1 Amoeba-resisting bacteria found in multilamellar bodies secreted by Dictyostelium
2 discoideum: social amoebae can also package bacteria
3
4 Valérie E. Paquet1,2 and Steve J. Charette1,2,3*
5
6 1. Institut de Biologie Intégrative et des Systèmes, Pavillon Charles-Eugène-Marchand,
7 Université Laval, Quebec City, QC, Canada
8 2. Centre de recherche de l’Institut universitaire de cardiologie et de pneumologie de
9 Québec, Hôpital Laval, Quebec City, QC, Canada
10 3. Département de biochimie, de microbiologie et de bio-informatique, Faculté des
11 sciences et de génie, Université Laval, Quebec City, QC, Canada
12
13 *Corresponding author:
14 Steve J. Charette, 1030 avenue de la medicine, Pavillon Marchand, local 4245, Université
15 Laval, Quebec City, QC, Canada, G1V 0A6, telephone: 1-418-656-2131, ext. 6914, fax:
16 1-418-656-7176, email: [email protected]
17
18 Running title (60 characters with space): Packaging of amoeba-resisting bacteria by D.
19 discoideum
20
21 Keywords (6): Multilamellar bodies; Dictyostelium discoideum; packaged bacteria,
22 amoeba-resisting bacteria, Cupriavidus, Rathayibacter
23
1 24 ABSTRACT
25 Many bacteria can resist phagocytic digestion by various protozoa. Some of these
26 bacteria (all human pathogens) are known to be packaged in multilamellar bodies
27 produced in the phagocytic pathway of the protozoa and that are secreted into the
28 extracellular milieu. Packaged bacteria are protected from harsh conditions, and the
29 packaging process is suspected to promote bacterial persistence in the environment. To
30 date, only a limited number of protozoa, belonging to free-living amoebae and ciliates,
31 have been shown to perform bacteria packaging. It is still unknown if social amoebae can
32 do bacteria packaging. The link between the capacity of 136 bacterial isolates to resist the
33 grazing of the social amoeba Dictyostelium discoideum and to be packaged by this
34 amoeba was investigated in the present study. The 45 bacterial isolates displaying a
35 resisting phenotype were tested for their capacity to be packaged. A total of seven isolates
36 from Cupriavidus, Micrococcus, Microbacterium, and Rathayibacter genera seemed to
37 be packaged and secreted by D. discoideum based on immunofluorescence results.
38 Electron microscopy confirmed that the Cupriavidus and Rathayibacter isolates were
39 formally packaged. These results show that social amoebae can package some bacteria
40 from the environment revealing a new aspect of microbial ecology.
41
42
2 43 INTRODUCTION
44 Free-living amoebae (FLAs) like Acanthamoeba spp. are mobile unicellular
45 protozoa that live in aquatic environments and feed on bacteria, fungi, and algae
46 (Rodriguez-Zaragoza, 1994). FLAs can colonize many man-made infrastructures that
47 provide a favorable environment for the proliferation of microorganisms, especially
48 where high bacterial population densities are found. Cooling towers (Pagnier et al.,
49 2009), air conditioners (Walker et al., 1986), and drinking water distribution systems
50 (Thomas & Ashbolt, 2011) are a few examples of man-made infrastructures where FLAs
51 grow (reviewed in (Siddiqui & Khan, 2012, Cateau et al., 2014) and regulate bacterial
52 population densities.
53
54 FLAs capture bacteria by phagocytosis and transfer them to lysosomal
55 compartments in the phagocytic pathway where they are usually digested by enzymes
56 (Siddiqui & Khan, 2012). However, some bacteria referred to as amoebae-resisting
57 bacteria (ARBs) are able to avoid or withstand enzymatic degradation in the phagocytic
58 pathway through various mechanisms and can survive amoeba predation and lodge inside
59 amoebae (Loret et al., 2008). ARBs include human pathogenic bacteria such as
60 Legionella, Chlamydia, and Mycobacteria. It has also recently been shown that the ARB
61 group includes non-pathogenic bacteria (Kebbi-Beghdadi & Greub, 2014).
62
63 ARBs can survive and grow within amoebae and may then escape by cell lysis or
64 exocytosis as free bacteria, or by being packaged in fecal pellets, which are usually
3 65 several concentric layers of lipid membranes known as multilamellar bodies (MLBs). The
66 secretion of packaged bacteria has been confirmed only for a number of human pathogens
67 (Legionella pneumophila, Salmonella enterica, Listeria monocytogenes, Helicobacter
68 pylori, and Escherichia coli O157:H7), but this process has been studied only with FLAs
69 and protozoa of the ciliate group (reviewed by Denoncourt et al. 2014).
70
71 Packaging provides bacteria with a number of advantages in unfavorable
72 conditions (Berk et al., 1998; Brandl et al., 2005, Gourabathini et al., 2008, Raghu
73 Nadhanan and Thomas, 2014). For example, Salmonella enterica bacteria packaged in
74 MLBs by the ciliate Tetrahymena are more resistant to low concentrations of calcium
75 hypochlorite than when they are in the planktonic state (Brandl et al., 2005). S. enterica
76 can even multiply inside pellets.
