Arch Microbiol (2001) 175 : 405–412 DOI 10.1007/s002030100279
O R I G I N A L PA P E R
Anke Apitz · Karl-Heinz van Pée
Isolation and characterization of a thermostable intracellular enzyme with peroxidase activity from Bacillus sphaericus
Received: 7 November 2000 / Revised: 14 March 2001 / Accepted: 23 March 2001 / Published online: 11 May 2001 © Springer-Verlag 2001
Abstract During a screening for bacteria producing enzymes with peroxidase activity, a Bacillus sphaericus strain was isolated. This strain was found to contain an intracellular enzyme with peroxidase activity. The native enzyme had a molecular mass of above 300 kDa and precipitated at a salt concentration higher than 0.1 M. Proteolytic digestion with trypsin reduced the molecular mass of the active enzyme to 13 kDa (dimer) or 26 kDa (tetramer) and increased its solubility, allowing purification to homogeneity. Spectroscopic investigations showed the enzyme to be a hemoenzyme containing heme c as the covalently bound prosthetic group. The enzyme was stable up to 90 °C and at alkaline conditions up to pH 11, with a pH optimum at pH 8.5. It could be visualized by activity staining after SDS-PAGE and showed activity with a number of typical substrates for peroxidases, e.g., 2,2′-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) diammonium salt, guaiacol and 2,4-dichlorophenol; however the enzyme had no catalase and cytochrome c peroxidase activity. Keywords Bacillus sphaericus · Peroxidase · Cytochrome c · Cytochrome c peroxidase · Heme c Abbreviations ABTS 2,2′-Azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) diammonium salt
Introduction Peroxidases are enzymes that are abundant in aerobic organisms. Together with catalases, one of their physiological roles is to scavenge H2O2 and to prevent oxidative damage. In contrast to catalases, peroxidases are also involved in a number of other physiological processes, such as growth and senescence, fruit development, and germinaA. Apitz · K.-H. van Pée (✉) Institut für Biochemie, Technische Universität Dresden, 01062 Dresden, Germany e-mail:
[email protected], Fax: +49-351-4635506
tion (Campa 1991). Unspecific peroxidases oxidize a large number of low-molecular weight aromatic compounds, such as phenols, hydroquinones, benzidine derivatives, flavonoids, and macromolecules, including lignin or humic substances. Thus, they are interesting enzymes for application in the pulp and paper, textile, and laundry industries, for the decomposition of pollutants, or for use as biosensors. Whereas catalases isolated from different organisms have similar structures, peroxidases can be divided into several structural classes. Welinder (1992) defined three structural classes of plant peroxidases, with class I consisting of mitochondrial yeast cytochrome c peroxidase, ascorbate peroxidases, and bacterial catalase-peroxidases, all of which are intracellular enzymes. Class II contains secretory fungal peroxidases, and class III classical secretory plant peroxidases. The enzymes belonging to class I are from a large variety of organisms and they have also different functions. Whereas much is known about cytochrome c peroxidase, which is a model enzyme for peroxidase reactions and structure (Erman and Vitello 1998), and about ascorbate peroxidases, which belong to seven different families (Jespersen et al. 1997), much less is known about intracellular bacterial catalase-peroxidases and their properties (Dunford 1999). The first catalase-peroxidase was isolated from Escherichia coli by Claiborne and Fridovich in 1979 (Claiborne and Fridovich 1979). Similar enzymes were later detected in many other bacteria, such as Halobacterium halobium (Brown-Peterson and Salin 1993), Bacillus stearothermophilus (Loprasert et al. 1990), and Streptomyces reticuli (Zou et al. 1999). These enzymes have a low heme content and both catalase and peroxidase activity. Nonetheless, some 22 years after the first description of bacterial catalase-peroxidases, very little is known about their physiological functions and only a few catalase-peroxidase genes have been cloned and sequenced (Dunford 1999; Loprasert et al. 1990; Zou et al. 1999). Furthermore, no three-dimensional structure has been elucidated so far. Some intracellular bacterial enzymes with peroxidase activity, but without catalase activity, have also been isolated: cytochrome c peroxidase from Pseudomonas fluo-
406
rescens (Ellfolk and Soininen 1970), non-heme hydroquinone peroxidase from Azotobacter beijerinckii (Nakamiya et al. 1997), and cytochromes c are known to have peroxidase activity (Vazquez-Duhalt 1999). The fact that much less is known about intracellular bacterial enzymes with peroxidase activity than about extracellular peroxidases is probably due to the lack of direct screening methods for intracellular peroxidases, which makes these enzymes more difficult to detect. To obtain more information about intracellular bacterial enzymes with peroxidase activity, soil samples were screened for bacteria producing intracellular enzymes with peroxidase activity. During this screening, a Bacillus sphaericus strain was isolated that produces an enzyme with peroxidase activity with unusual properties. Here we describe the purification and characterization of this enzyme.
