Enzymatically asymmetrised chiral building blocks for the synthesis of complex natural product analogues: The synthesis of dynemicin analogues from 2-(quinolin-4-yl)propane-1, 3-diol

FULL PAPER

DOI: 10.1002/ejoc.200901478

Enzymatically Asymmetrised Chiral Building Blocks for the Synthesis of

Complex Natural Product Analogues: The Synthesis of Dynemicin Analogues

from 2-(Quinolin-4-yl)propane-1,3-diol

Renata Riva,*^[a]Luca Banfi,^[a]Andrea Basso,^[a]Valentina Gandolfo,^[a]and

Giuseppe Guanti^[a]

Keywords: Antitumor agents / Diastereoselectivity / Nitrogen heterocycles / Alkynes / Enediynes

Full details of our synthetic approaches directed towards the

enantioselective synthesis of new dynemicin analogues each

containing a side-arm (a “handle”) incorporating a protected

alcohol are reported.

Introduction

Natural enediyne antibiotics are a small family of cyclic

derivatives endowed with very interesting biological activi-

ties.^[1–4]Their highly individual mode of action is due to

Bergmann cycloaromatisation, generating a diradical that

in turn brings about the cleavage of two complementary

DNA strands. The most important enediynes are character-

ised by 10-membered 3-ene-1,5-diyne cyclic structures em-

bedded in complex bridged polycyclic systems. They incor-

porate appropriate stabilizing moieties (“safety-locks”),

which prevent the Bergmann cycloaromatisation occurring,

and which are removed in vivo by appropriate “triggering”

events. The most renowned and potent enediyne, cali-

cheamycin γ₁^I, is currently in clinical use, under the brand

name mylotarg®, as its conjugate with a humanised anti-

CD33 antibody.

Scheme 1.

Other important natural products containing 10-mem-

bered 3-ene-1,5-diyne systems are dynemicin A^[5]and un-

Dynemicin and uncialamycin, each lacking an oligosac-

cialamycin^[6,7](Scheme 1), which share several common fea- charide unit, are structurally simpler than chalicheamicin

tures. They are each characterised by the presence of an and for this reason have been deemed particularly well

anthraquinone structure, a DNA-intercalating system typi- suited for the development of simplified analogues. First

cal of other very important anticancer agents. The stabiliz- Nicolaou,^[8–11]and then many others,^[12–19]including our

ing element (“safety-lock") is represented here by the epox- group,^[20,21]have designed various derivatives correspond-

ide, whereas the “trigger” is the quinone. Its bioreduction ing to the general structure 1 (Scheme 1), which are often

initiates a cascade of events culminating in the hydrolytic endowed with potent DNA-cleaving and/or cytotoxic activi-

opening of the epoxide, followed by facile Bergmann cyclo- ties. However, deletion of the naphthoquinone fragment

aromatisation, resulting in single and double cleavage of the calls for the development of a different type of trigger.^[22]

DNA strands.

The observation that the epoxide is quickly hydrolysed

when the tetrahydroquinoline nitrogen is free makes their

urethane derivatives ideal stabilizing elements (“safety-

locks”). The R¹group must be suitably designed in order

to allow cleavage of the carbamate under a variety of con-

trolled conditions. Previous work has also shown that maxi-

mum activity is obtained when one of R²and R³is hydro-

gen and neither is an OH or OR group, as in the case of

[a] Department of Chemistry and Industrial Chemistry,

University of Genova,

Via Dodecaneso 31, 16146 Genova, Italy

Fax: +39-010-3536118

E-mail: riva@chimica.unige.it

Supporting information for this article is available on the

WWW under http://dx.doi.org/10.1002/ejoc.200901478.

2768

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

dynemicin, but in contrast to uncialamycin. This presents

The success of this strategy, with the obtainment of a

some limitations in the choice of the synthetic strategies, compound corresponding to formula 4, had been already

excluding – for example – cyclisation of the ten-membered published in preliminary form.^[38]Now we report the full

ring through an intramolecular acetylide addition to a car- description of our efforts, including the (unsuccessful) ap-

bonyl group.

proach towards compounds 3 and 5 and the exploration of

In a previous paper we reported a novel synthetic ap- alternative routes for adducts 4. A thorough survey of the

proach to the synthesis of both diastereomers of com- influence of the protecting groups on the diastereoselectiv-

pounds 2,^[20]in which the two structural criteria quoted ity of the key acetylide addition to the quinoline nucleus is

above are satisfied. One of the two diastereomers, after con- also provided.

version into a β-sulfonylethyl carbamate, showed a very

We first studied the synthesis of compounds 3, each con-

taining a side arm with just one carbon atom (Scheme 3).

promising DNA-cleaving activity.^[21]

However, that synthesis gave a racemic mixture. Al- To that end we started from (S)-18, the efficient and enan-

though in most cases dynemicin analogues have been pre- tioselective synthesis of which (in 97% ee) had been already

pared as racemates, in two previous cases the synthesis of reported by us.^[39]The monoacetate (S)-18 was converted

enantiomerically pure compounds has been reported.^[23,24]into the three orthogonally diprotected 2-(quinolin-4-yl)-

Biological tests have indicated that the absolute configura- propane-1,3-diols 19–21 (Scheme 3). In a previous work^[40]

tions have an important influence on activity.

we had already described the yields and diastereoselectivi-

Moreover, because of the lack of suitable attaching ties of protecting-group-controlled Yamaguchi^[41,42]ad-

points (“handles”), it was not possible to modulate the bio- ditions of magnesium trimethylsilylacetylide to these deriva-

logical activity by joining DNA-complexing moieties as sur- tives. The best results were obtained with compound 19,

rogates of the missing anthraquinone intercalating sub- containing the triphenylmethyl (Tr) and acetyl groups,

structure. We^[25]and others^[1]have demonstrated that conju- which gave a 94% yield with a 70:30 diastereomeric ratio.

gation of simplified enediynes with DNA-intercalating In that work we also found that it was not possible to use

agents or minor groove binders can enhance their DNA- a quinoline bearing a free hydroxy group on one of the two

cleaving efficiencies.

side-arms, because of extensive decomposition during the

In order to overcome these limitations, an enantioselec- addition reaction.^[40]

tive synthesis of more complex analogues of 2, possessing

The synthesis was therefore continued with the major ad-

suitable attaching points, was designed. They are repre- duct 22a, after its chromatographic separation from 22b.

sented by general structures 3, 4 or 5 (Scheme 2).

The first task was the selective monodeprotection of one of

the two synthetically equivalent arms. The most obvious

approach involves removal of the triphenylmethyl group. To

our surprise, however, the usual conditions (pTSA in

MeOH) furnished 25a in only 28% yield, the main product

Results and Discussion

The retrosynthetic analysis is shown in Scheme 2, and being the diol 26 (61% yield). On switching to iPrOH the

entails, as key steps, the enzymatic asymmetrisation of pro- selectivity was even lower, the amount of diol increasing to

chiral 2-(quinolin-4-yl)propane-1,3-diol to give the mono- 68%, whereas in tBuOH no reaction occurred. Solvolysis

acetate 18,^[26–35]the diastereoselective addition of trimethyl- of an acetyl group under such mild conditions is rather un-

silyl acetylide to the quinolines 15–17, the homologation common, but intramolecular assistance from the other OH

of the aldehydes 9–11 to terminal alkynes, and finally the group is probably highly influential here. Milder protic ac-

Danishefsky cyclisation^[36]of the diiodoalkynes 6–8 under ids did not give better results, but we eventually succeeded

Stille conditions. This strategy would in principle allow the in obtaining 25a in 80% yield by use of zinc bromide in

preparation of all four possible stereoisomers with respect CH₂Cl₂/iPrOH (85:15).^[43]Because 50 equiv. of ZnBr₂were

to C-2 and to the stereogenic centre that bears the side required in order to drive the reaction to completion (owing

arm,^[37]and so stereochemical notations are deliberately to the existence of an equilibrium between the monoalcohol

omitted in Scheme 2.

and the trityl ether), however, the reaction was not very

Scheme 2.

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2769

R. Riva, L. Banfi, A. Basso, V. Gandolfo, G. Guanti

FULL PAPER

Scheme 3.

practical. Switching to a more labile trityl analogue [p- (iPr)₂] at –78 °C. The workup conditions were also very im-

methoxyphenyl diphenylmethyl (MMTR)] was also not par- portant. A quick acid wash, in order to remove excess

ticularly successful, because of poor yields in the Yamagu- amine prior to concentration to dryness, was essential.

chi addition leading to 24a^[40]and of poor selectivity in its These optimised conditions were always used by us

monodeprotection.

throughout the work described in this paper. The aldehyde

We therefore tried to remove the other protecting group 30a was found to be completely stable to epimerisation and

in 22a – the acetyl group – selectively. Conventional saponi- could even be chromatographed without problems. Under

fication with alkaline hydroxides was not possible because the same oxidation conditions the aldehyde 31a gave no epi-

of concurrent removal of the trimethylsilyl group from the merisation, but a 12% yield of the elimination product 36

alkyne, which was in turn detrimental for the subsequent was formed. Finally, the aldehyde 32a represented the most

Corey–Fuchs protocol. We thus tried enzymatic hydrolysis critical case. The crude product consisted of a mixture of

with a series of lipases. None of them accepted 22a as a 32a, its epimer 32b and the elimination product 36 in a

substrate, with the exception of CAL (Candida antarctica 62:18:20 ratio.

lipase). However, even at 60 °C and after long reaction

times, the reaction failed to reach completion, and the iso- eration of the necessary triple bond, was attempted first on

lated yields of 27a were never higher than 61%. the most stable aldehyde: compound 30a. However, we

The subsequent Corey–Fuchs protocol,^[45,46]for the gen-

These unsatisfactory results prompted us to prepare an- never succeeded in obtaining the desired dibromide 33a. As

other monoprotected derivative, compound 27a, by starting discussed in depth in the next section, this is probably the

from the Yamaguchi adduct 23a, notwithstanding the most crucial step of the whole synthetic sequence, because

slightly lower diastereoselectivity observed in its prepara- of the lability of the starting aldehydes, which are prone to

tion (60:40 instead of 70:30).^[40]The triethylsilyl group epimerisation, elimination and rearomatisation processes.

could be indeed selectively removed with HF at –18 °C in In particular, in the case of 30a, unlike in our previous

nearly quantitative yield to afford 27a.

work,^[21]the reaction did not take place at low temperature,

We also prepared the third alcohol 29a, starting from whereas on increasing the temperature to room temp. exten-

27a, through reprotection of the free hydroxy group as the sive decomposition took place, affording mainly the elimi-

methoxymethyl ether (MOM) in 86% yield [MOM-Cl, nation product 36. Reasoning that the steric bulk of the

EtN(iPr)₂, CH₂Cl₂, room temp.] followed by trityl cleavage trityl group could be responsible for this behaviour, we at-

(0.1  HCl in MeOH, room temp., 68%).^[44]This time we tempted the same protocol with the aldehydes 31a and 32a,

did not observe any formation of diol 26.

but with similar results, showing that the problem is more

Oxidation of these alcohols to the corresponding alde- electronic than steric: the β-alkoxy groups make these alde-

hydes is a delicate step, because of possible epimerisation. hydes less reactive and particularly sensitive to elimination/

The best conditions were found to be the use of a modified epimerisation reactions. Alternative methods^[47,48]for the syn-

Swern oxidation in the presence of Hünig’s base [EtN- thesis of terminal alkynes from these aldehydes also failed.

2770

www.eurjoc.org

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

access to two classes of simplified dynemicin analogues with

side-arms of different lengths.

Scheme 5 shows the preparation of both enantiomers of

the key silyl ether 41. The potential to generate both

enantiomers from a starting common intermediate is a typi-

cal property of asymmetrised 2-substituted propane-1,3-di-

ols, stemming from their latent C_Ssymmetry. Homologa-

tion of one of the two side-arms was achieved by S_N2 sub-

stitution of a tosylate by cyanide ion. The two-step protocol

(tosylation and substitution) worked without any problems

with silyl ether 40 to afford (S)-41 in high yields.^[40]The

reaction of the tosylate derived from the monoacetate 18,

on the other hand, was more tricky, being complicated by

a concurrent elimination reaction, especially when working

at 60 °C. Fortunately, when operating at room temp. this

side reaction was nearly suppressed. The nitrile (R)-42 was

then converted into (R)-41 by a high-yield protecting group

interchange. The enantiomeric purities of (S)- or (R)-43

were determined through formation of Mosher’s esters and

Scheme 4.

These unsatisfactory attempts prompted us to concen-

trate on the synthesis of compounds 4 and 5, each charac-

terised by a longer oxygenated side-arm (Scheme 2). We

reasoned that the alkene 37 could be an useful advanced

intermediate for both (Scheme 4). Compound 37 might in

principle be transformed into the alcohol 38, by hydrobor-

1

their examination by H NMR (ee 96% for both).

The synthesis was continued with (S)-41. Reduction with

ation/oxidation, or be degraded into alcohol 39 through DIBALH gave the aldehyde (S)-44, which was methyl-

oxidative cleavage of the double bond followed by re- enated to give (S)-45. This last reaction was more trouble-

duction. With a branched route we might therefore have some than expected. Although various alternative literature

Scheme 5. Reagents and conditions: a) TBDMSCl, imidazole, DMF, room temp.; b) KOH, MeOH, 0 °C; c) TsCl, pyridine, room temp.;

d) KCN, nBu₄NI, DMSO, 60 °C; e) KCN, nBu₄NI, DMSO, room temp.; f) MeONa (0.17 ), MeOH/THF, 0 °C, 80 min; g) HF, CH₃CN/

H₂O, 0 °C; h) DIBALH, –70 °C; i) [Ph₃PMe]Br, NaNH₂, THF, –78 °CǞ0 °C.

Scheme 6. Reagents and conditions: a) PhOCOCl, Me₃SiCϵCMgBr, THF, –78 °C; b) HF, CH₃CN/H₂O, 0 °C; c) (COCl)₂, DMSO,

EtN(iPr)₂, CH₂Cl₂, –78 °C; d) CBr₄, PPh₃, CH₂Cl₂, –78 °CǞ–40 °C; e) nBuLi, THF, –78 °C, 15 min.

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2771

R. Riva, L. Banfi, A. Basso, V. Gandolfo, G. Guanti

FULL PAPER

methods^[49]failed, we obtained a good yield by the “instant

ylide” method.^[50,51]Yamaguchi addition to the quinoline

(S)-45 (Scheme 6) afforded rather low stereoselection in re-

lation to previously studied chiral quinolines.^[40]Moreover,

the two diastereomers could be efficiently separated only

after removal of the silyl ether, on compounds 46a and 46b.

The stage was set for another crucial transformation: oxi-

dation to the aldehyde and conversion into a terminal al-

kyne by the Corey–Fuchs methodology, which was thor-

oughly explored with the major alcohol 46a. Whereas oxi-

dation under the modified Swern conditions described

above worked this time uneventfully without any epimer-

isation, treatment with CBr₄and PPh₃turned out to be

troublesome. The standard Corey–Fuchs conditions^[45]in-

volve pre-treatment of CBr₄with PPh₃(2 equiv.) to give

a dibromomethylphosphorane, followed by addition of the

aldehyde at 0 °C or room temp. Under these conditions,

however, no desired product 47a was obtained and the

starting material was completely destroyed. The addition of

buffering agents^[46]such as Et₃N or 2,6-lutidine, as well as

Scheme 7. Possible mechanism of the Corey–Fuchs reaction.

the use of zinc,^[52]did not change the situation. If the pre- would generate the ionic pair 54. Moreover, 51 could derive

formed phosphorane was cooled to –78 °C and treated with from 54 by phosphorus-centred nucleophilic substitution of

the aldehyde, no reaction took place until the temperature bromine by Br₃C^–. It should be noted that ionic pairs such

was raised: only at –20 °C did the aldehyde start to react, as 54 are universally considered the first products of inter-

but the desired product was also formed in low yield in this action between PPh₃and CBr₄or CCl₄and therefore the

case, with extensive decomposition.

key intermediates of the Appel reaction.^[67]However, these

On the other hand, rapid addition of PPh₃and CBr₄to two alternative reactions are probably reversible. In the ab-

the cooled (–78 °C) aldehyde solution brought about the sence of an aldehyde (or of a proton source as in the Appel

rapid, but incomplete, formation of the dibromide 47. The reaction), the equilibrium between 54 and 51 would soon

reaction was poorly reproducible and in all cases tended to be shifted towards the phosphorane 52 and the dibromide

stop at various degrees of conversion: warming to –40 °C 53 by the irreversible reaction of 51 with a second molecule

did not increase conversion, whereas warming to –20 °C or of PPh₃. The resulting phosphorane 52 would then react

higher temperatures brought about decomposition not only with an aldehyde through a “slow” Wittig mechanism to

of the starting material, but also of the product itself.

afford the final product (alkenyl dibromide). On the other

To the best of our knowledge, apart from a pioneering hand, if an aldehyde were present from the beginning, the

work by McKelvie,^[53]no in-depth study on the mechanism ionic pair 54 could react irreversibly with it to afford the

of this useful transformation has been carried out. A litera- intermediate 55, which upon intervention of a second PPh₃

ture search showed that this reaction was in most cases car- molecule would give the alkenyl dibromide, triphenylphos-

ried out at 0 °C or room temp. with reaction times from phane oxide and the dibromo triphenylphosphane 53.

30 min to 22 h,^[54–63]whereas in other cases the same reac-

The existence of these alternative mechanisms is not usu-

tion was reported to take place in less than one hour at ally an issue, because the “slow” path is convenient in most

–78 °C.^[64,65]This suggested two alternative pathways: one instances. In our case, however, the high lability of the dihy-

fast, taking place even at –78 °C, and another one slow, droquinoline system in the presence of an electrophilic rea-

going through the phosphorane 52 (Scheme 7). Two papers gent such as dibromotriphenylphosphane 53 makes only the

seemed particularly interesting to us: Bestmann reported “fast” mechanism productive.

that better yields could be obtained by addition of CBr₄to

Being confident that the scenario shown in Scheme 7 was

the mixture of aldehyde and PPh₃,^[66]whereas Weinreb, in correct, we reasoned that the “fast” mechanism would have

his total synthesis of licoricidin,^[64]faced a situation similar been favoured by low temperature and by the presence of

to ours – under the classical conditions (high temperature) the aldehyde from the beginning, but also by a lower con-

only aldehyde decomposition was observed, whereas, upon centration of PPh₃. Actually, it is PPh₃that promotes the

mixing all three reagents at –78 °C, a very fast reaction first irreversible step of path A, whereas in path B the sec-

(only 5 min) gave the expected product in good yields. The ond molecule of PPh₃is involved only at a later stage.

same result could be also achieved by using a mixture of

We thus treated the aldehyde at –78 °C with CBr₄

CHBr₃and KOtBu in the presence of PPh₃.

(2 equiv.) and then slowly added a solution of PPh₃

This latter evidence led us to propose the scenario shown (3 equiv.) in dropwise fashion. The temperature was then

in Scheme 7. Triphenylphosphane can react with CBr₄in allowed to rise to –40 °C (but not more!). We were pleased

two different ways: nucleophilic attack at the carbon atom to find that under these conditions a good and reproducible

would give the intermediate 51, whereas attack at a bromine yield of dibromides 47a and 47b could be obtained. We

2772

www.eurjoc.org

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

think that these conditions may be useful in all those cases tion of the stereoselectivity, they should have different elec-

in which the traditional methodology fails.

tronic and steric properties. In order to evaluate these ef-

The next elimination to afford the terminal alkynes 48a fects, and also to select the best combination of protecting

and 48b (Scheme 6) proceeded well, provided that exactly groups, we prepared the three different diprotected systems

2 equiv. of nBuLi were used at –78 °C, that the reaction was 59, 61 and 64 (Scheme 8). The nitrile (S)-57 was obtained

carried out under argon, and that it was quenched after 5– from the monoacetate 18 through conversion into (R)-56,^[40]

15 min. The presence of an excess of base or use of longer and subsequent homologation of the propane-1,3-diol sys-

reaction times led to deprotonation of the relatively acidic tem by the protocol described above (Scheme 5; replace-

proton at the C-2 position in the dihydroquinoline, resulting ment of a tosylate by cyanide). The synthesis of the related

in fast aromatisation.

nitriles (S)- and (R)-41 has already been shown in

With dienediynes 48a and 48b to hand we studied the Scheme 5, whereas (R)-62 was obtained from (R)-43 by pro-

hydroboration reaction, hoping to form the desired alcohols tection of the alcohol as the p-methoxybenzyl ether (PMB).

49. We performed several experiments with the major iso-

In principle, the cyano groups in the nitriles 57, 41 and

mer 48a, but unfortunately, and surprisingly, we did not 62 could be converted into the corresponding alcohols

succeed in obtaining the desired product. All attempts to either before or after the Yamaguchi reaction. However,

effect ozonolysis of the double bond also failed. On the preliminary experiments have shown that addition of tri-

other hand, a sequence of dihydroxylation of 48a with methylsilyl acetylide onto the nitriles 57 or 41 results in low

OsO₄, followed by NaIO₄treatment, led to the alcohol 50a, yields. Direct reduction of the nitriles 57, 41 and 62 to the

with concurrent desilylation of the alkyne. The overall yield, primary alcohols is not possible. We therefore employed a

however, was very low. Because we later succeeded in ob- two-step procedure, first reducing the nitriles to aldehydes

taining the same alcohol 50a through a more efficient route with DIBALH and then completing the reduction with

(see below), this approach was abandoned.

