(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Killer prey: Ecology reverses bacterial predation [1]

['Marie Vasse', 'Mivegec', 'Umr Cnrs', 'Ird', 'Um', 'Cnrs Montpellier', 'Francesca Fiegna', 'Institute For Integrative Biology', 'Eth Zürich', 'Zürich']

Date: 2024-02

Ecological variation influences the character of many biotic interactions, but examples of predator–prey reversal mediated by abiotic context are few. We show that the temperature at which prey grow before interacting with a bacterial predator can determine the very direction of predation, reversing predator and prey identities. While Pseudomonas fluorescens reared at 32°C was extensively killed by the generalist predator Myxococcus xanthus, P. fluorescens reared at 22°C became the predator, slaughtering M. xanthus to extinction and growing on its remains. Beyond M. xanthus, diffusible molecules in P. fluorescens supernatant also killed 2 other phylogenetically distant species among several examined. Our results suggest that the sign of lethal microbial antagonisms may often change across abiotic gradients in natural microbial communities, with important ecological and evolutionary implications. They also suggest that a larger proportion of microbial warfare results in predation—the killing and consumption of organisms—than is generally recognized.

Copyright: © 2024 Vasse et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Motivated by observations suggesting that M. xanthus predation could differ greatly on prey reared on a room temperature bench versus in a warmed incubator, we investigated how variation in the temperature at which bacterial prey grow prior to predator attack might alter the predation risk of prey. Using a phylogenetically diverse panel of bacteria previously demonstrated to fuel predatory growth by M. xanthus [ 18 , 39 ], we initially quantified the ability of M. xanthus to swarm through prey lawns reared at 3 temperatures (12, 22, or 32°C). Upon observing that Pseudomonas fluorescens eliminated M. xanthus swarming when the former had been previously reared at 22°C but not at 32°C, we quantified M. xanthus viable population size after interaction with P. fluorescens pregrown at the same 3 temperatures from several initial densities. Interaction with P. fluorescens grown at 22°C resulted in M. xanthus extinction. We then tested whether the killing factors produced by this sometime-prey species are cell-bound or diffusible secretions and whether they are proteinaceous or not (or require polypeptides to function) by testing for functional sensitivity to 95°C heat exposure. We further asked whether P. fluorescens functions as a predator after killing M. xanthus by testing whether nutrients from killed M. xanthus fuel P. fluorescens growth. Finally, we tested whether P. fluorescens secretions lethal to M. xanthus also kill a diverse panel of other potential prey species.

Heterogeneity in many abiotic factors such as temperature, prey nutrition, pH, oxygen availability, salinity, and surface or fluid viscosity is known or likely to influence microbial predator–prey interactions [ 21 – 26 ], and microbial antagonisms more broadly [ 27 – 30 ]. For example, predators and prey may be differentially sensitive to pH gradients [ 25 , 31 ] and temperature influences many bacterial traits relevant to predation, including cell division rates and motility behavior [ 32 , 33 ]. Temperature also impacts secondary metabolite production [ 34 , 35 ] and Type VI secretion [ 36 , 37 ], traits associated with a broad range of microbial antagonisms often not described as predatory, and can modulate the intensity of such antagonisms [ 38 ].

Predation pervades the microbial world [ 9 ] as well as animal communities, impacting microbial community composition and network structure and thereby influencing nutrient-cycling dynamics and important features of macroorganism biology. Behavioral modes of microbial predation vary greatly. Many eukaryotic predators of bacteria such as nematodes and many protists ingest whole prey cells [ 10 , 11 ], while predatory bacteria are too small to do so. In a virus-like life cycle, cells of the specialist bacterial predator Bdellovibrio bacteriovorus physically attach to a prey cell before invading it and reproducing inside [ 12 ]. Other bacteria kill prey without attachment using diffusible secretions [ 13 , 14 ]. In addition to using diffusible killing agents, the generalist bacterial predator Myxococcus xanthus employs highly effective predatory weapons that depend on contact with prey cells [ 15 , 16 ]. Perhaps best known for its formation of fruiting bodies in response to starvation [ 17 ], M. xanthus can prey on a very broad range of other microbes, including both gram-negative and gram-positive bacteria [ 18 ] and some fungi [ 19 ]. M. xanthus uses 2 synergistic motility systems to forage for prey across variable solid surfaces [ 20 ].

