Pseudomonas Keratitis: Protease IV Gene Conservation, Distribution, and Production Relative to Virulence and Other Pseudomonas Proteases

Abstract

purpose. To determine the distribution of the protease IV gene, the production of this and other proteases by multiple strains of Pseudomonas, and the virulence of a mutant specifically deficient in protease IV.

methods. The protease IV gene was cloned, its sequence analyzed, and its chromosomal location determined by pulse-field gel electrophoresis. Three PCR reactions were used to detect the protease IV gene in 30 Pseudomonas isolates and protease production was determined by Western blot analysis, colorimetric assay, and zymography. An allelic replacement mutant deficient in the protease IV gene was analyzed for enzyme production, corneal growth, and corneal virulence.

results. The protease IV gene was identified in all P. aeruginosa, but none of the non-aeruginosa strains tested. The protease IV genes of strains PA103-29 and PAO1 were in a common chromosomal site and had 98.5% sequence identity with variations occurring mainly in the promoter region. The protease IV activity of the 23 wild-type P. aeruginosa strains tested varied from 2.3 to 221.5 × 10⁻³ U/mg protein in the culture supernatant. Protease IV was produced by all P. aeruginosa wild-type strains. A protease IV–deficient mutant derived from strain PA103-29 had reduced virulence compared with its parent strain and unexpectedly produced alkaline protease.

conclusions. The protease IV gene and its product are common to P. aeruginosa, but not to other Pseudomonas species. Protease IV activity varies among P. aeruginosa strains, and a mutant specifically deficient in this activity produced alkaline protease and had reduced corneal virulence.

The bacterium Pseudomonas aeruginosa is a potent opportunistic pathogen frequently responsible for wound and burn infections as well as infections of the lungs, skin, urinary tract, external ear, and eye. ^{1 2 3 4 5 6 7 8} Immune-compromised individuals, such as cancer patients, are at particular risk of illnesses of high morbidity and mortality caused by this organism. ^{9 10 11 12 13} Contact lens users are susceptible to Pseudomonas keratitis, the most rapidly devastating bacterial infection of the cornea. ^{14 15 16 17 18 19 20}  

The virulence of P. aeruginosa for the cornea has been correlated with a variety of factors, ^{21 22 23 24 25 26 27 28 29} including the production of the extracellular enzyme protease IV. Production of this enzyme has been shown in several animal models to correlate with tissue damage during keratitis. ^{21 30 31} Protease IV has been shown to destroy a variety of host proteins, including such host defense factors as complement and IgG class immunoglobulin. ³² In addition, Wilderman et al. ³³ have reported that a protease with the exact size and N-terminal sequence of protease IV appears to contribute to virulence in a rat model of Pseudomonas lung infection. 

The information linking protease IV to corneal virulence has been based on the study of only one strain of P. aeruginosa, PA103-29. Information describing the distribution of this virulence-related gene and production of protease IV by a spectrum of Pseudomonas clinical isolates is not available. Proof of a wide distribution of this gene and production of this enzyme by multiple clinical isolates is essential to confirming the enzyme as a factor involved in the tissue damage that occurs in typical Pseudomonas infections. In the present study, 30 strains of Pseudomonas were analyzed for the presence of the protease IV gene and for production of protease IV. In addition, the production of other proteases relative to protease IV was analyzed. Finally, the virulence and enzyme production of a mutant specifically deficient in the protease IV gene was determined. Overall, the results indicate that protease IV is a commonly secreted product of P. aeruginosa, that the amount of enzyme produced is strain specific, and that a mutant deficient in protease IV initiates production of alkaline protease, yet has significantly reduced corneal virulence. 

Materials and Methods

Reagents

Enzymes and marker proteins were purchased from New England BioLabs (Beverly, MA), chemical reagents were purchased from Sigma-Aldrich (St. Louis, MO), and media from Difco Laboratories, Inc. (Detroit, MI). 

