining a crystal structure of the cyclization product, led to a shift in focus for this project back to optimization of and advancement of the synthesis of Core A. However, Matt Nelli is still making efforts towards completing the synthesis of Core B. Regarding the reactions detailed in this thesis, the first challenge faced in the synthesis of Core B related to formation of yne-acid 13. The carboxylation reaction has proven to be more fickle than other reactions due to a number of factors. First, benzylether 12 is prone to degradation if the solution of benzylether in tetrahydrofuran is not cooled properly. Using a -78 °C ice bath is crucial for the first step, and when the reaction is scaled up the vessel must be cooled longer before addition of n-butyllithium or degradation will occur. Additionally, the solution must be properly saturated with carbon dioxide gas in order for the reaction to be successful. This entails bubbling carbon dioxide directly into the solution for at least thirty minutes, then expelling nitrogen gas from the vessel to allow carbon dioxide to dominate the atmosphere. Finally, the reaction must not be allowed to warm above -20 °C during the addition of carbon dioxide. A minor adjustment was made to the procedure where the reaction vessel was raised to 0 °C rather than -20 °C, but this resulted in more degraded product that was nearly impossible to separate from desired product. If reaction procedures are followed with the proper techniques, the carboxylation results in high yield of pure to nearly-pure material without the need for further purification. 21 For Core A, the most obvious source of variety lies with the differentiation of the tails. Because the tail alcohols can be easily differentiated after deprotection (one is benzylic while the other is attached to an alkyl chain), a simple manganese dioxide oxidation should afford the benzaldehyde 35. With the benzylic aldehyde in place, the next step would be to install any number of amine and diamine moieties via condensation [Scheme 10]. This would afford the first analog 36 to be tested in biological assays. Installing the N-carbamoyl guanidine will likely prove to be more challenging. Currently, the plan is to convert the alcohol into an amine, followed by guanylation with a pseudo thioureau, or to attempt a direct Mitsunobu by starting with a N-carbamoyl guanidine, as seen in Scheme 11. A direct Mitsunobu would likely run into 36 37 Scheme 11: Future directions regarding installment of the N-carbamoyl guanidine moiety. Due to the unique nature of the N-carbamoyl guanidine moiety within bioactive natural products, installing the functional group will allow for further analysis of its binding properties and interactions within the cell. issues due to regiochemical difficulties or possibly steric hindrance, but the reaction would be worth attempting due to the directness of the route. Going back to the core synthesis, the bridge segment, in this case the aromatic region, can be substituted for any number of other groups. For example, Matt Nelli, the graduate student involved in the gold cyclization, has synthesized an analog of Core A with an alkyl chain rather 23 than the aromatic ring. Any moiety that can be converted into the iodo enal could be utilized as a bridge section of the core. For Core B, a number of analogs can also be synthesized. While differentiation of Core A analogs would likely happen late in the synthesis, Core B differentiation would occur earlier. Any number of hydroxamates can be made from various amino acids, and a variety of yne-acids can be made by substituting propargyl alcohol for other alkynyl compounds. Unfortunately, the final structure of Core B is not necessarily as malleable as Core A, although it does have different oxidation states than Core A that could prove useful. It is again important to note that Core B’s main purpose was to showcase the gold cyclization chemistry, and that analog development is a secondary goal. Because the synthetic routes for both cores were designed with analog development in mind, there are many ways the routes can be modified to produce multiple analogs. Overall, this project has resulted in the development of the desired Core A antibiotic candidate scaffold, as well as progress towards the synthesis of the Core B structure. Valuable insight into multiple reactions was gained as well. Generation of the aryl iodo allylicalcohol via the reductive iodination was a crucial