Abstract
Amphidinol 3 (AM3) is a super-carbon-chain compound isolated from the dinoflagellate Amphidinium klebsii. Although the absolute configuration of AM3 was determined in 1999 by instrumental analysis in combination with degradation of the natural product, it was a daunting task because of its limited availability from natural sources and presence of around 70% of chiral centers on the acyclic carbon chain. During the course of our synthetic studies of AM3, the originally proposed structure was revised, which was confirmed by the first total synthesis of AM3 in 2020, more than 20 years since its first discovery. A highly convergent strategy via the fragment assembly using Suzuki–Miyaura coupling and Julia–Kocienski olefination led to the successful total synthesis; however, it was not an easy task to assemble large segments by the Suzuki–Miyaura coupling. A number of experiments optimizing the reaction conditions including model systems revealed that the concentration of the aqueous cesium carbonate is crucial for the key step Suzuki–Miyaura coupling to proceed effectively.
You have full access to this open access chapter, Download chapter PDF
Keywords
- Structure revision
- Convergent synthesis
- Cross-metathesis
- Regioselective dihydroxylation
- Suzuki–Miyaura coupling
- Julia–Kocienski olefination
3.1 Introduction
Amphidinol 3 (AM3, 1) is a super-carbon-chain compound produced by the dinoflagellate Amphidinium klebsii (Fig. 3.1) [1]. AM3 is known to induce not only antifungal activity but also hemolysis by increasing membrane permeability. Determination of the stereochemistry was carried out in 1999 [2] by extensive NMR analysis including the JBCA method [3], the modified Mosher method [4], and degradation of the natural product. However, it was a difficult challenge to determine the stereochemistry because of the scarcity from natural sources as well as around 70% of chiral centers located on the flexible carbon chain. HPLC on a chiral column with UV detection was utilized to assign the absolute configuration at C2 as S by comparison of the retention times for the degradation product from 10 μg of AM3 with those for authentic samples. It is possible that the observed peak for the degradation product was that for an artifact. Both the relationship between C38–C39 and C50–C51 were assigned to be threo by the JBCA method, however, these are also ambiguous since some of the observed J values were categorized in the “medium” range, and not the “large” or “small” range. In our synthetic studies on AM3, the stereochemistry at C2, C32–C36, C38, and C51 was corrected. Based on the revised structure, the first total synthesis of AM3 was accomplished in 2020. It took over two decades to confirm the correct stereochemistry after structure determination of AM3 in1999. In this chapter, the long 15-year journey toward the structure revision and total synthesis of AM3 is described along with the difficulties encountered and the solutions to those problems [5,6,7,8,9,10,11,12,13].
Originally proposed structure (1999) of amphidinol 3. The absolute configurations at C2, C32–C36, C38, and C51 were later revised
3.2 Revision of the Absolute Configuration at C2: Synthesis of the C1–C14 Section
Due to the ambiguity of the stereochemistry at C2, we planned to synthesize the four possible diastereomers 2a-2d corresponding to the C1–C14 section and compare their 1H and 13C NMR data with those for AM3 (Fig. 3.2) [5]. For the efficient construction of the E-olefinic moieties, we envisaged extensive utilization of the cross-metathesis reaction. In this strategy, iodoolefin 3 [14] was utilized as the key intermediate in which the iodide moiety served as a masked terminal olefin in the first cross-metathesis with terminal olefin 4 to give 5 (Scheme 3.1). After reductive removal of the iodide moiety, the resulting terminal olefin 6 was subjected to the second cross-metathesis reaction with terminal olefin 7. Removal of the silyl groups afforded 2a. In an analogous manner, 2b-2d were prepared by changing the substrates to ent-3 and ent-7. However, the 1H NMR spectra of 2a-2d were completely identical, because the stereogenic centers are located 5 carbons apart. Differences in 13C NMR chemical shifts between AM3 and those of 2a-2d were also within the range of error, but those of 2b showed the smallest deviation from the natural product.
Synthesis of four possible diastereomers at C2, C6, and C10 corresponding to the C1–C14 section
Reprinted with permission from Ref. [5]. Copyright © 2008, American Chemical Society
Synthesis of the C1–C14 section via cross-metathesis.
Therefore, for confirmation of the stereochemistry at C2, AM3 (ca. 50 μg) was subjected to degradation by cross-metathesis with ethylene to give 8, which was compared with authentic samples 9 and ent-9 by GC–MS analysis using a chiral column, revealing that the stereochemistry at C2 should be revised to be R (Scheme 3.2) [5].
Revision of the absolute configuration at C2 via degradation of natural product by cross-metathesis and comparison with authentic samples
3.3 Revision of the Absolute Configuration at C51: Synthesis of the C43–C67 Section
For confirmation of the ambiguous absolute configuration at C51 by comparison of NMR data for AM3, we planned to synthesize the model compound 10a corresponding to the C43–C67 section and its epimer at C51 (10b) via Julia–Kocienski olefination (Fig. 3.3) [8].
