Báo cáo khoa học: Characterization of a recombinantly expressed proteinase K-like enzyme from a psychrotrophic Serratia sp.

Characterization of a recombinantly expressed proteinase K-like enzyme from a psychrotrophic Serratia sp. Atle Noralf Larsen1, Elin Moe1,2, Ronny Helland2, Dag Rune Gjellesvik3 and Nils Peder Willassen1,2

1 Department of Molecular Biotechnology, Institute of Medical Biology, Faculty of Medicine, University of Tromsø, Norway 2 The Norwegian Structural Biology Centre, University of Tromsø, Norway 3 Biotec Pharmacon ASA, Tromsø, Norway

Keywords Bioprospecting; proteinase K like; psychrotrophic; Serratia sp; stability

Correspondence N. P. Willassen, Department of Molecular Biotechnology, Institute of Medical Biology, University of Tromsø, N-9037 Tromsø, Norway Tel: +47 77 64 46 51 Fax: +47 77 64 53 50 E-mail: nilspw@fagmed.uit.no

(Received 8 September 2005, revised 26 October 2005, accepted 31 October 2005)

doi:10.1111/j.1742-4658.2005.05044.x

The gene encoding a peptidase that belongs to the proteinase K family of serine peptidases has been identified from a psychrotrophic Serratia sp., and cloned and expressed in Escherichia coli. The gene has 1890 base pairs and encodes a precursor protein of 629 amino acids with a theoretical molecular mass of 65.5 kDa. Sequence analysis suggests that the peptidase consists of a prepro region, a catalytic domain and two C-terminal domains. The enzyme is recombinantly expressed as an active (cid:1) 56 kDa peptidase and includes both C-terminal domains. Purified enzyme is con- verted to the (cid:1) 34 kDa form by autolytic cleavage when incubated at 50 (cid:1)C for 30 min, but retains full activity. In the present work, the Serratia peptidase (SPRK) is compared with the family representative proteinase K (PRK) from Tritirachium album Limber. Both enzymes show a relatively high thermal stability and a broad pH stability profile. SPRK possess superior stability towards SDS at 50 (cid:1)C compared to PRK. On the other hand, SPRK is considerably more labile to removal of calcium ions. The activity profiles against temperature and pH differ for the two enzymes. SPRK shows both a broader pH optimum as well as a higher temperature optimum than PRK. Analysis of the catalytic properties of SPRK and PRK using the synthetic peptide succinyl-Ala-Ala-Pro-Phe-pNA as sub- strate showed that SPRK possesses a 3.5–4.5-fold higher kcat at the tem- perature range 12–37 (cid:1)C, but a fivefold higher Km results in a slightly lower catalytic efficiency (kcat ⁄ Km) of SPRK compared to PRK.

Serine endo- and exo- peptidases are widespread in nature and found in viruses, archaea, bacteria and euk- aryotes. The biological importance of peptidases are clearly indicated by the fact that 2% of all genes encode peptidases (or their homologues) in all kinds of organisms [1]. Extracellular peptidases hydrolyse large proteins into smaller peptides for absorption by the cell, whereas intracellular peptidases play a major role in regulation of metabolism [2].

The families of chymo(trypsin) (S1) and subtilisin (S8) are regarded as the largest families of serine peptidases

[1]. The two families share a similar arrangement of the catalytic triad, the Asp, His and Ser residues, but display a totally different protein fold where the subtilisin clan has an a ⁄ b-fold and the (chymo)trypsin clan a b ⁄ b-fold. More than 600 members of the subtilisin-superfamily (S8 family) are currently known according to the MER- OPS peptidase database (http://merops.sanger.ac.uk/). Siezen and Leunissen (1997) subdivided the subtilisin- like serine peptidases or subtilases into six families based on sequence homology, where the subtilisin and protein- ase K are examples of family representatives.

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

47

Abbreviations AQUI, aqualysin I; PMSF, phenylmethylsulphonyl fluoride; PRK, proteinase K; SPRK, Serratia sp. peptidase; VPRK, Vibrio sp. PA44 peptidase.

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

The proteinase K family is a large family of secreted endopeptidases found in fungi, yeast and Gram-negative bacteria, where especially the bacterial enzymes show a high degree of sequence identity (> 55%) [3]. The bacterial endopeptidases in this family are probably synthesized as prepro enzymes along with a C-terminal extension beyond the catalytic domain as reported for some of these enzymes [4–6]. Proteinase K from the fungus Tritirachium album Limber (PRK) is also pro- duced as a prepro enzyme but lacks the C-terminal extension [7]. The prepeptide functions as a signal pep- tide and is cleaved off after translocation of the protein through the membrane [8–11]. The pro-peptide probably functions as an intramolecular chaperone to ensure the proper folding of the enzyme and is cleaved off by autolysis to give the fully active enzyme [12]. The C-terminal extension might be involved in extracellular secretion as reported for aqualysin I (AQUI) in Thermus thermophilus cells [13,14].

organisms from the marine environment with features that might be of commercial interest. The polar marine regions are characterized by their stabile low tempera- ture where the sea temperature rarely exceeds 4 (cid:1)C. Enzymes from microorganisms living in such harsh environment show in general a higher catalytic effi- ciency (kcat ⁄ Km) and lower stability against tempera- ture or pH than enzymes from microorganisms adapted to warmer climate. For enzymes that are secreted, and often submitted to high substrate concen- tration, an optimization of the catalytic activity (kcat) might be a more valid approach for adaptation to cold than optimization of kcat ⁄ Km since the contribution of Km becomes negligible at high substrate concentrations [24]. VPRK is the only peptidase from the proteinase K family that has been isolated and characterized from a psychrotrophic or psychrophilic source [4,25]. This peptidase showed the typical characteristics of enzymes adapted to cold by having an increased catalytic effi- ciency (and catalytic activity) and lower thermal stabil- ity compared to related mesophilic and thermophilic counterparts.

Two 3D structures of peptidases have been deter- mined from the proteinase K family, and includes PRK [15] and a peptidase from a psychrotrophic Vibrio sp. PA44 (VPRK) [16]. Disulfide bridges may contribute to the overall stability of proteins, and both PRK and AQUI of this family have been described to contain two disulfide bridges in different positions [15,17]. VPRK contains three disulfide bridges according to the struc- ture, where the two first disulfide bridges are located in the same position as suggested for AQUI. The third disulfide bridge present in the VPRK structure is located in the C-terminal part of the enzyme.

Bioprospecting of marine microorganisms in coastal seawater in Northern Norway resulted in a large col- lection of diverse cold adapted bacteria that serves as a basis for exploration of different enzymatic activities for industrial or biotechnological use. In this paper we present a serine peptidase of the proteinase K family isolated from a psychrotrophic bacterium originating from this bioprospecting, and we report some of its properties compared to the commercially available and mesophilic PRK.

Results

The subtilisin-like peptidases are dependent on cal- cium to maintain their stability, and PRK contains two calcium binding sites, one strong (Ca1) site and one weak (Ca2) site [15]. VPRK possesses three cal- cium binding sites, where one corresponds to Ca1 in PRK, one corresponds to the medium site in thermi- tase [18] whereas the third site is new and not identi- fied in other subtilases so far [16].

