Download PDF
Synlett 2024; 35(12): 1345-1360
DOI: 10.1055/a-2145-3647
account

Adventures in Total Synthesis – The Next Chapter

David Yu-Kai Chen∗

This work was supported by the National Research Foundation of Korea (NRF) via a grant funded by the Korean Government, Ministry of Science, ICT and Future Planning (MSIP) (2021R1A5A6002803) (Center for New Directions in Organic Synthesis) and Novartis.
› Further Information
 
 PDF Download Permissions and Reprints All articles of this category


Dedicated to Professor Ian Paterson and Professor K. C. Nicolaou

Abstract

This account summarizes the author’s endeavors in target-oriented synthesis at Seoul National University since 2011. A collection of the most celebrated molecules in total synthesis are revisited and the author’s solutions to these historical challenges are presented. In particular, the unique perception of their molecular frameworks and unprecedented bond-forming sequences form the basis of the newly developed strategies. Together with a ‘personal touch’ on these selected stories, the author hopes that this account will offer new insights and fresh perspectives for all levels of enthusiasts of target-oriented total synthesis.

1 Introduction

2.1 Synthesis of Strychnine

2.2 Synthesis of Actinophyllic Acid

2.3 Synthesis of Dendrobine

2.4 Synthesis of Communesin

2.5 Synthesis of Morphinans

2.6 Synthesis of Reserpine

2.7 Synthesis of Quinine and Quinidine

2.8 Synthesis of Haouamine

2.9 Future Work

3 Summary and Outlook

4 Abbreviations


#

Key words

molecular symmetry - natural products - synthetic strategy - target-oriented synthesis - total synthesis

Biographical Sketch

Zoom Image

David Chen was born in Taipei, Taiwan in 1976. He received his B.Sc. (Honors) from the University of Auckland (1997) and his Ph.D. under the supervision of Professor Ian Paterson at Cambridge University (1998–2001). After postdoctoral training under the direction of Professor K. C. Nicolaou at the Scripps Research Institute (2002–2003), he joined the Merck Research Laboratory at Rahway-New Jersey as a Senior Research Chemist (2003–2005). In March 2005, he was appointed as the Principal Investigator of the Chemical Synthesis Laboratory (CSL)@Biopolis under the Agency for Science, Technology and Research (A*STAR), Singapore. During his tenure in Singapore, he also held an Adjunct Associate Professorship at the division of Chemistry and Biological Chemistry (CBC), Nanyang Technological University, Singapore. In 2011, he relocated to Seoul, South Korea, where he was appointed as a professor of organic chemistry at Seoul National University.

1

Introduction

In 2011, I was honored to accept an invitation from the then Editor-in-Chief of Synlett, Professor Peter Vollhardt, to contribute a Synlett Account of my research work. Coincidentally, this was also at a transition point of my career when I relocated from A*Star (Singapore) to Seoul National University (SNU, South Korea). Twelve years later, as I was preparing for my sabbatical leave and much to my surprise, I received a follow-up invitation from Professor Vollhardt and again I felt honored and obliged to accept his offer. I was particularly drawn to the format of the account suggested by Professor Vollhardt on both occasions, which provided an opportunity to speak about my work from both a professional and a personal angle. In this context, let me apologize in advance to the readers of this account for the occasional digressions from scientific discussions as I ramble on about my personal experiences and viewpoints. Indeed, with more than 20 years of uninterrupted involvement in target-oriented synthesis and an ever-so-strong passion for this field, there is a lot that can be said from the mind and the heart, respectively. Looking back to when I first started, and like many investigators in target-oriented synthesis, there was a hunger to pursue the ‘hottest’ molecules, especially those not yet conquered by chemical synthesis. Whether this pursuit was to seek personal fulfillment or for professional reasons, I thoroughly enjoyed the journey, not to mention the fair share of trials and tribulations. The Synlett Account I contributed in 2011 provided an overview of my ‘6 years pursuit of molecules’ (Figure [1]),[1] and I am grateful for the generous support I received together with the heartfelt advice from Professor K. C. Nicolaou during this early stage of my independent career.

Zoom Image
Figure 1 Selected molecules investigated by the author from 2005 to 2011 at A*STAR, Singapore

While many in the field advocate that the pursuit of synthesis should not be treated as a race, I was never against the idea of ‘healthy competition’ as it can be a source of motivation for some. Clearly, opinions on this matter vary depending on whether you are an active competitor or merely a spectator, and the precise definition of ‘healthy competition’ is also subject to debate. As I embarked on a new scientific journey in South Korea back in 2011, it was clear to me that adjustments had to be made if I wished to remain on the competitor side of the game. At the same time, the word ‘competition’ gradually took on a different meaning under this new environment that motivated me to take a deeper look at target-oriented synthesis, particularly in terms of target selection, strategic design and experimental execution. I sincerely believe that our endeavors in target-oriented synthesis since 2011 have carried a different ‘flavor’ and have contributed something new to the community (Figure [2]), but I will leave it for the readers of this account to make their own decisions.