Materials and methods
Activities with 2,2′-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) diammonium salt (ABTS) and guaiacol as substrates were determined according to Bergmeyer (1974) in 0.1 M sodium phosphate buffer (pH 6.8), 1.7 mM ABTS, or 2.3 mM guaiacol, respectively, and 2.64 mM H2O2 following the increased absorbance at 405 nm with an extinction coefficient of 18.6 mM–1 cm–1 (ABTS assay) or at 436 nm using an extinction coefficient of 25.5 mM–1 cm–1 (guaiacol assay). One unit of peroxidase activity was defined as the formation of 1 µmol product per min. Catalase activity was assayed as described by Clairborne and Fridovich (1979) in 0.1 M potassium phosphate buffer (pH 7.0) by measuring the decomposition of H2O2 at 240 nm using an extinction coefficient of 43.6 M–1 cm–1 and an initial H2O2 concentration of 7.2 mM. Commercially available cytochrome c551 from Pseudomonas aeruginosa, horseradish peroxidase, and catalase from bovine liver were used as controls. Cytochrome c peroxidase activity was determined as described by Ellfolk and Soininen (1970) in 0.02 M sodium phosphate buffer (pH 7.2), 35 µM H2O2, and 11 µM cytochrome c from bovine liver. The absorbance was recorded at 550 nm. Protein concentrations were determined using the dye-binding method (Bradford 1976) with cytochrome c from bovine liver as the standard.
Isolation of bacteria, growth conditions, and enzyme production Bacteria producing enzymes with peroxidase activity were isolated from a soil sample collected near a hydrogen peroxide storage tank. These soil samples were used to inoculate 20 ml of Luria Bertani (LB) medium. After incubation at 30 °C on a rotary shaker for 3 days, 1 l of LB medium was inoculated with 1 ml of the mixed culture and incubated for 3 days at 30 °C. Cells were harvested by centrifugation (11,340×g), washed with saline, and disrupted using a mixer mill. Prior to harvesting the cells, samples were removed, diluted, plated out on LB agar plates, and incubated at 30 °C. The various single colonies obtained from 1 l of the mixed cultures that showed peroxidase activity were used to inoculate fresh medium to identify which bacterial colony was responsible for the production of peroxidase activity. The isolated bacterial strain was then sent to DSZM (Braunschweig, Germany) for identification. For purification of the enzyme, bacteria were grown in a 10-l fermenter containing 10 l of LB medium with stirring (150 rpm) and aeration (7 l min–1) for 2 days. Silicone anti-foam emulsion M-30 (Sigma, Deisenhofen, Germany) was added with the help of a pump when required. Production of the enzyme during fermentation was monitored by taking samples every 6 h and measuring the peroxidase activity in heat-treated crude extracts prepared from these samples by cell disruption with a mixer mill. Additionally, the growth rate was monitored by measuring the optical density at 600 nm, oxygen saturation, and pH. To investigate whether oxygen stimulated enzyme production, the bacteria were grown in a 10-l culture for 4 h under the conditions described above, resulting in an oxygen level of 1%. The culture was then flushed with nitrogen for 10 min and incubated for further 43 h without added oxygen. The cells from a 4-l sample were harvested and peroxidase activity was measured after cell disruption and heat treatment. The remaining culture was flushed with oxygen (20 l min–1 for 30 min) and incubated for further 24 h. After harvesting the bacteria, peroxidase activity was measured as described above. Enzyme assays and protein determination Peroxidase activity was measured according to Ishida et al. (1987) using 5 mM 2,4-dichlorophenol as the substrate in the presence of 1 mM 4-aminoantipyrine and 2.64 mM H2O2 and a suitable amount of enzyme in 0.1 M Tris-HCl buffer (pH 8.5) at 30 °C. The activity was calculated from the increased absorbance at 510 nm using an extinction coefficient of 13.6 mM–1 cm–1 (Loprasert et al. 1990).
