Motile mesophilic Aeromonas spp. are widely distributed in aquatic environments, including fresh and brackish water (1–3). Although Aeromonas spp. are abundant in fresh water, their numbers in seawater are extremely low (3, 4). However, in in vitro culture, Aeromonas can grow in medium containing NaCl at a concentration of 3.0%, which corresponds to that of seawater (4–6). It is therefore unclear why there are few Aeromonas in seawater. Monfort and Baleux examined the distribution and survival of Aeromonas spp. in brackish waters and about 500 m from the coast, and found that the number of Aeromonas spp. decreased markedly going from brackish water to the open lagoon (7). The rate of decrease was faster than that estimated from the dilution rate and they proposed that Aeromonas spp. are subject to marine stress. We deduced that Aeromonas has various agents that enable it to survive in seawater, and that these agents do not have the same effect in tube culture medium.
Extracellular protease is very important in enabling bacteria to survive in the environment, as these proteases degrade proteins into amino acids and oligopeptides that bacteria can use as nutrients. The representative proteases produced by A. sobria are ASP and AMP. In a previous study, we showed that the production of the bio-active ASP is influenced by NaCl in the surroundings. The ASP activity of culture supernatant is extremely low when the strain is cultured in medium containing 3.0% NaCl, compared to that found when cultured in medium containing 0.5% NaCl (8, 9). Studies of the mechanism of reduction of ASP activity by NaCl have revealed that the maturation pathway of ASP is disturbed in medium containing 3.0% NaCl (9).
A. sobria produces both AMP and ASP outside the cell. The metalloprotease gene of A. sobria 288 has previously been cloned and the sequence of this gene determined (GenBank Accession No. DQ784565). The deduced amino acid sequence of AMP is similar to that of the thermolysin family-type protease or elastase produced by Vibrionaceae and Pseudomonas aeruginosa (10–15). In this study, we examined the effect of NaCl in medium on the production of AMP.
A. sobria 288 (asp+, amp+) was used as a wild-type strain to analyze the production of protease. Our previous experiments showed that two kinds of proteases, ASP and AMP, are produced as extracellular proteases by the strain (9). Two isogenic mutant strains in which the serine protease gene (A. sobria 288 (asp−, amp+)) and metalloprotease gene (A. sobria 288 (asp+, amp−)) were deleted were prepared in the previous study (9). In this study, a deletion mutant cell in which both serine protease gene and metalloprotease gene (A. sobria 288 (asp−, amp−)) were deleted was prepared from A. sobria 288 (asp−, amp+) using suicide vector pXAC623 (9, 16). The desired mutation of the two protease genes was examined by Southern blot analysis by the method described in the previous report (9). The result showed that the metalloprotease gene had been correctly mutated in the prepared cell and that no other genes had been injured by the gene manipulation (data not shown).
The effect of NaCl on the proteolytic activity of Aeromonas was assayed using nutrient agar containing 3% skim milk supplemented with different concentrations of NaCl (final concentrations, 0.5%, 1.0%, 2.0%, and 3.0%). Strains were pre-cultured in NB (Eiken Chemical, Tokyo, Japan) at 37°C for 20 hr and these cultures were streaked onto nutrient agar plates containing 3% skim milk. The plates were incubated at 37°C for 48 hr. After incubation, production of active protease was assessed by the appearance of a transparent zone around the bacteria. This transparent zone is generated by the action of extracellular proteases, which degrade insoluble casein in the medium into soluble peptide.
In agar medium containing NaCl at concentrations of 0.5% and 1.0% three strains, namely A. sobria 288 (asp+, amp+), A. sobria 288 (asp−, amp+), and A. sobria 288 (asp+, amp−), created transparent zones, showing that both ASP and AMP have the ability to degrade casein. However, A. sobria 288 (asp−, amp−) did not create transparent zones. The fact that A. sobria 288 (asp−, amp−) did not form transparent zones shows that proteases other than ASP and AMP do not participate in the formation of these zones. It is also noteworthy that, in agar medium containing NaCl at 2.0% and 3.0%, no strains created a clear transparent zone.
