вход по аккаунту


Sequential Enzymatic Oxidation of Aminoarenes to Nitroarenes via Hydroxylamines.

код для вставкиСкачать
Enzyme Catalysis
Sequential Enzymatic Oxidation of Aminoarenes
to Nitroarenes via Hydroxylamines**
Robert Winkler and Christian Hertweck*
Aromatic nitro groups are relatively rare structural elements
in natural products, but are found in diverse types of natural
products, most of which are endowed with important biological properties. Prominent examples of bacterial nitroaryl
metabolites include the potent antibiotic chloramphenicol
(1),[1] the antifungal agent pyrrolnitrin (2),[2] and the cyto-
static, p-nitrobenzoate-derived polyketide aureothin (3).[3, 4]
The nitrophenanthrene derivative aristolochic acid C (4) is
the causative agent for ?Chinese herbs nephropathy?[5] and 2-
[*] R. Winkler, C. Hertweck
Leibniz Institute for Natural Products Research and
Infection Biology, HKI
Beutenbergstrasse 11a, 07745 Jena (Germany)
Fax: (+ 49) 3641-656-705
[**] We thank the Deutsche Forschungsgemeinschaft for financial
support in the priority program SPP1152 ?Evolution of Metabolic
Diversity? (HE 3469/2) and A. Perner for MS measurements.
Angew. Chem. Int. Ed. 2005, 44, 4083 ?4087
DOI: 10.1002/anie.200500365
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
nitrophenol (5) is a well-studied component of the aggregation?attachment pheromone of ticks. The fungal nitroaryl
compound stephanosporin (6) is the precursor of 2-chloro-4nitrophenol.[6] Important examples of natural nonaromatic
nitro compounds are nitropropionic acid[7] and the bacterial
signal metabolite hormaomycin.[8,9]
Surprisingly, the biosynthesis of aromatic nitro groups is
understood only to a small extent. One possible pathway
involves aromatic substitution of nitro groups, as is the case
for the unusual nitro alkaloid 1-nitroaknadinine (7) from
Stephania sutchuenensis, which occurs simultaneously with
the non-nitrated alkaloid. The proposed biosynthetic nitration step was mimicked in vitro with HNO3.[10] More recently,
it was reported that a NO synthase (NOS) takes part in the
nitration of the diketopiperazine thaxtomin (8), a phytotoxin
produced by the plant pathogen Streptomyces turgidiscabies.[11] Additional studies revealed that tryptophan is regioselectively nitrated by a complex of Deinococcus radiodurans
NOS and Trp?tRNA synthetase.[12]
Despite these important findings, it appears that biosynthesis of the vast majority of aromatic nitro compounds
proceeds by an alternate route: the enzymatic oxidation of
primary amines.[1, 13?15] In their studies with 18O labeling and
N NMR spectroscopy, Baxter and co-workers revealed that
both oxygen atoms of the fungal metabolite nitropropionate
are derived from molecular oxygen.[7, 16]
Conceivably, N-oxygenation could proceed in three steps
via hydroxylamine and nitroso intermediates (Scheme 1), or
both oxygen atoms could be introduced simultaneously by a
Scheme 1. Proposed biosynthetic pathway from PABA to PNBA via
p-hydroxylaminobenzoate (PHABA).
dioxygenase, yet no direct evidence for either hypothesis has
been given. The only existing studies on enzymatic nitro
group formation have been carried out with haloperoxidases.
These are not genuine N-oxygenases; they catalyze the
oxidation of amino groups only under very specific, nonnatural in-vitro conditions through the formation of reactive
peroxo species, with excess H2O2 and/or in the absence of
halide ions.[17?19] Furthermore, haloperoxidases such as CPOP are not naturally involved in the biosynthesis of nitro
compounds, as has been revealed by the analysis genes
responsible for pyrrolnitrin biosynthesis.[13, 20]
We recently cloned the N-oxygenase gene aurF from the
gene cluster encoding aureothin biosynthesis in Streptomyces
thioluteus.[21] Inactivation and complementation studies
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
revealed that AurF plays a crucial role in the formation of
the novel polyketide synthase starter unit p-nitrobenzoate
(PNBA) from p-aminobenzoate (PABA),[14] and that the
involvement of the cytochrome P450 monooxygenase AurH in
N-oxygenation can be ruled out.[22] Furthermore, aurFdeficient null mutant permitted mutasynthesis of the aureothin derivative, aureonitrile.[23] Aside of PrnD from the
pyrrolnitrin pathway in Pseudomonas pyrrocinia,[13] AurF is
only the second known N-oxygenase involved in the biosynthesis of a nitro-group-containing metabolite. Whereas PrnD
shows a clear Rieske motif, AurF appears to represent a novel
type of oxygenase, as it does not share any of the known
conserved oxidoreductase motifs (such as those for cofactor
binding, for example).
