Introduction

The genus Aeromonas are waterborne Gram-negative bacteria that are ubiquitous in water, including groundwater and chlorinated drinking water (LeChevallier et al., 1982; Kühn et al., 1997a, 1997b; Gavriel et al., 1998). Worldwide studies have demonstrated that Aeromonas spp are universally distributed and widely isolated from clinical and environmental samples (Nakano et al., 1990; Huys et al., 1996). Aeromonas comprise important human pathogens, including A. hydrophila, A. veronii and A. caviae (also known as A. punctata), which cause gastroenteritis, primary and secondary septicemia, and wound infections. Aeromonas strains isolated from water have been shown to possess virulence traits, such as adhesins, hemolysins and cytotonic enterotoxins, presumably involved with human pathogenicity (Handfield et al., 1996; Albert et al., 2000; Schubert, 2000). The prevalence of Aeromonas species in the aquatic environment has been recognized as a potential health risk, and some countries have adopted aeromonad counts as an additional indicator of water quality. For example, A. hydrophila has been placed on the US Environmental Protection Agency's Contaminant Candidate List of emerging pathogens in drinking water (Borchardt et al., 2003). Although it is clear that members of Aeromonas are primarily aquatic organisms, their natural host in the aquatic habitat is as yet unknown.

Among aquatic invertebrates, the nonbiting midges (Chironomidae, Diptera) are globally distributed, and they are the most abundant group of insects found in freshwater ecosystems (Armitage et al., 1995). They undergo four life stages, three of which are aquatic—egg, larva and pupa—and the last of which is the adult stage, when the insect emerges from the water into the air. After mating, the adult female lays a gelatin egg mass on the water's edge, with each mass containing hundreds of embedded eggs. Egg masses of Chironomus sp were found to harbor Vibrio cholerae, and they provide a natural reservoir for this bacterium (Broza and Halpern, 2001; Halpern et al., 2004, 2006). It was found that chironomid egg masses were destroyed in the presence of V. cholerae, and degradation of the egg masses was followed by failure of the eggs to hatch (Broza and Halpern, 2001). Hemagglutinin/protease, an extracellular enzyme of V. cholerae, was purified from V. cholerae O9 and from the pathogenic O1 and O139 serogroups as the enzyme responsible for the degradation of chironomid egg masses (Halpern et al., 2003). Furthermore, it was found that all V. cholerae strains, which were isolated from chironomid egg masses were able to degrade chironomid egg masses (Halpern et al., 2003, 2006, 2007). An interaction of host–pathogen population dynamics was found between the chironomid population and V. cholerae in their egg masses, and a peak in the chironomid population was followed by a peak in the V. cholerae population (Halpern et al., 2006).

The primary objective of this study was to explore bacteria other than V. cholerae that inhabit chironomid egg masses and are able to degrade the egg masses. Most of the isolates that have shown the ability to degrade the egg masses have been identified as Aeromonas species, demonstrating that Aeromonas species are natural inhabitants of chironomid egg masses. This study will allow us to understand the relationships between chironomids, V. cholerae and Aeromonas populations better. Moreover, since the prevalence of Aeromonas species in the aquatic environment has been recognized as a potential health risk, understanding the natural reservoir of Aeromonas will help in developing methods to monitor and control the bacteria in fresh and drinking water reservoirs.

Materials and methods

Sampling sites and environmental parameters

The study was performed in northern Israel near Haifa at two sampling sites: the Tivon waste stabilization pond (WSP; longitude 699, latitude 618) and the Kishon River (near Kiryat Haroshet; longitude 697, latitude 620). The sites were within a radius of 2 miles. Water temperature, pH (Scan Model2, Rajah Crescent, Singapore) and conductivity (El-Hamma Model TH-250, Mevo Hamma, Israel) were routinely monitored. Fluctuations of 11.3 and 8.1 °C in water temperature were measured across the sampling months in the Tivon WSP and the Kishon River, respectively. Maximal temperature (33.5 °C) was measured on June 23 in the Tivon WSP. The water temperatures were always higher in the Tivon WSP than the water temperatures in the Kishon River. A basic pH was measured in both the Tivon WSP (7.8±0.36) and the Kishon River (8.7±0.24). The conductivity that was measured in the Tivon WSP was between 1.5 and 1.8 mmoh cm–1 and it was three times lower than that of Kishon River (between 4.0 and 5.9 mmoh cm–1).

