Total Synthesis of Leupyrrins A1 and B1, Highly Potent Antifungal Agents from the Myxobacterium Sorangium cellulosum

A Journal of

Accepted Article

Title: Total Synthesis of Leupyrrins A1 and B1, Highly Potent Antifungal

Agents from the Myxobacterium Sorangium cellulosum

Authors: Sebastian Thiede, Paul R. Wosniok, Daniel Herkommer,

Thomas Debnar, Maoqun Tian, Tongtong Wang, Michael

Schrempp, and Dirk Menche

This manuscript has been accepted after peer review and appears as an

Accepted Article online prior to editing, proofing, and formal publication

of the final Version of Record (VoR). This work is currently citable by

using the Digital Object Identifier (DOI) given below. The VoR will be

published online in Early View as soon as possible and may be different

to this Accepted Article as a result of editing. Readers should obtain

the VoR from the journal website shown below when it is published

to ensure accuracy of information. The authors are responsible for the

content of this Accepted Article.

To be cited as: Chem. Eur. J. 10.1002/chem.201604445

Link to VoR: http://dx.doi.org/10.1002/chem.201604445

Supported by

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Total Synthesis of Leupyrrins A₁and B₁, Highly Potent Antifungal

Agents from the Myxobacterium Sorangium cellulosum

Sebastian Thiede,^[a]Paul R. Wosniok,^[a]Daniel Herkommer,^[a,b]Thomas Debnar,^[a,c]Maoqun Tian,^[a,d]

Tongtong Wang,^[a,e]Michael Schrempp,^[a]and Dirk Menche*^[a]

Dedicated to Prof. Dr. Dr. h.c. mult. Gerhard Bringmann on the occasion of his 65th birthday.

Abstract: Full details on the design, elaboration and application

of efficient strategies for the high-yielding total syntheses of

leupyrrins A₁and B₁, unique antifungal agents from the

myxobacterium Sorangium cellulosum are reported. A sequential

zirconocene-mediated diyne-cyclization and regioselective

opening of the zirconacyclopentadiene intermediate enabled a

concise entry into the unique dihydrofuran fragment, while

another domino reaction was developed for the butyrolactone

involving a one-pot lactol opening, stereoselective aldehyde

addition and in situ-lactonization. Furthermore, an innovative sp²-

sp³-cross coupling for pyrrole functionalization and an optimized

HATU-mediated amide coupling protocol of two elaborate

fragments were established. In addition, an unusual protective

group strategy, involving a Teoc-acetonide protected amine in

combination with tert-butyl and acetate esters, was successfully

elaborated. These tactics and strategies are generally useful and

may be applied also in the synthesis of other functionalized

compounds. Furthermore, it is expected that the material that was

obtained by these total syntheses will enable the further

exploration of the biological profile of these potent antifungal

agents.

for target identification or evaluation.^{[ 2 ]}Among these gliding

bacteria, the myxobacterium Sorangium cellulosum which

contains the largest bacterial genome sequenced to date has

proven the richest producer of structurally novel and diverse

metabolites.^[3]The leupyrrins^[4]hold a special place among these

fascinating structures, as they originate by one of the most

unusual biosyntheses^[5]for myxobacterial metabolites reported so

far.^{[ 6 ]}This includes an unusual construction of the densely

substituted γ-butyrolactone, presumably by an epoxide-opening

cascade, and

a

remarkable biosynthesis of the unique

dihydrofuran moiety bearing two exocyclic alkylidens by a Prins-

type cyclization.^{[ 7 ]}As shown in Figure 1 for the six natural

leupyrrins, i.e. leupyrrins A₁(1), A₂(5), B₁(3), B₂(6), C (2) and D

(4), these structures feature an 18 membered nonsymmetric

macrodiolide core that also contains a pyrrole, an oxazoline and a

diacid fragment as additional characteristic features together with

a side chain that is attached to the oxazoline moiety.^[4]The

natural leupyrrins differ in the methylation pattern of the terminal

hydroxyl and/or the olefination pattern, either in the side chain or

the diacid subunit. They contain up to seven stereogenic centers,

which have been recently assigned in our group by extensive

NMR-studies in combination with molecular modeling and

chemical methods.^[8]

Leupyrrin A₁has demonstrated highly potent antifungal properties

against Rhodotorula glutinis, Aspergillus niger, Botrytis cinerea,

and Mucor hiemalis with MIC-values in the submicromolar range

as well as antiproliferative activities against mouse fibroblast

cells.^[4]in addition, anti-HIV inhibitory effects have been reported

for leupyrrin B₁.^[9]In cellular assays, leurpyrrin A₁has been shown

to efficiently inhibit DNA, RNA, and protein syntheses. On the

other hand, no interaction with the respiratory chain, membrane

disruption or conductivity changes were observed, which led to

the suggestion that an unusual and so far undefined molecular

target may be involved.^[4]Very recently, human leukocyte

elastase has been discovered as a first molecular target of

leupyrrins A₁and B₁,^[10]which may however not explain the potent

antifungal activity of this unique class of natural products.

Introduction

Myxobacteria present one of the most prolific sources of

architecturally novel natural products,^[1]which often demonstrate

a wide variety of important potent biological activities. In many

cases biologically relevant targets are selectively addressed^[1e]

and consequently these compounds may become important tools

[a]

Kekulé-Institut für Organische Chemie und Biochemie

Universität Bonn

Gerhard-Domagk-Str. 1, 53121 Bonn, Germany

E-mail: dirk.menche@uni-bonn.de

[b]

[c]

Current address: GlaxoSmithKline, Medicines Research Centre,

Gunnels Wood Road, Stevenage, SG1 2NY (UK)

Current address: Dottikon Exclusive Synthesis AG, Dottikon

(Switzerland)

The important biological properties together with their unique

biosynthesis and intriguing molecular architectures have rendered

the leupyrrins attractive synthetic targets.^{[8,10, 11 ]}Herein, we

describe in full detail the design, evaluation and elaboration of the

various synthetic tactics that we pursued to access the unique

[d]

[e]

Current address: Scripps Research Institute, La Jolla (USA)

Current address: Institute of Quality Standard and Testing

Technology for Agro-products, Chinese Academy of Agricultural

Sciences, and Key Laboratory of Agro-food Safety and Quality,

Ministry of Agriculture, Beijing, China

12

]

three-dimensional architecture of the leupyrrins,^[

which

eventually culminated in first total syntheses of leupyrrins A₁and

B₁.

Supporting information for this article is available on the web under

wiley.com

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Figure 1. The leupyrrins: biosynthetically and structurally unique metabolites

from the myxobacterium Sorangium cellulosum with potent antifungal properties.

Results and Discussion

Retrosynthetic Analysis

As shown in Scheme 1, two main strategies were pursued to

access the unique macrodiolide core of leupyrrins A₁(1) and B₁

(3). These involve either a challenging ring closing metathesis to

forge the tri-substituted C5-C6 double bond (route A) or a

cyclodehydration (route B) in combination with more conventional

esterifications along the C4’-C25 or the C3-C1’ lactone. These

disconnections would either lead to a Northern and Southern

Fragment 7 and 8 or a Western and Eastern fragment 9 and 10,

respectively. These fragments in turn may originate from four

common subunits, i.e. a pyrrole, a dihydrofuran, a butyrolactone

and a diacid building block 11, 12, 13 and 14, which may be

connected by an oxazoline formation and esterification (route A)

or an sp²-sp³cross coupling and an alternative ester formation

(route B). Notably, this approach enables a high degree of

flexibility to this synthetic approach.

Scheme 1. Modular retrosynthetic analysis of leupyrrins A₁(1) and B₁(3) relying on two alternative disconnections leading either to a Northern and Southern

fragment (7 and 8) or a Western and Eastern fragment (9 and 10), which in turn may originate from four common subunits.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Fragment Syntheses

hydroxylation,^[14]which would enable the most concise entry into

the required substitution pattern. As shown in Scheme 3, we

studied the conversion of unsymmetrical alkene 21, which was

available from alkyne 20 by E-selective LAH-reduction of the

triple bond and sequential introduction of two different protective

groups, i.e. a PMB and a TBS ether. However, the crucial

oxidation reactions of 21 resulted in essentially no

regioselectivity giving the desired aminohydroxylated product 23

together with equimolar amounts of the here undesired isomer

22.^[15]While this unfavorable outcome could not be remedied by

modification of reaction conditions, 23 could still be elaborated to

alkyne 26 by homologation of the derived aldehyde 24 using the

Ohira-Bestmann reagent 25.^[16]Nevertheless, this route was not

efficient enough to enable preparation of sufficient material for

the projected total synthesis.

Furan Building Block

As shown in Scheme 2 (bottom part), our strategy to access the

unique dihydrofuran subunit 12 of the leupyrrins was based on a

zirconium-mediated cyclization of a 1,6-diyne of type 19.^[13]As

depicted for the simplified analogs, this relies on a sequential

process involving an intramolecular oxidative cyclization,

followed by a stepwise oxidative cleavage with two different

electrophiles. As previously reported, this strategy could indeed

be realized for diynes of type 15 which bear the original side

chain on one, and truncated version on the other side of the

diyne.^[11c]Importantly, excellent degrees of regioselectivity

(>20:1) were obtained in the oxidative cleavage of the

intermediate zirconacyclopentadiene 16 with NBS, which were

rationalized by DFT-calculations, suggesting a coordination of

the terminal OMe-ether of the original side chain to the low-

valent zirconium metal. Presumably, this leads to a movement of

the Cp-rings closer to C15 thus favoring a bromination at C18 in

agreement with the experimental results. These studies

demonstrated that the alkyne substituent at C15 must be

designed in a way that such coordination would be less

favorable.

Scheme 3. Unselective amino hydroxylation of alkene 21 and first preparation

of alkyne 26.

In order to overcome the lack of selectivity of 21, we decided to

evaluate alkenes with two electronically different substituents.

Gratifyingly, the Sharpless amino hydroxylation of enoate 28,

17

]

which was obtained from 27 by a known sequence,^[

proceeded with excellent regio- and enantioselectivities to give

the desired product 29, in agreement with a previous report

(Scheme 4).^[18]Also, the further elaboration towards aldehyde 30

could be readily reproduced. Subsequent conversion to alkyne

31 with the Ohira-Bestmann reagent 25, reduction of the ester

and PMB-protection of resulting terminal hydroxyl group

proceeded smoothly. Various methods were then studied for the

asymmetric addition to aldehyde 32. However, in all attempts^[19]

only moderate stereoselectivities were obtained. Subsequent

efforts to remedy this unfavorable outcome by an oxidation of

the resulting propargylic alcohol 33 and stereoselective

Scheme 2. One-pot zirconocene-mediated diyne-cyclization and regio-

selective opening of zirconacyclopentadienes to access simplified furan

fragments of type 17.^[11c]

Based on these first studies, we turned our efforts towards the

authentic fragments, which required an enantioselective access

to the appending aminoalcohol functionality. We first evaluated

the

Sharpless

protocol

for

an

asymmetric

amino

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

reduction were also not successful. Only the here undesired

(26R)-isomer 34 was selectively obtained by using the (R)-CBS-

reagent.^[20]Therefore, we decided to invert this configuration by

a Mitsunobu-reaction.²¹Finally, etherification with 35 gave the

authentic diyne building block 36 with the required substitution

pattern and configuration.

