Download PDF
Synlett 2024; 35(13): 1488-1499
DOI: 10.1055/a-2181-9876
account

Total Synthesis of Marine Macrolide Natural Products by the Macrocyclization/Transannular Pyran Cyclization Strategy

Haruhiko Fuwa  ∗

This work was supported by the Japan Society for the Promotion of Science (JSPS KAKENHI; Grant Nos. JP17K01941 and JP22K05336) and by a Chuo University Grant for Special Research.
› Further Information
 
 PDF Download Permissions and Reprints All articles of this category


Abstract

In this Account, we summarize the development of a new strategy for streamlined synthesis of tetrahydropyran-embedded macrolactones and its successful implementation to a 13-step synthesis of (–)-exiguolide and an 11-step synthesis of (+)-neopeltolide.

1 Introduction

2 Development of Macrocyclization/Transannular Pyran Cyclization Strategy

3 Total Synthesis of (–)-Exiguolide

4 Total Synthesis of (+)-Neopeltolide

5 Conclusions


#

Key words

synthetic methods - tandem reactions - total synthesis - macrolides - step-economy

Biographical Sketch

Zoom Image

Haruhiko Fuwa received his B.Sc. (1997), M.Sc. (1999), and Ph.D. (2002) degrees from the Department of Chemistry, School of Science, The University of Tokyo, under the guidance of the late Professor Kazuo Tachibana. He joined the group led by Professor Makoto Sasaki at the Graduate School of Life Sciences, Tohoku University. There, he was appointed as Assistant Professor (2006) and then promoted to Associate Professor (2009). He moved to the Department of Applied Chemistry, Faculty of Science and Engineering, Chuo University as Professor (2017). He received The Chemical Society of Japan Award for Young Chemists (2010) and The Young Scientists’ Award, The Commendation for Science and Technology by the Minister of Education, Sports, Science and Technology (2011).

1

Introduction

Macrolide natural products isolated from marine organisms, including cyanobacteria, dinoflagellates, sea hares, and sponges, are an important source of new chemotherapeutics for the treatment of human cancer and infectious diseases.[1] Unfortunately, marine macrolides are only available in very small amounts from natural producers, which are difficult to culture in large scales, and such compounds are also intractable to selective chemical modifications due to their complex structures. Thus, it has been believed that chemical synthesis is the most promising means to supply marine macrolides and their non-natural analogues in sufficient quantities.[2]

Total synthesis of macrolide natural products has been an area of long-standing interest for the synthetic community.[2] The common structural characteristic of macrolides is their macrocyclic skeletal structure, involving multiple stereogenic centers substituted with alkyl and/or oxygen functional group(s). It is also notable that five- or six-membered cyclic ether(s) may be embedded within their macrocyclic skeleton. The inherent challenges associated with macrolide synthesis are how to prepare stereochemically complex building blocks and how to build up macrocyclic skeletons through convergent assembly of building blocks.

Our group has been engaged in synthetic studies on marine macrolide natural products for more than 15 years. During this period, we completed the asymmetric total synthesis of (+)-neopeltolide (2008,[3] 2011[4]), (–)-aspergillide A (2010),[5] (–)-aspergillide B (2010),[5] (–)-exiguolide (2010),[6] (–)-lyngbyaloside B (2015),[7] (–)-iriomoteolide-2a (2018),[8] (–)-enigmazole A (2018),[9] and (–)-15-O-methylenigmazole A (2020)[9b] (Figure [1]). Some of these works contributed to the structural determination and/or biological evaluation of natural products.[6] [7] [8] However, we were not fully satisfied with the synthetic efficiency we could achieve. For example, the longest linear sequences required to complete our total syntheses of (–)-exiguolide and (–)-enigmazole A were 27 and 29 steps, respectively.[6,9]

Zoom Image
Figure 1 Structures of marine macrolide natural products synthesized in our group

This background prompted us to launch a project directed toward the development of a new strategy for streamlining the total synthesis of marine macrolide natural products, the outline of which is summarized in this Account.


