Journal of' General Microbiology (1989, 131, 191 1-191 6. Printed in Great Britain
Purification and Molecular and Catalytic Properties of Bromoperoxidase
from Streptomyces phaeochromogenes
By KARL-HEINZ VAN PEE AND FRANZ LINGENS*
Institut fur Mikrobiologie der Universitat Hohenheim, Garbenstrasse 30,
0-7000 Stuttgart 70, FRG
(Received 7 February 1985; revised 9 April 1985)
A bromoperoxidase has been isolated and purified from the chloramphenicol-producing
actinomycete Streptomyces phaeochromogenes. The purified enzyme was homogeneous as
determined by polyacrylamide gel electrophoresis. The prosthetic group of the bromoperoxidase
was ferriprotoporphyrin IX. Based on gel filtration results the molecular weight of the enzyme
was 147 000 & 3000. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis showed a
single band having the mobility of a 72500 molecular weight species. Therefore, in solution at
neutral pH, the bromoperoxidase behaved as a dimer. The isoelectric point was 4.0. The spectral
properties of the native and reduced enzyme are reported. The homogeneous enzyme also had
peroxidase and catalase activity.
Bromoperoxidases, enzymes which can use bromide ions in the presence of hydrogen
peroxide and a halogen acceptor for the catalytic formation of carbon-halogen bonds, have been
isolated from several marine algae (Ahern et al., 1980; Baden & Corbett, 1980; Manthey &
Hager, 198 1) and from the bacterium Pseudomonas aureofaciens (van PCe & Lingens, 1985). The
bacterial enzyme showed many differences from the algal enzymes. Surprisingly, although
chlorinated metabolites are produced by marine algae (Faulkner, 1977) and bacteria
(Neidleman, 1975), only brominating but not chlorinating enzymes have so far been isolated
from these sources. Therefore we decided to use the chloramphenicol-producing actinomycete
Streptomyces phaeochromogenes to obtain more information about bacterial halogenation.
Reagents. Hydrogen peroxide (30%, v/v) was purchased from Merck and t-butyl hydroperoxide (80%, v/v) was
from Fluka. Monochlorodimedone was prepared from dimedone by chlorination with sodium hypochloride
(Hager et al., 1966). o-Dianisidine (3,3'-dimethoxybenzidine) was from Sigma and L-isoleucine was a gift from
Degussa, Konstanz, FRG.
Organism and culture conditions. Streptomyces phaeochromogenes NRRL B-3559 was from Northern Regional
Research Laboratories, Peoria, Ill., USA. The mineral salt medium described by Chatterjee et al. (1983), with
glucose as the carbon source and L-isoleucine as the nitrogen source, was used. The chloramphenicol-producing
strain was grown in 2 litre Erlenmeyer flasks, containing 1 litre of medium, at 30 "C for 96 h on a rotary shaker.
Cells were harvested by centrifugation, yielding about 12 g (wet wt) 1-'.
Enzyme assays. Halogenating activity and peroxidase activity were measured essentially as described for the
bromoperoxidase from Pseudomonas aureofaciens (van Pke & Lingens, 1985).
Purijication of's. phueochromogenes bromoperoxidase. The crude extract was prepared by suspending one part of
cells (wet wt) in two parts of 50 mM-sodium acetate buffer, pH 5.5, and disrupting with a Branson sonifier 5-17 A
for six 30 s periods. The cell debris was removed by centrifugation for 30 min at 22 lOOg and 4 "C. Ammonium
sulphate was added to the crude extract to 40% saturation, and after stirring for 20min the precipitate was
removed by centrifugation and discarded. Ammonium sulphate was added to the supernatant to 70% saturation,
and stirred for 20 min. The precipitate was removed by centrifugation, dissolved in 50 mM-sodium acetate buffer,
pH 5.5, and dialysed against 5 litres 50 mM-sodium acetate buffer, pH 5.5, for 15 h. After centrifugation the
K.-H. VAN PBE AND F. LINGENS
dialysed enzyme was adsorbed to a DEAE-cellulose DE52 column (5 x 10 cm) equilibrated with 50 mM-sodium
acetate buffer pH 5.5. The sample was washed onto the column with 800 ml of this buffer and a 500 ml gradient of
0-0.6 M-KCl in 50 mM-sodium acetate buffer, pH 5.5, was applied. Then a further 250 ml0-6 M-KC~
sodium acetate buffer, pH 5.5, were passed through the column. Fractions (3.4 ml) were assayed for protein (Azg0)
and haloperoxidase activity. Those fractions (1 15-1 38), having an activity of more than 30% of the maximal
activity, were pooled.