Synthesis of the two possible diastereomers at C51 corresponding to the C43–C67 section via Julia–Kocienski olefination
The B-ring 15 was synthesized from the building block 3 via chemoselective cross-metathesis to give 11, where the iodoolefin moiety was utilized as a protected terminal olefin (Scheme 3.3) [6]. Conversion of the diene 11 to 12 was achieved via chemo- and diastereoselective Sharpless asymmetric dihydroxylation (AD) of the diene 11 in which the iodoolefin moiety remained intact and was utilized for the next Suzuki–Miyaura cross-coupling with pinacol boronate 13. The resulting diene 14 was converted to the B-ring 15 via Katsuki–Sharpless asymmetric epoxidation (AE) and acid catalyzed 6-endo cyclization. In an analogous sequence, its enantiomer ent-15 was synthesized from ent-3 [11].
Reprinted with permission from Ref. [6]. Copyright © 2009, American Chemical Society
Synthesis of the B-ring.
Sulfone 20 was synthesized based on a linchpin strategy via Negishi coupling of iodoolefin 16 and zinc reagent prepared from 17, followed by Migita–Kosugi–Stille coupling with iodoolefin 19 (Scheme 3.4) [10].
Reprinted with permission from Ref. [8]. Copyright © 2013, American Chemical Society
Synthesis of the two possible diastereomers at C51 corresponding to the C43–C67 section.
Coupling partners, aldehydes 21a and 21b corresponding to the diastereomer at C51, were prepared from the common intermediate 15 [8]. The aldehyde 21a was prepared via Sharpless AD of the olefin, whereas 21b was prepared via Katsuki–Sharpless AE, Payne and Pummerer rearrangements (cf. Scheme 3.9). The target compounds 10a and 10b were synthesized from the aldehydes 21a and 21b, respectively, via Julia–Kocienski olefination with the sulfone 20. Comparison of 1H and 13C NMR chemical shifts for 10a and 10b with those for the natural product revealed that deviations of the chemical shifts between 10a and AM3 were larger than those between 10b and AM3, suggesting that the stereochemistry at C51 should be corrected to be S.
3.4 Synthetic Studies of AM3 Based on Structural Revisions in 2008 and 2013
3.4.1 Retrosynthetic Analysis
Based on the revised structure at C2 (2008) and C51 (2013), the total synthesis was envisaged as shown in the retrosynthetic analysis in Scheme 3.5. The new target, AM3 (22) is to be synthesized via Suzuki–Miyaura coupling of 23 and 24 and Julia–Kocienski olefination with 20. Intriguing molecular structure and biological activity attract much attention of synthetic community, and a lot of synthetic studies have been reported [15] by the Cossy [16,17,18,19,20,21,22], Roush [23,24,25], Rychnovsky [26,27,28], Paquette [29,30,31], Crimmins [32], Evans [33], and Yadav [34] groups.
Revised structure of AM3 at C2 (2008) and C51 (2013) and retrosynthetic analysis
3.4.2 Synthesis of the C1–C29 Section
Synthesis of the C1–C20 section 33 is shown in Scheme 3.6 [9]. Coupling of lithium acetylide prepared from alkyne 26 and bis-Weinreb amide 25 and subsequent Noyori asymmetric hydrogen transfer reaction of the resulting ketone afforded 27. Reduction of the alkyne with Pd/C(en) to avoid hydrogenolysis of the PMB group and protection of the secondary alcohol gave Weinreb amide 28, which was coupled with the alkenyllithium prepared from iodoolefin 29 to furnish 30. CBS reduction of the ketone 30 and protection as a TBS ether afforded terminal olefin 31. The olefin was subjected to cross-metathesis with 0.33 eq of ent-7 to minimize homocoupling to afford 32 in 70% yield based on ent-7 with recovery of 31 (75%). Protection of the secondary alcohol, removal of the PMB group, and Parikh–Doering oxidation of the resulting primary alcohol gave aldehyde 33 (Dess–Martin oxidation gave 34% of 33 with unidentified byproducts).
Reprinted with permission from Ref. [9]. Copyright © 2015, American Chemical Society
Synthesis of the C1–C20 section.
The C21–C29 section 44 was synthesized from (R)-glycidol (34) (Scheme 3.7) [9]. Cross-metathesis of 35 derived from 34 and acrolein (36) with Hoveyda–Grubbs 2nd catalyst (that with Grubbs 2nd catalyst resulted in 19% yield) gave α,β -unsaturated aldehyde 37, which was subjected to intramolecular oxa-Michael reaction via hemiacetal formation [35] to furnish alcohol 38 after reduction of the aldehyde as a single diastereomer. Protecting group manipulation and oxidation of the hydroxy group at C24 afforded aldehyde 39, which was subjected to Brown asymmetric crotylation with 40 to furnish 41. Removal of the acetal in 41 in the presence of 1,3-propanediol [10] and subsequent protection as a TBS ether yielded 42. Hydroboration of the terminal olefin gave primary alcohol 43, which was converted to sulfone 44 via Mitsunobu reaction and oxidation with MCPBA of the resulting sulfide.
Reprinted with permission from Ref. [9]. Copyright © 2015, American Chemical Society
Synthesis of the C21–C29 section.
The C1–C29 section was synthesized as shown in Scheme 3.8 [9]. Julia–Kocienski olefination of aldehyde 33 and sulfone 44 with KHMDS in THF resulted in the formation of olefin 45 (E:Z = 20:1). One of the crucial steps, regio- and diastereoselective dihydroxylation, was successfully achieved by Sharpless AD to furnish 46 (dr = 13:1). The dihydroxylation occurred at the less hindered C20–C21 olefin compared to the C4–C5 and C8–C9 olefins located at neighboring TBSO groups. Oxidative removal of the NAP (2-naphthylmethyl) group with DDQ giving the primary alcohol followed by dehydration using the Nishizawa–Grieco protocol afforded terminal olefin 23.
