New b-Lactamase Inhibitory Protein ( BLIP-I ) from Streptomyces exfoliatus SMF 19 and Its Roles on the Morphological Differentiation *

A new b-lactamase inhibitory protein (BLIP-I) from Streptomyces exfoliatus SMF19 was purified and characterized. The molecular mass of BLIP-I was estimated to be 17.5 kDa by gel filtration fast protein liquid chromatography. The N-terminal sequence was NH2-Asn-SerGly-Phe-Ser-Ala-Glu-Lys-Tyr-Glu-Gln-Ile-Gln-Phe-Gly. BLIP-I inhibited Bacto® Penase (Difco), and plasmid encoded TEM-1 b-lactamase, whereas it did not inhibit Enterobacter cloacae b-lactamases. The Ki value of BLIP-I against TEM-1 b-lactamase was determined to be 0.047 nM. The gene (bliA) encoding BLIP-I protein was identified by screening a genomic library using an oligonucleotide probe with a sequence based on the Nterminal sequence of BLIP-I. Analysis of the nucleotide sequence revealed that the gene was 558 base pairs in length and encoded a mature protein of 157 amino acid residues preceded by a 29-amino acid signal sequence. Pairwise comparison of the deduced amino acid sequence showed 38% identity with BLIP of Streptomyces clavuligerus. Furthermore, the 49th amino acid residue of BLIP-I was identical to Asp-49 of BLIP that was characterized to be an important residue for the inhibitory activity of BLIP. A modified BLIP-I in which Asp-49 was replaced by alanine (D49A) was obtained by site-directed mutagenesis. The inhibitory activities of recombinant (r) BLIP-I and its D49A mutant derivative, expressed in Escherichia coli, were compared. The Ki value of rBLIP-I against TEM-1 b-lactamase was similar to that of wild-type BLIP-I, but the D49A mutation increased the Ki of rBLIP-I inhibition approximately 200fold. A disruption mutant of the bliA gene in S. exfoliatus SMF19 was obtained by replacing the wild-type bliA gene with a copy inactivated by inserting a hygromycin resistance gene. The disruption mutant showed a bald phenotype, indicating that the bliA gene plays a role in morphological differentiation.

␤-Lactam antibiotics including penicillins and cephalosporins inhibit the synthesis of the bacterial peptidoglycan layer (1).Widespread use of ␤-lactam antibiotics has led to the evolution of ␤-lactamase-mediated resistance in bacteria, which is now a serious threat to antibiotic therapy (2,3).
The incentive to maintain the efficacy of the classic ␤-lactams is great because they are still the antibiotics of choice in terms of having minimal side effects and a low cost of production (2).In addition to developing new ␤-lactams unable to be cleaved by ␤-lactamases, developing new potent inhibitors of ␤-lactamase is another strategy to circumvent ESBL-mediated inactivation of ␤-lactam antibiotics.Although ␤-lactamase inhibitors such as clavulanic acid and tazobacta have been used clinically (11), their wide application is somewhat hindered by the emergence of organisms bearing inhibitor-resistant ␤-lactamases (7,8,(12)(13)(14).
