Microbiology (1994), 140, 191-195 Printed in Great Britain

Purification, properties and heterologous

expression of formamidase from
Methylophilus methylotrophus
Neil R. Wyborn, Diana J. Scherrt and Colin W. Jones

Author for correspondence: Colin W. Jones. Tel. +44 533 523458. Fax: +44 533 523369.

Department of Biochemistry, The obligately methylotrophic bacterium Methy/ophi/us methylotrophus

University of Leicester,
Leicester LEI 7RH, UK
readily uses formamide as a source of nitrogen for growth. Physiological
investigations using batch, fed-batch and continuous cultures indicated that
the organism contains a discrete formamidase (formamide amidohydrolase;
EC 3.5.1.49), in addition to the previously characterized aliphatic amidase
(acylamide amidohydrolase; EC 3.5.1.4) which is specific to short-chain
aliphatic amides such as acetamide, propionamide and acrylamide.
Formamidase synthesis was induced by formamide and acetamide, and
repressed b y ammonia. The enzyme was purified using anion-exchange and
gel-filtration FPLC and shown to exhibit a narrow substrate specificity (high
activity with formamide, little activity with short-chain aliphatic amides, no
activity with urea). SDS-PAGE and gel-filtration FPLC showed that the enzyme
comprises a single type of subunit and probably exists as a homodimer. A
3.2 kbp Pstl restriction fragment of chromosomal DNA from M.
methyiotrophus was cloned in pUC19 and expressed in Escherichia coli JM109.
The purified gene product exhibited essentially identical properties to those of
the M. methylotrophus formamidase in terms of its chromatographic
behaviour, native M, (123000), subunit M, (51 000), Kcat(58 cf. 64 s-I), K,,, (1.6 cf.
2.1 mM), substrate specificity and N-terminal amino acid sequence (MKTIV-).

Keywords : MetbylophiluJ metbylotrophus, formamidase, molecular cloning, heterologous

expression

INTRODUCTION
A wide range of bacteria are able to utilize short-chain
aliphatic amides (e.g. acetamide, propionamide, acryl-
amide) as sources of nitrogen for growth by virtue of their
ability to hydrolyse these amides to ammonia and the
corresponding acid using an aliphatic amidase (acylamide
amidohydrolase; EC 3 . 5 . 1 .4). The obligately methylo-
trophic bacterium Metbylophilus metblotrophzls contains an
amidase of this type, the purification and biochemical
properties of which from wild-type and mutant strains of
this organism have been described in detail (Silman e t al.,
1989, 1991). The physiological and/or genetic regulation
and biochemical properties of aliphatic amidases from
other species of bacteria including Psezkdomonas aeruginosa
(Clarke, 1970, 1984; Ambler e t a/., 1987; Brammar e t al.,
1987 ; Clarke & Drew, 1988), Alcaligenes ezltrophus
(Friedrich & Mitrenga, 1981) and Brevibacterizkm sp. 312
t Presentaddress: Carbury Heme Ltd, Research and Development Centre,
University of Kent, Canterbury CT2 7PD, UK.
(Thiery etal., 1986; Maestracci etal., 1988) have also been
extensively investigated. Most of these enzymes are at best
only weakly induced by formamide, and exhibit only low
activities with formamide as substrate in spite of being
able to effect the rapid hydrolysis of other short-chain
aliphatic amides.
In contrast, relatively few species of bacteria have been
shown to hydrolyse formamide at high rates. There is
some evidence, mainly from physiological studies, that
Mycobacterium spp., Alcaligenes ezltrophus and Brevibacterizkm
sp. 312 contain discrete formamidases (formamide amido-
hydrolase; EC 3 . 5 . 1 .49) which are induced by both
formamide and acetamide (Draper, 1967 ; Friedrich &
Mitrenga, 1981 ; Maestracci e t a/., l988), but, as far as we
are aware, the biochemical properties of these enzymes
have not been described.
In this paper we show that hl. metbylotrophus contains a
discrete formamidase in addition to the previously char-
acterized short-chain aliphatic amidase (Silman e t al.,
1989, 1991 ; Carver & Jones, 1993). The physiological

