Metabolites of a blocked chloramphenicol producer

Article abstract of DOI:10.1021/np020306e

Addition of p-aminophenylalanine (4), an advanced biosynthetic precursor of the antibiotic chloramphenicol (5), to a Streptomyces venezuelae pabAB mutant (VS629) restored chloramphenicol production and led to formation of the non-chlorinated analogue corynecin II (6) and four acetanilide derivatives: p-(acetylamino)phenylalanine (7), p-(acetylamino)benzyl alcohol (13), p-(acetylamino)benzoic acid (14), and p-(acetylamino)phenol (acetaminophen, 16). Metabolite structures were deduced from NMR and MS-MS data and established by chromatographic and spectroscopic comparisons with authentic samples. Reference compound 13 was synthesized by reducing the acid chloride of 14. Shunt pathways are proposed to account for the formation of the metabolites from p-aminophenylalanine.

Full text of DOI:10.1021/np020306e

62
J . Nat. Prod. 2003, 66, 62-66
Meta bolites of a Block ed Ch lor a m p h en icol P r od u cer
Elizabeth A. Lewis,† Tamara L. Adamek,‡ Leo C. Vining,‡ and Robert L. White*,†
Department of Chemistry, Dalhousie University, Halifax, Nova Scotia, B3H 4J 3 Canada, and Department of Biology,
Dalhousie University, Halifax, Nova Scotia, B3H 4J 1 Canada
Received J uly 11, 2002
Addition of p-aminophenylalanine (4), an advanced biosynthetic precursor of the antibiotic chloramphenicol
(5), to a Streptomyces venezuelae pabAB mutant (VS629) restored chloramphenicol production and led to
formation of the non-chlorinated analogue corynecin II (6) and four acetanilide derivatives: p-
(acetylamino)phenylalanine (7), p-(acetylamino)benzyl alcohol (13), p-(acetylamino)benzoic acid (14), and
p-(acetylamino)phenol (acetaminophen, 16). Metabolite structures were deduced from NMR and MS-MS
data and established by chromatographic and spectroscopic comparisons with authentic samples. Reference
compound 13 was synthesized by reducing the acid chloride of 14. Shunt pathways are proposed to account
for the formation of the metabolites from p-aminophenylalanine.
Sch em e 1. Steps in the Biosynthesis of p-Aminobenzoic Acid
(3) and Chloramphenicol (5)
In Streptomyces venezuelae (ISP5230), the aromatic
metabolites p-aminobenzoic acid (3, PABA) and chloram-
phenicol (5) are derived from the shikimate pathway1 via
chorismic (1) and 4-amino-4-deoxychorismic acid (2, ADC).2
The aromatic ring in 3 is created by a pyridoxal phosphate-
catalyzed elimination of the pyruvyl side chain in 2,3
whereas alternative reactions catalyzed by the enzyme
system arylamine synthase4 lead to multistep synthesis of
L-p-aminophenylalanine (4). The plausible sequence of
functional group transformations proposed for conversion
of 4 to chloramphenicol (5) is supported by molecular
genetic evidence5 and incorporation of several isotopically
labeled substrates, including L-p-aminophenylalanine.6,7
Molecular genetic studies have determined that two
enzyme complexes, one (consisting of PabA and PabB) for
primary metabolism and another (the fused pabAB gene
product) for secondary metabolism, can catalyze conversion
of 1 to ADC (2) in S. venezuelae.8 The product of pabC
associates with ADC synthase to supply the lyase step in
the PABA synthase complex catalyzing p-aminobenzoic
acid (3) biosynthesis.9 Involvement of an ADC synthase in
chloramphenicol (5) biosynthesis was tested by disrupting
the fused pabAB gene in S. venezuelae ISP5230.10 The
mutant (S. venezuelae VS629) was able to produce chloram-
phenicol only when supplied with p-aminophenylalanine
(4), but high-performance liquid chromatography (HPLC)
of the supplemented cultures detected several additional
metabolites. The structures of these metabolites, deter-
mined in the present investigation, revealed shunt path-
ways of secondary metabolism in S. venezuelae.
Resu lts a n d Discu ssion
cultures of a blocked mutant in which pabAB had been
disrupted confirmed the role of p-aminophenylalanine in
production of the antibiotic by the pathway predicted from
isotopic labeling (Scheme 1). Chromatographic analyses of
the supplemented cultures indicated that additional me-
tabolites similar in some respects to chloramphenicol were
produced by the mutant cultures in yields related to the
amount of supplement provided, as well as to the time of
addition. To identify these metabolites, the products ex-
tracted with EtOAc from cultures (1 L) grown for 120 h
were fractionated initially by flash chromatography on
silica gel. Elution with a stepwise gradient of MeOH in
CHCl3 yielded, in addition to chloramphenicol (14.1 mg),
a second p-nitrophenylserinol derivative (6) identified as
Chloramphenicol (12-15 mg/L) was detected by bioassay
and HPLC analysis in cultures of the S. venezuelae mutant
VS629 grown in glucose-isoleucine production medium
supplemented with 1 mM p-aminophenylalanine. The
identity of the antibiotic was verified by purification from
EtOAc extracts of cultures and comparison with an au-
thentic sample by co-chromatography (TLC and HPLC).