A protein showing an inhibitory activity against Staphylococcus aureus ␤-lactamase was isolated from Streptomyces gedaensis (15).Thereafter, BLIPs from Streptomyces clavuligerus (16) and BLIP-II from Streptomyces exfoliatus SMF19 (17,18) were identified.BLIP from S. clavuligerus was a potent inhibitor of TEM-1 ␤-lactamase (K i ϭ 0.6 nM), and the crystal struc- ture of BLIP was determined, as well as the co-crystal with TEM-1 ␤-lactamase (19,20).It was reported that Asp-49 and Phe-142 of BLIP contribute to the inhibitory activity by interacting with Ser-70 of TEM-1 ␤-lactamase and mimicking the benzyl of penicillin G, respectively (21).BLIP-II and the gene (bliB) encoding BLIP-II were recently identified from S. exfoliatus SMF19, but the deduced amino acid sequence did not show similarity to that of BLIP (18).The eventual understanding of the inhibitory activity of BLIPs may lead to the development of new strategies for coping with the clinical occurrence of ␤-lactam antibiotic resistance.Although both of the BLIPs showed in vitro inhibitory activity against various ␤-lactamases, the biological roles of BLIP and BLIP-II are not well understood.Pairwise comparisons of the amino acid sequence of the two BLIPs did not show any evolutionary or structural relationship with other proteins (16,18).This report describes purification, characterization, and cloning of a gene (bliA) encoding the new ␤-lactamase inhibitory protein (BLIP-I) from S. exfoliatus SMF19.The bliA gene was also expressed at high levels in Escherichia coli to facilitate mutagenic study and to produce a large amount of functional BLIP-I.The contribution of certain amino acid residues to the inhibitory activity of TEM-1 ␤-lactamase was shown by site-specific mutagenesis and kinetic measurement.In order to see the biological importance of bliA in the production strain, an insertional inactivation mutation of bliA was made, and the phenotypic changes were demonstrated.
Strains of E. coli were grown in LB medium containing appropriate antibiotics for plasmid or cosmid maintenance (29).
Purification of BLIP-I-The culture broth from a 48-h culture of S. exfoliatus SMF19 was collected by centrifugation at 8,000 ϫ g for 20 min.Ammonium sulfate was added to the culture broth up to 60% saturation, and the precipitate was removed by centrifugation at 12,000 ϫ g for 20 min and discarded.Ammonium sulfate was then added to the supernatant to 80% saturation, and the resulting precipitate was collected by centrifugation at 12,000 ϫ g for 20 min.The protein precipitate was resuspended in 5 mM potassium phosphate buffer, pH 7.0, and dialyzed overnight against the same buffer.
The dialysate was loaded onto a 5 ϫ 20 cm hydroxyapatite column (Bio-Rad) equilibrated with 5 mM potassium phosphate buffer, pH 7.0, and eluted with a linear gradient of 5-300 mM potassium phosphate buffer, pH 7.0.Active fractions for ␤-lactamase inhibitory activity were pooled and concentrated by ultrafiltration with a PM-10 membrane (Amicon).
The sample obtained by hydroxyapatite chromatography was loaded onto a 5 ϫ 20-cm DEAE-Sephadex A-50 column (Sigma) that was equilibrated with 20 mM Tris-HCl buffer, pH 7.8, and the column was eluted with a linear gradient of 0 -1.0 M NaCl.Active fractions for ␤-lactamase inhibitory activity were pooled and concentrated by ultrafiltration with a PM-10 membrane (Amicon).
The concentrated active fractions were applied to a Superdex 75 column (Amersham Pharmacia Biotech) and eluted with 5 mM potassium phosphate buffer, pH 7.0, and then the pooled active fractions from Superdex 75 column were loaded onto a Mono Q column (Amersham Pharmacia Biotech) equilibrated with 20 mM Tris-HCl buffer, pH 7.8.The column was eluted with a linear gradient of 0 -1 M NaCl.The active fractions were pooled and loaded onto a Resource-PHE column (Amersham Pharmacia Biotech) equilibrated with 50 mM phosphate buffer, pH 7.0, supplemented with 40% (NH 4 ) 2 SO 4 .The column was eluted with a linear reverse phase gradient of 40 to 0% (NH 4 ) 2 SO 4 .Active fractions were desalted using a gel permeation chromatography column and then freeze-dried.
Characterization of Purified BLIP-I-The native molecular weight of BLIP-I was estimated using a Superose 12 FPLC gel filtration column equilibrated with 50 mM potassium phosphate buffer, pH 7.0.The column was calibrated using Bio-Rad gel filtration standards.Protein concentration was analyzed by the Bradford method (30).Purity of BLIP-I samples was assessed by SDS-PAGE, which was performed according to the method of Laemmli (31).