# 2

Development of Macrocyclization/Transannular Pyran Cyclization Strategy

In parallel with our efforts on the total synthesis of marine macrolides, we have been continuously developing synthetic methods that enable expedient, stereocontrolled construction of polyketide building blocks.[10] Among those synthetic methods, tandem[11] olefin cross-metathesis[12]/intramolecular oxa-Michael addition[13] is remarkable in that it tolerates a variety of functional groups and proceeds under mild conditions to provide 2,6-cis-substituted tetrahydropyrans III in a highly stereocontrolled manner from readily available olefinic alcohols I (Scheme [1]A).[10a] Here, the intramolecular oxa-Michael addition (II → III) is thought to be catalyzed by a ruthenium hydride species derived from decomposition of the second-generation Hoveyda–Grubbs complex, HG-II.[14] [15]

Zoom Image
Scheme 1 (A) Previous work: Tandem olefin cross-metathesis/intramolecular oxa-Michael addition. (B) Reaction sequence design in this work.

We were particularly intrigued with the possibility of expanding the scope of this reaction to an intramolecular format; namely, V → VI → VII (Scheme [1]B). In addition, we considered that vinyl ketones V would be readily synthesizable from propargylic alcohols IV via a Meyer–Schuster rearrangement[16] under Au/Mo combo catalysis.[17] The reaction sequence we designed forges the macrocyclic skeleton and the embedded tetrahydropyran ring consecutively, and is free from concession steps, such as protection/deprotection and oxidation/reduction. Moreover, integrating all these catalytic transformations into a single step, if possible, should enable step-economical synthesis of tetrahydropyran-embedded macrolactones VII from propargylic alcohols IV. We termed our designed reaction sequence a ‘macrocyclization/transannular pyran cyclization strategy’.[18] However, we were uncertain at the outset of this work whether transannular oxa-Michael addition of macrocyclic α,β-unsaturated ketones VI was actually feasible. Although intramolecular oxa-Michael addition has been used commonly in stereoselective synthesis of tetrahydropyran derivatives,[2c] [13] its use in a transannular format has been limited to specific cases.[19]

Nonetheless, we envisaged optimistically that transannular oxa-Michael addition of macrocyclic α,β-unsaturated ketones VI could be achieved by taking advantage of their preorganized conformation in which the unprotected hydroxy group would be positioned in close proximity to the α,β-unsaturated ketone moiety.[20]

Accordingly, in the middle of 2016, we set out to investigate the feasibility of our designed reaction sequence using model compounds. Yu Onodera, a second-year Master course student of Tohoku University at that time, was responsible for the early phase of the model investigation. We initially examined the stepwise synthesis of tetrahydropyran-embedded macrolactones 12a–k from propargylic alcohols 9a–k (Table [1]). Meyer–Schuster rearrangement of propargylic alcohols 9a–k under Akai conditions (Ph3PAuCl, AgOTf, MoO2(acac)2, toluene)[17] delivered vinyl ketones 10a–k in mostly good yields. Inter- and intramolecular oxa-Michael additions of the unprotected hydroxy group were not observed in any of the cases, despite the products having a highly electrophilic vinyl ketone moiety. Macrocyclic ring-closing metathesis[21] of 10a–k was carried out under the influence of HG-II in toluene at 45 °C, giving macrocyclic α,β-unsaturated ketones 11a–k in good yields. No traces of the transannular product were found in any of the cases, and 11a–k could be isolated by silica gel flash column chromatography. Finally, transannular oxa-Michael addition of 11a–k, which was deemed to be the most critical process at the outset of this study, was examined under cationic Au(I) catalysis. It was gratifying to see that transannular oxa-Michael addition of 11a–k proceeded promptly and cleanly to furnish tetrahydropyran-embedded macrolactones 12a–k in mostly excellent yields, with good to excellent diastereoselectivity, except for 12b and 12c.

From these preliminary experiments, we made several important observations. First, the unprotected hydroxy group within 10a–k and 11a–k did not react with the electrophilic α,β-unsaturated ketone moiety in the absence of an appropriate catalyst. This observation strongly suggested that there is no need to protect the hydroxy group throughout the reaction sequence. Second, transannular oxa-Michael addition of 11a–k did not take place under ruthenium catalysis, at least under the reaction conditions used here (toluene, 45 °C).