NaBH₄. Isolation of the intermediate aldehydes was essen-

The “allylic” route was mainly devised for the synthesis tial, because NaBH₄treatment of the crude products gave

of the alcohols 49, with longer arms, but the unexpected only poor yields, probably because the intermediate alumin-

failure of all hydroboration attempts frustrated this design. ium imines were hydrolysed only slowly in methanol. Fi-

However, the lessons learned during this approach, espe- nally, introduction of the second protecting groups gave

cially those relating to the Corey–Fuchs reaction, turned compounds 59, 61 and 64. In the case of 61 both enantio-

out to be crucial for the studies described next.

mers were prepared, thanks to the fact that both starting

The obvious alternative synthesis of compounds 4 in- materials – (S)- or (R)-41 – had been synthesised as de-

volves a Yamaguchi trimethylsilylacetylide addition onto a scribed in Scheme 5. The stereodivergency of the starting

quinoline of general formula 16 (Scheme 2), with two side monoacetate 18 should in principle also allow both enantio-

arms of different lengths (one and two carbon atoms), each mers of 59 and 64 to be generated, although this concept

bearing a protected alcohol. The choice of protecting has not been yet implemented.

groups was not trivial. They must be compatible with the

The results of diastereoselective Yamaguchi additions to

Yamaguchi addition, and the one on the shorter arm must 59, 61 and 64 are reported in Table 1. For comparison, and

be selectively removable in the presence of the other. In ad- in order to give a comprehensive picture, we also report the

dition, the one on the longer arm must be compatible with results of addition to the allyl derivative 45, as well as other

the Corey–Fuchs protocol and be removable smoothly at representative data previously collected and published by

the end of the synthesis. Finally, in order to allow modula- us.^[40]It is worth noting that the diastereoselection in these

Scheme 8. Reagents and conditions: a) TsCl, Et₃N, CH₂Cl₂, room temp.; b) KCN, nBu₄NI, DMSO, 100 °C; c) DIBALH, toluene/CH₂Cl₂,

–70 °C; d) NaBH₄, MeOH, –40 °CǞ–10 °C; e) TES-OTf, 2,6-lutidine, CH₂Cl₂, 0 °C; f) Ac₂O, pyridine/CH₂Cl₂, room temp.; g) NaH,

p(MeO)C₆H₄CH₂Cl, DMF, 0 °C; h) TBDMS-Cl, imidazole, DMF, room temp.

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2773

R. Riva, L. Banfi, A. Basso, V. Gandolfo, G. Guanti

Table 1. Protecting-group-controlled diastereoselective Yamaguchi reactions with various quinolines.^[a]

FULL PAPER

Entry

Starting quinoline

Addition products

R¹

R²

dr

% Yield

1

2

3

4

5

6

7

8

9

59

65a, 65b

66a, 66b

67a, 67b

46a, 46b

22a, 22b

23a, 23b

24a, 24b

see ref.^[40]

CH₂OTES

CH₂OAc

CH₂OTBDMS

CH=CH₂

OAc

OTES

OAc

Tr

82:18

74:26

61:39

70:30

60:40

70:30

84:16

68:32

79

95

93

77

94

87

44

98

87

61

TBDMS

PMB

TBDMS

Tr

MMTr

Tr

64

45

19

20

21

see ref.^[40]

H

TBDMS

[a] For the sake of clarity all additions are listed as if carried out on the enantiomer of starting quinoline indicated in the figure, although

some were actually carried out either on its enantiomer or on the racemic mixture.

additions represents long-range 1,4 asymmetric induction.

This purely steric model cannot fully explain our present

In addition, the starting stereogenic centre is located out- results, however. If we compare Entry 2 with Entry 9 or En-

side the ring, in a conformationally mobile appendage. Fi- try 2 with Entry 4, it is hard to explain the higher selectivity

nally, the two substituents R¹CH₂and R²OCH₂are differ- observed in Entries 2–9 by assuming that a methyl or an

entiated only two or three atoms away from the stereogenic allyl group is larger than a CH₂OAc group. The acetoxy

centre. The obtainment of diastereomeric ratios of around group therefore probably plays an active role, perhaps

5:1, as with compounds 59 and 61, must thus be considered through coordination of magnesium, favouring approach of

an exceptionally good result. A similar dr had previously the nucleophile from the same side of the ring. This assist-

been obtained only with the unsubstituted compounds (R¹ance seems to be operating only when the acetoxy group is

= H), and only when the oxygenated arm was protected as on the longer arm, as suggested by the lower dr obtained

the bulky trityl ether (see Entries 8 and 9). In a previous in the case of compound 19, notwithstanding the greater

paper^[40]we have proposed a model to explain the observed bulkiness of a trityl group in relation to a TBDMS

outcome (Figure 1). According to this model, the benzylic group.

hydrogen is directed toward the peri hydrogen, slightly tilted

away (30°) from the aromatic plane. Of the other two sub-

stituents, the larger (L) is orthogonal to the aromatic ring,

whereas the smaller (M) is on the other side of the plane,

but with a narrower angle (about 30° as the benzylic hydro-

gen). Obviously the nucleophile prefers to attack from the

face opposite to the “large” group.

With regard to the assessment of relative configuration,

compounds 65a, 65b, 66a, 66b, 67a and 67b were first of all

internally correlated through the transformations depicted

below in Schemes 9 and 10. The products 46a and 46b were

similarly chemically correlated to 66a and 66b through

transformation into the common intermediate 50a

(Scheme 6). Finally, the relative configurations of com-

pounds 22–24a and 24b had already been demonstrated.^[40]

We therefore had only to correlate 46 or 65–67 with 22–24,

which was achieved by finding a series of clear similarities

1

in their H NMR spectra. In particular we compared 22–

24 with compounds 46, 66 and 67 (in the case of 65a and

65b the two diastereoisomers could not be separated at this

level). These characteristic NMR features may again be ex-

plained by considering a preferred conformation with the

benzylic hydrogen directed towards the peri hydrogen, with

the C–H bond forming an angle of about 30° with the ring

plane (see Figure 1). In the addition products, it is reason-

able to assume that the orthogonal position (with respect

to the aromatic ring) is occupied by the group opposite to

the alkyne (the “large” group in isomer a and the “me-

dium” one in isomer b). This group should experience a

Figure 1. Model for explaining the diastereoselectivity.

2774

www.eurjoc.org

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

Scheme 9. Reagents and conditions: a) HF, CH₃CN/H₂O, –18 °C; b) pTSA, MeOH, room temp.; c) Candida antarctica lipase, vinyl acetate,

room temp.; d) HF, CH₃CN/H₂O, 0 °C; e) (COCl)₂, DMSO, EtN(iPr)₂, CH₂Cl₂, –78 °C; f) CBr₄, PPh₃, CH₂Cl₂, –78 °CǞ–40 °C;

g) nBuLi, THF, –78 °C; h) K₂CO₃, MeOH, 0 °C; i) TBDMS-Cl, imidazole, DMF, room temp.; j) mCPBA, CH₂Cl₂, room temp.;

k) I₂·morpholine, benzene; l) (Z)-Me₃Sn–CH=CH–SnMe₃, LiCl, Pd(PPh₃)₄, DMF (0.01 ), 70 °C.

shielding anisotropic effect by the aromatic ring and its pro-

tons should resonate upfield. This feature is common to all

the Yamaguchi adducts prepared previously with relative

configurations that were established by chemical corre-

lation, and is also perfectly reproduced in the new com-

pounds reported here. In the major isomers 46a, 66a and

67a the group opposite to the alkyne is the CH₂OR moiety,

and the CH₂OR protons therefore resonate, as multiplets,

at δ = 3.51–3.66, 3.51–3.68 and 3.35–3.55 ppm, whereas in

the minor isomers b the same protons fall at δ = 3.72–3.87,

3.78–3.88 and 3.58–3.75 ppm. On the other hand, in the

minor isomers 46b, 66b and 67b the group opposite to the

alkyne is the CH₂CH₂OR arm. The CH₂nearest to the

branching point in each case falls at δ = 2.20–2.47, 1.88–

2.10 and 1.65–1.98 ppm, whereas in the major isomers a the

same protons resonate at δ = 2.38–2.62, 1.95–2.21 and 1.85–

2.08 ppm. The relative configuration was further corrobo-

rated by the NMR spectrum of the final compound 74a, in

which the chemical shift of the proton indicated by an ar-

row (Scheme 9) is highly diagnostic.^[21]

In the cases of the adducts 66a, 66b, 67a and 67b the

pairs of diastereomers could be separated by chromatog-

raphy. For 65a and 65b, in contrast, separation was possible

only after removal of the TES group to give 68a and 68b.

In every case the synthesis was continued only on the major

diastereoisomer a.

Scheme 10. Reagents and conditions: a) DDQ, CH₂Cl₂/H₂O, room

For our purposes, the protecting group residing on the

temp.; b) (COCl)₂, DMSO, EtN(iPr)₂, CH₂Cl₂, –78 °C; c) CBr₄,

shorter arm should be removed first. In the case of 65a,

however, this was not possible, analogously with what had

PPh₃, CH₂Cl₂, –78 °CǞ–40 °C; d) nBuLi, THF, –78 °C; e) AgNO₃,

EtOH/H₂O, KCN, 0 °C; f) mCPBA, CH₂Cl₂, room temp.

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2775

R. Riva, L. Banfi, A. Basso, V. Gandolfo, G. Guanti

FULL PAPER

happened with 23a (see above). We therefore removed both pared the needed quinoline intermediate (R)-61 from (R)-

groups, confident of being able to acetylate the hydroxy 41 (Scheme 8). The syntheses of (S)-61 and of (R)-61 are

group on the longer arm selectively by enzymatic means. comparable: 49% and 37% overall yields, respectively

Candida antarctica lipase showed a moderate selectivity, (seven steps in both cases).

whereas other enzymes (e.g., Amano PS lipase) were found

For the preparation of the enantiomer of 74a we also

to be inert towards the diol. Although the moderate yield considered use of the PMB-protected quinoline (R)-64. Its

could probably be optimised, we found out that the same preparation from (S)-18 again involves seven steps and a

acetate 69a could be obtained more efficiently in just one 38% overall yield. However, its transformation into the fi-

step from the adduct 66a by desilylation. The synthesis of nal product ent-74a was expected to be more step-economi-

69a through the quinoline 61 and the Yamaguchi adduct cal (only nine steps instead of 10) and atom-economical,

66a is more convenient both in terms of step economy and thanks to the fact that the final TBDMS protection is in-

of the higher overall yields obtained (see also Table 1, En- stalled at the correct place from the beginning. Moreover,

try 1). On the other hand, the stereoselectivity is identical.

Compound 69a was converted in good overall yields into of the obtainment of the minor diastereoisomer ent-74b.

the alcohol (2R,3ЈR)-50a, taking advantage of the previous The transformation of the major Yamaguchi adduct 67a

the slightly lower diastereoselectivity could be useful in view

experience gained in the crucial Corey–Fuchs reaction and into the common intermediate 71a is shown in Scheme 10.

the subsequent nBuLi-promoted elimination. During this After oxidative removal of the PMB group, the usual Co-

latter reaction, we also observed the formation of variable rey–Fuchs protocol furnished compound 77a. Desilylation

amounts of the corresponding alcohol originating from at- was carried out this time with AgNO₃, because we were

tack of nBuLi at the acetyl group. We therefore preferred concerned about the stability of the TBDMS group in the

to treat the crude product directly with K₂CO₃in methanol presence of K₂CO₃. Compound (2S,3ЈS)-71a was obtained

to bring about complete ester solvolysis and simultaneous in good overall yield from 67a and in 25% overall yield

removal of the Me₃Si group. It is worth noting that these (six steps) from (R)-64 [for comparison, (2R,3ЈR)-71a was

quite mild conditions for alkyne desilylation are not general prepared from (S)-61 in 24% yield over seven steps]. When

for trimethylsilylalkynes,^[68]being a particular feature of we measured the [α]_Dof this intermediate, however, we

these systems.^[21]The enantiomeric excess of alcohol found a value of –108.8, whereas the previously obtained

(2R,2ЈR)-50a was determined by NMR analysis of its Mo- (2R,3ЈR)-71a had +244.3. In order to check this further, we

sher’s ester, which indicated an ee of about 94%, similar to also converted (2S,3ЈS)-71a into ent-72a, but again the [α]_D

that of the starting monoacetate 18. After reinstallation of was only –59.7 (72a had +129.2). From these values, and

the TBDMS group, a perfectly suited protecting group for from the measured ee of 94% for (2R,3ЈR)-50a, the optical

the handle, the synthesis was completed in three efficient purities of (2S,3ЈS)-71a and ent-72a would be only about

steps: epoxidation, introduction of the two alkynylic iodides 43%.

by use of iodine-morpholine complex,^[69]and finally Dani-

Evidently some racemisation had occurred during the

shefsky–Stille double cross-coupling^[36]with (Z)-bis-(tri- synthesis. Racemisation in the steps following the Yamagu-

methylstannyl)ethylene.^[70]For this last reaction we took ad- chi addition is very unlikely, because, with two stereogenic

vantage of the previously optimised reaction conditions,^[21]centres, epimerisation rather than racemisation would have

involving in particular the addition of anhydrous LiCl.^[71]

been observed. Moreover, the stereochemical integrity was

¹H NMR analysis of the enediyne 74a allowed confirm- checked at the level of the alcohol (R)-43. The critical step

ation of its relative configuration. In particular, it is well must therefore be either the introduction of the PMB pro-

known that in this diastereoisomeric series the CH indi- tecting group (which employs the strong base NaH) or the

cated by an arrow (Scheme 9) resonates at 3.0–3.1 ppm, be- nitrile reduction. From the [α]_Dvalues of compounds (R)-

cause it falls in the shielding cone of the aromatic ring, 62 (–40.3) and (R)-63 (–20.4), we feel that the critical step

whereas in the epimers it resonates at about 3.9–4.0 ppm.^[21]is the second one. Therefore, although this alternative com-

In 74a the chemical shift is indeed 3.08 ppm. Moreover, the bination of protecting groups has been demonstrated to

sign of optical rotation of 74a (+394.2) is the same as that work from the chemical point of view, partial loss of stereo-

of natural dynemicin and also of the simplified dynemicin chemical integrity is an important drawback.

analogues with the same absolute configuration.^[23,24,72]

For now the best route for both enantiomers of 74a re-

The preparation of 74a represents the first asymmetric mains that with use of TBDMS on the shorter arm and Ac

synthesis of a simplified dynemicin analogue. It is worth on the longer one. The good stereoselection, however,

noting that, to the best of our knowledge, enantiomerically makes this route not well suited for the synthesis of the two

pure simplified analogues 1 had been prepared in only two enantiomers of 74b. The data collected in Table 1 and the

cases.^[23,24]These methodologies did not involve true asym- interpretation of the stereochemical outcome might afford

metric synthesis, but the classical resolution of synthetic in- valuable suggestions on possible combinations of protecting

termediates. The synthesis reported here involves 17 steps groups that could lower, or even invert, the diastereoselec-

from monoacetate 18 with a remarkably good overall yield tivity. We should keep in mind, however, that the choice of

of 7.2%. Moreover, the synthetic route can be easily protection is not unrestricted: the experience gained from

adapted to the synthesis of the enantiomer, from the same all the efforts described here has shown that there are many

starting monoacetate (S)-18. We have indeed already pre- limitations, not only because of the need for orthogonality,

2776

www.eurjoc.org

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

to 0 °C, 22a (141 mg, 200 µmol) was added and the resulting yellow

solution was stirred at room temp. for 2 h. After quenching with

aqueous KH₂PO₄(0.5 ), the reaction was extracted with Et₂O,

followed by concentration and chromatography as above to give

25a (74 mg, 80%).

Compound 25a: R_f= 0.34 (PE/Et₂O 3:7). ¹H NMR (200 MHz,

CDCl₃): δ = 0.04 [s, 9 H, (CH₃)₃Si], 2.12 (s, 3 H, COCH₃), 3.33

(quint., J = 5.7 Hz, 1 H, 2Ј-H), 3.69 and 3.72 (AB part of an ABX

syst., J_AB= 11.2, J_AX= 6.0, J_BX= 4.4 Hz, 2 H, CH₂OH), 4.42 and

but also because introduction/removal of the blocking

groups must take account of the high susceptibilities of

compounds of the quinoline series to racemisation and of

those of the dihydroquinoline series to epimerisation or re-

aromatisation processes.

Conclusions

Despite the many problems that arose during this syn-

thetic study, we have succeeded in developing, in good over-

all yield, the first enantioselective synthesis of a simplified

dynemicin analogue not based on resolution methodolo-

gies.^[23,24]Moreover, the product obtained has a function-

alised side-arm potentially exploitable either for attachment

4.54 (AB part of an ABX syst., J_AB= 11.4, J_AX= 7.3, J_BX

=

5.5 Hz, 2 H, CH₂OAc), 5.94 and 6.03 (AB syst., J = 6.9 Hz, 1 H,

2-H and 3-H), 7.18–7.46 (m, 8 H), 7.77 (br. d, J = 7.0 Hz, 2 H, 5-

H or 8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –0.3 (CH₃Si),

21.0 (CH₃CO), 40.8 (C-2Ј), 44.9 (CHN), 61.9, 63.6 (CH₂OH and

CH₂OAc), 89.0, 100.8 (CϵC), 121.6 (ϫ2), 122.9, 125.1, 125.8,

of DNA-complexing moieties or for exploration of innov- 128.1, 129.4 (ϫ3, aromatic CH and C-3), 127.0, 132.7, 134.4, 150.9

(quat.), 151.9, 171.4 (C=O) ppm. IR: ν

= 3504, 3004, 2957,

˜

ative triggering devices.

max

2274, 1722, 1599, 1481, 1328, 1289, 1222, 1186, 1163, 1011,

842 cm^–1. GC-MS (standard conditions but with final temp.

290 °C): R_t= 12.22 min. MS: m/z (%) = 463 (6.3) [M]⁺, 386 (15),

347 (10), 346 (34), 252 (8.7), 236 (5.0), 226 (5.7), 192 (5.2), 180 (14),

151 (5.4), 117 (14), 94 (5.2), 77 (42), 75 (40), 73 (100), 65 (5.6), 59

(8.1), 51 (5.5), 45 (10), 43 (72). C₂₆H₂₉NO₅Si (463.60): calcd. C

67.36, H 6.31, N 3.02; found C 67.45, H 6.35, N 3.08.

Compound 26: R_f= 0.11 (PE/Et₂O 3:7). ¹H NMR (200 MHz,

CDCl₃): δ = 0.04 [s, 9 H, (CH₃)₃Si], 2.46 (br. s, 2 H, OH), 3.25

(centre of m, 1 H, 2Ј-H), 3.84 and 3.91 (AB part of an ABX syst.,

J_AB= 10.8, J_AX= 7.6, J_BX= 4.4 Hz, 2 H, CH₂OH), 4.15–3.98 (m,

2 H, CH₂OH), 5.92 and 5.94 (AB syst., J = 6.7 Hz, 2 H, 2-H and

3-H), 7.17–7.48 (m, 8 H), 7.75 (br. d, J = 7.0 Hz, 1 H, 5-H or 8-

H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = 0.3 (CH₃Si), 42.7 (C-

2Ј), 44.8 (CHN), 64.8, 64.9 (2 C, CH₂OH), 89.1, 101.5 (CϵC),

Experimental Section

NMR spectra were measured at room temp. in CDCl₃at 200 MHz

(¹H) or 50 MHz (¹³C), with TMS as internal standard for ¹H NMR

and the central peak of CDCl₃(at δ = 77.02 ppm) for ¹³C NMR

spectroscopy. Chemical shifts are reported in ppm (δ scale); cou-

pling constants are reported in Hertz. Peak assignments were made

with the aid of DEPT experiments. In ABX systems the proton A

is considered upfield and B downfield. GC-MS were carried out

with an HP-1 column (11.85 m long, 0.2 mm wide), electron impact

at 70 eV, and a mass temperature of about 170 °C. Only m/z Ͼ33

were detected. All analyses were performed (unless otherwise

stated) with a constant He flow of 0.9 mLmin^–1with an initial

temp. of 100 °C, init. time 2 min, rate 20 °Cmin^–1, final temp.

280 °C, inj. temp. 250 °C, det. temp. 280 °C. TLC analyses were 120.4, 121.6 (ϫ2), 123.0, 125.1, 125.9, 128.1, 129.4 (ϫ2), 129.6

carried out on silica gel plates and viewed under UV (254 nm) or

developed by dipping into a solution of (NH₄)₄MoO₄·4H₂O (21 g)

and Ce(SO₄)₂·4H₂O (1 g) in H₂SO₄(31 mL) and H₂O (469 mL)

and warming. R_fvalues were measured after elution of 7–9 cm.

