Recent progress toward the total synthesis of humulanolides

Tetrahedron 73 (2017) 3289e3303

Contents lists available at ScienceDirect

Tetrahedron journal homepage: www.elsevier.com/locate/tet

Tetrahedron report 1139

Recent progress toward the total synthesis of humulanolides Jing-Chun Han a, c, 1, Xin Liu a, b, 1, Jing Zhao b, Shaoping Li b, **, Chuang-Chuang Li a, * a

Department of Chemistry, South University of Science and Technology of China, Shenzhen, 518055, China State Key Laboratory of Quality Research in Chinese Medicine, Institute of Chinese Medical Sciences, University of Macau, Macao, China c Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian, 116023, China b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 23 March 2017 Received in revised form 25 April 2017 Accepted 2 May 2017 Available online 3 May 2017

The humulanolides are a series of sesquiterpene lactones, most of which have unique and challenging structure. The humulanolides have exhibited anticancer activity. The combinations of fascinating structural motifs and promising pharmacological properties have prompted significant interest in the synthetic community. In this review, we provide a summary of recent progress regarding the total synthesis of humulanolides. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Natural products Humulanolide Asteriscanolide Total synthesis

Contents 1. 2. 3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3290 Isolation and bioactivity of humulanolides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3290 Synthesis of asteriscanolide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3291 3.1. Synthetic study of asteriscanolide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3291 3.1.1. Pirrung's study of the bicyclo[6.3.0]undecane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3291 3.1.2. Sarkar's study of the bicyclo[6.3.0]undecane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3291 3.1.3. Mehta's study of the bicyclo[6.3.0]undecane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3291 3.1.4. BookereMilburn's study of the asteriscanolide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3291 3.1.5. Lange's study of the asteriscanolide core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3292 3.2. Total synthesis of asteriscanolide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3293 3.2.1. Wender's first total synthesis of asteriscanolide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3293 3.2.2. Paquette's synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3293 3.2.3. Snapper's synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3294 3.2.4. Kraft's synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3295 3.2.5. Yu's synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3295 Total syntheses of other humulanolides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3296 4.1. Trost's synthesis of (
)-asteriscunolide D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3296 4.2. Fernandes's synthesis of (
)-asteriscunolide C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3297 4.3. Total synthesis of aquatolide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3297 4.3.1. Hiemstra's synthesis of aquatolide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3297 4.3.2. Gu's synthesis of aquatolide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3299 4.4. Ito's synthesis of naupliolide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3299 4.5. Li's collective synthesis of humulanolides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3300

* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (S. Li), [email protected] (C.-C. Li). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.tet.2017.05.009 0040-4020/© 2017 Elsevier Ltd. All rights reserved.

3290

5.

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3301 Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3301 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3301

1. Introduction The field of natural product total synthesis has evolved considerably since the pioneering work of Woodward in the 1940s.1 A large number of natural products have been synthesized during the past eight decades, some of which have been marketed commercially as therapeutic agents to treat a variety of conditions. Natural products can be divided into specific chemical families depending on their structural characteristics that contain several other members.2 It is therefore important for synthetic chemists to familiarize themselves with the diverse range of structural frameworks associated with different natural product classes, so that they can develop an understanding of the pathways responsible for the biosynthesis of these materials. This understanding can also benefit synthetic chemists in terms of assisting them in the design of increasingly efficient synthetic approaches.3 In this review, we have provided a summary of recent work towards the total synthesis of humulanolides, including an overview of the synthetic strategies typically used to access these molecules, which will hopefully inspire future synthesis.

asteriscanolidenol (15).13 Several humulanolide compounds have been reported to exhibit anticancer activity. For example, asteriscunolide A (1) induces apoptosis in several human cancer cell lines,14 whereas asteriscunolides AeD exhibit moderate cytotoxic activities against neoplastic cell lines, including A-549 (human lung carcinoma), HT20 (human colon carcinoma), and MEL-28 (human melanoma) cells. Furthermore, asteriscunolide D (4) exhibits greater cytotoxicity than cisplatin.15 To the best of our knowledge, there have been no reports pertaining to the biological activities of sesquiterpene lactones 5, 6, 8 and 10, although it is envisaged that these molecules will exhibit some interesting biological properties. The newly isolated humulanolides (11e15) reported by Triana's group were assessed in terms of their cytotoxicity against HL-60 and MOLT-3

2. Isolation and bioactivity of humulanolides The humulanolides are a structurally complex series of sesquiterpene lactones that consist of a bridged cyclic butenolide connected to an 11-membered ring or bicyclo[6.3.0]undecane ring, and several compounds belonging to this structural class have been reported to show excellent bioactivity (Fig. 1). One of the most important contributors to the field of humulanolide research was San Feliciano. The first member of the humulanolide family of natural products to be isolated and fully characterized was asteriscunolide A (1). This compound contains an 11-membered ring and was first isolated from asteriscus aquaticus in 1982.4 Subsequent work involving the same plant species allowed for the isolation of asteriscunolides B (2), C (3) and D (4),5 and the stereochemical characteristics of asteriscunolides A, B, C and D were later confirmed by X-ray diffraction analysis.6 Following on from this early work, Bohlmann reported the isolation of a new compound containing a rather unusual bicyclo[6.3.0]undecane skeleton, which was subsequently named 6,7,9,10-tetradehydroasteriscanolide (5).7 The saturated version of this compound, asteriscanolide (6), was isolated soon after.8 Aquatolide (8) was first isolated in 1989 and assigned as a rare ladderane substructure; however, further isolation work and X-ray crystallographic analysis revealed this incorrect assignment, and the structure was subsequently revised to the uncommon bicyclo[2.1.1]hexane core shown in the figure.9 6,7,9,10-Tetrahydroasteriscunolide (9) was isolated in 2001 and its structure was initially established by two-dimensional NMR spectroscopy.10 The structure of this material was later confirmed by X-ray diffraction analysis using synthetic material.11 Naupliolide (10) was isolated as a novel tetracyclic skeleton in 2006, together with asteriscunolides AeD and 6,7,9,10tetradehydroasteriscanolide.12 Most recently, Triana et al. reported the isolation and characterization of several new members of the humulanolide family, namely 6b,7b-epoxyasteriscunolide A (11), 2a,3a-epoxyasteriscunolide C (12), 6b-hydroxy-asteriscunolide A (13), 6b-ethoxy-asteriscunolide A (14) and

Fig. 1. Humulanolides.