Reprinted with permission from Ref. [9]. Copyright © 2015, American Chemical Society
Synthesis of the C1–C29 section of AM3.
3.4.3 Synthesis of the C30–C52 Section
The precursors of the C30–C52 section 24, the A- and B-rings, were prepared from the common intermediate 47 as shown in Scheme 3.9 [11]. Synthesis of the A-ring commenced with Sharpless AD of 47 and subsequent protection of the hydroxy groups to give 48. Protecting group manipulation giving primary alcohol 49 followed by oxidation and Horner–Wadsworth–Emmons reaction with phosphonate 50 under Masamune–Roush conditions resulted in the formation of enone 51. After hydrogenation of the olefin, the resulting methyl ketone was converted to enol triflate 53 by treatment with KHMDS followed by Comins reagent 52. The enol triflate 53 was converted to iodide 54 via Stille reaction to give a stannane followed by treatment with iodine.
Reprinted with permission from Ref. [11]. Copyright © 2018, Wiley–VCH Verlag GmbH & Co. KGaA, Weinheim
Synthesis of the A- and B-rings.
Synthesis of the B-ring commenced with Katsuki–Sharpless AE of the allylic alcohol derived from 47. Payne rearrangement of epoxy alcohol 55 gave the sulfide with inversion of stereochemistry at C51, and protection of the resulting secondary alcohol gave 56. Pummerer rearrangement by treatment with MCPBA followed by trifluoroacetic anhydride and base resulted in the formation of an aldehyde, which was reduced with NaBH4 to furnish 57. After protection of the hydroxy group as a PMB ether, selective removal of the Bn group in the presence of the PMB group was achieved by treatment with Raney nickel under hydrogen, and oxidation of the resulting hydroxy group afforded aldehyde 58.
With the A- and B-rings in hand, coupling of these fragments was examined (Table 3.1). Tin-lithium exchange reaction of stannane 59 with n-BuLi (1 eq) in THF at − 78 °C and subsequent addition of aldehyde 58 in THF at − 78 °C afforded 60a in 33% yield accompanied by its diastereomer at C43 (60b) in 11% yield (entry 1). When iodoolefin 54 was treated with t-BuLi (2 eq) in Et2O at − 78 °C to generate the corresponding alkenyllithium followed by addition of aldehyde 58 in Et2O at − 78 °C, the yields of 60a (37%) and 60b (19%) increased slightly (entry 2). Although total yields of the products were improved by raising the temperature of the solution of 58 in Et2O to − 40 °C (entry 3), the ratio of the desired compound decreased, yielding 32% each of 60a and 60b. By changing the solvent of the aldehyde 58 to THF and lowering the temperature to − 78 °C, the desired 60a was obtained in 40% yield with 13% of 60b.
Attempts to improve the yield of the coupling product are shown in Scheme 3.10. Nozaki–Hiyama–Kishi coupling of aldehyde 58 and iodoolefin 54 or enol triflate 53 was unsuccessful via C42–C43 bond formation with recovery of 58 and terminal olefin derived form 54 or 53. Alternatively, cross-coupling reactions via C41–C42 bond formation by Suzuki–Miyaura coupling of iodoolefin 62 and alkylborane 64, Negishi coupling of iodoolefin 62 and alkylzinc reagent 65, or SN2 reaction of cuprate 63 and iodide 66, were examined, but in vain with recovery of the starting materials or formation of complex mixtures.
Unsuccessful coupling of the A- and B-rings
Coupling product 60a was converted to trisubstituted iodoolefin 24 corresponding to the C30–C52 section (Scheme 3.11) [11]. Protection of the hydroxy group at C43, removal of the TES group with TBAF/AcOH, and oxidation of the primary hydroxy group gave aldehyde 67. Alkynylation with Ohira–Bestmann reagent 68 gave a terminal alkyne, which was methylated via generation of a lithium acetylide with n-BuLi, followed by addition of MeI to furnish 69. Hydrozirconation of 69 with Schwartz reagent followed by treatment with iodine afforded the trisubstituted iodoolefin 24 corresponding to the C30–C52 section.
Synthesis of the C30–C52 section
3.4.4 Suzuki–Miyaura Coupling
With the essential intermediates in hand, we investigated the key Suzuki–Miyaura coupling step. First, a model experiment using terminal olefin 70 and iodoolefin 72 was examined (Scheme 3.12) [10]. Hydroboration of 70 with 9-BBN to generate alkylborane 71, followed by successive addition of aq Cs2CO3, iodoolefin 72, and palladium catalyst, afforded 73 in 75% yield. Encouraged by these results, Suzuki–Miyaura coupling of the C21–C29 section (23) and the C30–C52 section (24) was examined (Scheme 3.13). However, under identical conditions as those for the successful model study, Suzuki–Miyaura coupling did not proceed and no desired coupling product 74 was obtained. To investigate the scope and limitation of the Suzuki–Miyaura coupling, iodoolefin 72 corresponding to the C30–C40 section was utilized instead of 24 (Scheme 3.14). However, no coupling product was obtained under various conditions, e.g., AsPh3 added as an accelerating ligand, or TlOEt as a base as referenced in synthetic studies of palytoxin reported by Kishi [36]. As byproducts, olefin 75 or allene 76 was obtained. Alternatively, the C21–C29 section (77) was prepared by changing the protecting groups at C20, C21, C25, and C7 from TBS groups to cyclopentylidene acetals. However, Suzuki–Miyaura coupling of 77 did not proceed with iodoolefin 72.