Column chromatography was carried out with the “flash” method-

ology and 220–400 mesh silica. IR spectra were recorded as CHCl₃

solutions. Melting points are uncorrected. Petroleum ether (40–

60 °C) is abbreviated as PE. In extractive workup, aqueous solu-

tions were always reextracted thrice with the appropriate organic

solvent. Organic extracts were always dried with Na₂SO₄and fil-

tered, before evaporation of the solvent under reduced pressure. All

reactions employing dry solvents were carried out under nitrogen

(or argon where indicated). Lipase from recombinant Candida ant-

arctica was a kind gift from Novo Nordisk. Amano PS lipase was

a kind gift of Amano–Mitsubishi Italia.

(aromatic CH and C-3), 127.1, 133.1, 134.3, 150.8 (quat.), 151.9

(C=O) ppm. IR: ν = 3605, 3431, 3001, 2955, 2169, 1710, 1598,

˜

max

1481, 1452, 1381, 1162, 1135, 1005, 962, 841 cm^–1. GC-MS (usual

method but with final temp. 290 °C): R_t= 12.00 min. MS: m/z (%)

= 421 (5.9) [M]⁺, 347 (11), 346 (32), 345 (5.4), 344 (14), 328 (6.0),

280 (11), 264 (5.7), 252 (11), 236 (5.9), 226 (6.6), 206 (5.7), 180 (13),

172 (5.1), 165 (5.2), 156 (5.7), 155 (5.4), 151 (6.3), 115 (5.3), 97

(5.3), 94 (14), 83 (5.2), 77 (41), 75 (28), 73 (100), 65 (6.9), 59 (6.3),

51 (6.6), 45 (14), 43 (13), 39 (5.8). C₂₄H₂₇NO₄Si (421.56): calcd. C

68.38, H 6.46, N 3.32; found C 68.29, H 6.50, N 3.40.

Phenyl

(2R)-4-[(R)-1-Hydroxy-3-(triphenylmethoxy)prop-2-yl]-2-

[(trimethylsilyl)ethynyl]-1,2-dihydroquinoline-1-carboxylate (27a)

From Acetate 22a: The acetate 22a (85 mg, 120 µmol) was dispersed

in phosphate buffer (pH 7)/heptane (20 mL, 85:15). After addition

of lipase from Candida antarctica (87 mg) the mixture was vigor-

ously stirred at 60 °C for 31 h. The enzyme was filtered through a

celite pad and the crude product was extracted with Et₂O and

washed with brine. After solvent removal and chromatography (PE/

Et₂O 7:3), compound 27a (49 mg, 61%) was obtained as a white

foam together with 20 mg of unreacted starting material (24%).

Phenyl (2R)-4-[(S)-1-Acetoxy-3-hydroxyprop-2-yl]-2-[(trimethylsilyl)-

ethynyl]-1,2-dihydroquinoline-1-carboxylate (25a)

With pTSA in MeOH: A solution of 22a^[40](51 mg, 72.2 µmol) in

dry MeOH (2 mL) was cooled to 0 °C and treated with pTSA

(14 mg, 73.6 µmol). After 5 min the reaction mixture was allowed

to stir at room temp. for 5 h. After addition of solid NaHCO₃

(6.2 mg, 73.6 µmol), MeOH was removed under reduced pressure

without heating. The crude product was partitioned between water

and AcOEt, extracted with AcOEt and washed with brine. After

solvent removal, the mixture was chromatographed (PE/Et₂O 3:7

to Et₂O) to give 25a (9.4 mg, 28%) and 26 (18.5 mg, 61%) as white

foams.

From Triethylsilyl Ether 23a: This transformation has already been

reported.^[40]

Spectroscopic data for compound 27a have already been re-

ported.^[40]

Phenyl (2R)-4-[(S)-1-Methoxymethoxy-3-(triphenylmethoxy)prop-2-

yl]-2-[(trimethylsilyl)ethynyl]-1,2-dihydroquinoline-1-carboxylate

(28a): A solution of 27a (101 mg, 152 µmol) in dry CH₂Cl₂(5 mL)

was cooled to 0 °C and treated with N,N-diisopropylethylamine

With ZnBr₂: Dry ZnBr₂(2.252 g, 10 mmol) was dissolved in dry

CH₂Cl₂/iPrOH (85:15, 10 mL). After the mixture had been cooled

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2777

R. Riva, L. Banfi, A. Basso, V. Gandolfo, G. Guanti

FULL PAPER

(44 µL, 253 µmol) and chloromethyl methyl ether (17 µL,

224 µmol). The reaction mixture was then allowed to stir at room

temp. and the same amount of both reagents was added twice, after

20 and 30 h respectively, in order to achieve complete consumption

of the substrate. The resulting solution was partitioned between

water and Et₂O and extracted. After solvent removal and

chromatography (PE/Et₂O 9:1 to 8:2), 28a (92 mg, 85%) was ob-

tained as a white foam. R_f= 0.35 (PE/Et₂O 8:2). [α]_D= +188.1 (c

= 2.1, CHCl₃). ¹H NMR (200 MHz, CDCl₃): δ = 0.02 [s, 9 H,

evaporation with n-heptane was employed) gave the crude tosylate

(4.55 g). This was taken up in dry DMSO (15 mL), treated with

nBu₄NI (715 mg, 1.94 mmol) and KCN (1.864 g, 28.6 mmol) and

warmed at 60 °C for 2 h and 40 min. At this point the conversion

was complete (in order to separate 41 from the starting tosylate by

tlc it is necessary to use a 15 cm plate with PE/Et₂O/CH₂Cl₂4:3:3

as eluent and to carry out a double run). The solution was poured

into a mixture of water and brine (1:1). Extraction with Et₂O

(twice) and then with AcOEt (twice), followed by washing with

(CH₃)₃Si], 3.26–3.40 (m, 3 H, CHCH₂OTr), 3.28 (s, 3 H, OCH₃), brine, concentration and chromatography (PE/Et₂O 6:4 to 4:6),

3.86–4.02 (m, 2 H, CH₂OMOM), 4.58 and 4.58 (AB syst., J = gave pure (S)-41 as an oil (2.626 g, 84%). The preparation of this

7.2 Hz, 2 H, OCH₂OCH₃), 5.94 and 5.94 (AB syst., J = 7.2 Hz, 2

H, 2-H and 3-H), 6.91–7.35 (m, 23 H), 7.77 (br. d, J = 7.2 Hz, 1 in an overall yield of 79% without purification of the intermediates.

H, 5-H or 8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –0.2 R_f= 0.51 (PE/AcOEt 1:1). [α]_D= +33.9 (c = 1.3, CHCl₃). ¹H NMR

(CH₃Si), 40.2 (C-2Ј), 44.9 (CHN), 55.4 (OCH₃), 62.7, 67.7 (200 MHz, CDCl₃): δ = –0.02, –0.01 [2ϫs, 2 ϫ3 H, (CH₃)₂Si], 0.87

compound from the monoacetate 18 can be carried out (four steps)

(CH₂OTr and CH₂OMOM), 86.3 (OCPh₃), 88.5, 101.4 (CϵC),

96.9 (OCH₂OCH₃), 121.7 (ϫ3), 122.2, 123.2, 125.0, 125.6, 126.9

(ϫ3), 127.5, 127.7 (ϫ6), 128.6 (ϫ6), 129.2 (ϫ2, aromatic CH and

C-3), 125.2, 134.4, 134.5, 143.9 (ϫ 3), 150.9 (quat.), 151.8

[s, 9 H, (CH₃)₃C], 2.92 and 3.04 (AB part of an ABX syst., J_AB

16.7, J_AX= 6.6, J_BX= 6.0 Hz, 2 H, CH₂CN), 3.84–4.11 (m, 3 H,

CH₂OSi and 3Ј-H), 7.34 (d, J = 4.6 Hz, 1 H, 3-H), 7.63 (dt, J_d

1.4, J_t= 7.7 Hz, 1 H, 6-H), 7.76 (dt, J_d= 1.3, J_t= 7.6 Hz, 1 H, 7-

H), 8.02 (dd, J = 1.2, 8.2 Hz, 1 H, 5-H), 8.17 (dd, J = 0.6, 8.4 Hz,

1 H, 8-H), 8.90 (d, J = 4.6 Hz, 1 H, 2-H) ppm. ¹³C NMR (50 MHz,

CDCl₃): δ = –5.6 (CH₃Si), 18.2 [C(CH₃)₃], 19.5 (CH₂CN), 25.8

[(CH₃)₃C], 38.3 (C-3Ј), 64.4 (CH₂O), 118.1 (CN), 118.7 (C-3),

122.1, 127.2, 129.4, 130.8 (C-5, C-6, C-7, C-8), 126.5 (C-4a), 144.7

=

(C=O) ppm. IR: ν

= 3004, 2955, 2926, 2172, 1713, 1484, 1448,

˜

max

1379, 1325, 1288, 1196, 1020, 960, 842 cm^–1. GC-MS: unsuitable

for this analysis. C₄₅H₄₅NO₅Si (707.93): calcd. C 76.35, H 6.41, N

1.98; found C 76.55, H 6.36, N 1.90.

Phenyl (2R)-4-[(S)-1-Hydroxy-3-(methoxymethoxy)prop-2-yl]-2-

[(trimethylsilyl)ethynyl]-1,2-dihydroquinoline-1-carboxylate (29a): A

solution of 28a (221 mg, 312 µmol) in dry MeOH (2 mL) was co-

oled to 0 °C and treated with HCl (0.1  in MeOH, 100 µL). The

reaction mixture was then allowed to stir at room temp. and the

same amount of HCl was added twice, after 2 and 4 h, in order to

achieve complete consumption of the substrate. After addition of

aqueous NaHCO₃(5%) and evaporation of MeOH the crude prod-

uct was partitioned between water and Et₂O. After solvent removal

and chromatography (PE/Et₂O 24:76), 29a (99 mg, 68%) was ob-

(C-4), 148.5 (C-8a), 149.9 (C-2) ppm. IR: ν

= 2952, 2927, 2855,

˜

max

2248, 1718, 1592, 1571, 1462, 1387, 1362, 1191, 1103, 1006, 933,

833 cm^–1. GC-MS: R_t= 9.51 min. MS: m/z (%) = 326 (5.6) [M]⁺,

271 (5.6), 269 (100), 242 (14), 239 (6.9), 228 (7.9), 154 (15), 98 (12),

75 (8.0), 73 (6.8). C₁₉H₂₆N₂OSi (326.51): calcd. C 69.89, H 8.03, N

8.58; found C 70.0, H 8.3, N 8.4.

(R)-4-[(tert-Butyldimethylsilyl)oxy]-3-(quinolin-4-yl)butanenitrile

[(R)-41]: A solution of (R)-43 (130.4 mg, 0.61 mmol) in dry DMF

(2 mL) was cooled to 0 °C and treated with imidazole (67.9 mg,

0.99 mmol) and with tert-butyldimethylsilyl chloride (129.2 mg,

tained as a yellow foam. R_f= 0.37 (PE/Et₂O 24:76). [α]_D= +280.8

1

(c = 1.0, CHCl₃). H NMR (200 MHz, CDCl₃): δ = 0.04 [s, 9 H, 0.86 mmol). After 5 min the cooling bath was removed and the

(CH₃)₃Si], 2.27 (br. s, 1 H, OH), 3.32 (quint., J = 5.8 Hz, 1 H, 2Ј-

H), 3.42 (s, 3 H, OCH₃), 3.80–4.03 (m, 4 H, CH₂OH and CH₂O-

mixture was stirred overnight at room temp. Addition of saturated

aqueous NH₄Cl (20 mL) and saturated NaHCO₃(20 mL), followed

MOM), 4.72 (s, 2 H, OCH₂OCH₃), 5.93 and 5.96 (AB syst., J = by extraction with Et₂O, concentration and chromatography (PE/

6.7 Hz, 2 H, 2-H and 3-H), 7.17–7.48 (m, 8 H), 7.76 (br. d, J = AcOEt 7:3), gave pure (R)-43 as an oil (166.4 mg, 83%). [α]_D

8.4 Hz, 1 H, 5-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –0.3

–32.8 (c = 1.5, CHCl₃). The other analytical data were identical to

=

(CH₃Si), 41.4 (C-2Ј), 44.9 (CHN), 55.6 (OCH₃), 63.9, 69.0 those reported above for its enantiomer.

(CH₂OH and CH₂OMOM), 88.8, 100.9 (CϵC), 96.8 (OCH₂-

(R)-4-Acetoxy-3-(quinolin-4-yl)butanenitrile [(R)-42]: A solution of

the monoacetate 18^[39](10.86 g, 44.3 mmol, 97% ee) in dry CH₂Cl₂

(24 mL) was treated with pyridine (60 mL), cooled to 0 °C and

treated with p-toluenesulfonyl chloride (10.27 g, 53.9 mmol). After

10 min the cooling bath was removed and the solution was stirred

for 5 h at room temp. After quenching with H₂O (200 mL), the pH

was adjusted to 8 by addition of solid NaHCO₃, and the aqueous

phase was extracted once with Et₂O and twice with AcOEt. After

washing with brine, evaporation (in order to remove pyridine com-

pletely, azeotropic evaporation with n-heptane was employed) gave

the crude tosylate (13.50 g). This was taken up in dry DMSO

(60 mL), treated with nBu₄NI (3.173, 8.60 mmol) and KCN

OCH₃), 121.6 (ϫ2), 122.4, 123.0, 125.0, 125.7, 127.9, 129.3 (ϫ2,

aromatic CH and C-3), 127.2, 133.3, 134.3, 150.9 (quat.), 151.8

(C=O) ppm. IR: ν

= 3542, 3002, 2953, 2171, 1715, 1483, 1453,

˜

max

1379, 1327, 1304, 1187, 1106, 1018, 961, 842 cm^–1. GC-MS (usual

method but with final temp. 290 °C): R_t= 12.14 min. m/z = 465

(3.5) [M]⁺, 420 (6.3), 388 (15), 374 (5.0), 347 (8.3), 346 (28), 280

(11), 254 (6.2), 252 (10), 250 (6.5), 236 (5.9), 230 (5.6), 226 (5.8),

208 (5.2), 206 (5.2), 194 (6.4), 180 (14), 151 (6.0), 127 (5.5), 89 (5.5),

77 (32), 75 (16.2), 74 (6.4), 73 (71), 59 (8.8), 45 (100). C₂₆H₃₁NO₅Si

(465.61): calcd. C 67.07, H 6.71, N 3.01; found C 67.25, H 6.76, N

3.11.

(S)-4-[(tert-Butyldimethylsilyl)oxy]-3-(quinolin-4-yl)butanenitrile (5.593 g, 85.9 mmol) and stirred at room temp. for 27 h. The solu-

[(S)-41]: A solution of the alcohol (R)-40^[39]([α]_D= +35.6, c = 1.5

CHCl₃, ee = 97%, 3.054 g, 9.62 mmol) in dry CH₂Cl₂(10 mL) was

cooled to 0 °C and treated with pyridine (10 mL) and with p-tolu-

enesulfonyl chloride (2.608 g, 13.7 mmol). After 5 min the cooling

bath was removed and the mixture was stirred for 5 h at room temp.

After quenching with H₂O (50 mL), the pH was adjusted to 8 by

addition of solid NaHCO₃, and the aqueous phase was extracted

twice with Et₂O and once with AcOEt. After washing with brine,

evaporation (in order to remove pyridine completely, azeotropic

tion was poured into water (100 mL) and brine (100 mL). Extrac-

tion with Et₂O (once) and then with AcOEt (twice), followed by

washing with brine, concentration and chromatography (PE/AcOEt

2:8), gave pure (R)-42 as an oil (10.43 g, 93%). R_f= 0.47 (PE/

AcOEt 20:80). [α]_D= –13.8 (c = 2.1, CHCl₃). ¹H NMR (200 MHz,

CDCl₃): δ = 2.10 (s, 3 H, CH₃CO), 2.96 (d, J = 6.2 Hz, 2 H,

CH₂CN), 4.20–4.43 (m, 1 H, CHHOAc and 3Ј-H), 4.59 (dd, J =

4.4, 10.6 Hz, 1 H, CHHOAc), 7.33 (d, J = 4.8 Hz, 1 H, 3-H), 7.67

(dt, J_d= 1.6, J_t= 7.7 Hz, 1 H, 6-H), 7.79 (dt, J_d= 1.2, J_t= 7.7 Hz,

2778

www.eurjoc.org

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

1 H, 7-H), 8.08 (d, J = 8.8 Hz, 1 H, 5-H), 8.19 (dd, J = 1.2, 8.0 Hz,

1 H, 8-H), 8.94 (d, J = 4.8 Hz, 1 H, 2-H) ppm. ¹³C NMR (50 MHz,

8.13 (d, J = 5.8 Hz, 1 H, 5-H or 8-H), 8.17 (d, J = 5.4 Hz, 1 H, 5-

H or 8-H), 8.85 (d, J = 4.6 Hz, 1 H, 2-H), 9.83 (t, J = 1.4 Hz, 1

CDCl₃): δ = 20.0 (CH₂CN), 20.3 (CH₃CO), 35.0 (C-3Ј), 65.1 H, CHO) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –5.6 (CH₃Si),

(CH₂O), 117.2 (CN), 118.1 (C-3), 121.9, 127.3, 129.5, 130.5 (C-5, 18.1 [C(CH₃)₃], 25.7 [(CH₃)₃C], 35.5 (CH₂CH=), 41.4 (H-3Ј), 66.0

C-6, C-7, C-8), 126.1 (C-4a), 143.2 (C-4), 148.3 (C-8a), 149.8 (C-

(CH₂O), 116.9 (C=CH₂), 119.2 (C-3), 123.2, 126.3, 128.8, 130.3 (C-

5, C-6, C-7, C-8), 126.3 (C-4a), 135.8 (CH=CH₂), 148.4 (C-4),

2) ppm. IR: ν_max= 3000, 2963, 1741, 1592, 1571, 1420, 1384, 1366,

˜

1190, 1035, 907, 838 cm^–1. GC-MS: R_t= 8.68 min. MS: m/z (%) =

254 (20.0) [M]⁺, 194 (14.7), 182 (34.1), 181 (40.0), 179 (6.5), 172

(6.1), 155 (14.9), 154 (100), 153 (20.7), 129 (14.8), 127 (8.4), 101

148.9 (C-8a), 149.8 (C-2) ppm. IR: ν

= 3021, 2949, 2927, 2855,

˜

max

2729, 2474, 1724, 1589, 1570, 1462, 1386, 1360, 1200, 1101, 1005,

935, 831 cm^–1. GC-MS: R_t= 9.32 min. MS: m/z (%) = 329 (0.06)

(5.0), 43 (33.1). C₁₅H₁₄N₂O₂(254.28): calcd. C 70.85, H 5.55, N [M]⁺, 272 (8.4), 198 (7.4), 181 (15), 180 (100), 168 (8.2), 154 (9.3),

11.02; found C 70.7, H 5.7, N 10.85.

101 (9.8), 75 (13), 73 (8.9), 59 (9.2). C₁₉H₂₇NO₂Si (329.51): calcd.

C 69.26, H 8.26, N 4.25; found C 69.4, H 8.4, N 4.15.

(S)-4-Hydroxy-3-(quinolin-4-yl)butanenitrile [(S)-43]: A solution of

(S)-41 (93.5 mg, 286 µmol) in acetonitrile (2 mL) was cooled to

0 °C, and treated with aqueous HF (40%, 100 µL). The solution

was stirred at 0 °C for 5 h and was then treated with saturated

aqueous NaHCO₃(20 mL) and extracted with Et₂O. After washing

with brine, concentration and chromatography (AcOEt to AcOEt/

MeOH 96:4) gave pure (S)-43 as a white solid (58.2 mg, 95%); m.p.