Scheme 5. Concise synthesis of authentic furan 39 and unsuccessful efforts

to liberate the amino alcohol functionality.

Presumably, the diene fragment was too labile for a removal of

the Boc protective group.^{[ 24 ]}Therefore, we decided to first

remove the protective groups and install the oxazoline before

diyne cyclization. As shown in Scheme 6, deprotection of the

PMB-, the Boc and the acetonide group of cyclization precursor

36 could be realized in high yields (94%) using a solution of HCl

in MeOH.^{[ 25 , 26 ]}After selective reprotection of the terminal

hydroxyl as a TBS ether, we then focused on oxazoline

formation of 42 with a suitable pyrrole carboxylate, i.e. 43 or 46.

As 42 already contains the correctly configured amino-bearing

stereogenic center, our efforts had to focus on methods that

would proceed with retention. Accordingly, we first evaluated a

method by the group of Arndt^{[ 27 ]}via azide 45, which was

efficiently obtained from 42 by treatment with TfN₃in the

presence of CuSO₄.^[28]We then studied the esterification of 45

with pyrroles 43 and 46. However, despite considerable efforts

with various protocols, the desired compound 48 could not be

obtained. Instead, a distinct tendency of the pyrrole carboxylate

for self-condensation and anhydride formation was observed,

which may be rationalized by the pronounced nucleophilicity of

the carboxylate based on the electron rich pyrrole and the steric

hindrance of the free hydroxyl of 45. This unfavorable outcome

could also not be corrected by using the pyrrole anhydride

directly.

Scheme 4. Stereoselective synthesis of authentic diyne 36.

At this stage we submitted 36 to our zirconocene-mediated

sequential 3-step protocol. As shown in Scheme 5, the diyne

was treated with a freshly prepared solution of zirconocene at -

78 °C. After completing the formation of zirconacyclopentadiene

37 by warming to room temperature, the mixture was again

cooled to -78 °C and treated with NBS (2 eq) before hydrolysis.

Gratifyingly, following this sequence, the desired bromide 38

was obtained in high yields (88%) and excellent regioselectivity

(> 20:1). After replacement of the halide by a methyl group with

^tBuLi and Me₂SO₄we then evaluated the removal of the

protective groups, i.e. by first cleavage of the acetonide 39^[22]

with hydrochloric acid, obtained in situ from acetyl chloride and

MeOH, then liberation of the amine of 40.^[23]A broad variety of

conditions were evaluated, including various Brønsted (HCl, TFA,

TsOH) and Lewis acids (ZnBr₂, MgBr₂, SiCl₄) as well as

alternative reagents (TMSI, CAN). However, despite

considerably efforts, the deprotected product 41 neither with nor

without the terminal PMB group could be obtained.

In contrast, an alternative entry into the desired oxazoline,

involving firstly an amide formation and then dehydration was

successful. In detail, after HATU-mediated coupling of 42 with

pyrrole 43,^[29]condensation of 44 was effected by azeotropic

removal of water with toluene in the presence of catalytic

amounts of (NH₄)₂MoO₄.^[30]The resulting oxazoline 47 was then

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

submitted to our previously established oxidative zirconocene

mediated diyne cyclization protocol. However, despite

considerable efforts and variations of the reaction parameters,

no traces of the desired product 49 were obtained, suggesting

that the pyrrole moiety may not be compatible with the reaction

conditions.

as a ^tBu, PMB or TBDPS ether to give 54, 55, and 56,

respectively.

Scheme 7. Stereoselective synthesis of diynes of type 53 with various

protective groups based on an asymmetric alkyne addition.

However, in subsequent zirconocene-mediated cyclization

studies of these diynes, both the TBS and the TBDPS group

were shown to be too labile during this oxidative process and

only the deprotected product was obtained (R = H) in low yields

(12-24%). As shown in Scheme 8, good yields were only

obtained with a terminal PMB-group which proved stable under

the conditions giving furan 57 in 70% yield.

Scheme 6. Synthesis of a diyne-precursor bearing an appending oxazoline

pyrrole 47 and its evaluation in a zirconocene mediated cyclization.

Scheme 8. Zirconocene-mediated cyclization of diynes with

propargylic alcohol bearing different protective groups.

a terminal

Based on these results we reevaluated our strategy and focused

on an early stage introduction of the furan. In order to enable a

high degree of flexibility, diynes of type 53 bearing a propargylic

alcohol with different protective groups were targeted. As shown

in Scheme 7, their synthesis was initiated by an asymmetric

addition of alkyne 50 to isovaleraldehyde (32). Following a

protocol of Carreira^[31]in the presence of catalytic amounts of N-

methylephedrin (A) this coupling proceeded in high yields (84%)

and stereoselectivity (94% ee). After etherification of the

resulting alcohol 51 with tosylate 35^[11c,d]and removal of the

terminal TBS group of 52, the free hydroxyl of 53 was modified

We then analyzed, whether these substrates may be suitable for

a regioselective introduction of a bromide by a stepwise

treatment of the zirconacyclopentadiene intermediates with NBS

and NH₄Cl. However, as shown in Table 1, no selectivity was

obtained with the PMB protected substrate 55 giving the two

isomeric products 55a and b (R²= PMB) in equimolar amounts

(entry 3). Also, a low preference for one of the products was

t

obtained for the Bu-ether 54 (entry 2). In contrast, a certain

selectivity of the TBS-protected substrate 52 in favor of the

desired, deprotected isomer 53a was observed (ratio 5:1).

However, this compound could not be further elaborated due to

ensuing difficulties in reprotecting the terminal allylic alcohol as a

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

TBS ether, as the allylic alcohol proved too unstable under the

conditions.

Table 1. Zirconocene-mediated furan synthesis with variously protected

propargylic alcohols: evaluation of a regioselective opening.

Scheme 9. Concise synthesis of diyne 61 bearing an unprotected vicinal

amino alcohol functionality.

Entry

Substrate

Ratio^[a]

However, as shown in Table 2, treatment of TBS-protected

alcohol 60 with aqueous ammonia under microwave conditions

initially gave no conversion to 62 at 80 °C or 130 °C, possibly

due to the low solubility of the starting material in the aqueous

medium (entries 1 and 2), which could also not be remedied by

addition of EtOH (entry 3).

1

52 (R¹= TBS)

(53a/53b) 5:1 (R = H)

2

3

54 (R¹= ^tBu)

(54a/54b) 2:3 (R = ^tBu)

(55a/55b) 1:1 (R = PMB)

55 (R¹= PMB)

Table 2. Regioselective opening of epoxides 59 and 60 with ammonia.^[a]

[a] Only low yield of the products (25-40%) and variable amounts of the

starting material were obtained.

Entry

Substrate

Power / Max.

Temperature

Time

Yield

(Compound)

To circumvent the problems associated with the deprotection of

40 (Scheme 5) and 52 (Table 1) attempts were directed to

synthesize the fully unprotected aminodiole 61, thus offering

complete flexibility with regards to the protective group strategy.

Due to difficulties of our original oxazoline route (Scheme 6), we

also targeted an aminoalcohol with an opposite C13

configuration, which would enable application of more

conventional cyclodehydratization protocols for oxazoline

formation.^[32]The synthesis of 61 started from diyne 53 that had

already been obtained above in an efficient manner involving an

asymmetric Carreira addition (Scheme 7). Best results for the

oxidation were obtained with IBX in DMSO while other methods

(Parikh-Doering-Oxidation, DMP, IBX in EtOAc or activated

MnO₂) gave considerably lower yields (Scheme 9).

Homologation by an HWE-reaction^[33]with phosphonate B gave

unsaturated ester 58 which was reduced with DIBAL to the

corresponding allylic alcohol (not shown). Sharpless epoxidation

conditions (SAE) then gave epoxide 59 in good yields and

diastereoselectivites (dr = 15:1). The absolute configuration of

the epoxide was confirmed by ring opening with DIBAL-H

followed by Mosher ester analysis (see SI section). The primary

hydroxyl was then protected as TBS ether (80%). For

introduction of the amine we chose a regioselective epoxide

opening with ammonia under microwave conditions following a

literature precedent.^[34]

1

2

60

59

30 W / 80°C

40 W / 130 °C

40 W / 120 °C

40 W / 110 °C

8 min

15 min

20 min

10 min

-

3^[b]

4

-

63% (61)

71% (61)

82% (61)

86% (61)

5

6

7

[a] Reactions were conducted in aqueous ammonia, except for entry 3.

[b] Reaction was conducted in a 1:1 mixture of aqueous ammonia and EtOH.

Better results were obtained with the non-protected alcohol 59.

To enable useful degrees of conversion it was necessary to

create a finely dispersed emulsion by fast stirring and ultrasonic

irradiation, giving desired product 61 initially in 63% yield

together with small amounts of an unidentified side product

(entry 4). After optimization of the reaction parameters, by

reducing the reaction time (entry 5) and slightly lowering the

temperature (entries 6 and 7), the desired amino alcohol was

obtained in good yields (86%) and as a single regioisomer.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Based on our experiences during our first route (Scheme 5), we

now wanted to avoid protective groups that would require

strongly acidic conditions for deprotection. Therefore, we chose

fluoride cleavable Teoc and TBS groups for the amine and the

C25 alcohol and targeted 63. As shown in Scheme 10,

introduction of these protective groups proceeded smoothly

under standard conditions. For the secondary alcohol and the

remaining valency of the amine, an acetal was chosen. Notably,

N,O-acetals are known to be unstable without the electron

withdrawing effect of a carbamate, which would enable a joint

deprotection at a later stage of the synthesis. Moreover, it was

anticipated that an acetal group would reduce the flexibility of

the system, thus further impeding a possible coordination of the

carbamate to the zirconium.^{[ 35 ]}Gratifyingly, the zirconocene-

mediated oxidative cyclization of 64 and the subsequent opening

of resulting metallacyclopentadiene 65 with NBS proceeded with

excellent regioselectivity (>20:1) giving the desired bromide 66

in good yields (69%). Finally, the required methyl group of the

target compound 67 could effectively be installed by a Negishi

coupling.^{[36, 37]}In contrast to the unsuccessful deprotection of 40

(Scheme 6), first attempts on the selective deprotection of 67

showed promising results, confirming the suitability of the

protection group strategy (vide infra).

Butyrolactone Building Block

As shown in Scheme 11, three main strategies were evaluated

for synthesis of the unique butyrolactone fragment 13 of the

leupyrrins. The first one would rely on an asymmetric addition of

a suitable enolate or enolate surrogate 68 to an aldehyde of type

69 to forge the C3-C4 bond. Alternatively,

a selective

functionalization of an ester enolate 70, either by alkylations or

selective radical transfer methods was envisioned. Finally, an

adventurous tandem sequence was considered that would

involve a nucleophilic addition of an organometallic reagent 71

to

a chiral aldehyde 72 followed by an intramolecular

esterification of the resulting alkoxide.

Scheme 11. Synthetic strategies for construction of the butyrolactone

fragment of the leupyrrins.