Reprinted with permission from Ref. [10]. Copyright © 2017, Chemical Society of Japan
Suzuki–Miyaura coupling of the C21–C29 and C30–C40 sections.
Unsuccessful Suzuki–Miyaura coupling of the C1–C29 and C30–C52 sections
Unsuccessful Suzuki–Miyaura coupling of the C1–C29 and C30–C40 sections
Finally, the C21–C29 section 70 was utilized (Scheme 3.15). In this combination, Suzuki–Miyaura coupling of 70 and 24 proceeded to afford 78 but in moderate yield (51%). Since the coupling product 78 corresponding to the C21–C52 section was obtained, we proceeded to the synthesis of AM3. The TES group was selectively removed with TBAF/AcOH. Oxidation giving aldehyde, followed by Julia–Kocienski olefination with sulfone 79 afforded olefin 80 (E:Z = 11:1). The next step, regio- and diastereoselective dihydroxylation, was anticipated to be more difficult than that for 45 (Scheme 3.8) due to the presence of the additional trisubstituted and exo-olefins. As anticipated, Sharpless AD of 80 resulted in a low yield of diol 81 (28%), but the byproducts could not be identified.
Suzuki–Miyaura coupling of the C21–C29 and C30–C52 sections and synthesis of the C1–C52 section
3.5 Revision of the Absolute Configurations at C32–C36 and C38
In 2010, the similar compound to AM3, karlotoxin 2 (KmTx2, 82) was identified (Fig. 3.4) [37, 38]. However, the absolute configuration of the A- and B-rings was antipodal to that of AM3. Although the stereochemistry at C39 of AM3 was elucidated by the modified Mosher method, it was difficult to determine the relationship between C38 and C39 by the JBCA method. Therefore, there is a possibility that the relationship between C38 and C39 is not threo but erythro, namely the stereochemistry at C32–C36 and C38 of AM3 are antipodal (83).
Structure of karlotoxin 2, revised structure at C2 (2008) and C51 of AM3 (2013) and plausible structure of AM3
The absolute configurations around the bis-THP moiety of AM3 were determined by the modified Mosher method through degradation of the natural product via glycol cleavage with sodium metaperiodate and subsequent reduction, and esterification of the polyol as MTPA esters [2]. Therefore, we envisaged that it is possible to elucidate the stereochemistry around the bis-THP moiety by comparing the degradation product, (S)-MTPA ester 84 derived from AM3, with those from authentic samples, (S)-MTPA esters 86 and 88 derived from 85 and 87, respectively (Scheme 3.16).
Strategy for structure determination of the bis-THP moiety
A precursor of 85, the C31–C52 section 60a corresponding to the revised structure in 2013 (22), has already been synthesized (Table 3.1) [11]. As a precursor of 87, the C31–C52 section 96a corresponding to the plausible structure 83 was synthesized as shown in Scheme 3.17 [11]. The A-ring ent-15 was prepared from ent-3 in a similar manner to that shown in Scheme 3.3. Protection of the 1,2-diol moiety of ent-15 as a cyclopentylidene acetal followed by Sharpless AD of the olefin gave diol 89. The resulting 1,2-diol was also protected as a cyclopentylidene acetal, and the remaining secondary alcohol was converted to mesylate 90. Selective removal of the Bn group in the presence of the PMB group was achieved with Raney nickel under hydrogen, and subsequent treatment of the resulting primary alcohol with base furnished terminal epoxide 91 via inversion of the absolute configuration at C39. Epoxide opening of 91 with dilithium reagent 92 followed by protection as a TBS ether gave 93. Nickel catalyzed hydroalumination of the terminal alkyne 93 followed by addition of iodine furnished iodoolefin 94. After conversion of the protecting group from PMB to TES, treatment of the iodoolefin 95 with t-BuLi to generate an alkenyllithium, followed by addition of aldehyde 58, afforded coupling product 96a and its diastereomer 96b at C43 in 71% yield in a 1.7:1 ratio. The epimer was removed by silica gel column chromatography.
Reprinted with permission from Ref. [11]. Copyright © 2018, Wiley–VCH Verlag GmbH & Co. KGaA, Weinheim
Synthesis of the C31–C52 section corresponding to plausible structure.
Having synthesized the precursors of the authentic samples, degradation of 60a and 96a was carried out through (1) deprotection with HF∙Py, (2) glycol cleavage with HIO4 and subsequent reduction with NaBH4, and (3) esterification with (R)-MTPACl to give (S)-MTPA esters 86 and 88, respectively (Scheme 3.18) [11]. The 1H NMR data for the degradation product derived from AM3 (84) matched those for the authentic sample 88 (38S, 39R) but not 86 (38R, 39R). Therefore, the relative configuration between C38 and C39 is not threo but erythro, and the absolute configurations were revised to be 32S, 33R, 34S, 35S, 36S, and 38S.
Reprinted with permission from Ref. [11]. Copyright © 2018, Wiley–VCH Verlag GmbH & Co. KGaA, Weinheim
Preparation of authentic samples and comparison with the degradation product of AM3.