A PABA-to-PNBA oxidation assay with AurF appeared
as an ideal model system for gaining a deeper insight into
nitro group formation. Although the expression of AurF was
successful, all attempts to promote an in-vitro oxygenation
failed, probably the result of a missing and yet unknown
cofactor. It was reported earlier that cell-free extracts of
S. thioluteus are unable to carry out N-oxygenation.[24] For this
reason, aurF was expressed in the heterologous host S. lividans, and the oxygenation was performed as a whole-cell assay.
Comparison of the whole-cell activities of recombinant and
wild-type strains[24] implies that the heterologously produced
AurF exhibits a slightly lower conversion rate, albeit in the
same order of magnitude as that of the native enzyme.
However, as all attempts at genetic manipulation of the wildtype S. thioluteus have failed, this functional vector?host
system represents a valuable and flexible model system. A
host strain of S. lividans containing the empty vector and
phosphate buffer were used as negative controls in all
The putative PABA-derived hydroxylamine and nitroso
arene intermediates are known to be highly reactive and
unstable compounds. Therefore, the rapid and manipulationfree isolation and analysis of the corresponding benzoates
from aqueous solution was a challenge. To avoid the
extraction of labile intermediates, an at-line detection
method was established analogously to a protocol for the
analysis of acylated anilines from human whole blood.[25] The
supernatant of centrifuged samples (1 mL) was loaded
directly onto a RESOURCE RPC-18 column and eluted
with a gradient of aqueous Na2HPO4-buffered acetonitrile/
methanol. The method was thoroughly validated, and a very
low detection limit (signal/noise ratio 3:1) for PABA
(0.4 mg mL 1) and PNBA (0.6 mg mL 1) was detected with
UV/Vis spectroscopy at l = 280 nm. With this setup, the
course of the oxygenation reaction (PNBA formation and
PABA consumption) was observed. First, the N-oxygenation
reaction was relatively slow (> 12 h to reach completion).
Second, LC analyses revealed the simultaneous formation of
an intermediate with PNBA production. Its shorter retention
time relative to PABA and the UV/Vis spectrum suggested
that this intermediate could be p-hydroxylaminobenzoic acid
(PHABA; Figure 1).
Comparison with an authentic PHABA reference, prepared from PNBA with zinc and ammonium chloride in
aqueous NaOH solution,[26] unequivocally confirmed this
Angew. Chem. Int. Ed. 2005, 44, 4083 ?4087
Figure 3. Stoichiometry of the N-oxygenation reaction. Time-dependent
concentrations of PABA, PHABA, and PNBA under A) starvation conditions (secondary metabolism predominates) and B) growth (primary
metabolism predominates); ^ PABA, & PHABA, ~ PNBA, sum.
Figure 1. a) Chromatogram of PABA oxidation assay after 5 h; insets:
b) MS?MS data for 10; UV/Vis spectra for c) PHABA; d) PABA;
e) PNBA.
finding. Moreover, synthetic PHABA was readily transformed into PNBA when administered to the AurF-producing
strain, whereas the control strain was not capable of this
transformation. This result shows that not only the first
monooxygenation step, but also the second oxygen transfer is
catalyzed by AurF. In fact, PNBA formed from exogenously
supplied PHABA is produced approximately fivefold faster
than it is formed from PABA (Figure 2). This result provides
Figure 2. Rate of PNBA formation v with PABA or PHABA as substrates; & PNBA from PABA, & PNBA from PHABA.
the strongest evidence that PHABA is indeed a true
intermediate in nitro group formation and that the Noxygenation catalyzed by AurF occurs in a stepwise manner.
