The first total synthesis of the natural product angoluvarin

Article abstract of DOI:10.1016/j.tetlet.2010.08.037

The total synthesis of angoluvarin, a member of the dihydrochalcone family of natural products, is reported. Starting with 2-bromophenol, the synthesis was accomplished in eight steps with an overall yield of 2%. This represents the first reported synthes

Full text of DOI:10.1016/j.tetlet.2010.08.037

Tetrahedron Letters 51 (2010) 5497–5499
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
The first total synthesis of the natural product angoluvarin
⇑
Charles F. Nutaitis
Department of Chemistry, Lafayette College, Easton, PA 18042, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The total synthesis of angoluvarin, a member of the dihydrochalcone family of natural products, is
reported. Starting with 2-bromophenol, the synthesis was accomplished in eight steps with an overall
yield of 2%. This represents the first reported synthesis of angoluvarin.
Received 14 July 2010
Revised 11 August 2010
Accepted 12 August 2010
Available online 17 August 2010
Ó 2010 Elsevier Ltd. All rights reserved.
A number of C-benzylated dihydrochalcones isolated from the
genus Uvaria, of the plant family Annonaceae, have displayed a
wide variety of biological activity,1 such as in vivo activity against
P-388 lymphocytic leukemia, growth inhibition of human promy-
elocytic leukemia HL-60 cells, in vitro activity against human car-
cinoma cells of the nasopharynx, and antibacterial activity against
Staphylococcus aureus, Bacillus subtilis, Mycobacterium smegmatis,
Proteus micabilis, and Pseudomonas aeruginosa. Members of this
family and genus have also been extensively used in folk medicine
for the treatment of ailments associated with menstruation, dysen-
tery, and snakebites.2
HO
BnO
BnO
HO
OCH3
O
HO
OCH3
O
OH
OH
OH
2
1
BnO
BnO
BnO
HO
OCH3
In 1987, Hufford and Oguntimein3 isolated a new dihydrochal-
cone natural product from Uvaria angolensis which they named
angoluvarin. Similar to previously isolated dihydrochalcones, ang-
oluvarin also demonstrated antimicrobial activity againstB. subtilis,
S. aureus, and M. smegmatis. At the time of its isolation, angoluvarin
possessed the most complex structure of the known members of
this natural product class. In 1993, Achenbach4 and co-workers
reported that angoluvarin can also be isolated from Uvaria Leptoc-
ladon. In subsequent years even more complex naturally occurring
dihydrochalcones were isolated from Annonaceae.5 Although it has
been 23 years since the isolation and structural elucidation of ang-
oluvarin, no reports pertaining to its synthesis have appeared in
the literature. We now report the first total synthesis of the natural
product angoluvarin.
Our retrosynthetic analysis for the synthesis of angoluvarin is
depicted in Scheme 1. We envisioned a three component strategy
utilizing two commercially available compounds, 5-methoxyresor-
cinol (6) and hydrocinnamoyl chloride (4), as well as alcohol 5.
Friedel–Crafts alkylation of 5-methoxyresorcinol (6) with alco-
hol 5 could in theory lead to two different regioisomers. Alkylation
at C-4/C-6 is statistically favored, but alkylation at C-2, the desired
position, is sterically favored. Given the sterically crowded envi-
ronment of alcohol 5, as a result of the benzyloxy group situated
ortho to the reactive site, it was anticipated that formation of 3
O
HO
OCH3
+
Cl
OH
+
OH
6
BnO
OH
5
4
3
Scheme 1. Retrosynthetic analysis of angoluvarin.
would be favored over formation of the undesired regioisomer.
Subsequent acylation of 3 with hydrocinnamoyl chloride (4) was
expected to take place in the electron rich resorcinol ring to afford
protected angoluvarin 2. Although there are no regiochemical
considerations for C-acylation of the symmetric resorcinol nucleus,
O-acylation was considered to be a possible complicating factor.
However, it was hoped that reflux conditions would effect a
Fries-rearrangement to convert it to desired ketone 2.
The synthesis of key intermediate 5 began by benzyl-protection
of both o-bromophenol (7) and 3-bromo-4-hydroxybenzaldehyde
(8) in 90% and 72% yield, respectively (Scheme 2). Halogen-metal
exchange of 9 with sec-BuLi and subsequent quenching with alde-
hyde 10 afforded diphenylalcohol 11 in 81% yield. Surprisingly, the
presence of the additional bromine in 10 did not pose any major
problems due to halogen scrambling with the aryllithium reagent
generated from 9. Presumably the bulky ortho-benzyloxy groups
present in both reagents prevent close contact between the two
reactive sites, thus inhibiting the halogen scrambling. Reduction
of 11 with sodium borohydride/trifluoroacetic acid6 proceeded as
expected to provide 12 in 88% yield. Preparation of alcohol 5 can
⇑
Tel.: +1 610 330 5218; fax: +1 610 330 5714.
E-mail address: nutaitic@lafayette.edu
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.08.037