0001-84850 1994 SGM 191

Downloaded from www.microbiologyresearch.org by

IP: 122.15.200.115
On: Tue, 13 Oct 2015 05:23:53
N. R. W Y B O R N , D. J. S C H E R R a n d C. K'. J O N E S
regulation and biochemical properties of the enzyme are
described, together with the molecular cloning of the
formamidase structural gene and its expression in Escberi-
cbia coli.
Bacterial strains and plasmids. The bacterial strains and
plasmids used in this work were Metbylophilm metbylotrophw
(NCIB 10515), Escherichia coli JM109 [recA1 szlpE44 endAI
hsdRI7 gyrA96 F' (traD36 proAB' l a d q lacZAM15) relAl tbi
A(lac-proAB)] and pUC19 (amp lacz').
Maintenance of bacterial cultures. Stock cultures of M .
metbylotropbzls and E . coli JM109 were stored in 20% (v/v)
sterile glycerol at -70 "C and were routinely maintained on
methanol-mineral salts-ammonia agar plates (Silman e t al.,
1989) and M9 glucose-mineral saltsammonia agar plates
(Sambrook e t al., 1989) respectively.
Growth of M. methylotrophus.M . metbylotrophw was grown in
methanol-mineral salts medium at 37 "C in batch, fed-batch and
continuous culture with ammonia, formamide or acetamide as
the source of nitrogen essentially as described by Silman e t al.
(1989). During growth in continuous culture under formamide
limitation (37 "C, p H 7.0, D 0.05 h-'), the input formamide
concentration was 0.16 g 1-l. E. coli was grown in M9 glucose-
mineral salts medium (Sambrook e t al., 1989) at 37 "C in batch
culture with 20 mM ammonia, formamide or acetamide as the
source of nitrogen; during the growth of clones carrying
pUC19, the medium was supplemented with ampicillin (50 pg
ml-l). The M . metbylotropbzrs cell density (mg dry wt ml-l) was
measured from the optical density at 600 nm multiplied by 0-53,
and the E . coli cell density was measured from the optical density
at 680 nm multiplied by 0-63. Culture supernatants were
prepared by centrifuging culture samples for 1 min at maximum
speed in an MSE Microcentaur bench-top centrifuge; the
concentration of ammonia in the supernatants was measured
colorimetrically as described previously (Silman e t al., 1989).
Harvesting and preparation of washed or broken cells for
measurement of amidase activity. Cells were harvested by
centrifugation, then washed and resuspended in 0.1 M citric
acid/0.2 M Na,HPO, buffer pH 6.0 to a cell density of approxi-
mately 5 mg ml-l and, if required, disrupted by sonication
(Silman e t al., 1989).
Purification of formamidase. Harvested cells of M . metbylo-
tropbtks were washed and resuspended in 20 mM bis-Tris buffer
pH 6-8 to a cell density of approximately 20 mg dry wt ml-',
then disrupted by passage four times through an Aminco
French pressure cell at 8000-10000 p.s.i. (about 55-69 MPa).
The broken cells were centrifuged at 40000g for 20 min (4 "C)
to yield a cell-free extract which was subsequently centrifuged at
180000g for 75 min to produce a high-speed supernatant
essentially free of particulate material. The high-speed super-
natant was filtered through an Acrodisc filter (0.2 pm pore size;
Gelman), then loaded on to an FPLC Mono Q anion-exchange
column (Pharmacia) equilibrated with 20 mM bis-Tris buffer
pH 6.8, and eluted using a linear gradient of KC1 (0-1 M over
20 min) at a flow rate of 4 ml min-l. Fractions containing high
formamidase activity were pooled, concentrated by ultra-
filtration, loaded on to an FPLC Superose 6 gel-filtration
column equilibrated with 20 mM bis-Tris buffer, pH 6.8 con-
taining 100 m M KC1, and then eluted with the same buffer at a
flow rate of 0-2 ml min-l. The peak formamidase fraction was