The resulting 81 ml protein solution were dialysed against 5 litres 50 mM-Tris/HCl, pH 8.9, and passed onto a
DEAE-cellulose DE52 column (2.7 x 5 cm) equilibrated with 50 mM-Tris/HCl pH 8.9. The sample was washed
onto the column with 610 ml of this buffer and was then eluted with a linear gradient of 500 mlO-O.5 M-KCl in
50 mM-Tris/HCl, pH 8-9. Those fractions (57-69) having an activity of more than 20% of the maximal activity
were pooled and dialysed against 5 litres 5 mM-potassium phosphate buffer, pH 7.0, for 15 h. The dialysed protein
solution was adsorbed to a Bio-Gel HTP hydroxylapatite column (1.5 x 8 cm) equilibrated with 5 mM-potassium
phosphate buffer, pH 7.0. The sample was washed onto the column with 120 ml of this buffer and eluted with
200 ml of a linear gradient (5-50 mM) of potassium phosphate buffer pH 7.0. Those fractions (47-70) having more
than 25% of the maximal haloperoxidase activity were pooled and concentrated to 2.3 ml using an Amicon
concentrator with a PM-30 membrane. After dialysing against 1 litre sample buffer the enzyme was further
purified by preparative polyacrylamide gel electrophoresis on 7.5 % (w/v) gels under non-dissociating conditions at
pH 7.5 (Maurer, 1968).
The brownish band was cut out, immersed in 5 vols 10 mM-potassium phosphate buffer, pH 7.0, homogenized
and stirred for 2 h at 4 "C. After centrifugation the extraction was repeated and the acrylamide fragments were
removed by centrifugation. The supernatants were concentrated to 2 ml using an Amicon concentrator with a PM-
30 membrane and dialysed against 5 litres 10 mM-potassium phosphate buffer, pH 7-0, for 15 h.
Analysis of the purified bromoperoxidase by PAGE. Analytical PAGE under non-denaturing conditions at pH 8.9
was done in a horizontal 7.5% (w/v) polyacrylamide gel (Fehrnstrom & Moberg, 1977) with an LKB 21 17
Multiphor apparatus. SDS-PAGE was done by the method of Laemmli (1970).
Analytical isoelectric focusing in thin-layer plates of 5 % (w/v) polyacrylamide containing a 2.4% (w/v) solution
of ampholines in the pH range 3.5-9.5 was done using an LKB Multiphor system according to the manufacturer's
instructions (Winter et al., 1977). Gels were stained for protein with Coomassie Blue R250 and for peroxidase
activity with o-dianisidine. The PI of bromoperoxidase and the molecular weight of the subunits were determined
by comparing the migration of the protein with those of standard proteins.
Protein determination. Protein concentrations were measured by the method of Lowry using bovine serum
albumin as a standard.
Molecular weight determination. The molecular weight of bromoperoxidase was estimated by molecular sieve
chromatography with a 94 x 3 cm column of Sephadex G-200, standardized with ferritin (mol. wt 440000),
gamma globulin (1 50000), bovine serum albumin (68 OOO), and cytochrome c (1 3000). The column was
equilibrated with 50 mM-potassium phosphate buffer, pH 7.0.
Spectroscopy. Absorption spectra were recorded on a Uvicon 81 0 spectrophotometer (Kontron). Reduced
pyridine haemochromogen was prepared by the method described by Falk (1964).
in 50 mM-
Previous work has shown that bacterial bromoperoxidases can not be detected in crude
extracts or after ammonium sulphate fractionation (van PCe & Lingens, 1984, 1985). They only
can be detected after removal of interfering enzymes. In the case of bromoperoxidase from
Sit-eptomyces phaeochromogenes this was achieved by the use of a DEAE-cellulose DE52 ion-
exchange column with sodium acetate buffer, pH 5.5. The results of the purification procedure
are summarized in Table 1. The purified enzyme gave a single band on PAGE under non-
denaturing (RF 0.42) and denaturing conditions (Fig. l), when the gels were loaded with 20 pg
protein. The comparison with standard proteins yielded a molecular weight of 72500 for the
subunits (Fig. 1 b). In isoelectric focusing experiments the enzyme exhibited a single band when
20 pg protein was applied to the gel. The isoelectric point was estimated to be 4.0.
The visible absorption spectra of the native and reduced bromoperoxidase are given in Fig. 2.
The native bromoperoxidase spectrum had maxima at 404, 496 and 634 nm with millimolar
extinction coefficients ( E , ~ ) of 132.4, 18.3 and 11.0, respectively. The dithionite-reduced
bromoperoxidase had a Soret band at 438 nm with a E,~ of 85.2. The visible absorption
Bromoperoxidase from S. phaeochromogenes
Fig. 1. SDS-PAGE of bromoperoxidase from Streptomycesphaeochromogenes under non-denaturing (a)
and denaturing (b) conditions. Each gel contained 20 pg protein and was stained with Coomassie Blue
R250. The standard proteins in the left-hand lane of (b) were (molecular weights in parentheses):
phosphorylase B (92 500), bovine serum albumin (66200), ovalbumin (45 000), carbonic anhydrase
(31 000), soybean trypsin inhibitor (21 500) and lysozyme (14400).
Table 1. Summary of the pur$cation of bromoperoxidase from Streptomyces phaeochromogenes
The enzyme was purified from 275 g bacteria; -, could not be determined.