Bacterial Strains and Growth Conditions

All P. aeruginosa strains were grown in M9 medium, as described previously. ³² Escherichia coli strains XL-1 Blue MRF′, XLOLR, and DH5α were grown at 37°C in Luria-Bertani (LB) broth or NZY medium (5 g NaCl, 2 g MgSO₄ · 7H₂O, 5 g yeast extract, 10 g NZamine in 1 L dH₂O). P. aeruginosa strains PA103-29 and PA103-AP1 were kindly provided by Barbara H. Iglewski (University of Rochester, Rochester, NY) and strain PAO1 was provided by Dennis E. Ohman (Medical College of Virginia Campus of Virginia Commonwealth University, Richmond, VA). Strain BR-1 was purchased from the American Type Culture Collection (Manassas, VA). P. aeruginosa clinical isolates PA23, -70, and -178, were obtained from Charity Hospital, (New Orleans, LA). P. aeruginosa strains ALC1, ALC2, ALC3, 13004, 13010, 13014, 13036, 30009, 30099, 30132, 30244, 51003, 51005, 51131, 51036, and 51037 and the non-aeruginosa strains P. alcaligenes, P. mendocina, P. putida, P. otitidis 10150, and P. stutzeri 17001 and 17030, were provided by David W. Stroman (Alcon Laboratories, Fort Worth, TX). 

Cloning the Putative Protease IV Gene

Genomic DNA from P. aeruginosa strain PA103-29 was digested with BclI, and the DNA fragments were size fractionated by sucrose gradient centrifugation. ³⁴ DNA fragments in the range of 4 to 8 kb were ligated with lambda ZAP Express/BamHI vector DNA (0.5 μg; Stratagene, La Jolla, CA), and packaged using a lambda packaging system (Gigapack III Gold; Stratagene). The resultant phage particles were used to infect the host strain E. coli XL-1 Blue MRF′. Phage from plaques were blotted onto membranes (Nytran; Schleicher & Schuell, Keene, NH). A radioactive probe specific for the putative protease IV gene was synthesized by determining N-terminal amino acid sequences of purified protease IV fragments and matching these sequences to a region (bases 4,671,318–4,672,706) of the published sequence of the PAO1 genome (The Pseudomonas Genome Project; http://www.pseudomonas.com). From this sequence, primers for a PCR reaction were synthesized: 5′-GCCGGCTACCGCGACGGCTTC-3′ as the forward primer, and 5′-TCAGGGCGCGAAGTAGCGGGAG-3′ as the reverse primer. The PCR reaction used incubations at 97°C for 10 minutes, 90°C for 3 minutes (add Taq polymerase), and 35 cycles of 95°C for 1 minute, 60°C for 30 seconds, and 72°C for 1 minute. The PCR reaction yielded a 756-bp fragment that was purified and labeled with [α-³²P] dCTP, by using DNA labeling beads (Ready-To-Go; Amersham Pharmacia Biotech, Piscataway, NJ). This radioactive probe was hybridized to the alkali-denatured phage particles on the membrane according to the procedure of Sambrook et al. ³⁵ Phage particles from positive plaques were excised and allowed to infect E. coli strain XLOLR in the presence of helper phage to obtain a clone with a phagemid bearing a 5.88-kb genomic fragment. 

The cloned putative protease IV gene was subcloned from the phagemid into the plasmid vector pUCP20 ³⁶ by digesting with EcoRI (20 U), isolating the 3.3-kb band, and ligating it to predigested pUCP20 DNA (0.5 μg). The ligated DNA was transformed into E. coli DH5α cells (100 μL; Invitrogen-Gibco, Gaithersburg, MD) and clones selected on LB agar with carbenicillin (100 μg/mL). The chosen clone was designated DH5α/pUCP-PIV. 

Liquid Phase Colorimetric Assay

For E. coli DH5α/pUCP-PIV (with an insert) or DH5α/pUCP20 (vector without insert), supernatants were obtained from 500-mL cultures that were grown for 40 hours in LB broth with carbenicillin (100 μg/mL). For Pseudomonas isolates, supernatants were obtained from 10-mL cultures grown for 40 hours in M9 medium with supplements as described previously. ³² All supernatants were filtered through a 0.22-μm porosity filter (Corning Inc., Corning, NY) to remove any remaining bacteria, and the filtrate was concentrated with an ultrafiltration cell (10,000 kDa, YM10 filter; Amicon, Beverly, MA). Protease IV activity was assayed using tosyl-gly-pro-lys-p-nitroanilide, a colorimetric assay plasmin-specific chromogenic substrate (Chromozym PL; Sigma-Aldrich) as described by Caballero et al. ³⁷ For quantitation of activity, the change in optical density in Chromozym PL reactions was determined every 60 minutes and the units of protease IV activity per milligram of total protein calculated as described previously. ²¹  

SDS-Polyacrylamide Gel Electrophoresis

Denatured proteins and culture supernatants were electrophoresed through 10% or 12% SDS-polyacrylamide gels under reducing (1% β-mercaptoethanol) conditions, according to the method described by Sambrook et al. ³⁵  

Western Blot Analysis

Western blot analyses were performed as previously described. ³⁷ The blots were incubated with rabbit polyclonal anti-protease IV antibody (1:2000) at 4°C (Thibodeaux BA, et al. IOVS 2001;42:ARVO Abstract 1366), and developed with an alkaline phosphatase detection system (Sigma-Aldrich). Alternatively, blots were analyzed using the enhanced chemiluminescence (ECL) detection system (Amersham Pharmacia Biotech, Inc.), according to the instructions of the manufacturer. 

