Ability of Vibrio strains to attack and to lyse Alexandrium pacificum.

ND: not determined.

Ability of Vibrio Atlanticus LGP32 to degrade flagella, attack and lyse the targeted dinoflagellates spp. commonly found in the Mediterranean Sea.

ND not determined.

Oligonucleotide sequences of primers used for RNA expression analysis.

Dynamics of Alexandrium and Vibrio in the environment.

(A) Location of the monitoring station in the Thau Lagoon (southern France). (B) Mean abundance (DNA equiv.) of Vibrio spp. (16S) and Alexandrium spp. (A. pacificum ACT03 + A. tamarense ATT07). Vibrio cells (black line with diamond dot) and degraded Alexandrium cells (grey line with round dot) were evidence in the 0.2–0.8 μm fraction (free Vibrio fraction) in spring and autumn 2015. Living Alexandrium cells (grey line with roundot) but no plankton-associated Vibrio spp. (black line with diamond dot) were evidence in the 0.8–180 μm in spring and autumn. (C) Result of Akaike information criterion (AICc) models tested to explain the mean value of degraded Alexandrium cells (dead cells) in spring. (D) Wald test of the AICc model attributing the mean value of degraded cells of Alexandrium in spring to free Vibrio.

Incubation of Vibrio atlanticus LGP32 and Alexandrium pacificum ACT03 in enriched natural seawater (ENSW).

(A) A. pacificum ACT03 cultured alone (grey bar) and incubated with V. atlanticus LGP32 (black bar) in ENSW. (B) V. atlanticus LGP32 cultured alone (grey bar) and incubated with A. pacificum ACT03 (black bar) in ENSW.(C) Snapshot of the interaction between V. atlanticus LGP32-GFP cells (60-hour culture) and one cell of A. pacificum ACT03 taken at 8h00 of co-culture. (D) Chronological snapshots of the interaction (98 pictures, one per second). V. atlanticus LGP32 (small green cells) and A. pacificum ACT03 cell (large red cell). All experiments were done in triplicates. Asterisks indicate significant differences in a multiple comparison test (One-way ANOVA with post hoc Tukey test), *P ≤ 0.05.

Role of Vibrio atlanticus LGP32 starvation in the interspecific interaction process.

Experiments were conducted by incubatingA. pacificum ACT03 with V. atlanticus LGP32 previously grown for 12, 36, 60 and 126 h in Zobell medium. (A) Cumulative percentage of motile A. pacificum ACT03 cells. (B) Cumulative number of cells attacked by V. atlanticus LGP32 and (C) Cumulative cell lysis after 0, 15, 30, 45 and 60 minutes of interaction. Corresponding pictures showing (A1) Black arrows indicate unhooked and degrade flagellum from A. pacificum ACT03 flagellum, (B1) Chronological sequence of five snapshots showing V. atlanticus LGP32-GFP cells (60-hour culture) and A. pacificum ACT03 cells, during the first hour of their interaction. V. atlanticus LGP32 (small green cells), living A. pacificum ACT03 (large red cells) and dead A. pacificum ACT03 (large green cell). (C1) Black arrow 1 indicate vesicle formation on A. pacificum ACT03 cell and black arrow 2 indicate lysed A. pacificum ACT03 cell. All percentages were determined based on a minimum of 2,000 cells of A. pacificum ACT03. All experiments were done in triplicates. Asterisks indicate significant differences in a multiple comparison test (One-way ANOVA with post hoc Tukey test), ***P ≤ 0.001

Role of quorum sensing and the vibrioferrin iron uptake pathway in the interaction process.