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

leukemia cell lines, and the results revealed that 6b,7b-epoxyasteriscunolide A (11) showed moderate activity in both biological assays (IC50 values in the range of 4.1e5.4 mM). It is noteworthy that all of the humulanolides mentioned above were isolated from plants belonging to the same species, indicating that there could be a common intermediate involved in the biosynthesis of these compounds. San Feliciano and co-workers reported that asteriscunolide A (1) epimerized when it was heated at 200  C for 10 mins to give asteriscunolide C (3).5 More importantly, the isolation of compounds 5 and 10 from these plant species provides strong evidence of a biogenetic relation pathway involving the asteriscunolides and tricyclic sesquiterpene lactones such as asteriscanolide and related compounds. The cyclization of asteriscunolide C would give the intermediate carbocation I, which could proceed down two different pathways to give 5 or 10. Asteriscunolides AeD are also putative precursors of aquatolide, which could be formed via a photochemical [2 þ 2] reaction (Fig. 2).9b 3. Synthesis of asteriscanolide Asteriscanolide is the most famous member of the humulanolide family. Ever since it was first isolated in 1985, asteriscanolide has continued to attract considerable attention from the synthetic community because of its unique structural features. 3.1. Synthetic study of asteriscanolide 3.1.1. Pirrung's study of the bicyclo[6.3.0]undecane Synthetic studies toward the construction of the bicyclo[6.3.0] undecane core of asteriscanolide can be traced back to 1987, when Pirrung reported a novel synthetic method involving an intramolecular arene-alkyne photo-cycloaddition sequence. Notably, the irradiation of the substituted 5-aryl-1-pentyne compound 16 with UV light gave the bicyclo[6.3.0]undecane structure 17, which was treated with acid to give the corresponding ketone 18 containing the core skeleton of asteriscanolide (Scheme 1).16

3.1.2. Sarkar's study of the bicyclo[6.3.0]undecane Another synthetic study toward asteriscanolide was reported by Sarkar and co-workers in 1998, where the bicyclo[6.3.0]undecane core was constructed by sequential intramolecular ene-Michael and [3,3] sigmatropic rearrangement reactions.17 This synthesis started from the 1,6-diene 19, which underwent a thermal 5-(3,4)-ene cyclization reaction to give cyclopentanoid allylsilane 20 in 96% yield. The subsequent exposure of 20 to mCPBA, followed by the treatment of the crude epoxide product with silica gel in dichloromethane gave the desired lactone 21 in 46% yield. The methylenation of 21 with dimethyltitanocene18 afforded the corresponding enol ether 22. The thermolysis of 22 in a sealed tube gave the bicyclo[6.3.0]undecane core 23 in 36% yield (Scheme 2). 3.1.3. Mehta's study of the bicyclo[6.3.0]undecane In 1990, Mehta and co-workers reported a new strategy for the synthesis of asteriscanolide.19 The key step in this process involved the conversion of a [5-5-5]-tricyclic ring to a [5-8]-fused bicycle ring to allow for the construction of the highly functionalized bicycle[6.3.0]undecanedione core. The synthetic work (Scheme 3) started from dienone compound 25, which was synthesized from the pentacyclic-caged dione 24 by flash-vacuum pyrolysis (FVP).20 The thermal relocation of one of the enone moieties in 25 afforded the bis-enone 26 in 80% yield. The subsequent regioselective catalytic hydrogenation of 26 gave compound 27, which underwent a regioselective gem-dimethylation to give 28. The chemoselective thioketalisation of the least sterically hindered ketone in 28 gave the corresponding monothioketal 29, which was subjected to a reductive desulfurization reaction in metaleammonia milieu to yield a diastereomeric mixture (2:1) of alcohols 30a and 30b. The major diastereomer 30a was deoxygenated using the Barton protocol to give the tricyclic tetrasubstituted olefin 31, which was subjected to a rutheniumcatalyzed oxidation to give the 5,8-fused cis-bicyclic dione 32 with the core structure of asteriscanolide. 3.1.4. BookereMilburn's study of the asteriscanolide In 1994, BookereMilburn and co-workers reported their initial work toward the synthesis of asteriscanolide. Three years later, in 1997, they also reported a concise synthesis of 7-

Fig. 2. Biosynthetic pathway to bicyclo[6.3.0]undecane.

Scheme 1. Pirrung's study of the bicyclo[6.3.0]undecane.

3291

Scheme 2. Sarkar's study of the bicyclo[6.3.0]undecane.

3292

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

Scheme 3. Mehta's study of the bicyclo[6.3.0]undecane.

desmethylasteriscanolide.21 The key steps in this approach included an efficient intramolecular [2 þ 2] photocycloaddition and a Curtius rearrangement/oxidation/fragmentation sequence to yield the cyclooctane lactone core of asteriscanolide. The synthetic work in this particular study started with the reaction of dibromide 33 with 5,5-dimethylcyclopent-2-enol (34). The treatment of 34 with sodium hydride gave the corresponding alkoxide, which reacted with dibromide 33 to give ether 35 in 75% yield. The lithiumehalogen exchange reaction of 35 with t-butyllithium afforded the corresponding organolithium species, which was quenched with carbon dioxide to give the a,b-unsaturated carboxylic acid 36 in 81% yield. The stage was then set for the key intramolecular [2 þ 2] photocycloaddition reaction, which proceeded as anticipated under UV light irradiation conditions in acetonitrile using acetophenone as a sensitizer to give the desired cycloadduct 37 in 51% yield. Compound 37 was subsequently treated with diphenylphosphoryl azide in toluene, where it underwent a Curtius rearrangement to give isocyanate 38 in 89% yield (Scheme 4). The oxidation of isocyanate 38 with ruthenium tetroxide22 gave lactone 39, which was treated with sulfuric acid to give 7-desmethylasteriscanolide 40 as a 1:1 mixture of separable C9-epimers in 56% yield (over the two steps). To complete the synthesis of asteriscanolide, it was necessary to introduce a methyl group at C7. Unfortunately, however, the alkylation of 40 under

Scheme 4. BookereMilburn's study of the asteriscanolide.

various conditions afforded the undesired C9-monomethylated product 41, along with the dimethylated product 42, instead of the desired C7-monomethylated product. Compound 40 underwent a complex fragmentation process involving a base-catalyzed retro-Michael reaction to give enolate 43, which underwent an alkylation, followed by a transannular cyclization back to 41. This explains why the mono-alkylation of this substrate was observed at C9 rather than C7. 3.1.5. Lange's study of the asteriscanolide core Lange and Organ reported their work toward the synthesis of the core structure of asteriscanolide in 1996.23 The key step in this synthesis was an intermolecular [2 þ 2] photocycloaddition reaction, which was used to construct a tricyclic intermediate. This system subsequently underwent a fragmentation to give a [5-8] bicyclic ring system. The synthetic work in this particular study started from 2cyclopentenone and the trimethylsilyl derivative 46, which underwent an intermolecular [2 þ 2] photocycloaddition reaction to give the tricyclic adduct 47. The mono-methylation of 47 with