Zoom Image
Figure 2 Selected molecules investigated by the author from 2011 to 2022 at Seoul National University, South Korea

Without further ado, let me invite you to take a trip with me down my 12-year memory lane at SNU. Whether the following passages will bring you new perspectives in chemical synthesis or uncover similarities between your scientific journey and mine, I hope you enjoy reading this account as much as I have enjoyed sharing it with you.

2.1

Synthesis of Strychnine

The 22nd International Symposium in Organic Chemistry held at Cambridge University was one of the most memorable scientific meetings of my career. It was my first time giving a lecture at my alma mater, and more significantly, an opportunity to speak about my independent work in front of my graduate adviser, Professor Ian Paterson. It was also at the same symposium that Professor David MacMillan disclosed his group’s organocatalytic cascade-based synthesis of strychnine for the first time and coined the term ‘collective synthesis’.[2] Together with the ‘six-step synthesis of strychnine’ reported by Professor Chris Vanderwal earlier in the same year,[3] these newly completed strychnine syntheses not only brought this historical molecule back under the spotlight, but also rekindled my own synthetic interests. At the same time, I felt fairly confident of my knowledge on this topic having just taught strychnine to my graduate class at SNU.

I believe many of us have noticed the increasingly popular 3-dimensional drawings of chemical structures instead of the ‘planar’ representations with wedges and hashes. While these 3-dimensional renditions offer valuable insights for the understanding of chemical reactivity and selectivity, it is my impression that the motivation is primarily for esthetic appeal and adding a ‘wow factor’. In the case of strychnine with its polycyclic framework and array of stereocenters, it turns out that this daunting collection of structural challenges could be logically reduced to a single C7 quaternary center which dictates the relative stereochemical arrangement of the entire molecule. To this end, we envisaged that the pyrrolidine-based enal 10, which possesses both electrophilic and nucleophilic centers, may undergo sequential alkylation and arylation to accomplish this objective.

Zoom Image
Scheme 1 Synthesis of strychnine (1)

As outlined in Scheme [1], our synthesis commenced with the reaction between enal 10 and silyl ketene acetal 11.[4] I vividly remember personally initiating this investigation, and even with my outdated experimental skills and shaky hands, this reaction was found to proceed smoothly in the presence of a variety of Lewis acid promoters. Although we initially anticipated the generation of aldehyde 13 instead of TBS enol ether 12, this unexpected outcome not only facilitated the construction of the C7 quaternary center but also provided the mechanistic basis for an asymmetric entry to 12. We next turned our attention to the much anticipated C7 quaternary center formation using either TBS enol ether 12 or aldehyde 13, and after surveying several strategies we eventually found that TBS enol ether 12 underwent reaction with iodonium fluoride 14 to afford gram quantities of aldehyde 15 as a single stereoisomer.[5] While the successful preparation of aldehyde 15 represented a significant milestone in our strychnine campaign, reduction of its nitro functionality was problematic and we ultimately resorted to the use of titanium(III) chloride to afford the targeted spiro-imine 16. Advancing spiro-imine 16 to strychnine largely leveraged on chemistry already reported for related systems[6] and was further streamlined by application of one-pot reactions, non-aqueous workups, and bypass chromatographic purifications. With the completion of a ten-step total synthesis of strychnine, the remaining task was to develop an asymmetric variant of the reaction between enal 10 and silyl ketene acetal 11. Our early investigations in this direction were met with constant disappointments as a variety of Lewis acid mediated conditions in the presence of privileged chiral ligands failed to deliver any level of asymmetric induction. However, this observation together with the formation of TBS enol ether 12 instead of aldehyde 13 led to the suspicion that a ‘silyl-activation’ instead of ‘Lewis acid activation’ pathway may be operative, and that the asymmetric counterion-directed catalysis (ACDC) pioneered by Professor Benjamin List might provide a unique solution.[7] Much to our delight, this hypothesis was confirmed with chiral disulfonimide catalyst 22 to afford optically active 12 with modest enantioselectivity, representing one of the first applications of ACDC in target-oriented synthesis. Unfortunately, due to requiring elaborate catalyst preparation procedures and limited commercial availability, we were not able to perform a more extensive optimization of the reaction conditions (which also became the main criticism during peer review of this work).


# 2.2

Synthesis of Actinophyllic Acid

During my early days at SNU, I was actively exploring several options to expand my newly established research team. One such option was the exchange student scheme under the Campus Asia Program, which also provided an opportunity for me to reconnect with my friend Professor Hidetoshi Tokuyama at Tohoku University. Hideotoshi and I first became acquainted during our independent research on the total synthesis of haplophytine,[8] and ever since I always had great respect for the skill-set of his co-workers. Thanks to support from Hideotoshi, his graduate student Mr. Yoshii Yu arrived in my laboratory in the winter of 2012 and commenced on our synthetic studies towards actinophyllic acid.