Enzyme purification For the preparation of cell-free extracts, 50 g of cells (wet weight) were suspended in 150 ml 10 mM sodium carbonate-HCl buffer (pH 10.2) and disrupted by ultrasonication at 4 °C for 15 min with 5-s intervals. Cell debris was removed by centrifugation at 38,700×g and 4 °C. The resulting crude extract was dialyzed against 5 l of 10 mM sodium phosphate buffer (pH 6.8) for 18 h. To remove catalase activity present in the crude extract which strongly interfered with determination of peroxidase activity, the protein solution was incubated at 90 °C for 10 min. After cooling, 1 ml of a freshly prepared trypsin solution (5 mg ml–1 in 1 mM HCl) was added to every 10 ml of protein solution. After incubation for 2.5 h at 37 °C, the solution was again treated at 90 °C for 10 min to inactivate the trypsin, and denatured proteins were removed by centrifugation for 30 min at 38,700×g at 4 °C. The protein solution was then loaded onto a Q Sepharose fast-flow column (2.7×20 cm) equilibrated with 10 mM sodium phosphate buffer (pH 6.8) and washed with 900 ml of the same buffer. Protein was eluted with a 600-ml linear gradient from 0–1 M NaCl in 10 mM sodium phosphate buffer (pH 6.8); 6-ml fractions were collected and assayed for protein (A280), heme (A406), and peroxidase activity. Active fractions having an absorption ratio A406:A280 of more than 50% of the highest ratio were pooled and ammonium sulfate was added to a concentration of 1 M. The solution was applied onto a Phenyl-Sepharose CL-B4 column (2.7×12 cm) equilibrated with 1 M ammonium sulfate in 10 mM sodium phosphate buffer (pH 6.8). Elution was achieved with 360 ml of a linearly descending gradient of 1–0 M ammonium sulfate in buffer. Six-ml fractions were collected and assayed for protein, heme, and peroxidase activity. Those active fractions with the highest absorption ratio A406:A280 were pooled and concentrated at 4 °C to one fourth of the original volume using a Pall Filtron (Northborough, Mass.) concentrator with a 3-kDa membrane. The concentrated solution was split into ten samples that were individually passed through a HiLoad 16/60 Superdex 75 column (Pharmacia, Freiburg, Germany). The collected fractions (2 ml) were assayed as above. Two separate pools were formed from the active fractions (24–27 and 28–31, according to the absorption at 406 nm) obtained from the ten runs and loaded onto a Mono Q HR 5/5 column (Pharmacia, Freiburg, Germany) equilibrated with 10 mM sodium phosphate buffer (pH 6.8). After washing with 10 ml of buffer, proteins were eluted with 70 ml of a linear gradient of 0–0.5 M NaCl in buffer. The collected 1-ml fractions were assayed as described above and active fractions were pooled.
407 To remove salt, the obtained solution was concentrated and diluted three times by ultrafiltration. The desalted samples were further concentrated to one tenth of the original volume by freezedrying. After addition of sucrose, the samples were subjected to preparative polyacrylamide gel electrophoresis under non-denaturing conditions. The enzyme could be detected as a brownish band. After electrophoresis, this band was excised from the gel and eluted with an HSB-Elutor E51 (Biometra, Göttingen, Germany) filled with a low-salt solution (50 mM Tris, 50 mM glycine, pH 6.8) and a high-salt solution (3 M NaCl in low-salt solution). Elution was carried out twice for 30 min at 100 V. Eluted protein was desalted as described above and stored at –20 °C. Polyacrylamide gel electrophoresis Polyacrylamide gel electrophoresis under non-denaturing conditions at pH 7.5 was performed in 10% (w/v) polyacrylamide gels according to Maurer (1968). Peroxidase activity was detected in gels using the 2,4-dichlorophenol/4-aminoantipyrine assay. Protein was stained with Coomassie brilliant blue G-250. SDS-PAGE was performed in 16% (w/v) gels according to Schägger and von Jagow (1987). The protein standards (BioRad, Munich, Germany) used were triosephosphate isomerase (26.625 kDa), myoglobin (16.950 kDa), α-lactalbumin (14.437 kDa), aprotinin (6.512 kDa), insulin b-chain, oxidized (3.496 kDa), and bacitracin (1.423 kDa). Peroxidase activity was visualized using the 2,4-dichlorophenol/4-aminoantipyrine assay. Isoelectric focusing in 7.5% acrylamide thin-layer gels containing ampholines in the pH range of 3.5–9.5 or 3.5–5.0 was carried out using an LKB Multiphor system (Pharmacia, Freiburg, Germany) according to the manufacturer’s instructions. The protein standards (Serva, Heidelberg, Germany) contained amyloglucosidase (pH 3.5), glucose oxidase (pH 4.2), trypsin inhibitor (pH 4.5), β-lactoglobulin (pH 5.2 and 5.3), carbonic anhydrase (pH 6.0), myoglobin (pH 6.9 and 7.4), lectin (pH 7.8, 8.0, and 8.3), ribonuclease A (pH 9.5), and cytochrome c (pH 10.7). Molecular mass determination and partial amino acid sequence The molecular mass of the enzyme was estimated by molecular sieve chromatography using a HiLoad 16/60 Superdex 75 column (Pharmacia, Freiburg, Germany). The column was equilibrated with 10 mM sodium phosphate buffer (pH 6.8) and standardized with ovalbumin (44 kDa), myoglobin (17 kDa), and vitamin B12 (1.35 kDa) from BioRad (Munich, Germany). The amino-terminal amino acid sequence was determined after blotting the enzyme onto a polyvinylidine difluoride (PVDF) membrane using a TransBlot Cell (BioRad, Munich, Germany). The membrane was stained with Coomassie brilliant blue G-250 and the NH2-terminal amino acid sequence was determined by automated Edman degradation in a 471A gas-phase protein sequencer with a 140A solvent delivery system (Applied Biosystems, Foster City, Calif.).
