Enantioselective desymmetrisation of citric acid catalysed by the substrate-tolerant petrobactin biosynthetic enzyme AsbA

Article abstract of DOI:10.1039/b823147h

AsbA catalyses the highly enantioselective desymmetrisation of citric acid via ATP-dependent condensation with spermidine, as well as the condensation of citric acid with several spermidine analogues and the condensation of the citric acid analogue tricar

Full text of DOI:10.1039/b823147h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Enantioselective desymmetrisation of citric acid catalysed by the
substrate-tolerant petrobactin biosynthetic enzyme AsbAw
Daniel Oves-Costales, Lijiang Song and Gregory L. Challis*
Received (in Cambridge, UK) 24th December 2008, Accepted 28th January 2009
First published as an Advance Article on the web 16th February 2009
DOI: 10.1039/b823147h
AsbA catalyses the highly enantioselective desymmetrisation of
citric acid via ATP-dependent condensation with spermidine, as
well as the condensation of citric acid with several spermidine
analogues and the condensation of the citric acid analogue
tricarballylic acid with spermidine, suggesting that it may be a
useful biocatalyst for asymmetric synthesis.
As citric acid is a prochiral molecule, its desymmetrisation by
the AsbA-catalysed condensation with spermidine would yield
either the enantiomer 2a, the enantiomer 2b, or a mixture of
both. To elucidate the stereochemical outcome of this reaction
we envisioned exploiting si-citrate synthase, which catalyses the
enantiospecific addition of acetyl-CoA enolate to the si face of
the keto group of oxaloacetic acid followed by thioester hydro-
lysis, to prepare homochiral (3R)-[1,2-13C2]citric acid 3 from
[1,2-13C2]acetyl-CoA 4 (Fig. 2).8 We envisaged that the products
deriving from the AsbA-catalysed condensation of spermidine
with the labelled and unlabelled carboxyl groups of (3R)-
[1,2-13C2]citric acid 3 (i.e. those corresponding to the prochiral
carboxyl groups in unlabelled citric acid) could be discriminated
by 13C NMR spectroscopy. Thus, in a one-pot experiment, we
incubated si-citrate synthase, acetyl-CoA synthetase, purified
recombinant AsbA, [1,2-13C2]acetic acid, coenzyme A, ATP,
oxaloacetic acid and spermidine for a total of 6 hours at 22 1C
(see ESIw for details). The resulting 13C-labelled N8-citryl-
spermidine 2 was purified from the reaction mixture by reverse-
phase HPLC, and its identity was confirmed by HRMS (found
322.1899, calculated 322.1883), and 1- and 2-D NMR experi-
ments. The 13C NMR spectrum showed two doublets at 44.5
and 175.6 ppm (see ESIw). A coupling constant of 55 Hz was
observed for each doublet. These data showed that the major
product of the reaction (428 : 1 based on the signal to noise
ratio of the NMR spectrum) is either 2a or 2b and that the
AsbA-catalysed desymmetrisation of citric acid is highly
enantioselective. To determine which of the enantiomers 2a or
2b is the product of the labelling reaction we measured the 13C
NMR spectrum of a mixture of the labelled product and synthetic
racemic unlabelled N8-citryl-spermidine (Fig. 2). Together with
HMBC NMR data obtained for the pure 13C-labelled N8-citryl-
spermidine 2, this allowed us to unequivocally assign the identity
of this compound as (3R)-N8-citryl-spermidine 2a (Fig. 2).