Egg mass sampling

Egg masses were sampled twice each month between May and September 2005. Styrofoam boards (25 × 25 cm) were used as artificial oviposition sites for adult females. Egg masses along the side of the board were collected 24 h after introducing the board into the water habitat.

Isolation of V. cholerae and chironomid egg mass-degrading bacteria

Chironomid egg masses from both sampling sites were brought to the laboratory immediately after collection. Three egg masses from each sampling site were chosen randomly for isolation of the bacteria. Each freshly collected egg mass was treated separately to remove all the bacteria that were not firmly attached to the egg mass. After washing five times and vortexing for 1 min in sterile saline water, each egg mass was crushed in 1 ml saline, using a sterile glass homogenizer. The homogenate was diluted, and 0.1 ml from each dilution was then spread onto thiosulfate–citrate–bile salts (TCBS; Becton, Dickson and Company, Sparks, MD, USA) for isolation of V. cholerae strains. To screen for other bacterial species with the potential ability of degrading the egg masses, samples from each dilution were also spread onto the following media: (i) cetrimide agar base (Himedia, Mumbai, Maharashtra, India); (ii) modified cetrimide agar base (no glycerol; gelatin—15 g l–1); (iii) modified minimal broth Davis without dextrose (Becton, Dickson and Company)—gelatin and yeast extract were added, 15 g l–1 and 0.5 g l–1, respectively; and (iv) LB agar (Luria) (Himedia). Yellow colonies from the TCBS medium that were suspected to be V. cholerae were subcultured on LB agar and further examined as described below. Individual colonies from all the other four media were randomly picked and streaked onto LB agar to obtain single colonies. All isolated colonies were subcultured at least four times before further examination. Bacterial isolates were kept in LB with 30% glycerol (−80 °C).

Identification of V. cholerae

Yellow colonies from the TCBS medium that were suspected to be V. cholerae were subcultured on LB agar and then tested for oxidase (1% tetramethyl-p-phenylenediamine; Sigma, Saint Louis, MO, USA), as well as for the string test (0.5% sodium deoxycholate; Sigma). The identity of the isolates with positive results in the above tests was further verified by using the multiplex PCR assay in accordance with Nandi et al. (2000). The presence of toxin genes (in addition to ctxA, as described in Nandi et al., 2000) stn/sto, hlyA, hapA and ompU were determined in all V. cholerae strains. The primers used and the PCR procedure is described in Halpern et al. (2006) (Table 1).

Table 1 The presence of V. cholerae or Aeromonas species among the bacteria that were cultured from chironomid egg masses on all the sampling dates
Full size table

Egg mass-degradation activity

All the isolates (ca. 1100) were tested for their ability to degrade chironomid egg masses. Bacterial isolates were grown in 5 ml LB medium in a shaker incubator at 30 °C (or 37 °C for V. cholerae isolates) for 48 h. Culture supernatants were obtained by centrifugation (12 000 r.p.m. for 7 min) and then were filtered through a 0.45-μm pore size filter (Corning; Corning, New York, NY, USA), which resulted in a clear supernatant. Degradation activity was determined by using freshly collected or ethanol (70%)-rinsed chironomid egg masses. Degradation assay was performed by incubating one egg mass in 1.0 ml clear supernatant at 30 °C for up to 72 h. Each test was carried out at least three times. A reaction was considered positive when the egg mass structure disintegrated and a large excess of free eggs was found separated from the egg mass, as compared with the control LB medium (see Figure 1). Positive results were obtained between 10 and 40 h.

Figure 1
figure 1

Chironomid egg masses before (left) and after (right) 10 h incubation with the supernatant of a 48-h culture of Vibrio cholerae. The eggs were freed from the degraded gelatinous matrix.