The ability of p-aminophenylalanine to restore approxi-
mately 40% of wild-type chloramphenicol production in
* To whom correspondence should be addressed. Tel: (902) 494-6403.
Fax: (902) 494-1310. E-mail: Robert.White@Dal.Ca.
† Department of Chemistry.
‡ Department of Biology.
10.1021/np020306e CCC: $25.00
© 2003 American Chemical Society and American Society of Pharmacognosy
Published on Web 12/13/2002

Metabolites of a Blocked Chloramphenicol Producer
J ournal of Natural Products, 2003, Vol. 66, No. 1 63
corynecin II11 from its UV absorption maximum at 273 nm,
absence of acidic or basic functions, and chromatographic
behavior matching an authentic specimen. The yield varied
markedly with the age of the vegetative culture used as
inoculum for production cultures; with the standard in-
oculum it was 2.8-3.1 mg/L, but if producing cultures were
initiated from young, vigorously growing mycelium, it was
as high as 15 mg/L, slightly above that for chloramphenicol.
Fractions from flash chromatography collected immediately
after elution of corynecin II contained predominantly a
mixture of two substances. These were separated and
purified by chromatography on silica C18 to give 13 (2.6
mg, tR 4.6 min) and 16 (2.7 mg, tR 4.15 min).
values published for p-(acetylamino)benzoic acid,14,15 and
UV absorption at 264 nm is consistent with extended
conjugation provided by the carboxyl group. The APCI mass
spectrum of 14 displayed a strong signal for [M + H]+ at
m/z 180, and sequential loss of ketene and CO2 indicated
by peaks at m/z 138 and 94 in the MS-MS spectrum is
consistent with a carboxyl group in 14.
A hydroxyl group as the second substituent of 16 was
indicated in the APCI mass spectrum by [M + H]+ at m/z
152, only 18 mass units higher than the C6H4NHCOCH3
partial structure, and by extraction from EtOAc into
aqueous Na2CO3. This implied that 16 is the well-known
analgesic acetaminophen, a conclusion supported by agree-
ment between the 13C NMR data for 16 and values
published for acetaminophen,16 and the prominent loss of
ketene in the MS-MS spectrum of 16 and the published
MS-MS fragmentation of acetaminophen.17
Examination by HPLC (detector at 246 nm) of the
culture filtrate after extraction with EtOAc showed me-
tabolites with tR 3.40 and 4.00 min. The HPLC signal at tR
4.0 min was due to an unresolved mixture of 14 and 16,
which was separated and purified by preparative TLC on
silica gel. The extracted culture filtrate was acidified with
HCl to pH 3, concentrated 20-fold, and fractionated by
chromatography on a silica C18 column to yield 7 (33 mg,
tR 3.40 min).
The 1H NMR spectra of metabolites 7, 13, 14, and 16
showed an AA′BB′ pattern in the aromatic region and a
singlet at about δ 2.1 integrating in a 4:3 ratio, suggesting
a p-disubstituted aromatic ring and a methyl group bonded
to a carbonyl carbon. 13C NMR signals between δ 115 and
135 and at approximately δ 24 supported this conclusion,
but maximum UV absorption at 242 or 246 nm indicated
that 7, 13, and 16 differed from chloramphenicol and
corynecin II in not possessing the nitrobenzene chro-
mophore. A loss of 42 mass units observed in the MS-MS
spectrum for each compound is consistent with the genera-
tion of ketene on fragmentation of an N-acetyl group,12 and
the presence of a nitrogen fits with the odd molecular
masses. An acetylamino group and a para-disubstituted
benzene ring are common to each metabolite, indicating
that 7, 13, 14, and 16 differ in the second substituent on
the ring.
NMR, UV, and mass spectra were obtained for authentic
samples of 7, 13, 14, and 16. A sample of 7 was synthesized
previously,18 and a 1H-1H NOESY experiment demon-
strated that the acetyl and aromatic protons are in close
proximity. A reference sample for 13 was prepared by
sequentially treating p-(acetylamino)benzoic acid (14) with
thionyl chloride and NaBH4 on alumina. Mixtures of
sample and reference compounds eluted as a single HPLC
peak, and spectra of metabolites and standards were
identical, establishing 7 as p-(acetylamino)phenylalanine,
13 as p-(acetylamino)benzyl alcohol, 14 as p-(acetylamino)-
benzoic acid, and 16 as acetaminophen.