analysed using SDS-PAGE, which showed that the enzyme was
> 90 % pure, and was subsequently stored at -20 "C until
192
Downloaded from www.microbiologyresearch.org by
IP: 122.15.200.115
On: Tue, 13 Oct 2015 05:23:53
required. Purification of recombinant formamidase from E . coli
JM109(pNW3) was carried out as described above except that
cells (5 mg dry wt ml-l) were disrupted by sonication for
6 x 30 s periods in an MSE Soniprep 150 ultrasonic dis-
integrator, and peak fractions from the Mono Q step were
pooled, desalted, concentrated by ultrafiltration and re-
chromatographed using a shallower KC1 gradient (Q-o.35 M
over 24 min) at a flow rate of 4 ml min-l.
Measurement of formamidase activity. Formamidase activity
was determined colorimetrically by measuring the rate of
ammonia formation at pH 6-0 (Silman e t al., 1989) with 50 mM
formamide as substrate. In some experiments, formamide was
replaced by various short-chain aliphatic amides or urea (all
50 mM). Assays were normally carried out singly; duplicate
assays, when performed, were usually accurate to < 5 %.
Where appropriate, specific activities were expressed as the
mean & SEM (number of independent determinations). K , values
for formamide were measured using formamide concentrations
in the range 0-1-5.0 mM and calculated from plots of s / v against
s. Protein was assayed by the method of Bradford (1976) using
the Bio-Rad protein assay reagent according to the manu-
facturer's instructions. Bovine serum albumin was used as a
standard.
Polyacrylamide gel electrophoresis. Discontinuous SDS-
PAGE was carried out as described previously (Silman e t al.,
1989) using the procedures of Hames (1981). SDS-
polyacrylamide gels were stained for protein with Kenacid blue
R and subjected to image analysis using a GDS2000 gel
documentation system (Ultraviolet Products) linked to an IBM-
compatible PC.
Determination of N-terminal amino acid sequence. The N-
terminal amino acid sequence of the purified formamidase
(approximately 2 pg blotted on to a polyvinylidine difluoride
disc) was determined using an Applied Biosystems model 470
gas-phase sequencer.
DNA manipulations. Chromosomal D N A was prepared from
M . metbylotrophzls essentially by the method of Wilson (1990).
Samples of the D N A (10 pg) were restricted overnight at 37 *C
in a mixture containing 2 pl of restriction buffer (Sambrook e t
al., 1989) and 2 pl of PstI (10 U p1-'). The resulting restriction
fragments, together with samples of chromosomal DNA, were
separated by electrophoresis at 80 V for 3 h in 0.6 % (w/v)
agarose. Restriction fragments were removed from agarose gels
either by electroelution (Sambrook e t al., 1989) or using the
Geneclean I1 kit (Bio 101 Inc). Plasmid isolation was carried out
using the mini-prep procedure (Sambrook e t al., 1989). Trans-
formation of E. coli JM109 with pUC19 was carried out using
the RbClJCaCl, procedure (Kushner, 1978).
Chemicals and enzymes. Restriction endonuclease and T 4
ligase were obtained from BRL, and phosphatase from
Pharmacia ; all enzymes were used according to the manu-
facturers' instructions. Protein M, standards were obtained
from Sigma. All other chemicals used were obtained from BDH
or Sigma, and were of the highest grade obtainable.
RESULTS AND DISCUSSION
Physiological regulation of formamidase activity
M . methJylotropbzrswas grown in a methanol-mineral salts
medium in batch, fed-batch and continuous culture with
ammonia, formamide or acetamide as the nitrogen source.
Washed cells were assayed for amidase activity using
formamide or acetamide as substrates (Table 1). No
 Formamidase from Meth_ylophilzasmeth_ylotrophw