Ecological effects on predator–prey interactions can also be delayed. For example, maternal exposure to predation risk at one time can influence the degree of antipredator behavior displayed by offspring at a later time [ 8 ]. Investigating how predator–prey interactions change as a function of ecological context, whether contemporaneous or historical, is necessary for understanding how those interactions evolve in spatially and temporally heterogeneous habitats.

Ecological context strongly shapes the intensity and character of many biotic interactions [ 1 , 2 ], including between predators and prey [ 3 ]. Some ecological factors influence predator–prey interactions contemporaneously. For example, water temperature immediately influences the timing and duration of predator attacks in some fish [ 4 ], background versus prey coloration often determines prey detectability [ 5 ], and snow depth modulates the efficiency of wolf predation [ 6 ]. Most such examples illustrate how ecological context modulates the effectiveness of predation in a unidirectional predator–prey relationship. However, at least 1 study has found that abiotic ecology can even reverse the predominant direction of mutual predation; specifically, the majority direction of predation events between 2 amphipod species has been shown to reverse as a function of salinity [ 7 ].

Materials and methods

Strains M. xanthus strains GJV1, DK3470, A75, and Serengeti 01 were used in this study. GJV1 is a closely related derivative of the reference strain DK1622 [40], DK3470 is a mutant of DK1622 with a mutation in the dsp gene [41], and A75 [42] and Serengeti 01 [43] are natural isolates from Tübingen, Germany, and Serengeti National Park, Tanzania, respectively. Strains of Arthrobacter globiformis, Bacillus bataviensis, Curtobacterium citreum, Escherichia coli, Micrococcus luteus, and Rhizobium vitis used here are the same as those reported by Morgan and colleagues in 2010 [18]. Pseudomonas fluorescens strain SBW25 is from [44].

General media and culture conditions Unless specified otherwise, M. xanthus strains were inoculated from frozen stock onto CTT 1.5% agar (1% casitone; 8 mM MgSO 4 ; 10 mM Tris–HCl (pH 7.6); 1 mM KH 2 PO 4 , 1.5% agar [45]) plates 3 days prior to transfer of culture samples into 8 ml CTT liquid (identical to CTT agar except lacking agar) in 50-ml flasks. Liquid cultures were typically grown over 2 days with a dilution transfer after 1 day. All assays were initiated from log-phase cultures. For other species, samples from frozen stocks were inoculated directly into LB liquid. Unless specified otherwise, liquid cultures were incubated at 32°C with shaking at 300 rpm and plate cultures were incubated at 32°C, 90% relative humidity. Cell densities of bacterial populations were estimated with a TECAN Genios plate reader. Prior to resuspension to initiate assays, cultures were centrifuged at 4,472 × g, 15 minutes. Unless specified otherwise, all bacterial cultures were resuspended in M9 medium (1×M9 salts: 22 mM KH 2 PO 4 , 18.7 mM NH 4 Cl, 8.6 mM NaCl supplemented with 2 mM MgSO 4 and 0.1 mM CaCl 2 ), and experiments were run on M9cas agar (M9 medium supplemented with 0.3% casitone and 1.2% agar).

Swarming rates Around 25-ml aliquots of M9cas agar were poured into 9-cm petri plates and allowed to cool and solidify in a laminar flow hood without lids for 20 minutes, after which they were capped and stored overnight at room temperature. Centrifuged prey cultures were resuspended to a predicted density of approximately 5 × 106 cells/ml in M9 medium. From each resuspended culture (and one control containing only M9), 600-μl aliquots were placed on M9cas agar plates and distributed evenly with a sterile metal triangle. Plates were then left open without lids in a laminar flow hood for 60 minutes. Four plates for each prey species were then incubated at each of 3 temperatures (12, 22, and 32°C) for 22 hours. Incubator windows were covered to prevent light penetration. After incubation, the plates were kept at room temperature for 2 hours prior to addition of M. xanthus. Centrifuged cultures of M. xanthus strains GJV1, A75 and Serengeti 01 were resuspended to a predicted density of approximately 1010 cells/ml with M9 medium. For each temperature–prey combination (and for control plates), 20 μl of the resuspension were spotted in the middle of the plate and plates were then left open without lids in a laminar flow hood for 30 minutes before being incubated at 32°C and 90% relative humidity. M. xanthus swarm diameters were measured after 7 days of incubation (2 perpendicular diameters per swarm at random orientation). M. xanthus swarms of these strains are bright yellow in color. When no evidence of M. xanthus growth was observed (i.e., no yellow area) after 7 days even within the originally inoculated plate area, a diameter value of 0 was recorded. (This occurred only on some plates with P. fluorescens.)

