HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 28, 21002-21009, July 14, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/28/21002    most recent M002088200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Feltzer, R. E. Articles by Pierce, W. M. Search for Related Content PubMed PubMed Citation Articles by Feltzer, R. E. Articles by Pierce, W. M., Jr. Social Bookmarking                      What's this?

Alkaline Proteinase Inhibitor of Pseudomonas aeruginosa

INTERACTION OF NATIVE AND N-TERMINALLY TRUNCATED INHIBITOR PROTEINS WITH PSEUDOMONAS METALLOPROTEINASES^*

Rhona E. Feltzer^§, Robert D. Gray^¶, William L. Dean, and William M. Pierce Jr.^¶**

From the Departments of  Biochemistry and Molecular Biology, ^** Pharmacology and Toxicology, and ^¶ Ophthalmology and Visual Sciences, University of Louisville School of Medicine, Louisville, Kentucky 40292

Received for publication, March 13, 2000, and in revised form, April 12, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The apr locus of Pseudomonas aeruginosa encodes alkaline proteinase (APR), a member of the metzincin metalloendopeptidase superfamily, and an 11.4-kDa alkaline proteinase inhibitor (APRin). We describe here the expression in Escherichia coli and characterization of full-length and N-terminally truncated APRin proteins. Fluorescence and circular dichroism spectra indicated that the recombinant proteins were folded into native-like structures. Analytical ultracentrifugation showed that APRin was monomeric and formed a 1:1 complex with APR. Binding of wild-type APRin to APR occurred with association (k_on) and dissociation (k_off) rate constants of 0.29 ± 0.06 × 10⁶ M¹ s¹ and 1.15 ± 0.08 × 10⁶ s¹ to give an equilibrium dissociation constant (K_D) of ~4 × 10¹² M (25 °C, pH 7.0, ionic strength 2.4 M). The association rate decreased by ~2-fold in 20% glycerol and increased by ~3-fold in 0.1 M NaCl. The glycerol effect suggests a diffusion-limited reaction, and the small salt effect indicates that electrostatic interactions contribute little to binding. Deletion of residues 1-10, 1-6, or 6-10 abolished inhibition, and deletion of residues 1-2, 1-3, 1-4, and 1-5 resulted in a progressively decreased affinity of APRin for APR (K_D = 0.12 µM for the (1-5) mutant). Substitution of APRin residues 6-10 with a (Gly)₅ or (Pro)₅ linker restored inhibitory activity of the (6-10) mutant but with a 100- and 50-fold reduction in K_D. Log k_on for the full-length and truncated inhibitors correlated with the solvent-accessible surface area of their N-terminal regions, suggesting that increased interactions and/or desolvation of these residues in the transition state for binding contribute to the enhanced association rate. Treatment of APRin with pseudolysin, also secreted by P. aeruginosa, resulted in removal of residues 1-5. APRin was neither an inhibitor nor a substrate of other metzincins, including collagenase or gelatinases A or B.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Pseudomonas aeruginosa is a ubiquitous microorganism that is usually benign with respect to healthy individuals but pathogenic in a number of clinical settings, including surgical patients, burn victims, and persons with cystic fibrosis (1). Eye infection by P. aeruginosa is relatively common, particularly in wearers of contact lenses (2); if not promptly treated, these infections can result in corneal ulceration and visual impairment. Among the virulence factors of P. aeruginosa are two secreted metalloproteinases (3), elastase (pseudolysin, PL)¹ and alkaline proteinase (aeruginolysin, APR). Substrates for PL include extracellular matrix structural proteins, immunoglobulins, complement components, and ₁-proteinase inhibitor (3). APR degrades fibrinogen, fibrin (4), laminin (5), and -interferon (6); it also is capable of activating Hageman factor (7) and matrix metalloproteinases (MMPs) (8).

APR and PL were initially characterized by Morihora (9, 10). Subsequent structural comparison revealed that PL is related to thermolysin of Bacillus thermolyticus, and APR is homologous to the 50-kDa metalloproteinases secreted by Serratia marcescens and Erwinia chrysanthemi (11, 12). These enzymes constitute the serralysin branch of the metzincin superfamily of metalloendopeptidases (13). X-ray crystallography of APR (14) and SMP (15), the Serratia homolog of APR, shows that both consist of an N-terminal catalytic domain of about 230 residues and a C-terminal calcium binding domain of approximately 240 residues that appears to be required for proteinase secretion (16). The catalytic domain contains sequences characteristic of the metzincin superfamily of metalloendopeptidases, namely a zinc-binding motif (HEXXHXXGXXH) and a conserved Met located in a turn near the base of the metal binding pocket. Other metzincins include astacins, MMPs, reprolysins, and the adamalysins (12).

Serralysin secretors are also capable of producing specific inhibitory proteins of about 125 residues that are transported to the periplasm where the signal peptide is removed (17, 18). The three known members of this inhibitor family exhibit approximately 20% sequence identity and an additional 20% sequence similarity, including a conserved disulfide bond (19). The mature inhibitors from E. chrysanthemi and P. aeruginosa have N-terminal Ser, whereas the N terminus of the S. marcescens inhibitor is Gly; in addition, 8 of 15 N-terminal residues are identical among the three inhibitors.