5498
C. F. Nutaitis / Tetrahedron Letters 51 (2010) 5497–5499
OBn
OH
Ph
Br
OBn
OH
Ph
Br
Br
Br
Br
Br
K2CO3
K2CO3
acetone
acetone
CHO
CHO
9
7
8
10
BnO
BnO OH
1) sec-BuLi/THF
-78 ºC
Br
NaBH4
Br
9
TFA/Et2O
OBn
2) 10
OBn
11
12
1) sec-BuLi/THF/-78 ºC
2) 1-formylpiperidine
3) 3M HCl
4) 10% NaHCO3
BnO
BnO
NaBH4
CH3OH
OH
OBn
CHO
OBn
5
13
Scheme 2. Synthesis of key intermediate 5.
be accomplished in two ways from the organolithium reagent
derived from 12: direct conversion by quenching with solid
paraformaldehyde or reduction of intermediate aldehyde 13,
which is obtained by utilizing 1-formylpiperidine as the quenching
agent. The latter two-step approach proved to be the method of
choice, providing 5 in 70% (two-step yield) as opposed to 31% for
the one-step process. Attempts to purify aldehyde 13 by flash chro-
matography were plagued with low yields and decomposition
problems, as evidenced by additional spots appearing in the
column effluent upon analysis by thin layer chromatography. Fur-
thermore, it was found that a base wash is essential in the workup
of this reaction to prevent low yields and complex product mix-
tures. The workup requires acidification to convert the intermedi-
ate piperidinyl alcohol to the aldehyde, but apparently trace
amounts of residual acid leads to decomposition of the aldehyde.
However, if the required acidification step is immediately followed
by a wash with 10% aqueous sodium bicarbonate, and the subse-
quent reduction is performed without further purification of 13,
the yield and purity of 5 are vastly improved.
With key intermediate 5 in hand, attention was next focused on
Friedel–Crafts alkylation with 5-methoxyresorcinol (6). It was
quickly found that use of strong Lewis acids such as aluminum
chloride or ferric bromide led to complex product mixtures; proton
NMR indicated that benzyl-deprotection of both phenolic oxygens
of 5 had occurred. Two milder Lewis acids were explored, tin(IV)
chloride and zinc chloride, with the latter proving to be more effi-
cient, providing 3 in 34% yield. The regiochemistry for the alkyl-
ation of 5-methoxyresorcinol (6) was determined by proton NMR
spectroscopy. The symmetry of the product was evident by the
two proton upfield singlet at 5.91 ppm for the hydrogens at C-4/
C-6 of the resorcinol ring as well as a two proton singlet at
5.94 ppm representing the two equivalent phenolic hydrogens.
The undesired, non-symmetric isomer, which was not evident in
any of the column fractions isolated, would have produced two
doublets for the non-equivalent ring protons.
it in chloroform with a large excess of the acid chloride for 3 days
in the absence of a Lewis Acid catalyst, to afford protected angoluv-
arin 2. Acylation of the resorcinol ring was supported by the
appearance of a highly deshielded proton at 14.88 ppm which is
attributed to hydrogen bonding of the phenolic proton ortho to
the acyl group. Additionally the proton NMR spectrum now exhib-
ited only a single resorcinol ring proton as opposed to a two proton
signal for the resorcinol protons in 3. It was found that 2 was prone
to decomposition on standing for a few days, and all attempts at
complete purification were unsuccessful. As a result, 2 was only
partially purified by flash chromatography, primarily to separate
it from the substantial amount of unreacted hydrocinnamoyl chlo-
ride (4) present in the reaction product, and then immediately
deprotected with boron tribromide in methylene chloride to pro-
vide angoluvarin (1) in a 14% two-step yield and an overall
eight-step yield of 2% from o-bromophenol (7). Except for the
non-hydrogen bonded phenolic hydrogens, which are not com-
pletely visible due to peak-broadening, the 1H NMR spectral data
are in good agreement with the values reported for the natural
product; the 13C NMR spectral data are identical to reported values.
In summary an eight-step synthesis of angoluvarin (1) has been
accomplished. Although some of the later steps of the synthesis are
low yielding, it represents the first reported total synthesis of this
natural product that was isolated over 20 years ago. The synthetic
strategy developed in this synthesis should be applicable to more
complex members of this class of natural products for which total
syntheses have also not been reported.
Acknowledgment
The author thanks Lafayette College for supporting this work
and the mass spectrometry facility at Michigan State University
for providing the high resolution mass spectral data.
Supplementary data
Acylation of 3 with hydrocinnamoyl chloride (4) also proved to
be quite problematic; even relatively mild Lewis acids such as
tin(IV) chloride and zinc chloride lead to complete benzyl-
deprotection of the phenolic oxygens and very complex product
mixtures, as evidenced by TLC and proton NMR spectroscopy.
However, it was found that 3 could be slowly acylated by refluxing
General experimental, detailed experimental procedures, and
compound characterization data for all new compounds can be
found. Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.2010.08.037.