113.8–114.8 °C. [α]_D= +28.2 (c = 1, acetone). The enantiomeric

excess was determined by conversion into the Mosher’s esters and

was found to be 96%. R_f= 0.43 (AcOEt). ¹H NMR (200 MHz,

CDCl₃): δ = 2.86–3.15 (m, 2 H, CH₂CN), 3.40 (br. s, 1 H, OH),

4.00–4.20 (m, 3 H, CH₂OH and 3Ј-H), 7.29 (d, J = 4.7 Hz, 1 H, 3-

H), 7.57 (dt, J_d= 1.2, J_t= 6.8 Hz, 1 H, 6-H), 7.67 (dt, J_d= 1.2, J_t

= 7.0 Hz, 1 H, 7-H), 7.98 (d, J = 8.4 Hz, 2 H, 5-H and 8-H), 8.74

(d, J = 4.7 Hz, 1 H, 2-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ =

19.4 (CH₂CN), 38.2 (C-3Ј), 63.8 (CH₂O), 118.1 (CN), 118.6 (C-3),

122.1, 127.3, 129.6, 130.3 (C-5, C-6, C-7, C-8), 126.4 (C-4a), 144.8

(S)-1-[(tert-Butyldimethylsilyl)oxy]-2-(quinolin-4-yl)pent-4-ene [(S)-

45]: “Instant ylide” (methyltriphenylphosphonium bromide + so-

dium amide, Fluka cat. 69500, 468.9 mg, 1.10 mmol) was sus-

pended in dry THF (10 mL) under Ar, stirred for 15 min at room

temp. and cooled to –78 °C. A solution of aldehyde (S)-44

(208.7 mg, 0.63 mmol) in dry THF (1 mL) was added. After 4 h at

–78 °C the reaction was complete and it was quenched with AcOH

in Et₂O (1 , 2.2 mL). The mixture was stirred for 15 min at room

temp., treated with Et₃N (500 µL), evaporated to dryness and chro-

matographed immediately (PE/Et₂O 1:1 + 2% Et₃N to PE/Et₂O

3:7 + 2% Et₃N) to give pure (S)-45 as an oil (188 mg, 91%). Impor-

tant note: This reaction was found to be rather erratic. Although

we obtained good yields in five instances, in two other cases, al-

though the reaction seemed clean as usual in tlc, a poor yield was

obtained after chromatography. This annoying behaviour is proba-

bly due to decomposition pathways activated by the excess phos-

phorane and by silica. Rapid chromatography, in the presence of

Et₃N, is therefore essential. R_f= 0.74 (PE/AcOEt 1:1). [α]_D= +31.0

(C-4), 148.1 (C-8a), 149.8 (C-2) ppm. IR: ν

= 3608, 3401 (br),

˜

max

3003, 1709, 1592, 1571, 1499, 1417, 1357, 1175, 1063 cm^–1. GC-

MS: R_t= 8.52 min. MS: m/z (%) = 212 (55.4) [M]⁺, 182 (32.6), 181

(52.3), 179 (9.7), 155 (100.0), 154 (92.7), 153 (5.7), 143 (7.5), 142

(6.7), 129 (15.5), 128 (10.1), 127 (16.4), 126 (5.6), 115 (9.7), 102

(5.0), 101 (11.8), 77 (12.0), 76 (5.4), 75 (12.0), 63 (8.7), 51 (9.7), 50

(5.7), 39 (5.4). C₁₃H₁₂N₂O (212.25): calcd. C 73.56, H 5.70, N

13.20; found C 73.35, H 5.8, N 13.0.

1

(c = 1.3, CHCl₃). H NMR (200 MHz, CDCl₃): δ = –0.11 (s, 6 H,

CH₃Si), 0.80 [s, 9 H, (CH₃)₃C], 2.55 (dt, J_d= 14.4, J_t= 7.2 Hz, 1

H, CHHCH=), 2.75 (dt, J_d= 14.4, J_t= 5.8 Hz, 1 H, CHHCH=),

3.68–3.92 (m, 3 H, CH₂O and 3Ј-H), 4.90–5.12 (m, 2 H, C=CH₂),

5.72 (ddt, J_d= 10.0, 17.0, J_t= 5.0 Hz, 1 H, CH=CH₂), 7.31 (d, J

= 4.6 Hz, 1 H, 3-H), 7.56 (dt, J_d= 1.6, J_t= 7.7 Hz, 1 H, 6Ј-H),

7.71 (dt, J_d= 1.4, J_t= 7.7 Hz, 1 H, 7Ј-H), 8.12 (d, J = 9.0 Hz, 2

H, 5Ј-H and 8Ј-H), 8.85 (d, J = 4.6 Hz, 1 H, 2Ј-H) ppm. ¹³C NMR

(50 MHz, CDCl₃): δ = –5.6 (CH₃Si), 18.1 [C(CH₃)₃], 25.7

[(CH₃)₃C], 35.5 (CH₂CH=), 41.4 (C-2), 66.0 (CH₂O), 116.9

(C=CH₂), 119.2 (C-3Ј), 123.2, 126.3, 128.8, 130.3 (C-5Ј, C-6Ј, C-

7Ј, C-8Ј), 126.3 (4aЈ-C), 135.8 (CH=CH₂), 148.4 (C-4Ј), 148.9 (C-

(R)-4-Hydroxy-3-(quinolin-4-yl)butanenitrile [(R)-43]: A solution of

(R)-42 (210.3 mg, 0.83 mmol) in THF (5 mL) and MeOH (5 mL)

was cooled to 0 °C, and treated with a solution of MeONa in

MeOH (1 , 1.07 mL, 1.07 mmol). After the mixture had been

stirred for 1 h at 0 °C, a solution of AcOH in MeOH (1 , 1.25 mL)

and Et₃N (115 µL) were added. Concentration to dryness, followed

by chromatography (AcOEt to AcOEt/MeOH 96:4), gave pure (R)-

8aЈ), 149.8 (C-2Ј) ppm. IR: ν

= 3044, 2952, 2926, 2856, 1639,

˜

max

1589, 1570, 1462, 1361, 1193, 1103, 916, 831 cm^–1. GC-MS: R_t=

8.78 min. MS: m/z (%) = 327 (0.1) [M]⁺, 312 (2.6), 270 (100.0), 254

(6.0), 252 (7.4), 240 (3.9), 228 (9.2), 196 (33.7), 168 (6.9), 154 (11.7),

75 (9.7), 73 (9.9). C₂₀H₂₉NOSi (327.54): calcd. C 73.34, H 8.92, N

4.28; found C 73.4, H 8.9, N 4.25.

43 as a white solid (133 mg, 76%); m.p. 114.4–115.2 °C. [α]_D

=

–28.9 (c = 1.7, acetone). The other analytical data were identical

to those reported above for the enantiomer. The enantiomeric ex-

cess was determined by conversion into the Mosher’s esters and

was found to be 96%.

Phenyl (2R)- and (2S)-4-[(S)-1-Hydroxypent-5-en-2-yl]-2-[(trimeth-

ylsilyl)ethynyl]-1,2-dihydroquinoline-1-carboxylates (46a and 46b): A

solution of trimethylsilylacetylene (355 µL, 2.56 mmol) in dry THF

(4 mL) was cooled to 0 °C and treated with EtMgBr in Et₂O (3 ,

800 µL, 2.40 mmol). In another flask, the quinoline (S)-45

(419.5 mg, 1.28 mmol) was dissolved in dry THF (4 mL), cooled to

–78 °C and treated with the above solution of trimethylsilylethnyn-

ylmagnesium bromide (3.5 mL, about 1.75 mmol). After 1 min,

phenyl chloroformate (240 µL, 1.92 mmol) was added. The mixture

was stirred at –78 °C for 2.5 h and was then poured into saturated

aqueous NH₄Cl and extracted with Et₂O. Concentration and

chromatography gave an inseparable diastereoisomeric mixture of

the addition products (R_f= 0.41 and 0.43, PE/Et₂O 9:1, 540 mg,

(S)-4-[(tert-Butyldimethylsilyl)oxy]-3-(quinolin-4-yl)butenal [(S)-44]:

A solution of (S)-41 (855 mg, 2.62 mmol) in dry CH₂Cl₂(15 mL)

was cooled to –78 °C and treated with a solution of diisobutylalu-

minium hydride in toluene (1 , 4.3 mL). After 2 h the reaction

was quenched with saturated aqueous sodium potassium tartrate

(30 mL). After stirring for 1 h at room temp., extraction with Ac-

OEt gave, after concentration and chromatography (PE/Et₂O 1:9),

pure aldehyde (S)-44 as an oil (723 mg, 84%). R_f= 0.21 (PE/Et₂O

1

2:8). [α]_D= +33.7 (c = 1.7, CHCl₃). H NMR (200 MHz, CDCl₃):

δ = 0.04 and 0.06 (2ϫs, 2 ϫ3 H, CH₃Si), 0.84 [s, 9 H, (CH₃)₃C],

2.91–3.15 (m, 2 H, CH₂CHO), 3.77 and 3.95 (AB part of an ABX

syst., J_AB= 9.9, J_AX= 6.9, J_BX= 4.8 Hz, 2 H, CH₂O), 4.28–4.38

(m, 1 H, 3Ј-H), 7.28 (d, J = 4.6 Hz, 1 H, 3-H), 7.62 (dt, J_d= 1.4, 77 %). The diastereomeric ratio, determined by ¹H NMR, was

J_t= 6.2 Hz, 1 H, 6-H), 7.71 (dt, J_d= 1.4, J_t= 7.7 Hz, 1 H, 7-H),

60.5:39.5 (with the less polar product prevailing).

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2779

R. Riva, L. Banfi, A. Basso, V. Gandolfo, G. Guanti

FULL PAPER

This mixture was taken up in CH₃CN (7 mL) and cooled to 0 °C,

It was taken up in dry CH₂Cl₂(10 mL), cooled to –78 °C and

then aqueous HF (40%, 350 µL) was added. After 3 h the mixture treated with CBr₄(550 mg, 1.66 mmol). A solution of PPh₃

was treated with saturated aqueous NaHCO₃(17 mL) and ex-

tracted with Et₂O. After washing with brine, concentration and

chromatography (PE/Et₂O 6:4 to 1:1) gave pure (2R,2ЈS)-46a

(218 mg, 51%, R_f= 0.27, PE/Et₂O 1:1) and (2S,2ЈS)-46b (142 mg,

33%, R_f= 0.40, PE/Et₂O 1:1). The overall yield of 46a and 46b

from 45 was 65 %.

(595 mg, 2.27 mmol) in CH₂Cl₂(5 mL) was then slowly added

dropwise over 15 min. At the end of the addition, the temperature

was allowed to rise to –40 °C. After the system had been kept for

30 min at this temperature, Et₃N (1 mL) was added and the mixture

was poured into aqueous Na₂S₂O₃(0.4 , 30 mL). Extraction with

Et₂O, followed by washing with brine (20 mL) containing KH₂PO₄

(1 , 3 mL) and Na₂S₂O₃(0.4 , 3 mL) and by concentration, af-

forded a crude product, which was chromatographed (PE/Et₂O

95:5) to give pure 47a as a brownish oil (234 mg, 72%). R_f= 0.70

Compound (2R,2ЈS)-46a: [α]_D= +408.9 (c = 2.3, CHCl₃). ¹H NMR

(200 MHz, CDCl₃): δ = 0.04 (9 H, CH₃Si), 2.35–2.64 (m, 2 H,

CH₂CH=), 3.10 (quint., J = 6.2 Hz, 1 H, 2Ј-H), 3.56–3.82 (m, 2 H,

CH₂OH), 5.08 (dd, J = 1.0, 11.0 Hz, 1 H, C=CHH), 5.15 (dd, J =

1.0, 18.4 Hz, 1 H, C=CHH), 5.90 (ddt, J_d= 10.2, 18.4, J_t= 6.9 Hz,

1 H, CH=CH₂), 5.94 and 5.98 (AB syst., J = 7.0 Hz, 2 H, 2-H and

3-H), 7.15–7.46 (m, 8 H), 7.76 (br. d, J = 7.8 Hz, 1 H, 5-H or 8-

H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –0.3 (CH₃Si), 34.9

(CH₂–CH=), 41.0 (C-2Ј), 45.0 (CHN), 64.5 (CH₂OH), 89.0, 101.2

(CϵC), 117.1 (C=CH₂), 121.6 (ϫ 2), 122.5, 123.1, 124.9 (ϫ2),

125.8, 127.9, 129.4 (ϫ2, aromatic CH and C-3), 134.6 (CH=CH₂),

1

(toluene/CH₂Cl₂/Et₂O 70:29:1). H NMR (200 MHz, CDCl₃): δ =

0.04 (9 H, CH₃Si), 2.38–2.70 (m, 2 H, CH₂CH=), 3.79 (q, J =

7.9 Hz, 1 H, 3Ј-H), 5.08–5.23 (m, 2 H, C=CH₂), 5.87 (ddt, J_d

=

10.2, 17.2, J_t= 6.9 Hz, 1 H, CH=CH₂), 5.95 and 5.98 (AB syst., J

= 6.5 Hz, 2 H, 2-H and 3-H), 6.31 (d, J = 9.6 Hz, 1 H, CH=CBr₂),

7.16–7.48 (m, 8 H), 7.75 (br. d, J = 7.6 Hz, 1 H, 5-H or 8-H) ppm.

¹³C NMR (50 MHz, CDCl₃): δ = –0.2 (CH₃Si), 37.3 (CH₂CH=),

43.2 (C-3Ј), 44.9 (CHN), 89.1, 90.8, 100.9 (CϵC and CBr₂), 117.7

(C=CH₂), 121.7 (ϫ2), 122.3, 123.3, 125.0, 125.8, 128.0, 129.4 (ϫ2,

aromatic CH and C-3), 134.5 (CH=CH₂), 127.1, 134.2, 135.4, 151.0

(quat.), 139.9 (CH=CBr₂), 151.9 (C=O) ppm. GC-MS: R_t=

12.83 min. MS: m/z (%) = 587 (2.3), 585 (3.5), 583 (2.0) [M]⁺, 356

(2.1), 354 (3.1), 352 (1.7), 346 (5.4), 258 (4.1), 230 (6.9), 190 (4.4),

151 (7.2), 139 (18.5), 137 (14.3), 97 (4.7), 83 (4.8), 77 (35.9), 73

(100.0), 65 (8.6), 59 (6.1), 51 (5.7), 45 (6.2), 41 (7.3), 39 (5.2).

C₂₇H₂₇Br₂NO₂Si (585.40): calcd. C 55.40, H 4.65, N 2.39; found C

55.65, H 4.7, N 2.3.

127.6, 134.4, 135.2, 150.9 (quat.), 151.9 (C=O) ppm. IR: ν

=

˜

max

3597, 3040, 2957, 2169, 1712, 1637, 1593, 1482, 1450, 1379, 1327,

1300, 1242, 1162, 1138, 1004, 961, 914, 842 cm^–1. GC-MS: R_t=

12.45 min. MS: m/z (%) = 431 (3.4) [M]⁺, 354 (12.1), 347 (6.2), 346

(19.1), 338 (6.2), 254 (6.1), 252 (10.8), 226 (7.9), 206 (7.4), 180

(13.6), 178 (5.4), 167 (5.3), 166 (5.4), 151 (8.5), 97 (5.5), 94 (5.8),

77 (39.6), 75 (19.6), 73 (100.0), 65 (7.5), 59 (7.4), 45 (8.7), 41 (7.3),

39 (6.7). C₂₆H₂₉NO₃Si (431.60): calcd. C 72.35, H 6.77, N 3.25;

found C 72.4, H 6.8, N 3.2.

Phenyl (2S)-4-[(R)-1,1-Dibromohexa-1,5-dien-3-yl]-2-[(trimethyl-

silyl)ethynyl]-1,2-dihydroquinoline-1-carboxylate (47b): This com-

pound was prepared in 64% yield from the alcohol (2S,2ЈS)-46b by

the same procedure as described above for 47a. R_f= 0.65 (toluene/

CH₂Cl₂/Et₂O 70:29:1). ¹H NMR (200 MHz, CDCl₃): δ = 0.04 (9

H, CH₃Si), 2.20–2.58 (m, 2 H, CH₂CH=), 3.77 (dt, J_d= 5.4, J_t=

1

Compound (2S,2ЈS)-46b: [α]_D= –431.5 (c = 2.0, CHCl₃). H NMR

(200 MHz, CDCl₃): δ = 0.04 (9 H, CH₃Si), 2.37 (t, J = 6.7 Hz, 2

H, CH₂CH=), 3.11 (quint., J = 6.2 Hz, 1 H, 2Ј-H), 3.78–3.92 (m,

2 H, CH₂OH), 4.98–5.14 (m, 2 H, C=CH₂), 5.77 (ddt, J_d= 10.2,

17.0, J_t= 6.9 Hz, 1 H, CH=CH₂), 5.96 and 6.00 (AB syst., J =

6.8 Hz, 2 H, 2-H and 3-H), 7.15–7.48 (m, 8 H), 7.75 (br. d, J =

8.0 Hz, 1 H, 5-H or 8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ

= –0.3 (CH₃Si), 35.5 (CH₂CH=), 41.2 (C-2Ј), 44.7 (CHN), 64.4

(CH₂OH), 88.9, 101.5 (CϵC), 116.9 (C=CH₂), 121.5 (ϫ2), 121.9,

123.0, 125.0 (ϫ2), 125.7, 127.8, 129.3 (ϫ2, aromatic CH and C-

3), 135.7 (CH=CH₂), 134.3, 135.7, 150.9 (quat., the missing signal

for a quat. C atom is probably overlapped by one of the CH sig-

nals), 151.9 (C=O) ppm. GC-MS: R_t= 12.23 min. MS: m/z (%) =

431 (4.4) [M]⁺, 354 (11.6), 347 (5.8), 346 (18.0), 338 (5.5), 254 (5.3),

252 (9.2), 226 (7.4), 206 (6.5), 180 (11.3), 167 (5.2), 151 (7.2), 94

(6.2), 83 (5.0), 77 (35.7), 75 (22.4), 73 (100.0), 65 (6.8), 59 (7.8), 51

8.6 Hz, 1 H, 3Ј-H), 4.99–5.14 (m, 2 H, C=CH₂), 5.73 (ddt, J_d

=

9.6, 17.6, J_t= 6.6 Hz, 1 H, CH=CH₂), 5.93 (s, 2 H, 2-H and 3-H),

6.53 (d, J = 8.8 Hz, 1 H, CH=CBr₂), 7.14–7.47 (m, 8 H), 7.74 (br.

d, J = 7.8 Hz, 1 H, 5-H or 8-H) ppm. GC-MS: R_t= 12.92 min.

MS: m/z (%) = 587 (2.7), 585 (4.2), 583 (2.0) [M]⁺, 356 (3.2), 354

(4.7), 352 (2.7), 346 (5.7), 290 (5.4), 258 (6.0), 230 (8.3), 217 (5.2),

204 (9.4), 203 (5.5), 191 (5.3), 190 (6.1), 179 (5.3), 166 (5.4), 154

(4.9), 151 (9.8), 139 (24.0), 137 (17.1), 83 (5.1), 77 (39.4), 73 (100.0),

65 (8.9), 59 (6.5), 51 (6.7), 43 (5.7), 41 (8.5), 39 (5.0).

C₂₇H₂₇Br₂NO₂Si (585.40): calcd. C 55.40, H 4.65, N 2.39; found C

55.7, H 4.75, N 2.3.

(5.8), 45 (9.0), 43 (5.1), 41 (7.8), 39 (6.3). IR: ν

= 3587, 3007,

˜

max

Phenyl (2R)-4-[(R)-Hex-5-en-1-yn-3-yl]-2-[(trimethylsilyl)ethynyl]-

1,2-dihydroquinoline-1-carboxylate (48a): The dibromide (2R,3ЈR)-

47a (217.4 mg, 0.37 mmol) was dissolved in dry toluene (9 mL) un-

der Ar and cooled to –78 °C. This solution was rapidly treated

2956, 2169, 1713, 1636, 1594, 1482, 1450, 1379, 1324, 1302, 1248,

1162, 1135, 1000, 962, 910, 843 cm^–1. C₂₆H₂₉NO₃Si (431.60): calcd.

C 72.35, H 6.77, N 3.25; found C 72.5, H 6.8, N 3.15.