Zoom Image
Scheme 2 Synthesis of actinophyllic acid (2)

Prior to our involvement the only synthesis of actinophyllic acid had been reported by Professor Larry Overman’s laboratory, and featured their signature aza-Cope/Mannich rearrangement.[9] We knew from the outset that it would be difficult to rival Professor Overman’s elegant synthesis in terms of step-count or new reaction development, therefore our entry point was based on a strategically unique perception of the actinophyllic acid polycyclic framework. As outlined in Scheme [2], the [5.4.1]-azabicycle embedded within the actinophyllic acid core structure formed the basis of our synthetic investigation, and its synthesis was readily accomplished from trienone 23 through oxidative cleavage of the isolated alkene followed by sequential reductive aminations.[10] With diene 26 in hand we examined several alkene or diene specific transformations, among which the 1,3-dipolar cycloaddition with bromonitrile oxide derived from dibromo-oxime 27 delivered the most suitable intermediate for further synthetic manipulations.[11] Strategically, it is noteworthy that this transformation served a complexity-generating role that converted the symmetrical and easily accessible diene 26 into isoxazoline 28, and represents a possible entry point for accessing optically active intermediates. Advancing the synthesis next required an arylation as prerequisite for the late-stage indole synthesis, and we were pleased to accomplish this objective with the concurrent introduction of the adjacent pyrrolidine through a palladium-catalyzed aminoarylation reaction pioneered by Professor John Wolfe.[12] Unfortunately, as our actinophyllic acid campaign reached a climax with the successful preparation of aryl isoxazole 31, Yoshii’s one-year exchange program also came to an end and he had to return to Japan to continue his doctoral studies. After some discussions with Hideotoshi and Yoshii, we were pleased to have Yoshii continue working on this project, and much credited to Yoshii’s skills and commitments the actinophyllic acid project quickly resumed to full thrust at Tohoku University (in fact, Yoshii was concurrently handling two total synthesis projects at the time!). Reductive transformation of aryl isoxazole 32 to indole 33 required some optimization, and finally, the completion of actinophyllic acid was realized under the alkylation conditions reported by Professor Overman and co-workers.[9] I wish to highlight that during the course of our investigations Professor Stephen Martin[13] and Professor Ohyun Kwon[14] also independently reported their elegant syntheses of actinophyllic acid. Therefore I am grateful to the expert reviewers and the synthetic community for recognizing the scientific merit of our synthetic approach.


# 2.3

Synthesis of Dendrobine

Throughout my research career I have always had a strong ‘French connection’, and have benefitted immensely from their scientific contributions and valuable friendships. Therefore, I was more than delighted to accept an invitation from my former co-worker Dr. Philippe Peixoto to deliver a lecture at the 8th French Symposium on Total Synthesis, which would also be my first visit to France. Furthermore, it was a fitting occasion to share our newly completed dendrobine synthesis that had involved a French co-worker and was inspired by Philippe’s work with me back in Singapore.

Zoom Image
Scheme 3 Synthesis of dendrobine (3)
Zoom Image
Scheme 4 Synthesis of communesin (4)

In 2011 we reported a second-generation synthesis of echinopines that leveraged on the success of a one-pot ene-yne cycloisomerization/intramolecular Diels–Alder cascade (Scheme [3a]).[15] As an extension of this work, we became interested in other conjugated diene-specific transformations,[16] among which the Rh-catalyzed synthesis of substituted pyrrolidines, reported by Professor Zhi-Xiang Yu, and its potential application in the preparation of the dendrobine core structure was particularly appealing (Scheme [3b]).[17] As shown in Scheme [3c], our investigation began with the preparation of dienyne 40 through an organocuprate-based coupling reaction between dibromide 34 and alkynyl Grignard reagent 35 followed by reductive amination with enal 39.[18] Pd-catalyzed cycloisomerization of dienyne 40 under the conditions originally reported by Professor Barry Trost delivered triene 41 uneventfully,[19] but we were surprised by the sensitivity of this compound, especially upon prolonged storage. Much to our delight, the Rh-catalyzed pyrrolidine synthesis developed by Professor Zhi-Xiang Yu was feasible on triene 41, even on our first attempt, and afforded bicyclic diene 42 without any fine-tuning of the originally reported conditions. The realization of the Pd-catalyzed cycloisomerization and the Rh-catalyzed allylic CH-functionalization motivated us to develop a more streamlined process, especially considering the stability issue associated with triene 41. Along this line of thought we were able to merge the palladium and rhodium catalytic systems as a one-pot process to generate bicyclic diene 42 directly from dienyne 40, in doing so completely bypassing the isolation of sensitive triene 41. Next, advancing bicycle 42 involved the functionalization of its two isolated alkenes and setting the stage for the 6-membered ring formation. To this end, among the oxidative transformations examined, a double hydroboration–oxidation sequence proved most productive and delivered keto aldehyde 43 poised for an aldol-type cyclization. Although easily stated, the intramolecular aldol reaction of keto aldehyde 43 remained problematic for quite some time until the identification of the LiOH/ i PrOH conditions,[20] and enone 44 was obtained with an unexpected but fortuitous detosylation. Taking advantage of imine 44, reductive methylation smoothly delivered the reported Kende intermediate (45),[21] and in doing so completed a formal synthesis of dendrobine (3). Finally, an enantioselective organocuprate-based coupling between dibromide 34 and alkynyl Grignard reagent 35 under the conditions developed by Professor Ben Feringa provided an asymmetric entry to our established synthetic pathway.[22]