The dependence of metabolite formation on the presence
of p-aminophenylalanine (4) in the culture and its sensitiv-
ity to supplement concentration (vide supra) indicate that
the metabolites are derived from the amino acid. Support-
ing this are common structural features among the me-
tabolites, the role of 4 as an intermediate in both chloram-
phenicol (5)7 and corynecin (6)19 biosynthesis (Scheme 1),
and the formation of both products on addition of 4 to the
pabAB mutant (VS629). The pathway after the step blocked
in the pabAB mutant retains all of the necessary reactions,
including hydroxylation at the benzylic position to yield
p-aminophenylserine (8), another recognized intermediate
in chloramphenicol biosynthesis.7 The later steps of the
corynecin pathway in Corynbacterium hydrocarboclastus
have not been elucidated; however, the structural similari-
ties suggest a close parallel to the chloramphenicol path-
way with introduction of a propionyl unit instead of a
dichloroacetyl group by acylation.
Metabolite 7 gave a typical purple color reaction with
ninhydrin, and a fluorescence-quenching zone (Rf 0.32) by
TLC on silica gel F with BuOH-HOAc-H2O. These
together suggested an aromatic amino acid, and an ABX
pattern at chemical shifts typical for a CHCH2 unit in an
1
R-amino acid was evident in the H NMR spectrum. ESI-
MS analysis of 7 revealed [M + H]+ at m/z 223; under MS-
MS conditions, loss of ketene was observed along with
neutral losses of NH3, H2O, and CO characteristic of an
aromatic R-amino acid,13 such as p-acetylaminophenylala-
nine (7).
A metabolic grid incorporating steps of the chloram-
phenicol pathway and showing plausible reactions convert-
ing 4 to the acetylated metabolites 7, 13, 14, and 16 is
presented in Scheme 2 as alternative parallel pathways
connected by acetylation steps. Formation of the four p-N-
acetyl metabolites is consistent with acetyl transfer by an
unspecific enzyme acting on more than one of the possible
substrates shown on the left-hand side of Scheme 2, or with
acetylation as an early step in the catabolism of p-
aminophenylalanine (i.e., 7 f 9 f 11, 13, 14, and 16). The
formation of 7, as well as chloramphenicol and corynecin
II, indicates that 4 is a substrate for both acetylation and
hydroxylation reactions and that each potential route is
exploited. The ability of a chloramphenicol-producing
Streptomyces strain to acetylate the p-amino group of a
biosynthetic intermediate was implied earlier by the isola-
tion of 2-(S)-dichloroacetyl-3-(p-acetamidophenyl)alaninol.18
Metabolites 13 and 16 were not extracted from EtOAc
solution by either aqueous NaHCO3 or HCl, indicating that
carboxylic acid and amino functional groups were not
present. For 13, however, additional signals at δ 4.55 (s,
1
2H) and 64.5 in the H and 13C NMR spectra, respectively,
suggested a CH2OH group. An [M + H]+ ion at m/z 166,
and sequential fragmentations consistent with losses of
formaldehyde (m/z 136) and ketene (m/z 94) further sup-
ported p-(acetylamino)benzyl alcohol as metabolite 13.
1
The H NMR spectra of 14 and 16 in CD3OD displayed
only a methyl singlet at δ 2.1 and the aromatic AA′BB′
pattern and suggested a second substituent lacking non-
exchangeable protons. Simulated spectra indicated that
electron-withdrawing (CO2H) and electron-donating (OH)
substituents are most consistent with the chemical shifts
Formation of 13, 14, and 16 from 4 requires cleavage of
the aliphatic side chain, which becomes more feasible after
hydroxylation at the benzylic position. Both p-aminophen-
1
of the aromatic protons in 14 and 16, respectively. The H
and 13C NMR data collected for 14 are consistent with

64 J ournal of Natural Products, 2003, Vol. 66, No. 1
Lewis et al.
Sch em e 2. Proposed Pathways for the Metabolism of
p-Aminophenylalanine (4)
The degradative route in Scheme 2 parallels a proposed
mechanism for resistance to chloramphenicol of S. venezu-
elae 13s in physiological environments where the genes for
chloramphenicol biosynthesis are not expressed.32 After
prolonged exposure to such conditions, the organism de-
velops the ability to metabolically inactivate exogenous
chloramphenicol. Evidence to date supports the existence
of a chloramphenicol hydrolase33,34 hydrolyzing chloram-
phenicol (5) to p-nitrophenylserinol, which is then N-
acetylated,35 or degraded by other competing reactions to
p-nitrobenzoic acid and p-nitrobenzyl alcohol.32,34 These
products, in addition to p-aminobenzoic acid (3) and 3-hy-
droxy-4-acetamidobenzoic acid, were among the degrada-
tion products characterized from cultures of a soil bacte-
rium growing on chloramphenicol as the sole carbon and
nitrogen source.36
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r es. NMR spectra were
acquired on Bruker AC250F and AMX 400 spectrometers using
standard Bruker spectral acquisition parameters. Chemical
shifts (δ) in 1H and 13C NMR spectra are reported relative to
the central lines of CHD2OH in CD3OD (3.31 and 49.00 ppm,
respectively), HOD in D2O (4.80 ppm), or an acetone spike in
D2O (30.89 ppm).37 Positive ion atmospheric pressure chemical
ionization (APCI) and electrospray ionization (ESI) mass
spectra were acquired on a Micromass Quattro triple quad-
rupole mass spectrometer by flow injection using a syringe
pump. Sample solutions were introduced through a Rheodyne
valve equipped with a 10 µL sampling loop. The collision
energy for collision-induced dissociation experiments (CID)
was provided by argon gas at a nominal pressure of 1 × 10-4
Torr. UV-vis spectra were recorded in 1 cm cuvettes using a
Hewlett-Packard model hp8452a diode array spectrophotom-
eter. For HPLC a Beckman System Gold instrument was used.
