Synthesis of a Cn-Cn+6 building block common to important polyol,polyene antibiotics from a divinylcarbinol by a desymmetrizing sharpless epoxidation

FULL PAPER

DOI: 10.1002/ejoc.201300181

Synthesis of a Cⁿ–Cⁿ⁺⁶Building Block Common to Important Polyol,Polyene

Antibiotics from a Divinylcarbinol by a Desymmetrizing Sharpless Epoxidation

Luc Nachbauer^[a]and Reinhard Brückner*^[a]

Keywords: Diastereoselectivity / Enantioselectivity / Iodolactonization / Oxidation / Protective group / Wittig

rearrangement

A stereocontrolled synthesis of the enantiomerically pure ep- the bis(tert-butyl alkoxyacetate) 11 through a diastereoselec-

oxide 7b from propargyl ether 15 has been realized in 15

steps. Epoxide 7b represents a building block for the “east-

ern” moieties of the title compounds. Key steps in our ap-

tive [2,3]-Wittig rearrangement (Ǟsyn,syn-9), and a stereo-

and chemoselective iodolactonization (Ǟ35). The CO₂H

groups of dicarboxylic acid 37 were differentiated in a one-

pot bis-oxidation reaction. The latter entailed the novel

transformation of HO₂CCH₂-O-alkyl into AcOCH₂-O-alkyl.

proach were

a desymmetrizing Sharpless epoxidation

(Ǟanti,cis-16), the selective processing of the bis-enolate of

The basis for this expectation may be better than stan-

dard analogies. This is because the antimycotic activity^[11,12]

of the polyol,polyene macrolides may not hinge so much

upon the exact locations of the functional groups but on

shape. For amphotericin B (1), such an assessment stems

from attributing its bioactivity to the formation of potas-

sium ion channels through membranes.^[11]It was suggested

that the walls of these channels are a tubular array of 1:1

complexes wherein the hydrophobic backbone of the anti-

biotic is juxtaposed by a hydrophobic steroid alcohol.^[1e,11b]

The latter would be ergosterol in fungi or cholesterol in

Homo sapiens.^[13]In these complexes the steroids interact

with the membrane, which means that they define the ex-

terior of the channel.^[11]The polyol sections of the antibiot-

ics make up the channel’s interior. Comparing the ion-chan-

nel-forming propensities of amphotericin B (1) with de-

signed analogues thereof^[14]supported this concept and re-

vealed three facts: 1) Pairs of channels interact with one

another longitudinally so that they span the width of the

membrane,^[14a]2) the sugar moiety reinforces the binding of

the sterol by a hydrogen bridge,^{[14b,14d,14e]}and 3) the car-

boxylic acid moiety is not essential for antifungal ac-

tivity.^[14b,14e]Smaller-sized polyol,polyene antibiotics like

pimaricin (4) and rimocidin (5) do not make cell mem-

branes permeable for ions.^[15]In contrast, specific interac-

tions of pimaricin (4) with ergosterol inhibit endocytosis^[16]

as well as vacuole fusion in fungi.^[17]

Introduction

The polyol,polyene macrolides are a family of several

hundred secondary metabolites and are produced by bacte-

rial pathogens of the genus Streptomyces.^[1]Most polyol,

polyene macrolides show antifungal activity.^[1]From a clin-

ical point of view, the best-known polyol,polyene macro-

lides are amphotericin B (1)^[2]and nystatin A₁(2;^[3]

Scheme 1). The potential to keep fungi from food let pimar-

icin (4)^[4,5]become an important food preservative (E235).^[6]

Candidin (3)^[7]is structurally closely related to 1 and 2.

Compounds 1–4 have a tetrahydropyrancarboxylic acid

moiety (“eastern moiety”) in common. This structural mo-

tif is shared by rimocidin (5),^[8]genetically engineered nysta-

tin analogues,^[9]and a number of other polyol,polyene mac-

rolides.^[10]The tetrahydropyrancarboxylic acid moiety of

these compounds should assume, and does so^[2c]in 1, a

chair conformation with four equatorial substituents and

an axially oriented anomeric OH group. Considering this

and the abundance of this substructure, it is tempting to

suggest that the “eastern” moiety co-defines the 3D struc-

ture of said compounds. By extrapolation, one may wonder

whether inserting this “eastern” moiety between unnatural

polyol and polyene sections interconnected by an ester

bond could give rise to modified polyol,polyene macrolides

also acting as antibiotics.

To obtain (sub)structures that may or may not modify

the antifungal activity of the polyol,polyene macrolides, we

synthesized the “eastern” moiety of compounds 1–5 in a

novel fashion. The significant general interest in synthetic

polyol,polyene macrolactone antibiotics^[18–28]was a further

motivation for this work.

[a] Institut für Organische Chemie, Albert-Ludwigs-Universität,

Albertstr. 21, 79104 Freiburg, Germany

E-mail: reinhard.brueckner@organik.chemie.uni-freiburg.de

http://ww w .chemie.uni-freiburg.de/orgbio/brueck/w3br/

Supporting information for this article is available on the

WWW under http://dx.doi.org/10.1002/ejoc.201300181.

Eur. J. Org. Chem. 2013, 6545–6562

6545

L. Nachbauer, R. Brückner

FULL PAPER

Scheme 1. Top and center: Polyol,polyene macrolides 1–5 with a common tetrahydropyrancarboxylic acid, the “eastern moiety”. Bot-

tom: Synthon 6 for the “eastern moiety” and synthetic equivalent 7 thereof.

Our retrosynthetic analysis revealed the dioxane 12 as an

early precursor of building block 7b and the unsaturated

hydroxy ester 9 as the immediate precursor (cf. Scheme 2).

Results and Discussion

Retrosynthetic Analysis

The epoxide ring of 7b should stem from the homoallyl

alcohol moiety of 9. The relative configurations of the ste-

The desired synthetic equivalent 7 of the “eastern” moi-

reocenters in 9^[34]suggest that it is a promising substrate for

ety of the polyol,polyene antibiotics 1–5 should be incor-

a Mihelich epoxidation.^[34a]To achieve 7b, the hydroxy ester

porable into such compounds, for example, into rimocidin

moiety of compound 9 required shortening by an oxidative

cleavage. The extra carbon atom and the configuration of

its C–OH bond were implemented in 9 for two reasons:

(5), or analogues thereof. Accordingly, 7 was equipped

with an epoxide ring at one end (i.e., at C¹², using the

numbering of rimocidin from now on) and a latent OH

1) To control the steric course of epoxide formation and

group at the other (i.e., at C¹⁸; Scheme 2). These func-

2) to be able to obtain 9 by a [2,3]-Wittig rearrangement^[35]

of the bis-enolate 11, which is depicted as a dianion in

Scheme 2. The transformation 11Ǟ9 seemed reasonably

analogous to its antecessor lithio-cis-19aǞsyn,syn-20a,

tional groups should allow combinations with a “north-

ern”, that is, polyol fragment (through nucleophilic attack

upon C¹²) and a “southwestern”, that is, moderately oxy-

genated polyene moiety (through olefination of a C¹⁸alde-

which one of us studied previously (Table 1, entry 1^[36]). Re-

hyde). We prepared a similar building block 7c (Scheme 2)

lated precedents (Table 1, entries 2–4^[36]) suggested that we

best rearrange a tert-butyl ester containing a cis-configured

C=C double bond.

a long time ago,^[29]but there were certain deficiencies both

in its substitution pattern and in our synthetic route. As a

consequence, we have now synthesized two modified “east-

Closing the gap between the desired rearrangement sub-

ern” building blocks 7a^[30]and 7b (Scheme 2). Their sub-

strate dilithio-11 and the already-mentioned dioxane 12 re-

stituents are more adept for further elaboration, there is

quired the following structural changes (Scheme 2): 1) The

more stereocontrol, and the use of hexamethylphosphoric

PMB ether moieties of 12 had to be replaced by (tert-but-

oxycarbonyl)methyl ethers and 2) the underlying 1,3-diol

had to be incorporated into a pentanone ketal in lieu of the

triamide (HMPA) is avoided. Both 7a and 7b can be pre-

pared from the diol syn,cis-14, for which we already devel-

oped a short synthesis^[31,32](bottom part of Scheme 2^[33]).

benzaldehyde acetal, as we discovered later (cf. Table 2).

Transacetalization of syn,cis-14 with benzaldehyde di-

methyl acetal gave the dioxane 12 uneventfully.^[30b]Com-

Model Rearrangements, Epoxidations, and Functional

Group Modifications

pound 12 was then converted into both building block 7a

(retrosynthetic sketch: Scheme 2; synthesis: see ac-

companying paper^[30]) and building block 7b, which is the

Achieving stereocontrol in the Wittig rearrangement (di-

subject of this paper.

lithio-11Ǟ9; Scheme 2), epoxidizing stereoselectively, and

6546

www.eurjoc.org

Eur. J. Org. Chem. 2013, 6545–6562

Cⁿ-Cⁿ⁺⁶Building Block for Polyol,Polyene Antibiotics

Scheme 2. Retrosynthetic analyses of the synthetic equivalents 7a (ref.^[30]) and 7b (present study^[a]) of the “eastern” moiety 6 [rimocidin

(5) numbering adopted] of macrolides 1–5. Key building block 12 is a convergence point and emerged from our desymmetrizing route to

diol syn,cis-14.^[30a,31,32]Reagents and conditions: a) PhCH(OMe)₂(2.0 equiv.), CSA (2.5 mol-%), CH₂Cl₂, room temp., 1 h; 95% [see

ref.^[30b]for the enantiomeric diol, PhCH(OMe)₂, PPTS, DMF, 60 °C, 3 h; 95%]; b) 15 (2.3 equiv.), nBuLi (2.1 equiv.), THF, –78 °C,

60 min; addition of HCO₂Et (1.0 equiv.), –35 °C, 17 h; 89% (ref.^[31]94%); c) Red-Al^®(10.0 equiv.), toluene, –40 °C, 18 h; 46% over the

two steps (ref.^[31]85%); d) Ti(OiPr)₄(1.0 equiv.), d-(–)-DiPT (1.1 equiv.), 4 Å MS, CH₂Cl₂, –25 °C; addition of cis,cis-17, 1 h; addition of

tBuOOH (2.0 equiv.), 70 h; this mixture (anti/syn = 78:22; anti,cis-16 had Ͼ95% ee) was used in the next step without purification [ref.^[31]

71% (based on recovered starting material) of the diastereomeric mixture; anti,cis-16/syn,cis-16 Ͻ82:18; anti,cis-16 had 94–95% ee;^[31]

ref.^[32]69% of the diastereomeric mixture; anti,cis-16:syn,cis-16 Ͻ81:19; anti,cis-16 had 95% ee, syn,cis-16 had 39% ee^[32]]; e) Zn

(20 equiv.), Cu(OAc)₂·H₂O (1 equiv.), H₂O, room temp., 10 min, addition of AgNO₃(1 equiv.), 1 h, filtration, transfer of reductant into

MeOH/H₂O (1:1); addition of 18, 40 °C, 15 h; 76% (ref.^[31]75%, ref.^[32]82%). CSA = camphorsulfonic acid; DiPT = diisopropyl tartrate;

MS = molecular sieves; PG = protecting group; PMB = p-methoxybenzyl; PPTS = pyridinium p-toluenesulfonate; Red-Al^®

NaH₂Al(OCH₂CH₂OMe)₂; TBS = tert-butyldimethylsilyl.^[a]The synthesis of building block 7c is described in ref.^[29]

=

Eur. J. Org. Chem. 2013, 6545–6562

www.eurjoc.org

6547

L. Nachbauer, R. Brückner

FULL PAPER

Table 1. [2,3]-Wittig rearrangements of cis- vs. trans-configured α-(allyloxy)acetates 19a,b under the influence of an oxygenated allylic

stereocenter (*), ester dependency of the chemical yield, and geometry dependency of the asymmetric induction.^[36]

[a] Reagents and conditions: LDA (1.1 equiv.), TMEDA (5.5 equiv.), THF, –78 to –40 °C, 3 h. LDA = lithium diisopropylamide; TMEDA

= N,N,NЈ,NЈ-tetramethylethylenediamine.

Scheme 3. Conversion of l-malic acid into rearrangement substrate models 23a–d. Reagents and conditions: a) BH₃·Me₂S (3.2 equiv.),

B(OMe)₃(3.2 equiv.), THF, 0 °C to room temp., 72 h; 96% (ref.^[38]84% by using this method); b) benzaldehyde dimethyl acetal

(1.1 equiv.), CSA (5 mol-% equiv.), CH₂Cl₂, room temp., 20 h; 89%; c) SO₃·pyridine (5 equiv.), NEt₃(10 equiv.), DMSO, room temp., 2 h;

for minimizing the risk of racemization the product was not purified but used in the next step immediately; d) Ph₃P=CHCO₂Me

(1.5 equiv.), MeOH, room temp., 20 h; 55% cis-isomer over two steps (separated from a 90:10 cis/trans mixture); e) tert-butyl bromoacetate

–

(1.2 equiv.), Bu₄N⁺HSO₄, 50% NaOH/CH₂Cl₂(2:1), room temp., 1 h; 95%; f) DIBAH (2.1 equiv.), THF, 0 °C to room temp., 21 h; 85%;

g) dimethoxymethane/toluene 1:2, CSA (0.6 equiv.), 80 °C, 40 h; 55%; h) 2,2-dimethoxypropane (100 equiv.), CSA (0.4 equiv.), 50 °C, 3 h;

66%; i) 3,3-dimethoxypentane (10 equiv.), propan-3-one (200 equiv.), CSA (0.4 equiv.), 85 °C, 1 h; 78%. CSA = camphorsulfonic acid;

DIBAH = diisobutylaluminum hydride; DMSO = dimethyl sulfoxide.

removing the (tert-butoxycarbonyl)methoxy group, which tained from l-(–)-malic acid (20) as described previously.^[37]

would withstand the rearrangement step, were three chal- Following published procedures, (S)-butanetriol (21; 96%

lenges of our approach to building block 7b. We met the yield) was obtained from 20^[38]and subsequently the hy-

first two challenges in an exploratory study starting from droxy acetal 22 (89% yield).^[39]This compound was oxid-

the dioxanones 23a–d (Scheme 3), which model the diox- ized by using the SO₃·Pyr modification^[40]of the activation

anone 11, with which we deal later.

step of the Swern oxidation^[41]as we found this more conve-

Dioxanone 24 was obtained by a DIBAH reduction of nient for large-scale work. Wittig olefination of the re-

the unsaturated ester 25 (85% yield), which had been ob- sulting aldehyde in methanol^[37]gave the cis-configured es-

6548

www.eurjoc.org

Eur. J. Org. Chem. 2013, 6545–6562

Cⁿ-Cⁿ⁺⁶Building Block for Polyol,Polyene Antibiotics

Table 2. Influence of dioxane substituents on the asymmetric induction in ester enolate [2,3]-Wittig rearrangements.^[a]

[a] Reagents and conditions: a) LDA (1.2 equiv.), TMEDA (5.5 equiv.), THF, –78 °C to –30 °C, 16 h. LDA = Lithium diisopropylamide;

TMEDA = N,N,NЈ,NЈ-tetramethylethylenediamine. [b] Combined yield of syn^3,*,syn^2,3- and anti^3,*,syn^2,3-26. [c] The mol fraction of this

diastereomer in the crude product mixture was determined ¹H NMR spectroscopically. [d] 0.6–0.8 mmol of 23 were used. [e] 1.7–6.0 mmol

of 23 were used.

ter 25 as the major product (55% yield over the two steps).

Conversion into 24 (cf. above) and etherification with tert-

butyl bromoacetate under phase-transfer catalysis condi-

tions^[42]rendered the originally desired [2,3]-Wittig re-

arrangement substrate model, namely (allyloxy)acetate 23a

(95% yield). Shortly we realized that we also needed slightly

differently substituted rearrangement substrates. Accord-

ingly, we treated compound 23a, which is a benzaldehyde

acetal, with large excesses of three different dimethyl acetals

in the presence of 0.4–0.6 equiv. of camphorsulfonic acid.

This induced transacetalizations in yields of 55 (Ǟformal-

Figure 1. ORTEP plot of the crystal structure of syn^3,*,syn^2,3-26a

dehyde acetal 23b), 66 (Ǟacetonide 23c), and 78%

(Ǟpentan-3-one ketal 23d), respectively; no accompanying

transesterification was observed. These transacetalizations

increased our stock of rearrangement-prone (allyloxy)acet-

ates from one to four.

