Novel synthesis of (d,l) trans-chrysanthemic acid involving a β-diketone fragmentation

Article abstract of DOI:10.1016/S0040-4039(02)01270-4

Methyl (d,l) trans-chrysanthemate as well as its cis-diastereoisomer have been prepared from dimethyl dimedone, one of their isomers, in a few steps and with complete control of the relative stereochemistry.

Full text of DOI:10.1016/S0040-4039(02)01270-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 6167–6168
Novel synthesis of (d,l) trans-chrysanthemic acid involving a
b-diketone fragmentation
Alain Kriefa,* and Stephane Jeanmarta,b
aLaboratoire de Chimie Organique de Synthe`se, Department of Chemistry, Faculte´s Universitaires Notre-Dame de la Paix,
61 rue de Bruxelles, Namur 5000, Belgium
bFonds pour la Formation a` la Recherche dans l’Industrie et l’Agriculture, 5 rue d’Egmont, Bruxelles 1000, Belgium
Received 27 June 2002; accepted 28 June 2002
Abstract—Methyl (d,l) trans-chrysanthemate as well as its cis-diastereoisomer have been prepared from dimethyl dimedone, one
of their isomers, in a few steps and with complete control of the relative stereochemistry. © 2002 Elsevier Science Ltd. All rights
reserved.
We report a novel synthesis, from dimethyl dimedone 2,
of the methyl ester 1atrans of (d,l) trans-chrysanthemic
acid, a constituent of pyrethrin I, a natural insecticide
(Scheme 1).
(iii) an unusually high propensity of the trans-isopropyl-
idene enolate to be oxidised by the oxygen dissolved in
the DMSO (Scheme 2, steps d and e).
This transformation takes advantage of the easy oxida-
tive cyclisation of 2 to the bicyclic 1,3-diketone 3 ([i]
t-BuOK, THF, −78°C, [ii] Br2, pentane, 67%)1 and the
unexpectedly high propensity of the latter to produce
the d-hydroxy-g-keto-carboxylic acid 7trans on reaction,
performed in air, with sodium hydroxide in DMSO and
acid quench ([i] 6 equiv. NaOH, DMSO/H2O: 4/1,
70°C, 14 h, air [i] H3O+, 79% yield).
The transformation of 7trans to methyl chrysanthemate
1atrans was then readily achieved, via its methyl ester
(CH2N2, Et2O, 0°C, 100%) on reduction to the corre-
sponding diol 8trans (NaBH4, MeOH, 0°C, 3 h, 84%)
and subsequent Corey–Winter olefination3 ([i] S=
C(imid.)2, toluene, reflux, 4.5 h, 93% yield of 9trans, [ii]
P(OMe)3, 120°C, 24 h, 87% yield of 1atrans).
Scheme 1. (i) (a) t-BuOK, THF, −78°C to 40°C, (b) Br2,
pentane, 40°C, 1 h, (ii) (a) 6 equiv. NaOH, DMSO/H2O (4/1),
O2, 70°C, 14 h, (b) H3O+, (iii) (a) CH2N2, Et2O, 0°C, (b)
NaBH4, MeOH, 0°C, 3 h, (iv) SꢀC(imid.)2, toluene, reflux, 4.5
h, (v) P(OMe)2, 120°C, 24 h.
The transformation of 3 to 5trans reported above,
involves in the same pot an exceptional succession of
individual steps such as (i) 1,3-diketone fragmentation
(Scheme 2, step a),2 (ii) enolate isomerisations (Scheme
2, steps b and c) which finally lead to cis–trans isomeri-
sation on the cyclopropane ring and last but not least
* Corresponding author. Tel.: +32-81-724539; fax: +32-81-724536;
e-mail: alain.krief@fundp.ac.be
Scheme 2.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01270-4