Phenyl (2R)-4-[(R)-1,1-Dibromohexa-1,5-dien-3-yl]-2-[(trimethyl- with nBuLi in hexanes (1.6 , 510 µL, 816 µmol). After 15 min, the

silyl)ethynyl]-1,2-dihydroquinoline-1-carboxylate (47a): A CH₂Cl₂

solution of DMSO (1.4 , 1.58 mL, 2.21 mmol) was diluted with

dry CH₂Cl₂(5 mL), cooled under N₂to –78 °C, and treated with

a solution of (COCl)₂in CH₂Cl₂(2.08 , 663 µL, 1.38 mmol). After

10 min, a solution of the alcohol (2R,2ЈS)-46a (239 mg, 554 µmol)

in CH₂Cl₂(5 mL) was added. After a further 10 min, N-ethyldi-

isopropylamine (867 µL, 4.98 mmol) was added and the mixture

was stirred overnight at –78 °C. The mixture was rapidly poured

into an Erlenmeyer flask containing aqueous (NH₄)H₂PO₄(5%,

25 mL) and HCl (1 , 5 mL, pH 3). Extraction with PE/Et₂O 1:1,

washing with brine and concentration to dryness gave the crude

aldehyde (252 mg), which was used as such for the subsequent Co-

rey–Fuchs reaction.

reaction was quenched with AcOH in Et₂O (1 , 1.6 mL). After

15 min the solution was poured into brine (15 mL) and saturated

aqueous NaHCO₃(5 mL), and extracted with Et₂O. The crude

product was chromatographed (PE/Et₂O 9:1) to give pure 48a as a

yellowish oil (86.9 mg, 55 %). R_f= 0.39 (PE/Et₂O 9:1). [α]_D

=

+298.7 (c = 1.5, CHCl₃). ¹H NMR (200 MHz, CDCl₃): δ = 0.05 (9

H, CH₃Si), 2.27 (d, J = 2.6 Hz, 1 H, CϵCH), 2.46–2.74 (m, 2 H,

CH₂CH=), 3.72 (ddd, J = 2.2, 5.2, 7.6 Hz, 1 H, 3Ј-H), 5.08–5.25

(m, 2 H, C=CH₂), 5.97 (ddt, J_d= 10.6, 16.4, J_t= 7.0 Hz, 1 H,

CH=CH₂), 5.99 (d, J = 6.6 Hz, 1 H, 2-H), 6.22 (d, J = 6.6 Hz, 1

H, 3-H), 7.14–7.53 (m, 8 H), 7.76 (br. d, J = 7.6 Hz, 1 H, 5-H or

8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –0.2 (CH₃Si), 33.2 (C-

3Ј), 38.8 (CH₂CH=), 45.0 (CHN), 71.6, 84.4, 89.0, 101.7 (CϵC–

2780

www.eurjoc.org

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

H), 117.8 (C=CH₂), 121.7 (ϫ2), 123.0 (ϫ2), 124.8, 125.0, 125.8, 100 °C for 2 h and 50 min. The solution was quenched with satu-

127.9, 129.4 (ϫ2, aromatic CH and C-3), 134.6 (CH=CH₂), 134.0,

rated aqueous NH₄Cl and the pH was adjusted to 8 by addition

134.2, 151.0 (quat., the missing signal for a quat. C atom is proba- of aqueous (NH₄)H₂PO₄(5%). Extraction with Et₂O, followed by

bly overlapped by one of the CH signals), 151.9 (C=O) ppm. GC-

MS: R_t= 11.06 min. MS: m/z (%) = 425 (6.9) [M]⁺, 346 (75.9), 332

washing with brine, concentration and chromatography (PE/Et₂O

6:4 to 1:9) gave pure (S)-57 as a white foam (1.90 g, 76%). R_f=

(5.6), 316 (5.8), 230 (7.9), 220 (6.6), 204 (6.2), 194 (21.9), 151 (7.2), 0.27 (PE/Et₂O/CH₂Cl₂4:3:3). [α]_D= +46.6 (c = 1.2, CHCl₃). ¹H

139 (7.9), 83 (6.9), 77 (25.4), 75 (7.5), 74 (9.6), 73 (100), 59 (7.2),

45 (7.1), 39 (5.8). C₂₇H₂₇NO₂Si (425.59): calcd. C 76.20, H 6.39, N

3.29; found C 76.3, H 6.4, N 3.2.

NMR (200 MHz, CDCl₃): δ = 2.96 and 3.07 (AB part of an ABX

syst., J_AB= 16.8, J_AX= 7.4, J_BX= 5.7 Hz, 2 H, CH₂CN), 3.54 and

3.62 (AB part of an ABX syst., J_AB= 9.8, J_AX= 7.7, J_BX= 4.6 Hz,

2 H, CH₂OTr), 4.03 (quint., J = 6.5 Hz, 1 H, 3-H), 7.18–7.37 (m,

16 H, aromatics of Tr and 3Ј-H), 7.54 (dt, J_d= 1.4, J_t= 7.7 Hz, 1

H, 6Ј-H), 7.69–7.78 (m, 2 H, 5Ј-H and 7Ј-H), 8.14 (d, J = 8.8 Hz, 1

H, 8Ј-H), 8.84 (d, J = 4.4 Hz, 1 H, 2-H) ppm. ¹³C NMR (50 MHz,

CDCl₃): δ = 19.9 (CH₂CN), 36.7 (H-3), 64.9 (CH₂O), 87.3

(OCPh₃), 117.8 (CN), 118.4 (C-3), 122.1, 126.9, 129.3, 130.6 (C-5Ј,

C-6Ј, C-7Ј, C-8Ј), 126.4 (4aЈ-C), 127.2 (ϫ3), 127.9 (ϫ6), 128.4 (ϫ6)

(CH of Tr), 143.3 (ϫ3, quat. Tr), 144.3 (C-4Ј), 148.4 (C-8aЈ), 149.8

Phenyl (2S)-4-[(R)-Hex-5-en-1-yn-3-yl]-2-[(trimethylsilyl)ethynyl]-

1,2-dihydroquinoline-1-carboxylate (48b): This compound was ob-

tained in 61% yield from the dibromide (2S,3ЈR)-47b by the same

procedure as described above for 48a. R_f= 0.40 (PE/Et₂O 9:1).

1

[α]_D= –329.5 (c = 1.45, CHCl₃). H NMR (200 MHz, CDCl₃): δ

= 0.04 (9 H, CH₃Si), 2.31 (dt, J_d= 14.2, J_t= 7.7 Hz, 1 H

CHHCH=), 2.41 (d, J = 2.2 Hz, 1 H, CϵCH), 2.46–2.62 (m, 1 H,

CHHCH=), 3.89 (centre of m, 1 H, 3Ј-H), 4.96–5.15 (m, 2 H,

C=CH₂), 5.89 (ddt, J_d= 10.4, 17.2, J_t= 6.8 Hz, 1 H, CH=CH₂),

5.97 (d, J = 7.0 Hz, 1 H, 2-H), 6.42 (dd, J = 1.2, 6.6 Hz, 1 H, 3-

H), 7.14–7.46 (m, 8 H), 7.74 (br. d, J not measurable, 1 H, 5-H or

8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –0.2 (CH₃Si), 33.3 (C-

3Ј), 38.5 (CH₂CH=), 45.2 (CHN), 73.5, 83.4, 88.9, 100.8 (CϵCH),

117.3 (C=CH₂), 121.6 (ϫ 2), 122.9, 123.6, 125.0, 125.4, 125.8,

127.9, 129.4 (ϫ2, aromatic CH and C-3), 134.7 (CH=CH₂), 133.7,

134.6, 151.0 (quat., the missing signal for a quat. C atom is proba-

bly overlapped by one of the CH signals), 151.9 (C=O) ppm. IR:

(C-2Ј) ppm. IR: ν

= 2957, 2253, 1733, 1448, 1361, 1243, 1171,

˜

max

1068, 972, 907 cm^–1. GC-MS (usual method but with final temp.

290 °C): R_t= 16.50 min. MS: m/z (%) = 454 (1.2) [M]⁺, 244 (25),

243 (100), 196 (6.1), 195 (14), 183 (6.7), 166 (5.6), 165 (35), 156

(6.1), 155 (6.0), 154 (1.0), 105 (22), 77 (12). C₃₂H₂₆N₂O (454.56):

calcd. C 84.55, H 5.77, N 6.16; found C 84.75, H 5.80, N 6.22.

(S)-3-(Quinolin-4-yl)-4-[(triphenylmethyl)oxy]butan-1-ol [(S)-58]: A

solution of (S)-57 (1.45 g, 3.19 mmol) in dry CH₂Cl₂(15 mL) was

cooled to –78 °C and treated with DIBALH (1  in toluene,

4.1 mL). After the system had been kept for 5 h at –78 °C, saturated

aqueous Rochelle’s salt solution was added and the mixture was

stirred at room temp. until two clear, well separated layers were

obtained. After extraction with Et₂O, followed by washing with

brine and solvent evaporation, the crude aldehyde was taken up in

methanol (20 mL), cooled to 0 °C and treated with NaBH₄

(241 mg, 6.38 mmol, added portionwise). After 2 h the reaction was

quenched with aqueous saturated NH₄Cl and the pH was adjusted

to 8 by addition of aqueous (NH₄)H₂PO₄(5%). Extraction with

Et₂O, concentration and chromatography (PE/Et₂O 1.9 to pure

Et₂O) gave (S)-58 as a white foam (498 mg, 34%). R_f= 0.35 (PE/

Et₂O 4:96). [α]_D= +34.4 (c = 1.2, CHCl₃). ¹H NMR (200 MHz,

CDCl₃): δ = 1.94–2.38 (m, 2 H, CHCH₂), 3.43 (d, J = 5.8 Hz, 2 H,

CH₂OTr), 3.48–3.67 (m, 2 H, CH₂OH), 3.99 (centre of m, 1 H, 3-

H), 7.16–7.30 (m, 16 H, aromatics of Tr and 3Ј-H), 7.54 (dt, J_d=

1.0, J_t= 7.7 Hz, 1 H, 6Ј-H), 7.71 (dt, J_d= 1.2, J_t= 7.7 Hz, 1 H,

7Ј-H), 8.11 (t, J = 6.6 Hz, 2 H, 5’-H and 7Ј-H), 8.81 (br. s, 1 H, 2Ј-

H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = 35.2 (CHCH₂), 36.5 (H-

3), 60.4, 67.3 (CH₂OTr and CH₂OH), 86.9 (OCPh₃), 118.9 (C-3Ј),

123.3, 126.4, 129.1, 130.2 (C-5Ј, C-6Ј, C-7Ј, C-8Ј), 127.9 (4aЈ-C),

127.0 (ϫ3), 127.7 (ϫ6), 128.5 (ϫ6) (CH of Tr), 143.7 (ϫ3, quat.

ν

_max= 3303, 3004, 2957, 1720, 1594, 1480, 1453, 1378, 1326, 1303,

˜

1202, 1134, 1000, 971, 917, 844 cm^–1. GC-MS: R_t= 10.9 min. MS:

m/z (%) = 425 (8.9) [M]⁺, 220 (5.4), 194 (17.8), 154 (5.1), 139 (7.0),

83 (7.1), 77 (22.6), 75 (8.8), 74 (8.8), 73 (100), 59 (8.0), 51 (5.1), 45

(8.4), 41 (5.8), 39 (5.5). C₂₇H₂₇NO₂Si (425.59): calcd. C 76.20, H

6.39, N 3.29; found C 76.5, H 6.5, N 3.2.

(S)-3-(Quinolin-4-yl)-4-[(triphenylmethyl)oxy]butanenitrile [(S)-57]

Formation of the Tosylate: A solution of the alcohol (R)-56^[40]

(2.70 g, 6.06 mmol) in dry pyridine (14 mL) was cooled to 0 °C and

treated with p-toluenesulfonyl chloride (3.47 g, 18.2 mmol). It was

then allowed to stir at room temp. for 4 h. The solution was parti-

tioned between water and Et₂O and the pH was adjusted to 7 by

addition of aqueous NaHCO₃(5%). After extraction and evapora-

tion of the solvent, residual pyridine was azeotropically removed

with n-heptane and the crude product was purified by chromatog-

raphy with PE/Et₂O 1:1 to 1:9 to give the expected tosylate as a

white foam (3.30 g, 91%). R_f= 0.31 (PE/Et₂O/CH₂Cl₂4:3:3). [α]_D

1

= +20.8 (c = 1.8, CHCl₃). H NMR (200 MHz, CDCl₃): δ = 2.38

(s, 3 H, CH₃), 3.49 (d, J = 6.2 Hz, 2 H, CH₂OTr), 3.98 (quint., J

= 6.0 Hz, 1 H, 2Ј-H), 4.52 and 4.55 (AB part of an ABX syst., J_AB

= 10.1, J_AX= 6.4, J_BX= 6.0 Hz, 2 H, CH₂OTs), 7.02 (d, J = 4.8 Hz,

1 H, 3-H), 7.12–7.72 (m, 22 H, aromatics of Tr and Ts and 5-H, 6-

H, 7-H), 8.07 (dd, J = 8.8 Hz, 1 H and 1.0, 8-H), 8.70 (d, J =

4.4 Hz, 1 H, 2-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = 21.5

(CH₃), 39.5 (H-3), 63.0, 69.5 (CH₂OTr and CH₂OTs), 87.1

(OCPh₃), 119.1 (C-3), 122.4, 126.6, 129.0, 130.4 (C-5, C-6, C-7, C-

8Ј), 126.8 (C-4a), 127.1 (ϫ3), 127.8 (ϫ6), 128.4 (ϫ6) (CH of Tr),

127.6 (ϫ2), 129.6 (ϫ2) (CH of Ts), 132.3 (SO₂C of Ts), 143.3 (ϫ3,

quat. Tr), 143.7 (CH₃C of Ts), 144.7 (C-4), 148.3 (C-8a), 149.6 (C-

Tr), 148.3 (C-8aЈ), 149.4 (C-4Ј), 149.8 (C-2Ј) ppm. IR: ν_max= 3405,

˜

2953, 1589, 1441, 1259, 1069, 898 cm^–1. GC-MS (standard condi-

tions but with final temp. 290 °C): R_t= 16.54 min. MS: m/z (%) =

258 (1.2) [M – 201]⁺, 244 (23), 243 (100), 200 (5.5), 165 (26), 154

(6.6), 105 (10), 77 (5.4). C₃₂H₂₉NO₂(459.58): calcd. C 83.63, H

6.36, N 3.05; found C 83.80, H 6.31, N 3.15.

(S)-2-(Quinolin-4-yl)-4-[(triethylsilyl)oxy]-1-[(triphenylmethyl)oxy]-

butane [(S)-59]: A solution of (S)-58 (181 mg, 394 µmol) in dry

CH₂Cl₂(4 mL) was cooled to 0 °C and treated with 2,6-lutidine

(101 µL, 867 µmol) and triethylsilyl triflate (125 µL, 552 µmol). Af-

ter 3 h the solution was partitioned between water and Et₂O and

the pH was adjusted to about 8 by addition of saturated aqueous

NaHCO₃. After extraction with Et₂O and solvent removal, residual

2,6-lutidine was azeotropically removed with n-octane. Chromatog-

2) ppm. IR: ν_max= 2952, 2927, 2859, 1592, 1570, 1445, 1364, 1307,

˜

1167, 1072, 975, 897 cm^–1. GC-MS: unsuitable for this analysis.

C₃₈H₃₃NO₄S (599.74): calcd. C 76.10, H 5.55, N 2.34; found C

76.25, H 5.70, N 2.38.

Formation of the Cyanide: A solution of the above tosylate (3.30 g,

5.50 mmol) in dry DMSO (15 mL) was treated with nBu₄NI raphy with PE/Et₂O 6:4 to 1:1 gave (S)-59 as a yellow oil (107 mg,

1

(407 mg, 1.10 mmol) and KCN (1.29 g, 28.6 mmol) and warmed at

47%). R_f= 0.43 (PE/Et₂O 1.1). [α]_D= +21.4 (c = 1.7, CHCl₃). H

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2781

R. Riva, L. Banfi, A. Basso, V. Gandolfo, G. Guanti

FULL PAPER

NMR (200 MHz, CDCl₃): δ = 0.48 [q, J = 7.9 Hz, 6 H, (CH₃CH₂)₃- (R)-4-[(4-Methoxybenzyl)oxy]-3-(quinolin-4-yl)butanenitrile [(R)-

Si], 0.85 [t, J = 7.9 Hz, 9 H, (CH₃CH₂)₃Si], 1.99–2.28 (m, 2 H, 62]: A solution of the alcohol (R)-43 (3.713 g, 17.5 mmol) in dry

CHCH₂), 3.34–3.65 (m, 5 H, CHCH₂OTr and CH₂OTES), 4.02 DMF (60 mL) was cooled to 0 °C and treated portionwise with

(quint., J = 6.6 Hz, 1 H, 2-H), 7.14–7.28 (m, 16 H, aromatics of Tr NaH in mineral oil (60%, 705 mg, 17.5 mmol). After cessation of

and 3Ј-H), 7.53 (dt, J_d= 1.0, J_t= 7.4 Hz, 1 H, 6Ј-H), 7.71 (dt, J_dgas evolution, p-methoxybenzyl chloride (2.70 mL, 19.2 mmol) was

= 1.2, J_t= 7.6 Hz, 1 H, 7Ј-H), 8.13 (br. d, J = 8.4 Hz, 2 H, 5Ј-H added. After 5 min the cooling bath was removed and the suspen-

and 7Ј-H), 8.82 (br. d, J = 4.8 Hz, 1 H, 2Ј-H) ppm. ¹³C NMR sion was stirred for 30 min. After quenching with saturated NH₄Cl

(50 MHz, CDCl₃): δ = 4.3 [(CH₃CH₂)₃Si], 6.7 [(CH₃CH₂)₃Si], 35.1 (80 mL), the mixture was extracted with Et₂O (twice) and AcOEt

(CHCH₂), 36.1 (H-2), 60.3, 67.3 (CH₂OTr and CH₂OTES), 86.6 (once). The organic extracts were washed with water and then with

(OCPh₃), 118.9 (C-3Ј), 123.6, 126.1, 128.9, 130.2 (C-5Ј, C-6Ј, C-7Ј, brine. Concentration and chromatography gave pure (R)-62 as an

C-8Ј, C-4aЈ), 126.9 (ϫ3), 127.8 (ϫ6), 128.6 (ϫ6, CH of Tr), 143.8

(ϫ3, quat. Tr), 148.4, 149.7, 149.8 (C-8aЈ, C-4Ј, C-2Ј) ppm. IR:

oil (5.23 g, 90%). R_f= 0.68 (AcOEt). [α]_D= –40.3 (c = 1.8, CHCl₃).

¹H NMR (200 MHz, CDCl₃): δ = 2.94 and 3.01 (AB part of an

ABX syst., J_AB= 14.9, J_AX= 4.5, J_BX= 4.2 Hz, 2 H, CH₂CN),

3.78 and 3.88 (AB part of an ABX syst., J_AB= 9.6, J_AX= 7.8, J_BX

= 4.2 Hz, 2 H, CH₂OPMB), 3.81 (s, 3 H, OCH₃), 4.05–4.21 (m, 1

H, 3-H), 4.51 (s, 2 H, ArCH₂), 6.88 (d, J = 8.4 Hz, 2 H, H ortho

to OMe), 7.24 (d, J = 8.4 Hz, 2 H, H meta to OMe), 7.35 (d, J =

4.6 Hz, 1 H, 3Ј-H), 7.61 (dt, J_d= 1.4, J_t= 7.7 Hz, 1 H, 6Ј-H), 7.75

(dt, J_d= 1.4, J_t= 7.5 Hz, 1 H, 7Ј-H), 7.97 (d, J = 8.4 Hz, 1 H, 5Ј-

H or 8Ј-H), 8.17 (dd, J = 0.6, 8.4 Hz, 1 H, 5Ј-H or 8Ј-H), 8.89 (d,

J = 4.6 Hz, 1 H, 2Ј-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ =

20.16 (CH₂CN), 36.2 (C-3), 55.2 (OCH₃), 70.6, 73.2 (CH₂O), 113.9

(C ortho to OMe), 117.8 (CCH₂O), 118.9 (C-3Ј), 122.0, 127.2,

129.4, 130.7 (C-5Ј, C-6Ј, C-7Ј, C-8Ј), 126.4 (C-4aЈ), 129.4 (C meta

to OMe), 130.7 (CCH₂O), 144.5 (C-4Ј), 148.4 (C-8aЈ), 150.0 (C-2Ј),

ν

= 2953, 2871, 1587, 1447, 1386, 1187, 1073, 898, 872 cm^–1.

˜

max

GC-MS: unsuitable for this analysis. C₃₈H₄₃NO₂Si (573.84): calcd.

C 79.54, H 7.55, N 2.44; found C 79.70, H 7.52, N 2.61.

(S)-4-[(tert-Butyldimethylsilyl)oxy]-3-(quinolin-4-yl)butanol [(S)-60]:

The alcohol (S)-41 (1.98 g, 6.06 mmol) was converted into the alde-

hyde (S)-44 by the procedure already described above. The crude

aldehyde was taken up in dry MeOH (15 mL), cooled to –40 °C

and treated with NaBH₄(1.193 g, 31.5 mmol). The temperature

was allowed to rise slowly to 0 °C over 2 h. The reaction was

quenched with saturated aqueous NH₄Cl. Extraction with Et₂O

(twice) and AcOEt (twice), washing with brine, concentration and

chromatography (PE/AcOEt 3:7) gave pure (S)-60 (1.254 g, 62%).

R_f= 0.15 (Et₂O). [α]_D= +48.0 (c = 1.1, CHCl₃). ¹H NMR

(200 MHz, CDCl₃): δ = –0.04 (s, 6 H, CH₃Si), 0.83 [s, 9 H,

(CH₃)₃C], 1.97–2.42 (m, 2 H, CH₂CH₂OH), 3.55–3.94 (m, 5 H,

CH₂O and 3-H), 7.30 (d, J = 4.6 Hz, 1 H, 3Ј-H), 7.57 (dt, J_d= 1.3,

J_t= 7.7 Hz, 1 H, 6Ј-H), 7.71 (dt, J_d= 1.6, J_t= 7.7 Hz, 1 H, 7Ј-H),

8.08–8.21 (m, 2 H, 5Ј-H and 8Ј-H), 8.84 (d, J = 4.6 Hz, 1 H, 2Ј-

H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –5.7 (CH₃Si), 18.1

159.4 (COMe) ppm. IR: ν_max= 2956, 2861, 2836, 1667, 1610, 1592,

˜

1570, 1502, 1462, 1418, 1360, 1301, 1192, 1093, 1030 cm^–1. GC-

MS: R_t= 11.86 min. MS: m/z (%) = 332 (18.1) [M]⁺, 331 (9.0), 156

(9.6), 154 (6.3), 136 (5.8), 121 (100.0), 78 (8.0), 77 (10.2).