(at 100 K).^[43]

configuration (Figure 2). The same anti^3,*,syn^2,3structure

was attributed to the minor rearrangement products 26a–c

[2,3]-Wittig rearrangements of dioxanones 23a–d were in-

duced under conditions similar to those established for the

rearrangement of the related dioxolanones cis- and trans-

19a,b (Table 1^[36]): Enolate formation with LDA/TMEDA

(ca. 1:4) at –78 °C followed by exposure to –30 °C (rather

than –40 °C) for 16 h (Table 2). The rearrangement prod-

ucts 26a–d were isolated in yields of around 40% when we

worked on a sub-mmol scale, but the yields almost doubled

when mmol quantities of 23 were used. The dioxanone-sub-

stituted esters 23a–d rearranged with diastereoselectivities

of 82:0:18:0 to 94:0:6:0. In contrast, the dioxolanone-substi-

tuted esters cis-19a,b had rearranged with a diastereoselec-

tivity of 100:0:0:0.^[36]Each major rearrangement product

26 was assigned the syn^3,*,syn^2,3-configuration. The major

diastereomer of compound 26a was configured in this man-

ner as determined by X-ray crystal structure analysis^[43]

(Figure 1), whereas the major isomers of the rearrangement

products 26b–d were labeled syn^3,*,syn^2,3due to their plaus-

ible structural similarity to 26a and syn^3,*,syn^2,3-20a,b. The

1

because of their plausible similarity to 26d and H NMR

chemical shift and J_vicvalue analogies.^[44]

Figure 2. ORTEP plot of the crystal structure of anti^3,*,syn^2,3-26d

(at 292 K).^[43]

Continuing our investigation of the model systems be-

consistency of various ¹H NMR chemical shifts and J_vicyond the rearrangement stage was handicapped by our in-

values corroborate the veracity of these assignments.^[44]

ability to free the desired syn^3,*,syn^2,3isomers of com-

The X-ray analysis^[43]of crystals of the minor rearrange- pounds 26a–c from between 6 and 18% rel-% of the respec-

ment product 26d proved that it possesses the anti^3,*,syn^2,3tive anti^3,*,syn^2,3isomer by flash chromatography.^[45]We

Eur. J. Org. Chem. 2013, 6545–6562

www.eurjoc.org

6549

L. Nachbauer, R. Brückner

Table 3. Direct epoxidation of the homoallylic alcohol 26d (CH₂Cl₂, room temp., 66 h for entries 1–3, 24 h for entry 4).

FULL PAPER

Entry Oxidant, additive

Conversion [%]^[b]

50

Reisolated syn,syn-26d [%]

Isolated 27+epi-27 [%]

Ratio 27/epi-27^[b]

1

2

3

4

tBuOOH (1.5 equiv.),

VO(acac)₂(3 mol-%)

tBuOOH (1.5 equiv.),

VO(OEt)₃(5 mol-%)

cumyl-OOH (1.5 equiv.),

VO(OiPr)₃(2 mol-%)

mCPBA (2 equiv.)

42

21^[c]

85:15

ca. 80

35

–

0

29^[c]

–

73:27

82:18

66:34

100

51^[c]

[a] The configuration of the epoxide ring was inferred from the transition-state model of the Mihelich epoxidation.^[34a][b] Determined

1

by H NMR analysis. [c] Yield of the 27/epi-27 mixture after chromatographic purification. mCPBA = meta-chloroperoxybenzoic acid.

overcame this obstacle when we included substrate 23d in

our study. Its enolate rearranged to give a 91:9 mixture of

syn^3,*,syn^2,3-26d and anti^3,*,syn^2,3-26d. Fortunately, syn^3,*,

syn^2,3-26d was readily purified by flash chromatography on

silica gel.^[45]

Our next objective was to convert the vinyl group of

compound syn^3,*,syn^2,3-26d into an epoxide in a diastereo-

selective manner. First, we attempted to do this by epoxid-

ation (Table 3). Mihelich epoxidations with tBuOOH^[34a]or

CumylOOH^[46](entries 1–3) were expected to work well, as

discussed in the retrosynthetic analysis. Indeed, their dia-

stereoselectivities (85:15–73:27) would have been acceptable

except for two aggravating factors: 1) The epoxide dia-

stereomers 27 and epi-27 were inseparable by flash

chromatography on silica gel^[45]and 2) the epoxidation of

syn^3,*,syn^2,3-26d could not be pushed beyond 50% conver-

sion. Replacing MihelichЈs additive VO(acac)₂by VO-

Scheme 4. Unintentional iodolactonization of compound syn,syn-

(OEt)₃(which subsists longer under the reaction condi-

26d at the expense of the attempted formation of an iodo-carbon-

ate. Reagents and conditions: a) nBuLi (1.05 equiv.), THF, –78 °C,

tions^[47]) or by VO(OiPr)₃had no beneficial effect. Increas-

ing the epoxidation time did not help either because the 30 min; dry stream of CO₂, –40 °C, 15 min; addition of I₂

(4.0 equiv.), THF, 0 °C, 16 h, ca. 15% iodolactone 30 (impure) +

epoxide started to decompose. mCPBA proved to be a bet-

48% reisolated syn,syn-26d.

ter oxidant for syn^3,*,syn^2,3-26d, rendering epoxides 27 and

epi-27 in 51% yield (Table 3, entry 4). However, the dia-

stereocontrol was poor (66:34).

Iodolactonization reactions are an indirect way of epoxi-

To avoid these limitations we attempted to involve a dizing unsaturated esters. This is because iodolactones can

neighboring group effect in the epoxide formation. As a be cleaved to give iodohydrins, which cyclize to give epox-

homoallyl alcohol, syn^3,*,syn^2,3-26d might be susceptible to ides.^[48]In the case in hand, this epoxide would hopefully

iodo-carbonate formation.^[48]The latter would be expected be 29. With this possibility in mind, we deliberately in-

to lead to the isomer 28 with three equatorially oriented volved the ester group of substrate syn^3,*,syn^2,3-26d in an

substituents (Scheme 4). By using a protocol for a related iodolactonization (Scheme 5) by using Bartlett’s “condi-

transformation,^[49]we deprotonated the homoallylic alcohol tions for thermodynamic ring-closure control”.^[50]Accord-

syn^3,*,syn^2,3-26d with nBuLi, bubbled dry CO₂through the ingly we treated syn^3,*,syn^2,3-26 in acetonitrile with an ex-

solution, added iodine, and allowed the mixture to react at cess of both iodine and NaHCO₃.^[51]Our substrate reacted

room temp. for 16 h. Disappointingly, we retrieved 48% of to completion and furnished a 75:25 mixture of dia-

the starting material. However, we also isolated about 15% stereomerically pure iodolactone 30 and diastereomerically

of contaminated iodolactone 30. Thus, the ester group of pure iodo ether 31a or 31b.^[52]The cyclization of syn^3,*,-

syn^3,*,syn^2,3-26d had exerted a neighboring group effect syn^2,3-26d became chemoselective when we added 1.0 equiv.

with respect to the iodonium ion intermediate.

of LiI to the reaction mixture. Under these conditions, the

6550

www.eurjoc.org

Eur. J. Org. Chem. 2013, 6545–6562

Cⁿ-Cⁿ⁺⁶Building Block for Polyol,Polyene Antibiotics

Scheme 5. Intentional iodolactonization of compound syn,syn-26d and final steps of our study of a model system towards the C¹²–C¹⁸

fragment analogue 32. Reagents and conditions: a) I₂(3 equiv.), NaHCO₃(5 equiv.), acetonitrile, –40 °C, 20 h; room temp., 30 h; 75:25

mixture of 30 and 31^[52](according to ¹H NMR analysis), which was not purified; b) I₂(6 equiv.), NaHCO₃(5 equiv.), LiI (1 equiv.),

acetonitrile, –20 °C, 16 h; 85% 30; c) LiOH (3.0 equiv.), Ag₂O (0.6-fold molar amount), THF/H₂O (4:1), 0 °C to room temp., 15 min,

then dil. HCl, 0 °C; quant. (without purification); d) Pb(OAc)₄(1.5 equiv.), THF, 0 °C to room temp.; 85% (without purification); e) Na-

ClO₂(4.0 equiv.), NaH₂PO₄(4.5 equiv.), 2-methylbut-2-ene (8.0 equiv.), acetone/H₂O (1:1), 0 °C to room temp., 1 h; f) TMS-diazomethane

(excess), benzene/MeOH (7:2), room temp., 10 min; 27% over the four steps from 30. TMS = trimethylsilyl.

iodo etherification subsided completely and the iodolacton-

Hydrolysis of iodolactone 30 with LiOH in aq. THF fur-

ization proceeded to 30 as diastereoselectively as before nished the epoxide-containing carboxylic acid 29 in quanti-

(85% yield). We have no direct evidence that the newly es- tative yield (Scheme 5). The presence of a stoichiometric

tablished stereocenter of iodolactone 30 possesses the con- amount of Ag^Iions, added as Ag₂O, in the hydrolysis mix-

figuration indicated in Scheme 5, however, there is good an- ture was essential for success; they scavenge the iodide ions

cillary evidence. We iodolactonized the benzaldehyde-pro- expelled during epoxide formation. In the absence of Ag^I,

tected rearrangement product syn^3,*,syn^2,3-26a under iden- the anion of epoxy acid 29 also formed, as revealed by TLC

tical conditions to those used on syn^3,*,syn^2,3-26d, which (the TLC spot of iodolactone 30 was replaced by the iden-

provided a single iodolactone isomer 30a (85% yield). Un- tical product spot whether Ag₂O was present or not). How-

like 30 it was crystalline and thus its 3D structure could be ever, as soon as HCl was added, in preparation of an extrac-

determined by X-ray crystallography^[43](Figure 3).

tive work-up, this anion was protonated to give the epoxy

acid 29; the latter tended to react with the iodide ions and

thereby reverted to the iodolactone 30.

The α-hydroxy acid 29 was cleaved by Pb(OAc)₄

(Scheme 5). The resulting aldehyde 33 was very sensitive

towards acid and base because its epoxide ring was poised

for opening by β-elimination, which probably delivered a

formyl-conjugated allylic alcohol. Accordingly, oxidation to

the corresponding carboxylic acid had to be effected under

neutral conditions. This was achieved by the Lindgren pro-

cedure.^[53]Without purification the resulting carboxylic acid

was methylated with TMS-diazomethane^[54]to furnish ep-

oxide 32.

At this stage we terminated our work on the model sys-

tem. Having optimized a number of crucial steps, in par-

Figure 3. ORTEP plot of the crystal structure of iodolactone 30a

(at 292 K).^[43]

Eur. J. Org. Chem. 2013, 6545–6562

www.eurjoc.org

6551

L. Nachbauer, R. Brückner

FULL PAPER

ticular, the [2,3]-Wittig rearrangement and the iodolactoniz- here onward until we reached epoxy acid 37 we could em-

ation, we felt ready to tackle the “eastern” moiety building

block 7b.

ploy essentially identical reaction conditions to those used

in our model study. Thus, the following steps were per-

formed: Bis(etherification) with tert-butyl bromoacetate

(Ǟ80% 34), transacetalization with 3,3-dimethoxypentane/

pentan-3-one (Ǟ60% 11), bis(ester enolate) formation and

[2,3]-Wittig rearrangement (Ǟ68% syn^3,*,syn^2,3-9 and

17 rel-%^[56]of anti^3,*,syn^2,3-9 as the only other dia-

stereomer; it proved separable by flash chromatography on

Synthesis of the C¹²–C¹⁸Building Block

The PMB ether groups of dioxanone 12 (95.0% ee; for

its preparation, see Scheme 2) were removed with DDQ in

the presence of H₂O^[55]to furnish diol 13 (Scheme 6). From

Scheme 6. Synthesis of the C¹²–C¹⁸building block 7b. Reagents and conditions: a) DDQ (2.5 equiv.), CH₂Cl₂/H₂O (18:1), room temp.,

2 h; 85%; b) tert-butyl bromoacetate (3.0 equiv.), Bu₄N⁺HSO₄(0.7 equiv.), 50% NaOH/CH₂Cl₂(2:1), room temp., 1 h; 80%; c) 3,3-

–

dimethoxypentane (7 equiv.), pentan-3-one (100 equiv.), CSA (0.4 equiv.), 80 °C, 1 h; 60%; d) LDA (2.5 equiv.), TMEDA (10 equiv.),

1

THF, –78 to –30 °C, 15 h; 68% syn^3,*,syn^2,3-9 after separation from 17% of anti^3,*,syn^2,3-9, ds = 85:15 according to H NMR analysis

of the crude product; e) I₂(6.0 equiv.), LiI (1.25 equiv.), NaHCO₃(5.0 equiv.), CH₃CN, –15 °C, 16 h; 71%, ds = 100:0; f) LiOH·H₂O

(6.0 equiv.), Ag₂O (0.6 equiv.), THF/MeOH/H₂O (10:5:3), room temp., 60 min; 95% crude product, which was neither subjected to

chromatography nor characterized by ¹H NMR; g) Pb(OAc)₄(3.0 equiv.), Cu(OAc)₂(5 mol-%), visible light (tungsten lamp, 150 W),

benzene/THF (1:1), 5 °C, 20 min; h) NaBH₄(5.0 equiv.), MeOH, 0 °C to room temp., 2 h; 46% over the three steps from iodolactone 35;

i) TBSOTf (1.5 equiv.), 2,6-lutidine (2.0 equiv.), CH₂Cl₂, –78 °C, 1 h; j) K₂CO₃(1.5 equiv.), MeOH, room temp., 30 min; 67% over two

steps. CSA = camphorsulfonic acid; DDQ = 2,3-dicyano-5,6-dichlorobenzoquinone; TMEDA = N,N,NЈ,NЈ-tetramethylethylenediamine;

TBS = tert-butyldimethylsilyl; TMS = trimethylsilyl.

6552

www.eurjoc.org

Eur. J. Org. Chem. 2013, 6545–6562

Cⁿ-Cⁿ⁺⁶Building Block for Polyol,Polyene Antibiotics

silica gel;^[45]note that one ester moiety remained un-

changed, as desired), diastereo- and chemoselective iodolac-

tonization in the presence of I₂and LiI (Ǟ71% 35), and

LiOH-mediated and Ag₂O-assisted conversion into epoxy

acid 37 (95% as a crude product, which was prone to de-

composition and too polar to be chromatographed rapidly).

An unusual and novel step in Scheme 6 is the oxidative

degradation of the bis(carboxylic acid) 37 by treatment in

benzene/THF^[57](1:1) at 5 °C with a mixture of Pb(OAc)₄

(stoichiometric) and Cu(OAc)₂(catalytic) while irradiating

with a tungsten lamp.^[58]Under these conditions, the

–CH(OH)CO₂H moiety of 37 rendered a –CH=O group,

which represents a routine glycol acid cleavage; concomi-

tantly, the HO₂C–CH₂O– group of 37 was converted into an

AcO–CH₂O– moiety. This suggests that this HO₂C–CH₂O–

group was decarboxylated oxidatively and the resulting

carboxonium ion CH₂=O⁺– scavenged by an acetate ion

from one of the heavy metal salts.^[59]The degradation of

HO₂C–CH₂O– to AcO–CH₂O– in the bis(carboxylic acid)

37 proceeded smoothly and went to completion within

20 min only when the reaction mixture was irradiated.^[60]In

the absence of light the substrate failed to react during as

much as 4 h (at room temp.) or the reaction was sluggish

and led to numerous side-products (2 h at 80 °C). The de-

gradation product 36 is an O,O-acetal and concomitantly a

type of acycal. For this reason we skipped purification by

flash chromatography until after the next step (Ǟ38; see

below).

Scheme 7 presents literature precedents for the trans-

formation 37Ǟ36. To the best of our knowledge, to date,

no α-(alkoxy)acetic acid has been degraded by treatment

with Pb(OAc)₄or with a reagent mixture containing Pb-

(OAc)₄. Only two α-(aryloxy)acetic acids (40 and 41) have

been degraded in this manner delivering formaldehyde O,O-

acetals^[61]akin to 36. More distant analogies to the trans-

formation 37Ǟ36 are the Pb(OAc)₄-mediated degradations

of N-acylglycins like 41^[62]or 42,^[63]which contains the reac-

tive moiety in duplicate. These substrates provided formal-

dehyde N,O-acetals. Some of these oxidation reactions were

performed at an elevated temperature, but the assistance of

light had not been tested.

Scheme 7. Copper-free Pb^IV-mediated degradations of α-(aryloxy)-

acetic acids 40 and 41 and of N-protected α-amino acids 42 and 43

from refs.^{[61a,61b,62,63]}None of these reactions invoked support by

Cu(OAc)₂and/or light, whereas the degradation of our α-(alkyl-

oxy)acetic acid 37 was most effective when both were present

(Scheme 6). TBS = tert-butyldimethylsilyl.

With the stereocenters established and the backbone

shortened as required, the aldehyde 36 was subjected to the

final transformation, starting with a reduction (Scheme 6).