Purification of Recombinant Protease IV

Protease IV was purified from E. coli culture supernatant by ionic exchange and molecular sieve chromatography as described by Engel et al. ³²  

Pulse-Field Gel Electrophoresis Analysis

Approximately 3 to 5 × 10⁸ bacterial cells of strains PAO1 and PA103-29 were embedded in 0.75% agarose blocks (InCert; FMC Bioproducts, Rockland, ME) and lysed with lysozyme and detergent, and the DNA was digested with the restriction enzyme SpeI. ³⁸ Pulse-field gel electrophoresis (PFGE) was performed on a commercial system (Chef Mapper Xa; Bio-Rad, Richmond, CA) with autoalgorithms set to separate 25- to 550-kb fragments in 12 hours (Hobden J, Wayne State University Medical School, Detroit, MI, personal communication, March 2000). Southern transfer to a membrane was accomplished according to the manufacturer’s instructions (Nytran; Schleicher & Schuell). The probe that was used to identify the protease IV gene was synthesized as described in the section on Cloning the Putative Protease IV Gene. 

DNA Sequencing

The 5.88-kb insert of PA103-29 genomic DNA in the original phagemid clone was sequenced using a terminator cycle sequencing kit (Thermo Sequenase Radiolabeled Terminator Cycle Sequencing; United States Biochemicals [USB], Cleveland, OH), according to the manufacturer’s instructions. 

Analysis of DNA and Protein Sequence Data

DNA and protein sequence data were analyzed with the Expert Protein Analysis System (ExPASy) proteomic server of the Swiss Institute of Bioinformatics, Geneva, Switzerland (http://www.expasy.ch). 

Distribution of the Protease IV Gene

Genomic DNA from all strains of Pseudomonas tested was purified with a genomic DNA preparation kit (Qiagen, Valencia, CA). PCR analysis was conducted as described earlier for synthesis of the protease IV gene probe, using three sets of primers derived from the sequence of the protease IV gene from P. aeruginosa strain PA103-29: set A (forward primer; 5′-ACCGCGCGCTTCCCGATCAGC-3′; reverse primer, 5′-TCAGGGTGCGAAGTAGCGGGAG-3′); set B (forward primer; 5′-GCCGGCTACCGCGACGGCTTC-3′, reverse primer; same as set 1), and set C (forward primer; 5′-CTGGAACTCAAGCGCACTCCG-3′, reverse primer; same as sets 1 and 2). A 10-μL aliquot from each PCR reaction was electrophoresed through a 1% agarose gel in TAE buffer (0.04 M Tris-acetate, 0.001 M EDTA) and stained with ethidium bromide. 

Zymography of Extracellular Proteases

Zymography was conducted as described previously. ³⁷  

Allelic Replacement Mutagenesis

The phagemid clone containing the protease IV gene was digested with FseI/BglII, which eliminated 98% of the protease IV gene, and a tetracycline resistance cassette was inserted into the remaining protease IV gene sequence. This construct plus flanking Pseudomonas DNA sequences was digested with SmaI, ligated into the gene replacement vector pEX100T and introduced into PA103-29 by triparental mating, as described by Pillar et al. ³⁹ Clones resistant to tetracycline that grew in the presence of 10% sucrose were selected and tested by PCR, Western blot, and a colorimetric assay using a chromogenic substrate (Chromozym PL; Sigma-Aldrich) to ensure that the protease IV gene had been inactivated. 

