When we investigated the efficiency of the mutase column alone
with a feed of 1 mM HAB, the substrate was unstable due to auto-
oxidation, which made quantification of the conversion efficiency
difficult. One of the advantages to a continuous reaction system
is the rapid conversion of the unstable HAB intermediate into
ortho-aminophenol.
(1.04 ¡ 0.029 mM) was obtained continuously for 24 h,
demonstrating 100% conversion efficiency and a product forma-
tion rate of 0.24 mg h21 mg21 total protein. The zinc becomes
oxidized over time. When the column was repacked with new zinc,
the system could be continued for a further 24 h.
The flow-through system described can be applied to the
transformation of a wide variety of nitroarene substrates into the
corresponding aminophenolic products while bypassing many of
the current limitations of whole cell biocatalysis. The transforma-
tion of antibiotics with nitro functional groups to the correspond-
ing aminophenols may provide a simple method for synthesising
novel antibiotic analogs. The high efficiency and regiospecificity of
the reported system provides an attractive alternative to conven-
tional chemical synthesis. In addition, the immobilized enzyme can
be recovered and subsequently reused.{
The biosynthesis of antibiotics using intact bacterial cells is
inherently limited due to the biocidal properties of the product. An
immobilized enzyme reaction system therefore provides an
attractive alternative. The use of the zinc/mutase cascade was
therefore investigated with the antibiotic chloramphenicol (N-[2-
hydroxy-1-(hydroxymethyl)-2-(4-nitrophenyl) ethyl]-2,2-dichloro-
acetamide) as a model system (Fig. 2). Analogs of chloramphenicol
that lack the nitro substituent have been investigated,6 but the
formation of an aminophenol analog has not been reported.
Preliminary investigation demonstrated that the nitro group of
chloramphenicol was reduced to a hydroxylamino derivative by
reaction with zinc and the identity of the product was confirmed
by LC–MS analysis (data not shown). When the hydroxylamino
derivative of chloramphenicol was incubated with partially purified
HAB mutase B, the product was not converted to the expected
aminophenol. HAB mutase A, however, converted the hydro-
xylamino derivative to the corresponding aminophenol (N-[2-(4-
amino-3-hydroxy-phenyl)-2-hydroxy-1-hydroxy methyl-ethyl]-2,
2-dichloro-acetamide). This was the first observation of a
difference in substrate specificity between the two enzymes. The
aminophenol was purified from this reaction by HPLC and
characterised by LC–MS (molecular ion, m/z 308) and NMR.{
The success of the batch reaction with chloramphenicol led us to
investigate the synthesis of the product using the continuous zinc/
mutase column system as described above for the nitrobenzene
model system. When an aqueous solution of chloramphenicol
(1 mM) was pumped through the two columns at a flow rate
of 0.25 ml min21, the corresponding aminophenol analog
This work was funded by the US Air Force Office of Scientific
Research. HRL was supported by a postdoctoral fellowship from
the Oak Ridge Institute for Science and Education (US
Department of Energy).
Heather R. Luckarift, Lloyd J. Nadeau and Jim C. Spain*
Air Force Research Laboratory, 139 Barnes Drive, Suite #2, Tyndall
AFB, FL, 32403-5323, USA. E-mail: jim.spain@tyndall.af.mil;
Fax: 850 283 6090; Tel: 850 2836058
Notes and references
{ Biosilica immobilization of partially purified mutase enzyme ({) was
performed as described previously.5 Mutase enzyme activity was deter-
mined as described previously.2b For continuous flow experiments, columns
(XK-16/20, Pharmacia Biotech) were packed with (a) zinc (5 g, 40 mesh)
(total 1 ml volume), (b) immobilized mutase from a 5 ml reaction mixture
(containing approx. 10 mg protein) and 5 g of glass beads (60/80 mesh,
Alltech) (total 10 ml volume). Substrate was pumped through the system at
a fixed flow rate and the eluate collected for analysis. The entire apparatus
was maintained at 30 uC. Substrates were dissolved in water containing
NH4Cl (40 mM) and sparged with argon to ensure an anaerobic
environment. The pH of the reaction remained at approximately pH 7.4
throughout. Reactants and products of nitrobenzene conversion were
monitored by reverse-phase HPLC on a Supelco ABZ column with an
acetonitrile/water (+0.1% trifluoroacetic acid) gradient. Reactants and
products of chloramphenicol transformation were resolved by ion pair
chromatography on a Luna C18 column (150 6 2 mm, Phenomenex) with
an acetonitrile/triethylamine (10 mM, adjusted to pH 5 with acetic acid)
gradient.
1 (a) M. Keitmann, R. Sezi and A. Weber, US Patent 6 320 081; (b)
L. J. Nadeau, Z. He and J. C. Spain, J. Ind. Micro. Biotech., 2000, 24,
301; (c) V. Kadiyala, L. J. Nadeau and J. C. Spain, Appl. Environ.
Micro., 2003, 69, 11, 6520.
2 (a) S. F. Nishino and J. C. Spain, Appl. Environ. Micro., 1993, 59, 8,
2520; (b) J. K. Davis, G. C. Paoli, Z. He, L. J. Nadeau, C. C. Somerville
and J. C. Spain, Appl. Environ. Micro., 2000, 66, 7, 2965.
3 B. S. Furniss, A. J. Hannaford, P. W. G. Smith and A. R. Tatchell,
Vogel’s Textbook of Practical Organic Chemistry, 1989, Longman
Scientific and Technical and John Wiley & Sons, New York.
4 F. Gelman, J. Blum and D. Avnir, J. Am. Chem. Soc., 2002, 124, 48,
14460; O. Pamies and J. E. Backvall, Curr. Opin. Biotech., 2003, 14, 407.
5 H. R. Luckarift, J. C. Spain, R. R. Naik and M. O. Stone, Nat. Biotech.,
2004, 22, 2, 211; R. R. Naik, M. M. Tomczak, H. R. Luckarift,
J. C. Spain and M. O. Stone, Chem. Commun., 2004, 1684.
6 M. D. Corbett and B. R. Chipko, Antimicrob. Agents Chemother., 1978,
13, 2, 193; T. Izard, Protein Sci., 2001, 10, 1508.
Fig. 2 Structure of chloramphenicol (A) and an aminophenol analog (B).
384 | Chem. Commun., 2005, 383–384
This journal is ß The Royal Society of Chemistry 2005