The resulting alcohol 38 was protected as the tert-butyldi-

methylsilyl ether 39. Methanolysis of the (acetoxy)methoxy

moiety of this compound provided the C¹²–C¹⁸building

block 7b (31% overall yield over the five steps from the

iodolactone 35).

The C¹²–C¹⁸building block 7b contains a CH₂OSiR₃

side-chain where the model compound 32 (Scheme 5) bears

a CO₂Me group. This change was necessary to suppress the

acidity of the 14-H atom, which would have made model

32 but not building block 7b incompatible with our final

transformation: epoxide-opening by treatment with the li-

thio derivative of model dithiane 44 (Scheme 8). The latter

mimics the hypothetical “northern fragment” of any of the

Scheme 8. Coupling of the C¹²–C¹⁸building block 7b with the

model 1,3-dithiane 44. Reagents and conditions: a) 44 (4.0 equiv.),

nBuLi (3.5 equiv.), THF, 0 °C, 30 min; HMPA (4.0 equiv.), –78 °C,

10 min; addition of 7b (1.0 equiv.), –40 °C, 16 h; 70% 45 separated

from recovered 44 (78% of the 3.0 equiv., which we used in excess).

polyol/polyene macrolides 1–5, for example, the “northern HMPA = N,N,NЈ,NЈ,NЈЈ,NЈЈ-hexamethylphosphoric triamide.

Eur. J. Org. Chem. 2013, 6545–6562

www.eurjoc.org

6553

L. Nachbauer, R. Brückner

FULL PAPER

Hickl, F. Tönnies, and A. Siegel, Institut für Organische Chemie,

University of Freiburg. IR spectra: Perkin–Elmer Paragon 1000.

Optical rotations were measured with a Perkin–Elmer 341 polari-

meter at 589 nm and 20 °C and calculated by using the Drude

equation: [α]_D= (α_expϫ100)/(cϫd); rotational values are the

average of five measurements of α_expin a given solution of the

corresponding sample. Melting points were measured with a Dr.

Tottoli apparatus (Büchi). The ee values were determined by chiral

HPLC with a Chiralpak AD-H column (0.46ϫ25 cm; Daicel

Chemical Ind. Ltd.) by G. Fehrenbach, Institut für Organische

Chemie, University of Freiburg.

fragment” of rimocidin (5). This dithiane was deprotonated

in THF solution with nBuLi. After the addition of HMPA,

epoxide 7b was added to the reaction mixture. The hydroxy-

alkylated dithiane 45 was isolated in 70% yield and sepa-

rated from 78% of recovered excess dithiane.

Conclusions

A universal Cⁿ–Cⁿ⁺⁶building block 7b for the synthesis

of macrolide antibiotics 1–5 (Scheme 1) has been synthe-

sized from propargyl ether 15. The route comprises 15 steps

in the longest linear sequence. The average yield was 71%

per step and the overall yield 2.0%. The enantiopurity was

created in a desymmetrizing Sharpless epoxidation of divi-

nylcarbinol cis,cis-17 (Ǟepoxy alcohol anti,cis-16; Ͼ95%

ee). Another key step was the stereocontrolled [2,3]-Wittig

rearrangement of the bis(enolate) of diester 11. This reac-

tion exhibited a diastereoselectivity of 85:15:0:0. The epox-

ide ring of target molecule 7b incorporates a stereogenic C–

O bond, which was established by a perfectly diastereoselec-

tive iodolactonization. The latter only completely domi-

nated over an otherwise competing iodoetherification reac-

tion if iodine and LiI were present. This was concluded

from the study of a model system, which gave 31.

[(4R,6S)-6-{(S)-2-(tert-Butyldimethylsiloxy)-1-[(R)-oxiran-2-yl]-

ethyl}-2,2-diethyl-1,3-dioxan-4-yl]methanol (7b):

A tBuO₂CCH₂ether moiety was carried through the

major part of our synthesis, only removing it after cleavage At –78 °C, TBSOTf (40 μL, 181 μmol, 1.5 equiv.) was added drop-

wise to a solution of alcohol 38 (40 mg, 120 μmol, 1.0 equiv.) and

2,6-lutidine (28 μL, 240 μmol, 2.0 equiv.) in CH₂Cl₂(2.5 mL). After

1 h, H₂O (3 mL) was added and the temperature was raised to

room temp. The mixture was extracted with Et₂O (3ϫ3 mL). The

combined organic layers were washed with brine (4 mL), dried with

Na₂SO₄, and filtered.

of the tBuO₂C group by an innovative two-step procedure:

an oxidative degradation (37Ǟ36) and a methanolysis

(39Ǟ7b). The oxidative degradation was effected by treat-

ment with Pb(OAc)₄(stoichiometric), Cu(OAc)₂(substo-

ichiometric), and visible light. The functional group in-

terconversion HO₂CCH₂O-alkylǞAcOCH₂O-alkyl, which

was induced thereby, seems to be unprecedented. It may

warrant further study with respect to developing an orthog-

onal protecting group.

The viability of building block 7b for macrolide antibi-

otic total synthesis was established by combining it in 70%

yield with an umpoled aldehyde, namely the lithiated dithi-

ane 44.

After evaporation of the solvent under reduced pressure, the resi-

due (39) was dissolved in MeOH. Powdered K₂CO₃(25 mg,

180 μmol, 1.5 equiv.) was added. After stirring vigorously for

30 min aqueous phosphate buffer (pH = 7.1, 0.5 m, 10 mL) was

added. The resulting mixture was extracted with AcOEt (3ϫ5 mL).

The combined organic layers were washed with brine (5 mL), dried

with Na₂SO₄, filtered, and concentrated in vacuo. The residue was

purified by flash chromatography^[45](1.5 cm; petroleum ether/Et₂O,

7:3) to yield 7b as a colorless wax (fractions 4–13, 30 mg, 67% over

the two steps from 38). [α]²_D⁰= +11.0 (c = 1.20, CHCl₃, 10 cm). H

1

Experimental Section

NMR (400.1 MHz, C₆D₆): δ = 0.06 and 0.08 [2ϫs, 2 ϫ3 H,

3

†

General: Reactions were performed under N₂in glassware that had

been dried under vacuum at heat-gun temperature. THF was

freshly distilled over potassium prior to use and CH₂Cl₂from

CaH₂. Petroleum ether had a boiling range of 30–50 °C. Products

were purified by flash chromatography^[45]on Merck silica gel 60

(0.040–0.063 mm), yields refer to analytically pure samples. ¹H

NMR [TMS (δ = 0.00 ppm) as an internal standard in CDCl₃;

C₆HD₅(δ = 7.16 ppm) as an internal standard in C₆D₆]: Varian

Mercury VX 300, Bruker Avance 400, and Bruker DRX 500. ¹³C

NMR [TMS (δ = 0.00 ppm) as an internal standard in CDCl₃;

C₆HD₅(δ = 128.06 ppm) as an internal standard in C₆D₆]: Bruker

Avance 400 and Bruker DRX 500. Assignments of ¹H and ¹³C

NMR resonances refer to the IUPAC nomenclature except within

substituents (where primed numbers are used) or when indicated

Si(CH₃)₂], 0.79 (t, J_CH2= 7.5 Hz, 3 H, 2Ј-CH₂-CH₃), 0.96 (dd,

A

B

A

B

‡

³J_CH

=

³J_CH

= 7.5 Hz, 3 H, 2Ј-CH₂-CH₃), 0.97 [s, 9 H,

H

2

SiC(CH₃)₃], 1.04 (ddd, J_5Јeq,5Јax= 12.6, J_5Јeq,4Ј= J_5Јeq,6Ј= 2.5 Hz,

1 H, 5Ј-H^eq), 1.16 (dddd, J_2,3= 8.5, J_2,1B= 6.7, J_2,4Ј= 5.4, J_2,1A

=

2

3.8 Hz, 1 H, 2-H), 1.33 (ddd, J_5Јax,5Јeq= 12.4, J_5Јax,4Ј= J_5Јax,6Ј

11.9 Hz, 1 H, 5Ј-H^ax), AB signal (δ_A= 1.61, δ_B= 1.65, J_AB

13.9 Hz, in addition split by J_A,CH3= J_B,CH3= 7.4 Hz, 2 H, 2Ј-

‡

3

†

CH₂-CH₃), 1.71 (q, J_CH3= 7.5 Hz, 2 H, 2Ј-CH₂-CH₃), 2.38 (dd,

²J_4cis,4trans= 5.4, J_4cis,3= 2.6, 1 H, 4-H^cis), 2.56 (dd, J_4trans,4cis

=

2

5.4, J_4trans,3= 3.9 Hz, 1 H, 4-H^trans), 3.02 (ddd, J_3,2= 8.5, J_3,4trans

= 3.9, J_3,4cis= 2.6 Hz, 1 H, 3-H), AB signal (δ_A= 3.35, δ_B= 3.42,

J_AB= 11.2 Hz, in addition split by J_A,6Ј= 5.8, J_B,6Ј= 3.5 Hz, 2 H,

1ЈЈ-H₂), 3.62 (dddd, J_6Ј,5Јax= 11.7, J_6Ј,1ЈЈA= 5.8, J_6Ј,1ЈЈB= 3.5,

J_6Ј,5Јeq= 2.4 Hz, 1 H, 6Ј-H), AB signal (δ_A= 3.80, δ_B= 3.94, J_AB

explicitly. NMR measurements: Dr. M. Keller, F. Reinbold, M. = 9.9 Hz, in addition split by J_A,2= 3.8, J_B,2= 6.7 Hz, 2 H, 1-

Schonhardt, Institut für Organische Chemie, University of Frei-

burg. MS measurements: Dr. J. Wörth, C. Warth, Institut für Or-

ganische Chemie, University of Freiburg. Combustion analyses: E.

CH₂), 4.14 (ddd, J_4Ј,5Јax= 11.9, J_4Ј,2= 5.4, J_4Ј,5Јeq= 2.5 Hz, 1 H,

4Ј-H)ppm. ¹³C NMR (100.6 MHz, C₆D₆): δ = –5.43 and –5.40

‡

†

[Si(CH₃)₂], 7.19 (2Ј-CH₂-CH₃), 8.25 (2Ј-CH₂-CH₃), 18.42

6554

www.eurjoc.org

Eur. J. Org. Chem. 2013, 6545–6562

Cⁿ-Cⁿ⁺⁶Building Block for Polyol,Polyene Antibiotics

1135, 915, 745 cm^–1. C₂₄H₄₂O₈(458.59): calcd. C 62.86, H 9.23;

found C 63.01, H 9.26.

†

[SiC(CH₃)₃], 22.68 (2Ј-CH₂-CH₃), 26.06 [SiC(CH₃)₃], 30.42 (C-5Ј),

‡

31.51 (2Ј-CH₂-CH₃), 47.66 (C-4), 50.39 (C-3), 50.67 (C-2), 60.61

(C-1), 66.18 and 66.24 (C-1ЈЈ, C-4Ј), 69.43 (C-6Ј), 101.75 (C-

Minor Diastereomer (anti^3,*,syn^2,3-9): [α]²_D⁰= –1.10 (c = 1.09,

CHCl₃, 10 cm). ¹H NMR (400.1 MHz, CDCl₃/TMS): δ = 0.87 (dd,

2Ј) ppm. IR (film): ν = 3455, 2955, 2930, 2880, 2860, 1465, 1380,

˜

1360, 1250, 1165, 1105, 980, 940, 840, 775, 665 cm^–1. C₁₉H₃₈O₅Si

(374.59): calcd. C 60.92, H 10.22; found C 60.71, H 10.47.

³J = 7.5 Hz, 3 H, 2Ј-CH₂-CH₃), 0.91 (dd, ³J = 7.5 Hz, 3 H, 2Ј-

†

‡

CH₂-CH₃), 1.12 (ddd, ²J_5Јax,5Јeq= 12.8, J_5Јax,4Ј= J_5Јax,6Ј= 11.8 Hz,

1 H, 5Ј-H^ax), 1.46 [s, 9 H, C(CH₃)₃], 1.47 (s, 9 H, tert-butyl), 1.53

tert-Butyl (2S,3S)-3-{(4S,6R)-6-[(2-tert-Butoxy-2-oxoethoxy)meth-

yl]-2,2-diethyl-1,3-dioxan-4-yl}-2-hydroxypent-4-enoate (syn^3,*,

syn^2,3-9, Major Diastereomer) and tert-Butyl (2R,3R)-3-{(4S,6R)-6-

[(2-tert-Butoxy-2-oxoethoxy)methyl]-2,2-diethyl-1,3-dioxan-4-yl}-

2-hydroxypent-4-enoate (anti^3,*,syn^2,3-9, Minor Diastereomer):

2

(ddd, J_5Јeq,5Јax= 13.0, J_5Јeq,4Ј= J_5Јeq,6Ј= 2.5 Hz, 1 H, 5Ј-H^äq), AB

signal (δ_A= 1.61, δ_B= 1.66, J_AB= 14.2 Hz, in addition split by

‡

J_A,CH3= J_B,CH3= 7.4 Hz, 2 H, 2Ј-CH₂-CH₃), AB signal (δ_A

=

1.78, δ_B= 1.94, J_AB= 14.9 Hz, in addition split by J_A,CH3= J_B,CH3

†

= 7.3 Hz, 2 H, 2Ј-CH₂-CH₃), 2.40 (ddd, J_3-4= J_3,4Ј= 9.8, J_3,2

2.2 Hz, 1 H, 3-H), 2.90 (d, J_2-OH,2= 3.8 Hz, 1 H, 2-OH), AB signal

(δ_A= 3.50, δ_B= 3.54, J_AB= 10.2 Hz, in addition split by J_A,4Ј

=

4.5, J_B,4Ј= 5.6 Hz, 2 H, 1ЈЈ-H₂), AB signal (δ_A= 4.01, δ_B= 4.06,

J_AB= 16.4 Hz, 2 H, 1ЈЈЈ-H₂), 3.98–4.12 (m, 2 H, 4Ј-H, 6Ј-H), 4.51

(dd, J_2,2–OH= 5.1, J_2,3= 2.2 Hz, 1 H, 2-H), 5.11 (ddd, J_5cis,4= 17.1,

4

²J_5cis,5trans= 2.2, J_5cis,3= 0.6 Hz, 1 H, 5-H^cis), 5.15 (dd, J_5trans,4

=

10.4, J_5trans,5cis= 2.2 Hz, 1 H, 5-H^trans), 5.62 (ddd, J_4,5cis= 17.1,

2

J_4,5trans= J_4,3= 10.2 Hz, 1 H, 4-H) ppm. ¹³C NMR (100.6 MHz,

‡

†

CDCl₃/TMS): δ = 7.13 (2Ј-CH₂-CH₃), 7.89 (2Ј-CH₂-CH₃), 22.16

(2Ј-CH₂-CH₃), 28.13 and 28.17 [2ϫC(CH₃)₃], 31.15 (C-5Ј), 32.13

(2Ј-CH₂-CH₃), 54.12 (C-3), 66.06 (C-6Ј), 68.29 (C-4Ј), 69.10 (C-

†

‡

At –78 °C, the allyl ether derivative 11 (2.29 g, 5.00 mmol, 1 equiv.)

in THF (25 mL) was added during 20 min to a solution prepared

from diisopropylamine (2.25 mL, 16.0 mmol, 3.2 equiv.) and nBuLi

(2.4 m in hexanes, 5.2 mL, 12.5 mmol, 2.5 equiv.) in THF (25 mL).

Tetramethylethylenediamine (7.5 mL, 50.0 mmol, 10 equiv.) was

added 20 min later. Another 20 min later the temperature was al-

lowed to rise to –30 °C. After 15 h the reaction was quenched by

the addition of half-satd. NH₄Cl (50 mL) and the mixture was ex-

tracted with tBuOMe (3ϫ50 mL). The combined organic layers

were washed with H₂O (50 mL) and brine (50 mL), dried with

Na₂SO₄, filtered, and concentrated in vacuo. The crude product

contained an 81:19 mixture of syn^3,*,syn^2,3-9 and anti^3,*,syn^2,3-9

2), 69.40 (C-1ЈЈЈ), 74.94 (C-1ЈЈ), 81.49 and 82.37 [2ϫC(CH₃)₃],

102.15 (C-2Ј), 119.97 (C-5), 132.13 (C-4), 169.76 and 174.18 (C-1,

C-2ЈЈЈ) ppm. IR (film): ν = 3505, 2980, 2940, 2885, 1750, 1725,

˜

1460, 1395, 1370, 1280, 1255, 1225, 1165, 1135, 995, 970, 930, 850,

755 cm^–1. C₂₄H₄₂O₈(458.59): calcd. C 62.86, H 9.23; found C

62.94, H 9.30.

tert-Butyl 2-{(Z)-3-[(4S,6R)-6-{(2-tert-Butoxy-2-oxoethoxy)methyl}-

2,2-diethyl-1,3-dioxan-4-yl]allyloxy}acetate (11):

1

(determined by H NMR signal integration). Purification by flash

chromatography^[45](6 cm; petroleum ether/Et₂O, 3:1) afforded the

major diastereomer syn^3,*,syn^2,3-9 (fractions 14–39, 1.55 g, 68%,

colorless oil) pure and the minor diastereomer anti^3,*,syn^2,3-9 (frac-

tions 9–13, 319 mg, 17%, colorless oil) also pure.

