syntheses of galbulimima alkaloids · 2 syntheses of galbulimima alkaloids several syntheses of...

REVIEW 3763

Syntheses of Galbulimima AlkaloidsSyntheses of Galbulimima AlkaloidsUwe Rinner,* Christoph Lentsch, Christian AichingerInstitute of Organic Chemistry, University of Vienna, Währinger Straße 38, 1090 Vienna, AustriaFax +43(1)4277 9521; E-mail: [email protected] Received 28 May 2010; revised 1 July 2010

SYNTHESIS 2010, No. 22, pp 3763–3784xx.xx.2010Advanced online publication: 15.09.2010DOI: 10.1055/s-0030-1258251; Art ID: E27410SS© Georg Thieme Verlag Stuttgart · New York

Abstract: The bark of the rain forest tree Galbulimima belgraveanahas been identified as a rich source of fascinating natural productsand so far 28 unique alkaloids have been isolated. The chemical in-terest in Galbulimima alkaloids started with the discovery of thepromising biological activities of himbacine. Fifteen years after thefirst total synthesis of himbacine, this class of natural product stillinspires, with the structurally more complex congeners himandrine,himgaline, and G.B. 13 as highly challenging synthetic targets. Thisreview article summarizes and discusses all syntheses of Galbulimi-ma alkaloids published to date.

1 Introduction2 Syntheses of Galbulimima Alkaloids2.1 Class I Alkaloids2.2 Class II Alkaloids2.3 Class III Alkaloids3 Conclusion

Key words: Galbulimima alkaloids, natural products, total synthe-sis, Diels–Alder reaction, heterocycles, fused-ring systems

1 Introduction

The rain forest tree commonly called white magnolia orpigeonberry ash (Figure 1), endemic in Northern Austra-lia and Papua New Guinea, was first botanically describedin 1887 by F. Mueller and was originally named Eupoma-tia belgraveana.1 Ever since this first disclosure, botanistshave argued about the correct taxonomic classificationand the plant was allocated different synonyms, the mostcommon name being Himantandra belgraveana.2,3 Todaythe name Galbulimima belgraveana is used and this namehas been given as synonym to Galbulimima baccata byauthors who consider the genus monotypic. Recently,Jessup reported that Galbulimima belgraveana is endemicto Papua New Guinea and Indonesia while Galbulimimabaccata is found in Northern Queensland, Australia.4 Therain forest tree is categorized as a member of the familyHimantandraceae in the order Magnoliales.

Traditional medicine of the native populations of PapuaNew Guinea, Malaysia, and Northern Australia made useof this rain forest tree. The bark was chewed – often incombination with leaves of Homalomena sp. – to inducevision and dream-like states, for divination and religiousrites, or it was swallowed to reduce abdominal pains.5–7

Powdered bark in combination with wild tobacco and gin-ger served as natural treatment for hair lice.7

In 1948, Webb recorded that the bark of Galbulimimabaccata is rich in alkaloids and several compounds wereisolated.8 In the following years, a total of 28 alkaloidswere revealed, of which 22 were structurally character-ized. The newly isolated compounds were either namedwith the prefix ‘him’, as the authors were under the im-pression that the botanically correct name for white mag-nolia was Himantandra belgraveana, or they were givennumbers. Based on their structural properties, the Galbu-limima alkaloids are subdivided in four classes; the struc-turally characterized alkaloids (classes I to III) areoutlined in Figure 2, and the miscellaneous compounds(class IV) include the alkaloids for which characterizationand thus categorization have not yet been achieved. Onlyrecently, Mander published the structures of three Galbu-limima alkaloids which had not been characterized previ-ously.9 While himgrine (5) possesses the structuralfeatures of a Galbulimima class I alkaloid, G.B. 16 (24) ismore complex and could be classified as a class III alka-loid despite the missing C9–C20 bond. G.B. 17 (25) clear-ly stands out, as the structure of this natural product isunique among Galbulimima constituents.

Atoms and ring systems are labeled in Figure 2. The num-bering for class I alkaloids is consistent throughout thechemical literature with only a few exceptions. There is nouniform system for the atom numbering of class II andclass III alkaloids, so we have adopted the method sug-gested by Ritchie and Taylor.10

Figure 1 Galbulimima baccata11

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.

3764 U. Rinner et al. REVIEW

Synthesis 2010, No. 22, 3763–3784 © Thieme Stuttgart · New York

The synthetic interest in Galbulimima alkaloids beganwith the discovery of their biological properties. Himba-cine (2) was identified as a potent muscarinic receptorantagonist12,13 and a possible candidate for the treatmentof senile dementia associated with Alzheimer’s dis-ease.14,15 Furthermore, derivatives of himbacine are cur-rently in clinical trial as antithrombotic agents.16,17

Despite the high potential of himbacine, the biologicalproperties of most other Galbulimima alkaloids remainunexplored.

A characteristic feature of all the Galbulimima alkaloidsis a trans-decalin system. While the piperidine ring is at-tached to the decalin moiety via a carbon tether in class Ialkaloids, the heterocycle is part of a complex fused-ringsystem in the class II and class III alkaloids. Althoughclass II and III alkaloids are structurally more complex,the main carbon skeleton remains basically identical forall members of this family of natural products.

The structure of Galbulimima alkaloids has been inten-sively studied. Degradation studies revealed the structureof himbacine (2)18,19 before the absolute configurationwas determined by X-ray crystal structure analysis.20 Sub-sequently, the structure of Galbulimima class II alkaloid

himbosine (7) was established via X-ray crystal structureanalysis of the corresponding hydrobromide21 and thestructures of himandrine (6) and himandridine (8) weredetermined by degradation studies and X-ray crystalstructure analysis.22,23 The structurally related Galbulimi-ma class II alkaloids were assigned based on these find-ings without proof of the absolute configuration.10,24

Because of the similarity of the carbon skeleton, it was as-sumed that subtypes of this family of natural productswould also share the absolute configuration of the decalinsystem and the structures of himbadine (23), himgaline(21) and G.B. 13 (22) were assigned accordingly.25,26 In2006, Movassaghi27 corrected the stereochemistry of G.B.13 from 2R to 2S and this report also motivated Willis andMander28 to reevaluate this group of natural products. Af-ter extensive X-ray studies, they were able to prove, un-ambiguously, that the C2 stereochemistry is S in all ofthese alkaloids. Apart from that, they found the C10 andC15 stereocenters of more complex class II and class IIIalkaloids to be antipodal to those in himbacine (2) and re-lated class I congeners. This highly interesting result is infull agreement with all total synthetic efforts published todate.

Uwe Rinner studied chem-istry at the Technical Uni-versity in Graz. Aftergraduation he moved toGainesville, Florida (USA)to pursue graduate studies inthe area of total synthesis ofAmaryllidaceae constitu-ents with Prof. Hudlicky. In

2003, he received is Ph.D.from the University of Flor-ida and moved to BrockUniversity, Ontario (Cana-da) for postdoctoral studies.After returning to Austria,Uwe Rinner joined the re-search group of Prof.Johann Mulzer and was ac-

tive in the field of diterpenechemistry before he startedhis own research group in2007. His current interestsinclude the synthesis ofditerpenes and the prepara-tion of other biologically in-teresting natural products.

Christoph Lentsch wasborn in 1981 in Zams, Aus-tria. He studied chemistry atthe University of Vienna,where he received his M.Sc.

in 2006 under the supervi-sion of Prof. WaltherSchmid. Currently he is car-rying out his Ph.D. studiesunder the guidance of Prof.

Johann Mulzer in the fieldof natural product total syn-thesis.

Christian Aichinger wasborn in 1984 in St. Pölten,Austria. He studied chemis-try at the University of Vi-enna, where he received hisM.Sc. in 2008 under the su-pervision of Prof. Michael

Widhalm, working on im-proved methods to accessbinaphthyl derivatives. Cur-rently, he is working to-wards his Ph.D. in theRinner group, studying thetotal synthesis of euphosali-

cin. His research interestsinclude transition-metal-mediated coupling reactionsand the synthesis of biologi-cally important naturalproducts and derivatives.

Biographical Sketches

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.

REVIEW Syntheses of Galbulimima Alkaloids 3765


In a first speculation about the biosynthesis of Galbulimi-ma alkaloids, Taylor and Ritchie18,26 suggested that thesenatural products are derived from either nine acetates andone pyruvate unit or ten acetates and a one-carbon unit.Movassaghi27,29,30 published a hypothetical biogenesiswhich explains the generation of class II and class III Gal-bulimima alkaloids. Mander suggested biosynthetic path-ways for the formation of G.B. 16 and G.B. 17.9 Asidefrom these reports, no further information about the bio-synthesis of Galbulimima alkaloids has appeared.

2 Syntheses of Galbulimima Alkaloids

Several syntheses of Galbulimima alkaloids have ap-peared in the literature over the last few years and these ef-forts are listed in Table 1. Despite the growing interest inthis highly fascinating class of natural products, synthesesof Galbulimima alkaloids have not been summarized pre-viously.

This review article concentrates on total and formal syn-theses of members of this interesting class of natural prod-ucts, and all of the syntheses outlined in Table 1 are

discussed within this article. Reports describing the prep-aration of derivatives or simplified models for biologicalactivity studies are not included. For clarity and better un-derstanding, key steps are depicted in blue color withinthe schemes.

2.1 Class I Alkaloids

Syntheses of class I alkaloids are discussed in detail be-low. All synthetic efforts utilize Diels–Alder reactions asthe key step to establish the decalin unit of the naturalproduct. As each of the approaches varies by the bondsformed during the cycloaddition step, different strategiesfor the construction of the Diels–Alder precursors wereapplied (Scheme 1).

Hart, Kozikowski, 1995 (Himbacine, Himbeline)

In 1995, Hart and Kozikowski31,42 reported the first totalsynthesis of himbacine via himbeline, as outlined inScheme 2. The key step in this synthetic effort is the for-mation of the trans-decalin system by an intramolecularLewis acid promoted Diels–Alder reaction of thioester 29.

Figure 2 Classification of Galbulimima alkaloids

Class I Galbulimima alkaloids:

O

NR

O

H

H

H

H

1 himbeline; R = H2 himbacine; R = Me

O

N

O

H H

H

3 himgravine

O

HN

O

H

H

H

H

4 himandravine

Class II Galbulimima alkaloids:

OBz OMe

N

MeO2C

HO

H

H

6 himandrine

R1 R2

N

MeO2C

R4

H

HR3

Alkaloid

7 himbosine8 himandridine9 G.B. 110 G.B. 211 G.B. 312 G.B. 413 G.B. 514 G.B. 615 G.B. 716 G.B. 817 G.B. 918 G.B. 1019 G.B. 1120 G.B. 12

OAcOCOPhOAcOAcOHOHOHOCOPhOCOPhOHOAcOAcOHOAc

OCOPhOMeOCOPhOAcOHOHOHOMeOMeOMeOMeOMeOHOAc

OAcOHOAcOAcOAcOCOPhOHOAcOHHHHHH

OAcOHOHOAcOAcOAcOAcOHOAcOHOHOAcOAcOAc

Class III Galbulimima alkaloids:

OH

N

HO

H

H

H

21 himgaline

O

NR

HO

H

H

H

22 G.B. 13; R = H23 himbadine; R = Me

Me

A B C

D

1

3 3a

5

8 99a

2

12

17

11

1

4

20

2

7

912

18

6

3

8

1517 AB

C

D

E

AB

C

D

E

O

N

O

H

HOH

Me

O

N

O

H

H

H

5 himgrine

24 G.B. 16

N

H

H

H

HO

O

H

O

25 G.B. 17

20

19

R1 R2 R3 R4

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



The reaction sequence started with cycloheptene (26)which was converted into 7,7-dimethoxyheptanal via ozo-nolysis in the presence of methanol following a proceduredescribed by Schreiber.43 Subsequent Wittig olefinationgave a,b-unsaturated ester 27 as a 2:1 mixture of geomet-rical isomers. Deprotonation of this material with lithiumdiisopropylamide afforded the corresponding enolate,which reacted with O-protected (S)-2-hydroxypropanal togive tetrahydropyranyl ether 28. By exposing alkene 28 tosunlight in the presence of iodine, the ratio of the desiredtrans-isomer could be increased to 32:1. The unsaturatedg-lactone was obtained by cleavage of the acid-labile tet-

rahydropyran group in methanol and elimination of thehydroxy functionality.

