Abstract
Little is known on the spreading capacities of Limnomonas gaiensis across freshwater lakes in Northern Europe. In this study, we show that the species could successfully be aerosolized from water sources by bubble bursting (2-40 particles.cm−3), irrespectively of its density in the water source or of the jet velocity used to simulate wave breaking. The species viability was impacted by both water turbulences and aerosolization. The survival rate of emitted cells was low, strain-specific, and differently impacted by bubble busting processes. The entity “microalga and bionts” could produce ethanol, and actively nucleate ice (principally ≤−18 °C) mediated soluble ice nucleation active proteins, thereby potentially impacting smog and cloud formation. Moreover, smallest strains could better cope with applied stressors. Survival to short-term exposure to temperatures down to −21 °C and freezing events further suggest that L. gaiensis could be air dispersed and contribute to their deposition.
Similar content being viewed by others
Introduction
The microalga Limnomonas gaiensis inhabits freshwater lakes in Northern Europe. This member of the Chlamydomonas phylogroup has recently been morphologically and genetically described1. The species has been isolated from unconnected water systems in Northern Europe1 and presents key features for organismal dispersal and local adaptation characterized by a potential to acclimate to a wide range of pH2. However, its spreading capacity is not known.
Airborne Chlamydomonas species have been reported from a wide range of geographical locations3, with successful dispersal4,5,6 including in the snow species C. nivalis7. Because of the distance and absence of water connectivity between lakes where L. gaiensis occurs, we hypothesized that air dispersal may play a role in its spreading.
Aquatic microalgae are aerosolized by water surface abrasion3, via wind-friction and breaking wave crests generating spume drops8, or by bubble bursting producing film or jet droplets. Microalgal aerosolization has been reported over land and ocean, with variations according to location, wind conditions3,9, organismal density in the water source, and growing conditions10,11. To date, less than a hand-full emission fluxes are available4,12,13,14 reaching up to 3 × 103 cells.m−3 in microalgae and 4 × 105 cells.m−3 in picomicroalgae (0.2–2 µm). Moreover, processes governing microalgal aerosolization are still poorly characterized.
Microalgal aerosolization has retained attention because of their capacities at interacting with the atmosphere, adapting morphologically to survive atmospheric conditions15, dispersing to new environments, and being a source of sanitary risks for the environment and the society3,9,16. Proliferation of airborne microalgae, such as Chlamydomonas spp.16, can lead to severe environmental and sanitary issues both indoors16,17 and outdoors3,9,18. Moreover, some can proliferate in lakes contaminated by toxic cyanobacteria6, a similarity with L. gaiensis1.
Microalgae can produce volatile organic compounds (VOCs) that are important for atmospheric chemistry19,20,21,22. They can also actively nucleate ice from below −6 °C mediated the production of ice nucleation active (INA) compounds23 and from below −23 °C through the production of INA exudates24. More specifically, certain Chlamydomonas sp. can actively nucleate ice25 at temperature comprised between −8 and −17 °C. Therefore, produced microalgal VOCs and INA molecules can have potential impact on atmospheric processes such as clouds formation, and their own deposition.
Aerosolized microalgae are not expected to stay airborne for prolonged periods of time due to their typically large sizes3. For this reason, their impact on climate and spreading mediated air transportation has been thought to be negligible. Surprisingly, some microalgae have been reported far away from their potential sources, even in remote locations such as Antarctica3,26,27,28. Additionally, using back trajectory analysis, studies have shown that long and viable transportation of microalgae was feasible15,29,30,31. The long air transportation gives rise to several opportunities of interactions with solar radiation and increases their potential to act as so-called giant cloud condensation nuclei (CCN) forming the seed for liquid cloud droplets to form. The role of airborne microalgae as giant CCN has previously been suggested28 but is so far not well understood.
This study aims to infer if L. gaiensis can be emitted into the atmosphere via bubble busting processes and cope with stresses associated with the emission, and to identify biotic compounds that could influence atmospheric processes. We investigated its (i) aerosolization capacity (i.e., relative magnitudes of aerosolization fluxes) for different types of jets, (ii) effect on atmospheric chemistry (i.e., production of VOCs and of ice nucleation active compounds from −4 down to −21 °C), and (iii) viable capacities after aerosolization and freezing events. Results from our study may provide novel insights in microalgae dispersal and their role as primary biological aerosol particles (PBAPs) in the atmosphere.
Results
Pilot study
In a pilot study (Experiment (Exp)1–2, Table 1), we investigated the impact of bubble bursting (Single Jet (SJ) versus Multiple Jets (MJ)), at different water flow intensities (SJ7, SJ9, MJ5, MJ3), on the emission of VR66-07 and R86-47. In all scenarios, results showed that microalgae were successfully aerosolized by bubble bursting.
Direct measurement of the emitted microalgae (EM) ensemble (i.e., cells, cell fragments and other microalgal material) was estimated using the Optical Particle Spectrometer (OPS) (Supplementary Figs. 1 and 2). Exp1 data was excluded from the emission flux analysis due to a malfunctioning in the OPS pump. Exp2 showed a constant emission of aerosolized particles over time in SJ and MJ. The flow rates SJ9 and MJ3 yielded the largest concentrations of emitted particles (Supplementary Fig. 2). In the light of these results, we applied SJ9 and MJ3 to the successive experiments.
The abundance of EM, captured onto filters over Exp1-2, was assessed using Lugol fixed samples. In all samples, numerous cell fragments were visible. Unexpectedly, the number of intact EM was higher after SJ (2467 cells ± 336) than after MJ (1800 cells ± 255) (one-way ANOVA F(1,6) = 10.0, p = 0.020), in both strains (F(1,6) = 0.67, p = 0.45). We suspected that the filtration pressure may have damaged EM integrity and affected the results. To avoid bias, an alternative technique in liquid phase was used in Exp5–7 to capture EM.
Indirect estimation of the number of EM was calculated from Lugol fixed samples collected from the water tank (Supplementary Figs. 1–3), showing a significant decrease in the abundance of aquatic microalgae (−73.8% ± 0.4 cells on average; Kruskal–Wallis X2(4) = 21.94, p < 0.001, Dunn test p < 0.001), in both strains (X2(1) = 0.004, p = 0.95). Despite the percentage of EM after MJ (51.2% ± 19.9) was higher than after SJ (41.7% ± 22.9), there was no significant difference between the two treatments (Fisher Exact probability test, p = 0.26), possibly due to the high standard deviation retrieved from the counts. Discrepancies regarding Lugol counts over time are known32, and therefore a more systematic approach was used to estimate abundances, e.g., counting replicates of a sample phase within a day and applying the same dilution factors for all replicates.
Aerosolization
The OPS data confirmed that both strains were emitted under SJ9 and MJ3 (Table 1 and Fig. 1). The total concentration of EM ensemble ranged from 2 to 40 particles.cm−3 and was relatively constant in all treatments and strains (Fig. 1). The emission fluxes, ranging 0.8–7.9 × 104 m−2.s−1, were comparable between the two strains (one-way ANOVA F(1,10) = 2.16, p = 0.17) and were systematically, but only marginally significantly smaller under SJ9 (2.4 × 104 m−2.s−1 ± 1.3 × 104) than MJ3 (4.8 × 104 m−2.s−1 ± 2.4 × 104) (Fig. 1, Kruskal–Wallis X2(1) = 3.69, p = 0.055). Moreover, there was no clear relation between the emission flux and microalgal abundance in the water tank (Supplementary Fig. 4).
The time series illustrating the total concentration of aerosolized particles (a, c) under SJ9 (blue) and MJ3 (orange) treatments in two strains of Limnomonas gaiensis, i.e., VR66-07 (a, b) and R86-47 (c, d), in Experiment 2 (crosses), Experiments 3–4 (circles), Experiments 5–6 (upside-down triangles), and Experiment 7 (upside-up triangles). Additionally, the mean flux is illustrated for each of the two strains (b,d), for both treatments, together with one standard deviation denoting the variability over time (n = 52–58).
