Aims: To determine whether the extracellular products (ECPs) from Aeromonas hydrophila, a frog bacterial pathogen that is resistant to skin antimicrobial peptides of three different frog species Xenopus laevis, Litoria aurea and Litoria raniformis, can modulate the activity of these peptides.
Methods and Results: ECPs were collected from cultures of Klebsiella pneumoniae, a pathogen susceptible to skin antimicrobial peptides of all three tested frog species, and from cultures of Aer. hydrophila. They were tested for protease activity and for inhibition of the antimicrobial activity of natural peptide mixtures and single peptides of all three frog species against Escherichia coli ATCC 25922. ECPs from cultures of Aer. hydrophila grown for 16, 24 and 36 h showed protease activity and inhibited the antibacterial activity of all peptides against E. coli ATCC 25922. In contrast, the ECPs from cultures of Kl. pneumoniae neither had protease activity nor inhibited the activity of any peptides.
Conclusion: The proteolytic ECPs of Aer. hydrophila have the ability to inhibit the skin antimicrobial peptides of frogs.
Significance and Impact of the Study: The results of this study provide new information on the association of ECPs with the resistance of Aer. hydrophila to frog antimicrobial peptides.
The bacterium Aeromonas hydrophila and some other Gram-negative bacterial species including Chryseobacterium indologenes, Chryseobacterium meningosepticum, Citrobacter freundii, Klebsiella pneumoniae, Proteus mirabilis, Pseudomonas aeruginosa and Serratia liquefaciens have been associated with bacterial dermatosepticemia, a fatal infectious disease of frogs (Glorioso et al. 1974; Nyman 1986; Taylor et al. 2001). They are opportunistic pathogens that are normally found in the skin and gut microbiota of healthy frogs but are known to cause disease in animals with compromised immune defences (Hird et al. 1981; Carey and Bryant 1995; Taylor et al. 2001). Pearson (1998) suggested that extracellular products (ECPs) with protease activity from these bacteria might have ability to inhibit certain immune mechanisms by proteolytic degradation of different protein components.
The only effective immune defence of frog skin against invading pathogens is the antimicrobial peptides of the innate immune system. Inhibition of these peptides by the ECPs could enable pathogens to evade elimination (Simmaco et al. 1998; Rollins-Smith et al. 2002, 2005; Woodhams et al. 2007). Evidence for a role of proteolytic ECPs in the resistance of bacteria to skin peptides has been found in previous studies of Rollins-Smith et al. (2002) and Schadich (2009). It was shown that the bacterium most frequently isolated from diseased frogs, Aer. hydrophila, is resistant to skin peptides of four different frog species including African Clawed Frogs (Xenopus laevis) and three Australian Litoria species: Litoria aurea, Litoria raniformis and Litoria ewingii. Three other pathogens, Cit. freundii, Kl. pneumoniae and Ps. aeruginosa, the frog saprophytic bacterium Lactococcus lactis and a standard reference strain of Escherichia coli ATCC 25922 were susceptible to the skin peptides of three of these frog species X. laevis, L. aurea and L. raniformis (Schadich 2009). To date, the mechanisms of resistance of frog bacterial pathogens to skin peptides have not been investigated.
The aim of this study was to determine whether the ECPs from Aer. hydrophila can modulate the action of skin peptides. ECPs were tested against skin peptides of three different frog species: X. laevis, L. aurea and L. raniformis.
Materials and methods
Heart isolates of Aer. hydrophila and Kl. pneumoniae collected from diseased wild L. ewingii were accessed from the University of Canterbury, School of Biological Sciences collection of frog bacterial isolates as described in Schadich (2009).
Peptide mixtures collected from skin secretion of X. laevis, L. aurea and L. raniformis, magainin 2, a single peptide of X. laevis (GIGKFLHSAKKFGKAFVGEIMNS) (Sigma Chemical, St Louis, MO, USA), aurein 2·1, a single peptide of L. aurea and L. raniformis (GLLDIVKKVVGAFGSL.CO-NH2) (Sigma-Genosys, Castle Hill, NSW, Australia) and a nonamphibian peptide polymyxin B, a single peptide of Bacillus polymyxa (cyclic octapeptide) (Sigma Chemical) were from stock solutions as described in Schadich (2009).
Bacterial cultures and production and preparation of ECPs
Bacterial isolates were grown in 18-h shake culture in 50 ml of tryptone soya broth (TSB; Oxoid, Basingstoke, UK) in 100 ml Erlenmeyer flasks at 30°C. A volume of 100 μl of this bacterial broth was inoculated into 250 ml of tryptone soya broth in 500 ml Erlenmeyer flasks and incubated as mentioned above for 36 h. Samples were removed at 0, 8, 16, 24 and 36 h and examined for bacterial growth by determining the optical density at 610 nm. The bacterial suspension was then centrifuged at 10 000 g for 20 min at 4°C and the supernatant filter was sterilized using a 0·22-μm Millipore membrane filter. The cell-free supernatant was used as a source of the ECPs. Three independent experiments were performed for each bacterial isolate.
Protease detection on agar media
Proteases produced by bacterial cells growing on Mueller–Hinton agar containing 10% (w/v) skimmed milk were assessed as described by Castro-Escarpulli et al. (2003).