The temperature stability was determined by incubating the purified enzyme in 10 mM sodium phosphate buffer (pH 6.8) at different temperatures for 10 min. After cooling, the remaining activity was measured using the 2,4-dichlorophenol/4-aminoantipyrine assay. For determination of the temperature stability at 90 °C, the enzyme solution was incubated at 90 °C for 10, 20, 30, and 60 min, and the remaining activity was measured as above. Chemicals H2O2 (30% solution), catalase from bovine liver, and horseradish peroxidase were obtained from Merck (Darmstadt, Germany). Cytochrome c from bovine liver was from Fluka (Neu Ulm, Germany). Trypsin from bovine pancreas and cytochrome c551 from Pseudomonas aeruginosa were from Sigma (Deisenhofen, Germany). Q Sepharose fast-flow and Phenyl Sepharose CL-B4 were purchased from Pharmacia (Freiburg, Germany). All other reagents were of analytical grade.
Results Isolation and identification of bacteria with peroxidase activity and enzyme production Several bacteria producing intracellular enzymes with peroxidase activity were isolated from the soil sample. In the crude extract of one of these bacteria, much higher peroxidase activity was detected after heat treatment than in the extracts of the other isolated strains. This bacterium was identified as a new Bacillus sphaericus strain by physiological assays, partial 16S rRNA sequencing, and fatty acid analysis. The stationary-growth phase was reached after 36 h. During the exponential-growth phase, oxygen was depleted and during a period of 18 h the culture remained under oxygen-limited conditions for several hours. Optimum enzyme production was reached when the oxygen concentration had again increased to 60%, which was the case after 42 h (Fig. 1). Heat-treated crude extracts prepared from bacteria cultivated under oxygen-limited
Prosthetic group and metal content Absorption spectra were recorded on an UV4 spectrophotometer (Unicam, Kassel, Germany). Reduced pyridine hemochromes were prepared by the method of Falk (1964). Metal determinations were carried out with an energy-dispersive X-ray fluorescence apparatus, system 77 (Finnigan International, San Jose, Calif.). Prior to determination, the purified enzyme was dialyzed against 10 mM sodium phosphate buffer (pH 6.8) for 16 h which was used as a blank. pH optimum and temperature stability The pH optimum was determined in Britton-Robinson universal buffer between pH 5 and 11 using purified enzyme.
Fig. 1 Enzyme production during bacterial growth in a 10-l fermenter. The graph shows growth of bacteria as optical density at 600 nm (∇), specific peroxidase activity (▲), and oxygen concentration (❍) during fermentation of the Bacillus sphaericus strain
408 Table 1 Purification of the enzyme with peroxidase activity from B. sphaericus using 750 g of cells (wet weight). n.d. Not determined (the activity of the crude extract could not be determined reproducibly)
Purification step
Total protein (mg)
Total activity (U)
Specific activity (U mg–1)
Purification (fold)
Recovery (%)
Crude extract Heat treatment Trypsin digestion Q Sepharose Phenyl Sepharose Superdex 75 Mono Q Preparative PAGE
50,000 50,000 10,480 850 134 30 12 2.5
n.d. 4,700 29,400 14,500 6,323 2,350 1,046 526
n.d. 0.1 2.8 17.1 47.2 78.3 87.2 210.4
n.d. 1 28 171 472 783 827 2,104
n.d. 100 626 309 135 50 22 11
conditions had a specific activity of only about 14% of that of an extract obtained from bacteria grown as described above. Enzyme purification Purification of the enzyme to homogeneity was achieved by a seven-step procedure. In the crude extract, hardly any peroxidase activity was found. Furthermore, catalase strongly interfered with the peroxidase activity, which did not allow reproducible determination of peroxidase activity in the crude extract. After the first heat-treatment step, reliable activity data were obtained. The enzyme was purified about 2,100-fold with a yield of 11%. From 750 g of cells (wet weight), 2.5 mg of homogeneous enzyme were obtained (Table 1). The native enzyme precipitated irreversibly when salt was present in concentrations above 0.1 M. However, after proteolytic digestion with trypsin, the resulting protein containing the peroxidase activity was soluble in the presence of high salt concentrations. In addition to the change in solubility, the molecular mass decreased from above 300 kDa to about 13 kDa (Fig. 2) and the activity increased six-fold. Since the trypsin digestion resulted in a number of proteins with peroxidase activity, the one with the highest activity and the highest ratio of A406:A280 was further purified. During chromatography on
Fig. 3 SDS-PAGE of the purified enzyme from B. sphaericus. Lane 1 Activity staining with 2,4-dichlorophenol/4-aminoantipyrine, lane 2 Coomassie brilliant blue staining, M protein standards
