Long-term survive of Aliarcobacter butzleri in two models symbiotic interaction with Acanthamoeba castellanii

Aliarcobacter butzleri (formerly known as Arcobacter butzleri) is an emerging foodborne zoonotic pathogen that establishes in vitro endosymbiotic relationships with Acanthamoeba castellanii, a free-living amoeba. Previously, we described that this bacterium acts as an endocytobiont of A. castellanii, surviving for at least 10 days in absence of bacterial replication. Thus, the aim of this study was to evaluate the ability of A. butzleri to survive as a long-term endosymbiont of A. castellanii for 30 days in two models of symbiotic interaction with A. castellanii: i) endosymbiotic culture followed by gentamicin protection assay and ii) transwell co-culture assay. The results allow us to conclude that A. butzleri is capable of surviving as an endosymbiont of A. castellanii for at least 30 days, without multiplying, under controlled laboratory conditions. In addition, in the absence of nutrients and as both microorganisms remain in the same culture, separated by semi-permeable membranes, A. castellanii does not promote the survival of A. butzleri, nor does it multiply. Our �ndings suggest that the greater survival capacity of A. butzleri is associated with their endosymbiont status inside A. castellanii, pointing out the complexity of this type of symbiotic relationship. Signi�cant differences were observed in the percentage of viable intra-amoebic bacteria dependent on the MOI at which the tests were performed. For MOI 1:40, 1.75% (SD ± 0.34) of viable intra-amoebic bacteria was observed for ATCC 49619 and 1.62% (SD ± 0.21) for HF 2810. On the other hand, when performing the infection at a MOI of 1:100, a percentage of viable bacteria of 11.26% (SD ± 3.4) and 11.93% (SD ± 2.7) was obtained for the ATCC 49619 and HF 2810 strains, respectively. The results observed allow us to suggest that the culture of A. butzleri in endosymbiosis with A. castellanii has a more stable behavior at a MOI greater than or equal to 100 from 500,000 amoebas/mL. Therefore, this bacteria/amoeba ratio would be optimal for these tests.

). This bacterium acts as an endocytobiont of A. castellanii due to its ability to establish itself within amoebic vacuoles, surviving for at least 10 days in absence of bacterial replication. The adhesion of A. butzleri to amoebas mainly involves membrane-associated galactose receptors present on the amoebic membrane, and the PI3K and RhoA pathways are involved in the bacterial internalization, where the tyrosine kinase-induced actin polymerization signal is essential in Acanthamoeba-mediated bacterial uptake. Furthermore, A. butzleri requires a biphasic transcriptional pattern of agellar and putative virulence genes to establish an endosymbiotic relationship with A. castellanii.
Despite advances in understanding the events involved between these microorganisms during the early stages of endosymbiosis, there are no studies that contribute to describe the behavior of A. butzleri as a long-term endosymbiont of A. castellanii. Moreover, the effect that amoebic factors would promote on A. butzleri in the long term under controlled laboratory conditions remains unknown. Therefore, the aim of this study was to analyze the ability of A. butzleri to survive and multiply in two models of long-term symbiotic interaction with A. castellanii for 30 days.
In this study, A. castellanii T4 genotype originally isolated from a patient with keratitis (European Centre for Disease Prevention and Control -ECDC -Collection, BP 91/2760) was used in all the experiments.
Trophozoites were maintained in log phase at 25 °C in T-25 tissue culture asks with 10 mL of modi ed peptone-yeast-glucose (PYG) medium. Prior to the assays a suspension of 5 × 10 6 amoebae/mL in PBS 1X was adjusted as previously described by Medina et al. (2019). A. butzleri ATCC 49616 and native strain HF28, isolated from an environment water source, were used in all experiments. The routine cultures were performed on plates of Blood Agar Base No. 2 (Oxoid TM ) supplemented with 5% sheep blood, incubated for 24 h at 30 °C under aerobic conditions. Prior to the assays a suspension of each strain culture in exponential growth phase was adjusted to 5 × 10 7 bacteria/mL in PBS 1 × as previously in a total volume of 1 mL PBS 1X. A second suspension of 6 × 10 4 amoebas was inoculated with 2,4 × 10 6 bacteria (MOI 40) in a total volume of 1 mL PBS 1×. Later, both samples were incubated for 1 h at 25°C. A bacteria-free amoebas suspension was included as control. The intracellular survival of A. butzleri was determined by the gentamicin protection assay with the addition of 50 μg/mL gentamicin during 1 h at 25º C to kill extracellular bacteria. Subsequently, gentamicin was removed by twice step of centrifugation (2000 rpm × 10 min). The obtained pellet was suspended in 1 mL of PBS with 10 μg/mL of gentamycin on the at side of the 3mL cell culture tube (Thermo Scienti c™) and kept at an inclination of approximately 45° at 25°C for 1, 2, 5, 10, 15, 20, 25 and 30 days, followed by gentamicin treatment (50 μg/mL gentamycin at 30 °C for 1 h) to kill possible extracellular bacteria.
