New Results

Volatile Organic Compound Specialists in the Phycosphere

Vaishnavi G. Padaki, Xavier Mayali, Peter K. Weber, Stephen J. Giovannoni, Kaylene Abraham, Kerry Jacobs, Lindsay Collart, Kimberly H. Halsey
doi: https://doi.org/10.1101/2024.06.14.599129

Abstract

Labile dissolved organic carbon (LDOC) in the oceans accounts for ∼ ¼ of global photosynthesis and turns over with a half-life of about one day, fueling one of the largest engines of microbial heterotrophic production on the planet. Volatile organic compounds (VOCs) are poorly constrained components of LDOC. We detected 78 VOCs totaling 18.5 μM in cultures of the model diatom Phaeodactylum tricornutum, including hydrocarbons usually found in petroleum. In five individual cocultures with bacteria adapted to grow with this diatom, 1 to 66 VOCs were depleted. Two of the most active VOC consumers, Marinobacter and Roseibium, had more VOC oxidation genes, and attached to the diatom, suggesting VOC specialism. Diatom photosynthesis increased by up to 29% in the presence of VOC consumers, indicating that VOC consumption by heterotrophic bacteria in the phycosphere – a region of rapid organic carbon oxidation that surrounds phytoplankton cells – could impact global rates of primary production.

Main

The phycosphere, the layer of water immediately surrounding an algal cell, is the site of complex interactions between phytoplankton and bacterioplankton, which influence carbon and nutrient cycling in marine ecosystems 1–3. Microbial activities in the phycosphere are distinct from processes occurring in bulk water, a result of bacterial taxa with specialized adaptations populating the region of concentrated chemicals surrounding phytoplankton 4,5. Direct chemical transfer from phytoplankton to bacterioplankton is enhanced by their occupation of the narrow zone of the phycosphere, which allows access to metabolites before they diffuse into the bulk seawater 2,6. The ecology of the phycosphere is similar to its terrestrial analog, the rhizosphere; phylogenetically related bacterial taxa are found across both systems 7, chemotaxis plays a central role in accessing the narrow zone of high chemical concentration 8,9, and the chemicals transferred from the primary and secondary producers are functionally similar 2. For example, primary metabolites, such as sugars, amino acids, and vitamins, as well as metabolite precursors, such as organosulfur compounds (dimethylsulfide (DMS) and dimethylsulfoniopropionate) 10,11, are used as bacterial growth substrates in the phycosphere and rhizosphere 12,13.

Volatile organic compounds (VOCs) are an important subset of the labile dissolved organic compounds (LDOC) produced by phytoplankton, but VOCs are often overlooked in metabolomic studies because most are not captured by common methods of analysis such as solid phase extraction and liquid chromatography. VOCs are relatively low in molecular weight, hydrophobic, and can diffuse across cell membranes. Some VOCs, such as isoprene and DMS, are produced by phytoplankton in response to stress 14–16, while others (e.g., acetaldehyde and methyl iodide) are intermediate chemicals in metabolic pathways 17,18. Due to their diffusive properties, these VOCs may escape biochemical transformation and cross the algal cell membrane into the phycosphere. Other VOCs, including aromatic benzenoids such as benzene, toluene, and ethylbenzene/xylene (collectively known as BTEX), are produced by phytoplankton via the shikimate and non-mevalonate pathways 19,20, but the specific reactions involved in BTEX formation are not known.

Research on VOC transfer between algae and bacteria is sparse but suggests VOCs released by algae can be growth substrates for marine bacteria. A wide range of VOCs were detected during the growth of the diatom, Thalassiosira pseudonana, and a subset of those supported the growth of Pelagibacter ubique HTCC1062, a member of ubiquitous SAR11 bacterioplankton 18,21. These publications suggested that VOCs were “public goods” released by the diatom constitutively and available to any organism able to metabolize them. HTCC1062 metabolizes isoprene, acetone, acetaldehyde, DMS, and cyclohexanol, which are some of the most well-studied VOCs for their roles in sea-air exchange and atmospheric chemistry 18. BTEX are produced by some algae 19, and are growth substrates for a wide variety of bacteria isolated from oilfields 22,23, soils 24, groundwater 25, and wastewater 26. Less is known about BTEX consumption by marine bacteria, which is likely an important sink that limits BTEX emissions into the atmosphere 27. Genetic mechanisms underlying bacterial metabolism of some VOCs (e.g., methanol, DMS, acetone, acetaldehyde) have been elucidated 18,28,29. The widespread expression of some of the genes involved, such as acetone/cyclohexanone monooxygenase, suggests marine bacteria have an important role in mediating VOC sea-air emissions 18. However, the full diversity of VOCs used by bacteria in the oceans is unknown.

The bacterial community in the phycosphere of marine microalgae may be uniquely adapted to VOC metabolism, as is the case for methylotrophs in the phyllosphere 30. The phycosphere of a modest-sized (∼20 μm) diatom extends about 50 – 2000 μm away from the cell2. In this region, chemicals are modeled to be >50% higher than in bulk seawater 2, where VOC concentrations are in the pM to nM range. VOC concentration in the phycosphere is partly controlled by the rate of diffusion from the algal source. The rate of diffusion decreases with increased molecular size and decreased hydrophobicity and likely increases with phytoplankton growth rate and bacterial VOC uptake. Some bacteria occupy the phycosphere through chemotaxis or physical attachment to algal cells 31,32. Chemotactic bacteria experience transient exposures to a dynamic organic chemical cocktail in the phycosphere 12, where the composition depends on the algal producer 1,33, environment 34,35, and antagonistic and mutualistic interactions 12,36,37, requiring bacterial consumers to efficiently shift physiologies to match resource availability. Strategies bacteria use to navigate carbon acquisition as they traverse phycosphere boundaries are not known but may depend on pool sizes and the diversity of regulatory behavior systems for the uptake and metabolism of various substrates 38. Bacteria that can adhere to algae would, in principle, optimize diffusion-mediated VOC uptake by maximizing areal coverage of VOC flux directly from the algal source.

Here, we investigated VOC transfer from the model diatom, Phaeodactylum tricornutum, to phycosphere bacteria previously isolated from a large-scale P. tricornutum production pond. The bacteria varied in taxonomy, DOC uptake, and attachment to the diatom 12,39,40. Results show significant VOC transfer from P. tricornutum to its phycosphere bacteria, fueling primary and secondary production. Extrapolated to larger scales, our findings support the perspective that phycosphere bacteria are a sink for up to 29% of primary production and play an important role in limiting VOC accumulation in the surface ocean.