# 2.4

Synthesis of Communesin

As we continue to celebrate the remarkable achievements in science, sadly, the extraordinary individuals behind these outstanding achievements are also leaving us as each day goes by. One of those extraordinary individuals was the late Professor Carlos Barbas, who passed away in 2014 after a courageous battle against cancer. Although I never had the opportunity to interact with Professor Barbas during my postdoctoral work at The Scripps Research Institute, his science definitely had a long-lasting impact on my research, as featured in our communesin synthesis.

The communesin structure contains several thorny challenges from the synthetic standpoint, most notably the C7–C8 all-carbon quaternary centers embedded within the complex polycyclic framework. In this context, we were encouraged by the numerous reports of asymmetric organocatalytic syntheses of oxindoles bearing quaternary stereocenter(s), many of which originated from the late Professor Barbas’s laboratory.[23] As shown in Scheme [4], our synthesis began with the reaction between oxindole aldehyde 46 and oxindole azide 48 in the presence of silyl-prolinol catalyst 49.[24] The enantioselectivity of this process was extensively optimized, and the success of this reaction represented the first example of an organocatalytic coupling reaction that afforded the communesin relative stereochemistry. Next, a skeletal rearrangement was carried out to transform the bis-oxindole connectivity into the communesin partial ring framework, as depicted by the conversion from bis-oxindole 51 into pentacyclic amidine 52. Despite being a highly efficient transformation, there was a great deal of anxiety due to a number of possible product structures not easily distinguishable by routine NMR analysis and our bold decision to continue the synthesis without full structural assurance. Pressing on with the speculated structure 52, our attention next turned to a directed aryl CH-functionalization reaction and the preparation of the oxalamide substrate 55 suggested by our model studies.[25] It was from this study that we isolated the unexpected compound 55a during chromatographic purification of oxalamide 55 using acetone as the eluent, and this serendipitous finding also provided a timely resolution to all structural ambiguities through an X-ray crystallographic analysis of enamide 55a. The CH-functionalization-based coupling between oxalamide 55 and methyl acrylate proceeded beyond all our expectations, and this remarkable transformation constitutes one of the most efficient and complex examples in target-oriented synthesis to date. Guided by the previously reported communesin synthesis, enoate 56 was further elaborated into the tertiary alcohol derivative 64 for the installation of the remaining ring systems and the completion of communesin.[26] Although we were able to accomplish these objectives and in doing so brought our communesin campaign to a conclusion, the lengthy sequence of late-stage functional group transformations was undoubtedly a shortcoming of our synthesis and did not escape the watchful eyes of the expert reviewers.

Zoom Image
Scheme 5 Synthesis of morphinans – core synthesis phase

# 2.5

Synthesis of Morphinans

Academic teaching is one of the most fulfilling aspects of being a professor, and sometimes even a seemingly trivial undergraduate course can be a source of inspiration for new projects. In this context, it was during my course on stereoinduction in target-oriented synthesis that two elegant and unconventional examples of chirality transfer caught my attention, namely the synthesis of longithorone by Professor Mathew Shair[27] and the preparation of rhazinilam by Professor Armen Zakarian.[28] The application of ‘point-to-axial’ (longithorone, Scheme [5a]) and ‘axial-to-point’ (rhazinilam, Scheme [5b]) chirality transfer in total synthesis was relatively uncommon at the time, and to step up the challenge, we pondered whether a series of these unconventional forms of stereoinduction could be logically sequenced and applied in target-oriented synthesis.