Installation of the thioester and formation of the trienewas achieved by cleavage of the acetal followed by Wittigolefination. Diels–Alder cyclization of triene 29 underthermal conditions afforded a 3:2 mixture of the desiredendo-cycloadduct 30 and the corresponding exo-deriva-tive. The endo/exo selectivity could be improved by addi-tion of Lewis acids. The best results (endo/exo = 20:1)were obtained with a heterogeneous promoter preparedfrom diethyl aluminum chloride and silica gel.

The synthetic strategy envisaged the introduction of thepiperidine ring via Julia–Lythgoe olefination and the fol-lowing steps were devoted to the installation of the sul-fone required for this olefination reaction. Reduction ofthe thioester with Raney nickel under concomitant stereo-selective hydrogenation of the remaining double bond inthe decalin system (C9–C9a), and conversion of the pri-mary alcohol into the corresponding tosylate and dis-placement with thiophenol gave sulfide 31. In order tocircumvent compatibility issues of a deprotonated sulfoneand a lactone present in one species, the lactone moiety in31 was converted into a methyl acetal by reduction withdiisobutylaluminum hydride and exposure to boron tri-fluoride etherate in methanol (the original erroneous ste-reochemical assignment of the acetal carbon atom31 wascorrected in the full paper published in 1997).42 Subse-quently, sulfur was oxidized by means of m-chloroperoxy-benzoic acid to afford sulfone 32 in good overall yield.The required coupling partner for the Julia–Lythgoe olefi-

Table 1 Syntheses of Galbulimima Alkaloids

Alkaloid class Compound Author Publication year Number of steps Overall yield

class I himbacine/himbeline Hart, Kozikowski31 1995 19 (himbeline)20 (himbacine)

5.4% (himbeline)3.8% (himbacine)

class I himbacine/himbeline Chackalamannil32 1996 11 (himbeline)12 (himbacine)


class I himbacine (formal synthesis) Terashima33 1999 18 12.6%

class I himandravine Chackalamannil34 2001 12 17.0%

class I himbacine (formal synthesis) De Clercq35 2002 13 6.6%

class I himbacine (formal synthesis) Sherburn36 2003 26 2.4%

class I himbacine/himbeline Baldwin37 2005 12 (himbeline)13 (himbacine)


class II himandrine Movassaghi29 2009 28 0.7%

class III (±)-G.B. 13 Mander38 2003 30 0.2%

class III G.B. 13/himgaline Chackalamannil39 2006 31 (G.B. 13)33 (himgaline)

0.5%0.3%

class III G.B. 13 Movassaghi27 2006 19 1.7%

class III ent-G.B. 13/ent-himgaline Evans40 2007 31 (ent-G.B. 13)32 (ent-himgaline)

1.0% (ent-G.B. 13)0.9% (ent-himgaline)

class III (±)-G.B. 13 Sarpong41 2009 18 1.2%

Scheme 1 Different strategies in the syntheses of himbacine (2)

O

NMe

H

O

H

H

H

H

2 himbacine

O

OHart/Kozikowski 1995De Clercq 2002Baldwin 2005

O

O

R

Chackalamannil 1996

O

O

O

O

Terashima 1999 Sherburn 2002

R R

R

R

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



nation reaction – aldehyde 36 – was prepared from com-mercially available (R)-piperidinecarboxylic acid (34).Reduction followed by N- and O-protection gave tert-bu-tyldimethylsilyl ether 35. After stereoselective methyla-tion, the silyl ether was removed and oxidation usingLey’s protocol44 completed the reaction sequence to alde-hyde 36.

The first synthesis of himbacine was completed by thecoupling of sulfone 32 and aldehyde 36, which after treat-ment with sodium-amalgam delivered trans-alkene 33. Itis interesting to note that the original synthetic plan de-vised by Hart and Kozikowski proposed a Julia–Lythgoeolefination between aldehyde 37 and sulfone 38 as shownin Scheme 3. When this protocol failed, the problem wassolved by preparing coupling partners with reversed func-tionalities.

Oxidation of the protected lactol with Jones reagent andcleavage of the Boc group on the piperidine nitrogen af-forded himbeline (1) which was converted into himbacine(2) by N-methylation under reductive alkylation condi-tions. Overall, himbeline was prepared in 17 steps and5.4% yield and himbacine was obtained in 18 steps and3.8% yield.

Chackalamannil, 1996 (Himbacine, Himbeline)

Shortly after the publication of the first synthesis of him-bacine by Hart and Kozikowski, Chackalamannil32 report-ed his route to the alkaloid. The key step is anintramolecular Diels–Alder reaction of a substrate whichbears the complete carbon framework of the natural prod-uct as shown in Scheme 4. The stereochemistry of this re-action was controlled by the C3-methyl group attached tothe projected g-lactone. Conformation 45b is energetical-ly favored, as the methyl group in 45a adds considerableallylic strain. In this intramolecular cyclization reaction,the piperidinyl-substituted diene system acts as dieno-phile. The authors suggest that the vinylcyclohexenyl

Scheme 2 Hart and Kozikowski’s synthesis of himbacine (2)

1. O3, MeOH2. NaHCO3, Me2S, MeOH

3. Ph3P=CHCO2Me

72%

OMeMeO

O

OMe

1. LDA, THF–HMPA, (S)-MeCH(OTHP)CHO

3. Amberlyst-154. Ph3P=CHCOSEt

44% (from 27)

O

O

COSEt

O

O

H

O SEt

H

H

75%

26 27 29

30

O

O

H

PhS

H

H31

H

1. Raney-Ni2. TsCl, py3. t-BuOK, DMSO, PhSH

72%O

OMe

H

PhO2S

H

H32

H

1. DIBAL-H2. BF3·OEt2, MeOH

3. MCPBA, NaHCO3

88%

O

OMe

HH

H

33

H

NH

HO O

NBoc

OTBS

NBoc

O

1. BH3⋅Me2S2. (Boc)2O, NaOH

3. TBSCl, imidazole

84%

1. s-BuLi, TMEDA; MeI2. TBAF

3. TPAP, NMO

57%

NBoc

1. n-BuLi; 362. 6% Na(Hg), MeOH, Na2HPO4

46%

O

O

HH

H

1 himbeline; R = H

H

NR

SiO2,Et2AlCl

34 35 36

1. Jones reagent2. TFA

87%

37% aq HCHO,MeCN, NaBH3CN; 70%2 himbacine; R = Me

OMeMeO

O

OMe28

HOOTHP

1. TsOH, MeOH2. MsCl, Et3N

2. I2, hν, CH2Cl2

99a

Scheme 3 Attempted Julia–Lythgoe olefination of aldehyde 37 andsulfone 38

O

O

H

O

H

H

37

H

O

OMe

HH

H

33

H

NBoc

NBoc

38

+

Julia–Lythgoeolefination

SO2Ph

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



moiety is more likely to adopt the required cisoid confor-mation.

Amine 39 was obtained by chiral resolution of commer-cially available 2-methylpiperidine with L-tartaric acid.Boc-protection of the nitrogen and installation of an alde-hyde functionality via treatment of the carbamate withsec-butyllithium and addition of dimethylformamide wasfollowed by Takai iodovinylation to afford vinyl iodide40.45 This material was used in a palladium-mediated cou-pling reaction with (S)-but-3-yn-2-ol (41) and the triplebond in alkyne 42 was reduced to the corresponding dou-ble bond via Lindlar hydrogenation. Esterification of alco-hol 43 with acid 44 (accessible in three steps fromcyclohexane carbaldehyde)46 cleanly gave key intermedi-ate 45 which was used in the intramolecular Diels–Alderreaction. As pointed out earlier, this intramolecular pro-cess only afforded the exo-adduct with a trans-fused lac-

tone ring attached to the decalin system. As reactionconditions only afforded partial isomerization to the de-sired cis-fused material, the cycloaddition product wastreated with 1,8-diazabicyclo[5.4.0]undec-7-ene, result-ing in complete epimerization at C9a. Reprotection of thepiperidinyl nitrogen was necessary as the high reactiontemperature required for the Diels–Alder reaction also re-sulted in thermally induced cleavage of the Boc function-ality. Stereoselective hydrogenation of the decalin doublebond in 46 from the less hindered face of the moleculewith Raney nickel and cleavage of the Boc group affordedhimbeline (1) which was converted into himbacine (2) fol-lowing the reductive alkylation protocol applied by Hartand Kozikowski.31

Chackalamannil thus presented a highly efficient route tohimbacine and the synthesis is the shortest and highest-yielding process published so far. Only 12 steps (11 fromBoc-protected methyl piperidine) were required to elabo-rate this piperidine alkaloid from inexpensive and readilyavailable starting materials, and the author reported a9.7% overall yield.

Terashima, 1999 (Himbacine)

In 1999, Terashima and co-workers14,33 reported a synthe-sis of himbacine (2) based on an intermolecular Diels–Alder reaction of tetrahydroisobenzofuran 51 and chiralbutenolide 58 as outlined in Scheme 5.

Following a procedure developed by Kanematsu,47 50 wasobtained starting from propargyl ether 47 by an intramo-lecular Diels–Alder reaction and base-catalyzed ringopening sequence of cycloadduct 49. The hydroxy func-tionality in 50 was converted into a leaving group andeliminated. Hydrogenation finally delivered the couplingpartner for the key cycloaddition reaction. Butenolide 58was accessible from tetrahydropyranyl-protected (S)-2-hydroxypropanal (56) by Still–Gennari olefination andacid-catalyzed cyclization in large quantities.48 Reactionof diene 51 and butenolide 58 in diethyl ether containinglithium perchlorate afforded the exo-cycloadduct 52 assole product. Hydrogenation of the double bond from thesterically less hindered face of the decalin derivative 52and base-mediated opening of the oxygen bridge gave alk-ene 53. Isomerization of the double bond and hydrogena-tion cleanly afforded alcohol 54.

Installation of the piperidinyl moiety was accomplishedvia the Julia–Lythgoe olefination protocol which had al-ready proven successful in the work of Hart andKozikowski. Therefore, a one-carbon elongation was re-quired, which, after protection of the lactone as methylacetal, was carried out by oxidation of the secondary alco-hol and subsequent Wittig olefination. Hydroboration ofthe double bond delivered primary alcohol 55 which wasconverted into sulfone 32. The synthesis was completed inanalogy to Hart and Kozikowski’s protocol with signifi-cantly higher reported yields.

