Conversion from arenes having a benzene ring to those having a picolinic acid by simple growing cell reactions using Escherichia coli that expressed the six bacterial genes involved in biphenyl catabolism

Article abstract of DOI:10.1021/ja044850g

The comprehensive bioconversion of aromatic compounds with a benzene ring to a picolinic acid was achieved with a recombinant Escherichia coli strain that expressed the six genes involved in biphenyl catabolism, these being the bphA1(2072)A2A3A4 genes encoding the evolved biphenyl dioxygenase, the bphB gene encoding dihydrodiol dehydrogenase, and the bphC gene encoding catechol 2,3-dioxygenase. Copyright

Full text of DOI:10.1021/ja044850g

Published on Web 10/29/2004
Conversion from Arenes Having a Benzene Ring to Those Having a Picolinic
Acid by Simple Growing Cell Reactions Using Escherichia coli that
Expressed the Six Bacterial Genes Involved in Biphenyl Catabolism
Kazutoshi Shindo,*,† Ayako Osawa,† Ryoko Nakamura,† Yukiko Kagiyama,† Shohei Sakuda,‡
Yoshikazu Shizuri,§ Kensuke Furukawa,| and Norihiko Misawa§
Department of Food and Nutrition, Japan Women’s UniVersity, 2-8-1, Mejirodai, Bunkyo-ku,
Tokyo 112-8681, Japan, Department of Applied Biological Chemistry, Graduated School of Agricultural and
Life Sciences, The UniVersity of Tokyo, 1-1-1, Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan,
Marine Biotechnology Institute, 3-75-1, Heita, Kamaishi-shi 026-0001, Japan, and Graduate School of Bioresource
and BioenVironmental Sciences, Kyushu UniVersity, 6-10-1, Hakozaki, Fukuoka-shi 812-8581, Japan
Received August 25, 2004; E-mail: kshindo@fc.jwu.ac.jp
Scheme 1. Catabolic Pathway from Biphenyl (I) to the
Meta-Cleavage Compound (II)
Aromatic compounds having a picolinic acid (pyridine-2-
carboxylic acid) in the molecules are rare in nature and usually
difficult to chemically synthesize. Using the multiple biocatalytic
functions of cells may, however, enable us to achieve “chemical
synthesis” of a series of organic compounds that are difficult or
impractical to synthesize chemically. We report here a system for
the comprehensive bioconversion of a series of aromatic compounds
with a benzene ring to those with a picolinic acid, which was
performed by simple growing cell reactions using recombinant
microbes. We constructed a recombinant Escherichia coli strain
expressing the six genes involved in biphenyl catabolism, these
being the bphA1(2072)A2A3A4 genes encoding the evolved biphe-
nyl dioxygenase, the bphB gene encoding dihydrodiol dehydro-
genase, and the bphC genes encoding catechol 2,3-dioxygenase.
Their enzymatic functions are shown in Scheme 1.1
the liquid phase was applied to a high-performance liquid chro-
matographic (HPLC) analysis for further purification of the
converted products. The structure of each purified product was
1
determined by HRFAB-MS and by H and 13C NMR data.9
The products converted from the respective substrates were
surprisingly found to be aromatic compounds, in which a benzene
ring had been replaced with a picolinic acid, as shown in Table 1.
Product 11 from (trans)-chalcone showed a reduction of ∆1,2 due
to endogenous activity in the E. coli host.10 Except for the product
from biphenyl, 6-phenylpyridine-2-carboxylic acid (1), all of the
remaining products were novel compounds according to the CAS
database. Each substrate-to-product conversion ratio was in the
range of 13-84% as shown in Table 1.