As we described in a previous report, serine protease is synthesized in the presence of 3% NaCl, but the maturation pathway does not proceed successfully in salty medium containing 3% NaCl (9). This disturbance by NaCl of the production of serine protease resulted in disappearance of the transparent zone around A. sobria 288 (asp+, amp−) in agar medium containing 2.0% and 3.0% NaCl (Fig. 1, strain 3). Similarly, A. sobria (asp−, amp+) did not create a transparent zone in agar medium containing 2.0% and 3.0% NaCl (Fig. 1, strain 2). This suggests that AMP is not synthesized, or that any AMP synthesized does not take an active form, in the presence of NaCl at concentrations of 2.0% and more.
The above result was obtained from experiments using solid medium. Subsequently, we examined the effect of NaCl on production of AMP in liquid medium. The medium used was nutrient broth containing NaCl at 0.5%, 1.0%, 1.5%, 2.0%, and 3.0%. A 500 μl sample of overnight cultured solution of A. sobria 288 was inoculated into 50 ml nutrient broth containing various concentrations of NaCl. The bacterial suspensions were cultured at 37°C for 24 hr with shaking (140 rpm). At the times indicated in Figure 2, bacterial growth was measured by absorbance of the culture at 620 nm. Bacterial growth was not affected by NaCl at a concentration of 1.5% or less, and was slightly, but not significantly, reduced at 2%. In medium containing 3.0% NaCl, growth dropped to about 50% of that in medium containing 0.5% NaCl.
At 6 and 12 hr, 2 ml culture samples were centrifuged to separate the cells from the culture supernatant. AMPs in the supernatant were detected by Western blotting. As a comparison, ALH in these samples was also detected by Western blotting. In the previous experiment, it was shown that production of ALH is not significantly affected by NaCl in the environment, that ALH emerges as a precursor in culture medium, and that the precursor is then processed to the active form by proteases produced by the cell itself (9). As shown in Figure 3, all samples from 6-hr culture contained comparable amounts of the precursor form of ALH, however, in the samples from 12-hr culture, the band for the mature (active) form of ALH was detected in medium containing 0.5% and 1.0% NaCl. In samples from medium containing 1.5% and more, both the precursor and active forms of ALH were present. This means that the ability of A. sobria to produce ALH was not suppressed by NaCl in an environment up to 3.0% NaCl, but the protease activity required to change the precursor of ALH to the active form was weak in samples prepared from medium containing 1.5% and more.
In contrast, bands for AMP, both the intermediate and active forms (17), were detected in the lanes for the samples from medium containing 0.5% and 1.0% NaCl. The band was not detected in the lane of the sample from medium containing 1.5% and more, indicating that the production of AMP is suppressed by NaCl in the medium.
From these observations, we concluded that A. sobria can produce ALH in liquid medium containing 3.0% or 0.5% NaCl. However, A. sobria loses the ability to produce AMP in medium containing 1.5% or more NaCl.
In order to examine the effect of NaCl in medium on suppression of production of AMP in A. sobria, we measured the amount of amp mRNA of A. sobria 288. The samples for the assay were prepared by culture for 6 hr at 37°C in medium containing various concentrations of NaCl (0.5%, 1.0%, 1.5%, 2.0%, 3.0%), and dot blot assay was used for measurement of mRNA. As a control, the amount of 16S rRNA was measured.
The total RNA in the cells was purified. The purified RNAs were dissolved in RNase-free water, and a portion of RNA solution mixed with an equal volume of denaturation buffer (0.18 M sodium citrate, 1.8 M sodium chloride, 14.8% formaldehyde [pH 7.0]), according to the manufacturer's protocol for the DIG system (Roche Diagnostics, Mannheim, Germany). Each mixture containing 10 μg of RNA was spotted onto nylon membranes, and the membrane baked for 30 min at 120°C. The probe for RNA of amp gene was prepared by PCR using a DNA labeling kit (Roche Diagnostics). The probe covers from the 186th to 369th amino acid residues from the amino terminal of AMP. The probe for 16S rRNA covering from the first to 792nd of the nucleotide residues of 16S rRNA gene was prepared in the same way. The specific primers for 16S rRNA were designed from the genome sequence of A. hydrophila ATCC7966 (GenBank Accession Number CP000462). A. sobria 288 (asp+, amp+) genome DNA was used as template for these PCRs. RNA on the nylon membranes was hybridized with the probe, and hybridization signals detected according to the manual supplied with the DIG Nucleic Acid Detection Kit (Roche Diagnostics). The detection of chemiluminescence and signal quantification were carried out by LAS3000mini (Fujifilm, Tokyo, Japan).