3.6 Total Synthesis of AM3 Based on the Revised Structure in 2018
We moved on to total synthesis of AM3 (83) to verify the structure revised in 2018 (Scheme 3.19) [13]. In an analogous retrosynthetic analysis as shown in Scheme 3.5, 83 is to be synthesized via Suzuki–Miyaura coupling of 23 and 97, and Julia–Kocienski olefination with 20. However, Suzuki–Miyaura coupling of 23 and 97 might be problematic as mentioned in Sect. 3.4.4.
Reprinted with permission from Ref. [13]. Copyright © 2020, American Chemical Society
Revised structure of AM3 at C32–C36 and C38 and retrosynthetic analysis.
Therefore, model experiments were carried out to optimize the reaction conditions. In place of the polyol segment 23, a simple terminal olefin with a linear 15-carbon chain (98) was used (Scheme 3.20). However, even in this simple substrate, Suzuki–Miyaura coupling with iodoolefin 72 did not gave desired product 99 under the standard conditions for coupling of small molecules. After considerable experimentation, we found that the concentration of aq Cs2CO3 had a strong effect on this reaction. When the concentration of aq Cs2CO3 was changed from 3 to 1 M, 99 was formed in 60% yield.
Model study of Suzuki–Miyaura coupling
We then moved on to the Suzuki–Miyaura coupling of 70 and 97 as a model study (Scheme 3.21) [13]. Synthesis of 97 commenced with protection of the secondary alcohol at C43 of 96a as a TBS ether. The TES group was selectively removed and the resulting hydroxy group was oxidized to an aldehyde. Alkynylation with Ohira–Bestmann reagent 68 furnished terminal alkyne 100. Methylation of 100 with LHMDS and MeI, followed by palladium catalyzed hydrostannylation and iodination, afforded trisubstituted iodoolefin 97. Suzuki–Miyaura coupling of 70 with a shorter polyol moiety was examined. Hydroboration of 70 with 9-BBN, and successive addition of 3 M Cs2CO3, 70 in DMF, and Pd(PPh3) at room temperature furnished coupling product 101 but in low yield (29%). When the concentration of aq Cs2CO3 was changed from 3 to 1 M, the yield of 101 was improved to 45%, which was lower than that with its counterpart 24 (51%) (Scheme 3.15). The reaction was carried out with 3 M Cs2CO3, and after the addition of 70 and Pd(PPh3), H2O was added to the reaction mixture, diluting the concentration of aq Cs2CO3 from 3 to 1 M. As a result, Suzuki–Miyaura coupling proceeded rapidly and was completed in 10 min, affording 101 in high yield (80%).
Reprinted with permission from Ref. [13]. Copyright © 2020, American Chemical Society
Model study of Suzuki–Miyaura coupling.
Finally, Suzuki–Miyaura coupling of 23 and 97 was carried out (Scheme 3.22) [13]. Although the Suzuki–Miyaura coupling with 3 M Cs2CO3 gave the coupling product 102 in 42% yield, that with 1 M Cs2CO3 improved the yield to 77%. Thus, connection of the large segments, terminal olefin 23 (MW 1844) and iodoolefin 97 (MW 1747) was successfully achieved by Suzuki–Miyaura coupling under the optimized conditions to afford the long-cherished compound 102 (MW 3396).
Reprinted with permission from Ref. [13]. Copyright 2020 American Chemical Society
Suzuki–Miyaura coupling of the C1–C29 and C30–C52 sections.
The results of the Suzuki–Miyaura coupling can be rationalized as shown in Fig. 3.5, while the reaction mechanism is still controversial. Hydroboration of 23 or 70 with 9-BBN affords alkylborane A. Oxidative addition of Pd(0) catalyst to iodoolefin 97 furnished Pd(II) iodide complex C. It is reported that ligand exchange from I– to OH– generates highly reactive hydroxo Pd(II) complex D, which reacts with alkylborane A via coordination of the oxygen atom to the borane to give the coupling products 102 or 101 via transition state E, respectively. It is reported that the reaction rate for D is to be 104 times faster than that for C (k4 ≫ k2) [39]. The complex D would be formed at the interfacial surface of the organic (ORG) and aqueous (AQ) phases, or AQ. Therefore, formation of D might be retarded due to the low accessibility of C in AQ (conditions a). This salting out effect due to the size and hydrophobicity of C results in the low yield of the coupling products 102 (42%) and 101 (29%). Formation of D might be accelerated by diluting the aq Cs2CO3 available to react with A giving 102 in 77% yield (conditions b). However, in the case of A generated from 70, formation of borate B would be accelerated under conditions b to inhibit the reaction with D, giving 101 in 45% yield. Dilution of aq Cs2CO3 (3M to 1 M) by adding water was carried out at the final step (conditions c). Therefore, A could react with D prior to the formation of B (k3 > k1) to afford 101 in 80% yield.
Plausible reaction process for Suzuki–Miyaura coupling
On the other hand, it is reported that B is more reactive with C than A to give E via elimination of I– [40]. However, the present results suggest that the reaction proceeds via D, which reacts with A directly. If B is more reactive than A, coupling with 70 could afford better yields than 23 under both conditions a and b. Because it is more favorable to form B from A for 70 possessing a shorter carbon chain than 23 due to its higher accessibility in AQ.