Scheme 4 Chackalamannil’s synthesis of himbacine (2)

NH

NBocNBoc

I

OH

1. (Boc)2O, NaOH2. s-BuLi, TMEDA, Me2NCHO

3. CrCl2, CHI3

41%

OH

PdCl2(PhCN)2,piperidine, CuI

81%

NBoc

OH

NBoc

O

O

COOH

NBoc

O

O

NR

O

O

Lindlar, H2 95%

DEAC, DMAP, TEMPO (1%)

91%1. toluene, 186 °C, TEMPO (1%)2. DBU3. (Boc)2O, NaOH

60%

H H

H

H H

HH

1. Raney-Ni, H22. TFA

1 himbeline; R = HHCHO, MeCN, NaBH3CN; 78% 2 himbacine; R = Me

72%

46

39 40 42

45 43

44

41

45a 45b

9a

9a

3

3

O

O

R

O

O

RA1,3

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



The natural product was obtained in 18 linear steps with12.6% overall yield. As several steps of the endgame wereadopted from earlier studies, this synthetic effort consti-tutes a formal synthesis of the natural product. The ele-gance of the synthesis with a nice intermolecular Diels–Alder key step suffers from the tedious C1 elongation se-quence. Terashima also employed the reaction sequenceto prepare derivatives of the natural product for furtherstructure–activity relationship studies.

Chackalamannil, 2001 (Himandravine)

As himandravine (4) and himbeline (1) only differ in thestereochemistry at C13, synthetic routes to himbeline (1)can be modified in a way to grant access to himandravine(4) as well. In 2001, Chackalamannil34 reported the firsttotal synthesis of himandravine (4) based on results ob-tained in his earlier work on the epimeric alkaloid. Thepreparation of 4 was carried out in complete analogy tothe synthesis of himbacine (2) and the only difference wasthe epimerization of starting material 36 with triethyl-amine on silica gel to aldehyde 59 as outlined inScheme 6. The alkaloid was obtained after 12 syntheticoperations (11, if counting from Boc-protected piperidine36) in 17.0% yield overall.

Scheme 6 Epimerization of aldehyde 36 in Chackalamannil’s syn-thesis of himandravine (4)

DeClercq, 2002 (Himbacine)

A Diels–Alder strategy similar to that already employedby Hart and Kozikowski31 in their first total synthesis ofhimbacine was used by De Clercq35,49 who presented aformal synthesis of the Galbulimima alkaloid in 2002(Scheme 7).

Aldehyde 36 was converted into alkyne 60 with deproto-nated (trimethylsilyl)diazomethane in good yield and sub-jected to a Sonogashira coupling reaction with vinyliodide 66, easily obtained by syn-hydroalumination ofcommercially available octa-1,7-diyne (65). Stille cou-pling of iodide 61 with stannylated butenolide 68 deliv-ered the precursor for the key cyclization reaction. Thecycloaddition was accomplished in refluxing toluene aftera reaction time of three days, resulting in a complex mix-ture of isomers in 80% total yield. The double bond be-tween C9 and C9a in conjugation to the ester functionalitywas reduced with magnesium in methanol and intermedi-ate 64 was obtained in 31% yield over two steps.De Clercq decided to convert the alkyne into an alreadyknown intermediate. Thus, the lactone moiety was re-duced and converted into its methyl ketal before the triplebond was reduced under dissolving metal conditions af-fording alkene 33 in excellent yield. As this material wasdescribed in Hart and Kozikowski’s synthesis of himba-cine, the reduction concludes the formal synthesis of thealkaloid.

De Clercq thus presented an approach towards Galbulimi-ma alkaloids that is highly valuable for obtaining variousderivatives for SAR studies. All major isomers isolatedafter the Diels–Alder/reduction sequence were furtherconverted into himbacine analogues and used in biologi-

Scheme 5 Terashima’s synthesis of himbacine (2)

O OO O

HO

O

O O

C

t-BuOK, 83 °C

98%

1. MsCl, Et3N2. DBU, 80 °C

39%

3. Pd/C (10%), H2, 0 °C

O

O

O

H

H

O

O

H

O

O

H

H

O

OMe

H

H

47 48 49 50 51

O

55 54 53 52

OHH

H

OHH

H

OH

H

H

O

OMe

H

H32

SO2Ph

H

H

O

O

H

H

H

HNBoc

CHO

5 M LiClO4, Et2O, 58, r.t. 58%

1. Pd/C (10%), H22. LiN(TMS)2

NR O

OCOOMe

OTHPO

OTHP

88%

1. DBU, 100 °C2. PtO2, H2

80%

1. DIBAL-H2. BF3·OEt2, MeOH3. TPAP, NMO

4. Ph3PCH3I, NaN(TMS)25. BH3⋅THF; H2O2, NaOH

44%1. MsCl, Et3N2. PhSH, t-BuOK3. MCPBA

82%

1 himbeline; R = H 37% aq HCHO,MeCN, NaBH3CN; 91%2 himbacine; R = Me

1. n-BuLi, 362. 5% Na(Hg), MeOH, Na2HPO4

3. Jones reagent4. TFA

66%

56 57 58

36

Still–Gennariolefination H3O+

NBoc

CHO

silica gel, Et3N

83% NBoc

CHO

himandravine (4)cf. Scheme 4

36 59

13 13

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



cal investigations. However, the low selectivity of theDiels–Alder reaction is a drawback and obstacle for an ef-ficient preparation of himbeline (1) and himbacine (2).

Sherburn, 2003 (Himbacine)

In 2003, Sherburn36 published a formal synthesis of him-bacine (2) with an intramolecular Diels–Alder reactionand a radical cyclization as key steps. The alkene moietyformed in the Diels–Alder reaction is utilized as reactivehandle for the radical process. The synthesis is outlined inSchemes 8 and 9.

Stille coupling of vinylstannane 69 and dibromide 70, de-rived from (S)-lactic acid in three steps,50 afforded diene71. Protection of the free hydroxy functionality and cleav-age of the silyl ether paved the way for the preparation ofDiels–Alder precursor 72 which was obtained after ester-ification of the primary alcohol with acryloyl chloride.The thermal cycloaddition reaction was carried out in re-fluxing chlorobenzene and delivered a mixture of cis- andtrans-fused bicyclic reaction products; this mixture wasconverted into the desired cis-fused adduct 73 in quantita-tive yield by exposure to 1,8-diazabicyclo[5.4.0]undec-7-ene. The piperidinyl side chain was attached by way of aStille coupling between stannane 78 and the vinyl bro-mide moiety in 73 and the corresponding E-alkene 74 wasobtained in high yield.

The following steps were devoted to the installation of theselenoate ester required for the radical cyclization. Thiswas achieved by cleavage of the methoxymethyl ether fol-lowed by oxidation of the primary alcohol under Swernand Pinnick51 conditions and exposure of the carboxylicacid to diphenyl diselenide and tributyl phosphine. For the

radical cyclization reaction, different reaction conditionsand mediators were employed. The desired cyclizationproduct 76 could be isolated without evidence of the C4epimer, thus indicating the high degree of stereocontrol inthis reaction. However, the main product in this key stepwas identified to be a mixture of E- and Z-isomers of ke-tone 77. As the maximum yield of the desired product didnot exceed 13%, the approach was abandoned.

As delocalized propargylic radicals react almost exclu-sively at the propargylic site, the modified approach askedfor the preparation of alkyne 80 which was achieved bycoupling of stannane 79 with bromide 73 as outlined inScheme 9. In analogy to the procedure described above,the methoxymethyl group in 80 was cleaved and the hy-droxy functionality was oxidized to the correspondingcarboxylic acid by applying Swern and subsequentPinnick51 oxidation protocols. Esterification with diphe-nyl diselenide and tributyl phosphine again proceededsmoothly and selenoate 81 was isolated in 51% overallyield from methoxymethyl ether 80. All cyclization ef-forts resulted in approximately 1:1 mixtures of alkyne 83and allene 82; the best results were obtained when the re-action was carried out in refluxing benzene with a totalyield of 90% and 43% yield of the desired alkyne 83.

Epimerization of C4a by treatment of ketone 83 with 1,8-diazabicyclo[5.4.0]undec-7-ene was followed by Luchereduction. The Barton–McCombie deoxygenation proto-col required installation of a thionocarbonate. Unexpect-edly, all attempts to introduce the thionocarbonate failedand when more forcing conditions were applied, the cor-responding secondary chloride was obtained. As pointedout by Sherburn, this transformation is unprecedentedwith thionocarbonates but has been observed with xan-

Scheme 7 De Clerq’s formal synthesis of himbacine (2)

NBoc

CHO

NBoc

NBoc

O

O

NBoc

O

O

36

4

I

I

I

NBoc NBoc

O

O

LDA, TMSCHN2

75%

DIBAL-H, 50 °C;I2, –50 °C

86%

OH

CO2Et

O

Bu3SnO

n-Bu3SnH, Pd(PPh3)4

80%

65 66

67 68

60

61 62 63

64

66, Pd(PPh3)4,piperidine 68, CuCl, DMF

toluene, 110 °C,3 days

total yield: 80%(mixture of isomers)

63% 80%

Mg, MeOH

31%(from 62)

H

H

H

H

NBoc

O

OMe33

H

H

H

H

1. DIBAL-H2. BF3·OEt2, MeOH3. Li, liquid NH3, t-BuOH

92%

Hart, Kozikowski,see Scheme 2

himbacine(2)

mixture ofdiastereomers

9

9a

9

9a

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



thates.52 As dechlorination proceeds under same reactionconditions as removal of the xanthate or thionocarbonate,the formation of a chloride might have easily been over-looked in previous deoxygenation protocols.

The formal synthesis of himbacine (2) was concluded bysuch a radical dechlorination, resulting in the formation ofalkyne 64 which had been reported as intermediate inDe Clercq’s preparation of the natural product.

Sherburn’s synthesis clearly suffered from unexpectedproblems such as migration of the double bond in the firstapproach and formation of the allene in the second ap-proach which significantly reduces the overall efficiencyof the published formal synthesis. The utilization of bothbromines in the starting material for coupling reactions,and the use of the double bond generated in the Diels–Alder reaction for the radical cyclization are worth men-tioning. However, the installation of the selenoate ester re-quires several synthetic operations and preparation of theunfunctionalized six-membered ring is tedious. Counting,

26 operations are necessary to prepare the natural productwith a yield of 2.6% overall.

Baldwin, 2005 (Himbacine, Himbeline, Himandravine)

The latest synthetic contribution to class I Galbulimimaalkaloids was published by Baldwin in 2005.37 The ap-proach mimics the assumed biosynthetic pathway and fea-tures an intramolecular Diels–Alder reaction similar to thestrategies presented by Hart and Kozikowski and byDe Clercq accompanied by simultaneous formation of thepiperidine ring.