X-ray crystallography of a complex between SMP and the inhibitor of E. chrysanthemi (Inh) reveals that Inh is folded into an eight-stranded -barrel with an N-terminal trunk of 10 residues (19). Residues 1-5 occupy part of the extended active site of the proteinase, thereby preventing access of the substrate. Residues 6-10 form a linker that connects the N-terminal proteinase-binding peptide to the body of the -barrel. The backbone carbonyl of Ser-1 interacts with the catalytic zinc; the Ser-2 side chain occupies the S₁'-binding site and also forms a hydrogen bond to the carboxyl end of the catalytic Glu, whereas Leu-3 occupies the S₂' recognition site. Penetration of the trunk region further than 5 residues into the substrate binding cleft appears to be prevented by the -barrel, which itself interacts with the proteinase near its Met turn (19). Peptide mimetics of the trunk at concentrations up to about 100 µM do not inhibit APR, thus demonstrating that the barrel is essential for inhibitory activity.²

The inhibitors of E. chrysanthemi and S. marcescens have been reported to inhibit their target proteinases with K_I values of 1-10 (17) and 0.7 µM (20), respectively; the inhibitor from P. aeruginosa has not previously been characterized with respect to its interaction with APR. Because of the potential role of this proteinase in the development of Pseudomonas keratitis (21-23), we initiated an investigation of the interaction between the two proteins. Our goal was to develop a recombinant expression system for APRin and to characterize its interaction with APR. After finding that the mature inhibitor exhibited an unexpectedly high affinity for APR (K_D 4 pM), we undertook an assessment of the role of the N terminus in determining enzyme-inhibitor affinity. In addition, because of the structural similarity between the catalytic domain of the serralysins and that of the matrix metalloproteinases, we investigated possible interactions between APRin and other members of the metzincin family. Finally, we found that PL specifically removed 5 N-terminal residues from APRin, thereby essentially inactivating it.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Construction of APRin Expression Plasmids-- Plasmids for expressing APRin and APRin mutants in Escherichia coli were developed using the pET22b(+) expression system (24). This vector was designed for isopropyl-1-thio--D-galactopyranoside-inducible expression of recombinant proteins with an N-terminal pelB leader for export to the periplasm followed by processing to remove the signal peptide. Four full-length and seven truncation mutants were prepared for this study. Full-length APRin derivatives include wt APRin, a mutant protein with Ser-1 of the P. aeruginosa protein replaced by Gly of the S. marcescens homolog (S1G), and two mutants with residues 6-10 replaced by (Pro)₅ or (Gly)₅ linkers_. Truncated APRins include proteins with deletion of residues 1-2, 1-3, 1-4, 1-5, 6-10, 1-6, and 1-10.

DNA encoding APRin or mutant APRins was generated by PCR with genomic DNA from strain PA-28 of P. aeruginosa as template (16). Synthetic oligonucleotides (listed in Table I) with 5' or 3' sites compatible with the MscI- (blunt) and BamHI-cloning sites of the expression vector were used as primers. A typical PCR contained 50 ng of EcoRI-digested template DNA, 1 µM each of sense and antisense primer, 50 µM dNPTs, 4 mM MgCl₂, and 10% Me₂SO (25) in 100 µl. The sample was initially denatured at 97 °C for 5 min, after which Taq DNA polymerase (2.5 units) was added, and DNA amplification was carried out in an MJ thermocycler programmed for 35 cycles at 95 °C (30 s), 50 °C (30 s), and 72 °C (60 s). After the completion of the amplification program, the reaction was incubated for an additional 7 min at 72 °C to increase the yield of PCR product. In most instances, the PCR products generated with primers containing 5' MscI and 3' BamHI sites were cloned into a pT7-Blue vector using reagents and protocols supplied in a PCR cloning kit (Novagen, Madison WI). Competent E. coli cells (XL1-Blue or NovaBlue) were transformed, and colonies with an insert-containing plasmid were selected by standard screening techniques (26). The APRin coding region was excised by restriction digestion, gel-purified, and ligated into the MscI/BamHI-cloning site of the pET22b(+) expression vector. For four of the constructs, a 5' blunt-ended PCR product with a 3' BamHI site was ligated directly into the expression vector.

In three of the plasmids, DNA sequence analysis of the expression plasmids revealed a non-conservative single base mutation. These mutations occurred toward the 3' end of the coding region and happened to be flanked either by unique StuI and HindIII restriction sites or between a pair of PstI sites. The mutations were repaired by inserting DNA excised from a different plasmid previously shown by sequence analysis to contain the desired wt sequence.

The (1-2) protein was prepared by digesting an S2R APRin mutant overnight at room temperature with 1.5 µg/ml trypsin (Sigma) followed by addition of 1.5 µg/ml soybean trypsin inhibitor (Sigma). Electrospray ionization mass spectrometry and SDS-PAGE confirmed that trypsin cleavage occurred exclusively at Arg-2; the resulting preparation was assayed without further purification. The presence of trypsin/trypsin inhibitor did not affect the APR kinetic assay.