Sephadex 75, the dimeric form of the enzyme was separated from the tetrameric form which was further purified. The tetrameric form of the purified enzyme gave a single band on polyacrylamide gels under denaturing (Fig. 3) and non-denaturing conditions (not shown). The enzyme could be detected on gels run under non-denaturing and also under denaturing conditions by its brownish color, or could be made visible by activity staining using the 2,4dichlorophenol/4-aminoantipyrine assay (Fig. 3). The isoelectric point was estimated to be at pH 3.5. Spectral properties and partial amino acid sequence
Fig. 2 Gel filtration of trypsin-digested protein solution from Bacillus sphaericus. Extracts containing peroxidase activity were analyzed on Bio-Silect SEC-250 with 10 mM sodium phosphate buffer, pH 6.8, before (bold line) and after (thin line) trypsin treatment
The visible spectrum of the native enzyme showed a Soret band at 406 nm and a weak broad band at 356. Reduction of the enzyme with dithionite resulted in a shift of the Soret band to 416 nm and in the appearance of bands at 521 and 550 nm (Fig. 4). When reduced in alkaline pyridine, a spectrum with bands at 414, 520, and 550 nm was obtained (not shown). This spectrum is very similar to that of heme c.
409
Fig. 4 UV/VIS absorption spectra of the enzyme with peroxidase activity from B. sphaericus. UV/VIS spectra of oxidized (dotted line) and dithionite-reduced (solid line) enzyme at a concentration of 70 µg ml–1 were recorded
The enzyme showed hardly any absorbance at 280 nm, suggesting that it contains only very few aromatic amino acids. This is consistent with the lack of aromatic amino acids in the first 32 amino acids of the amino-terminal end (NH2-EIAAEHEGGGETTTE(E)TDAEADGSALVQ(S)(C)IG). Comparison of this amino acid sequence with sequences available in standard databases revealed no similarity to peroxidases, catalase-peroxidases, cytochrome c peroxidases, cytochrome c oxidases, and cytochromes c, including microperoxidases, with the exception of the cytochrome c553 precursor from Desulfovibrio vulgaris strain Hildenborough. However, the similarity and the identity with the cytochrome c553 precursor were only 34 and 28%, respectively. Physical properties The molecular mass of an enzyme subunit was estimated from SDS gel electrophoresis to be 6.5 kDa (Fig. 3). From molecular sieve chromatography on HiLoad 16/60 Superdex 75, two enzymes with peroxidase activity with molecular masses of 13 kDa and 26 kDa were obtained, representing the dimeric and tetrameric forms of the enzyme with the tetrameric form as the more active one. From the absorbance of the 550 nm band, a heme content of two molecules of heme per molecule of enzyme subunit was calculated. This is in accordance with the content of 2.2 atoms of iron per molecule of enzyme subunit. In addition to iron, the enzyme also contains 0.8 atoms of copper and 0.5 atoms of zinc per molecule of enzyme subunit.
Fig. 5 Influence of the H2O2 concentration on the peroxidase activity of the purified enzyme from B. sphaericus; (--), 4.4 mM, (-·-) 8.8 mM, (_ _ _) 26.4 mM, (___) 44 mM, (···) 88 mM
Initial peroxidase activity increased with increasing H2O2 concentrations; however, with higher concentrations of H2O2, the reaction stopped after a short incubation time (Fig. 5). Peroxidase activity of cytochrome c551 from Pseudomonas aeruginosa also increased at H2O2 concentrations between 2.64 and 26.4 mM, whereas peroxidase activity of horseradish peroxidase did not increase under these conditions. Catalase activity was not detectable and the enzyme did not show any cytochrome c peroxidase activity with cytochrome c from bovine liver as the substrate. Typical inhibitors of hemoproteins such as cyanide and azide inhibited peroxidase activity. However, rather high inhibitor concentrations were required and the enzyme still showed 80% of its activity at inhibitor concentrations of 30 mM. Tween 20 at 1 and 2% inhibited the activity by 20 and 50%, respectively. Methanol and ethanol at 20% inhibited the reaction by 50 and 65%, respectively. Temperature stability and pH optimum The enzyme was very stable at high temperatures. Incubation in a boiling water bath for 10 min decreased the ac-
Substrate specificity and inhibition of the reaction The isolated enzyme had very low substrate specificity and accepted a variety of different hydrogen donors. The reactions with a few typical substrates for peroxidases, such as 2,4-dichlorophenol, ABTS, and guaiacol, was studied in some detail. The Km values for H2O2 and these organic substrates were determined to be 5.26×103 µM, 224 µM, 1.64 µM, and 1.83 µM, respectively.
