Total synthesis of (-)-conocarpan and assignment of the absolute configuration by chemical methods

Article abstract of DOI:10.1039/b704211f

(-)-Conocarpan (1) was synthesized by a method based on radical cyclization, and the absolute configuration was established by chemical degradation; the original 2R,3R-assignment to (+)-conocarpan should be reversed, as suggested by a later chiroptical study of model 2,3-dihydrobenzofurans. The Royal Society of Chemistry.

Full text of DOI:10.1039/b704211f

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Total synthesis of (2)-conocarpan and assignment of the absolute
configuration by chemical methods{
Derrick L. J. Clive* and Elia J. L. Stoffman
Received (in Cambridge, UK) 20th March 2007, Accepted 19th April 2007
First published as an Advance Article on the web 1st May 2007
DOI: 10.1039/b704211f
low (ca. 40%), or extensive (ca. 25%) stereochemical scrambling
(2)-Conocarpan (1) was synthesized by a method based on
radical cyclization, and the absolute configuration was
established by chemical degradation; the original 2R,3R-
assignment to (+)-conocarpan should be reversed, as suggested
by a later chiroptical study of model 2,3-dihydrobenzofurans.
occured.
A neolignan1,2 called (+)-conocarpan,3 and assigned the structure
and absolute stereochemistry 1, was first isolated3 from timber
used for underwater construction, the purpose of the investigation
being to identify substances that made the wood resistant to attack
by a variety of marine organisms. The compound has also been
extracted from a number of other plant sources.2,4–6
We report a route that gives optically active (2)-conocarpan
(ee 88%)16 and we prove that the absolute configuration of natural
(+)-conocarpan2,3,4a,6 is 2S,3S (2).
p-Tosyloxybenzaldehyde (5)17 was homologated by Horner–
Emmons–Wadsworth olefination (5 A 6) and reduced to allylic
alcohol 7, which was then subjected to Sharpless asymmetric
epoxidation12 (7 A 8a, 93%, er 94.5 : 5.518). Enantiomeric
enrichment via crystallization (3 times) of the derived p-nitro-
benzoate from EtOH gave epoxide 8a9 with an er of 98.9 : 1.1.18
Epoxide opening under basic conditions with the sodium salt of
iodophenol 919 produced diol 10 [64%, or 83% after correction for
recovered 8a9 (23%)],20,21 and the diol segment was then converted
into an olefin (10 A 11 A 12) by mesylation (100%) and treatment
with NaI in refluxing 2-butanone [65%, or 80% corrected for
recovered dimesylate (19%)]. At this point, radical cyclization
under standard conditions (slow addition of stannane, AIBN,
PhMe, 80 uC) gave 13 (69%) with little, if any, of the corresponding
cis isomer being formed. The Wittig reaction with Ph3PEt+I2/
t-BuLi then afforded a Z : E mixture of alkenes 14. Equilibration,
mediated by PdCl2(MeCN)2 (24 h),22 generated in near quantita-
tive yield material that was largely the E isomer (E : Z = 96.9 : 3.1),
but this ratio could be improved by prolonged exposure (10 days)
to the catalyst and, under these conditions (85% yield), no Z
isomer could be detected (1H NMR, 400 MHz). Finally, the tosyl
group was removed (95%) by reaction with Na(Hg) to release (2)-
conocarpan, mp 120–123 uC [lit.4a 133–135 uC; lit.4g 124–126 uC],
[a]D 282.2 (c 1.4 MeOH) [lit.4a 122 (c 1.03 MeOH)].4a,23
Examination of our synthetic material by chiral HPLC16 showed
that it had an ee of 88% (i.e. enantiomeric ratio = 94 : 6). We did
not establish the stage at which the enantiomeric ratio was eroded
from 99 : 1 (for 8a9) to 94 : 6 (for 1), but suspect that the allylic-
benzylic ether 12 is involved as, in the earlier steps, scrambling at
the benzylic position would lead to diastereoisomers—which were
not observed.24
Conocarpan is toxic to mosquito larvae,4g,4f although the
relevance of this property to malaria control7 has not been
established, and it shows antitrypanosomal, antibacterial, anti-
fungal and photoprotective activity.8 The 2R,3R configuration
(1) was assigned3 by comparing the CD curve of conocarpan
acetate with CD curves of reference compounds having
more highly oxygenated aromatic rings; however,
a later
chiroptical study of 2,3-dihydrobenzofurans fused to the steroid
nucleous9 indicated that this assignment should be reversed
(as in 2).