Table 7- Formamidase activities of M. rnethylotrophus contains a separate formamidase (induced by formamide

grown in batch, fed-batch and continuous culture
.................................. . ....................
M. meth_ylotroph,w was grown in batch, fed-batch (p 0.06 h-l) and
continuous ( D 0-05 h-l) culture in methanol-mineral salts
medium at 37 "C with ammonia, acetamide or formamide as the
nitrogen source. Cells were harvested and assayed for amidase
activity at 37 "C as described in Methods. Activities are means of
one to three independent determinations.
Culture Nitrogen Limiting Amidase activity
source nutrient [pmol min-l
(mg cells)-']
Formamide Acetamide
Batch Ammonia - 0
Formamide - 0.018
Acetamide - 0.014
Fed-batch Formamide Formamide 0.333
Continuous Formamide Formamide 0.943
Acetamide Acetamide 1.809
amidase activity was exhibited by cells grown on am-
monia, but low activities approaching the calculated rate
at which ammonia must be formed in order to support the
observed growth rates were obtained with formamide as
substrate following growth on formamide, and with both
formamide and acetamide as substrates following growth
on acetamide (the concentrations of ammonia in these
latter two cultures, formed as the result of formamide and
acetamide hydrolysis, were > 2 mM). Very much higher
activities with formamide as substrate were exhibited by
cells grown in fed-batch or continuous culture under
formamide limitation, and also in continuous culture
under acetamide limitation or dual methanol-acetamide
limitation (J. Mills & C. W. Jones, unpublished), where
the concentrations of ammonia, produced as a result of
amide hydrolysis, were undetectably low. It was con-
cluded from these results that M . meth_ylotrophw probably
and acetamide, repressed by ammonia), in addition to the
previously described aliphatic amidase (induced by aceta-
mide, repressed by ammonia, much greater activity with
acetamide and acrylamide as substrates than with form-
amide as substrate; Silman e t al., 1989, 1791; N. R.
Wyborn & C. W. Jones, unpublished).
Purification and properties of formamidase
Formamidase was purified using anion-exchange and gel-
filtration FPLC from cell extracts prepared from M .
metl5Jylotropha.sgrown in continuous culture ( D 0-05 h-l)
under formamide limitation (Table 2). Subsequent purifi-
cations showed the enzyme to have a higher specific
0 activity than originally determined and a K , for form-
0-003 amide of 2.1 mM. SDS-PAGE and gel-filtration FPLC
0-016 gave Mr values of 51 000 and 123000 respectively for the
dissociated and native forms of the purified formamidase,
suggesting that it exists either as a homodimer (a,) or, less
0-053
probably, a homotrimer (a,) ; the N-terminal amino acid
1-940
sequence was MKTIV-. Based on a native Mr of 102000
(a,) and a specific activity of 37.4 pmol min-' (mg
protein)-', the purified enzyme was calculated to have a
Kcat of 64 s-'. In contrast to its high activity with
formamide (100 Yo), the enzyme exhibited very low
activities with other short-chain aliphatic amides includ-
ing propionamide (11 Yo), acetamide (5 Yo), butyramide
(3 Yo) and acrylamide (1 Yo),and showed no activity with
urea. O n the basis of its subunit Mr, oligomeric structure,
N-terminal amino acid sequence and substrate specificity
it was concluded that the formamidase is a different
enzyme from the short-chain aliphatic amidase previously
purified from this organism (Silman e t a/., 1991).
The major physiological role of formamidase in M .
metl5Jylotroph.w is clearly to hydrolyse formamide with the
production of ammonia, which can be used as a source of
nitrogen for growth, and formate. However, in contrast
to short-chain aliphatic acids such as acetate and acrylate,
which are used only poorly by this organism (Carver &
Jones, 1993), formate is readily oxidized to carbon dioxide

Table 2. Purification of formamidase from M. rnethylotrophus

M. methylotrophta was grown in continuous culture at 37 "C under formamide limitation ( D 0.05 h-l).
Washed cells were disrupted using a French press, then centrifuged to produce a high-speed
supernatant from which the formamidase was purified using anion-exchange (Mono Q) and gel-
filtration (Superose 6) FPLC as described in Methods. Formamidase activity was measured at 37 "C
with formamide as substrate.