For N-terminal sequencing, the purified BLIP-I was electrophoresed on Tricine/SDS-PAGE and electroblotted onto polyvinylidene difluoride membrane (Millipore) in transfer buffer A (10 mM CAPS and 10% methanol, pH 11.0).The N-terminal sequence was determined by the method of Edman degradation using a MilliGen/Biosearch ProSequencer 6600 (Millipore).
Cloning of the bliA Gene-General methods for DNA manipulations in Streptomyces spp.and E. coli were performed as described by Hopwood et al. (23) and Sambrook et al. (29).The genomic library of S. exfoliatus SMF19 was prepared as described (18).
A 45-mer, mixed oligonucleotide probe (5Ј-AATTC(G/C)GGCT-TCTC(G/C)GC(G/C)GAGAA GTACGAGCAGATCCAGTTCGGC-3Ј) was designed based on the sequence of the N-terminal 15 amino acids of BLIP-I (Genosys).E. coli colonies containing recombinant cosmids were screened for hybridization to the oligonucleotide probe.The probe was labeled with the ECL 3Ј-oligolabeling kit (Amersham Pharmacia Biotech).Southern hybridization, colony hybridization, and physical mapping of the selected recombinant plasmid were performed using standard techniques (29).
Both DNA strands of the insert of the selected recombinant plasmid were sequenced by the dideoxy method of Sanger et al. (32) using Sequenase 2 and 7-deaza-GTP (Promega) in place of dGTP to relieve compressions.The DNA sequence of each fragment was determined repeatedly to provide unequivocal results in regions of high G ϩ C content.The sequences of DNA fragments were assembled into a complete sequence using the DNASIS program (Hitachi).The nucleotide sequence data were also analyzed with the DNASIS program to identify restriction sites and regions of dyad symmetry.The location of the open reading frame was confirmed with the Frameplot program (33,34).
Construction of a Recombinant Plasmid for Expressing bliA-To produce functional BLIP-I in E. coli, an expression system was developed using pET-30a(ϩ) vector (Novagen) that has an N-terminal His tag and enterokinase cleavage site.The structural gene of bliA encoding mature BLIP-I was first amplified by PCR.The primers are BLIPEXP1 (a top strand primer), 5Ј-GT CGCGGGTACCGACGACGACGACAAGA-ATTCGGGCTTTTCGGCC-3Ј, and BLIPEXP2 (a bottom strand primer), 5Ј-GGCGTCGGATCCTCAGGTCAGGCTGCGCTGGTAGCGGTA CGTCAG-3Ј.The primers were designed to incorporate KpnI and BamHI sites, respectively.The KpnI/BamHI-digested PCR fragment was inserted between KpnI and BamHI sites of pET-30a(ϩ), creating pSMF1130.The positioning of bliA in pSMF1130 allowed the gene to be expressed by induction of the T7lac promoter with IPTG, and the His 6 tag facilitated the purification of BLIP-I using an appropriate nickel-or cobalt-based affinity column.The mature BLIP-I protein can then be obtained by using enterokinase cleavage.
Construction of a Recombinant Plasmid for Expressing the D49A Mutant Form of bliA-A recombinant plasmid to produce the D49A mutant form of BLIP-I was constructed.Site-directed mutagenesis of Asp-49 to Ala was accomplished using the Altered Sites II in vitro mutagenesis system (Promega).An EcoRI-SacI fragment of pSMF1130 containing the Asp-49 coding region was cloned into pAlter-1 (pSMF11302), and then site-specific mutagenesis of Asp to Ala was carried out as described in the technical manual.The mutagenic oligonucleotide used was 5Ј-GAGTCGGGCGCCTACGCCCCC-3Ј (underlined was the NarI site).Plasmid DNA was isolated from candidate mutant clones and screened by digestion with NarI, incorporated in the mutagenic oligonucleotide.The corresponding codons of NarI site in the wild-type bliA gene were GGCGCG, not digested with NarI.The plasmid carrying presumptive mutant bliA gene was named pSMF11307.The EcoRI-SacI fragment from pSMF11307 was cloned into EcoRI/ SacI-digested pSMF1130 (named pSMF11308), and then the SacI fragment of bliA purified from pSMF1130 was inserted into SacI-digested pSMF11308 in the proper orientation (to give pSMF11309).Finally, BglII-BamHI fragments of pSMF1130 and pSMF11309 were cloned into pUC18, and their sequences were confirmed by the dideoxy chain termination method.
