*2.3. Co-Culture Experiments*

2.3.1. *L. pneumophila* Lens co-cultivated with Amoeba Strains

The mean initial amount of amoeba-internalized bacteria at 22 ◦C was 16 ± 0.5% (16% in *A. castellanii*, 15% in *W. magna* C2c Maky, and 16% in *W. magna* Z503). Seen at 37 ◦C, a mean bacterial uptake of 20 ± 5.5% was observed (15% in *A. castellanii*, 26% in *W. magna* C2c Maky, and 18% in *W. magna* Z503).

A significant decrease (*p* < 0.05) in the number of intracellular *L. pneumophila* Lens per *W. magna* C2c Maky cell was observed after 24 h (5-fold and 10-fold reduction at 22 ◦C and 37 ◦C, respectively), while the level remained nearly constant for *A. castellanii* at 22 ◦C and 37 ◦C and for *W. magna* Z503 at 22 ◦C with no significant difference between T0 and T0 + 24 h (*p* > 0.05) (Figure 3). Occurring at T0 + 96 h (Figure 3), the percentage of intracellular *L. pneumophila* Lens per *W. magna* C2c Maky cell was reduced by 48 ± 0.3% at 22 ◦C and 77 ± 1.2% at 37 ◦C, and an increase was observed for *W. magna* Z503 (9-fold at 22 ◦C and 5-fold at 37 ◦C) and *A. castellanii* (19-fold at 22 ◦C and 50,000-fold at 37 ◦C). Observed at 37 ◦C, a small number of *A. castellanii* cells were still alive (5.6 × 10<sup>2</sup> ± 5.9 × 10<sup>2</sup> amoebas/mL), demonstrating that amoeba cell lysis occurred following the intracellular multiplication of *L. pneumophila* Lens.

**Figure 3.** Comparison of the evolution of the number of intracellular *L. pneumophila* cells (Lens, Paris,

and Philadelphia) per amoeba cell (*A. castellanii, W. magna* C2c Maky, and *W. magna* Z503). Results are expressed as the mean +/− 95% CI (Confidence Interval based on the standard error of the mean). (**a**) *L. pneumophila* number per *A. castellanii* cell at 22 ◦C (n = 9 for Lp Lens and Paris, n = 15 for Lp Philadelphia); (**b**) *L. pneumophila* number per *A. castellanii* cell at 37 ◦C (n = 9); (**c**) *L. pneumophila* number per *W. magna* cell (C2c and Z503) at 22 ◦C (n = 9 for Lp Lens and Paris, n = 15 for Lp Philadelphia); (**d**) *L. pneumophila* number per *W. magna* cell (C2c and Z503) at 37 ◦C (n = 9).

Considering the number of *L. pneumophila* Lens at 22 ◦C and 37 ◦C, a significant increase (*p* < 0.05) was obtained when the bacterium was co-cultivated with *W. magna* Z503 and *A. castellanii*, and this was not observed when *L. pneumophila* Lens was cultivated alone or in the presence of *W. magna* C2c Maky (Figure 4a,b), demonstrating an intracellular multiplication of *L. pneumophila* Lens in *W. magna* Z503 and *A. castellanii* as the bacterium was unable to multiply by itself in the coculture medium (Figure 1a,b).

**Figure 4.** Comparison of the evolution of the number of *L. pneumophila* cells in the presence or absence of amoeba cells (alone, or in presence of *A. castellanii*, *W. magna* C2c Maky, or *W. magna* Z503). Results are expressed as the mean +/− 95% CI (Confidence Interval based on the standard error of the mean). (**a**) *L. pneumophila* Lens at 22 ◦C (n = 9); (**b**) *L. pneumophila* Lens at 37 ◦C (n = 9); (**c**) *L. pneumophila* Paris at 22 ◦C (n = 9); (**d**) *L. pneumophila* Paris at 37 ◦C (n = 9); (**e**) *L. pneumophila* Philadelphia at 22 ◦C (n = 15); (**f**) *L. pneumophila* Philadelphia at 37 ◦C (n = 9).