77
78 The social amoeba Dictyostelium discoideum is a bacterial predator that lives in
79 damp forest floors. The virulence traits and host-pathogen relationships of more than 20
80 pathogenic bacterial species have been studied using this amoeba as a model (Cosson &
81 Soldati, 2008, Bonifait et al., 2011, Dallaire-Dufresne et al., 2011). D. discoideum is
82 often compared to a macrophage-like organism that shares many proteins, such as
83 lysosomal hydrolases involved in intracellular killing, that are found in specialized
84 phagocytic cells in mammals (Cosson & Lima, 2014). D. discoideum produces (Mercanti
85 et al., 2006) and secretes large amounts of MLBs when fed digestible bacteria (Paquet et
86 al., 2013). While no studies on bacteria packaging by D. discoideum have been
4 87 published, inert polystyrene beads can be packaged in D. discoideum MLBs in presence
88 of digestible bacteria (Denoncourt et al., 2014).
89
90 We propose that D. discoideum has also the capacity to package ARBs in MLBs.
91 In the present study, 136 bacterial strains of various genera and environments were tested
92 for their capacity to resist D. discoideum predation and to determine whether these newly
93 identified ARBs are packaged in expelled MLBs. As expected, some ARBs were
94 packaged in D. discoideum MLBs and were secreted into the extracellular milieu.
95
5 96 MATERIALS AND METHODS
97 Amoebae
98 D. discoideum DH1-10 cells (Cornillon et al., 2000) were grown at 21°C in HL5
99 medium supplemented with 15 µg/mL of tetracycline (Mercanti et al., 2006). The cells
100 were subcultured twice a week in fresh medium to prevent the cultures from reaching
101 confluence. They were also grown on bacterial lawns as described below.
102
103 Bacteria
104 Klebsiella aerogenes was a kind gift from Pierre Cosson (Geneva University,
105 Switzerland), 19 bacterial isolates were provided by Martin Filion (Moncton University,
106 Canada) (Filion et al., 2004), and 78 bacterial isolates were provided by Janet Martha
107 Blatny et al. (Norwegian University of Science and Technology, Norway) (Dybwad et
108 al., 2012). All the other isolates used in the present study were from a drinking water
109 distribution network model (Berthiaume et al., 2014) or were obtained from ATCC or
110 USDA. Stock cultures were stored at -80°C in LB (EMD, Canada) supplemented with
111 15% glycerol. As needed, the stock cultures were thawed and were inoculated on Tryptic
112 Soy Agar (TSA) (EMD, Canada) plates, which were incubated at 25°C, typically for two
113 days, before being used for the experiments.
114
115 Predation resistance assay
116 Bacterial isolates grown on TSA plates were resuspended in 3 mL of LB, and the
117 OD at 595 nm was adjusted to 1. The resuspended bacteria (300 µL) were plated on three
118 different nutrient media (HL5: bacto peptone (Oxoid) 14.3 g L-1, yeast extract 7.15 g L-1,
6 -1 -1 -1 119 maltose monohydrate 18 g L , Na2HPO4.2H2O 0.65 g L , KH2PO4 0.5 g L , and bacto
-1 -1 -1 -1 120 agar 20 g L ); SM: bacto peptone 10 g L , yeast extract 1 g L , KH2PO4 2.2 g L ,
-1 -1 -1 1/10 121 K2HPO4 1 g L , MgSO4 1 g L , and bacto agar 20 g L ); or SM (the ingredients for
122 SM were all diluted 1/10 except for the bacto agar). The plates were allowed to dry under
123 sterile conditions to obtain bacterial lawns.
124
125 The tetracycline from the amoeba cell culture maintenance was removed by
126 medium replacement, and the D. discoideum cells were resuspended in fresh HL5 with no
127 antibiotic before counting them in a hemacytometer chamber. Serial dilutions were
128 prepared in HL5 medium to obtain the following D. discoideum cell concentrations:
129 500,000; 50,000; 5,000; 500, 50, and 5 cells per 5 µL. The bacterial lawns were spotted
130 with 5 µL of the serial D. discoideum dilutions. The plates were allowed to dry and were
131 incubated at 21°C for 7 days. They were examined visually for plaque formation on days
132 1, 3, and 7. The isolates that did not allow the growth of amoebae were considered as
133 ARBs.
134
135 Bacteria/amoebae co-cultures
136 The identified ARBs were co-cultured alone or were mixed in a final volume of
137 300 µL with digestible K. aerogenes (Ka), which is known to stimulate the production of
138 MLBs (Paquet et al., 2013), and with 30 prewashed D. discoideum cells. The mixtures
139 were spread on SM agar plates. Serial Ka:ARB ratios ([99:1], [9:1] [1:1], [1:9], and
140 [1:99], in a total volume of 300 µL), based on an OD adjusted to 1, were used to
141 determine the best conditions for D. discoideum growth on bacterial co-cultures. The
7 142 plates were incubated at 21°C for 14 days and were examined visually for phagocytic
143 plaque formation, bacterial colonies within the phagocytic plaques, or all other
144 anomalous growth on days 3, 9, and 14.
145
146 Production of packaged and secreted ARBs
147 Potential packaged bacteria deduced from the bacteria/amoebae co-culture results
148 were mixed in a final volume of 300 µL with Ka using the best ratio determined from
149 previous experiments and were plated on SM1/10 agar. Drops (5 µL) containing 100,000
150 D. discoideum cells were spotted on the bacterial lawns. The plates were allowed to dry
151 and were incubated for 3 or 4 days at 21°C to obtain large phagocytic plaques. Samples
152 from the peripheries of the phagocytic plaques were collected using sterile tips. The
153 samples were gently diluted in fresh SM1/10 medium and were processed for
154 immunofluorescence (IF) or transmission electron microscopy (TEM) as described
155 below.