Catalytic enantioselective desymmetrisation of meso compounds
is a powerful strategy in asymmetric synthesis for the generation
of homochiral organic compounds. Most of these processes are
catalysed by homochiral metal-based catalysts.1 There are
relatively few examples of such transformations catalysed by
enzymes.2 Probably the most widely-exploited strategy for
enantioselective desymmetrisation of meso compounds using
enzymes involves the use of hydrolases to catalyse stereoselective
hydrolytic and transesterification reactions.2 Another strategy
is the use of redox processes, such as the asymmetric Baeyer–
Villiger oxidation of prochiral cyclic ketones by FADH2-
dependent monooxygenases.2,3
Petrobactin 1 is a mixed catechol–citrate siderophore found in
numerous Bacillus species4 that has been shown to be required
for growth and virulence of B. anthracis in iron-depleted media
in a mouse model.5 Importantly, recent investigations have
shown that petrobactin is able to evade siderocalin, a protein
from the innate immune system that can sequester the apo- and
ferric forms of different siderophores and thus disrupt the
bacterial iron-uptake process.6 These data suggest that inhibitors
targeting the biosynthesis and uptake of 1 may be useful anti-
anthrax agents. The production of petrobactin in B. anthracis is
directed by the asbABCDEF gene cluster (Fig. 1), which encodes
a unique combination of nonribosomal peptide synthetases
(NRPSs) and NRPS-independent siderophore (NIS) synthetases
and has been extensively studied over the last two years.7 The
biosynthetic roles of each of the enzymes encoded by this gene
cluster are now relatively well understood (Fig. 1). We recently
showed that the NIS synthetase AsbA catalyses the ATP-
dependent condensation of citric acid with N8 of spermidine to
afford N8-citryl-spermidine 2.7c This raised an interesting
question about the stereoselectivity of this desymmetrisation
reaction. Here we report that it is highly enantioselective and
that AsbA can catalyse similar reactions with an analogue of
citric acid and several analogues of spermidine.
Department of Chemistry, University of Warwick, Coventry, UK CV4
7AL. E-mail: g.l.challis@warwick.ac.uk; Fax: +44 (0)247 652 4112;
Tel: +44 (0)247 657 4024
w Electronic supplementary information (ESI) available: Experimental
procedures and spectroscopic data. See DOI: 10.1039/b823147h
We have previously reported that the AsbA-catalysed con-
densation reaction occurs via adenylation of the prochiral
carboxyl groups in citric acid and subsequent nucleophilic
attack of N8 of spermidine on the activated carbonyl carbon
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 1389–1391 | 1389

View Article Online
Fig. 1 Organisation of the asbABCDEF gene cluster that directs petrobactin production in B. anthracis (genes encoding NIS synthetases are in
grey and genes encoding NRPS-like proteins are in black) and the reactions catalysed by the encoded petrobactin biosynthetic enzymes.
Based on the above data it seems reasonable to assume
that the stereochemical outcome of the reaction would be
unaffected by substitution of spermidine with other di- and
triamines. Thus, we incubated purified recombinant AsbA,
citric acid, Mg2+ and ATP with several spermidine analogues
at 37 1C for 90 minutes. We also carried out control reactions
using heat-inactivated AsbA. LC-MS analyses indicated that
in all cases compounds with m/z corresponding to those
expected for the N8-citryl-spermidine analogues (absent from
the control reactions) were produced. The reactions where
significant quantities of analogues were produced were scaled-
up and the products were partially purified using reverse-phase
HPLC. The identities of the resulting analogues were con-
firmed as 5–8 by HRMS and MS/MS analyses (Fig. 3 and
ESIw). Although LC-MS analysis of incubations using
1,4-butanediamine, 1,3-propanediamine and N1-(3,4-dihydroxy-
benzoyl)-spermidine in place of spermidine showed traces of
compounds with m/z expected for analogues 9, 10 and 11
(Fig. 3), no attempt to purify these was undertaken due to the
small amounts of material produced.9 Taken together the data
on the stereochemical outcome of the AsbA-catalysed
condensation of citric acid and spermidine, and the ability of
AsbA to catalyse condensation of citric acid with several
di- and triamine analogues of spermidine suggest that this
enzyme may be a versatile biocatalyst for the preparation of
highly enantiomerically-enriched citrate derivatives via a novel
Fig. 2 Elucidation of stereocontrol in the AsbA-catalyzed condensa-
type of desymmetrisation process.