Test for M. xanthus killing of P. fluorescens reared at 32°C For our test of whether M. xanthus kills P. fluorescens reared at 32°C on M9cas agar, media and culture-handling protocols were the same as in the swarming assays described above except in the following respects. The killing test was performed on M9cas agar in 50-ml glass flasks rather than in petri dishes to allow shaking with resuspension buffer (see details below). A 10-μl aliquot of resuspended P. fluorescens culture was inoculated onto M9cas agar the day before addition of M. xanthus. The aliquot was not spread across the plate (as the 600-μl aliquots in the swarming assays were) but was rather allowed to grow into a small circular lawn within the originally inoculated spot area. After incubation for 24 hours at 32°C, a 50-μl aliquot of M. xanthus resuspension (approximately 1010 cells/ml, as in the swarming assays) was inoculated across the top of and immediately surrounding the circular P. fluorescens lawn that had grown up overnight. For the control treatment to which M. xanthus was not added, a 50-μl aliquot of M9 liquid was added in the same manner. The flasks were then incubated for 4 days before P. fluorescens was harvested and dilution plated into LB 0.5% soft agar, in which M. xanthus does not grow. P. fluorescens was harvested by adding 1-ml M9 liquid to each flask, scraping the bacteria into suspension with a loop and mixing the suspension by repeated pipetting to disperse P. fluorescens cells. To suspend any cells remaining on the agar after the above procedure, 2 ml of additional M9 liquid and 10 glass beads were then added to each flask and flasks were shaken for 1 hour at 300 rpm, 32°C prior to dilution plating.

Test for effects of P. fluorescens rearing temperature and inoculum population size on M. xanthus DK3470 survival One day prior to inoculation, 10-ml aliquots of M9cas agar were poured into 50-ml glass flasks and allowed to solidify without flask covers for 30 minutes in a laminar flow hood. Flasks were then capped and kept at room temperature overnight. Centrifuged cultures of P. fluorescens were resuspended to predicted densities of approximately 5 × 105, approximately 5 × 106, and approximately 5 × 107 cells/ml with M9 liquid. A 200-μl aliquot of resuspended culture was inoculated into each agar flask and then spread across the agar surface by gentle rotation. Flasks were then kept open without covers for 60 minutes in a laminar flow hood, after which they were capped and flasks of each inoculum population size treatment were incubated at 3 temperatures (12, 22, and 32°C) for 22 hours. After incubation at different temperatures, all flasks were kept at room temperature for 2 hours prior to either addition of M. xanthus or assessment of P. fluorescens population size. For assays of M. xanthus strain DK3470 population size, 20-μl aliquots of DK3470 culture resuspended to approximately 1010 cells/ml (approximately 2 × 108 cells) were spotted in the middle of each flask; flasks were then left open for 30 minutes in a laminar flow hood. Flasks were then harvested at one of 2 time points: either immediately or after 7 days of incubation at 32°C. To harvest and disperse DK3470, 10 glass beads and 1-ml M9 liquid were added to each flask and flasks were shaken at 300 rpm, 32°C for 15 minutes before the resulting suspensions were dilution plated into CTT 0.5% soft agar containing gentamicin (10 μg/ml), which prevents growth of P. fluorescens but not M. xanthus. Colonies were counted after 7 days of incubation at 32°C, 90% relative humidity.

