Semisynthesis of macrocarpal c and analogues by selective dehydration of macrocarpal a or b

Article abstract of DOI:10.1021/np400869z

Macrocarpals A and C are structurally related compounds that have been extracted from different Eucalyptus species. Although macrocarpal C is of biological interest, its isolation in pure form is difficult to achieve. We report herein an efficient method for the semisynthesis of macrocarpal C by selective exo-dehydration of another member of the macrocarpal family, macrocarpal A. We also report the semisynthesis of three new macrocarpal structures derived from either macrocarpal A or B.

Full text of DOI:10.1021/np400869z

Article
pubs.acs.org/jnp
Semisynthesis of Macrocarpal C and Analogues by Selective
Dehydration of Macrocarpal A or B
Julien Alliot,† Edmond Gravel,*,† Laurent Larquetoux,‡ Marc Nicolas,‡ and Eric Doris*,†
†CEA, iBiTecS, Service de Chimie Bioorganique et de Marquage, 91191 Gif-sur-Yvette, France
‡Les Laboratoires Pierre Fabre, Centre de Dev
́
eloppement de Chimie Industrielle, 16 Rue Jean Rostand, 81600 Gaillac, France
S
* Supporting Information
ABSTRACT: Macrocarpals A and C are structurally related
compounds that have been extracted from different Eucalyptus
species. Although macrocarpal C is of biological interest, its
isolation in pure form is difficult to achieve. We report herein
an efficient method for the semisynthesis of macrocarpal C by
selective exo-dehydration of another member of the macrocarpal family, macrocarpal A. We also report the semisynthesis of three
new macrocarpal structures derived from either macrocarpal A or B.
acrocarpals are “mixed” polyketide−diterpenoid natu-
that of macrocarpals investigated here.5 While the structure of
Mrally occurring compounds that have been isolated from macrocarpal B (1b) was elucidated by single-crystal X-ray
plants of the Eucalyptus genus.1 Macrocarpals possess an
diffraction, that of the other macrocarpals were assigned using
multidimensional NMR techniques. However, in the two
independent reports mentioned above, NMR analyses were
performed in different deuterated solvents, which rendered
direct structure comparison difficult to achieve. In particular,
confusion emerged6 in the case of macrocarpals G3 and C,4 to
which similar planar structures were attributed using 2D NMR
analysis in methanol-d4 and pyridine-d5, respectively. However,
in 1997, the picture became clearer with the completion of the
first total synthesis of macrocarpal C. Indeed, the stereo-
controlled synthesis of macrocarpal C7,8 afforded a synthetic
sample that exhibited spectroscopic data identical to those of
natural macrocarpal C but surprisingly also to those of
macrocarpal G. Hence, macrocarpals G and C share the same
structure, 2a.
unusual skeleton that can be subdivided in two domains: a left-
hand domain comprising a phloroglucinol dialdehyde moiety
(common to all macrocarpals) and a right-hand terpenoid
domain. Members of the macrocarpal family mainly differ by
the stereochemistry of the C-1′ isobutyl side chain and by the
presence or absence of a tertiary C-7 hydroxy group, which is
replaced, in some cases, by an exo-double bond. Several
macrocarpals have been extracted from Eucalyptus species
(Figure 1) since the first isolation of macrocarpal A (1a) in
1990 by Aida, Hori, Ohashi, and co-workers.2 The structure of
1a was unambiguously assigned by X-ray crystallography. A
couple of years later, two groups simultaneously reported the
isolation and characterization of other macrocarpals, namely,
macrocarpals B, C, D, E, F, and G.3,4 Macrocarpals H, I, and J
were also reported, but their terpenoid-like domain differs from
Macrocarpal C exhibits several interesting biological proper-
ties that are either common to other macrocarpals, e.g.,
antibacterial and antiviral activities, or specific to 2a, e.g.,
anorectic effect.9 A major limitation in the general utilization of
2a in biological evaluations is related to its arduous isolation
from the plant. On the contrary, 1a can be readily crystallized
from a methanolic extract. As 1a and 2a are structurally related,
it is conceivable that regioselective dehydration of the tertiary
alcohol moiety of 1a could provide straightforward access to 2a.
We thus report here our approach for the selective dehydration
of 1a into 2a (Scheme 1, path a). We also describe the
semisynthesis of novel macrocapal structures derived from 1a
and 1b by selective dehydration of the C-7 hydroxy group to
afford either the exo- (2) or endo-dehydrated (3 and/or 4)
macrocarpal structures (Scheme 1, paths a and b, respectively).