The mean number of EM captured by filters (solid and dry phase, Exp1–4) was 2586 cells (± 876), and numerous broken cells and debris were visible. Captured cell number did not differ significantly between treatments (Kruskal-Wallis X2(1) = 0.53, p = 0.47), nor strains (one-way ANOVA F(1,10) = 1.04, p = 0.33). The number of EM captured by impingers (liquid phase, Exp5–7) was at least 10 times higher than using filters (Table 1), with few to no broken cells visible. However, the proportion of cells captured by the two devices were fairly similar as the microalgal abundance in the emission source was >10 times higher in Exp5–7 than in Exp1–4 (Table 1). Captured cell numbers by impingers did not differ significantly between treatments (Kruskal–Wallis X2(1) = 0.017, p = 0.90), but between strains (Kruskal–Wallis X2(1) = 14.63, p < 0.001).
Indirect measurement confirmed the aerosolization of both strains from the water source with 32.8% (± 24.6) of EM. Despite a slightly higher EM rate under SJ (41.1% ± 25.1) than MJ (24.5% ± 23.3), this emission was not significantly affected by the choice of treatment (two-way ANOVA F(1,8) = 1.21, p = 0.30) nor by strains (F(1,8) = 0.58, p = 0.47), or by strains and treatment (F(1,8) = 0.062, p = 0.81). The fluctuation in microalgal abundance in the water tank (Table 2) was not significantly affected by treatments (Kruskal–Wallis X2(1) = 2.37, p = 0.12) or strains (X2(1) = 0.009, p = 0.93). Notable fluctuations in microalgal abundances in the water tank were observed between experiments (Kruskal-Wallis X2(6) = 32.59, p < 0.001) as a result of different microalgal load in the water tank (X2(4) = 38.99, p < 0.001), in particular between Exp1–4 and Exp5–7 (Dunn Test p < 0.05). When 107 cells were loaded in the tank (Exp1–4), the microalgal abundances in the water tank remained higher under SJ than MJ (Kruskal–Wallis X2(1) = 12.81, p < 0.001), resulting in lower emission fluxes under SJ, as observed in the OPS data (Fig. 1). However, when higher microalgal abundances were loaded (Exp5–7), there was no significant difference between treatment (Kruskal–Wallis X2(1) = 0.11, p = 0.74), despite systematically higher fluxes were recorded by the OPS under MJ (Fig. 1).
Biovolume
Strain biovolume was estimated from the water source and filters (Exp2–4, Fig. 2) to investigate if size selection occurred during emission. Both strains showed the same trend in all experiments (two-way ANOVA strains: F(1,6) = 2.60, p = 0.16, experiments: F(1,6) = 0.036, p = 0.86). A significant decrease in biovolume was observed between the starter (original culture) and the treatments (F(4,6) = 11.91, p = 0.0051). The cell biovolume was not affected by SJ or MJ in organisms collected on filters (TukeyHSD p = 0.99) nor in the water tank (TukeyHSD p = 0.99). But difference in biovolume was observed between strains under treatment (one-way ANOVA F(1,8) = 6.374, p = 0.036), with an average smaller biovolume in R86-47 (24.0 µm3 ± 11.6) than in VR66-07 (42.3 µm3 ± 10.5).
Box plot of the biovolume values measured in two strains of L. gaiensis: VR66-07 (Experiment 4) and R86-47 (average of Experiments 2–3). Number of observations are comprised between 6 and 50 per strain and treatment. Significant differences between strains through the experiment was observed (a vs b, p = 0.036).
Revival capacities after aerosolization
To understand the constrains of treatment and recapture on L. gaiensis potential for air dispersal, the revival capacity of EM was investigated. None of the inoculates collected from filters were able to revive or show vital signs (growth or movement) under condition favoring growth over a two-month period of incubation. Instead, numerous fragmented cells and debris were visible. Results suggested that only a small proportion of EM may be viable, or, that the EM were negatively impacted by filtration pressure.
Inoculates collected from impingers showed 10.4% (± 4.8; 57/548 inoculates) survival success, suggesting that capture in liquid phase may break less cells and that a small portion of microalgal strains could be airborne dispersed. The two strains reacted differently to bubble bursting treatments, with a lower survival rate in VR66-07 (4.9%; 9/184 inoculates; Exp5) than in R86-47 (13.2% ± 1.0, Exp6-7; 48/364 inoculates) (Z-test X2(1)Exp5-Exp6 = 5.78, p = 0.016; X2(1)Exp5-Exp7 = 7.67, p < 0.01). More specifically, significant differences were observed for MJ (VR66-07: 0% (0/92 inoculates) vs R86-47: 14.8% ± 0.5 (27/182 inoculates); Fisher’s Exact Test for Count Data pMJ < 0.0001), but not SJ (VR66-07: 9.8% (9/92 inoculates) vs R86-47: 11.6% ± 2.5 (21/182 inoculates), strains: 11.0% ± 2.1 (30/274 inoculates), Fisher’s Exact Test for Count Data pSJ = 0.81). Results were reproducible between experiments (X2(1)Exp6-Exp7 = 0.06, p = 0.81) and treatments (Z-test X2(1)MJ < 0.001, p = 1.0; X2(1)SJ = 0.27, p = 0.61).
Cell vitality
Organism survival capacity was also assessed using live/dead stains in samples collected from the water tank and after emission. The Neutral Red vital stain was used to infer the proportion of intact organisms. It efficiently stains several microalgal taxa23,33 but did not properly stain L. gaiensis as the orange-red signal of the dye was masked by the microalgal cup-shaped morphology of the chloroplast, hiding signals from underneath vacuoles and cytoplasm.
Propidium iodine was used to assess the fraction of dead/damaged organisms (Exp7, Supplementary Table 1). The percentage of dead cells was low in the starter (7.0% ± 0.5). It significantly increased to 25.5% (± 3.5) after water homogenization (Fisher’s Exact Test for Count Data p < 0.001) and remained constant across treatments in the aquatic population (SJ: 28.4% ± 2.3, MJ: 27.9% ± 1.6; Fisher’s Exact Test for Count Data pHomogenization-Treatment = 0.87 and pSJ-MJ = 1.0). The dynamic of dead cell occurrence investigated using flow cytometry (Fig. 3) confirmed a net diminution of the intact cell population (outside the gate) towards dead cells (inside the gate), principally after water homogenization (Fig. 3a, b), indicating that water turbulence, regardless of SJ or MJ (Fig. 3c, d), had a negative impact on cell survival. The percentage of dead cells was very high in the emitted fraction (97.1% ± 4.2), corroborating with the low revival rate retrieved from the inoculates. There was a marginally significantly higher proportion of dead cells under SJ than MJ (Fisher’s Exact Test for Count Data p = 0.06). However, this tendency is to take with caution because the total number of cells detected by the flow cytometer in the emitted fraction was quite low and in the range of the detection threshold.
Dead cells stained by propidium iodide (PI+; cluster on the right) with an indication of the percentage of dead cells among chlorophyll-containing cells (ntotal). Treatments include the initial culture (a, ntotal = 0.9 × 109 cells), after water homogenization (b, ntotal = 1.6 × 109 cells), after SJ9 treatment (c, ntotal = 2.4 × 109 cells), and after MJ3 treatment (d, ntotal = 2.4 × 109 cells).
VOCs
VOCs production is of relevance for their potential effect on atmospheric chemistry (e.g., climate, cloud formation, see Discussion). Gaseous emissions of L. gaiensis natural entity (i.e., microalga and bionts, see Methods) were investigated over the experimental phases. VOCs in the m/z range 21–200 were explored and the emission of m/z 47.05 (ethanol) under stress was found. Despite ethanol is used for cleaning purposes and its mixing ratio in the laboratory air varied substantially between experiments, L. gaiensis entity was able to produce ethanol under stress (Fig. 4).
Panels a and c illustrate results for two experiments with the Limnomonas gaiensis strain VR66-07 over Experiments 4 and 5, respectively, and panels b and d show the results with strain R86-47 over Experiments 3 and 6, respectively. Only the VOC m/z 47.05 was detected and corresponds to the protonated signal for ethanol. Five different phases of the experiments are highlighted in different colors, e.g., emission from MilliQ water (gray dots, n = 212–1200), emissions from bubbling MilliQ water (black dots, n = 757–2134), emissions during first mixing (homogenization, purple dots, n = 351–685), emissions from MJ3 (orange dots, n = 3360–3720) and SJ9 (blue dots, n = 3180–3670) treatment.