We also examined the capacity of AsbA to utilise citric acid
tion of citric acide and spermidine. The grey arrow in 2a shows a key
correlation observed in the HMBC spectrum of purified 13C-labelled
N8-citryl-spermidine. Portions of the 13C NMR spectrum of a mixture
at 37 1C for 90 minutes with spermidine, Mg2+, ATP and
of labelled 2a and racemic synthesised 2 revealing doublets flanking
analogues. Thus, we incubated purified recombinant AsbA
each of the following carboxylic acids: 2-ketoglutaric acid,
the signals for C-1 and C-2 are shown below the reaction scheme.
3-ketoglutaric acid, D-glutamic acid, L-glutamic acid,
atom.7c The data reported here indicate that the adenyl-
DL-isocitric acid, glutaric acid and tricarballylic acid. LC-MS
analyses of the reaction mixtures indicated that among the
citrate analogues tested only tricarballylic acid (an analogue of
citric acid lacking the C-3 hydroxyl group) is a productive
substrate for AsbA. The resulting compound was partially
ation of citric acid must be highly enantioselective. Only the
pro-R carboxyl group reacts with ATP and subsequent
reaction of the resulting adenylate with spermidine affords
(3S)-N8-citryl-spermidine 2a.
ꢀc
This journal is The Royal Society of Chemistry 2009
1390 | Chem. Commun., 2009, 1389–1391

View Article Online
procedures and provides direct access to homochiral products
from commercially available starting materials. It will be
interesting to investigate further spermidine analogues in the
AsbA-mediated catalytic enantioselective desymmetrisation of
citric acid to see whether an expanded range of compounds
can be produced. It will also be interesting to examine whether
other NIS synthetases possess similar stereoselectivity.
This work was supported by a grant from the BBSRC
(grant ref. BB/FO13760/1).
Notes and references
1 (a) M. Willis, J. Chem. Soc., Perkin Trans. 1, 1999, 1765–1784;
(b) T. Rovis, in New Frontiers in Asymmetric Catalysis, ed.
M. Koichi and M. Lautens, John Wiley & Sons, Hoboken, USA,
2007, ch. 9, pp. 275–311.
2 E. Garcia-Urdiales, I. Alfonso and V. Gotor, Chem. Rev., 2005,
105, 313–354.
3 M. D. Mihovilovic, Curr. Org. Chem., 2006, 10, 1265–1287.
4 A. T. Koppisch, S. Dhungana, K. K. Hill, H. Boukhalfa,
H. S. Heine, L. A. Colip, R. B. Romero, Y. Shou,
L. O. Ticknor, B. L. Marrone, L. E. Hersman, S. Iyer and
C. E. Ruggiero, BioMetals, 2008, 21, 581–589.
5 S. Cendrowski, W. MacArthur and P. Hanna, Mol. Microbiol.,
2004, 51, 407–417.
6 R. J. Abergel, M. K. Wilson, J. E. L. Arceneaux, T. M. Hoette,
R. K. Strong, B. R. Byers and K. N. Raymond, Proc. Natl. Acad.
Sci. U. S. A., 2006, 103, 18499–18503.
7 (a) J. Y. Lee, B. K. Janes, K. D. Passalacqua, B. F. Pfleger,
N. H. Bergman, H. Liu, K. Hakansson, V. S. Ravindranadh,
C. A. Courtney, S. Cendrowski, P. C. Hanna and
D. H. Sherman, J. Bacteriol., 2007, 189, 1698–1710;
(b) B. F. Pfleger, J. Y. Lee, R. V. Somu, C. C. Aldrich,
P. C. Hanna and D. H. Sherman, Biochemistry, 2007, 46,
4147–4157; (c) D. Oves-Costales, N. Kadi, M. J. Fogg, L. Song,
K. S. Wilson and G. L. Challis, J. Am. Chem. Soc., 2007, 129,
8416–8417; (d) D. Oves-Costales, N. Kadi, M. J. Fogg, L. Song,
K. S. Wilson and G. L. Challis, Chem. Commun., 2008, 4034–4036;
(e) A. T. Koppisch, K. Hotta, D. T. Fox, C. E. Ruggiero,
C. Y. Kim, T. Sanchez, S. Iyer, C. C. Browder, P. J. Unkefer
and C. J. Unfeker, J. Org. Chem., 2008, 73, 5759–5765;