As shown in Scheme [5c], our synthesis commenced with the preparation of biaryl systems 69/69a followed by an investigation of their atropisomeric properties.[29] Our initial idea was to introduce ‘point’ stereochemistry through nucleophilic addition to biaryl aldehyde 68 (prepared via an unconventional CH-activation process),[30] and this newly formed ‘point’ stereochemistry would induce an ‘atrope’ stereochemical property about the biaryl axis. To this end, allyl Grignard addition to aldehyde 68 smoothly delivered two readily separable stereoisomers 69 and 69a, and their silylated derivatives 70 and 70a proved much more configurationally stable about the biaryl axis under thermal conditions. Next, as part our design, we planned to capitalize on the newly formed atrope stereochemical property about the biaryl axes in 70 and 70a to induce new ‘point’ stereocenters through an intramolecular Diels–Alder reaction of the intermediate masked o-quinones 71 and 71a. Gratifyingly, treatment of phenols 70 and 70a under hypervalent iodine conditions smoothly generated the masked o-quinones 71 and 71a, which underwent intramolecular Diels–Alder reactions engaging the pendent alkene to afford tetracyclic intermediates 72 and 72a. We also performed several control studies to validate that the stereocontrol during the intramolecular Diels–Alder process was entirely governed by the ‘atrope’ stereochemical property of the biaryl axis and not the benzylic-OTBS ‘point’ stereocenter. Tetracycles 72 and 72a contained the characteristic phenanthrene backbone and the all-carbon quaternary center found in the morphinan family of natural products,[31] and hence form the basis for an application in target-oriented synthesis. This ‘functionalization phase’ of our synthetic studies entailed three key events, namely (i) rupture of the [2.2.2]-bicyclic domain in 73; (ii) formation of the dihydrobenzofuran; and (iii) construction of the morphinan piperidine ‘belt’. Although we were able to achieve these objectives as shown in Scheme [6], I have to confess this work definitely leaves plenty of room for improvement in terms of synthetic efficiency and novelty.

Zoom Image
Scheme 6 Synthesis of morphinans – core functionalization phase

# 2.6

Synthesis of Reserpine

The chemistry department open-day is always a festive event and a great opportunity to interact with the new faces in the department. As part of the open-day activities, each lab gives a short presentation that usually includes a brief historical overview of their respective research fields. It was during this time that I noticed that Professor R. B. Woodward and Professor Gilbert Stork were the only individuals that had succeeded in completing the famous ‘trifecta’ of total syntheses that included strychnine,[32] reserpine,[33] and quinine.[34] This ‘non-scientific’ observation somehow became a motivation for me to embark on the synthesis of reserpine (and quinine, see later), which also coincided with our long-standing interest in desymmetrization-based syntheses of complex molecules.

Zoom Image
Scheme 7 Synthesis of reserpine (6)

A common feature of past reserpine syntheses is the thoughtful preparation of a highly substituted and stereochemically defined reserpine E-ring precursor prior to assembling the 6-methoxytryptamine domain.[35] In this context, we pondered if a greatly simplified and symmetrical cyclohexyl system with no apparent stereochemical information could be utilized and strategically transformed into a fully decorated reserpine E-ring. As outlined in Scheme [7a], our synthesis commenced with the preparation of symmetrical [2.2.1]-bicyclic enol ether 88 and its coupling with 6-methoxytryptamine (89) through a Pictet–Spengler reaction.[36] It is worth-noting that this coupling reaction renders the two olefinic carbons in 90 diastereotopic due to the newly created C3 stereocenter, and differentiation of these diastereotopic carbon atoms represents an example of a process generally referred to as ‘local desymmetrization’.[37] Indeed, upon oxidative cleavage of alkene 92 and hydrogenolytic removal of the benzyl-carbamate-protected aliphatic amine, conformational bias resulted in the selective formation of imine 96, which underwent further hydrogenation to afford pentacyclic aldehyde 97 as a single isomer. Although we initially hoped this ‘local desymmetrization’ process would directly afford the desired anti stereochemical arrangement possessed by reserpine, syn-isomer 99 was readily isomerized and recycled to afford an ample quantity of the anti-isomer 3-epi-99. Advancing pentacyclic intermediate 3-epi-99 next called for a stereoselective and contra-steric hydroxylation, which was realized with the l-proline/nitrosobenzene combination to afford α-hydroxy ketone 101 as a single isomer.[38] The seemingly trivial methylation of α-hydroxy ketone 101 was extensively studied and ultimately achieved with a concomitant and stereoselective ketone reduction through an unconventional thiomethylation–desulfurization sequence to provide alcohol 103.[39] Finally, introduction of the trimethoxybenzoyl ester followed by silica-gel-mediated carbamate removal completed the total synthesis of reserpine (6). An asymmetric entry to our newly developed synthetic pathway was made possible through the venerable Noyori asymmetric transfer hydrogenation of imine 105 (Scheme [7b]),[40] and as part of our continued interest in late-stage CH-functionalization, compound 106 (prepared from unsubstituted tryptamine following the same reaction sequence instead of the considerably more costly 6-methoxytryptamine (89)) was found to undergo CH-borylation followed by methanolysis to afford Boc-reserpine 107 (Scheme [7c]).[41]


# 2.7

Synthesis of Quinine and Quinidine

Following our work on strychnine and reserpine, and continuing our quest to complete the Woodward–Stork ‘trifecta’, we next turned our attention to quinine. In this context, recognizing the structural relationship between quinine and quinidine, we felt that a local-desymmetrization-based strategy may again provide a unique solution to this historical synthetic problem.