Results

Phaeodactylum tricornutum produced a wide range of VOCs in culture

During exponential growth of P. tricornutum, 78 m/z signals were detected by proton transfer reaction mass spectrometry (PTR-MS) at concentrations higher than the media control (Benjamini-Hochberg, Q-value < 0.1, n = 6). VOCs accumulated to a total of 85.1 pM cell-1 and represented nine chemical functional groups (Fig. 1). Hydrocarbons represented about 60% of the VOCs produced in the exponential phase (54.2 pM cell-1; Supp. table 1). The benzenoid class of aromatic hydrocarbons was most abundantly produced, including BTEX compounds (m/z 79.11, m/z 93.07, and m/z 107.08), respectively 19, and C H + (m/z 149.13). Note that in PTR-MS, VOCs are detected at their protonated ion masses. Alcohols were the second most abundantly produced group of VOCs, contributing up to 11% of the total VOC pool (10.20 pM cell-1). Methanol (m/z 33.03) and, likely, ethanol (m/z 47.04) made up the vast majority of the accumulated alcohols (Fig.1, Supp. table 1). Consistent with reports that P. tricornutum is a halocarbon producer 41,42, m/z 124.96, corresponding to ethyl bromide, was detected in the exponential phase. The well-studied VOCs acetone (m/z 59.05), acetaldehyde (m/z 45.03), and DMS (m/z 68.07) were also detected in the exponential phase (Fig. 1, Supp. table 1).

Figure 1:
Figure 1:

VOC production in P. tricornutum. a. Cell-normalized concentrations of m/z values (rows) ranging from low (light blue) to high (orange) concentrations produced during exponential and stationary phases. Black are m/z values that were not significantly higher than the media control. Functional group classifications for each m/z+1 value are shown in the rightmost column, b. P. tricornutum growth (n=6 independent cultures). Arrows indicate sampling days for VOCs.

After P. tricornutum entered the stationary growth phase, 58 m/z signals were detected in concentrations higher than the media control (Benjamini-Hochberg, Q-value < 0.1, n = 6) and accumulated to 9.01 pM cell-1 (Fig. 1), which was only about 10% of the VOC accumulation during the exponential phase (Fig. 1, Supp. table 1). VOCs detected in both phases of growth were consistently lower in concentration in the stationary phase. Slower VOC production in response to cessation of growth, and loss of VOCs from the vented growth flasks, are the most likely explanations for the concentration differences between growth phases. Hydrocarbons still represented the majority of the VOCs (84%) but were only 7.59 pM cell-1. Sixteen of the 58 m/z signals were unique to the stationary phase, but many of these had unknown identities.

The most abundant VOC in both exponential and stationary phases was C H + (m/z 149.13) in the benzenoid superclass and corresponds to dictyopterene, an algal sesquiterpenoid pheromone. Chemical isomers of dictyopterene include the aromatic hydrocarbons, neopentyl-benzene, 4-tert-butyl-toluene, and 1-ethyl-2-propylbenzene, which have been identified in plant leaves 43. Forty-two m/z signals, including BTEX, were detected in both exponential and stationary phases (Fig. 1).

VOCs were depleted in P. tricornutum-bacteria cocultures

Five P. tricornutum-bacteria cocultures were grown to identify VOCs produced by the diatom that decreased in concentration in the presence of a bacterium, which would suggest those VOCs were consumed by the bacterium. Alternatively, decreased VOC concentrations in a coculture could indicate VOC production was inhibited in the presence of the bacterium. The five P. tricornutum-bacteria cocultures (herein “PT-bacteria genus name”) each showed distinct VOC depletion patterns during exponential growth when compared to axenic P. tricornutum (Fig. 2a, Supp. Fig. 1; PERMANOVA, p-value = 0.03, n = 6). Of the 78 m/z signals produced by P. tricornutum in exponential phase, 31 to 59 were significantly depleted in four of the five cocultures (Benjamini-Hochberg, Q-value cut-off ≤ 0.1, n = 6) (Fig. 2a). VOCs were most strongly depleted in PT-Marinobacter (7.09 pM cell-1), followed by PT-Stappia (6.51 pM cell-1), PT-Roseibium (5.52 pM cell-1) and PT-Rhodophyticola (4.54 pM cell-1) (Fig. 2a, Fig. 4, Supp. table 1). About half the number of m/z signals were depleted in the PT-Roseibium compared to PT-Marinobacter. Marinobacter and Roseibium were observed to be attached to P. tricornutum (Fig. 3) 39, thus the cell-normalized VOC depletion data in PT-Marinobacter and PT-Roseibium are based on estimates of bacteria cell densities using flow cytometry for the free-living and scanning electron microscopy for the attached bacteria (see Methods). In PT-Yoonia, only one m/z signal, 45.03, corresponding to acetaldehyde, was significantly depleted (0.5 pM cell-1 (Fig. 2a)). There were 14 m/z signals that did not show significant depletion in any of the five PT-bacteria cocultures. A lack of depletion could mean that the bacteria could not take up or metabolize those VOCs, or those VOCs were maintained at a steady state concentration by equal rates of production and consumption in the coculture. Of those, 8 m/z signals belonged to the nitrogen-containing class of VOCs (Supp. table 6). Furthermore, neither methanol nor acetone were depleted in any of the five cocultures.

Supp. fig. 1:
Supp. fig. 1:

Principal coordinate analysis of depleted VOCs in P tricornutum-bacteria cocultures in exponential (left) and stationary (right) growth phases.

Figure 2:
Figure 2:

Phycosphere bacteria demonstrate widely ranging VOC depletion patterns, a. m/z+1 depletion (rows) in each P. tricornutum-bacteria coculture (columns are labeled with the bacteria grown with P. tricornutum) in exponential (a) and stationary phase (b) relative to axenic R tricornutum in each of the growth phases. The magnitude of depletion is shown ranging from low (light blue) to high (yellow). Pearson correlation distances were used to cluster depletion patterns between PT-bacteria cocultures (top axes). Pie charts show the proportion of the total depleted VOC concentration in each functional group.(c) Growth of each bacteria (red lines) in coculture with P. tricornutum (blue lines), (n=6). Arrows indicate sampling day for VOCs. d. Genetic potential for hydrocarbon metabolism in the phycosphere bacteria. Number of gene copies is shown for each bacteria. Abbreviations: alkB (alkane-1 -monooxygenase), bnzA (benzene dioxygenase alpha-subunit), bnzB (benzene dioxygenase beta-subunit), almA (flavin-binding monooxygenase), aor (aldehyde oxidoreductase), acmA (acetone monooxygenase).

Figure 3.
Figure 3.

Scanning electron microscopy images showing (a) Marinobacter attached to P. tricronutum in a flat orientation and (b) Roseibium attached to P. tricornutum in polar orientation.

Figure 4:
Figure 4:

Photosynthetic carbon fixation in P. tricornutum in the presence and absence of VOC sinks (hydrocarbon trap or each bacteria in coculture). Points are short term (10 min) carbon fixation rates for each culture or coculture. Box plots show the mean and upper and lower quartile divisions; whiskers show 95% confidence limits. *p-value < 0.01; t-test. Inset shows the total bacterial cell-normalized VOC depletion in each P. tricornutum-bacteria coculture.