Melting points (uncorrected) were measured in open capillaries
using a Gallenkamp melting point apparatus. Chromatogra-
phy was performed using TLC-grade silica gel H (Merck) and
Bakerbond C18 silica gel (40 µm, J . T. Baker). NaBH4-Alox
reagent38 was prepared by mixing NaBH4 (1 g in 1 mL of H2O)
with a stirred sample of solid alumina (10 g, 80-200 mesh,
Fisher). The solid was dried under reduced pressure for 3 days
and stored over phosphorus pentoxide.
Micr oor ga n ism s, Med ia , a n d In cu ba tion Con d ition s.
Streptomyces venezuelae ISP5230 (wild-type chloramphenicol
producer)39 and VS629 (pabAB mutant blocked in chloram-
phenicol production)10 were maintained at -20 °C as stock
spore suspensions in 20% glycerol. Spores dispersed in H2O
(10 mL) from agar cultures (MYM agar for ISP5230 and TO
agar for VS629) were filtered to remove mycelial fragments
pelleted by centrifugation and resuspended in 20% aqueous
glycerol. For detection or isolation of metabolites, cultures of
S. venezuelae VS62910 were grown in glucose-isoleucine me-
dium40 supplemented with 1 mM p-aminophenylalanine. Shaken
ylserine (8), an intermediate of the chloramphenicol path-
way, and its acetylated analogue 9 are potential substrates
for retro-aldol reaction, or after oxidation of the secondary
alcohol, a retro-Claisen reaction. A similar degradation of
p-aminophenylalanine involving benzylic hydroxylation
before cleavage has been proposed to account for the
incorporation of a C6C1 unit into the nitroaromatic moiety
of the antibiotic aureothin in Streptomyces thioluteus.20
Also, a catalytic antibody-catalyzed retro-aldol cleavage of
p-acetaminophenylserine has been reported,21 and an al-
dolase acting upon a range of phenylserine derivatives has
been detected in Streptomyces tendae22 and isolated from
Streptomyces amakusaensis.23,24 The enzyme is specific for
the 2S,3R configuration of the substrate, the same absolute
configuration as that of chloramphenicol (1) and its ad-
vanced precursors.
An aldehyde (11) formed by a retro-aldol cleavage would
readily undergo reduction or oxidation to form metabolites
13 and 14. Alternatively, a retro-Claisen cleavage would
generate a coenzyme A ester,25 from which 14 could be
obtained by hydrolysis and 13 by reduction. The formation
of 4-nitrobenzyl alcohol from 4-aminobenzoic acid in a
chloramphenicol-producing Streptomyces26 supports the
feasibility of carboxyl group reduction. The formation of
acetaminophen (16) by hydroxylation and acetylation of
aniline has been demonstrated in several species of Strep-
tomyces.27,28 In S. venezuelae ATCC10712, however, aniline
was only acetylated.29 Alternatively, oxidative decarboxy-
lation of a benzoic acid derivative, as in the flavin mo-
nooxygenase-catalyzed conversion of p-aminobenzoic acid
(3) to p-aminophenol (15) by the mushroom Agaricus
bisporus,30,31 would account for the formation of 16.
cultures (25 mL) were started from
a 1% (v/v) washed
vegetative inoculum grown in GNY medium32 and incubated
for 7-10 days at 27 °C and 200 rpm in 250 mL Erlenmeyer
flasks. Samples of culture broth were removed aseptically and
assayed by HPLC.
HP LC An a lysis. Samples (20 µL) were injected onto a C18
column (45 × 4.6 mm, Beckman Ultrasphere ODS) and
routinely eluted with stepped linear gradients from 0 to 50%
(4 min) and 50 to 100% (2 min) of MeOH in 0.1 M sodium
acetate buffer (pH 4.75). This was followed by 100% MeOH (1
min), a gradient returning to 100% buffer (1 min), and 100%
buffer alone (2 min) to reequilibrate the column before the next
sample was injected. In some early HPLC analyses a slightly
modified gradient was used, and H2O replaced the buffer.