156
157 Immunofluorescence
158 The samples containing suspended cells and material from the peripheries of
159 phagocytic plaques were allowed to adhere to glass coverslips for 3 h and were then fixed
160 in 4% paraformaldehyde for 30 min. The coverslips were rinsed with PBS 1X (1.9 mM
161 NaH2PO4 + H2O; 8.1 mM Na2HPO4 + 2 H2O; 154 mM NaCl, pH 7.4) containing 40 mM
162 NH4Cl to stop the fixation and then with PBS 1X. The cells were permeabilized for 2 min
163 with methanol at -20 °C, and the coverslips were rinsed with PBS 1X and then with PBS
164 1X containing 0.2 % bovine serum albumin (PBS-BSA) at room temperature for at least 5
8 165 min to block non-specific binding sites. The adherent cells were then incubated for 45
166 min with the H36 antibody (Mercanti et al., 2006) diluted 1:1000 in PBS-BSA and then
167 with Alexa 568-coupled anti-mouse IgG secondary antibody (diluted 1:400; Invitrogen,
168 Canada) and 2.5 µg/mL of DAPI (4,6-di-amidino-2-phenylindole diluted in PBS-BSA)
169 for 30 min at room temperature in the dark. The coverslips were washed at least three
170 times with PBS-BSA between each step. The coverslips were mounted on glass slides
171 using Prolong Gold (Invitrogen). Images were acquired using an Axio Observer Z1
172 microscope equipped with an Axiocam camera (Carl Zeiss, Canada).
173
174 Transmission electron microscopy
175 Samples from the bacteria/amoebae co-cultures and material from the peripheries
176 of the phagocytic plaques were collected using sterile tips and were fixed for 3 h in 0.1 M
177 sodium cacodylate buffer (pH 7.3) containing 2 % glutaraldehyde and 0.3 % osmium
178 tetroxide. They were washed three times with sodium cacodylate buffer and were
179 dehydrated for 5 min in 30 % ethanol, 5 min in 50 % ethanol, 5 min in 70 % ethanol, 10
180 min in 95 % ethanol, and 1 h in 100 % ethanol. The samples were then embedded in
181 Epon resin and were incubated overnight at 37 °C followed by 3 days at 60 °C. Very thin
182 slices (60 to 80 nm) were cut and were stained for 8 min with 0.1 % lead citrate and then
183 for 5 min with 3 % uranyl acetate. They were then examined using a transmission
184 electron microscope (JEOL 1230) at 80 kV.
185
186
9 187 RESULTS AND DISCUSSION
188
189 Predation resistance assay
190 D. discoideum is probably the simplest system for assessing bacterial virulence
191 (Hilbi et al., 2007, Froquet et al., 2009). Because medium richness may have an impact
192 on the results of predation resistance assays (Froquet et al., 2007, Filion & Charette,
193 2014), our assays were performed using three different media of varying composition and
194 richness (HL5, SM, SM1/10).
195
196 Phagocytic plaques, which are bacteria-free zones due to amoeba grazing, are
197 produced when amoebae are spotted on lawns of digestible bacteria (Figure 1).
198 Phagocytic plaques were not observed in the presence of ARBs or were observed only for
199 the highest D. discoideum cell concentrations (Figure 1C and D) (Filion & Charette,
200 2014). Ka is used routinely in many phagocytic experiments to feed D. discoideum,
201 which is why we used it as a positive control for amoeba predation (Figure 1B) (Froquet
202 et al., 2009).
203
204 We considered that the isolates were ARBs when 500 or fewer D. discoideum
205 cells were unable to produce phagocytic plaques on the bacterial lawn for at least one of
206 the media tested. For example, it is the case for Cupriavidus sp. and Microbacterium sp.
207 isolates shown in Figure 1C and D. Isolates that allowed the growth of the amoebae with
208 an initial inoculum of 500 D. discoideum cells per drop or less were considered sensitive
209 to amoeba predation and were rejected for subsequent experiments.
10 210 A total of 136 bacterial isolates were screened with the amoeba predation assay to
211 identify those that were potential ARBs. All the experiments were performed twice, and
212 45 isolates were considered as D. discoideum resisting bacteria and, as such, potential
213 candidates for the packaging process (see Table S1).
214 The newly discovered ARBs were not specific to one phylum but belonged to various
215 clades distributed throughout the prokaryotes, which was in agreement with a study by
216 Moliner et al. (Moliner et al., 2010). Table 1 presents the ARBs discovered in the present
217 study. Our results suggested that the adaptation of bacteria to avoid digestion during
218 phagocytosis is widespread in bacteria. Moreover, the term ARB cannot be generalized
219 and be applied to an entire genus or species since bacteria from the same genus or species
220 did not display the same resistance to predation (Table 1).
221
222 Triple co-cultures
223 The 45 newly identified ARBs were co-cultured with digestible bacteria (Ka) and
224 D. discoideum. The goal of this experiment was to assess the growth of amoebae on
225 digestible bacteria (Ka) in the presence of ARBs to determine whether the ARBs were
226 toxic for the amoebae, making it impossible for them to produce packaged bacteria. All
227 the phagocytic plaques with a profile similar to the positive control, that is, with a large
228 bacteria-free zone (black arrow, Figure 2A) due to extensive amoeba growth, were
229 rejected. Similarly, co-cultures where no amoeba growth occurred, as for the negative
230 control, were also rejected. For example, all the Ka:Luteibacter anthropic ratios produced
231 small phagocytic plaques compared to the plaques produced by amoebae grown only on
232 Ka, suggesting that L. anthropic was toxic to the amoebae or markedly limited their
11 233 growth (black arrow, Figure 2B). Conversely, the presence of bacterial colonies in the
234 middle of grazing plaques (black arrow at top, Figure 2C) or substantial growth of the
235 ARB around phagocytic plaques (black arrow at the bottom, Figure 2C) indicated that the
236 ARB was resistant to predation and had no obvious toxicity for D. discoideum. One
237 possibility is that the bacteria passed through the phagocytic pathway and were expelled
238 as packaged bacteria, which then began to grow and form colonies. Three Cupriavidus
239 and 17 other isolates displayed this profile (Table 2). Thus based on the unusual growth
240 pattern of amoebae on their lawns, 20 isolates were considered as ARBs and were
241 retained in order to determine whether they were packageable.
