Impact of airborne algicidal bacteria on marine phytoplankton blooms

Abstract

Ocean microbes are involved in global processes such as nutrient and carbon cycling. Recent studies indicated diverse modes of algal-bacterial interactions, including mutualism and pathogenicity, which have a substantial impact on ecology and oceanic carbon sequestration, and hence, on climate. However, the airborne dispersal and pathogenicity of bacteria in the marine ecosystem remained elusive. Here, we isolated an airborne algicidal bacterium, Roseovarius nubinhibens, emitted to the atmosphere as primary marine aerosol (referred also as sea spray aerosols) and collected above a coccolithophore bloom in the North Atlantic Ocean. The aerosolized bacteria retained infective properties and induced lysis of Gephyrocapsa huxleyi cultures.This suggests that the transport of marine bacteria through the atmosphere can effectively spread infection agents over vast oceanic regions, highlighting its significance in regulating the cell fate in algal blooms.

Keywords: Gephyrocapsa huxelyi; aerial infection; airborne pathogen; marine bioaerosols; oceanic blooms; ocean–atmosphere interactions.

Graphical Abstract

Figure 1

Experimental setup to test the pathogenicity of the collected aerosolized marine bacteria on coccolithophore bloom; marine aerosols were pulled from ~30 m above mean sea level; in one inlet, air samples were split into an OPC, and a Spot-Sampler, collecting air into FSW for periods of ~12 h; from the other inlet, aerosols were collected for periods of ~12 h on PVDF filters for genomic analysis and on polycarbonate filters for SEM-EDS analysis (A); Spot-Sampler’s air samples and controls were cocultivated with healthy G. huxleyi cultures, with and without antibiotics (B); demised cultures were used for the isolation and identification of bacterial strains; these bacteria were reintroduced to healthy G. huxleyi cultures (C); in parallel, airborne samples and blanks were used for genomic analysis that included both 16S rRNA amplicon sequencing and quantitative PCR for the calculation of the bacterial aerosolization fraction (D); created with BioRender.com.

Figure 2

Sampling location and ocean surface PIC concentration. Map showing the sampling location (centred at 48.38°N 7.06°W) with 48-h air mass back trajectories arriving at 50 m height at noon (A); Composite image from MODIS aboard the Aqua satellite between May 29 and June 2, 2019, of particulate inorganic carbon (PIC) levels during sampling. The location of R/V Tara is indicated by circles, showing its position within the bloom area (circled with dashed line) during the sampling period and its departure from the bloom on the night of June 2 (B).

Figure 3

Meteorological conditions during the 5-day sampling period; wind speed measured ~30 m above mean sea level (arrows show the direction and represent the 1-h mean, the shadow represents the standard deviation) and RH measured ~7 m above mean sea level (rhombus) (A); air temperature measured ~7 m above mean sea level (squares) and SST measured between 0.5 and 2 m depth (circles) (B); aerosol composition determined by SEM-EDS for geometrical diameters (D_geo) > 0.3 μm (C); the filled rectangle above the panels correspond to the time duration inside the bloom, as circled in Fig. 2B; the lighter (predominantly day) and darker (predominantly night) bars in panles (A) and (B) represent the measuring period for each filter.

Figure 4

Roseovarius nubinhibens isolated from air samples kills various natural G. huxleyi strains in the lab; detailed time courses of G. huxleyi cells (A—NCMA379; B—Eh53; C—Eh55; D—Eh60, more information is given in Table S5) incubated with (circles) and without (squares, as a control) air sample-isolated R. nubinhibens and the bacterial cell growth in the monocultures ( rhombus), represented by the difference between the infected and the control cultures; the symbol denotes calcified strains.

Figure 5

Abundance of marine aerosols, R. nubinhibens, and G. huxleyi in the AMBL; The concentration of marine aerosols above the bloom for 0.5 < D_op < 1.3 μm (orange), and 1.6 < D_op < 32 μm (brown) (A), as well as the abundance of the MraZ gene of R. nubinhibens (circles) and the Cox3 gene (triangles) from G. huxleyi in air filter samples (B), and the MraZ gene of R. nubinhibens and the calcified G. huxleyi cell counts per litre of the water samples collected at ~15 m depth (C); results in (B) and (C) represent the average of three replicates (n = 3), and the error bars are 1σ; the shaded bars represent predominantly nighttime samples, and signify the measuring period for each filter; Ex. Bloom refers to the area outside of the G. huxleyi bloom.