Ethanol emissions were low when the tank was filled only with water (water controls) and higher when the L. gaiensis entity was present (treatments, Fig. 4). A strong initial decrease in ethanol concentration was recorded over time when only water was in the tank (water condition, Fig. 4 Exp5-6), reflecting the replacement of laboratory air in the tank headspace by filtered air. The ethanol signal strongly increased when organisms were homogenized (mix condition, Fig. 4). A steady increase in ethanol emission was observed with SJ9, characterized by a slope of 0.006 ppb.s−1 (Exp3) and 0.040 ppb.s−1 (Exp6) in R86-47 and of 0.002 ppb.s−1 (Exp4) and 0.006 ppb.s−1 (Exp5) in VR66-07. It also increased during MJ3 in R86-47 with a slope of 0.001 ppb.s−1 (Exp3) and 0.03 ppb.s−1 (Exp6) but remained constant in VR66-07.
Frozen fraction and IN activity
The capacity of L. gaiensis entity to actively nucleate ice was investigated using droplet-freezing assays in four strains1. Freezing events and the number of ice nucleating particles (INP) (total fraction, Figs. 5, 6) were estimated when exposed to a gradient spanning −4 to −21 °C. Strains originating from Lake Ryssbysjön started freezing at −8 °C, i.e., R86-45 at >−8 °C with 2.1 × 10−6 INP.cell−1 and R86-47 at <−8 °C with 3.3 × 10−6 INP.cell−1. Strains from Lake Västra Ringsjön were active at lower subzero temperatures, i.e., VR66-10 at <−17 °C with 2.5 × 10−5 INP.cell−1 and VR66-07 at <−18 °C with 8.2 × 10−6 INP.cell−1. In R86-47 the IN activity remained low, between −8 and −17 °C ( ≤ 3.9 × 10−6 INP.cell−1), sometimes below the detection limit (<−12 °C) and started to increase again at <−17 °C ( ≤ 1.5 × 10−4 INP.cell−1). In all strains, half of the replicates were frozen (frozen fraction (FF) of 0.5) from −18 down to −21 °C (Fig. 5). At −21 °C, all replicates were frozen (FF = 1) in R86-45 (Fig. 5a). In the three other strains (Fig. 5b–d), FF reached 0.95 in VR66-10, 0.88 in R86-47 and 0.75 in VR66-07. At −21 °C the number of INP was 1.01 × 10−4 (± 0.4 × 10−4) INP.cell−1 on average, reaching 8.2 × 10−5 INP.cell−1 in R86-45, 1.5 × 10−4 INP.cell−1 in R86-47, 5.2 × 10−5 INP.cell−1 in VR66-07, and 1.2 × 10−4 INP.cell−1 in VR66-10 (Fig. 6). Results indicated that L. gaiensis entity could be IN active at rather low temperatures, almost negligeable compared to known INA PBAPs (e.g., P. syringae, our positive control, ≤−6 °C) and abiotic particles (≤−12 °C).
The four investigated strains of Limnomonas gaiensis, R86-45 (a), R86-47 (b), VR66-07 (c), and VR66-10 (d) exposed to a decreasing gradient of −1 °C from −4 to −21 °C. The total fraction is indicated in green. Treatments included either protein denaturation by heat (non-proteinaceous, orange), molecule size selection by filtration on 0.22 µm pore membrane (soluble, blue), or a combination of both treatments (soluble non-proteinaceous, purple). The number of replicates was of 32 droplets of 20 µL of either microalgae (strain), MWC (negative control, black) or Pseudomonas syringae active >-4 °C (positive control, gray). The error bars show the 95% confidence interval. Each data point is the synthesis of a total of 52 to 64 replicates per strain and treatment, and of 116 replicates per control.
The four investigated strains of Limnomonas gaiensis, R86-45 (a), R86-47 (b), VR66-07 (c), and VR66−10 (d) exposed to a decreasing gradient of −1 °C from −4 to −21 °C. Data are normalized to the negative control. The total fraction is indicated in green. Treatments included either protein denaturation by heat (non-proteinaceous, orange), molecule size selection by filtration on 0.22 µm pore membrane (soluble, blue), or a combination of both treatments (soluble non-proteinaceous, purple). The error bars show the 95% confidence interval for 52–64 replicates per strain and treatment.
The nature of the component responsible for IN activity was investigated using heat and filtration treatments (Fig. 5). There was a significant effect of treatments on strain activity (Kruskal–Wallis X2(15) = 26.0, p = 0.04, Fig. 6), and in particular an effect of heat (Dunn Test p = 0.002), suggesting that the INA compounds were proteinaceous. Filtration treatment was performed to decipher if the proteinaceous INA compounds were soluble (size <0.22 µm). Strains R86-45 was significantly affected by the filtration treatment (Figs. 5a and 6a, Dunn Test p = 0.03), indicating that its INA compounds were not soluble. However, in the three other strains, the effect of filtration was not significant (Dunn Test p = 0.24, p = 0.63, p = 0.61, respectively), suggesting soluble INA compounds (Figs. 5b–d and 6b–d). When both the total and the soluble fractions were heated, there was a significant drop in the IN activity, below the detection limits (Figs. 5–6), confirming the proteinaceous origin of the INA compounds. In summary, results indicated that the IN activity was triggered by INA proteins that were either soluble (R86-47, VR66-07, VR66-10) or associated to the organisms (R86-45).
Additionally, the peak of IN activities in R86-47 between −8 and −11 °C and in VR66-10 from −12 down to −17 °C were investigated (Fig. 6b, d). R86-47 was sensitive to filtration (Dunn Test p = 0.09) but not to heat treatment (Dunn Test p = 0.41), indicating that its INA compounds were non-proteinaceous possibly associated to the organism (e.g., exudates). VR66-10 showed IN activity only in the soluble fraction over this temperature interval, but not in the total fraction. Moreover, a significant reduction of activity between the soluble and the soluble heated fractions indicated that the soluble INA compounds were proteinaceous (Kruskal-Wallis X2(1) = 8.25, p = 0.004, Dunn Test p = 0.004). Results suggested that the filtration pressure in VR66-10 may have affected the integrity of some cells and released soluble INA proteins in the filtrated fractions.
Revival capacities after cold exposure and freezing events
The revival capacity of L. gaiensis strains was investigated after exposure to thermal stress gradually decreasing down to −21 °C and an incubation period of up to three weeks at conditions favoring growth. The four strains were able to cope with up to 8 h exposure to these subzero temperatures. About half of the replicates revived (183/384 replicates). R86-47 showed the highest survival rate (59–100%) and R86-45 the lowest (0–63%). There was a general decrease in the revival percentage with the increment in cell abundance (Supplementary Fig. 5). The highest percentage of revival was observed in the “youngest tested culture”, having the lowest abundance (Supplementary Table 2) and undergoing exponential growth phase (color-based observation). Interestingly, a third of the revival (58/183 replicates) occurred in wells that experienced freezing, suggesting that L. gaiensis strains could survive freezing events.
Discussion
This study revealed that L. gaiensis could be aerosolized from freshwater sources through bubble bursting processes and survive aerosolization and exposure to cold temperatures, and that the microalga-biont particles could react to aerosolization conditions producing VOCs and actively nucleate ice at warm subzero temperatures. Thus, L. gaiensis is a potential candidate for air transportation and dispersal, with possible effects on atmospheric processes. We provide, for the first time, estimates of L. gaiensis emission fluxes.
Microalgal aerosolization has previously been reported. Chlamydomonas-like species were detected in air samples, snow depositions, and in artificial water tanks following air deposition. In this study, we showed that L. gaiensis could be aerosolized from water sources under bubble bursting using either a single or multiple jets. The emitted concentration of the microalgae ensemble was constant and reached up to 4 × 107 cells.m−3. This emission rate is far higher than natural emission rate records on microalgae (≤3 × 103 cells.m−3)4 and picomicroalgae (≤3.7 × 105 cells.m−3)12.