Hydrocarbons were the most highly depleted group of VOCs in all PT-bacteria cocultures except PT-Yoonia. Depending on the bacterium present, 11-19 of the depleted m/z signals were classified as hydrocarbons and made up to 70% of the total VOC pool depleted in the cocultures. PT-Marinobacter depleted the most hydrocarbons (5.95 pM cell-1), followed by PT-Stappia (5.44 pM cell -1), PT-Roseibium (4.48 pM cell-1) and PT-Rhodophyticola (3.94 pM cell -1) (Fig. 2a). Monoaromatic hydrocarbons, including BTEX, C H +, and phenylacetylene (m/z 103.05), were the most strongly depleted hydrocarbons in the PT-bacteria cocultures (Fig. 2a).

In stationary growth phase, the range of depleted VOCs in the PT-bacteria cocultures was lower, from 0.64-4.97 pM cell-1 (average 2.96 ± 0.66 pM cell-1). The depletion patterns between PT-bacteria cocultures in stationary phase were not distinct from one another (Fig. 2b, PERMANOVA p-value = 0.22, n = 6). In total, 24 to 55 m/z signals were depleted in the PT-bacteria cocultures (Benjamini-Hochberg, Q-value < 0.1, n = 6). The number of m/z signals depleted in PT-Yoonia expanded from 1 in the exponential phase to 24 in the stationary phase, but the amounts of the individual VOCs depleted were low (Fig. 2b, Supp. Table 2 and 3). In stationary phase, 9-16 m/z signals depleted in PT-bacteria cocultures were classified as hydrocarbons, and these VOCs were the majority of the depleted VOC pools (80-92%). VOCs classified as carboxylic acids were the second-most depleted group, followed by ketones (Fig. 2a and b). Except for hydrocarbons, the magnitude of VOC depletion in PT-bacteria cocultures did not mirror VOC abundances in axenic P. tricornutum in either growth phase (Fig. 1, Supp. tables 1, 2, and 3).

VOC transfer to bacteria comprised up to 30% of primary production

P. tricornutum growth rates, chlorophyll content, and photosynthetic efficiencies were unaffected by the presence of any of the bacterial species in the PT-bacteria cocultures (Supp. Table 4). To assess the impact of bacterial VOC uptake on P. tricornutum primary production, short-term 14CO2-uptake rates, approximating gross carbon production (GCP) were measured. GCP in PT-bacterial co-cultures were higher in PT-Marinobacter, PT-Stappia, and PT-Rhodophyticola than in axenic P. tricornutum (Fig. 4). A hydrocarbon trap that stripped VOCs from the axenic P. tricornutum culture, simulating a highly efficient bacterial VOC sink, caused GCP to increase 24.9% compared to axenic P. tricornutum with no trap (Fig. 4; p = 0.025, n = 6). GCP in PT-Marinobacter, PT-Rhodophyticola, and PT-Stappia were 20.1% to 29.3% higher than in axenic P. tricornutum, similar to the hydrocarbon trap-control, and corresponding to 0.27 to 0.41 nmol C cell-1 h-1 transferred to the bacteria in the form of VOCs (p-values < 0.02, t-tests, n=3) (Fig. 4). Consistent with the minimal amount of VOC depleted in PT-Yoonia there was no significant difference in CO2-uptake rates between PT-Yoonia and axenic P. tricornutum (p-values 0.23, t-test, n=3). Interestingly, GCP was not stimulated in PT-Roseibium until Roseibium entered the second exponential growth phase in its diauxic growth pattern (Fig. 2; note lag between days 2 and 4). This result suggests that Roseibium shifted carbon uptake from primarily LDOC early in culture growth to primarily VOCs later in culture growth.

VOC depletion correlates with genome content for hydrocarbon metabolism

Of the five bacteria tested, Marinobacter and Roseibium genomes harbor the greatest range of hydrocarbon metabolism genes. For example, genes that initiate hydrocarbon oxidation were identified in Marinobacter and Roseibium, including rhdA, encoding ring hydroxylating dehydrogenase, almA, encoding flavin-binding monooxygenase involved in long-chain alkane degradation, and alkB, encoding alkane monooxygenase. rhdA is in a superclass of nonheme iron enzymes that convert aromatic hydrocarbons to dihydrodiols in the presence of dioxygen and NADH 44. The broad substrate range for rhdA includes benzene, and the benzene-alkyl substituted hydrocarbons toluene, ethylbenzene, and xylene, consistent with BTEX depletion profiles in Marinobacter and Roseibium. Benzene dioxygenases are a subclass of rhdA that oxidize benzene and toluene to their corresponding dihydrodiols. We manually curated the rhdA hits for bnzA, encoding the α-subunit of the benzene 1,2 dioxygenase hydroxylase based on the conservation of the active site motif Cys-X1-His-15-to-17 aa-Cys-X2-His (where X is any amino acid) 45. bnzA was identified in Marinobacter, Rhodophyticola, and Roseibium (Fig. 2d). Marinobacter also encodes bnzB, the β-subunit of benzene 1,2 dioxygenase hydroxylase (Fig. 2d). The presence of both bnzA and bnzB subunits increases hydroxylation activity compared to the activity of either of the subunits alone 46. Genes facilitating hydrocarbon oxidation in Stappia were not identified. Acetone/cyclohexanone monooxygenase (acmA) was identified in Marinobacter, Roseibium, and Stappia. The broad substrate range of acmA likely includes cyclic ketones, such as valerophenone (m/z 163.11) and E-jasmone (m/z 165.13), corresponding to m/z signals that were depleted in the PT-Marinobacter, PT-Roseibium, and PT-Stappia. All five bacteria encode one or two copies of aldehyde dehydrogenase (aor), consistent with acetaldehyde (m/z 45.03) depletion in all five cocultures (Fig. 2).

Marinobacter and Rhodophyticola exhibit distinct benzene metabolism

We selected benzene as a representative hydrocarbon to directly test incorporation into biomass by Marinobacter and Rhodophyticola. These bacteria differed in the amounts of benzene depleted in the PT-bacteria cocultures, in their genetic capacities for hydrocarbon metabolism and chemotaxis, in their attachment physiologies (Fig. 2, 3, 5) 39, and could be grown without P. tricornutum in media prepared from filtrate collected from P. tricornutum in exponential growth (‘PTspent’; see methods). Marinobacter and Rhodophyticola grew to higher abundances in PT spent medium with added benzene vs. no added benzene (Fig. 5).

Figure 5:
Figure 5:

Benzene incorporation into Marinobacter and Rhodophyticola biomass (a,c) Bacterial cell densities in the presence (solid lines) and absence (dashed lines) of 13C benzene in ASW (blue), Ptspent (olive), and coculture with P. tricornutum (orange) Asterisks indicate treatments with significant increases in cell densities in the presence of benzene, (b, d) 13C enrichment in Marinobacter and Rhodophyticola in ASW (Marinobacter only), Ptspent, and in coculture with P. tricornutum. Cnet is the fraction of biomass labelled from 13C benzene, (e) Linear regression of bacterial 13C benzene enrichment in PT-Marinobacter (top) and PT-Rhodophyticola (bottom). Student t-test *p-value < 0.05. (f) Expanded y-axis showing bacterial Cnet values in cocultures and inset SEM images of Marinobacter attached to P. tricornutum and Rhodophyticola unattached.

