Ecological significance of extracellular vesicles in modulating host-virus interactions during algal blooms

Abstract

Extracellular vesicles are produced by organisms from all kingdoms and serve a myriad of functions, many of which involve cell-cell signaling, especially during stress conditions and host-pathogen interactions. In the marine environment, communication between microorganisms can shape trophic level interactions and population succession, yet we know very little about the involvement of vesicles in these processes. In a previous study, we showed that vesicles produced during viral infection by the ecologically important model alga Emiliania huxleyi, could act as a pro-viral signal, by expediting infection and enhancing the half-life of the virus in the extracellular milieu. Here, we expand our laboratory findings and show the effect of vesicles on natural populations of E. huxleyi in a mesocosm setting. We profile the small-RNA (sRNA) cargo of vesicles that were produced by E. huxleyi during bloom succession, and show that vesicles applied to natural assemblages expedite viral infection and prolong the half-life of this major mortality agent of E. huxleyi. We subsequently reveal that exposure of the natural assemblage to E. huxleyi-derived vesicles modulates not only host-virus dynamics, but also other components of the microbial food webs, thus emphasizing the importance of extracellular vesicles to microbial interactions in the marine environment.

Fig. 1. Extracellular vesicles from natural E. huxleyi populations.

A bloom of E. huxleyi was induced in a mesocosm setup at the Marine Biological Station Espegrend, Norway (60°16′11 N; 5°13′07E) in May-June 2018. Abundance of calcified E. huxleyi cells (a) was measured by flow cytometry, and EhV concentration (b) was measured by qPCR targeting the major capsid protein gene (mcp). Average ± SE of the four mesocosm bags are presented for (a) and of bags 2 and 4 in (b). Abundance of cells and virions was also measured in the surrounding fjord water as a “blank” control (empty circles). An asterisks indicate the times at which samples were taken for vesicle extraction; arrows indicate sampling times for the experiments presented in Fig. 2a, b; cross indicates time of sampling of EhV for the experiment presented in Fig. 2c. c Workflow for vesicle sRNA profiling. Sampling (1): 25 l water samples were collected on the days marked with an asterisk in (a) from each bag using a peristaltic pump and a 200 µm nylon pre-filter. Concentration (2): the water samples were combined and filtered through two subsequent filters (GF/C and 0.45 µm PVDF). Samples were then concentrated to ~500 ml on a 100 kDa TFF cartidge and stored at 4 °C in the dark. Separation and analysis (3): Once at the home lab, the samples were further concentrated on 100 kDa Amicon-ultra filters, and separated on an Optiprep gradient. After speparation and cleaning, vesicles were subjected to RNase treatment to eliminate extra-vesicular RNA. sRNA within the vesicles was then extracted and sequenced. Workflow was created with BioRender.com. d Vesicles were collected from the natural assemblages at the time points indicated by an asterisk in (a) and the sRNA cargo was sequenced. sRNA sequences were aligned to E. huxleyi target genes as indicated. sRNAs that target the same genes were also found in vesicles from lab cultures of (i) uninfected E. huxleyi CCMP2090, (ii) infected cultures, (iii) Both uninfected and infected cultures or (iv) not found in vesicles from lab cultures of E. huxleyi CCMP2090. Read counts were scaled to one million reads mapped to the E. huxleyi transcriptome, log2 transformed and compared across time points.

Fig. 2. Extracellular vesicles modulate viral infection of E. huxleyi and prolong the half-life of EhV in natural assemblages.

Vesicles derived from lab cultures of E. huxleyi CCMP2090 were incubated with natural microbial populations from (a) the early bloom phase (day 14, blue arrow in Fig. 1a) or (b) demise phase (day 20, red arrow in Fig. 1a). Abundance of nanophytoplankton (including calcified and non-calcified E. huxleyi), calcified E. huxleyi, EhV-like particles, Synechococcus, and bacteria was measured by flow cytometry 48 h post-treatment. In (a), significant difference between the treated and untreated populations was observed only for bacteria (***p value = 0.001). For untreated n = 10, for vesicle-treated n = 30. In (b), significant reduction in nanophytoplankton abundance (***p value = 0.0003), concomitant with elevation in EhV-like particle count (**p value = 0.009) and Synechococcus abundance (***p value = 1.03 × 10⁻⁷) was observed. For both treated and untreated n = 30. At least 10,000 events were counted for each gate. In (a) and (b), p value was calculated using two tailed t test with equal variance. c Half-life of infectious EhV from the mesocosm bags, measured by the most probable number (MPN) method. EhV was sampled from the bags during the demise phase of the bloom (day 18, green cross in Fig. 1a). Vesicles from infected (^VirusVesicles) or uninfected (^ControlVesicles) lab cultures of E. huxleyi CCMP2090 were added to natural EhV at a ratio of 10 vesicles per EhV-like particle. For the ^Controlvesicle treatment, the decay was so fast we could not detect infectivity in more than one time point. Therefore, the minimum detectable infectivity values were used in the subsequent time points in order to calculate the maximum possible half-life. Average ± SE is presented. ***p value < 0.001 for each treatment compared to the untreated control, using ANOVA with Dunnett’s post-hoc test, n = 3.