Zoom Image
Scheme 8 Synthesis of quinine (7) and quinidine (8)
Zoom Image
Scheme 9 Synthesis of haouamine (9)

As shown in Scheme [8], our synthesis commenced with a site-selective oxidative cleavage of norbornadiene derivative 109 followed by reductive amination of the resulting dialdehyde 110 with p-methoxybenzylamine to afford bicyclic amine 111.[42] Conversion of bicyclic amine 111 into Weinreb amide 113 and its reaction with lithiated quinoline derivative 114 provided ketone 115, an advanced intermediate containing the entire carbon backbone of the target molecules. Analogous to our reserpine studies, reduction of ketone 115 generated a hydroxy-bearing stereocenter, and in doing so rendered the two olefinic carbons in 116 diastereotopic. From common intermediate 116, two strategically contrasting pathways were realized, with each involving a local-desymmetrization process either before or after quinuclidine formation. In our first approach, a two-stage oxidative transformation of alcohol 116 afforded lactone 118 through the intermediacy of hemiacetal 117, and the selective formation of lactone 118 (instead of 118a) implied that the two olefinic carbons in 116 had been successfully differentiated. Functional group transformation of lactone 118 afforded mesylate 120, which underwent cyclization to furnish trisubstituted quinuclidine 121 as a common intermediate to access either quinine (7) or quinidine (8) through selective carbon–carbon bond cleavage (deformylation).[43] On the other hand, local desymmetrization was also realized after quinuclidine formation as illustrated by divinyl intermediate 126 and its conversion into iodide 127 through an intramolecular iodoetherification. Advancing iodide 127 afforded quinuclidine intermediate 128 analogous to trisubstituted quinuclidine 121, which underwent a similar carbon–carbon bond cleavage (deformylation) to provide an alternative synthetic entry to quinidine (8). Finally, we dedicated this work in memory of the life and legacy of Professor Gilbert Stork who passed away in late 2017.


# 2.8

Synthesis of Haouamine

Group meetings present a wonderful learning platform for both the student and the advisor, and can also serve as the birthplace of exciting new projects. Several years ago during a literature review session on the synthesis of haouamine, a student presenter concluded that the biaryl linkage within the highly strained cyclophane cannot be constructed as the macrocyclization event. While contemplating this synthetic conundrum, I recalled making a wager to find a solution for this elusive bond formation and hence our studies towards haouamine were initiated. As we shall see, although we have not yet answered this challenge, several findings along our investigative pathway have proved equally valuable compared to our initial objective.

Our synthesis of haouamine commenced with the preparation of the quaternary-center-containing amino-alcohol 141 (Scheme [9]).[44] Inspired by the protocol originally developed by Professor Jianbo Wang,[45] a Rh-catalyzed diazo-insertion reaction between benzocyclobutanol 131 and diazoester 132 smoothly afforded the quaternary-center-containing methyl ester 133. In contrast to the substrate scope originally reported by Professor Wang, the benzyloxy substituent was strategically chosen to permit the conversion into indene 136 through the intermediacy of keto-ester 135. Introduction of the nitrogen atom of haouamine was achieved through an intramolecular aziridination reaction of sulfamate 138,[46] and the obtained aziridine 139 was exhaustively reduced to afford the targeted amino-alcohol intermediate 141. Forging the tetrahydropyridine domain of haouamine onto amino-alcohol 141 was readily accomplished through a three-step process (amidation, intramolecular aldol condensation, and amide reduction),[47] which also became a generalized reaction sequence applied to the preparation of several potential precursors to study the haouamine cyclophane system (for example, in the preparation of tosylate 151 and boronic ester 155). At this point, we were in the position to put our proposal to the test and examine the macrocyclization at the biaryl linkage of haouamine. Recognizing the severe strain associated with the haouamine cyclophane system,[48] we first examined the feasibility of transition-metal-mediated intramolecular cross-coupling of several cyclohexyl systems containing sp3 hybridized carbon centers as arene precursors (for example, boronic ester 155). Unfortunately, all of our macrocyclization studies in this direction were met with disappointment, not to mention the elaborate chemistry involved in synthesizing some of these macrocyclization precursors.

One of the most valuable lessons I have learnt from managing total synthesis projects is to make timely decisions in the best interest of the morale of my co-workers. As such, we turned to a greatly simplified system and a reported site of macrocyclization just to have a ‘feel’ of the system, as illustrated in the preparation of macrocycle 152. With macrocycle 152 readily synthesized, my co-worker suggested that it may be possible to utilize this model system in the actual haouamine synthesis through a site-selective oxygenation followed by aromatization. I have to confess that I was not too keen on this idea, especially considering the numerous oxidative labile sites in macrocycle 152 (benzylic, olefinic, aromatic, and nitrogen), but after witnessing the failure of my cyclophane formation strategy I was reluctant to refuse my co-worker’s proposal. It was definitely a memorable day when macrocycle 152 was found to undergo a smooth and site-selective oxidation to afford enone 154, the reported precursor to haouamine A (9). Furthermore, additional control experiments suggested that the success of this transformation is most likely attributed to the strain associated with macrocycle 152, which accelerated the desired oxidation process over other potential side reactions. After this work had been published, I received a kind message from my former co-worker (now Professor Doron Pappo) praising our courageous design and execution of this late-stage oxidation – little does Doron know that I was as surprised as he was on the success of this ‘designed’ transformation!