Expression and Purification of Recombinant Proteins-- APRin proteins were expressed in E. coli strain BL21(DE3). A culture derived from a colony harboring the expression plasmid was grown at 37 °C in LB broth (1 liter) containing 50 µg/ml ampicillin. Isopropyl-1-thio--D-galactopyranoside (1 mM) was added at a culture A₆₁₀ of 0.6-1.0; growth was continued for either 4 h at 37 °C or overnight at 30 °C. The culture was centrifuged (10,000 × g, 4 °C), the cell pellet washed with 30 mM Tris-HCl, 20% sucrose, 1 mM EDTA, pH 8.0, and then suspended in ice-cold 5 mM MgSO₄ to induce osmotic shock. After stirring for 10 min, phenylmethylsulfonyl fluoride was added to a final concentration of 100 µM. The suspension was centrifuged, and the supernatant was brought to 20 mM Tris, pH 7.4, and incubated overnight at 4 °C with 100 units of Omnicleave^TM endonuclease (Epicentre Technologies) to digest polynucleotides that otherwise co-purified with the inhibitor protein. The digest was concentrated by pressure filtration using an Amicon YM-10 membrane and then loaded on a Bio-Rad Hi-Q anion exchange cartridge previously equilibrated with 20 mM Tris-HCl, pH 7.4. Unbound material was eluted with equilibration buffer, and bound material was eluted with a linear gradient of NaCl (0 to 0.1 M) in equilibration buffer. Fractions containing inhibitor were concentrated and re-chromatographed on a Sephacryl S-200 column (1 × 32 cm) equilibrated with 20 mM Tris-HCl, 0.1 M NaCl, pH 7.4. Inhibitor-containing fractions were concentrated, and the NaCl was reduced to <0.01 M by two cycles of dilution with 20 mM Tris-HCl, pH 7.4, followed by ultrafiltration. The resulting APRin preparations were homogeneous as assessed by SDS-PAGE in a Tris-Tricine buffer system (27). The mass of each recombinant protein was verified by electrospray ionization mass spectrometry, which also served to demonstrate the absence (<1%) of heterogeneity resulting from imprecise removal of the signal peptide.

APRin concentrations were estimated with an ₂₈₀ of 19.6 mM¹ cm¹ calculated from the aromatic amino acid and cystine content (28). Comparison of the UV absorption spectra of native and denatured APRin in 6 M GdnHCl showed that ₂₈₀ for the folded protein was about 10% less than the calculated value; this value was used to estimate inhibitor concentrations in kinetic experiments.

Metalloproteinases-- Crystalline APR and PL were purchased from Nagase Biochemicals Ltd., Fukuchiyama, Japan. The proteinases were purified to apparent homogeneity as assessed by SDS-PAGE using the ion exchange procedure described above for the inhibitor proteins. APR and PL concentrations were estimated from their absorbance at 280 nm using E_1 cm^1% values 16.0 and 14.5, respectively (8, 9). Interstitial collagenase was purified from serum-free culture medium of phorbol ester-stimulated human umbilical vein endothelial cells (29), and human gelatinases A and B were from Oncogene Research Products, Cambridge, MA.

Kinetic Assays-- The thioester peptolide Ac-Pro-Leu-Gly-(2-mercapto-4-methylpentanoyl)-Leu-Gly ethyl ester (TPS, Bachem Bioscience) was used as substrate for APR and the MMPs (30). Assays were conducted at 25 °C in the presence of dithiodipyridine (DTDP, Sigma) as a thiol indicator (31) in microtiter plates using a Molecular Devices Spectramax 250 plate reader set to 324 nm. The assay buffer was 50 mM MOPS, pH 7.0, 5 mM CaCl₂, 0.001% BSA, 500 µM DTDP and NaCl at 0.1 or 2.4 M. To obtain reproducible kinetics, we treated the assay wells with silanizing agent (Sigmacote) and added a low concentration of BSA (Sigma) to the reaction. Multiple reactions were initiated simultaneously from the same stock solutions by dispensing 100 µl of enzyme solution into an equal volume of substrate ± inhibitor using a multichannel pipette. Blank reactions without proteinase were run to correct for spontaneous hydrolysis of the substrate. The reaction between DTDP and the thiol product was shown not to be rate-limiting with 500 µM DTDP and up to 400 µM substrate.

Data Analysis-- Kinetic constants were estimated by nonlinear least squares analysis using the computer program Dynafit of Kuzmic (32) in conjunction with reaction Scheme I





to analyze a family of progress curves obtained with a range of inhibitor to enzyme ratios. K_s and k_cat values were set to their experimentally determined values, and the highest inhibitor concentration was fixed at the value calculated from its absorbance at 280 nm. Trial values of the association and dissociation rate constants, along with a zero time A₃₂₄ offset, enzyme concentration, and the inhibitor concentration for each run were adjusted to produce the optimum fit for the data set as a whole. The standard deviation of kinetic constants derived from a set of eight kinetic runs was generally <10%. Unless noted, the values of the kinetic constants reported represent means of at least three independent experiments at six different inhibitor concentrations.

Spectroscopic Methods-- Fluorescence emission spectra were recorded at room temperature with a Perkin-Elmer LS50B instrument set for excitation at 280 nm. Protein concentrations were 0.5 µM in 20 mM Tris-HCl, 50 mM NaCl, pH 7.4, with or without 4 M GdnHCl. CD spectra were measured at room temperature in a cuvette of 0.02-cm path length with a Jasco J-710 spectropolarimeter that was calibrated with d-camphor sulfonic acid. Protein concentrations were 20-40 µM in 20 mM Tris-HCl, pH 7.4. Four spectra were collected from 240 to 190 nm at 1-nm intervals, the results averaged and converted to values based on the mean residue weight of each protein. The fitting program CDsstr of Johnson (33) was used to estimate secondary structure from the CD data. Emission and CD spectra were corrected by subtraction of a buffer blank.