This was in line with our previous finding that tandem olefin cross-metathesis/intramolecular oxa-Michael addition of olefinic alcohols proceeded inefficiently in toluene even under elevated temperature conditions.[22] Third, transannular oxa-Michael addition of 11a–k could be efficiently achieved under cationic Au(I) catalysis, supporting the possibility of reaction integration.

Together, these observations encouraged us to examine tandem Meyer–Schuster rearrangement/macrocyclic ring-closing metathesis/transannular oxa-Michael addition of propargylic alcohols 9a–k under Au/Ru catalysis. Gratifyingly, in most cases, the tandem synthesis proceeded cleaner than expected, and provided macrolactones 12a–k in higher yields when compared to the stepwise synthesis, thus clearly demonstrating the advantage of omitting chromatographic isolation of intermediates 10a–k and 11a–k. The diastereomer ratio of 12a–k obtained by the tandem synthesis corresponded to that of 12a–k prepared by the stepwise synthesis.

It seems that the stereochemical consequence of transannular oxa-Michael addition depends on the conformational property of the precursor macrocyclic α,β-unsaturated ketones 11a–k. While acid-catalyzed intramolecular oxa-Michael addition is known to give 2,6-cis-substituted tetrahydropyrans kinetically,[23] the stereochemical course of transannular oxa-Michael addition of 11b and 11c was apparently complicated by the macrocyclic constraint. This observation was actually a disappointment to us, but we did not take it seriously. Many of natural macrolides have multiple stereogenic centers and substituents along the macrocyclic backbone; such structural elements should have significant impact on the transition states of transannular oxa-Michael addition.[24]

Table 1 Stepwise/Tandem Synthesis of Tetrahydropyran-Embedded Macrolactones from Propargylic Alcoholsa

Propargylic alcohols

Meyer–Schuster

Macrocyclic RCM

Transannular oxa-Michael

Tandem reaction

Substrate

n

X, Y

Product

Yield (%)

Product

Yield (%)

Product

Ring size

Yield (%)

cis/trans

Product

Ring size

Yield (%)

cis/trans

9a

1

H, H

10a

79

11a

81

12a

14

87

 86:14

12a

14

79

 81:19

9b

1

O(CH2)2O

10b

N/Ab

11b

71c

12b

14

98

 58:42

12b

14

53

 47:53

9c

1

OTIPS, H

10c

75

11c

94

12c

14

88

 57:43

12c

14

52

 58:42

9d

3

H, H

10d

85

11d

83

12d

16

88

>95:5

12d

16

67

>95:5

9e

3

O(CH2)2O

10e

N/Ab

11e

63c

12e

16

90

>95:5

12e

16

69

 92:8

9f

5

H, H

10f

92

11f

71

12f

18

90

>95:5

12f

18

72

>95:5

9g

7

H, H

10g

89

11g

77

12g

20

87

>95:5

12g

20

75

 94:6

9h

H, H

10h

83

11h

84

12h

18

89

>95:5

12h

18

72

 94:6

9i

O(CH2)2O

10i

71

11i

72

12i

18

91

>95:5

12i

18

63

 92:8

9j

H, H

10j

77

11j

79

12j

18

88

>95:5

12j

18

75

 91:9

9k

O(CH2)2O

10k

84

11k

82

12k

18

98

 88:12

12k

18

68

 88:12

a Meyer–Schuster rearrangement: (Ph3P)AuCl (5 mol%), AgOTf (5 mol%), MoO2(acac)2 (5 mol%), toluene, r.t., 0.5–1 h. Macrocyclic RCM: HG-II (10–15 mol%), toluene (1 mM), 45 °C, 24 h. Transannular oxa-Michael: (Ph3P)AuCl (5 mol%), AgOTf (5 mol%), MoO2(acac)2 (5 mol%), toluene, r.t., 0.5 h. Tandem reaction: (Ph3P)AuCl (10–15 mol%), AgOTf (10–15 mol%), MoO2(acac)2 (10–15 mol%), toluene (0.1 M), r.t., 0.5 h; then HG-II (15 mol%), toluene (1 mM), 45 °C, 24 h.

b Yield not attained.

c Yield for two steps.