Bioprospecting in coastal waters in Northern Norway resulted in a large collection of cold adapted (psychro- philic and psychrotrophic) bacteria. The bacterial strains were isolated and cultivated at 4 (cid:1)C, and the API ZYM system (BioMerieux, Paris, France) was chosen in order to study the enzymatic activities originating from these strains (unpublished data).

One of the marine bacteria showing peptidase activ- ity was closely related to Serratia proteamaculans of the Serratia genus belonging to the Enterobacteriaceae based on 16S rDNA analysis. The bacterium does not grow at 37 (cid:1)C, but grows well below 30 (cid:1)C indicating psychrotrophic nature.

Identification and analysis of the peptidase gene

PRK possesses a broad substrate specificity, but pre- fers to cleave peptide bonds after aliphatic and aroma- tic amino acids [19,20]. PRK is reported to be very stable even in presence of denaturants like urea and SDS. Cleavage of protein substrates by PRK is in fact stimulated by SDS [21]. The enhanced activity in the presence of SDS is probably due to denaturation of the protein substrate which in turn leads to increased accessibility for cleavage. Because of these features, PRK is typically used in procedures for inactivation of RNases and DNases during nucleic acid extraction [22,23].

Degenerate primers were constructed on the basis of sequence alignment of proteinase K-like multiple

Bioprospecting has become increasingly important in order to search for interesting genes, biomolecules and

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

48

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

the

full

‘libraries’,

restriction enzyme

enzymes from Gram-negative bacterial sources, and the codon usage in the sequences from Vibrio alginolyt- icus [26] and Alteromonas sp.O7 [5] were taken into account. A (cid:1) 200-bp fragment was generated by PCR and the sequence of this fragment was used for con- struction of PCR primers for genome walking (Gen- ome WalkerTM Kit, Clontech, Palo Alto, CA, USA). By using several different primers described in Experi- mental procedures and genome walking on the differ-

length ent sequence of the Serratia sp. peptidase (SPRK) gene was identified and found to be 1890 bp long, encoding a protein of 629 amino acids with a theoretical molecular mass of 65.5 kDa. The nucleotide sequence and deduced amino acid sequence is shown in Fig. 1. The peptidase sequence can be divided into a 22-resi- due presequence, a (cid:1) 100–105 residue pro-sequence, a catalytic domain of (cid:1) 280 residues and two C-terminal

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

49

Fig. 1. Nucleotide sequence and deduced amino acid sequence of the Serratia sp. gene encoding the precursor form of the peptidase. The catalytic residues Asp (D), His (H) and Ser (S) are bolded; N-terminal residues of the catalytic domain are under- lined. The preregion is indicated in red, the pro-region in black, catalytic domain in blue and the C-terminal domains in violet. The assumed start of the second C-terminal domain is indicated with a black arrow.

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

domains (repeated sequences) that are (cid:1) 220–225 resi- dues long (including linker-region between the catalytic and C-terminal domains) as indicated in Fig. 1. Data- base searches revealed that the deduced amino acid sequence showed high identity to other enzymes in the proteinase K family of serine peptidases, especially with sequences from Gram-negative bacterial sources. Sequences from cold adapted as well as sequences of included. mesophilic and thermophilic origin are

Figure 2A shows a multiple sequence alignment gener- ated by clustalx [27] of some of these sequences belonging to the proteinase K family, and the number- ing in this alignment is used throughout the Results and Discussion. In addition to the mesophilic family representative, PRK from the fungus T. album [7], sequences from Alteromonas sp. O7 [5], V. alginolyticus [26] and V. cholera [28] (mesophilic), T. aquaticus [6] (thermophilic), Pseudoalteromonas sp. AS11 (Genebank

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

50

Fig. 2. (A) Multiple alignment of the full length peptidase sequences from Serratia sp. (SPRK), Pseudoalteromonas sp. AS-11, Alteromonas sp. O-7, Vibrio sp. PA44, V. alginolyticus, V. cholera, Thermus aquaticus aqualysin I (AQUI) and Tritirachium album proteinase K (PRK). Blue is 100% sequence identity, red is 80–99% while green is 60–79% sequence identity. The catalytic domain from position 145 to 429. (B) Multiple alignment of the C-terminal sequences from Serratia sp. (SPRK), Pseudoalteromonas sp. AS-11, Alteromonas sp. O-7, Vibrio sp. PA44, V. alginolyticus, V. cholera, T. aquaticus (AQUI) belonging to the proteinase K family of serine peptidases. In addition, C-terminal sequences of zinc metolloproteases from V. cholera S01, Helicobacter pylori, V. anguillarum, V. vulnifucus and V. parahaemolyticus are included. Blue is 100% sequence identity, red is 80–99% while green is 60–79% sequence identity. Both alignments are generated using ClustalX.

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

accession number: BAB61726) and Vibrio sp. PA44 [4] (cold-adapted) are included.

Fig. 3. SDS ⁄ PAGE (4–12% Bis-Tris) of purified SPRK. Lanes 1 and 4: SeeBlue(cid:2) standard (Invitrogen); Lane 2: Purified SPRK without addition of PMSF prior to SDS ⁄ PAGE analysis; Lane 3: Purified SPRK (PMSF inhibited); Lane 5: Heat treated (50 (cid:1)C) and purified SPRK (PMSF inhibited).

The catalytic domain is well conserved, especially the sequences around the catalytic triad (D183, H216 and S373). There are three disulfide bridges present in the VPRK structure [16]. The two first disulfide brid- ges observed in VPRK are in agreement with sugges- tions made for AQUI [17], and are formed between C213-C245 and C314-C345. Serratia sp. peptidase pos- sesses cysteines in the equivalent sequence positions as VPRK and AQUI, hence these disulfide bridges are probably present in SPRK. PRK has its disulfide brid- ges positioned elsewhere (C178-C270, C325-C399) [15]. Based on Fig. 2A, one or two extended C-terminal region(s) of the peptidase sequences are common within the bacterial subgroup of the proteinase K fam- ily. Database search on the second C-terminal domain (CII) of SPRK revealed that this region shows > 43% sequence identity with C-terminal region(s) in most of the other sequences in this alignment. The exception is the sequence from T. aquaticus which only shows about 15% identity. In addition, several sequences of metallopeptidases originating from pathogenic organ- isms have a C-terminal region showing > 43% sequence identity with CII of SPRK. Figure 2B shows a multiple alignment of the C-terminal sequences ori- ginating from peptidase sequences of the proteinase K family along with some of the metallopeptidase sequences.

Expression and purification

sp. peptidase was purified approximately sixfold with a total yield of (cid:1) 0.7 mg. Serratia sp. peptidase is expressed as a (cid:1) 56-kDa protein, but after purification five bands at 56, 45, 34, 28.5 and 22 kDa appear when analysing the purified sample by SDS ⁄ PAGE as shown in Fig. 3 (lane 3). The purified sample was incubated with 1 mm (final concentration) phenylmethylsulfonyl fluoride (PMSF) to inhibit autolytic degradation prior to analysis on SDS ⁄ PAGE. If the peptidase sample was not treated with PMSF during preparation for electro- phoresis, the major band observed in the gel corres- lane 2). No ponds to the 34-kDa protein (Fig. 3, proteins above this size could be observed, although some weak degradation products could be detected.