Ultracentrifugation-- Sedimentation equilibrium analyses were carried out on samples at 20 °C in assay buffer with 0.1 or 2.4 M NaCl but without BSA or DTDP using a Beckman model L5-75B ultracentrifuge equipped with a Prep UV scanner. Equilibrium was achieved by over-speeding at 1.5 times the final speed for 4 h followed by centrifugation overnight at 13,000 or 18,000 rpm. The apparent molecular weight (M) of the macromolecular species was estimated from the slope (M(1 ())) of plots of lnA₂₈₀ versus r². Partial specific volumes () were calculated from residue partial specific volumes and the amino acid composition of each protein. Buffer density () was measured with a Paar density meter.

Mass Spectrometry-- Mass spectra were obtained using either electrospray ionization quadrupole mass spectrometry or matrix-assisted laser desorption ionization-time of flight mass spectrometry. The electrospray ionization quadrupole mass spectrometry instrument was a Micromass Quattro LC electrospray ionization triple quadrupole mass spectrometer with an orthogonal array ion source. Single quadrupole and positive ion data were collected. Samples were infused directly as a solution in 50% aqueous acetonitrile, 0.1% CF₃COOH. Scanning data were collected for m/z = 200-2000. The instrument was calibrated using purified horse heart myoglobin and data were analyzed using the maximum entropy algorithm provided with the Masslynx software. Matrix-assisted laser desorption ionization-time of flight mass spectrometry data were acquired using a Micromass TofSpec 2E instrument operating in the linear mode. The matrix used was sinapinic acid; the solid support was stainless steel, and a nitrogen (337 nm) laser was used. Time of flight data were collected for the m/z range 500-100,000. Internal standard calibration was performed using purified cytochrome c, myoglobin, and trypsinogen. Protein samples were desalted and concentrated by adsorption to C₁₈ Zip Tips (Millipore) and eluted with 50% aqueous acetonitrile, 0.l% CF₃COOH prior to analysis.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Protein Expression and Purification-- A representative purification of full-length S1G-APRin is summarized in Fig. 1A and shows the UV absorbance elution profile of concentrated osmotic shock solution chromatographed on an anion exchange cartridge. The electrophoretic analysis in Fig. 1B shows that the inhibitor eluted in fractions 13-15 at approximately 50 mM NaCl. These fractions were concentrated and subjected to molecular sieve chromatography on Sephacryl S-200 (Fig. 1C) to remove a small amount of contaminating protein. Analysis of these fractions by SDS-PAGE (Fig. 1D) showed that inhibitor protein eluted in pure form in fractions 23-26. Similar results were obtained with all the APRin proteins.

View larger version (64K): [in this window] [in a new window]   Fig. 1.   Purification of S1G-APRin. A, A280 and A260 elution profiles of concentrated osmotic fluid containing S1G-APRin chromatographed on a HiQ cartridge as described under "Experimental Procedures." A linear gradient of 0-100 mM NaCl was initiated at fraction 10. Fraction volume was 6.5 ml. B, electrophoretic analysis by SDS-PAGE of starting material and fractions from ion exchange chromatography in A. The lanes contained the following: 1, osmotic shock solution; 2, concentrated osmotic shock solution; 3, unbound fraction from the ion exchange cartridge; 4-13, fractions 3, 5, 10-17; 14, purified APRin mutant S1G/V100E (1 µg); 15, molecular weight markers. C, elution profile of pooled and YM-10 concentrated fractions 13-15 from the ion exchange step loaded on a Sephacryl S-200 column and eluted with 20 mM Tris-HCl, 0.1 M NaCl, pH 7.4. D, electrophoretic analysis of fractions from the molecular sieve chromatographic step. The lanes contained the following: 1, concentrated fractions 13-15 from ion exchange cartridge; 2, flow-through from the YM-10 concentrator; 3, Sepharose fraction 10; 4, fraction 15; 5-12, fractions 19-26; 13, fraction 30; 14, APRin mutant S1G/V100E (1 µg); 15, molecular weight markers. Fractions 23-26 were pooled and concentrated by pressure filtration.

Physical Characterization of APRin and Mutant APRins-- The fluorescence emission and low UV CD spectra of mutant and wt APRins were determined as a means comparing the proteins in terms of tertiary and secondary structure. The inhibitors have 3 Trp residues (at positions 15, 53, and 61 in APRin) that should be solvent-inaccessible based on the crystal structure of Inh (19); thus, the fluorescence emission maxima of the folded proteins should be blue-shifted compared with Trp residues in an aqueous environment (34). Under non-denaturing conditions, the recombinant proteins exhibited comparable emission maxima (332 ± 2 nm); when the proteins were denatured (4 M GdnHCl), their emission maxima shifted to ~358 nm, a value characteristic of the indole fluorophore in an aqueous medium. The relative emission intensities of the folded proteins were the same within about 5%. We conclude from these characteristics that the recombinant APRins are folded into compact proteins of similar tertiary structure.

We utilized low UV CD spectra to compare secondary structures of the APRin proteins. As illustrated in Fig. 2, A and B, the CD spectra of all APRin proteins exhibited a maximum near 188 nm, a minimum near 200 nm, and a small peak near 230 nm. The low UV regions are dominated by contributions from the peptide backbone, and the 230-nm feature likely involves the single disulfide bond. The spectra were analyzed with the program CDsstr (33) to estimate the relative amounts of various secondary structural elements present. For simplicity, only the fitted spectrum for wt APRin resulting from this analysis is shown in Fig. 2A; the fitted and experimental spectra were virtually superimposed.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Circular dichroism spectra of APRin proteins. A, spectra of full-length APRin derivatives. B, spectra of truncated APRin mutants. Data points represented by filled circles represent fitted values for wt APRin using the program CDsstr (33), and the estimated percentages of secondary structures are given in the text. For clarity, only the fitted spectrum of wt APRin is shown.