Starting from cycloheptene (26), Baldwin repeated thefirst six steps of Hart and Kozikowski’s route to aldehyde84. Conversion into a,b-unsaturated aldehyde 86 was ef-fected by reaction with aldimine 85, as traditional Wittigolefination protocols unexpectedly delivered the desiredproduct only in low yield. Chiral phosphonate 92 was de-rived from Boc-protected methyl piperidine 90 throughruthenium tetraoxide oxidation and reaction with lithiateddimethyl methylphosphonate. The key cyclization precur-sor 87 was accessed via Horner–Emmons reaction ofaldehyde 86 and phosphonate 92. Exposure to trifluoro-

Scheme 8 Sherburn’s first approach towards himbacine (2)

OTBSBr

Br

+SnBu3

OTBS

BrOHPd2(dba)3, AsPh3

72%

O

BrMOMO

O

1. MOMCl, DIPEA2. TBAF3. acryloyl cloride, Et3N

69 70 71

82%

O

BrMOMO

OH H

H

NBoc

SnBu3

1. PhCl, reflux BHT (0.05 equiv)2. DBU

75%

O

MOMO

OH H

H

NBoc

78, Pd(PPh3)4,CuCl, LiCl, DMSO

73 72

78

74

himbacine (2)

86%

O

OH H

H

NBoc

75

PhSe O

O

OH H

H

NBoc

76

O

OH H

H

NBoc

77

+

O OHH

1. PPTS, t-BuOH, reflux2. (Boc)2O, CH2Cl23. (COCl)2, DMSO; Et3N

4. Pinnick ox.5. Ph2Se2, n-Bu3P, DMF

OH

Ph3SnH, Et3B,air, PhH, reflux

42%65% total yield77/76 = 4:1

44

Scheme 9 Sherburn’s formal synthesis of himbacine (2)

NBoc

79SnBu3

NBoc

O

MOMO

OH H

H

80

NBoc

O

OH H

H

81

O

O

OH H

H

82 47%

O

+

NBoc

O

OH H

H

83 43%

O C

HNBoc

H

NBoc

O

OH H

H

64

HH

H

Pd(PPh3)4, CuCl, LiCl, DMSO, 73

1. PPTS, t-BuOH,2. (Boc)2O3. Swern ox.4. Pinnick ox.5. Ph2Se2, n-Bu3P

Bu3SnH, AIBN

PhSe

85%

51%

total yield: 90%

1. DBU2. NaBH4, CeCl3⋅7H2O3. C6F5OCSCl, DMAP4. Bu3SnH, AIBN

De Clercq,see Scheme 7

himbacine (2)

65%

4a

4a

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



acetic acid in dichloromethane, followed by cyanoboro-hydride reduction, gave the desired tetracycle 88 as aninseparable mixture of C13 epimers. Reprotection of thenitrogen and hydrogenation of the trisubstituted doublebond afforded lactones 33 and 89 which were separatedby column chromatography. The two diastereomers 33and 89 were then elaborated into himbacine (2; from lac-tone 33) and himandravine (4; from lactone 89) as out-lined in Scheme 10.

In his effort to improve upon Hart and Kozikowski’s syn-thesis of himbacine, Baldwin presented a highly divergentapproach towards all class I Galbulimima alkaloids. Thekey step not only establishes the decalin system but alsodelivers the piperidine unit of the natural products. Giventhat the key Diels–Alder cycloaddition delivers a mixtureof epimers, the strategy does not allow the preparation ofa single alkaloid in high yield.

2.2 Class II Alkaloids

Only one synthesis of a class II Galbulimima alkaloid hasbeen presented to date. The preparation of himandrine,published by Movassaghi in 2009, is described below.

Movassaghi, 2009 (Himandrine)

In 2006, Movassaghi reported the synthesis of G.B. 13which is outlined in the discussion of Galbulimima classIII alkaloids (section 2.3). The preparation of himandrine(6) can be regarded as a follow-up project to build astructure of higher complexity and is closely related toMovassaghi’s earlier work. As similar strategies havebeen applied, a direct comparison of these two synthesesmight be of interest.

The synthesis of the natural product follows the assumedbiosynthetic pathway as outlined by Movassaghi.27 Ac-cordingly, the N–C9 bond is formed at a late stage of thesynthesis by spirocyclization of pentacycle 93 that isobtained by oxidation of a,b-unsaturated ester 94(Scheme 11). The heterocyclic ring is introduced througha formal [3+3]-cycloaddition reaction of tricyclic ketone95 and alaninol-derived piperidine derivative 96. The keysteps in the preparation of tricyclic ketone 95 are an aldolcondensation and a Diels–Alder reaction.

The starting material for this synthesis was obtained viaproline-catalyzed a-oxidation of hept-6-enal followingthe procedure developed by MacMillan.53 Monosilylationof diol 97, methylation of the secondary hydroxy groupwith Meerwein salt and subsequent desilylation delivered

Scheme 10 Baldwin’s synthesis of himbacine (2) and himandravine (4)

6 steps,Hart, Kozikowski,

see Scheme 2

49%O

O

O

O

O

O

O

O

ONHBoc

O

O

NH

O

O

NBoc

O

O

NBoc

+

60%

(TMS)2CHCH=Nt-Bu (85),ZnBr2; ZnCl2, H2O

NBoc NBoc

O

NHBoc

(MeO)2P

O

92, DIPEA,LiCl, MeCN

TFA; NaBH3CN

50%

H

H H

H

1. (Boc)2O, Et3N2. H2, PtO2

22% from 87

26 84 86 87

888933

90 91

92

H

H

H

HH

HH

H

O

O

NH

H

H

H

H

O

O

N

himbacine (2)

H

HH

H

TFA80%1. TFA2. HCHO, NaBH3CN60%

himandravine (4)

RuO4, NaIO4

39%

BuLi(MeO)2P(O)Me

54%

1313 13

O

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



alcohol 98 in excellent overall yield (Scheme 12). The hy-droxy moiety was oxidized and converted into the corre-sponding dibromide by means of tetrabromomethane andtriphenylphosphine. Suzuki coupling of this material withboronic acid 105 delivered bromide 99. Next, the 2-azeti-dinone moiety was introduced; this served as maskingfunctionality for the C16-carbonyl group and simulta-neously forced the diene system into the s-cis conforma-tion required for the Diels–Alder cycloaddition. Beforethe key step could be carried out, the silyl ether was con-verted into the corresponding silyl enol ether 100 bycleavage of the tert-butyldimethylsilyl group, oxidation ofthe alcohol and treatment of the corresponding ketonewith tert-butyldimethylsilyl triflate in the presence of tri-ethylamine. Terminal alkene 100 was subjected to a ruthe-nium-catalyzed cross-metathesis reaction with acrolein,and the resulting Diels–Alder precursor 101 was heated to95 °C in acetonitrile containing butylated hydroxytoluene(BHT) and N,N-diethylaniline to afford trans-decalin 102in 75% yield via an endo transition state. Tricyclic lactone95 was obtained by titanium tetrachloride promotedMukaiyama aldol reaction and subsequent exposure toMartin sulfurane.54

The nitrogen heterocycle was introduced via a formal[3+3]-cycloaddition reaction30 as shown in Scheme 13.Iminium hydrochloride 96, readily available from L-alan-inol,27 was lithiated and subjected to a copper-promoted1,4-addition reaction with pentenone 95. Iminoketone in-termediate 106 tautomerized to enamine 107 which react-ed in a subsequent nucleophilic 1,2-addition to yieldpentacycle 108. The reduction of the imino functionalityproceeded under complete stereocontrol and the resultingamine was protected as carbamate. Hydrolysis of the N-vinyl lactam released the masked ketone (109) and pavedthe way for the introduction of the methoxycarbonyl func-tionality present in the natural product. After a variety ofconditions had been probed, Vilsmeier formylation was

found to be the method of choice and vinyl ether 111 wasobtained by intramolecular trapping with the hydroxyfunctionality at C20. The close vicinity of the C20 hy-droxy functionality and C18 becomes obvious when thestructure is redrawn as shown for Vilsmeier adduct 111.

Oxidation of vinyl ether 111 with 2,3-dichloro-5,6-dicy-ano-1,4-benzoquinone afforded the a,b-unsaturated alde-hyde which was further oxidized to the carboxylic acid.The methyl ester was formed by exposure to diazo-methane and the carbamate was cleaved with iodotrimeth-ylsilane.55 As this protocol also converted the hydroxyfunctionality at C20 into the trimethylsilyl ether, an addi-tional deprotection step with triethylamine trihydrogenfluoride became necessary to reveal amino keto ester 94which served as key intermediate for the completion ofthe synthesis of the natural product.

The key spirocyclization reaction was accomplished bytreatment of amino keto ester 94 with N-chlorosuccinim-ide in acetonitrile. Presumably, this highly efficient reac-tion proceeded via chlorination of dienol 112 at C17 andsubsequent intramolecular attack of the nitrogen at C9with concomitant loss of the chlorine atom. An interesting

Scheme 11 Retrosynthetic analysis of Movassaghi’s synthesis ofhimandrine (6)

OBz OMe

N

MeO2C

HO

H

HOMe

NH

MeO2C

HO

H

H

N OMe

OH

HO OMe

NH

MeO2C

HO

H

H

himandrine (6) 93

94

95

N–C9 bond formation

OX

spiro-cyclization

oxidation

formal [3+3]cycloaddition

carboxylation

N⋅HCl+

L-alaninolderived

aldol

Diels–Alder

96

O

Scheme 12 Movassaghi’s synthesis of himandrine (6) – part 1

OH OMe OMeBr

OTBS

1. TBSCl, imidazole2. proton sponge, Me3O·BF4

86%

3. HCl, MeOH

1. DMSO, DIPEA, SO3·py2. CBr4, Ph3P

OH OH

97 98 99

OMeN

OTBS

100

3. 105, Tl2CO3 Pd(PPh3)4

63%

O

1. 2-azetidinone, CuI, K2CO3, (MeNHCH2)22. TBAF3. DMSO, DIPEA, SO3·py4. TBSOTf, Et3N

56%

OMeN

OTBS

101

O

O

acrolein, Grubbs–Hoveyda

OMeNO

TBSO

O

H

H

85%

OMeNO

H

HO

BHT, N,N-diethyl-aniline, 95 °C

75%

1. TiCl4, –78 °C2. Martin sulfurane

57%

102 95

14 14

20

20

OTBS OTBS

B

O

O

OTBS

B

OH

OH

NaIO4,NH4OAc

acetone, H2O

95%103 104 105

pinacol-borane

CH2Cl268%

16

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



mechanistic insight is shown in Scheme 14. N-Chloropentacycle 115 was formed in small amounts during thecourse of the cyclization reaction but disappeared after thereaction neared completion. Dissolution of the isolated N-chloro compound in acetonitrile did not result in the for-mation of spirocycle 114 even after prolonged reactiontime. However, when small amounts of amino keto ester94 were added, 114 was obtained. These results suggestthat the nitrogen of pentacycle 115 is not basic enough tofacilitate deprotonation at C9 and formation of dienol 112and a stronger base such as 94 is required for the reactionto proceed via a nitrogen-to-chlorine halogen transfer. Re-duction of the ketone and benzoylation of the hydroxyfunctionality concluded the synthesis of the natural prod-uct.56

Movassaghi thus reported a beautiful synthesis of thishighly complex Galbulimima alkaloid. He succeeded inmodifying and advancing an already existing protocol(preparation of G.B. 13 as discussed below) and was ableto demonstrate that the proposed biosynthetic pathway toGalbulimima alkaloids is plausible. The formal [3+3] cy-cloaddition and the spirocyclization are clearly highlightsof this synthesis.

2.3 Class III Alkaloids

In 2003, Mander38 reported the first total synthesis of aclass III Galbulimima alkaloid with the preparation ofG.B. 13. In subsequent years, Movassaghi,27 Chackala-mannil,39 Evans,40 and Sarpong41 contributed to this fieldwith syntheses of G.B. 13 (22) and himgaline (21).