242
243 Bacteria packaging by D. discoideum
244 The next step was to determine whether D. discoideum cells were able to package
245 ARBs. Based on previous packaging assays by Gourabathini et al. with E. coli O157:H7
246 and the ciliate Tetrahymena pyriformis (Gourabathini et al., 2008), packaged bacteria
247 released on a rich medium are able to grow inside the package and break out. Indeed,
248 packaged bacteria are likely a transitory state, allowing the bacteria to survive in harsh
249 conditions (Berk et al., 1998, Marciano-Cabral & Cabral, 2003) until they are released
250 into an environment that is more favorable for bacterial growth. Packaged ARBs were not
251 observed during the triple co-culture experiments using rich medium even after a long
252 period of time probably due to growth of potentially packaged bacteria. On the other
253 hand, starvation media (Smith et al., 2010), which contains only few nutriments to
254 prevent bacterial growth have been also tried, but they induce the multicellular
255 development of amoebae despite the presence of digestible bacteria (data not shown).
12 256 Again, no packaged bacteria were seen because active vegetative D. discoideum cells are
257 required for the packaging process to occur.
258
259 The stimulation of bacteria packaging and secretion was also studied using diluted
260 nutrient agar (SM1/10) to avoid rapid bacterial growth following exocytosis that could
261 break up the packages. We observed amoebae on mixed bacterial lawns of digestible
262 bacteria and ARBs (see ratios and strains in Table 2). Samples collected at the peripheries
263 of the phagocytic plaques were examined by IF with the H36 antibody (Mercanti et al.,
264 2006) and by TEM.
265
266 A sample containing potential packaged bacteria had to display combined DAPI and
267 H36 antibody-positive staining for structures smaller than amoebae but bigger than free-
268 living bacteria (data not shown) due to packaging of bacteria. DAPI would reveal the
269 presence of bacteria in the structures. On its side, H36 antibody has been shown in a
270 previous study to be a specific marker of MLBs by binding to a protein still not
271 characterized (Paquet et al, 2013). The magenta arrows in Figure 3 point to bacteria
272 packages measuring 2 to 3 µm in diameter, and the black arrow indicates a D. discoideum
273 cell. Of the 20 potential candidates tested by IF, three Cupriavidus isolates, two
274 Micrococcus luteus isolates, and one isolate each of Rathayibacter tritici and
275 Microbacterium oxydans presented features suggesting that they were packaged by D.
276 discoideum.
277
13 278 The same co-culture protocol was performed on several samples to formally confirm
279 the presence of expelled packaged bacteria by TEM. For the control condition shown on
280 Figure 4, D. discoideum produced (white arrow, Figure 4B) and secreted empty MLBs
281 (black arrow, Figure 4C) in the presence of digestible bacteria on SM1/10. However,
282 D. discoideum produced fewer MLBs on SM1/10 than on rich HL5 medium (Paquet et al.,
283 2013). Despite this, Cupriavidus sp. and R. tritici were found inside secreted MLBs when
284 they were co-cultured with amoeba and digestible bacteria (Figure 4E, F, and I). The
285 TEM observations revealed that some of the tested bacteria could be packaged by D.
286 discoideum.
287
288 Interestingly, R. tritici accumulated inside the amoebae, with up to 50 undigested
289 bacteria visible inside each D. discoideum cell (Figure 4H). It is not clear whether the
290 accumulation was due to rapid bacterial growth inside the amoebae, the inhibition of the
291 exocytic process, or a combination of both. While the mechanism involved is not known,
292 this result suggested that bacteria can also survive in harsh environments by residing
293 inside amoebae. The intracellular survival in protozoa of many bacteria has been
294 described in the past (reviewed in Denoncourt et al., 2014). Many bacteria of the genus
295 Rathayibacter are phytopathogens of terrestrial plants (Hahn et al., 2003, Schaad &
296 Schuenzel, 2010), and it is likely that amoebae and these soil bacteria interact.
297
298 We showed that the packaging of bacteria is possible by D. discoideum amoeba model
299 and that the phenomenon is not restricted to specific genera. Indeed, both Gram-negative
300 and -positive bacteria from various environments, including soil and water, were trapped
14 301 inside the MLBs. Moreover, the outcome of various isolates from a same genera or even
302 a same species regarding packaging is fairly variable. For example, 23 strains of M.
303 luteus were tested using the predation assay and 9 were identified as ARBs, two of which
304 were packaged in MLBs based on the IF results. Thirteen Pseudomonas strains were also
305 tested using the predation assay. While 4 displayed an ARB phenotype, none was
306 packaged in MLBs. These results indicated that bacterial adaptive evolution with respect
307 to protozoa is complex, as has been shown by the farming of different strains of
308 Burkholderia sp. by non-farmer D. discoideum (DiSalvo et al., 2015). Given this, it
309 would be difficult to predict whether a given bacterial isolate can be packaged or can
310 resist predation by a specific protozoan without in vitro testing. It would thus be
311 interesting to determine whether the same ARBs are packaged by different wild-type
312 strains of D. discoideum or other protozoa.