To realize the first approach, a suitably protected aldehyde of

type 69 with a chiral quaternary center was required. As shown

in Scheme 12, its synthesis started from dimethyl malonate 73,

which was hydroxyalkylated and then underwent a pig liver

esterase-(PLE)-catalyzed resolution giving monoacid acid 74 in

91% yield and an ee of 82 % following a known procedure.^[38a,b]

The acid was then converted into the corresponding anhydride,

reduced to the respective alcohol and then selectively oxidized

to the desired aldehyde 75 with IBX.^[38c]In a rationale to avoid

potential disadvantages of the tert-butyl-protecting group due to

its Lewis acid lability we also prepared the corresponding TBS

protected aldehyde 76. Its synthesis involved esterification of 74

with BnOH, Lewis acid mediated cleavage of the tert-butyl group,

protection of the resulting alcohol 77 as TBS ether and liberation

of the acid by hydrogenolysis. Finally, application of the same

procedure as before to acid 78 gave TBS-protected aldehyde 76

in 81% yield.

Scheme 10. Completion of the synthesis of the fully functionalized furan

building block 67.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

However, the substrate influence of the quaternary stereocenter

in the α-position of 75 was only moderate upon addition of vinyl

magnesium bromide as well as alternative vinylating reagents.^[41]

The required allylic alcohol 82 could only be obtained together

with the here undesired C3-epimer 81 in a 3:1 ratio. This ratio

could also not be improved by addition of various Lewis acids

(MgBr₂and BF₃·OEt₂) or remedied by an oxidation-/reduction-

sequence. As shown in Table 3, the intermediate ketone 86

could not be selectively converted to the desired isomer 82.

Table 3. Evaluation of a selective reduction of ketone 86.

entry

reagents

T [°C]

d.r. (81:82)

yield [%]

1

2

3

4

CeCl₃∙7H₂O; NaBH₄

[(R)-CBS]; BH₃∙SMe₂

[(R)-CBS]; catecholborane

0

1:1^[a]

--

n. d.^[b]

--^[c]

-78

RT

Scheme 12. Stereoselective preparation of aldehydes 75 and 76 bearing a

vicinal quaternary chiral center.

--

--^[c]

[a] Determination of selectivity by ¹H-NMR-analysis of the crude product; [b]

n.d.: not determinated; [c] no conversion, quantitative recovery of starting

material.

With aldehydes 75 and 76 in hand, we then studied the key anti-

aldol-reaction. However, all efforts to functionalize 75 or 76 (not

shown) to aldol product 80 either with the Crimmins reagent

79^[39]or alternative procedures (not shown)^[40]were unsuccessful

(Scheme 13). Only decomposition of the starting material,

Finally, in the course of these investigations, we discovered that

an iridium catalyzed allylation developed by Krische^[42]enabled a

very efficient homologation of aldehyde 75. Using catalyst 85 in

the presence of ligand 84, the homoallyalic alcohol 83 was

obtained in high yield (85 %) and excellent diastereoselectivity

(>20:1).

t

presumably by cleavage of the Bu group and/or no conversion

was observed. We then evaluated a substrate controlled

asymmetric vinylation, as the resulting products may be directly

converted to the desired butyrolactone by an one-pot

dihydroxylation-lactonization sequence (see also Scheme 14).

Scheme 14. Conversion of homoallylic alcohol 83 to butyrolactone fragment

90.

Scheme 13. Stereoselective homologation of aldehyde 75.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

As shown in Scheme 14, alcohol 83 was then converted to the

desired butyrolactone fragment 90 in a six-step sequence, which

first involved TES-protection of the free hydroxyl, ozonolysis with

reductive workup and elimination of the resulting alcohol 87 to

the alkene 88 following the protocol by Grieco.^[43]The alkene 88

was then be directly converted to the desired butyrolactone 90 in

high yield by the Sharpless protocol^{[ 44 ]}involving a one-pot

dihydroxylation and regioselective -lactone-formation of the

intermediate diol. Notably, only one diastereomer was obtained

regardless of the chiral additive used. At this stage, the full

stereochemistry of 90 was confirmed by detailed NMR-analysis.

An alternative route via enantiopure epoxide 89 could not be

realized as the allylic alcohol 82 proved too unreactive under the

conditions conventionally employed for a Sharpless epoxidation.

While in principle this route enabled

a

modular and

stereoselective entry into the required butyrolactone fragment of

the leupyrrins and also confirmed the originally proposed relative

configuration of this segment, the overall sequence with 11

steps in the longest linear sequence was considered to be too

lengthy. Therefore, we decided to evaluate alternative strategies

before proceeding with the further elaboration of this building

block.

As shown in Scheme 15, an alternative sequence involved a

diastereoselective hydroxyalkylation of lactone 97, which may be

obtained by two different routes. The first sequence started from

45

]

known D-Xylose (91) derivative 92^[

and involved TBS

protection of the primary alcohol and stereoselective methylation

of 93. Alternatively, the methyl bearing stereogenic center was

first installed by an asymmetric Crimmins syn aldol reaction of

95 with aldehyde 94.^{[ 46 ]}Simultaneous Lewis acid mediated

transesterification of the auxiliary and removal of the acetonide

group of aldol product 96 led to the desired lactone, which was

subsequently TBS protected at the primary hydroxyl. Gratifyingly,

this

key

intermediate

could

then

be

selectively

hydroxymethylated towards the desired isomer 98 by treatment

with LDA and paraformaldehyde.^[11i]Minor amounts of the here

undesired epimer could be readily removed by a simple

recrystallization. Leaving the secondary hydroxyl of 97

unprotected was crucial for the transformation due to the strong

elimination tendency of the substrate towards the corresponding

enoate (see also Scheme 16). At this stage, a suitable protective

group, e.g. a MOM ether could be installed on one or both free

hydroxyls giving either 99 or 101, before the TBS ether could be

selectively cleaved to access either 100 or 102 for further

elaboration (vide infra).

In a parallel fashion, also a more adventurous sequence that is

based on a stereoselective radical reaction was studied. As

shown in Scheme 16, this involves epoxide 106, which was

obtained in five steps from aldehyde 94 involving an asymmetric

aldol reaction with ester 103, subsequent methylation and

elimination of the amine. The resulting enoate 104 then

condensed to the -lactone after removal of the acetonide.

Finally, TBS protection and epoxidation with TBHP under basic

conditions gave 106. A highly chemo-, regio- and stereoselective

conversion of epoxide 106 towards butyrolactone 107 could then

be developed.^[11i]This innovative approach for -chiral

butyrolactone synthesis relies on a 3-step tandem sequence

involving a regioselective epoxide opening with a titanocene(III)-

catalyst obtained in situ from Cp₂TiCl₂and Bu₃SnH, an

intermolecular hydrogen atom transfer with Bu₃SnH and a -

47

]

bond metathesis to liberate again the Ti(III)-catalyst.^[

Unfortunately, the methylation of 107 to 108 could not be

realized, due to the facile elimination towards the corresponding

enoate.

Scheme 15. Concise synthesis of butyrolactone fragments by stepwise

asymmetric alkylations.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

During these studies, it became quickly apparent that this

alcohol was very unreactive. Only with DMP an oxidation to the

required product aldehydes 110 and 111 was observed.

However, the obtained yields (entries 1 and 2) were not

reproducible (42-90% and 25-87%) and this high variance could

also not be remedied by use of freshly prepared reagent or

conducting the reaction at high dilution. The low reactivity of this

specific alcohol became even more pronounced with the MOM-

analogs 100 and 102 which could not be oxidized at all to

aldehydes 112 and 113 (entries 3 and 4).

Accordingly, we continued our route with the TES and TBS

protected aldehydes 110 and 111. As shown in Scheme 17, we

then converted them to ketones 114 and 115. In detail,

treatment with MeMgCl gave a 1:1 diastereomeric mixture of the

corresponding secondary alcohols, which were directly oxidized

to methyl ketones 114 and 115. Along this sequence partial

deprotection of the TES ether 110 was observed, resulting in

only low yields (21%) over these two steps while acceptable

conversions were obtained for the TBS congener (64%). At this

stage, 116 was obtained by deprotection of 115.

Scheme 16. An innovative Ti(III)-catalyzed radical transfer of epoxide 106 to

access -chiral butyrolactone 107 with high selectivity and the attempted

elaboration towards 108.

Nevertheless, at this stage we had established two routes to the

authentic butyrolactone fragment with various protective groups

(see Schemes 14 and 15). We then focused on the further

elaboration of the terminal hydroxyl to the desired building block

13 (Scheme 11). This would require an oxidation of the terminal

primary alcohol to an aldehyde (Table 4). Accordingly, the TES

protected alcohol 90 as well as the TBS analog 109 were

subjected to different oxidation protocols (Swern, Parikh-Doering,

IBX, DMP, Stahl).

Scheme 17. Conversion of the aldehydes 110 and 111 to methyl ketones 114-

116.

Next, installation of the required methylene group was

investigated. This proved again to be very challenging, in

agreement with the very low reactivity of this position, as already

noted above (see Table 4).

Table 4. Oxidation of butyrolactones to the corresponding aldehydes.^[a]

As shown in Table 5, the olefination of ketones 114, 115, and

116 was evaluated with a broad range of reagents. However,

none of the standard methylenation methods, such as Wittig,^[48]

Tebbe,^[49]Peterson^[50]or Takai-Lombardo^[51]olefination resulted

in the formation of respective olefins 117-119 (entries 1-15),

possibly due to steric hindrance or inherent enolate formation

under the basic reaction conditions. Finally, only treatment of

ketone 115 under base-free conditions using the Petasis

reagent^[52]gave rise to 118 in low yields (30%, entry 16).

Also, trapping of the enolate as the corresponding vinyl triflate

which may then be used for a cross-coupling reaction to access

the olefination products, remained unsuccessful. Again, the

ketone proved too unreactive for conversion to the vinyl triflate

under various conditions (Tf₂O, Comins reagent, PhNTf₂).

entry

substrate (R¹; R²)

product (R¹; R²)

yield [%]

1

2

3

4

90 (TES; ^tBu)

109 (TBS; ^tBu)^[b]

100 (H; MOM)

110 (TES; ^tBu)

111 (TBS; ^tBu)

112 (H; MOM)

44-90^[c]

25-87^[c]

--^[d]

102 (MOM; MOM)

113 (MOM; MOM)

--^[d]

[a] conditions: DMP (5.0 eq), NaHCO₃(10 eq), CH₂Cl₂, RT, 3-5 h [b] TBS ether

109 was obtained from the TES ether 90 in a three step sequence, involving

CSA mediated removal of the TES group, TBS protection of both hydroxyls

and selective removal of the primary TBS ether with CSA (3 steps: 81%): see

Experimental details. [c] yield and reisolated starting material varied

undefinably [d] no turnover was observed

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Table 5. Methenylation of butyrolactone fragments.

entry

1-3

substrate

reagent (eq)

NaHMDS

(2.0),

MePPh₃I

(2.0)

solvent

THF

T [°C]

t [h]

16

yield

--^[b]

114

115

116

-78→RT

114

115

116

114

115

116

Tebbe

reagent (2.0)

4-6

7-10

11

THF

0

2

--^[a]

--^[b]

KO^tBu (1.2),

MePPh₃I

(1.2)

n-BuLi (2.0),

MePPh₃I

(2.0)

0→RT

-78

16

115

0.25

Scheme 18. Unsuccessful carbometallation of alkynes 121 and 122.