reaction in the synthesis of Core A, and an important observation was made in determining that heat promoted generation of what is believed to be the beta-elimination alkyne byproduct during the copper cyclization. The challenges faced in using common peptide coupling reagents for the Core B precursor emphasized the fact that even ‘straightforward’ reactions can be more challenging than they first seem, and that careful purification of compounds and proper analysis of NMR data are crucial skills in any total synthesis endeavor. Changing from the carboxamide to the hydroxamate in an effort to promote the gold cyclization highlighted the important role electronegativity plays in these cyclizations, 24 and the determination that the resulting product displayed O cyclization rather than N cyclization demonstrate the limitations of NMR spectroscopy. In addition to the synthetic knowledge gained during this project, several transferable skills regarding analysis and effective laboratory technique were gained. Development of antibiotic candidates hypothesized to inhibit bacteria via a novel mechanism of action and advancement of the knowledge within the synthetic organic community are two valuable results from this project. Because antibiotic resistance is becoming increasingly prevalent worldwide, all help is needed in finding new ways to inhibit bacteria, and being able to contribute to the wealth of knowledge regarding synthetic reactions at the same time is valuable as well.20 Both cores show great promise for analog development, and once biological testing can begin, more knowledge regarding P site inhibition will be gained. Researching new antibiotics is a crucial endeavor, and utilizing organic synthesis in antibiotic candidate development is a valuable pursuit. 25 addition of methyl iodide, the reaction was warmed to room temperature and stirred overnight. Upon completion, water was added to the reaction to quench any remaining sodium hydride. The reaction was washed with ether, and the ether underwent a sodium bicarbonate extract. Following ether extraction, the aqueous layers were collected and acidified to pH 2. The acidified aqueous layers were then extracted with ethyl acetate, and the ethyl acetate extract was washed with sodium thiosulfate before being dried over sodium sulfate and concentrated via rotary evaporator, which provided Compound 15 as a white solid (quantitative yield). IR (thin-film) 2969, 2934, 2361, 2338, 1740, 1697, 1661 cm-1; 1H NMR (500 MHz, CDCl3) δ 4.14 (dd, J = 28.9, 10.4 Hz, 1H), 2.87 (s, 3H), 2.25 (d, J = 27.1 Hz, 1H), 1.47 (s, 9H), 1.03 (d, J = 6.5 Hz, 3H), 0.92 (d, J = 6.7 Hz, 3H) ppm; 13C NMR (125 MHz, CDCl3) δ 155.7, 81.0, 80.6, 65.1, 28.3, 27.5, 20.1, 19.7, 19.0 ppm; Tert-butyl (S)-(1-amino-3-methyl-1-oxobutan-2-yl)(methyl)carbamate (16): Triethylamine was added to a reaction vessel, followed by tetrahydrofuran, to make a solution that was cooled to 0 °C before addition of Compound 15 (1 equiv.). Carbonyldiimidazole (CDI) was then added to the reaction in one portion WITH VENT. After warming to room temperature and stirring for 1 hour, aqueous ammonia was added to the reaction WITH VENT, and the reaction was stirred until starting material was consumed. Saturated ammonium chloride was added to the reaction, followed by an ethyl acetate wash. The organic layers were washed with sodium bicarbonate, 1 M hydrochloric acid, and brine before being dried over sodium sulfate and concentrated on a rotary 33 evaporator. The resulting solid was purified via column chromatography with an eluent of 50-60% ethyl acetate:hexanes to give Compound 16 as a white solid (67% yield). IR (thin-film) 3388, 3204, 2967, 2932, 2361, 2339, 1665 cm-1; 1H NMR (500 MHz, CDCl3) δ 6.18 (s, 1H), 5.34 (s, 1H), 4.09 (d, J = 11.1 Hz, 1H), 2.81 (s, 3H), 2.26 (m, 1H), 1.69 (s, 1H), 1.48 (s, 9H), 0.98 (d, J = 6.5 Hz, 3H), 0.88 (d, J = 6.7 Hz, 3H) ppm; 13C NMR (125 MHz, CDCl3) δ 206.9, 80.4, 64.1, 30.9, 30.2, 28.4, 25.9, 19.8, 18.5 ppm; Tert-butyl (S)-(1-(methoxyamino)-3-methyl-1-oxobutan-2-yl)(methyl)carbamate (26): Triethylamine was added to a reaction vessel, followed by tetrahydrofuran, to make a solution that was cooled to 0 C before addition of Compound 15 (1 equiv.). CDI was then added to the reaction in one portion WITH VENT. After warming to room temperature and stirring for 1 hour, nmethylhydroxylamine hydrochloride was added to the reaction WITH VENT, and the reaction was stirred until starting material was