C₂₁H₂₀N₂O₂(332.40): calcd. C 75.88, H 6.06, N 8.43; found C 75.7,

H 6.1, N 8.4.

[C(CH₃)₃], 25.7 [(CH₃)₃C], 35.3 (CH₂CH=), 38.7 (H-3), 60.5, 67.0 (R)-4-[(4-Methoxybenzyl)oxy]-3-(quinolin-4-yl)butanol [(R)-63]:

(CH₂O), 118.9 (C-3Ј), 123.1, 126.4, 129.0, 130.1 (C-5Ј, C-6Ј, C-7Ј,

C-8Ј), 127.6 (C-4aЈ), 148.2 (C-4Ј), 149.5 (C-8aЈ), 149.7 (C-2Ј) ppm.

This compound was prepared from (R)-62 in 66% yield by the same

procedure as described above for (S)-60. R_f= 0.33 (AcOEt). [α]_D

= –20.4 (c = 2.1, CHCl₃). H NMR (200 MHz, CDCl₃): δ = 1.81

(br. s, 1 H, OH), 1.95–2.31 (m, 2 H, CH₂CH₂OH), 3.53–3.83 (4 H,

CH₂OH and CH₂OPMB), 3.80 (s, 3 H, OCH₃), 3.99 (quint., J =

6.5 Hz, 1 H, 3-H), 4.46 (s, 2 H, ArCH₂), 6.85 (d, J = 8.8 Hz, 2 H,

H ortho to OMe), 7.19 (d, J = 8.8 Hz, 2 H, H meta to OMe), 7.27

1

IR: ν

= 3614, 3382 (br), 2951, 2881, 2856, 1588, 1571, 1463,

˜

max

1390, 1361, 1242, 1100, 906, 831 cm^–1. GC-MS: R_t= 9.60 min. MS:

m/z (%) = 316 [M – 15]⁺(2.5), 300 (2.7), 274 (100.0), 244 (39.1),

241 (5.7), 200 (17.3), 182 (44.2), 180 (15.4), 170 (32.8), 168 (11.6),

167 (22.1), 156 (13.5), 154 (17.3), 105 (27.9), 89 (5.7), 75 (45.6), 73

(14.8). C₁₉H₂₉NO₂Si (331.52): calcd. C 68.83, H 8.82, N 4.22; (d, J = 4.8 Hz, 1 H, 3Ј-H), 7.53 (dt, J_d= 1.2, J_t= 7.7 Hz, 1 H, 6Ј-

found C 70.0, H 8.9, N 4.15.

H), 7.71 (dt, J_d= 1.4, J_t= 7.6 Hz, 1 H, 7Ј-H), 8.12 (dd, J = 1.0,

8.8 Hz, 2 H, 5Ј-H and 8Ј-H), 8.82 (d, J = 4.8 Hz, 1 H, 2Ј-H) ppm.

¹³C NMR (50 MHz, CDCl₃): δ = 36.4 (CH₂CH₂OH), 36.9 (C-3),

55.3 (OCH₃), 60.8, 73.1, 73.5 (CH₂O), 113.9 (C ortho to OMe),

118.8 (C-3Ј), 123.0, 126.6, 129.1, 130.4 (C-5Ј, C-6Ј, C-7Ј, C-8Ј),

127.3 (C-4aЈ), 129.4 (C meta to OMe), 130.4 (CCH₂O), 148.5 (C-

(S)-4-Acetoxy-1-[(tert-butyldimethylsilyl)oxy]-2-(quinolin-4-yl)-

butane [(S)-61]: A solution of the alcohol (S)-60 (1.101 g,

2.95 mmol) in dry CH₂Cl₂(8 mL) was treated with pyridine (4 mL)

and acetic anhydride (550 µL). After 1 h at room temp. the reaction

was complete, and H₂O (10 mL) was added. After stirring for

30 min, the mixture was diluted with saturated aqueous NaHCO₃

and extracted with AcOEt. The organic extracts were washed with

brine and the solvents were evaporated to dryness. The residue was

taken up three times with CH₂Cl₂/n-heptane and concentrated

again in order to remove pyridine completely. The crude product

4Ј), 149.2 (C-8aЈ), 150.0 (C-2Ј), 159.4 (COMe) ppm. IR: ν

=

˜

max

3416 (br), 2952, 2862, 1610, 1587, 1570, 1501, 1460, 1391, 1360,

1301, 1204, 1072, 800 cm^–1. GC-MS: R_t= 11.85 min. MS: m/z (%)

= 337 (12.4) [M]⁺, 336 (4.2), 276 (6.1), 156 (5.2), 154 (8.4), 121

(100.0), 77 (5.3). C₂₁H₂₃NO₃(337.41): calcd. C 74.75, H 6.87, N

4.15; found C 74.8, H 6.9, N 4.1.

1

(1.260 g) was pure by tlc, GC-MS and H NMR analyses and was

therefore used as such for the preparation of 66a and 66b. R_f=

(R)-4-[(tert-Butyldimethylsilyl)oxy]-1-[(4-methoxybenzyl)oxy]-2-(qui-

0.59 (Et₂O). ¹H NMR (200 MHz, CDCl₃): δ = –0.08 (s, 6 H, nolin-4-yl)butane [(R)-64]: This compound was prepared in 90%

CH₃Si), 0.81 [s, 9 H, (CH₃)₃C], 1.93 (s, 3 H, CH₃CO), 2.04–2.45

yield from (R)-63 by the procedure reported above for the synthesis

(m, 2 H, CH₂CH₂OAc), 3.76–4.21 (m, 5 H, CH₂O and 2-H), 7.34 of (R)-41. Chromatography was done with PE/Et₂O 3:7 to 15:85.

1

(d, J = 4.8 Hz, 1 H, 3Ј-H), 7.58 (dt, J_d= 1.6, J_t= 7.7 Hz, 1 H, 6Ј- R_f= 0.38 (PE/Et₂O 2:8). [α]_D= –10.3 (c = 1.5, CHCl₃). H NMR

H), 7.72 (dt, J_d= 1.5, J_t= 7.6 Hz, 1 H, 7Ј-H), 8.13 (dt, J_d= 8.2,

(200 MHz, CDCl₃): δ = –0.09 and –0.07 (2ϫs, 2 ϫ3 H, CH₃Si),

J_t= 1.8 Hz, 2 H, 5Ј-H and 8Ј-H), 8.87 (d, J = 4.8 Hz, 1 H, 2Ј- 0.85 [s, 9 H, (CH₃)₃C], 1.85–2.28 (m, 2 H, CH₂CH₂O), 3.42–3.77

H) ppm. GC-MS: R_t= 9.87 min. MS: m/z (%) = 358 (1.3)

[M – 15]⁺, 317 (9.8), 316 (39.7), 274 (32.9), 242 (5.0), 182 (25.3),

170 (6.1), 168 (8.6), 167 (13.5), 154 (12.0), 117 (100.0), 89 (9.0), 75

(50.3), 73 (29.7), 59 (8.5), 58 (5.1), 43 (28.7).

(4 H, CH₂O), 3.79 (s, 3 H, OCH₃), 4.05 (quint., J = 6.6 Hz, 1 H,

2-H), 4.41 (s, 2 H, ArCH₂), 6.83 (d, J = 8.6 Hz, 2 H, H ortho to

OMe), 7.13 (d, J = 8.6 Hz, 2 H, H meta to OMe), 7.32 (d, J =

4.6 Hz, 1 H, 3Ј-H), 7.53 (dt, J_d= 1.2, J_t= 7.6 Hz, 1 H, 6Ј-H), 7.71

2782

www.eurjoc.org

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

(dt, J_d= 1.4, J_t= 7.6 Hz, 1 H, 7Ј-H), 8.12 (dd, J = 0.8, 8.4 Hz, 1

H, 5Ј-H or 8Ј-H), 8.18 (d, J = 9.2 Hz, 1 H, 5Ј-H or 8Ј-H), 8.84 (d,

J = 4.6 Hz, 1 H, 2Ј-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –5.5

(CH₃Si), 18.2 [C(CH₃)₃], 25.9 [C(CH₃)₃], 35.5 (CH₂CH₂OH), 35.8

aromatic and olefinic CH), 128.3, 134.2, 136.2, 151.0 (quat.), 151.9,

170.9 (C=O) ppm. IR: ν

= 3003, 2953, 2892, 2854, 1721, 1594,

˜

max

1483, 1379, 1325, 1209, 1108, 962, 836 cm^–1. GC-MS: R_t=

14.25 min. MS: m/z (%) = 591 (1.3) [M]⁺, 534 (2.5), 514 (6.6), 346

(C-2), 55.2 (OCH₃), 60.6, 72.8, 72.9 (CH₂O), 113.7 (C ortho to (38.2), 206 (4.7), 180 (6.8), 155 (7.2), 151 (10.1), 117 (37.6), 89

OMe), 118.9 (C-3Ј), 123.5, 126.2, 128.9, 130.2 (C-5Ј, C-6Ј, C-7Ј, C- (22.5), 77 (17.4), 75 (39.4), 73 (100.0), 59 (7.4), 57 (5.9), 43 (27.5).

8Ј), 127.9 (C-4aЈ), 129.1 (C meta to OMe), 130.2 (C–CH₂O), 148.5

(C-4Ј), 149.5 (C-8aЈ), 150.0 (C-2Ј), 159.2 (COMe) ppm. IR: ν

C₃₃H₄₅NO₅Si₂(591.89): calcd. C 66.96, H 7.66, N 2.37; found C

70.0, H 7.7, N 2.35.

=

˜

max

2951, 2927, 2855, 1610, 1588, 1571, 1505, 1462, 1386, 1360, 1300,

1193, 1091 cm^–1. GC-MS: R_t= 13.13 min. MS: m/z (%) = 451

(0.04) [M]⁺, 436 (0.1), 394 (2.5), 242 (1.6), 168 (1.8), 167 (1.6), 154

(2.0), 121 (100.0), 77 (2.7), 73 (2.8). C₂₇H₃₇NO₃Si (451.67): calcd.

C 71.80, H 8.26, N 3.10; found C 72.0, H 8.4, N 3.05.

Phenyl (2S)- and (2R)-4-{(R)-4-[(tert-Butyldimethylsilyl)oxy]-1-[(4-

methoxybenzyl)oxy]but-2-yl}-2-[(trimethylsilyl)ethynyl]-1,2-dihydro-

quinoline-1-carboxylates (67a and 67b): These compounds were pre-

pared from (R)-64 (546.3 mg) by the procedure described above for

66a and 66b. Chromatography (PE/Et₂O 9:1 to 75:25) afforded pure

67a (545.3 mg), pure 67b (180.6 mg) and mixed fractions (41.2 mg).

The overall yield was 93%. The diastereoisomeric ratio, determined

Phenyl (2R)- and (2S)-4-{(S)-4-Acetoxy-1-[(tert-butyldimethylsilyl)-

oxy]but-2-yl}-2-[(trimethylsilyl)ethynyl]-1,2-dihydroquinoline-1-carb-

oxylates (66a and 66b): A solution of trimethylsilylacetylene

(685 µL, 4.95 mmol) in dry THF (7.6 mL) was cooled to 0 °C and

treated with EtMgBr in Et₂O (3 , 1.65 mL, 4.95 mmol). In an-

other flask, crude quinoline (S)-61 (616.6 mg, 1.65 mmol) was dis-

solved in dry THF (9 mL), cooled to – 78 °C and treated with the

solution of trimethylsilylethynylmagnesium bromide. After 1 min,

phenyl chloroformate (625 µL, 4.95 mmol) was added. The mixture

was stirred at –78 °C for 1.3 h and the temperature was then al-

lowed to rise to –30 °C over 1 h. The mixture was then poured into

saturated aqueous NH₄Cl and extracted with Et₂O. The organic

extracts were washed with NaOH (to remove phenol) and then with

brine. After concentration, the crude product was treated with a

few drops of Et₃N and immediately chromatographed with PE/Ac-

OEt 8:2 containing Et₃N (0.5%) to give 66a (757 mg), mixed frac-

tions of 66a and 66b (190 mg) and pure 66b (92 mg). The mixed

fractions were chromatographed again, this time with PE/Et₂O 7:3,

to give 66a (112 mg) and 66b (78 mg).

1

by H NMR on the crude product, was 73:27 (67a prevailing).

Compound (2S,2ЈR)-67a: R_f= 0.58 (PE/Et₂O 8:2). [α]_D= –89.7 (c

= 1.1, CHCl₃). ¹H NMR (200 MHz, CDCl₃): δ = 0.03 [s, 9 H,

(CH₃)₃Si], 0.03 and 0.04 [2ϫs, 2 ϫ3 H, (CH₃)₂Si], 0.92 [s, 9 H,

(CH₃)₃C], 1.74–1.93 (m, 1 H, CHHCH₂OSi), 1.98–2.18 (m, 1 H,

CHHCH₂OSi), 3.24–3.70 (m, 3 H, CH₂OPMB and 2Ј-H), 3.60–

3.80 (m, 2 H, CH₂OSi), 3.76 (s, 3 H, CH₃O), 4.41 (s, 2 H, ArCH₂),

5.91 and 5.94 (AB syst., J = 6.6 Hz, 2 H, 2-H and 3-H), 6.83 (d, J

= 8.8 Hz, 2 H, H ortho to OMe), 7.10–7.43 (m, 9 H), 7.53 (dd, J

= 1.0, 7.6 Hz, 1 H, 5-H or 8-H), 7.73 (br. d, J = 7.4 Hz, 1 H, 5-H

or 8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –5.33, –5.27, –0.2

(CH₃Si), 18.3 [C(CH₃)₃], 26.0 [(CH₃)₃C], 34.4 (CH₂CH₂OSi), 35.4

(CHCH₂CH₂O), 45.0 (CHN), 55.2 (OCH₃), 60.5, 72.5, 73.1

(CH₂O), 88.6, 101.5 (CϵC), 113.7 (ϫ2), 121.7 (ϫ3), 123.4, 124.9

(ϫ2), 125.7, 127.5, 129.0 (ϫ2), 129.4 (ϫ2, aromatic CH and H-3),

128.4, 130.4, 134.3, 136.3, 151.0, 159.1 (quat.), 151.9 (C=O) ppm.

C₃₉H₅₁NO₅Si₂(700.00): calcd. C 69.91, H 7.67, N 2.09; found C

70.2, H 7.8, N 2.0.

Although 66b was analytically pure, 66a was contaminated with

diphenyl carbonate (12.5%). Full analytical characterisation of this

epimer was therefore performed only after conversion into

(2R,2ЈS)-69a. The calculated overall yield was 95%, whereas the

diastereoisomeric ratio (GC) was 82:18 (66a prevailing).

Compound (2R,4ЈR)-67b: R_f= 0.67 (PE/Et₂O 8:2). [α]_D= +100.4 (c

= 1.1, CHCl₃). ¹H NMR (200 MHz, CDCl₃): δ = –0.06 and 0.03

[2 ϫ s, 2 ϫ 3 H, (CH₃)₂Si], 0.02 [s, 9 H, (CH₃)₃Si], 0.85 [s, 9 H,

(CH₃)₃C], 1.64–1.99 (m, 2 H, CH₂CH₂OSi), 3.26 (quint., J =

Compound (2R,2ЈS)-66a: R_f= 0.47 (PE/Et₂O 8:2). ¹H NMR 6.3 Hz, 1 H, 2Ј-H), 3.50–3.72 (m, 4 H, CH₂OPMB and CH₂OSi),

(200 MHz, CDCl₃): δ = 0.00 [s, 9 H, (CH₃)₃Si], 0.07 [s, 9 H, (CH₃)₂-

3.80 (s, 3 H, CH₃O), 4.45 and 4.50 (AB syst., J = 11.6 Hz, 2 H,

Si], 0.84 [s, 9 H, (CH₃)₃C], 1.89–2.10 (m, 1 H, CHHCH₂OAc), 2.10 ArCH₂), 5.93 and 5.98 (AB syst., J = 7.0 Hz, 2 H, 2-H and 3-H),

(s, 3 H, CH₃CO), 2.15–2.34 (m, 1 H, CHHCH₂OAc), 3.13 (centre

of m, 1 H, 2Ј-H), 3.55 and 3.71 (AB part of an ABX syst., J_AB

9.9, J_AX= 6.7, J_BX= 4.5 Hz, 2 H, CH₂OSi), 4.09–4.41 (m, 2 H,

CH₂OAc), 5.96 and 5.99 (AB syst., J = 6.6 Hz, 2 H, 2-H and 3-

H), 7.18–7.53 (m, 8 H), 7.78 (br. d, J = 7.8 Hz, 1 H, 5-H or 8-

H) ppm. GC-MS: R_t= 13.81 min. MS: m/z (%) = 534 [M – 57]⁺

6.86 (d, J = 8.4 Hz, 2 H, H ortho to OMe), 7.10–7.50 (m, 10 H),

7.72 (br. d, J = 8.0 Hz, 1 H, 5-H or 8-H) ppm. ¹³C NMR (50 MHz,

CDCl₃): δ = –5.5, –0.2 (CH₃Si), 18.2 [C(CH₃)₃], 25.9 [(CH₃)₃C],

34.7 (CH₂CH₂OSi), 36.0 (CHCH₂CH₂O), 44.9 (CHN), 55.2

(OCH₃), 60.6, 72.4, 73.0 (CH₂O), 88.5, 101.5 (CϵC), 113.7 (ϫ2),

121.6 (ϫ3), 123.5, 124.9 (ϫ2), 125.7, 127.5, 129.4 (ϫ4, aromatic

=

(12.2), 514 (4.0), 400 (3.7), 346 (25.2), 278 (6.1), 206 (5.9), 180 (6.7), CH and H-3), 128.2, 130.6, 134.2, 136.6, 151.0, 159.1 (quat.), 151.9

151 (13.8), 147 (5.2), 117 (48.0), 115 (5.3), 89 (22.2), 77 (15.5), 75

(37.0), 73 (100.0), 59 (6.6), 57 (5.3), 43 (22.4).

(C=O) ppm. C₃₉H₅₁NO₅Si₂(700.00): calcd. C 69.91, H 7.67, N

2.09; found C 70.0, H 7.7, N 2.1.

Compound (2S,4ЈS)-66b: R_f= 0.40 (PE/Et₂O 8:2). [α]_D= –220.6 (c

= 1.1, CHCl₃). ¹H NMR (200 MHz, CDCl₃): δ = 0.06 [s, 9 H,

(CH₃)₃Si], 0.08 and 0.10 [2ϫs, 2 ϫ3 H, (CH₃)₂Si], 0.94 [s, 9 H,

Phenyl (2R)-4-[(S)-4-Hydroxy-1-(triphenylmethoxy)but-2-yl]-2-[(tri-

methylsilyl)ethynyl]-1,2-dihydroquinoline-1-carboxylate (68a): Com-

pounds 65a and 65b were prepared from (S)-59 in 79% yield and

(CH₃)₃C], 1.78–2.04 (m, 1 H, CHHCH₂OAc), 1.99 (s, 3 H, 82:18 diastereoisomeric ratio by the procedure described above for

CH₃CO), 2.00–2.20 (m, 1 H, CHHCH₂OAc), 3.07 (quint., J = 66a and 66b. Chromatography gave an inseparable mixture of the

6.3 Hz, 1 H, 2Ј-H), 3.78 and 3.88 (AB part of an ABX syst., J_ABtwo diastereoisomers. This mixture was directly subjected to Et₃Si

= 9.9, J_AX= 6.0, J_BX= 5.4 Hz, 2 H, CH₂OSi), 3.99–4.21 (m, 2 H,

CH₂OAc), 5.98 and 6.00 (AB syst., J = 6.5 Hz, 2 H, 2-H and 3-

H), 7.18–7.51 (m, 8 H), 7.77 (br. d, J = 7.4 Hz, 1 H, 5-H or 8-

H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –5.4, –0.2 (CH₃Si), 18.2

cleavage, carried out by the same procedure as already described

for the synthesis of 27a. Chromatography (PE/Et₂O 6:4 to 1:1) gave

pure 68a in 54% yield (66% taking the starting diastereoisomeric

ratio into account). Compound 68b was not isolated. R_f= 0.60

[C(CH₃)₃], 20.9 (CH₃CO), 25.9 [(CH₃)₃C], 30.6 (CH₂CH₂OAc), (PE/Et₂O 4:6, R_f= of 68b = 0.47). [α]_D= +218.0 (c = 2.1, CHCl₃).