Fig. 6 pH optimum of the enzyme with peroxidase activity from B. sphaericus. The pH optimum was determined in Britton-Robinson universal buffer
410
tivity by only 18%, while after incubation of the enzyme at 90 °C for 1 h 95% of the activity still remained. The enzyme was active over a wide pH range, from pH 5.0–11.0 with a pH optimum at pH 8.5 (Fig. 6).
Discussion The screening procedure used led to the isolation of a number of bacteria that produced enzymes with peroxidase activity which showed activity at alkaline pH values and which were resistant to higher temperatures. Often, peroxidase activity could only be detected after heat treatment, which reduced the interference of catalase with the assay for peroxidase activity. One of the enzymes showed much higher peroxidase activity and was more heat-resistant than the enzymes found in other bacteria. This enzyme was therefore purified and further characterized. For the purification of the enzyme, a procedure had to be developed taking the unusual properties of the enzyme into account. Since the enzyme already precipitated in the presence of 0.1 M salt, all chromatographic methods employing higher salt concentrations, such as ion-exchange chromatography or hydrophobic-interaction chromatography, could not be used. On the other hand, the stability of the enzyme against high temperature seemed to be a property that could be exploited for its purification. However, incubation of the crude extract at 90 °C did not result in precipitation of protein and had thus no purifying effect. But, since heat treatment resulted in the inactivation of catalase, which interfered with the determination of peroxidase activity, it was nonetheless very useful as the first step in the purification procedure. The problem of precipitation in the presence of salt could be overcome by protease digestion with trypsin. Incubation with trypsin resulted not only in digestion of other proteins, but also in digestion of the enzyme, reducing its size from 300 kDa to 13 kDa. Additionally, the part of the enzyme responsible for precipitation in the presence of salt was removed and peroxidase activity increased six-fold. Since after this treatment the enzyme no longer precipitated at higher salt concentrations, ion-exchange and hydrophobic-interaction chromatography could now be used. The trypsin digestion did not result in a single protein with peroxidase activity, but, in addition to the 13-kDa enzyme, gave rise to a population of differently sized enzymes with peroxidase activity. Thus, not only activity measurements, but also the ratio between the absorbance of the heme group at 406 nm and the protein absorbance at 280 nm was used for fractionation. The obtained homogeneous enzyme has a very high A406:A280 ratio of 27, which is due to the absence of aromatic amino acids (Fig. 4). Since the purified enzyme is only a fragment of less than a twentieth of the original enzyme, it is not known whether this is also true for the native enzyme as it exists in bacterial cells. The small size of the active fragment makes this enzyme an interesting object to study the structural features and components that are actually required for peroxidase activity. It has been shown that protoporphyrin IX in other
hemoproteins, e.g., hemoglobin and myoglobin, is responsible for the peroxidase activity shown by these enzymes; however, these activities are much lower than the activities of peroxidases (Grisham and Everse 1991). The spectral properties of the enzyme show that the chromophore is a heme c and not a protoporphyrin IX as known from bacterial catalase-peroxidases and plant as well as fungal peroxidases (Dunford 1999; Brown-Peterson and Salin 1993). This is especially evident from the reduced pyridine hemochrome with its absorption bands at 520 and 550 nm (Falk 1964). Heme c can be found in cytochromes c (Pettigrew and Moore 1987), cytochrome c peroxidase (Zahn et al. 1997), and in a subunit of cytochrome caa3 terminal oxidase (Tanaka et al. 1996). Miki and Orii (1986) reported cytochrome c peroxidase activity of cytochrome c oxidase, and cytochromes c are also known to have peroxidase activity (Vazquez-Duhalt 1999). The finding that the isolated enzyme shows the same dependence of peroxidase activity on H2O2 concentration as cytochrome c551 from Pseudomonas aeruginosa, and that the peroxidase activity of this enzyme also increases six-fold by heat treatment at 90 °C for 10 min indicates that the isolated enzyme could actually be a cytochrome c. The fact that the native enzyme precipitates already at moderate salt concentrations suggests that the enzyme could be