During the model investigation of the macrocyclization/transannular pyran cyclization strategy, the principal investigator (H.F.) started his own laboratory at Chuo University in April 2017 with one Ph.D. and five undergraduate students, including Mami Oda and Kei Iio. M.O. contributed to the successful completion of the model investigation by the middle of 2018, and then K.I. joined the project to initiate the total synthesis of (–)-exiguolide (1).


# 3

Total Synthesis of (−)-Exiguolide

(–)-Exiguolide (1) is a macrolide natural product that was isolated by Ohta and co-workers from a rare marine sponge Geodia exigua Thiele, collected off the Amami-Oshima, Kagoshima Prefecture, Japan.[25] The structure, including the relative configuration, was proposed on the basis of extensive 2D NMR analysis, J-based configurational analysis, and molecular modeling. The absolute configuration was eventually determined by Lee and co-workers through the total synthesis of the unnatural enantiomer, (+)-exiguolide.[26a] Our group completed the total synthesis of the natural enantiomer, (–)-exiguolide,[6a] and identified its potent anticancer activity against human non-small cell lung cancer cells.[6b] Due to the complex structure and anticancer activity, this natural product has been an intriguing target for the synthetic community.[26] [27] [28]

Our synthetic blueprint toward (–)-exiguolide (1) is summarized in Scheme [2]. On the basis of our previous synthesis,[6] 1 would be available from macrolactone 13a or 13b via an asymmetric Horner–Wadsworth–Emmons (HWE) olefination[29] and a Suzuki–Miyaura reaction.[30] We envisioned that 13a could be traced back to alkoxy ketone 14 by considering a transannular Kishi reduction.[31] In turn, 14 should be synthesizable from propargylic alcohol 15 by the macrocyclization/transannular pyran cyclization strategy. Another synthesis plan could also be envisioned by taking advantage of the pseudo-symmetric structure of the methylene bis-tetrahydropyran moiety. Thus, 13b would be obtainable from alkoxy ketone 16 through a transannular Kishi reduction, and the latter could be accessed from propargylic alcohol 17 according to the macrocyclization/transannular pyran cyclization strategy. Due to the lack of relevant precedents, we were uncertain whether the transannular Kishi reduction was an appropriate retrosynthetic disconnection in these situations. Also, we were concerned whether the macrocyclization/transannular pyran cyclization strategy would actually work in complex situations. However, our synthetic blueprint appeared worthwhile pursuing because, if successful, it would enable the total synthesis of 1 with remarkable synthetic efficiency.

Zoom Image
Scheme 2 Synthetic blueprint toward (–)-exiguolide
Zoom Image
Scheme 3 Total synthesis of (–)-exiguolide by the macrocyclization/transannular pyran cyclization strategy; Part 1
Zoom Image
Scheme 4 Total synthesis of (–)-exiguolide by the macrocyclization/transannular pyran cyclization strategy; Part 2: (A) Outline of the synthesis of key building blocks, (B) Completion of the total synthesis.

K.I. directed his efforts to the synthesis of key intermediate 13a from propargylic alcohol 15 (Scheme [3]). Meyer–Schuster rearrangement of 15 gave vinyl ketone 18 in 77% yield. Tandem macrocyclic ring-closing metathesis/transannular oxa-Michael addition of 18 was problematic due to the moderate reactivity of 18 and required screening of the reaction conditions. The reaction using G-II in DCE at 40 °C gave alkoxy ketone 14 in 18% yield, along with the intermediate macrocyclic α,β-unsaturated ketone 19 in 50% yield, while reaction under the catalysis of HG-II in DCE (40–60 °C) delivered 14 in 48% yield with recovery of unreacted 18 in 23% yield. Eventually, it was found that treatment of 18 with Zhan-1B [32] in DCE at 60 °C delivered alkoxy ketone 14 in 60% yield as a single diastereomer (d.r. >95:5), and this was the most successful result. Finally, transannular Kishi reduction of 14 was attempted using Et3SiH/BF3·OEt2, giving a single isolable product upon purification by flash column chromatography. However, the 1H NMR spectrum of the isolated product did not correspond to that of authentic 13a synthesized in our previous work.[6] Full spectroscopic characterization of the isolated product revealed that Kishi reduction of 14 proceeded with completely reversed diastereoselectivity to deliver 9-epi-13a. This was a nightmare – upon synthesis planning, we actually failed to pay attention to the stereochemical outcome of the Kishi reduction since this reaction is known to be a robust method for reducing six-membered ring oxocarbenium ions to give 2,6-cis-substituted tetrahydropyrans as a consequence of a stereoelectronic effect. In the present case, however, the α-face of an intermediate oxocarbenium ion generated from 14 should be shielded by the macrocyclic backbone. Accordingly, the reduction occurred from less hindered β-face of the oxocarbenium ion and afforded 9-epi-13a as the sole isolable product. An important lesson learned from this bitter experience was that we need to take special care for the effect of substrate conformation on stereodifferentiating reactions at post-macrocyclization stages.