Pseudomonas Keratitis

New Zealand White rabbits (2.0–3.0 kg) used in these studies were maintained according to the guidelines put forth in the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Rabbits were anesthetized and injected intrastromally with 1000 colony-forming units (CFU) of log phase bacteria, as described previously. ⁴⁰ At 20, 24, and 27 hours postinfection (PI), rabbit eyes underwent slit lamp examination (SLE) scoring on a scale of 0 for normal and 28 for the eye with maximum inflammation. No eyes were allowed to go beyond an SLE score of 20. At 24 or 27 hours PI, rabbits were killed, and the average log number of CFU per cornea per group (n = 6 eyes per group) was determined as described previously. ⁴⁰  

Statistical Analysis

Probabilities were determined for CFU and SLE data as previously described, ⁴⁰ and P ≤ 0.05 was considered significant. 

Results

The Protease IV Gene

Based on the amino acid sequence of purified protease IV, the putative protease IV gene was identified and cloned. The resultant plasmid, with the cloned 3.38-kb fragment, pUCP-PIV, was expressed in E. coli, and the extracellular protease activity of this E. coli was compared with that of protease IV from P. aeruginosa. Both enzymes shared the following properties: (1) digestion of the chromogenic substrate (Chromozym PL; Sigma-Aldrich) in the presence of EDTA but not of TLCK (N-α-p-tosyl-l-lysine chloromethyl ketone; Fig. 1A ); (2) a positive reaction with antibody to protease IV on Western blot with a band at approximately 30 kDa (Fig. 1B) , and (3) an N-terminal amino acid sequence of A-G-Y-R-D-G-F (data not shown). No such protease activity was detected in cultures of E. coli with the vector plasmid without the protease IV insert. 

The sequence of the cloned protease IV gene from PA103-29 (GenBank Accession Number AY062882; http://www.ncbi.nlm.nih.gov/Genbank; provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD) was compared to the published sequence of the corresponding region of PAO1 (Fig. 2) . ⁴¹ The two DNA sequences exhibited 98.5% homology overall. The differences in the two genes include 10 of 199 bases in the promoter region (5%), 21 of 1389 bases in the open reading frame (1.5%), one amino acid in the signal sequence (23), and three amino acids in the propeptide domain (57, 102, and 137). There were no amino acid differences between these strains in the secreted and enzymatically active protease IV molecule. 

The protease IV genes of PA103-29 and PAO1 were found by PFGE to be located on chromosomal fragments having the same electrophoretic mobility (Fig. 3) . 

Distribution of the Protease IV Gene in Pseudomonas aeruginosa and Other Pseudomonas Species

The genomic DNA of 24 P. aeruginosa strains, as well as six non-aeruginosa strains of Pseudomonas, was analyzed by PCR using three different combinations of primers, each specific for a different region of the protease IV gene. As shown in Figure 4 , all P. aeruginosa strains examined were found to contain the protease IV gene. Furthermore, each of the three primer combinations yielded DNA fragments of the correct size, as deduced from the sequence data of the protease IV gene from P. aeruginosa strains PAO1 and PA103-29 (set A; 1047 bp, set B; 756 bp, and set C; 381 bp). In contrast, the non-aeruginosa organisms, P. alcaligenes, P. mendocina, P. putida, P. otitidis, and P. stutzeri (strains 17001 and 17030), were found to lack PCR products or produced products of incorrect length relative to what was expected according to the sequence data. 

Production of Protease IV by P. aeruginosa Strains

Concentrated culture supernatants from 24 strains of P. aeruginosa were examined by Western blot analysis with anti-protease IV antibody. Protease IV was detected in the culture media of 23 of 24 strains tested (Fig. 5) . In all cases, and under reducing conditions, protease IV migrated at the apparent molecular size of 31 kDa. The only P. aeruginosa strain without the 31-kDa band was PA103-API (Fig. 5 , lane 4), a strain known to have a mutation that reduces the protease IV activity by 90%. ^{42 43} None of the six culture supernatants from the non-aeruginosa strains examined contained protease IV (data not shown). 

Protease IV Activity of Pseudomonas Strains

The P. aeruginosa strains analyzed differed in the amounts of protease IV (units) produced per milligram of total extracellular protein (Fig. 6) . P. aeruginosa strains PA103-29, PAO1, ALC1, ALC2, ALC3, PA23, 70, 13004, 13010, 13036, 30009, 30099, 30244, 51037, 51003, 51005, and 51131 produced from 2.3 to 50 × 10⁻³ U/mg of protein, whereas strains PA103, BR-1, 178, 13014, 30132, and 51036 produced from 80 to 221.5 × 10⁻³ U/mg of protein. The mutant strain PA103-AP1 had no detectable enzyme activity. No protease IV activity was detected in the culture supernatants of the five non-aeruginosa species of Pseudomonas. 