CDsstr gives the percentage of -helix, 3₁₀-helix, -strand, turns, polyproline II, and other secondary structural elements based on the CD spectra of collection of standard globular proteins (33). All of the recombinant proteins exhibited a similar distribution of these structural elements based on analysis of the CD spectra. For the proteins in Fig. 2, the percentage of -helix was 2.3 ± 1.3%, 3₁₀-helix was 6.3 ± 0.9%, -strand was 17.0 ± 4.0%, turn was 15.6 ± 1.9%, polyproline II was 12.8 ± 1.2%, and other was 45.7 ± 1.6%, where the uncertainties indicate the standard deviation of the mean for the 10 proteins. Unfortunately, the CD spectrum of Inh is not available to compare with the spectra of the APRins. However, based on analysis of the crystal structure of the Inh·SMP complex (19, 35), this inhibitor contains 11% -helix, 53% -strand, 22% turns, and 14% of other secondary structures. Thus, the APRins appear to differ in secondary structure from Inh, particularly in -structures. Whether these apparent differences in secondary structure are real or result from uncertainties in the CD analysis cannot be assessed until a three-dimensional structure of APRin becomes available. Nevertheless, the CD and fluorescence data together indicate that the conformation of the recombinant full-length and N-terminal truncation mutants in this study are similar.

The aggregation state of APR and its complex with full-length APRin and selected truncation mutants was assessed by equilibrium ultracentrifugation. Radial distribution plots of logA₂₈₀ versus r² for APR and APRin alone were linear (data not shown), demonstrating that both proteins sedimented as a single homogeneous species; the calculated molecular weights were within 2% of the expected values (49,467 and 11,386, respectively). Linear logarithmic radial distribution plots were also observed when mixtures of APR and wt or S1G APRin were analyzed. The apparent molecular weights of 55,500 and 55,670 calculated from the sedimentation data in low and high salt, respectively, are similar to the value expected for a 1:1 complex. For the low affinity truncated APRins, the logarithmic sedimentation profiles were generally nonlinear; however, it was not feasible to assess the extent of complex formation quantitatively because the relatively small UV absorption coefficient of APRin (17.6 mM¹ cm¹) compared with APR (80 mM¹ cm¹) precluded accurate determination of bound and free inhibitor at different positions in the centrifuge cell (36).

Kinetics Studies-- APR catalyzed the hydrolysis of the thioester peptolide TPS previously developed as a chromogenic substrate for collagenase (30). Reversed phase high performance liquid chromatography showed that APR cleaved TPS exclusively at the thioester bond (data not shown). Spectrophotometric progress curves for hydrolysis of TPS by APR were fit to a Michaelis-Menten mechanism to give K_s and k_cat values. K_s was affected by salt concentration, decreasing from 845 ± 88 µM in 0.1 M NaCl to 132 ± 41 µM in 2.4 M NaCl; k_cat (11.7 ± 3.1 s¹) was not significantly affected by ionic strength. The resulting 6.4-fold increase in k_cat/K_s at high ionic strength was useful because it allowed determination of the kinetics of TPS hydrolysis with lower APR concentrations than was possible at low ionic strength.

Fig. 3A shows a family of kinetic curves obtained with APR in the presence of increasing APRin concentrations. In the presence of APRin, the initial rate of TPS hydrolysis was the same as that observed without inhibitor; however, over time, inhibition of substrate hydrolysis became evident and was complete within about an hour at the higher ratios of inhibitor to enzyme. The rate of TPS hydrolysis was constant over the 4-h assay time in the absence of APRin.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Inhibition of APR by S1G-APRin. A, inhibition of APR by increasing concentrations of S1G-APRin. Experimental data points are superimposed on progress curves calculated using the best-fitting kinetic constants below. Concentrations of S1G-APRin (top curve to bottom) were as follows: 0, 0.25, 0.5, 0.75, 1.0, 1.5, 2.5, and 0 nM, respectively. Data points in the bottom curve represent background substrate hydrolysis without APR. Reactions were initiated by mixing 100 µl of APR with 100 µl of substrate/inhibitor solution pre-equilibrated at 25 °C. Other conditions were as follows: [S] = 300 µM, [APR] = 0.5 nM in 0.05 M MOPS, 2.5 M NaCl, 5 mM CaCl2, pH 7.0, 0.001% BSA. The 324 due to product formation was determined independently to be 0.00936 µM1 for the 200-µl reaction volume. Best-fitting values of kon and kh (rate constants for complex formation and spontaneous substrate hydrolysis) were 0.47 ± 0.003 × 106 M1 s1 and 0.208 ± 0.003 × 105 s1 as determined with Dynafit (32) with Ks (175.3 µM) and kcat (12.6 s1) fixed at the indicated values determined independently with the same reagent solutions. B shows the residuals as a percentage of the A324 values in A. The graphs correspond from top to bottom to kinetic curves 1-8 in A.

The slow binding inhibition exhibited in this experiment is compatible with at least two mechanisms (37). In the first, enzyme and inhibitor interact in a bimolecular step to form an inactive EI complex (Scheme I); the magnitude of the association and dissociation rate constants, in conjunction with the concentration of the protein reactants, determines the time for enzyme and inhibitor association and thus the appearance of inhibition. In a more complex variation of this mechanism, enzyme and inhibitor initially form an EI complex that subsequently undergoes a conformational change to produce a high affinity, fully inhibited EI' complex.