(A) Effect of V. atlanticus LGP32 cell density on the attack process. A. pacificum ACT03 cells (1.103 cells mL-1) were incubated with V. atlanticus LGP32 grown for 60 hours in Zobell medium at concentrations ranging from 5.103 to 5.105 cells mL−1 (black bars). For comparison, A. pacificum ACT03 incubated with V. atlanticus LGP32 grown for 12 hours in Zobell medium at concentrations ranging from 5.103 to 4.106 cells mL−1 (grey bars). The image on the bars indicates either unaffected algae (live red algae) or algae attacked by vibrio’s (algae covered with green vibrio’s) during the interaction (B) CqsS, luxM, luxN, luxS, and luxP quorum sensing and PvsA, PvuB and PvuA2 vibrioferrin pathway genes expression in V. atlanticus LGP32 grown for 12, 36 and 60 h in Zobell medium. (C) Effect of V. atlanticus LGP32 mutants on the attacked process. Experiments were conducted by incubating A. pacificum ACT03 with V. atlanticus LGP32, V. atlanticus LGP32 tagged with GFP, V. atlanticus LGP32 washed with ENSW or V. atlanticus LGP32 mutant ΔPvuB, ΔluxM, ΔluxR and ΔluxS previously grown 60 h in Zobell media (control), The percentage of A. pacificum ACT03 attacked was determined during the first 30 min of exposure. (D) Effect of V. atlanticus LGP32 cultures media composition on the attacked process. Experiments were conducted by incubating A. pacificum ACT03 with V. atlanticus LGP32 grown 60 h in Zobell media supplemented with H3BO4 or FeCl3. The results were compared with an exposure to V. atlanticus LGP32 grown 60 h in Zobell media. All percentages were determined based on a minimum of 2,000 cells of A. pacificum ACT03. All experiments were done in triplicates. Asterisks indicate significant differences in a multiple comparison test (One-way ANOVA with post hoc Tukey test), **P ≤ 0.01, ***P ≤ 0.001

Schematic representation of a putative strategy developed by Vibrio spp. to feed on Alexandrium spp. and G. catenatum in the environment.

(A) Vibrio in the environment when subjected to starvation secrete non-protein lytic compounds. (B) Some of these lytic compounds degrade the flagella, immobilizing the alga (immobilization stage). (C) Then Vibrio swims and clusters around its prey (attack stage). (D) Lytic compounds released by Vibrio where able to concentrate around the algae cells, thereby lysing the algae (killing stage). (E) Feeding on the released nutrients, Vibrio multiply and then spread in the environment. Yellow clouds: Lytic compound release by Vibrio, Grey clouds: Algal nutrients released upon lysis.

Time and dose-dependent effects of the V. atlanticus LGP32 culture supernatant on A. pacificum ACT03 motility.

(A) A time dependence experiment was conducted by incubating A. pacificum ACT03 for 1, 5 or 20 h with 1/1000 v/v (1 µL/mL) of culture supernatant from V. atlanticus LGP32 previously grown for 60 h in Zobell culture media. (B) A dose dependence experiment was conducted by incubating A. pacificum ACT03 for 1 h with 1/1000 to 1/20 v/v (1-50 µL/mL) of culture supernatant from V. atlanticus previously grown for 60 h in Zobell media. The percentage of motile A. pacificum ACT03 was determined after 1 hours of exposure. All percentages were determined based on a minimum of 2,000 cells of A. pacificum ACT03. Error bars represent the standard deviation of the mean of three independent experiments. Asterisks indicate significant differences in a multiple comparison test (One-way ANOVA with post hoc Tukey test), *P ≤ 0.05, ***P ≤ 0.001.

Vibrio atlanticus LGP32 proteome analysis following nutrient stress.

(A) Example of 2D gel, the numbers in white on the gel 4-7 correspond to the number and position of the protein spots analyzed. (B) Proteins identified by LC-MS/MS as differentially represented in the 2D gel comparative approach following nutrient stress. ND: Not determined; ENSW (artificial seawater).

Quorum sensing and the vibrioferrin iron uptake pathway in Vibrio.

(A) Putative quorum sensing (QS) pathways at low and high cell density in Vibrio according to Lami et al.(Lami, 2019). (B) Genetic organization of the vibrioferrin utilization gene cluster on V. atlanticus LGP32 chromosome 2. The Pvu and Pvs operons are involved in the secretion and the transport of ferric vibrioferrin and biosynthesis of vibrioferrin, respectively. Arrows indicate the transcriptional directions of the genes. VSII1126, VSII1137 and VSII1129 corresponding to PvuB, PvuA2 and PvsA genes respectively.