New coumarins from Harbouria trachypleura: Isolation and synthesis

Article abstract of DOI:10.1016/S0040-4039(01)01355-7

Two known furanocoumarins, (+)-oxypeucedanin and (+)-saxalin, two known coumarins, umbelliferone and (+)-epoxysuberosin, and two new coumarins (+)-trachypleuranin-A and (±)-trachypleuranin-B, were isolated from the methanol extract of Harbouria trachypleu

Full text of DOI:10.1016/S0040-4039(01)01355-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6491–6494
New coumarins from Harbouria trachypleura: isolation and
synthesis
Nathan R. Guz,a,* Peter Lorenzb and Frank R. Stermitza
aDepartment of Chemistry, Colorado State University, Fort Collins, CO 80523, USA
bInstitute for Medical Neurobiology, Otto von Guericke University, Magdeburg D-39120, Germany
Received 19 June 2001; revised 17 July 2001; accepted 18 July 2001
Abstract—Two known furanocoumarins, (+)-oxypeucedanin and (+)-saxalin, two known coumarins, umbelliferone and (+)-epoxy-
suberosin, and two new coumarins (+)-trachypleuranin-A and ( )-trachypleuranin-B, were isolated from the methanol extract of
Harbouria trachypleura. An efficient five-step asymmetric synthesis of (+)-trachypleuranin-A was performed using a sequential
[3,3] Claisen–Cope rearrangement and a Shi asymmetric epoxidation. ( )-Trachypleuranin-B, a chlorinated coumarin, was
synthesized from ( )-epoxysuberosin. © 2001 Elsevier Science Ltd. All rights reserved.
Our research laboratories are currently investigating
bacterial MDR (multidrug resistance) efflux pump
inhibitors of the gram-positive bacterium Staphylococ-
cus aureus.1 We recently extended the scope of this
project to include the gram-negative bacteria Pseu-
domonus aeruginosa and Escherichia coli. A preliminary
investigation with E. coli mutants showed that cou-
marins may potentiate the action of anthraquinones as
MDR efflux pump inhibitors.2 Harbouria trachypleura
(A. Gray) J. Coulter & Rose belongs to the Apiaceae
(Umbelliferae), a family of plants known to often
contain coumarins. H. trachypleura, commonly called
R1
OR
R3
R2
O
O
O
O
O
OH Cl
O
1: R1 = R3 = H, R2 = OH
2: R1 = H, R2 = OMe, R3 =
5: R =
OH OMe
O
6: R =
3: R1 = H, R2 = OMe, R3 =
4: R1 = H, R2 = OMe. R3 =
7: R1 = R2 = OMe, R3 =
OH OH
14: R =
14a: R=
OH Cl
O
O
OH OMe
OH Cl
OH OMe
15: R =
8: R1 = R2 = OMe, R3 =
13: R1 = R3 = H, R2 = OMe
Figure 1. Coumarins and furanocoumarins from Harbouria trachypleura (1–6) and other representative structures.
Keywords: Apiaceae; Harbouria trachypleura; coumarins; furanocoumarins; Shi asymmetric epoxidation; [3,3] Claisen–Cope rearrangement.
* Corresponding author. Tel.: +1-970-491-6399; fax: +1-970-491-5610; e-mail: nrg@lamar.colostate.edu
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01355-7

6492
N. R. Guz et al. / Tetrahedron Letters 42 (2001) 6491–6494
1
whiskbroom parsley, is approximately 30–50 cm tall
and grows on the foothills of the Colorado and Wyo-
ming Front Range.3 Our assays showed that H. trachy-
pleura does not contain bacterial MDR efflux pump
inhibitors of E. coli; however, the plant belongs to a
monotypic genus and this distinction warranted its
further investigation. In this letter, we report the isola-
tion of four coumarins (1–4) and two furanocoumarins
(5, 6) (Fig. 1) from the methanol extract of H. trachy-
pleura and the synthesis of 2, 3, and 4.
Isolates were analyzed by H, 13C, COSY and DEPT
NMR as well as ES− and FAB+ mass spectrometry.
Spectral data for 7-methoxycoumarins with derivatized
prenyl side chains are well documented in the literature.
The structures for the unknown coumarins were tenta-
tively assigned as 2 and 4 based upon data for similar
compounds 7 and 8.4,5 Compounds 1, 3,6,7 48,9 and
610,11 are characterized in the literature and all spectra
were in excellent agreement with previously reported
isolations.
Isolation
Synthesis and structure verification of (+)-2, (+)-3, and
( )-4
The dried aerial parts of H. trachypleura (185.6 g) were
immersed in 1 L of distilled MeOH for one day, the
MeOH filtered, and the process repeated another time.