Molecular characteristics

lane 3). Conversion of

The gene encoding SPRK was cloned into the pBAD ⁄ gIII vector (Invitrogen) for recombinant expression in Escherichia coli TOP10. The presequence of SPRK was not included in the construct as a signal sequence is provided in this vector. Small-scale expression was compared at 37 (cid:1)C, 30 (cid:1)C and 22 (cid:1)C, but peptidase activity could only be detected at 22 (cid:1)C. Large-scale expression was therefore performed at 22 (cid:1)C.

Some characterized bacterial enzymes in the proteinase K family that have a C-terminal extension have pre- viously been shown to include several bands on a SDS ⁄ PAGE gel after purification [5,25], as seen with SPRK (Fig. 3, the enzyme sample from the (cid:1) 56-kDa protein to the 34-kDa protein readily took place when incubating the enzyme at 50 (cid:1)C (Fig. 4). No decrease in enzyme activity,

The purification of SPRK includes ion exchange, hydrophobic interaction chromatography and gel filtra- tion and the scheme is summarized in Table 1. Serratia

Table 1. Purification scheme of SPRK expressed in E. coli.

Step Volume (ml) Total activity (U) Total protein (mg) Yield (%) Purification (fold) Activity (U ⁄ ml) Protein concentration (mg ⁄ mL) Specific activity (U ⁄ mg)

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

51

Periplasmic fraction Q-sepharose Phenyl-seph. Source 15Q Superdex 75 435 160 45 22.5 1.5 1.23 2.52 5.61 8.82 41.33 0.08 0.15 0.26 0.16 0.46 534 403 252 198 62 34.4 23.8 11.7 3.7 0.7 15.5 16.9 21.5 54 88.6 100 75 47 37 12 1 1.1 1.4 3.5 5.7

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

Fig. 5. pH stability of SPRK and PRK. Enzymes were preincubated for 24 h at 22 (cid:1)C at various pHs. One hundred percent activity refers to the pH value with highest activity. (r), SPRK; (n), PRK.

Fig. 4. Processing of the purified SPRK. SDS ⁄ PAGE (4–12% Bis- Tris) showing the effect of incubation at 50 (cid:1)C on the apparent molecular mass. PMSF is added to a final concentration of 1 mM at each time point to inhibit enzyme activity. Lane 1: SeeBlue(cid:2) stand- ard; Lane 2–8: Enzyme sample heated to 50 (cid:1)C in time intervals ranging from 0 to 45 min.

[29], we

that

however, was observed during incubation (results not shown). Based on the results shown in Fig. 4, together with analysis of other enzymes in the same family; Alteromonas sp. O-7 [5], Vibrio sp. PA44 [4,25] and T. aquaticus the bands at suggest (cid:1) 56 kDa and (cid:1) 45 kDa refer to a peptidase form including two C-terminal domains and one C-terminal domain, respectively. The protein band at (cid:1) 34 kDa refers to the ‘mature’ peptidase containing the catalytic domain only.

temperatures ranging from 4 to 80 (cid:1)C in 15 min. PRK was slightly more stable than SPRK, and had a half- life of 30 min at 70 (cid:1)C while SPRK had a half-life of 19 min at this temperature. Stability of SPRK and PRK towards SDS was measured by preincubating the enzymes with various concentrations of SDS (0.1, 0.25, 0.5 and 1.0%) at 37 (cid:1)C and 50 (cid:1)C for 30 min, and the results are shown in Fig. 6. At 37 (cid:1)C there were no sig- nificant difference between the two enzymes, both hav- ing (cid:1) 90% residual activity even in presence of 1% SDS. Significant differences in stability between the two enzymes appeared at 50 (cid:1)C. Serratia sp. peptidase still had 90% residual activity in the presence of 1% SDS, while PRK only had (cid:1) 19%. Stability of SPRK and PRK towards EDTA was tested by preincubating the enzymes at 37 (cid:1)C and 50 (cid:1)C for 120 min, and the results are shown in Fig. 7. At 37 (cid:1)C, PRK is

To verify the experiment shown in Fig. 4, and to obtain the ‘mature’ form of the peptidase, a periplas- mic extract of SPRK was submitted to the same purifi- cation procedure as described previously with one major exception: the concentrated sample (3 mL) was heated to 50 (cid:1)C for 30 min before application on a Superdex 75 (2.6 ⁄ 60) column. Figure 3, lane 5 shows the SDS ⁄ PAGE after gel filtration (the sample was treated with PMSF as described previously). One single band corresponding to a protein of (cid:1) 34 kDa was present in the gel. As conversion to the 34-kDa protein or ‘mature’ form readily took place at 50 (cid:1)C, only the ‘mature’ form of SPRK was used during further characterization experiments.

Stability

), SPRK 37 (cid:1)C; ( Fig. 6. Stability of SPRK and PRK towards SDS at 37 (cid:1)C and 50 (cid:1)C. The enzymes were preincubated for 30 min in buffer containing 0.1%, 0.25%, 0.5% and 1% SDS. One hundred percent activity refers to enzyme samples incubated at the selected temperatures during the experiments without SDS present. ( ), PRK 37 (cid:1)C; ( ), SPRK 50 (cid:1)C; ( ), PRK 50 (cid:1)C.

The pH stability of SPRK and PRK was compared by preincubating the enzymes for 24 h at 22 (cid:1)C in buffers of different pH. PRK was stable in the pH range from pH 4 to 12, while SPRK had optimal stability in the range from pH 5.5 to 9.5 (Fig. 5). Temperature stabil- ity was measured by preincubating SPRK and PRK at

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

52

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

Fig. 9. Temperature optimum for activity of SPRK and PRK. Enzyme assay was performed in the temperature range of 20–75 (cid:1)C. One hundred percent activity refers to the temperature value with the highest activity. (r), SPRK; (n), PRK.

Fig. 7. Stability of SPRK and PRK towards EDTA at 37 (cid:1)C and 50 (cid:1)C. Enzymes were incubated at the selected temperatures in calcium free buffers containing 10 mM EDTA, and sampled after 15, 30, 45, 60, 90 and 120 min. One hundred percent (0 min) resid- ual activity refers to enzyme sample incubated on ice. (r), SPRK 37 (cid:1)C; (n), PRK 37 (cid:1)C; (d), SPRK 50 (cid:1)C; (m), PRK 50 (cid:1)C.

The temperature optimum was determined to be 70 (cid:1)C for SPRK, and 55 (cid:1)C for PRK (Fig. 9). Protein- ase K exhibits a broad optimum with > 90% activity in the temperature range 40–70 (cid:1)C.

unaffected by the presence of EDTA, while SPRK had (cid:1) 60% residual activity. At 50 (cid:1)C, SPRK was totally inactivated after 120 min, while PRK retained (cid:1) 50% residual activity.

Effect of SDS and EDTA on activity

pH and temperature optimum

The pH optimum for activity of SPRK and PRK was determined by measuring the enzyme activity towards suc-Ala-Ala-Pro-Phe-pNA at different pH values. Ser- ratia sp. peptidase had a broad pH optimum with the highest activity in the range pH 8–11, and an apparent optimum at pH 10.5; PRK had the highest activity in the range pH 8–9.5, and an apparent optimum at pH 8 (Fig. 8).

The effect of SDS on activity of SPRK and PRK was measured by addition of 0.1, 0.25, 0.5 and 1.0% SDS (final concentrations) in the standard assay buffer. Both enzymes were inhibited by addition of SDS dur- ing activity measurements, and showed (cid:1) 30% of the maximum activity in presence of 1% SDS (Table 2). The effect of EDTA on activity was measured by including EDTA (10 mm) in a calcium-free assay buf- fer. EDTA had no inhibitory effect on the activity of the enzymes (Table 2).