The bphA1 gene encoded the large (R) subunit of the iron-sulfur
protein of biphenyl dioxygenase which contains the substrate-
binding site.2 The evolved bphA1 (2072) gene,3 which had been
constructed by DNA shuffling between bphA1 from the biphenyl-
degrading soil bacteria, Pseudomonas pseudoalcaligenes KF707 and
Burkholderia sp. LB400, encoded a large subunit with a broad
substrate range.4 This gene was inserted into the SacI/BamHI site
As an example of the structural identification, the molecular
formula of product 6 from flavanone was determined to be
C15H11O4N by HRFAB-MS. The presence of a carboxylic acid
function in 6 was demonstrated by treating 6 with phenasylbromide
and triethylamine to give the phenasyl ester of 6. An analysis of
the DQF COSY and HMQC spectra of 6 showed the A and C rings
in flavanone to have been completely preserved, while the signals
due to the B ring had disappeared. Except for the signals due to
the A and C rings, vicinal coupled sp2 methine signals (C-2′ (δH
8.02, δC 124.3) - C-3′ (δH 8.07, δC 138.7) - C-4′ (δH 7.87, δC
124.3)) and three sp2 quarternary carbons (δC 148.8, δC 157.3, δC
5
of plasmid pJHF18 carrying the bphA2A3A4BC genes derived
from P. pseudoalcaligenes KF707 6 to construct plasmid pSHF1072.
All the six genes in this plasmid were oriented to have transcrip-
tional read-through starting from bphA1 (2072) by the lac promoter
of vector pUC118 upon induction by IPTG.7
We selected biphenyl, 2-phenyl naphthalene, 3-phenylindanone,
2-phenylquinoline, and 2-phenylbenzoxazole, and the six flavonoids,
flavanone, flavone, 6-hydroxyflavanone, 6-hydroxyflavone, 7-hy-
droxyflavanone, and (trans)-chalcone, as substrates for the biotran-
formation experiments.8 E. coli JM109 harboring pSHF1072 was
grown in an LB medium7 containing 150 µg/mL of ampicillin (Ap)
at 30 °C with reciprocal shaking (175 rpm) for 6-7 h until the
absorbance at OD 600 nm had reached approximately 1. Eight
milliliters of this culture was inoculated into 100 mL of an M9
medium7 with 150 µg/mL of Ap, 1 mM of IPTG, 0.4% (w/v)
glucose, and 10 mg of each substrate in a Sakaguchi flask, and
co-cultivated at 30 °C with reciprocal shaking (175 rpm) for 2 days.
To extract the converted products as well as the substrates, a volume
of methanol equal to that of the culture medium was added to the
co-culture and mixed for 30 min. After centrifuging to remove cells,
1
166.1 (carboxylic acid)) were observed in the H and 13C NMR
spectra of 6. The 1H-13C long-range coupling from H-3′ to δC 148.8
(C-5′) and δC 157.3 (C-2′) in the HMBC spectrum and the vicinal
spin couplings of J2′,3′ (7.7 Hz) and J3′,4′ (7.6 Hz) confirmed the
presence of a pyridine ring composed of the carbons of C-2′ to
C-5′. The linkage of C-2′ to the C ring and of C-5′ to carboxylic
acid was determined from the 13C chemical shift data. These
findings collectively indicated 6 to be 6-oxo-6-(4-oxo-chroman-2-
yl)-pyridine-2-carboxylic acid.
The products by the catabolic enzymes, BphA1(2072)A2A3A4BC,
would have been aromatic compounds with a meta-cleavage group,
i.e., ones with 2-hydroxymuconic ꢀ-semialdehyde (Scheme 1). Since
we were unable to detect such meta-cleavage products, these
compounds are considered to have been unstable in the co-culture,
† Japan Women’s University.
‡ The University of Tokyo.
§ Marine Biotechnology Institute.
| Kyushu University.
9
15042
J. AM. CHEM. SOC. 2004, 126, 15042-15043
10.1021/ja044850g CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Bioconversion of Various Aromatic Compounds with
Phenyl Groups by Co-culture with the Cells of E. coli Carrying the
bphA1 (2072), bphA2, bphA3, bphA4, bphB, and bphC Genesa
Figure 1. Scaled-up test on the substrate by using flavanone. Product (g)
represents wt of the purified picolinic acid compound. Yield (%) shows
the ratio, purified product/substrate added.
substantiate that the substrate content was not at too low a level,
we conducted scaled-up experiments on the substrate using fla-
vanone as the example, the results being shown in Figure 1. It was
found to easily scale-up the substrate by 20-fold (0.2% w/v) with
more than a moderate level of yield (40-80%). We have reported
in the present study a simple method for producing various picolinic
acid-incorporating compounds (which are substituents at the C-6
position) from aromatic compounds with a benzene ring by ordinary
co-culture with recombinant E. coli and each substrate. The
comprehensive production of such organic chemicals, which are
very difficult or impractical to chemically synthesize, could intend
to serve as new starting materials for the chemical synthesis of
pharmaceuticals, agrochemicals, and other industrially useful
compounds.
a The percentage values in parentheses represent the conversion ratio of
each substrate to the corresponding product, which was measured by HPLC.