Diels–Alder reactions are key steps in the construction ofthe decalin unit in all syntheses of class III Galbulimimaalkaloids. However, strategies for the elaboration of theremaining framework vary significantly. The following

Scheme 13 Movassaghi’s synthesis of himandrine (6) – part 2

N·HCl +

OMeNO

H

HO

95

N OMe

NH

H

106

O

O

N OMe

NHH

H

O

O

N OMe

NH

H

O

HO

n-BuLi;CuBr·SMe2

96 107 108

O OMe

NRH

H

HO 109 R = Cbz

1. NaBH42. ClCO2Bn, DIPEA, K2CO33. TsOH

O OMe

NRH

H

111 R = Cbz

POCl3, DMF

71%

41%from 95

O

O OMe

NHH

H

HO 94

1. DDQ, SiO2, MeCN, H2O2. Pinnick ox.3. CH2N24. TMSI, 2,6-dibutyl-4-methylpyridine5. Et3N·(HF)3

MeO2C

OH OMe

NHH

H

HO 112

MeO2C

O OMe

NH

H

HO 114

MeO2C

NCS,MeCN

89%

1. NaBH4, EtOH2. BzCl, py, 7 d

78%

36%

NCbz

HO

O OMeH

H

NCbz

O OMeH

H

O

=

110111

O OMe

NHH

H

HO 113

MeO2CCl

9

17

9

17

9 himandrine(6)

NMeMe

20

2020

20

18

17

18

Scheme 14 Mechanistic considerations on the key spirocyclization

94

112

113 114

O OMe

NClH

H

HO 115

MeO2C

NCS

9

94

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



section summarizes and discusses synthetic efforts to-wards G.B. 13 and himgaline.

Mander, 2003 [(±)-G.B. 13]

The strategy of the first racemic synthesis of G.B. 13 (22),published by Mander in 2003,38 is shown in Scheme 15.The approach is based on the utilization of tricyclic alkene118 as dienophile in a Diels–Alder reaction with tert-bu-tyldimethylsilyl-protected enol ether 117. The aromaticportion of cycloadduct 116 is subsequently converted intothe nitrogen heterocycle of the natural product. Key inter-mediate 118 was prepared from acid 119 which can be ac-cessed via reduction of dimethoxy benzoic acid (120) withlithium in liquid ammonia and subsequent alkylation ofthe intermediary dianion.

Scheme 15 Retrosynthetic analysis of Mander’s synthesis of (±)-G.B. 13 (22)

Dimethoxy benzoic acid 120 was subjected to lithium inliquid ammonia and subsequently alkylated with m-meth-oxybenzyl bromide to afford bis-enol ether 11957 whichwas treated with sulfuric acid in acetone to deliver tricycle121 in good yield (Scheme 16). Decarboxylation and pro-tection of the tertiary hydroxy functionality as its meth-oxymethyl ether was followed by diazotation employinga protocol developed by Regitz58 to furnish diazo ketone122. Photolysis in the presence of hexamethyldisilazaneresulted in Wolff ring contraction and delivered amide123 which was dehydrated to the corresponding nitrile byreaction with trichloroacetyl chloride. The alkene moietywas established by deprotonation of the nitrile and trap-ping of the anion with diphenyl diselenide and a subse-quent oxidation and elimination sequence. Starting from

bis-alkene 119, key intermediate 118 was prepared ineight steps and 34% yield overall.

Lewis acid catalyzed Diels–Alder cycloaddition of alkene118 with silyl enol ether 117 delivered the desired endo-adduct 125 in excellent 87% yield as outlined inScheme 17. After hydrolysis of the enol ether, the ketonewas reduced and the corresponding hydroxy functionalitywas protected (126). Next, the aromatic portion had to beconverted into the piperidine ring. Treatment of nitrile126 with lithium in ammonia resulted in reductive cleav-age of the cyano functionality59 and subsequent additionof ethanol to the reaction mixture delivered the Birch re-duction product. Hydrolysis of the methyl enol ether witha catalytic amount of hydrochloric acid in methanol gavethe corresponding enone 127 which was epoxidized to de-liver the precursor required for the Eschenmoser fragmen-tation. As all attempts to directly convert the enone intoepoxide 128 failed, the ketone was reduced to the allylicalcohol and reoxidized after successful epoxidation withm-chloroperoxybenzoic acid.

When 128 was subjected to traditional Eschenmoserconditions, alkyne 129 was obtained in only low yields.Better results were obtained when p-nitrobenzenesulfo-nylhydrazide was used.

Reductive amination of ketone 129 was unsuccessful.However, treatment of this alkynyl ketone with an excessof hydroxylamine afforded bis-oxime 130. Such hydro-aminations of nonactivated alkynes has long depended onthe application of transition-metal catalysis or the use ofstrong acids. In 1989, Pradhan60,61 reported the formationof a piperidine ring from alkynyl ketones by reaction withhydroxylamine, and most recently, Mayrargue62 andBeauchemin63,64 contributed to this highly interestingfield with the development of milder reaction conditionsand the presentation of possible mechanisms. Reductivecyclization of bis-oxime 130 was achieved with sodiumborohydride and zirconium tetrachloride and the hydrox-ylamine was reduced by means of zinc and acetic acid be-fore the nitrogen was protected as the trifluoroacetamide

NHH

H

G.B. 13 (22)

O

HO

OH

H=

OTBS

CO2H

MeO

+

117

118116

119

Diels–Alder

Birch reduction

Eschenmoserfragmentation

reductive cyclization

OMe

MeO

HO

NH

H

H

OH

HMOMO

CN

MeO

MOMO CN

MeO

Birchalkylation

HO2C

MeO OMe

120

Scheme 16 Mander’s synthesis of (±)-G.B. 13 (22) – part 1

MeO

OHO

CO2H

MeO

OMOMO

N2

119

H2SO4, acetone

80%

1. AcOH, H2O2. MOMCl, DMAP, DIPEA

3. NaH, HCOOEt4. 124, Et3N, MeCN

SO2N3O2N

121 122

123118

124

hν, (Me3Si)2NH,0 °C; H3O+

1. Cl3CCOCl, Et3N2. KDA, Ph2Se2; H2O2, THF

59%from 121

73%

NH2

OMOMO

MeO

CNMOMO

MeO

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



(131). Cleavage of the methoxymethyl ethers was fol-lowed by oxidation of the secondary alcohol and reprotec-tion of the tertiary hydroxy functionality. Enolization ofthe ketone and dehydrosilylation of the intermediary silylenol ether afforded a,b-unsaturated ketone 132. The syn-thesis was completed by cleavage of the oxygen and nitro-gen protecting groups to furnish G.B. 13.

Mander thus presented an interesting approach toward thenatural product and demonstrated the utility of benzenoidsynthons. Problems with the epoxidation of enone 127 aswell as the methoxymethyl deprotection/reprotection se-quence during the endgame prolonged the synthesis ofG.B. 13. Overall, the alkaloid was prepared in an elegantand efficient manner.

Movassaghi, 2006 (G.B. 13)

The preparation of G.B. 13 and the assignment of theabsolute stereochemistry of the natural product wasMovassaghi’s first contribution27 to the interesting field ofGalbulimima alkaloids. During the course of this synthet-ic effort, Movassaghi developed strategies that also result-ed in the successful preparation of the class II alkaloidhimandrine,29 published in 2009 and discussed above.

The retrosynthetic analysis is outlined in Scheme 18. Dis-connection between C5 and C20 simplifies the alkaloid to

imine 133 which is accessed by conjugate addition of sub-strate 134 to build the C ring of the natural product. Thepiperidine ring can be traced back to L-alaninol-derivedimine 96 which is linked to aldehyde 135 via a condensa-tion reaction. The key step in the formation of the decalinsystem is a Diels–Alder cycloaddition. The decalin unitwas prepared as a racemic mixture; however, as piperi-dine 96 was synthesized in enantiopure form, the absoluteconfiguration of the natural product could be defined.

As outlined in Scheme 19, dibromide 136 was coupledwith boronic acid 105 in a Suzuki coupling before the ox-azolidin-2-one moiety was attached to deliver triene 137in excellent yield. This oxazolidin-2-one moiety served asthe masking group for a ketone which was installed laterin the sequence but also, as noted in Movassaghi’s com-munication describing the synthesis of himandrine,29

forces the diene into the s-cis-configuration required forthe key Diels–Alder reaction to establish the decalin core.

The silyl ether in 137 was cleaved, and the hydroxy func-tionality was oxidized and converted into the correspond-ing silyl enol ether before the tetraene was subjected tocross-metathesis with acrolein to deliver Diels–Alder pre-cursor 138. The key cyclization step proceeded smoothlyand trans-decalin 135 was obtained in good yield with anendo/exo ratio greater than 20:1. This reaction sequence

Scheme 17 Mander’s synthesis of (±)-G.B. 13 (22) – part 2

NRH

HMOMO

MOMO

Yb(thd)3

1. TBAF2. LAH3. MOMCl, DMAP, DIPEA

1. LAH2. MCPBA3. DMP, NaHCO3

OTBS

+

p-NO2C6H4SO2NHNH2,py, EtOH, THF

=NR

H

HO

MOMO

H2NOH·HCl,py, 100 °C

1. ZrCl4, NaBH42. Zn, AcOH3. TFAA, Et3N

117

NHH

HO

HO

1. HCl (dil.)2. DMP3. MOMCl, DMAP, DIPEA

1. K2CO3, H2O, 60 °C2. HCl (dil.)

4. LDA, TMSCl5. Pd(OAc)2, DMSO

131 132 G.B. 13 (22)

1. Li, NH3; EtOH2. HCl, MeOH

87% 67% 55%

77%

76%

32% (from 129)

47%

33%

R = COCF3 R = COCF3

118

CNMOMO

MeO

125

MOMO

MeO

OTBS

CN

H H

127

MOMO

O

MOMO

H

H H

H

126

MOMO

MeO

MOMO

CN

H H

H

H

128

MOMO

O

MOMO

H

H H

H

H

O129

MOMO

MOMO

H

H H

H

H

O130

MOMO

MOMO

H

H H

H

H

(HO)N(HO)N

131N

MOMO

MOMO

H

H H

H

H

COCF3

H

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



was successfully applied to a more complex substrate inthe preparation of himandrine (Scheme 12).

Enantiopure iminium hydrochloride 96 was deprotonatedand added to racemic aldehyde 135. Subsequent dehydra-tion using Martin sulfurane afforded E-alkene 134 as amixture of diastereomers (for ease of reading, the undes-ired isomers are not shown in Scheme 19; the diastereo-mers were separated after formation of pentacycle 142).Vinyl bromide 139 was obtained after exposure of silylenol ether 134 to N-bromosuccinimide. Next, the C ringwas formed. Heating of crude bromide 139 with tributyl-tin hydride and 2,2¢-azobis(isobutyronitrile) generated aradical at C21 which reacted with C7 in a 5-exo-trig cy-clization reaction to tetracycle 140. Silyl enol ether cleav-age also triggered a clean cyclization of the enamine ontothe newly formed carbonyl functionality. Reduction of theimine intermediate finally produced pentacycle 142. Atthis point, the separation of the diastereomers generatedfour steps earlier was accomplished by flash column chro-matography. Subsequently, the enone had to be formed;therefore, the nitrogen was protected as a carbamate andthe compound was treated with excess p-toluenesulfonicacid and 2-iodoxybenzoic acid (IBX).65 The synthesis wascompleted by N-deprotection with iodotrimethylsilane55

followed by acidic workup.