NaHMDS

(1.2),

MePPh₃Br

(1.2)

TMSCH_2-

MgCl (5.0),

CeCl₃(2.0)

CH₂I₂(5.0),

Zn (9.0),^[c]

TiCl₄(1.0)

Cp₂TiMe₂

(1.0)

12

115

Et₂O

THF

-78→RT

16

--^[a]

As shown in Scheme 19, an adventurous sequential process

was planned to implement this strategy. This approach relied on

lactol 126, which in turn was accessible from commercial

methyl-oxosuccinate 125 in enantiopure form involving an

enzymatic enantioselective reduction.^[55]In detail, a three-step

tandem process was envisioned that would involve a metal or

base-enhanced ring opening of lactol 126 (step 1), the addition

of an organometallic reagent to the resulting aldehyde 127 to

install the required functionality at C4 (step 2) and an

intramolecular lactone formation by addition of the resulting

alkoxide 128 to the ester (step 3). Gratifyingly, this challenging

process could indeed be realized and addition of

isopropenylmagnesium bromide to lactol 126 resulted in clean

formation of the desired butyrolactone 129 with the required

functionality at C4 in high yields (74%) and excellent

diastereoselectivity (> 20:1). At this stage, the configuration at

C2 and C3 was unambiguously assigned by an X-ray structure

analysis of side product 130 originating from a double addition of

the Grignard reagent. During this tandem sequence, the

configuration at C2 had a critical influence on this process, as

the corresponding C2 epimer of 126 did not react under the

same reaction conditions. The stereoselectivity of the aldehyde

addition may be explained by a Felkin-Anh type transition state

or a cyclic Cram model. Finally, building block 129 was

transformed into derivative 131 bearing a free hydroxyl, as

required for the RCM approach as well as to the corresponding

acetate 132 that was used for the macrolactonization pathway

(vide infra).

13

14

115

60

16

2

--^[a]

THF/

CH₂Cl₂

RT

--^[a]

15

16

115

THF

65

16

Cp₂TiMe₂

(1.2)

30%

(118)

PhMe

[a] No reaction: Recovery of the starting material. [b] Decomposition. [c] Zn

was activated with HCl before use.

As an alternative to homologate the butyrolactone fragment by a

methylenation reaction, we also studied the functionalization of

the corresponding alkyne 120. As shown in Scheme 18,

conversion of aldehyde 110 to alkyne 120 with the Ohira-

Bestmann reagent (25) was accompanied with deprotection of

the TES ether. After introduction of either a TES or a TBS group,

the resulting protected analogs 121 and 122 were then

submitted to various carbometalation reaction to access

stannanes of type 124.^[53]However, despite considerable efforts,

no conversion could be observed. Also, a two-step route via the

vinyl iodide 123 failed due to the extremely low reactivity of C5

of the butyrolactone,^[54]in agreement our previous experiences.

Despite our efforts, only 118 was obtained as a useful precursor

for further reactions. Also only small amounts of the substrate

were available due to the lengthy reaction sequence. Therefore,

at this stage we re-evaluated our approach to the butyrolactone

fragment. In order to circumvent the low reactivity at C5 of the

butyrolactone, on an earlier introduction of a functional handle at

this position was planned.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Table 6. Cross metathesis of 132 to boronate 136.^[a]

entry

boron

133

134

135

catalyst^[a]

196

solvent

neat

T [°C]

100

80

T [h]

22

21

19

22

23

24

39

23

39

Yield (E:Z)

n.c.^[b]

1

2

3

4

5

6

7

8

9

196

neat

196

neat

83 % (1.7:1)

84 % (9:1)

47 % (14:1)

64 % (>20:1)

58 % (8:1)

n.c.^[b]

196

toluene

196

197

100

80

197

195

70

198

80

n.c.^[b]

[a] Different catalyst loadings ranging from 0.15 to 0.3 equiv. were applied. For

catalyst structures, see Figure 2. – [b] n.c.: no conversion.

Pyrrole Building Block

In a rational to enable a high degree of flexibility for fragment

coupling, two types of pyrrole carboxylate building blocks were

targeted that either bear a terminal alkene for the proposed

RCM-reaction or a terminal methylbromide for the projected

cross-coupling reactions.

Scheme 19. Concise synthesis of the authentic butyrolactone fragment 132 by

an innovative three-step tandem process, which proceeds with excellent

selectivity.

As shown in Scheme 20, synthesis of the former type was

initiated by an iridium-catalyzed direct borylation of commercial

pyrrole corboxylate 137.^[60]The resulting boronate 138 was then

homologated to the terminal alkene 140. Optimum conditions for

this cross coupling involved treatment of 138 with allylbromide

(139) (1.5 eq) in the presence of Pd(PPh₃)₄(0.1 eq) and K₃PO₄

as base (2 eq) in a dioxane/water mixture. While in principle this

compound may be used for cross and ring closing metatheses

reactions, only low reactivity was anticipated for this substrate.^[61]

Inspired by a report of Robinson,^[56]we also prepared the

corresponding prenyl-analog 142 by cross metathesis with

amylene (141) in the presence of Grubbs-II catalyst. Subsequent

saponifications of 140 and 142 were then effected with NaOH in

a MeOH/water mixture at slightly elevated temperatures (50 °C).

The amine groups of the resulting free acids (43 and 143) may

then be protected as a Boc-amide, as shown for the conversion

of 43 to 46.

With fragment 132 in hand, we then studied the cross

metathesis of 132 to the desired boronate 136. As shown in

Table 6, we initially evaluated the Hoveyda-Grubbs-II catalyst for

the reaction of lactone 132 with different boronates (entry 1-3).

However, only in the case of disubstituted boronate 135

formation of the desired product 136 in good yield (83 %) but

unsatisfying E/Z-ratio was observed (entry 3).^[56]Therefore, we

changed the approach and used toluene as a solvent, which

increased the ratio up to 9:1 (entry 4). Running the reaction at

lower temperature (80 °C) led to an improvement of the ratio to

14:1 but decreased the yield to 47 % (entry 5). Changing the

catalyst from 196 to the new generation catalyst 197^{[ 57]}and

heating to 100 °C led to the desired isomer in acceptable yield

and excellent E/Z-ratio (entry 6). On the other hand, running the

reaction at lower temperatures had no beneficial effect (entry 7).

In order to evaluate the influence of some other catalysts, i.e.

195^[58]and 198^[59]was also investigate, but unfortunately no

conversion was observed (entries 8 and 9).

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Diacid Building Block

For the synthesis of the diacid building block, a stereoselective

reduction was first envisioned as described by Burk and co-

workers as shown in Scheme 22.^{[ 64 ]}After the itaconic acid

derivative 151 was obtained through

a

Stobbe-type

condensation of 150 with isobutyraldehyde, the stereoselective

hydrogenation was realized with a Rhodium(DuPHOS) catalyst

under high pressure, making the desired methyl ester 152

available in only two steps. While esterification of the free acid

moiety then proceeded smoothly, selective methyl ester

saponification of the resulting diesters 153, 154 and 155 was

unsuccessful, possibly due to the high steric demand of the

isobutyl group neighboring the methyl ester.

Scheme 20. Syntheses of pyrrole building blocks with a terminal alkene for

RCM-approaches.

As shown in Scheme 21, synthesis of a pyrrole carboxylate with

a terminal alkylbomide started from methyl carboxylate 144^[62]

,

which was Boc-protected in the presence of DMAP to obtain 146.

Subsequent

side

chain

bromination

gave

terminal

methylbromide 148^[63]in high yields. To maintain a high degree

of flexibility in the protecting group strategy, an acid labile

protecting group in addition to the base labile ethyl ester was

envisioned. Therefore

a

tert-butyl protecting group was

introduced via hydrolysis of 144 and subsequent esterification

with tert-butyl-trichloroacetimidate under Lewis acidic conditions

then gave 145. Following the same sequence involving Boc-

protection and bromination of 147, also the analogous terminal

methylbromide 149 was accessible in high yields.

Scheme 22. First approach to diacid building blocks of leupyrrin A₁(1).

To maintain a high degree of flexibility in the protecting group

strategy, an acid labile protecting group in addition to the base

labile methyl ester 152 was selected. Accordingly, an Evans

auxiliary based approach was chosen (see Scheme 23).^{[ 65 ]}

Acylation of Evans-type auxiliary 159 with 4-methyl valeroyl

chloride

(160)

and

subsequent

auxiliary

controlled

diastereoselective nucleophilic substitution with benzyl- or tert-

butyl 2-bromoacetate (162 and 163) gave the desired esters

164 and 165 in acceptable yields.^[66]

Scheme 21. Syntheses of pyrrole building blocks with a terminal bromide for

cross coupling reactions.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Scheme 24. Synthesis of the diacid building block 177 of leupyrrin B₁.

Fragment Coupling: Evaluation of an RCM Approach

As discussed above in our general retrosynthetic approach,

evaluation of an RCM approach to close the macrocylic ring of

the leupyrrins required Northern fragments 181 or 182. As

shown in Scheme 25, their synthesis was based on an amide

coupling of deprotected analogue 178 of furan segment 67 with

the two pyrrole building blocks 43 and 143. During our

deprotection studies of 67 to access 178 we realized that a

selective removal of the Teoc and the acetonide group in the

presence of the primary TBS group was possible, in agreement

with our protective group design. In detail, the Teoc group was

removed first, rendering the acetonide group unstable which was

then cleaved in situ.^[70]The desired monoprotected derivative

178 may be obtained by treatment with TASF^[71]at carefully

controlled temperatures. In order to guarantee a useful yield

(66%), careful extraction with a polar solvent mixture (ethyl

acetate / iso-propanol = 10:1) had to be applied to ensure

complete transport of the product to the organic phase. We then

focused on the amide formation. While a multitude of mild

methods for this purpose have been described, the applicability

of a number of these methods was impeded by the additional

free hydroxyl group present in 178. Accordingly, we evaluated

the coupling reagent HATU,^[29]which proved compatible with the

free OH-group enabling a selective amide formation with either

43 or the dimethyl analog 143 in good yields (80% for 179 and

79% for 180). This method proved to be superior to a likewise

tested alternative using DEPBT.^{[ 72 ]}While this reaction in

principle enabled an effective access to the amides, the yields

were variable and also a slight excess of the pyrroles were

required. While this was tolerable for coupling of comparably

less precious pyrroles 43 and 143, this deficiency became more

precarious on more elaborate fragments, requiring a further

optimization (see below). Finally, oxazoline formation was best

Scheme 23. Improved syntheses of diacid building blocks of leupyrrin A₁(1).

Regioselective cleavage of the Evans auxiliary in basic H₂O₂

gave free acids 166 and 169 in good to high yields. While the

latter was already suitable for application in the total synthesis,

166 was used in a protection, deprotection sequence, yielding

ester 168 which bears the tert-butyl group on the sterically more

hindered position. Additionally, Evans auxiliary protected ester

167 was envisioned as a possible alternative to 154 due to the

high degree of selectivity that was obtained under H₂O₂

mediated hydrolysis. While 167 was not accessible from 164

through hydrogenation because of the reduction of the Evans

auxiliary prior to the benzyl ester, acidic cleavage of the tert-

butyl ester of 165 gave 167 in very good yields.

As shown in Scheme 24, two different olefination approaches to

diacid building block 177 of leupyrrin B₁(3) were evaluated.