38.7 (CHCH₂CH₂O), 44.8 (CHN), 62.7, 65.8 (CH₂O), 88.5, 101.3

¹H NMR (200 MHz, CDCl₃): δ = –0.01 [s, 9 H, (CH₃)₃Si], 1.87–

(CϵC) 121.6 (ϫ3), 123.1, 124.9 (ϫ2), 125.7, 127.6, 129.4 (ϫ2, 2.06 (m, 2 H, CH₂CH₂O), 3.13–3.29 (m, 3 H, CH₂OTr and 2Ј-H),

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2783

R. Riva, L. Banfi, A. Basso, V. Gandolfo, G. Guanti

FULL PAPER

3.68 (t, J = 6.2 Hz, 2 H CH₂OH), 5.94 (s, 2 H, 2-H and 3-H), 6.88– deacetylated adduct, was taken up in absolute MeOH (10 mL), co-

7.36 (m, 12 H), 7.43 (br. d, J = 8.2 Hz, 1 H, 5-H or 8-H), 7.75 (br. oled to 0 °C and treated with anhydrous K₂CO₃(33 mg). After

d, J = 7.2 Hz, 1 H, 5-H or 8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): stirring at room temp. for 2 h, the solution was poured into satu-

δ = –0.3 (CH₃Si), 33.9 (CH₂CH₂OH), 36.2 (CHCH₂CH₂O), 44.9 rated aqueous NH₄Cl (30 mL) and extracted with Et₂O. The or-

(CHN), 60.7, 66.2 (CH₂O), 86.5 (CPh₃), 88.8, 101.4 (CϵC) 121.7 ganic extracts were washed with brine, dried (Na₂SO₄), concen-

(ϫ2), 121.9, 123.3, 125.0, 125.1, 125.6, 126.9 (ϫ3), 127.6, 127.7 trated and chromatographed (CH₂Cl₂/Et₂O 9:1) to give pure 50a

(ϫ6), 126.6 (ϫ6), 129.2 (ϫ2, aromatic and olefinic CH), 128.0,

as an oil (114 mg, 57%). R_f= 0.52 (AcOEt). [α]_D= +282.6 (c =

134.4, 136.8, 143.9 (ϫ3), 150.9 (quat.), 151.9 (C=O) ppm. IR: ν

1.2, CHCl₃). ¹H NMR (200 MHz, CDCl₃): δ = 1.89–2.23 (m, 2 H,

˜

max

= 3471 (br), 3012, 2956, 1711, 1598, 1482, 1448, 1378, 1325, 1301, CH₂CH₂O), 2.25 (d, J = 2.6 Hz, 1 H, CϵCH), 2.28 (d, J = 2.6 Hz,

1241, 1070, 1019, 962, 906, 841 cm^–1. C₄₄H₄₃NO₄Si (677.90): calcd. 1 H, CϵCH), 3.82–4.03 (m, 3 H, CH₂OH, CHCH₂CH₂O), 6.02

C 77.96, H 6.39, N 2.07; found C 77.75, H 6.45, N 2.0.

(dd, J = 2.2, 6.6 Hz, 1 H, 2-H), 6.25 (d, J = 6.6 Hz, 1 H, 3-H),

7.14–7.47 (m, 7 H), 7.62 (dd, J = 1.5, 7.7 Hz, 1 H, 5-H or 8-H),

7.77 (br. d, J = 7.4 Hz, 1 H, 5-H or 8-H) ppm. ¹³C NMR (50 MHz,

CDCl₃): δ = 30.0 (CHCH₂CH₂O), 37.3 (CH₂CH₂O), 44.1 (CHN),

60.3 (CH₂O), 71.7, 72.1, 79.7, 84.4 (CϵCH) 121.6 (ϫ 2), 122.1

(broad), 123.5, 124.8, 125.0, 125.8, 128.2, 129.4 (ϫ2, aromatic and

olefinic CH), 126.2, 134.0, 135.1, 150.8 (quat.), 151.9 (C=O) ppm.

Phenyl (2R)-4-[(S)-4-Acetoxy-1-(hydroxy)but-2-yl]-2-[(trimethyl-

silyl)ethynyl]-1,2-dihydroquinoline-1-carboxylate (69a): A solution

of the Yamaguchi adduct (2R,2ЈS)-66a (648 mg, 87.5 % pure,

0.96 mmol) in acetonitrile (6 mL) was cooled to 0 °C and treated

with aqueous HF (40%, 275 µL). The solution was stirred at 0 °C

for 5 h and then treated with saturated aqueous NaHCO₃(20 mL)

and extracted with Et₂O. After washing with brine, concentration

and chromatography (PE/Et₂O 2:8 to 1:9) gave pure (2R,2ЈS)-69a

(384 mg, 84%). R_f= 0.37 (PE/Et₂O 2:8). [α]_D= +450.8 (c = 1.37,

CH₂Cl₂). ¹H NMR (200 MHz, CDCl₃): δ = 0.04 [s, 9 H, (CH₃)₃Si],

1.90–2.28 (m, 2 H, CH₂CH₂OAc), 2.07 (s, 3 H, CH₃CO), 3.16

(quint., J = 6.2 Hz, 1 H, 2Ј-H), 3.55–3.82 (m, 2 H, CH₂OH), 4.07–

4.35 (m, 2 H, CH₂OAc), 5.94 and 6.00 (AB syst., J = 6.8 Hz, 2 H,

2-H and 3-H), 7.14–7.50 (m, 8 H), 7.77 (br. d, J = 7.6 Hz, 1 H, 5-

H or 8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –0.4 (CH₃Si),

20.9 (CH₃CO), 29.5 (CH₂CH₂OAc), 38.1 (CHCH₂CH₂O), 44.9

(CHN), 62.5, 64.8 (CH₂O), 89.1, 100.9 (CϵC) 121.5 (ϫ2), 122.4,

122.9, 124.9 (ϫ2), 125.7, 127.9, 129.3 (ϫ2, aromatic and olefinic

CH), 127.5, 134.3, 135.2, 150.8 (quat.), 151.8, 171.0 (C=O) ppm.

IR: ν_max= 3320 (broad), 3303, 3004, 2954, 2882, 1714, 1593, 1484,

˜

1453, 1380, 1328, 1302, 1261, 1161, 1136, 1024, 974, 939, 907 cm^–1.

GC-MS: R_t= 11.31 min. MS: m/z (%) = 357 (23.6) [M]⁺, 356 (7.7),

329 (9.3), 313 (9.3), 312 (17.0), 280 (100.0), 274 (62.6), 230 (19.7),

228 (6.9), 221 (6.3), 220 (13.3), 218 (13.4), 217 (21.6), 216 (11.7),

206 (13.5), 204 (25.3), 203 (18.6), 202 (18.5), 192 (18.4), 191 (32.3),

190 (30.5), 189 (14.4), 180 (9.4), 178 (15.1), 176 (8.7), 166 (10.2),

165 (18.3), 164 (13.1), 163 (14.1), 154 (29.9), 152 (13.4), 139 (12.4),

128 (11.0), 127 (12.0), 115 (10.5), 77 (95.0), 65 (33.7), 63 (16.2), 51

(33.7), 43 (36.9), 39 (35.1). C₂₃H₁₉NO₃(357.40): calcd. C 77.29, H

5.36, N 3.92; found C 77.1, H 5.5, N 3.8.

Phenyl (2R)-4-[(R)-5-[(tert-Butyldimethylsilyl)oxy]pent-1-yn-3-yl]-2-

ethynyl-1,2-dihydroquinoline-1-carboxylate (71a): This compound

was prepared from (2R,3ЈR)-50a by the same procedure as de-

scribed above for the synthesis of (R)-41. Chromatography was car-

ried out with PE/Et₂O 9:1 to 8:2; yield 93%. R_f= 0.63 (PE/Et₂O

IR: ν

= 3485 (br), 2955, 1725, 1594, 1483, 1452, 1375, 1300,

˜

max

1192, 1008, 907 cm^–1. GC-MS: R_t= 12.59 min. MS: m/z (%) = 477

(1.8) [M]⁺, 400 (7.5), 346 (36.4), 294 (5.1), 278 (6.2), 226 (7.4), 206

(6.4), 194 (4.9), 180 (14.5), 156 (7.9), 151 (6.9), 117 (10.5), 94 (6.2),

77 (35.3), 75 (35.1), 73 (100.0), 59 (6.3), 45 (9.5), 43 (67.5).

C₂₇H₃₁NO₅Si (477.62): calcd. C 67.90, H 6.54, N 2.93; found C

67.75, H 6.65, N 2.85.

1

7:3). [α]_D= +244.3 (c = 1, CHCl₃). H NMR (200 MHz, CDCl₃):

δ = 0.12 (s, 3 H, CH₃Si), 0.13 (s, 3 H, CH₃Si), 0.97 [s, 9 H,

(CH₃)₃C], 1.90 (ddt, J_d= 9.9, 13.6, J_t= 4.0 Hz, 1 H, CHHCH₂O),

2.07–2.25 (m, 1 H, CHHCH₂O), 2.247 (d, J = 2.2 Hz, 1 H, CϵCH),

2.252 (d, J = 2.4 Hz, 1 H, CϵCH), 3.76–4.03 (m, 3 H, CH₂OSi,

CHCH₂CH₂O), 6.04 (dd, J = 2.0, 6.7 Hz, 1 H, 2-H), 6.28 (d, J =

6.7 Hz, 1 H, 3-H), 7.14–7.47 (m, 7 H), 7.68 (dd, J = 1.2, 7.8 Hz, 1

H, 5-H or 8-H), 7.78 (br. d, J = 8.0 Hz, 1 H, 5-H or 8-H) ppm.

¹³C NMR (50 MHz, CDCl₃): δ = –5.42, –5.36 (CH₃Si), 18.3 [(Si–

CCH₃)₃], 25.9 [C(CH₃)₃], 29.6 (CHCH₂CH₂O), 38.4 (CH₂CH₂O),

44.1 (CHN), 60.3 (CH₂OSi), 71.2, 71.9 (CϵCH) 79.8, 84.7

(CϵCH), 121.6 (ϫ2), 121.9 (broad), 123.6, 124.7, 125.0, 125.8,

128.1, 129.4 (ϫ2, aromatic and olefinic CH), 126.4 (broad), 134.0,

Phenyl (2R)-4-[(R)-5-Acetoxy-1,1-dibromopent-1-en-3-yl]-2-trime-

thylsilylethynyl-1,2-dihydroquinoline-1-carboxylate (70a): This com-

pound was prepared in 77% yield from (2R,2ЈS)-69a by the pro-

cedure described above for 47a. Chromatography (PE/Et₂O 8:2 to

6:4) afforded pure (2R,3ЈS)-70a as an oil. R_f= 0.54 (PE/Et₂O 6:4).

1

[α]_D= +230.7 (c = 2.1, CHCl₃). H NMR (200 MHz, CDCl₃): δ =

0.06 [s, 9 H, (CH₃)₃Si], 1.90–2.30 (m, 2 H, CH₂CH₂OAc), 2.11 (s,

3 H, CH₃CO), 3.88 (dt, J_d= 5.6, J_t= 9.2 Hz, 1 H, 3Ј-H), 4.21 (t,

J = 6.4 Hz, 2 H, CH₂OAc), 5.96 and 6.00 (AB syst., J = 6.9 Hz, 2

H, 2-H and 3-H), 6.28 (d, J = 10.0 Hz, 1 H, CH=CBr₂), 7.15–7.51

(m, 8 H), 7.76 (br. d, J = 7.8 Hz, 1 H, 5-H or 8-H) ppm. ¹³C NMR

(50 MHz, CDCl₃): δ = –0.3 (CH₃Si), 21.1 (CH₃CO), 31.9

(CH₂CH₂OAc), 40.5 (CHCH₂CH₂O), 44.9 (CHN), 61.9 (CH₂O),

89.3, 91.4, 100.7 (CϵC and CBr₂) 121.6 (ϫ2), 122.1, 123.3, 125.0

(ϫ2), 125.8, 128.2, 129.4 (ϫ2, aromatic CH and H-3), 126.8, 134.3,

135.4, 150.9 (quat.), 139.5 (CH=CBr₂), 151.9, 170.9 (C=O) ppm.

135.4, 150.9 (quat.), 151.9 (C=O) ppm. IR: ν

= 3303, 3000,

˜

max

2948, 2927, 2854, 1716, 1593, 1485, 1379, 1329, 1302, 1259, 1162,

1107, 977, 938 cm^–1. C₂₉H₃₃NO₃Si (471.66): calcd. C 73.85, H 7.05,

N 2.97; found C 73.6, H 7.1, N 2.8.

Phenyl (2S)-4-[(S)-5-[(tert-Butyldimethylsilyl)oxy]pent-1-yn-3-yl]-2-

ethynyl-1,2-dihydroquinoline-1-carboxylate (71a): The dibromide

(2S,3ЈS)-76a (761.6 mg, 1.08 mmol) was converted into (2S,3ЈS)-

77a by the general procedure already described for (R,R)-48a. The

product was chromatographed with PE/Et₂O 95:5 to 9:1, to give

(2S,3ЈS)-77a (535 mg). ¹H NMR, however, showed that it was con-

taminated with starting material (9.5%), which could not be sepa-

IR: ν

= 3002, 2958, 2871, 1716, 1594, 1484, 1452, 1380, 1327,

˜

max

1225, 1109, 1003, 962 cm^–1. C₂₈H₂₉Br₂NO₄Si (631.43): calcd. C

53.26, H 4.63, N 2.22; found C 53.5, H 4.8, N 2.2.

Phenyl (2R)-2-Ethynyl-4-[(R)-5-hydroxypent-1-yn-3-yl]-1,2-dihy- rated chromatographically. This compound was thus not fully char-

droquinoline-1-carboxylate (50a): The dibromide (2R,3ЈR)-70a

(353.6 mg, 0.56 mmol) was converted into phenyl (2R)-4-[(R)-5-

acetoxypent-1-yn-3-yl]-2-(trimethylsilyl)ethynyl-1,2-dihydroquino-

line-1-carboxylate by the general procedure already described for

(R,R)-48a. This crude product, also containing the corresponding

acterised, but directly converted into 71a. The mixture (0.98 mmol)

was dissolved in EtOH (96%, 14.6 mL) and cooled to –15 °C. It

was treated with a solution of AgNO₃(333 mg, 1.96 mmol) in H₂O

(1.4 mL). A precipitate suddenly formed. After 1 h at –15 °C the

reaction was complete. Then a solution of KCN (376 mg,

2784

www.eurjoc.org

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

5.77 mmol) in H₂O (2.2 mL) was added. The mixture was stirred

immediately chromatographed (PE/Et₂O 8:2 to 7:3) to give pure

at 0 °C for 15 min and was then diluted with H₂O (50 mL) and

73a as a white foam (357 mg, 85%). R_f= 0.67 (toluene/CH₂Cl₂/

extracted rapidly with Et₂O. The organic extracts were immediately Et₂O 90:5:5), 0.46 (PE/Et₂O 7:3). [α]_D= +74.7 (c = 2.5, CH₂Cl₂).

washed with a mixture of KH₂PO₄(1 , 20 mL) and brine (20 mL).

Drying (Na₂SO₄), concentration and chromatography (PE/Et₂O 9:1

to 8:2) gave pure (2S,3ЈS)-71a as a foam (308.2 mg, 60% from 76a).

[α]_D= –108.8 (c = 1.7, CHCl₃). The optical purity was estimated

to be 42%. All other analytical and spectroscopic data were iden-

tical to those of the enantiomer.

¹H NMR (200 MHz, CDCl₃): δ = 0.13 (s, 3 H, CH₃Si), 0.14 (s, 3

H, CH₃Si), 0.97 [s, 9 H, (CH₃)₃C], 1.60–1.80 (m, 1 H, CHHCH₂O),

2.00–2.20 (m, 1 H, CHHCH₂O), 3.80–3.97 and 4.04–4.18 (2 m, 2

ϫ 2 H, m, CH₂OSi, CHCH₂, 3-H), 6.04 (d, J = 2.8 Hz, 1 H, 2-H),

7.05–7.44 (m, 7 H), 7.57 (br. d, J = 7.6 Hz, 1 H, 5-H or 8-H), 7.70

(d, J = 7.8 Hz, 1 H, 5-H or 8-H) ppm. ¹³C NMR (50 MHz, CDCl₃):

δ = –5.4, –5.3 (CH₃Si), 14.1, 14.3 (CI), 18.3 [(SiCCH₃)₃], 26.0

[C(CH₃)₃], 32.2 (CHCH₂CH₂O), 35.5 (CH₂CH₂O), 45.5 (CHN),

56.1 (C-4), 60.1 (CH₂OSi), 63.0 (C-3), 87.8, 91.1 (CϵCI), 121.5

(ϫ2), 125.7, 125.8, 127.4, 127.7, 128.7, 129.4 (ϫ2, aromatic CH),

125.3, 135.1, 151.0 (quat.), 152.9 (C=O) ppm. C₂₉H₃₁I₂NO₄Si

(739.46): calcd. C 47.10, H 4.23, N 1.89; found C 47.35, H 4.4, N

1.9.

Phenyl (2S,3S,4R)-4-[(R)-5-[(tert-Butyldimethylsilyl)oxy]pent-1-yn-

3-yl]-3,4-epoxy-2-ethynyl-2H-quinoline-1-carboxylate (72a): A solu-

tion of (2R,3ЈR)-71a (306.3 mg, 0.65 mmol) in dry CH₂Cl₂(6 mL)

was cooled to 0 °C, and treated with meta-chloroperbenzoic acid

(80% purity) (288 mg, 1.33 mmol). The solution rapidly turned red,

and a precipitate formed. After 1 h the temperature was raised to

room temp., and the mixture was stirred at this temperature for

20 h. After the system had again been cooled to 0 °C, Me₂S (93 µL)

and saturated aqueous NaHCO₃(18 mL) were added. Extraction

with Et₂O, washing with brine, concentration and chromatography

(PE/Et₂O 8:2) gave pure 72a as a white foam (296.9 mg, 94%). R_f

Enediyne 74a: A solution of the diiodide 73a (79.0 mg, 107 µmol)

in dry DMF (10 mL) was degassed and put under argon. Freshly

dried lithium chloride (23 mg, 543 µmol) and Pd(PPh₃)₄(20 mg,

17.3 µmol) were rapidly added, and the mixture was again put un-

der argon through a series of vacuum/Ar cycles. A solution of (Z)-

bis(trimethylstannyl)ethylene (33 µL, 139.4 µmol) in dry DMF

(10 mL) was transferred into a dropping funnel. After the diiodide

solution had been warmed to 70 °C, the stannylene solution was

slowly added over 1 h. After stirring at the same temperature for

1 h, the orange solution was poured into H₂O (50 mL) and ex-

tracted four times with Et₂O. The organic extracts were washed

with H₂O (40 mL) and brine, dried (Na₂SO₄), concentrated to dry-

ness and immediately chromatographed (PE/Et₂O 9:1 to 75:25) to

give pure 74a as a foam (30.7 mg, 56%). R_f= 0.52 (PE/Et₂O 70:30).

A side product (the acyclic doubly stannylated enediyne, 11.2 mg)

= 0.57 (PE/Et₂O 7:3), 0.57 (toluene/CH₂Cl₂/Et₂O 90:5:5). [α]_D

=

1

+129.2 (c = 2, CH₂Cl₂). H NMR (200 MHz, CDCl₃): δ = 0.12 (s,

3 H, CH₃Si), 0.14 (s, 3 H, CH₃Si), 0.97 [s, 9 H, (CH₃)₃C], 1.71 (ddt,

J_d= 10.8, 13.6, J_t= 3.2 Hz, 1 H, CHHCH₂O), 2.04–2.20 (m, 1 H,

CHHCH₂O), 2.19 (d, J = 2.6 Hz, 1 H, CϵCH), 2.22 (d, J = 2.2 Hz,

1 H, CϵCH), 3.80–4.03 (m, 3 H, CH₂OSi, CHCH₂CH₂O), 4.12 (d,

1 H, 3-H), 5.92 (t, J = 2.7 Hz, 1 H, 2-H), 7.05–7.44 (m, 7 H), 7.57

(br. d, J = 8.0 Hz, 1 H, 5-H or 8-H), 7.76 (d, J = 7.0 Hz, 1 H, 5-

H or 8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –5.43, –5.37

(CH₃Si), 18.3 [(SiCCH₃)₃], 25.9 [C(CH₃)₃], 29.9 (CHCH₂CH₂O),

35.5 (CH₂CH₂O), 43.7 (CHN), 55.9 (C-4), 60.0 (CH₂OSi), 62.8 (C-

3), 72.2, 74.2 (CϵCH) 77.4, 80.9 (CϵCH), 121.5 (ϫ 2), 125.7,

125.8, 127.5 (broad), 127.8, 128.6, 129.3 (ϫ2, aromatic CH), 125.5

1

was also obtained (R_f= 0.76). [α]_D= +394.2 (c = 1.5, CHCl₃). H

NMR (200 MHz, CDCl₃): δ = 0.09 (s, 6 H, CH₃Si), 0.91 [s, 9 H,

(CH₃)₃C], 2.30–2.55 (m, 2 H, CH₂CH₂O), 3.08 (dd, J = 6.2, 7.4 Hz,

1 H, CHCH₂CH₂O), 3.70–3.92 (m, 2 H, CH₂O), 3.88 (d, J =

2.9 Hz, 1 H, 3-H), 5.65 (dt, J_d= 9.9, J_t= 2.6 Hz, 1 H, CH=CH),

5.80 (dd, J = 0.8, 9.9 Hz, 1 H, CH=CH), 5.88 (dd, J = 1.6, 2.9 Hz,

1 H, 2-H), 7.10–7.28 (m, 4 H), 7.29–7.42 (m, 3 H), 7.55 (br. d, J =

7.8 Hz, 1 H, 5-H or 8-H), 7.85 (dd, J = 1.3, 7.9 Hz, 1 H, 5-H or

8-H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –5.34, –5.28 (CH₃Si),

18.3 [(SiCCH₃)₃], 25.9 [C(CH₃)₃], 33.1 (CH₂CH₂O), 40.0

(CHCH₂CH₂O), 46.0 (CHN), 59.5 (C-4), 61.3 (CH₂OSi), 69.8 (C-

3), 88.3, 91.1, 92.0, 102.0 (CϵC), 121.3, 121.6 (ϫ2), 125.3, 125.5,

125.8, 128.2, 129.2, 129.4 (ϫ3, aromatic or alkenylic CH), 127.0,

(broad), 135.3, 151.0 (quat.), 153.4 (C=O) ppm. IR: ν

= 3303,

˜

max

3004, 2953, 2927, 2854, 1721, 1593, 1489, 1380, 1323, 1288, 1204,

1085, 970, 915 cm^–1. GC-MS: R_t13.1. m/z 472 (0.5) [M – 15]⁺, 430

(56.4), 337 (7.4), 336 (7.0), 310 (7.0), 309 (7.9), 308 (20.2), 294

(14.3), 290 (10.5), 280 (5.5), 278 (6.0), 274 (6.0), 262 (9.1), 244 (6.1),

230 (5.2), 228 (5.7), 218 (14.7), 217 (17.3), 206 (10.2), 204 (15.0),

191 (16.1), 180 (9.1), 178 (10.6), 151 (27.7), 140 (10.2), 115 (13.6),

89 (23.8), 83 (13.3), 77 (73.4), 75 (63.4), 73 (100.0), 65 (12.1), 59

(19.9), 57 (15.6), 51 (16.7), 41 (10.2), 39 (11.5). C₂₉H₃₃NO₄Si

(487.66): calcd. C 71.42, H 6.82, N 2.87; found C 71.3, H 6.8, N

2.8.