A portion (9.61 g) of the total crude MeOH extract
(18.10 g) was subjected to VLC (vacuum liquid chro-
matography, gradient hexanes to EtOAc) to yield 14
fractions. Fractions 5–7 contained coumarins as deter-
The synthesis of 2 and 3 is outlined in Scheme 1. The
commercially available 2-hydroxy-4-methoxybenzalde-
hyde (9) was used as a starting point. A Williamson
ether
reaction
with
4-bromo-2-methyl-2-butene
afforded the phenyl prenyl ether 10 in 80% yield. Still’s
reagent12 was used under thermodynamic conditions to
smoothly yield the corresponding ester 11 in 96% yield
and 77:23 E/Z selectivity. The optimization of E/Z
selectivity was not investigated because of the subse-
quent cyclization reaction of 11. This cyclization was
achieved by refluxing 11 in PhN(Et)2 to yield the corre-
sponding 6-prenylated coumarin suberosin (12) in good
yield. This reaction utilizes the well-known sequential
[3,3] Claisen–Cope rearrangement of phenyl prenyl
ethers.13,14 The coumarin 12 was targeted using a simi-
lar strategy by Mali and co-workers under less efficient
conditions.15 Cairns, Harwood and Astles also synthe-
sized 12 in their investigations of 6-substituted prenyl,
geranyl, and farnesyl analogues of 1.16 They, however,
used the coumarin herniarin (13) as a starting point.
This sequential [3,3] Claisen–Cope rearrangement has
also been utilized in the syntheses of other natural
products as well.16–19
1
mined by TLC and H NMR. Fraction 7 was subjected
to flash column chromatography (CC, 97:3 CH2Cl2/
(CH3)2CO) to yield 10 mg of umbelliferone (1). Com-
bined fractions 5 and 6 were also subjected to CC (97:3
to 4:1 CH2Cl2/(CH3)2CO) to yield two fractions A (frs.
5–10) and B (frs. 11–16). Both frs. A and B were
chromatographed separately on C-18 VLC (1:1 MeOH/
H2O to 100% MeOH, in-house prepared C-18) to
provide twelve fractions each. Fr. A1 yielded 4 mg of a
new coumarin, 2. Fr. A2 yielded 7 mg of epoxy-
suberosin (3). Fr. A3 yielded a mixture of another new
coumarin (4) and saxalin (5). These isolates were sepa-
rated by preparative TLC using 3:2 hexanes/EtOAc to
yield 3 mg of 4 and 2 mg of 5. Fr. B1 yielded 75 mg of
oxypeucedanin (6). Final isolates (2, 3, and 6) were
subjected to preparative TLC (97:3 CH2Cl2/(CH3)2CO)
to remove minor impurities.
O
O
O
O
i
H
H
ii
OMe
MeO
OH
MeO
MeO
O
9
10
11
iii
MeO
OH
O
iv
v
MeO
O
O
MeO
O
O
MeO
O
O
2
3
12
Scheme 1. Synthesis of (+)-trachypleuranin-A: (i) 4-bromo-2-methyl-2-butene, K2CO3, acetone, reflux, 1.5 h, 80%; (ii)
(CF3CH2)2P(O)CH2CO2Me, NaH, THF, −30°C, 45 min, 98%, 77:23 E/Z; (iii) N,N-diethylaniline, reflux, 4 h, 81%; (iv) Shi
asymmetric epoxidation, 82% ee, 85% yield; (v) TsOH, MeOH, 15 min, 99%.

N. R. Guz et al. / Tetrahedron Letters 42 (2001) 6491–6494
6493
With 12 in hand, the olefin was subjected to Shi asym-
natural product from a higher plant. Because of the
small quantity of isolated 4, it was possible that the
sample concentration of 4 was not high enough to
obtain an optical rotation. To verify the presence or
lack of optical rotation, a circular dichroism experiment
was performed. The spectrum obtained from this exper-
iment showed no circular dichroism scanning from 254
to 450 nm. It was thus concluded that 4 was a racemic
isolate.
metric epoxidation conditions20 to yield the desired
1
epoxide (+)-3 in 82% yield. The H and 13C spectral
data of the synthetic epoxide were essentially identical
to those of an authentic sample from H. trachypleura.
The optical rotation of the synthetic was +17 (c=0.01,
CHCl3). Gray reported an optical rotation of +34 (c=
0.07, CHCl3) for a sample of 3 isolated from Coleonema
album7 and the isolate from H. trachypleura had an
optical rotation of +27 (c=0.006, CHCl3). While the
optical rotation of the synthetic epoxide was only in
fair agreement with that of the isolate, the determina-
tion of the isolate’s absolute configuration was the main
goal of utilizing the asymmetric epoxidation. Extensive
attempts to determine the % ee of the epoxidation
reaction using chiral HPLC, GC and europium-based
chiral NMR shift reagents were met with failure. It was
thus decided that the % ee of the epoxidation would be
determined from the final product using the same chiral
analytical techniques.