Zymography of Protease Activity

The activities of the major P. aeruginosa extracellular proteases (i.e., protease IV, elastase B, and alkaline protease) were visualized by gelatin zymography (Fig. 7) . Protease IV was present and active in the extracellular medium of all the P. aeruginosa strains (except PA103-AP1). Protease IV, as described previously, ³⁷ aggregates under nonreducing conditions into a high molecular mass complex (>200 kDa) that remains near the top of the separating gel. Unlike protease IV, elastase B and alkaline protease production was variable among the P. aeruginosa strains. Under these conditions, some strains failed to produce alkaline protease (PA103-29, PA103, PA103-AP1, PA178, 51036, 51037, and 51131), and some did not produce elastase B (PA103-29, PA103, PA103-AP1, BR-1, PA13036, 30099, 51036, 51037, 51003, 51005, and 51131). There were also bands of unknown proteinase activity secreted by strains PA23, 178, 13004, and 30009 that could be uncharacterized proteases or could be partial degradation of known extracellular proteases. 

Role of Protease IV

That protease IV was present in all strains of P. aeruginosa suggests that this enzyme was conserved and probably provided an important function to the organism. To test this point, a PA103-29 deletion mutant of the protease IV gene was constructed. PCR analysis of the genome and Western blot and colorimetric assays of the culture supernatant of the mutated strain demonstrated that the protease IV gene and its product were no longer present (Fig. 8) . Zymography demonstrated that the parent strain produced protease IV whereas the protease IV–deficient mutant produced alkaline protease, an enzyme not previously noted to be produced by PA103-29 (Fig. 9) . 

The abilities of the parent PA103-29 and its protease-IV–deficient mutant strain to replicate in the rabbit cornea and to mediate corneal damage were compared. The parent and mutant each grew to approximately 7 log CFU per cornea (P = 0.859) by 27 hours PI (Table 1) . The parent produced SLE scores at 20, 24, and 27 hours PI that were significantly higher than those of the protease IV–deficient mutant (P = 0.0249, 0.0183, and 0.0356, respectively). 

To confirm the role of protease IV in corneal virulence, a plasmid coding for active protease IV (pUCP-PIV) was transformed into the protease IV–deficient mutant. This transformed strain, the protease IV rescued strain, was then tested in vivo for its growth and corneal virulence. The rescued and parent strains produced equivalent SLE scores (P = 0.505), having scores of approximately 9.0 by 24 hours PI (Table 2) . The demonstration of restored virulence for the rescued strain occurred despite its slightly diminished ability to grow in the cornea. The rescued strain grew to only 7.1 log CFU, whereas the parent strain grew to 7.29 log CFU (P = 0.045). 

Discussion

The present study demonstrates that the protease IV gene is commonly found and expressed in strains of P. aeruginosa and that the protein is processed and secreted as an extracellular enzyme. PCR, Western blot analysis, zymography, and a colorimetric assay demonstrated the protease IV gene and its product in all strains of P. aeruginosa tested, except PA103-AP1, a mutant strain with reduced protease activity. ^{42 43} The production of protease IV by all strains is in contrast to alkaline protease and elastase B, which were produced only by a subset of P. aeruginosa strains. The production of protease IV by all P. aeruginosa strains is also in contrast to the six strains of non-aeruginosa species of Pseudomonas that showed no protease IV by PCR, Western blot, and enzyme assays. 

The selective expression of virulence genes has been noted by several researchers. ^{44 45 46} Lomholt et al. ⁴⁵ reported that despite the fact that the lasA and lasB genes were present in all 145 P. aeruginosa isolates examined, 17 isolates did not exhibit elastase A activity, whereas 4 isolates failed to express elastase B. They suggest that because virulence factors are regulated by quorum sensing, it may be that in a particular niche certain gene products are not needed, which may lead to clonal selection for nonexpression. One must be aware, however, that what triggers the expression of virulence genes in vitro may be different from what happens in vivo, and certain strains may require additional signals or interactions that may not be present in vitro for expression of those genes. 

The production of protease IV by all P. aeruginosa strains suggests that this enzyme has an important role in the life of this organism. Inactivation of the protease IV gene was accompanied by the production of alkaline protease, an enzyme not normally produced by strain PA103-29. Gambello and Iglewski ⁴⁷ demonstrated that strain PA103 does not have the positive regulatory gene needed to allow expression of elastase and alkaline protease. The expression of alkaline protease was unexpected and suggests that bacteria possessing the ability to express another protease, such as alkaline protease, were the ones capable of growth once protease IV production was no longer possible. In no case was there detection of a clone deficient in all the major proteases (alkaline protease, protease IV, elastase B, and elastase A). 