While various methods are reported for the dehydration of
tertiary alcohols, regioselectivity of the process remains a
Received: October 15, 2013
Figure 1. Examples of macrocarpal structures.
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
dx.doi.org/10.1021/np400869z | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Scheme 1. Possible Dehydration Pathways of Macrocarpals
Figure 2. Structure of propylphosphonic anhydride (T3P).
While at room temperature no reaction occurred (entry 6),
macrocarpal C (2a) was cleanly obtained as a major product
(9:1 ratio) by regioselective exo-dehydration of macrocarpal A
(1a) in refluxing THF. Indeed, pure 2a was recovered in 41%
yield after 2.5 h of reaction time (entry 7). The reaction was
quenched at ca. 50% conversion (implying the recovery of ca.
50% of 1a) because of the slow isomerization of the Δ7,15 exo-
olefinic bond of compound 2a into the Δ7,8 endo-double bond
of compound 3a. This isomerization process toward the most
substituted endo-double bond was ascribed to the in situ
formation of propylphosphonic acid that likely catalyzed the exo
to endo migration of the olefinic bond. To minimize this
rearrangement, the reaction was carried out in the presence of a
base, e.g., pyridine, but side-products and a lower conversion
were observed due to the addition of pyridine onto T3P. To
avoid this side-reaction, we used the more hindered 2,5-di-tert-
butylpyridine (DTBpyr), which was found to be efficient as
mild base. Near quantitative conversion of 1a into 2a was
indeed observed upon using DTBpyr with the formation of
only minute amounts of the endo-dehydrated product 3a (95:5,
exo/endo). Macrocarpal C (2a) was isolated pure in 87% yield
(entry 8).
challenge. The exo-dehydration of macrocarpal A (1a) was
nevertheless investigated, and some elimination conditions
were screened. The use of thionyl chloride in combination with
a base, e.g., pyridine, at different temperatures afforded a
mixture of endo- (3a) and exo-dehydrated (2a) products (Table
1). The influence of low temperature on the regioselectivity of
Table 1. Conditions Investigated for the Selective exo-
Dehydration of Macrocarpal A
As the above process was found to be quite efficient for the
regioselective exo-dehydratation of macrocarpal A, the T3P-
mediated process was also applied to macrocarpal B (1b), a C-
1′ epimer of 1a. Accordingly, when 1b was reacted with T3P
under the above conditions (heating in THF in the presence of
DTBPyr), a novel macrocarpal structure, 2b, was produced by
selective exo-dehydration of the tertiary C-7 hydroxy group of
1b. Compound 2b, which has no literature precedent, was
obtained pure in 89% yield. Again, when no base was used, slow
isomerization of 2b into 3b was observed.
entry
conditions
products exo/endo
yield of 2a
1
2
3
4
5
6
SOCl2, pyr, −50 °C, THF
SOCl2, pyr, −15 °C, THF
SOCl2, pyr, rt, THF
SOCl2, pyr, Δ, THF
Martin’s sulfurane, rt, CH2Cl2
T3P, rt, THF
no reaction
26%
0.8:1
1:1
19%
2:1
27%
1:1
37%
no reaction
41%
a
With the latter observation in mind, we conceived that novel
macrocarpal structures, incorporating a Δ7,8 endo-olefinic
moiety, could be accessed by in situ T3P-dehydration/
isomerization. Accordingly, the reaction of macrocarpal A
(1a) with T3P in refluxing THF led to an increase in the
formation of the endo-isomer 3a. Although the proportion of 3a
was >50% after 6 h of reflux in THF, some exo-product 2a
remained unaffected. Extended reaction time did not improve
the product distribution but led to some degradation.
Nevertheless, we found that the target transformation could
be directly catalyzed by a strong acid instead of the in situ
produced propylphosphinic acid. Macrocarpals A and B were
hence reacted with sulfuric acid in substoichiometric amounts
in THF for 1 h to provide, after the usual workup, hitherto
unknown macrocarpal structures 3a and 3b, respectively, which
were isolated in high yields: 90% from 1a and 78% from 1b.
The latter incorporated an olefinic group whose Δ7,8 endo-
nature was confirmed by 2D NMR experiments. This result is
in agreement with the formation of the thermodynamically
favored most substituted alkene as predicted by Zaitsev’s rule.