The ECPs of each culture were incubated at 30°C for 1 h with the three different protease inhibitors, phenyl methyl sulfonyl fluoride (PMSF; 1 mmol l−1), ethylene diamine tetra-acetic acid (EDTA; 2 mmol l−1) and leupeptin (0·1 mmol l−1), which are specific for proteases of Aeromonas species.
Effects of high temperature on thermostability of ECPs
The stability of proteases at elevated temperature was measured by heating 300 μl of the ECPs of each culture to 90°C for 15 min. After heat treatment, the protease activity was measured as described above.
Inhibition of skin peptide activity by ECPs
The ability of the ECPs to inhibit the activity of peptide mixtures and single peptides was determined by bio-assay as described by Devine et al. (1999). A volume of 150 μl of peptides at a concentration of 300 μg ml−1 in ddH2O were incubated with 150 μl of ECPs or ECPs that had been inactivated either by incubation at 90°C for 1 h or in the presence of protease inhibitors. Control reactions included 150 μl of ddH2O. Following incubation, 50 μl of each peptide/ECP mixture was transferred to 96 well microtiter plates. Their activity against the easily manipulated bacterium E. coli ATCC 25922 was tested by growth inhibition assay as described by Rollins-Smith et al. (2002). Five replicate reactions and three independent assays were performed for all tested peptide/ECP mixtures.
Secretion of proteases on agar media
Aeromonas hydrophila was shown to produce protease as indicated by zones of clearing around colonies while no protease activity was detected for Kl. pneumoniae (Supporting Fig. S1).
Protease activity of ECPs
The presence of protease activity in the ECPs of Aer. hydrophila was dependent on the incubation time of cultures (Fig. 1a). No activity was detected in the ECPs from 0 and 8 h cultures, while the activity of the ECPs from the 16, 24 and 36 h cultures increased with incubation time of the cultures (Fig. 1a). The activity was lost upon treatment with protease inhibitors. No protease activity was detected in the ECPs of Kl. pneumoniae (Fig. 1b). Heat-treated ECPs of both bacterial cultures did not show protease activity (Fig. 1).
ECP-mediated inhibition activity of skin peptides
Aeromonas hydrophila ECPs were shown to inhibit the peptide mixtures and the single peptides from X. laevis, L. aurea and L. raniformis (Table 1; Data for single peptides not shown). The activity was dependent upon the incubation time of Aer. hydrophila in culture. Activity was only found in cultures grown for 16 h and longer (Table 1). It is interesting to note that Aer. hydrophila ECPs did not affect the activity of polymyxin B (Table 1); however, polymyxin B has disulfide bonds in its structure, which may confer greater stability to proteases as shown by Paulus and Gray (1964). The ECPs of Kl. pneumoniae which were not proteolytic in nature had no effect on the frog peptides. Both heat treatment and protease inhibitors abolished any inhibitory effects on the skin peptides.
Table 1. Effect of incubation time of Aeromonas hydrophila cultures on the ability of extracellular products (ECPs) to inhibit the activity of peptide mixtures of Xenopus laevis, Litoria aurea and L. raniformis against Escherichia coli ATCC 25922
Treatment with ECPs from cultures incubated for different time (h)
Frog peptide mixture
+ denotes complete inhibition of peptide.
− denotes no peptide inhibition.
The ECPs of Aer. hydrophila, which are proteolytic in nature, appear to be responsible for the resistance of this bacterium to the skin peptides of three frog species: X. laevis, L. aurea and L. raniformis. Significantly, the ECPs of Aer. hydrophila, a pathogen resistant to peptide mixtures and single peptides of all three frog species, completely inhibited the activity against E. coli ATCC 25922 of the peptides of these frog species (Table 1). The ECPs were shown to be thermolabile (Fig. 1a). Moreover, the ECPs of Kl. pneumoniae, a pathogen susceptible to peptide mixtures and single peptides of all three species, did not show proteolytic activity (Fig. 1b) and did not inhibit the activity of the frog skin peptides against E. coli ATCC 25922 of peptide mixtures or single peptides of all three species. The virulence of Aer. hydrophila would, therefore, appear to be associated with the production of proteases and this is consistent with the hypothesis of Pearson (1998) that Aer. hydrophila could have mechanisms to evade destruction by immune defences and thereby ensure a sufficient number of bacterial cells required for tissue invasion in the initial phase of disease.
Aeromonas hydrophila is commonly found in diseased animals of many different frog species (Hird et al. 1981; Taylor et al. 2001), which might suggest that production of proteases evolved initially as a trait required for adaptation to the skin environment of their host. This is supported by the fact that protease-mediated resistance to skin peptides could enable Aer. hydrophila to survive on the skin. These proteases might be specific to host peptides as the inhibition of the activity of polymyxin B, a nonfrog antimicrobial peptide, was not observed (Table 1).
The resistance mechanisms of opportunistic bacteria to skin peptides of the innate immunity of lower vertebrates such as frogs and fish might be species specific; however, variations in these mechanisms might be expected to occur among isolates. One direction of future research would be to compare resistance to skin peptides between different isolates and different species.
This research was supported by a Royal Society of New Zealand Marsden Grant (M1069). The authors thank Dru Mason (University of Canterbury, New Zealand) for useful comments on this manuscript.