The Papain Inhibitor (SPI) of Streptomyces mobaraensis Inhibits Bacterial Cysteine Proteases and Is an Antagonist of Bacterial Growth

Abstract

A novel papain inhibitory protein (SPI) from Streptomyces mobaraensis was studied to measure its inhibitory effect on bacterial cysteine protease activity (Staphylococcus aureus SspB) and culture supernatants (Porphyromonas gingivalis, Bacillus anthracis). Further, growth of Bacillus anthracis, Staphylococcus aureus, Pseudomonas aeruginosa, and Vibrio cholerae was completely inhibited by 10 μM SPI. At this concentration of SPI, no cytotoxicity was observed. We conclude that SPI inhibits bacterial virulence factors and has the potential to become a novel therapeutic treatment against a range of unrelated pathogenic bacteria.

TEXT

Papain (EC 3.4.22.2) belongs to clan A of the cysteine protease enzymes and is the eponym for the C1 family of proteases, which comprises a large number of endopeptidases, although fewer exopeptidases. All members of the papain family contain cysteine and histidine at their active site, forming a catalytic dyad. Papain and related cysteine proteases are widely distributed in the plant kingdom and are believed to act as virulence/defense factors for both hosts and pathogens (1). Cysteine proteases are also found in bacteria and are known to be virulence factors involved in bacterial pathogenicity (2). Further, papain-like hydrolases are involved in peptidoglycan turnover in both Gram-negative and Gram-positive bacteria (3, 4), with disruption of amide-hydrolyzing autolysins in Bacillus subtilis, leading to defective cell wall division and thereby affecting bacterial viability (5). More recently, the growing resistance of microorganisms to conventionally used antibiotics has meant that cysteine proteases have attracted attention as possible targets for antimicrobial therapy (6). Indeed, several cysteine proteases have been identified as potential targets for such therapy, including the papain-like staphopains A and B from Staphylococcus aureus (7), streptopain (exotoxin B) from Streptococcus pyogenes (3), the gingipains RGP and KGP from Porphyromonas gingivalis (2), PrtH (FDF) from Tannerella forsythia (8), and YopT from Yersinia enterocolitica (9).

Recently, we described a novel, heat-resistant protein (Streptomyces papain inhibitor [SPI]) from Streptomyces mobaraensis that inhibited the activity of the cysteine protease papain, as well as (to a lesser extent) the activities of cysteine protease bromelain and the serine protease trypsin in the nanomolar range (10). In the present report, we describe investigations into the inhibitory activities of SPI on bacterial cysteine proteases, i.e., potential virulence factors, and on the growth capabilities of a range of bacterial pathogens. Our results indicate that SPI has the ability to inhibit secreted bacterial cysteine proteases, as well as bacterial growth, and represent a first step in confirming SPI as a potential broad-spectrum antibacterial agent.

SPI.

Preparation of highly purified SPI was performed as described previously (10). Briefly, Streptomyces mobaraensis DSM 40847^T (IPCR 16-22) was cultured at 42°C for 30 h. The heated cell-free supernatant (70°C, 30 min) was separated consecutively by Fractogel EMD trimethylaminoethyl (adsorption at pH 9.0, elution at pH 6.0) followed by two Fractogel EMD SO₃⁻ chromatographies, both performed at pH 4.0.

Protease FRET assay.

A specialized fluorescent resonance energy transfer (FRET) reporter assay was used to determine the effect of SPI on bacterial proteolytic activity, as previously described by Kaman et al. (11). Basically, the assay utilizes a short d-amino acid peptide probe as a substrate for proteolytic enzymes. This short peptide probe is linked to both fluorescent reporter and quencher molecules, such that cleavage of the short peptide probe by specific proteases results in the spatial separation of reporter and quencher molecules, with a concomitant increase in the fluorescent signal upon excitation of the probe. For measurement of proteolytic activity, SPI concentrations between 10 and 1,000 nM were used and tested in the presence of 16 μM FRET substrate. The peptide sequence of the FRET substrate used varied per organism and was based on data present in previously published articles (Table 1). For analysis, purified enzymes, or 0.2-μm-filtered bacterial culture supernatants, were incubated with SPI and FRET probe at 37°C for 60 min in a total volume of 50 μl. Residual proteolytic activity was determined in 2-min time intervals during 60 min of incubation at 37°C. Fluorescence intensity was continuously measured at 530 nm (excitation wavelength of 485 nm) by using a FluoStar Galaxy spectrometer (BMG laboratories, Offenburg, Germany). Negative controls included reaction mixtures prepared in the absence of SPI or brain heart infusion broth (BHI). After each measurement the relative fluorescence (RF) per minute was calculated, and the RF/minute value of the sample without SPI (control) was corrected to 100%.

Table 1.