Production and Purification of rBLIP-I and the D49A Mutant-The plasmids pSMF1130 and pSMF11309 were transformed into E. coli BL21(DE3)pLysS.Overnight culture of each was grown with shaking in 40 ml of Luria-Bertani (LB) medium at 37 °C in the presence of 34.5 g/ml chloramphenicol and 50 g/ml kanamycin.The 40-ml overnight cultures were used to inoculate 2-liter amounts of LB medium containing 34.5 g/ml chloramphenicol and 50 g/ml kanamycin.The cultures were then grown with shaking at 37 °C until A 600 ϭ 0.5-0.7.For induction, IPTG was added to each culture to a final concentration of 1 mM, and the cultures were then allowed to grow an additional 3 h.Following the 3-h induction, the cells were harvested and resuspended in 50 ml of ice-cold binding buffer (5 mM Tris-HCl, pH 8.0, and 500 mM NaCl).The cells were then sonicated to disrupt the cells and shear chromosomal DNA, and insoluble material was pelleted by centrifugation.The insoluble material was resuspended in 50 ml of binding buffer containing 6 M urea, and any residual insoluble material was removed by ultracentrifugation after 1-h incubation at 4 °C.The supernatant was purified using a 1-ml HisTrap column (Amersham Pharmacia Biotech) under denaturing conditions according to the manufacturer's instructions.Binding buffer containing 10 mM imidazole was utilized to remove loosely bound protein from the column.His-tagged rBLIP-I and His-tagged D49A mutant were eluted using binding buffer containing 200 mM imidazole, pH 8.0.The purified His-tagged proteins were cleaved by enterokinase (Novagen) to remove the His tag, and then rBLIP-I and D49A mutant were purified by a Mono Q chromatography.Fractions were examined by SDS-PAGE to estimate purity and yield.
␤-Lactamase Inhibitory Activity Assay-␤-Lactamase inhibitory activity of culture broth or samples collected during purification was analyzed at pH 7.0 and 37 °C by the modified iodometric assay method (35).␤-Lactamase inhibitory activity was calculated as follows: ␤-lactamase inhibitory activity (%) ϭ ((A Ϫ B)/A) ϫ 100, where A is the ␤-lactamase activity without the inhibitor, and B is the ␤-lactamase activity with the inhibitor.One unit of ␤-lactamase inhibitory activity was defined as the amount of inhibitor needed for 50% inhibition of 10,000 IU of Bactopenase (Difco).
Kinetic parameters of BLIP-I, rBLIP-I, and the D49A mutant were determined by fitting the data to the equation of Dennis and co-workers (21,36) or Lineweaver-Burk plot.Varying concentrations of BLIP-I were incubated with 1 nM preparations of various ␤-lactamases for 2 h at 25 °C in 0.05 M phosphate buffer, pH 7.0.Following the 2-h incubation, cephaloridine was added to a concentration of 10 -200 M. The final volume for the reaction was 2.5 ml.Hydrolysis of cephaloridine was monitored at A 260 on a Shimazu UV-160 spectrophotometer.The extinction coefficient used for cephaloridine was ⌬⑀ ϭ 10,200 M Ϫ1 cm Ϫ1 (37).Plots of the concentration of free ␤-lactamase versus inhibitor concentration were fitted by nonlinear regression analysis to the equation.From the equation, apparent equilibrium dissociation constants (K i *) were determined (21,36).