Test for effect of P. fluorescens inoculum population size on postgrowth population size For assays of P. fluorescens population size after overnight growth at one of 3 temperatures, assays that were performed separately from the above-described assays with DK3470, 10 glass beads and 1 ml of M9 liquid were added and the flasks shaken at 300 rpm, 32°C for 15 minutes to disperse P. fluorescens populations. Samples were then dilution plated into LB 0.5% soft agar, and P. fluorescens colonies were counted after 2 days of incubation at 32°C, 90% relative humidity.

Test for diffusion of the killing compound(s) produced by P. fluorescens DK3470 was inoculated onto CTT hard agar from frozen stock 4 days prior to inoculation in 8-ml CTT liquid in a 50-ml Erlenmeyer flask. The resulting liquid culture was grown for approximately 24 hours at 32°C, 300 rpm, diluted into fresh medium and grown for another 24 hours before being centrifuged and resuspended to approximately 5 × 109 cells/ml in either supernatant from buffer suspensions of P. fluorescens (prepared as described below) or control buffer. The resuspended cultures were incubated for 6 hours at 32°C and then dilution plated into CTT 0.5% soft agar. Plates were incubated for 3 to 5 days before colonies were counted. To prepare P. fluorescens supernatants, P. fluorescens was inoculated from frozen stock into 8-ml LB liquid and then incubated overnight at 32°C, 300 rpm. After dilution to approximately 5 × 107 cells/ml, 200-μl aliquots of the resulting culture were spread across the surfaces of 10-ml aliquots of M9cas agar that had been poured into 50-ml flasks 24 hours before. P. fluorescens populations and control flasks containing only M9cas agar were then incubated at 12, 22, and 32°C. After approximately 24 hours, 0.8-ml M9 liquid and approximately 10 sterile glass beads were added to flasks, which were then shaken at 300 rpm, 32°C for 15 minutes before 800 μl of each of the resulting culture suspensions were removed from the flask and centrifuged (5,000 rpm, 15 minutes). After centrifugation, supernatant was filter sterilized with 0.2-μm filters. Each sample of filtered supernatant was separated into 2 equal subsamples, one of which was heated at 95°C for 45 minutes while the other was kept at room temperature.

Test for effects of P. fluorescens on M. xanthus when pregrowth of both species and their interaction all occurred at either 22 or 32°C One day prior to inoculation, 10-ml aliquots of M9cas agar were poured into 50-ml flasks and allowed to solidify for 60 minutes in a laminar flow hood, after which flasks were capped and kept at room temperature overnight. Centrifuged cultures of P. fluorescens, previously grown in LB liquid at 22°C or 32°C, were resuspended to predicted densities of approximately 5 × 107 cells/ml with M9 liquid. A 200-μl aliquot of resuspended culture was spread across the agar surface in each flask by gentle rotation. As a control treatment, 200 μl M9 liquid was spread instead of P. fluorescens culture. Flasks were then kept open without covers for 60 minutes in a laminar flow hood, after which they were closed and each was incubated for approximately 28 hours at the same temperature as its source liquid culture had been grown. DK3470 previously grown on CTT agar was grown in CTT liquid at either 22 or 32°C for approximately 2 days before 20-μl aliquots resuspended to approximately 1 × 1010 cells/ml were placed on agar cultures of P. fluorescens that had been incubated at the same temperature as the corresponding M. xanthus culture. Flasks were then left open for 30 minutes in a laminar flow hood before being incubated for 24 hours at the same temperature as both species in the flask had been incubated prior to interaction. Cultures were then resupended and dilution plated into CTT 0.5% soft agar with gentamicin (10 μg/ml) to count DK3470 population sizes.