Although poor selectivity was obtained in the HWE-coupling of

phosphonate 172^{[ 67 ]}with aldehyde 174 (test system), efforts

were directed to coupling with 173.^[68]Unfortunately, only traces

of the desired diester 175 were observed. In a different

approach, a Wittig reaction of ylide 170 with ketoacid 171

proceeded smoothly giving the desired diester 175^[69]in high

yield and good selectivity. Careful saponification of 175 with

LiOH then gave the desired monoacid 177 in 87% yield.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

performed using DAST^{[ 73 ]}as a mild condensation reagent,

before the primary hydroxyl was removed with TASF giving 181

and 182 in 66% and 61% yield over these two steps. At this

stage, the relative anti-configuration of the oxazoline ring was

confirmed by NOE studies.

Scheme 26. Synthesis of the Southern fragment 185.

Scheme 25. Synthesis of the Northern fragments 181 and 182.

With the Northern and Southern fragments 181/182 and 185 in

hand, a union of these subunits by an esterification was required

before the macrocyclic ring by a challenging ring closing

metathesis could be evaluated. Initial attempts with the

Yamaguchi reagent resulted in only moderate yields due to

significant amounts of a trichlorobenzoate ester derivative of the

Northern fragment that was likewise formed. In contrast,

esterification of 181 and 182 with acid 185 by using the Shiina

protocol^[76]proceeded smoothly, giving the desired esters 188

and 189 in useful yields (Scheme 27). TASF mediated

deprotection of 189 also gave the corresponding alcohol 190.

As shown in Scheme 26, various strategies were pursued to

access the Southern fragments 185 or 186 bearing either a TBS

or a ^tBu-group at the primary hydroxyl group. The TBS-protected

compound was best prepared by a high yielding esterification of

butyrolactone 131 with acid 168 following the Yamaguchi

procedure^{[ 74 ]}and subsequent deprotection of the ^tBu-group

using a large excess of TMSOTf und 2,6-lutidine.^[75]In contrast,

alternative routes relying on diesters 184 and 187, which were

obtained by coupling of 131 with 167 and 119 with 152 were

thwarted by difficulties for selective cleavage of the sterically

hindered ester adjacent to the isopropyl group.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

However, no formation of the desired products 1 or 191 were

observed, also when a light stream of argon was blown through

the mixture to move the equilibrium towards the desired product

by removal of the side product 2-methyl propene.^[77]The use of

perflourated benzene^[78]or heating to 80 °C likewise had no

beneficial effect (entry 2). Also for substrates 188 and 189, no

conversion was observed with the second generation Hoveyda-

Grubbs-catalyst 196 (entry 7). Neither catalysts 192 and 199

(entries 3 and 10), which had been developed for sterically

hindered systems,^[57]nor fast initiating catalysts 193 and 198^[59]

(entries 4 and 9) lead to product formation. Catalyst 195 (entry

6) which was especially designed for the synthesis of

trisubstituted olefins^{[ 79 ]}and highly active catalysts 194 and

197^[57]were also evaluated, but were likewise unsuccessful in

the transformation.

Scheme 27. Esterification of the Northern and Southern fragments to give

RCM-precursors 188 and 189.

With these RCM precursors in hand we studied the key RCM

reaction to forge the macrocyclic core of leupyrrin A₁(1). As

shown in Table 7, we initially evaluated the Grubbs-II catalyst

(entry 1) in CH₂Cl₂for the reaction of 188-190.

Table 7. Attempted RCM reactions to protected leupyrrins 191 and 192.^[a]

entry

substrate

188-190

188-189

catalyst^[a]

Grubbs II

192

solvent

CH₂Cl₂

C₆F₆

T [°C]

40

T [h]

72

yield

n.c.^[b]

1

2

60-80

0-25

3

C₆F₆

4

193

CH₂Cl₂

C₆F₆

5

194

60-80

0-25

6

195

C₆F₆

7

196

C₆F₆

8

197

C₆F₆

9

198

CH₂Cl₂

C₆F₆

Figure 2. Metathesis catalysts that were evaluated for the RCM reaction.

10

199

60-80

[a] Different catalyst loadings ranging from 0.3 to 1.2 equiv. were applied.

For catalyst structures, see Figure 2. [b] n.c.: no conversion.

At this stage, we decided not to pursue this RCM strategy any

longer and focus on our alternative approach involving a

Western and Eastern fragment instead.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Total synthesis of Leupyrrins A₁and B₁by a Macro-

lactonization

toluene, DMF and ethers were suitable solvents in combination

with carbonates or fluorides as bases. During our evaluation, it

became also apparent that the catalytic system had a high

impact on the reaction and best results were obtained with

As discussed above (Scheme 1), connection of the Western and

Eastern fragment was planned by an adventurous oxazoline

formation of two highly elaborate substrates together with a

macrolactonization. In principle either strategy may be used to

close the macrocycle thus adding considerable flexibility to our

approach.

[83]

[84]

Pd(dppf)Cl₂(entry 5) and the complex Pd(^tBu₃P)₂(entry 3)

giving the desired coupling product with yields over 50%

together

with two side products arising from

a

protodehalogenation and dimerization of the pyrrole. We then

analyzed the choice of solvent, temperature and base in

combination with the catalyst Pd(P^tBu₃)₂in more detail. Optimum

results were obtained at 40 °C in THF or 1,4-dioxane (entries 8,

12), while at higher or lower temperatures (entries 10, 11) or

other solvents (entries 7, 9) lower degrees of conversion were

obtained. Furthermore, increasing the equivalents of the pyrrole

from 1.3 to 2.0 equivalents as well as the catalyst loading from

0.1 to 0.15 also resulted in a slight increase of the yield (entry

13), while replacement of Cs₂CO₃by CsF had an unfavorable

effect (entry 14). In our study, water proved essential, despite

several reports on water free Suzuki-reactions.^[85]Otherwise, the

reaction was very slow and only very low degrees of conversion

could be obtained (entry 15).

As discussed above, synthesis of the Western fragment required

a

challenging sp²-sp³cross coupling between

a suitable

boronate 136 and bromopyrroles 148 or 149. While a number of

Suzuki reactions at the benzylic position of aromatic and

heteroaromatic compounds such as pyridines,^[80]furans,^[81]or

thiophenes^[82]have been described, an example with a pyrrole

appeared not to have been reported before our studies.

Therefore, we decided to first evaluate optimum conditions for

the coupling of pyrrole 148 with simplified model boronate 135

before analyzing the authentic fragment.

Table 8. Optimization of the Suzuki reaction of simplified boronate 135 with

bromopyrrole 148.^[a]

entry

catalyst, base

solvent

T

water

yield

[°C]

60

1

2

Pd(PPh₃)₄, Cs₂CO₃

Pd(OAc₂) / JohnPhos,

Cs₂CO₃

THF

50 eq.

25%^[b]

16%^[b]

3

4

5

6

Pd(P^tBu₃)₂, Cs₂CO₃

Pd(dtbpf)Cl₂, Cs₂CO₃

Pd(dppf)Cl₂, Cs₂CO₃

(XPhos) Pd(II)-

THF

60

50 eq.

66%^[c]

46%^[c]

57%^[b]

34%^[b]

phenetylamin chlorid,

Cs₂CO₃

7

Pd(P^tBu₃)₂, Cs₂CO₃

Pd(P^tBu₃)₂, CsF

DMF

60

RT

90

40

60

50 eq.

-

<25%^[b]

65%^[b]

46%^[b]

56%^[b]

34%^[b]

78%^[b]

83%^[b,d]

47%^[b]

<25%^[b]

8

dioxane

toluene

THF

9

10

11

12

13

14

15

dioxane

THF

Scheme 28. Coupling of authentic fragments gives the desired sp²-sp³

coupled product, together with side products resulting from dimerization and

butyrolactone ring cleavage.

THF

Pd(P^tBu₃)₂, Cs₂CO₃

THF

[a] conditions: 135 (1.0 eq), 148 (1.3 eq), Pd(P^tBu₃)₂(0.1 eq), base (see Table,

1.3 eq), H₂O (see Table), 20 h. [b] Yield was determined by NMR-

measurements of the crude product with an internal standard. [c] Isolated

yields. [d] conditions: 148 (2.0 eq), Pd(P^tBu₃)₂(0.15 eq).

With these optimized conditions in hand we then focused on the

coupling of an authentic butyrolactone. As shown in Scheme 28,

first studies were done with fragment 201, which was available

by cross metathesis from ester 183 and 135. Boronate 201 was

then coupled with an excess of 148 using the reagent

combination developed above, resulting in the formation of

desired product 202. However, two further unwanted side

As shown in Table 8, this model reaction was studied with

different catalysts, solvents, bases and temperatures. In first

studies, we quickly realized that alcohols as solvents as well as

strong bases in the presence of water had to be excluded due to

facile ether formation of the pyrrole bromide. Accordingly,

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

products were obtained which could be assigned by NMR

analyses to be the dimerized butyrolactone 203 and the

aldehyde product 204, where the butyrolactone ring had been

opened.

Formation of these side products may be explained by oxygen

present in the reaction mixture. As shown in Scheme 29,

Amatore, Jutand et al. have discovered that traces of oxygen

may result in formation of a palladium(II)-peroxocomplex of type

A,^[86]which may then react with a boronate. The resulting peroxo

adduct B may then undergo a transmetalation to C, before

substitution with water may give a perboronic acid F together

with a palladium-species D where the residue that was originally

attached to the boronate is now directly linked to the Pd(II)-

catalyst. This may then undergo a Suzuki reaction with the

starting boronate to give the observed dimerized product 203 by

reductive elimination of the transmetallated intermediate E. The

perboronic acid F in turn may then oxidize another molecule of

the original boronate to give a hydroxylated product 205.

According to this mechanism, this hydroxylated and the

homodimerized products are usually formed in a 1:1 ratio.

Scheme 30. Mechanistic proposal for formation of aldehyde 204 involving a

retro-Michael type ring opening of the butyrolactone.

In agreement with this mechanistic explanation, formation of

these side products could be reduced by careful exclusion of

oxygen, involving degasing of all starting materials and solvents

by a „Freeze-Pump-Thaw“-procedure. In contrast, increasing the

temperature to 50 °C or using thallium carbonate as base^[87]did

not improve the outcome of the reaction.

At this stage, all coupling reactions had been successfully

evaluated and in principle, fragment 202 may be further

elaborated into the final product. This would involve

a

condensation with furan 178 by oxazolin formation and

subsequent macrolactonization between C4‘ and C25. However,

at this stage it was anticipated, that this coupling may pose

considerable

difficulties,

since

esterifications had proven difficult (Scheme 27).

Scheme 29. Mechanistic rationale to explain the formation of the

homodimerized product 203 and an oxidation of the boronate to intermediate

205 (see Scheme 30).

As shown in Scheme 30, in our case the alcohol 205 may then

undergo further rearrangement, involving a retro-Michael type

reaction. The resulting ring opened carboxylate 206 would then

be expected to substitute the reactive heterobenzylic bromide

149 to give the ester 204, as experimentally observed.

Scheme 31. Synthesis of the Western building block 208/209.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

In contrast, the corresponding esterifications along the C1’-C3

bond proceeded much more readily. It was also anticipated, that

after oxazoline formation of 202 with 178 the required seco acid

may possibly be difficult to access, as a selective deprotection of

the primary TBS group at C25 and the Bu-ester at C4’ may be

challenging.

A slight improvement of the yield to 71% could be effected by

increasing the amount of base to 1.6 equivalents, as well as

increasing the substrate and water concentrations. These

modifications also led to a decrease of the aldehyde side

product formation to negligible 3%. Joint removal of the ^tBu- and

the Boc-group then proceeded smoothly with TMSOTf/lutidine^[88]

giving the desired Western fragment 208. In addition, the

azabenzotriazole derivative 209 was prepared, which was

important for optimization of the amide coupling reaction (see

below).