Furthermore, the first oxygenation reaction to hydroxylamine
appears to be the rate-limiting step for the overall reaction
in vivo.
To gain deeper insight into the course and stoichiometry
of the N-oxygenation reaction, the fates of PABA, PHABA,
and PNBA were monitored by LC (Figure 3). Under starvation conditions (A), the biotransformation of PABA to PNBA
clearly predominates. The concentrations of both hydroxylamine and PNBA increase while the concentration of PABA
decreases; the sum of all three remains constant. After three
Angew. Chem. Int. Ed. 2005, 44, 4083 ?4087
days, however, before all exogenously supplied PABA is
converted into PNBA, cell lysis with concomitant release of
new substrates initiates the continuation of cell division and
growth. At this stage (B), primary metabolism sets in with
predominant consumption of PABA for the biosynthesis of
folic acid. This finding is in full accord with our earlier
observation that feeding PABA to a growing culture (log
phase) results in a significant decrease in the transformation
of PABA into PNBA.[27] Moreover, the same phenomenon,
the initiation of near-complete PABA consumption after
three days, was observed in a control experiment with the
strain lacking the aurF gene. Contrary to this, the formation of
PNBA is detectable in cells expressing aurF even without
supplemented PABA, which indicates that PABA is withdrawn from primary metabolism. Crosstalk between primary
and secondary metabolism takes place as a result of the dual
role of PABA as precursor for the polyketide synthase starter
unit PNBA as well as for folic acid. Furthermore, the course
and exact stoichiometry of the reaction under starvation
conditions strongly support the key role of PHABA in nitro
group formation.
Despite the high sensitivity of the analytical method, the
proposed nitroso intermediate 9 (see Scheme 1), which
resulted from the subsequent oxygen transfer to PHABA,
could not be detected in the medium. The absence of a free
nitroso intermediate in the reaction mixture may be rationalized by the rapid and possibly spontaneous turnover to the
nitro compound. However, at l = 360 nm it was possible to
identify small amounts of another metabolite that derived
from administered PABA and PHABA, respectively, which
implies the intermediacy of a nitroso compound. MSn
analyses and comparison with a synthetic sample proved
that this novel metabolite is identical to azoxybenzol-4,4?dicarboxylic acid (10). According to a number of mechanistic
investigations, azoxybenzene derivatives are formed by the
condensation of hydroxylamines with nitroso compounds, or
by the dimerization of nitrosobenzenes (Scheme 2).[17, 28, 29] In
both cases, nitroso benzoate intermediates are involved.
However, it was also possible to observe the nonenzymatic
formation of 10 from synthetic PHABA in phosphate buffer
which seems to result from a spontaneous oxidation of
PHABA under aerobic conditions. Nonetheless, the kinetic
data clearly show that AurF also promotes the oxidation of
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Scheme 2. Formation of azoxy compound 10 from PHABA.
PHABA to PNBA. Interestingly, 10 was previously identified
as a metabolite of the insect parasitic zygomycete Entomophthora virulenta.[30, 31] The detection of 10 as a by-product in
PABA oxidation thus not only supports the existence of
PHABA, but also provides the first model for the biosynthesis
of this unusual fungal azoxy metabolite.
Many speculations have been made on the biosynthesis of
nitro compounds from amines,[15] yet until now, no hard
evidence has been reported. Through the extraction-free atline detection of PHABA, and by successful transformation of
synthetic PHABA to PNBA, we provide herein the first direct
evidence for a stepwise oxidation of aromatic amines to nitro
groups via p-hydroxylamine as a key intermediate. Our
results clearly demonstrate that AurF is a monooxygenase,
which not only catalyzes the formation of hydroxylamine, but
also the subsequent oxidation to the aromatic nitro group,
presumably through the same catalytic mechanism
(Scheme 1). Thus, regarding the intermediates, the sequential
N-oxygenation by AurF represents the reverse of the stepwise
enzymatic reduction of nitroaromatic compounds, as in the
case of biodegradation of TNT.[32]
Experimental Section
Strains and cultivation: For heterologous expression, aurF was
amplified by PCR from cosmid pHJ48, cloned into the expression
vector pWHM4* downstream of the constitutive ermE* promoter.