313
314 Lastly, the present study showed that some ARBs are packaged in MLBs and are
315 secreted by D. discoideum in laboratory conditions. Amoeba/bacteria interactions are
316 ubiquitous in natural as well as in man-made environments such as in municipal drinking
317 water storage tank sediments (Lu et al., 2015), the floating and fixed biofilms of spring
318 recreation areas (Hsu et al., 2011), and the surface water of warm water systems and
319 cooling towers (Kuiper et al., 2006). As such, it is likely that bacteria packaging occurs in
320 real conditions, not just in the laboratory.
15 321 CONCLUSION
322 The resistance to predation of 136 bacterial isolates was assessed using a standardized
323 D. discoideum predation assay. Forty-five of these isolates displayed an ARB phenotype
324 and were co-cultured with digestible bacteria to stimulate MLB production. Twenty
325 potential candidates were retained based on this screening. The bacteria packaging of
326 seven isolates by D. discoideum was suggested by IF and confirmed for two isolates by
327 TEM. This is the first study to show that D. discoideum can package bacteria. These
328 results open the way to a better understanding of the role of ARBs in microbial ecology
329 and their persistence in many environments.
330
331
16 332 FUNDING
333 This work was supported by grants to S. J. C. from the Fonds de la Recherche du Québec
334 – Nature et Technologies (FRQNT) [2014-PR-173418], the Chaire de pneumologie de la
335 fondation J.-D. Bégin de l’Université Laval, the Fonds Alphonse L’Espérance de la
336 fondation de l’IUCPQ, and the Establishment of young researchers - Juniors 1 program of
337 the Fonds de la Recherche du Québec en Santé (FRQS) [20004].
338
17 339 Acknowledgements
340 We are grateful to P. Cosson (University of Geneva, Switzerland) for the antibodies and
341 bacterial strains. We warmly thank the teams of J. M. Blatny (FFI, Norway) and M.
342 Filion (University of Moncton, Canada) as well as the USDA, who provided many
343 bacterial strains. We thank A. Denoncourt and A. Vincent (Université Laval, Canada) for
344 their critical reading of the manuscript and Richard Janvier (Plateforme de microscopie,
345 IBIS, Université Laval, Canada) for acquiring the transmission electron
346 microphotographs.
347
18 348 REFERENCES
349 Berk SG, Ting RS, Turner GW & Ashburn RJ (1998) Production of respirable vesicles 350 containing live Legionella pneumophila cells by two Acanthamoeba spp. Appl Environ 351 Microbiol 64: 279-286. 352 Berthiaume C, Gilbert Y, Fournier-Larente J, Pluchon C, Filion G, Jubinville E, Serodes 353 JB, Rodriguez M, Duchaine C & Charette SJ (2014) Identification of dichloroacetic 354 acid degrading Cupriavidus bacteria in a drinking water distribution network model. J 355 Appl Microbiol 116: 208-221. 356 Bonifait L, Charette SJ, Filion G, Gottschalk M & Grenier D (2011) Amoeba host model 357 for evaluation of Streptococcus suis virulence. Appl Environ Microbiol 77: 6271-6273. 358 Brandl MT, Rosenthal BM, Haxo AF et al. (2005) Enhanced survival of Salmonella 359 enterica in vesicles released by a soilborne Tetrahymena species. Appl Environ 360 Microbiol 71:1562-1569. 361 Cateau E, Delafont V, Hechard Y & Rodier MH (2014) Free-living amoebae: what part 362 do they play in healthcare-associated infections? J Hosp Infect 87: 131-140. 363 Cornillon S, Pech E, Benghezal M, Ravanel K, Gaynor E, Letourneur F, Bruckert F & 364 Cosson P (2000) Phg1p is a nine-transmembrane protein superfamily member 365 involved in dictyostelium adhesion and phagocytosis. J Biol Chem 275: 34287-34292. 366 Cosson P & Soldati T (2008) Eat, kill or die: when amoeba meets bacteria. Curr Opin 367 Microbiol 11: 271-276. 368 Cosson P & Lima WC (2014) Intracellular killing of bacteria: is Dictyostelium a model 369 macrophage or an alien? Cell Microbiol 16: 816-823. 370 Dallaire-Dufresne S, Paquet VE & Charette SJ (2011) [Dictyostelium discoideum: a 371 model for the study of bacterial virulence]. Can J Microbiol 57: 699-707. 372 Denoncourt AM, Paquet VE & Charette SJ (2014) Potential role of bacteria packaging by 373 protozoa in the persistence and transmission of pathogenic bacteria. Front Microbiol 374 5: 240. 375 DiSalvo S, Haselkorn TS, Bashir U, Jimenez D, Brock DA, Queller DC & Strassmann JE 376 (2015) Burkholderia bacteria infectiously induce the proto-farming symbiosis of 377 Dictyostelium amoebae and food bacteria. Proc Natl Acad Sci U S A 112: E5029-5037. 378 Dybwad M, Granum PE, Bruheim P & Blatny JM (2012) Characterization of airborne 379 bacteria at an underground subway station. Appl Environ Microbiol 78: 1917-1929. 380 Filion G & Charette SJ (2014) Assessing Pseudomonas aeruginosa virulence using a 381 nonmammalian host: Dictyostelium discoideum. Methods Mol Biol 1149: 671-680. 382 Filion M, Hamelin RC, Bernier L & St-Arnaud M (2004) Molecular profiling of 383 rhizosphere microbial communities associated with healthy and diseased black spruce 384 (Picea mariana) seedlings grown in a nursery. Appl Environ Microbiol 70: 3541-3551. 385 Froquet R, Lelong E, Marchetti A & Cosson P (2009) Dictyostelium discoideum: a model 386 host to measure bacterial virulence. Nat Protoc 4: 25-30. 387 Froquet R, Cherix N, Burr SE, Frey J, Vilches S, Tomas JM & Cosson P (2007) 388 Alternative host model to evaluate Aeromonas virulence. Appl Environ Microbiol 73: 389 5657-5659. 390 Gourabathini P, Brandl MT, Redding KS, Gunderson JH & Berk SG (2008) Interactions 391 between food-borne pathogens and protozoa isolated from lettuce and spinach. Appl 392 Environ Microbiol 74: 2518-2525.