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

3293

iodomethane gave 48, which was treated with sodium borohydride to give lactone 49 via sequential reduction and lactonization reactions. Cleavage of the silyl ether in 49 with TBAF afforded the corresponding alcohol, which underwent a spontaneous fragmentation to generate compound 50. It is noteworthy that this strategy only required four steps to allow for the efficient construction of the tricyclic core of asteriscanolide (Scheme 5). 3.2. Total synthesis of asteriscanolide 3.2.1. Wender's first total synthesis of asteriscanolide The first enantioselective total synthesis of asteriscanolide was reported by Wender and co-workers in 1988.24 Wender's group applied a nickel-catalyzed intramolecular [4 þ 4] cycloaddition to an unactivated bis-diene to achieve the construction of the tricyclic core of asteriscanolide. Wender's total synthesis started with the addition of isopropenyl lithium to acrolein (Scheme 6) to give alcohol 51, which was esterified with isobutyric anhydride before being subjected to a LDA-mediated Claisen rearrangement to give the diene acid 52. A two-step procedure involving the reduction of the carboxylic acid moiety in 52 with LiAlH4, followed by the Swern oxidation of the resulting alcohol gave aldehyde 53 in 88% yield (over the two steps). The subsequent addition of lithium vinylacetylide to aldehyde 53 afforded the propargyl alcohol 54 as a racemic mixture. To prepare optically pure alcohol 54, Wender's group subjected the racemic material to a Swern oxidation, followed by the reduction of the resulting tert-alkyl alkynyl ketone with Darvon alcoholmodified LiAlH4 to give R-55 in excellent yield (97%, >98% ee). Then hydroalumination of the alkyne in R-55 with Red-Al, followed by a stannylation at
78  C with Me3SnCl, gave the vinyltin compound 57, along with the 1,3-rearranged isomer 56. Both of these compounds were treated with n-butyllithium, and the resulting organolithium compound was carboxylated at the internal position to provide the desired lactone 58. With the bis-diene 58 in hand, Wender's group proceeded to investigate the key nickel-catalyzed cycloaddition to the [5-8]-bicyclic ring system. In contrast to their previous studies,25 compound 58 had substituents on both dienes, an electron-withdrawing group at the internal position of one diene and one olefin in a pre-existing ring. Despite

Scheme 6. Wender's synthesis of (þ)-asteriscanolide.

these challenges, compound 58 underwent the desired intramolecular cycloaddition reaction to give compound 59 in 67% yield with excellent stereoselectivity. Having succeeded in their construction of the bicyclo[6.3.0] undecane core of asteriscanolide, Wender's group proceeded toward the completion of their total synthesis. The conjugated reduction of the unsaturated lactone in 59 with copper hydride afforded the saturated lactone 60 in 74% yield with good stereoselectivity. Finally, the selective hydroboration of the alkene moiety in 60, followed by an in situ oxidation with chromium,26 provided (þ)-asteriscanolide with the correct stereochemistry at C7.

Scheme 5. Lange's study of the asteriscanolide core.

3.2.2. Paquette's synthesis The second total synthesis of asteriscanolide was reported in 2000 by Paquette and co-workers, who used two key steps to achieve the construction of its bicyclo[6.3.0]undecane core.27 The first of these two steps was a MichaeleMichael reaction sequence, allowing for the enantioselective formation of the two fivemembered rings of the target molecule. The second step was a

3294

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

ring-closing metathesis reaction to provide the eight-membered ring. This sequence was initiated from 61, which was subjected to a lithiumehalogen exchange reaction, followed by the condensation of the resulting vinyllithium reagent with (S)-(
)-menthyl p-toluenesulfinate to give compound 62. The MichaeleMichael reaction of 62 was investigated under a variety of conditions, and the desired product 63 was obtained as the major product in 38% yield when 62 was treated with potassium carbonate in THF. Compound 63 was then treated with Raney nickel to allow for the cleavage the carbonesulfur bond, as well as the selective hydrogenation of the C]C double bond to give 64. Paquettes's group subsequently conducted sequential enolation/triflation and Stille coupling reactions to give diene 66 in excellent yield. The reduction of 66 with LiAlH4, followed by the conversion of the resulting primary alcohol to the corresponding iodide, gave 67 in good yield. The coppercatalyzed substitution of the iodide moiety in 67 with methallylmagnesium chloride proceeded smoothly to afford 68 in 98%

Scheme 7. Paquette's synthesis of (þ)-asteriscanolide (6).

yield (Scheme 7). Paquette's group then proceeded to investigate the second key step in their synthetic strategy. In the presence of Grubbs' catalyst, compound 68 readily underwent a ring-closing metathesis reaction to give compound 69 in 93% yield. Compound 69 was subsequently subjected to a selective photooxygenation reaction according to the conditions reported by Hasty and Kearns,28 followed by the reduction of the resulting intermediate, giving 70 in 61% overall yield. The DesseMartin periodinane oxidation of the alcohol, followed by the Pd/C-catalyzed hydrogenation of the exocyclic C]C double bond, afforded the desired compound 71 in good yield. The regioselective oxidation of 71 with ruthenium tetroxide, gave (þ)-asteriscanolide (6) in 63% yield. 3.2.3. Snapper's synthesis Snapper and co-workers also reported a total synthesis of (þ)and (
)-asteriscanolide in 2000, which proceeded over a nine-step linear sequence.29 Snapper's group used two sequential reactions as the key steps in their construction of the core structure of

Scheme 8. Snapper's synthesis of (þ)-asteriscanolide.

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

3295

asteriscanolide. The first of these reactions was an intramolecular cyclobutadiene cycloaddition, which provided facilely access to a highly functionalized trisubstituted cyclobutene. This cyclobutene system was then subjected to a ring-opening metathesis cascade with ethylene, followed by a Cope rearrangement to give the core structure of 6. This work started from the iron-complexed cyclobutadiene 75, which was prepared by the photolysis of the commercial pyrone 74 (Scheme 8). The subsequent reduction of 75 using a mixture of LiAlH4 and BF3$Et2O resulted in the formation of 76, which was subjected to an electrophilic aminomethylation reaction to give the p-substituted cyclobutadiene complex 77, together with the corresponding o-substituted product (67% yield, 3:1). Synthon 73 was synthesized from ketone 72 through sequential Saegusa oxidation and CBS reduction reactions in 56% yield and 94% ee. Compound 77 was then methylated, and the resulting quaternary amine reacted with alcohol 73 to give ether 78. The subsequent intramolecular cyclobutadiene cycloaddition of 78 afforded compound 79 in 63% yield, which was treated with the Grubbs second-generation catalyst in refluxing benzene under an atmosphere of ethylene to give cyclooctadiene 81 in 74% yield. This process proceeded via the dialkenyl cyclobutane intermediate 80, which underwent a Cope rearrangement to provide the desired tricyclic ring system in 81. The oxidation of 81 with PCC provided Wender's lactone 59 in 79% yield, which was subjected to Wender's sequence to complete Snapper's total synthesis of (þ)-asteriscanolide (6).