Fraction Total Specific Total Yield Purification

protein activity activity ("/.) (-fold)
(mg) [pmol min-' (pmol min-l)
(mg protein)-']

Cells 578 0.65 378 100 1.0

Supernatant 180 1-12 202 54 1.7 1

Mono Q 38 5.05 193 51 7.7

Superose 6 17.80 105 28 27.2

193

Downloaded from www.microbiologyresearch.org by

IP: 122.15.200.115
On: Tue, 13 Oct 2015 05:23:53
N. R. W Y B O R N , D. J. S C H E R R a n d C. W. J O N E S

via the action of an NAD+-linked formate dehydrogenase can therefore be predicted to contain approximately
and the resulting NADH is available for oxidation by the 1.4 kbp of DNA.
respiratory chain with the concomitant production of The E . coli JMl09 clone containing pNW3 grew well in
ATP (Patchett e t al., 1985; Jones e t al., 1987). It is likely, batch culture in glucose-mineral salts medium supple-
therefore, that formamide not only provides ammonia as mented with ampicillin and containing either ammonia,
a source of nitrogen for growth, but may also contribute formamide or acetamide as the nitrogen source. Washed
to the energy economy of the cell. cells from all three cultures exhibited substantial form-
amidase activity (Table 3), even though the concentration
Molecular cloning and expression of formamidase of ammonia in each culture at the time of harvesting was
> 0.5 mM (either via inclusion in the original medium or
Chromosomal DNA from M . rneth_ylotrophzlswas restricted as a result of amide hydrolysis during growth). Form-
with PstI and the restriction digest was subjected to amidase activity of cells grown in glucose-mineral
agarose-gel electrophoresis. DNA fragments in the ap- salts-acetamide medium was not enhanced by the pres-
proximate size range 1.9-4-0 kbp were excised from the ence of IPTG, and no formamidase activity was detected
gel and ligated into pUC19, which was then used to in washed cells of E. coli JM109 or JM109(pUC19) (i.e. in
transform E. coli JM109. Potential transformants were host cells alone or in host cells which c o n t a i c J the
spread on glucose-mineral salts-ammonia agar plates plasmid vector but not the 3.2 kbp insert). Washed cells of
supplemented with ampicillin and IPTG X-Gal and + E. coli JM109(pNW3) exhibited the substrate specificity
incubated at 37 "C for 24 h ; at the end of this period, all
white recombinant colonies were streaked on to glucose-
mlneral salts-acetamide agar plates supplemented with 1 2 3 4
ampicillin and incubated at 37 "C for 36 h. The 18 fastest-
growing streaks were tested for growth in batch culture in
glucose-mineral salts-acetamide medium supplemented
with ampicillin. Five of these were able to grow rapidly
under these conditions, but the remainder failed to grow
indicating that they probably appeared as false positives
on the original agar plates (possible due to the presence of
trace amounts of ammonia in the growth medium
resulting from spontaneous hydrolysis of acetamide
and/or to the very low putative acetamidase activity of E.
coli JM109 penicillin amidase). Restriction analysis
showed the presence of pUC19 plus a 3-2 kbp insert in all
five clones suggesting genetic homogeneity; one of these
clones harbouring such a plasmid (designated pNW3) was
chosen for further work. The 3.2 kbp insert was suf-
ficiently large to carry the formamidase structural gene
since the latter encodes a protein of Adr 51 000 and

Table 3. Formamidase activity of E. cofi JM109(pNVV3)

following growth under various conditions
........................................................................................................................................................
E. coli JM109(pUC19) and JM109(pNW3) were grown in batch
culture in glucose-mineral salts medium supplemented with
ampicillin and either ammonia, formamide or acetamide as the
nitrogen source. Washed cells were assayed for amidase activity
with formamide as substrate as described in Methods.