In summary, Movassaghi presented a highly efficient ap-proach towards G.B. 13 (22). As already mentioned, strat-

Scheme 18 Retrosynthetic analysis of Movassaghi’s synthesis ofG.B. 13 (22)

O

NH

HO

H

H

NH

H

H

H

O

H

H

G.B. 13 (22) 133

134135

N

N.HCl+

L-alaninolderived

Diels–Alder

96

iminereduction

carbonyladdition

D-ringsynthesis

N

O

O

CHO

TBSO

TBSO

N

condensation

N

O

O

O

O conjugate addition

C

Scheme 19 Movassaghi’s synthesis of G.B. 13 (22)

Br

Br N

O

O

TBSO

1. Pd(PPh3)4, TlCO3, 1052. CuI, K2CO3, oxazolidin-2-one, (MeNHCH2)2

1. TBAF, THF2. MnO2, CH2Cl23. TBSOTf, Et3N

4. Grubbs–Hoveyda, acrolein

N

O

O

TBSO

N

O

TBSO

O

CHO

toluene,90 °C

H

H

NH

HO

H

H

140 134

N

TBSO

N

N

OO

O

TBSO

1. 96, n-BuLi2. Martin sulfurane

NBS,NaHCO3

NHH

H

141

N

O

O

O

Et3N·(HF)3then NaBH4

NHH

H

142

N

O

O1. ClCO2Bn, Na2CO32. IBX, TsOH3. TMSI; HCl; NaOH

71%

136 137 138 13569%

82%

69%

Bu3SnH,AIBN

55%(2 steps)

total yield: 70%desired isomer: 36%

46%

HO

O

O

H

H

139

TBSO

N

N

O

Br

NHH

HO

HO G.B. 13 (22)

HCl⋅N

96

OTBS

B(OH)2

105

2021

2021

C77

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



egies developed during this synthetic effort weresuccessfully incorporated in the preparation of himan-drine three years later.

Chackalamannil, 2006 (Himgaline)

The first synthesis of himgaline (21) was reported byChackalamannil in 2006.39,66 The alkaloid was obtainedfrom G.B. 13 via an aza-Michael reaction, thus providingproof for Movassaghi’s assumption and observation of theease of interconversion between himgaline and G.B. 13.

Scheme 20 outlines the retrosynthetic analysis. G.B. 13(22) can be obtained from 143 through acid-catalyzed N-deprotection, lactone opening and decarboxylation. Sim-plification of the piperidine ring and removal of the oxida-tion around C17 leads back to 144. Precursor 145 isaccessible from 146, which Chackalamannil had alreadyprepared in an earlier synthetic effort towards Galbulimi-ma class I alkaloids.16

Scheme 20 Retrosynthetic analysis of Chackalamannil’s synthesisof himgaline (21)

As shown in Scheme 21, tetrahydropyranyl-protected (R)-but-3-yn-2-ol (147) was converted into the pentynoic acidbenzyl ester by deprotonation and reaction with benzylchloroformate. Hydrolysis of the tetrahydropyran and es-terification with dienoic acid 44 afforded 148. After hy-drogenation using Lindlar’s catalyst, a thermally inducedDiels–Alder cycloaddition followed by epimerization tothe cis-lactone by treatment with 1,8-diazabicyc-lo[5.4.0]undec-7-ene gave decalin 146. Hydrogenation ofthe double bond resulted in simultaneous cleavage of the

benzyl ester. The acid moiety was converted into the cor-responding acid chloride and this functionality was re-duced to the aldehyde by reaction with tributyltin hydridein the presence of palladium.67 The terminal double bondwas introduced via Wittig chemistry, delivering alkene149.

Next, the C ring of the natural product was prepared. Thelactone was reduced and the primary alcohol was selec-tively protected as a silyl ether before the methyl ketonewas formed by oxidation of the secondary alcohol. Ozo-nolysis of the terminal double bond permitted chain exten-sion to the a,b-unsaturated benzyl ester. a-Bromination ofthe silyl enol ether at C21 with N-bromosuccinimide ledto 151, the precursor for the radical ring-closure reaction.

The radical-promoted formation of the C ring proceededwith high diastereoselectivity. Hydrogenation deliveredacid 152 which was treated with N,N¢-dicyclohexylcarbo-diimide and Meldrum’s acid (155) and esterified to yield145. Cleavage of the silyl ether under acidic conditionswas followed by addition of zinc triflate to effect the clo-sure of the D ring.

Diketone 153 was obtained after conjugate addition of b-keto ester 144 to methyl vinyl ketone followed by deben-zylation and decarboxylation. Construction of the piperi-dine ring started with selective reductive amination of 153with (R)-a-methylbenzylamine and sodium cyanoborohy-dride. Subsequent N-debenzylation by hydrogen transferand a second cyanoborohydride reduction completed thesequence. Protection of the free amine as a trifluoroacet-amide delivered hexacycle 154.

Scheme 22 shows the continuation of the synthesis. Thetetrahydrofuran ring in 154 was oxidized to the corre-sponding lactone by means of ruthenium tetroxide68 andthe C17–C19 double bond was introduced via formationof the thioether at C17 (a to the ester), and a subsequentoxidation and elimination sequence. With the double bondin place, the stage was set for the introduction of the ke-tone moiety at C16. Allylic bromination was followed bydisplacement of the bromine by treatment with silver tri-fluoroacetate. The intermediary trifluoroacetate was hy-drolyzed and the resulting secondary alcohol wasoxidized with Dess–Martin periodinane to deliver ketone143 in good overall yield.

When subjected to 6 M hydrochloric acid in a microwavereactor at 100 °C, ketone 143 afforded G.B. 13 (22) in ex-cellent 80% yield. The reaction sequence included cleav-age of the trifluoroacetamide, aza-Michael addition ontothe conjugated double bond, thermally induced decarbox-ylation and retro-Michael reaction.

Treatment of G.B. 13 (22) with scandium triflate in chlo-roform and subsequent reduction with sodium borohy-dride yielded 16-epi-himgaline. The problem was solvedby the use of an internally coordinated reducing agent:with sodium (triacetoxy)borohydride, himgaline (21) wasobtained in 60% yield.

NH

HOH

HO

aza-Michael

himgaline(21)

G.B. 13(22)

144 143

reductive amination

radical ringclosure

145

H

H

NCOCF3

OH

H

O

O

NH

OH

H

HO

BnO2C

H

H

O

O

Diels–Alder

146

CO2Bn

O

O H

H H

TIPSO

O

O COOBn

H

H

17

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



Chackalamannil thereby successfully employed Diels–Alder adduct 146, used in previous efforts towards deriv-atives of himbacine (2), as the starting point for the stereo-selective preparation of G.B. 13 (22) and himgaline (21).The microwave-promoted reaction sequence leading toG.B. 13 clearly represents the highlight of this work. Adrawback, however, is the highly linear synthetic strategyand the oxidation and reduction sequences, especiallywith regard to the lactone moiety which is eventually re-

moved to afford the natural product. The first preparationof himgaline (21) from G.B. 13 (22) proves observationsby Mander describing the formation of oxohimgalinefrom G.B. 13 under acidic conditions.26

Evans, 2007 (ent-G.B. 13, ent-Himgaline)

With an auxiliary-controlled, enantioselective Diels–Alder reaction as one of the key steps, Evans reported the

Scheme 21 Chackalamannil’s synthesis of himgaline (21) – part 1

OTHP

O

HO

1. n-BuLi, BnOCOCl2. Dowex, MeOH

3. 4-pyrrolidinylpyridine, DCC, 44

O

CO2Bn

O

44

O

CO2Bn

O

H

H

H

1. H2, Lindlar2. o-xylene, 185 °C3. DBU

O

O

H

H

H

1. H2, PtO22. SOCl2 3. Bu3SnH, (Ph3P)4Pd

H

4. MePPh3Br, PhLi

H

H

H

H

O

1. LAH2. TIPSOTf, Et3N3. DMP

H

H

H

H

COOBn

O

Br

1. O3, Zn/AcOH cat. AgNO32. BnOCOCH2P(O)(OEt)2 NaHMDS

3. TMSOTf, Et3N4. NBS, THFH

H H

O

O

1. Bu3SnH, AIBN, PhH, reflux

O O

O O155

1. DCC, 155 DMAP

N

O

H

HH

H

154

O

H

HBnO2C

O

1. HCl (aq)2. Zn(OTf)2, CHCl3, Δ

O

OBn

O

H

H

O

1. MVK, NaOEt2. H2, Pd/C3. EtOH, H3O+

O

1. (R)-α-methylbenzylamine2. Na(CN)BH3, AcOH3. HCO2NH4, MeOH, Pd(OH)2, reflux

4. Na(CN)BH3, AcOH5. (CF3CO)2O, Et3N

H

O

CF3

147 148 146 149

145 151 150

144 153

45%

65%56%

90%

78%74% from 151

70%

77%

61%

HH

C

21 21H

H H

O

HOOC152

C

21

TIPSOTIPSOTIPSO

2. H2, Pd/C

TIPSO

2. BnOH, PhH, reflux

Scheme 22 Chackalamannil’s synthesis of himgaline (21) – part 2

NH

H

HO 158

154

N

O

H

HH

H

156O

CF3

O

1. NaIO4, RuCl3·H2O2. LHMDS, Me2S23. NaIO4, MeOH

4. toluene, 100 °CN

O

H

HH

H

143O

CF3

O O

1. NBS, AIBN2. AgOCOCF3

3. NaHCO34. DMP

NH

HO

H

HH

H

157

O

HOOC

HOOC

O

NHH

H

HO

HOOC

O

NH

H

HO 159

O

NHH

H

HO G.B. 13 (22)

O

=

NH

H

HO himgaline (21)

OH

1. Sc(OTf)3, cat. HCl2. Na(OAc)3BH, AcOH

HCl (6 M)microwave

80% – CO2

56% 77%

60%

17

1916

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



synthesis of ent-G.B. 13 and ent-himgaline in 2007.40 Asshown in Scheme 23, ent-G.B. 13 was simplified to 160,which should be accessible from decalin 161. With thisdisconnection of two rings, the complexity is significantlyreduced. Further carbon–carbon bond disconnections ofthe side-chains and dihydroxylation of the double bond fi-nally lead back to 162, which is easily obtained through aDiels–Alder reaction of a linear precursor.

Vinylogous Horner–Wadsworth–Emmons olefination of6,6-dimethoxyhexanal (163)43 with phosphonate 173 wasfollowed by diisobutylaluminum hydride reduction of theester and subsequent protection of the primary alcohol asa silyl ether (Scheme 24). The chiral auxiliary was in-stalled by reaction of the unmasked aldehyde with oxazo-lidinone 174 in a second Horner–Wadsworth–Emmonsolefination. When treated with dimethylaluminum chlo-ride at –30 °C in toluene, triene 164 cleanly affordedtrans-decalin 165 as a single diastereomer in good yield.69

Scheme 23 Retrosynthetic analysis of Evans’ synthesis of ent-him-galine (ent-21) and ent-G.B. 13 (ent-22)

ent-G.B. 13 (ent-22)

160

H

H

161

enaminealdol

H

H

OO

O

ROOC

OBocHN

NHH

HO

HO

H

NH

HO

H

O

Xc

162

Diels–Alder

Michaeladdition

O

RO

Scheme 24 Evans’ synthesis of ent-himgaline (ent-21) and ent-G.B. 13 (ent-22)

OHC

MeO OMe

TBDPSO

N OO

TBDPSO

O

1. 173, LiOH2. DIBAL-H3. TBDPSCl, DMAP

4. LiBF4, wet MeCN5. 174, LiClO4, DIPEA

Me2AlCl, PhMe, –30 °C

H

H

NO

O

O

Bn

Bn

CHO

TBDPSO

H

HO

O

1. K2OsO4, acetone, pH 7 buffer2. Me2C(OMe)2, TsOH3. LiSEt4. DIBAL-H

(EtO)2PO

OEt

173

NO

O O

P(OMe)2

O

Bn 174

160

H

H

167

OO

OH

OHBocHN

NH

HO

H

OH

H

H

O

1. LiClO4, DIPEA, 922. DIBAL-H3. TBAF

163 164 165 166

H

H

170

OO 1. DMP, NaHCO3

2. allyldiazo- acetate, SnCl2

OO

NBnBoc

O

O

H

H

O

O

BocHN

NHBoc

=

1. Pd(PPh3)4, morpholine2. DBU3. Pd(OH)2, H24. DMP, NaHCO3

LiOMe, LiClO4

OO

(MeO)2P

O

NHBoc

O

92

TFA (20%);NaHCO3;4 Å MS

O

NHH

HO

H1. AcOH2. NaBH3CN3. DMP, NaHCO3

HO

1. BnOCOCl, Na2CO32. IBX, TsOH3. TMSI

AcOH;NaBH(OAc)3

20%

81% 83%

63%

52%

62%

32% from 160

O

172

1617

1617

H

H

168

OO

OH

OHBocBnN

1. TBSOTf, 2,6-lutidine2. NaH, BnBr3. TBAF

H

H

169

OO

OO

NBnBoc

O

O

ent-G.B. 13 (ent-22)

NHH

HO

HO

H

ent-himgaline (ent-21)

NH

HOH

HO

H

171

2020

20

205

5 5

1617

C

5

16 17

19

17

19

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



Dihydroxylation of the C16–C17 double bond by reactionwith potassium osmate and subsequent protection of thevicinal cis-diol afforded aldehyde 166 after removal of thechiral auxiliary.