Natural fluxes at lake scale are difficult to calculate as little is known on the typical limnic concentrations of L. gaiensis or Chlamydomonas sp. One of the reasons is that abundances in limnic surveys are often provided as dry weight or biomass34,35, qualitative measurements35, or binary data (presence/absence) with ecological information36. Emission rates were estimated in studies using either natural and mock phytoplankton community assemblages or culture enrichments loaded in tanks with microalgal concentrations up to ~103−107 cells.mL−1, at the upper ranges of natural phytoplankton blooms10,11,37,38. In our study, we used similar concentration ranges spanning ~103−108 cells.mL−1. In our cell counts, we noticed lower emission of microalgae (3.2%) in densely populated experiments (1012 cells, Exp5) than in less dense water sources ( ~ 107 cells, 44.8% of EM, Exp1–4). The abundance contributes but does not fully explain the higher emission rate retrieved in L. gaiensis, suggesting that other intrinsic and extrinsic factors play a role in microalgal aerosolization (e.g., droplet characteristics, organism size, salinity, growth, and habitat conditions).
Particles aerosolization potential can be influenced by organismal properties39. Jet droplets can up-concentrate 8–307 times available aquatic microalgal density and this up-concentration is taxa-specific11. Organism size can play a role, as the highest emission estimates were reported in picophytoplankton12. Organism growth and the production of exudates can also increase the number of aerosols10. Exudates change the viscosity of the water surface and could affect the number of generated droplets40. Unlike many microalgal taxa (e.g., cyanobacteria, diatoms), L. gaiensis is not known for mucilage or exudates production1. Thus, a combination of organism size and exudate production from the microalgae entity could influence organismal density load per droplet, and thus their emission rate. This hypothesis calls for further investigations, as the load of L. gaiensis per droplet was not investigated, falling behind the scope of this study.
The concentration and size of generated aerosols can differ by the technique and the intensity of the stressor used to produce them41,42. In our study, we used bubble bursting, an aerosolization technique known to produce a broad size distribution and being able to efficiently eject microalgae up to 13 cm into the air37. We induced bubble bursting using single and multiple jets, and generated bubbles that covered the whole water surface area of the tank. Pietsch et al.42 showed a positive correlation between wind speed (water turbulence) and flux of aerosolized aquatic microorganisms. The systematic but marginally lower number of EM in SJ9 than in MJ3, suggests that the microalgal emission would be greater under conditions comparable to the multiple jets setup. While this difference was not significant, we observed it both when investigating the full microalgae ensemble (OPS data) and when following the cell abundance in the water tank. Higher emissions using SJ compared to MJ had previously been observed for the used tank when analyzing water containing synthetic sea salt43. The authors observed a steady increase in particle flux with increasing water velocity, which is not in agreement with the constant fluxes observed in this study when using freshwater microalgae (Supplementary Fig. 2).
Aerosolized microalgae are candidate propagules for air dispersal. Along their journey, they are subject to diverse stressors. Water turbulences generated during bubble-bursting experiments accounted for the loss of about one-fourth of the aquatic population. The homogenization step led to a net diminution of the mean cell biovolume and an increase of at least 18.5% in the number of dead cells. Larger cells were not observed during treatments in the aquatic samples or on filters after aerosolization. Instead, a high number of broken cells and debris were visible from filters. It is possible that larger organisms have been damaged by applied pressures both in the water and on filters, or removed from the water surface by water turbulences, towards the bottom of the water tank. The latter corroborates with simulations44 showing that, against expectations, larger particles would sediment faster under turbulent flow. Ross45 showed that particles sedimented in well-homogenized surface mixed layer (SML) tend to resuspend at the bottom of the SML where higher turbulences occur. Here we used a water tank with a small SML (~27.1–30.1 cm) and sampled the top 2 cm. However, organisms with larger and elongated shaped have a higher encounter rate, facilitating agglomeration and sedimentation46. Moreover, to avoid deleterious effects of strong turbulence, motile phytoplankton species tend to evade turbulent layers47. Therefore, it is possible that cells with larger biovolume would be flushed and trapped at the bottom of the SML, a natural exclusion facilitated by the microalgal tendency to attach to particles and surfaces1.
Limnomonas gaiensis was further negatively selected during aerosolization with ~97.6% of dead cells retrieved from the emitted fraction. A small fraction of EM (2.4%) maintained their viability, some of which revived in conditions favoring growth. Interestingly, the survival rate was higher when the organisms were collected in liquid phase (~10.8%) than on solid-dry phase (0%). Similarly, higher recovery rates of airborne microorganisms were found when using liquid impinger than filters48. On filters, desiccation and osmolysis could have drastically reduced survival chances. Impinger technique appeared to be less destructive with few to no broken cells and debris. This suggests that a low proportion of L. gaiensis can generate viable airborne propagules, and that the duration of their transportation, i.e., period of organism exposure to additional atmospheric stressors49, may play a key role in determining their capacity and distance of dispersal.
The airborne survival of L. gaiensis is affected by subzero temperatures, at times by freezing events. In a previous study23, we showed that airborne and aquatic microalgae, among which Chlorophyceae, can cope with temperatures down to −28 °C. Limnomonas gaiensis could tolerate gradual and shorter exposure down to −21 °C and experience freezing (this study). Despite negative selection, some L. gaiensis could survive thermal stress during air transportation and wet depositions.
Aerosolization survival to dispersal can be strain dependent. Airborne microalgae can adjust their physiological conditions to the environment, strain-specifically producing thicker cell wall and carotenoids to cope with UV exposure and reactive oxygen stress15. Strains of L. gaiensis were impacted by aerosolization treatments, especially when using multiple jets. Interestingly, there was a significant difference in the number of EM between strains. Also, the strain with the larger biovolume (VR66-07) had the highest emission rate when the microalgal load was ~107 cells, whereas R86-47 had not only the smaller biovolume, but it also demonstrated a higher emission rate when the aquatic microalgal load was >109 cells, and a better survival rate both after emission and freezing. Additionally, the negative trend between the percentage of revived organisms and the condition of microalgal growth (density, age) suggests that cell abundance and physiology may play an important role in the species survival capacity. The physiological response under aerosolization and freezing differed between strains, despite organismal concentration and growth phase. VR66-07 and R86-47 had similar revival capacities after cold exposure (Z-test X2(1) = 7.45, p = 0.006) and their entity produce INA soluble proteins active below −17 °C, whereas R86-45 was less efficient at coping with cold temperature exposure (Z-test X2(1)VR66-07- R86-45 = 20.58 and X2(1)R86-47- R86-45 = 42.98, p < 0.001, respectively) and its entity produced non-soluble INA proteins active from −8 °C. To decipher the mechanisms behind L. gaiensis tolerance to atmospheric stressors, results call for complement morphological and physiological investigations.
Even though estimations suggest that air abundance and invasion risk of L. gaiensis are rather low, the microalga-biont natural complex could influence, to a certain extent, its environment producing VOCs and being IN active at warm subzero temperatures.
Emissions by aquatic organisms have been studied carefully, particularly regarding the production of dimethyl sulfide50, a molecule that plays a role in the formation of warm clouds (e.g., CLAW hypothesis51,52). Our study shows that L. gaiensis entity can produce low amounts of ethanol under stress induced by bubble bursting. Gas-phase ethanol reacts in the atmosphere with hydroxyl radicals to form acetaldehyde, thereby potentially contributing to ozone and smog formation. This observation is of great interest as the sources of ethanol in the atmosphere remain so-far poorly quantified.