In summary, we reported here the semisynthesis of
macrocarpal C (2a) from macrocarpal A (1a). The process
involved the selective T3P-mediated exo-dehydration of 1a,
7
T3P, Δ, THF
9:1
8
T3P, DTBPyr, Δ, THF
95:5
87%
a
The reaction was stopped at ca. 50% conversion.
the process was studied and was found to have little impact on
the product distribution. The reaction did not occur at −50 °C
(Table 1, entry 1), but afforded a nearly stoichiometric mixture
of the two alkenes (exo/endo) when increasing the temperature
to −15 °C (entry 2). The latter observation also applies for the
reaction at room temperature (entry 3). Better results were
obtained by reacting 1a with SOCl2 in refluxing THF (entry 4),
as a 2:1 ratio in favor of the expected exo-dehydrated product
2a was obtained. However, under harsher conditions, some
unidentified degradation products were detected. Notably the
isomeric Δ6,7 endo-olefin 4a could not be detected. The most
substituted and hence thermodynamically more stable Δ7,8
endo-olefin 3a resulted from H-8 elimination. The use of
Martin’s sulfurane10 afforded the Δ7,15 and Δ7,8 isomers in equal
amounts (entry 5). We finally turned our attention to the use of
propylphosphonic anhydride (T3P), a reagent that was initially
developed as a peptide coupling agent11 (Figure 2).
B
dx.doi.org/10.1021/np400869z | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
→ 99:1) to afford either 3a (endo-A) (from 1a, 90%) or 3b (endo-B)
(from 1b, 78%).
which was also applied to macrocarpal B (1b). Reaction
conditions were further adjusted to reverse the selectivity of the
dehydration process and permit access to macrocarpal
analogues bearing a Δ7,8 endo-double bond. Three hitherto
unknown semisynthetic derivatives of macrocarpals, i.e.,
compounds 2b, 3a, and 3b, were hence synthesized.
Compound 3a: [α]20D −20.5 (c 0.06, EtOH); IR (neat) νmax 2924,
1622, 1454, 1376, 1308, 1185 cm−1; 1H NMR (methanol-d4, 400
MHz) δ 0.55 (1H, m, H-4), 0.68−0.76 (4H, m, H-12 and H-2), 0.81
(3H, d, J = 5.9 Hz, H-4′), 0.82 (3H, d, J = 5.8 Hz, H-4′), 1.05 (3H, s,
H-13), 1.06 (3H, s, H-14), 1.10−1.22 (2H, m, H-2′ and H-3′), 1.41−
1.50 (2H, m, H-10 and H-5), 1.52 (3H, s, H-15), 1.72 (1H, m, H-5),
1.85 (1H, m, H-10), 2.10−2.31 (4H, m, H-6 and H-9), 2.34 (1H, m,
H-2′), 2.51 (1H, brs, H-1), 3.50 (1H, dd, J = 5.1 Hz, 14.6 Hz, H-1′),
10.11 (1H, s, H-22), 10.12 (1H, s, H-23); 13C NMR (methanol-d4,
100 MHz) δ 16.8, 20.0, 21.9, 22.2, 22.3, 24.2, 24.9, 27.4, 28.3, 29.1,
30.3, 31.5, 35.3, 35.7, 36.9, 37.2, 45.8, 49.7, 106.0, 106.1, 111.2, 126.7,
138.4, 168.6, 170.3, 171.0, 192.9, 193.1; negative HRESMS m/z
453.2637 (calcd for C28H37O5, 453.2641 [M − H]−).
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
determined using the sodium D line (589 nm) on a Perkin-Elmer
341 polarimeter. IR spectra were recorded on a Perkin-Elmer System
1
2000 FT-IR. H and 13C NMR spectra were recorded on a Bruker
Avance DPX 400 spectrometer at 400 and 100 MHz, respectively.
Chemical shifts (δ) are expressed in ppm, and coupling constant (J) in
hertz. Chemicals were purchased from Aldrich. Reactions were carried
out using dry solvents. THF was distilled from sodium/benzophenone.
