The depletion mechanism can actuate bacterial aggregation by self-produced exopolysaccharides and determine species distribution and composition in bacterial aggregates

Bacteria causing chronic infections are often found in cell aggregates suspended in polymer secretions, and aggregation may be a factor in infection persistence. One aggregation mechanism, called depletion aggregation, is driven by physical forces between bacteria and polymers. Here we investigated whether the depletion mechanism can actuate the aggregating effects of P. aeruginosa exopolysaccharides for suspended (i.e. not surface attached) bacteria, and how depletion affects bacterial inter-species interactions. We found cells overexpressing the exopolysaccharides Pel and Psl, but not alginate remained aggregated after depletion-mediating conditions were reversed. In co-culture, depletion aggregation had contrasting effects on P. aeruginosa’s interactions with coccus- and rod-shaped bacteria. Depletion caused S. aureus (cocci) and P. aeruginosa (rods) to segregate from each other, S. aureus to resist secreted P. aeruginosa antimicrobial factors, and the species to co-exist. In contrast, depletion aggregation caused P. aeruginosa and Burkholderia sp. to intermix, enhancing type VI secretion inhibition of Burkholderia by P. aeruginosa, leading to P. aeruginosa dominance. These results show that in addition to being a primary cause of aggregation in polymer-rich suspensions, physical forces inherent to the depletion mechanism can actuate the aggregating effects of self-produced exopolysaccharides and determine species distribution and composition of bacterial communities.


Introduction 30 31
At sites of chronic infection, bacteria are often found within cell aggregates suspended in 32 polymer-rich host secretions such as mucus, pus, sputum and others (1-3). Aggregated growth is 33 thought important because it can increase the ability of bacteria to survive environmental stresses 34 such as pH and osmotic extremes, as well as host-derived and pharmaceutical antimicrobials (4, 35 5). Bacterial aggregation also affects disease-relevant phenotypes such as bacterial invasiveness, 36 virulence factor production, and resistance to phagocytic uptake (6-10). 37 38 Bacteria can aggregate via bridging aggregation, which occurs when adhesions, polymers, or 39 other molecules bind cells to one another. Another general yet underappreciated mechanism is 40 depletion aggregation (11). Depletion aggregation occurs in environments containing high 41 concentrations of non-adsorbing polymers (12,13). Such conditions exist in the cytoplasm of 42 eukaryotic cells (11), cystic fibrosis airways (14), wounds (15), biofilm matrices (16), and others 43 settings. Depletion aggregation is initiated when bacteria spontaneously come into close contact 44 with each other (Fig 1A), causing the polymers in between cells to become restricted in their 45 configurational freedom, and thus decreasing their entropy. When polymers spontaneously move 46 out from in between bacterial cells (17) a polymer concentration gradient is established across 47 adjacent bacterial cells, producing an osmotic imbalance (i.e., the depletion force) that physically 48 holds the aggregate together (Fig 1B and C) (18). 49 50 While definitions and terminology can vary among investigators, biofilm formation and 51 depletion aggregation can be differentiated by two factors. First, biofilms are generally 52 considered a phenomenon of surface-attached bacteria (19-23), whereas depletion aggregation 53 operates on cells suspended in polymer solutions. Second, biofilm formation is driven by 54 bacterial activity (19,21,23) whereas depletion aggregation is a consequence of physical forces 55 generated when high concentrations of polymers are present. If bacteria and polymer 56 concentrations are high enough, aggregation via depletion will occur as default and obligatory 57 outcome unless mechanisms like mechanical disruption or bacterial motility produce stronger 58 counteracting forces. The dependency on environmental conditions also means that a reduction 59 Polymer-mediated depletion aggregation caused cocci shaped species (S. aureus) to segregate 131 from rods (P. aeruginosa and B. cenocepacia). In some cases, entire aggregates appeared 132 composed of single species. In other cases, sections of mixed-species aggregates were composed 133 primarily of either the rod or cocci-shaped species (Fig 3A and B). In contrast, depletion 134 aggregation caused bacteria with similar cell shapes (i.e. differentially labeled P. aeruginosa 135 with P. aeruginosa, or P. aeruginosa with E. coli) to intermix (Fig 3C and D). Similar results 136 were seen using mixtures of formalin-killed P. aeruginosa and S. aureus, and formalin-killed P. 137 aeruginosa and 2 µm diameter spherical beads similarly sized as S. aureus (Fig S2A and B). Similar to previous studies, (39-43) we found that wild-type P. aeruginosa severely inhibited S. 155 aureus in non-aggregated broth co-cultures (Fig 4A), and inhibition was diminished if P. 156 aeruginosa's main quorum sensing systems were genetically inactivated (i.e. ΔlasR/rhlR PAO1; 157 p<0.01) (Fig 4A, compare white bars). However, in co-cultures exposed to PEG 35 kDa to 158 induce depletion aggregation, wild-type P. aeruginosa killing of S. aureus was reduced by over 159 10-fold (Fig 4A, black bars). 160