135.7, 151.0 (quat.), 152.9 (C=O) ppm. IR: ν

= 3028, 3002,

˜

max

2951, 2927, 2855, 1723, 1599, 1378, 1321, 1187, 1107, 907 cm^–1.

C₃₁H₃₃NO₄Si (511.68): calcd. C 72.77, H 6.50, N 2.74; found C

72.65, H 6.55, N 2.7.

Phenyl (2R,3R,4S)-4-[(S)-5-[(tert-Butyldimethylsilyl)oxy]pent-1-yn-

3-yl]-3,4-epoxy-2-ethynyl-2H-quinoline-1-carboxylate (ent-72a):

This compound was obtained from (2S,3ЈS)-71a by the same pro-

cedure as described above for its enantiomer. [α]_D= –59.7 (c = 2,

CH₂Cl₂). The optical purity was estimated to be 42%.

Phenyl (2S)-4-[(R)-4-[(tert-Butyldimethylsilyl)oxy]-1-hydroxybut-2-

yl]-2-[(trimethylsilyl)ethynyl]-1,2-dihydroquinoline-1-carboxylate

(75a): A solution of (2S,2ЈR)-67a (842 mg, 1.26 mmol) in CH₂Cl₂

(50 mL) was treated with aqueous KH₂PO₄(0.1 , 6.4 mL) and

dichlorodicyanobenzoquinone (DDQ, 438 mg, 1.93 mmol). After

stirring for 1 h at room temp., the reaction mixture was quenched

with aqueous NaHCO₃(5%) and extracted with CH₂Cl₂. Washing

with brine and concentration gave a crude product, which was

chromatographed (PE/Et₂O 6:4 to 4:6) to give pure (2S,2ЈR)-75a as

a foam (550 mg, 80%). R_f= 0.44 (PE/Et₂O 4:6). [α]_D= –146.3 (c

= 1.1, CH₂Cl₂). ¹H NMR (200 MHz, CDCl₃): δ = 0.04 [s, 9 H,

(CH₃)₃Si], 0.08 (s, 3 H, CH₃Si), 0.10 (s, 3 H, CH₃Si), 0.94 [s, 9 H,

(CH₃)₃C], 1.96 (q, J = 5.9 Hz, 2 H, CH₂CH₂O), 2.73 (t, J = 6.4 Hz,

Phenyl (2S,3S,4R)-4-[(R)-5-[(tert-Butyldimethylsilyl)oxy]-1-iodo-

pent-1-yn-3-yl]-3,4-epoxy-2-iodoethynyl-2H-quinoline-1-carboxylate

(73a): Iodine (868 mg, 3.42 mmol) was suspended in dry benzene

(15 mL) in the dark, and treated with morpholine (895 µL,

10.26 mmol). The mixture was stirred for 40 min at room temp.

The resulting suspension of an orange precipitate in a relatively

clear solution was treated with a solution of the diyne 72a

(277.8 mg, 570 µmol) in benzene (5 mL). After stirring for 46 h at

room temp., the mixture was poured into Na₂S₂O₃(0.4 , 35 mL)

and extracted three times with Et₂O (pH = 9). The organic extracts

were washed in order with: a) NaH₂PO₄(1 , 30 mL) + citric acid

(0.5 , 7 mL, pH = 3), b) Na₂S₂O₃(0.1 , 20 mL), and c) brine. 1 H, OH), 3.20 (quint., J = 4.9 Hz, 1 H, CHCH₂CH₂O), 3.52–3.96

After drying (Na₂SO₄) and concentration, the crude product was (m, 4 H, CH₂O), 5.93 and 5.96 (AB system, 2 H, 2-H and CH=C),

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2785

R. Riva, L. Banfi, A. Basso, V. Gandolfo, G. Guanti

FULL PAPER

[12] R. Unno, H. Michishita, H. Inagaki, Y. Baba, T. Jomori, T.

Nishikawa, M. Isobe, Chem. Pharm. Bull. 1997, 45, 125–133.

[13] R. Unno, H. Michishita, H. Inagaki, Y. Suzuki, Y. Baba, T.

Jomori, T. Nishikawa, M. Isobe, Bioorg. Med. Chem. 1997, 5,

883–901.

[14] R. Unno, H. Michishita, H. Inagaki, Y. Suzuki, Y. Baba, T.

Jomori, M. Moku, T. Nishikawa, M. Isobe, Bioorg. Med.

Chem. 1997, 5, 903–919.

[15] R. Unno, H. Michishita, H. Inagaki, Y. Suzuki, Y. Baba, T.

Jomori, T. Nishikawa, M. Isobe, Bioorg. Med. Chem. 1997, 5,

987–999.

[16] P. A. Wender, C. K. Zercher, S. Beckham, E. M. Haubold, J.

Org. Chem. 1993, 58, 5867–5869.

7.14–7.44 (m, 7 H), 7.51 (dd, J = 1.5, 7.7 Hz, 1 H, 5-H or 8-H),

7.76 (br. d, J = 7.4 Hz, 1 H, 5-H or 8-H) ppm. ¹³C NMR (50 MHz,

CDCl₃): δ = –5.4 [(CH₃)₂Si], –0.2 [(CH₃)₃Si], 18.3 [(SiCCH₃)₃], 25.9

[C(CH₃)₃], 34.5 (CH₂CH₂O), 39.4 (CHCH₂CH₂O), 45.0 (CHN),

61.4, 65.7 (CH₂O), 88.7, 101.3 (CϵC), 121.6 (ϫ2), 121.8 (broad),

123.3, 124.9 (ϫ2), 125.7, 127.8, 129.4 (ϫ2, aromatic or olefinic

CH), 127.7, 134.4, 136.2, 150.9 (quat.), 151.9 (C=O) ppm. IR: ν

˜

max

= 3387 (broad), 3004, 2952, 2927, 2875, 2855, 1714, 1592, 1484,

1379, 1327, 1302, 1244, 1080, 963, 837 cm^–1. C₃₁H₄₃NO₄Si₂

(549.85): calcd. C 67.72, H 7.88, N 2.55; found C 67.9, H 7.8, N

2.45.

Phenyl (2S)-4-[(S)-5-[(tert-Butyldimethylsilyl)oxy]-1,1-dibromopent-

1-en-3-yl]-2-trimethylsilylethynyl-1,2-dihydroquinoline-1-carboxylate

(76a): This compound was prepared in 75% yield from (2S,2ЈR)-

75a by the procedure described above for 47a. Chromatography

(PE/Et₂O 95:5 to 85:15) afforded pure (2S,3ЈS)-76a as an oil. R_f=

0.68 (PE/Et₂O 80:20). [α]_D= –92.6 (c = 2.1, CHCl₃). ¹H NMR

(200 MHz, CDCl₃): δ = 0.04 [s, 9 H, (CH₃)₃Si], 0.06 and 0.09 [2ϫs,

2ϫ3 H, (CH₃)₂Si], 0.94 [s, 9 H, (CH₃)₃CSi], 1.77–2.14 (m, 2 H,

[17] S. C. Sinha, L. S. Li, G. P. Miller, S. Dutta, C. Rader, R. A.

Lerner, Proc. Natl. Acad. Sci. USA 2004, 101, 3095–3099.

[18] M. P. Hay, W. R. Wilson, W. A. Denny, Bioorg. Med. Chem.

Lett. 1999, 9, 3417–3422.

[19] M. P. Hay, W. R. Wilson, W. A. Denny, Bioorg. Med. Chem.

Lett. 1995, 5, 2829–2834.

[20] G. Guanti, R. Riva, Chem. Commun. 2000, 1171–1172.

[21] G. Guanti, R. Riva, Org. Biomol. Chem. 2003, 1, 3967–3976.

[22] M. E. Maier, F. Bosse, A. J. Niestroj, Eur. J. Org. Chem. 1999,

1–13.

CH₂CH₂OSi), 3.71 (t, J = 5.8 Hz, 2 H, CH₂OSi), 3.97 (dt, J_d

=

9.4, J_t= 7.2 Hz, 1 H, 3Ј-H), 5.95 (s, 2 H, 2-H and 3-H), 6.38 (d, J

= 9.8 Hz, 1 H, CH=CBr₂), 7.15–7.44 (m, 7 H), 7.57 (dd, J = 1.5,

8.0 Hz, 1 H, 5-H or 8-H), 7.75 (br. d, J = 7.8 Hz, 1 H, 5-H or 8-

H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = –5.4, –0.2 (CH₃Si), 18.2

[ C( C H₃)₃] , 2 6 . 0 [(CH₃)₃C ], 36 . 6 (CH₂CH₂O Si ), 39 . 9

(CHCH₂CH₂O), 44.9 (CHN), 59.9 (CH₂O), 89.0, 90.4, 101.1 (CϵC

and CBr₂) 121.7 (ϫ2), 121.9, 123.6, 124.9 (ϫ2), 125.8, 128.0, 129.4

(ϫ2, aromatic CH and H-3), 127.3, 134.2, 135.9, 150.9 (quat.),

[23] K. C. Nicolaou, Y. P. Hong, W. M. Dai, Z. J. Zeng, W. Wras-

idlo, J. Chem. Soc., Chem. Commun. 1992, 1542–1545.

[24] T. Nishikawa, M. Yoshikai, K. Obi, T. Kawai, R. Unno, T.

Jomori, M. Isobe, Tetrahedron 1995, 51, 9339–9352.

[25] L. Banfi, A. Basso, E. Bevilacqua, V. Gandolfo, G. Giannini,

G. Guanti, L. Musso, M. Paravidino, R. Riva, C. Scapolla,

Bioorg. Med. Chem. 2008, 16, 3501–3518.

[26] For pioneering work on enzymatic monoacetylation of prochi-

ral diols see refs [27–35].

[27] U. Ader, D. Breitgoff, P. Klein, K. E. Laumen, M. P. Schneider,

Tetrahedron Lett. 1989, 30, 1793–1796.

[28] G. M. Ramos Tombo, H. P. Schar, X. Fernandez, I. Busquets,

O. Ghisalba, Tetrahedron Lett. 1986, 27, 5707–5710.

[29] Y. Terao, M. Murata, K. Achiwa, T. Nishio, M. Akamtsu, M.

Kamimura, Tetrahedron Lett. 1988, 29, 5173–5176.

[30] Y. F. Wang, J. J. Lalonde, M. Momongan, D. E. Bergbreiter,

C. H. Wong, J. Am. Chem. Soc. 1988, 110, 7200–7205.

[31] C. J. Barnett, T. M. Wilson, Tetrahedron Lett. 1989, 30, 6291–

6294.

140.6 (CH=CBr ), 151.9 (C=O) ppm. IR: ν_max= 3003, 2951, 2855,

˜

2

1712, 1595, 1475, 1419, 1381, 1329, 1193, 1040, 924 cm^–1

.

C₃₂H₄₁Br₂NO₃Si₂(703.65): calcd. C 54.62, H 5.87, N 1.99; found

C 54.8, H 6.0, N 1.95.

Acknowledgments

We wish to thank Ms. Laura Bondanza for her valuable collabora-

tion in this work and the University of Genova, and the Italian

Ministero dellЈUniversità e della Ricerca (PRIN 2000) for financial

assistance.

[32] K. Mori, N. Chiba, Liebigs Ann. Chem.1989, 957–962.

[33] Z. F. Xie, H. Suemune, K. Sakai, J. Chem. Soc., Chem. Com-

mun. 1988, 1638–1639.

[34] G. Guanti, L. Banfi, E. Narisano, J. Org. Chem. 1992, 57,

1540–1554.

[35] P. Grisenti, P. Ferraboschi, A. Manzocchi, E. Santaniello, Tet-

rahedron 1992, 48, 3827–3834.

[1] G. Guanti, L. Banfi, A. Basso, R. Riva, in Polyynes (Ed.: F.

Cataldo), CRC, Boca Raton (FL), 2006, pp. 453–491.

[2] A. L. Smith, K. C. Nicolaou, J. Med. Chem. 1996, 39, 2103–

2117.

[36] S. J. Danishefsky, M. D. Shair, J. Org. Chem. 1996, 61, 16–44.

[37] Although compounds 3–5 each have four stereogenic centres,

the relative configuration of these in the oxirane ring is fixed

for steric reasons. Moreover, it would be very difficult to obtain

compounds in which the epoxide oxygen was on the same side

of the enediyne ring.

[3] K. C. Nicolaou, W. M. Dai, Angew. Chem. Int. Ed. Engl. 1991,

30, 1387–1416.

[4] K. C. Nicolaou, A. L. Smith, E. W. Yue, Proc. Natl. Acad. Sci.

USA 1993, 90, 5881–5888.

[5] M. Konishi, H. Ohkuma, T. Tsuno, T. Oki, G. D. Vanduyne, J.

Clardy, J. Am. Chem. Soc. 1990, 112, 3715–3716.

[6] J. Davies, H. Wang, T. Taylor, K. Warabi, X. H. Huang, R. J.

Andersen, Org. Lett. 2005, 7, 5233–5236.

[7] K. C. Nicolaou, J. S. Chen, H. J. Zhang, A. Montero, Angew.

Chem. Int. Ed. 2008, 47, 185–189.

[8] K. C. Nicolaou, W. M. Dai, Y. P. Hong, S. C. Tsay, K. K. Bald-

ridge, J. S. Siegel, J. Am. Chem. Soc. 1993, 115, 7944–7953.

[9] K. C. Nicolaou, W. M. Dai, S. C. Tsay, V. A. Estevez, W. Wras-

idlo, Science 1992, 256, 1172–1178.

[10] K. C. Nicolaou, W. M. Dai, S. V. Wendeborn, A. L. Smith, Y.

Torisawa, P. Maligres, C. K. Hwang, Angew. Chem. Int. Ed.

Engl. 1991, 30, 1032–1036.

[11] K. C. Nicolaou, A. L. Smith, S. V. Wendeborn, C. K. Hwang,

J. Am. Chem. Soc. 1991, 113, 3106–3114.

[38] L. Banfi, A. Basso, V. Gandolfo, G. Guanti, R. Riva, Tetrahe-

dron Lett. 2004, 45, 4221–4223.

[39]

[40]

[41]

[42]

[43]

[44]

[45]

L. Banfi, G. Guanti, A. Mugnoli, R. Riva, Tetrahedron: Asym-

metry 1998, 9, 2481–2492.

G. Guanti, S. Perrozzi, R. Riva, Tetrahedron: Asymmetry 2002,

13, 2703–2726.

R. Yamaguchi, E. Hata, K. Utimoto, Tetrahedron Lett. 1988,

15, 1785–1788 .

R. Yamaguchi, Y. Nakazono, K. Mituyosi, Tetrahedron Lett.

1983, 17, 1801–1804.

F. Waldheimer, S. De Bernardini, C. A. Leach, T. Tamm, Helv.

Chim. Acta 1982, 65, 2472–2475.

The introduction of the MOM group at the quinoline level (e.g.

on 18 or 56) was not successful.

E. J. Corey, P. L. Fuchs, Tetrahedron Lett. 1972, 13, 3769–3772.

2786

www.eurjoc.org

Eur. J. Org. Chem. 2010, 2768–2787

Complex Natural Compound Analogues

[46] D. Grandjean, P. Pale, J. Chuche, Tetrahedron Lett. 1994, 35,

3529–3530.

[47] O. Gebauer, R. Brueckner, Synthesis 2000, 588–602.

[48] S. Ohira, Synth. Commun. 1989, 19, 561–564.

[49] T. Okazoe, K. Takai, K. Utimoto, J. Am. Chem. Soc. 1987,

109, 951–953.

[50] R. M. Pike, D. W. Mayo, S. S. Butcher, D. J. Butcher, R. J. Hin-

kle, J. Chem. Educ. 1986, 63, 917–918.

[51] M. Schlosser, B. Schaub, Chimia 1982, 36, 396–397.

[52] E. Levas, D. Raulet, Bull. Soc. Chim. Fr. 1971, 71, 2598–2606.

[53] F. Ramirez, N. B. Desai, N. McKelvie, J. Am. Chem. Soc. 1962,

84, 1745–1747.

[61]

[62]

W. Boland, U. Niedermeyer, L. Jaenicke, H. Goerisch, Helv.

Chim. Acta 1985, 68, 2062–2073.

F. Marcacci, G. Giacomelli, R. Menicagli, Gazz. Chim. Ital.

1980, 110, 195–200.

[63]

[64]

F. E. Ziegler, J. S. Tung, J. Org. Chem. 1991, 56, 6530–6537.

M. C. McIntosh, S. M. Weinreb, J. Org. Chem. 1993, 58, 4823–

4832.

C. Caderas, R. Lett, L. E. Overman, M. H. Rabinowitz, L. A.

Robinson, M. J. Sharp, J. Zablocki, J. Am. Chem. Soc. 1996,

118, 9073–9082.

H. J. Bestmann, H. Frey, Liebigs Ann. Chem. 1980, 2061–2071.

A. Rolf, Angew. Chem. Int. Ed. Engl. 1975, 14, 801–811.

L. Banfi, A. Basso, G. Guanti, M. Paravidino, R. Riva, C.

Scapolla, Arkivoc 2006, vi, 15–39.

L. Banfi, A. Basso, G. Guanti, Tetrahedron 1997, 53, 3249–

3268.

T. N. Mitchell, A. Amamria, H. Killing, D. Rutschow, J. Or-

ganomet. Chem. 1986, 304, 257–265.

D. K. Moss, J. D. Spence, M. H. Nantz, J. Org. Chem. 1999,

64, 4339–4343.

A. G. Myers, N. J. Tom, M. E. Fraley, S. B. Cohen, D. J. Ma-

dar, J. Am. Chem. Soc. 1997, 119, 6072–6094.

Received: December 18, 2009

[65]

[66]

[67]

[68]

[54] I. V. Ekhato, C. H. Robinson, J. Org. Chem. 1989, 54, 1327–

1331.

[55] E. Piers, A. V. Gavai, J. Org. Chem. 1990, 55, 2374–2379.

[56] J. D. White, G. L. Bolton, A. P. Dantanarayana, C. M. J. Fox,

R. N. Hiner, R. W. Jackson, K. Sakuma, U. S. Warrier, J. Am.

Chem. Soc. 1995, 117, 1908–1939.

[57] H. Frey, G. Kaupp, Synthesis 1990, 931–934.

[58] L. F. Tietze, T. Neumann, M. Kajino, M. Pretor, Synthesis

1995, 1003–1006.

[59] A. G. M. Barrett, M. L. Boys, T. L. Boehm, J. Org. Chem.

1996, 61, 685–699.

[60] S. Sugiyama, M. Honda, T. Komori, Liebigs Ann. Chem. 1990,

1069–1078.

[69]

[70]

[71]

[72]

Published Online: April 6, 2010

Eur. J. Org. Chem. 2010, 2768–2787

www.eurjoc.org

2787

Article Doi

Enzymatically asymmetrised chiral building blocks for the synthesis of complex natural product analogues: The synthesis of dynemicin analogues from 2-(quinolin-4-yl)propane-1, 3-diol

DOI: 10.1002/ejoc.200901478

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejoc.200901478

Full text of DOI:10.1002/ejoc.200901478

Products guided by the article

R&D Labs maybe for 1233057-61-8

Relevant to this article

Hot Product