a Distribution of swimming velocity (left axis) of two environmental isolates (Pseudoalteromonas sp. ASV39 and ASV16) in the presence and absence of 10 µM DMSPd. Overnight cell cultures were diluted to 106 cells mL−1 and amended with 10 µM (final concentration) of DMSPd (ntotal = 2.3 × 104 and 2.4 × 103 cells for Pseudoalteromonas sp. ASV39 and ASV16, respectively). A second culture in artificial seawater (ASW) served as control (ntotal = 1.6 × 104 and 4.2 × 103 cells for Pseudoalteromonas sp. ASV39 and ASV16, respectively). Cells from each culture were tracked by video-microscopy, between 15 and 40 min after DMSPd amendment (dark shade), and between 40 and 75 min (light shade). Box plots inside the violins indicate the quartiles of corresponding measurements. Minimum and maximum values are indicated by endpoints; centre dot denotes median values; whiskers denote 1.5 × the interquartile range; fine line extends to outlier measurements. Results show no significant difference in swimming velocity between the DMSPd-amended treatment and the DMSPd-free control (Fig. S7, ANOVA (one-sided), p > 0.05). b–d Chemotactic index (IC, coloured bars, left axis) in response to 10 mg mL−1 laminarin in laboratory experiments in the presence of different concentrations (0, 0.1, 1 and 10 µM) of (b) methionine (one methyl group), (c) choline (three methyl groups), and (d) homocysteine (no methyl group) for Pseudoalteromonas sp. ASV16. Asterisks denote chemotactic index significantly larger than the filtered seawater controls (ANOVA (one-sided), p < 0.05, all p-values are reported in Supplementary Data 23–25). Bars in a darker shade indicate a significantly stronger chemotactic response to laminarin compared to the controls without methionine, choline or homocysteine, respectively (ANOVA, p < 0.05, all p-values are reported in Supplementary Data 23–25). Each treatment was replicated across three different ISCAs (n = 3 biologically independent experiments, denoted by individual dots). Bar plots represent the mean ± SD. e Schematic of DMSPd-enhanced chemotaxis to laminarin. In a given timeframe (t1 to t2), cells will cover larger distances along a laminarin gradient when DMSPd is present (left), than when DMSPd is absent (right). Our data suggest that this behaviour is induced because DMSPd acts as a precursor for methylation of the chemotaxis machinery, favoring the cell’s adaptation to increasing polymer concentrations. Source data are provided as a Source Data file.