FRET-based peptide substrates used in this study

Peptide source	Organism	Protease	Substrate	Sequencec
Purified enzyme	Staphylococcus aureus	Staphopain Ba	BikKam17	FITC-Phe-Arg-KDbc
	Clostridium histolyticum	Collagenaseb	BikKam18	FITC-Ala-Ala-Gly-Pro-Ala-Ala-KDbc
Culture supernatant	Bacillus anthracis	Unknown	BikKam1	FITC-Leu-DLeu-KDbc
	Porphyromonas gingivalis	Gingipain K/Ra	BikKam13	FITC-Arg-DArg-KDbc

^aCysteine protease.

^bMetalloprotease.

^cFITC, fluorescein isothiocyanate; KDbc, N^ε-Lys-DABCYL; DABCYL, 4-(N,N-dimethylamino)azobenzene-4′-carboxylic acid.

Inhibition of cysteine protease activity by SPI.

The ability of SPI to inhibit cysteine protease activity was tested using purified enzyme, staphopain B (SspB; BioCentrum, Krakow, Poland), which is a member of the papain clan CA cysteine proteases. A purified metalloprotease, collagenase derived from Clostridium histolyticum (Sigma, Zwijndrecht, The Netherlands), was used as a non-cysteine protease control. The enzymes were used in the FRET assay at a final concentration of 60 μg/ml and 80 μg/ml in phosphate-buffered saline (PBS), respectively, in the presence or absence of SPI (1,000 nM). All experiments were performed in triplicate.

Inhibition of bacterial culture supernatant protease activity by SPI.

The inhibitory effect of SPI was tested on 0.2-μm-filtered culture supernatants of Porphyromonas gingivalis and Bacillus anthracis after overnight culture in BHI. The supernatants were prepared as described by Kaman et al. (11). All experiments were performed in triplicate.

Inhibition of bacterial growth.

The antimicrobial activity of SPI was evaluated against Bacillus anthracis Vollum strain (ATCC 14578), Staphylococcus aureus USA300, Vibrio cholerae serotype O1 (ATCC 14035), and Pseudomonas aeruginosa PAO1 (ATCC 15692). All bacteria were grown overnight in 5 ml BHI broth at 35°C. The following day, bacteria were cultured to early-logarithmic phase by transferring 100 μl of the overnight culture into 5 ml BHI medium, followed by incubation for 2 h at 35°C with shaking at 200 rpm. Subsequently, 100 μl of the early-logarithmic culture (1:50 in BHI) was incubated with 100 μl serially diluted SPI (in BHI) or acetate buffer (pH 4.0) at 35°C and with shaking at 200 rpm. The growth rate was then measured spectrophotometrically at 600 nm at 30-min time intervals for 12 h by using the BioScreen apparatus (Thermo Fisher Scientific, Breda, The Netherlands). After incubation, the samples were plated onto Columbia blood agar plates to determine whether the effect of SPI on bacterial growth was bacteriostatic or bactericidal. Experiments were performed in triplicate at the biosafety level 3 facility of TNO Defense, Security and Safety (Rijswijk, The Netherlands).

Cytotoxicity assay.

The cytotoxicity of SPI on RAW264.7 cells was evaluated in a lactate dehydrogenase (LDH) assay (Roche, Almere, The Netherlands). Cells were incubated for 2 h with culture medium (low control [LC]), 1% Triton X-100 (high control [HC]), SPI, or acetate buffer at 37°C and 5% CO₂. After the incubation, 100-μl aliquots of the supernatants were taken and used in the LDH assay according to the manufacturer's description.

Inhibition of cysteine protease activity by SPI.

A concentration of 1,000 nM SPI was sufficient to inhibit the proteolytic activity of purified cysteine protease SspB by approximately 50% (Fig. 1). In contrast, the metalloprotease control enzyme, collagenase from C. histolyticum, remained active at approximately 95% of the level of activity of the no-SPI control. Interestingly, SPI had a significant inhibitory effect on both types of enzyme activity, though inhibition of SspB reached much greater significance.

Fig 1.

Influence of the papain inhibitor SPI from Streptomyces mobaraensis on the proteolytic activity of purified SspB (S. aureus) and collagenase (C. histolyticum) enzymes. A 16 μM peptide substrate was incubated with 60 μg/ml SspB or 80 μg/ml collagenase in the absence or presence of 1,000 nM SPI. Fluorescence was measured at 37°C for 60 min. Error bars represent the standard errors of the means from 3 independent experiments. **, P < 0.01; *, P < 0.05.

Inhibition of bacterial culture supernatant protease activity by SPI.

Observations on the inhibitory effects of SPI on P. gingivalis and B. anthracis overnight culture supernatants indicated that a significant decrease in supernatant protease activity occurred (P < 0.05) when we used an SPI concentration of 800 nM in culture supernatants. However, this effect was much more pronounced for P. gingivalis than B. anthracis, with an SPI concentration of 10 nM achieving a significant inhibition of protease activity. The mean residual protease activity of P. gingivalis and B. anthracis substrates after 1 h of incubation with 1,000 nM SPI was 3% and 45%, respectively (Fig. 2). The 50% inhibitory concentrations (IC₅₀s) for P. gingivalis supernatant protease was 40 nM SPI and 1,000 nM SPI for B. anthracis. Additionally, a dose-response curve was observed in both sets of experiments.