Interestingly, we found higher concentrations of released ethanol using single than multiple jets. Ethanol is produced through the fermentation of microalgal polysaccharides such as glycogen, cellulose and starch, the latter being accumulated in green microalgae (as Chlamydomonas sp.) in 50% up to 70% of their dry weight53,54. Starch accumulation enhancement can be triggered under microelement limitation coupled with air bubbling54, diverting energy away from cell growth55. Ethanol can serve as a carbon source for other microbes and acts in the microalgae as antioxidant or as stimulant of metabolite accumulation and for the production of lipid, fatty acids, and biomass56. While ethanol can be beneficial for the microalgae, it can also alter cell metabolism, oxidative stress response, and growth. This effect is species-dependent and has been monitored in different taxa in the range of 0.39 to 16 g L−1. In our study, the concentration of ethanol released in the air was low (≤2 × 10−4 g L−1), suggesting that ethanol would rather be a stimulating defense mechanism than an inhibitor of organismal activity.
In the atmosphere, cloud condensation and ice nucleation properties of PBAPs can favor the formation of clouds and wet deposition28, affecting the distance of air transportation of microorganisms (including microalgae) over geographic scales3. Even though they are a minority among aerosols, emitted marine PBAPs, of a size of 0.5–10 µm, can act as giant CCN, influencing precipitation57. Additionally, ice nucleation in microalgae has been reported in the past (see papers herein), including in Chlamydomonas-like species. The L. gaiensis entity could actively nucleate ice at temperatures ≤−18 °C, triggered by soluble INA proteins of a size <0.22 µm (in 3/4 strains). Their activity occurred at similar temperatures as in other aquatic microalgal species (−18 to −24 °C, e.g., Peridinium aciculiferum and Microcystis sp.) with higher efficiency, triggered by a lower number of IN particles ( ~ 10−5 INP.cell−1 in L. gaiensis versus ~10−1–10−2 INP.cell−1)23. However, this activity occurred at lower temperatures than reported in Chlamydomonas sp. (−8 and −17 °C)25. The low IN activity detected in R86-45 from −8 °C (2.1 × 10-6 INP.cell−1) was possibly trigged by cell-attached INA proteins.
As microalgae live in close relationship with their bionts58, we suspect that the microalgal entity, rather than the microalgal alone, is aerosolized. It is therefore possible that the cell-attached INA proteins were not of microalgal origin but were produced by the few prokaryotes attached to the microalgal cells or present in the phycosphere. It has been proposed that attached-bionts can be IN active for the entity59 or contribute to the microalga IN activity25. Nevertheless, microalgal axenic and non-axenic cultures can be active in the same temperature ranges23. It would be interesting to perform complement analyses on R86-45 to decipher the source of the activation.
The low activity observed in the L. gaiensis entity suggests that its contribution to atmospheric processes is negligeable in comparison to efficient PBAPs (active from −1 °C)60 and inorganic particles in general (<−15 °C)28,61 present in the atmosphere. However, the ice nucleation properties of L. gaiensis entity could confer advantages contributing to its survival and to reducing its residence time in the atmosphere.
The distribution of Chlamydomonas sp. covers a wide range of biogeographic regions around the world3, but very little is known on their atmospheric abundance. Microalgal airborne concentrations are usually very low, i.e., up to 104 cells.m−3 over the North Atlantic Ocean14 and hundreds of cells.m-3 locally9,62, sometimes below detection limit63. Only a fraction of the emitted microorganisms can remain airborne (10% after 4 days after emission), allowing long distance travel up to 104 km of 0.5-5 µm unicellular eukaryotes14. Their emission and chance of long-distance transportation can be increased under conditions stimulating small droplet emission42. Additionally, environmental selection pressures, dilution factors during dissipation and transportation, and species selective requirements to freshwater habitats may negatively select the pool of viable aerosolized propagules. Generated data in the present study can serve as inputs in models to assess the potential transportation range of L. gaiensis. Altogether, the capacity of L. gaiensis to be aerosolized, to possibly produce ethanol and INA molecules, to cope with cold temperatures, freezing events, and microbial exposure even to toxic species, to carry bionts (reported here and from the literature), and its potential for spreading and acclimation1,2, makes this species a non-negligeable actor in the environment and the society9,16. Available knowledge calls for further investigations on L. gaiensis survival capacities to atmospheric stressors, its biont diversity, and an evaluation of the viable distance of its propagule dispersal.
Methods
Strains
Strains of the freshwater microalga Limnomonas gaiensis (VR66-07 and R86-47) originating from forest lakes in Sweden1 were grown in MWC medium64 in a controlled climate chamber at 15 °C, with a 12 h dark:12 h light regime and 50 µmol photons.m-2.s-1 light intensity. Strains were non-axenic (Fig. 7), mimicking the species natural entity (i.e., microalga and bionts). Prokaryotes were present in the medium of culture, dispatched as single cells or as groups of cells; few were attached to L. gaiensis; and others were present in its phycosphere. Two additional non-axenic strains (VR66−10 and R86-45), originated from the same forest lakes in Sweden1, were used for investigating the effect of exposure to warm subzero temperatures (see below). The four strains are available at the Culture Collection of Algae at the University of Göttingen, Germany (SAG 2636−2639).
a: VR66-07, b: R86-47, c: VR66-10, and d: R86-45. Scale bar: 10 µm. Microalgal cells appeared red because of their chlorophyll autofluorescence, and their DNA in blue stained by DAPI. Prokaryotes, smaller in size, appeared blue by DNA staining and green in b (universal probe).
Aerosolization settings
A stainless-steel tank was used to investigate the strain aerosolization simulating natural bubble bursting, as described in ref. 65. The tank was filled with 10 L of MilliQ water (EMD Millipore, 18.2 MΩ.cm at 25 °C, 2 ppb TOC). Bubbles were generated either by a single centered 4 mm plunging jet (SJ) or by multiple jets (MJ) composed of eight nozzles of 2 mm each. Water flow rates were set to 6.0 and 6.2 L min−1 (SJ7 and SJ9, respectively, named after the pump setting) in the SJ treatment and 8.7 and 11.4 L min−1 (MJ3 and MJ5, respectively) in the MJ treatment. Flow velocities translated to 8 m.s−1 under SJ- and 6–7 m.s−1 under MJ treatment43. A concentration of 106–1011 microalgal cells.L−1 was loaded in the tank (Table 1), mimicking microalgal concentrations naturally occurring in lake during a bloom of Chlamydomonas sp.34 and previously used in aerosolization experiments38. Emitted gases and particles were sampled from the headspace (ca. 5 L), above the water column.
Emission fluxes
Continuous aerosolized concentrations of particles in the size range 0.30–10 µm were measured using an Optical Particle Spectrometer (OPS 3330, TSI, flow rate: 1 L.min−1). A silica gel diffusion dryer was placed in front of the instrument to assess the concentration of dry particles. The total concentration of particles was then used to calculate the emission flux of the “microalgae ensemble” (i.e., cells, cell fragments and other microalgal associated material). The flux (F) was estimated using Eq. 1, taking into the account the sweep air through the system (Qair), total concentration of particles (NOPS), and the tank surface area (Atank of 3.6 × 10−2 m2)43. Potential wall and bottom effects were not considered. Data was analyzed using AIM Software 10 (TSI).
Volatile organic compounds
Over the duration of each experiment (see Table 1), continuous measurements of the VOCs were taken in the m/z range 21–200 using a Proton Transfer Reaction Time-of-Flight mass spectrometer (PTR-ToF-MS 4000, Ionicon Analytik), operated under standard conditions with an E/N ratio of 132 Td with a drift tube volage of 820 or 900 V, a drift tube pressure of 3.3 or 3.4 mbar and a drift tube temperature of 80 or 60 °C. The PTR-ToF-MS was directly connected to the tank analyzing VOCs in the air above the water (headspace) with a resolution of 1 Hz. Data were analyzed with PTR-MS Viewer 3.2.12 (Ionicon).
Microalgal abundance in the water tank
Microalgal abundance in the tank was evaluated over time and treatment to account for organismal airborne emission. A volume of 1 mL of aqueous sample was collected in triplicate from the tank after each treatment phase (i.e., starters, homogenized tank, SJ, MJ), fixed in neutral Lugol solution (4% final concentration, Bie & Berntsen A/S, Denmark), homogenized, and stored at 4 °C, in the dark, until further analyses. Cell abundance was assessed using a Sedgewick rafter counting chamber (S52, Graticules Optics Limited, United Kingdom) and an inverted microscope (AXIOVERT 135 M, Zeiss), magnification 100 or 200x.