Kinetics

activity)

and kcat ⁄ Km (catalytic

To investigate if there were any differences in kcat (cat- efficiency) alytic between SPRK and the mesophilic PRK, kinetic experiments using the synthetic substrate suc-Ala-Ala- Pro-Phe-pNA was performed at 12, 22 and 37 (cid:1)C. The

Table 2. Effect of SDS and EDTA on activity for SPRK and PRK at 22 (cid:1)C.

Inhibitor Concentration SPRK (% relative activity) PRK (% relative activity)

SDS

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

53

EDTA 0.10% 0.25% 0.50% 1.00% 10 mM 85 67 49 30 100 87 77 56 32 100 Fig. 8. pH optimum of SPRK and PRK. Enzyme assay was per- formed at 22 (cid:1)C in different buffers from pH 5.5–11, and activity towards Suc-Ala-Ala-Pro-Phe-pNA was measured. One hundred per- cent activity refers to the pH value with the highest activity. (r), SPRK; (n), PRK.

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

Table 3. Kinetic parameters for the hydrolysis of suc-AAPF-pNA at 12 (cid:1)C, 22 (cid:1)C and 37 (cid:1)C for SPRK and PRK.

SPRK PRK

Substrate kcat Km kcat ⁄ Km kcat Km kcat ⁄ Km

175 364 827 2,36 2,46 2,72 74 148 304 51 88 180 0,48 0,46 0,52 106 191 346 S-AAPF-pNA12 (cid:1)C S-AAPF-pNA22 (cid:1)C S-AAPF-pNA37 (cid:1)C

kinetic parameters of SPRK and PRK are shown in Table 3. Serratia sp. peptidase had a 3.5–4.5 fold higher kcat at all temperatures tested. On the other hand, SPRK had a fivefold higher Km (lower binding affinity) leading to a slightly lower catalytic efficiency at the selected temperatures compared to PRK.

Discussion

identity with the other sequences, although its cata- lytic domain has 60% sequence identity. Database searches using CII from SPRK revealed an interest- ing feature as several metallopeptidase also showed identity with CII of SPRK. > 43% sequence Recently, it has been shown that a metallopeptidase from V. anguillarum with a similar C-terminal region (C-terminal sequence is shown in Fig. 2B) is import- ant for virulence in Atlantic salmon [31]. In addition, the C-terminal domain of a metallopeptidase from V. vulnificus with > 50% sequence identity to CII of SPRK, is shown to be essential for efficient attach- ment to protein substrates or erythrocyte membranes [32]. The question arises why peptidases from the bacterial subgroup of the PRK family and the metal- loproteases have one similar C-terminal domain or two (repeated) domains as seen for peptidase sequen- ces from Alteromonas sp. O7, Pseudoalteromonas sp. AS11 and the Serratia sp.? From the information discussed above one might speculate that the C-ter- minal domains of SPRK could have an additional reported for AQUI, and may function than that function in attaching the peptidase to cellular surfa- ces or protein substrates.

It has also been suggested that

[13,14].

identity compared to the

Disulfide bridges may contribute to the overall sta- bility of proteins, and some peptidases of this family are known to contain cysteine residues involved in disulfide bridges. Proteinase K and AQUI are both described to have two disulfide bridges, but at differ- ent positions in the structure [15,17], whereas the VPRK structure revealed the presence of three disul- fide bridges [16]. One or more of the disulfide bonds is shown to be essential for maintaining the active conformation of VPRK, since cleavage of the disul- fides lead to inactivation of the enzyme [25]. The first two disulfide bonds in VPRK are the same as suggested for AQUI and should also be present in SPRK. Based on the sequence alignment in Fig. 2A, it seems that all members of the bacterial subgroup contain these two disulfide bonds. Attempts to stabil- ize the cysteine-free subtilisin BPN¢ by introducing disulfide bridges in structurally analogous positions to those in PRK showed no stabilizing effect [33]. However, stabilizing subtilisin E by introducing an S–S bond (positioning C213–C245 compared to Fig. 2A) was successfully performed using AQUI as a template molecule [34]. This introduction did not affect the catalytic efficiency of the enzyme, and may therefore be a suitable target for site-directed muta- genesis in order to create a more temperature labile SPRK.

The two peptidases PRK and SPRK possess both high thermal and pH stability. Proteinase K was stable

Based on 16S rDNA sequencing, the gene encoding a PRK-like serine peptidase was isolated from a bac- terial strain most closely related to S. proteamaculans the Serratia genus. The gene was found to be of 1890 bp long, encoding a precursor protein of 629 amino acids with a theoretical molecular mass of 65.5 kDa. The deduced amino acid sequence of the peptidase gene revealed that the peptidase consists of an N-terminal prepro sequence, a catalytic domain and two C-terminal domains (repeated sequences). The presequence (22 residues) helps to guide the pro- tein into the periplasmic space [10], while the pro- sequence ((cid:1) 100–105 residues) assist the peptidase to achieve its correct folding [12]. The catalytic domain consists of (cid:1) 280 residues. The function of the C-ter- minal domains in SPRK is ((cid:1) 220–225 residues) unknown, but may be necessary for extracellular secretion as reported for AQUI in T. thermophilus the cells C-terminal pro-sequence may play a role in translo- cation across both the cytoplasmic and outer mem- branes [30]. In the case of the psychrotrophic Vibrio peptidase (VPRK), the wild-type enzyme was secre- ted into the medium as a 47-kDa peptidase with the C-terminal domain intact, and was converted to the 36-kDa ‘mature’ form by mild heat treatment [25]. Furthermore, VPRK has also been recombinantly expressed in E. coli and showed similar molecular characteristics to those of the wild-type enzyme [4]. The C-terminal region (CII) of SPRK shows > 43% sequence corresponding region of the bacterial members in the PRK family compared here (Fig. 2B). The only exception is the region of AQUI which has (cid:1) 15% C-terminal

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

54

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

characterization of

twofold higher kcat than its mesophilic homologue subtilisin Carlsberg, but has on the other hand a higher Km, leading to a more or less preserved cata- lytic efficiency [36]. These results differ somewhat the psychrotrophic from the VPRK that possesses both higher kcat and kcat ⁄ Km ratio at moderate (15–45 (cid:1)C) temperatures compared to mesophilic (PRK) and thermophilic (AQUI) coun- terparts [25].

over the whole pH range tested from pH 4–12 and had a half-life of 30 min at 70 (cid:1)C, while SPRK possessed highest stability from pH 5.5)9.5 and had a half-life of 19 min at 70 (cid:1)C. An interesting feature was the differ- ence in stability between the two peptidases toward SDS and EDTA (Fig. 6 and 7). Serratia sp. peptidase was clearly more stable against SDS at 50 (cid:1)C, but showed stability similar to that of PRK at 37 (cid:1)C. Pro- teinase K is, on the other hand, significantly more sta- ble towards EDTA at both 37 (cid:1)C and 50 (cid:1)C, indicating that SPRK is more dependent on calcium for stability.

To elucidate the differences in stability and activity between SPRK and PRK, a high-resolution structure of SPRK is needed. The catalytic domain of SPRK has been crystallized and the crystal structure was compared with the already known structure of PRK and will be published in an accompanying paper in FEBS [37].