19 393 Hahn MW, Lunsdorf H, Wu Q, Schauer M, Hofle MG, Boenigk J & Stadler P (2003) 394 Isolation of novel ultramicrobacteria classified as actinobacteria from five freshwater 395 habitats in Europe and Asia. Appl Environ Microbiol 69: 1442-1451. 396 Hilbi H, Weber SS, Ragaz C, Nyfeler Y & Urwyler S (2007) Environmental predators as 397 models for bacterial pathogenesis. Environ Microbiol 9: 563-575. 398 Hsu BM, Huang CC, Chen JS, Chen NH & Huang JT (2011) Comparison of potentially 399 pathogenic free-living amoeba hosts by Legionella spp. in substrate-associated 400 biofilms and floating biofilms from spring environments. Water Res 45: 5171-5183. 401 Kebbi-Beghdadi C & Greub G (2014) Importance of amoebae as a tool to isolate 402 amoeba-resisting microorganisms and for their ecology and evolution: the Chlamydia 403 paradigm. Environ Microbiol Rep 6: 309-324. 404 Kuiper MW, Valster RM, Wullings BA, Boonstra H, Smidt H & van der Kooij D (2006) 405 Quantitative detection of the free-living amoeba Hartmannella vermiformis in surface 406 water by using real-time PCR. Appl Environ Microbiol 72: 5750-5756. 407 Loret JF, Jousset M, Robert S, Saucedo G, Ribas F, Thomas V & Greub G (2008) 408 Amoebae-resisting bacteria in drinking water: risk assessment and management. Water 409 Sci Technol 58: 571-577. 410 Lu J, Struewing I, Yelton S & Ashbolt N (2015) Molecular Survey of Occurrence and 411 Quantity of Legionella spp., Mycobacterium spp., Pseudomonas aeruginosa and 412 Amoeba Hosts in Municipal Drinking Water Storage Tank Sediments. J Appl 413 Microbiol. 119:278-88 414 Marciano-Cabral F & Cabral G (2003) Acanthamoeba spp. as agents of disease in 415 humans. Clin Microbiol Rev 16: 273-307. 416 Mercanti V, Charette SJ, Bennett N, Ryckewaert JJ, Letourneur F & Cosson P (2006) 417 Selective membrane exclusion in phagocytic and macropinocytic cups. J Cell Sci 119: 418 4079-4087. 419 Moliner C, Fournier PE & Raoult D (2010) Genome analysis of microorganisms living in 420 amoebae reveals a melting pot of evolution. FEMS Microbiol Rev 34: 281-294. 421 Pagnier I, Merchat M & La Scola B (2009) Potentially pathogenic amoeba-associated 422 microorganisms in cooling towers and their control. Future Microbiol 4: 615-629. 423 Paquet VE, Lessire R, Domergue F, Fouillen L, Filion G, Sedighi A & Charette SJ (2013) 424 Lipid composition of multilamellar bodies secreted by Dictyostelium discoideum reveals 425 their amoebal origin. Eukaryot Cell 12: 1326-1334. 426 Raghu Nadhanan R, Thomas CJ (2014) Colpoda secrete viable Listeria monocytogenes 427 within faecal pellets. Environ Microbiol 16:396-404. 428 Rodriguez-Zaragoza S (1994) Ecology of free-living amoebae. Crit Rev Microbiol 20: 429 225-241. 430 Schaad N & Schuenzel E (2010) Sensitive molecular diagnostic assays to mitigate the 431 risks of asymptomatic bacterial diseases of plants. Crit Rev Immunol 30: 271-275. 432 Siddiqui R & Khan NA (2012) Biology and pathogenesis of Acanthamoeba. Parasit 433 Vectors 5: 6. 434 Smith EW, Lima WC, Charette SJ & Cosson P (2010) Effect of starvation on the 435 endocytic pathway in Dictyostelium cells. Eukaryot Cell 9: 387-392. 436 Thomas JM & Ashbolt NJ (2011) Do free-living amoebae in treated drinking water 437 systems present an emerging health risk? Environ Sci Technol 45: 860-869.