Flash chromatography was carried out on Kieselgel 60 (230−240
mesh, Merck), and analytical TLC was performed on Merck precoated
silica gel (60 F254); visualization was done with UV and/or heating
with a solution of 5−7% phosphomolybdic acid in EtOH. HRMS
Compound 3b: [α]20 +47.1 (c 0.1, EtOH); IR (neat) νmax 2924,
D
1623, 1452, 1376, 1308, 1184 cm−1; 1H NMR (methanol-d4, 400
MHz) δ 0.56 (1H, m, H-4), 0.76 0.84 (4H, m, H-2 and H-4′), 0.86
(3H, d, J = 6.4 Hz, H-4′), 1.05 (3H, s, H-12), 1.06 (3H, s, H-13), 1.07
(1H, m, H-10), 1.17 (3H, s, H-14), 1.22 (1H, m, H-3′), 1.33−1.42
(2H, m, H-2′ and H-10), 1.52 (3H, s, H-15), 1.55 (1H, m, H-5), 1.73
(1H, m, H-5), 2.07−2.24 (4H, m, H-6 and H-9), 2.38 (1H, td, J =
12.9, 3.2 Hz, H-2′), 2.44 (1H, m, H-1), 3.32 (1H, dd, J = 13.1, 3.9 Hz,
H-1′), 10.10 (1H, s, H-22), 10.12 (1H, s, H-23); 13C NMR
(methanol-d4, 100 MHz) δ 15.1, 17.5, 18.2, 19.2, 20.9, 22.6, 23.5,
26.0, 26.7, 27.7, 29.7, 30.3, 35.6, 35.7, 37.1, 40.4, 48.9, 49.1, 104.5,
104.6, 109.9, 125.4, 138.3, 166.9, 168.7, 169.5, 191.5, 191.6; negative
HRESMS m/z 453.2634 (calcd for C28H37O5, 453.2641 [M − H]−).
́
spectra were recorded at “Service de Spectrometrie de Masse de
l’Institut de Chimie des Substances Naturelles” in Gif-sur-Yvette
(France).
General Procedure for exo-Dehydration of 1a or 1b. Under
N2, macrocarpal (1a or 1b, 100 mg, 0.212 mmol, 1 equiv) was
dissolved in anhydrous THF (10 mL) before T3P (1 mL of a 50%
solution in EtOAc, 8 equiv) and DTBPyr (380 μL, 8 equiv) were
added. The reaction mixture was heated at reflux for 6 h. It was then
quenched with 1 M HCl (8 mL), and Et2O (8 mL) was added. The
organic phase was collected, and the aqueous layer was extracted with
Et2O (2 × 8 mL). The combined organic layers were washed with
brine (5 mL), dried over Na2SO4, filtered, and concentrated under
vacuum. The crude residue was purified over silica (CH2Cl2/HOAc,
99.9:0.1 → 97:3) to give 2a (from 1a, 87%) or 2b (from 1b, 89%).
Compound 2a (macrocarpal C): [α]20D −25.8 (c 0.1, EtOH); IR
(neat) νmax 3204, 2952, 2867, 1627, 1446, 1307, 1185 cm−1; 1H NMR
(methanol-d4, 400 MHz) δ 0.62−0.72 (2H, m, H-4 and H-2), 0.78
(3H, d, J = 6.2 Hz, H-4′), 0.80 (3H, d, J = 6.2 Hz, H-4′), 0.81 (3H, s,
H-12), 0.95 (1H, m, H-5), 1.01 (3H, s, H-14), 1.09 (3H, s, H-13),
1.18−1.23 (2H, m, H-3′ and H-2′), 1.30−1.43 (2H, m, H-10 and H-
1), 1.67 (1H, m, H-9), 1.79 (1H, m, H-9), 1.96−2.08 (2H, m, H-6 and
H-5), 2.22−2.47 (4H, m, H-8, H-6, H-10 and H-2′), 3.40 (1H, dd, J =
3.6 Hz, 12.8 Hz, H-1′), 4.64 (1H, s, H-15), 4.70 (1H, s, H-15), 10.11
(1H, s, H-22), 10.12 (1H, s, H-23); 13C NMR (methanol-d4, 100
MHz): δ 17.4, 21.0, 22.4, 23.9, 24.9, 26.8, 28.2, 28.5, 28.6, 29.3, 29.5,
36.0, 37.4, 38.7, 40.5, 50.6, 51.7, 52.5, 106.1, 106.1, 106.2, 111.3, 156.5,
168.7, 170.4, 170.9, 192.9, 193.1; negative HRESMS m/z 453.2623
(calcd for C28H37O5, 453.2641 [M − H]−).
ASSOCIATED CONTENT
■
S
* Supporting Information
Copies of 1H and 13C NMR for compounds 2a, 2b, 3a, and 3b.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: edmond.gravel@cea.fr. Fax: +33 169 08 79 91. Tel:
+33 169 08 84 84.