The specific loss of protease IV production by an engineered mutation was accompanied by a significant decrease in corneal virulence. This confirms earlier studies correlating protease IV production to virulence in animal models of keratitis. ^{21 30} The results also agree with the finding that the production of protease IV, a serine protease, in Pseudomonas putida bestowed corneal virulence on this organism, which normally lacks such virulence. ^{40 48} The decrease in virulence mediated by the protease IV–deficient mutant was significant even though the mutant produced significant amounts of alkaline protease. This finding is in agreement with that of Pillar et al., ³⁹ who found that the loss of alkaline protease production was not accompanied by a significant decline in corneal virulence. However, the alkaline protease–deficient mutant analyzed by Pillar et al. still produced protease IV, an enzyme that could have maintained the corneal virulence of the mutant. Still unresolved is the role of specific proteases in corneal virulence. 

The detection of the protease IV gene in all the P. aeruginosa strains analyzed by PCR and the uniformity of the PCR generated fragments using three different primer sets suggest that this gene is conserved. This finding agrees with the results of Wilderman et al., ³³ who showed that the supernatant of several strains of P. aeruginosa contained a protein with electrophoretic mobility equivalent to protease IV. These findings also agree with the report of Twining et al. that described protease IV activity in several isolates of P. aeruginosa (Twining SS, et al. IOVS 1993;34:ARVO Abstract 713). 

Differences in protease IV activity in the culture supernatants of P. aeruginosa strains could relate to rates of gene transcription or to factors other than the control of gene expression. Variations in the sequences of PA103-29 and PAO1 were greatest in the promoter areas of the gene (approximately 5%). Overall, the DNA sequence of the protease IV gene was conserved in the open reading frame for the two strains (98.5% identity), and the gene was located in the same DNA fragment in each strain. However, strains could differ in their ability to secrete this protease, or the proteases produced could differ in their enzymatic activity. Both sequence analysis of protease IV genes and enzyme kinetics measurements of protease purified from more strains are needed to understand these differences in the production of enzyme activity. 

The non-aeruginosa strains studied did not have a DNA sequence that could be recognized as protease IV and nor did they have the antigenic and enzymatic properties of protease IV. The protease IV gene could be absent because of the acquisition of virulence factors by P. aeruginosa, which is thought to have been a gradual process involving the lateral transfer of genes from other species by transposable elements, bacteriophage, and/or plasmids. ^{49 50 51} The protease IV gene shares sequence similarities with other bacterial proteases—in particular, endoproteinase Arg-C and lysyl endopeptidase of Lysobacter enzymogenes and protease I of Achromobacter lyticus. ⁴⁸ Non-aeruginosa strains apparently did not acquire the protease IV gene, although they may have many other genes in common with P. aeruginosa. Non-aeruginosa strains of Pseudomonas without protease IV are not recognized as ocular pathogens. Traidej et al. ⁴⁰ have demonstrated that expression of protease IV in P. putida confers corneal virulence. 

Protease IV has been shown to be a lysine-specific protease enzymatically distinct from other proteases produced by P. aeruginosa. ^{32 37} This enzyme contributes to virulence in ocular models of infection ²¹ and can damage ocular tissue and host defense proteins. ^{31 32 40} The present study demonstrates that protease IV is an extracellular product of all wild-type strains of P. aeruginosa tested and its loss resulted in a decline in corneal virulence. The conservation of this protease suggests that it is active in a critical function of the organism in nature and that it is commonly available to contribute to ocular virulence. 

 Supported by National Eye Institute grant EY12961.

 Submitted for publication September 23, 2003; revised October 2, 2003; accepted October 5, 2003.