consumed. Saturated ammonium chloride was added to the reaction, followed by an ethyl acetate wash. The organic layers were washed with sodium bicarbonate, 1 M hydrochloric acid, and brine before being dried over sodium sulfate and concentrated on a rotary evaporator. The resulting solid was purified via column chromatography with an eluent of 50-60% ethyl acetate:hexanes. This process gave Compound 26 as a slightly pink, extremely viscous solid (75% yield). IR (thin-film) 3222, 2967, 2935, 2361, 2339, 1661 cm-1; 1H NMR (500 MHz, CDCl3) δ 9.11 (s, 1H), 3.85 (d, J = 11.3 Hz, 1H), 3.75 (s, 3H), 2.84 (s, 3H), 2.39-2.24 (m, 1H), 1.46 (s, 9H), 0.95 (d, 34 undesired O cyclization product. However, the structure elucidation work was done by Matthew Nelli and is therefore not included in these experimentals. 38 References 1. Germe, T., et. al.; (2018). A new class of antibacterials, the imidazopyrazinones, reveal 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. structural transitions involved in DNA gyrase poisoning and mechanisms of resistance. Nucleic Acids Research, 46(8), 4325-4325. Payne, D. J., et. al.; (2015). Time for a change: Addressing R&D and commercialization challenges for antibacterials. Philosophical Transactions of the Royal Society B: Biological Sciences, 370(1670), 20140086-20140086. Hu, C. (2018, July 21). Pharmaceutical companies are backing away from a growing threat that could kill 10 million people a year by 2050. Kapoor, G., Saigal, S., & Elongavan, A. (2017). Action and resistance mechanisms of antibiotics: A guide for clinicians. Journal of Anaesthesiology, Clinical Pharmacology, 33(3), 300-305. Wright, P. M., Seiple, I. B., & Myers, A. G. (2014). The Evolving Role of Chemical Synthesis in Antibacterial Drug Discovery. Angewandte Chemie International Edition, 53(34), 8840-8869. Noller, H. F. (1991). Ribosomal RNA and Translation. Annu. Rev. Biochem., 60, 191227. Monro, R. E., Cerna, J., & Marcker, K. A. (1968). Ribosome-catalyzed peptidyl transfer: substrate specificity at the P-site. Proceedings of the National Academy of Sciences, 61(3), 1042–1049. Spizek, J., & Rezanka, T. (2017). Lincosamides: Chemical structure, biosynthesis, mechanism of action, resistance, and applications. Biochemical Pharmacology, 133, 20– 28. Mast, Y., & Wohlleben, W. (2014). Streptogramins – Two are better than one! International Journal of Medical Microbiology, 304(1), 44-50. Yamaguchi, H., & Tanaka, N. (1966). Inhibition of Protein Synthesis by Blasticidin SI. Studies with Cell-free Systems from Bacterial and Mammalian Cells. The Journal of Biochemistry, 60(6), 632–642. Guoquing, N., and Huarong, T. (2015). Trends in Microbiology, 23 (2), 110-19 Ben-Shem, A., Jenner, L., Yusupova, G., & Yusupov, M. (2010). Crystal Structure of the Eukaryotic Ribosome. Science, 330(6008), 1203-1209. Serrano, C. M., & Looper, R. E. (2011). Synthesis of Cytimidine through a One-Pot Copper-Mediated Amidation Cascade. Organic Letters, 13(19), 5000-5003. Katayama, N., Fukusumi, S., Funabashi, Y., Iwahi, T., & Ono, H. (1993). TAN-1057 AD, new antibiotics with potent antibacterial activity against methicillin-resistant Staphylococcus aureus. Taxonomy, fermentation and biological activity. The Journal of Antibiotics, 46(4), 606-613. Funabashi, Y., Tsubotani, S., Koyama, K., Katayama, N., & Harada, S. (1993). A new anti-MRSA dipeptide, TAN-1057 A. Tetrahedron, 49(1), 13-28. Sokolov, V. V., Kozhushkov, S. I., Nikolskava, S., Belov, V. N., Es-Sayed, M., & Meijere, A. D. (1998). Total Synthesis of TAN-1057 A/B, a New Dipeptide Antibiotic from Flexibacter sp. PK-74. European Journal of Organic Chemistry, 1998(5), 777-783. 39 17. Svidritskiy, E., Ling, C., Ermolenko, D. N., & Korostelev, A. A. (2013). Blasticidin S inhibits translation by trapping deformed tRNA on the ribosome. Proceedings of the National Academy of Sciences, 110(30), 12283-12288. 18. Ma, D., & Xia, C. (2001). CuI-Catalyzed Coupling Reaction of β-Amino Acids or Esters with Aryl Halides at Temperature Lower Than That Employed in the Normal Ullmann Reaction. Facile Synthesis of SB-214857. Organic Letters, 3(16), 2583-2586. 19. Girard, A. -L., Enomoto, T., Yokouchi, S., Tsukano, C., & Takemoto, Y. (2011). Control of 6-Exo and 7-Endo Cyclizations of Alkynylamides using Platinum and Bismuth Catalysts. Chemistry – An Asian Journal, 6(6), 1321-1324. 20. Wilson, D. N. (2013). Ribosome-targeting antibiotics and mechanisms of bacterial resistance. Nature Reviews Microbiology, 12(1), 35-48. 40