Having succeeded in the Suzuki–Miyaura coupling of 23 and 97 giving 102, we approached the endgame of the total synthesis (Scheme 3.23) [13]. The PMB group of 102 was removed with DDQ (75%) and the resulting primary alcohol was subjected to Swern oxidation to furnish 103 (quant). Coupling of aldehyde 103 and sulfone 20 by Julia–Kocienski olefination. KHMDS was added to a mixture of 103 and 20 in a mixed solvent (THF: toluene = 7.5:1) at − 78 °C and warmed up to room temperature to afford the olefin 104 in 67% yield (E:Z = 2:1). Use of THF as a single solvent improved both the yield to 79% and the E:Z ratio to 4:1. The E:Z ratio was improved to 10:1 when the solvent was changed to THF/HMPA = 4:1 (79% yield). Presumably, the polar solvents prevent coordination of K+ in the intermediate. Removal of all silyl groups and acetals was carefully carried out due to the labile nature of the product, which is prone to decomposition under acidic conditions. Thus, treatment of 104 with HF·Py in THF to remove silyl groups, and subsequential addition of (CH2OH)2 and MeOH to accelerate the removal of acetals afforded a mixture of AM3 (83) and mono-acetal 105 and/or 106. The mixture was subjected again to the same conditions to afford 83 in 58% (purified by HPLC). The 1H and 13C NMR data including the specific rotation for the synthetic sample were in good accordance with those for AM3; therefore, the structure of AM3 revised in 2018 was confirmed.
Reprinted with permission from Ref. [13]. Copyright © 2020, American Chemical Society
Total synthesis of AM3.
3.7 Synthetic Efficiency
The first and highly convergent total synthesis of AM3 (83) has been accomplished through the assemblage of the polyol (23), bis-THP (97), and polyene (20) segments in only 5 steps. Generally, in the case of total syntheses of natural products, the efficiency is evaluated by the longest linear sequence (LLS). The LLS was 40 steps in the synthesis of 83. Although recent progress on short-step total syntheses of diterpenoids is remarkable, the number of steps might be highly dependent on the MW of the target molecules, making it difficult to evaluate the synthetic efficiency (SE) based simply on this parameter. Therefore, we proposed a new concept to evaluate the efficiency in the synthesis of medium molecular-weight molecules such as AM3 (Fig. 3.6) [41]. We defined an index SELLS which is calculated as MW divided by LLS, meaning how MW increases on average per single step. The LLS synthetic efficiency for AM3 is calculated to be SELLS(AM3) = 1328/40 = 33.2. This value is comparable to that for synthesis of a small molecule (MW 332) in10 steps. On the other hand, another important factor for evaluating the total SE is the total steps (TS), because a number of total syntheses of natural products have been executed in a convergent manner. To evaluate the total efficiency, we defined another index SETS, which is calculated as MW divided by TS, meaning how MW increases per single step on average in total. Since TS of the total synthesis is 112 steps [13], the TS synthetic efficiency of AM3 is calculated to be SETS(AM3) = 1328/112 = 11.9. These parameters, SELLS and SETS, might be useful indices to assess the synthetic efficiency in the total syntheses of medium molecular-weight molecules.
Reprinted with permission from Ref. [41]. Copyright © 2020, Chemical Society of Japan
Synthetic efficiency.
3.8 Conclusion
It was a long 15-year journey to reveal the correct structure of AM3 and to achieve its first total synthesis, which was accomplished by the power of synthetic organic chemistry. Although progress on computational chemistry and data science is remarkable and they have been applied to structure determination, total synthesis is indisputable approach to determine and verify structures including absolute configurations, particularly in the case of complex and large natural products that are only available in small quantities from natural sources. Comparison of 1H and 13C NMR data for the natural product with those for the synthetic samples apparently proves the structure. During our synthetic studies of AM3, a new aspect in Suzuki–Miyaura coupling was discovered in the case of medium-sized molecules, and a new parameter to evaluate synthetic efficiency (SE) that is suitable for total syntheses of medium-sized molecules was proposed.
References
Paul G K, Matsumori N, Konoki K, Sasaki M, Murata M, Tachibana K (1996) Structure of amphidinol 3 and its cholesterol-dependent membrane perturbation: A strong antifungal metabolites produced by dinoflagellate, Amphidinium klebsii. In: Harmful and Toxic Algal Blooms, Yasumoto T, Oshima Y, Fukuyo Y (eds) Intergovernmental Oceanographic Commission of UNESCO, Paris, pp. 503–506.