 Disclosure: A. Caballero, None; B. Thibodeaux, None; M. Marquart, None; M. Traidej, None; R. O’Callaghan, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

(A) Colorimetric peptide assay of native protease IV from P. aeruginosa strain PA103-29, and concentrated culture supernatants from clones DH5α/pUCP20 and DH5α/pUCP-PIV. Purified protease IV (100 μg/mL) or equal protein amounts of culture supernatant from clones DH5α/pUCP20 and DH5α/pUCP-PIV (10 μL total volume) were incubated at 37°C for 1 hour with a chromogenic substrate in reaction buffer in the presence or absence of EDTA (100 mM) or TLCK (2 mM). Cleavage of the chromogenic peptide was measured by the increase in OD₄₁₀, with a plate reader. The data presented are from six determinations, and the error bars represent the mean ± SEM. Concentrated culture supernatant in reaction buffer was used for background subtraction. PIV, protease IV; DH5α/pUCP20, culture supernatant from parent vector without insert; DH5α/pUCP-PIV, culture supernatant from vector with protease IV gene. (B) Western blot analysis of native protease IV from P. aeruginosa strain PA103-29 and concentrated culture supernatants from clones DH5α/pUCP20 and DH5α/pUCP-PIV. Purified protease IV (3 μg) or culture supernatants (30 μL) from clones DH5α/pUCP20 and DH5α/pUCP-PIV containing equal amounts of protein were electrophoresed under reducing conditions in a 12% acrylamide gel. The proteins were transferred to a polyvinylidene (PVDF) membrane and incubated overnight at 4°C with a 1:2000 dilution of polyclonal rabbit anti-PIV antisera. The protein bands were visualized with alkaline-phosphatase–conjugated goat anti-rabbit antibody (1:8000). Molecular masses are in kilodaltons. Lane 1: protease IV and its breakdown product; lane 2: culture supernatant from DH5α/pUCP20 (vector without insert); lane 3: culture supernatant from DH5α/pUCP-PIV (vector with protease IV gene).

Figure 2.

 

Sequence analysis of the protease IV gene from P. aeruginosa strain PA103-29. The protease IV gene was sequenced and translated into protein. Analysis of the protein sequence was conducted online (ExPASy). Dashed underscore: signal sequence; bold amino acids, transmembrane domain; arrow: start of mature protease; solid underscore, serine protease active site sequential motif; boxed nucleotides or amino acids, locations where PA103-29 differs from PAO1 at the genomic or protein level, with the differing nucleotide in PAO1 on top of the nucleotide box, and the differing amino acid on the bottom of the amino acid box.

Figure 3.

 

PFGE and Southern blot analysis of genomic DNA from PAO1 and PA103-29. Genomic DNA was digested with SpeI, and the fragments were resolved in a 1% agarose gel by PFGE (A). After transfer onto a nylon membrane, the genome fragments were probed with a radioactive probe specific for the protease IV gene (B). Lane 1: PAO1; lane 2: PA103-29.

Figure 4.

 

PCR analysis of genomic DNA from P. aeruginosa and other Pseudomonas species for the presence of the protease IV gene. Genomic DNA (500 ng) from 24 P. aeruginosa strains and six non-aeruginosa strains of Pseudomonas was used as a template in PCR reactions containing three different 5′ primer combinations specific for three regions of the protease IV gene; Primer sets A, B, and C yielded products equivalent to three-quarters, one half, and one quarter of the gene, respectively. Numbers are those above each primer set: 1, PA103-29; 2, PAO1; 3, PA103; 4, PA103-AP1; 5, BR-1; 6, PA ALC1; 7, PA ALC2; 8, PA ALC3; 9, PA23; 10, PA70; 11, PA178; 12, PA13004; 13, PA13010; 14, PA13014; 15, PA13036; 16, PA30009; 17, PA30099; 18, PA30132; 19, PA30244; 20, PA51036; 21, PA51037; 22, PA51003; 23, PA51005; 24, PA51131; 25, P. alcaligenes; 26, P. mendocina; 27, P. putida; 28, P. otitidis 10150; 29, P. stutzeri 17001; 30, P. stutzeri 17030. The molecular weight marker is φX174 DNA-HaeIII digest; from top to bottom: 1353, 1078, 872, 603, and 310 bases.

Figure 5.

 

Western blot analysis of concentrated culture supernatants from P. aeruginosa strains for the presence of protease IV. Twenty microliters of concentrated culture supernatants (5 mg/mL) from 24 P. aeruginosa strains were examined with rabbit polyclonal anti-PIV antibody, using the ECL detection system. Lane 1: PA103-29; lane 2: PAO1; lane 3: PA103; lane 4: PA103-AP1; lane 5: BR-1; lane 6: PA ALC1; lane 7: PA ALC2; lane 8: PA ALC3; lane 9: PA23; lane 10: PA70; lane 11: PA178; lane 12: PA13004; lane 13: PA13010; lane 14: PA 13014; lane 15: PA13036; lane 16: PA30009; lane 17: PA30099; lane 18: PA30132; lane 19: PA30244; lane 20: PA51036; lane 21: PA51037; lane 22: PA51003; lane 23: PA51005; lane 24: PA51131.