membrane-associated. Sone and Toh (1994) suggested that several small soluble cytochromes c from gram-positive bacteria are derived from membrane-bound forms by proteolysis, since the soluble fractions of the Bacillus strains used did not contain any cytochromes. Membrane-bound cytochromes c could be made soluble by a controlled digestion with trypsin (Trumpower and Katki 1975). This method was used by von Wachenfeldt and Hederstedt (1993) to isolate the heme domain of Bacillus subtilis cytochrome c550 in water-soluble form. The tryptic fragment comprised 74 C-terminal amino acid residues of cytochrome c550, and the spectroscopic and electrochemical properties of this heme domain were found to be very similar to that of the complete cytochrome c, but did not show any significant sequence similarity to the enzyme described here. There is a sequence similarity to the cytochrome c553 precursor from Desulfovibrio vulgaris (Blackledge et al. 1995), with amino acids at the end of the amino-terminal sequence of the isolated enzyme that could be part of the heme binding domain. Several bacterial cytochrome c peroxidases have been described (Arciero and Hooper 1994; Ellfolk and Soininen 1970; Hu et al. 1998; Pettigrew 1991; Villalain et al. 1984; Zahn et al. 1997). They are di-heme enzymes with molecular masses ranging from 33.7 to 44 kDa. Some of these also show peroxidase activity with non-specific peroxidase substrates such as guaiacol (Zahn et al. 1997) or the non-physiological horse-heart cytochrome c (Arciero and Hooper 1994; Hu et al. 1998). Zahn et al. (1997) suggested that many di-heme cytochromes c might be peroxidases, but we could not detect any enzymatic activity with cytochrome c. Additionally, we found no sequence similarity to bacterial cytochrome c peroxidases. There is also
411
a difference in the substrate specificity of the isolated enzyme compared to those of bacterial catalase-peroxidases. Whereas the enzymes from Bacillus stearothermophilus, Pellicularia filamentosa, and Escherichia coli are highly specific for 2,4-dichlorophenol and show no activity with guaiacol (Poole et al. 1986), the enzyme from Bacillus sphaericus accepts guaiacol, ABTS, and many other substrates, but has no catalase activity, as also described for the peroxidase from Flavobacterium meningosepticum (Koga et al. 1999). Additional differences to peroxidases are the very high temperature stability and the alkaline pH optimum. The temperature stability of the native enzyme as well as that of the isolated fragment is very unusual for a heme peroxidase, but is very similar to that of cytochrome c551 from Pseudomonas aeruginosa. From the obtained data, it is obvious that the isolated enzyme is not a bacterial catalase-peroxidase, whereas the similar properties of the enzyme with those of cytochromes c suggest that is actually a cytochrome c with peroxidase activity, although there is no report of a copper- and zinccontaining cytochrome c. Nearly all known peroxidases are stable and active at acidic or neutral pH values. Thus, the isolated enzyme is an interesting candidate for applications in which higher pH values are required. Cherry et al. (1999) reported that the peroxidase from Coprinus cinereus had to be mutated extensively to allow this enzyme to be used as a detergent additive under laundry conditions at pH 10.5, high temperature, and high peroxide concentration. In this regard, the enzyme described here seems to be a good candidate for improvement for the use in this and other technical processes. Acknowledgements We thank C. Mühlbach for screening and isolation of the bacterial strain, Dr. V. Cercassov, Universität Hohenheim, Germany, for metal content analysis, and Dr. B. Hauer, BASF, Ludwigshafen, Germany, for the partial amino acid sequence determination. This work was supported by the Max-BuchnerStiftung and the Fonds der Chemischen Industrie.
References Arciero DM, Hooper AB (1994) A di-heme cytochrome c peroxidase from Nitromonas europaea catalytically active in both the oxidized and half-reduced states. J Biol Chem 269:11878– 11886 Bergmeyer HU (1974) Methods of enzymatic analysis. Verlag Chemie, Weinheim Blackledge MJ, Medvedeva S, Pocin M, Guerlesquin F, Bruschi M, Marion D (1995) Structure and dynamics of ferrocytochrome c553 from Desulfovibrio vulgaris studied by NMR spectroscopy and restrained molecular dynamics. J Mol Biol 345:661–681 Bradford MM (1976) A rapid and sensitive method for the quantitation of micro quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254 Brown-Peterson NJ, Salin ML (1993) Purification of a catalaseperoxidase from Halobacterium halobium: characterization of some unique properties of the halophilic enzyme. J Bacteriol 175:4197–4202