161
Our previous finding that depletion aggregation caused marked antibiotic tolerance in P. 162 aeruginosa (14) led us to hypothesize that depletion-mediated tolerance explained P. 163 aeruginosa-S. aureus co-existence in aggregates. We tested this by exposing dispersed and 164 depletion-aggregated S. aureus to filter-sterilized culture P. aeruginosa supernatant found that 165 dispersed S. aureus were ~10-fold more sensitive to killing after 18 hrs (Fig 4B). Control 166 experiments indicate that PEG did not diminish the antimicrobial activity of P. aeruginosa 167 supernatants (see Fig S3), and that the inhibitory effects were mediated by quorum-controlled 168 factors ( Fig 4C). These results indicate that depletion aggregation can promote co-existence of To test this, we mixed P. aeruginosa which capable of T6SS with Burkholderia thailandensis, a 184 TSS-susceptible rod-shaped Gram-negative bacterium (47). In dispersed conditions, no P. 185 aeruginosa-B. thailandensis antagonism was apparent over 24 hours, as the ratio P. aeruginosa 186 to B. thailandensis remained unchanged (Fig 5A). In contrast, P. aeruginosa outcompeted B. 187 thailandensis in depletion aggregates as measured by viable counts (Fig 5A) and visually 188 assessing differentially-labeled species (Fig 5B). Notably the competitive advantage of P. 189 aeruginosa was eliminated by genetically inactivating TSS (i.e. PAO1 ΔclpV1 (Fig 5C)). Taken  Surface attachment is thought to be fundamental to biofilm formation; sensing and adhering to 209 surfaces induces physiological responses important in biofilm growth, and attachment keeps 210 nascent biofilm-forming cells from dispersing (from random movement or fluid flows) before the 211 matrix binds them together (23). Our previous work and current experiments raise the possibility 212 that the depletion mechanism might serve somewhat similar functions as attachment surfaces. 213 Previously we found that like surface attachment (57), depletion aggregation can induce stress 214 responses in P. aeruginosa that mediate antibiotic tolerance (14). Our current experiments show 215 that depletion aggregation also brings suspended cells together and can promote adhesion by 216 self-produced polymers. One important caveat is that in the conditions used here, 217 exopolysaccharide overexpression was required as P. aeruginosa PAO1 capable of producing 218 "wild-type" levels of polysaccharides did not exhibit matrix-mediated adhesion even after long 219 periods of depletion aggregation. Notably, mutant strains constitutively expressing EPS can be 220 isolated from infected CF subjects (58) The effect of depletion aggregation to intermix species with similar shapes, and segregate species 234 dissimilar shapes could have wide ranging effects. One consequence we demonstrated is 235 enhanced efficacy of TSS-mediated inhibition of rod shaped Burkholderia sp. by rod-shaped P. 236 aeruginosa, as TSS is dependent upon species intermixing and cell-to-cell contact (Fig 6B). Our study had several limitations. For example, we used a non-biological polymer (PEG) at a 250 specific concentration (30% w/vol) with a defined molecular weight (PEG 35 kDa) to induce 251 depletion aggregation as use of a defined polymer limited variability and the transparency of 252 PEG enhanced microscopy. While it is possible that biological polymers could produce different 253 results, our previous work shows that depletion aggregation by DNA and mucin at 254 concentrations found at infection sites cause similar aggregate morphology and antibiotic 255 tolerance phenotypes as PEG (14). We also recognize that varying polymer size and molecular 256 weight will affect the strength of the aggregating force, and these variables were not examined 257 here. An additional limitation was that our experiments used laboratory strains and a handful of

Acknowledgments 276
We are grateful to Joseph Mougous for sharing the clpV1 mutant and Burkholderia strains.