Revival capacities after aerosolization
Emitted organisms were recaptured for a period of 52-61 min on filters (PTFE hydrophilic membranes, 47 mm diameter, 0.20 µm pore size, H020A047A, Advantec) connected to a pump with a flow rate of ~2 L.min−1 or in 50 mL of sterile MWC using an impinger (NS 29/32 25×250 mm, Assistant, Duran), following Grinshpun et al.66 recommendations. After the collection period, filters were subdivided, under sterile conditions. Half of the filter was resuspended in 1 mL sterile MWC medium and fixed in 4% Lugol solution to assess cell abundance. A fourth of the filter was resuspended in 0.5 mL sterile MWC and dispatched in replicates in 96-well sterile microplates (Sarstedt, Germany) filled with 250 µL of sterile MWC, and incubated at 15 °C under light conditions favoring growth (as mentioned above) up to a period of two months. The last fourth of the filter was stored at 4 °C in the dark. For the impinger, a volume of 2 mL of homogenized liquid phase was fixed in duplicates in 4% Lugol solution and stored at 4 °C in the dark until cell abundance assessment, and a volume of 250 µL up to 500 µL was dispatched in replicates in either 96- or 48-well sterile microplates (Sarstedt and Cellstart® Greiner Bio-one) respectively. Inoculated plates were closed with parafilm to prevent wells from drying out and were incubated in a controlled culture chamber at 15 °C, 12 h dark:12 h light regime and 50 µmol photons.m−2.s−1 light intensity. To assess organismal revival through time, inoculates were monitored up to 2 months of incubation using an inverted microscope. Organismal revival and growth were regularly and qualitatively monitored for cell coloration, cell movement (swimming or pendulum movement1) and density using an inverted microscope (Axiovert 135 M, Zeiss).
Microalgal biovolume
Microalgal biovolume (B, µm3) was determined following Eq. 267 for prolate spheroid organisms, considering the organism width (W, in µm) and length (L, in µm).
Cell vitality
Microalgal viability was tested using Neutral Red vital staining for microscopy (72210 Sigma-Aldrich) using 0.2–20% final concentration and 1:50,000 solution, following Zetsche and Meysman33 recommendations and Tesson and Šantl-Temkiv23 protocol. The dye accumulates in the cytoplasm and/or vacuoles of live organisms68 and provides an orange-red coloration to viable organisms.
Additionally, the concentration of total microalgae and of dead microalgae over treatments were estimated from a volume of 200 µL of homogenized culture and analyzed by flowcytometry using a NovoCyte 3000 flow cytometer equipped with a 405 nm, 488 nm, and 640 nm laser (Aligent, Santa Clara, CA). Note that performance and stability of the NovoCyte lasers and detectors are daily checked using NovoCyte QC Particles (Agilent). In each sample, a volume of 1.5 µL of propidium iodide (PI; 1 mg.mL−1, Sigma) was used to specifically stain dead cells. All samples were vortex shortly and loaded along negative controls (medium and autoclaved distilled water) on a 24-tube loader. Analysis was performed using a flow rate of 14 µL.min−1, a sample time of 1.26 min per sample, a 20 µL volume load, a series of 3 rinses between each sample, a homogenization of 1 cycle per sample with a speed of 1000 rpm for 10 sec, a forward scatter threshold for height set >50,000, and excitement with two lasers at 405 nm and 488 nm. Because the signal from both lasers was similar, we here show data from the 488 nm laser for comparison with available body of literature. Generated data was analyzed using FlowJoTM version 10.8.1 (Becton Dickinson & Company 2006-2021).
Ice nucleation activity, frozen fraction, and revival capacities after cold exposure and freezing events
Ice nucleation activity and freezing tolerance in L. gaiensis were investigated using a droplet-freezing assay69 in VR66-07 and R86-47, and in VR66-10 and R86-45, two strains collected from the same lakes1. Prior to the analyses, strains were grown at 4 °C, 50 µmol photons.m−2.s−1, 12 h dark: 12 h light. Twenty to thirty-two replicates of 20 µL of each of the four strains were loaded into 384-well plates (VWR International, Lutterworth, United Kingdom). Each plate also contained 20 to 32 replicates of INA Pseudomonas syringae R10.79 cells70 suspended in MWC (positive control) and 20–32 replicates of MWC (negative control). We used heat and filter treatments following the protocol described in ref. 23 to assess the nature of the INA compounds. A subset of cultures was filtered on polycarbonate membrane (Q-max syringe, Frisenette) to isolate particles of a size inferior to 0.22 µm (i.e., soluble fraction). A denaturation step by heat shock, of 10 min incubation at 100 °C, was used to reveal if the soluble particles and the total INA particles (soluble and particulate) had a proteinaceous origin71. Twenty to thirty-two replicates of 20 µL of each treatment were loaded alongside with the controls. The freezing profiles were determined using two biological replicates for a total of 52-64 replicates per treatment and per strain (n = two experiments × 20–32 replicates) and of 116 replicates for each control (n = two experiments × duplicate × 20–32 replicates), for each of the investigated temperature. Plates were incubated for 30 min in an environmental climate chamber (Binder MK115, Tuttingen, Germany) through a gradient of −4 to −21 °C, with an increment of −1 °C. The initial microalgal concentration span from 0.2 to 13.5 × 105 cells.mL−1 (340–26,933 cells per inoculate). The fraction of frozen wells per condition and strains was monitored visually across the temperature gradient.
The frozen fraction (FF, i.e., proportion of frozen replicates for a given treatment and strain) was calculated using the equation from ref. 72. The FF of the negative control was subtracted to the FF of each sample. The number of ice nucleating particles (INP) per replicate was calculated using a modified equation from Vali72, described in23, taking into the account the concentration of microalgae per replicate. Negative values and measurements with less than three consecutive occurrences above the detection limit were removed from the FF and INP datasets. INP data provided in Fig. 6 were normalized to the negative controls. Graphical analyses were performed using the R software version 4.2.073 and the packages ggplot274, scales version 1.2.075, dplyr76, and grid77. Investigated plates were then incubated at 4 °C under conditions favoring growth. Growth in each inoculates was monitored as mentioned above over a period of 23 days of incubation.
Bionts
The presence of microalgal epibionts was investigated using Fluorescent In Situ Hybridization and fluorescent counter-staining. Cells were fixed by mixing one part of cellular growth with three parts of cold 4% PFA and incubating for 12 h or 24 h at 4 °C. Fixed samples were centrifugated at 14,000×g for 5 min at 4 °C and the pellet was washed in 1x phosphate-buffer saline (PBS) three times. Cells were resuspended in a 1:1 PBS:96% ethanol mixture and stored at -20 °C. Prior to cell hybridization, cells were filtered on a polycarbonate membrane filter (0.2 μm pore size). Cell hybridization was performed using EUB-MIX probe78,79 (0.5 ng DNA.µL-1) and the negative control NON-EUB probe80, both labeled with the fluorescent dye Atto-488 in 35% formamide for 2 h at 46 °C. Cells were then counterstained using 4′,6-diamidino-2-phenylindole (DAPI, 1 µg.mL−1), and mounted in Citifluor:Vecta Shield (4:1 v:v) medium. Microscopic imaging was performed using a fluorescence microscope at 400x magnification (Eclipse Ni, Nikon, Axiovert 200 M, Zeiss) and analyzed using the imaging software NIS-Elements (v.4.50, Nikon Instruments Inc., Melville, NY, USA).
Statistics and reproducibility
Statistical analyses were performed in R73. Comparison of means was performed using one-way and two-way ANOVA upon the assumption of homogeneous variances, tested using the F- and Levene tests, and a normal distribution of residuals, tested using the Shapiro-Wilk normality test. The post-hoc range test Tukey’s honestly significant difference (HSD) was used for multiple pairwise comparison with a confidence level of 95%. Upon inequality of variances, the non-parametric Kruskal–Wallis rank-sum test was used to determine if differences between groups were statistically significant, with a significant level alpha of 5%. Post hoc multiple pairwise comparisons were then performed using Dunn test and the R package FSA version 0.9.381. Comparison of proportions was performed using Z-test with continuity correction for small sample size.