Fig 2.

Influence of the papain inhibitor SPI from Streptomyces mobaraensis on the proteolytic activity of P. gingivalis and B. anthracis culture supernatants. A 16 μM peptide substrate was incubated with 40.2 μl culture supernatant and various SPI concentrations. Fluorescence was measured at 37°C for 60 min. Error bars represent the standard errors of the means from 3 independent experiments. ***, P < 0.001; **, P < 0.01; *, P < 0.05.

Inhibition of microbial growth.

Figure 3 demonstrates that the addition of 10 μM SPI to bacterial cells in the early-logarithmic growth phase was able to completely inhibit the growth of B. anthracis, S. aureus, V. cholerae, and P. aeruginosa. Further, the growth of B. anthracis, S. aureus, and V. cholerae was also delayed and reduced at a concentration of 5 μM SPI. For all organisms, a dose-dependent effect was observed between SPI concentration and inhibition of bacterial growth, although this effect was not linear in scale. Culture of the bacteria surviving the 10 μM SPI experiments on Columbia blood agar generated visible growth for B. anthracis, S. aureus, and P. aeruginosa after 72 h of incubation, although no growth was observed for V. cholerae. These results indicate that SPI may generate either a bacteriostatic or a bactericidal effect on bacterial growth, dependent on the bacterial species.

Fig 3.

Inhibition of bacterial growth by the papain inhibitor SPI from Streptomyces mobaraensis. (A) Bacillus anthracis; (B) Staphylococcus aureus; (C) Pseudomonas aeruginosa; (D) Vibrio cholerae. Bacteria were cultured at 35°C in the presence of SPI at 0 μM, 1.25 μM, 2.5 μM, 5 μM, and 10 μM. Error bars represent the standard errors of the means from 3 independent experiments.

Cytotoxicity of SPI.

SPI was evaluated for its toxicity to RAW264.7 cells at similar concentrations as used in the growth assays. The concentration at which inhibition of bacterial growth was observed, 10 μM, appeared to be noncytotoxic. Also, the buffer in which SPI was diluted showed no cytotoxicity (all data not shown).

During submerged culture, the bacterium Streptomyces mobaraensis secretes inhibitory proteins that are active against a variety of endoproteases, including the ubiquitous cysteine protease papain (10). The isolation of one of these endoprotease inhibitors, the Streptomyces papain inhibitor, or SPI, allowed experiments to be performed to investigate the effect of this cysteine protease inhibitor on the proteolytic activity and growth of clinically relevant pathogenic bacteria.

When we used internally quenched FRET peptide substrates, and competitive conditions, SPI inhibited the proteolytic activity of purified enzyme, as well as several cell culture supernatants, most likely via the inhibition of important bacterial cysteine protease enzymes. However, these results presume that SPI preferentially inhibits the activity of cysteine proteases. Interestingly, however, genes encoding papain-like cysteine proteases are actually absent in B. anthracis (12), although it is known that B. anthracis produces NlpC/p60 cell wall hydrolases, which are related to cysteine proteases (5, 13). These cell wall hydrolases could also be a target for SPI-mediated protease inhibition. Further, our results indicate that SPI has an inhibitory effect on several types of bacterial proteases, including metalloproteases. However, the efficiency of inhibition may vary, dependent on the type of protease being inhibited.

The addition of SPI to growing bacterial cells resulted in a dose-dependent reduction in bacterial growth, a process which may be facilitated by inhibition of cell wall hydrolases. These enzymes are part of the cysteine peptidase family and play an important role in bacterial growth, particularly due to their association with cell wall turnover and recycling (4, 14). For example, papain-related amidases are involved in cell wall degradation (13), while cell wall construction also requires the action of serine proteases, so-called DD-peptidases (15). This could be one mechanism by which SPI inhibits bacterial growth, although additional experiments are required in order to confirm or deny this hypothesis. In any case, our results indicate that the cysteine protease inhibitor SPI is a good candidate for further research, particularly with respect to its mode of action and the range of bacterial species it inhibited. From our own experiments, SPI appears to be effective as an inhibitory agent against a wide range of important and clinically relevant bacterial species, including species comprising Gram-positive cocci, Gram-positive rods, and Gram-negative rod phenotypes, without being cytotoxic. The work suggests that SPI may have the potential to become a novel broad-spectrum antimicrobial agent for the treatment of clinically relevant infectious diseases. Additionally, SPI is a glutamine and a lysine donor protein of transglutaminases (10), a property that could be exploited to facilitate the preparation of SPI-containing antimicrobial protein sponges and foils for use in medical applications.