As shown in Scheme 32, synthesis of the Eastern fragment 212

was initiated by selective removal of the primary TBS ether of

furan 67 by a procedure that had already been established

above (see Scheme 5), involving catalytic amounts of HCl

prepared in situ from AcCl and MeOH.^[26]Esterification of the

resulting alcohol 210 with acid 169 proceeded in high yields with

the Shiina protocol involving 2-methyl-6-nitrobenzoic acid

anydride (MNBA).^[76]Again this procedure proved more effective

as compared to the likewise tested Yamaguchi method, in

agreement with our experiences above (see Scheme 27). Finally,

TASF^[71]mediated cleavage of the Teoc and acetonide group

proceeded in high yields and with almost full conversion. In

addition to the desired C25-ester 212, small amounts of the

C13-ester as well as the corresponding amide were obtained,

which resulted from acyl migration. Further attempts to remove

the minor amounts of the C13-ester at this stage were

accompanied with further transesterification processes.

Therefore, we decided to continue the sequence with this

mixture and removing the side product at a later stage of the

sequence.

t

Therefore, at this stage we decided that an alternative fragment

coupling that would rely on a macrolactonization along the C1’-

C3 bond may be more promising. Such a strategy, that would

also be more convergent, would rely on Western and Eastern

building blocks 208 and 212 (Schemes 31 and 32). Scheme 31

shows our synthesis of the Western fragment 208 by coupling of

136 and 149. As depicted, we chose an acetate protective group

for the C3 hydroxyl. Based on our previous experiences, it was

expected that an ester at this position may be more readily

cleaved as compared to the ester linkage between C4‘ and C25.

Therefore, it was hoped that a selective saponification would be

possible at a late stage of our route. Also, other protective

groups that are routinely used in complex target synthesis, e.g.

t

silyl-, benzyl-, PMB-, Bu-ether, had to be excluded due to

incompatibility with the other protective groups or with the whole

molecular framework. As protective group for the pyrrolic acid

t

149, a Bu-ester was chosen, as we already knew that this may

be cleaved under very mild conditions (TMS-triflate, 2,6-lutidine).

Furthermore, these conditions would also result in cleavage of

the Boc-group and thus enabling a concomitant removal of these

two groups. As shown in Scheme 31, the Suzuki coupling of 136

with 149 could be effected by our previously established protocol

involving treatment with Pd(P^tBu₃)₂and Cs₂CO₃in the absence

of oxygen and led to product 207.

As shown in Scheme 33, coupling of the Western and Eastern

fragment was initiated by a challenging amide formation in which

a HOAt mediated protocol was likewise applied.^[89]While this

procedure enables a selective coupling of acid 208 with amine

212, the obtained yields were variable and strongly depended on

the purity of the starting material, which in turn could only be

purified with difficulties due to facile and unfavorable

transesterification reaction. Also, a slight excess (1.3 eq) of the

precious Western fragment 208 was required. In contrast,

separate preparation of the HOAt derivative 209, before coupling

with the amine 212 proved more beneficial giving amide 213 in a

reliable fashion in excellent yield (96%). Also equimolar amounts

of the two coupling partners could be used, which further

underlines the importance of this modification. Based on our

previous experiences (see Scheme 25), ring formation

proceeded smoothly with DAST^[73]giving oxazoline 214. At this

stage, also minor amounts of the C13-ester still present could be

readily removed. For conversion to the seco acid 215, we first

t

removed the Bu-ester. Treatment with TMSOTf/lutidine then

gave the acid in essentially quantitative yields and high purity to

allow for directly submitting it to a subsequent selective removal

of the acetate group. After optimization of reaction conditions

with the simplified model compound 132 (not shown), selective

ester cleavage of the C3-acetate in the presence of the C4’-

ester could be affected by treatment with a 0.33 M solution of

potassium carbonate in a 1:1:1 THF/MeOH/H₂O mixture within

two hours.

Scheme 32. Synthesis of the Eastern fragment 212.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Under these conditions, no cleavage of the C4’-ester was

observed and the desired seco acid was obtained from 214 in

84% yield over these two steps.^[90]

Scheme 34. Completion of the total synthesis of leupyrrin A₁.

After having completed the total synthesis of leupyrrin A₁, we

then turned our attention to apply these strategies also to a total

synthesis of leupyrrin B₁. As shown in Scheme 35, this involved

esterification of the identical furan segment 210 with the

unsaturated diacid building block 177, which proceeded in high

yield using the Shiina procedure. Also, the TASF protocol could

be readily implemented to remove the acetonide and Teoc

protective group of the resulting ester 217. However, in contrast

to our leupyrrin A₁route, coupling of the amine 218 with the

pyrrole carboxylate 208 proved challenging and only small

degrees of conversion could be obtained (19%). This situation

could again be remedied by using the azabenzotriazole

derivative 209 resulting in the formation of amide 219 in high

yields. Furthermore, equimolar amounts of the two building

blocks were used which add to the significance of this variant of

the original procedure. After DAST-mediated oxazoline

formation, the seco acid 220 could again be liberated with

TMSOTf and K₂CO₃using the protocols developed for leupyrrin

A₁. Finally, the total synthesis was completed by a high yielding

macrocyclization using the Shiina procedure with high dilution

(92%) and TASF mediated deprotection giving the desired target

natural product leupyrrin B₁(3) in an effective manner. All

spectroscopic data of our synthetic sample were likewise in full

agreement with the originally reported data as well as with an

authentic sample of the natural product, thus likewise confirming

Scheme 33. Oxazoline formation and preparation of seco acid 216.

As shown in Scheme 34, we were pleased that the

marolactonization of the seco acid 215 could indeed be realized

by applying again the Shiina protocol. In detail, a highly dilute

solution of MNBA and DMAP in dichloromethane (1 mM) was

treated very slowly over several hours with a solution of seco

acid 215 in the presence of molecular sieves giving the desired

macrodiolide 216 in high yield (92%). Finally, completion of the

total synthesis of leupyrrin A₁(1) was realized by removing the

TBS-ether at C1 with TASF.^[71]Treatment of 216 in acetonitrile

with two batches of TASF in acetonitrile solutions first at 0 °C

and then at RT gave the target natural product leupyrrin A₁(1) in

76% yield. All spectroscopic data were in full agreement with the

originally reported data as well as with an authentic sample of

the natural product, thus fully confirming the unique structure

and full stereochemistry of this unique macrodiolide.^[4]

the full structure and configuration of this macrodiolide.^[4,Fehler!

Textmarke nicht definiert.]

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

intermediate enabled

a

concise entry into the unique

dihydrofuran fragment with two appending alkyliden subsitutents.

The butyrolacton subunit in turn was prepared by a one-pot

lactol opening and highly stereoselective aldehyde addition by

means of a Grignard reagent and subsequent lactonization of

the resulting alkoxide. Further key steps include an

advantageous sp²-sp³Suzuki-cross coupling for C6/C7 and an

optimized HATU-mediated amide coupling protocol of two

elaborate fragments. It is expected that all these reaction types

will find further applications in related strategies.

Along these syntheses, considerable time and efforts had to be

invested in a few critical steps. The best method to close the

macrocyclic core proved to be a macrolactonization along the

C1’-C3 bond. This ester proved to be more reactive as

compared to the C4’-C25-bond of the diacid fragment. This

reactivity was successfully exploited by using an unusual

acetate protective group at this stage which could be effectively

cleaved in the presence of the C4’-ester. In addition, tert-butyl

esters were widely used as orthogonal, Lewis acid labile

protection groups, allowing mild deprotection conditions that

were crucial for the labile, more elaborate fragments. An efficient

protective group strategy was also established for the

aminoalcohol functionality, which involved a Teoc-acetonide

group enabling a fluoride mediated one-pot cleavage of these

two groups. Finally, establishing an efficient strategy to the

butyrolactone fragment proved very challenging. During this

campaign, also two efficient methods for -chiral butyrolactone

syntheses have been established involving either an

adventurous highly chemo-, regio-, and stereoselective radical

transfer or effective asymmetric alkylations that purely rely on

substrate control.

In general, the high overall yields of the sequences and ready

access to the building blocks renders the reported route

amenable to analogues and enabled an access to sufficient

material to further explore the promising biological potential of

these unique macrodiolides.

Acknowledgements

Scheme 35. Completion of the total synthesis of leupyrrin B₁(3).

Generous financial support by the DFG (ME 2756/7-1 and the

SFB 813) is most gratefully acknowledged. Furthermore, we

thank Andreas J. Schneider (University of Bonn) for excellent

HPLC-support and Dr. Gregor Schnakenburg for X-ray structure

analysis of 130. Also we thank the Chinese Scholarship Council

(stipend to Tongtong Wang).

Conclusions

In summary, efficient strategies for the concise total syntheses

of the leupyrrins A₁and B₁, structurally unique secondary

metabolites from the myxobacterium Sorangium cellulosum

have been designed, elaborated and implemented. The

syntheses of these most prominent representatives of this

biosynthetically highly unusual class of antifungal natural

products have been accomplished in 21 steps and 8.2% for

leupyrrin A₁and 4.0% for leupyrrin B₁. These first total

syntheses also establish unambiguously the relative and

absolute configuration of these potent antifungal agents. Notable

features of their synthesis include two innovative and

adventurous sequential processes. One of them, a zirconocene-

mediated cyclization of a highly elaborate substrate followed by

a regioselective opening of the resulting zirconacyclopentadiene

Keywords: total synthesis • natural products • leupyrrin •

domino reactions • antifungal agent

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

Entry for the Table of Contents

FULL PAPER

The total syntheses of leupyrrins A₁

and B₁were accomplished by an

expedient and high-yielding strategy.

During this campaign, generally useful

protocols were developed, including

innovative one-pot sequences to

access the furan and butyrolactone

moieties, an sp²-sp³-coupling for

pyrrole functionalization and an

optimized HATU-mediated amide

coupling protocol.

Total Synthesis *

Sebastian Thiede, Paul R. Wosniok,

Daniel Herkommer, Thomas Debnar,

Maoqun Tian, Tongtong Wang, Michael

Schrempp, and Dirk Menche*

Page No. – Page No.

Total Synthesis of Leupyrrins A₁and

B₁, Highly Potent Antifungal Agents

from the Myxobacterium Sorangium

cellulosum

* one-pot sequences, sp²-sp³Suzuki-cross coupling

[1] a) G. Höfle, H. Reichenbach, in: Sekundärmetabolismus bei Mikroorganismen (Eds. W. Kuhn, H.-P. Fiedler) Attempto Verlag, Tübingen, 1995, 61-78; b) H.

Reichenbach, G. Höfle, in: Drug Discovery from Nature (Eds. S. Grabley, R. Thierecke) Springer Verlag, Berlin, 1999, 149-179; c) D. Menche, Nat. Prod. Rep.

2008, 25, 905-918; d) S. C. Wenzel, R. Müller, R. Nat. Prod. Rep. 2009, 26, 1385-1407; e) K. J. Weissman, R. Müller, Nat. Prod. Rep. 2010, 27, 1276-1295;

f) T. F. Schäberle, F. Lohr, A. Schmitz, G. M. König, Nat. Prod. Rep. 2014, 31, 953-972.