A chloramphenicol reference sample (200 µg/mL in 25%
aqueous MeOH) was injected as a control with each batch of
samples. Its retention time varied as the method and HPLC

Metabolites of a Blocked Chloramphenicol Producer
J ournal of Natural Products, 2003, Vol. 66, No. 1 65
solvents were altered, but was kept between 6.0 and 7.0 min;
the peak area at 273 nm was used to estimate the chloram-
phenicol concentration in each sample. Unknown metabolites
were estimated as chloramphenicol equivalents by comparing
their peak areas with that of the reference. For some samples,
the detector wavelength differed from 273 nm.
5.2 Hz, J BX ) 7.6 Hz, J AB ) 14.6 Hz), 2.15 (3H, s); partial 13C
NMR (D2O spiked with acetone-d6, 100.6 MHz) δ 173.7, 138.0
(C1′), 133.4 (C4′), 131.4 (C2′,C6′), 123.1 (C3′,C5′), 57.4 (C2),
37.4 (C3), 24.5 (CH3); ESI+MS (MeOH-H2O, 1:1, trace HCO2H,
30 µL/min, 15 V cone) m/z 223 [M + H]+; CID of m/z 223 (30
eV) m/z 206 (44), 177 (100), 164 (41).
TLC An a lysis. Metabolites extracted from cultures with
EtOAc, fractions collected by column chromatography, and
reference compounds were applied to 1 mm layers of silica gel
60 F254 coated on 5 × 20 or 20 × 20 cm glass plates (Merck).
The chromatograms were developed with either CHCl3-MeOH
(9:1, v/v) or butanol-acetic acid-H2O (12:3:5, v/v/v) and air-
dried. The plates were viewed under UV light (254 nm) to
detect fluorescence-quenching zones and heated at 80 °C after
spraying with ninhydrin to detect amino acids.
Standard p-(acetylamino)-L-phenylalanine was prepared
previously:18 UV (H2O) λmax 244 nm; 1H NMR (D2O, 250.1 MHz)
δ 7.40-7.27 (4H, AA′BB′ system, major splittings of 8.5 Hz),
3.94 (1H, X of ABX, J AX ) 5.2 Hz, J BX ) 7.9 Hz), 3.24 and 3.08
(2H, AB of ABX, J AX ) 5.2 Hz, J BX ) 7.9 Hz, J AB ) 14.6 Hz),
2.15 (3H, s); 13C NMR (D2O spiked with acetone-d6, 100.6 MHz)
δ 175.7, 173.8, 137.8 (C1′), 133.8 (C4′), 131.4 (C2′,C6′), 123.3
(C3′,C5′), 57.6 (C2), 37.7 (C3), 24.3 (CH3); 1H-1H NOESY
correlation (D2O, 250.1 MHz) δ 2.15 (s, 3H)/7.39 apparent d,
1H) and 7.29 (apparent d, 1H); ESI+MS (MeOH-H2O, 1:1,
trace HCO2H, 30 µL/min, 15 V cone) m/z 223 [M + H]+; CID
of m/z 223 (30 eV) m/z 206 (41), 177 (100), 164 (48).
p-(Acetyla m in o)ben zyl Alcoh ol (13). UV (MeOH) λmax
246 nm; 1H NMR (CD3OD, 250.1 MHz) δ 7.53-7.28 (4H,
AA′BB′ system, major splittings of 8.6 Hz), 4.55 (2H, s), 2.12
(3H, s); partial 13C NMR (CD3OD, 100.6 MHz) δ 128.6
(C2,C6), 121.1 (C3,C5), 64.9 (CH2OH), 23.8 (CH3); APCI+MS
(MeOH, 50 µL/min, 15 V cone) m/z 166 [M + H]+; CID of m/z
166 (30 eV) m/z 148 (9), 136 (40), 106 (24), 94 (100), 77 (12),
43 (19), 31 (15); HREIMS m/z 165.0791 (calcd for C9H11NO2,
165.0790).
Syn th esis of 13. Thionyl chloride (0.052 mL, 0.7 mmol) was
added to a mixture of p-acetamidobenzoic acid (0.105 g, 0.58
mmol), DMF (5 µL, 0.06 mmol), and THF (15 mL) under N2 in
a 10 mL flask fitted with a CaCl2 drying tube. After stirring
at ambient temperature overnight, the mixture was concen-
trated by vacuum distillation. The pale yellow, solid residue
of 4-(acetylamino)benzoyl chloride was dissolved in THF (5 mL)
and added to a suspension of NaBH4-Alox38 (1 g) in THF (15
mL) in a flask fitted with a CaCl2 drying tube. The reaction
mixture was stirred at ambient temperature overnight and
filtered. The solid residue was washed with THF (10 mL). The
wash and the filtrate from the reaction mixture were com-
bined, dried over anhydrous MgSO4, and concentrated in
vacuo. The residue was sublimed, yielding a white solid (40
mg, 42%): mp 118.5-119 °C [lit.42 119-122 °C]; UV (MeOH)
λmax 246 nm; 1H NMR (CD3OD, 250.1 MHz) δ 7.53-7.27 (4H,
AA′BB′ system, major splittings of 8.6 Hz), 4.55 (2H, s,), 2.11
(3H, s); 13C NMR (CD3OD, 62.9 MHz) δ 171.6 (NHCdO), 139.1,
138.5, 128.6 (C2,C6), 121.1 (C3,C5), 64.9 (CH2), 23.8 (CH3);
APCI+MS (MeOH, 0.2 mL/min, 15 V cone) m/z 166 [M + H]+;
CID of m/z 166 (35 eV) m/z 148 (1), 136 (33), 106 (26), 94 (100),
77 (13), 43 (13), 31 (17).