Full size image

Identification of egg mass-degrading bacteria

Universal bacterial primers: 8F (5′-CACGGATCCAGACTTTGAT(C/T)(A/C)TGGCTCAG-3′) and 1512R (5′-GTGAAGCTTACGG(C/T)TAGCTTGTTACGACTT-3′), based on Escherichia coli positions, were used to amplify internal fragments of 16S rRNA gene (Felske et al., 1997). Eight microliters of a bacterial suspension was transferred to a sterile thin-walled PCR tube. One microliter of each primer (20 pmole μl–1) and 10 μl of PCR master mixture (ReddyMix, ABgene, Epsom, UK) were added to the tube to a final reaction volume of 20 μl. Initial DNA denaturation was performed at 94 °C for 4 min followed by 33 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C for 50 s, and elongation at 72 °C for 2 min, and then a final elongation step at 72 °C for 10 min. The amplified PCR products were purified with the AccuPrep PCR purification kit (BIONEER, Daedeok-gu, Deajeon, South Korea). Purified PCR products were sequenced at Technion Medical School (Haifa, Israel) directly by the dideoxynucleotide chain-termination method using a DNA sequencer (ABI PRISM 3100) with BigDye terminator reagents (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. The sequencing was performed by using 8F primer and for several isolates—the reverse 1512 primer was used as well. This resulted in data for approximately 750 or 1504 bp, respectively. Newly determined sequences were compared to those available in the GenBank (www.ncbi.nlm.nih.gov) databases, using the standard nucleotide–nucleotide BLAST program (BLASTN; www.ncbi.nlm.nih.gov), to ascertain their closest relatives. These sequence data have been submitted to the GenBank database under accession numbers EF634206–EF634234 (see also Supplementary Table S1). A phylogenetic tree was generated by using the neighbor-joining method with NJPlot (MEGA 3), based on alignments from CLUSTAL W. The bootstrap values obtained were from 1000 iterations.

Identification of isolates using specific primers for Aeromonas

Since most isolates were identified as Aeromonas species while using 16S rRNA gene sequencing, the identification of the remaining isolates was done by using specific primers for the genus Aeromonas and for four species in this genus: A. hydrophila, A. punctata, A. veronii and A. schubertii. All nucleotide primers were targeted to the 16S-23S intergenic spacers, in accordance with Kong et al. (1999). To confirm amplicon production, the mixture was analyzed by electrophoresis on 1.5% agarose gel, followed by staining with ethidium bromide and visualization under ultraviolet light.

Direct identification of Aeromonas spp presence in egg masses using PCR

DNA was extracted from individual egg masses that were sampled in Tivon WSP as follows: Each egg mass was first incubated with proteinase K and then DNA was extracted by using Invisorb Spin Plant Mini Kit (Invitec, Berlin, Germany), according to the manufacturer's instructions. The presence of Aeromonas was detected in the extracted DNA by using aero-F and aero-R primers (Kong et al., 1999), specific for the detection of members of the Aeromonas genus.

Genotypes associated with virulence in isolates of Aeromonas species

The presence of the following genes coding for the virulence factors, was determined in all Aeromonas isolates: act/aerA/hlyA by using one set of primers AHCF1/AHCR1; alt and ast genes for cytotonic enterotozins; ahyB genes for elastase; pla/lipH3/apl-1/lip genes for phospholipase; and fla gene for flagellin (Kingombe et al., 1999; Sen and Rodgers, 2004). The presence of the genes act/aerA/hlyA and ast; fla and alt; ahyB and pla/lipH3/apl-1/lip was tested simultaneously in the same reaction mixture, according to Sen and Rodgers (2004).