Nitrogen Formamidase
source activity 0.00 0.00 1.24
[pmol min-'
(mg cells)-'] .........................................................................................................................................................
Figrn1. E. coli JM109, JM109(pUC19) and JMlOg(pNW3) were
grown in batch culture in glucose-mineral salts medium with
JM109(pUC19) Ammonia 0 ammonia and/or acetamide as nitrogen source; the medium
Ammonia + formamide 0 was supplemented with ampicillin for the growth of
Ammonia + acetamide 0 JM109 (pUC19) and JM109 (pNW3). Cellular proteins were
subjected to SDS-PAGE and then stained with Kenacid blue R.
JM109(pNW3) Ammonia 0.500 k0-049 (8) The numbers under the tracks indicate whole cell formamidase
Formamide 0.553 k0.049 (8) activities [pmol min-I (mg cells)-1]. Tracks: 1, JM109; 2,
Acetamide 1*25Ok0-049 (17) JM109(pUC19); 3, JMlOg(pNW3); 4, M. methylotrophus
formamidase.

Downloaded from www.microbiologyresearch.org by

IP: 122.15.200.115
On: Tue, 13 Oct 2015 05:23:53

expected of Ad. metbylotrophus formamidase, i.e. activities
with acetamide and other aliphatic amides were < 8 % of
the activity with formamide, and no activity was observed
with urea. SDS-PAGE of cellular proteins from E. coli
JM109(pNW3) showed the presence of a protein of
Mr 51000 which was absent from E. coli JM109 or
JMl09(pUC19) (Fig. 1). It was concluded from these
results that expression of fmd in E. coli JM109(pNW3) is
at least partially constitutive (i.e. substantial activity is
expressed in the absence of formamide and acetamide;
contrast M . metbylotrophus) and probably occurs off its
own promoter in the insert rather than offthe lac promoter
in the vector.
Unequivocal confirmation that the M . methylotropbus
formamidase had been expressed in E. coli was sought by
purifying the enzyme from E. cali JM109(pNW3) using
anion-exchange and gel-filtration FPLC as used pre-
viously for purification of the enzyme from M . metbylo-
truphus. The purified enzyme exhibited a very similar
specific activity L34-4cf. 37.4 pmol min-l (mg protein)-'],
Kcat (58 cf. 64 s-') and K , (1.6 cf. 2.1 mM) to that of the
M . metbylotropbus formamidase, and the two enzymes were
identical in terms of their chromatographic properties,
subunit and native Adr values (51 000 and 123000 respec-
tively), substrate specificities (formamidase % propion-
amide/acetamide/butryamide/acrylamide > urea) and
N-terminal amino acid sequences (MKTIV-). It was
concluded from these results that the M . metbylotrophus
formamidase had been successfully expressed in E. coli. A
molecular genetic analysis of the putative fmd operon is
currently underway.
ACKNOWLEDGEMENTS
T h e authors arc indebted t o Mark Carver for bringing the likely
presence of a formamidase i n M.metbyylotrophus t o their attention,
to Nigel Silman for help and advice with the first part of the
work, t o Jim Mills and Steve Williams for useful discussions,
and to Katherine Lilley and Elizabeth Cavanagh for carrying
out the N-terminal amino acid sequencing. N. R. W. was
supported by an SERC-CASE studentship in collaboration with
Z E N E C A BioProducts.
REFERENCES
Ambler, R. P., Auffret, A. D. & Clarke, P. H. (1987). The amino acid
sequence of the aliphatic amidase from Pseudomonas aerzlginosa.
FEBS Lett 215, 285-290.
Bradford, M. M. (1976). A rapid and sensitive method for the
quantitation of microgram quantities of protein using the principle
of protein-dye binding. Anal Biochem 72, 248-254.