After this, K.I. made a detour to (–)-exiguolide (1) through a stereoselective reduction of alkoxy ketone 14. Bromination of the resultant alcohol 20 with inversion of the C9 stereogenic center followed by deprotection of the PMB group led to bromo alcohol 21. Transannular cycloetherification of 21 was easily achieved in refluxing Et3N/CH3CN to afford 13a in a stereospecific manner, thereby intercepting our previous synthesis of (–)-1 (20 steps from a commercially available material).

In 2020, the Ishihara group at Nagasaki University published a paper on an elegant total synthesis of (–)-exiguolide (1), which was the shortest synthesis of 1 at that time (19 steps from a commercially available material).[26h] Almost simultaneously with the publication of the Ishihara synthesis of (–)-1, Daichi Mizukami, a first-year Master course student at that time, started to investigate the synthesis of macrolactone 13b with great enthusiasm. D.M. thoroughly optimized the synthesis of key building blocks 22, 23, and 24 from commercially available materials 25, 26, and 27, respectively (Scheme [4]A). Assembly of the building blocks led to propargylic alcohol 17 (8 steps from 26 or 27). The convergency of the synthesis of 17 was maximized by setting the point of convergence at the latest stage possible.[33]

Now the stage was set for macrocyclization/transannular pyran cyclization of propargylic alcohol 17 (Scheme [4]B). Meyer–Schuster rearrangement of 17 proceeded cleanly in the presence of IPrAuCl, AgOTf, and MoO2(acac)2 in toluene at room temperature to deliver vinyl ketone 29, which was isolated by silica gel flash column chromatography (81%). Tandem macrocyclic ring-closing metathesis/transannular oxa-Michael addition of 29 was found to be best achieved in the presence of G-II in CH2Cl2 (2 mM) at 40 °C, giving rise to alkoxy ketone 16 in 72% yield from 29 with greater than 95:5 diastereoselectivity. In contrast, the tandem reaction of 29 under the influence of HG-II or Zhan-1B in DCE (2–5 mM) at 40 °C gave unsatisfactory results; under these conditions, alkoxy ketone 16 was delivered in only 20–25% yield with recovery of the intermediate 30 in 53–57% yield. Transformation of propargylic alcohol 17 into alkoxy ketone 16 without isolating vinyl ketone 29 was then examined. Upon completion of Meyer–Schuster rearrangement of 17, the reaction mixture was directly exposed to G-II in CPME/DCE (1:1, v/v, 2 mM) at 40 °C, affording alkoxy ketone 16 in 81% yield from 17 with greater than 95:5 diastereoselectivity. In this case, CPME/DCE (1:1, v/v) was the solvent of choice for the macrocyclic ring-closing metathesis/trans­annular oxa-Michael addition sequence. When the same reaction sequence was carried out in DCE (2 mM) at 40 °C, alkoxy ketone 16 was obtained in only 49% yield from 17. We assume that the Lewis basicity of CPME modulates the Lewis acidity of cationic Au(I) complex, IPrAu+OTf−, and suppresses undesired decomposition of acid-labile intermediates 29 and 30. Nonetheless, it was suggested that IPrAu+OTf− was still responsible for transannular oxa-Michael addition of 30 under these conditions, as treatment of isolated 29 with G-II in CPME/DCE (1:1, v/v, 2 mM) at 40 °C delivered 30 in 89% yield, along with only 2% of 16.