[2] For examples, see: a) D. M. Bollag, P. A. McQueney, J. Zhu, O. Hensens, L. Koupal, J. Liesch, M. Goetz, E. Lazarides, C. M. Woods, Cancer Res. 1995, 55,

2325-2333; b) I. Nickeleit, S. Zender, F. Sasse, R. Geffers, G. Brandes, I. Sörensen, H. Steinmetz, S. Kubicka, T. Carlomagno, D. Menche, I. Gütgemann, J.

Buer, A. Gossler, M. P. Manns, M. Kalesse, R. Frank, N. P. Malek, Cancer Cell 2008, 14, 23-35; c) L. S. Schneider, K. von Schwarzenberg, T. Lehr, M. Ulrich,

R. Kubisch-Dohmen, J. Liebl, D. Trauner, D. Menche, A. M. Vollmar, Cancer Res. 2015, 75, 2863-2874; d) D. Herkommer, S. Dreisigacker, G. Sergeev, F.

Sasse, H. Gohlke, D. Menche, ChemMedChem 2015, 10, 470-489; e) J. P. Gölz, S. Bockelmann, K. Mayer, H.-J. Steinhoff, H. Wieczorek, M. Huss, J. P.

Klare, D. Menche, ChemMedChem 2016, 11, 420-428.

[3] S. Schneiker, O. Perlova, O. Kaiser, K. Gerth, A. Alici, M. O. Altmeyer, D. Bartels, T. Bekel, S. Beyer, E. Bode, H. B. Bode, C. J. Bolten, J. V. Choudhuri, S.

Doss, Y. A. Elnakady, B. Frank, L. Gaigalat, A. Goesmann, C. Groeger, F. Gross, L. Jelsbak, J. Kalinowski, C. Kegler, T. Knauber, S. Konietzny, M. Kopp, L.

Krause, D. Krug, B. Linke, T. Mahmud, R. Martinez-Arias, A. C. McHardy, M. Merai, F. Meyer, S. Mormann, J. Munoz-Dorado, J. Perez, S. Pradella, S.

Rachid, G. Raddatz, F. Rosenau, C. Ruckert, F. Sasse, M. Scharfe, S. C. Schuster, G. Suen, A. Treuner-Lange, G. J. Velicer, F. J. Vorholter, K. J. Weissman,

R. D. Welch, S. C. Wenzel, D. E. Whitworth, S. Wilhelm, C. Wittmann, H. Blocker, A. Puhler, R. Müller, Nat. Biotechnol. 2007, 25, 1281-1289.

[4] H. B. Bode, H. Irschik, S. C. Wenzel, H. Reichenbach, R. Müller, G. Höfle, J. Nat. Prod. 2003, 66, 1203-1206.

[5] a) H. B. Bode, S. C. Wenzel, H. Irschik, G. Höfle, R. Müller, Angew. Chem. Int. Ed. 2004, 43, 4163-4167; b) M. Kopp, H. Irschik, K. Gemperlein, K. Buntin, P.

Meiser, K. J. Weissman, H. B. Bode, R. Müller, Mol. BioSyst. 2011, 7, 1549-1563.

[6] a) K. J. Weissman, R. Müller, Bioorg. Med. Chem. 2009, 17, 2121-2136; b) C. T. Walsh, Nat. Chem. Biol. 2015, 11, 620-624.

[7] a) H. J. Prins, Chemisch Weekblad 1919, 16, 64, 1072, 1510; b) E. Arundale, L. A. Mikeska, Chem. Rev. 1952, 51, 505–555.

[8] D. Herkommer, S. Thiede, P. R. Wosniok, S. Dreisigacker, M. Tian, T. Debnar, H. Irschik, D. Menche, J. Am. Chem. Soc. 2015, 137, 4086-4089.

[9] J. P. Martinez, B. Hinkelmann, E. Fleta-Soriano, H. Steinmetz, R. Jansen, J. Diez, R. Frank, F. Sasse, A. Meyerhans, Microb. Cell Fact. 2013, 12:85.

[10] S. Thiede, P. R. Wosniok, D. Herkommer, A.-C. Schulz-Fincke, M. Gütschow, D. Menche, Org. Lett. 2016, 18, 3964-3967.

[11] a) W. Liu, H. Jiang, L. Huang, Org. Lett. 2010, 12, 312-315; b) L. Zhang, X. Xu, J. Tan, L. Pan, W. Xia, Q. Liu, Chem. Comm. 2010, 46, 3357-3359; c) T.

Debnar, S. Dreisigacker, D. Menche, Chem. Commun. 2013, 49, 725-727; d) T. Debnar, T. Wang, D. Menche, Org. Lett. 2013, 15, 2774-2777; e) N. Thies, M.

Gerlach, E. Haak, Eur. J. Org. Chem. 2013, 7354-7365; f) D. Herkommer, B. Schmalzbauer, D. Menche, Nat. Prod. Rep. 2014, 31, 456-467; g) G.-Q. Liu, C.-

H. Yang, Y.-M. Li, J. Org. Chem. 2015, 80, 11339-11350; h) P. D. Morse, D. A. Nicewicz, Chem. Sci. 2015, 6, 270-274; i) M. Schrempp, S. Thiede, D.

Herkommer, A. Gansäuer, D. Menche, Chem. Eur. J. 2015, 21, 16266-16271; j) Y.-C. Wong, Z. Ke, Y.-Y. Yeung, Org. Lett. 2015, 17, 4944-4947; k) N.

Rajesh, B. Manisha, J. Ranjith, P. R. Krishna, RSC Advances 2016, 6, 6058-6064.

[12] Part of these results have been previously communicated, see Refs. [7, 9, 10c, 10d, 10i].

[13] For representative references on diyne cyclizations, see: a) E. Negishi, F. E. Cederbaum, T. Takahashi, Tetrahedron Lett. 1986, 27, 2829-2832; b) E. Negishi,

S. J. Holmes, J. M. Tour, J. A. Miller, F. E. Cederbaum, D. R. Swanson, T. Takahashi, J. Am. Chem. Soc. 1989, 111, 3336-3346; c) S. L. Buchwald, R. B.

Nielsen, Chem. Rev. 1988, 88, 1047-1058; d) S. L. Buchwald, R. B. Nielsen, J. Am. Chem. Soc. 1989, 111, 2870-2874; e) H. Ubayama, Z. Xi, T. Takahashi,

Chem. Lett. 1998, 27, 517-518; f) V. H. Gessner, J. F. Tannaci, A. D. Miller, T. D. Tilley, Acc. Chem. Res. 2011, 44, 435-446.

[14] A. E. Rubin, K. B. Sharpless, Angew. Chem. Int. Ed. 1997, 36, 2637-2640.

[15] Separation of the undesired regioisomer 22 was achieved via HPLC after subsequent protection with dimethoxypropane.

[16] S. Müller, B. Liepold, G. J. Roth, H. J. Bestmann, Synlett 1996, 521-522.

[17] E. C. Shuter, H. Duong, C. A. Hutton, M. D. McLeod, Org. Biomol. Chem., 2007, 5, 3183-3189.

[18] J. A. Bodkin, M. D. McLeod, J. Chem. Soc., Perkin Trans. 1 2002, 2733-2746.

[19] a) N. K. Anand, E. M. Carreira, J. Am. Chem. Soc. 2001, 123, 9687-9688; b) R. Takita, K. Yakura, T. Ohshima, M. Shibasaki, J. Am. Chem. Soc. 2005, 127,

13760-13761; c) B. M. Trost, J. D. Sieber, W. Qian, R. Dhawan, Z. T. Ball, Angew. Chem. Int. Ed. 2009, 48, 5478-5481.

[20] E. J. Corey, S. Shibata, R. K. Bakshi, J. Org. Chem. 1988, 53, 2861–2863.

[21] O. Mitsunobu, Y. Yamada, Bull. Chem. Soc. Japan, 1967, 40, 2380–2382.

[22] 39 is obtained as an inseparable 4:1 mixture containing the dehaloginated product.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

[23] In situ analysis (NMR, MS) confirmed the successful deprotection of 39 while no evidence of 41 could be observed in the subsequent reaction.

[24] a) M. Dieckmann, S. Rudolph, S. Dreisigacker, D. Menche, J. Org. Chem. 2012, 77, 10782-10788; b) M. Kretschmer, M. Dieckmann, P. Li, S. Rudolph, D.

Herkommer, J. Troendlin, D. Menche, Chem. Eur. J. 2013, 19, 15993-16018.

[25] In contrast, no product formation was observed with ZnBr₂, AcCl/MeOH, TFA or aqueous HCl.

[26] A. Nudelman, Y. Bechor, E. Falb, B. Fischer, B. A. Wexler, A. Nudelman, Synth. Commun. 1998, 28, 471-474.

[27] M. Riedrich, S. Harkal, H.-D. Arndt, Angew. Chem. Int. Ed. 2007, 46, 2701-2703.

[28] a) E. J. Sorensen, C. A. Guerrero, Org. Lett. 2011, 13, 5164-5167; b) P. B. Alper, S.-C. Hung, C.-H. Wong, Tetrahedron Lett. 1996, 37, 6029-6032.

[29] L. A. Carpino, J. Am. Chem. Soc. 1993, 115, 4397-4398.

[30] A. Sakakura, R. Kondo, K. Ishihara, Org. Lett. 2005, 7, 1971-1974.

[31] a) S. Katukojvala, K. N. Barlett, S. D. Lotesta, L. J. Williams, J. Am. Chem. Soc. 2004, 126, 15348-15349; b) D. E. Frantz, R. Fa, E. M. Carreira, J. Am.

Chem. Soc. 2000, 122, 1806-1807.

[32] a) P. Lafargue, A. Dodi, M. Ponchant, C. Garcia, M. Le Cavorsin, J.-F. Pujol, J.-P. Lellouche, Biorg. Med. Chem. 1994, 8, 827-835; b) A. J. Phillips, Y. Uto, P.

Wipf, M. J. Reno, D. R. Williams, Org. Lett. 2000, 2, 1165-1168; c) P. Wipf, C. P. Miller, Tetrahedron Lett. 1992, 33, 907-910; d) C. Lamberth, J. Prakt. Chem.

2000, 342, 518-522.

[33] W. S. Wadsworth, W. D. Emmons, J. Am. Chem. Soc. 1961, 83, 1733–1738.

[34] a) B. Olofsson, P. Somfai, J. Org. Chem. 2002, 67, 8574-8583; b) U. M. Lindström, B. Olofsson, P. Somfai, Tetrahedron Lett. 1999, 40, 9273-9276.

[35] Acceptable selectivities were obtained, albeit favoring the wrong isomer.

[36] J. Shi, X. Zeng, E. Negishi, Org. Lett., 2003, 5, 1825-1828.

[37] In contrast, deprotonation with ^tBuLi and treatment with Me₂SO₄gave only 18% product and 55% debrominated starting material.

[38] a) J. H. Jones, D. W. Thomas, R. M. Thomas, M. E. Wood, Synth. Commun. 1986, 16, 1607-1610; b) M. Luyten, S. Müller, B. Herzog, R. Keese, Helv. Chim.

Acta 1987, 70, 1250-1254; c) K. Prantz, J. Mulzer, J. Chem. Eur. J. 2010, 16, 485-506.

