Correspondence: Estelle Cateau, Laboratoire de Parasitologie et Mycologie Médicale, 2 rue de la Milétrie, BP 577, 86021 Poitiers Cedex, France. Tel.: +33 5 49 44 31 56; fax: +33 5 49 44 39 08; e-mail: firstname.lastname@example.org
Acinetobacter baumanii, which may be found in water, is an important emerging hospital-acquired pathogen. Free-living amoebae can be recovered from the same water networks, and it has been shown that these protozoa may support the growth of other bacteria. In this paper, we have studied potential relationships between A. baumanii and Acanthamoeba species. Two strains of A. baumanii isolated from hospital water were co-cultivated with the trophozoites or supernatants of two free-living amoebae strains: Acanthamoeba castellanii or Acanthamoeba culbertsoni. Firstly, the presence of the amoebae or their supernatants induced a major increase in A. baumanii growth, compared with controls. Secondly, A. baumanii affected only the viability of A. culbertsonii, with no effect on A. castellanii. Electron microscopy observations of the cultures investigating the bacterial location in the protozoa showed persistence of the bacteria within cyst wall even after 60 days of incubation. In our study, the survival and growth of A. baumanii could be favored by Acanthamoeba strains. Special attention should consequently be paid to the presence of free-living amoebae in hospital water systems, which can promote A. baumanii persistence.
Members of the genus Acinetobacter are ubiquitous microorganisms and A. baumannii can survive for long periods under a wide range of environmental conditions (Maragakis & Perl, 2008). According to Peleg et al. (2008), even if species of this genus do not necessarily have their habitat in the environment, no systematic study has been performed concerning the occurrence of the different species and their natural habitats still remain to be determined. In a hospital environment, on the other hand, it has been conclusively proven that a water system can be a reservoir for this bacterium (Huang et al., 2008). Improved understanding of the reservoirs and routes of transmission of this bacterium is indeed needed in the effective operation of prevention and control.
The objective of our study is to analyze the relationships between two strains of Acanthamoeba (Acanthamoeba castellanii and Acanthamoeba culbertsoni) and two strains of A. baumanii in order to investigate whether Acanthamoeba could influence the growth and/or survival of this bacterium.
Materials and methods
Strains and growth conditions
Acanthamoeba castellanii ATCC 30234 and A. culbertsoni ATCC 30171 were grown in 150-cm2 tissue culture flasks in PYG broth at 27 °C (Schuster, 2002). When cells formed a monolayer, the trophozoites were harvested by tapping the flasks and washed three times in Page's modified Neff's amoeba saline (PAS, containing in 1 L of distilled water, 120 mg NaCl, 4 mg MgSO4·7H2O, 4 mg CaCl2·2H2O, 142 mg Na2HPO4 and 36 mg KH2PO4). For experiments carried out in 96-well microtiter plates, amoebae were used at a final cell concentration of 5 × 105 mL−1 in PAS or in filtered tap water (0.22 μm).
Two antibiotic-sensitive strains of A. baumanii (named Ab1 and Ab2) were isolated from water of Poitiers Teaching Hospital (France). These bacteria were grown on Mueller–Hinton agar slants at 37 °C for 2 days, then harvested and washed in PAS medium to obtain a bacterial concentration of 5 × 104 mL−1 in this medium or in filtered tap water (0.22 μm) before use.
Co-cultivation of Acanthamoeba and A. baumanii
Each well of a 96-well microplate was filled with 100 μL of the amoebic trophozoites suspension (containing 5 × 104 cells). The amoebae (A. castellanii or A. culbertsoni) were allowed to adhere to the wells for 2 h at 27 °C. PAS (100 μL) containing 5 × 102 bacteria (multiplicity of infection of 0.01) were added in the wells and incubation was carried out during 24, 48 and 72 h at 27 °C. Controls were performed by incubating bacteria in PAS without amoebae. The same experiments were carried out in filtered tap water.
After incubation (24, 48 or 72 h), the co-cultures were passed five times through a 27-gauge needle to lyse the amoebae. Experiments had previously been performed with A. baumanii alone to ensure that this passage did not affect the viability of the bacteria. Serial dilutions of the lysates were plated on Mueller–Hinton medium and incubated at 37 °C for 48 h to evaluate CFU. After 24, 48 and 72 h of incubation, a microscopical examination of the culture using trypan blue staining was also carried out in order to determine the viability of amoebae.