The resulting plasmid, pRW01, was introduced into Streptomyces
lividans ZX1. A strain containing only the plasmid, but not the AurFencoding insert, S. lividans ZX1/pWHM4*, was used as control. Both
strains were stored as glycerol cell banks at 80 8C. A cell bank
volume of 1 mL was used to inoculate 100 mL soy 2G medium (soy
flour (15 g L 1), glucose (15 g L 1), NaCl (5 g L 1), CaCO3 (1 g L 1),
KH2PO4 (0.3 g L 1), pH adjusted to 6.9 prior to sterilization)
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
containing thiostrepton (25 mg mL 1). Growth proceeded for 3 days
at 28 8C in a 500-mL Erlenmeyer flask on a rotary shaker at 180 rpm.
Seed culture (10 mL) was grown in 1-L Erlenmeyer flasks containing
250 mL B 6 (glucose (20 g L 1), oatmeal (20 g L 1), yeast extract
(3 g L 1), NaCl (3 g L 1), CaCO3 (3 g L 1), FeSO4�H2O (0.57 g L 1),
MnCl2�H2O (0.6 g L 1)) containing thiostrepton (25 mg mL 1) for
4 days at 28 8C on a rotary shaker at 180 rpm.
The deduced amino acid sequence of AurF has been deposited
under accession number CAE 02601 at EMBL/GenBank.
N-Oxygenation assay: Culture broth of S. lividans ZX1/pRW01
was filtered through MN640 filter paper, and portions of wet cell mass
(20 g) were resuspended in 500-mL Erlenmeyer flasks with 100 mL
PBS (Na2HPO4�H20 (1.44 g L 1), KH2PO4 (0.24 g L 1), NaCl
(8.00 g L 1), KCl (0.20 g L 1)), pH 7. After shaking for 15 min on a
rotary shaker at 28 8C and 180 rpm, blank samples were taken, and
substrates were added from a stock solution ( 20 mg mL 1) in
methanol to yield a final concentration of 0.07 mg mL 1. The
enzymatic reaction was measured ?at-line? by chromatographic
determination of PABA, PHABA, and PNBA from 1-mL samples.
Additional samples were frozen at 20 8C for later analysis.
LC Measurement of PABA, PHABA, PNBA: LC analysis was
performed on an KTAexplorer system equipped with a
RESOURCE RPC-18 column (3 mL, Amersham Biosciences).
Samples (1 mL) were centrifuged for 1 min at 12 500 rcf and the
supernatant was subsequently injected into a 100-mL sample loop
through a 0.22-mm PES syringe filter. The chromatography method
starts with 5 CV eluent A (CV = column volume), followed by a
gradient of 5 CV from 0 to 100 % eluent B, and finally 5 CV eluent B,
each with a flow rate of 2 mL min 1. Eluent A consists of acetonitrile
(12 % v/v) in Na2HPO4 (10 mm), pH 3; eluent B is methanol. This
methanol gradient is essential for the elution of PNBA, and also
permits the detection of more hydrophobic compounds. UV/Vis
absorption was monitored at l = 280, 325, and 360 nm. These
detection wavelengths are based on results from previous UV/Vis
scans of possible substrates and products. Peaks were identified and
quantified by comparison with external reference standards. Peak
fractionation was used to isolate 4,4-diazoxybenzoate for MS?MS
Received: January 31, 2005
Revised: March 23, 2005
Published online: June 1, 2005
Keywords: azoxides � enzyme catalysis � natural products �
nitro compounds � oxygenases
[1] J. He, N. Magarvey, M. Piraee, L. C. Vining, Microbiology 2001,
147, 2817.
[2] K. Arima, H. Imanaka, M. Kousaka, A. Fukuda, G. Tamura,
Agric. Biol. Chem. 1964, 28, 575.
[3] Y. Hirata, H. Nakata, K. Yamada, K. Okuhara, T. Naito,
Tetrahedron 1961, 252.
[4] R. Cardillo, C. Fuganti, D. Ghiringhelli, D. Giangrasso, P.
Grasselli, Tetrahedron Lett. 1972, 4875.
[5] J. P. Cosyns, Drug Saf. 2003, 26, 33.
[6] M. Lang, P. Spiteller, V. Hellwig, W. Steglich, Angew. Chem.