Again, gentamycin was removed by repeating the centrifugation step twice (2000 rpm × 2 min). Amoebas were counted to detect possible variations in the number of trophozoites in all the infection times. Pellets were suspended in 1 mL of PBS 1 × containing sodium deoxycholate 0.5% and incubated at 30 °C for 10 min to lyse amoebae and release intracellular bacteria. The samples were then washed with PBS 1 × by centrifugation at 2000 rpm for 2 min to remove the sodium deoxycholate. The viable bacteria were counted according to the protocol previously described by Medina et al. (2019). Brie y, 250 μL of sample was loaded into the rst well of each row in a 96-well plate, and ten-fold serial dilutions were made using a multichannel pipette by transferring 20 μL from initial column into 180 μL of medium in the next column, mixing ten times, and repeating the process until completion of six dilutions. Pipette tips were changed between dilutions. Then, three replicates of 10 μL from each of the six dilutions were plated onto plates of Blood Agar Base No. 2 (Oxoid TM ) supplemented with 5% sheep blood. The plates were incubated for 24 h at 30 °C under aerobiotic conditions. Then, colonies were enumerated at the latest dilution where growth was observed. Colony count was calculated by multiplying the number of colonies observed by the corresponding dilution. Additionally, the amoebae infection percentage was calculated by counting in the Neubauer chamber at the times of infection.
Transwell co-culture assays were performed as previously described by Laskowski-Arce and Orth (2008) with modi cations. Brie y, a suspension of 5 × 10 7 bacteria was deposited at the bottom of a 24-well tissue culture plate. Then, a 0.2 μm pore size transwell membrane (Anopore™ -NUNC™ Cell Culture) was inserted into each well of the 24-well tissue culture dish. Later, a suspension of 5 × 10 5 amoebas was added over the transwell membrane. The nal volume of co-culture was adjusted to 2 mL PBS 1X. The transwell co-cultures were incubated in humid chamber at 25°C for 1, 2, 5, 10, 15, 20, 25 and 30 days. The viable bacteria were counted as mentioned above, with the following modi cations: prior to the bacterial count, the integrity of the transwell membrane was veri ed by seeding a volume of 10 μL from upper of the well onto plates of Blood Agar Base No. 2 (Oxoid TM ) supplemented with 5% sheep blood, incubated for 24 h at 30 °C under aerobiotic conditions. Then, the transwell membrane was removed and the well was exposed at -20 ° C for 3 minutes in order to detach the bacterial monolayer. The bacterial suspension was washed by centrifugation (2000 rpm × 2 min) and the obtained pellet was suspended in 1 mL of PBS. Then, the viable bacteria were counted according to the protocol mentioned above.
For both endosymbiotic culture and transwell co-cultures assays, experimental units were done in duplicate (biological duplicate), and each duplicate was performed in triplicate. Data were analyzed using Student's t test and one-way ANOVA with GraphPad software. Values of p < 0.05 were considered statistically signi cant.
When evaluating the viability and multiplication capacity of A. butzleri as an endosymbiont of A. castellanii at different multiplicity of infection (MOI), it was observed that both strains of A. butzleri have the capacity to survive, for at least 30 days, under controlled conditions in endosymbiosis with A. castellanii, while the survival capacity of this bacterium was observed in all trials (Figure 1). Previously, the ability of A. butzleri to survive inside A. castellanii for 240 hours has been demonstrated, suggesting that it can resist amoeba digestion processes (Villanueva et al., 2016). From the results of this study, the ability of A. butzleri to survive, for at least 30 days, as an endosymbiont of A. castellanii is demonstrated for the rst time. In this sense, Balczun & Scheid (2017) proposed that the relationship between microorganism/FLA would promote evolutionary processes associated with the development of pathogenicity and adaptation to human macrophages.
Signi cant differences were observed in the percentage of viable intra-amoebic bacteria dependent on the MOI at which the tests were performed. For MOI 1:40, 1.75% (SD ± 0.34) of viable intra-amoebic bacteria was observed for ATCC 49619 and 1.62% (SD ± 0.21) for HF 2810. On the other hand, when performing the infection at a MOI of 1:100, a percentage of viable bacteria of 11.26% (SD ± 3.4) and 11.93% (SD ± 2.7) was obtained for the ATCC 49619 and HF 2810 strains, respectively. The results observed allow us to suggest that the culture of A. butzleri in endosymbiosis with A. castellanii has a more stable behavior at a MOI greater than or equal to 100 from 500,000 amoebas/mL. Therefore, this bacteria/amoeba ratio would be optimal for these tests.