4&70 % (N HJ2 SO4
Bio-Gel HTP hydroxyl-
- - -
spectrum of the ferrous enzyme had a peak at 557 nm with a shoulder at 592 nm. The E , , , ~ values
were 11.5 and 6.1, respectively. When reduced in alkaline pyridine, bromoperoxidase yielded a
haemochromogen whose spectrum was identical with that of protohaem IX (Fig. 3). On the
basis of the absorbance at 557 nm ( E , ~ 34.4; Falk, 1964), haem content of 0-91 molecules per
dimeric molecule of bromoperoxidase was calculated. The purified enzyme had an A,,4/A274
ratio of 0.4 which is consistent with the low haem content.
Reactions catalysed by bromoperoxidase
Bromoperoxidase in the presence of hydrogen peroxide and bromide catalysed the
bromination of monochlorodimedone, but not the chlorination or fluorination of this organic
substrate. When hydrogen peroxide was substituted by t-butyl hydroperoxide, no brominating
K.-H. VAN PkE AND F. LINGENS
Fig. 2. Absorption spectra of native and reduced bromoperoxidase. The spectra of native ( - )
reduced (----) bromoperoxidase (1.3 PM) were recorded in 25 mhl-potassium phosphate buffer, pH 7.0.
Fig. 3. Pyridine haemochromogen spectrum. Bromoperoxidase was present at 1.2 PM final
concentration. The method of haemochromogen preparation is described in Methods.
activity could be detected. The enzyme activity, however, was not inhibited by t-butyl
hydroperoxide. Bromoperoxidase in the absence and presence of bromide was able to catalyse
the oxidation of o-dianisidine, pyrogallol and o-phenylendiamine. Catalase activity could also be
measured under the same conditions as peroxidase and haloperoxidase activity. With t-butyl
hydroperoxide instead of hydrogen peroxide neither peroxidase nor catalase activity could be
pH optima o f the dgerent catalytic activities o f bromoperoxidase
The pH optima of the three catalytic activities of this bromoperoxidase were slightly different
(Fig. 4). The optima for the brominating and peroxidase activities were at pH 5.0, whereas the
optimum for the catalase activity was at pH 7.0.
When crude extracts from Streptomyces phaeochromogenes were incubated in the presence of
monochlorodimedone, hydrogen peroxide and bromide, no enzyme reaction could be detected.
However, when the supernatant solution was incubated in the presence of o-dianisidine and
hydrogen peroxide, oxidation of the substrate was evident. The same phenomenon was observed
with extracts from the bacterium Pseudomonas aureofaciens (van Pee & Lingens, 1985) and from
the marine alga Penicillus lamourouxii (Baden & Corbett, 1980). In the latter case, however,
bromoperoxidase activity could already be detected after precipitation and concentration by
ammonium sulphate fractionation. With extracts from S. phaeochromogenes ammonium
sulphate fractionation and ion-exchange chromatography had to be done before brominating
activity was detectable. In the case of bromoperoxidase from Ps. aureofaciens even a second ion-
exchange chromatography step was necessary before brominating activity could be measured.
This demonstrates one of the problems with the isolation of bacterial haloperoxidases, which is
probably due to the fact that crude extracts contain a very efficient catalase. This catalase
Bromoperoxidase jrom S. phaeochromogenes
.- .s 80
Fig. 4. Brominating, peroxidase and catalase activities of bromoperoxidase as a function of pH. The
brominating activity was measured by the monochlorodimedone assay (0).
determined by the 0-dianisidine assay (0). Catalase activity was measured by following the
decomposition of hydrogen peroxide at 240 nm (0).
phosphate buffer. 100% activity corresponded to the maximal activity.
Peroxidase activity was
Buffering was achieved with 0.1 M-potassium
Table 2. Comparison of bromoperoxidase from Streptomyces phaeochromogenes with
bromoperoxidase from Pseudomonas aureo faciens, with an algal bromoperoxidase
from Penicillus capitatus, and with chloroperoxidase from Caldariomyces fumago
10-3 x M ~ I .
Spectrum peak wave-
Oxidation of o-dianisidine
pH optimum (bromination)
* Data from Manthey & Hager (1981).
f Data from Baden & Corbett (1980).
$ Shoulder of peak.
competes with bromoperoxidase for hydrogen peroxide. Therefore bromoperoxidase activity
can only be detected after removal of the catalase. The molecular weight of the bromoperoxidase
from S. phaeochromogenes is very similar to that of the other bacterial bromoperoxidases already
described (Table 2), but quite different from those of the eukaryotic bromoperoxidases from
marine algae (Baden & Corbett, 1980; Manthey & Hager, 1981) and that of the chloroperoxidase
from the fungus Caldariomyces fumago (Morris & Hager, 1966). Both bacterial bromo-
peroxidases form a dimer, consisting of two identical subunits, a characteristic they share with
only one of the described algal bromoperoxidases. The spectral properties of the bacterial
bromoperoxidases parallel those of the algal bromoperoxidases and chloroperoxidase. The
prosthetic group of all these enzymes is protoporphyrin IX. Like the eukaryotic haloperoxidases
the two bacterial bromoperoxidases also have catalase and peroxidase activity, but their
brominating activity is rather low and the two other activities are very high. Like the algal