Horner–Wadsworth–Emmons olefination of 166 with N-Boc-protected ketophosphonate 9270 was followed by re-duction of the ketone functionality and desilylation to af-ford allylic alcohol 167. Next, the nitrogen wasbenzylated after intermediary protection of the two hy-droxy functionalities and 168 was obtained in good yield.Both hydroxy groups were oxidized with Dess–Martinperiodinane and the keto ester required for the Michaeladdition was introduced via a Roskamp reaction71 of thealdehyde. The desired b-keto ester 170 was observed;however, the compound was formed only as minor reac-tion product and the main product was identified to beenol ester 169. This observation can easily be explainedby conjugate addition of the enol oxygen of the b-keto es-ter. Fortunately, the process proved to be reversible uponexposure to basic conditions, and the desired Michael ad-duct was obtained after treatment with a mixture of lithi-um perchlorate and lithium methoxide.72 Decarboxylationwas achieved by addition of morpholine and Pd(PPh3)4

and the ketone at C16 was installed by 1,8-diazabicyc-lo[5.4.0]undec-7-ene promoted elimination of the ace-tonide to the corresponding allylic alcohol which washydrogenated with concomitant cleavage of the N-benzylgroup, and oxidized to afford triketone 171.

Next, the piperidine ring was to be formed and the re-quired ring-closure reaction proceeded smoothly uponcleavage of the carbamate under acidic conditions, and af-ter dehydration imine 160 was obtained. The stage was setfor the intramolecular enamine aldol addition which tookplace under acidic conditions. Reduction of the iminiumfunctionality with sodium cyanoborohydride also resultedin reduction of the keto functionality and the alcohol hadto be reoxidized with Dess–Martin periodinane to deliverintermediate 172.

The synthesis of ent-G.B. 13 was completed by installa-tion of the C17–C19 double bond which was achieved byprotection of the amine as a carbamate, oxidation bymeans of 2-iodoxybenzoic acid and then deprotection ofthe nitrogen. The conversion of ent-G.B. 13 to ent-himga-line proceeded in the same fashion as described byChackalamannil39 by exposure to acetic acid and reduc-tion with sodium (triacetoxy)borohydride.

Evans’ synthesis of the natural product is another exampleof the high versatility of chiral oxazolidinone as auxiliaryin enantioselective Diels–Alder reactions. The decalinsystem is prepared in a short synthetic sequence. As thesynthesis suffers from several unexpected problems, how-ever, the elaboration of the remaining ring systems re-quired more steps than originally anticipated.

Sarpong, 2009 [(±)-G.B. 13]

The latest and also shortest total synthesis of racemic G.B.13 (22) was published by Sarpong in 2009.41 The synthe-sis is based on the utilization of a bromopyridine (178) assynthon for the piperidine ring of the natural product(Scheme 25). The piperidine is obtained by hydrogena-tion of pyridine 175 at a late stage of the synthesis. Dis-connection of the C5–C20 bond results in 176 whichshould be assembled via 1,2-addition of ketone 177 and178. The key step in the formation of tricyclic ketone 177is a Diels–Alder cycloaddition reaction.

An endo-selective Lewis acid catalyzed Diels–Alder reac-tion of silyl enol ether 117 and a,b-unsaturated ketone 179delivered, after in situ epimerization, cycloadduct 181.73

Cyclopentenone 177 was obtained by retro-Diels–Alderreaction effected by flash vacuum pyrolysis in excellentyield (Scheme 26).

Scheme 25 Retrosynthetic analysis of Sarpong’s synthesis of (±)-G.B. 13 (22)

NHH

H

HO G.B. 13 (22)

O

NH

H

HO 175

O

N

OMe

H

HO

Br

OH

H

N

Br

OMe

+

pyridine reduction

Rh-catalyzedketone hydroarylation

1,2-addition

O

Diels–Alder

176

178

177

H

H

520

520

H H

HH H

HO

Scheme 26 Sarpong’s synthesis of (±)-G.B. 13 (22) – part 1

OTBS

O

+

OTBS

O

OTBS

O

Yb(tmhd)3110 °C

600 °C

179

117

181177

H

H

H

HHH

HH

180

OTBS

O

H

HHH

H

85%

86%

H H

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



Next, the pyridine ring was installed (Scheme 27): 1,2-ad-dition of lithiated 178, available by methylation of com-mercially available 3-bromo-6-hydroxy-2-picoline,74

proceeded with complete diastereocontrol, presumablybecause of the steric influence of the hydrogen at C9, and182 was obtained after hydrolysis of the silyl enol ether.In order to establish the correct oxygenation pattern, an al-lylic hydroxy group transposition had to be performed.After standard procedures failed to give the desired prod-uct, Larson and Sarpong41 found that the desired 1,3-transposition was successfully accomplished usingParikh–Doering conditions75 and 176 was obtained afterhydrogenation of the double bond using platinum oxide ascatalyst. The hydrogenation reaction proceeded from thea-face with a diastereomeric ratio greater than 95:5 in fa-vor of the desired material. To form the crucial C5–C20bond, Larson and Sarpong attempted halogen–metal ex-change after oxidation to the C20 ketone. As no productcould be obtained via this route, the bromine was substi-tuted for a boronic acid ester. After C20 oxidation andepimerization to the cis-fused pentanone (yielding 183), avariety of palladium catalysts were investigated. Unfortu-nately, those, too, failed to deliver the desired product.When the researchers finally turned to cationic rhodiumcatalysts, they met with success: [Rh(cod)(MeCN)2]

+BF4

in the presence of triethylamine gave 184 in 77% yield.According to the authors, this is the first example of an ad-dition of an aryl boron species onto an unactivated ketoneusing rhodium(I) catalysis.

Next, a methyl group was installed at the pyridine ring atC2 instead of the methoxy functionality. This wasachieved by cleavage of the methyl ether using sodiumethanethiolate in N,N-dimethylformamide at 120 °C fol-lowed by triflation of the resultant pyridine and cross-cou-pling via exposure to trimethylaluminum and Pd(PPh3)4.At this point, the pyridine was hydrogenated to establishthe piperidine moiety and the reaction afforded the desired

material in good diastereoselectivity from the exo-face.The synthesis of the natural product was completed byprotection of the nitrogen as a carbamate, installation ofthe C17–C19 double bond by oxidation with 2-iodoxy-benzoic acid, and final cleavage of the carbamate.

Sarpong thus presented a beautiful and concise total syn-thesis of racemic G.B. 13 and clearly points out the advan-tage of the hydrogenation of pyridine in the formation ofpiperidine motifs in the preparation of complex alkaloids.In comparing Sarpong’s addition of lithiated pyridine 178to enone 177 with Movassaghi’s annulation chemistrywith imine 96, both authors introduce the same carbonframework with different synthetic strategies. Althoughthe natural product was prepared in racemic form, the syn-thesis could be modified by starting with an enantioselec-tive Diels–Alder76 to allow the generation of enantio-merically enriched material. Besides the brevity of thissynthetic effort, the rhodium-catalyzed 1,2-addition iswell worth mentioning.

3 Conclusion

Although the chemical interest in Galbulimima alkaloidsclearly started with, and benefited from, the discovery ofthe promising biological properties of himbacine, re-search in this area has also been motivated by the synthet-ic challenges of these structurally complex andcompelling natural products. After the first synthesis ofhimbacine by Hart and Kozikowski in 1995, Galbulimimaclass I alkaloids became the main research focus and near-ly a full decade had passed before the total synthesis of aclass III alkaloid appeared in the literature. Today, severalgroups have contributed to this field and current researchactivity centers around the class II and class III alkaloids.

All syntheses utilize Diels–Alder reactions as key steps toestablish the decalin system or to generate the cyclohex-

Scheme 27 Sarpong’s synthesis of (±)-G.B. 13 (22) – part 2

O

OH

N

Br

MeO

1. LDA, 1782. HCl; K2CO3 N

OMe

H

HO

BrHO

N

OMe

H

HO

pinBO

1. SO3⋅py, py, DMSO then pH 7 buffer2. H2, PtO2

1. (Bpin)2, Pd2(dba)3 PCy3HBF4

2. DMP3. Et3N, SiO2

N

OMe

H

HO

NH

HO

HO

[Rh(cod)(MeCN)2]BF4Et3N, PhMe, 80 °C

177

1. NaSEt, DMF, 120 °C2. Tf2O, py3. Me3Al, Pd(PPh3)4

NHH

HO

HO

1. Rh/Al2O3, H22. BnOCOCl, NaHCO33. IBX, TsOH, DMSO

G.B. 13(22)

4. TMSI; HCl; NaOH

175 184

183

58% 51%

51%

77%

54%47%

H

O

N

Br

MeO

182

H

=H

H

HHO

H

HO

17

19

1919

H

HHH

HH

H

HHHHH

176

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



ane-fused g-lactone. Chiral piperidine derivatives areused in syntheses of Galbulimima class I alkaloids and thestereochemical information of C2 can always be tracedback to lactic acid as an inexpensive chiral pool startingmaterial. The approaches differ mainly in the differentbonds formed through the key Diels–Alder cycloadditionreaction and the reaction type employed to couple the pi-peridinoyl side chain to the decalin system.

Syntheses of Galbulimima class II and III alkaloids aremore complex as these targets contain a highly fused ringsystem. Mander was first to publish the preparation of ra-cemic G.B. 13. The conversion of an aromatic moiety intothe piperidine ring is interesting and demonstrated thehigh potential of alkyne functionalization with hydroxyl-amine for alkaloid synthesis. Chackalamannil utilizedfragments prepared during earlier approaches towardsGalbulimima class I alkaloids and used well-establishedreaction sequences to prepare the natural product. Evans’contribution to this field clearly proves the versatility ofchiral auxiliaries in enantioselective Diels–Alder reac-tions. Sarpong and Movassaghi presented highly concisesyntheses of Galbulimima alkaloids, and both were alsoable to develop new synthetic strategies such as the formal[3+3]-cycloaddition reaction (Movassaghi) or the rhodi-um-catalyzed hydroarylation reaction (Sarpong).

This review article summarizes approximately fifteenyears of research on Galbulimima alkaloids and coversthis field from the first successful synthesis of himbacineto the preparation of G.B. 13, himandrine, and himgaline.Recent contributions are indicative of the ongoing interestin this area and there is no doubt that further interestingcontributions will be made.