C. F. Nutaitis / Tetrahedron Letters 51 (2010) 5497–5499
3. Hufford, C. D.; Oguntimein, B. O. J. Org. Chem. 1987, 52, 5286–5288.
5499
References and notes
4. Nkunya, M. H. H.; Weenen, H.; Renner, C.; Waibel, R.; Achenbach, H.
Phytochemistry 1993, 32, 1297–1300.
5. (a) Anam, E. M.; Ekpa, O. D.; Gariboldi, P. V.; Morah, F. N. I.; Dosunmu, M. I.
Indian J. Chem. 1993, 32B, 1051–1054; (b) Anam, E. M. Indian J. Chem. 1994, 33B,
204–205; (c) Anam, E. M. Pharm. Pharmacol. Lett. 1994, 3, 241–244; (d) Anam, E.
M. Indian J. Chem. 1997, 36B, 57–60.
1. (a) Cole, J. R.; Torrance, S. J.; Wiedhopf, R. M. J. Org. Chem. 1976, 41, 1852–1855; (b)
Lasswell, W. L., Jr.; Hufford, C. D. J. Org. Chem. 1977, 42, 1295–1302; (c) Ichimaru,
M.;Nakatani, N.; Takahashi, T.; Nishiyama, Y.; Moriyasu, M.; Kato, A.; Mathenge, S.
G.; Juma, F. D.; Ngana, J. N. Chem. Pharm. Bull. 2004, 52, 138–141; (d) Nakatani, N.;
Ichimaru, M.; Moriyasu, M.; Kato, A. Biol. Pharm. Bull. 2005, 28, 83–86.
2. (a) Leboeuf, M.; Cavé, A.; Mhaumik, P. K.; Mukherjee, B.; Mukherjee, R.
Phytochemistry 1982, 21, 2783–2813; (b) Parmar, V. S.; Tyagi, O. D.; Malhotra,
A.; Singh, S. K.; Bisht, K. S.; Jain, R. Nat. Prod. Rep. 1994, 11, 219–224.
6. Gribble, G. W.; Leese, R. M.; Evans, B. E. Synthesis 1977, 172–176.