There has been some debate in the literature on
whether epoxide-opened coumarins are actual plant
constituents or artifacts of isolation. Large quantities of
toddalolactone (7), an analogue of 2, were found in a
Soxhlet extraction of Toddalia asiatica with refluxing
methanol.23 It was believed that a thermal initiated
opening of 5-OMe epoxysuberosin and subsequent
trapping of the opened oxirane intermediate yielded 7.
The authors showed that in a reextraction of T. asiatica
with supercritical CO2, 7 was in fact a component of
the plant, although smaller quantities of 7 were isolated
with this extraction method. This problem was not
anticipated with H. trachypleura because the extraction
was performed at room temperature. Another literature
report details the isolation of 14a, the acetonide of 14,
from the acetone extracts of Eremocitrus glauca.24 It
was shown experimentally that the naturally occurring
citric acid in E. glauca catalyzed the formation of 14a.24
During the extraction of H. trachypleura, a similar acid
catalyzed reaction could have opened the epoxide of 3
to yield 2. It was assumed that if this reaction occurred
with 3 it would have also happened with the major
furanocoumarin constituent of the plant (6) and thus 15
be easily found in the plant extract. The furanocou-
The final product was easily synthesized using catalytic
tosic acid in distilled methanol to give (+)-2 in 99%
yield. The epoxide opening should occur without
racemization of the secondary alcohol because of the
tertiary carbocation formed as a reactive intermediate.
The data for synthetic 2 were in excellent agreement
with those of the isolate from H. trachypleura. The
optical rotation of the isolate was +34 (c=0.003,
CHCl3) and the synthetic +28 (c=0.01, CHCl3) thus
confirming the absolute structure of the isolate as (R)-
(+)-2.21
The final % ee of 2 was also unobtainable after exten-
sive attempts at chiral HPLC and GC and europium-
based chiral NMR shift reagents. A Mosher ester of 2
was thus synthesized using 2, (+)-MTPA, oxalyl chlo-
ride and catalytic DMF in dry CH2Cl2. The product
was analyzed by C-18 HPLC (HP 1090a, Adsorbo-
sphere C-18 5 micro, 250×4.5 mm column, 1:1 acetoni-
trile/water, 330 nm) and an 82% ee was determined for
2.
1
marin 15, however, was not isolated or seen in any H
NMR spectra. It was also possible that 2 was formed
by a silica-gel catalyzed reaction with methanol, but
stirring a sample of 3 with methanol and silica gel for
24 h at room temperature gave only starting material
1
by TLC and H NMR. We concluded that 2 and by the
same reasoning 4 are in fact constituents of H. trachy-
pleura and not artifacts of isolation.
The synthesis of ( )-4 (Scheme 2) was elaborated
through a racemic epoxidation of 12 with mCBPA in
89% yield. The chlorohydrin ( )-4 was prepared from
12 using concentrated HCl in CHCl3 for 30 minutes
(72% yield). The 1H and 13C NMR spectral data of
synthetic ( )-4 were identical to those of the natural
product.22
Acknowledgements
This research was supported by the Colorado State
University Agricultural Experiment Station and by
grant CHE 9619213 from the National Science Foun-
dation. We thank Jon C. Lorenz and J. David Warren
from Professor Yian Shi’s research group for their help
optimizing epoxidation procedures and the use of their
chiral HPLC and GC equipment.
It is interesting to note that of the pro-chiral isolates (2,
3, 4, 5, and 6), all but one (4) possessed optical activity.
Furthermore, 4 is a rare example of a chlorinated
Cl
OH
O
ii
i
12
MeO
O
O
MeO
O
O
3
4
Scheme 2. Synthesis of ( )-trachypleuranin-B; (i) mCPBA, CH2Cl2, 30 min, rt, 89%; (ii) conc. HCl, CHCl3, 30 min, rt, 72%.

6494
N. R. Guz et al. / Tetrahedron Letters 42 (2001) 6491–6494
14. Zeigler, F. E. In Comprehensive Organic Synthesis; Trost,
References
B. M.; Flemming, I.; Paquette, L. A., Eds.; Pergamon
Press: Oxford, 1991; Vol. 5, pp. 875–898.
15. Mali, R. S.; Sandu, P. K.; Manekar-Tilve, A. J. Chem.
Soc., Chem. Commm. 1994, 251.
1. (a) Guz, N. R.; Stermitz, F. R.; Johnson, J. B.; Beeson,
T. D.; Willen, S.; Hsiang, J.-F.; Lewis, K. J. Med. Chem.
2001, 44, 261; (b) Guz, N. R.; Stermitz, F. R. J. Nat.
Prod. 2000, 63, 1140; (c) Stermitz, F. R.; Tawara-Mat-
suda, J.; Lorenz, P.; Mueller, P.; Zenewicz, L.; Lewis, K.