The data set of Fig. 3A was analyzed using Dynafit (32) in conjunction with the mechanism in Scheme I (with k_off set to zero³), to give a rate constant k_on for enzyme-inhibitor association of 0.47 ± 0.02 × 10⁶ M¹ s¹. The residual plot in Fig. 3B indicates that the data are well represented by the single-step binding mechanism, thus indicating that this simple process adequately describes the kinetics of inhibitor binding. Comparable association rate constants were obtained with TPS concentrations from 50 to 450 µM and APR concentrations from 0.01 to 1 nM, as expected for a reaction occurring in a single bimolecular step. Decreased ionic strength (to 0.1 M) resulted in approximately a 3-fold increase in k_on (to 1.7 ± 0.2 × 10⁶ M¹ s¹), whereas addition of 20% glycerol resulted in a decrease of about 2-fold in the association rate constant (to 0.22 ± 0.1 × 10⁶ M¹ s¹).

The binding of wt and S1G APRin to APR was essentially irreversible even with the lowest practical concentration of reactants; this irreversibility prevented estimation of k_off by fitting progress curves. However, we were able to determine the rate of EI dissociation by dilution of the pre-formed complex into an activity assay solution. Fig. 4 shows that in such an experiment, enzyme-catalyzed hydrolysis of TPS was not immediately apparent, thereby showing that initially all APR was in an inhibited complex. The rate of TPS hydrolysis increased over time and became constant within about 6 h, reflecting the equilibrium concentration of free APR. Fitting the data of Fig. 4 to Scheme I gave a value for k_off of 2.29 ± 0.04 × 10⁶ s¹; combining the association and dissociation rate constants resulted in a calculated K_D of 4 × 10¹² M for S1G-APRin. A similar K_D was calculated for wt APRin (Table II).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Dissociation of S1G-APRin from APR. APR (50 nM) and APRin (65 nM) were incubated in 50 mM MOPS, 0.1 M NaCl, 5 mM CaCl2, pH 7.0. After 60 min at room temperature, a 2-µl aliquot of the mixture was diluted into 200 µl of assay mix containing 2.4 M NaCl and 400 µM substrate at 25 °C. The data points shown were corrected for spontaneous substrate hydrolysis and represent the mean of duplicate experiments. A dissociation constant koff of 2.29 ± 0.04 × 106 s1 was estimated by least squares analysis using Dynafit (32) and reaction Scheme I. Essentially identical kinetics was observed when the reactants were preincubated at high ionic strength (2.4 M).

Inhibition of APR by N-terminally Truncated Mutants-- We constructed a series deletion mutants to test the hypothesis that inhibition of APR by APRin depends on the presence of an intact N-terminal region. Mutant proteins lacking 10 or 6 N-terminal residues at concentrations up to 10 µM did not inhibit APR. Deletion of 2-5 N-terminal residues gave inhibitors with progressively decreased affinity (Table II). The effect of N-terminal truncation is illustrated in Fig. 5 for (1-5) APRin. By using Dynafit and Scheme I in conjunction with these data, we determined values for both k_on and k_off from a set of progress curves. Compared with wt APRin, k_on decreased ~1,000-fold and k_off increased by ~30-fold to give an increase in K_D from 4 pM to 0.12 µM. Surprisingly, deletion of residues 6-10 (to give a trunk with the same number of residues as (1-5) but with a wt N terminus) did not inhibit. To explore this further, we prepared APRin derivatives with wt residues 6-10 (-Leu-Ser-Ala-Ser-Asp-) replaced by a linker consisting of an equivalent number of Gly or Pro residues. We anticipated that (Gly)₅ should produce a flexible "hinge" between the N-terminal end of the trunk and the barrel of the inhibitor and that (Pro)₅ should produce a linker with less conformational mobility. Both restored inhibition to the (6-10) mutant but with the affinity reduced by 100-fold for the (Gly)₅ mutant and 50-fold for the (Pro)₅ mutant. The kinetic and equilibrium constants for all APRin proteins in this study are summarized in Table II.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Inhibition of APR by (1-5)-APRin. Concentrations of inhibitor were 0, 0.5, 1, 2, 3, and 4 µM (top to bottom). Other conditions were as follows: [S] = 350 µM and [APR] = 1 nM. Progress curves were corrected for spontaneous substrate hydrolysis prior to fitting. Ks and kcat were fixed at their experimentally determined values. For the particular data set shown, best-fitting values of kon and koff are 489 ± 10 M1 s1 and 4.19 ± 0.03 × 105 s1, to give a KD of 0.39 µM for this experiment.

Interaction of APRin with Other Metalloproteinases-- Because of the structural homology between APR and the MMPs, it was of interest to determine if APRin inhibits or possibly is inactivated by the MMPs. We found that 10 µM S1G-APRin did not inhibit interstitial collagenase, gelatinase A, or gelatinase B when tested with TPS as substrate. Furthermore, as assessed by mass spectrometry, none of these MMPs degraded or inactivated full-length APRin when incubated overnight with nM concentrations of MMP.