Racemic conocarpan is readily available by biomimetic oxida-
tion3 or by manganese(III)-mediated radical cyclization,10 but
preparation of a single enantiomer is a much more complex task
because of the fragility of the C(2)–O bond that is part of the para-
oxygenated benzylic subunit, and the fact that Sharpless asym-
metric epoxidation—a potentially ideal method for setting the C(2)
absolute stereochemistry—does not work well11,13 (see below) for
compounds of type 3 having an electron-releasing para-oxygen
substituent.15 We have also found that benzylic alcohols of type 4,
available by Evans asymmetric aldol condensation, are unsatisfac-
tory substrates for displacement with 4-[(t-butyldimethylsilyl)
oxy]phenol, after activation of the benzylic hydroxyl in 3 under
Mitsunobu conditions, or by triflation or mesylation; the yield was
We were unable to obtain satisfactory crystals of a heavy atom
derivative of conocarpan or of 8a9, and so the fact that our
synthetic material appeared to be (and, in the event, is) the
enantiomer of the dextrorotatory natural product caused us to
check the absolute stereochemistry of the key epoxide 8a9. We
Chemistry Department, University of Alberta, Edmonton, Alberta,
Canada T6G 2G2. E-mail: derrick.clive@ualberta.ca
{ Electronic supplementary information (ESI) available: Experimental and
characterization data. See DOI: 10.1039/b704211f
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 2151–2153 | 2151

View Article Online
Scheme 2 aCorrected for recovered 19 (23%).
(t-BuMe2SiOSO2CF3, S-collidine, 100%) and the double bond
was saturated (H2, Rh-Al2O3, THF, 100%) to afford 19. Then the
tosyl group was removed [19 A 20, Na(Hg), MeOH, 60% or 88%
after correction for recovered 19 (23%)]. Triflation of the phenolic
hydroxyl (20 A 21) and hydrogenolysis27 (H2, Pd–C, Et3N,
EtOAc, 100%) gave 22. Finally, desilylation afforded 23 having
[a]D 229.3 (c 1.23, CHCl3) [lit.28 245.6 (c 1.3, CHCl3)],
corresponding to an er of 82 : 18. We did not identify the point
at which there is erosion of optical purity; our starting epoxide 8a9
had an er of 98.1 : 1.1. The absolute stereochemistry of
levorotatory 23 has been established unambiguously,28,29 and so
the experiments summarized in Scheme 2 establish that the
stereochemical course of the asymmetric epoxidation (7 A 8a)
proceeds in the desired sense; consequently, natural (+)-conocar-
pan must have the absolute stereochemistry shown in 2, as
indicated by the chiroptical study of Antus et al.9 The absolute
stereochemistry of (+)-conocarpan has been related to another
natural product30 by chemical interconversion; consequently that
assignment should also be reversed, as should the configuration of
natural (2)-conocarpan.5
Scheme 1 aCorrected for recovered 11 (19%). bCorrected for recovered
8a9 (23%).
could find no example in the literature—with proof of the
stereochemical outcome—of Sharpless asymmetric epoxidation of
a styrene derivative having a para oxygen substituent,15 and so we
checked that our substrate 7 behaves in the expected manner
during the epoxidation.14,25 Accordingly, optically active 8a9 was
converted into alcohol 17 [[a]22D +4.63 (c 0.57, CHCl3)] (Scheme 2)
by successive treatment with MsCl/Et3N and NaI26 in refluxing
DME. The hydroxyl group of 17 was silylated
We thank the Natural Sciences and Engineering Research
Council of Canada for financial support.