■
*E-mail: eric.doris@cea.fr. Fax: +33 169 08 79 91. Tel: +33 169
08 80 71.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
J.A. thanks the CEA and “Les Laboratoires Pierre Fabre” for a
CTCI Ph.D. grant. The “Service de Chimie Bioorganique et de
Marquage” belongs to the Laboratory of Excellence in Research
on Medication and Innovative Therapeutics (LabEx LERMIT).
Compound 2b: [α]20 +9.8 (c 0.1, EtOH); IR (neat) νmax 3423,
D
2951, 2864, 1632, 1446, 1305, 1179 cm−1; 1H NMR (methanol-d4, 400
MHz) δ 0.62−0.83 (5H, m, H-4, H-2 and H-4′),0.85 (3H, d, J = 6.5
Hz, H-4′), 1.00 (1H, m, H-5), 1.10−1.25 (11H, m, H-13, H-14, H-10,
H-12, and H-3′), 1.27−1.40 (2H, m, H-1 and H-2′), 1.55 (1H, m, H-
9), 1.63 (1H, m, H-10), 1.83 (1H, m, H-9), 1.96 (1H, m, H-6), 2.04
(1H, m, H-5), 2.38 (3H, m, H-8, H-6 and H-2′), 3.33 (1H, m, H-1′),
4.62 (1H, s, H-15), 4.69 (1H, s, H-15), 10.11 (1H, s, H-22), 10.12
(1H, s, H-23); 13C NMR (methanol-d4, 100 MHz) δ 17.5, 19.8, 20.7,
21.6, 24.8, 27.1, 27.3, 28.0, 28.1, 29.4, 31.7, 36.5, 40.6, 41.1, 42.7, 51.0,
54.8, 56.4, 105.6, 106.0, 106.1, 111.5, 157.0, 168.5, 170.4, 171.0, 193.1,
193.3; negative HRESMS m/z 453.2619 (calcd for C28H37O5,
453.2641 [M − H]−).
REFERENCES
■
(1) (a) Eschler, B. M.; Pass, D. M.; Willis, R.; Foley, W. J. Biochem.
Syst. Ecol. 2000, 28, 813−824. (b) Moore, B. D.; Foley, W. J.; Wallis, I.
R.; Cowling, A.; Handasyde, K. A. Biol. Lett. 2005, 1, 64−67.
(2) Murata, M.; Yamakoshi, Y.; Homma, S.; Aida, K.; Hori, K.;
Ohashi, Y. Agric. Biol. Chem. 1990, 54, 3221−3226.
(3) Yamakoshi, Y.; Murata, M.; Shimizu, A.; Homma, S. Biosci.
Biotech. Biochem. 1992, 56, 1570−1576.
(4) Nishizawa, M.; Emura, M.; Kan, Y.; Yamada, H.; Ogawa, K.;
Hamanaka, N. Tetrahedron Lett. 1992, 33, 2983−2986.
(5) Osawa, K.; Yasuda, H. J. Nat. Prod. 1996, 59, 823−827.
(6) Ghisalberti, E. L. Phytochemistry 1996, 41, 7−22.
(7) Tanaka, T.; Mikamiyama, H.; Maeda, K.; Ishida, T.; In, Y.; Iwata,
C. Chem. Commun. 1997, 2401−2402.
(8) Tanaka, T.; Mikamiyama, H.; Maeda, K.; Iwata, C.; In, Y.; Ishida,
T. J. Org. Chem. 1998, 63, 9782−9793.
General Procedure for endo-Dehydration of 1a or 1b. Under
N2, macrocarpal (1a or 1b, 50 mg, 0.106 mmol, 1 equiv) was dissolved
in anhydrous THF (4 mL), and three drops of concentrated H2SO4
were added. After stirring for 1 h at room temperature, H2O (5 mL)
and Et2O (5 mL) were added. The aqueous layer was extracted with
Et2O (2 × 5 mL). The combined organic layer was washed with brine
(5 mL), dried over Na2SO4, filtered, and concentrated under vacuum.
The crude product was purified over silica (CH2Cl2/HOAc, 99.9:0.1
C
dx.doi.org/10.1021/np400869z | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(9) Pass, D. M.; Foley, W.; Bowden, B. J. Chem. Ecol. 1998, 24,
1513−1527.
(10) Pooppanal, S. S. Synlett 2009, 850−851.
(11) Basavaprabhu; Vishwanatha, T. M.; Panguluri, N. R.;
Sureshbabu, V. V. Synthesis 2013, 45, 1569−1601.
D
dx.doi.org/10.1021/np400869z | J. Nat. Prod. XXXX, XXX, XXX−XXX