To determine whether APR is capable of removing the N-terminal residue of APRin as reported for SMP and Inh (19), we incubated S1G-APRin overnight with approximately equimolar APR. Analysis by mass spectrometry (Fig. 6A) showed that APR did not significantly modify the inhibitor, with less than ~1% of the N-1 truncation product detected. In contrast, overnight incubation of S1G-APRin with a catalytic amount of PL resulted in removal of residues 1-5 (Fig. 6B). Hydrolysis proceeded with initial loss of residues 1 and 2 followed by removal of residues 3-5. This degradation was prevented by 0.5 µM HONH-COCH₂CH₂CO-Phe-Ala-NH₂, a known inhibitor of PL and the MMPs (38, 39). There was no evidence by mass spectrometry or SDS-PAGE of additional degradation of the inhibitor by PL.

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Mass spectrometry of S1G-APRin after treatment with APR or PL. A, S1G-APRin (17 µM) was incubated in assay buffer at room temperature for 18 h in the absence (dashed line) or presence of 15 µM APR (solid line). Ten percent CF3COOH was then added to give a final concentration of 0.1%, and the samples were desalted prior to mass spectrometry with a C18-ZipTip and eluted with 50% aqueous acetonitrile, 0.1% CF3COOH. B, S1G-APRin (24 µM) was incubated in assay buffer at room temperature for 18 h in the absence (dashed line) or presence (solid line) of 7 nM PL. Reactions were quenched and prepared for analysis as described above for A.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The results of this study show that APRin interacts with its cognate proteinase in a reversible reaction that follows bimolecular kinetics as assessed by measurement of the time-dependent loss of catalytic activity. Implicit in this assertion is the reasonable assumption that the rate of inhibition of APR reflects the rate of its complex formation with APRin. We were unable to detect changes in UV absorbance or fluorescence that could provide an independent assessment of complex formation. However, the invariance of the apparent second order rate constant measured over a range of reactant concentrations supports the assumption that inhibition and binding are commensurate.

The association rate was sensitive to the viscosity of the medium as expected for a diffusion-limited reaction. The 2-fold decrease in k_on that accompanied the increase in viscosity brought about by 20% glycerol is approximately that expected for a purely diffusion-limited reaction according to the Smoluchowski equation (40). With respect to ionic strength, the relatively small decrease in k_on (~3-fold) associated with the large increase in salt concentration (24-fold) suggests that electrostatic interactions between the reactants are not a major determinant of the rate of complex formation. The small effect of ionic strength on the binding of APRin to APR can be contrasted with the 100-fold decrease in association rate constant for the barnase-barstar interaction associated with a 10-fold increase in ionic strength (41). Ionic strength effects of similar magnitude have been noted with other electrostatically controlled protein association reactions including hirudin with thrombin (42) and cytochrome c with cytochrome b₅ (43). In these systems there was clear indication that mutual attraction of interfacial regions controls the rate of association by pre-aligning the reactants to increase the number of effective collisions (44, 45).

The effect of trunk length on the kinetics of APR-APRin association may be interpreted in terms of the Brownian dynamics model of protein-protein interaction (46). In this model, reaction partners collide in random orientations at a diffusion-limited rate. They become transiently trapped in a solvent cage that allows time for rotation and multiple close-range collisions to occur, thereby increasing the probability that reactive surfaces become aligned and guide productive bonding before the complex dissociates. A favorable effect of trunk length on the association rate could result from the increased probability for productive interactions provided by the increased surface area of a longer trunk. If this weakly bonded complex is viewed as a transition state, then the increased surface area provided by a long trunk will increase the stability of suboptimally aligned states leading to final docking. Fig. 7A shows that there is a strong correlation (r = 0.95) between log k_on and total surface area (side chain plus backbone) of the 5-residue trunk proposed to interact with APR. This relationship suggests that the number of potential contacts between reactants does influence the stability of the transition state for association.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Relationship between log kon or free energy of binding and solvent-accessible surface area of the N-terminal trunk of wt and mutant APRins. A, solvent-accessible surface area (side chains plus backbone) of residues comprising the binding region of APRin or APRin-deletion mutants was calculated from data in Creighton (48). The line was derived by linear regression (correlation coefficient 0.9495). B, non-polar solvent-accessible surface area of the APRin derivatives was calculated from data in Ref. 48. The slope of the regression line is 10 cal/mol/Å2 (correlation coefficient 0.7684).

The Gibbs free energy change associated with binding of APRin and APRin mutants also correlates with accessible surface area of the non-polar regions trunk, but the correlation is weaker (r = 0.76) than that for the rate of association. As shown in Fig. 7B, a plot of G of binding versus solvent-accessible non-polar surface area of the residues of APRin believed to interact with APR is approximately linear with a slope of ~10 cal/mol/Ų for the regression line. Free energy relationships such as this are usually correlated with desolvation of non-polar surfaces on binding and hence are related to hydrophobicity. The free energy change varies between 25 and 50 cal/mol/Ų depending on the experimental system (45). The value of ~10 cal/mol/Ų in Fig. 7B probably represents a minimal estimate of the contribution of hydrophobicity to binding because proteinase surfaces were not included in the estimation of solvent-accessible area, yet these sites may also release water molecules to the bulk solvent on binding inhibitor. The effect of NaCl on the apparent K_s for the non-polar substrate TPS (a 6.5-fold decrease) is also consistent with a role for hydrophobicity in ligand binding at the substrate-binding site. Interestingly, the interface between the inhibitor -barrel and the Met-turn region of the proteinase contains 16 water molecules (19), suggesting that desolvation in this region may make less of a contribution to the energetics of binding.

