Five complementary tests were performed to evaluate the ability of the modified cells to assemble functional curli: 1) determination of the percentage of adherent cells to polystyrene in 24 wells-plates, 2) crystal violet staining of biofilm formed on polystyrene in 24 wells-plates, 3) ability to bind the congo red, 4) biofilm maximum thickness measurement and biovolumes quantification of GFP-tagged biofilm observed with a confocal microscopy and 5) curli structure observation using Transmission Electron Microscopy (MET).
Adhesion test and curli production
Figure 1 : Engineered bacteria Percentage of adhesion
csgA-knockout E. coli strain was transformed with BBa_CsgA-WT (BBa_K1404006); BBa_CsgA-His1 (BBa_K1404007); BBa_CsgA-His2 (BBa_K1404008). The corresponding positive and negative controls are Wild-type E.coli curli producing strain transformed with with the empty vector and csgA--knockout E. coli strain transformed with the empty vector respectively.
Strains with our parts, the positive and negative controls were cultivated in 24-wells microplate in M63 Mannitol during 24H at 30°C. The supernatant was removed and the OD600 measured, then the bacteria forming the biofilm were resuspended and the OD600 measured in order to estimate the number of cells (See protocol for more details). The percentage of adhesion was calculated as follow :
(OD600 of the biofilm)/ (OD600 of the supernatant + OD600 of the biofilm)
Significant differences are indicated using uppercase letters, and different letters indicate significant differences (Tukey’s test, p < 0.05)
These results show that the percentage of adhesion is similar between the strains containing the three parts and the positive control, thus tagged CsgA were still functional. CsgA with one or two tags from the P70 promoter were sufficient to form thick biofilms.
Figure 2 : Engineered bacteria Biofilm formation
The cells were grown as described as figure 1.
The supernatant was removed and the remaining biofilm was fixed to the microplate by heat treatment at 80°C during 1H. The violet crystal solution was added in each well in order to stain the cells and the wells were washed with water to remove crystal violet in excess (See protocol for more details).
Violet crystal staining shows that the strain containing the three parts could form a biofilm like the positive control, thus tagged CsgA were still functional. CsgA with one or two tags from the P70 promoter were sufficient to form thick biofilms.
Figure 3 : Engineered bacteria curli production
Strains are the same as in figure 1.
Strains with our parts, the positive and negative control were cultivated in M63 Mannitol at 30°C and 180rpm. After centrifugation, the supernatant was removed and the cell pellet was resuspended in the Congo Red solution, in order to specifically stain the curli. The samples were centrifuged again and the pellets observed (See protocol for more details).
Congo Red staining shows that the CsgA with one or two tag from the P70 promoter allows to form curli fiber which are able to bind congo red.
For the Confocal Laser Scanning Microscopy biofilm acquisitions, all the strains were cultivated in 96-wells microplate in M63 Mannitol during 16H at 30°C (See Protocol for details). See results in Figure 4.
Figure 4: Engineered bacteria biofilm characterization and quantification using Confocal Laser Scanning Microscopy
All the strains used are constitutively fluorescent to allow detection with confocal laser microscopy (ZEISS LSM510 META, 40X/1.3OILDIC, laser Argon 4 lines 30 W 458 nm, 477 nm, 488 nm, 514 nm, See Protocol). Positive control/CsgA+ (Wild-type E. coli curli producing strain); Negative control/CsgA- (csgA-knockout E. coli strain); BBa_CsgA (BBa_K1404006); BBa_CsgAHis1 (BBa_K1404007); BBa_CsgAHis2 (BBa_K1404008). A) Biofilm sections obtained by Z-stack acquisitions. B) Biofilm 3D reconstruction using IMARIS® from acquisitions in A). C) Bio-volume quantification and maximum of thickness measurement using COMSTAT2 (ImageJ). The strain marked with a star is significantly different from all others (Tukey’s test, p<0.05).
As no strains carrying our parts show a significant difference with the positive control, our parts insertion doesn’t modify the biofilm formation property. The His-Tag and His2-Tag engineered CsgA doesn’t disturb the curli formation.
Transmission Electron Microscopy
For the Transmission Electron Microscopy, all the strains were cultivated in conditions that allow the formation of the curli : 48h (for a 50 mL culture), 28⁰C temperature and low agitation. The ammonium molybdate was used as a negative colorant (Microscope: MET PHILIPS CM120).
Figure 5: Engineered bacteria curli structure observation using Transmission Electron Microscopy
The bacteria cultures we used are a csgA-knockout strain as negative control, and the three engineered csgA constructions, the WT, the His1-Tag and His2-Tag.
The images show that there is no significant difference between our positive control and our constructions. Thus, we can conclude that our parts insertions don’t affect the structure of the amyloid fibers and the configuration of the curli formation.
General Conclusion
The expression of CsgA derivatives from the p70 csg promoter carried by the psb1c3 plasmid leads to functional CsgA, and allows E. coli to stick and form biofilm. Moreover, our results show that the addition of one or two His-Tag on C-term of CsgA doesn’t disturb the normal properties of curli (sturdiness, adhesion, structure and folding of CsgA).