Scheme 2. Proposed Routes for Generating the Aromatic
Compound with a Picolinic Acid from a Compound with a
2-Hydroxymuconic ꢀ-Semialdehyde
Acknowledgment. This work was supported by Biotechnology
and Medical Technology Development Department of New Energy
and Industrial Technology Development Organization (NEDO).
Supporting Information Available: Methods for HPLC analysis
1
and purification of the converted products, and HRFAB-MS, H- and
13C NMR data of 1-11. This material is available free of charge via
the Internet at http://pubs.acs.org.
References
(1) The microbial biodegradation of biphenyl and polychlorinated biphenyls
(PCBs) has been reviewed: (a) Furukawa, K. Curr. Opin. Biotechnol.
2000, 11, 244-249. (b) Furukawa, K.; Suenaga, H.; Goto, M. J. Bacteriol.
2004, 186, 5189-5196.
and rapidly converted to the picolinic acid compounds in a
nonenzymatic manner, the speculated routes being shown in Scheme
2. Ammonia is very likely to have been derived from NH4Cl (1
g/L; 30-40-fold equivalent amounts of the substrates) in the M9
medium. Asano et al. have also found that the meta-cleavage
compound, 2-hydroxymuconic ꢀ-semialdehyde, was biotransformed
from the monocyclic aromatic molecule, catechol, through the cell
suspension of E. coli which expressed a bacterial catechol 2,3-
dioxygenase gene (similar to bphC), and then was chemically
converted to picolinic acid by such severe conditions as stirring in
25% ammonia water or autoclaving with ammonium phosphate.11
We also carried out the biotransformation experiments of flavanone
with different concentrations of NH4Cl in the co-culture to
determine its required concentration.
(2) Carredano, E.; Karlsson, A.; Kauppi, B.; Choudhury, D.; Parales, R. E.;
Parales, J. V.; Lee, K.; Gibson, D. T.; Eklund, H.; Ramaswamy, S. J.
Mol. Biol. 2000, 296, 701-712.
(3) DDBJ/EMBL/GenBank accession no. AB085748.
(4) Misawa, N.; Shindo, K.; Takahashi, H.; Suenaga, H.; Iguchi, K.; Okazaki,
H.; Harayama, S.; Furukawa, K. Tetrahedron 2002, 58, 9605-9612.
(5) Hirose, J.; Suyama, A.; Hayashida, S.; Furukawa, K. Gene 1994, 138,
27-33.
(6) DDBJ/EMBL/GenBank accession no. M83673.
(7) Sambrook, J.; Fritsch, E. F.; Maniatis, T. Molecular Cloning:
A
Laboratory Manual; Cold Spring Harbor Laboratory: Cold Spring Harbor,
New York, 1989
(8) The chemicals used as substrates were purchased from Aldrich Chemical
or Tokyo Kasei Kogyo. The respective substrates were each dissolved in
a small volume of dimethyl sulfoxide (DMSO) and added to the culture.
(9) The structures were analyzed by their mass spectral data [MS (FAB) and
HRMS (FAB); JEOL JMS-700T] and nuclear magnetic resonance (NMR)
spectral data (400 MHz, Bruker AMX400).
(10) Shindo, K.; Kagiyama, Y.; Nakamura, R.; Hara, A.; Ikenaga, H.; Furukawa,
K.; Misawa, N. J. Mol. Catal. B: Enzym. 2003, 23, 9-16.
(11) Asano, Y.; Yamamoto, Y.; Yamada, H. Biosci. Biotechnol. Biochem. 1994,
58, 2054-2056.
Consequently, the picolinic acid product started to be generated
with 0.35 g/L of NH4Cl and reached a plateau between 0.75 and
10 g/L of NH4Cl.
We used 10 mg of each substrate per 100 mL of culture (0.01%
w/v) in all of the foregoing bioconversion experiments. To
JA044850G
9
J. AM. CHEM. SOC. VOL. 126, NO. 46, 2004 15043