3.2.4. Kraft's synthesis Kraft and co-workers reported their total synthesis of asteriscanolide in 2000.30 The key steps in this synthesis included an intermolecular PausoneKhand cycloaddition and a ring-closing metathesis reaction. It is noteworthy that the latter of these two steps resulted in the formation of an eight-membered ring with an “in-out” intrabridgehead relationship. Kraft's synthesis started with the preparation of alkynoate 82 from 3-butyn-1-ol (Scheme 9). The treatment of 82 with dicobalt octacarbonyl furnished the desired hexacarbonyldicobalt complexed alkyne 83, which underwent an intermolecular PausoneKhand cycloaddition in DCM to give the highly functionalized cyclopentenone 84 in good yield with high regioselectivity. Methylation at the a-position relative to the carbonyl group, followed by the removal of the TBS group with HF/pyridine, gave ketone 85, which was reduced to the corresponding allylic alcohol. Subsequent lactonization and TBS-protection steps afforded compound 86, which was hydrogenated over Pd/C to allow for the reduction of the double bond to afford the desired all-cis diastereomer 88 (90%). The reduction of lactone 88 gave a mixture of lactols, which underwent a Wittig reaction to give the corresponding alkene 89. The Jones oxidation of the primary alcohol gave the corresponding carboxylic acid, which was heated at reflux with 2 N HCl in acetone to give the bicyclic lactone 90. The subsequent ozonolysis of alkene 90 gave the corresponding aldehyde 91, which was treated with (E)-crotylSnnBu3 to give (syn, syn)alcohol 92a, along with its isomer 92b (92a:92b ¼ 8:1). The protection of the free alcohol moiety in 92a as a TES ether, followed by an allylation reaction, gave the required RCM precursor 93. The subsequent treatment of diene 93 with Grubbs' catalyst (50 mol%) in CH2Cl2 gave the inside-outside tricycle 94 in 92% yield. The desilylation of 94 with tetrabutylammonium (triphenylsilyl) difluorosilicate (TBAT) in refluxing acetonitrile gave alcohols 96 (30%) and 95 (62%). The last two steps in this synthesis involved the hydrogenation of the double bond, followed by the oxidation of the secondary alcohol to give a ketone using Ley oxidation condition to afford (±)
asteriscanolide.

Scheme 9. Krafft's total synthesis of (±)-asteriscanolide.

3.2.5. Yu's synthesis Yu and co-workers published their total synthesis of (þ)-asteriscanolide in 2011.31 The key steps in this particular synthesis included an elegant Rh(I)-catalyzed [(5 þ 2) þ 1] cycloaddition for

3296

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

the efficient construction of the [6.3.0] carbocyclic core, as well as an alkoxycarbonyl radical annulation to afford the bridging butyrolactone ring (Scheme 10). The Rh(I)-catalyzed [(5 þ 2) þ 1] cycloaddition of enevinylcyclopropanes to CO was developed by Yu's group.32 Their total synthesis began with the catalytic asymmetric alkynylation of aldehyde 97 with cyclopropylacetylene (98) in the presence of Zn(OTf)2 and chiral ligand 100, which gave propargylic alcohol 99 in 90% yield and 94% ee. The reduction of the triple bond in 99 with Red-Al provided the corresponding allylic alcohol, which was protected with a TBS group to yield ene-vinylcyclopropane 101 as the key precursor. The subsequent treatment of 101 with [Rh(CO)2Cl]2 (5 mol%) in toluene gave ketone 102 in 70% yield with excellent diastereoselectivity (dr > 95:5). However, the hydrogen atom at C1 was in a trans configuration with respect to the bridgehead hydrogen atoms at C2 and C9. The subsequent enolation of ketone 102 with trifluoromethanesulfonic anhydride and 2,6-di-tert-butyl-4-methylpyridine gave the corresponding enol triflate 103, along with its regioisomer (88:12). Compound 103 was then subjected to an iron-catalyzed cross-coupling reaction to give cyclooctadiene 104 in 58% yield over two steps. The selective epoxidation of the trisubstituted olefin with mCPBA afforded epoxide 105 with excellent diastereoselectivity (dr > 95:5). This material was then treated with diethylaluminum 2,2,6,6tetramethylpiperidide, resulting in the regioselective formation of 106 in 86% yield over two steps. Then authors investigated the possibility of inverting the configuration at C1. The protection of the free alcohol in 106 with PMBCl/NaH, followed by the deprotection of the TBS group with TBAF, gave 107. The oxidation of the alcohol in 107 to the corresponding ketone and the reduction of this ketone with a bulky reducing reagent (DIBAL-H) provided 108 with the desired configuration. Yu's group then studied the alkoxycarbonyl radical annulation for the construction of the bridged butyrolactone ring. Selenocarbonate 109 was synthesized by the reaction of 108 with triphosgene and PhSeH. The subsequent radical cyclization of 109 in the presence of AIBN and n-Bu3SnH gave the desired tricyclic compound 110 as a single stereoisomer in 95% yield. Unfortunately, the configuration at C3 was the opposite of that found in the target molecule. In an attempt to complete their total synthesis, Yu's group investigated the possibility of epimerizing the C3 stereocenter in 110. Following the failure of a deprotonation/protonation sequence, they turned to an alternative strategy. The reduction of the lactone in 110 with DIBAL-H gave a mixture of hemiacetals, which were smoothly converted to enol ether 111 with methanesulfonyl chloride and triethylamine in 85% yield over two steps. The subsequent hydrogenation of 111 over Pd/C allowed for the successful installation of the hydrogen at C3 with the correct configuration. This step also allowed for the cleavage of the PMB group and hydrogenation of the exo double bond with 66:34 diastereoselectivity. The subsequent treatment of this mixture with DesseMartin periodinane gave two separable tricyclic ketones, including 114 with the desired configuration at C7 and its C7 epimer 113. Pleasingly, the enolation of 113, followed by the acidic hydrolysis of the resulting enol triflate, allowed for the partial formation of the desired tricyclic ketone 114. The final step in this total synthesis was the same as that reported by Paquette for the installation of the last carbonyl group, thereby completing the total synthesis of (þ)-asteriscanolide. 4. Total syntheses of other humulanolides 4.1. Trost's synthesis of (
)-asteriscunolide D Scheme 10. Yu's total synthesis of (þ)-asteriscanolide.

With the exception of asteriscanolide, prior to 2010, there were

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

3297

very few reports pertaining to the synthesis of other compounds belonging to the humulanolide family. Trost and co-workers completed the first total synthesis of asteriscunolide D in 2012 according a nine-step linear sequence that avoided the use of protecting group.33 The key step in this synthesis was a dimethyl(methylthio)sulfonium tetrafluoroborate (DMTSF)-mediated cyclization for the construction of the 11-membered ring in the target molecule. The initial absolute stereochemical configuration was established using a Zn-ProPhenol-catalyzed enantioselective addition, which provided a g-hydroxy propiolate moiety as a handle for the formation of the butenolide via a Ru-catalyzed alkene
alkyne coupling. Trost's synthesis of (
)-asteriscunolide D started from dimer 115, which was subjected to a HWE olefination to give enone 116. The Swern oxidation of the alcohol in 116 gave the corresponding aldehyde 117, which underwent an enantioselective addition reaction to give g-hydroxy propiolate 118 with good yield and enantiopurity. The treatment of 118 with catalytic CpRu(CH3CN)3PF6 and allyl alcohol provided direct access to butenolide 119 bearing the required aldehyde appending. Sequential thioacetal and Z-selective TMS-enol ether forming reactions afforded substrate 121 for the DMTSF-mediated cyclization test. The cyclization of compound 121 provided the medium-sized ring with nearly complete diastereoselectivity in 32%e42% yield. The final stage in this synthesis involved a Lewis acid-promoted thioether elimination to provide (
)-asteriscunolide D as the only product (Scheme 11).

