Role of a single MCP in evolutionary adaptation of Shewanella putrefaciens for swimming in planktonic and structured environments

Abstract

Bacteria can adapt to their environments by changing phenotypic traits by mutations. However, improving one trait often results in the deterioration of another one, a trade-off that limits the degree of adaptation. The gammaproteobacterium Shewanella putrefaciens CN-32 has an elaborate motility machinery comprising two distinct flagellar systems and an extensive chemotaxis array with 36 methyl-accepting chemotaxis sensor proteins (MCPs). In this study, we performed experimental selection on S. putrefaciens for increased spreading through a porous environment. We readily obtained a mutant that showed a pronounced increase in covered distance. This phenotype was almost completely caused by a deletion of 24 bp from the chromosome, which leads to a moderately enhanced production of a single MCP. Accordingly, chemotaxis assays under free-swimming conditions and cell tracking in soft agar showed that the mutation improved navigation through nutritional gradients. In contrast, further increased levels of the MCP negatively affected spreading. The study demonstrates how moderate differences in the abundance of a single MCP can lead to an efficient upgrade of chemotaxis in specific environments at a low expense of cellular resources.IMPORTANCEExperimental evolution experiments have been used to determine the trade-offs occurring in specific environments. Several studies that have used the spreading behavior of bacteria in structured environments identified regulatory mutants that increase the swimming speed of the cells. While this results in a higher chemotaxis drift, the growth fitness decreases as the higher swimming speed requires substantial cellular resources. Here we show that rapid chemotaxis adaptation can also be achieved by modifying the chemotaxis signal input at a low metabolic cost for the cell.

Keywords: MCP; chemotaxis; evolution; flagellar motility; flagellar system; signal perception.

Fig 1

Isolation of an S. putrefaciens CN-32 spreading mutant by experimental evolution. (A) Procedure of isolation. An S. putrefaciens strain was allowed to spread on LB soft agar for 24 h. Cell material was isolated from the outer fringes of the spreading zone and re-inoculated on soft agar. The procedure was repeated 14 times. (B) Comparison of the parent and the evolved strain (G14) on soft agar. (C) Quantification of the spreading radius of four independent experiments. The asterisks indicate the significance according to a pairwise.t.test (P < 0.001).

Fig 2

Cell morphology and flagellation of the G14 mutant. Quantification of (A) length of cells, (B) percentage of flagellated cells, and (C) length of filaments, each in liquid media. n.s., not significant according to a pairwise.t.test; the number of cells is indicated by n.

Fig 3

Mapping of the dominant mutation to MCP_0378. (A) Cartoon of gene region. A 24 bp region was found to be deleted upstream of SO_0384, encoding an MCP (MCP_ 384). (B) Quantification of SO_384 expression by integration of luxCDABE directly downstream of the structural gene. Light units emitted by the indicated strains were determined in exponentially growing cultures. (C) Comparison of produced MCP_0387-mCherry in the wild type and the Δ24 mutant backgrounds by western blotting. The panel below shows the same region of the Coomassie-stained gel prior to blotting as loading control. For the full uncut gel and western blot, see Fig. S3. (D) Localization of MCP_0387-mCherry by fluorescence microscopy. Shown are micrographs of cells expressing the hybrid fusion gene from the chromosome. The upper two panels show the phase contrast image, and the lower two show the corresponding mCherry fluorescence image, where the position of the cells is outlined. The scale bar equals 5 µm. For an image including the untagged wild type, see Fig. S4. (E) Quantification of the polar mCherry fluorescence intensity. (B and D) The asterisks display the significance using a pairwise.t.test (P < 0.01).

Fig 4

Effects of MCP_0387 levels on chemotaxis. (A) Quantification of the swimming speed during free swimming by 3D tracking. Shown is the speed distribution of the wild type (left panel; red; n = 281), the Δ24Δ0387 mutant (middle panel; green; n = 935), and the Δ24 mutant (right panel; blue; n = 265). (B) Quantification of wild type and mutant spreading at different LB conditions as indicated. Shown are the results from at least three biological replicates. (C) Quantification of wild type and mutant spreading under different media conditions (LB, LM, and PBS buffer plus casamino acids) as indicated. Shown are the results from at least three biological replicates. (D) Quantification of the chemotactic drift in controlled gradients as indicated. Shown are the results of at least three biological replicates after 1 (red), 2 (green), and 3 (blue) hours of incubation. The asterisks display the significance using a pairwise.t.test (*, P < 0.05; ***, P > 0.01; n.s., not significant).

Fig 5

MCP_0387 affects spreading in soft agar. 3D track projection of wild-type cells under planktonic conditions (A) and in soft agar (B). The side of each square at the bottom is 100 µm. (C) The mean distance traveled in 100 ms (“step size”) of the indicated cells in soft agar (wild type, red; Δ24Δ0387, green; Δ24, blue). (D) Standard deviation in step size (wild type, red; Δ24Δ0387, green; Δ24, blue). The number of tracks quantified was 9,931 for the wild type, 11,393 for the Δ24Δ0387 mutant, and 1,709 for the Δ24 mutant. Further parameters can be found in Fig. S6 and S7.

Fig 6

Effect of MCP_0378 levels on growth. Growth of the indicated strains in (A) LB and (B) 4M media. Shown are the results of three biological replicates. The wild type is displayed in red, the Δ24 mutant is in blue, and the Δ24Δ0387 mutant is in green.