When analyzing the behavior of A. butzleri ATCC 49619 as an endosymbiont of A. castellanii (for MOI 1:40) signi cant differences were observed in all counts after the rst day of endosymbiosis ( Figure 1A). In contrast, no signi cant changes were observed in the count of viable intra-amoebic bacteria in the rst 10 days after the initiation of endosymbiosis in the cultures made at MOI of 1:100 ( Figure 1B). Regarding the A. butzleri HF 2810 strain, signi cant differences were observed in the number of CFU/mL obtained after the rst day of endosymbiosis initiation for MOI 1:40 ( Figure 1C), while for the second trial (MOI 1:100), no signi cant differences were observed in the count of viable intra-amebic bacteria until day 5 after endosymbiosis began ( Figure 1D). Finally, the results indicate that both strains of A. butzleri do not undergo multiplication in endosymbiosis with A. castellanii since the count of viable intra-cellular bacteria (CFU/mL) decreased as the infection time increased, independent of the MOI to which the cultures were made. This nding suggests that, under in vitro culture conditions for 30 days, A. butzleri is capable of surviving but not multiplying within A. castellanii and can be considered a stable relationship over time. Furthermore, we previously showed that during the early stages of endosymbiosis, there is not colocalization between amoebic vacuoles containing A. butzleri and mitochondria or ER vesicles of A. castellanii, which would contribute. Therefore, the lack of replication of A. butzleri may be associated with its inability to access nutrients that are found in endoplasmic reticulum vesicles and mitochondria.
In this context, two mechanisms have been proposed to explain the interaction of amoebas with bacteria. In the rst of them, the amoeba acts as a reservoir where the bacteria are able to evade the defenses of the amoeba, which allows its survival and multiplication (Kebbi-Beghdadi and Greub, 2014). The second mechanism proposes that the amoeba acts as a "Trojan horse", allowing the bacterium to remain viable without multiplying in number, which would indicate that the microorganism has mechanisms that protect it from the lysosomal action of the amoeba (Ridwane et al., 2021). Based on this evidence, the observed results suggest that A. castellanii would participate as a transmission vehicle for A. butzleri.
In order to determine the viability and multiplication capacity of A. butzleri in the presence of A. castellanii, co-cultures using semi-permeable membranes were performed, allowing both microorganisms to be kept in the same culture but in separate compartments. This interaction model has been used to describe the symbiotic relationships between various bacteria and FLA. The results show that both strains of A. butzleri remain viable for at least 30 days (under controlled conditions) in the presence of A. castellanii ( Figure 2). For both strains, no signi cant differences were observed until day 25 of infection between the number of viable bacteria maintained in co-culture with respect to the control, which corresponded to a suspension of bacteria, free of amoebas, incubated under the same conditions as the co-cultures. On the other hand, on day 30 of infection signi cant differences were observed with respect to the control. As reported in the rst interaction model analyzed, no multiplication of A. butzleri strains was observed in the presence of A. castellanii (Figure 2).
Available evidence regarding this second analyzed model suggests that the extracellular survival of bacteria would depend on a diffusible factor produced and secreted by the amoeba, which highlights the possibility that these microorganisms coexist in the environment, being considered as an interaction of a high level of complexity and dependent on each species. Thus, the observed results show that A. castellanii does not promote the survival of A. butzleri, nor its multiplication, as both microorganisms remain in the same culture separated by semi-permeable membranes. This indicates that, in the absence of nutrients, the amoeba would not produce any diffusible factor that positively or negatively alters the survival of the bacteria, at least during the rst 25 days of incubation. Therefore, the greater survival capacity of A. butzleri is associated with the endosymbiont status of A. castellanii, which shows the complexity of this type of symbiotic relationship.
Finally, the results of this study allow us to conclude that A. butzleri is capable of surviving as an endosymbiont of A. castellanii for at least 30 days, without multiplying, under controlled laboratory conditions. In addition, in the absence of nutrients and as both microorganisms remain in the same culture, separated by semi-permeable membranes, A. castellanii does not promote the survival of A. butzleri, nor does it multiply.  Figure 1 Endosymbiotic culture assays. Graphics represent the results indicating the count of viable intra-amoebic bacteria expressed as log CFU/mL. a: A. butzleri ATCC 49616 (MOI 1:40); b: A. butzleri ATCC 49616 (MOI 1:100); c: A. butzleri HF28 (MOI 1:40); d: A. butzleri HF28 (MOI 1:100). Data were analyzed using Student's t test through Graph Pad software. (* p < 0.05; ** p < 0.01; *** p < 0.001) Figure 2 Transwell co-culture assays. Graphics represent the results indicating the count of viable bacteria expressed as log CFU/mL. a: A. butzleri ATCC 49616; b: A. butzleri HF28. Data were analyzed using Student's t test through Graph Pad software. (* p < 0.05; ns: not signi cant).