Murata M, Matsuoka S, Matsumori N, Paul G. K, Tachibana K (1999) Absolute configuration of amphidinol 3, the first complete structure determination from amphidinol homologues: Application of a new configuration analysis based on carbon-hydrogen spin-coupling constants. J. Am. Chem. Soc. 121: 870–871.https://doi.org/10.1021/ja983655x
Matsumori N, Kaneno D, Murata M, Nakamura H, Tachibana K (1999) Stereochemical determination of acyclic structures based on carbon−proton spin-coupling constants. A method of configuration analysis for natural products. J. Org. Chem. 64: 866–876. https://doi.org/10.1021/jo981810k
Ohtani I, Kusumi T, Kashman Y, Kakisawa H (1991) High-field FT NMR application of Mosher's method. The absolute configurations of marine terpenoids. J. Am. Chem. Soc. 113: 4092−4096. https://doi.org/10.1021/ja00011a006
Oishi T, Kanemoto M, Swasono R, Matsumori N, Murata M (2008) Combinatorial Synthesis of the 1,5−Polyol System Based on Cross Metathesis: Structure Revision of Amphidinol 3. Org. Lett. 10: 5203−5206. https://doi.org/10.1021/ol802168r
Kanemoto M, Murata M, Oishi T (2009) Stereoselective synthesis of the C31−C40/C43−C52 unit of amphidinol 3. J. Org. Chem. 74: 8810–8813. https://doi.org/10.1021/jo901793f
Manabe Y, Ebine M, Matsumori N, Murata M, Oishi T (2012) Confirmation of the absolute configuration at C45 of amphidinol 3. J. Nat. Prod. 75: 2003−2006. https://doi.org/10.1021/np300604w
Ebine M, Kanemoto M, Manabe Y, Konno Y, Sakai K, Matsumori N, Murata M, T Oishi (2013) Synthesis and structure revision of the C43−C67 part of amphidinol 3. Org. Lett. 15: 2846−2849. https://doi.org/10.1021/ol401176a
Tsuruda T, Ebine M, Umeda A, Oishi T (2015) Stereoselective synthesis of the C1-C29 part of amphidinol 3. J. Org. Chem. 80: 859−871. https://doi.org/10.1021/jo502322m
Ebine M, Takada Y, Yanai N, Oishi T (2017) Synthesis of a truncated analog of amphidinol 3 corresponding to the C21–C39/C52–C67 section. Chem. Lett. 46: 662−664. https://doi.org/10.1246/cl.160985
Wakamiya Y, Ebine M, Murayama M, Omizu H, Matsumori N, Murata M, Oishi T (2018) Synthesis and stereochemical revision of the C31-C67 fragment of amphidinol 3. Angew. Chem. Int. Ed. 57: 6060−6064. https://doi.org/10.1002/anie.201712167
Koge T, Wakamiya Y, Ebine M, Oishi T (2018) Synthesis of an analog of amphidinol 3 corresponding to the C31–C67 section. Heterocycles 96: 1197−1202. https://doi.org/10.3987/COM-18-13927
Wakamiya Y, Ebine M, Matsumori N, Oishi T (2020) Total synthesis of amphidinol 3: A general strategy for synthesizing amphidinol analogues and structure-activity relationship study. J. Am. Chem. Soc. 142: 3472−3478. https://doi.org/10.1021/jacs.9b11789
Katayama S, Koge T, Katsuragi S, Akai S, Oishi T (2018) Flow synthesis of (3R)- and (3S)-(E)-1-iodohexa-1,5-dien-3-ol: Chiral building blocks for natural product synthesis. Chem. Lett. 47: 1116–1118. https://doi.org/10.1246/cl.180475.
For a review, see: Bensoussan C, Rival N, Hanquet G, Colobert F, Reymond S, Cossy J (2014) Isolation, structural determination and synthetic approaches toward amphidinol 3. Nat. Prod. Rep. 31: 468–488. https://doi.org/10.1039/c3np70062c
BouzBouz S, Cossy J (2001) Chemoselective cross-metathesis reaction. Application to the synthesis of the C1–C14 fragment of amphidinol 3. Org. Lett. 3: 1451–1454. https://doi.org/10.1055/10.1021/ol0157095
Cossy J, Tsuchiya T, Ferria L, Reymond S, Kreuzer T, Colobert F, Jourdain P, Marko I E (2007) Efficient syntheses of the polyene fragments present in amphidinols. Synlett 14: 2286–2288. https://doi.org/10.1055/s-2007-985601
Colobert F, Kreuzer T, Cossy J, Reymond S, Tsuchiya T, Ferrié L. Markó I E, Jourdain P (2007) Stereoselective synthesis of the C(53)–C(67) polyene fragment of amphidinol 3. Synlett 2351–2354. https://doi.org/10.1055/s-2007-985601
Cossy J, Tsuchiya T, Reymond S, Kreuzer T, Colobert F, Markó I (2009) Convergent synthesis of the C18–C30 fragment of amphidinol 3. Synlett 16: 2706–2710. https://doi.org/10.1055/s-0029-1217754
Rival N, Hazelard D, Hanquet G, Kreuzer T, Bensoussan C, Reymond S, Cossy L, Colobert F (2012) Diastereoselective synthesis of the C17–C30 fragment of amphidinol 3. Org, Biomol. Chem. 10: 9418–9428. https://doi.org/10.1039/c2ob26641e
Rival N, Hanquet G, Bensoussan C, Reymond S, Cossy J, Colobert F (2013) Diastereoselective synthesis of the C14–C29 fragment of amphidinol 3. Org. Biomol. Chem. 11: 6829–6840. https://doi.org/10.1039/c3ob41569d
Bensoussan C, Rival N, Hanquet G, Colobert F, Reymond S, Cossy J (2013) Iron-catalyzed cross-coupling between C-bromo mannopyranoside derivatives and vinyl Grinard reagent: toward the synthesis of the C31–C52 fragment of amphidinol 3. Tetrahedron 69: 7759–7770. https://doi.org/10.1016/j.tet.2013.05.067