4.2. Fernandes's synthesis of (
)-asteriscunolide C In 2013, Fernandes and co-workers reported the total synthesis of (
)-asteriscunolide C.34 The key step in this strategy was a late stage ring-closing metathesis reaction, which allowed for the conversion of a 12-membered ring to a more strained 11membered ring. Notably, this synthetic route was convergent. Fernandes's synthesis started from the commercially available 1,5-pentane diol (Scheme 12). The mono-protection of 1,5-pentane diol gave 123, which was subjected to sequential oxidation and amethylenation reactions to give aldehyde 124. The Pinnick oxidation of 124 gave the acid fragment 125 in high yield. The MOMprotection of D-pantolactone gave compound 126 in 95% yield. The subsequent reduction of the lactone ring gave the corresponding aldehyde, which underwent a Wittig olefination to give 127 in 85% yield over two steps. The Swern oxidation of 127, followed by a HWE-olefination, gave the unsaturated ester 128 in 87% yield over two steps. The subsequent treatment of 128 with the carbanion of (EtO)2P(O)CH2CH3, followed by the removal of the MOM group, gave ketophosphonate 129 in 88% yield over two steps. With the two key fragments in hand, Fernandes's group used Yamaguchi conditions to couple them together to give ester 130 in 94% yield. They then proceeded to investigate several ring closing conditions. Following the failure of a RCMeHWE sequence, the authors investigated the possibility of reversing this sequence. The acid-mediated deprotection of the TBS group gave alcohol 131 in 92% yield. The subsequent DMP oxidation of the alcohol to the corresponding aldehyde, followed by an intramolecular HWEolefination resulted in an inseparable 1:1 mixture of compounds 132 and 133 in 66% yield. This mixture was then treated with Grubbs II catalyst to give asteriscanolide C in 90% yield from 132. However, this reaction failed to afford any asteriscunolide D from 133.

Scheme 11. Trost's synthesis of (
)-asteriscunolide D.

4.3. Total synthesis of aquatolide Aquatolide, which consists of a highly strained bicyclo[2.1.1] hexane structure, represents a challenging target for synthetic chemistry. Prior to 2015, there were no reports in the literature pertaining to the synthesis of this molecule. Since then, only two groups have completed the total synthesis of this complex molecule, including Heimstra's group35 in 2015 and Gu's group36 in 2016. Both of these groups adopted a similar strategy involving the early construction of the bicyclo[2.1.1]hexane, followed by a latestage ring closing reaction to give the eight-membered ring.

4.3.1. Hiemstra's synthesis of aquatolide The first total synthesis of aquatolide was reported by Hiemstra and co-workers in 2015.35 The key step in this synthesis was the intramolecular [2 þ 2]-photocycloaddition of an allene onto an a,bunsaturated d-lactone to give the bicyclo[2.1.1]hexane skeleton. Another important step was an intramolecular Mukaiyama-type aldol reaction to close the challenging eight-membered ring.

3298

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

Scheme 13. Hiemstra's total synthesis of aquatolide.

Scheme 12. Fernandes's synthesis of (
)-asteriscunolide C.

The synthesis of this compound started with the preparation of acid 136 (Scheme 13). The crossed-aldol reaction of propynal and isobutyraldehyde with KOtBu and Ti(OiPr)4 gave the corresponding aldehyde, which was immediately converted to the corresponding methyl acetal 137. The esterification of the acid moiety in 137 with alcohol 136 using dicyclohexylcarbodiimide and N,N4-dimethylaminopyridine, followed by a deprotection step, gave ester 138 in 92% yields over two steps, which was subjected to an intramolecular HWE reaction to give pentenolide 139. Finally, the

acetylene was readily converted to the allene 140 for the intramolecular [2 þ 2]-photocycloaddition. The key reaction was then investigated. The irradiation of 140 with 300-nm light resulted in the formation of the desired product 141 with the required bicyclo [2.1.1]hexane skeleton in 77% yield. The subsequent hydroboration of the sterically hindered terminal alkene in 141 with BH3 and H2O2 gave the primary alcohol 142 in 36% yield with good stereoselectivity. The low yield of this reaction was attributed to the competitive reduction of the carbonyl group of the lactone. Alcohol 142 was then oxidized before being treated with ethylmagnesium bromide to give a mixture of secondary alcohols, which were applied directly to a hydrogenolysis reaction. The crude product was oxidized with excess DMP to produce ketoaldehyde 143 as a single product in 49% yield over four steps. The aldehyde moiety in 143 was then chemoselectively converted to the corresponding methyl acetal 144 in 81% yield using cerium(III) chloride. The regioselective enolate silylation of the ketone in 144 gave the crude silyl enol ether, which was treated with BF3$Et2O to provide a mixture of stereoisomers, followed by an elimination in the presence of p-toluenesulfonic acid, to complete the first

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

3299

reported total synthesis of aquatolide with 59% yield of the last three steps. 4.3.2. Gu's synthesis of aquatolide In 2016, Gu and co-workers reported the total synthesis of aquatolide using a photoinduced Wolff ring contraction reaction for the construction of challenging bicyclo[2.1.1]hexane core.36 They also used a late-stage intramolecular NozakieHiyamaeKishi vinylation reaction to allow for the construction of the eightmembered enone. Gu's synthesis started from the known bromo ester 145, which was readily prepared from 2,5-norbornadiene.37 The elimination of hydrogen bromide from 145 using DBU gave the corresponding alkene, which was alkylated with iodide 146 to provide compound 147 stereoselectively. The subsequent reduction of the ester moiety in 147 with LiAlH4, followed by sequential ketal deprotection and silylation reactions gave ketone 148 in 61% overall yield. The adimethylation of 148 proceeded smoothly via a two-step process to give compound 149 in 99% yield. The reduction of the carbonyl group in 149 with LiAlH4, followed by an acid-mediated desilylation, gave diol 150 in 91% yield over two steps. The initial treatment of this diol with TEMPO and PhI(OAc)2, followed by the addition of N-hydroxylamine 152 after the complete consumption of diol 150, gave the tricyclic compound 153 in 63% yield (Scheme 14). Compound 153 was then subjected to sequential dihydroxylation and oxidation reactions to give the dicarbonyl compound 154 in 93% overall yield. The condensation of compound 154 with tosyl hydrazine, followed by a detosylation step over basic Al2O3 gave the diazo compound 155, which was irradiated with a 125-W highpressure mercury lamp in a mixture of THF and H2O to give the bicyclo[2.1.1]hexane compound 156 in 80% yield. The chemoselective reduction of the carboxylic acid moiety in 156 with NaBH4 proceeded via a mixed anhydride intermediate, and the resulting alcohol was protected with TBSOTf to give 157. The reductive cleavage of the benzyl group using Pd/C, followed by the oxidation of the resulted free hydroxyl group, gave aldehyde 158, which was converted to 159 in moderate yield via the Bestmann (Z)-vinyl iodide.38 The acid-mediated cleavage of the TBS group in 159, followed by the oxidation of the alcohol, afforded aldehyde 160 in 41% yield, along with the undesired isomer 161 in 53% yield. Pleasingly, stereoisomer 161 could be partially converted to the required aldehyde 160 using Et3N and silica gel. Finally, the treatment of 160 with CrCl2 in the presence of a catalytic amount of NiCl2 in diluted DMSO, followed by the oxidation of the resulting alcohol, gave aquatolide in 43% yield for last two steps, along with the starting material 160 (50%). 4.4. Ito's synthesis of naupliolide

Scheme 14. Gu's synthesis of aquatolide.

