Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Filters applied. Clear all
. 2020 Apr 23;11(1):1942.
doi: 10.1038/s41467-020-15693-z.

Single-cell Bacterial Transcription Measurements Reveal the Importance of Dimethylsulfoniopropionate (DMSP) Hotspots in Ocean Sulfur Cycling

Free PMC article

Single-cell Bacterial Transcription Measurements Reveal the Importance of Dimethylsulfoniopropionate (DMSP) Hotspots in Ocean Sulfur Cycling

Cherry Gao et al. Nat Commun. .
Free PMC article


Dimethylsulfoniopropionate (DMSP) is a pivotal compound in marine biogeochemical cycles and a key chemical currency in microbial interactions. Marine bacteria transform DMSP via two competing pathways with considerably different biogeochemical implications: demethylation channels sulfur into the microbial food web, whereas cleavage releases sulfur into the atmosphere. Here, we present single-cell measurements of the expression of these two pathways using engineered fluorescent reporter strains of Ruegeria pomeroyi DSS-3, and find that external DMSP concentration dictates the relative expression of the two pathways. DMSP induces an upregulation of both pathways, but only at high concentrations (>1 μM for demethylation; >35 nM for cleavage), characteristic of microscale hotspots such as the vicinity of phytoplankton cells. Co-incubations between DMSP-producing microalgae and bacteria revealed an increase in cleavage pathway expression close to the microalgae's surface. These results indicate that bacterial utilization of microscale DMSP hotspots is an important determinant of the fate of sulfur in the ocean.

Conflict of interest statement

The authors declare no competing interests.