Tests for growth of P. fluorescens on nutrients from M. xanthus killed by P. fluorescens and for effects of P. fluorescens rearing temperature on M. xanthus DK3470 and P. fluorescens population sizes after interaction One day prior to inoculation, 0.5-ml aliquots of M9cas agar were placed into 48-well plate wells and allowed to solidify for 60 minutes in a laminar flow hood at room temperature, after which the plates were covered and kept at room temperature overnight. Centrifuged cultures of P. fluorescens were resuspended to predicted densities of approximately 5 × 105 cells/ml with M9 liquid. A 10-μl aliquot of resuspended culture was inoculated into each well with agar and then spread across the agar surface by gentle rotation. Plates were then kept open for 60 minutes in a laminar flow hood, after which they were closed and incubated at 22 or 32°C for approximately 24 hours. To estimate P. fluorescens population size after the 24 hours of growth but prior to addition of M. xanthus, lawns grown at each temperature were harvested by pipetting-resuspension with 0.5-ml M9 liquid, and then samples were dilution plated into LB 0.5% soft agar. To initiate interaction between M. xanthus and P. fluorescens after the 24-hour period of P. fluorescens lawn growth, overnight M. xanthus DK3470 cultures were centrifuged and resuspended in M9 liquid to a density of approximately 2 × 1010 cells/ml. One sample of each resuspension was dilution plated into CTT 0.5% soft agar containing gentamicin (10 μg/ml) to estimate viable population size. Twenty μl aliquots of resuspended M. xanthus cultures or of M9 liquid as controls were spotted in each well on top of the P. fluorescens lawn; plates were then left open for 30 minutes in a laminar flow hood before being incubated at 32°C, 90% relative humidity for another 24 hours. Wells were harvested after 24 hours by adding 0.5 ml of M9 liquid and resuspended by pipetting; the resulting suspensions were dilution plated into LB 0.5% soft agar (which allows growth of only P. fluorescens) and CTT 0.5% soft agar containing gentamicin (10 μg/ml, which allows growth of only M. xanthus). Colonies were counted after 3 or 7 days at 32°C, 90% relative humidity for P. fluorescens and M. xanthus, respectively.

Test for growth of P. fluorescens on nutrients <0.2 μm in size derived from M. xanthus killed by P. fluorescens To test if P. fluorescens could grow on readily diffusible nutrients released by M. xanthus incubated in P. fluorescens supernatant, P. fluorescens supernatant was harvested following the same protocol described above, except flasks with P. fluorescens lawns were incubated overnight only at 22°C. M. xanthus DK3470 cells from exponential phase cultures were spun down and resuspended to a density of approximately 5 × 109 cells/ml in the P. fluorescens supernatant. These resuspensions with M. xanthus, as well as supernatant without M. xanthus, were then incubated for 6 hours, after which (i) 10-μl samples from the treatment with DK3470 were plated onto CTT 1.5% hard agar to test for M. xanthus growth after subsequent incubation and (ii) 200 μl of M9 liquid culture of P. fluorescens at densities of approximately 106 and approxiamtely 108 cells/ml were added to both supernatant treatments (with and without DK3470 resuspended cells). The resulting P. fluorescens cultures were then dilution plated onto LB 1.5% soft agar immediately and after 24, 48, and 96 hours to determine population size.

Test for P. fluorescens supernatant effects on other bacterial species Erlenmeyer flasks with M9cas agar were prepared as in previous experiments. Centrifuged cultures of P. fluorescens were resuspended to a predicted density of 5 × 107 cells/ml with M9 liquid, and 200 μl of resuspended samples (or of M9 liquid for the controls) were inoculated into each agar flask and then spread over the agar surface by gentle rotation. Flasks were subsequently kept open for 60 minutes in a laminar flow hood before being closed and incubated at 22°C for 22 hours. To harvest P. fluorescens, 10 glass beads and 0.8 ml of M9 liquid were added to flasks, which were then shaken at 300 rpm, 32°C for 15 minutes. Suspensions were then centrifuged to harvest supernatants, which were then sterilized with a 0.2-μm filter. Centrifuged cultures of A. globiformis, B. bataviensis, C. citreum, E. coli, M. luteus, M. xanthus strain DK3470, P. fluorescens, and R. vitis were resuspended to a predicted density of 5 × 109 cells/ml with the filtered supernatant, or M9 liquid in the controls. The resulting cell suspensions were incubated at 32°C, 90% relative humidity for 6 hours before they were dilution plated into CTT (M. xanthus) or LB (other species) 0.5% soft agar. Colonies were counted after 2 to 4 days of incubation.

pH of P. fluorescens supernatant Supernatants from P. fluorescens cultures on agar medium incubated at 22°C and 32°C were harvested as previously described. pH values of the supernatants and of M9 medium were measured with a Mettler Toledo pH meter for 3 independent replicate cultures for each temperature treatment.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002454

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.