[39] a) M. T. Crimmins, P. J. McDougall, Org. Lett. 2003, 5, 591-594; b) M. T. Crimmins, B. W. King, E. A. Tabet, J. Am. Chem. Soc. 1997, 119, 7883-7884.

[40] a) S.-i. Kiyooka, K. A. Shahid, Tetrahedron Asym. 2000, 11, 1537-1542; b) S.-i. Kiyooka, K. A. Shahid, F. Goto, M. Okazaki, Y. Shuto, J. Org. Chem. 2003,

68, 7967-7978.

[41] a) S. D. Burke, D. M. Armistead, F. J. Schoenen, J. M. Fevig, Tetrahedron 1986, 42, 2787-2801; b) G. E. Keck, M. B. Andrus, D. R. Romer, J. Org. Chem.

1991, 56, 417-420; c) T. Hirose, T. Sunazuka, S. Tsuchiya, T. Tanaka, Y. Kojima, R. Mori, M. Iwatsuki, S. Ōmura, S. Chem. Eur. J. 2008, 14, 8220-8238.

[42] I. S. Kim, M.-Y. Ngai, M. J. Krische, J. Am. Chem. Soc. 2008, 130, 14891-14899.

[43] P. A. Grieco, S. Gilman, M. Nishizawa, J. Org. Chem. 1976, 41, 1485-1486.

[44] a) H. C. Kolb, M. S. VanNieuwenhze, K. B. Sharpless, K. B. Chem. Rev. 1994, 94, 2483-2547; b) H. Becker, K. B. Sharpless, Angew. Chem. Int. Ed. 1996,

35, 448-451; c) A. B. Zaitsev, H. Adolfsson, Synthesis 2006, 1725-1756; d) Y. Takeda, J. Shi, M. Oikawa, M. Sasaki, Org. Lett. 2008, 10, 1013-1016.

[45] N. C. Chaudhuri, A. Moussa, A. Stewart, J. Wang, R. Storer, Org. Process Res. Dev. 2005, 9, 457-465.

[46] M. T. Crimmins, B. W. King, E. A. Tabet, K. Chaudhary, J. Org. Chem. 2001, 66, 894–902.

[47] A. Gansäuer, M. Pierobon, H. Bluhm, J. Am. Chem. Soc. 1998, 120, 12849-12859.

[48] a) Y. Oikawa, T. Tanaka, K. Horita, I. Noda, N. Nakajima, N. Kakusawa, T. Hamada, O. Yonemitsu, Chem. Pharm. Bull. 1987, 35, 2184-2195; b) K. Tomooka,

T. Ezawa, H. Inoue, K. Uehara, K. Igawa, J. Am. Chem. Soc. 2011, 133, 1754-1756.

[49] a) F. N. Tebbe, G. W. Parshall, G. S. Reddy, J. Am. Chem. Soc. 1978, 100, 3611-3613; b) S. H. Pine, R. J. Pettit, G. D. Geib, S. G. Cruz, C. H. Gallego, T.

Tijerina, R. D. Pine, J. Org. Chem. 1985, 50, 1212-1216.

[50] a) D. Ager, Org. React. 1990, 38, 1-223; b) B. M. Trost, M. T. Rudd, J. Am. Chem. Soc. 2005, 127, 4763-4776.

[51] a) J.-i. Hibino, T. Okazoe, K. Takai, H. Nozaki, Tetrahedron Lett. 1985, 26, 5579-5580; b) Y. Du, X. Lu, J. Org. Chem. 2003, 68, 6463-6465.

[52] a) N. A. Petasis, E. I. Bzowej, J. Am. Chem. Soc. 1990, 112, 6392-6349; b) A. B. Smith, I. G. Safonov, R. M. Corbett, J. Am. Chem. Soc. 2001, 123, 12426-

12427.

[53] a) A. B. Smith, P. R. Verhoest, K. P. Minbiole, M. Schelhaas, J. Am. Chem. Soc. 2001, 123, 4834-4836; b) A. B. Smith, K. P. Minbiole, P. R. Verhoest, M.

Schelhaas, J. Am. Chem. Soc. 2001, 123, 10942-10953; c) J. D. White, P. R. Blakemore, N. J. Green, E. B. Hauser, M. A. Holoboski, L. E. Keown, C. S.

Nylund Kolz, B. W. Phillips, J. Org. Chem. 2002, 67, 7750-7760; d) A. Gopalarathnam, S. G. Nelson, Org. Lett. 2005, 8, 7-10; e) J. D. White, L. Quaranta, G.

Wang, J. Org. Chem. 2007, 72, 1717-1728.

[54] a) C. Spino, C. Godbout, C. Beaulieu, M. Harter, T. M. Mwene-Mbeja, L. Boisvert, J. Am. Chem. Soc. 2004, 126, 13312-13319; b) F. Yang, J. J. Newsome, D.

P. Curran, J. Am. Chem. Soc. 2006, 128, 14200-14205; c) T. Ito, M. Ito, H. Arimoto, H. Takamura, D. Uemura, Tetrahedron Lett. 2007, 48, 5465-5469.

[55] G. Fronza, C. Fuganti, P. Grasselli, L. Malpezzi, A. Mele, J. Org. Chem. 1994, 3487-3489.

[56] Z. J. Wang, W. R. Jackson, A. J. Robinson, Org. Lett. 2013, 15, 3006-3009.

[57] A. Michrowska, R. Bujok, S. Harutyunyan, V. Sashuk, G. Dolgonos, K. Grela, J. Am. Chem. Soc 2004, 126, 9318–9325.

[58] A. Fürstner, L. Ackermann, B. Gabor, R. Goddard, C. W. Lehmann, R. Mynott, F. Stelzer, O. R. Thiel, Chem. Eur. J. 2001, 7, 3236-3253.

[59] T. Ung, A. Hejl, R. H. Grubbs, Y. Schrodi, Organometallics 2004, 23, 5399-5401.

[60] T. Ishiyama, J. Takagi, Y. Yonekawa, J. F. Hartwig, N. Miyaura, Adv. Synth. Catal. 2003, 345, 1103-1106.

[61] A. K. Chatterjee, T.-L. Choi, D. P. Sanders, R. H. Grubbs, J. Am. Chem. Soc. 2003, 125, 11360-11370.

[62] T. P. Curran, M. T. Keaney, J. Org. Chem. 1996, 61, 9068-9069.

[63] also possible: T. Ashkenazi, D. Pinkert, A. Nudelman, A. Widberg, B. Wexler, V. Wittenbach, D. Flint, Pest. Manag. Sci. 2007, 63, 974-1001.

[64] M. J. Burk, F. Bienewald, M. Harris, A. Zanotti-Gerosa, Angew. Chem. Int. Ed. 1998, 1931-1933.

[65] 164 and 169 were prepared on smaller scale according to: M. S. Hoekstra, D. M. Sobieray, M. A. Schwindt, T. A. Mulhern, T. M. Grote, B. K. Huckabee, V. S.

Hendrickson, L. C. Franklin, E. J. Granger, G. L. Karrick, Org. Process Res. Dev. 1997, 1, 26-38.

[66] Diastereomer of 165 was not observed. Diastereomer of 164 could be separated via column chromatography.

[67] Phosphonate 172 was obtained in 2 steps from 4-methylvaleric acid see: a) P.-Y. Géant, J. Martínez, X. J. Salom-Roig, Eur. J. Org. Chem. 2011, 1300-1309;

b) D. K. Donald, M. M. Hann, J. Saunders, H. J. Wadsworth, Thiol Based Collagenase Inhibitors. U.S. Patent 703,973, Februar 21, 1985. (see SI section)

[68] Aldehyde 173 was obtained in 2 steps from tert-butylbromoacetate see: K. Long, T. A. Edwards, A. J. Wilson, Bioorg. Med. Chem 2013, 21, 4034-4040.

[69] The configuration of 12 was assigned by NOE correlations.

[70] Alternatively, also a full deprotection of 67 and reprotection of the primary hydroxyl before amide formation was evaluated. However, the obtained yields were

lower as compared to the sequence described in Scheme 25.

[71] K. Scheidt, H. Chen, B. C. Follows, S. R. Chemler, S. Coffey, W. R. Roush, J. Org. Chem. 1998, 3263, 6436-6437.

[72] H. Li, X. Jiang, Y. Ye, C. Fan, T. Romoff, M. Goodman, Org. Lett. 1999, 1, 91-94.

[73] A. J. Phillips, Y. Uto, P. Wipf, M. J. Reno, D. R. Williams, Org. Lett. 2000, 2, 1165-1168.

[74] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull. Chem. Soc. Jpn. 1979, 52, 1989-1993.

[75] a) A. Trzeciak, W. Bannwarth, Synthesis 1996, 1433-1434; b) D. A. Evans, H. P. Ng, D. L. Rieger, J. Am. Chem. Soc. 1993, 11446-11459.

[76] Shiina, I.; Kubota, M.; Oshiumi, H.; Hashizume, M. J. Org. Chem. 2004, 69, 1822-1830.

[77] H. Waldmann, T.-S. Hu, S. Renner, S. Menninger, R. Tannert, T. Oda, H.-D. Arndt, Angew. Chem. Int. Ed. 2008, 47, 6473-6477.

[78] C. Samojlowicz, M. Bieniek, A. Zarecki, R. Kadyrovb, K. Grela, Chem. Commun. 2008, 6282-6284.

10.1002/chem.201604445

Chemistry - A European Journal

FULL PAPER

[79] A. Fürstner,L. Ackermann, B. Gabor, R. Goddard, C. W. Lehmann, R. Mynott, F. Stelzer, O. R. Thiel, Chem. Eur. J. 2001, 7, 3236-3253.

[80] J. R. Schmink, M. T. Tudge, Tetrahedron Lett. 2013, 54, 15-20.

[81] M. Kuriyama, M. Shinozawa, N. Hamaguchi, S. Matsuo, O. Onomura, J. Org. Chem. 2014, 79, 5921-5928.

[82] M. Shimizu, Y. Tomioka, I. Nagao, T. Hiyama, Synlett 2009, 2009, 3147-3150.

[83] H. Doucet, Eur. J. Org. Chem. 2008, 2013-2030.

[84] G. C. Fu, Acc. Chem. Res. 2008, 41, 1555-1564.

[85] B. Tao, D. W. Boykin, Tetrahedron Lett. 2002, 43, 4955-4957.

[86] C. Adamo, C. Amatore, I. Ciofini, A. Jutand, H. Lakmini, J. Am. Chem. Soc. 2006, 128, 6829-6836.

[87] Y. Hoshino, N. Miyaura, A. Suzuki, Bull. Chem. Soc. Jpn. 1988, 61, 3008-3010.

[88] D. A. Evans, H. P. Ng, D. L. Rieger, J. Am. Chem. Soc. 1993, 11446-11459.

[89] L. A. Carpino, J. Am. Chem. Soc. 1993, 115, 4397-4398.

[90] The solvent had a critical effect on the kinetics of the cleavage of 214. Instantaneous saponification was observed in MeOH, while addition of water

substantially slowed down the reaction. Also, addition of THF led to a decrease of the reaction rate.

Article Doi

Total Synthesis of Leupyrrins A1 and B1, Highly Potent Antifungal Agents from the Myxobacterium Sorangium cellulosum

DOI: 10.1002/chem.201604445

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.201604445

Full text of DOI:10.1002/chem.201604445

Products guided by the article

R&D Labs maybe for 1437250-26-4

Relevant to this article

Hot Product