Brammar, W. J., Charles, I. G., Matfield, M., Cheng-Pin, L., Drew,
R. E. & Clarke, P. H. (1987). The nucleotide sequence of the amiE
gene of Pseudomonas aerr@zosa.FEBS Lett 215, 291-294.
Carver, M. A. & Jones, C. W. (1993). Production and applications
of amidase from Metbylophilus metbylotrophus. In Microbial Growth on
Downloaded from www.microbiologyresearch.org by
IP: 122.15.200.115
On: Tue, 13 Oct 2015 05:23:53
Formamidase from Metbyyluphilas methylotruphzls
C, Compounds, pp. 365-379. Edited by J. C. Murrell & D. P. Kelly.
Andover : Intercept.
Clarke, P. H. (1970). The aliphatic amidases of Psewdomonas aeru-
ginosa. A d v Microb Pbysiol4, 179-222.
Clarke, P. H. (1984). Amidases of Pseudomonas aeruginosa. In Micro-
organisms as Model Sjstems for Stztdying Evalzttkon, pp. 187-231.
Edited by R. P. Mortlock. New York: Plenum Press.
Clarke, P. H. & Drew, R. (1988). An experiment in enzyme
evolution. Studies with Pseudomonas aeruginosa amidase. Bioscience
Rev 8, 103-120.
Draper, P. (1967). The aliphatic acylamide amidohydrolase of
Mycobacterzum smegmatis: its inducible nature and relation to acyl
transfer to hydroxylamine. J Gen Microbiol 46, 111-123.
Friedrich, C. G. & Mitrenga, G. (1981). Utilization of aliphatic
amides and formation of two different amidases by ,4lcaligenes
eutrophus. ] Gen Microbiol 125, 367-374.
Hames, B. D. (1981). An introduction to polyacrylamide gel
electrophoresis. In Gel Electrophoresis of Proteins; A Practical
Approach, pp. 1-91. Edited by B. D. Hames & D. Rickwood.
Oxford: IRL Press.
Jones, C. W., Greenwood, J.A., Burton, S.M., Santos, H. &
Turner, D. L. (1987). Environmental regulation of methanol and
formaldehyde metabolism in h!ktbylopbilzls metLylotrophus. J Gen
Microbiol 133, 1511-1519.
Kushner, 5. R. (1978). An improved method for transformation of
E. coli with ColEl-derived plasmids. In Genetic Engineering, pp.
17-23. Edited by H. B. Boyer & S. Nicosia. Amsterdam: Elsevier-
North Holland.
Maestracci, M., Bui, K., Thiery, A., Arnaud, A. & Galzy, P. (1988).
The amidases from a Brevibacterizlm strain : study and applications.
A d v Biochem Eng-Biotechnol36, 67-1 15.
Patchett, R. A., Quilter, J. A. & Jones, C. W. (1985). Energy
conservation in whole cells of the methylotrophic bacterium
Metbylophilus metbylotrophzls. Arch Microbioll41, 95-1 02.
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). illolecular Cloning;
A Laborator_y iVanual. Cold Spring Harbor, N Y : Cold Spring
Harbor Laboratory.
Silman, N. J., Carver, M. A. & Jones, C. W. (1989). Physiology of
amidase production by Methjlophilus metLylotrophus: isolation of
hyperactive strains using continuous culture. J Gen Microbiol 135,
3153-3164.
Silman, N. J., Carver, M. A. & Jones, C. W. (1991). Directed
evolution of amidase in iVletLylopbilus methylotrophus : purification
and properties of amidases from wild-type and mutant strains. J Gen
Microbioll37, 169-178.
Thiery, A., Maestracci, M., Arnaud, A., Galzy, P. & Nicolas, M.
(1986). Purification and properties of an acylamide amidohydrolase
(EC 3 . 5 . 1 .4)with a wide activity spectrum from Brevtbacterizlm sp.
R312. J Basic Microbiol5, 299-31 1.
Wilson, K. (1990). Preparation of genomic D N A from bacteria. In
Czlrrent Protocols in Molecuhr Biology, Unit 2.4. Edited by F. M.
Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman,
J. A. Smith & K. Struhl. New York: John Wiley.
Received 7 July 1993; revised 25 August 1993; accepted 31 August 1993.
195