Major Diastereomer (syn^3,*,syn^2,3-9): [α]²_D⁰= –3.7 (c = 0.80, CHCl₃,

10 cm). ¹H NMR (400.1 MHz, CDCl₃/TMS): δ = 0.83 (dd,

At 80 °C, a mixture of benzylidene derivative 34 (1.31 g,

2.73 mmol, 1 equiv.), camphorsulfonic acid (253 mg, 1.09 mmol,

0.4 equiv.), 3-pentanone (30 mL, 80 mmol, 100 equiv.), and 3,3-di-

A

B

A B

³J_CH

=

³J_CH

= 7.4 Hz, 3 H, 2Ј-CH₂-CH₃^†), 0.88 (dd,

H

³J_CH

= 7.4 Hz, 3 H, 2Ј-CH₂-CH₃^‡), 1.41 (ddd, methoxypentane (2 mL, excess) was stirred for 1 h. The brown solu-

A

B

A

B

H

²J_5Јax,5Јeq= J_5Јax,4Ј= J_5Јax,6Ј= 12.0 Hz, 1 H, 5Ј-H^ax), 1.46 [s, 9 H, tion was neutralized by adding half-satd. aqueous NaHCO₃

C(CH₃)₃], 1.48 [s, 9 H, C(CH₃)₃], 1.55–1.65 (2ϫm, 3 H, 2Ј-CH₂-

(50 mL). After extraction with tBuOMe (2ϫ50 mL), the combined

‡

CH₃and 5Ј-H^eq), AB signal (δ_A= 1.83, δ_B= 1.85, J_AB= 9.1 Hz, organic layers were washed with brine (50 mL), dried with Na₂SO₄,

†

in addition split by J_A,CH3= J_B,CH3= 7.4 Hz, 2 H, 2Ј-CH₂-CH₃), filtered, and concentrated in vacuo. The residue was purified by

2.47 (ddd, J_3,4= 9.4, J_3,4Ј= 6.2, J_3,2= 3.0 Hz, 1 H, 3-H), 3.23 (d,

J_2-OH,2= 3.8 Hz, 1 H, 2-OH), AB signal (δ_A= 3.50, δ_B= 3.56, J_AB

= 10.2 Hz, in addition split by J_A,4Ј= 4.6, J_B,4Ј= 5.7 Hz, 2 H, 1ЈЈ-

H₂), AB signal (δ_A= 4.01, δ_B= 4.06, J_AB= 16.4 Hz, 2 H, 1ЈЈЈ-H₂),

flash chromatography^[45](5 cm; cyclohexane/AcOEt, 9:1) to yield

the title compound as a yellowish oil (741 mg, 60%). [α]²_D⁰= +4.23

20

and [α] = +14.5 (in both instances c = 1.12 in CDCl₃, 10 cm).

365

¹H NMR (400.1 MHz, CDCl₃/TMS): δ = 0.85 (t, ³J = 7.5 Hz, 3 H,

3

A

B

3

A

B

†

4.08–4.16 (m, 2 H, 4Ј-H, 6Ј-H), 4.30 (dd, J_2,2–OH= 3.7, J_2,3

=

CH₃), 0.85 (dd, J_CH

= J_CH

= 7.5 Hz, 3 H, 2-CH₂-CH₃),

H

2

4

3

‡

2

2.8 Hz, 1 H, 2-H), 5.13 (ddd, J_5cis,4= 17.2, J_5cis,5trans= 2.1, J_5cis,30.90 (t, J_CH2= 7.4 Hz, 3 H, 2-CH₂-CH₃), 1.33 (ddd, J_5ax,5eq

=

= 0.6 Hz, 1 H, 5-H^cis), 5.19 (dd, J_5trans,4= 10.3, ²J_5trans,5cis= 2.1 Hz, J_5ax,4= J_5ax,6= 12.1 Hz, 1 H, 5-H^ax), 1.47 [s, 9 H, C(CH₃)₃], 1.48

1 H, 5-H^trans), 5.85 (ddd, J_4,5cis= 17.2, J_4,5trans= 10.3, J_4,3= 9.7 Hz, [s, 9 H, C(CH₃)₃], 1.51 (ddd, ²J_5eq,5ax= 13.3, J_5eq,4= J_5eq,6= 2.5 Hz,

‡

1 H, 4-H) ppm. ¹³C NMR (100.6 MHz, CDCl₃): δ = 7.00 (2Ј-CH₂- 1 H, 5-H^eq), 1.61 (q, ³J_CH3= 7.4 Hz, 2 H, 2-CH₂-CH₃), AB signal

‡

†

CH₃), 8.14 (2Ј-CH₂-CH₃), 22.33 (2Ј-CH₂-CH₃), 28.18 and 28.18

[2 ϫ C(CH₃)₃], 30.85 (C-5Ј), 31.13 (2Ј-CH₂-CH₃^‡), 52.55 (C-3),

68.05 and 69.05 (C-4Ј, C-6Ј), 69.39 (C-1ЈЈЈ), 71.66 (C-2), 74.86 (C-

(δ_A= 1.84, δ_B= 1.86, J_AB= 14.8 Hz, in addition split by J_A,CH3

J_B,CH3= 7.4 Hz, 2 H, 2-CH₂-CH₃), AB signal (δ_A= 3.49, δ_B

=

†

3.55, J_AB= 10.1 Hz, in addition split by J_A,6= 4.6, J_B,6= 5.7 Hz,

1ЈЈ), 81.59 and 82.48 [2ϫC(CH₃)₃], 102.07 (C-2Ј), 119.52 (C-5), 2 H, 1ЈЈ-H₂), AB signal (δ_A= 3.92, δ_B= 3.97, J_AB= 15.7 Hz, 2 H,

132.86 (C-4), 169.77 and 172.7 (C-1, C-2ЈЈЈ) ppm. IR (film): ν =

3490, 2975, 2940, 1750, 1730, 1460, 1395, 1370, 1255, 1225, 1160,

2ЈЈЈЈ-H₂), AB signal (δ_A= 4.00, δ_B= 4.06, J^gem= 16.5 Hz, 2 H,

1ЈЈЈ-H₂), 4.12 (dddd, J_6,5ax= 11.9, J_6,1ЈЈB= 5.4, J_6,1ЈЈA= 4.8, J_6,5äq

˜

Eur. J. Org. Chem. 2013, 6545–6562

www.eurjoc.org

6555

L. Nachbauer, R. Brückner

FULL PAPER

= 2.6 Hz, 1 H, 6-H), AB signal (δ_A= 4.18, δ_B= 4.21, J_AB

12.6 Hz, in addition split by J_A,2Ј= 6.4, J_A,3Ј= 1.49, J_B,2Ј= 6.2,

J_B,3Ј= 1.5 Hz, 2 H, 1Ј-H₂), 4.69 (ddd, J_4,5ax= 11.4, J_4,3Ј= 7.4,

=

instances c = 1.02 in CDCl₃, 10 cm); m.p. 80–83 °C. ¹H NMR

2

(400.1 MHz, CDCl₃/TMS): δ = 1.54 [ddd, J_5eq,5ax= 13.3, J_5eq,4

=

2

J_5eq,6= 2.7 Hz, 1 H, 5-H^eq], 1.71 [ddd, J_5ax,5eq= 13.3, J_5ax,4

J_4,5eq= 2.6 Hz, 1 H, 4-H), AB signal (δ_A= 5.56, δ_B= 5.65, J_AB

=

J_5ax,6= 11.4 Hz, 1 H, 5-H^ax], 2.00 and 2.23 (2ϫbr. s, 2ϫ1 H, 1Ј-

OH, 1ЈЈ-OH), AB signal (δ_A= 3.64, δ_B= 3.70, J_AB= 11.8 Hz, in

addition split by J_A,6= 6.3, J_B,6= 3.4 Hz, 2 H, 1ЈЈ-H₂), 4.02 (dddd,

J_6-5ax= 11.4, J_6-1ЈЈA= 6.1, J_6-1ЈЈB= 3.3, J_6-5eq= 2.7 Hz, 1 H, 6-H),

AB signal (δ_A= 4.19, δ_B= 4.29, J_AB= 13.3 Hz, in addition split

4

11.3 Hz, in addition split by J_A,4= 7.3, J_A,1ЈA

=

⁴J_A,1ЈB= 1.1,

J_B,1ЈA= J_B,1ЈB= 6.0 Hz, A: 3Ј-H, B: 2Ј-H) ppm. ¹³C NMR

‡

†

(100.6 MHz, CDCl₃): δ = 6.64 (2-CH₂-CH₃), 8.19 (2-CH₂-CH₃),

†

22.40 (2-CH₂-CH₃), 28.12 and 28.13 [2ϫC(CH₃)₃], 31.14 (2-CH₂-

‡

CH₃), 33.31 (C-5), 64.77 (C-4), 67.01 (C-1Ј), 67.72 and 67.78 (C-

by J_A,4= 6.2, J_B,4= 6.7 Hz, 2 H, 1Ј-H₂), 4.70 (dddd, J_4,5ax= 11.3,

4

6, C-2ЈЈЈЈ), 69.32 (C-1ЈЈЈ), 74.75 (C-1ЈЈ), 81.54 and 81.58 J_4,3Ј= 7.2, J_4,5eq= 2.8, J_4,2Ј= 1.2 Hz, 1 H, 4-H), AB signal (δ_A

=

[2ϫC(CH₃)₃], 101.95 (C-2), 127.60 (C-2Ј), 133.95 (C-3Ј), 169.53 5.61, δ_B= 5.77, J_AB= 11.2 Hz, in addition split by J_A,4= 7.2,

4

and 169.70 (C-2ЈЈЈ, C-1ЈЈЈЈ) ppm. IR (film): ν = 2975, 2940, 2880, ⁴J_A,1ЈA= J_A,1ЈB= 1.4, J_B,1ЈB= 6.7, J_B,1ЈA= 6.2, J_B,4= 1.2 Hz, A:

˜

1750, 1460, 1430, 1395, 1370, 1300, 1230, 1165, 1130, 1040, 965,

845, 700, 655 cm^–1. C₂₄H₄₂O₈(458.59): calcd. C 62.86, H 9.23;

found C 62.88, H 9.15.

3Ј-H, B: 2Ј-H), superimposed by 5.60 (s, 1 H, 2-H), 7.31–7.40 and

7.48–7.56 (2 ϫ m, 5 H, 2 ϫ 2^Ar-H, 2 ϫ 3^Ar-H, 4^Ar-H) ppm. ¹³C

NMR (100.6 MHz, CDCl₃/TMS): δ = 32.61 (C-5), 58.93 (C-1Ј),

65.48 (C-3ЈЈ), 73.14 (C-4), 77.13 (C-6), 100.93 (C-2), 126.31, 128.4

1, and 129.16 (2ϫC-2_Ar, 2ϫC-3_Ar, C-4_Ar), 131.31 (C-1ЈЈ), 131.82

(2R,4R,6S)-4-[(4-Methoxybenzyloxy)methyl]-6-[(Z)-3-(4-meth-

oxybenzyloxy)prop-1-enyl]-2-phenyl-1,3-dioxane (12):^[64]

(C-2ЈЈ), 138.08 (C-1^Ar) ppm. IR (film): ν = 3375, 3035, 2920, 2870,

˜

1650, 1395, 1335, 1310, 1215, 1130, 1105, 1030, 1010, 840, 765,

720 cm^–1. C₁₄H₁₈O₄(250.12): calcd. C 67.18, H 7.25; found C

66.91, H 7.31.

(Z)-(2R,4S)-1,7-Bis(4-methoxybenzyloxy)hept-5-ene-2,4-diol

(syn,cis-14):^[65]

Camphorsulfonic acid (57 mg, 9.30 mmol, 2.5 mol-%) was added

in one portion to a solution of diol syn,cis-14 (3.95 g, 9.81 mmol,

1.0 equiv.) and benzaldehyde dimethyl acetal (2.9 mL, 20 mmol,

2.0 equiv.) in CH₂Cl₂(60 mL). The mixture was stirred for 1 h at

room temp., neutralized by the addition of half satd. aqueous

NaHCO₃(60 mL), and extracted with tBuOMe (2ϫ60 mL). The

combined organic layers were washed with brine (50 mL), dried

with MgSO₄, filtered, and concentrated in vacuo. The residue was

purified by flash chromatography^[45](cyclohexane/AcOEt, 5:1) to

The 78:22 mixture of the epoxides anti,cis- and syn,cis-16 (11.8 g,

29.4 mmol, containing 20.9 mmol of anti,cis-16) was dissolved in

toluene (120 mL) and cooled to –40 °C. A Red-Al^®solution (3.4 m

in toluene, 62 mL, 210 mmol, 10-fold molar amount) was added

dropwise under vigorous stirring. The solution was stirred for 18 h

at the same temperature; the reaction was quenched by pouring it

into aqueous Na/K tartrate solution (half-satd., 250 mL). After 2 h

vigorous stirring, the mixture was extracted with tBuOMe

(4ϫ100 mL). The combined organic layers were washed with brine

(200 mL), dried with Na₂SO₄, filtered, and concentrated in vacuo.

The residue was purified by flash chromatography^[45](8 cm; cyclo-

hexane/AcOEt, 1:1) to yield syn,cis-14 as a colorless oil (6.62 g,

46% from divinylcarbinol cis,cis-17; ee (HPLC): 95%).^{[66] 1}H NMR

1

yield a colorless oil (4.57 g, 95%). H NMR (300.1 MHz, CDCl₃/

TMS): δ = 1.57–1.62 (m, 2 H, 5-H₂), AB signal (δ_A= 3.47, δ_B

=

3.62, J_AB= 10.2 Hz, in addition split by J_A,4= 4.8, J_B,4= 5.9 Hz,

2 H, 1ЈЈ-H₂), 3.79 (s, 1 H, OCH₃), 3.80 (s, 1 H, OCH₃), 4.06 (m_c,

1 H, 4-H), 4.11 (m_c, 1 H, 3Ј-H), AB signal (δ_A= 4.43, δ_B= 4.46,

J_AB= 11.4 Hz, 2 H, benzyl-H₂), AB signal (δ_A= 4.50, δ_B= 4.53,

J_AB= 11.7 Hz, 2 H, benzyl-H₂), 4.61 (ddd, J_6,1ЈЈ≈ J_6,5A≈ J_6,5B

≈

7.0 Hz, 1 H, 6-H), 5.55 (s, 1 H, 2-H), 5.63–5.74 (m, 2 H, 1Ј-H, 2Ј-

H), AAЈBBЈ signal centered at δ = 6.86 and 7.25 ppm, superim-

posed by another AAЈBBЈ signal centered at δ = 6.87 and

7.26 ppm, (8 H, 2ϫC₆H₄), 7.46–7.51 (2ϫm, 5 H, C₆H₅) ppm.