Notes and references
1 O. R. Gottlieb, Phytochemistry, 1972, 11, 1537–1570 especially p 1546.
2 P. Sartorelli, P. J. C. Benevides, R. M. Ellensohn, M. V. A. F. Rocha,
P. R. H. Moreno and M. J. Kato, Plant Sci., 2001, 161, 1083–1088.
3 T. Hayashi and R. H. Thomson, Phytochemistry, 1975, 14, 1085–1087.
2152 | Chem. Commun., 2007, 2151–2153
This journal is ß The Royal Society of Chemistry 2007

View Article Online
4 E.g. (a) H. Achenbach, J. Gross, X. A. Dominguez, G. Cano, J. V. Star,
L. D. C. Brussolo, G. Mun˜oz, F. Salgado and L. Lo´pez,
Phytochemistry, 1987, 26, 1159–1166; (b) X. A. Dominiguez,
C. Rombold, J. V. Star, H. Achenbach and J. Gross, Phytochemistry,
1987, 26, 1821–1823; (c) A. Arnone, V. Di Modugno, G. Nasini and
I. Venturini, Gazz. Chim. Ital., 1988, 118, 675–682; (d) H. Achenbach,
W. Utz and X. A. Dominguez, Phytochemistry, 1993, 34, 835–837; (e)
A. M. Rimando, J. M. Pezzuto, N. R. Farnsworth, T. Santisuk and
V. Reutrakul, Nat. Prod. Lett., 1994, 4, 267–272; (f) C. B. Bernard,
H. G. Krishnamurty, D. Chauret, T. Durst, B. J. R. Philoge`ne,
P. Sa´nchez-Vindas, C. Hasbun, L. Poveda, L. San Roman and
J. T. Arnason, J. Chem. Ecol., 1995, 21, 801–814; (g) D. C. Chauret,
C. B. Bernard, J. T. Arnason and T. Durst, J. Nat. Prod., 1996, 59,
152–155; (h) H. Achenbach, W. Utz, B. Lozano, E. M. G. Touche´ and
S. Moreno, Phytochemistry, 1996, 43, 1093–1095; (i) Nguyen
Hoang Anh, H. Ripperger, Tran Van Sung and G. Adam,
Phytochemistry, 1996, 42, 1167–1169; (j) Nguyen Hoang Anh,
H. Ripperger, A. Porzel, Tran Van Sung and G. Adam,
Phytochemistry, 1997, 46, 569–571; (k) P. J. C. Benevides, P. Sartorelli
and M. J. Kato, Phytochemistry, 1999, 52, 339–343; (l) S. A. S. Silva,
J. C. M. De Castro, T. G. Da Silva, E. V. L. Da-Cunha, J. M. Barbosa-
Filho and M. S. Da Silva, Nat. Prod. Lett., 2001, 15, 323–329.
12 (a) Y. Gao, R. M. Hanson, J. M. Klunder, S. Y. Ko, H. Masamune and
K. B. Sharpless, J. Am. Chem. Soc., 1987, 109, 5765–5780; (b) We used
(2)-diisopropyl tartrate and checked that the material so-labeled was
indeed levorotatory.
13 See also: (a) ref. 14, page 19; (b) M. G. Finn and K. B. Sharpless, in
Asymmetric Synthesis, ed. J. D. Morrison, Academic Press, Orlando,
1985, vol. 5, pp 247–308, especially pp 287–288.
14 T. Katsuki and V. S. Martin, Org. React., 1996, 48, 1–299.
15 We can find only one example of Sharpless epoxidation of a styrene
derivative with a para-oxygen substituent (and no additional substi-
tuents) on the benzene ring, but it is a kinetic and not a preparative
experiment: S. S. Woodard, M. G. Finn and K. B. Sharpless, J. Am.
Chem. Soc., 1991, 113, 106–113.
16 Determined by chiral HPLC: Chiralcel OD column (0.46 6 15.0 cm);
90 : 10 heptane–isopropanol; flow rate 0.6 mL min21; 40 uC; detection at
210 nm.
17 P. C.-M. Mao, J.-F. Mouscadet, H. Leh, C. Auclair and L.-Y. Hsu,
Chem. Pharm. Bull., 2002, 50, 1634–1637.
18 Determined by chiral HPLC: Chiralpak AD-RH (150 6 4.6 mm), 1 : 1
MeCN-water, flow 0.5 mL min21, detection at 232 nm. Baseline
separation of a racemic sample; retention times 11.9 min and 14.3 min.
19 Y. M. Choi-Sledeski, D. G. McGarry, D. M. Green, H. J. Mason,
M. R. Becker, R. S. Davis, W. R. Ewing, W. P. Dankulich,
V. E. Manetta, R. L. Morris, A. P. Spada, D. L. Cheney,
K. D. Brown, D. J. Colussi, V. Chu, C. L. Heran, R. S. Morgan,
R. G. Bentley, R. J. Leadley, S. Maignan, J.-P. Guiloteau,
C. T. Dunwiddie and H. W. Pauls, J. Med. Chem., 1999, 42,
3572–3587.
20 Opening of related epoxides with phenols did not work in non-aqueous
solvents, and diastereoselection in acidic organic solvents was poor; no
epoxide opening occurred in basic organic media.
21 Cf H. Tanaka, I. Kato and K. Ito, Chem. Pharm. Bull., 1987, 35,
3603–3608.
22 J. Yu, M. J. Gaunt and J. B. Spencer, J. Org. Chem., 2002, 67,
4627–4629.
