1448 Journal of Natural Products, 2008, Vol. 71, No. 8
Mason et al.
Scheme 1a
benzoxazin-4-ones have been previously accomplished from sec-
ondary and tertiary amides.11
In conclusion we have shown unequivocally that the originally
proposed structure of cephalandole A (1) was incorrectly assigned,
and the revised structure 7 should be given this appropriate name.
It would be well advised to state that a reinvestigation of C. gracilis
should be performed in order to establish the true structure and
origin of cephalandole B, whether structure 1 is a true natural
product and whether 2 is or is not an artifact.
Experimental Section
General Experimental Procedures. NMR spectra were recorded
1
on a Bruker DPX 300 instrument operating at 300.1 MHz for H and
75.5 MHz for 13C, or a JEOL Eclipse 500 spectrometer operating at
500.2 MHz for 1H and 125.7 MHz for 13C (where indicated), using the
residual solvent signal as reference. Assignments are based on standard
1H, APT, COSY, HMQC, and HMBC experiments. IR spectra were
acquired on a Thermo Nicolet Avatar 330 FT-IR instrument. Elemental
analyses were performed by H. Kolbe Mikroanalytisches Laboratorium,
Mu¨lheim an der Ruhr, Germany. Melting points were determined in
open capillary tubes on a Bu¨chi B-545 melting point apparatus. The
microwave reaction was performed using an Emrys Optimizer (Personal
Chemistry) (max. 300 W). LCMS (ESI) analyses were performed with
a Shimadzu SCL-10Avp UV/vis detector at 254 nm, a HPLC System,
a SunFire 5 µm SB-C8, 2.1 × 50 mm column, and a Perkin-Elmer
SCIEX API 1500EX mass spectrometer. Chemicals and solvents were
obtained from commercial sources and used as received, with the
exception of THF, which was distilled from sodium and benzophenone,
and DMF, which was stored over activated 4 Å molecular sieves. All
reactions were performed under a nitrogen atmosphere. Chromatography
was performed using silica gel (40-63 µm).
2-[(1-Phenylsulfonyl-1H-indol-3-ylcarbonyl)amino]methyl Ben-
zoate (5). Methyl anthranilate (18.9 g, 125.0 mmol) was added to a
solution of freshly prepared 1-phenylsulfonylindol-3-ylcarboxylic acid
chloride (4)4 (16.00 g, 50.0 mmol), in CH2Cl2 (300 mL), and the
resulting mixture was stirred at room temperature overnight. The
reaction mixture was diluted with CH2Cl2 (100 mL) and washed with
aqueous HCl (2 M, 2 × 100 mL), water (100 mL), and brine (100
mL), and the combined organic portions were dried over MgSO4.
Evaporation of the solvent gave the crude product, which was triturated
with diethyl ether. The solid was collected by filtration and washed
with EtOAc/heptane (1:5). Recrystallization from absolute ethanol gave
5 (14.97 g, 69%), in three crops, as a white, amorphous solid, mp
175-176 °C; IR (neat) 1697, 1679, 1607, 1590, 1533 cm-1; 1H NMR
(DMSO-d6) δ 11.11 (1H, br s), 8.63 (1H, s), 8.27 (1H, dd, J ) 8.2, 0.8
Hz), 8.19-8.16 (1H, m), 8.13-8.10 (2H, m), 8.00-7.95 (2H, m),
7.78-7.72 (1H, m), 7.69-7.62 (3H, m), 7.48-7.36 (2H, m), 7.29-7.23
(1H, m), 3.87 (3H, s); 13C NMR (DMSO-d6) δ 164.6 (C), 161.0 (C),
139.0 (C), 136.4 (C), 135.2 (CH), 134.1 (C), 133.8 (CH), 130.6 (CH),
130.1 (CH), 130.1 (CH), 128.7 (CH), 127.8 (C), 127.1 (CH), 127.1
(CH), 125.8 (CH), 124.5 (CH), 123.7 (CH), 122.1 (CH), 121.9 (CH),
119.1 (C), 117.0 (C), 113.1 (CH), 52.5 (CH3); MS (ESI+) m/z 435 [M
+ H]+; HRMS (FAB+) m/z calcd for C23H18N2O5S + H 435.1015,
found 435.1012 [M + H]+.
a Reagents and conditions: (i) methyl anthranilate, CH2Cl2, rt, 18 h, 69%;
(ii) NaOH, CH2Cl2, MeOH, H2O, 50 °C, 2 h, 100%; (iii) SOCl2, reflux, 2 h,
70%.
experiments (all performed in acetone-d6) are included in Table 1,
along with those obtained for the synthetic compound 1 and the
data from the work of Wu et al.1a At this point we would like to
point out the difference in the numbering of the benzoxazinone
portions of structures of 1 and 7, as well as the numbering from
the paper by Wu et al.1a The numbering systems in our work were
chosen since they follow the IUPAC rules for nomenclature of
organic compounds. The indole portion of both structures is
indicated as the secondary system with the use of the “prime”
annotation.