Benzene uptake and subsequent incorporation into biomass by Marinobacter and Rhodophyticola were quantified using nanoSIMS. We incubated each bacterium with and without 13C benzene (60 µM in Marinobacter cultures, 36 µM in Rhodophyticola cultures) and 15N leucine (50 nM) in f/2+Si artificial seawater medium (ASW), PT spent medium, and in coculture with P. tricornutum. Marinobacter increased its growth in ASW with benzene added as the sole source of carbon (Fig. 5a). Associated Cnet values of incorporation, measuring the fraction of a cell’s C originating from benzene, increased over the incubation period, reaching 0.03 to 0.15 (average 0.065) on day 7 (Fig. 5b). Marinobacter reached higher cell densities in PTspent with added benzene compared to PTspent with no benzene or ASW with benzene (Fig. 5a (p-value <0.05, n = 3). Consistent with benzene-stimulated growth in PTspent, Marinobacter Cnet values in that treatment ranged from 0.02 to 0.17, averaging 0.07 on day 4. Marinobacter cell densities were unaffected by benzene addition to PT-Marinobacter. Surprisingly, Marinobacter did not incorporate 13C benzene in the presence of live P. tricornutum cells. Marinobacter Cnet values in the coculture hovered around zero across the 10-day incubation, even in the presence of 60 µM externally-supplied benzene (Fig. 5a,b). Thus, Marinobacter did not incorporate measurable quantities of 13C-benzene in the presence of live P. tricornutum cells. Nnet values from leucine in all Marinobacter treatments, including the coculture, were low but significant (<0.01, Supp. Fig. 5), confirming bacterial activity.

Despite significantly higher Rhodophyticola cell densities in PTspent with 36 µM benzene added compared to the no benzene-added control (Fig. 5c). 13C-benzene addition to PTspent resulted in low Cnet values, and only 2% of the population showed enrichment in 13C benzene on day 6 (Fig. 5d). Rhodophyticola may take up and oxidize benzene for energy rather than biomass generation. No significant difference was observed in Rhodophyticola cell densities grown in the PT coculture with or without added 13C-benzene. Similar to Marinobacter, Cnet values in Rhodophyticola grown in the PT coculture were very low. Rhodophyticola showed significant 15N enrichment from leucine in both PTspent and PT-Rhodophyticola treatments (Supp. Fig. 5).

Discussion

Phycospheres are recognized as energy-rich regions of heterotrophic activity where microbes adapted to this niche benefit from proximity to high concentrations of the organic carbon resources produced by phytoplankton cells. VOCs are rapidly diffusing molecules that easily traverse cell membranes and, due to the introduction of new measurement technologies, are increasingly recognized as complex and important products of phytoplankton metabolism. Here, we show that the phycosphere concept applies to VOCs and that they appear to be an important component of photosynthetic carbon transfer to heterotrophic microbial populations in the region immediately surrounding phytoplankton cells. We also show that some phycosphere bacteria have traits that can be interpreted as evidence of VOC specialism, enabling them to access these chemicals whose concentrations are expected to rapidly decrease with distance from their source.

VOCs were primarily produced during active P. tricornutum growth, when over a quarter of photosynthetically fixed carbon was emitted from P. tricornutum as VOCs. Hydrocarbons were 64% of the total VOC produced, with BTEX accounting for 13% of the emitted VOC pool, accumulating to 11.8 μM. BTEX production by phytoplankton has been proposed to be of sufficient magnitude to influence atmospheric chemistry 19, unless bacterial VOC consumption prevents BTEX accumulation in the surface ocean 27. The VOC concentrations we measured could be underestimated because the cultures were grown in vented flasks. However, in the absence of bacterial sinks, internal VOC concentrations in axenic P. tricornutum cells may reach equilibria with the surrounding medium.

We measured concentrations of VOCs in diatom-bacteria cocultures relative to diatoms alone to investigate the potential for VOC uptake by bacteria. In the case of benzene, isotopic labeling confirmed that the bacteria were consumers, but for other VOCs, we cannot eliminate the possibility that VOC depletion in coculture was caused by inhibition of VOC production by P. tricornutum. P. tricornutum photosynthesis (GCP) was stimulated in cocultures where depletion of a wide range of VOCs was also observed. Some marine bacteria are known to metabolize VOCs, including BTEX 21,47, and we show that the bacteria that caused VOC depletion also harbor genes for some of the depleted compounds; for other depleted compounds, genetic loci associated with consumption are unknown and cannot be ruled out.

The data we report here indicate a range of VOC specialism (Table 1). Four of the five bacterial strains harbor multiple hydrocarbon oxidation genes that encode enzymes with broad substrate ranges. Marinobacter and Roseibium, which consumed large amounts of hydrocarbons, had the most genes for hydrocarbon oxidation. Marinobacter and Roseibium encode few genes for complex polysaccharide metabolism, and in previous research, incorporated low amounts of macromolecular DOC into biomass 39. In our experiments, Rhodophyticola and Stappia, which harbor fewer hydrocarbon oxidation genes, depleted smaller amounts of individual hydrocarbons but a wide range. Mayali et al. reported that Stappia, Rhodophyticola, and Yoonia used complex DOC in the 2023 study, in agreement with their genetic capacities for oligosaccharide, starch, and cellulose degradation. Thus, for this set of traits, among a set of five heterotrophic isolates from P. tricornutum ponds, two strains appeared to be VOC specialists: they used more VOCs and less carbohydrate DOM, and as discussed below, they were motile, and attached to P. tricornutum. Two strains used lesser amounts of VOCs, are active metabolizers of carbohydrate DOM, are not motile, and did not attach; and one strain used carbohydrate DOM, only a single VOC, and attached.

View this table:
Table 1: Traits of bacteria used in this study

We looked for evidence of phycosphere occupation to understand whether, in the strains we studied, it was linked to VOC use. The VOC specialists, Marinobacter and Roseibium, and one strain that did not use VOCs, Yoonia, attached to P. tricornutum, but the other strains did not. Marinobacter attached to the diatom in a flat orientation, while Roseibium attached to P. tricornutum using polar adhesion (Supp. Fig. 2). It has been reported that chemotaxis and motility are key traits of hydrocarbon-degrading bacteria. Volatile aromatic hydrocarbons, including toluene and benzene derivatives, have been shown in previous work to be effective bacterial chemoattractants, even in the presence of complex organic matter, such as petroleum, crude oil, and diesel 48–50. Marinobacter and Roseibium encode canonical chemotaxis che genes 39, but chemoattractant compounds have not been identified in these bacteria 8,51 (Table 1).

Supp. fig. 2:
Supp. fig. 2:

Final cell densities measured in benzene dose experiments. Each bacterium was inoculated into Ptspent (filled symbols) and ASW media (open symbols) with initial benzene concentrations 0-180 µM. Roseibium (circles), Marinobacter (triangles), and Rhodophytico/a (squares).