J. Nat. Prod. 2000, 63, 1146; (d) Stermitz, F. R.; Tawara-
Matsuda, J.; Lorenz, P.; Zenewicz, L.; Lewis, K. Proc.
Natl. Acad. Sci. USA 2000, 97, 1433; (e) Hsieh, P.-C.;
Siegel, S. A.; Rogers, B.; Davis, D.; Lewis, K. Proc. Natl.
Acad. Sci. USA 1998, 95, 6602.
16. Cairns, N.; Harwood, L. M.; Astles, D. P. J. Chem. Soc.,
Chem Comm. 1986, 1264.
17. Harwood, L. M.; Loftus, G. C.; Oxford, A.; Thomson,
C. Synth. Commun. 1990, 20, 649.
18. Noungoue-Tchamo, D.; Brown, D.; Mariotte, A. Nat.
Prod. Lett. 1995, 7, 73.
19. Gester, S.; Metz, P.; Zierau, O.; Vollmer, G. Tetrahedron
2001, 57, 1015.
20. Wang, Z.-X.; Tu, Y.; Frohn, M.; Zhang, J.-R.; Shi, Y. J.
Am. Chem. Soc. 1997, 119, 11224. Optimum results were
with method B at 25°C.
2. Unpublished results.
3. Collection occurred near Horsetooth Mountain Park,
west of Fort Collins, CO, USA on 5 May 1998. A
voucher specimen (c77735) is deposited in the Depart-
ment of Biology, Colorado State University and iden-
tified by Jun Wen, Department of Biology.
4. Reisch, J.; Strobel, H. Pharmazie 1982, 37, 862–863.
5. Ishii, H.; Kobayashi, J.; Ishikawa, M.; Haginiwa, J.;
Ishikawa, T. Yakugaku Zasshi 1991, 111, 365–375.
6. Nayar, M. N. S.; Bhan, M. K. Phytochemistry 1972, 11,
3331.
7. Gray, A. I. Phytochemistry 1981, 20, 1711–1713.
8. Avramenko, L. G.; Nikonov, G. K. Khim. Prir. Soedin.
1971, 7, 830.
9. Harkar, S.; Razdan, T. K.; Waight, E. S. Phytochemistry
1984, 23, 419.
10. Lemmich, J.; Pedersen, P. A.; Nielsen, B. E. Phytochem-
istry 1971, 10, 3333.
11. Perel’son, M. E.; Sheinker, Y. N.; Syrova, G. P. Khim.
Prir. Soedin. 1971, 5, 576.
12. Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 23, 4405.
13. Claisen, L.; Tietze, E. Justus Leibigs Ann. Chem. 1926,
449, 81.
21. 1H NMR (400 MHz, CDCl3): l 1.22 (s, gem-Me), 1.25 (s,
gem-Me), 2.48 (bs, OH), 2.51 (dd, J=14.0, 10.0 Hz), 2.99
(dd, J=14.0, 1.6 Hz), 3.28 (s, OMe), 3.72 (br d, J=10
Hz), 3.91 (s, ArOMe), 6.25 (d, J=9.4 Hz), 6.81 (s, ArH),
7.37 (s, ArH), 7.64 (d, J=9.4 Hz). 13C NMR (100 MHz,
CDCl3): l 19.2, 20.7, 31.9, 49.2, 55.9, 76.0, 98.7, 112.0,
113.0, 125.5, 129.5, 143.6, 154.8, 160.7, 161.5. Mp=113–
114°C (uncorrected). HRFAB+ calcd 293.1389, found
293.1399. [h]D=+34 (c=0.006, CHCl3), synthetic [h]D=
+28 (c=0.01, CHCl3).
22. 1H NMR (400 MHz, CDCl3): l 1.67 (s, gem-Me), 1.71 (s,
gem-Me), 2.25 (d, J=5.2 Hz, OH), 2.62 (dd, J=14, 10
Hz), 3.16 (dd, J=10, 2 Hz), 3.77 (m), 3.92 (s, OMe), 6.27
(d, J=9.6 Hz), 6.83 (s), 7.34 (s), 7.64 (dd, J=9.6 Hz). 13
C
NMR (100 MHz, CDCl3): l 27.6, 29.0, 32.6, 56.0, 75.2,
78.2, 98.9, 112.2, 113.3, 124.5, 129.7, 143.4, 154.9, 160.6,
161.3. Mp=121–122°C (uncorrected) HRFAB+ calcd
297.0894, found 297.0884. [h]D=+0 (c=0.003, CHCl3).
23. Ishii, H.; Tan, S.; Wang, J. P.; Chen, I. S.; Ishikawa, T.
Yakugaku Zasshi 1991, 111, 376–385.
24. Dreyer, D. L.; Huey, P. F. Phytochemistry 1974, 13, 1237.