The metabolite 2-aminophenol has previously been suggested
in another natural system, as an intermediate formed by decar-
boxylation of 3-hydroxyanthranilic acid, during biosynthesis of the
antibiotic LL-C10037R produced by a strain of Streptomyces.7 Since
several of the alkaloids derived from Cephalanceropsis gracilis
incorporate an anthranilic acid subunit, the identification of the 2H-
1,4-benzoxazin-2-one unit within the structure of cephalandole A
(7) is therefore not implausible, as it could possibly be biosynthe-
sized from 2-aminophenol and a suitable indolic precursor related
to indole-3-acetic acid. To the best of our knowledge, the true
cephalandole A (7) represents the first example of a natural product
featuring a 3-substituted 2H-1,4-benzoxazin-2-one core, although
there are many examples of other naturally occurring benzoxazi-
noids.8
Next, a synthesis of cephalandole B (2) was devised, in order to
unambiguously confirm its structure. The commercially available
compound indole-3-carboxylic acid (8) was converted to its acid
chloride by reaction with an excess of oxalyl chloride in CH2Cl2
in the presence of a catalytic amount of DMF (Scheme 2).9 The
resulting intermediate was in turn treated with methyl anthranilate
in THF to afford the desired product 2. The 1H and 13C NMR data
for cephalandole B (2) in CDCl3 were in excellent agreement with
those reported for the natural product.1a Cephalandole B (2) was
also formed directly upon heating 1 in methanol. The outcome of
this reaction suggests that the methanol extraction conditions
described by Wu et al.1a could have converted the proposed
structure 1 into 2, provided that it ever existed. In such a case,
cephalandole B would not be a natural product but an artifact
produced by ring opening of the proposed natural product 1.
Furthermore, the quinazolinone compound 9 was synthesized by
heating 1 with formamide under microwave irradiation,10 displaying
the utility of the 4H-3,1-benzoxazin-4-one core in the preparation
of 3H-1,3-quinazolin-4-ones (Scheme 3). Finally, the ring-opened
amide (10) was prepared by reacting 1 with ammonium acetate in
DMF at 70 °C. When compound 10 was heated in xylenes with a
catalytic amount of p-toluenesulfonic acid monohydrate, cyclization
was achieved, exclusively giving 1 and not the expected quinazoli-
none 9. Similar acid-catalyzed cyclizations leading to 4H-3,1-
2-[(1H-Indol-3-ylcarbonyl)amino]benzoic Acid (6). Compound 5
(2.173 g, 5.0 mmol) was hydrolyzed in a 1:1 mixture of MeOH and
CH2Cl2 (80 mL) and aqueous NaOH (6 M, 10 mL), at reflux for 1 h.
The solvents were then evaporated and the residue was redissolved in
water (200 mL). Acidification to pH 2 with 3 M HCl gave a precipitate,
which was collected by filtration and coevaporated three times with
toluene, giving 6 as an amorphous, white powder (1.40 g, 100%), mp
229-230 °C (lit.4 248 °C); IR (neat) 3302, 1658, 1581, 1522 cm-1
;
1H NMR (DMSO-d6) δ 13.68 (1H, br s), 11.90 (1H, br s), 11.83 (1H,
s), 8.78 (1H, d, J ) 8.5 Hz), 8.23 (1H, dd, J ) 6.6, 1.5 Hz), 8.07-8.03
(2H, m), 7.66-7.60 (1H, m), 7.54-7.51 (1H, m), 7.26-7.11 (3H,m);
13C NMR (DMSO-d6) δ 170.1 (C), 162.9 (C), 142.1 (C), 136.6 (C),
134.2 (CH), 131.3 (CH), 128.8 (CH), 125.4 (C), 122.5 (CH), 121.8
(CH), 121.0 (CH), 120.5 (CH), 119.7 (CH), 115.4 (C),112.3 (CH), 111.2
(C); MS (ESI+) m/z 281 [M + H]+.
2-(1H-Indol-3-yl)-4H-3,1-benzoxazin-4-one (1). Compound 6 (1.44
g, 5.13 mmol) was heated at reflux in thionyl chloride (20 mL) for
2 h. Evaporation to dryness and crystallization of the residue from
acetonitrile gave the crude product, which was recrystallized once more