Supp. fig. 3:
Supp. fig. 3:

Cnet vs Nnet in Marinobacter and Rhodophytico/a measured in three different treatments.

One of the most surprising results in this study was the absence of 13C-benzene cellular incorporation in Marinobacter grown in coculture with P. tricornutum, despite strong 13C-benzene incorporation in Marinobacter grown in the absence of P. tricornutum (i.e., in PTspent and ASW) and significant 12C-benzene depletion in PT-Marninobacter. We repeated the experiment and confirmed the results. These unexpected results suggest that in coculture with P. tricornutum, Marinobacter obtained benzene, and possibly other hydrocarbons, directly from the diatom rather than from the bulk medium. Marinobacter and Roseibium were sometimes observed attached to P. tricornutum. Bacterial attachment or entry into the phycosphere 8 could facilitate efficient VOC transfer directly from the diatom to the bacteria, before the chemicals diffuse into the bulk medium. In coculture, direct VOC uptake from the diatom to Marinobacter may occur via their physical interaction, enabling passive VOC diffusion across the diatom and bacterial membranes. However, FadL outer membrane transporters can boost BTEX metabolism in some Gram-negative hydrocarbon degraders 52, and the Marinobacter strain used in this study encodes a FadL protein with 72% nucleotide identity to the well-characterized FadL in Escherichia coli. In the presence of the diatom, Marinobacter may regulate FadL to optimize use of a wide range of substrates delivered from P. tricornutum. However, FadL substrate specificity and kinetics are not yet known, and FadL may become saturated, inhibiting BTEX uptake (i.e., added 13C-benzene in our experiment) from the bulk medium. Bacteria also exhibit varying preferences for BTEX substrates 53 and the benzenoid collection of hydrocarbons. Benzene may be of lower value to Marinobacter compared to the full collection of VOC substrates from growing P. tricornutum, or benzene may be oxidized to CO2 in the coculture. Nevertheless, the contrasting 13C labeling results in Marinobacter suggest attachment dynamics in the P. tricornutum phycosphere may have an important role in VOC uptake.

In the model set of phytoplankton and associated bacteria we investigated, phycosphere resources were exploited by bacteria employing varying strategies: VOC specialists (Marinobacter and Roseibium) that were motile and attached, generalists capable of VOC and macromolecular uptake (Rhodophyticola and Stappia), and macromolecular specialists (Yoonia). We propose that VOC specialists use chemotaxis to find and colonize the phycosphere, where proximity to sources of low molecular weight, rapidly diffusing VOCs offers advantages 8. Unknown is the duration of attachment or how hydrocarbon uptake is differentially regulated in the presence and absence of P. tricornutum. The free-living generalists, Rhodophyticola and Stappia, used VOCs and carbohydrate DOC and had fewer hydrocarbon oxidation genes.

Collectively, VOC use by phycosphere bacteria was surprisingly broad in the range of compounds used and accounted for up to 29% of primary production by P. tricornutum. These findings add support to previous reports of VOC uptake by bacteria and stimulation of photosynthesis in phytoplankton co-cultured with VOC oxidizers. Previous reports show that balanced production and consumption maintain low VOC concentrations in the oceans and have associated VOC consumption with specialized metabolism in abundant, streamlined, non-motile cells 54–57 but in the set of bacteria we studied some of the most active VOC utilizers were motile and attached to the phytoplankton. Here, we show that the flux can be intercepted by VOC specialists in the phycosphere, reducing ocean VOC accumulation and air-sea transfer. If the proposed model for interpreting these results, that heterotroph consumption stimulates photosynthesis because phytoplankton replenish lost VOC resources, is correct, then it is conceivable that the mechanisms we described result in a global increase in ocean photosynthesis and are an important conduit of direct transfer of GCP to bacteria that does not rely on the influence of protistan grazing and virus predation.

Materials and methods

Culture growth and P. tricornutum physiology

Phaeodactylum tricornutum strain CCMP 2561 was maintained axenically and in separate co-cultures with Roseibium sp. 13C1, Yoonia sp. 4BL, Marinobacter sp. 3-2, Rhodophyticola sp. 6CLA, and Stappia sp. ARW1T. Cultures were grown under 12h:12h light: dark cycles (60-70 µmole photons m-2 s-1), at 19°C in f/2+Si medium (ASW) 58. DAPI (NucBlue, Thermo Fischer) and fluorescence microscopy were used to check P. tricornutum axenicity. Bacterial and diatom growth were determined in triplicate. Cell densities were measured using a GUAVA flow cytometer (Millipore; Billerica, MA, USA) 1-2 h into the light phase. VOCs, chlorophyll content (Chla), and photosynthetic efficiency (Fv/Fm) were measured 4-6 h into the light phase. Chla was determined from filtering 2-5 ml of culture (GF/F, Whatman, 25 mm), extracting in 5 ml 90% acetone, and storing at −20° C for 24 h. Chla was quantified by spectrophotometer (Shimadzu; Kyoto, Japan) 59. Fv/Fm was measured by fast repetition-rate fluorometer following 10 min dark acclimation 60. Short-term carbon fixation rates (approximating gross carbon production) were measured as in Moore et al. (2020). Axenic P. tricornutum was grown in a closed system with culture headspace recirculating at 80 ml min-1 by peristaltic pump and either directed back into the culture medium or passed first through a hydrocarbon trap (Marineland Black Diamond Activated Carbon) and then into the medium. P. tricornutum-bacteria cocultures were grown in the same manner with no hydrocarbon trap. All flasks were recirculated for 24 h prior to inoculation with the cultures to remove background VOCs. Samples for 14CO2-uptake (10 ml) were collected by syringe during the P. tricornutum exponential phase (2-4 × 105 cells ml-1), spiked with 2 μCi of 14C-sodium bicarbonate and incubated 20 min in light or dark.

VOC measurement by PTR-TOF/MS

VOCs were quantitatively measured in axenic P. tricornutum and cocultures by PTR-TOF/MS (Ionicon, Austria), as in Moore et al. (2020). Autoclaved ASW and HPLC-grade water were measured alongside the samples as blank controls. Culture, ASW, or HPLC water (100 ml) was transferred to a dynamic stripping chamber 56and bubbled with breathing-grade air passed first through a hydrocarbon trap and then through a glass frit in the bottom of the chamber at 50 ml min-1 for 5 min to strip VOCs from the sample. PTR-TOF/MS data were analyzed using PTR viewer 3.4.3 (Ionicon Analytik). Calibration and binning of m/z signals were performed as in Moore et al. (2020). By using correlation plots and GLOVOCS 43, contaminants, fragments, duplicates, and water clusters were removed; m/z signals were putatively identified using GLOVOCS and Ionicon Viewer 3.4.1.

Benzene uptake and nanoSIMS analyses

Benzene dose experiments were performed on Rhodophyticola, Marinobacter, and Roseibium. Stappia could not be reliably grown in the absence of P. tricornutum, and Yoonia showed no depletion of benzene in coculture. A saturated benzene solution was prepared in ASW and added to P. tricornutum-spent media (prepared from 0.2 µm filtered spent media collected from axenic P. tricornutum grown to 2-4 × 105 cells ml-1 and diluted in 1:10 with ASW) to achieve benzene concentrations 0-180 µM with no headspace. Cultures were maintained at 19°C in the dark.