p-(Acetyla m in o)ben zoic Acid (14). The isolated metabo-
lite was purified by preparative-layer chromatography (silica
gel, toluene-acetic acid-H2O, 20:6:0.2). After development,
the fluorescence-quenching band at Rf 0.25 was eluted into
dry CH3CN, which was removed in vacuo to yield a white
solid: UV (MeOH) λmax 264 nm; 1H NMR (CD3OD, 250.1 MHz)
δ 7.99-7.65 (4H, AA′BB′ system, major splittings of 8.9 Hz),
2.15 (3H, s); partial 13C NMR (CD3OD, 100.6 MHz) δ 131.7
(C2,C6), 120.1 (C3,C5), 24.0 (CH3); APCI+MS (MeOH, 0.2 mL/
min, 25 V cone) m/z 180 [M + H]+; CID of m/z 180 (30 eV) m/z
138 (14), 94 (100), 77 (41), 43 (45)); HREIMS m/z 179.0588
(calcd for C9H9NO3, 179.0582).
Extr a ction a n d Isola tion . Cultures (typically 500 mL total
volume) were filtered to remove the mycelium, and the filtrate
at pH 7.5 was extracted with EtOAc (3 × 200 mL). After
evaporation to dryness, the extract was taken up in a mini-
mum volume of CHCl3 and applied to a dry column (4 × 2.3
cm) of silica gel in a 15 mL sintered glass funnel;41 solvent
was drawn through the column, and substances were eluted
in fractions by successive 5 mL solvent additions: CHCl3 (6
fractions) was followed by 95:5 CHCl3-MeOH (12 fractions)
and then 90:10 CHCl3-MeOH (4 fractions). Separation of
metabolites in the extract was monitored by HPLC analysis
of each fraction after it had been evaporated and redissolved
in a measured volume of H2O-MeOH (3:1). Fractions contain-
ing the same metabolite or metabolite mixture were combined.
Chloramphenicol was the predominant metabolite in frac-
tions 11-13 and was purified by successive crystallization from
CHCl3 and H2O (colorless needles, mp 152 °C; unchanged on
admixture with an authentic specimen). Corynecin II was the
main component in fractions 14-16, but was mixed with the
trailing edge of the preceding chloramphenicol band and the
leading edge of another metabolite (see later). For further
purification of corynecin II in pooled fractions 14-16, reversed-
phase chromatography on a column (95 × 1.0 cm) of silica C18
was used with a linear gradient of MeOH in H2O as the eluting
solvent. Fractions (3.0 mL) were collected, monitored for
absorbance at 273 nm, and examined by HPLC. The metabolite
recovered from a well-resolved absorbance peak (fractions 44-
45) by evaporation of the solvent in vacuo co-chromatographed
with corynecin II (HPLC, tR 5.5 min: TLC, Rf 0.29) in the
standard systems used.
Monitoring the eluates from flash chromatography by HPLC
with the detector at 246 nm showed two dominant peaks (tR
4.15 and 4.6 min) in each of fractions 17-21, with a steady
decrease in the relative intensity of the peak at tR 4.15.
Fractions 17-21 were pooled, evaporated to dryness, and
applied in the minimum volume of H2O-MeOH (3:1) to a 95
× 1.0 cm silica C18 column. Elution with a linear gradient of
MeOH in H2O and evaporation of fractions 34-35 in vacuo
gave metabolite 16 (2.7 mg; tR 4.15 min), while fractions 38-
39 gave metabolite 13 (2.6 mg; tR 4.6 min).
Metabolites retained in culture filtrates extracted with
EtOAc at neutral pH were recovered by acidifying the aqueous
phase with HCl to pH 3, concentrating 20-fold in vacuo, and
fractionating the clarified concentrate by chromatography on
a silica C18 column (20 × 1.6 cm). Fractions (5 mL) were
collected during elution with a shallow gradient of MeOH-
H2O (3:1) in H2O. Metabolites in the eluate were monitored
by UV absorbance and HPLC (detector at 246 nm). Fractions
21-31 were evaporated in vacuo to give metabolite 14 (24.6
mg; tR 4.00 min). Fractions 41-60 were evaporated in vacuo
to give metabolite 7 (33.6 mg; tR 3.35 min). Metabolites 7 and
14 were also isolated from the acidified culture filtrate
concentrate by preparative TLC on silica gel GF254 using
repeated irrigation with the solvent mixture toluene-acetic
acid-H2O (20:5:1) to separate components (Rf values for 7,
0.02; 16, 0.05; 13, 0.08; 14, 0.13).