23 The 1H and 13C NMR data for our sample matched the values reported
in ref. 4g, 4l and 4c extremely closely, but showed some differences from
the values reported in ref. 4a.
24 For a related, and sensitive, para-alkoxy allylic carbonate, see
compound (R)-6 in: P. A. Evans and D. K. Leahy, J. Am. Chem.
Soc., 2003, 125, 8974–8975.
25 T. Katsuki and K. B. Sharpless, J. Am. Chem. Soc., 1980, 102,
5974–5976.
26 Normally, Zn would be required to react with the intermediate iodo
epoxide formed from the mesylate. Direct conversion of an iodo epoxide
to an allylic alcohol by iodide ion has been suggested in a mechanistic
scheme: J. Barluenga, J. L. Fernandez-Simon, J. M. Concellion and
M. Yus, J. Chem. Soc., Perkin Trans. 1, 1989, 77–80.
27 A. P. Kozokowski, W. Tu¨ckmantel and C. George, J. Org. Chem., 2000,
65, 5371–5381.
28 Optical rotation: A. Kamal, M. Sandbhor and K. V. Ramana,
Tetrahedron: Asymmetry, 2002, 13, 815–820.
29 Levorotatory material corresponds to an S configuration: R. MacLeod,
F. J. Welch and H. S. Mosher, J. Am. Chem. Soc., 1960, 82, 876–880.
30 H. Achenbach, W. Utz, A. Usubillaga and H. A. Rodriguez,
Phytochemistry, 1991, 30, 3753–3757.
5 Levorotatory conocarpan: (a) B. Freixa, R. Vila, E. A. Ferro, T. Adzet
and S. Can˜igueral, Planta Med., 2001, 67, 873–875; (b) M. R. G. Vega,
M. G. de Carvalho, J. R. Velandia and R. Braz-Filho, Rev. Latinoam.
Quim., 2001, 29, 63–72.
6 Conocarpan isolated from the leaves of Piper regnelli (ref. 2); the roots
of Krameria cystisoides (ref. 4a), Krameria tomentosa (ref. 4l) and
Krameria triandra (ref. 4c); and the stems of Anorgeissus acuminata
(ref. 4e) is reported to be dextrorotatory; our own synthetic material, of
proven absolute configuration 1, is levorotatory and the CD curve of its
acetate has negative De at 260 nm. The acetate of conocarpan isolated
from timber (ref. 3) is reported to have positive De at 260 nm and so the
parent conocarpan must be dextrorotatory. Consequently all these plant
sources afford material of the same absolute configuration. With the
following exception, other references to conocarpan that we have
examined do not give the specific rotation. Conocarpan isolated from
the leaves of Piper fulvescens C. DC (ref. 5) is reported to have [a]21
=
D
2108.26.
7 R. P. Tripathi, R. C. Mishra, N. Dwivedi, N. Tewari and S. S. Verma,
Curr. Med. Chem., 2005, 12, 2643–2659 and references therein.
8 (a) Antitrypanosomal activity: P. S. Luize, T. Ueda-Nakamura,
B. P. D. Filho, D. A. G. Cortez and C. V. Nakamura, Biol. Pharm.
Bull., 2006, 29, 2126–2130; (b) antibacterial activity: G. L. Pessini,
B. P. D. Filho, C. V. Nakamura and D. A. G. Cortez, Mem. Inst.
Oswaldo Cruz, 2003, 98, 1115–1120; (c) photoprotective activity:
M. Carini, G. Aldini, M. Orioli and R. M. Facino, Planta Med.,
2002, 68, 193–197; (d) antifungal activity: M. P. De Campos, C. V. Filho,
R. Z. Da Silva, R. A. Yunes, S. Zacchino, S. Juarez, R. C. Bella Cruz
and A. Bella Cruz, Biol. Pharm. Bull., 2005, 28, 1527–1530; (e)
G. L. Pessini, B. P. D. Philo, C. V. Nakamura and D. A. G. Cortez,
J. Braz. Chem. Soc., 2005, 16, 1130–1133.
9 T. Kurta´n, E. Baitz-Ga´cs, Z. Majer and S. Antus, J. Chem. Soc., Perkin
Trans. 1, 2000, 453–461.
10 B. B. Snider, L. Han and C. Xie, J. Org. Chem., 1997, 62, 6978–6984.
11 When we tried the standard catalytic epoxidation (ref. 12) we observed
very little, if any, reaction with 2 (R = Me or t-BuMe2Si).
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 2151–2153 | 2153