Campa A (1991) Biological role of plant peroxidases: known and potential function. In: Everse J, Everse KE, Grisham M (eds) Peroxidases in chemistry and biology, vol II. CRC, Boca Raton, pp 25–50 Cherry JR, Lamsa MH, Schneider P, Vind J, Svendsen A, Jones A, Pedersen AH (1999) Directed evolution of a fungal peroxidase. Nat Biotechnol 17:379–384 Clairborne A, Fridovich I (1979) Purification of the o-dianisidine peroxidase from Escherichia coli. J Biol Chem 254:4245–4252 Dunford HB (1999) Heme peroxidases. Wiley-VCH, New York Ellfolk N, Soininen R (1970) Pseudomonas cytochrome c peroxidase. Acta Chem Scand 24:2126–2136 Erman JE, Vitello LB (1998) Cytochrome c peroxidase: a model heme protein. J Biochem Mol Biol 31:307–327 Falk JE (1964) Porphyrins and metalloporphyrins. Elsevier, Amsterdam Grisham MB, Everse J (1991) Prooxidant activity of hemoglobin and myoglobin. In: Everse J, Everse KE, Grisham M (eds) Peroxidases in chemistry and biology, vol I. CRC, Boca Raton, pp 335–344 Hu W, De Smet L, van Driessche G, Bartsch RG, Meyer TE, Cusanovich MA, van Beeumen J (1998) Characterization of cytochrome c556 from the purple phototrophic bacterium Rhodobacter capsulatus as a cytochrome-c peroxidase. Eur J Biochem 258:29–36 Ishida A, Futamura N, Matsusaka T (1987) Detection of peroxidase activity and its localisation in the forespore envelopes of Bacillus cereus. J Gen Appl Microbiol 33:27–32 Jespersen HM, Kjaersgard IVH, Ostergaard L, Welinder KG (1997) From sequence analysis of three novel ascorbate peroxidases from Arabidopsis thaliana to structure, function and evolution of seven types of ascorbate peroxidase. Biochem J 326:305– 310 Koga S, Ogawa J, Choi Y-M, Shimizu S (1999) Novel bacterial peroxidase without catalase activity from Flavobacterium meningosepticum: purification and characterization. Biochim Biophys Acta 1435:117–126 Loprasert S, Urabe I, Okada H (1990) Overproduction and singlestep purification of Bacillus stearothermophilus peroxidase in Escherichia coli. Appl Microbiol Biotechnol 32:690–692 Maurer HR (1968) Disk-Electrophorese: Theorie und Praxis der diskontinuierlichen Polyacrylamidgelelectrophorese, Walter de Gruyter, Berlin Miki T, Orii Y (1986) Cytochrome c peroxidase activity of bovine heart cytochrome oxidase incorporated in liposomes and generation of membrane potential. J Biochem 100:735–745 Nakamiya K, Ooi T, Kinoshita S (1997) Non-heme hydroquinone peroxidase from Azotobacter beijerinckii HM121. J Ferm Bioeng 84:14–21 Pettigrew GW (1991) The cytochrome c peroxidase of Paracoccus denitrificans. Biochim Biophys Acta 1058:25–27 Pettigrew GW, Moore GR (1987) Cytochromes c. Biological aspects. Springer, Berlin, Heidelberg, New York Poole RK, Baines BS, Appleby CA (1986) Haemoprotein b-590 (Escherichia coli), a reducible catalase and peroxidase: evidence for its close relationship to hydroperoxidase I and a ‘cytochrome a1b’ preparation. J Gen Microbiol 132:1525–1539 Schägger H, von Jagow G (1987) Tricine-sodium dodecyl sulfatepolyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem 166:368– 379 Sone N, Toh H (1994) Membrane-bound Bacillus cytochromes c and their phylogenetic position among bacterial class I cytochromes c. FEMS Microbiol Lett 122:203–210 Tanaka T, Inoue M, Sakamoto J, Sone N (1996) Intra- and intercomplex cross-linking of subunits in the quinol oxidase supercomplex from thermophilic Bacillus PS3. J Biochem 119:482– 486 Trumpower BL, Katki A (1975) Controlled digestion with trypsin as a structural probe for n-terminal peptide of soluble and membranous cytochrome c1. Biochemistry 14:3635–3642
412 Vazquez-Duhalt R (1999) Cytochrome c as a biocatalyst. J. Mol. Catal. B: Enzymatic 7:241–249 Villalain J, Moura I, Liu MC, Payne WJ, LeGall J, Xavier AV, Moura JJG (1984) NMR and electron-paramagnetic-resonance studies of a dihaem cytochrome from Pseudomonas stutzeri (ATCC 11607) (cytochrome c peroxidase). Eur J Biochem 141: 305–312 Von Wachenfeldt C, Hederstedt L (1993) Physico-chemical characterisation of membrane-bound and water-soluble forms of Bacillus subtilis cytochrome c-550. Eur J Biochem 212:499–509
Welinder KG (1992) Superfamily of plant, fungal and bacterial peroxidases. Curr Opin Struct Biol 2:388–393 Zahn JA, Arciero DM, Hooper AB, Coats JR, DiSpirito AA (1997) Cytochrome c peroxidase from Methylococcus capsulatus Bath. Arch Microbiol 168:362–372 Zou P, Borovok I, Ortiz de Orué Lucana, D, Müller D, Schrempf H (1999) The mycelium-associated Streptomyces reticuli catalase-peroxidase, its gene and regulation by FurS. Microbiology 145:549–559