Pleasingly, transannular Kishi reduction of alkoxy ketone 16 was achieved cleanly by its exposure to BF3·OEt2 in Et3SiH/CH2Cl2 (1:4, v/v, –78 to 0 °C), giving rise to macrolactone 13b in 79% yield with >95:5 d.r. In contrast to what we experienced in the synthesis of 9-epi-13a, transannular Kishi reduction of 16 proceeded with complete diastereoselection favoring the desired 2,6-cis-isomer, as predicted by molecular mechanics calculations on the corresponding oxocarbenium ion intermediate. At this point, the present synthesis successfully intercepted our previous synthesis,[6] and only three steps were left to reach the target. Dess–Martin oxidation of 13b (96%), followed by asymmetric HWE olefination using chiral phosphonate 31, gave α,β-unsaturated ester 32 (93%, Z/E 82:18). The minor (E)-isomer was removed by flash column chromatography using silica gel. Finally, Suzuki–Miyaura coupling of 32 with pinacolboronate 33 furnished (–)-exiguolide (1) in 71% yield. The present synthesis was accomplished in only 13 steps from a commercially available material and therefore represents the shortest synthesis of 1 hitherto reported.[28]

Zoom Image
Scheme 5 Total synthesis of (+)-neopeltolide and (+)-9-epi-neopeltolide by the macrocyclization/transannular pyran cyclization strategy: (A) Outline of the synthesis of key building blocks; (b) Completion of the total synthesis.

# 4

Total Synthesis of (+)-Neopeltolide

(+)-Neopeltolide (2) was isolated by Wright and co-workers from a Jamaican deep-water sponge that belongs to the family Neopeltidae.[34] The structure, including relative configuration, was assigned on the basis of 2D NMR analysis. However, the relative configuration was eventually reassigned through total synthesis and, at the same time, the absolute configuration of (+)-neopeltolide was unambiguously established as that shown in structure 2.[35] [36] Wright et al. reported that this natural product had potent antiproliferative activity against several human cancer cell lines and also significant antifungal activity against fungal pathogen Candida albicans. It was later revealed by Kozmin et al. that neopeltolide targets the complex III of the mitochondrial electron transport chain to inhibit oxidative phosphorylation.[37] Our group described that a synthetic analogue, (–)-8,9-dehydroneopeltolide, is a potent anti-austerity agent against energetically starved human tumor cells,[38] [39] and that (–)-8,9-dehydroneopeltolide accumulates rapidly in the mitochondria and the endoplasmic reticulum.[40] [41] Given the complex structure and intriguing biological activities, (+)-neopeltolide (2) has attracted immense attention from the synthetic community; more than 20 papers have been published to date on total and formal syntheses of 2.[42] [43] [44] Indeed, this natural product has served as a test ground for new methods for the synthesis of tetrahydropyrans and tetrahydropyran-embedded macrolactones.[2c]

In our preliminary attempts at the synthesis of 14- to 20-membered macrolactones, the observed diastereomer ratio was clearly lower for 14-membered macrolactones 12a–c compared to those of 16-, 18-, and 20-membered 12d–k (Table [1]). Nonetheless, we anticipated that our macrocyclization/transannular pyran cyclization strategy should still be applicable to 14-membered macrolide natural products having multiple substituents along the macrocyclic skeleton. It appeared that (+)-neopeltolide (2) was an ideal target to test this hypothesis.

Kazuki Nakazato, a fourth-year undergraduate student who joined our group in April 2019, was the key person in our total synthesis of (+)-neopeltolide (2) by the macrocyclization/transannular pyran cyclization strategy. The synthesis of important building blocks 34 and 35, outlined in Scheme [5]A, was carried out in six steps each. Esterification of these compounds and in situ acid treatment provided propargylic alcohol 36.