13C-benzene uptake was evaluated using nanoSIMS in six different treatments grown in Teflon-sealed borosilicate vials (Marinobacter) or airtight Nalgene bottles (Rhodophyticola). Fifty nM 98% 15N leucine (Cambridge Isotopes, MA, USA) was added to all treatments as an independent measure of cell activity (see Supp. table 9). Sample preparation for nanoSIMS and a brief explanation of isotopic measurement are explained elsewhere (Supp. document 4). A CAMECA NanoSIMS 50 was used to measure the isotopic composition of individual cells, explained in detail elsewhere 61.

SEM imaging

Samples (5 ml) of Marinobacter and Rhodophyticola cocultured with P. tricornutum were collected in mid-exponential phase, treated with 1.25 ml of 2X 5% glutaraldehyde + 2% paraformaldehyde (PFA) in 0.1 M sodium cocadylate buffer (NaCB), incubated for 24 h, filtered on a 0.2 µm polycarbonate filter, and washed three times with NaCB + Milli-Q water. The filtered cells were subjected to critical point drying and then vacuum-dried and gold-coated. Imaging was done using a detector with a 5.1 mm working distance, 1.8 kV voltage, and 20-30 µm aperture diameter at the Electron Microscope facility, LPSC, Oregon State University. The number of bacteria adhered to P. tricornutum in coculture was estimated from 50 diatom cells.

Genetic potential for hydrocarbon metabolism

Whole genome sequences of each bacteria were acquired from NCBI (GB accession numbers: Marinobacter: PRJNA500125; Rhodophyticola: PRJNA441682; Yoonia: PRJNA441685; Stappia: PRJNA441689; Roseibium: PRJNA441686). Geneious Prime v2023.2.1 and Hidden Markov Models were used to search genomes for proteins initiating hydrocarbon oxidation (i.e., alkB, Pfam(PF00487); almA, Pfam(PF00743); rhdA, Pfam(PF00848); acmA, Pfam(PF00743); aldH, Pfam(PF00171) 62. Each hit for alkB and rhdA was manually curated using Geneious Prime v2023.2.1. Each hit was aligned using MUSCLE protein alignment in Geneious prime v2023.2.1.

Statistics

Statistical analyses and figure construction were done in R studio v.4.1.1, with ggplot2 and Complex heatmap packages 63,64. Paired student t-tests were used to identify significant differences between treatments. Benjamini-Hochberg corrections (Q-value cut-off ≤ 0.1) eliminated false positive m/z signals. Principle component analysis and Permutational Multivariate Analysis of Variance (PERMANOVA) were used to evaluate VOC accumulation and depletion patterns between cultures.

Author contributions

K.H.H., X.M., and V.G.P. conceived of the study. V.G.P. conducted all experiments with assistance from K.A., K.J., and L.C. and wrote the first manuscript draft with K.H.H. NanoSIMS analyses were conducted by X.M. and P.K.W. All authors contributed to final version of the manuscript.

Acknowledgments

We thank Christina Ramon for lab assistance and Nicholas Baetge and Chih Ping Lee for assistance with data analysis. This research was supported by award ID OCE1948163 to K.H.H. and S.J.G. from the National Science Foundation, Biological Oceanography Program. Work at LLNL was Supported by the US Department of Energy’s (DOE) Genomic Science Program through the LLNL μBioSpheres Science Focus Area grant # SCW1039 and performed under the auspices of the DOE under contract DE-AC52-07NA27344.

Footnotes

  • Figure 2 revised: Due to spelling error and missing keys.