Results

Chironomid egg mass-degrading bacteria

A total number of approximately 1100 bacterial colonies were isolated from chironomid egg masses between May and September 2005. All isolates were examined for their ability to degrade chironomid egg masses. The supernatants of all the isolates that were identified as V. cholerae (82 isolates) were able to degrade the egg masses within 10 h at the most (Figure 1). Only 45 out of all the rest of the isolates (1018), which were randomly picked from the different culture media and were not identified as V. cholerae, were able to degrade the egg masses. These degrading bacteria were identified either by the 16S rRNA gene or by using specific primers for Aeromonas species. Thirty-three isolates were identified as A. punctata, nine as A. veronii, one as A. hydrophila, one as Pseudomonas otitidis and one as P. pseudoalcaligenes (see Figure 2 and Supplementary Tables S1 and S2). The time required for the egg mass degradation by the 48-h supernatant ranged between 15 and 40 h, and there was no clear association with particular bacterial species (Supplementary Tables S1 and S2). The phylogenetic relationships between the egg mass-degrading isolates and their closest relatives from public databases are shown in Figure 2.

Figure 2
figure 2

Phylogenetic tree of chironomid egg mass-degrading bacteria isolated from the egg masses. The tree shows the relationships based on partial sequences of the 16S ribosomal RNA gene. Strains in bold were previously isolated from chironomid egg masses (Halpern et al., 2007). The accession numbers of the isolates and the time required for egg mass degradation are listed in Supplementary Table S1. The sequences alignments were performed using the CLUSTAL W program, and the tree was generated using the neighbor-joining method with Kimura 2 parameter distances in MEGA 3 software. Bootstrap values (from 1000 replicates) greater than 50% are shown at the branch points. The bar indicates 2% sequence divergence.

Full size image

During the 5 month period, 11 samplings were performed at the two sampling sites. Table 1 summarizes the presence of V. cholerae and/or Aeromonas species among the bacteria that were cultured from chironomid egg masses on all the sampling dates. DNA was extracted from egg masses that were collected on all the sampling dates from the Tivon WSP, but not from the Kishon River. The primers aero-F and aero-R were used to amplify a 550-bp amplicon, demonstrating the presence of Aeromonas sp in all the sampled egg masses from the Tivon WSP (except for sampling 4, see Table 1). All the egg masses were washed and vortexed several times before they were cultured or before DNA was extracted. Thus, the detected or isolated bacteria were attached to the egg masses. No enrichment procedures were used to isolate V. cholerae. Likewise, no enrichment or selective media were used for the isolation of Aeromonas species.

Genotype of V. cholerae isolates

None of the strains (n=82) that were isolated from the egg masses and examined in this study contained the genes for the cholera toxin (ctxA) or for non-O1 heat-stable enterotoxin (stn/sto). The genes for El Tor-like hemolysin (hlyA) and outer membrane protein (ompU), which have been associated with enteropathogenicity, were found in 70% and 16% of the strains, respectively. The genotypes were found in various combinations (see Table 2). All isolates possessed the structural gene hapA, which is responsible for soluble hemagglutinin/protease production (see Table 2). The supernatants of all the strains examined in this study demonstrated the ability to degrade chironomid egg masses.

Table 2 Genotypic traits of representative isolates of V. cholerae non-O1/non-O139 examined in this study (n=82)
Full size table

Virulence factors in Aeromonas species

Aeromonas bacteria that were isolated from the chironomid egg masses were screened for the presence of genes for six virulence factors: act/aerA/hlyA genes for a cytotoxic enterotoxin and hemolytic activity; ast and alt genes for cytotonic enterotoxins; ahpB gene for elastase; pla/lipH3/apl-1/lip genes for lipase; and fla gene for flagella. The most prevalent genes were alt and ahpB, which were found in 78% and 76% of the isolates, respectively (see Table 3, Figure 3). The genes act/aerA/hlyA, fla and pla/lipH3/apl-1/lip were also quite prevalent among the isolates (65%, 59% and 43%, respectively). The most infrequent gene was ast (see Table 3, Figure 3). It was found only in one isolate, which was identified as A. hydrophila (isolate 10T1D19) (see Figure 2 and Supplementary Table S1). The isolate 9T1LB56 was the only one that did not possess any of the six examined virulence factors. Although the virulence genotypes were found in different combinations, five of the six examined virulence factors were positive (ast was negative) in the most widespread genotype (26% of all isolates) (see Table 3).