Figure 6.

 

Protease IV activity in concentrated culture supernatants from P. aeruginosa strains, corresponding to the lanes in Figure 5 . The protease IV activity of concentrated culture supernatants from 24 strains of P. aeruginosa equalized for protein concentration were examined for protease IV activity, by using a chromogenic substrate. Ten microliters of each supernatant (2 mg/mL) was incubated at 37°C for 6 hours in the reaction mix in the presence of 100 mM EDTA. The activity was recorded every hour and expressed as units of enzyme activity per milligram total protein. Bars represent the mean of protease activity units of six replicates.

Figure 7.

 

Total gelatinase activity of concentrated culture supernatants from P. aeruginosa strains. Total gelatinase activity of culture supernatants from 24 strains of P. aeruginosa, corresponding with the lanes in Figure 5 , was analyzed in a 10% gelatin zymogram. Five microliters of culture supernatants, equalized for protein concentration (2 mg/mL), were loaded in each lane. Purified protease IV (PIV), elastase B (Las B), and alkaline protease (AP), were used as molecular mass markers.

Figure 8.

 

Analysis of the PA103-29 protease-IV–deficient mutant. The PA103-29 protease-IV–deficient mutant was analyzed by PCR (A), Western blot (B), and colorimetric enzyme assay (C). (A) Genomic DNA from the parental strain PA103-29 and the protease IV deletion mutant PA(IV−) were analyzed by PCR with primers specific for the lasA, PIV, lasB, and apr genes. Primers for protease IV were derived from our PAO1. The remainder of the primers were designed using sequence information available in GenBank. Lane A: lasA; lane B: PIV; lane C: lasB; lane D: apr. (B) Concentrated culture supernatants from the parental strain PA103-29 and the protease IV deletion mutant strain were examined with rabbit polyclonal anti-PIV antibody using an alkaline phosphatase detection system. Lane 1: protease IV; lane 2: protease IV deletion mutant; lane 3: PA103-29; lane 4: protease IV. (C) Concentrated culture supernatants from the parental strain PA103-29 and the protease IV deletion mutant strain PA(IV−) were tested for their ability to cleave the chromogenic substrate. Cleavage of the chromogenic peptide was measured by the increase in OD₄₁₀. Purified protease IV was used as a control.

Figure 9.

 

Gelatinase activity of concentrated culture supernatants from the protease IV deletion mutant PA(IV−) and its parent strain PA103-29. Total gelatinase activity of culture supernatants from PA103-29 and the protease IV deletion mutant PA(IV−) were analyzed in a 10% gelatin zymogram. Twenty microliters of culture supernatants equalized for protein concentration (2 mg/mL) were loaded in each lane. Purified alkaline protease, elastase B, and protease IV were used as molecular mass markers. Lane 1: alkaline protease; lane 2: elastase B; lane 3: protease IV; lane 4: PA103-29; lane 5: PA(IV−).

Table 1.

 

SLE Scores of Corneas Infected with the Parental Strain PA103-29 and the Protease IV Deletion Mutant PA(IV−)

Strain	SLE*					CFU, †
		20 h	24 h	27 h
PA103-29	4.30 ± 0.42	8.86 ± 1.04	12.38 ± 1.19	7.07
PA(IV−)	2.61 ± 0.19	5.42 ± 0.56	9.88 ± 1.03	7.04
P	0.0249	0.0183	0.0356	0.859

*  Mean SLE score ± SEM; n = 6 eyes per strain.

†  Mean CFU ± SEM expressed as log10 at 27 hours after infection.

Table 2.

 

SLE Scores of Corneas Infected with the Parental Strain PA103-29 and the Protease IV Rescued Strain PA(R)

Strain	SLE*			CFU, †
		20 h	24 h
PA103-29	6.70 ± 0.75	9.56 ± 0.76	7.29
PA(R)	6.66 ± 0.57	8.94 ± 0.56	7.10
P	0.9659	0.5052	0.045

*  Mean SLE score ± SEM; n = 6 eyes per strain.

†  Mean CFU ± SEM expressed as log10 at 24 hours after infection.

The authors thank Judy M. Moreau, Steven Delvisco, Kiana Nelson, Julian Reed, Dalia Girgis, and Joseph Dajcs for technical assistance.