References

  1. 1.↵
    Fu, H., Uchimiya, M., Gore, J. & Moran, M. A. Ecological drivers of bacterial community assembly in synthetic phycospheres. Proc Natl Acad Sci U S A 117, 3656–3662 (2020).
    OpenUrlAbstract/FREE Full Text
  2. 2.↵
    Seymour, J. R., Amin, S. A., Raina, J.-B. & Stocker, R. Zooming in on the phycosphere: the ecological interface for phytoplankton–bacteria relationships. Nat Microbiol 2, 17065 (2017).
    OpenUrl
  3. 3.↵
    Mitchell, J. G., Okubo, A. & Fuhrman, J. A. Microzones surrounding phytoplankton form the basis for a stratified marine microbial ecosystem. Nature 1985 316:6023 316, 58–59 (1985).
    OpenUrlCrossRefWeb of Science
  4. 4.↵
    Seymour, J. R., Ahmed, T., Durham, W. M. & Stocker, R. Chemotactic response of marine bacteria to the extracellular products of Synechococcus and Prochlorococcus. Aquatic Microbial Ecology 59, 161–168 (2010).
    OpenUrl
  5. 5.↵
    Smriga, S., Fernandez, V. I., Mitchell, J. G. & Stocker, R. Chemotaxis toward phytoplankton drives organic matter partitioning among marine bacteria. Proc Natl Acad Sci U S A 113, 1576–1581 (2016).
    OpenUrlAbstract/FREE Full Text
  6. 6.↵
    Kogure, K., Simidu, U. & Taga, N. Bacterial attachment to phytoplankton in sea water. J Exp Mar Biol Ecol 56, 197–204 (1981).
    OpenUrlCrossRef
  7. 7.↵
    Ramanan, R. et al. Phycosphere bacterial diversity in green algae reveals an apparent similarity across habitats. Algal Res 8, 140–144 (2015).
    OpenUrl
  8. 8.↵
    Raina, J. B. et al. Chemotaxis increases metabolic exchanges between marine picophytoplankton and heterotrophic bacteria. Nature Microbiology 2023 8:3 8, 510–521 (2023).
    OpenUrl
  9. 9.↵
    Ugolini, G. S. et al. Microfluidic approaches in microbial ecology. Lab Chip 24, 1394–1418 (2024).
    OpenUrl
  10. 10.↵
    Durham, B. P. et al. Cryptic carbon and sulfur cycling between surface ocean plankton. Proc Natl Acad Sci U S A 112, 453–457 (2015).
    OpenUrlAbstract/FREE Full Text
  11. 11.↵
    Nouchi, I., Hosono, T. & Sasaki, K. Seasonal changes in fluxes of methane and volatile sulfur compounds from rice paddies and their concentrations in soil water. Plant Soil 195, 233–245 (1997).
    OpenUrl
  12. 12.↵
    Brisson, V., et al. Identification of Effector Metabolites Using Exometabolite Profiling of Diverse Microalgae. mSystems 6, (2021).
  13. 13.↵
    Russo, A., Pollastri, S., Ruocco, M., Monti, M. M. & Loreto, F. Volatile organic compounds in the interaction between plants and beneficial microorganisms. J Plant Interact 17, 840–852 (2022).
    OpenUrl
  14. 14.↵
    Dawson, R. A. et al. The microbiology of isoprene cycling in aquatic ecosystems. Aquatic Microbial Ecology 87, 79–98 (2021).
    OpenUrl
  15. 15.
    Shemi, A., Ben-Dor, S., Rotkopf, R., Dym, O. & Vardi, A. Phylogeny and biogeography of the algal DMS-releasing enzyme in the global ocean. ISME Communications 3, (2023).
  16. 16.↵
    Asher, E., Dacey, J. W., Ianson, D., Peña, A. & Tortell, P. D. Concentrations and cycling of DMS, DMSP, and DMSO in coastal and offshore waters of the Subarctic Pacific during summer, 2010-2011. J Geophys Res Oceans 122, 3269–3286 (2017).
    OpenUrl
  17. 17.↵
    Pozzer, A. C., Gómez, P. A. & Weiss, J. Volatile organic compounds in aquatic ecosystems – Detection, origin, significance and applications. Science of the Total Environment 838, (2022).
  18. 18.↵
    Moore, E. R., Weaver, A. J., Davis, E. W., Giovannoni, S. J. & Halsey, K. H. Metabolism of key atmospheric volatile organic compounds by the marine heterotrophic bacterium Pelagibacter HTCC1062 (SAR11). Environ Microbiol 24, 212–222 (2022).
    OpenUrl
  19. 19.↵
    Rocco, M. et al. Oceanic phytoplankton are a potentially important source of benzenoids to the remote marine atmosphere. Communications Earth & Environment 2021 2:1 2, 1–8 (2021).
    OpenUrlCrossRef
  20. 20.↵
    Misztal, P. K. et al. Atmospheric benzenoid emissions from plants rival those from fossil fuels. Scientific Reports 2015 5:1 5, 1–10 (2015).
    OpenUrl
  21. 21.↵
    Moore, E. R., Davie-Martin, C. L., Giovannoni, S. J. & Halsey, K. H. Pelagibacter metabolism of diatom-derived volatile organic compounds imposes an energetic tax on photosynthetic carbon fixation. Environ Microbiol 22, 1720–1733 (2020).
    OpenUrl
  22. 22.↵
    Wu, H. jun et al. Metagenomic analysis reveals specific BTEX degrading microorganisms of a bacterial consortium. AMB Express 13, (2023).
  23. 23.↵
    Bacosa, H. P., Mabuhay-Omar, J. A., Balisco, R. A. T., Omar, D. M. & Inoue, C. Biodegradation of binary mixtures of octane with benzene, toluene, ethylbenzene or xylene (BTEX): insights on the potential of Burkholderia, Pseudomonas and Cupriavidus isolates. World J Microbiol Biotechnol 37, 1–8 (2021).
    OpenUrlCrossRef
  24. 24.↵
    Hocinat, A., Boudemagh, A., Ali-Khodja, H. & Medjemadj, M. Aerobic degradation of BTEX compounds by Streptomyces species isolated from activated sludge and agricultural soils. Arch Microbiol 202, 2481–2492 (2020).
    OpenUrl
  25. 25.↵
    Khodaei, K., Nassery, H. R., Asadi, M. M., Mohammadzadeh, H. & Mahmoodlu, M. G. BTEX biodegradation in contaminated groundwater using a novel strain (Pseudomonas sp. BTEX-30). Int Biodeterior Biodegradation 116, 234–242 (2017).
    OpenUrl
  26. 26.↵
    Jiang, B. et al. Biodegradation of Benzene, Toluene, Ethylbenzene, and o-, m-, and p-Xylenes by the Newly Isolated Bacterium Comamonas sp. JB. Appl Biochem Biotechnol 176, 1700–1708 (2015).
    OpenUrl
  27. 27.↵
    Halsey, K. H. & Giovannoni, S. J. Biological controls on marine volatile organic compound emissions: A balancing act at the sea-air interface. Earth Sci Rev 240, 104360 (2023).
    OpenUrl
  28. 28.↵
    Rix, G. D. et al. Analysis of Essential Isoprene Metabolic Pathway Proteins in Variovorax sp. Strain WS11. Appl Environ Microbiol 89, (2023).
  29. 29.↵
    Halsey, K. H. et al. Biological cycling of volatile organic carbon by phytoplankton and bacterioplankton. Limnol Oceanogr 62, 2650–2661 (2017).
    OpenUrl
  30. 30.↵
    Yurimoto, H. & Sakai, Y. Interaction between C1-microorganisms and plants: contribution to the global carbon cycle and microbial survival strategies in the phyllosphere. Biosci Biotechnol Biochem 87, 1–6 (2022).
    OpenUrl
  31. 31.↵
    Samo, T. J. et al. Attachment between heterotrophic bacteria and microalgae influences symbiotic microscale interactions. Environ Microbiol 20, 4385–4400 (2018).
    OpenUrlCrossRef
  32. 32.↵
    Di Costanzo, F., Di Dato, V. & Romano, G. Diatom–Bacteria Interactions in the Marine Environment: Complexity, Heterogeneity, and Potential for Biotechnological Applications. Microorganisms 2023, Vol. 11, Page 2967 11, 2967 (2023).
    OpenUrl
  33. 33.↵
    Wear, E. K., Carlson, C. A., Windecker, L. A. & Brzezinski, M. A. Roles of diatom nutrient stress and species identity in determining the short- and long-term bioavailability of diatom exudates to bacterioplankton. Mar Chem 177, 335–348 (2015).