All the experiments were reproduced three times, each time in duplicate.
Incubation of infected amoebae in encystment medium
Amoebae were infected with A. baumanii Ab1 strain as described above, but the experimentations were performed in flasks, and after 24 h of incubation, the co-cultures were transferred into encystment medium as described previously (Bouyer et al., 2007). This medium was chosen so as to allow cyst formation and to mimic conditions of poor nutrient availability. The CFU of A. baumanii were then numbered after 3, 5, 7, 11, 30 and 60 days of incubation at 27 °C. In addition, samples of suspensions (with Ab1 only) were examined after 2 h, 1, 3, 11 and 60 days by electron microscopy.
Effect of amoeba supernatant on bacterial growth
Trophozoites (5 × 105 mL−1) of each strain were incubated at 27 °C for 72 h in PAS or filtered water. The amoebae were then pelleted by gentle centrifugation (1000 g–10 min) in order to prevent lysis, and each strain of A. baumanii (5 × 103 mL−1) was incubated at 27 °C for 48 h in the resultant filtered (0.45 μm) supernatant. After incubation, the growth potential of the bacteria was determined by plating serial dilutions of the suspension on Mueller–Hinton medium to determine CFU counts. Controls were performed with A. baumanii incubated in PAS or in filtered water without supernatant.
The potential internalization of bacteria was investigated by electron microscopy of infected amoebae. After 2 h, 1, 3, 11 and 60 days, a sample of the co-cultivation in PAS or in encystment medium (A. castellanii or A. culbertsoni with the strain A. baumanii Ab1) was incubated for 1 h in phosphate buffer 0.1 M, containing 4% glutaraldehyde at 4 °C. Cells were washed four times in phosphate-buffered saline and post-fixed with 1% OsO4 in phosphate buffer 0.1 M for 1 h at 4 °C. The sample was dehydrated in an acetone series and embedded in araldite resin. Sections were stained with uranyl acetate and lead nitrate before examination with a Jeol 1010 transmission electron microscope.
Data were analyzed by anova and a Scheffe's test.
Results and discussion
Co-cultivation of Acanthamoeba and A. baumanii
Acinetobacter baumanii can be isolated from many surfaces in hospitals (Beggs et al., 2006; Peleg et al., 2008) and recovered from hospital water (Simor et al., 2002; Huang et al., 2008). Moreover, free-living amoebae are also frequently isolated from the same aquatic environment. It has been previously shown that free-living amoebae can interact with a great variety of microorganisms (Greub & Raoult, 2004). In this work, we investigated the relationships between A. baumanii and two Acanthamoeba species. In the co-cultures in PAS medium, the presence of A. castellanii or A. culbertsoni induced a major increase in A. baumanii growth, as compared with bacterial growth without amoebae (Fig. 1). The results of these co-cultures were similar in filtered tap water (data not shown). In addition, the viability of A. castellanii was not affected by the conditions of co-culture with A. baumanii as shown by trypan blue exclusion experiments (Table 1). As for Shigella sp. (Saeed et al., 2009), the relationships between these microorganisms may consequently be considered symbiotic. Acanthamoeba culbertsoni/A. baumanii co-culture induced a decrease in the viability of the amoeba, whatever the incubation medium used (PAS or filtered water); nevertheless, when incubated alone in the experimentation medium, the mortality of this amoeba was already high after 72 h (Table 1). This is not the first time that a differential effect of a bacterium has been observed on various amoebae. Dey et al. (2009) recently showed that amoebae are not all equally permissive with regard to L. pneumophila and that some amoebae strains can be particularly resistant to this bacterium.