20 438 Walker PL, Prociv P, Gardiner WG & Moorhouse DE (1986) Isolation of free-living 439 amoebae from air samples and an air-conditioner filter in Brisbane. Med J Aust 145: 440 175. 441
21 442 Table 1. Taxonomic grouping of new ARBs identified by the predation assay Gram Class a Order Familia Genera Species No. of No. of isolates ARB tested isolates Positive Actino Actinomycetales Microbacteriaceae Microbacterium Microbacterium sp. 8 3 Rathayibacter Rathayibacter tritici 1 1 Micrococcaceae Kocuria Kocuria sp. 17 2 Micrococcus Micrococcus luteus 23 9 Nocardiaceae Rhodococcus Rhodococcus sp. 6 6 Streptomycetaceae Streptomyces Streptomyces 1 1 luridiscabiei Micrococcales Promicromono- Cellulosimicrobium Cellulosimicrobium 1 1 sporaceae funkei Bacilli Bacillales Paenibacillaceae Paenibacillus Paenibacillus larvae 1 1 Staphylococcaceae Staphylococcus Staphylococcus sp. 9 1 Lactobacillales Leuconostocaceae Weissella Weissella confusa 1 1 Negative Alpha - - - - 1 1 Rhizobiales Rhizobiaceae Sinorhizobium Sinorhizobium sp. 2 1 Beta Burkholderiales Burkholderiaceae Burkholderia Burkholderia sp. 3 3 Cupriavidus Cupriavidus sp. 5 4 Comamonadaceae Comamonas Comamonas 1 1 koreensis Oxalobacteraceae Duganella Duganella 1 1 zoogloeoides Gamma Enterobacteriales Enterobacteriaceae Escherichia Escherichia coli 3 2 Serratia Serratia grimesii 1 1 Pseudomonadales Pseudomonadaceae Pseudomonas Pseudomonas sp. 13 4 Xanthomonadales Xanthomonadaceae Luteibacter Luteibacter anthropi 1 1 443 a Actino = Actinobacteria; Alpha = Alphaproteobacteria; Beta = Betaproteobacteria; Gamma = Gammaproteobacteria
22 444 Table 2. ARBs identified after co-culture assays as potential candidates for bacteria 445 packaging. Strains Ratio KA:ARB Observations and comments Cupriavidus basilensis 1:1 Based on the morphology and Cupriavidus sp. 9:1 color of the colonies at the center and periphery of the Micrococcus luteus (Norway) 9:1 phagocytic plaques Micrococcus luteus US4 1:9 A few fruiting bodies, with Rathayibacter tritici 9:1 colored spores at the top. Rhodococcus erythropolis US1 1:9 Rhodococcus erythropolis US2 9:1 Rhodococcus fascians US1 9:1 Rhodococcus fascians US2 1:1 Cupriavidus necator US1 1:1 Several colonies within the Duganella zoogloeoides 1:1 phagocytic plaques. Kocuria kristinae 1:9 Microbacterium oxydans US1 9:1 Micrococcus luteus 9:1 Micrococcus luteus 8_4_14 x2 1:9 Micrococcus luteus D_1_6 x2 1:9 Micrococcus luteus US3 1:9 Rhodococcus erythropolis 1:1 Rhodococcus pyridinovorans 1:9 Cellulosimicrobium funkei 1:1 Unusual growth on agar. 446 447 448
23 449 FIGURE LEGENDS
450 Figure 1. Predation resistance assay. A. Serial dilutions of D. discoideum cells
451 (500,000 to 5 cells/5 µL) were spotted counter clockwise on bacterial lawns on HL5 agar
452 plates. The plates were incubated for 7 days. The negative control (HL5 medium only)
453 was spotted in the middle of the lawn. B. Klebsiella aerogenes is sensitive to predation by
454 amoebae. It was used as a positive control for amoeba predation. Cupriavidus sp. (C) and
455 Microbacterium sp. (D) were resistant to predation and were considered as potential
456 ARBs.
24 457 Figure 2. Triple co-cultures. Example of potential ARB isolates co-cultured with
458 digestible bacteria (Ka) and 30 D. discoideum cells on SM agar. A. A lawn of Ka was
459 used as positive control for phagocytic plaque formation (clear zones in the bacterial
460 lawn; black arrow). B. A lawn of co-cultured Ka and Luteibacter anthropic [ratio 1:9].
461 After the same incubation time, the amoebae were unable to farm the bacterial lawn, and
462 the plaques (black arrow) were much smaller than those of the negative control. This
463 bacterial species was not retained for subsequent analyses. C. A lawn of co-cultured Ka
464 and Cupriavidus sp. [ratio 1:9]. Pigmented colonies corresponding to the Cupriavidus sp.
465 can be seen in the middle of the phagocytic plaques (upper black arrow). Pigmented
466 colonies can also seen around the plaques (lower black arrow). This isolate was
467 considered as an ARB.
468
25 469 Figure 3. Immunofluorescence of bacteria packaged by D. discoideum. Material from
470 the peripheries of phagocytic plaques on lawns of co-cultured bacteria (see ratio in Table
471 2) on SM1/10 agar spotted with D. discoideum were processed for IF and were observed
472 under an epifluorescence microscope. For each ARB tested, the differential interference
473 contrast (DIC) is shown on the left while DAPI (blue), which targets the DNA of bacteria
474 and amoebae, and the H36 antibody (red), which targets MLBs and the amoeba
475 membrane, staining are presented on the right. D. discoideum (black arrow in A)
476 produced and secreted a few packaged Cupriavidus sp. (magenta arrow) into the
477 extracellular milieu. The bacteria shown on the images (A and B) were coated and
478 recognized by the H36 antibody. In C and D, only a fraction of the M. luteus and R. tritici
479 cells were in H36-positive structures.
480
26 481 Figure 4. Transmission electron microscopy of bacteria packaged and secreted by D.
482 discoideum. The peripheries of phagocytic plaques from co-cultured bacteria (see ratio in
483 Table 2) on SM1/10 agar spotted with D. discoideum were processed and were observed
484 by TEM. A, D, and G. Bacteria grown alone on rich medium. B and C. D. discoideum
485 produces (white arrow) and secretes (black arrow) MLBs with digestible bacteria on
486 SM1/10. No Ka were seen inside the MLBs. E, F, and I. Cupriavidus sp. and R. tritici
487 were packaged by D. discoideum and were exocytosed into the extracellular milieu. H.