Inclusion and ethics statement
The research included local researchers throughout the research process; its relevance has been discussed among co-authors and with colleagues. Roles, responsibilities, and research plan were agreed amongst collaborators ahead of the research. The research was not severely restricted nor prohibited in the setting of the researchers. The research was based on microalgae cultures and was not approved by local ethics as it does not involve work on humans, animals, or dangerous substances. The research was undertaken to the higher standards and in accordance with Aarhus University and Danish regulations. The research did not result in stigmatization, incrimination, discrimination, or otherwise personal risk to participants. The research did not involve major health, safety, security or other risk to researchers, and basic laboratory safety measures were performed according to regulations at Aarhus University. No biological materials, cultural artefacts or associated traditional knowledge were produced in this research. Local and regional research relevant to our study were considered in citations.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
The datasets generated during and/or analyzed during the current study are available in the Supplementary data file and from the corresponding authors on reasonable request. Investigated strains are available at the Culture Collection of Algae at the University of Göttingen, Germany (SAG 2636-2639).
Change history
14 August 2023
A Correction to this paper has been published: https://doi.org/10.1038/s42003-023-05228-9
References
Tesson, S. V. M. & Pröschold, T. Description of Limnomonas gen. nov., L. gaiensis sp. nov. and L. spitsbergensis sp. nov. (Chlamydomonadales, Chlorophyta). Diversity 14, 481 (2022).
Tesson, S.V.M. Physiological responses to pH in the freshwater microalga Limnomonas gaiensis. J. Basic Microbiol. https://doi.org/10.1002/jobm.202300107 (2023).
Tesson, S. V. M. et al. Airborne microalgae: insights, opportunities and challenges. Appl. Environ. Microbiol. 82, 1978–1991 (2016).
Brown, R. M. Jr, Larson, D. A. & Bold, H. C. Airborne algae: their abundance and heterogeneity. Science 143, 583–585 (1964).
Genitsaris, S., Moustaka-Gouni, M. & Kormas, K. A. Airborne microeukaryote colonists in experimental water containers: diversity, succession, life histories and established food webs. Aquat. Microb. Ecol. 62, 139–152 (2011).
Genitsaris, S. et al. Air-dispersed aquatic microorganisms show establishment and growth preferences in different freshwater colonisation habitats. FEMS Microbiol. Ecol. 97, fiab122 (2021).
Marshall, W. A. & Chalmers, M. O. Airborne dispersal of antarctic terrestrial algae and cyanobacteria. Ecography 20, 585–594 (1997).
Andreas, E. L. et al. The spray contribution to net evaporation from the sea: a review of recent progress. Bound. Layer. Meteorol. 72, 3–52 (1995).
Wiśniewska, K., Lewandowska, A. & Śliwińska-Wilczewska, S. The importance of cyanobacteria and micro- algae present in aerosols to human health and the environment–Review study. Environ. Int. 131, 104964 (2019).
Alpert, P. A. et al. The influence of marine microbial activities on aerosol production: a laboratory mesocosm study. J. Geophys. Res. -Atmos. 120, 8841–8860 (2015).
Marks, R. et al. Rising bubbles as mechanism for scavenging and aerosolization of diatoms. J. Aerosol Sci. 128, 79–88 (2019).
Murby, A. L. & Haney, J. F. Field and laboratory methods to monitor lake aerosols for cyanobacteria and microcystins. Aerobiologia 32, 395–403 (2016).
Gregory, P. H., Hamilton, E. D. & Sreeramulu, T. Occurrence of the alga Gloeocapsa in the air. Nature 176, 1270 (1955).
Mayol, E. et al. Resolving the abundance and air-sea fluxes of airborne microorganisms in the North Atlantic Ocean. Front. Microbiol. 5, 557 (2014).
Chiu, C.-S. et al. Mechanisms protect airborne green microalgae during long distance dispersal. Sci. Rep. 10, 13984 (2020).
Genitsaris, S., Kormas, K. A. & Moustaka-Gouni, M. Airborne algae and cyanobacteria: occurrence and related health effects. Front. Biosci. 3, 772–787 (2011).
Chu, W.-L., Tneh, S.-Y. & Ambu, S. A survey of airborne algae and cyanobacteria within the indoor environment of an office building in Kuala Lumpur, Malaysia. Grana 52, 207–220 (2013).
Sharma, N. K. et al. Airborne algae: their present status and relevance. J. Phycol. 43, 615–627 (2007).
Kameyama, S. et al. Application of PTR-MS to an incubation experiment of the marine diatom Thalassiosira pseudonana. Geochem. J. 45, 355–363 (2011).
Achyuthan, K. E. et al. Volatile metabolites emission by in vivo microalgae—an overlooked opportunity. Metabolites 7, 39 (2017).
Rocco, M. et al. Oceanic phytoplankton are a potentially important source of benzenoids to the remote marine atmosphere. Commun. Earth Environ. 2, 175 (2021).
Mussatto, S. I. et al. Technological trends, global market, and challenges of bio-ethanol production. Biotechnol. Adv. 28, 817–830 (2010).
Tesson, S. V. M. & Šantl-Temkiv, T. Ice nucleation activity and Aeolian dispersal success in airborne and aquatic microalgae. Front. Microbiol. 9, 2681 (2018).
Wilson, T. W. et al. A marine biogenic source of atmospheric ice-nucleating particles. Nature 525, 234–238 (2015).
Kvíderová, J., Hájek, J. & Worland, R. M. The ice nucleation activity of extremophilic algae. Cryo Lett. 34, 137–148 (2013).
Broady, P. A. Diversity, distribution and dispersal of Antarctic terrestrial algae. Biodivers. Conserv. 5, 1307–1335 (1996).
Kristiansen, J. Dispersal of freshwater algae—a review. Hydrobiologia 336, 151–157 (1996).
Després, V. et al. Primary biological aerosol particles in the atmosphere: a review. Tellus B: Chem. Phys. Meteorol. 64, 15598 (2012).
Wiśniewska, K., Śliwińska-Wilczewska, S. & Lewandowska, A. The first characterization of airborne cyanobacteria and microalgae in the Adriatic Sea region. PLoS ONE 15, e0238808 (2020).
Dillon, K. P. et al. Cyanobacteria and algae in clouds and rain in the area of puy de Dôme, central France. Appl. Environ. Microbiol. 87, e01850–20 (2021).
Noirmain, F. et al. Interdisciplinary strategy to assess the impact of meteorological variables on the biochemical composition of the rain and the dynamics of a small eutrophic lake under rain forcing. Biogeosciences 19, 5729–5749 (2022).
Williams, O. J., Beckett, R. E. & Maxwell, D. L. Marine phytoplankton preservation with Lugol’s: a comparison of solutions. J. Appl. Phycol. 28, 1705–1712 (2016).
Zetsche, E.-M. & Meysman, F. J. R. Dead or alive? Viability assessment of micro- and mesoplankton. J. Plankton Res. 35, 493–509 (2012).
Similä, A. Spring development of a Chlamydomonas population in Lake Nimetön, a small humic forest lake in southern Finland. Hydrobiologia 161, 149–157 (1988).
Gustafsson, S. & Cronberg, G. Sammanställning Och Utvärdering Av Planktonsamhället i Nio Skånska Sjoöar. Vol. 15, p. 86. (Länsstyrelsen i Skåne län, 2012).
Hällfors, G. Checklist of Baltic Sea Phytoplankton Species. Balt. Sea Environ. Proc. 95, 210 (2004).
Schlichting, H. E. Ejection of microalgae into the air via bursting bubbles. J. Allergy Clin. Immunol. 53, 185–188 (1974).
Ickes, L. et al. The ice-nucleating activity of Arctic sea surface microlayer samples and marine algal cultures. Atmos. Chem. Phys. 20, 11089–11117 (2020).
Perrott, P. et al. Preferential aerosolization of bacteria in bioaerosols generated in vitro. J. Appl. Microbiol. 123, 688–697 (2017).