Insertional Inactivation of BLIP-I-A bliA::hyg mutant of S. exfoliatus SMF19 was constructed by inserting a hygromycin resistance (hyg) gene into the middle of bliA.A 2-kb KpnI-XhoI fragment from pSMF1101 containing the complete bliA gene together with 800 bp of upstream and 800 bp of downstream sequence was subcloned into pZErO-2, creating pSMF11041.The hyg gene from pJOE829 was amplified by PCR.The primers are BLIPDIS1 (a top strand primer), 5Ј-G-TCGCACAATTGGGCGGTGGCGTACACCGT-3Ј, and BLIPDIS2 (a bottom strand primer), 5Ј-GAGCTGCAATTGAAGCTCGGCCGACCACCC-3Ј, which are designed to contain an MfeI site at the ends of the amplified PCR fragment.The resulting PCR product was digested with MfeI and then was inserted into the EcoRI site of pSMF11041, creating pSMF11042.A 3.5-kb KpnI-SphI fragment from pSMF11042 was ligated to KpnI/SphI-digested pIJ702, and then transformed into S. lividans.A transformant able to grow in TSB medium containing hygromycin (50 g ml Ϫ1 ) and thiostrepton (50 g ml Ϫ1 ) was selected.The plasmid from the colony, pSMF11043, was purified and introduced into S. exfoliatus SMF19 by transformation.A disruption mutant defective in bliA gene resulting from gene conversion by homologous recombination was isolated as described by Aidoo et al. (38).The transformant harboring pSMF11043 was cultured in minimal medium containing only hygromycin (50 g ml Ϫ1 ) for 5 rounds, and then the recombinants of Hyg r and thiostrepton-sensitive (Thio s ) colonies unable to grow in the minimal medium containing thiostrepton (50 g ml Ϫ1 ) were selected as disruption mutants.The insertional inactivation of bliA was confirmed by Southern hybridization using the 1.2-kb EcoRI-KpnI fragment labeled with the ECL direct nucleic acid labeling system as a probe.

Purification and Characterization of BLIP-I-A new ␤-lacta-
mase inhibitory protein (BLIP-I) was purified from the culture supernatant of S. exfoliatus SMF19.Final recovery of BLIP-I upon complete purification was 29.8% with a purification fold of 36.1 (Table I).The mass was estimated to be 17.5 kDa by gel filtration FPLC, and the N-terminal amino acid sequence of the purified BLIP-I was determined to be NH 2 -Asn-Ser-Gly-Phe-Ser-Ala-Glu-Lys-Tyr-Glu-Gln-Ile-Gln-Phe-Gly.Purified BLIP-I from S. exfoliatus SMF19 was clearly separable from BLIP-II, indicating that S. exfoliatus SMF19 produced two different BLIPs (Fig. 1, A and B).
The inhibitory activity of BLIP-I against various types of ␤-lactamases was determined (Table II).Plasmid-encoded TEM-1 ␤-lactamase and Bacillus cereus ␤-lactamases were clearly inhibited by BLIP-I, whereas ␤-lactamases of Enterobacter cloacae were not inhibited.The K i value of BLIP-I against TEM-1 ␤-lactamase was determined to be 0.047 nM (Table II and Fig. 2), which is comparable to the potency of BLIP from S. clavuligerus (19,21).
Cloning and Analysis of BLIP-I Gene-Among 2,000 E. coli colonies picked as a genomic library, pSMF1101 was identified as a recombinant cosmid carrying the bliA gene by colony hybridization and Southern hybridization.The cosmid was digested with various enzymes, and a BamHI-ApaI 1.5-kb fragment that hybridized to the probe was subcloned into the plasmid, pZErO-2, to give the recombinant plasmid pSMF1106.