#### 2.3.2. *L. pneumophila* Paris Co-Cultivated with Amoeba Strains

Occurring at 22 ◦C, we reported a mean *L. pneumophila* Paris uptake by amoebas of 24 ± 1.5% (25% in *A. castellanii*, 23% in *W. magna* C2c Maky, and 23% in *W. magna* Z503). The initial mean amount of cells internalized by amoebas decreased to 14 ± 5.0% at 37 ◦C (9% in *A. castellanii*, 19% in *W. magna* C2c Maky and 13% in *W. magna* Z503).

A significant decrease of the number of intracellular *L. pneumophila* Paris per amoeba cell (*p* < 0.05) first was observed in the three amoebas after 24 h, with the exception of *A. castellanii* at 37 ◦C (8-fold for *W. magna* C2c Maky, 3-fold for *W. magna* Z503, and 9-fold for *A. castellanii* at 22 ◦C and 19-fold for *W. magna* C2c Maky, 11-fold for *W. magna* Z503, and 2-fold for *A. castellanii* at 37 ◦C) (Figure 3). This decrease was maintained until the end of the experiment (T0 + 96 h) only by *W. magna* C2c Maky, and the percentage of intracellular *L. pneumophila* Paris per amoeba cell was reduced by 79 ± 2% at 22 ◦C and 98 ± 0.1% at 37 ◦C (*p* < 0.05). The opposite was observed for *W. magna* Z503 and *A. castellanii* at 22 ◦C and 37 ◦C, as the decrease measured after 24 h was not maintained. Seen at 48 h, the level of intracellular *L. pneumophila* Paris per amoeba cell began to increase until it reached 4-fold and 3-fold more bacteria per amoeba cell than that observed at T0 for *W. magna* Z503 and *A. castellanii*, respectively at 22 ◦C. Observed at 37 ◦C for *W. magna* Z503, the number of intracellular *L. pneumophila* Paris per amoeba cell at T0 + 96 h was 5-fold the ratio observed at 24 h, but it did not reach the initial ratio. Regarding *A. castellanii*, a strong increase was observed at both temperatures, and the initial ratio was slightly increased by 3-fold at 22 ◦C (*p* > 0.05) and strongly increased by 60,000-fold at 37 ◦C (*p* < 0.05). Furthermore, the correlation between the increase in *L. pneumophila* Paris and the low concentration of viable *A. castellanii* (5.6 × 10<sup>2</sup> ± 5.9 × 10<sup>2</sup> cells/mL) after 96 h indicated that a high intracellular multiplication of *L. pneumophila* Paris occurred that was followed by a release of bacteria in the medium after *A. castellanii* death.

Considering the number of *L. pneumophila* Paris at 22 ◦C, a significant increase (*p* < 0.05) was obtained when the bacterium was co-cultured with *W. magna* Z503 and *A. castellanii*, and this was not observed when *L. pneumophila* Paris was cultured alone or in the presence of *W. magna* Z503 at 37 ◦C and

*W. magna* C2c Maky at both 22 ◦C and 37 ◦C (Figure 4c,d), demonstrating an intracellular multiplication of *L. pneumophila* Paris in *W. magna* Z503 and *A. castellanii* at 22 ◦C and only in *A. castellanii* at 37 ◦C as the bacterium was unable to multiply by itself in the coculture medium (Figure 1a,b).

#### 2.3.3. *L. pneumophila* Philadelphia Co-Cultivated with Amoeba Strains

The mean bacterial internalization by amoebas was 9 ± 1.1% (9% in *A. castellanii,* 10% in *W. magna* C2c Maky, and 7% in *W. magna* Z503) at 22 ◦C, and the initial amount of internalized cells by amoebas increased to 17 ± 3.8% (19% in *A. castellanii*, 20% in *W. magna* C2c Maky, and 13% in *W. magna* Z503).

Occurring at 22 ◦C, a rapid and significant (*p* < 0.05) decrease in the number of intracellular *L. pneumophila* per amoeba cell was observed within 24 h (20-fold for *A. castellanii*, 11-fold for *W. magna* C2c Maky, and 10-fold for *W. magna* Z503) in the three amoebas (Figure 3). Then, a slow but significant (*p* < 0.05) decrease continued until the death of more than 99% of intracellular *L. pneumophila* Philadelphia in all cases. Even if this decrease could be attributed to the bacterial death in the coculture medium, the experiment demonstrated the absence of intra-amoeba multiplication of *L. pneumophila* Philadelphia necessary for survival at 22 ◦C.