Fig. 1
Fig. 1. Single-cell measurements of DMSP degradation pathway expression.
a, b Plasmids transformed into R. pomeroyi DSS-3 contain three components: dmdA reporter (222 bp promoter region); dddW reporter (500 bp promoter region); and constitutive yfp expression (strong, synthetic promoter PA1/04/03). YFP signal was used as a proxy for plasmid copy number and metabolic activity. Transcriptional terminators (represented by T) were placed between promoter fusion cassettes to prevent transcriptional read-through. To control for spectral bias caused by fluorescent protein choice (RFP or TFP), we constructed two R. pomeroyi reporter strains—Regular (a) and Goofy (b)—in which the colors of fluorescent proteins fused to dmdA and dddW promoter regions were interchanged. Vector backbone: pBBR1MCS-2 with origin of replication pBBR1 (open circles). c Schematic of a single microfluidic device used for time-lapse DMSP experiments. Each time-lapse DMSP experiment used one microfluidic device containing nine observation chambers for parallel incubation of a single reporter strain (Regular or Goofy; a, b) with different concentrations of DMSP. Glucose was used as negative control. White squares in each observation chamber represent the seven fields of view (200 μm × 200 μm) imaged at each time point. d Representative phase contrast and fluorescence images of a single cell (strain Goofy) over time in the presence of 1 mM DMSP. Scale bar, 2 μm.
Fig. 2
Fig. 2. DMSP concentration-dependent upregulation of dmdA and dddW.
a, b Mean fluorescence signals of dmdA (demethylation) (a) and dddW (cleavage) (b) reporters in response to different concentrations of DMSP. One representative replicate experiment of reporter strain Goofy is shown (data from additional replicate experiments (n = 3 for each reporter strain) are shown in Supplementary Fig. 6). Spectral leakage correction, background subtraction, and a threshold on YFP intensity were applied (see Methods and Supplementary Note 1). RFP and TFP signals of each cell were normalized by the mean YFP signal at each time point of each experimental condition. Data points and error bars represent means ± s.e.m. of cells (error bars may be smaller than markers). c, d Average end-point fluorescence levels of dmdA (c) and dddW (d) reporters. For each replicate experiment, baseline signal (glucose) was subtracted at each time point, fluorescence signals over the final five time points (~20.4–24 h) were averaged, then normalized by the corresponding end-point fluorescence signal at 250 μM DMSP. Data points and error bars represent means ± s.d. of six total replicates (n = 3 for strain Regular and n = 3 for strain Goofy combined).
Fig. 3
Fig. 3. DMSP concentration modulates relative expression of dddW and dmdA.
a Cleavage-to-demethylation pathway ratio was calculated at each DMSP concentration for strains Regular (RFP/TFP) and Goofy (TFP/RFP). High variability of fluorescence output amongst replicate experiments at ≥10 μM DMSP prevented the comparison of pathway reporters within each color (Supplementary Fig. 5). Average fluorescence signals at time points at which dmdA expression is mid-exponential for each DMSP concentration (shown in Supplementary Fig. 6), or at the second time point for glucose and 1 μM DMSP conditions, were used for ratio calculation. Close agreement between strains Regular and Goofy at ≥10 μM DMSP suggests that fluorescence ratios are close to true pathway expression ratios. The deviation between strains Regular and Goofy of ratios in glucose and 1 μM DMSP may be due to low fluorescence signals; importantly, ratios calculated within-color (0.3–0.5; Supplementary Fig. 5) and across-color (0.15–1.0; Fig. 3) in glucose showed consistent results (similar values at 1 μM DMSP). Pathway ratios from the phycosphere experiment (also shown in Fig. 4f) were calculated using TFP signals of reporter strains (reporting either dmdA or dddW expression in strains Regular or Goofy, respectively) normalized by constitutive YFP signals, and were plotted against modeled phycosphere DMSP concentrations (Supplementary Fig. 13). Data points and error bars of DMSP concentration experiments are slightly offset in the x-direction for presentation clarity, and represent means ± s.d. of replicate experiments (n = 3 for strain Regular; n = 3 for strain Goofy). Error bars of the phycosphere experiment represent the variance of the ratio of normalized cleavage and demethylation fluorescence signals (Supplementary Note 1). Error bars may be smaller than markers. b Inset represents the same data as a, plotted on a linear scale on the x-axis to show the saturating relationship between DMSP concentration and cleavage-to-demethylation ratio.
Fig. 4
Fig. 4. Gene expression patterns in a natural DMSP hotspot.
a, b Representative images of co-incubation between DMSP-producing microalgae, Breviolum CCMP2459 (photosynthetic pigment, orange), and engineered bacteria, R. pomeroyi, constitutively expressing YFP (white) and fluorescently reporting dmdA (a demethylation, strain Regular-TFP, magenta) or dddW (b cleavage, strain Goofy-TFP, blue) expression. Fluorescence signals are false-colored. Representative concentric rings (widths, 20 pixels = 1.6 μm) that bin distances from the center (red dots) of Breviolum cells for fluorescence quantification are shown at 15 and 30 μm. Scale bar, 5 μm. c, d Quantification of YFP (constitutive) and TFP (dmdA (c), or dddW (d)) fluorescence at each distance (r) from the center of Breviolum cells (mean radius = 3.3 μm, red dotted line; n = 33). Fluorescence upregulation of YFP (c, d) and TFP (dddW; d) were detectable up to r = 18.6 μm (gray shading), at which modeled DMSP concentration was 35 nM (Supplementary Fig. 13). YFP of the cleavage pathway reporter (b, d) was brighter than that of the demethylation pathway reporter (a, c), probably due to differences in metabolic activity levels in each bacterial culture. Bacteria nearest to the surface of the phytoplankton (r = 4.2 μm, the first concentric ring) appeared dimmer than expected, possibly due to spectral interference from photosynthetic pigments. Data points and error bars represent mean ± s.d. of images (n = 15  Breviolum cells for dmdA (c); n = 18 for dddW (d)), calculated at each concentric ring (image analysis methods in Supplementary Note 3). e TFP fluorescence of each cell-containing pixel normalized by the average YFP fluorescence of cell-containing pixels within the corresponding concentric circle to remove the effect of metabolic activity differences on fluorescence intensities. Direct comparisons between demethylation and cleavage after normalization revealed that demethylation expression was higher than cleavage expression at all distances from the phytoplankton. Data points and error bars represent mean ± s.d. of normalized fluorescence of images, calculated at each concentric ring. f Average normalized TFP intensity (e) of strain Goofy (dddW, n = 18) divided by that of strain Regular (dmdA, n = 15) at each distance from the phytoplankton. Error bars represent the variance of the ratio of normalized cleavage and demethylation fluorescence signals (Supplementary Note 3).

Similar articles

See all similar articles


    1. Archer SD, Widdicombe CE, Tarran GA, Rees AP, Burkill PH. Production and turnover of particulate dimethylsulphoniopropionate during a coccolithophore bloom in the northern North Sea. Aquat. Microb. Ecol. 2001;24:225–241. doi: 10.3354/ame024225. - DOI
    1. Howard EC, et al. Bacterial taxa that limit sulfur flux from the ocean. Science. 2006;314:649–652. doi: 10.1126/science.1130657. - DOI - PubMed
    1. Moran MA, Durham BP. Sulfur metabolites in the pelagic ocean. Nat. Rev. Microbiol. 2019;17:665–678. doi: 10.1038/s41579-019-0250-1. - DOI - PubMed
    1. Zubkov MV, et al. Linking the composition of bacterioplankton to rapid turnover of dissolved dimethylsulphoniopropionate in an algal bloom in the North Sea. Environ. Microbiol. 2001;3:304–311. doi: 10.1046/j.1462-2920.2001.00196.x. - DOI - PubMed
    1. Simó R, Archer SD, Pedrós-Alió C, Gilpin L, Stelfox-Widdicombe CE. Coupled dynamics of dimethylsulfoniopropionate and dimethylsulfide cycling and the microbial food web in surface waters of the North Atlantic. Limnol. Oceanogr. 2002;47:53–61. doi: 10.4319/lo.2002.47.1.0053. - DOI