Table 3 Genotypic traits of Aeromonas spp isolated from chironomid egg masses (n=43)
Full size table
Figure 3
figure 3

The frequency of the virulence genes among A. veronii (n=9) and A. caviae (n=33) isolated from chironomid egg masses.

Full size image

Discussion

Chironomid egg masses are known to be a natural reservoir of V. cholerae (Broza and Halpern, 2001; Halpern et al., 2004, 2006, 2007). While exploring for bacteria other than V. cholerae that can degrade the egg masses, we demonstrated that chironomid egg masses harbor Aeromonas species along with V. cholerae. Aeromonas species were isolated from the egg masses throughout most of the samplings and were further shown to be inhabitants of the egg masses by using specific primers for Aeromonas sp (Table 1).

In a recently published study, Aeromonas species were isolated from chironomid egg masses when the bacterial diversity of the egg masses was studied (Halpern et al., 2007). Both in Halpern et al. (2007) and in the current study, egg masses were thoroughly washed to obtain only bacteria that were attached, probably as a biofilm, to the egg mass and no selective or enrichment methods were used for the isolation of Aeromonas species. Thus, Aeromonas species seem to be inhabitants of chironomid egg masses.

The following virulence-associated genes were detected in V. cholerae isolates from chironomid egg masses: hemagglutinin/protease (hapA, 100%); hemolysin (hlyA, 70%); and outer membrane protein (ompU, 16%) (Table 2). Cholera toxin (ctxA) and the non-O1 heat-stable enterotoxin (stn/sto) were not detected. A very similar pattern (hapA 100%, hlyA 72% and ompU 28%) was observed in a previous study examining V. cholerae isolates from egg masses (Halpern et al., 2006).

The following virulence-associated genes were detected in Aeromonas in the current study: 78% (alt), 76% (ahpB), 65% (act/aerA/hlyA), 59% (fla), 43% (pla/lipH3/apl-1/lip) and 2% (ast) (see Table 3, Figure 3). These results (except for the ast frequencies) are quite similar to those obtained by Sen and Rodgers (2004): 88% (ahyB), 88% (lip), 59% (fla), 43% (alt), 70% (act) and 30% (ast). The genes were found in various combinations in both studies. The results show that Aeromonas bacteria found in chironomid egg masses possess a wide variety of virulence-related genes, as was shown by Sen and Rodgers (2004) for bacteria that were isolated from drinking water.

Chironomid infestation in drinking water supply systems is an existing problem in Israel (Halpern et al., 1999, 2002) and in other countries in the world, for example, the United Kingdom (Williams, 1974), the United States (Bay 1993; Alexander et al., 1997), China (Sun et al., 2007) and Japan (Tanaka et al., 2004). In Japan, the source of chironomids in tap water contamination was suggested to be the larvae, which live in the water purification plant (Tanaka et al., 2004). Chironomids may disseminate pathogenic species of Aeromonas between drinking water reservoirs, as was suggested for V. cholerae (Broza et al., 2005). Furthermore, the gelatinous egg mass matrix may protect Aeromonas that inhabit the egg masses from the disinfecting agents, which are used in the drinking water systems.

No correlation was found between the concentration of Aeromonas spp and fecal indicator organisms (Legnani et al., 1998). However, Aeromonas species showed a seasonal distribution, as was found for chironomids (Holmes et al., 1996; Halpern et al., 2004, 2006). In surface water, Aeromonas abundance peaks in the warm summer and fall months (Burke et al., 1984; Chauret et al., 2001). The seasonal increase in Aeromonas detection in the drinking water supply was found to match the peak occurrence of clinical isolates (Burke et al., 1984).

Since the prevalence of Aeromonas species in the aquatic environment has been recognized as a potential health risk (Borchardt et al., 2003), understanding the natural reservoir of Aeromonas will help in developing methods to monitor and control the bacteria in fresh and drinking water reservoirs. More work is needed to survey Aeromonas abundance in chironomid egg masses in different freshwater habitats, including drinking water reservoirs, so as to identify the secreted factor that degrades the egg mass and to monitor the population dynamics of chironomids, Aeromonas and V. cholerae, and their dependency.