Occurring at 37 ◦C, a similar rapid decrease in the number of intracellular *L. pneumophila* per amoeba was observed within 24 h for all three amoebas (20-fold for *A. castellanii*, 10-fold for *W. magna* C2c Maky, and 92-fold for *W. magna* Z503). Then, di fferential behaviours were observed depending on the amoeba strains. Regarding *W. magna* C2c Maky, the significant decrease (*p* < 0.05) continued until the death of more than 99.99% of the intracellular *L. pneumophila* Philadelphia per amoeba cell (Figure 3d). Concerning *W. magna* Z503, a decrease also was observed up to 97% elimination of intracellular *L. pneumophila* Philadelphia per amoeba cell after 96 h (*p* < 0.05) (Figure 3d). To contrast, for *A. castellanii*, a significant increase (*p* < 0.05) in intracellular *L. pneumophila* Philadelphia per amoeba cell appeared after 48 h, demonstrating an intra-amoeba multiplication up to 2600-fold at the end point (Figure 3c).

Considering the number of *L. pneumophila* Philadelphia at 22 ◦C, a significant decrease (*p* < 0.05) was obtained in all cases (Figure 4e), while at 37 ◦C, a significant increase (*p* < 0.05) was observed when *L. pneumophila* Philadelphia was cultured in the presence of *A. castellanii* (Figure 4f). This demonstrated an intracellular multiplication of *L. pneumophila* Philadelphia *A. castellanii* at 37 ◦C, as the bacterium was unable to multiply by itself in SCYEM medium (Figure 1a,b).

#### *2.4. Microscopic Observations of Intracellular L. pneumophila Philadelphia at 37* ◦*C*

Microscopic observations were performed at T0, T0 + 48 h, and T0 + 96 h. Occurring at T0, excess intracellular *L. pneumophila* Philadelphia bacteria were observed in the presence of the three amoebas (Figure 5A,D,G). Regarding *A. castellanii* at 48 h, a strong bacterial multiplication was observed (Figure 5B) which was not observed for both *W. magna* strains (Figure 5E,H). Occurring at 96 h, lysis of *A. castellanii* after intracellular bacterial multiplication was clearly evident (Figure 5C), and only a small amount of amoeba lysis could be observed for both *W. magna* strains (Figure 5F,I).

#### *2.5. Statistical Comparison of Amoeba Behavior*

Analysis of variance tests (ANOVA) were performed to determine if *W. magna* C2c Maky interacted with *L. pneumophila* in a significantly di fferent manner compared to interactions with the two other amoebas.

Concerning the three bacterial strains, T0 data obtained in the presence of the three amoebas were not statistically di fferent at 22 ◦C (*p* > 0.05); however, at 37 ◦C, a significant di fference in behaviour (*p* < 0.05) was detected at T0.

Pairwise comparisons (Dunn test) established that at 72 h and 96 h at both temperatures and with the three legionella strains, *W. magna* C2c Maky behaviour was statistically di fferent from that of the two other amoeba strains (Table 1). This significant di fference was observed even after 24 h with strain Paris at both temperatures, and at 22 ◦C for strain Lens. Statistical tests provided evidence

that *W. magna* C2c Maky behaved differently compared to *W. magna* Z503 and *A. castellanii* cells in the presence of *Legionella* strains.

**Figure 5.** Optical microscopy observation using Gimenez staining of *A. castellanii* (**A**–**C**), *W. magna* C2c Maky (**D**–**F**), and *W. magna* Z503 (**G**–**I**) infected with *L. pneumophila* Philadelphia at 37 ◦C. Photos of the co-cultures were acquired at T0 (**A**,**D**,**G**), T0 + 48 h (**B**,**E**,**H**), and T0 + 96 h (**C**,**F**,**I**).


**Table 1.** Statistical analysis of the behaviour of the three amoeba strains in the presence of the three *Legionella* strains at 22 ◦C and 37 ◦C. Significant differences for *W. magna* C2c Maky are highlighted in yellow.