Naupliolide is another synthetically challenging molecule belonging to the humulanolide family. In terms of its structural characteristics, this compound consists of a three-membered ring, a five-membered ring, an eight-membered ring and a butyrolactone. Ito and co-workers reported the first total synthesis of this molecule in 2016.39 The key steps in their synthesis included a SimmonseSmith cyclopropanation of an allyl alcohol, a radical cyclization of an aldehyde with a cyclopropane ring and the construction of an eight-membered ring using a ring-closing metathesis reaction. Ito's synthesis started with the stereoselective construction of the tricyclic lactone (Scheme 15). The known cyclic acetal 16240 was oxidized using DesseMartin periodinane to give the corresponding aldehyde, which was subjected to a Z-selective HWE reaction to give the unsaturated ester 163 as an inseparable mixture of cis and trans isomers at the phenyl group. The reduction of the ethyl ester

3300

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

thermodynamically more stable trans isomer. A similar result also occurred with 164b. The TBDPS protection of the free hydroxy group in 165, followed by the reduction of the benzylidene acetal, gave 167 as a single diasteroisomer. The Swern oxidation of 167, followed by the Z-selective olefination of the resulting aldehyde, gave the unsaturated ester 168 in 74% yield as a single isomer. The cleavage of the TBDPS group, followed by the oxidation of the resulting alcohol, gave the radical cyclization precursor 169, which was treated with SmI2 in t-BuOH, affording the tricyclic lactone 170 as a single isomer in 61% yield. The reductive cleavage of the benzyl group using Pearlman's catalyst, followed by butenylation at the a-position of the carbonyl group, gave the undesired alcohol 172 (dr ¼ 19:1) in 56% yield. Following an extensive period of screening, Ito's group found that the formation of the enolate with LDA, followed by protonation with solid silica gel, gave the best result for the formation of the desired alcohol 173 over the undesired 172 in a 5:1 ratio. The Swern oxidation of 173, followed by the addition of isopropenylmagnesium bromide to the resulting aldehyde, gave two separable diastereoisomers 174a and 174b (ratio of 2:1) at the secondary hydroxy group in 58% overall yield over two steps. This mixture was subjected to a RCM reaction using the Grubbs secondgeneration catalyst to afford the tetracyclic compounds 175 as a mixture in 50% yield. Finally, the Swern oxidation of 175 resulted in the formation of (±)-naupliolide in 76% yield. 4.5. Li's collective synthesis of humulanolides

Scheme 15. Ito's synthesis of naupliolide.

moiety in 163 with DIBAL-H afforded the cis and trans isomers 164a and 164b in 43% and 42% yields, respectively. The SimmonseSmith cyclopropanation of cis-164a bearing the Z-olefin afforded cis-165a and trans-165b in 18% and 54% yields, respectively. This result demonstrated that the phenyl group in the 1,3-dioxolane epimerized under the cyclopropanation conditions to give the

In 2014, Li and co-workers reported their collective synthesis of several humulanolides. Li's group synthesized seven different humulanolides in 7e9 steps without the need for protecting groups.11 The key step in this strategy was a ring-closing/ringopening/ring-closing metathesis cascade reaction for the construction of the challenging 11-membered ring and bridged butenolide moieties in asteriscunolide D. A biosynthetic method was used to synthesize the remaining six natural products, including photoinduced isomerization, transannular Michael and transannular MoritaeBayliseHillman-type reactions. Li's synthesis started from the known diol 177,41 which was prepared in two steps from D-pantolactone.42 The regioselective oxidation of the primary alcohol in 177 with TEMPO/BAIB afforded the corresponding aldehyde 178 in 75% yield, which was subjected to an olefination reaction with Wittig reagent 179 to provide the desired diene 180 in 75% yield. The treatment of the Weinreb amide 180 with isopropenylmagnesium bromide afforded triene 181 in 63% yield, which reacted with mixed anhydride 182 under Yamaguchi's condition to give ester 183 in 67% yield (Scheme 16). With tetraene 183 in hand, Li's group proceeded to investigate the key RCM/ROM/RCM cascade reaction. Compound 183 readily underwent the RCM/ROM/RCM cascade reaction in the presence of the HoveydaeGrubbs second-generation catalyst in refluxing toluene to afford (
)-asteriscunolide D in 36% yield. A biosynthetic approach was then adopted to investigate the use of (
)-asteriscunolide D for the collective synthesis of several other humulanolides. The regio- and stereoselective hydrogenation of asteriscunolide D with Wilkinson's catalyst afforded 6,7,9,10tetrahydroasteriscunolide exclusively in 99% yield. This compound could be considered as the transannular Michael reaction precursor of asteriscanolide. Pleasingly, the treatment of this compound with DBU resulted in the formation of asteriscanolide. For the synthesis of tetradehydroasteriscanolide, the transannular MoritaeBayliseHillman-type reaction of asteriscunolide D in the presence of NaOMe gave compound 184, which underwent a BF3$Et2O-mediated elimination reaction to give 6,7,9,10-

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

3301

Table 1 Total syntheses of different humulanolides.

Scheme 16. Li's collective synthesis of seven humulanolides.

tetradehydroasteriscanolide in 65% yield over two steps. Furthermore, the irradiation of asteriscunolide D in a photochemical reactor resulted in asteriscunolides AeC. This biosynthetic study of naupliolide and aquatolide was proceeded by the use of various Lewis acid and irradiation conditions, which failed to provide anything of interest. Li and co-workers developed a highly concise and asymmetric synthetic strategy for the construction of a series of humulanolides in 7e9 steps from the commercially available compound D-pantolactone. Critically, this strategy avoided the need for any protecting groups, allowing for the first reported synthesis of 6,7,9,10tetrahydroasteriscunlide and 6,7,9,10-tetradehydroasteriscanolide. 5. Conclusions Ever since they were first isolated, the humulanolides have continued to attract considerable attention from the synthetic chemistry community, as exemplified by the large number of total syntheses reported for different members of this family (Table 1).

Years

Group

Target

Steps

Yield

1988 2000 2000 2000 2011 2012 2013 2014

Wender Paquette Snapper Krafft Yu Trost Fernandes Li

2015 2016 2016

Hiemstra Ito Gu

Asteriscanolide Asteriscanolide Asteriscanolide Asteriscanolide Asteriscanolide Asteriscunolide D Asteriscunolide C Asteriscunolide D Asteriscunolide A Asteriscunolide B Asteriscunolide C 6,7,9,10-Tetra-hydroasteriscunlide Asteriscanolide 6,7,9,10-Tetrade-hydroasteriscanolide Aquatolide Naupliolide Aquatolide

13 13 9 19 19 9 12 7 8 8 8 8 9 9 16 18 22

2.7% 4.2% 3.9% 12% 3.8% 9.8% 16% 5.3% 2.4% 0.95% 0.95% 5.2% 3.3% 3.4% 2.2% 1.1% 1.1%