(300.1 MHz, CDCl₃/TMS): AB signal (δ_A= 1.55, δ_B= 1.70, J_AB

=

14.1 Hz, in addition split by J_A,4= 4.4, J_B,2= 9.9, J_A,2= 2.8, J_B,4

= 8.7 Hz, 2 H, 3-H₂), 3.09 and 3.12 (2ϫbr. s, 2 H, 2-OH, 4-OH),

AB signal (δ_A= 3.34, δ_B= 3.40, J_AB= 9.4 Hz, in addition split by

J_A,2= 6.8, J_B,2= 4.2 Hz, 2 H, 1-H₂), 3.79 and 3.80 (2ϫs, 2 ϫ3 H,

2ϫOCH₃), 3.95 (m_c, 1 H, 2-H), 4.00–4.14 (m, 1 H, 7-H₂), AB

signal (δ_A= 4.43, δ_B= 4.46, J_AB= 11.4 Hz, 2 H, benzyl-H₂), 4.47

(s, 2 H, benzyl-H₂), 4.66 (ddd, J_4,5= J_4,3B= 8.1, J_4,3A= 4.5 Hz, 1

H, 4-H), 5.55–5.72 (m, 2 H, 5-H, 6-H), AAЈBBЈ signal centered at

δ = 6.87 and δ = 7.25 (8 H, 2ϫC₆H₄) ppm,

(Z)-3-[(2R,4S,6R)-6-(Hydroxymethyl)-2-phenyl-1,3-dioxan-4-yl]-

prop-2-en-1-ol (13):

cis-(2S,3R,4R)-2,3-Epoxy-1,7-bis(4-methoxybenzyloxy)hept-5-en-

4-ol (anti,cis-16, Major Diastereomer) and

Aqueous phosphate buffer (pH = 7.1, 0.5 m, 10 mL) and DDQ

(5.27 g, 23.3 mmol, 2.5 equiv.) were added to a solution of the PMB

ether derivative 12 (4.57 g, 9.31 mmol, 1.0 equiv.) in CH₂Cl₂

(180 mL). The mixture quickly turned dark green and then yellow

after 2 h. It was filtered and the filtrate was washed with half-satd.

aqueous NaHCO₃(2 ϫ 50 mL). The aqueous phase was re-ex-

tracted with CH₂Cl₂(5ϫ20 mL) and the combined organic layers

were washed with brine (50 mL), dried with Na₂SO₄, filtered, and

concentrated in vacuo. The crude was purified by flash chromatog-

cis-(2R,3S,4R)-2,3-Epoxy-1,7-bis(4-methoxybenzyloxy)hept-5-en-

4-ol (syn,cis-16, Minor Diastereomer):^[69]

raphy^[45](8 cm; cyclohexane/AcOEt/NEt₃, 25:75:0.5) to yield a col-

orless oil (19.8 g, 85%). [α]²_D⁰= +42.1; [α]

= +172.4 (in both

20

365

6556

www.eurjoc.org

Eur. J. Org. Chem. 2013, 6545–6562

Cⁿ-Cⁿ⁺⁶Building Block for Polyol,Polyene Antibiotics

At –25 °C, Ti(OiPr)₄(11.2 mL, 37.9 mmol, 1.05 equiv.) was added

to a suspension of d-(–)-DiPT (7.8 mL, 39.7 mmol, 1.1 equiv.) and

1,7-Bis(4-methoxybenzyloxy)hepta-2,5-diyn-4-ol (18):^[69]

4 Å molecular sieves (16.6 g) in CH₂Cl₂(35 mL). [We found the

[67]

presence of 1.0 equiv. of Ti(OiPr)₄

advantageous during the

Sharpless epoxidation of divinylcarbinol cis,cis-17: the C=C bonds

of this substrate reacted sluggishly, which is not unexpected in view

of its cis configuration.^[68,71]] After 1 h of stirring at room temp., a

solution of diene cis,cis-17 (13.88 g, 36.1 mmol, 1.0 equiv.) in

CH₂Cl₂(500 mL) was added dropwise at –25 °C. After 1 h agita-

tion, a solution of tBuOOH (4.7 m in CH₂Cl₂, 15.4 mL, 72.2 mmol,

2.0 equiv.) was added and the suspension was stirred for 72 h at

–25 °C. The reaction was quenched by cautious addition of a Fe^II

solution (156 g of FeSO₄and 57 g of citric acid in 500 mL of water)

under vigorous agitation. After 30 min, the CH₂Cl₂layer was sepa-

rated and the aqueous suspension was extracted with tBuOMe

(4ϫ100 mL). The combined organic layers were concentrated to

200 mL and treated with aqueous NaOH/NaCl (15 g of NaOH,

2.5 g of NaCl in 45 mL of H₂O). After filtration through a

13ϫ3 cm pad of Celite, the organic layer was separated, washed

with brine (50 mL), dried with Na₂SO₄, filtered, and concentrated

in vacuo. The crude was purified by flash chromatography^[45]

(10 cm; cyclohexane/AcOEt, 2:1) to yield a colorless oil containing

anti,cis- and syn,cis-16 in a 78:22 ratio (11.79 g, 29.4 mmol, con-

taining 20.9 mmol of anti,cis-16). The crude was submitted to re-

duction (Ǟsyn,cis-14) without further purification.

A solution of nBuli (2.5 m in hexanes, 39 mL, 97 mmol, 2.1 equiv.)

was added dropwise to a solution of 15 (18.75 g, 106.4 mmol,

2.3 equiv.) in THF (400 mL) at –78 °C. After stirring for 1 h at this

temperature, ethyl formate (3.73 mL, 46.3 mmol, 1.0 equiv.) was

added and the solution was stirred at –35 °C for 17 h. The reaction

was quenched by the addition of cold water (600 mL) and the mix-

ture was extracted with tBuOMe (4ϫ200 mL). The combined or-

ganic layers were washed with brine (300 mL), dried with MgSO₄,

filtered, and concentrated in vacuo. The residue was purified by

flash chromatography^[45](10 cm; cyclohexane/AcOEt, 3:1) to afford

18 (15.60 g, 89%) as a colorless oil. ¹H NMR (300.1 MHz, CDCl₃/

TMS): δ = 2.32 (d, J_OH,4= 7.2, 4-OH), 3.80 (s, 2ϫOCH₃), 4.19 (d,

4

⁴J_1,4. = J_7,4= 1.8, 1-H₂, 7-H₂), 4.53 (s, 2ϫbenzyl-H₂), 5.23 (d,

J_4,OH= 7.3, 4-H), AAЈBBЈ signal centered at δ = 6.87 and δ = 7.27

(8 H, 2ϫC₆H₄) ppm.

tert-Butyl 2-[(Z)-3-{(2R,4S,6R)-6-[(2-tert-Butoxy-2-oxoethoxy)-

methyl]-2-phenyl-1,3-dioxan-4-yl}allyloxy]acetate (34):

(Z,Z)-1,7-Bis(4-methoxybenzyloxy)hepta-2,5-dien-4-ol (cis,cis-

17):^[69]

At room temp., 50% aqueous NaOH (200 mL) was added to a

solution of diol 13 (2.69 g, 10.7 mmol, 1 equiv.) and tert-butyl bro-

moacetate (4.70 mL, 32.1 mmol, 3.0 equiv.) in CH₂Cl₂(100 mL).

The mixture was vigorously stirred. Bu₄N·HSO₄(2.54 g,

The first step consisted in the preparation of Cu^II- and Ag^I-acti-

vated zinc reactant.^[70]Cu(OAc)₂·H₂O (16.1 g, 80.4 mmol, 7.49 mmol, 0.7 equiv.) was added. After 1 h the mixture was ex-

1.0 equiv.) was added in one portion to a suspension of Zn dust

(105 g, 1.61 mol, 20 equiv.) in water (700 mL) at room temp. After

vigorous stirring at this temperature for 10 min, AgNO₃(13.65 g,

80.4 mmol, 1.0 equiv.) was carefully added over 30 min (careful,

very exothermic reaction occurring!). After stirring for 1 h at room

temp., the suspension was filtered under an inert atmosphere main-

tained by placing over the apparatus an inverted funnel through

which N₂was passed. The zinc cake was successively washed with

water (500 mL), methanol (500 mL), acetone (500 mL), and tBu-

OMe (500 mL). The metal powder was suspended in water/meth-

anol (1:1, 700 mL) and a solution of dialkyne 18 (30.6 g,

tracted with tBuOMe (3ϫ70 mL). The combined organic layers

were washed with half-satd. aqueous NH₄Cl (100 mL) and brine

(100 mL), dried with Na₂SO₄, filtered, and concentrated in vacuo.

Flash chromatography^[45](4 cm; cyclohexane/AcOEt, 5:1) afforded

the title compound as a colorless oil (4.09 g, 80%). [α]²_D⁰= +29.5

(c = 1.03, CHCl₃, 10 cm). H NMR (499.9 MHz, CDCl₃/TMS): δ

= 1.46 and 1.48 [2ϫs, 2 ϫ9 H, 2ϫC(CH₃)₃], 1.61–1.70 (m, 2 H,

5-H₂), AB signal (δ_A= 3.64, δ_B= 3.74, J_AB= 10.4 Hz, in addition

1

split by J_A,4= 4.5, J_B,4= 5.9 Hz, 2 H, 1ЈЈ-H₂), AB signal (δ_A

3.94, δ_B= 3.99, J_AB= 16.2 Hz, 2 H, 1ЈЈЈЈ-H₂), AB signal (δ_A

4.03, δ_B= 4.08, J_AB= 16.4 Hz, 2 H, 1ЈЈЈ-H₂), 4.17 (dddd, J_4,5ax

=

80.4 mmol) in methanol (50 mL) was added at room temp. The 9.9, J_4,1ЈЈB= 5.6, J_4,1ЈЈA= J_4,5eq= 4.3 Hz, 1 H, 4-H), AB signal (δ_A

resulting mixture was vigorously stirred at 40 °C for 15 h. The reac-

tion mixture was cooled to room temp., filtered through a

13ϫ3 cm pad of Celite, and the filter cake was washed with tBu-

OMe (200 mL). The filtrate and washings were washed with half-

satd. aqueous NaHCO₃(300 mL) and brine (300 mL), dried with

MgSO₄, filtered, and the solvents evaporated. The residue was puri-

fied by flash chromatography^[45](10 cm; cyclohexane/AcOEt, 3:2)

= 4.22, δ_B= 4.26, J_AB= 12.7 Hz, in addition split by J_A,2Ј= 3.7,

J_A,3Ј= 1.2, J_B,2Ј= 3.6, J_B,3Ј= 1.1 Hz, 2 H, 1Ј-H₂), 4.71 (dddd, J_6,5ax

= 10.0, J_6,3Ј= 5.0, J_6,5eq= 3.8, J_6,2Ј= 1.3 Hz, 1 H, 6-H), 5.60 (s, 1

H, 2-H), 5.64–5.76 (m, 2 H, 2Ј-H, 3ЈH), 7.28–7.36 and 7.45–7.51

(2 ϫ m, 5 H, 2 ϫ 2^Ar-H, 2 ϫ 3^Ar-H, 4^Ar-H) ppm. ¹³C NMR

(125.7 MHz, CDCl₃/TMS): δ = 28.17 and 28.19 [2 ϫC(CH₃)₃],

33.47 (C-5), 67.09 (C-1Ј), 67.92 (C-2ЈЈЈЈ), 69.39 (C-1ЈЈЈ), 73.24 (C-

6), 74.06 (C-1ЈЈ), 75.96 (C-4), 81.70 [2ϫC(CH₃)₃], 100.74 (C-2),

1

to yield a yellow oil (23.5 g, 76%). H NMR (300.1 MHz, CDCl₃/

TMS): δ = 2.38 (br. s, 1 H, 4-OH), 3.79 (s, 6 H, 2ϫOCH₃), AB 126.32 (2ϫC-2^Ar),* 128.17 (C-2Ј), 128.24 (2ϫC-3^Ar),* 128.83 (C-

signal (δ_A= 4.01, δ_B= 4.08, J_AB= 12.4 Hz, in addition split by

J_A,vic= 4.9 Hz, J_B,vic= 5.0 Hz, 4 H, 1-H₂, 7-H₂), 4.42 (s, 2 H, 1ЈЈЈЈ)* ppm; *: assignment corroborated by HMBC experiment. IR

4^Ar),* 132.90 (C-3Ј), 138.32 (C-1^Ar),* 169.57 (C-2ЈЈЈ),* 169.68 (C-

2ϫbenzyl-H₂) 5.14 (t, J_4,3= J_4,5= 6.9 Hz, 1 H, 4-H), 5.59–5.73

(m, 4 H, 2-H, 3-H, 5-H, and 6-H), AAЈBBЈ signal centered at δ =

6.87 and δ = 7.25 (8 H, 2ϫC₆H₄) ppm.

(film): ν = 2980, 1745, 1455, 1390, 1365, 1305, 1225, 1125, 1020,

˜

955, 845, 755, 700, 570 cm^–1. C₂₆H₃₈O₈(478.58): C 65.25, H 8.00;

found C 64.95, H 8.00.

Eur. J. Org. Chem. 2013, 6545–6562

www.eurjoc.org

6557

L. Nachbauer, R. Brückner

FULL PAPER

tert-Butyl 4-{(3S,4S,5R)-6-[(4S,6R)-(2-tert-Butoxy-2-oxoethoxy)-

methyl]-2,2-diethyl-1,3-dioxan-4-yl}-3-hydroxy-5-(iodomethyl)-4,5-

dihydrofuran-2(3H)-one (35):

At –25 °C, I₂(5.15 g, 20.3 mmol, 6.0 equiv.) was added in one por-

tion to a suspension of homoallyl alcohol syn^3,*,syn^2,3-9 (1.55 g,

3.38 mmol, 1.0 equiv.), NaHCO₃(1.42 g, 16.9 mmol, 5.0 equiv.),

and LiI (565 mg, 4.22 mmol, 1.25 equiv.) in CH₃CN (85 mL). The

temperature was allowed to rise to –15 °C. The mixture was stirred

for 16 h. The reaction was quenched by adding satd. aqueous

Na₂S₂O₃until complete discoloration occurred. The mixture was

extracted with tBuOMe (3ϫ25 mL). The combined organic layers

were washed with H₂O (50 mL) and with brine (50 mL), dried with

Na₂SO₄, filtered, and concentrated in vacuo. Purification by flash

chromatography^[45](5 cm; petroleum ether/Et₂O, 1:1) afforded 35

as a yellowish oil (1.26 g, 71%). [α]²_D⁰= +4.7 (c = 1.10, CHCl₃,

10 cm). ¹H NMR (400.1 MHz, CDCl₃/TMS): δ = 0.83 (dd,

At 5 °C and exposed to a spot-light (visible, 150 W), a solution of

the compound 37 (63 mg, 0.189 mmol, 1 equiv.) in THF (2 mL)

was added dropwise during 5 min to a solution of Pb(OAc)₄(80%

in AcOH, 314 mg, 0.567 mmol, 3 equiv.) and Cu(OAc)₂(2 mg,

9.5 μmol, 5 mol-%) in benzene (2 mL). After stirring/irradiating for

another 15 min the reaction was quenched by adding ethylene

glycol (0.1 mL) and phosphate buffer (pH = 7.1, 0.5 m, 4 mL). The

resulting mixture was extracted with Et₂O (4ϫ4 mL). The com-

bined organic layers were washed with brine (5 mL), dried with

Na₂SO₄, filtered, and concentrated in vacuo. We obtained 36 as a

colorless oil (53 mg, 85% from 35 over the two steps) but could

not purify it due to its incompatibility with silica gel. ¹H NMR

3

A

B

A

B

(300.1 MHz, CDCl₃/TMS): δ = 0.83 (dd, J_CH

=

³J_CH

=

H

A

B

A

B

†

3

³J_CH

=

³J_CH

= 7.5 Hz, 3 H, 2Ј-CH₂-CH₃), 0.85 (t, J_CH2

†

3

A

B

H

7.5 Hz, 3 H, 2Ј-CH₂-CH₃), superimposed by 0.87 (dd, J_CH

H

= 7.4 Hz, 3 H, 2Ј-CH₂-CH₃^‡), 1.45 (ddd, J_5Јax,5Јeq= J_5Јax,4Ј

J_5Јax,6Ј= 12.1 Hz, 1 H, 5Ј-H^ax), 1.48 [s, 9 H, C(CH₃)₃], 1.57 (q,

=

2

A

B

‡

2

³J_CH

= 7.4 Hz, 3 H, 2Ј-CH₂-CH₃), 1.35–1.55 (m, H, 5Ј-H₂),

H

‡

superimposed by 1.51–1.67 (m, 2 H, 2Ј-CH₂-CH₃), 1.78–1.09 (m,

‡

2

³J_CH3= 7.5 Hz, 2 H, 2Ј-CH₂-CH₃), 1.58 (ddd, J_5Јeq,5Јax= 12.6,

†

2 H, 2Ј-CH₂-CH₃), 2.10 (s, 2ЈЈЈЈ-H₃), 2.37 (br. s, 1 H, 1-OH), 2.12

J_5Јeq,4Ј= J_5Јeq,6Ј= 2.7 Hz, 1 H, 5Ј-H^eq), AB signal (δ_A= 1.66, δ_B

=

(ddd, J_2,3= 7.5, J_2,4Ј= 5.3, J_2,1= 1.6 Hz, 1 H, 2-H), 2.59 (dd,

2

²J_4cis,4trans= 4.8, J_4cis,3= 2.7 Hz, 1 H, 4-H^cis), 2.92 (dd, J_4trans,4cis

1.98, J_AB= 14.9 Hz, in addition split by J_A,CH3= J_B,CH3= 7.3 Hz,

‡

= 4.8, J_4cis,3= 4.0 Hz, 1 H, 4-H^trans), 3.36 (ddd, J_3,2= 7.7, J_3,4trans

= 4.0, J_3,4cis= 2.7 Hz, 1 H, 3-H), AB signal (δ_A= 3.57, δ_B= 3.66,

J_AB= 10.5 Hz, in addition split by J_A,6Ј= 4.7, J_B,6Ј= 5.7 Hz, 2 H,

2 H, 2Ј-CH₂-CH₃), 2.47 (ddd, J_4,3= 8.7, J_4,4Ј= 3.9, J_4,5= 3.5 Hz,

1 H, 4-H), 2.58 (d, J_3OH,3= 5.5 Hz, 1 H, 3-OH), AB signal (δ_A

3.41, δ_B= 3.44, J_AB= 10.6 Hz, in addition split by J_A,5= 4.5, J_B,5

= 5.9 Hz, 2 H, 1ЈЈЈЈ-H₂), AB signal (δ_A= 3.53, δ_B= 3.59, J_AB

=

1ЈЈ-H₂), 4.01–4.12 (m, 1 H, 6Ј-H), 4.44 (ddd, J_4Ј,5Јax= 11.7, J_4Ј,2

=

10.1 Hz, in addition split by J_A,6Ј= 4.7, J_B,6Ј= 5.4 Hz, 2 H, 1ЈЈ-

H₂), AB signal (δ_A= 4.02, δ_B= 4.05, J_AB= 16.4 Hz, 2 H, 1ЈЈЈ-H₂),

4.21 (dddd, J_6Ј,5Јax= 11.3, J_6Ј,1ЈЈA= J_6Ј,1ЈЈB= 5.1, J_6Ј,5Јeq= 2.6 Hz,

1 H, 6Ј-H), 4.41 (ddd, J_4Ј,5Јax= 12.0, J_4Ј,4= 4.0, J_4Ј,5Јeq= 3.0 Hz,

1 H, 4Ј-H), 4.62 (dd, J_3,4= 8.8, J_3,3OH= 5.5 Hz, 1 H, 3-H), 4.68

(ddd, J_5,1ЈЈЈЈA= J_5,1ЈЈЈЈB= 5.4, J_5,4= 3.3 Hz, 1 H, 5-H) ppm. ¹³C

5.5, J_4Ј,5Јeq= 2.8 Hz, 1 H, 4Ј-H), 5.29 (s, 2 H, 1ЈЈЈ-H₂), 9.84 (d, J_1,2

= 1.7 Hz, 1 H, 1-H) ppm.

