Systematic identification of novel regulatory interactions controlling biofilm formation in the bacterium Escherichia coli

Abstract

Here, we investigated novel interactions of three global regulators of the network that controls biofilm formation in the model bacterium Escherichia coli using computational network analysis, an in vivo reporter assay and physiological validation experiments. We were able to map critical nodes that govern planktonic to biofilm transition and identify 8 new regulatory interactions for CRP, IHF or Fis responsible for the control of the promoters of rpoS, rpoE, flhD, fliA, csgD and yeaJ. Additionally, an in vivo promoter reporter assay and motility analysis revealed a key role for IHF as a repressor of cell motility through the control of FliA sigma factor expression. This investigation of first stage and mature biofilm formation indicates that biofilm structure is strongly affected by IHF and Fis, while CRP seems to provide a fine-tuning mechanism. Taken together, the analysis presented here shows the utility of combining computational and experimental approaches to generate a deeper understanding of the biofilm formation process in bacteria.

Figure 1

Flagella-biofilm transcriptional regulatory network. The principal nodes and paths, which drive the flagella-biofilm network were mapped using Cytoscape 3.4.1. The network was analyzed by using degree (A) and out-degree (B). Size of the nodes (circles) indicates number of interactions of the nodes. Color scale are from blue-green-yellow to orange indicating from low to high degree values.

Figure 2

Flagella-biofilm transcriptional regulatory network. The principal nodes and paths, which drive the flagella-biofilm network were analyzed using Cytoscape 3.4.1. The network was analyzed by using betweenness centrality and edge betweenness centrality algorithms. Size of the nodes (circles) indicates betweenness analysis and width of the Edges (lines connections the circles) indicates the edge betweenness analysis. Color scale are from blue-green-yellow to orange indicating from low to high betweenness values.

Figure 3

Effect of CRP, IHF and Fis GRs over the promoter activity of rpoS, rpoE and flhD. Promoter activity assay of (A) pMR1-PrpoS, (B) pMR1-PrpoE, and (C) pMR1-PflhD were evaluated in E. coli BW25113 wild type (blue line), ∆ihf (red line) and ∆fis (green line) in 96well plate as described in methods in the absence (left panel) or presence (right panel) of 0.4% of glucose. GFP fluorescence was measured each 20 minutes at 37 °C during 8 hours in static conditions and normalized by OD600. Solid lines represent the average from three independent experiments while dashed lines are the upper and lower limits of standard error of the mean (S.E.M).

Figure 4

Effect of CRP, IHF and Fis GRs on the promoter activity of csgD, fliA and yeaJ. GFP promoter activity assay of (A) pMR1-PcsgD, (B) pMR1-PfliA, and (C) pMR1-PyeaJ were evaluated in E. coli BW25113 wt (blue line), ∆ihf (red line) and ∆fis (green line) in 96well plate as described in methods in the absence (left panel) or presence (right panel) of 0.4% of glucose. GFP fluorescence was measured each 20 minutes at 37 °C during 8 hours in static conditions and normalized by OD600. Solid lines represents the average from three independent experiments while dashed lines are the upper and lower limits of standard error of the mean (S.E.M).

Figure 5

Experimental data integration to the Flagella-biofilm transcriptional regulatory network. Promoter activity values were transformed into activation or repression connections and were loaded into Cytoscape 3.4.1. An organic algorithm was used to the properly cluster visualization. The network was analyzed by use the betweenness centrality and edge betweenness centrality measurements. Size of the nodes (circles) indicates betweenness analysis and width of the Edges (lines connections the circles) indicates the edge betweenness analysis. Scale colors from red to bright to dark indicate the high to low values.

Figure 6

Effect of GRs in the motility program at 24 h. Motility phenotype of E. coli BW25113 wild-type and mutant strains were evaluated by cell motility assay at 24 h in the presence or absence of glucose as depicted. Divergent motility capability is observed between the different conditions, proving the effect of the GRs CRP, IHF and Fis to modulate the motility program. The results are representative of 3 independent experiments.

Figure 7

Capability of E. coli and mutant strains to develop adherence and mature biofilm. (A) Adherence capability of E. coli BW25113 wild-type, ∆ihf and ∆fis were evaluated in 96-well plate using violet crystal method. Comparisons of adherence capability of BW25113 wt and mutant strains at 30 and 37 °C are shown. Vertical bars are standard deviations calculated from three independent experiments. (B) Mature biofilm formation of E. coli BW25113 wt, ∆ihf and ∆fis strains were perform using congo red plate assay. Comparisons of the mature biofilm morphological characteristics of wild type and mutant strains at 37 °C in the presence or absence of glucose is shown. Dashed line, Zone III; white line, zone II; black line, zone I; arrows represent wrinkles and clefts structures. The results are representative of 3 independent experiments.