Standard 4-(acetylamino)benzoic acid was the product of an
undergraduate laboratory experiment:43 mp 252-253.5 °C
1
(H2O) [lit.43 250-252 °C]; UV (MeOH) λmax 264 nm; H NMR
(CD3OD, 250.1 MHz) δ 7.98-7.65 (4H, AA′BB′ system, major
splittings of 8.9 Hz), 2.15 (3H, s); 13C NMR (CD3OD, 100.6
MHz) δ 171.9, 169.6, 144.3 (C4), 131.7 (C2,C6), 127.1 (C1),
120.1 (C3,C5), 24.0 (CH3); APCI+MS (MeOH, 0.2 mL/min, 25
V cone) m/z 180 [M + H]+; CID of m/z 180 (30 eV) m/z 138
(10), 94 (100), 77 (49), 43 (45).
p-(Acetyla m in o)p h en ol (16): UV (MeOH) λmax 246 nm; 1H
NMR (CD3OD, 250.1 MHz) δ 7.32-6.70 (4H, AA′BB′ system,
major splittings of 8.9 Hz), 2.08 (3H, s); partial 13C NMR (CD3-
OD, 100.6 MHz) δ 123.4 (C3,C5), 116.2 (C2,C6), 23.5 (CH3);
APCI+MS (MeOH, 50 µL/min, 20 V cone) m/z 152 [M + H]+;
p-(Acetyla m in o)p h en yla la n in e (7): UV (H2O) λmax 242
nm; 1H NMR (D2O, 250.1 MHz) δ 7.40-7.27 (4H, AA′BB′
system, major splittings of 8.2 Hz), 3.96 (1H, X of ABX, J AX
5.2 Hz, J BX ) 7.6 Hz), 3.25 and 3.09 (2H, AB of ABX, J AX
)
)

66 J ournal of Natural Products, 2003, Vol. 66, No. 1
Lewis et al.
(16) Newmark, R. A.; Hill, J . R. J . Magn. Reson. 1976, 21, 1-7.
(17) Ayrton, J .; Plumb, R.; Leavens, W. J .; Mallett, D.; Dickins, M.; Dear,
G. J . Rapid Commun. Mass Spectrom. 1998, 12, 217-224.
(18) Wat, C.-K.; Malik, V. S.; Vining, L. C. Can. J . Chem. 1971, 49, 3653-
3656.
CID of m/z 152 (35 eV) m/z 110 (75), 93 (41), 65 (56), 43 (100);
HREIMS m/z 151.0638 (calcd for C8H9NO2, 151.0633).
Standard p-(acetylamino)phenol (Eastman Chemicals): mp
166-167 °C (H2O); UV (MeOH) λmax 248 nm; 13C NMR (CD3-
OD, 62.9 MHz) δ 171.4 (CdO), 155.3 (C1), 131.6 (C4), 123.4
(C3,C5), 116.2 (C2,C6), 23.5 (CH3); APCI+MS (MeOH, 50 µL/
min, 20 V cone) m/z 152 [M + H]+; CID of m/z 152 (35 eV) m/z
110 (68), 93 (35), 65 (67), 43 (100).
(19) Nakano, H.; Tomita, F.; Suzuki, T. Agric. Biol. Chem. 1976, 40, 207-
212.
(20) Cardillo, R.; Fuganti, C.; Ghiringhelli, D.; Giangrasso, D.; Grasselli,
P.; Santopietro-Amisano, A. Tetrahedron 1974, 30, 459-461.
(21) Tanaka, F.; Oda, M.; Fujii, I. Tetrahedron Lett. 1998, 39, 5057-5060.
(22) Evans, D. R.; Herbert, R. B.; Baumberg, S.; Cove, J . H.; Southey, E.
A.; Buss, A. D.; Dawson, M. J .; Noble, D.; Rudd, B. A. M. Tetrahedron
Lett. 1995, 36, 2351-2354.
(23) Bycroft, M.; Herbert, R. B.; Ellames, G. J . J . Chem. Soc., Perkin Trans.
1 1996, 2439-2442.
(24) Herbert, R. B.; Wilkinson, B.; Ellames, G. J . Can. J . Chem. 1994, 72,
114-117.
Ack n ow led gm en t. We thank the Natural Sciences and
Engineering Research Council of Canada for financial support;
Drs. M. Lumsden and D. L. Hooper for NMR spectra from the
Atlantic Region Magnetic Resonance Centre; and Dr. J . S.
Grossert for the mass spectra.
(25) Schnitzler, J .-P.; Madlung, J .; Rose, A.; Seitz, H. U. Planta 1992, 188,
594-600.