{[(4R,6S)-2,2-Diethyl-6-{(S)-2-hydroxy-1-[(R)-oxiran-2-yl]ethyl}-

1,3-dioxan-4-yl]methoxy}methyl Acetate (38):

‡

NMR (100.6 MHz, CDCl₃): δ = 6.80 (2Ј-CH₂-CH₃), 7.62 (C-1ЈЈЈЈ),

†

8.11 (2Ј-CH₂-CH₃), 22.48 (C-5Ј), 28.14 [C(CH₃)₃], 30.06 (2Ј-CH₂-

†

‡

CH₃), 30.96 (2Ј-CH₂-CH₃), 47.69 (C-4), 64.45 (C-4Ј), 67.39 (C-

3), 67.64 (C-6Ј), 69.29 (C-1ЈЈЈ), 74.42 (C-1ЈЈ), 78.00 (C-5), 81.68

[C(CH₃)₃], 102.35 (C-2Ј), 169.64 and 175.73 (C-2ЈЈЈ, C-2) ppm. IR

(film): ν = 3445, 2975, 2935, 2880, 1785, 1745, 1460, 1365, 1225,

˜

1135, 960, 845, 740 cm^–1. HRMS (EI, 70 eV): calcd. for C₁₈H₂₈IO₈

[M – C₂H₅]⁺499.08290; found 499.08220 (Δ = –1.4 ppm).

{[(4R,6S)-2,2-Diethyl-6-{(S)-1-[(R)-oxiran-2-yl]-2-oxoethyl}-1,3-

dioxan-4-yl]methoxy}methyl Acetate (36):

At 0 °C, aqueous LiOH [prepared from LiOH·H₂O (48 mg,

1.1 mmol, 6.0 equiv.) and H₂O (0.6 mL)] and Ag₂O (31 mg,

130 μmol, 0.6 equiv.) were added to a solution of iodolactone 35

(98 mg, 190 μmol, 1.0 equiv.) in THF/methanol (2:1, 1.2 mL). The

mixture was stirred at room temp for 1 h, cooled to 0 °C, and acidi-

fied (pH ≈ 3) by the cautious addition of ice-cold HCl (10% in

H₂O). The aqueous layer was saturated with NaCl and extracted

with Et₂O (6ϫ1.5 mL). The combined organic layers were dried

with Na₂SO₄, filtered, and concentrated in vacuo to yield crude 37

as a colorless oil (63 mg, 95%, no further purification).

At 0 °C, NaBH₄(47 mg, 820 μmol, 5.0 equiv.) was added in five

portions to a solution of the crude aldehyde 36 (53 mg, 160 μmol,

1.0 equiv.) in dry methanol (45 mL). After stirring at room temp.

for 2 h, aqueous phosphate buffer (pH = 7.1, 0.5 m, 2 mL) was

added. The resulting mixture was extracted with AcOEt (4ϫ2 mL).

The combined organic layers were washed with brine (2 mL), dried

with Na₂SO₄, filtered, and concentrated under reduced pressure.

The residue was purified by flash chromatography^[45](1.5 cm; cy-

6558

www.eurjoc.org

Eur. J. Org. Chem. 2013, 6545–6562

Cⁿ-Cⁿ⁺⁶Building Block for Polyol,Polyene Antibiotics

†

2

clohexane/AcOEt, 3:2) to yield 38 as a colorless oil (fractions 11–

20, 29 mg, 87 μmol, 46% over the three steps from iodolactone

CH₂-CH₃), 2.35 (dd, J_1A,1B= 15.0, J_1A,2= 1.3 Hz, 1 H, 1-H^A),

2.31–2.50 (m, 4 H, 4ЈЈ-H₂, 6ЈЈ-H₂), superimposed by 2.47 (ddd,

²J_{1ЈЈЈA,1ЈЈЈB}= 14.1, J_{1ЈЈЈA,2ЈЈЈB}= 12.3, J_{1ЈЈЈA,2ЈЈЈA}= 4.5 Hz, 1 H, 1ЈЈЈ-

H^A), 2.64 (dd, ²J_1B,1A= 14.9, J_1B,2= 8.2 Hz, 1 H, 1-H^B), 2.68 (ddd,

1

derivative 35). H NMR (400.1 MHz, CDCl₃/TMS): δ = 0.85 (dd,

A

B

3

A

B

†

³J_CH

= J_CH

= 7.5 Hz, 3 H, 2Ј-CH₂-CH₃), superimposed

H

3

A

B

A

B

‡

by 0.88 (dd, J_CH

=

³J_CH

= 7.4 Hz, 3 H, 2Ј-CH₂-CH₃), ²J_{1ЈЈЈB,1ЈЈЈA}= 14.1, J_{1ЈЈЈB,2ЈЈЈA}= 12.3, J_{1ЈЈЈB,2ЈЈЈB}= 4.7 Hz, 1 H, 1ЈЈЈ-

H

2

1.40 (ddt, J_2,3= 7.4, J_2,4Ј= 5.4, J_2,1A= 4.6 Hz, 1 H, 2-H), 1.41

H^B), 3.00 (ddd, J_{2ЈЈЈA,2ЈЈЈB}= J_{2ЈЈЈA,1ЈЈЈB}= 12.7, J_{2ЈЈЈA,1ЈЈЈA}= 4.4 Hz,

2

(ddd, J_5Јax,5Јeq= 12.8, J_5Јax,4Ј= 12.0, J_5Јax,6Ј= 11.0 Hz, 1 H, 5Ј- 1 H, 2ЈЈЈ-H^A), 3.23 (ddd, J_{2ЈЈЈB,2ЈЈЈA}= J_{2ЈЈЈB,1ЈЈЈA}= 12.8, J_{2ЈЈЈB,1ЈЈЈB}

H^ax), 1.54 (ddd, J_5Јeq,5Јax= 12.6, J_5Јeq,4Ј= J_5Јeq,6Ј= 2.6 Hz, 1 H,

= 4.6 Hz, 1 H, 2ЈЈЈ-H^B), 3.42–3.52 (m, 2 H, 1ЈЈЈЈ-H₂), 3.62 (d, J_2-

2

5Ј-H^eq), AB signal (δ_A= 1.58, δ_B= 1.63, J_AB= 14.6 Hz, in addition

= 3.9 Hz, 1 H, 2-OH), 3.63 (dddd, J_6Ј,5Јax= 11.7, J_{6Ј,1ЈЈЈЈA}

split by J_A,CH3= J_B,CH3= 7.4 Hz, 2 H, 2Ј-CH₂-CH₃), AB signal J_{6Ј,1ЈЈЈЈB}= J_6Ј,5Јeq= 2.5 Hz, 1 H, 6Ј-H), AB signal (δ_A= 3.94, δ_B

=

OH,2Ј

‡

(δ_A= 1.82, δ_B= 1.88, J_AB= 14.8 Hz, in addition split by J_A,CH3

=

3.99, J_AB= 10.5 Hz, in addition split by J_A,3= 6.3, J_B,3= 4.9 Hz,

†

J_B,CH3= 7.4 Hz, 2 H, 2Ј-CH₂-CH₃), 2.09 (s, 2ЈЈЈЈ-H₃), 2.37 (br. s, 2 H, 4-H₂), 4.29 (ddd, J_4Ј,5Јax= 11.9, J_4Ј,3= 4.9, J_4Ј,5Јeq= 2.40, 1

2

1 H, 1-OH), 2.65 (dd, J_4cis,4trans= 4.8, J_4cis,3= 2.8, 1 H, 4-H^cis), H, 4Ј-H), 4.87 (dddd, J_2,1B= 8.2, J_2,2–OH= 4.0, J_2,3= J_2,1A

=

2.88 (dd, ²J_4trans,4cis= 4.8, J_4cis,3= 4.0 Hz, 1 H, 4-H^trans), 3.20 (ddd, 2.0 Hz, 1 H, 2-H), 7.06 (m_c, 1 H, 4^Ar-H), 7.17 (m_c, 2 H, 2ϫ3^Ar

-

J_3,2= 7.3, J_3,4trans= 4.1, J_3,4cis= 2.8 Hz, 1 H, 3-H), AB signal (δ_A

= 3.59, δ_B= 3.67, J_AB= 10.5 Hz, in addition split by J_A,6Ј= 4.5,

J_B,6Ј= 5.8 Hz, 2 H, 1ЈЈ-H₂), 3.87 (m_c, 2 H, 1-H₂), 4.07 (dddd, J_6Ј,5Јax18.33 [SiC(CH₃)₃], 22.59 (2Ј-CH₂-CH₃), 25.58 (C-5ЈЈ), 26.02 and

= 11.5, J_6Ј,1ЈЈB= 5.8, J_6Ј,1ЈЈA= 4.5, J_6Ј,5Јeq= 2.6 Hz, 1 H, 6Ј-H), 26.11 (C-4ЈЈ, C-6ЈЈ), 26.08 [SiC(CH₃)], 30.73 (C-5Ј), 31.44 (2Ј-CH₂-

H), 7.29 (m_c, 2 H, 2ϫ2^Ar-H) ppm. ¹³C NMR (125.6 MHz, C₆D₆):

‡

†

δ = –5.36 (2ϫSiCH₃), 7.21 (2Ј-CH₂-CH₃), 8.22 (2Ј-CH₂-CH₃),

†

‡

4.20 (ddd, J_4Ј,5Јax= 11.7, J_4Ј,2= 5.6, J_4Ј,5Јeq= 2.6 Hz, 1 H, 4Ј-H), CH₃), 31.86 (C-2ЈЈЈ), 41.17 (C-1ЈЈЈ), 44.88 (C-1), 52.40 (C-3), 53.61

AB signal (δ_A= 5.30, δ_B= 5.30, J_AB= 6.2 Hz, 2 H, 1ЈЈЈ-H₂) ppm.

(C-2ЈЈ), 60.27 (C-4), 66.18 (C-1ЈЈЈЈ), 67.39 (C-2), 68.28 (C-4Ј), 69.53

¹³C NMR (100.6 MHz, CDCl₃/TMS): δ = 6.90 (2Ј-CH₂-CH₃^‡),

(C-6Ј), 102.25 (C-2Ј), 126.09 (C-4^Ar), 128.71 (2 ϫC-3^Ar), 129.00

8.05 (2Ј-CH₂-CH₃), 21.05 (C-2ЈЈЈЈ), 22.24 (2Ј-CH₂-CH₃), 30.79 (2ϫC-2^Ar), 142.76 (C-1^Ar) ppm. IR (film): ν = 3455, 2955, 2930,

†

˜

‡

(C-5Ј), 31.07 (2Ј-CH₂-CH₃), 47.24 (C-4), 48.17 (C-2), 51.17 (C-3),

2880, 2860, 1465, 1380, 1360, 1255, 1165, 1105, 980, 940, 835, 775,

61.58 (C-1ЈЈ), 67.65 (C-6Ј), 68.21 (C-4Ј), 73.44 (C-1), 89.54 (C-1ЈЈЈ), 665 cm^–1. C₁₉H₃₈O₅Si (374.59): C 60.92, H 10.22; found C 60.71,

102.07 (C-2Ј), 170.57 (C-1ЈЈЈЈ) ppm. IR (film): ν = 3460, 2975,

H 10.47.

˜

2940, 2885, 1745, 1465, 1370, 1230, 1165, 1125, 1015, 925,

835 cm^–1.

Supporting Information (see footnote on the first page of this arti-

cle): Experimental details of the Wittig rearrangement model study,

NMR comparison of the rearrangement products, NMR spectra,

X-ray data.

(2S,3R)-4-(tert-Butyldimethylsiloxy)-3-[(4S,6R)-2,2-diethyl-6-

(hydroxymethyl)-1,3-dioxan-4-yl]-1-[2-(2-phenylethyl)-1,3-dithian-

2-yl]butan-2-ol (45):

Acknowledgments

The authors express their gratitude to Dr. J. Geier and Dr. M.

Keller, Institut für Organische Chemie, University of Freiburg for

performing the X-ray diffraction analyses.

[1] For reviews, see: a) J. M. T. Hamilton-Miller, Bacteriol. Rev.

1973, 37, 166–196; b) J.-M. Beau, G. Lukacs, in: Recent Pro-

gress in the Chemical Syntheses of Antibiotics (Ed.: M. Ohno),

Springer, Berlin, 1990, p. 137–179; c) S. Omura, H. Tanaka,

Macrolide Antibiotics: Chemistry, Biochemistry and Practice

(Ed.: S. Omura), Academic Press, New York, 1984, p. 351–404;

d) S. D. Rychnovsky, Chem. Rev. 1995, 95, 2021–2040; e) S. B.

Zotchev, Curr. Med. Chem. 2003, 10, 211–223.

[2] a) For the isolation from Streptomyces nodosus, see: J. Vander-

putte, J. L. Wachtel, E. T. Stiller, Antibiot. Ann. 1956, 587–591;

for connectivity and configuration, see: b) W. Mechlinski, C. P.

Schaffner, P. Ganis, G. Avitabile, Tetrahedron Lett. 1970, 11,

3873–3876; c) P. Ganis, G. Avitabile, W. Mechlinski, C. P.

Schaffner, J. Am. Chem. Soc. 1971, 93, 4560–4564.

[3] For the isolation from Streptomyces noursei, see: a) E. L.

Hazen, R. Brown, Science 1950, 112, 423–430; b) E. L. Hazen,

R. Brown, Proc. Soc. Exp. Biol. Med. 1951, 76, 93–98; for con-

nectivity, see: c) C. N. Chong, R. W. Rickards, Tetrahedron

Lett. 1970, 11, 5145–5149; d) E. Borowski, J. Zielin´ski, L. Fal-

kowski, T. Zimin´ski, J. Golik, P. Kołodziejczyk, E. Jereczek,

M. Gdulewicz, Yu. Shenin, T. Kotienko, Tetrahedron Lett.

1971, 12, 685–690; e) for the relative configuration, see: J.-M.

Lancelin, F. Paquet, J.-M. Beau, Tetrahedron Lett. 1988, 23,

2827–2830; for the absolute configuration, see: f) K. C. Nico-

laou, K. H. Ahn, Tetrahedron Lett. 1989, 30, 1217–1220; g) J.-

M. Lancelin, J.-M. Beau, Tetrahedron Lett. 1989, 30, 4521–

4524.