Asteriscanolide, in particular, has been the subject of various synthetic studies, including five total syntheses. It is noteworthy that these works have featured several elegant transformations, including intramolecular [4 þ 4] cycloaddition, fragmentation, RCM, [(5 þ 2) þ 1] cycloaddition and transannular Michael addition reactions, all of which have provided a platform for the synthesis of other complex natural products containing eight-membered rings. Asteriscunolides have also received considerable attention as a result of the interesting biological activities. Three total syntheses have been reported for the construction of these systems involving the use of several interesting transformations, including a dimethyl(methylthio)sulfonium tetrafluoroborate (DMTSF)-mediated cyclization, HWE-RCM (12-membered ring to 11-membered ring) and RCM/ROM/RCM cascade reaction. Aquatolide represents one of the most challenging targets in the humulanolide family, as evidenced by the fact that the first total synthesis of this compound appeared almost 30 years after it was first isolated. Notably, the total synthesis of this compound involve the intramolecular [2 þ 2]-photocycloaddition of an allene, as well as a photoinduced Wolff ring contraction reaction. Naupliolide is another challenging molecule, the core of which has been synthesized using SimmonseSmith cyclopropanation and SmI2-promoted radical cyclization reactions. For natural product families, there can be no doubt that the development of a collective synthesis represents one of the most efficient strategies for the diversity-oriented construction of numerous analogs. In this regard, research towards the collective synthesis of humulanolides has provided access to seven members of the humulanolide family. Despite considerable progress in this area, new opportunities still remain that will continue to be exploited as new chemistries are developed. Acknowledgment This work was supported by the Natural Science Foundation of China (Grant nos. 21402084, 21522204, 21472081 and 21672095), Guangdong Science and Technology Department (2016A050503011), the Shenzhen Science and Technology Innovation Committee (JSGG20160301103446375 and KQTD2015071710315717). References 1. Woodward RB, Doering WE. J Am Chem Soc. 1944;66:849. 2. Han JC, Li CC. Synlett. 2015;26:1289e1304.

3302

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303

3. Shi YB, Gao SH. Tetrahedron. 2016;72:1717e1735. 4. San Feliciano A, Barrero AF, del Corral JMMiguel, Ledesma E, S anchezFerrando F. Asteriscunolide A: humulanolide from. Tetrahedron Lett. 1982;23(30):3097e3100. 5. San Feliciano A, Barrero AF, Medarde M, Miguel del Corral JM, Aramburu nchez-Ferrando F. Asteriscunolides A, B, C and D, the first humuAizpiri A, Sa lanolides; Two pairs of conformationally stable stereoisomers. Tetrahedron. 1984;40(5):873e878. 6. Feliciano Arturo San, Barrero Alejandro F, Medarde Manuel, et al. The stereochemistry of asteriscunolides. Tetrahedron. 1985;41(23):5711e5717. 7. El Dahmy S, Jakupovic J, Bohlmann F, Sarg TM. New humulene derivatives from asteriscus graveolens. Tetrahedron. 1985;41(2):309e316. 8. San Feliciano A, Barrero AF, Medarde M, et al. Asteriscanolide. A sesquiterpene lactone with a new natural skeleton. Tetrahedron Lett. 1985;26(19):2369e2372. 9. (a) San Feliciano AS, Medarde M, Delcorral JMM, Aramburu A, Gordaliza M, Barrero AF. Tetrahedron Lett. 1989;30:2851e2854; (b) Lodewyk MW, Soldi C, Jones PB, et al. J Am Chem Soc. 2012;134: 18550e18553. 10. Chaari A, Jannet HB, Mighri Z, Robinot C, Kunesch N. Isolation and structure determination of a new sesquiterpene lactone from Nauplius aquaticus. Nat Product Lett. 2001;15(6):419e423. 11. Han JC, Li F, Li CC. Collective synthesis of humulanolides using a metathesis cascade reaction. J Am Chem Soc. 2014;136(39):13610e13613. 12. Akssira Mohamed, Mellouki Fouad, Salhi Ali, et al. Naupliolide, a sesquiterpene lactone with a novel tetracyclic skeleton from Nauplius graveolens subsp. odorus. Tetrahedron Lett. 2006;47(38):6719e6721. 13. Triana J, Eiroa JL, Morales M, et al. Sesquiterpenoids isolated from two species of the Asteriscus alliance. J Nat Prod. 2016;79(5):1292e1297. vez F. Naturally 14. Negrín G, Eiroa JL, Morales M, Triana J, Quintana J, Este occurring asteriscunolide A induces apoptosis and activation of mitogenactivated protein kinase pathway in human tumor cell lines. Mol Carcinog. 2010;49(5):488e499. lez AG, García-Gr 15. Rauter AP, Branco I, Bermejo J, Gonza avalos MD, Feliciano AS. Bioactive humulene derivatives from Asteriscus vogelii. Phytochemistry. 2001;56(2):167e171. 16. Pirrung Michael C. Intramolecular arene-alkyne photocycloadditions. J Org Chem. 1987;52(8):1635e1637. 17. Sarkar Tarun K, Gangopadhyay Pulak, Ghorai Binay K, Nandy Sandip K, Fang Jim-Min. Cyclopentanoid allylsilanes in synthesis: a facile construction of the 5e8 fused carbon framework of asteriscanolide. Tetrahedron Lett. 1998;39(45):8365e8366. 18. Petasis Nicos A, Bzowej Eugene I. Titanium-mediated carbonyl olefinations. 1. Methylenations of carbonyl compounds with dimethyltitanocene. J Am Chem Soc. 1990;112(17):6392e6394. 19. (a) Mehta G, Murty AN. J Org Chem. 1990;55:3568e3572; (b) Mehta G, Umarye JD. Tetrahedron Lett. 2001;42:8101e8104. 20. (a) Mehta G, Reddy AV, Srikrishna A. Tetrahedron Lett. 1979;20:4863e4866; (b) Mehta G, Srikrishna A, Reddy AV, Nair MS. Tetrahedron. 1981;37: 4543e4559. 21. (a) Booker-Milburn KI, Cowell JK, Harris LJ. Tetrahedron Lett. 1994;35: 3883e3886; (b) Booker-Milburn KI, Cowell JK. Tetrahedron Lett. 1996;37:2177e2180; (c) Booker-Milburn KI, Cowell JK, Harris LJ. Tetrahedron. 1997;53: 12319e12338. 22. (a) Carlsen PHJ, Katsuki T, Martin VS, Sharpless KB. J Org Chem. 1981;46: 3936e3938. For the oxidation of related cyclobutane containing ethers see:; (b) Kojima T, Inouye Y, Kakisawa H. Chem Lett. 1985:323e326; (c) Berkowitz WF, Amarasekara AS, Perumattam JJ. J Org Chem. 1987;52: 1119e1124. 23. Lange Gordon L, Organ Michael G. Use of cyclic b-keto ester derivatives in photoadditions. Synthesis of (±)-Norasteriscanolide. J Org Chem. 1996;61(16): 5358e5361. 24. Wender Paul A, Ihle Nathan C, Correia Carlos RD. Nickel-catalyzed intramolecular [4þ4] cycloadditions. 4. Enantioselective total synthesis of (þ)-asteriscanolide. J Am Chem Soc. 1988;110(17):5904e5906. 25. (a) Wender PA, Ihle NC. Tetrahedron Lett. 1987;28:2451e2454; b) Wender PA, Snapper ML. Tetrahedron Lett. 1987;28:2221e2224; (c) Wender PA, Ihle NC. J Am Chem Soc. 1986;108:4678e4679. 26. Piancatelli G, Scettri A, D'Auria M. Pyridinium chlorochromate: a versatile oxidant in organic synthesis. Synthesis. 1982;1982(04):245e258. 27. Paquette Leo A, Tae Jinsung, Arrington Mark P, Sadoun Aladin H. Enantioselective double Michael addition/cyclization with an oxygen-centered nucleophile as the first step in a concise synthesis of natural (þ)-Asteriscanolide. J Am Chem Soc. 2000;122(12):2742e2748. 28. Hasty Noel M, Kearns David R. Mechanisms of singlet oxygen reactions. Intermediates in the reaction of singlet oxygen with dienes. J Am Chem Soc. 1973;95(10):3380e3381. 29. Limanto John, Snapper Marc L. Sequential intramolecular cyclobutadiene cycloaddition, ring-opening metathesis, and Cope Rearrangement: total syntheses of (þ)- and (
)-Asteriscanolide. J Am Chem Soc. 2000;122(33): 8071e8072. 30. (a) Krafft ME, Cheung YY, Juliano-Capucao CA. Synthesis. 2000:1020e1026; (b) Kraft ME, Cheung YY, Abboud KA. J Org Chem. 2001;66:7443e7448. 31. (a) Liang Y, Jiang X, Yu ZX. Chem Commun. 2011;47:6659e6661; (b) Liang Y, Jiang X, Fu XF, et al. Chem Asian J. 2012;7:593e604.