# 2.9

Future Work

Before concluding these scientific discussions, I wish to share a few words on the potential future work for each of the aforementioned projects (Figure [3]). In our strychnine synthesis, the demonstration of asymmetric counterion-directed catalysis (ACDC) in target-oriented synthesis represents an important advance, but the moderate enantioselectivity demands a systematic investigation of disulfonimide catalysts (Figure [3a]). An asymmetric version of our actinophyllic acid synthesis based on the nitrile oxide 1,3-dipolar cycloaddition appears challenging, however, the pioneering allylic substitution chemistry developed by Professor Barry Trost may provide a unique solution (Figure [3b]).[49] Although the Pd/Rh-catalyzed cycloisomerization/CH-activation cascade represented the highlight of our dendrobine synthesis, the late-stage intramolecular aldol condensation of keto aldehyde 43 inspired a more ambitious organocatalysis cascade (Figure [3c]). In our communesin synthesis, the lengthy synthetic sequence that transformed enoate 56 into tertiary alcohol 64 could be significantly improved with an alternative CH-functionalization coupling partner (Figure [3d]). On the same note of step-economy, advancing functionalized tetracycle 72 into the morphinan family of natural products will require significant redesign together with the incorporation of modern synthetic methodologies (Figure [3e]). The successful implementation of local desymmetrization in our reserpine synthesis could be made more attractive by directly accessing the desired anti stereochemical arrangement, and our preliminary studies suggested that engaging the indole nitrogen instead of the aliphatic nitrogen may provide a solution (Figure [3f]). The local desymmetrization strategy featured in the quinine and quinidine synthesis could be made more attractive by starting from a symmetrical quinuclidine 165 followed by a directed and desymmetrizing CH-functionalization to introduce the quinoline domain (Figure [3g]). Finally, we remain optimistic with our original macrocyclization strategy through thoughtfully designed non-aryl precursors, alternative methods of carbon–carbon bond formation, or a more ambitious aromatic ring-distortion approach (Figure [3h]).

Zoom Image
Figure 3 Proposed follow-up studies for: (a) strychnine, (b) actinophyllic acid, (c) dendrobine, (d) communesin, (e) morphinans, (f) reserpine, (g) quinine and quinidine, and (h) haouamine

#
# 3

Summary and Outlook

Revisiting these projects a decade later has certainly brought back a great deal of nostalgia together with fresh new perspectives. First and foremost, you may have noticed the non-scientific entry point to each project together with a personal aside regarding these research endeavors. At the same time, whether it was from years of experience, sheer luck, or our unimaginative approach, we noticed the time invested to realize the key idea behind each project was usually significantly less compared to completing the target molecule itself. In fact, this observation seems not uncommon based on literature surveys and private communications amongst peers, being often independent of the structural class of the target molecule. Therefore, it raises an interesting question on the scientific merit and return-of-investment of the ‘decoration phase’ of the project once the key idea has been realized (besides for structural validation, biological investigations, etc.). Personally, completing the target molecule in its entirety brings a sense of closure, and more importantly, continues my admiration for the sophistication of chemical synthesis in imitating Nature’s wonders.

A few years back during an e-mail exchange with my graduate advisor Professor Ian Paterson, I was asked if I had been using Professor Nicolaou’s ‘Classics in Total Synthesis’ for my target selection. This question resonates with my long-standing belief that the creative element behind target-oriented synthesis is just as rich and vibrant for previously synthesized molecules. At the same time, I hope target-oriented synthesis will enjoy its freedom to the fullest in the years to come and is not bound by certain expectations, quantitative benchmarks, and the pressure to find applications. As was famously said by Sir Robert Robinson, “It is in the course of attack of the most difficult problems, without consideration of eventual applications, that new fundamental knowledge is most certainly garnered.” However, speaking from personal experiences, I am quite well aware that these personal viewpoints may not be welcomed by some members of the community:

“Currently, synthesis itself is still important; however, it is not the biggest interest anymore especially for molecules already made before. By this reviewer’s understanding, XX synthesis paper should be one of the following: 1. the first synthesis of complex molecule; 2. divergent synthesis; 3. molecules with intriguing bioactivity; 4. development/demonstration of new methodology.” An excerpt taken from a past peer-review;[50] XX = journal name.

As I was drafting my research proposals for new projects to be initiated after my sabbatical leave, I could hardly contain my excitement at sharing these ideas with my co-workers and the rest of the synthetic community. Indeed, all these projects are still based on previously synthesized molecules, and undoubtedly, there will be a fair share of scientific and non-scientific challenges. No matter what lies ahead, I hope each one of us continues to enjoy research to the fullest, pursues our own dreams, and finds fulfillment in life and beyond. Let me end this account with a slightly modified quote from Mark Twain: “The two most important days of a scientist’s life are the day you started doing research and the day you find out why.”