(f) B. F. Pfleger, Y. Kim, T. D. Nusca, N. Maltseva, J. Y. Lee,
C. M. Rath, J. B. Scaglione, B. K. Janes, E. C. Anderson,
N. H. Bergman, P. C. Hanna, A. Joachimiak and
D. H. Sherman, Proc. Natl. Acad. Sci. U. S. A., 2008, 105,
17133–17138; (g) D. T. Fox, K. Hotta, C. Y. Kim and
A. T. Koppisch, Biochemistry, 2008, 47, 12251–12253.
Fig. 3 Structures of citric acid and tricarballylic acid derivatives
produced using analogues of spermidine and citric acid in AsbA-
catalysed reactions. Diagnostic fragment ions observed in MS/MS
spectra of compounds 5–8 and 12 and calculated (C) and found (F)
parent ion m/z values are indicated.
purified by reverse-phase HPLC from a scaled-up incubation
and HRMS and MS/MS analyses were consistent with struc-
ture 12 (Fig. 3 and ESIw). From these experiments we conclude
that AsbA is significantly more selective towards citric acid
than spermidine as a substrate. It would be interesting to
investigate whether AsbA shows the same high degree of
enantioselectivity in the desymmetrisation of tricarballylic acid
as in the desymmetrisation of citric acid. The lack of a
convenient method for the preparation of homochiral labelled
tricarballylic acid prevented us from pursuing this question in
the present study.
8 D. P. Bloxham, D. C. Parmelee, S. Kumar, R. D. Wade,
L. H. Ericsson, H. Neurath, K. A. Walsh and K. Titani, Proc.
Natl. Acad. Sci. U. S. A., 1981, 78, 5381–5385.
9 N1-(3,4-dihydroxybenzoyl)-spermidine is a potential intermediate in
petrobactin biosynthesis. We previously reported that we could
detect no activity with this spermidine analogue as a substrate for
AsbA.7c Here we employed lower concentrations of this compound
than in our previous experiments. High concentrations of the
compound appear to inhibit the enzyme, as suggested by the
decreased amounts of 11 produced with higher concentrations of
N1-(3,4-dihydroxybenzoyl)-spermidine. We employed our pre-
viously reported7d AMP release assay to measure the relative rates
of product formation using concentrations of ATP and citric acid
presumed to be well above the Km, with different concentrations of
N1-(3,4-dihydroxybenzoyl)-spermidine. In all cases the measured
rates were comparable to the background rate of AMP forma-
tion resultingfrom ATP hydrolysis and much lower than the
In conclusion, we have shown that the petrobactin bio-
synthetic enzyme AsbA catalyses the highly enantioselective
desymmetrisation of citric acid via the condensation of one of
its prochiral carboxyl groups with spermidine. The enzyme can
also catalyse the condensation of several di- and triamine
analogues of spermidine with citric acid and the condensation
of the citric acid analogue tricarballylic acid with spermidine.
These data indicate that AsbA could be developed into a
useful and novel biocatalyst for the preparation of homochiral
citric acid derivatives. To the best of our knowledge there are
only two other methods for the preparation of homochiral
citric acid derivatives. One involves the regio- and enantio-
selective esterase-catalysed hydrolysis of citrate triesters.10 The
other involves resolution by repeated fractional crystallisation
of the brucine salts of 1,2-diesters of citric acid.11 The bio-
transformation reported here is highly complimentary to these
rates observed using spermidine as
a substrate. These data
provide further support for our previous conclusion7c,d that
N1-(3,4-dihydroxybenzoyl)-spermidine is not a significant inter-
mediate in petrobactin biosynthesis.
10 R. Chenevert, B. Ngatcha, T. Beatrice, Y. Rose, S. Yannick and
D. Goupil, Tetrahedron: Asymmetry, 1998, 9, 4325–4329.
11 R. J. Bergeron, M. Xin, R. E. Smith, M. Wollenweber,
J. S. McManis, C. Ludit and K. A. Abboud, Tetrahedron, 1997,
53, 427–434.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 1389–1391 | 1391