The nucleotide sequence of bliA encoding BLIP-I was determined and is available from GenBank (accession number AF201389).The correct reading frame was initially recognized by the presence of a DNA sequence that corresponded precisely to the N-terminal amino acid sequence of BLIP-I.By analysis using the DNASIS program and Frameplot program (33,34), the GTG codon (nucleotide 496) was identified as the most likely initiation codon because it is preceded by a potential ribosome-binding site (nucleotide 485, GAAGGA; 39) and followed by a typical signal sequence observed in other secretory proteins of Streptomyces spp.(16).The open reading frame apparently terminated at a TGA codon at nucleotide 1054.Its overall G ϩ C content was 66.5 mol %; the values for codon positions 1-3 were 55.6, 51.3, and 92.5 mol %, respectively.
Based on these predicted start and stop codons, bliA (558 bp) encodes a protein of 186 amino acids.The deduced precursor a One unit of BLIP-I activity is defined as that amount of material that gives 50% inhibition of 10,000 IU Bactopenase used in the standard spectrophotometer ␤-lactamase inhibition assay.
protein has a 29-amino acid leader peptide and gives a mature protein of 157 amino acid residues, of which 12.7% are acidic, 8.9% are basic, and 49% are hydrophobic.The calculated molecular mass of the mature form of BLIP-I is 17,319 Da which corresponds closely to that estimated by molecular mass determination (17.5 kDa).The leader sequence of the BLIP-I protein precursor contains three positively charged amino acids, two arginine residues at positions 5 and 8, and one lysine residue at position 3, followed by a long hydrophobic stretch (residues 9 -24).The composition of leader sequence was similar to those of other secreted proteins from Streptomyces spp.including BLIP of S. clavuligerus (16, 18, 40 -42).
A BLAST search of the deduced amino acid sequence of bliA showed 38% identity to BLIP of S. clavuligerus but no significant similarities to BLIP-II of S. exfoliatus SMF19.bliA and bliB are not adjacent on the chromosome since the recombinant cosmids containing bliA did not hybridize to a probe specific for BLIP-II and vice versa.This indicates that S. exfoliatus SMF19 produces two different ␤-lactamase inhibitory proteins.The alignment of the deduced amino acid sequence of BLIP-I with BLIP is shown in Fig. 3. Identical amino acids were distributed throughout the molecule.Of particular note, the 48 -50 residues in the amino acid sequence of BLIP-I are GDY, and this sequence is identical to the ␤-hairpin of BLIP (boxed in Fig. 3) which contributes to the inhibitory activity.However, residues next to the loop were not identical.
Production of Wild-type BLIP-I and the D49A Mutant-The recombinant plasmid, pSMF1130 was introduced into E. coli BL21(DE3)pLysS, and expression of bliA was induced with IPTG.From the SDS-PAGE patterns of clarified cell sonicates, it was apparent that a 30-kDa protein was produced in large amounts only in the culture of pSMF1130 with IPTG present.The protein was detected in the insoluble fraction, indicating that the protein accumulated in inclusion bodies.The insoluble protein was solubilized with 6 M urea and then purified under denaturing conditions using a His-trap column.The purified His-tagged protein was easily refolded by stepwise removal of the urea, and then the soluble mature rBLIP-I was purified by removing the His tag with enterokinase cleavage (Fig. 4).Approximately 10 mg of Ͼ90% pure BLIP-I could be isolated for every 2 liters of culture using this strategy.
The alignment of the deduced amino acid sequence of BLIP and BLIP-I revealed that the Asp-49 residue of BLIP-I was identical to that of BLIP.The crystal structure of BLIP complexed with TEM-1 ␤-lactamase showed that Asp-49 of BLIP makes strong hydrogen bond contacts with four conserved residues in the TEM-1 active site pocket as follows: Ser-130, Lys-234, Ser-235, and Arg-244 (20).The importance of Asp-49 was confirmed by mutagenic studies (21).Thus, site-specific mutagenesis was carried out to examine the contribution of Asp-49 to the inhibitory activity in BLIP-I.A mutagenic recombinant plasmid for production of the D49A mutant form of BLIP-I was constructed and expressed as described.The D49A mutant protein was also detected in the insoluble fraction and purified under denaturing conditions by the same method used for rBLIP-I purification.The D49A mutant was purified to Ͼ90% homogeneity.The purified D49A mutant form of BLIP-I migrated a little faster than rBLIP-I in SDS-PAGE, presumably due to the single amino acid change (Fig. 4).