Since there were significant differences in kcat and Km between the two enzymes, kinetic studies to iden- tify possible differences in the substrate-binding region will be initiated. This knowledge will further be used in redesign of SPRK to yield an enzyme with higher catalytic efficiency and lower temperature stability.

Experimental procedures

Materials

It was difficult to get reproducible pH optimum measurements for PRK in different 0.1 m buffers. Nev- ertheless, the results indicate that SPRK possesses a broader (and higher) pH optimum for activity than PRK (Fig. 8). Interestingly, SPRK also showed a higher temperature optimum for activity (Fig. 9). Pro- teinase K has a broad temperature optimum with only minor difference in activity in the temperature range from 40–70 (cid:1)C. PRK has also previously been des- cribed to have a broad temperature optimum profile with more than 80% of the maximum activity in the range of 20–60 (cid:1)C (with an apparent optimum at 37 (cid:1)C) [35]. Serratia sp. peptidase shows the same tem- perature and similar pH optimum for activity as repor- ted for the peptidase from Alteromonas sp. O-7 [5]. Serratia sp. peptidase and PRK were unaffected by the presence of EDTA, while the activity of both were inhibited in the presence of SDS (Table 2). Protein- ase K has previously been demonstrated to exhibit similar effects of EDTA and SDS on activity when act- ing on small substrates [20,21].

from Amersham Biosciences

The Genome WalkerTM kit was from Clontech (Palo Alto, CA, USA). Restriction enzyme NcoI was from New Eng- land Biolabs (Beverly, MA, USA). Escherichia coli TOP10 [F- mcrA n(mrr-hsdRMS-mcrBC) u80lacZnM15 nlacX74 deoR recA1 araD139 n(araAleu)7697 galU galK rpsL endA1 nupG] and expression vector pBAD ⁄ gIII were from Invitrogen (Carlsbad, CA, USA). Q-Sepharose FF, Phenyl sepharose FF, Hi-Prep Desalting, Source 15Q and Super- dex 75 were (Uppsala, Sweden). Suc-Ala-Ala-Pro-Phe-pNA and PRK were from Sigma Aldrich (St. Louis, MO, USA) and Finnzymes (Espoo, Finland), respectively.

No significant differences in pH or temperature sta- bility ⁄ optimum were found between purified samples of the unprocessed (56 kDa) and processed (34 kDa) SPRK (data not shown); this is in accordance with analysis performed with the peptidase from the Vibrio sp. PA44 [25].

16SrDNA sequencing

activity)

and Km (substrate

temperatures

the moderate

Bacterial genomic DNA was purified by using Qiaquick DNA purification kit (Qiagen, Germany) according to manufacturer’s protocol. Polymerase chain reaction was performed with 50 ng template DNA, 0.2 mm dATP, dCTP, dGTP and dTTP, 0.2 lm upstream primer (5¢-AGA GTTTGATCMTGGCTCAG-3¢) and downstream primer (5¢-GGTTACCTTGTTACGACTT-3¢) and 1 U Taq poly- merase (Promega). PCR amplification was carried out at 95 (cid:1)C for 5 min, 30 cycles of 95 (cid:1)C for 30 s, 53 (cid:1)C for 30 s and 72 (cid:1)C for 1 min, and a final extension step of 72 (cid:1)C for 7 min.

Significant differences in the kinetic parameters, kcat (catalytic binding), between the two peptidases were observed. Serratia sp. peptidase had a much higher kcat (3.5–4.5 fold) than PRK at tested (12 (cid:1)C, 22 (cid:1)C and 37 (cid:1)C), and the difference in kcat between the two enzymes increased slightly with increasing temperature. Serratia sp. peptidase exhib- ited a much higher Km at the same temperatures (fivefold), leading to a slightly lower catalytic effi- ciency in SPRK. Similar effects have been observed in subtilisin S39 from the psychrophilic Antarctic Bacillus TA39 when hydrolysing the substrate suc- shows FAAF-pNA. The

psychrophilic

enzyme

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

55

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

Isolation of genomic DNA from Serratia sp.

Construction of expression vector

Genomic DNA was isolated as described by Chen and Kuo [38] for use in identification of the peptidase gene.

Generation of an (cid:1) 200-bp fragment of the peptidase gene

and

The peptidase gene lacking the first 66 bp (encoding the pre- sequence) was cloned into pBAD ⁄ gIII B expression vector (Invitrogen). PCR was performed in 50 lL containing 1 ng of genomic DNA as template, 0.2 mm dATP, dCTP, dGTP and dTTP, 0.2 lm of upstream primer (OP17: 5¢-GA AAAACCATGGTGAATGAATACCAAGCGACT-3¢) and downstream primer (NP7: 5¢-CAATCTCCATGGCTAG TAGCTTGCACTCAG-3¢) containing a NcoI restriction site and 1 U of Pfu-polymerase. PCR amplification was carried out at 94 (cid:1)C for 5 min, 30 cycles at 94 (cid:1)C for 30 s, 60 (cid:1)C for 1 min and 3 min at 72 (cid:1)C and a final extension step at 72 (cid:1)C for 5 min. PCR products were purified using Qiaquick PCR Purification Kit (Qiagen), digested with 10 U NcoI (New England Biolabs), ligated into NcoI digested pBAD ⁄ gIII B expression vector using T4-DNA-ligase and transformed into competent TOP10 E. coli cells.

Polymerase chain reaction was carried out in a final vol- ume of 50 lL containing 1 ng of bacterial genomic DNA template, 10 mm Tris ⁄ HCl pH 9.0 (25 (cid:1)C), 50 mm as KCl, 0.1% Triton X-100, 0.2 mm dATP, dCTP, dGTP and dTTP, 0.4 lm upstream primer (5¢-GACTGTAA CGGTCATGGYACMAYGT-3¢) and downstream primer (5¢-CCGCCACCCAAACTCATRTTRGC-3¢) 1.5 U Taq-polymerase (Promega). PCR-amplification was per- formed at 94 (cid:1)C for 7 min, 30 cycles at 94 (cid:1)C for 30 s, 60 (cid:1)C for 80 s and 2 min at 72 (cid:1)C, and a final extension step at 72 (cid:1)C for 5 min.

DNA sequencing

Full length gene identification

DNA sequencing was performed with the Amersham Phar- macia Biotech Thermo Sequenase Cy5 Dye Terminator Kit, ALFexpressTM DNA Sequencer and ALFwin Sequence Analyser version 2.10 according to the manufacturer’s instructions. Gels were made with ReprogelTM Long Read and Reproset UV-polymerizer. All items were from Amer- sham Biosciences (Uppsala, Sweden).

Expression and fermentation of SPRK in E. coli

Genomic DNA was treated according to the Genome WalkerTM kit manual (Clontech) with four different blunt end restriction enzymes; EcoRV, DraI, PvuII and SspI each giving rise to a genome walking ‘library’. The following gene specific primers were used to obtain the full length sequence: OP5, 5¢-GACTGTAACGGTCATGGYACMAYGT-3¢; OP6, 5¢-GATGAAAATCCTAACCTCTCCCCCGCACAG- 3¢; OP7, 5¢-ACTGCACCTACGGCGGGTCGTTGGTACG TG-3¢; NP4, 5¢-GACACCGTAGGTTGAGCCGCCAATC GTCCC-3¢; NP5, 5¢-CTTTAACTTGTTGGGCACTGG CATTG-3¢; NP6, 5¢-TTGATCGATTCTGTCTATGCCC CA-3¢ along with the adaptor primers: AP1 (5¢-GTAATAC GACTCACTATAGGGC-3¢) and AP2 (5¢-ACTATAGGG CACGCGTGGT-3¢).

