P. putida are attracted to higher salinity.

(a) A microfluidic assay to evaluate cells under NaCl gradients. (b) Change in the total cell population normalized by the initial cell population over time in the presence (blue) and absence (black) of NaCl gradients. Symbols represent experimental data (blue triangles for NaCl gradients, black circles for no gradient) while solid lines represent numerical simulations. Shaded regions denote the standard deviation across three simulations. (c,d) Local distribution of cells over time (c) without (Movie S1) or (d) with NaCl gradients (Movie S2). The color code represents the number of cells in the i-th bin, bi (t), normalized by the total number of cells from the beginning, . Although the heat map only shows one experiment here, the observed patterns were reproduced across multiple replicates, with similar trends

P. putida aligns along salt gradients.

(a,b) Unfiltered cell trajectories recorded over 25 s of duration for different times (0–25, 75–100, 150–175 s) in the (a) absence and (b) presence of NaCl gradients. (c,d) Distribution of cell mean positions along the x axis at successive 25 s time intervals (Early: 0–25 s, Mid: 75–100 s, Late: 150–175 s) in the (c) absence and (d) presence of NaCl gradients. The white box in (c,d) indicates interquartile range with median shown as a stripe and mean as a dot. (e,f) Probability density function distribution of cell’s directions of instantaneous velocity vectors in the (e) absence and (f) presence of NaCl gradients. Scale bar in (a) is 100 μm.

P. putida run faster and straighter toward salt.

(a) Comparison of run speeds averaged over the entire angles, 0° (-15° < θ < 15°), and 180° (165° < θ < 195°). (b,c) Distributions of average run speed in the (b) absence and (c) presence of NaCl gradients. (d) Instantaneous run speed in the absence and presence of NaCl gradients (e) Comparison of average run straightness for entire angles, 0° (-15° < θ < 15°), and 180° (165° < θ < 195°). (f,g) Distributions of average run straightness in the (f) absence and (g) presence of NaCl gradients. (h) Tumble rates measured in the absence and presence of NaCl gradients. The inset compares tumble rates for runs oriented up gradient and down gradient. Error bars in (a, d, e) represent standard deviation.

Non-uniform diffusiophoresis steers P. putida toward salt.

(a) An illustration of cell rotation due to non-uniform diffusiophoresis. (b,c) Experimental run trajectories (40 randomly chosen) repositioned to start at the origin in the (b) absence (Movie S3) and (c) presence (Movie S4) of NaCl gradients. The color code represents run straightness S . (d,e) Numerical simulation run trajectories repositioned to start at the origin in the (d) absence (Movie S5) and (e) presence (Movie S6) of diffusiophoretic drift. (f,g) Angle correlation function 〈cos(Δθ)〉 in (f) experimental run trajectories and (g) simulation run trajectories. Shaded regions indicate ± standard error of the mean (SEM.) (h,i) MSD anisotropy ratio for (h) experimental run trajectories and (i) simulation run trajectories.

Salt gradients disperse cells toward toxic contaminant.

(a) Simulating diffusiophoretic bioaugmentation of toluene-contaminated pore. Cells suspended in low salinity water are injected into the channel that is filled with high salinity water in the presence of toluene on the gel-sided channel, thereby creating dual chemical (salt and toluene) gradients. (b) An image of the microfluidic channel in the presence of toluene. Toluene is false-colored. (c,d) Trajectories of cells in the (c) absence (Movie S7) and (d) presence of NaCl gradients. (Movie S8) Scale bars in (b,d) are 50 μm.