Kinetic Analysis of rBLIP-I and the D49A Mutant Form of BLIP-I-A comparison of K i values between wild-type BLIP-I and rBLIP-I was carried out by fitting the data to the nonlinear equation ( 21) in order to determine whether the expression system affects the inhibitory activity of BLIP-I.The K i values of

TABLE II
The inhibition kinetics between BLIP-I and various ␤-lactamases K i value of BLIP-I against TEM-1 is obtained by fitting to the equation ( 21), and the other K i values against various ␤-lactamases are obtained by linear least squares fitting to Lineweaver-Burk plot.

␤-Lactamase
Source wild-type BLIP-I and rBLIP-I were determined to be 0.047 and 0.062 nM, respectively (Fig. 2 and Table III).The results indicate that the heterologous expression system used in this study is effective for the production of the functional BLIP-I.Interestingly, the inhibitory activity of purified His-tagged rBLIP-I was determined to be 0.82 nM (Fig. 2 and Table III), which represents a 10-fold increase in K i value compared with rBLIP-I of TEM-1 ␤-lactamase.This suggests that the His-tagged signal sequence composed of 42 amino acids does have some effect, but the tagged precursor rBLIP-I form still showed the inhibitory activity with the K i value in the nanomolar range.
To examine the contribution of Asp-49 to BLIP-I inhibitory activity, the K i value of purified D49A mutant of BLIP-I was compared with that of rBLIP-I.The K i value of the D49A mutant was determined to be 10 nM, an approximately 200-fold increase in K i value as compared with that of rBLIP-I.This indicates that Asp-49 of BLIP-I was important for the inhibitory activity.The K i value of the His-tagged precursor form of D49A mutant was determined to be 32 nM, which means that the His-tagged signal sequence also had some effect in the inhibitory activity in the mutant.However, the reduction in inhibitory activity was minimal, considering the size of the tag (42 amino acids).
The Biological Importance of BLIP-I in S. exfoliatus SMF19 -A bliA::hyg mutant of S. exfoliatus SMF19, called SMF110451, was developed to investigate the biological importance of bliA.The disruption plasmid construct, pSMF11043, was introduced into S. exfoliatus SMF19 by transformation.After 5 rounds of culture in minimal medium supplemented with hygromycin (50 g ml Ϫ1 ), 24 independent isolates show- ing Hyg r and Thio s phenotype were selected.To confirm the insertional inactivation of bliA in SMF110451, Southern hybridization of genomic DNAs purified from SMF19 and SMF110451 was carried out.The 4.6-kb PstI fragment of S. exfoliatus SMF19 was replaced by a 6.1-kb PstI fragment in SMF110451 (Fig. 5).These results indicated that the 1.5-kb hyg was successfully inserted into the middle of bliA gene in the chromosome of S. exfoliatus SMF19.The biological effect of bliA inactivation in S. exfoliatus was investigated by culturing the wild-type strain and SMF110451 on minimal medium agar plates (Fig. 6A).The difference in the morphological differentiation was most striking.The wild-type strain showed typical differentiation process.After 2 days of incubation, colony surfaces were covered with spore chains.On the other hand, SMF110451 showed a bald appearance, and the surfaces of the colonies showed only the presence of substrate mycelium even after 2 weeks.The morphological differentiation of SMF110451 was restored by extracellular complementation; the bliA::hyg mutant showed the sporulation in the neighboring region of the wild-type strain (Fig. 6B).These results indicate that BLIP-I may function in the morphological differentiation process and represent the first report showing the biological importance of a ␤-lactamase inhibitory protein.