Small-scale expression was performed at 37, 30 and 22 (cid:1)C in 1-L baffled shake flasks containing 100 mL Luria– Bertani (LB) medium with 20 mm glucose and 50 lgÆmL)1 ampicillin. A 10-mL preculture of E. coli TOP10 pBAD ⁄ gIIIB containing the SPRK gene was used as inoculum, and induced with 0.1% arabinose. Fermentation was per- formed in a 15-L Chemap CF 3000 fermentor (Switzer- land). A 200-mL preculture of E. coli TOP10 pBAD ⁄ gIIIB containing the SPRK gene was inoculated to 7 L of 2· LB- medium supplemented with 20 mm glucose and 50 lgÆmL)1 ampicillin. Cells were grown until no glucose could be detected (OD600)2.5). Gene expression was induced by 0.1% arabinose and cells were grown further for 12 h at 22 (cid:1)C. Cells were harvested and centrifuged at 5000 g for 15 min at 4 (cid:1)C.

Purification of SPRK

Bacterial cell pellet was resuspended in 10% of the ori- ginal volume (700 mL from 7 L culture) in 20% sucrose, 0.1 m Hepes, 1 mm EDTA. Freshly made lysozyme was added to a final concentration of 0.5 mgÆmL)1. The solu- tion was incubated 30 min at 22 (cid:1)C, and centrifuged for

Nested PCR was carried out in a final volume of 50 lL in 20 mm containing 1 lL of a genome walking ‘library’ Tris ⁄ HCl pH 8.8 (25 (cid:1)C), 10 mm KCl, 10 mm (NH4)2SO4, 2 mm MgSO4, 0.1% Triton X-100, 0.1 mgÆmL)1 nuclease free BSA, 0.2 mm dATP, dCTP, dGTP and dTTP, 0.2 lm gene specific primer and adaptor primer and 1 U Pfu-poly- merase (Promega). PCR-amplification was done at 94 (cid:1)C for 2 min, 7 cycles at 94 (cid:1)C for 30 s, 55 (cid:1)C for 30 s and 4 min at 72 (cid:1)C, 30 cycles at 94 (cid:1)C for 30 s, 50 (cid:1)C for 30 s and 4 min at 72 (cid:1)C and a final extension step at 72 (cid:1)C for 5 min. The final product of this first PCR reaction (1 lL) was used as template in a secondary or nested PCR reaction in 20 mm Tris ⁄ HCl pH 8.8 (25 (cid:1)C), 10 mm KCl, 10 mm (NH4)2SO4, 2 mm MgSO4, 0.1% Triton X-100, 0.1 mgÆmL)1 nuclease free BSA, 0.2 mm dATP, dCTP, dGTP and dTTP, 0.2 lm gene specific primer and adaptor primer and 1 U Pfu-poly- merase (Promega) and 94 (cid:1)C for 2 min, 30 cycles of 94 (cid:1)C for 30 s, 55 (cid:1)C for 1 min and 4 min at 72 (cid:1)C and a final extension step of 72 (cid:1)C for 5 min.

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

56

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

SDS ⁄ PAGE

SDS ⁄ PAGE was performed using The Nupage(cid:2) Pre-Cast gel system (Invitrogen) according to manufacturer’s proto- col. Gels were stained with SimplyBlueTM SafeStain (Invi- trogen). The standard used is SeeBlue(cid:2) Pre-Stained Protein Standard (Invitrogen). Enzyme samples were treated with 1 mm (final concentration) of the serine peptidase inhibitor for 30 min at phenylmethylsulphonyl fluoride (PMSF) room temperature prior to analysis on 4–12% NuPAGE(cid:2) Novex Bis-Tris Gels. The effect of temperature on auto- catalytic cleavage was performed using purified SPRK incu- bated at 50 (cid:1)C. Samples were taken after 1, 5, 10, 15, 30 and 45 min and analysed by SDS ⁄ PAGE. Approx. 3 lg of protein sample was added to each well.

pH and temperature stability

30 min at 5000 g at 4 (cid:1)C. The supernatant contained the periplasmic fraction of the proteins. CaCl2 was added to a final concentration of 10 mm before freezing at )20 (cid:1)C. Periplasmic fraction (450 mL) was filtrated through glass- wool, and then applied to a Q-Sepharose FF (5.0 ⁄ 10, 50 mm in diameter and 10 cm in height) equlibrated in buffer A (25 mm Hepes pH 8.0, 1% glycerol, 10 mm CaCl2). The column was washed with three column vols (CV) of buffer A and the proteins were eluted over four CV isocratic 30% buffer B (buffer A + 1.0 m NaCl) and isocratic 100% buffer B over two CV. Fractions contain- ing peptidase activity were pooled. Ammonium sulphate (80 mL, 3 m) was added to the pooled Q-Sepharose frac- tion (160 mL), and applied to a phenyl sepharose FF (high substituted, 1.6 ⁄ 10) column. The column was equli- brated with buffer C (25 mm Hepes pH 8.0, 10 mm CaCl2, 1% glycerol, 1 m ammonium sulphate). The col- umn was washed with three CV of buffer C, and pro- teins were eluted using two isocratic steps; 25% buffer A over seven CV and 100% buffer A over seven CV. The peptidase containing fractions collected in the last step were pooled. The pooled fraction (45 mL) after phenyl sepharose was applied to a Hi-Prep Desalting (26 ⁄ 10) column equlibrated in buffer A. The protein fraction after the desalting step (62.5 mL) was applied to a Source 15Q (2.6 ⁄ 3.5) equlibrated with buffer A. The column was washed with three CV of buffer A, and bound proteins were eluted with a linear gradient of 0–100% buffer B over 10 CV. Fractions containing pepti- dase activity were pooled. Using an Amicon Ultra the Source 15Q fraction (22.5 mL) (15 mL, Millipore) were concentrated to 2.2 mL, and applied to the Super- dex 75 (2.6 ⁄ 60) column equlibrated with buffer A +15% buffer B. Fractions containing peptidase activity were pooled.

The effect of pH on stability was determined by preincubat- ing SPRK ((cid:1)5 lgÆmL)1) and proteinase K (PRK, (cid:1)10 lgÆmL)1) for 24 h at 22 (cid:1)C in the following buffers containing 10 mm CaCl2: 25 mm sodium acetate (pH 4.0– 5.5), 25 mm Mes ⁄ NaOH (pH 5.5–6.5), 25 mm Mops ⁄ - NaOH (pH 6.5–7.5), 25 mm Tris ⁄ HCl (pH 7.5–8.5), 25 mm diethanolamine ⁄ HCl (pH 8.5–9.5), 25 mm piperazine ⁄ HCl (pH 9.5–10), 25 mm glycine ⁄ NaOH (pH 10–12). The effect of temperature on stability was determined by incubating SPRK ((cid:1)5 lgÆmL)1) and PRK ((cid:1)10 lgÆmL)1) at tempera- tures ranging from 4 to 80 (cid:1)C for 15 min in 25 mm Hepes pH 8.0, 10 mm CaCl2. The samples were kept on ice prior to and after incubation. Temperature stability at fixed temperature (70 (cid:1)C) was also performed to estimate the half-life. Residual activities towards Suc-AAPF-pNA were determined under standard assay conditions. One hundred percent activity refers to samples kept on ice during the whole experiment.