Walls, P. L. L. et al. Moving with bubbles: a review of the interactions between bubbles and the microorganisms that surround them. Integr. Comp. Biol. 54, 1014–1025 (2014).
Lewis, E.R. & Schwartz, S.E. Geophysical Monograph 152 (eds E.R. Lewis and S.E. Schwartz). (American Geophysical Union, 2004).
Pietsch, R. B. et al. Wind-driven spume droplet production and the transportation of Pseudomonas syringae from aquatic environments. PeerJ 6, e5663 (2018).
Butcher, A. C. Sea Spray Aerosols: Results from Laboratory Experiments, 190 (University of Copenhagen, Faculty of Science, Department of Chemistry, 2013).
Riuz, J., Macías, D. & Peters, F. Turbulence increases the average settling velocity of phytoplankton cells. Proc. Natl Acad. Sci. USA 101, 17720–17724 (2004).
Ross, O. N. Particles in motion: how turbulence affects plankton sedimentation from an oceanic mixed layer. Geophys. Res. Lett. 33, L10609 (2006).
Arguedas-Leiva, J.-A. et al. Elongation enhances encounter rates between phytoplankton in turbulence. Proc. Natl Acad. Sci. USA 119, e2203191119 (2022).
Sengupta, A., Carrara, F. & Stocker, R. Phytoplankton can actively diversify their migration strategy in response to turbulent cues. Nature 543, 555–558 (2017).
Griffin, D. W. et al. Observations on the use of membrane filtration and liquid impingement to collect airborne microorganisms in various atmospheric environments. Aerobiologia 27, 25–35 (2011).
Attard, E. et al. Effects of atmospheric conditions on ice nucleation activity of Pseudomonas. Atmos. Chem. Phys. 12, 10667–10677 (2012).
Keller, M. D. Dimethyl Sulfide Production and Marine Phytoplankton: The Importance of Species Composition and Cell Size. Biol. Oceanogr. 6, 375–382 (1989).
Charlson, R. et al. Oceanic phytoplankton, atmospheric sulphur, cloud albedo and climate. Nature 326, 655–661 (1987).
Ayers, G. P. & Cainey, J. M. The CLAW hypothesis: a review of the major developments. Environ. Chem. 4, 366–374 (2007).
Özçimen, D. & İnan, B. Biofuels - Status and Perspective (ed. Biernat K.). IntechOpen, Chapter 7 https://doi.org/10.5772/59305 (2015).
de Farias Silva, C. E. & Bertucco, A. Bioethanol from microalgae and cyanobacteria: A review and technological outlook. Process. Biochem. 51, 1833–1842 (2016).
Grossman, A. Acclimation of Chlamydomonas reinhardtii to its nutrient environment. Protist 151, 201–224 (2000).
Miazek, K. et al. Effect of organic solvents on microalgae growth, metabolism and industrial bioproduct extraction: a review. Int. J. Mol. Sci. 18, 1429 (2017).
Rudich, Y., Khersonsky, O. & Rosenfeld, D. Treating clouds with a grain of salt. Geophys. Res. Lett. 29, 2060 (2002).
Amin, S. A., Parker, M. S. & Armbrust, E. V. Interactions between diatoms and bacteria. Microbiol. Mol. Biol. Rev. 76, 667 (2012).
D’Souza, N. A. et al. Diatom assemblages promote ice formation in large lakes. ISME J. 7, 1632–1640 (2013).
Hoose, C. & Möhler, O. Heterogeneous ice nucleation on atmospheric aerosols: a review of results from laboratory experiments. Atmos. Chem. Phys. 12, 9817–9854 (2012).
Kanji, Z. A. et al. Overview of ice nucleating particles. Meteorol. Monogr. 58, 1–1.33 (2017).
Wiśniewska, K., Śliwińska-Wilczewska, S. & Lewandowska, A. Airborne microalgal and cyanobacterial diversity and composition during rain events in the southern Baltic Sea region. Sci. Rep. 12, 2029 (2022).
Roy-Ocotla, G. & Carrera, J. Aeroalgae: responses to some aero-biological questions. Grana 32, 48–56 (1993).
Guillard, R. R. L. & Lorenzen, C. J. Yellow-green algae with chlorophyllide c. J. Phycol. 8, 10–14 (1972).
King, S. M. et al. Investigating primary marine aerosol properties: CCN activity of sea salt and mixed inorganic–organic particles. Environ. Sci. Technol. 46, 10405–10412 (2012).
Grinshpun, S. A. et al. Effect of impaction, bounce and reaerosolization on the collection efficiency of impingers. Aerosol. Sci. Technol. 26, 326–342 (1997).
Hillebrand, H. et al. Biovolume calculation for pelagic and benthic microalgae. J. Phycol. 35, 403–424 (1999).
Crippen, R. W. & Perrier, J. L. The use of neutral red and Evans blue for live-dead determinations of marine plankton (with comments on the use of rotenone for inhibition of grazing). Biotech. Histochem. 49, 97–104 (1974).
Ling, M. L. et al. Effects of ice nucleation protein repeat number and oligomerization level on ice nucleation activity. J. Geophys. Res. Atmos. 123, 1802–1810 (2018).
Šantl-Temkiv, T. et al. Characterization of airborne ice-nucleation-active bacteria and bacterial fragments. Atmos. Environ. 109, 105–117 (2015).
Christner, B. C. et al. Ubiquity of biological ice nucleators in snowfall. Science 319, 1214 (2008).
Vali, G. Quantitative evaluation of experimental results on heterogenous freezing nucleation of supercooled liquids. J. Atmos. Sci. 28, 402–409 (1971).
R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2022).
Wickham, H. ggplot2: Elegant Graphics for Data Analysis. (Springer-Verlag New York, 2016).
Wickham, H. & Seidel, D. Scales: Scale Functions for Visualization. https://scales.r-lib.org, https://github.com/r-lib/scales (2022).
Wickham, H. et al. dplyr: A Grammar of Data Manipulation. https://dplyr.tidyverse.org, https://github.com/tidyverse/dplyr (2022).
Murrell, P. R Graphics (Chapman & Hall/CRC Press, 2005).
Amann, R. I. et al. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56, 1919–1925 (1990).
Daims, H. et al. The domain-specific probe EUB338 is insufficient for the detection of all bacteria: development and evaluation of a more comprehensive probe set. Syst. Appl. Microbiol. 22, 434–444 (1999).
Manz, W. et al. Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacteria: problems and solutions. Syst. Appl. Microbiol. 15, 593–600 (1992).
Ogle, D. et al. FSA: Simple Fisheries Stock Assessment Methods. https://github.com/fishR-Core-Team/FSA (2022).
Acknowledgements
The authors thank Merete Bilde for providing access to the aerosolization tank and facility, and feedback on the manuscript. The authors also thank Kasper V Kristiensen and Anders Feilberg for their constructive discussion and access to the PTR-ToF-MS. The authors are grateful to the Aarhus University, Biology department Microbiology and Aquatic Biology units for access to environmental climate chamber and laboratory facilities. The authors thank Corina Wieber for providing an aliquot of Pseudomonas syringae and Jesper Lundsgaard Wulff and Lars Borregaard Pedersen for technical support. This research and S.T. were funded by the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement no. 754513 and The Aarhus University Research Foundation. B.R. was funded by a research grant from Villum Fonden (42128) and Novo Nordisk Foundation (NNF19OC0056963).
Author information
Authors and Affiliations
Contributions
S.V.M.T. contributed to conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, visualization, writing original draft, review, and editing. M.B. contributed to investigation, methodology, and writing review & editing. B.R. contributed to conceptualization, formal analysis, investigation, methodology, visualization, writing review, and editing.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Communications Biology thanks Naveen Sharma, Sylwia Śliwińska-Wilczewska, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Tobias Goris and David Favero.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Tesson, S.V.M., Barbato, M. & Rosati, B. Aerosolization flux, bio-products, and dispersal capacities in the freshwater microalga Limnomonas gaiensis (Chlorophyceae). Commun Biol 6, 809 (2023). https://doi.org/10.1038/s42003-023-05183-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s42003-023-05183-5
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