488 More than 50 undigested R. tritici can be seen inside a D. discoideum cell.
27
Table S1. Compilation of the results of the predation resistance assays. One hundred thirty- six soil and water isolates were plated on three types of medium: HL5 = rich medium, SM = nutrient medium, and SM1/10 = nutrient-poor medium. Serial dilutions of D. discoideum cells (500,000 to 5 cells/5 µL) were spotted and spread on bacterial lawns. The plates were incubated for 7 days at 21°C. The phagocytic plaques were counted to determine the resistance of each isolate to predation by D. discoideum. The potential ARBs (underlined in yellow) are located in the magenta spectrum. The predation-sensitive strains are located in the cyan spectrum. A brown box indicates that the bacterial isolate did not grow on that medium after two or more tries.
Legend: > 500,000 > 50,000 > 5,000 > 500 > 50 > 5 No resistance ARBs No growth
Isolates HL5 SM SM1/10 Aeromonas hydrophila
M15918-11 Alcaligenes faecalis Alphaproteobacterium Arthrobacter humicola Arthrobacter koreensis Arthrobacter tumbae Brevundimonas vesicularis Burkholderia ambifaria HSJ1 Burkholderia ambifaria variant Burkholderia thailendensis Cellulosimicrobium funkei Clavibacter michiganensis Comamonas koreensis Corynebacterium callunae Cupriavidus basilensis Cupriavidus necator-US1 Cupriavidus necator-US2 Cupriavidus sp. Curtobacterium pusillum Dietzia cinnamea Duganella zoogloeoides Endophytic bacterium Enhydrobacter aerosaccus Ensifer adhaerens P43 Erwinia tasmaniensis Escherichia coli B/R Escherichia coli BL21 Isolates HL5 SM SM1/10 Escherichia coli MC1061 Exiguobacterium indicum Flavobacterium sp. Frigoribacterium sp. Ev.-gws-
26 Gordonia alkanivorans Janibacter limosus Klebsiella aerogenes Kocuria kristinae Kocuria kristinae-US1 Kocuria kristinae-US2 Kocuria kristinae-US3 Kocuria kristinae-US5 Kocuria kristinae-US6 Kocuria kristinae-US7 Kocuria palustris Kocuria rosea Kocuria sp.8_1_14 Kocuria sp.32_3_20 Kocuria sp. D_1_23 Kocuria sp. 56_3_23_x1 Kocuria sp.56_2_16 Kocuria sp. 72_1_15 Kocuria sp. 48_5_11 Kocuria sp. 1_3_18B Luteibacter anthropi Microbacterium sp. Microbacterium esteraromaticum Microbacterium hatanonis Microbacterium lacus Microbacterium oleivorans Microbacterium oxydans Microbacterium oxydans-US1 Microbacterium phyllosphaerae Micrococcus luteus Micrococcus luteus (Norway) Micrococcus luteus 24_4_18 Micrococcus luteus 8_4_14_X2 Isolates HL5 SM SM1/10 Micrococcus luteus D_1_6_x2 Micrococcus luteus 25_5_4 Micrococcus luteus 1_1_24 Micrococcus luteus 8_4_15_x2 Micrococcus luteus D_3_15 Micrococcus luteus 61_5_26 Micrococcus luteus 37_4_14 Micrococcus luteus 8_5_8 Micrococcus luteus 48_3_19 Micrococcus luteus 48_5_10 Micrococcus luteus 4_3_25 Micrococcus luteus 8_5_6 Micrococcus luteus 4_4_11 Micrococcus luteus 4698 Micrococcus luteus-US1 Micrococcus luteus-US2 Micrococcus luteus-US3 Micrococcus luteus-US4 Micrococcus luteus-US6 Ochrobactrum intermedium Oerskovia paurometabola Paenibacillus larvae 3558 Paracoccus yeei Pectobacterium cypripedii Pedobacter agri Phyllobacterium sp. ORS 1420 Planococcus rifietoensis Plantibacter flavus Pseudomonas asplenii isolate 1 Pseudomonas asplenii isolate 2 Pseudomonas CT107 Pseudomonas fluorescence Pseudomonas fulva Pseudomonas koreensis Pseudomonas poea Pseudomonas psychrotolerans Pseudomonas putida Pseudomonas sp. Isolates HL5 SM SM1/10 Pseudomonas sp. LBUM-636 Pseudomonas sp. LBUM-677 Pseudomonas stutzeri Ralstonia sp. Rathayibacter tritici Rhodococcus erythropolis Rhodococcus erythropolis-US1 Rhodococcus erythropolis-US2 Rhodococcus fascians-US1 Rhodococcus fascians-US2 Rhodococcus pyridinivorans Rhodospirulum rubrum Roseomonas mucosa Rothia amarae Rothia nasimurium Serratia grimesii Serratia marcescens Sinorhizobium meliloti Sphingomonas paucimobilis Sphingomonas sanguinis Staphylococcus cohnii Staphylococcus aureus Staphylococcus epidermidis Staphylococcus equorum Staphylococcus haemolyticus Staphylococcus kloosii Staphylococcus lentus Staphylococcus saprophyticus Staphylococcus succinus Streptomyces luridiscabiei Variovorax paradoxus Wautersia eutropha Weissella confusa Yersinia ruckeri RS41