Protein determination

Stability against SDS and EDTA

Protein concentrations were determined with Bio-Rad Pro- tein Assay based on the method of Bradford [39] and according to the microtiter plate protocol as described by the manufacturer using BSA as standard.

Standard enzyme assay

SPRK activity was routinely measured using Suc-Ala-Ala- Pro-Phe-pNA (Sigma Aldrich) as substrate. The peptidase assay was carried out in a total volume of 250 lL, contain- ing 10 lL enzyme solution and 240 lL of standard assay buffer (0.1 m Tris ⁄ HCl pH 8.0, 10 mm CaCl2, 1% DMSO, 1 mm substrate) for 2 min at 22 (cid:1)C. Activity was measured by increase in absorbance at 405 nm (e ¼ 8480 m)1Æcm)1) using the Vmax kinetic microplate reader (Molecuar Devices Co., CA). One unit of enzyme hydrolyses 1 lmol of sub- strate per minute at 22 (cid:1)C.

The effect of SDS on the stability of SPRK ((cid:1)3 lgÆmL)1) and PRK ((cid:1)7.5 lgÆmL)1) was determined by incubating the enzymes in 25 mm Hepes, 1 mm CaCl2 pH 8.0 at 37 (cid:1)C and 50 (cid:1)C for 30 min containing various SDS concentra- tions (0.1, 0.25, 0.5 and 1%). Residual activities towards Suc-AAPF-pNA were determined using standard assay con- ditions. One hundred percent activity refers to enzyme sam- ples incubated at the selected temperatures without SDS present. The effect of EDTA on the stability of SPRK ((cid:1)3 lgÆmL)1) and PRK ((cid:1)7.5 lgÆmL)1) was determined by incubating the enzymes in 25 mm Hepes pH 8.0, 10 mm EDTA at 37 (cid:1)C and 50 (cid:1)C and samples were collected after 15, 30, 60, 90 and 120 min. At each time point, the sample was incubated on ice for 5 min and remaining activity toward Suc-AAPF-pNA was determined. One hundred percent activity (0 min) is sample incubated on ice for

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

57

standard assay

5 min before measuring activity under conditions.

to fit

regression analysis

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

pH and temperature optimum

linear plots

concentrations. The parameters Km and kcat were estimated by nonlinear the Michaelis– Menton equation by using the SigmaPlot Kinetic module (Systat Software Inc., Richmond, CA). When Km values reached the upper limit for the substrate concentrations, also manual such as Eadiee–Hofstee and Hanes–Woolf were used as verification of the nonlinear regression analysis.

Database searches and prediction

blast (http://www.ncbi.nlm.nih.gov/blast/) was used to search for related protein sequences of SPRK. N-terminal presequence was predicted by Signal P [10].

Acknowledgements

The present study was supported by the national Functional Genomics Programme (FUGE) in The Research Council of Norway and Biotec Pharmacon ASA. We wish to thank Jonas Jakobsen and Sigmund Sperstad for technical assistance.

References

The effect of pH on the activity of SPRK ((cid:1)3.75 lgÆmL)1) and PRK ((cid:1)7 lgÆmL)1) towards 1 mm Suc-AAPF-pNA was determined at 22 (cid:1)C using the following buffers con- taining 1% (v ⁄ v) DMSO and 10 mm CaCl2: 0.1 m sodium acetate ⁄ HCl (pH 4.0–5.5), 0.1 m Mes ⁄ NaOH (pH 5.5–6.5), 0.1 m Mops ⁄ NaOH (pH 6.5–7.5), 0.1 m Tris ⁄ HCl (pH 7.5– 8.5), 0.1 m diethanolamine ⁄ HCl (pH 8.5–9.5), 0.1 m pipera- (pH 9.5–10), 0.1 m glycine ⁄ NaOH (pH 10–12). zine ⁄ HCl One hundred percent activity refers to the pH value with highest measured activity. Temperature optimum was per- formed using a thermostatted Perkin Elmer Lamda15 uv ⁄ vis Spectrophotometer. The peptidase assay was carried out in a total volume of 500 lL, containing 20 lL diluted enzyme solution (in 25 mm Hepes, 10 mm CaCl2 pH 8.0) and 480 lL of 0.1 m Tris ⁄ HCl, 0.0005% Triton X-100, 10 mm CaCl2 and 0.5% DMSO (pH adjusted to 8.0 for all temperatures). The temperature range used was 20–75 (cid:1)C, and the activity of SPRK ((cid:1)0.75 lgÆmL)1) and PRK ((cid:1)1 lgÆmL)1) towards 0.5 mm and 0.1 mm Suc-AAPF- pNA, respectively, was measured at the selected tempera- tures.

1 Rawlings ND, Tolle DP & Barrett AJ (2004) MEROPS: the peptidase database. Nucleic Acids Res 32, Database issue, D160–164.

Effect of SDS and EDTA on activity

2 Rao MB, Tanksale AM, Ghatge MS & Deshpande VV (1998) Molecular and biotechnological aspects of micro- bial proteases. Microbiol Mol Biol Rev 62, 597–635. 3 Siezen RJ & Leunissen JA (1997) Subtilases: the super- family of subtilisin-like serine proteases. Protein Sci 6, 501–523.

4 Arnorsdottir J, Smaradottir RB, Magnusson OT,

Thorbjarnardottir SH, Eggertsson G & Kristjansson MM (2002) Characterization of a cloned subtilisin-like serine proteinase from a psychrotrophic Vibrio species. Eur J Biochem 269, 5536–5546.

5 Tsujibo H, Miyamoto K, Tanaka K, Kawai M, Tainaka K, Imada C, Okami Y & Inamori Y (1993) Cloning and sequence of an alkaline serine protease-encoding gene from the marine bacterium Alteromonas sp. strain O-7. Gene 136, 247–251.

The effect of SDS on the activity of the two peptidases (3 lgÆ mL)1 SPRK; 7.5 lgÆmL)1 PRK) was determined using 0.1, 0.25, 0.5 and 1.0% SDS (final concentrations) in the standard assay buffer, and activity towards Suc- AAPF-pNA was measured using standard assay condi- to enzyme tions. One hundred percent activity refers samples assayed using standard conditions without SDS. The effect of EDTA on activity was determined using 10 mm EDTA in the standard assay buffer without CaCl2 present. Enzyme samples were diluted 2500-fold in 25 mm to a concentration of Hepes pH 8.0 without CaCl2 3 lgÆmL)1 (SPRK) and 7.5 lgÆmL)1 (PRK) before activity towards Suc-AAPF-pNA was measured. Enzyme diluted in Hepes buffer containing 1 mm CaCl2 and standard assay conditions without EDTA was used as controls, and refers to 100% activity.

6 Kwon ST, Terada I, Matsuzawa H & Ohta T (1988)

Kinetic studies

Nucleotide sequence of the gene for aqualysin I (a ther- mophilic alkaline serine protease) of Thermus aquaticus YT-1 and characteristics of the deduced primary struc- ture of the enzyme. Eur J Biochem 173, 491–497. 7 Gunkel FA & Gassen HG (1989) Proteinase K from Tritirachium album Limber. Characterization of the chromosomal gene and expression of the cDNA in Escherichia coli. Eur J Biochem 179, 185–194.