    OpenUrl
  34. 34.↵
    Liu, S. et al. Opportunities and challenges of using metagenomic data to bring uncultured microbes into cultivation. Microbiome 10, (2022).
  35. 35.↵
    Wang, Y., Zhao, Y., Bollas, A., Wang, Y. & Au, K. F. Nanopore sequencing technology, bioinformatics and applications. Nature Biotechnology 2021 39:11 39, 1348–1365 (2021).
    OpenUrlCrossRefPubMed
  36. 36.↵
    Van Tol, H. M., Amin, S. A. & Virginia Armbrust, E. Ubiquitous marine bacterium inhibits diatom cell division. ISME J 11, 31–42 (2017).
    OpenUrl
  37. 37.↵
    Helliwell, K. E., Shibl, A. A. & Amin, S. A. The Diatom Microbiome: New Perspectives for Diatom-Bacteria Symbioses. The Molecular Life of Diatoms 679–712 (2022) doi:10.1007/978-3-030-92499-7_23/COVER.
    OpenUrlCrossRef
  38. 38.↵
    Balakrishnan, R., de Silva, R. T., Hwa, T. & Cremer, J. Suboptimal resource allocation in changing environments constrains response and growth in bacteria. Mol Syst Biol 17, 10597 (2021).
    OpenUrl
  39. 39.↵
    Mayali, X. et al. Single-cell isotope tracing reveals functional guilds of bacteria associated with the diatom Phaeodactylum tricornutum. Nat Commun 14, 5642 (2023).
    OpenUrl
  40. 40.↵
    Kim, H. et al. Bacterial response to spatial gradients of algal-derived nutrients in a porous microplate. The ISME Journal 2021 16:4 16, 1036–1045 (2021).
    OpenUrl
  41. 41.↵
    Lang, X. P., He, Z., Yang, G. P. & Dai, G. Physiological responses and altered halocarbon production in Phaeodactylum tricornutum after exposure to polystyrene microplastics. Ecotoxicol Environ Saf 268, 115702 (2023).
    OpenUrl
  42. 42.↵
    Cabrita, M. T., Vale, C. & Rauter, A. P. Halogenated Compounds from Marine Algae. Marine Drugs 2010, Vol. 8, Pages 2301-2317 8, 2301–2317 (2010).
    OpenUrl
  43. 43.↵
    Yáñez-Serrano, A. M. et al. GLOVOCS-Master compound assignment guide for proton transfer reaction mass spectrometry users. Atmos Environ 244, 117929 (2021).
    OpenUrl
  44. 44.↵
    Jakoncic, J., Jouanneau, Y., Meyer, C. & Stojanoff, V. The catalytic pocket of the ring-hydroxylating dioxygenase from Sphingomonas CHY-1. (2006) doi:10.1016/j.bbrc.2006.11.117.
    OpenUrlCrossRefPubMed
  45. 45.↵
    Tan, H. M., Tang, H. Y., Joannou, C. L., Abdel-Wahab, N. H. & Mason, J. R. The Pseudomonas putida ML2 plasmid-encoded genes for benzene dioxygenase are unusual in codon usage and low in G + C content. Gene 130, 33–39 (1993).
    OpenUrlCrossRefPubMedWeb of Science
  46. 46.↵
    Bagnéris, C., Cammack, R. & Mason, J. R. Subtle Difference between Benzene and Toluene Dioxygenases of Pseudomonas putida. Appl Environ Microbiol 71, 1570 (2005).
    OpenUrlAbstract/FREE Full Text
  47. 47.↵
    Mangwani, N., Kumari, S. & Das, S. Involvement of quorum sensing genes in biofilm development and degradation of polycyclic aromatic hydrocarbons by a marine bacterium Pseudomonas aeruginosa N6P6. Appl Microbiol Biotechnol 99, 10283–10297 (2015).
    OpenUrlCrossRef
  48. 48.↵
    Olson, M. S., Ford, R. M., Smith, J. A. & Fernandez, E. J. Quantification of bacterial chemotaxis in porous media using magnetic resonance imaging. Environ Sci Technol 38, 3864–3870 (2004).
    OpenUrlPubMed
  49. 49.
    Lacal, J. et al. Bacterial chemotaxis towards aromatic hydrocarbons in Pseudomonas. Environ Microbiol 13, 1733–1744 (2011).
    OpenUrlCrossRefPubMedWeb of Science
  50. 50.↵
    Matilla, M. A., Gavira, J. A. & Krell, T. Accessing nutrients as the primary benefit arising from chemotaxis. Curr Opin Microbiol 75, 102358 (2023).
    OpenUrl
  51. 51.↵
    Bell, W. & Mitchell, R. Chemotactic and growth responses of marine bacteria to algal extracellular products. Biol. Bull. 143, 265–277 (1972).
    OpenUrlCrossRefWeb of Science
  52. 52.↵
    Hearn, E. M., Patel, D. R. & Van Den Berg, B. Outer-membrane transport of aromatic hydrocarbons as a first step in biodegradation. Proc Natl Acad Sci U S A 105, 8601–8606 (2008).
    OpenUrlAbstract/FREE Full Text
  53. 53.↵
    Chicca, I. et al. Degradation of BTEX mixture by a new Pseudomonas putida strain: role of the quorum sensing in the modulation of the upper BTEX oxidative pathway. Environmental Science and Pollution Research 27, 36203–36214 (2020).
    OpenUrlCrossRef
  54. 54.↵
    Sun, J. et al. One Carbon Metabolism in SAR11 Pelagic Marine Bacteria. PLoS One 6, e23973 (2011).
    OpenUrlCrossRefPubMed
  55. 55.
    Simó, R., Cortés-Greus, P., Rodríguez-Ros, P. & Masdeu-Navarro, M. Substantial loss of isoprene in the surface ocean due to chemical and biological consumption. Communications Earth & Environment 2022 3:1 3, 1–8 (2022).
    OpenUrlCrossRef
  56. 56.
    Davie-Martin, C. L., Giovannoni, S. J., Behrenfeld, M. J., Penta, W. B. & Halsey, K. H. Seasonal and Spatial Variability in the Biogenic Production and Consumption of Volatile Organic Compounds (VOCs) by Marine Plankton in the North Atlantic Ocean. Front Mar Sci 7, (2020).
  57. 57.↵
    Halsey, K. H. et al. Biological cycling of volatile organic carbon by phytoplankton and bacterioplankton. Limnol Oceanogr 62, 2650–2661 (2017).
    OpenUrl
  58. 58.↵
    Guillard, R. R. L. & Ryther, J. H. STUDIES OF MARINE PLANKTONIC DIATOMS: I. CYCLOTELLA NANA HUSTEDT, AND DETONULA CONFERVACEA (CLEVE) GRAN. Can J Microbiol 8, 229–239 (1962).
    OpenUrlCrossRefPubMedWeb of Science
  59. 59.↵
    Ritchie, R. J. Consistent Sets of Spectrophotometric Chlorophyll Equations for Acetone, Methanol and Ethanol Solvents. Photosynth Res 89, 27–41 (2006).
    OpenUrlCrossRefPubMedWeb of Science
  60. 60.↵
    Kolber, Z. S., Prášil, O. & Falkowski, P. G. Measurements of variable chlorophyll fluorescence using fast repetition rate techniques: defining methodology and experimental protocols. Biochimica et Biophysica Acta (BBA) - Bioenergetics 1367, 88–106 (1998).
    OpenUrl
  61. 61.↵
    Dekas, A. E. et al. Characterizing Chemoautotrophy and Heterotrophy in Marine Archaea and Bacteria With Single-Cell Multi-isotope NanoSIP. Front Microbiol 10, 455124 (2019).
    OpenUrl
  62. 62.↵
    Apweiler, R. et al. The Universal Protein Resource (UniProt) 2009. Nucleic Acids Res. 37, D169–D174 (2008).
    OpenUrlCrossRefPubMed
  63. 63.↵
    Gu, Z. Complex heatmap visualization. iMeta 1, (2022).
  64. 64.↵
    Gu, Z., Eils, R. & Schlesner, M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32, 2847–2849 (2016).
    OpenUrlCrossRefPubMed
Back to top
Posted June 17, 2024.
Download PDF
Email
Volatile Organic Compound Specialists in the Phycosphere
Vaishnavi G. Padaki, Xavier Mayali, Peter K. Weber, Stephen J. Giovannoni, Kaylene Abraham, Kerry Jacobs, Lindsay Collart, Kimberly H. Halsey
bioRxiv 2024.06.14.599129; doi: https://doi.org/10.1101/2024.06.14.599129
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area