PEG-induced depletion aggregation of bacteria 288
For PEG-induced depletion aggregation, bacteria were added at the indicated densities to either 289 LB diluted 4:6 with distilled water or LB diluted with 50% PEG 35 kDa (w/vol) prepared in 290 distilled water to ensure that nutrient concentrations were the same in dispersed and aggregated 291 conditions. LB was diluted with water or 50% w/vol PEG 35 kDa for all experiments described 292 unless noted otherwise. Cultures were then incubated on a roller (60 rpm) at 37°C unless 293 indicated otherwise. 294 295 Aggregate reversibility assays 296 The indicated bacterial strains were grown overnight in full-strength LB. One hundred µl of 297 overnight cultures were used to inoculate 3 ml of LB+PEG 35 kDa. After 18-h of growth, 100 µl 298 of the indicated cultures were removed to a 1.5 ml tube containing 900 µl of either 1x PBS or 299 PBS supplemented with 30% w/vol PEG 35 kDa and vortexed. Imaging was performed on 50 µl 300 culture aliquots pre-and post-dilution using a Leica DM1000 LED microscope by spotting onto 301 a glass slide. Aggregate dispersal was scored by eye by comparing to undiluted control cultures. 302 303 Bacterial competition assays 304 S. aureus SH1000 (73) and P. aeruginosa PAO1 (64) were grown overnight at 37°C with 305 shaking in LB broth. S. aureus and P. aeruginosa were pelleted and resuspended at 10 8 CFU/ml 306 in fresh LB broth. One hundred µl of each culture was added to 2 ml LB supplemented with 307 either 30% w/vol PEG (35 kDa or 2 kDa) where indicated. Bacteria were grown in co-culture for 308 18 h and viable bacteria were enumerated by serial dilution and plating on LB plates. Colony 309 morphology was used to differentiate P. aeruginosa from S. aureus. 310

311
For experiments investigating the effects of quorum-regulated antimicrobials on S. aureus 312 killing, P. aeruginosa PAO1 or ΔlasR/rhlR (68) were grown overnight at 37°C with shaking in 313 50 ml LB broth in a 250 ml flask. Bacteria were removed by centrifugation (10 minutes, 9,000 x 314 g) and supernatants were filter sterilized using bottle top vacuum filters with 0.2 µm pore size 315 (Millipore). PEG 2 kDa or 35 kDa was added to these supernatants to a final concentration of 316 30% w/vol where indicated. S. aureus was inoculated into P. aeruginosa supernatants at 10 8 317 CFU/ml and cultured for 6 h at 37°C on a roller at 60 rpm. Viable S. aureus were enumerated by 318 serial dilution and plating onto LB agar plates. 319 320 To investigate TSS mediated killing, P. aeruginosa PAO1, ΔclpV1 (46), and B. thailandensis 321 E264 (71) were grown overnight at 37°C with shaking in LB broth. Bacteria were resuspended in 322 fresh LB at 10 9 CFU/ml. One hundred µl containing 1x10 8 CFU P. aeruginosa PAO1 or ΔclpV1 323 and 100 µl containing 2.0x10 7 CFU B. thailandensis were added to 800 µl LB or the indicated 324 polymer solutions and incubated in co-culture for 24 h at 37°C on a roller at 60 rpm. Viable 325 bacteria were enumerated by serial dilution and plating on LB plates. Colony morphology was 326 used to differentiate P. aeruginosa from B. thailandensis. For fluorescent imaging of aggregates, 327 strains PAO1 or ΔclpV1 constitutively expressing GFP (PAO1 attTn7::GFP, (69)) were co-328 cultured with B. thailandensis E264 attTn7::mCherry for 24 hours (47). Image analysis is 329 described below. TFP in PBS at a concentration of 10 9 CFU/ml. Formaldehyde (16%, Thermo) was added slowly 338 to bacteria while vortexing to a final concentration of 4% vol/vol. Bacteria were allowed to fix 339 for 30 minutes with constant mixing to prevent bacteria from clumping. Cells were then 340 centrifuged for 10 minutes at 9,000 x g, washed twice with PBS, and resuspended in 1 ml PBS. 341 Complete bacterial killing was confirmed by plating fixed bacteria on LB agar. One hundred µl 342 of the indicated fixed strains were added to 2 ml PBS or PBS+30% PEG 35 kDa. Bacteria were 343 incubated in a 37°C in a roller at 60 rpm and visualized at the indicated times using a Zeiss LSM 344 510 confocal laser-scanning microscope. Image series were processed using Volocity 345 (Improvision). 346

17.
Schwarz  Depletion aggregates were then diluted 10X with PBS and representative images were acquired 595 immediately pre-and immediately post-dilution. See also Figure S1 and Movie S1. Scale bar 40 596 µm. 597 598 Figure S1. Depletion aggregation was induced with 30% w/vol PEG 35 kDa for 18 hours. P. 599 aeruginosa PAO1 depletion aggregates were then diluted 10X with additional PEG 35 kDa. Depletion aggregation was induced with PEG 35 kDa using combinations of dead formalin-fixed 610 cocci and rods. Fluorescent microscopy was used to image aggregates after 18-h of growth. Note 611