At 0 °C, nBuLi (2.4 m in hexanes, 135 μL, 326 μmol, 3.5 equiv.) was

added dropwise to a solution of dithiane 44 (84 mg, 370 μmol,

4.0 equiv.) in THF (1.5 mL). The resulting mixture was stirred for

30 min. At –78 °C, a solution of epoxide 7b (35 mg, 93 μmol,

1.0 equiv.) in THF (1.5 mL)/HMPA (67 μL, 370 μmol, 4.0 equiv.)

was added. After stirring for 16 h at –40 °C, the reaction mixture

was quenched by the addition of aqueous satd. NH₄Cl. After ex-

traction with Et₂O (3ϫ2 mL), the combined organic layers were

dried with Na₂SO₄, filtered, and concentrated in vacuo. Flash

chromatography^[45](1.5 cm; petroleum ether/Et₂O, 7:3) of the resi-

due afforded the title compound as a colorless oil (fractions 30–45,

39 mg, 70%). Excess dithiane 44 was also isolated (fractions 2–6,

49 mg, 78% of initial excess of this reagent). [α]²_D⁰= +11.0 (c =

1

1.20, CDCl₃, 10 cm). H NMR (499.6 MHz, C₆D₆): δ = 0.07 and

3

A

B

A B

0.08 (2ϫs, 2 ϫ3 H, 2ϫSiCH₃), 0.77 (dd, J_CH

=

³J_CH

=

H

†

3

7.5 Hz, 3 H, 2Ј-CH₂-CH₃), 0.93 (t, J_CH2= 7.4 Hz, 3 H, 2Ј-CH₂-

CH₃), 0.97 [s, 9 H, SiC(CH₃)₃], 1.17 (ddd, ²J_5Јeq,5Јax= 12.7, J_5Јeq,4Ј

‡

= J_5Јeq,6Ј= 2.4 Hz, 1 H, 5Ј-H^eq), 1.42–1.59 (m, 2 H, 5ЈЈ-H₂), super-

imposed by 1.51–1.62 (m, 1 H, 5Ј-H^ax), superimposed by 1.56–1.64

3

(m, 1 H, 1ЈЈЈЈ-OH), superimposed by 1.57 (q, J_CH3= 7.7 Hz, 2 H,

2Ј-CH₂-CH₃), 1.67 (dddd, J_3,4A= 6.8, J_3,4B= 4.8, J_3,4Ј= 4.5, J_3,2

‡

= 1.9 Hz, 1 H, 3-H), AB signal (δ_A= 1.68, δ_B= 1.73, J_AB

=

14.6 Hz, in addition split by J_A,CH3= J_B,CH3= 7.4 Hz, 2 H, 2Ј-

Eur. J. Org. Chem. 2013, 6545–6562

www.eurjoc.org

6559

L. Nachbauer, R. Brückner

FULL PAPER

[4] a) For the isolation from Streptomyces natalensis, see: A. P.

Struyk, I. Hoette, G. Drost, J. M. Waisvisz, T. van Eek, J. C.

Hoogerheide, Antibiot. Ann. 1957–1958, 5, 878–885; b) for

cconnectivity, see: B. T. Golding, R. W. Richards, W. E. Meyer,

J. B. Patrick, M. Barber, Tetrahedron Lett. 1966, 7, 3551–3557;

c) for the absolute configuration, see: J.-M. Lancelin, J.-M.

Beau, J. Am. Chem. Soc. 1990, 112, 4060–4061.

[11]

For reviews, see: a) J. Bolard, Biochim. Biophys. Acta - Reviews

on Biomembranes 1986, 864, 257–304; b) S. C. Hartsel, C.

Hatch, W. Ayenew, J. Liposome Res. 1993, 3, 377–408; c) ref.^[1e]

S. Hartsel, J. Bolard, Trends Pharmacol. Sci. 1996, 17, 445–

449.

Amphotericin B (1) has a higher affinity to ergosterol than to

cholesterol in lipid bilayers: a) T. Teerlink, B. de Kruijff, R. A.

Demel, Biochim. Biophys. Acta - Biomembranes 1980, 599, 484–

492; b) I. Fournier, J. Barwicz, P. Tancrède, Biochim. Biophys.

Acta - Biomembranes 1998, 1373, 76–86; c) M. J. Paquet, I.

Fournier, J. Barwicz, P. Tancrède, M. Auger, Chem. Phys. Lipids

2002, 11, 1–11.

a) M. Szpilman, D. M. Cereghetti, J. M. Manthorpe, N. R.

Wurtz, E. M. Carreira, Chem. Eur. J. 2009, 15, 7117–7128; b)

D. S. Palacios, T. M. Anderson, M. D. Burke, J. Am. Chem.

Soc. 2007, 129, 13804–13805; c) A. A. Volmer, A. M. Szpilman,

E. M. Carreira, Nat. Prod. Rep. 2010, 27, 1329–1349; d) M. P.

Croatt, E. M. Carreira, Org. Lett. 2011, 13, 1390–1393; e) D. S.

Palacios, I. Dailey, D. M. Siebert, B. C. Wilcock, M. D. Burke,

Proc. Natl. Acad. Sci. USA 2011, 108, 6733–6738.

For pimaricin (4), see: a) B. De Kruijff, W. J. Gerritsen, A. Oer-

lemans, R. A. Demel, L. L. M. van Deenen, Biochim. Biophys.

Acta - Biomembranes 1974, 339, 30–43; for Rimocidin (5), see:

b) V. W. Dennis, N. W. Stead, T. E. Andreoli, J. Gen. Physiol.

1970, 55, 375–400; c) W. A. Zygmunt, Appl. Microbiol. 1966,

14, 953–956.

M. R. Van Leeuwen, E. A. Golovina, J. Dijksterhuis, J. Appl.

Microbiol. 2009, 106, 1908–1918.

Y. M. te Welscher, L. Jones, M. R. van Leeuwen, J. Dijkster-

huis, B. de Kruijff, G. Eitzen, E. Breukink, Antimicrob. Agents

Chemother. 2010, 54, 2618–2625.

For reviews, see: a) T. Oishi, T. Nakata, Synthesis 1990, 635–

645; b) C. Schneider, Angew. Chem. 1998, 110, 1445–1448; An-

gew. Chem. Int. Ed. 1998, 37, 1375–1378; c) J. Dougherty, M.

Zhou, G. A. O’Doherty, Chemtracts: Org. Chem. 2005, 18,

349–363; d) S. E. Bode, M. Wolberg, M. Müller, Synthesis

2006, 4, 557–588; e) D. M. Cereghetti, E. M. Carreira, Synthe-

sis 2006, 914–942; f) A. M. Szpilman, E. M. Carreira, Angew.

Chem. 2010, 122, 9786–9823; Angew. Chem. Int. Ed. 2010, 49,

9592–9628.

[12]

[13]

[5] The compound known as “pimaricin” has also been called nat-

amycin (H. Brik, Analytical Profiles of Drug Substances 1981,

10, 513–561) and tennecetin (P. V. Divekar, J. L. Bloomer, J. F.

Eastham, D. F. Holtman, D. A. Shirley, Antibiot. Chemother.

1961, 11, 377–380).

[6] a) For a web site on pimaricin, see: http://ww w .

[14]

[15]

codexalimentarius.net/gsfaonline/additives/

details.html?id=208&lang=en; b) for pimaricin as a food addi-

tive, see: F. Aguilar, U. R. Charrondiere, B. Dusemund, P. Galt-

ier, J. Gilbert, D. M. Gott, S. Grilli, R. Guertler, J. Koenig, C.

Lambré, J.-C. Larsen, J.-C. Leblanc, A. Mortensen, D. Parent-

Massin, I. Pratt, I. M. C. M. Rietjens, I. Stankovic, P. Tobback,

T. Verguieva, R. A. Woutersen, EFSA J. 2009, 7, 1412–1427;

(http://ww w .efsa.europa.eu/en/efsajournal/pub/1412.htm).

[7] a) For the isolation from Streptomyces viridoflavus, see: W. A.

Taber, L. C. Vining, S. A. Waksman, Antibiot. Chemother.

1954, 4, 455–461; b) for connectivity, see: E. Borowski, L. Fal-

kowski, J. Golik, J. Zielin´ski, T. Zimin´ski, W. Mechlin´ski, E.

Jereczek, P. Kołodziejczyk, H. Adlercreutz, C. P. Schaffner, S.

Neelakantan, Tetrahedron Lett. 1971, 12, 1987–1992; c) for the

absolute configuration, see: J. Pawlak, P. Sowinski, E. Borow-

ski, P. Gariboldi, J. Antibiot. 1993, 46, 1598–1603.

[8] a) For the isolation from Streptomyces rimosus, see: J. W. Davis-

son, F. W. Tanner, A. C. Finlay, I. A. Solomons, Antibiot.

Chemother. 1951, 1, 289–290; b) for connectivity, see: A. C.

Cope, R. K. Bly, E. P. Burrows, O. J. Ceder, E. Ciganek, B. T.

Gillis, R. F. Porter, H. E. Johnson, J. Am. Chem. Soc. 1962, 84,

2170–2181; c) for the absolute configuration, see: P. Sowin´ski,

J. Pawlak, E. Borowski, P. Gariboldi, J. Antibiot. 1995, 48,

1288–1291.

[16]

[17]

[18]

[9] T. Brautaset, H. Sletta, A. Nedal, S. E. F. Borgos, K. F. De-

gnes, I. Bakke, O. Volokhan, O. N. Sekurova, I. D. Treshalin,

E. P. Mirchink, A. Dikiy, T. E. Ellingsen, S. B. Zotchev, Chem.

Biol. 2008, 15, 1198–1206.

[19]

For the total synthesis of amphotericin B (1), see: a) K. C. Nic-

olaou, T. K. Chakraborty, Y. Ogawa, R. A. Daines, N. S.

Simpkins, G. T. Furst, J. Am. Chem. Soc. 1988, 110, 4660–4672;

b) K. C. Nicolaou, R. A. Daines, J. Uenishi, W. S. Li, D. P. Pa-

pahatjis, T. K. Chakraborty, J. Am. Chem. Soc. 1988, 110,

4672–4685; c) K. C. Nicolaou, R. A. Daines, T. K. Chakra-

borty, J. Am. Chem. Soc. 1988, 110, 4685–4696; d) K. C. Nico-

laou, R. A. Daines, Y. Ogawa, T. K. Chakraborty, J. Am.

Chem. Soc. 1988, 110, 4696–4705; for the total synthesis of the

methyl ester of 35-deoxyamphotericin B, see: e) A. M. Szpil-

man, D. M. Cereghetti, N. R. Wurtz, J. M. Manthorpe, E. M.

Carreira, Angew. Chem. 2008, 120, 4407–4410; Angew. Chem.

Int. Ed. 2008, 47, 4335–4338; for the total synthesis of the

methyl ester of 35-deoxyamphotericin B, see: f) A. M. Szpil-

man, J. M. Manthorpe, E. M. Carreira, Angew. Chem. 2008,

120, 4411–4414; Angew. Chem. Int. Ed. 2008, 47, 4339–4342;

for the formal total synthesis of amphotericinolide B, see: g)

D. Boschelli, T. Takemasa, Y. Nishitani, S. Masamune, Tetrahe-

dron Lett. 1985, 26, 5239–5242; h) R. M. Kennedy, A. Abiko,

S. Masamune, Tetrahedron Lett. 1988, 29, 447–450; i) R. M.

Kennedy, A. Abiko, T. Takemasa, H. Okumoto, S. Masamune,

Tetrahedron Lett. 1988, 29, 451–454; g) for the relay synthesis

of unnatural glycosides of amphoteronolide B, see ref.^[14c]

For the total synthesis of the aglycon of rimocidin (5), see: a)

G. K. Packard, Y. Hu, A. Vescovi, S. D. Rychnovsky, Angew.

Chem. 2004, 116, 2882–2886; Angew. Chem. Int. Ed. 2004, 43,

2822–2826; b) A. B. Smith III, M. A. Foley, S. Dong, A. Orbin,

J. Org. Chem. 2009, 74, 5987–6001; c) for the preparation of

the C¹-C¹⁸fragment of rimocidin, see: A. B. Smith III, S. M.

Pitram, M. J. Fuertes, Org. Lett. 2003, 5, 2751–2754.

[10] For example, the following polyol,polyene macrolides, ampho-

tericin A: a) for isolation from Streptomyces nodosus, see: J.

Vandeputte, J. L. Wachtel, E. T. Stiller, Antibiot. Ann. 1955–

1956, 587–591; for connectivity, see: b) P. Sowin´ski, J. K. Paw-

lak, E. Borowski, T. Iwashita, J. Antibiot. 1985, 38, 175–180;

c) A. Aszalos, A. Bax, N. Burlinson, P. Roller, C. McNeal, J.

Antibiot. 1985, 38, 1699–1713; lucensomicin: d) for the isola-

tion from Streptomyces lucensis, see: F. Arcamone, C. Bertaz-

zoli, G. Canevazzi, A. Dimarco, M. Ghione, A. Grein, Giorn.

Microbiol. 1957, 4, 119–128; e) for connectivity, see: G. Gaudi-

ano, P. Bravo, A. Quilico, Tetrahedron Lett. 1966, 7, 3559–3565;

G. Gaudiano, P. Bravo, A. Quilico, Tetrahedron Lett. 1966, 7,

3567–3571; tetrin A and B: f) for the isolation from Streptomy-

ces Illinois, see: D. Gottlieb, H. L. Pote, Phytopathology 1960,

50, 817–822; g) for the connectivity of tetrin A, see: R. C. Pan-

dey, V. F. German, Y. Nishikawa, K. L. Rinehart Jr, J. Am.

Chem. Soc. 1971, 93, 3738–3747; h) for the connectivity of te-

trin B, see: K. L. Rinehart Jr., W. P. Tucker, R. C. Pandey, J.

Am. Chem. Soc. 1971, 93, 3747–3751; tetrin C: i) for the isola-

tion from Streptomyces sp. GK9244 and connectivity, see:G.

Ryu, W.-C. Choi, S. Hwang, W.-H. Yeo, C.-S. Lee, S.-K. Kim,

J. Nat. Prod. 1999, 62, 917–919; HA-2–91: j) for the isolation

from Streptomyces sp. UK10, see: T. E. Gupte, S. R. Naik,

Hind. Antibiot. Bull. 1991, 33, 1–6; k) for connectivity, see: T. E.

Gupte, R. K. Nanda, N. R. Chatterjee, S. R. Naik, Indian J.

Chem. Sect. B 2000, 39, 936–940; 3874 H1, 3874 H2, and 3874

H3: l) for the isolation from Streptomyces sp. HAG 003874 and

connectivities, see: L. Vértesy, W. Aretz, E. Ehlers, S. Hawser,

D. Isert, M. Knauf, M. Kurz, M. Schiell, M. Vogel, J. Wink,

J. Antibiot. 1998, 51, 921–928.

[20]

6560

www.eurjoc.org

Eur. J. Org. Chem. 2013, 6545–6562

Cⁿ-Cⁿ⁺⁶Building Block for Polyol,Polyene Antibiotics

1

19

[21]

was also feasible by using Sharpless’ catalytic conditions, but

distinctly slower. Separation by flash chromatography on silica

gel^[45]was tedious, particularly on a large scale. We circum-

vented this problem by treating the 78:22 anti,cis-16/syn,cis-

16 mixture with a 10-fold molar amount of Red-Al^®(toluene,

–40 °C). Under these conditions, diastereomer syn,cis-16 re-

mained unchanged, whereas diastereomer anti,cis-16 gave the

diol syn,cis-14. The resulting mixture was easily separated by

flash chromatography on silica gel.^[45]The yield of syn,cis-14

was 46% over the two steps from cis,cis-17 and its ee was

Ͼ95% according to chiral HPLC.

For vanadium(V)-catalyzed epoxidation reactions of β-substi-

tuted γ,δ-unsaturated alcohols in which C^αand C^βpossess the

same relative configurations as in 7b, see: a) E. D. Mihelich, K.

Daniels, D. J. Eickhoff, J. Am. Chem. Soc. 1981, 103, 7690–

7692; b) C. Kouklovsky, S. V. Ley, S. S. P. Marsden, Tetrahe-

dron Lett. 1994, 35, 2091–2094; c) M. Oikawa, T. Ueno, H.

Oikawa, A. Ichihara, J. Org. Chem. 1995, 60, 5048–5068; d) H.

Oikawa, Y. Yoneta, T. Ueno, M. Oikawa, T. Wakayama, A.

Ichihara, Tetrahedron Lett. 1997, 38, 7897–7900; e) E. Me-

nager, E. Merifield, M. Smallridge, E. J. Thomas, Tetrahedron

1997, 53, 9377–9392; f) J. D. White, G. Wang, L. Quaranta,

Org. Lett. 2003, 5, 4109–4112; g) K. Prantz, J. Mulzer, Chem.

Eur. J. 2010, 16, 485–506.

For reviews, see: a) T. Nakai, K. Mikami, Chem. Rev. 1986, 86,

885–902; b) R. Brückner, Kontakte (Darmstadt) 1991, 2, 3–14;

c) R. Brückner, Kontakte (Darmstadt) 1991, 3, 3–15; also see:

d) R. Brückner, Heteroatom Manipulation (Ed.: E. Win-

terfeldt), in: Comprehensive Organic Synthesis (Eds.: B. M.