Note Added in ProofAfter this article was accepted for publication, two syntheses ofGalbulimima alkaloids were reported.77,78

Acknowledgment

The authors thank Prof. Edda Gössinger for helpful discussions andcareful proofreading of the manuscript.

References

(1) Sprague, T. A. J. Bot. 1922, 60, 129.(2) Bailey, I. W.; Nast, C. G.; Smith, A. C. J. Arnold Arbor.

1943, 24, 190.(3) Bullock, A. A. Kew Bull. 1957, 409.(4) Jessup, L. W. In Flora of Australia, Australian Biological

Resources Study, Vol. 2; CSIRO Publishing: Canberra, 2007, 11.

(5) Glick, L. B. Ethnology 1967, 6, 31.(6) Thomas, B. J. Psychoactive Drugs 2005, 37, 109.(7) Holdsworth, D.; Sakulas, H. Int. J. Crude Drug Res. 1986,

24, 31.(8) Webb, L. J. C.S.I.R. Bull. 1948, 232.(9) Mander, L. N.; Willis, A. C.; Herlt, A. J.; Taylor, W. C.

Tetrahedron Lett. 2009, 50, 7089.

(10) Ritchie, E.; Taylor, W. C. In The Alkaloids, Vol. 13; Manske, R. H. F., Ed.; Academic Press: New York/London, 1971, 227.

(11) Picture of Galbulimima baccata published with permission from CSIRO Publishing; Scienceimage file number DA7199; Photographer: Andrew Ford.

(12) Darroch, S.; Taylor, W.; Choo, L. K.; Mitchelson, F. Eur. J. Pharmacol. 1990, 183, 1720.

(13) Darroch, S. A.; Taylor, W. C.; Choo, L. K.; Mitchelson, F. Eur. J. Pharmacol. 1990, 182, 131.

(14) Takadoi, M.; Katoh, T.; Ishiwata, A.; Terashima, S. Tetrahedron 2002, 58, 9903.

(15) Miller, J. H.; Aagaard, P. J.; Gibson, V. A.; Mckinney, M. J. Pharmacol. Exp. Ther. 1992, 263, 663.

(16) Chackalamannil, S.; Xia, Y.; Greenlee, W. J.; Clasby, M.; Doller, D.; Tsai, H.; Asberom, T.; Czarniecki, M.; Ahn, H. S.; Boykow, G.; Foster, C.; Agans-Fantuzzi, J.; Bryant, M.; Lau, J.; Chintala, M. J. Med. Chem. 2005, 48, 5884.

(17) Chackalamannil, S.; Xia, Y. Expert Opin. Ther. Pat. 2006, 16, 493.

(18) Pinhey, J. T.; Ritchie, E.; Taylor, W. C. Aust. J. Chem. 1961, 14, 106.

(19) Ritchie, E.; Taylor, W. C. In The Alkaloids, Vol. 9; Manske, R. H. F., Ed.; Academic Press: New York/London, 1967, 529.

(20) Fridrichsons, J.; Mathieson, A. M. Acta Crystallogr. 1962, 15, 119.

(21) Lovell, F. M. Proc. Chem. Soc., London 1964, 58.(22) Mander, L. N.; Ritchie, E.; Taylor, W. C. Aust. J. Chem.

1967, 20, 981.(23) Mander, L. N.; Ritchie, E.; Taylor, W. C. Aust. J. Chem.

1967, 20, 1021.(24) Guise, G. B.; Mander, L. N.; Prager, R. H.; Rasmusse, M.;

Ritchie, E.; Taylor, W. C. Aust. J. Chem. 1967, 20, 1029.(25) Mander, L. N.; Prager, R. H.; Rasmusse, M.; Ritchie, E.;

Taylor, W. C. Aust. J. Chem. 1967, 20, 1473.(26) Mander, L. N.; Prager, R. H.; Rasmusse, M.; Ritchie, E.;

Taylor, W. C. Aust. J. Chem. 1967, 20, 1705.(27) Movassaghi, M.; Hunt, D. K.; Tjandra, M. J. Am. Chem. Soc.

2006, 128, 8126.(28) Willis, A. C.; O’Connor, P. D.; Taylor, W. C.; Mander, L. N.

Aust. J. Chem. 2006, 59, 629.(29) Movassaghi, M.; Tjandra, M.; Qi, J. J. Am. Chem. Soc. 2009,

131, 9648.(30) Movassaghi, M.; Chen, B. Angew. Chem. Int. Ed. 2007, 46,

565.(31) Hart, D. J.; Wu, W. L.; Kozikowski, A. P. J. Am. Chem. Soc.

1995, 117, 9369.(32) Chackalamannil, S.; Davies, R. J.; Asberom, T.; Doller, D.;

Leone, D. J. Am. Chem. Soc. 1996, 118, 9812.(33) Takadoi, M.; Katoh, T.; Ishiwata, A.; Terashima, S.

Tetrahedron Lett. 1999, 40, 3399.(34) Chackalamannil, S.; Davies, R.; McPhail, A. T. Org. Lett.

2001, 3, 1427.(35) Gao, L. J.; Waelbroeck, M.; Hofman, S.; Van Haver, D.;

Milanesio, M.; Viterbo, D.; De Clercq, P. J. Bioorg. Med. Chem. Lett. 2002, 12, 1909.

(36) Wong, L. S. M.; Sherburn, M. S. Org. Lett. 2003, 5, 3603.(37) Tchabanenko, K.; Adlington, R. M.; Cowley, A. R.;

Baldwin, J. E. Org. Lett. 2005, 7, 585.(38) Mander, L. N.; McLachlan, M. M. J. Am. Chem. Soc. 2003,

125, 2400.(39) Shah, U.; Chackalamannil, S.; Ganguly, A. K.; Chelliah, M.;

Kolotuchin, S.; Buevich, A.; McPhail, A. J. Am. Chem. Soc. 2006, 128, 12654.

(40) Evans, D. A.; Adams, D. J. J. Am. Chem. Soc. 2007, 129, 1048.

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.



(41) Larson, K. K.; Sarpong, R. J. Am. Chem. Soc. 2009, 131, 13244.

(42) Hart, D. J.; Li, J.; Wu, W. L.; Kozikowski, A. P. J. Org. Chem. 1997, 62, 5023.

(43) Schreiber, S. L.; Claus, R. E.; Reagan, J. Tetrahedron Lett. 1982, 23, 3867.

(44) Griffith, W. P.; Ley, S. V.; Whitcombe, G. P.; White, A. D. J. Chem. Soc., Chem. Commun. 1987, 1625.

(45) Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108, 7408.

(46) Tanikaga, R.; Nozaki, Y.; Tamura, T.; Kaji, A. Synthesis 1983, 134.

(47) Hayakawa, K.; Yamaguchi, Y.; Kanematsu, K. Tetrahedron Lett. 1985, 26, 2689.

(48) Ito, Y.; Kobayashi, Y.; Kawabata, T.; Takase, M.; Terashima, S. Tetrahedron 1989, 45, 5767.

(49) Van Cauwenberge, G.; Gao, L. J.; Van Haver, D.; Milanesio, M.; Viterbo, D.; De Clercq, P. J. Org. Lett. 2002, 4, 1579.

(50) Marshall, J. A.; Xie, S. P. J. Org. Chem. 1995, 60, 7230.(51) Bal, B. S.; Childers, W. E.; Pinnick, H. W. Tetrahedron

1981, 37, 2091.(52) Barany, G.; Schroll, A. L.; Mott, A. W.; Halsrud, D. A.

J. Org. Chem. 1983, 48, 4750.(53) Brown, S. P.; Brochu, M. P.; Sinz, C. J.; MacMillan, D. W.

C. J. Am. Chem. Soc. 2003, 125, 10808.(54) Martin, J. C.; Arhart, R. J. J. Am. Chem. Soc. 1971, 93, 4327.(55) Jung, M. E.; Lyster, M. A. J. Chem. Soc., Chem. Commun.

1978, 315.(56) Graening, T. Nachr. Chem. 2009, 57, 1203.(57) Hook, J. M.; Mander, L. N. J. Org. Chem. 1980, 45, 1722.(58) Regitz, M.; Ruter, J. Chem. Ber./Recl. 1968, 101, 1263.(59) Arapakos, P. G. J. Am. Chem. Soc. 1967, 89, 6794.(60) Pradhan, S. K.; Akamanchi, K. G.; Divakaran, P. P.

Tetrahedron Lett. 1983, 24, 5017.(61) Pradhan, S. K.; Akamanchi, K. G.; Divakaran, P. P.;

Pradhan, P. M. Heterocycles 1989, 28, 813.

(62) Avril, J. L.; Marcot, B.; Coquillay, M.; De Rango, C.; Moskowitz, H.; Mayrargue, J. New J. Chem. 1999, 23, 743.

(63) Moran, J.; Gorelsky, S. I.; Dimitrijevic, E.; Lebrun, M. E.; Bedard, A. C.; Seguin, C.; Beauchemin, A. M. J. Am. Chem. Soc. 2008, 130, 17893.

(64) Beauchemin, A. M.; Moran, J.; Lebrun, M. E.; Seguin, C.; Dimitrijevic, E.; Zhang, L.; Gorelsky, S. I. Angew. Chem. Int. Ed. 2008, 47, 1410.

(65) Nicolaou, K. C.; Zhong, Y. L.; Baran, P. S. J. Am. Chem. Soc. 2000, 122, 7596.

(66) Shah, U.; Chackalamannil, S.; Ganguly, A. K. Total Synthesis of Galbulimima Alkaloids: Himgaline and GB13; Vdm Verlag Dr. Müller: Saarbrücken, 2008.

(67) Four, P.; Guibe, F. J. Org. Chem. 1981, 46, 4439.(68) Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B.

J. Org. Chem. 1981, 46, 3936.(69) Evans, D. A.; Chapman, K. T.; Bisaha, J. J. Am. Chem. Soc.

1988, 110, 1238.(70) Tchabanenko, K.; Chesworth, R.; Parker, J. S.; Anand, N.

K.; Russell, A. T.; Adlington, R. M.; Baldwin, J. E. Tetrahedron 2005, 61, 11649.

(71) Holmquist, C. R.; Roskamp, E. J. J. Org. Chem. 1989, 54, 3258.

(72) Raban, M.; Noe, E. A.; Yamamoto, G. J. Am. Chem. Soc. 1977, 99, 6527.

(73) Borsato, G.; De Lucchi, O.; Fabris, F.; Lucchini, V.; Frascella, P.; Zambon, A. Tetrahedron Lett. 2003, 44, 3517.

(74) Haudrechy, A.; Chassaing, C.; Riche, C.; Langlois, Y. Tetrahedron 2000, 56, 3181.

(75) Parikh, J. R.; Doering, W. V. E. J. Am. Chem. Soc. 1967, 89, 5505.

(76) Hu, Q. Y.; Zhou, G.; Corey, E. J. J. Am. Chem. Soc. 2004, 126, 13708.

(77) Synthesis of G.B. 13 and G.B. 16: Zi, W.; Yu, S.; Ma, D. Angew. Chem. Int. Ed. 2010, 49, 5887.

(78) Synthesis of G.B. 13: McLachlan, M. M. W.; O’Conner, P. D.; Fairweather, K. A.; Willis, A. C.; Mander, L. N. Aust. J. Chem. 2010, 63, 742.

Dow

nloa

ded

by: U

nive

rsitä

t Wie

n. C

opyr

ight

ed m

ater

ial.

syntheses of galbulimima alkaloids · 2 syntheses of galbulimima alkaloids several syntheses of...

Documents