DISCUSSION
The development of novel inhibitors for ␤-lactamases would provide new options for the treatment of bacterial infections.Furthermore, understanding how ␤-lactamase inhibitory proteins interact with ␤-lactamases could facilitate the development of novel inhibitors active against ESBLs.
In this work, we have purified and cloned a new ␤-lactamase inhibitory protein (BLIP-I), initially thought to be 44 kDa (17), but now clearly identified as a 17.5-kDa protein.BLIP-I was different from BLIP-II identified from the same strain (18), and pairwise comparison of deduced amino acid of BLIP-I showed similarity to BLIP.An expression system was developed to produce functional BLIP-I and identify amino acid residues important for its BLIP-I inhibitory activity.By using this expression system, His-tagged rBLIP-I could be purified to Ͼ90% homogeneity in one step with a yield of 40 mg of His-tagged rBLIP-I from every 2 liters of culture.The K i value of rBLIP-I compared well with that of BLIP-I, although the K i value of His-tagged rBLIP-I indicated a 10-fold reduction in the inhibitory activity compared with BLIP-I.This indicates that the expression system developed in this study is effective in expressing bliA to yield large amounts of functional BLIP-I.
The identification of amino acids responsible for inhibition and those critical for binding specificity will pinpoint residues that could be targeted for engineering BLIP-I mutants with higher inhibitory activity for different ␤-lactamases.Amino acid residues 48 -50 were found to be identical to the inhibition hairpin of BLIP, and the D49A mutant of BLIP-I reduced the inhibitory activity of BLIP-I approximately 200-fold.This indicates that as is the situation with BLIP, Asp-49 does make an important contribution to the inhibition activity in BLIP-I, and the inhibition hairpin structure may be conserved in BLIP-I.
On the other hand, the mutagenic study of BLIP showed that Phe-142 is important for inhibition of TEM-1 ␤-lactamase (21).However, it is not easy to pinpoint the residue in BLIP-I that is equivalent to Phe-142 of BLIP.Perhaps, Phe-135 and Phe-141 of BLIP-I could be important.X-ray crystallographic studies on the interaction between BLIP-I and ␤-lactamases will be necessary to identify the portions of the molecule critical for the inhibition and will also contribute to additional information to facilitate the engineering of tighter, smaller inhibitors for these ␤-lactamases to circumvent widespread bacterial resistance.
The biological significance of ␤-lactamase inhibitory proteins remained uncertain (16,18).In this work, the insertional inactivation of bliA conferred a bald phenotype on S. exfoliatus SMF19, which indicates that BLIP-I may be involved in the morphological differentiation process.The morphological differentiation of Streptomyces has been defined genetically by the isolation of bld and whi mutants (43,44).BLIP-I may represent a new locus for differentiation since pairwise comparison of the deduced amino acids sequence of BLIP-I did not show any similarity to bld and whi genes.However, the precise way that BLIP-I might be involved in differentiation is not clear at the moment.One hypothesis is that it functions as a regulator of cell wall synthesis by interacting with penicillin-binding proteins.Studies exploring this possibility and aimed at elucidating the regulatory mechanism are now in progress.

aFIG. 2 .FIG. 3 .
FIG. 2. The determination of K i value of rBLIP-I and D49A mutant against TEM-1 ␤-lactamase.BLIP inhibitory activity is expressed as the remaining concentration of free ␤-lactamase at varying inhibitor concentrations.TEM-1 concentration is 1 nM, and cephaloridine concentration is 70 M for all experiments.The lines represent the nonlinear regression fit of the data to the equation (21, 36) to calculate the K i .Open circles, wild-type BLIP-I; filled circles, rBLIP-I; filled squares, His-tagged rBLIP-I; open squares, D49A; and filled triangles, His-tagged D49A.

TABLE I
Purification of BLIP-I

TABLE III
(21)determination of K i values of BLIP-I and D49A mutant against TEM-1 All values are expressed in nM.Error limits are the standard deviation of the parameter values obtained by non-linear least squares fitting to the equation(21).