2001, 113, 1749; Angew. Chem. Int. Ed. 2001, 40, 1704.
[7] R. L. Baxter, A. B. Hanley, H. W. S. Chan, S. L. Greenwood,
E. M. Abbot, I. J. McFarlane, K. Milne, J. Chem. Soc. Perkin
Trans. 1 1992, 2495.
[8] E. Rssner, A. Zeeck, W. A. Knig, Angew. Chem. 1990, 102, 84;
Angew. Chem. Int. Ed. Engl. 1990, 29, 64.
[9] M. Brandl, S. I. Kozhushkov, B. D. Zlatopolskiy, P. Alvermann,
B. Geers, A. Zeeck, A. de Meijere, Eur. J. Org. Chem. 2004, 123.
[10] X. K. Wang, Y. R. Zhao, T. F. Zhao, S. Lai, C. T. Che, Phytochemistry 1994, 35, 263.
Angew. Chem. Int. Ed. 2005, 44, 4083 ?4087
[11] J. A. Kers, M. J. Wach, S. B. Krasnoff, J. Widom, K. D. Cameron,
R. A. Bukhalid, D. M. Gibson, B. R. Crane, R. Loria, Nature
2004, 429, 79.
[12] M. R. Buddha, T. Tao, R. J. Parry, B. R. Crane, J. Biol. Chem.
2004, 279, 49 567.
[13] S. Kirner, P. E. Hammer, D. S. Hill, A. Altmann, I. Fischer, L. J.
Weislo, M. Lanahan, K.-H. van Pee, J. M. Ligon, J. Bacteriol.
1998, 180, 1939.
[14] J. He, C. Hertweck, J. Am. Chem. Soc. 2004, 126, 3694.
[15] W. Steglich, B. Fugmann, S. Lang-Fugmann, Rmpp Encyclopedia of Natural Products, Georg Thieme, Stuttgart, 2002.
[16] R. L. Baxter, S. L. Greenwood, J. Chem. Soc. Chem. Commun.
1986, 175.
[17] N. Itoh, N. Morinaga, T. Kouzai, Biochem. Mol. Biol. Int. 1993,
29, 785.
[18] M. D. Corbett, B. R. Chipko, A. O. Batchelor, Biochem. J. 1980,
187, 893.
[19] S. Kirner, K.-H. van Pee, Angew. Chem. 1994, 106, 346; Angew.
Chem. Int. Ed. Engl. 1994, 33, 352.
[20] S. Kirner, S. Krauss, G. Sury, T. Lam, J. M. Ligon, K.-H. van Pee,
Microbiology 1996, 142, 2129.
[21] J. He, C. Hertweck, Chem. Biol. 2003, 10, 1225.
[22] J. He, M. Mller, C. Hertweck, J. Am. Chem. Soc. 2004, 126,
16 742.
[23] M. Ziehl, J. He, H.-M. Dahse, C. Hertweck, Angew. Chem. 2005,
117, 1226; Angew. Chem. Int. Ed. 2005, 44, 1202.
[24] S. Kawai, K. Kobayashi, T. Oshima, F. Egami, Arch. Biochem.
Biophys. 1965, 112, 537.
[25] R. M. Lindsay, A. M. McLaren, J. D. Baty, J. Chromatogr. 1988,
433, 292.
[26] H. Bauer, S. M. Rosenthal, J. Am. Chem. Soc. 1944, 66, 611.
[27] R. Winkler, C. Hertweck, unpublished results.
[28] G. A. Russel, E. J. Geels, J. Am. Chem. Soc. 1965, 87, 122.
[29] I. K. Kim, J. Whang, J. Korean Chem. Soc. 1976, 20, 56.
[30] N. Claydon, J. F. Grove, J. Chem. Soc. Perkin Trans. 1 1978, 171.
[31] N. Claydon, J. Invertebr. Pathol. 1978, 32, 319.
[32] N. G. McGormick, F. E. Feeherry, H. S. Levinson, Appl. Environ. Microbiol. 1976, 31, 949.
Angew. Chem. Int. Ed. 2005, 44, 4083 ?4087
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Без категории
Размер файла
372 Кб
oxidation, hydroxylaminen, aminoarenes, enzymatic, sequential, via, nitroarenes
Пожаловаться на содержимое документа