Tandem Meyer–Schuster rearrangement/macrocyclic ring-closing metathesis/transannular oxa-Michael addition of propargylic alcohol 36 required reaction conditions screening, mainly by varying catalyst loading, solvent, and reaction temperature, for optimization (Scheme [5]B). Eventually, Meyer–Schuster rearrangement of 36 was efficiently achieved under the influence of IPrAuCl (11 mol%), AgOTf (12 mol%), and MoO2(acac)2 (11 mol%) in toluene at room temperature, to provide transient vinyl ketone 37, which was then exposed to Zhan-1B (16 mol%) in DCE (20 mM) at 40 °C to furnish macrocycle 39 in 69% yield as a single diastereomer, after purification by flash column chromatography using silica gel. Lowering the loading of cationic Au(I) complex and substrate concentration resulted in incomplete conversion of intermediate 38. The present tandem reaction could also be achieved efficiently using HG-II, giving 39 in 65% yield. Interestingly, however, the reaction under the catalysis of G-II provided 38 in 71% yield as the sole isolable product, even in the presence of a cationic Au(I) species. The diastereoselectivity of transannular oxa-Michael addition of 38 was estimated to be 91:9 in a separate experiment, which was in marked contrast to that observed for model compounds 11b and 11c.

Takai methylenation[45] of 39 provided exo-olefin 40 (84%). Hydrogenation of the olefin and hydrogenolysis of the benzyl ether delivered alcohol 41 in 93% yield with 79:21 d.r. at C9. Finally, Mitsunobu reaction[46] of 41 with α,β-unsaturated carboxylic acid 42 afforded (+)-neopeltolide (2) in 94% yield. The present synthesis was completed in only 11 steps from (S)- or (R)-epichlorohydrin and in 23 total steps, and therefore represents the shortest synthesis of 2. (+)-9-epi-Neopeltolide (9-epi-2), a synthetic diastereomer of (+)-2 with potent antiproliferative activity,[44a] [b] was also synthesized from exo-olefin 40 through hydrogen-atom transfer reduction,[47] hydrogenolysis of the benzyl ether, and Mitsunobu esterification with 42. Importantly, the present synthesis was readily scalable and afforded (+)-2 and (+)-9-epi-2 in 40 and 63 mg in one batch, respectively. The present synthesis thus demonstrated the versatility of our macrocyclization/transannular pyran cyclization strategy.[48]


# 5

Conclusions

In this Account, we summarized our recent efforts on the development of a macrocyclization/transannular pyran cyclization strategy toward the concise total syntheses of marine macrolide natural products. Integration of three catalytic reactions, Meyer–Schuster rearrangement, macrocyclic ring-closing metathesis, and transannular oxa-Michael addition, in a consecutive manner enabled rapid access to tetrahydropyran-embedded macrolactones of various ring sizes from readily available linear precursors. The synthetic versatility of our newly developed strategy was highlighted by the total synthesis of (–)-exiguolide (13 steps) and (+)-neopeltolide (11 steps).[49]


#
#

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

The author thanks Daichi Mizukami, Kazuki Nakazato, Kei Iio, Mami Oda, and Yu Onodera for their significant contribution to the work summarized in this Account. The author acknowledges Professor Makoto Sasaki for supporting the preliminary phase of this work.


Corresponding Author

Haruhiko Fuwa
Department of Applied Chemistry, Faculty of Science and Engineering, Chuo University
1-13-27 Kasuga, Bunkyo-ku, Tokyo 112-8551
Japan.   

Publication History

Received: 02 September 2023

Accepted after revision: 26 September 2023

Accepted Manuscript online:
26 September 2023

Article published online:
02 November 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 Structures of marine macrolide natural products synthesized in our group
Zoom Image
Scheme 1 (A) Previous work: Tandem olefin cross-metathesis/intramolecular oxa-Michael addition. (B) Reaction sequence design in this work.
Zoom Image
Scheme 2 Synthetic blueprint toward (–)-exiguolide
Zoom Image
Scheme 3 Total synthesis of (–)-exiguolide by the macrocyclization/transannular pyran cyclization strategy; Part 1
Zoom Image
Scheme 4 Total synthesis of (–)-exiguolide by the macrocyclization/transannular pyran cyclization strategy; Part 2: (A) Outline of the synthesis of key building blocks, (B) Completion of the total synthesis.
Zoom Image
Scheme 5 Total synthesis of (+)-neopeltolide and (+)-9-epi-neopeltolide by the macrocyclization/transannular pyran cyclization strategy: (A) Outline of the synthesis of key building blocks; (b) Completion of the total synthesis.