Hydrolysis of the p-nitroanilide derivative Suc-AAPF-pNA was determined at 405 nm. All assays were performed in 0.1 m Tris ⁄ HCl pH 8.0, 0.0005% (v ⁄ v) Triton X-100, 5% (v ⁄ v) DMSO at 12, 22 and 37 (cid:1)C (pH adjusted for each temperature). Substrate concentration was in the range from 0.05 mm to 5 mm with minimum eight different

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

58

8 Bendtsen JD, Nielsen H, von Heijne G & Brunak S

(2004) Improved prediction of signal peptides: SignalP 3.0. J Mol Biol 340, 783–795.

24 Feller G & Gerday C (1997) Psychrophilic enzymes: molecular basis of cold adaptation. Cell Mol Life Sci 53, 830–841.

9 von Heijne G (1983) Patterns of amino acids near signal-sequence cleavage sites. Eur J Biochem 133, 17–21.

25 Kristjansson MM, Magnusson OT, Gudmundsson HM, Alfredsson GA & Matsuzawa H (1999) Properties of a subtilisin-like proteinase from a psychrotrophic Vibrio species comparison with proteinase K and aqualysin I. Eur J Biochem 260, 752–760.

26 Deane SM, Robb FT, Robb SM & Woods DR (1989)

10 Nielsen H, Engelbrecht J, Brunak S & von Heijne G (1997) Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng 10, 1–6.

Nucleotide sequence of the Vibrio alginolyticus calcium- dependent, detergent-resistant alkaline serine exopro- tease A. Gene 76, 281–288.

11 Izard JW & Kendall DA (1994) Signal peptides: exqui- sitely designed transport promoters. Mol Microbiol 13, 765–773.

12 Shinde U & Inouye M (2000) Intramolecular chaper-

ones: polypeptide extensions that modulate protein fold- ing. Semin Cell Dev Biol 11, 35–44.

27 Thompson JD, Higgins DG & Gibson TJ (1994) CLUS- TAL W: improving the sensitivity of progressive multi- ple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucl Acids Res 22, 4673–4680.

28 Heidelberg JF, Eisen JA, Nelson WC, Clayton RA,

13 Lee YC, Koike H, Taguchi H, Ohta T & Matsuzawa H (1994) Requirement of a COOH-terminal pro-sequence for the extracellular secretion of aqualysin I (a thermo- philic subtilisin-type protease) in Thermus thermophilus. FEMS Microbiol Lett 120, 69–74.

Gwinn ML, Dodson RJ, Haft DH, Hickey EK, Peter- son JD, Umayam L, et al (2000) DNA sequence of both chromosomes of the cholera pathogen Vibrio cholerae. Nature 406, 477–483.

14 Kim DW, Lee YC & Matsuzawa H (1997) Role of the COOH-terminal pro-sequence of aqualysin I (a heat- stable serine protease) in its extracellular secretion by Thermus thermophilus. FEMS Microbiol Lett 157, 39–45.

15 Betzel C, Pal GP & Saenger W (1988) Three-dimen-

29 Kurosaka K, Ohta T & Matsuzawa H (1996) A 38 kDa precursor protein of aqualysin I (a thermophilic subtili- sin-type protease) with a C-terminal extended sequence: its purification and in vitro processing. Mol Microbiol 20, 385–389.

30 Kim DW & Matsuzawa H (2000) Requirement for

sional structure of proteinase K at 0.15-nm resolution. Eur J Biochem 178, 155–171.

the COOH-terminal pro-sequence in the translocation of aqualysin I across the cytoplasmic membrane in Escherichia coli. Biochem Biophys Res Commun 277, 216–220.

31 Denkin SM & Nelson DR (2004) Regulation of

Vibrio anguillarum empA metalloprotease expression and its role in virulence. Appl Environ Microbiol 70, 4193–4204.

32 Miyoshi S, Wakae H, Tomochika K & Shinoda S

16 Arnorsdottir J, Kristjansson MM & Ficner R (2005) Crystal structure of a subtilisin-like serine proteinase from a psychrotrophic Vibrio species reveals structural aspects of cold adaptation. FEBS J 272, 832–845. 17 Kwon ST, Matsuzawa H & Ohta T (1988) Determina- tion of the positions of the disulfide bonds in aqualysin I (a thermophilic alkaline serine protease) of Thermus aquaticus YT-1. J Biochem 104, 557–559.

18 Teplyakov AV, Kuranova IP, Harutyunyan EH,

(1997) Functional domains of a zinc metalloprotease from Vibrio vulnificus. J Bacteriol 179, 7606–7609. 33 Mitchinson C & Wells JA (1989) Protein engineering of

Vainshtein BK, Frommel C, Hohne WE & Wilson KS (1990) Crystal structure of thermitase at 1.4 A resolu- tion. J Mol Biol 214, 261–279.

disulfide bonds in subtilisin BPN¢. Biochemistry 28, 4807– 4815.

19 Morihara K & Tsuzuki H (1975) Specificity of Protei- nase K from Tritirachium album Limber for synthetic peptides. Agric Biol Chem 39, 1489–1492.

20 Ebeling W, Hennrich N, Klockow M, Metz H, Orth

HD & Lang H (1974) Proteinase K from Tritirachium album Limber. Eur J Biochem 47, 91–97.

34 Takagi H, Takahashi T, Momose H, Inouye M, Maeda Y, Matsuzawa H & Ohta T (1990) Enhancement of the thermostability of subtilisin E by introduction of a dis- ulfide bond engineered on the basis of structural com- parison with a thermophilic serine protease. J Biol Chem 265, 6874–6878.

35 Bajorath J, Saenger W & Pal GP (1988) Autolysis and inhibition of proteinase K, a subtilisin-related serine proteinase isolated from the fungus Tritirachium album Limber. Biochim Biophys Acta 954, 176– 182.

21 Hilz H, Wiegers U & Adamietz P (1975) Stimulation of proteinase K action by denaturing agents: application to the isolation of nucleic acids and the degradation of ‘masked’ proteins. Eur J Biochem 56, 103–108. 22 Herrmann BG & Frischauf AM (1987) Isolation of genomic DNA. Methods Enzymol 152, 180–183.

23 Sweeney PJ & Walker JM (1993) Proteinase K. Methods

36 Narinx E, Baise E & Gerday C (1997) Subtilisin from psychrophilic antarctic bacteria: characterization and

Mol Biol 16, 305–311.

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

59

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme

site-directed mutagenesis of residues possibly involved in the adaptation to cold. Protein Eng 10, 1271– 1279.

38 Chen WP & Kuo TT (1993) A simple and rapid method for the preparation of gram-negative bacterial genomic DNA. Nucl Acids Res 21, 2260.

39 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248–254.

37 Helland R, Larsen AN, Smalas AO & Willassen NP (2006) The 1.8 A˚ crystal structure of a proteinase K- like enzyme from a psychrotroph Serratia species. FEBS J 273, 61–71.

FEBS Journal 273 (2006) 47–60 ª 2005 The Authors Journal compilation ª 2005 FEBS

60

A. N. Larsen et al. Characterization of a Serratia proteinase K-like enzyme