# 4

Abbreviations

Ac = acetyl

Ac2O = acetic anhydride

AcOH = acetic acid

BAIB = (diacetoxyiodo)benzene

Bn = benzyl

Boc = tert-butoxycarbonyl

Boc2O = di-tert-butyl dicarbonate

B2Pin2 = bis(pinacolato)diboron

Bz = benzoyl

Cbz = carboxybenzyl

COD = 1,5-cyclooctadiene

COE = cyclooctene

DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene

DCE = 1,2-dichloroethane

DDQ = 2,3-dichloro-5,6-dicyano-1,4-benzoquinone

Dibal-H = diisobutylaluminum hydride

DMAP = N,N-(dimethylamino)pyridine

DMP = Dess–Martin periodinane

DMSO = dimethylsulfoxide

dppp = 1,3-bis(diphenylphosphino)propane

EDC = N-(3-dimethylaminopropyl)-N-ethylcarbodiimide

FG = functional group

HBPin = pinacolborane

KHMDS = potassium bis(trimethylsilyl)amide

LDA = lithium diisoproprylamide

l-Selectride = lithium tri-sec-butylborohydride

MeSO2PT = 1-(methylsulfonyl)-5-phenyl-1H-tetrazole

MOM = methoxymethyl

Ms = methanesulfonyl

MsCl = methanesulfonyl chloride

NBS = N-bromosuccinimide

NIS = N-iodosuccinimide

NMO = N-methylmorpholine N-oxide

OAc = acetate

OTf = trifluoromethanesulfonate

PCC = pyridinium chlorochromate

Ph = phenyl

PIFA = [bis(trifluoroacetoxy)iodo]benzene

PivOH = 2,2-dimethylpropionic acid

PPh3 = triphenylphosphine

PMB = para-methoxybenzyl

PPTs = pyridinium p-toluenesulfonate

iPr = isopropyl

(IPr)AuCl = 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene gold(I) chloride

Rh2(Oct)4 = rhodium(II) octanoate, dimer

RuCl[(S,S)-TsDPEN](p-cymene) = [N-[(1S,2S)-2-(amino-κN)-1,2-diphenylethyl]-4-methylbenzenesulfonamidato-κN]chloro[(1,2,3,4,5,6-η)-1-methyl-4-(1-methylethyl)benzene]-ruthenium

TBAF = tetra-n-butylammonium fluoride

TBS = tert-butyldimethylsilyl

Teoc = 2-(trimethysilyl)ethyl carbonyl

Tf = trifluoromethansulfonyl

TFA = trifluoroacetic acid

TMDSO = 1,1,3,3-tetramethyldisiloxane

Ts = para-toluenesulfonyl

TsCl = para-toluenesulfonyl chloride

TsOH = para-toluenesulfonic acid

TMS = trimethylsilyl chloride


#
#

Conflict of Interest

The author declares no conflict of interest.

Acknowledgment

I am honored to have worked alongside a very small group of enthusiastic young researchers at SNU, where most of them having had no prior exposure to organic synthesis, and to be a part of their professional and personal development over the years. At the same time, I am grateful to the support and understanding from my colleagues at SNU for me to function as the only foreign faculty member in the chemistry department. Most of all, I would like to express my sincere gratitude to my wife Sowon and my children Jony and Katy for their unconditional love and endless tolerance.


Corresponding Author

David Yu-Kai Chen
Department of Chemistry, Seoul National University
Gwanak-1 Gwanak-ro, Gwanak-gu, Seoul 08826
South Korea   

Publication History

Received: 30 June 2023

Accepted after revision: 31 July 2023

Accepted Manuscript online:
31 July 2023

Article published online:
13 September 2023

© 2023. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany


   Permissions and Reprints All articles of this category
Zoom Image
Zoom Image
Figure 1 Selected molecules investigated by the author from 2005 to 2011 at A*STAR, Singapore
Zoom Image
Figure 2 Selected molecules investigated by the author from 2011 to 2022 at Seoul National University, South Korea
Zoom Image
Scheme 1 Synthesis of strychnine (1)
Zoom Image
Scheme 2 Synthesis of actinophyllic acid (2)
Zoom Image
Scheme 3 Synthesis of dendrobine (3)
Zoom Image
Scheme 4 Synthesis of communesin (4)
Zoom Image
Scheme 5 Synthesis of morphinans – core synthesis phase
Zoom Image
Scheme 6 Synthesis of morphinans – core functionalization phase
Zoom Image
Scheme 7 Synthesis of reserpine (6)
Zoom Image
Scheme 8 Synthesis of quinine (7) and quinidine (8)
Zoom Image
Scheme 9 Synthesis of haouamine (9)
Zoom Image
Figure 3 Proposed follow-up studies for: (a) strychnine, (b) actinophyllic acid, (c) dendrobine, (d) communesin, (e) morphinans, (f) reserpine, (g) quinine and quinidine, and (h) haouamine