32. (a) Wang YY, Wang JX, Su JC, et al. J Am Chem Soc. 2007;129:10060e10061; (b) Huang F, Yao ZK, Wang Y, Wang YY, Zhang JL, Yu ZX. Chem Asian J. 2010;5: 1555e1559; (c) Wender PA, Gamber GG, Hubbard RD, Zhang L. J Am Chem Soc. 2002;124: 2876e2877. 33. Trost BM, Burns AC, Bartlett MJ, Tautz T, Weiss AH. Thionium ion initiated medium-sized ring formation: the total synthesis of asteriscunolide D. J Am Chem Soc. 2012;134(3):1474e1477. 34. Fernandes RA, Chavan VP. A 12-membered to a strained 11-membered ring: first stereoselective total synthesis of (-)-asteriscunolide C. Chem Commun Camb Engl. 2013;49(32):3354e3356. 35. Saya JM, Vos K, Kleinnijenhuis RA, van Maarseveen JH, Ingemann S, Hiemstra H. Org Lett. 2015;17:3892e3894. 36. Wang Bin, Xie Yuanzhen, Yang Qin, Zhang Guozhu, Gu Zhenhua. Total synthesis of aquatolide: Wolff ring contraction and late-stage NozakieHiyamaeKishi medium-ring formation. Org Lett. 2016;18(20):5388e5391. 37. (a) Peel R, Sutherland JK. J Chem Soc Chem Commun. 1974:151e153; (b) Grieco PA, Pogonowski CS, Burke SD, et al. Am Chem Soc. 1977;99: 4111e4118. 38. (a) Bestmann HJ, Bomhard A. Angew Chem Int Ed. 1982;21:545e546; (b) Stork G, Zhao K. Tetrahedron Lett. 1989;30:2173e2174. 39. Abe H, Morishita T, Yoshie T, Long K, Kobayashi T, Ito H. The total synthesis of (±)-Naupliolide: a tetracyclic sesquiterpene lactone. Angew Chem Int Ed Engl. 2016;55(11):3795e3798. llar Geiser, Jime nez-V 40. Juaristi Eusebio, Díaz Francisco, Cue azquez Hugo A. Conformational analysis of 5-substituted 1,3-dioxanes. 7. Effect of lithium bromide Additiony,1. J Org Chem. 1997;62(12):4029e4035. 41. (a) Storer RI, Takemoto T, Jackson PS, Brown DS, Baxendale IR, Ley SV. Chem Eur J. 2004;10:2529e2547; (b) Dieckmann M, Kretschmer M, Li PF, Rudolph S, Herkommer D, Menche D. Angew Chem Int Ed. 2012;51:5667e5670. , Rey Allan W. Controlled diastereoselection in 2-lithio-1,3-dithiane 42. Roy Rene additions onto a-substituted g-lactols. Model studies toward bryostatins from (R)-pantolactone. Can J Chem. 1991;69(1):62e69.

Jing-Chun Han was born in Rizhao, Shandong province, P. R. of China. He received his B.Sc. in applied chemistry from Nanchang Hangkong University in 2008, and he completed his doctorate under the direction of Professors Zhen Yang and Chuang-Chuang Li at Peking University in 2013. He then spent two years (2013e1015) at the South University of Science and Technology of China as research fellow. He is currently a postdoctoral working with Professor Chuang-Chuang Li at the South University of Science and Technology of China. His research interests lie in the total synthesis of biologically active natural products.

Xin Liu was born in Jiangxi province, P. R. of China. He received his B.Sc. in applied chemistry from Jiangxi University of T.C.M in 2011, and obtained his M.Sc. degree in pharmacochemistry from Jinan University in 2014. Now, he is a PhD student of the joint program of the University of Macau (UM) and South University of Science and Technology of China (SUSTech) and his supervisors are the Prof. Shao-ping Li and Prof. Chuang-Chuang Li. The theme of his research program is the total synthesis of biologically active natural products.

Jing Zhao, was born in Chengdu, Sichuan Province of P. R. China. She received her B.Sc. in Chinese Medicines at Chengdu University of Chinese Medicines in 2004 and completed her doctorate under the direction of Professor Ping Li at China Pharmaceutical University in 2009. After two years of postdoctoral training with Professor Luqi Huang at China Academy of Chinese Medical Sciences. She joined University of Macau in 2011. Her current research interests include quality control of medicine and food dual purposes materials and online bioassay of their active components.

J.-C. Han et al. / Tetrahedron 73 (2017) 3289e3303 Shao-ping Li, was born in Feixi, Anhui Province of P. R. China. He completed his doctorate under the direction of Professor Ping Li at China Pharmaceutical University in 2000. After two years of postdoctoral training with Professor Quan Zhu at Nanjing University of Chinese Medicine and Professor Karl W. K. Tsim at the Hong Kong University of Science and Technology, he joined University of Macau in 2003. Her current research interests include discovery of active components and quality control of Chinese medicines, as well as herbal glycol-analysis.

3303 Chuang-Chuang Li, was born in Beiliu, Guangxi province P. R. of China. He received his B.Sc. in applied chemistry at China Agricultural University in 2001 and completed his doctorate under the direction of Professor Zhen Yang at Peking University in 2006. After two years of postdoctoral training with Professor Phil S. Baran at The Scripps Research Institute, he joined Peking University Shenzhen Graduate School in 2008. In 2014, he moved to the Department of Chemistry at South University of Science and Technology of China. His current research interests include the development of new synthetic methods and total synthesis of biologically active natural products.