Trost, I. Fleming), Pergamon Press, Oxford, UK, 1991, vol. 6,

p. 873–908; e) K. Mikami, T. Nakai, Synthesis 1991, 594–604;

f) T. Nakai, K. Mikami, Org. React. 1994, 46, 105–210.

a) R. Brückner, H. Priepke, Angew. Chem. 1988, 100, 285–286;

Angew. Chem. Int. Ed. Engl. 1988, 27, 278–280; b) R. Brückner,

Chem. Ber. 1989, 122, 703–710.

a) For the synthesis of the C –C polyol fragment of nystatin

A₁(2), see: I. Kadota, Y. Hu, G. K. Packard, S. D. Rychnovsky,

Proc. Natl. Acad. Sci. USA 2004, 101, 11992–11995; b) for the

synthesis of the C¹-C¹³fragment of nystatin A₁(2), see: G.

Solladié, N. Wilb, C. Bauder, J. Org. Chem. 1999, 64, 5447–

5452.

1

19

[22]

[23]

[24]

For the synthesis of the C –C polyol fragment of candidin

(3), see ref.^[21a]

For the total synthesis of pimaricin (4), see: A. J. Duplantier,

S. Masamune, J. Am. Chem. Soc. 1990, 112, 7079–7081.

For the total synthesis of RK-397, see: a) S. A. Burova, F. E.

McDonald, J. Am. Chem. Soc. 2004, 128, 2495–2500; b) S. E.

Denmark, S. Fujimori, J. Am. Chem. Soc. 2005, 127, 8971–

8973; c) M. J. Mitton-Fry, A. J. Cullen, T. Sammakia, Angew.

Chem. 2007, 119, 1084–1088; Angew. Chem. Int. Ed. 2007, 46,

1066–1070; for the synthesis of the C^11–28fragment of RK-397,

see: d) S. A. Burova, F. E. McDonald, J. Am. Chem. Soc. 2002,

124, 8188–8189; e) C. Schneider, F. Tolksdorf, M. Rehfeuter,

Synlett 2002, 2098–2100; f) S. Gerber-Lemaire, A. T. Carmona,

K. T. Meilert, P. Vogel, Eur. J. Org. Chem. 2006, 891–900; for

the formal total synthesis of RK-397, see: g) H. Guo, M. S.

Mortensen, G. A. O’Doherty, Org. Lett. 2008, 10, 3149–3152.

For the total synthesis of mycotycin A, see: a) C. S. Poss, S. D.

Rychnovsky, S. L. Schreiber, J. Am. Chem. Soc. 1993, 115,

3360–3361; for the formal total synthesis of mycotycin A, see:

b) A. B. Smith III, S. M. Pitram, Org. Lett. 1999, 1, 2001–2004;

c) S. D. Dreher, J. L. Leighton, J. Am. Chem. Soc. 2001, 123,

341–342.

For the total synthesis of roxaticin, see: a) S. D. Rychnovsky,

R. C. Hoye, J. Am. Chem. Soc. 1994, 116, 1753–1765; b) D. A.

Evans, B. T. Connell, J. Am. Chem. Soc. 2003, 125, 10899–

10905; c) S. B. Han, A. Hassan, I. S. Kim, M. J. Krische, J.

Am. Chem. Soc. 2010, 132, 15559–15561; for the synthesis of

the C¹²-C³⁰fragment of roxaticin, see: d) S. D. Rychnovsky, C.

Rodriguez, J. Org. Chem. 1992, 57, 4793–4795; for the synthesis

of the C¹¹-C²⁶fragment of roxaticin, see: Y. Mori, M. Asai, A.

Okumura, H. Furukawa, Tetrahedron 1995, 51, 5299–5314.

For the total synthesis of roflamycoin, see: a) S. D. Rychnov-

sky, U. R. Khire, G. Yang, J. Am. Chem. Soc. 1997, 119, 2058–

2059; for the total synthesis of the spiroketal derivative of rofla-

mycoin, see: b) S. D. Rychnovsky, G. Griegraber, R. Schlegel,

J. Am. Chem. Soc. 1994, 116, 2621–2622; for the synthesis of

the C^12–35fragment of roflamycoin, see: c) B. H. Lipshutz, R.

Moretti, R. Crow, Tetrahedron Lett. 1989, 30, 15–18.

[34]

[25]

[26]

[35]

[36]

[37]

[38]

B. Herradón, Tetrahedron: Asymmetry 1991, 2, 191–194.

P. J. Kocien´ski, C. Yeates, S. D. A. Street, S. F. Campbell, J.

Chem. Soc. Perkin Trans. 1 1987, 2183–2187.

R. C. Corcoran, Tetrahedron Lett. 1990, 31, 2101–2104.

J. P. Parikh, W. von E. Doering, J. Am. Chem. Soc. 1967, 89,

5505–5507.

[27]

[28]

[39]

[40]

[41]

[42]

A. J. Mancuso, S.-L. Huang, D. Swern, J. Org. Chem. 1978, 43,

2480–2482.

For the formation of ethers using phase-transfer catalysis, see:

H. H. Freedman, R. A. Dubois, Tetrahedron Lett. 1975, 16,

3251–3254.

For the total synthesis of dermostatin A, see: a) C. J. Sinz, S. D.

Rychnovsky, Angew. Chem. 2001, 113, 3324–3327; Angew.

Chem. Int. Ed. 2001, 40, 3224–3227; b) C. J. Sinz, S. D. Rych-

novsky, Tetrahedron 2002, 58, 6561–6576.

3,

CCDC-849003 (for syn *,syn^2,3–26a), -849005 (for anti^3,*,

[43]

syn^2,3–26d), and -849004 (for 30a) contain the supplementary

crystallographic data for this paper. These data can be obtained

free of charge from The Cambridge Crystallographic Data

Centre via www.ccdc.cam.ac.uk/data_request/cif.

[29]

[30]

R. Brückner, Tetrahedron Lett. 1988, 29, 5747–5750.

a) R. Kramer, R. Brückner, Eur. J. Org. Chem. 2013, DOI:

10.1002/ejoc.201300182 (following contribution in this jour-

nal); the article describes the synthesis of the mirror image of

the C¹²-C¹⁸building block 7a; b) the mirror image of the diox-

ane 12 of Scheme 2 was first described by R. Kramer, Ph.D.

Dissertation, University of Freiburg, Germany, 2007, p. 151–

152, 447–449.

T. Berkenbusch, R. Brückner, Synlett 2003, 1813–1816.

R. Kramer, T. Berkenbusch, R. Brückner, Adv. Synth. Catal.

2008, 350, 1131–1148.

Our preparation of diol syn,cis-14 from divinylcarbinol cis,cis-

17 via epoxy alcohol anti,cis-16 meant a modification of earlier

protocols^[31,32]and an adoption of the protocol optimized in

the enantiomeric series.^[30a]However, the yield here (46%

syn,cis-14 over the two steps from cis,cis-17) was less than the

yield in ref.^[30a](59% ent-syn,cis-10 over the two steps from

cis,cis-17). In essence, cis-divinylcarbinol cis,cis-17 was submit-

ted to asymmetric epoxidation under Sharpless’ stoichiometric

conditions^[68,71](plus molecular sieves). After 70 h at –25 °C, a

78:22 mixture of the desired epoxy alcohol anti,cis-16 and its

undesired diastereomer syn,cis-16 was obtained. The reaction

[44]

[45]

These data are available in the Supporting Information.

W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43, 2923–

2925.

V

[46]

[47]

For V -catalyzed epoxidations of achiral homoallyl alcohols

with CumylOOH, see: W. Zhang, H. Yamamoto, J. Am. Chem.

Soc. 2007, 129, 286–287.

[31]

[32]

VO(OEt)₃was a better catalyst than VO(acac)₂for the epoxid-

ation of homoallylic alcohols, see: ref.^[29]The same holds true

for the epoxidation of allylic alcohols, see footnote 21 in: T.

Katsuki, A. W. M. Lee, P. Ma, V. S. Martin, S. Masamune,

K. B. Sharpless, D. Tuddenham, F. J. Walker, J. Org. Chem.

1982, 47, 1373–1378.

For selected iodocarbonate formations from homoallyl

alcohols by in-situ formed lithium homoallylic carbonates, see:

a) A. Bongini, G. Cardillo, M. Orena, G. Porzi, S. Sandri, J.

Org. Chem. 1982, 47, 4626–4633; b) M. Majewski, D. L. J.

Clive, P. C. Anderson, Tetrahedron Lett. 1984, 25, 2101–2104;

c) R. Tirado, J. Prieto, J. Org. Chem. 1993, 58, 5666–5673.

[33]

[48]

Eur. J. Org. Chem. 2013, 6545–6562

www.eurjoc.org

6561

L. Nachbauer, R. Brückner

FULL PAPER

[49] The alcohol 22d was deprotonated by nBuLi (1.05 equiv.) in

THF at –78 °C while stirring for 30 min. After warming to

–40 °C, dry CO₂was bubbled through the solution. After

15 min iodine (4 equiv.) was added. The resulting mixture was

kept at 0 °C for 16 h.

[50] The intramolecular opening of an iodonium ion by a CO₂H

group, that is, the formation of an (iodomethyl)-γ-lactone in

the absence of a base, occurs reversibly, that is, under thermo-

dynamic control: P. A. Bartlett, J. Myerson, J. Am. Chem. Soc.

1978, 100, 3950–3952.

[51] The presence of NaHCO₃was necessary to avoid cleavage of

the ketal group through HI; the latter is the stoichiometric side-

product of iodolactonization reactions. Fortunately, NaHCO₃

was unlikely to shift the diastereocontrol from thermodynamic

(desired) to kinetic (undesired) (as would be the case if the

substrate was an unsaturated acid^[50]).

37Ǟ36. This is because hydrocarbon carboxylic acids react to

give a carbenium ion (i.e., no carboxonium ion as when starting

from 37, 40, 41 or 46 and no iminium ion as when starting

from 42 or 43). The carbenium ion is usually deprotonated

by the acetate ion (and not nucleophilically attacked like the

mentioned carboxonium or iminium ions). This renders 1-ole-

fins. However, significant amounts of carbenium ion (with or

without an intervening Wagner–Meerwein rearrangement)/

acetate recombination products resulted from analogous decar-

boxylations of cyclobutanecarboxylic acid: J. K. Kochi, J. D.

Bacha, J. Org. Chem. 1968, 33, 2746–2754. This indicates a

strain-induced deceleration of cyclobutene formation.

[60] Light-supported Pb(OAc)₄decarboxylation reactions of ali-

phatic carboxylic acids without a heteroatom α to the CO₂H

group have been performed by using in CCl₄and I₂. This fur-

nished C₁-shortened aliphatic iodides: D. H. R. Barton, H. P.

Faro, E. P. Serebryakov, N. F. Woolsey, J. Chem. Soc. 1965,

2438–2444.

[61] a) Y. Ohishi, T. Nakanishi, Chem. Pharm. Bull. 1983, 31, 3418–

3423; b) J. E. H. Buston, A. Coop, R. Keady, M. G. Moloney,

R. M. Thompson, J. Chem. Res. (M) 1994, 1101–1116.

[62] For an example of a Pb(OAc)₄-mediated degradation of an N-

arylglycinamide, see: O. Süs, Justus Liebigs Ann. Chem. 1949,

564, 137–140.

[63] For an example of a Pb(OAc)₄-mediated degradation of an N-

alkylglycinamide, see: J. Kalvoda, M. Kessler, R. Lattmann, G.

Ramos, U.S. Pat. 5317100, 2011.

[52] We could not determine the ring size or elucidate the configu-

ration at the starred stereocenter of iodo ether 31 because it

was inseparable from the iodolactone 30.

[53] B. O. Lindgren, T. Nilsson, Acta Chem. Scand. 1973, 27, 888–

890.

[54] For the synthetic procedure, see: N. Hashimoto, T. Aoyama, T.

Shioiri, Chem. Pharm. Bull. 1981, 29, 1475–1478.

[55] a) Y. Oikawa, T. Yoshioka, O. Yonemitsu, Tetrahedron Lett.

1982, 23, 885–888; b) for a precedent for the deprotection of a

PMB ether in the presence of a benzaldehyde acetal, see: M.

Nilsonn, T. Norberg, Carbohydr. Res. 1988, 183, 71–82.

[56] This selectivity (ds = 85:15) did not reach the level (ds = 91:9)

of the model rearrangement 23dǞ syn^3,*,syn^2,3–26. Similarly,

we effected [2,3]-Wittig rearrangements with the phenyl and

dimethyl ketal analogues of diethyl ketal 37. These proceeded

with ds = 67:33 and 89:11, respectively.

[57] THF was added because the dicarboxylic acid 37 is insoluble

in pure benzene. Nonetheless, we do not consider THF an es-

sential ingredient of the reaction mixture because the model α-

alkoxycarboxylic acid 46 was defunctionalized in pure benzene

under the best conditions of a model study, which is summa-

rized in ref.^[58]

[58] The decarboxylative degradation of compound 37 was per-

formed under the best conditions, which we established for the

decarboxylative degradation of the model α-alkoxyacetic ester

46 (except that we needed to modify the solvent^[57]). That reac-

tion proceeded in 84% yield with Pb(OAc)₄(1.2 equiv.),

Cu(OAc)₂(0.1 equiv.), and visible-light irradiation (tungsten

lamp, 150 W) in pure benzene at room temp. during 15 min.

The concomitant use of catalytic Cu^IIand light prevented the

formation of byproducts.

[64] For the synthesis and full characterization of the mirror image

of 12 (¹H and ¹³C NMR, IR, elemental analysis), see ref.^[30a]

[65] For the reaction conditions for the preparation of the mirror

image of diol syn,cis-14 and full characterization (¹H and ¹³C

NMR, IR, elemental analysis), see ref.^[31]

[66] The ee of the epoxide syn,cis-14 was determined by chiral

HPLC [Chiralpak AD-H column; n-heptane/EtOH, 70:30 (v/

v), flow-rate 1.0 mLmin^–1; 15 °C isothermal; λ_{UV detector}

=

227 nm; syn,cis-14: t_R

= 21.0 min; ent-syn,cis-14: t_R

17.6 min].

[67] a) T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc. 1980, 102,

5974–5976; b) R. M. Hanson, K. B. Sharpless, J. Org. Chem.

1986, 51, 1922–1925; c) Y. Gao, R. M. Hanson, J. M. Klunder,

S. Y. Ko, H. Masamune, J. Am. Chem. Soc. 1987, 109, 5765–

5780.

[68] For reviews, see: a) A. Pfenniger, Synthesis 1986, 89–116; b) T.

Katsuki, V. S. Martin, Org. React. 1996, 48, 1–299; c) T. Kat-

suki, in: Comprehensive Asymmetric Catalysis (Eds: A. Pfaltz,

E. N. Jacobsen, H. Yamamoto), Springer, Berlin, 1999, p. 621–

648; d) J. A. Johnson, K. B. Sharpless, in: Catalytic Asymmetric

Synthesis (Ed.: I. Ojima), 2nd ed., Wiley-VCH, Weinheim, Ger-

many, 2000, p. 231–286.

[69] For the synthesis, ¹H and ¹³C NMR spectra, IR spectrum, and

elemental analysis, see ref.^[32]

[70] This reductant was introduced by: W. Boland, N. Schroer, C.

Sieler, Helv. Chim. Acta 1987, 70, 1025–1040.

[71] Note added in proof (August 23, 2013): Having learnt to pro-

ceed from divinylcarbinol cis,cis-17 to the diol syn,cis-14 on the

multi-gram scale reproducibly, we did not study whether the

catalytic asymmetric desymmetrization conditions for other di-

vinylcarbinols (Z. Li, W. Zhang, H. Yamamoto, Angew. Chem.

2008, 120, 7630–7632; Angew. Chem. Int. Ed. 2008, 47, 7520–

7522) apply to the anti-selective desymmetrization cis,cis-

17Ǟanti,cis-16, too.

[59] Aliphatic carboxylic acids without a heteroatom α to the CO₂H

group are also decarboxylated by Pb(OAc)₄and Cu(OAc)₂in

benzene solutions: either at 30 °C if 350Ϯ50 nm light is shone

into the reactant solution or at 80 °C in the dark (J. D. Bacha,

J. K. Kochi, Tetrahedron 1968, 24, 2215–2226). Such decarbox-

ylation reactions take a different course than the oxidation

Received: February 2, 2013

Published Online: September 17, 2013

6562

www.eurjoc.org

Eur. J. Org. Chem. 2013, 6545–6562

Article Doi

Synthesis of a Cn-Cn+6 building block common to important polyol,polyene antibiotics from a divinylcarbinol by a desymmetrizing sharpless epoxidation

DOI: 10.1002/ejoc.201300181

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejoc.201300181

Full text of DOI:10.1002/ejoc.201300181

Products guided by the article

R&D Labs maybe for 103773-79-1

Relevant to this article

Hot Product