Table 1. Evaluation of the amoebae mortality after 24, 48 and 72 h of co-cultivation with Acinetobacter baumanii (Ab1 and Ab2 strains) in PAS medium
Percentage of amoeba mortality
8 ± 1.3
9 ± 0.9
9.4 ± 2.4
8 ± 1.1
7 ± 1.3
10 ± 1.9
10 ± 2
12 ± 2.1
11 ± 1.9
7 ± 1.1
14 ± 2.3
19 ± 3.6
50 ± 5.6
72 ± 5.9
80 ± 5.9
46 ± 4.9
74 ± 6.4
81 ± 7.2
The results of electron microscopy have indicated the location of the bacterium. After 2 h of co-culture, bacteria were recovered only in the medium with intact trophozoites. After 1 and 3 days, bacteria were found in vacuoles in the cytoplasm of amoebae trophozoites. At that time, a few cysts were observed, without any bacteria (Fig. 2). It has previously been mentioned that A. castellanii enhances growth of some microorganisms (Greub & Raoult, 2004) and Abd et al. (2005) have shown that Vibrio cholerae O139 may be located in the cytoplasm of A. castellanii trophozoites as well as in the cysts of this amoeba. Moreover, it has recently been proven that some eucaryotes including Acanthamoeba species may prolong the survival of Campylobacter for at least 4 weeks (Axelsson et al., 2010). In addition, it had previously been shown by molecular techniques that A. baumanii could survive and be isolated in co-cultivation with Acanthamoeba polyphaga (Pagnier et al., 2008) or A. castellanii (Thomas et al., 2008), but no microscopic observation has shown the interactions between the microorganisms.
Acinetobacter baumanii and amoebae supernatants
We demonstrated that not only can the presence of amoebae facilitate bacterial development, but supplementation with the A. culbertsoni or A. castellanii supernatants obtained in PAS may likewise lead to increased bacterial counts (Fig. 3). The results were similar with supernatants obtained from filtered tap water (data not shown). This could be due, particularly for A. culbertsoni, to the death and the lysis of amoeba, especially when they were co-cultivated with A. baumanii, which could provide nutrients for the bacteria to grow.
Incubation of infected amoebae in encystment medium
Among microorganisms related to amoebae, some bacteria that may be human pathogens have evolved in a way that allows them to resist destruction by protozoa either because they are not internalized or else because they are able to survive, grow and exit amoebae following internalization (Greub & Raoult, 2004). We have evaluated the growth and survival of the bacterium in a poor medium such as encystment medium with and without amoebae. The bacterial count showed that the presence of amoebae (A. castellanii or A. culbertsoni) allows for increased bacterial growth, while A. baumanii in the same medium without amoebae is able to survive, but at lower concentrations. After 60 days in this medium, the survival of the bacteria is favored by the presence of amoebae (Fig. 4) In electron microscopy after 11 days of incubation, some cysts already contained intracellular A. baumanii, located only in the space between the double walls (Fig. 2), which is similar to the location of Pseudomonas in Acanthamoeba astronyxis (Marciano-Cabral & Cabral, 2003), Mycobacterium avium in A. polyphaga (Steinert et al., 1998), Mycobacterium sp. (Sharbati-Tehrani et al., 2005; Ben Salah & Drancourt, 2010), V. cholerae (Abd et al., 2005) or Vibrio mimicus (Abd et al., 2010) in A. castellanii. The significance of both this location within the cyst structures, but outside of the cytoplasm and the fact that the bacteria seem clustered together remain to be determined. The survival of other bacteria such as F. tularensis or Shigella sp. in A. castellanii cysts has also been reported, but the bacteria are intracellular and not associated with the outer surface (Abd et al., 2003; Saeed et al., 2009). According to Ben Salah & Drancourt (2010), the cellulase encoded by some bacteria may play a role in their exocyst location. Moreover, this location could allow the bacterium to more rapidly escape from the cyst.
In this study, we have shown that the presence of A. castellanii or A. culbertsoni may allow increase of A. baumanii growth, whatever the co-culture medium, PAS or filtered water. The presence of A. baumanii did not influence the viability of A. castellanii, but did dramatically decrease the viability of A. culbertsoni. When the cells were incubated in a hostile medium such as the encystment medium, the presence of the amoebae (A. castellanii or A. culbertsoni) increased the viable counts of bacteria, even after 60 days of incubation.
Considering that the cyst can confer resistance against some biocides and disinfection treatments (Greub & Raoult, 2003; Coulon et al., 2010; Marciano-Cabral et al., 2010), bacterial intracellular position can consequently protect it from adverse conditions. Moreover, similar studies may be conducted with strains resistant to antibiotics in order to evaluate, as regards Mycobacterium smegmatis, the potential intracellular persistence of such strains (Sharbati-Tehrani et al., 2005). The ability of A. baumanii to grow and survive intracellularly in Acanthamoeba species may be one factor that could enhance bacterial survival in aquatic environments and networks. Hence, in hospital water taps, special attention should be paid to the presence of free-living amoebae, which can promote survival of this pathogenic bacteria.