Flamme E M, Roush W R (2005) Synthesis of the C(1)–C(25) fragment of amphdinol 3: Application of the double-allylboration reaction for synthesis of 1,5-diols. Org. Lett. 7: 1411–1414. https://doi.org/10.1021/ol050250q
Hicks J D, Flamme E M, Roush W R (2005) Synthesis of the C(43)–C(67) fragment of amphidinol 3. Org. Lett. 7: 5509–5512. https://doi.org/10.1021/ol052322j
Hicks J D, Roush W R (2008) Synthesis of the C(26)–C(42) and C(43)–C(67) pyran-containing fragments of amphidinol 3 via a common pyran intermediate. Org. Lett. 10: 681–684. https://doi.org/10.1021/ol703042q
de Vicente J, Betzemeier B, Rychnovsky S D (2005) A C-glycosidation approach to the central core of amphidinol 3: synthesis of the C39–C52 fragment. Org. Lett. 7: 1853–1856. https://doi.org/10.1021/ol050477l
de Vicente J, Huckins J R, Rychnovsky S D (2006) Synthesis of the C31–C67 fragment of amphidinol 3. Angew. Chem. Int. Ed. 45, 7258–7262. https://doi.org/10.1002/anie.200602742
Huckins J R, de Vicente J, Rychnovsky S D (2007) Synthesis of the C1–C52 fragment of amphidinol 3, featuring a β-alkoxy alkyllithium addition reaction. Org. Lett. 9: 4757–4760. https://doi.org/10.1021/ol7020934
Paquette L A, Chang S-K (2005) The polyol domain of amphidinol 3. A stereoselective synthesis of the Entire C(1)–C(39) Sector. Org. Lett. 7: 3111–3114. https://doi.org/10.1021/ol0511833
Chang S-K, Paquette L A (2005) Synthesis of the skipped polyene chain and its neighboring highly oxygenated pyran ring en route to delivering the C(43)–C(67) subsector of amphidinol 3. Synlett 19: 2915–2918. https://doi.org/10.1055/s-2005-918960
Bedore M W, Chang S-K, Paquette L A (2007) Development of an access route to the C31–C52 central core of amphdinol 3. Org. Lett. 9: 513–516. https://doi.org/10.1021/ol062875+
Crimmins M T, Martin T J, Martinot T A (2010) Synthesis of the bis-tetrahydropyran core of amphidinol 3. Org. Lett. 12: 3890–3893. https://doi.org/10.1021/ol1015898
Grisin A, Evans P A (2015) A highly convergent synthesis of the C1–C31 polyol domain of amphidinol 3 featuring a TST-RCM reaction: confirmation of the revised relative stereochemistry. Chem. Sci. 6: 6407–6412. https://doi.org/10.1039/c5sc00814j
Yadav J S, Gopalarao Y, Chandrakanth D, Reddy B V S (2016) Stereoselective synthesis of the C(1)–C(28) fragment of amphidinol 3. Helv. Chim. Acta 99: 436–446. https://doi.org/10.1002/hlca.201500281
Evans P A, Grisin A, Lawler M (2012) Diastereoselective construction of syn-1,3-dioxanes via a bismuth-mediated Two-component hemiacetal/oxa-conjugate addition reaction. J. Am. Chem. Soc. 134: 2856–2859. https://doi.org/10.1021/ja208668u
Uenishi J, Beau J M, Armstrong R W, Kishi Y (1987) Dramatic rate enhancement of Suzuki diene synthesis. Its application to palytoxin synthesis. J. Am. Chem. Soc. 109: 4756–4758. https://doi.org/10.1021/ja00249a069
Peng J, Place A R, Yoshida W, Anklin C, Hamann M T (2010) Structure and Absolute Configuration of Karlotoxin-2, an Ichthyotoxin from the Marine Dinoflagellate Karlodinium veneficum. J. Am. Chem. Soc. 132: 3277–3279. https://doi.org/10.1021/ja9091853
Waters A L, Oh J, Place A R, Hamann M T (2015) Stereochemical studies of the karlotoxin class using NMR spectroscopy and DP4 chemical-shift Analysis: Insights into their mechanism of action. Angew. Chem. Int. Ed. 54: 15705–15710. https://doi.org/10.1002/anie.201507418
Carrow B P, Hartwig J F (2011) Distinguishing between pathways for transmetalation in Suzuki–Miyaura reactions. J. Am. Chem. Soc. 133: 2116–2119. https://doi.org/10.1021/ja1108326
Matos K, Soderquist J A (1998) Alkylboranes in the Suzuki–Miyaura coupling: Stereochemical and mechanistic studies. J. Org. Chem. 63: 461–470. https://doi.org/10.1021/jo971681s
Oishi T (2020) Structure determination, chemical synthesis, and evaluation of biological activity of super carbon chain natural products. Bull. Chem. Soc. Jpn. 93: 1350−1360. https://doi.org/10.1246/bcsj.20200151
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Copyright information
© 2024 The Author(s)
About this chapter
Cite this chapter
Oishi, T. (2024). A Long Journey Toward Structure Revision and Total Synthesis of Amphidinol 3. In: Nakada, M., Tanino, K., Nagasawa, K., Yokoshima, S. (eds) Modern Natural Product Synthesis. Springer, Singapore. https://doi.org/10.1007/978-981-97-1619-7_3
Download citation
DOI: https://doi.org/10.1007/978-981-97-1619-7_3
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-97-1618-0
Online ISBN: 978-981-97-1619-7
eBook Packages: Chemistry and Materials ScienceChemistry and Material Science (R0)