6168
A. Krief, S. Jeanmart / Tetrahedron Letters 43 (2002) 6167–6168
Figure 1.
The related potassium enolate is much less sensitive to
such an oxidation, since potassium hydroxide instead
leads, under identical conditions, to the g-keto acid 11trans
(R=H) in very good yield (84%, 11trans/7trans: 94/6, Fig.
1).† The d-hydroxy-g-keto-carboxylic acid 7trans can be
nevertheless produced in almost quantitative yield if the
reaction is carried out under a slight pressure of oxygen
(6 equiv. KOH, oxygen, DMSO/H2O: 4/1, 70°C, 14 h,
[ii] H3O+, 98%).
Scheme3. (i)hw, benzene, 20°C, 9h, (ii)m-CPBA, dioxane/H2O,
20°C, 3 h, (iii) (a) CH2N2, Et2O, 0°C, (b) BH3·Me2S, toluene,
0°C, 0.5 h, (iv) SꢀC(imid.)2, toluene, reflux, 14 h, (v) 1,3-
dimethyl-2-phenyl-[1,3,2]diazaphospholidine, 40°C, 8 h, (vi) (a)
1.5 equiv. NaIO4, MeOH, phosphate.
All attempts to preserve on 7 or 9 (Fig. 1) the cis-configu-
ration present on 3 were unsuccessful. Therefore the
synthesis of methyl cis-chrysanthemate 1acis and of its
dibromovinylanalogue1bcis whichispartofdeltamethrin,
the most active pyrethroid insecticide, became a real
challenge.
reaction using triphenyl phosphine and carbon tetra-
bromide (20°C, 85% in 1bcis).4
Oxidation of the enol lactone 14, readily available by
photochemical rearrangement of 32 (Scheme 3), with
m-CPBA (2 equiv., dioxane/H2O: 4/1, 20°C, 3 h) directly
affords the d-hydroxy-g-keto-carboxylic acid 7cis (R=H)
in good yield (74%) with complete control of the cis-
configuration.‡,§
The synthesis of enantiopure methyl (1R)-trans-chrysan-
themate 1a* requires enantioselective ring opening of
trans
3 reminiscent of a reaction successfully performed on the
related bicyclic anhydride 12.5 The synthesis of the
(1R)-cis-cyclopropyl esters 1a* and 1b* requires a
cis
Norrish type I6 enantioselective rearrangecmis ent. To our
knowledge, both transformations have not yet been
achieved. We are working towards these ends.
Reduction of the corresponding cis-ester 7cis (R=Me)
using sodium borohydride under the conditions described
above for its trans-analogue, was troublesome. It requires
a longer time (NaBH4, MeOH, 18 h) and unfortunately
yield a mixture of the desired diol 8cis and the correspond-
ing lactone 12 (55/45 in 95% overall yield, Fig. 1).
References
1. Krief, A.; Surleraux, D.; Frauenrath, H. Tetrahedron Lett.
1988, 29, 6157–6160.
The synthesis of the required diol 8cis can be nevertheless
achieved on reduction of 7cis (R=Me) with the boron
hydride–dimethyl sulfide complex (BH3·SMe2, toluene,
0°C, 0.5 h, 91%). Its transformation to:
“ Methyl cis-chrysanthemate 1acis has been achieved by
reduction of the corresponding thiocarbonate 8cis
(S=C(imid.)2, toluene, 110°C, 14 h, 81% then 1,3-
dimethyl-2-phenyl-[1,3,2] diazaphospholidine, 40°C, 8
h, 83% in 1acis);3
“ Methyl dibromovinyl cis-chrysanthemate 1bcis
requires sequential cleavage of the diol (1.5 equiv.
NaIO4, MeOH, phosphate buffer, 20°C, 1 h, 61%)4a
leading to the aldehyde 12 followed by a Wittig-type
2. Okada, T.; Kamogawa, K.; Kawanisi, M.; Nozaki, H.
Bull. Chem. Soc. Jpn. 1970, 43, 2908–2911.
3. (a) Corey, E. J.; Winter, R. A. E. J. Am. Chem. Soc. 1963,
85, 2677–2678; (b) Block, E. Org. React. 1984, 30, 457–
566; (c) Corey, E. J.; Hopkins, P. B. Tetrahedron Lett.
1982, 23, 1979–1982.
4. (a) Krief, A.; Dumont Tetrahedron Lett. 1988, 29, 1083–
1084; (b) De Vos, M. J.; Krief, A. J. Am. Chem. Soc. 1982,
104, 4282–4283.
5. Bolm, C.; Schiffers, I.; Vinter, C. L.; Gerlach, A. J. Org.
Chem. 2000, 65, 6984–6991.
6. Horspool, W. R. Photochemistry 2001, 32, 49–73.
† Lithium hydroxide produces a 1/1 mixture of 7trans and 12trans but in
very poor yield (10%) even after a longer reaction time.
‡ The synthesis of 11cis has been effectively achieved from 14 and (i)
lithium methylate in the presence of boron trimethylborate (6.4
equiv. each, MeOH, 20°C, 7 days, 92% yield after acid hydrolysis),
(ii) barium hydroxide (aq. THF, 20°C, 1.5 h, 93% yield after acid
hydrolysis), (iii) methanol in the presence of catalytic amounts of a
Lewis acid (cat. p-TSA, toluene/MeOH: 10/1, 110°C, 24 h, 72%
yield).
§ Reaction of 14 with sodium hydroxide (6 equiv. NaOH, aq. DMSO,
70°C, 1 h) or lithium methylate in methanol (1 equiv. MeOLi,
MeOH, 20°C, 6 h) did not provide the cis-stereoisomer 9cis: (90% of
9trans cis/trans 00/100 and 81% of 9, cis/trans 71/29, respectively).