The results of this study also show that the interaction of APR with its cognate inhibitor is unexpectedly strong in comparison to the two other serralysin-inhibitor pairs. Our data demonstrate that the APR·wtAPRin complex is formed with a K_D of ~4 pM compared with reported affinities of the SMP·SMPI complex of 0.7 µM (20) and the E. chrysanthemi complex of 1-10 µM (17). It is perhaps significant that the secondary structure of wt APRin provided by CD indicates that its backbone conformation may differ from that of Inh, particularly with respect to the amount of -structure present. These possible structural differences might provide stabilizing interactions, perhaps in the barrel region, that are not available to the SMP·Inh complex.

The reduced affinity of the truncated inhibitors and the lack of inhibition engendered by deletion of 6 or 10 N-terminal residues supports the inference from crystallography only 5 N-terminal residues of the trunk can bind to the proteinase. Ser-2 of Inh occupies S1' of SMP, and Leu-3 and Leu-5 also contact the proteinase, presumably at S2' and S4' sites (19). The weak inhibition observed with (1-5)⁴ and the lack of inhibition by (6-10) can be rationalized if APR has an S4' site that interacts specifically with the hydrophobic residues such as Leu-5, which happens to be conserved in the three family members. The N-terminal residue of the (6-10) mutant is Ser, which may not bind strongly enough to the S4' site to stabilize a complex. On the other hand, N-terminal position in the (1-5) mutant is Leu and could be accommodated at this site (refer to Table II for a summary of the N-terminal sequences).

Finally, we found that PL, also a secreted metalloproteinase of P. aeruginosa, could essentially inactivate APRin by removing its 5 N-terminal residues. In retrospect, this action of PL is not surprising in view of the proclivity of this enzyme for cleaving at bulky hydrophobic residues (47). Whether inactivation of APRin is physiologically significant or not is unclear because the function of the inhibitor has not been defined with certainty. It has been suggested that serralysin inhibitors protect the host from proteolysis during proteinase export (17). From a teleological standpoint, it would appear advantageous for an organism to have a mechanism for inactivating an inhibitor that could be deleterious if released where it could neutralize a survival factor. Thus, inactivating APRin released for example as a result of cell death would benefit the colony as a whole. In this regard, use of APRin as a specific inhibitor of APR in treating Pseudomonas keratitis may require development of a modified molecule that resists the action of PL and possibly other proteinases present in infected corneas.

	ACKNOWLEDGEMENTS

We thank Dr. P. V. Liu for the P. aeruginosa strains used in this study, Ned Smith and Dr. Jian Cai for mass spectrometry, Dr. Ron Gregg for DNA sequencing, and Dr. Arno Spaola for the hydroxamate metalloproteinase inhibitor. The CD spectrophotometer was purchased with funds from National Science Foundation Grant BIR-91-19404 and the University of Louisville Research Foundation. The Mass Spectrometry Laboratory is supported in part by National Institutes of Health Grant 1 S10 RR11368-01A1 (to W. M. P.), the State of Kentucky Physical Facilities Trust Fund, the University of Louisville School of Medicine, and the University of Louisville Research Foundation.

	FOOTNOTES

^* This work was supported in part by National Institutes of Health Grant EY09722 (to R. D. G. and C. A. Paterson).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Recipient of a University of Louisville Graduate Fellowship.

To whom correspondence should be addressed: Dept. of Biochemistry and Molecular Biology, University of Louisville, Louisville, KY 40292. Tel.: 502-852-5226; Fax: 502-852-6222; E-mail: rdgray@louisville.edu.

Published, JBC Papers in Press, April 17, 2000, DOI 10.1074/jbc.M002088200

² R. D. Gray and R. E. Feltzer, unpublished data.

³ When k_off was included as an adjustable parameter in the fitting procedure, its value consistently reached the lower limit (10⁹ s¹) set in the program, thereby suggesting that under the conditions of the assay the reaction was kinetically irreversible.

⁴ It is conceivable that the inhibition observed with the (1-5) mutant could have resulted from the presence of a small amount (~1%) of a contaminating high affinity inhibitor, possibly resulting from incomplete or inaccurate processing of the recombinant pelB-APRin protein. However, no contaminants were detected by mass spectrometry under conditions where a level of 1% of a contaminant could be observed. In addition, (1-5)-APRin generated by treating wt APRin with PL exhibited the same inhibitory properties as the recombinant preparation.

	ABBREVIATIONS

The abbreviations used are: PL, pseudolysin from P. aeruginosa; APR, alkaline proteinase from P. aeruginosa; APRin, alkaline proteinase inhibitor from P. aeruginosa; BSA, bovine serum albumin; DTDP, 4,4'-dithiodipyridine; Inh, APRin homolog from E. chrysanthemi; MMP, matrix metalloproteinase; MOPS, 3-(N-morpholino)propanesulfonic acid; PCR, polymerase chain reaction; PAGE, polyacrylamide gel electrophoresis; SMP, APR homolog from S. marcescens; SMP inhibitor, homolog of APRin from S. marcescens; TPS, Ac-Pro-Leu-Gly-(2-mercapto-4-methylpentanoyl)-Leu-Gly ethyl ester; wt, wild-type; (1-n) APRin, wt APRin with residues 1 to n deleted; (Gly)₅ and (Pro)₅ designate APRin mutants in which residues 6-10 were replaced by the indicated amino acid residues, Tricine, N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine; GdnHCl, guanidine hydrochloride.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg