Asymmetric total synthesis of smyrindiol employing an organocatalytic aldol key step

Article abstract of DOI:10.3762/bjoc.8.123

The first organocatalytic asymmetric synthesis of smyrindiol, by using an (S)-proline catalyzed enantioselective intramolecular aldol reaction as the key step, is described. Smyrindiol was synthesized from commercially available 2,4-dihydroxybenzaldehyde in 15 steps, with excellent stereoselectivity (de = 99%, ee = 99%). In the course of this total synthesis a new and mild coumarin assembly was developed.

Full text of DOI:10.3762/bjoc.8.123

Asymmetric total synthesis of smyrindiol employing
an organocatalytic aldol key step
Dieter Enders*, Jeanne Fronert, Tom Bisschops and Florian Boeck
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2012, 8, 1112–1117.
Institute of Organic Chemistry, RWTH Aachen University,
Landoltweg 1, 52074 Aachen, Germany
doi:10.3762/bjoc.8.123
Received: 08 May 2012
Accepted: 25 June 2012
Published: 18 July 2012
Email:
Dieter Enders* - enders@rwth-aachen.de
* Corresponding author
This article is part of the Thematic Series "Organocatalysis".
Guest Editor: B. List
Keywords:
aldol reaction; asymmetric synthesis; organocatalysis; proline;
smyrindiol
© 2012 Enders et al; licensee Beilstein-Institut.
License and terms: see end of document.
Abstract
The first organocatalytic asymmetric synthesis of smyrindiol, by using an (S)-proline catalyzed enantioselective intramolecular
aldol reaction as the key step, is described. Smyrindiol was synthesized from commercially available 2,4-dihydroxybenzaldehyde in
15 steps, with excellent stereoselectivity (de = 99%, ee = 99%). In the course of this total synthesis a new and mild coumarin
assembly was developed.
Introduction
Furocoumarins are a group of compounds that are structurally
derived from psoralen or angelicin (Figure 1) [1]. Naturally
occurring furocoumarins are mainly found in plants of the
Apiaceae and Rutaceae families [2] and are used in the treat-
ment of vitiligo, psoriasis and other skin diseases [3]. Some
furocoumarins exhibit vasodilatory, antifungal and antibacterial
activity [4,5].
Smyrindiol (1), also called (+)-(2’S,3’R)-3-hydroxymarmesin
[7], is a linear dihydrofurocoumarin, which was isolated from
the roots of Smyrniopsis aucheri by Dzhafarov et al. in 1992 [6]
Figure 1: Furocoumarins.
and from the roots of Brosimum gaudichaudii by Vilegas et al.
1112

Beilstein J. Org. Chem. 2012, 8, 1112–1117.
Scheme 1: Synthesis of smyrindiol (1) by Grande et al.
Scheme 2: Synthesis of smyrindiol by Snider et al.
Results and Discussion
in 1993 [7]. Its 1'-O-glucoside was isolated in 1982 by
Lemmich et al. from the roots of Angelica archangelica [4]. Retrosynthetic analysis
Smyrindiol has shown antifungal and antibacterial effects [5]. Previously the proline-catalyzed intramolecular aldol reaction
of O-acetonyl-salicylaldehydes was described by our research
The first synthesis of smyrindiol was described by the group of group (Scheme 3) [10].
Grande [8]. Starting with the naturally occurring dihydrofuro-
coumarin (−)-prantschimgin (2), the hydroxy group in 3-pos-
ition was introduced by a Cr(VI)-mediated benzylic oxidation,
followed by a diastereoselective sodium borohydride reduction
(Scheme 1).
A few years later Snider et al. described the, up to now, sole
Scheme 3: Proline-catalyzed intramolecular aldol reaction of
O-acetonyl-salicylaldehydes.
total synthesis of smyrindiol using an enantiomerically pure
epoxy aldehyde [9]. The two possible diastereoisomers resulting
from the addition of the epoxy aldehyde to the coumaryl
Grignard intermediate, occurred with low diastereoselectivity, We envisaged that this 5-enolexo aldolization could be utilized
with a slight preference (18% versus 15%) towards the anti- to construct the 1,3-diol moiety of smyrindiol, if a coumarin
isomer xanthoarnol (Scheme 2).We now wish to present the derivative of the O-acetonyl-salicylaldehyde were to be used.
first diastereo- and enantioselective, organocatalytic, asym- The addition of a methyl group to the carbonyl of the aldol
metric synthesis of smyrindiol.
product would yield smyrindiol (Scheme 4).
Scheme 4: First retrosynthetic analysis.
1113

Beilstein J. Org. Chem. 2012, 8, 1112–1117.
Scheme 5: Attempted proline catalyzed aldol reaction.
Unfortunately, all attempts to conduct the intramolecular aldol of the aldol product 7 with subsequent protection of the 1,3-
reaction with the ketoaldehyde 4 failed (Scheme 5). Instead of diol. The substrate for the aldol key step 8 could be prepared
the expected hydroxy ketone 3, we obtained a complex reaction starting with commercially available 2,4-dihydroxybenzalde-
mixture. This was probably the result of unwanted intermolec- hyde (9) by a selective iodination of the 5-position of the
ular side reactions of the keto group of one molecule on the aromatic ring, protection of the hydroxy group in the 4-position,
coumarin system of another molecule, as an NMR analysis of and 2-O-acetonylation of the selectively protected salicylalde-
the reaction mixture showed that an aldehyde signal was still hyde (Scheme 6).
present after the reaction and had, thus, not taken part in the
aldol reaction. For this reason, we decided to construct the The total synthesis of smyrindiol is depicted in Scheme 7.
coumarin system after the proline catalyzed aldol reaction had Firstly, the O-acetonyl-salicylaldehyde 13, as the substrate for
taken place.
the aldol reaction, was synthesized in five steps. Iodination of
commercially available 2,4-dihydroxybenzaldehyde (9) with
We envisaged that the preparation of the coumarin system of iodine monochloride in acetic acid [11] furnished a mixture of
smyrindiol (1) could be achieved by Lindlar reduction of the different iodination products. After the addition of water to the
propiolate ester 5, in which the 1,3-diol would be protected as reaction mixture, only the 5-iodo derivative 10 precipitated
an acetonide, whereupon lactonization would take place simul- from the solution and could be separated from the other isomers
taneously. Alkyne 5 could be synthesized through a Sono- by filtration in 56% yield. We decided to protect the 4-hydroxy
gashira coupling with iodide 6. This acetonide could be formed group as an allyl ether, as we found that other protecting groups
by addition of a methyl anion equivalent to the carbonyl group were either not compatible with the reaction conditions towards
Scheme 6: Second retrosynthetic analysis.
1114

Beilstein J. Org. Chem. 2012, 8, 1112–1117.
Scheme 7: Asymmetric total synthesis of smyrindiol (1).
acetonide 15 or were found to be impossible to cleave after- butylammonium iodide mediated deprotection of the salicylic
wards while leaving the rest of the molecule intact.
hydroxy group. The selectivity of this deprotection is caused by
the chelating effect of the carbonyl oxygen towards titanium
[12], directing the Lewis acidic species to the oxygen in the
2-position. In spite of this circuitous protection–deprotection
procedure, the yield is excellent (92% over two steps) and the
Since we were unable to conduct a selective mono-allylation of
the hydroxy group in the 4-position, both hydroxy groups were
allylated, followed by a selective, titanium tetrachloride/tetra-n-
1115

Beilstein J. Org. Chem. 2012, 8, 1112–1117.
procedure is straightforward. The mono-allylated phenol must The TBS group could now be cleaved by the addition of tetra-
then be converted to the O-acetonylaldehyde 13.
butylammonium fluoride in THF. When subjected to Lindlar
catalyst and 1 atm of hydrogen, the resulting ethyl (2-hydroxy-
The alkylation of the monoprotected salicylaldehyde 12 with phenyl)propiolate 20 could be reduced to the corresponding
1-bromo- or chloroacetone under basic conditions led to side ortho-hydroxy-(Z)-cinnamate, which ring closed immediately to
products due to a subsequent base-catalyzed aldol reaction. For coumarin 21. Deprotection of the acetonide under acidic condi-
this reason, the alkylation was conducted under basic condi- tions proved to be difficult due to the acid labile nature of the
tions with 2-methoxyallyl bromide as a masked acetonylating 1,3-diol. However, indium(III) catalysis in acetonitril in the
reagent. Hydrolysis of the methyl vinyl ether with dilute acid presence of water was found to cleave the acetonide selectively,
liberated the ketone derivative 13. The synthesis of this starting yielding smyrindiol (1), whose spectroscopic data were iden-
material for the envisaged intramolecular aldol reaction could tical to those published in the literature [9].
be conducted on a multigram scale without the need for column
chromatography.
Conclusion
In summary, we have developed the first asymmetric organocat-
Similar to the previously reported aldol reaction (Scheme 3), the alytic total synthesis of smyrindiol, using an (S)-proline
(S)-proline catalyzed 5-enolexo aldol key step of this synthesis catalyzed 5-enolexo aldol reaction as the key step. The
could be performed in good yield (71%). To our delight, we diastereo- and enantioselectivity is virtually complete (de 99%,
found in this case an exceedingly high diastereo- (99%) and ee 99%), and the title compound was obtained in 15 steps in an
enantioselectivity (99%) for the aldol reaction, furnishing the overall yield of 6.3%. All steps were performed under mild
aldol product 14 as a single stereoisomer.
conditions with short reaction times. Our novel total synthesis
should allow the synthesis of larger quantities of the natural
The introduction of the methyl group by using the Imamoto compound without having to rely on natural sources. Needless
protocol [13] employing cerium(III) chloride and methyl- to mention, the unnatural enantiomer could be synthesized if
lithium at −78 °C, was unreliable with highly varying yields of (R)-proline were to be used as the organocatalyst. In addition,
the 1,3-diol. This irreproducibility is probably due to the hetero- the Sonogashira/Lindlar reduction/lactonization sequence opens
geneous nature of the reaction mixture. Fortunately, the use of a new efficient and flexible entry to the coumarin core of other
Knochel's published modification [14] of this reaction using a natural products.
lanthanum(III) chloride bis(lithium chloride) complex solution
and a methyl Grignard reagent proved to be robust and
Supporting Information
produced the desired 1,3-diol 15 in high yields (87%). The 1,3-
diol was found to be sensitive towards condensation to the
Supporting Information File 1
benzofuran system; thus, for further modifications it was
protected as acetonide 16 in a moderate yield (66%) using 2,2-
dimethoxypropane under PPTA catalysis.
Experimental procedures and characterization of
compounds.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-8-123-S1.pdf]
As we turned towards the Sonogashira reaction of the iodide 16
with a propiolic acid derivative, we found that the allyl
protecting group was cleaved readily by the palladium present
in the reaction mixture. Since the Sonogashira reaction did not
occur with the unprotected ortho-iodophenol 17, we decided to
reprotect the phenol as a tert-butyldimethylsilyl (TBS) ether,
using TBS chloride and DBU. Reprotection was necessary,
Supporting Information File 2
NMR-spectra and chromatograms.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-8-123-S2.pdf]
since employing the TBS protected ketoaldehyde 8 (PG = TBS) Acknowledgements
in the proline catalyzed aldol reaction led to complete conden- This work was supported by the Deutsche Forschungsgemein-
sation to the corresponding benzofuran.
schaft (SPP 1179 “Organocatalysis”). We thank the former
Degussa AG and the BASF SE for the donation of chemicals.
Propiolic acid esters are known to be problematic substrates for
Sonogashira reactions, due to side reactions [15]. For this
reason, we used an orthoester, which coupled smoothly with
iodide 18 and could be transformed into the aryl alkynoate 19
under mild acidic conditions, leaving the acetonide unscathed.
References
1. Moss, G. P.; Smith, P. A. S.; Tavernier, D. Pure Appl. Chem. 1995, 67,
1307–1375. doi:10.1351/pac199567081307
1116

Beilstein J. Org. Chem. 2012, 8, 1112–1117.
2. Hoffmann, D. Medical Herbalism: The Science Principles and Practices
of Herbal Medicine; Healing Arts Press: Rochester, Vermont, 2003.
3. Nore, P.; Honkanen, E. J. Heterocycl. Chem. 1980, 17, 985–987.
doi:10.1002/jhet.5570170527
4. Lemmich, J.; Havelund, S.; Thastrup, O. Phytochemistry 1983, 22,
553–555. doi:10.1016/0031-9422(83)83044-1
5. Ngameni, B.; Kuete, V.; Simo, I. K.; Mbaveng, A. T.; Awoussong, P. K.;
Patnam, R.; Roy, R.; Ngadjui, B. T. S. Afr. J. Bot. 2009, 75, 256–261.
doi:10.1016/j.sajb.2008.11.006
6. Dzhafarov, Z. R.; Kuliev, Z. A.; Vdovin, A. D.; Kuliev, A. A.;
Malikov, V. M.; Ismailov, N. M. Chem. Nat. Compd. 1992, 28, 27–31.
doi:10.1007/BF00629787
7. Vilegas, W.; Pozetti, G. L.; Harumi Yariwake Vilegas, J. H. J. Nat. Prod.
1993, 56, 416–417. doi:10.1021/np50093a015
8. Jiménez, B.; Grande, M. C.; Anaya, J.; Torres, P.; Grande, M.
Phytochemistry 2000, 53, 1025–1031.
doi:10.1016/S0031-9422(99)00524-5
9. Zou, Y.; Lobera, M.; Snider, B. B. J. Org. Chem. 2005, 70, 1761–1770.
doi:10.1021/jo047974k
10.Enders, D.; Niemeier, O.; Straver, L. Synlett 2006, 3399–3402.
doi:10.1055/s-2006-956454
11.Kaur, N.; Xia, Y.; Jin, Y.; Dat, N. T.; Gajulapati, K.; Choi, Y.;
Hong, Y.-S.; Lee, J. J.; Lee, K. Chem. Commun. 2009, 1879–1881.
doi:10.1039/b823340c
12.Tsuritani, T.; Shinokubo, H.; Oshima, K. Tetrahedron Lett. 1999, 40,
8121–8124. doi:10.1016/S0040-4039(99)01717-7
13.Takeda, N.; Imamoto, T. Org. Synth. 1999, 76, 228–233.
14.Krasovskiy, A.; Kopp, F.; Knochel, P. Angew. Chem., Int. Ed. 2006, 45,
497–500. doi:10.1002/anie.200502485
15.Anderson, K. W.; Buchwald, S. L. Angew. Chem., Int. Ed. 2005, 44,
6173–6177. doi:10.1002/anie.200502017
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.8.123
1117