As shown in Figure 4a, the probe for 16S rRNA reacted with all samples and the density of the each spot was comparable, showing that 16S rRNA was well transcribed in every condition and useful as an internal standard. In contrast, the reaction of the probe for amp became weak in parallel with increasing NaCl in the medium. That is, the total RNA prepared from the culture in medium containing 0.5% and 1.0% NaCl reacted with the probe for amp, but no reaction was detected in samples prepared from cells cultured in medium containing 1.5% or more NaCl. We measured the density of each spot with LAS3000mini (Fujifilm). As shown in Figure 4b, quantification of signals revealed that the amount of amp mRNA in cells grown in medium containing 1.5% NaCl was decreased to 30% of that found when they were grown in 0.5% NaCl. This means that amp was transcribed well in medium containing 0.5% NaCl, but not in medium containing 1.5% or more NaCl. Suppression of transcription of amp by NaCl causes inhibition of production of AMP in A. sobria.
In this study, we found that AMP was not produced in medium containing 1.5% or more NaCl, although cells grow well in such a salty medium (Fig. 2). Exotoxins of bacteria are thought to be important for bacterial growth and survival in the environment. Some exotoxins, such as cytolysin and protease, destroy host cells in order to take nutrients from these cells (18, 19). Loss of the ability to produce protease induces loss of ability to decompose proteins. This leads to the conclusion that A. sobria cannot decompose protein in seawater. Decomposition of proteins is important in that it enables bacteria to survive in nature, because they utilize amino acids and oligopeptides yielded by such decomposition. This may account for the fact that the number of Aeromonas in seawater is low compared to that in freshwater.
Bacteriovorous predators, for example protozoa such as ciliates and flagellates, are predators of bacteria in the natural world, and bacterial growth and survival are subject to constraints by predators (20–23). Some bacterial proteases function to protect the bacteria from attack by predators (24–26). For example, the proteases of Vibrio (20) and Pseudomonas (21, 26) have been reported to be protectors against predator grazing. It has been also reported that Vibrio vulnificus, which inhabits marine water, can produce a metalloprotease homologous with AMP in liquid medium containing 2.0% NaCl (14). The ability to produce thermolysin-type metalloprotease may be related to the native habitat of bacteria. Further studies are necessary to clarify the relationship between production of AMP by Aeromonas and the habitat of the bacteria.
Our study has revealed that synthesis of amp mRNA is regulated by NaCl in the surroundings. The amount of amp mRNA decreased in A. sobria 288 cultured in liquid medium containing 1.5% or more NaCl, indicating that the amp gene is not transcribed under such salty conditions. Ørmen et al demonstrated that transcription of the hemolysin gene in A. hydrophila is induced by NaCl (27). We confirmed that the mRNA of the alh gene encoding ALH in A. sobria increases in proportion to the concentration of NaCl (data not shown). This result indicates that transcription of the amp gene is specifically inhibited by NaCl, whereas transcription of the alh gene is positively induced by NaCl. Further study is required to show whether alteration in expression of amp is responsible for the ecology of Aeromonas in the environment.
The amino acid sequence of AMP is similar to that of metalloprotease secreted by Vibrionaceae. It is well known that production of these homologous metalloproteases is regulated via the quorum-sensing system, a cell density-dependent mechanism (28–30). Moreover, it has been reported that A. hydrophila has a quorum-sensing system and that exoprotease production in A. hydrophila is influenced by this system (31). Another of our experiments showed that AMP is not produced in in vitro culture until it enters the middle logarithmic phase (data not shown). In a quorum-sensing system, autoinducers play an important role. It may be that the function of the autoinducer of A. sobria is specifically lost when these bacteria are cultured in medium containing 1.5% or more NaCl, although the bacteria can grow in 1.5% and 2.0% NaCl, as well as in 0.5% NaCl.
Inhibition of production of AMP by NaCl is a result of dysfunction in transcription, whereas that of serine protease is due to disruption of the maturation pathway (9). The step in production of ASP which is affected by NaCl is different from the step affected in production of AMP. This indicates that production of these proteases might be controlled by individual mechanisms.