(26) Siddiqueullah, M.; McGrath, R.; Vining, L. C.; Sala, F.; Westlake, D.
W. S. Can. J . Biochem. 1968, 46, 9-14.
(27) J in, H. J .; Park, A. K.; Lee, S. S. Arch. Pharm. Res. 1992, 15, 41-47.
(28) Theriault, R. J .; Longfield, T. H. Appl. Microbiol. 1967, 15, 1431-
1436.
(29) Lee, S. S.; J in, H. J .; Son, M. W. Arch. Pharm. Res. 1992, 15, 30-34.
(30) Tsuji, H.; Oka, T.; Kimoto, M.; Hong, Y.-M.; Natori, Y.; Ogawa, T.
Biochim. Biophys. Acta 1996, 1309, 31-36.
(31) Tsuji, H.; Ogawa, T.; Bando, N. Sasaoka, K. J . Biol. Chem. 1986, 261,
13203-13209.
(32) Malik, V. S.; Vining, L. C. Can. J . Microbiol. 1970, 16, 173-179.
(33) Malik, V. S.; Vining, L. C. Can. J . Microbiol. 1971, 17, 1287-1290.
(34) Mosher, R. H.; Ranade, N. P.; Schrempf, H.; Vining, L. C. J . Gen.
Microbiol. 1990, 136, 293-301.
(35) Gottlieb, D.; Robbins, P. W.; Carter, H. E. J . Bacteriol. 1956, 72, 153-
156.
(36) Lingens, F.; Eberhardt, H.; Oltmanns, O. Biochim. Biophys. Acta
1966, 130, 345-354.
Refer en ces a n d Notes
(1) Emes, A.; Floss, H.-G.; Lowe, D. A.; Westlake, D. W. S.; Vining, L. C.
Can. J . Microbiol. 1974, 20, 347-352.
(2) Teng, C.-Y. P.; Ganem, B.; Doktor, S. Z.; Nichols, B. P.; Bhatnagar,
R. K.; Vining, L. C. J . Am. Chem. Soc. 1985, 107, 5008-5009.
(3) Green, J . M.; Nichols, B. P. J . Biol. Chem. 1991, 266, 12971-12975.
(4) Francis, M. M.; Westlake, D. W. S. Can. J . Microbiol. 1979, 25, 1408-
1415.
(5) He, J .; Magarvey, N.; Piraee, M.; Vining, L. C. Microbiology 2001,
147, 2817-2829.
(6) Siddiqueullah, M.; McGrath, R.; Vining, L. C.; Sala, F.; Westlake, D.
W. S. Can. J . Biochem. 1967, 45, 1881-1889.
(7) McGrath, R.; Vining, L. C.; Sala, F.; Westlake, D. W. S. Can. J .
Biochem. 1968, 46, 587-594.
(8) Chang, Z.; Sun, Y.; He, J .; Vining, L. C. Microbiology 2001, 147, 2113-
2126.
(9) Green, J . M.; Nichols, B. P.; Matthews, R. G. In Escherichia coli and
Salmonella Cellular and Molecular Biology, 2nd ed.; Neidhardt, F.
C., Ed.; ASM Press: Washington, 1996; Vol. 1, Chapter 41, pp 665-
673.
(37) Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. J . Org. Chem. 1997, 62,
7512-7515.
(38) Santaniello, E.; Farachi, C.; Manzocchi, A. Synthesis 1979, 912.
(39) Stuttard, C. J . Gen. Microbiol. 1982, 128, 115-121.
(40) Doull, J .; Ahmed, Z.; Stuttard, C.; Vining, L. C. J . Gen. Microbiol.
1985, 131, 97-104.
(10) Brown, M. P.; Aidoo, K. A.; Vining, L. C. Microbiology 1996, 142,
1345-1355.
(11) Shirahata, K.; Hayashi, T.; Deguchi, T.; Suzuki, T.; Matsubara, I.
Agric. Biol. Chem. 1972, 36, 2229-2232.
(12) Tu, Y.-P.; Harrison, A. G. J . Am. Soc. Mass Spectrom. 1998, 9, 454-
462.
(41) Harwood, L. M. Aldrichim. Acta 1985, 18, 25.
(42) Byramova, N. E.; Mochalova, L. V.; Belyanchikov, I. M.; Matrosovich,
M. N.; Bovin, N. V. J . Carbohydr. Chem. 1991, 10, 691-700.
(43) Kremer, C. B. J . Chem. Educ. 1956, 33, 71-72.
(13) Dookeran, N. N.; Yalcin, T.; Harrison, A. G. J . Mass Spectrom. 1996,
31, 500-508.
(14) Bennett, J .; Delmas, M.; Maire, J . C. Org. Magn. Reson. 1969, 1, 319-
327.
(15) Rossi, C.; Picchi, M.; Tiezzi, E.; Corbini, G.; Corti, P. Magn. Reson.
Chem. 1990, 28, 348-354.
NP020306E