Super-Resolution Fluorescence Microscopy Study of the Production of K1 Capsules by Escherichia coli: Evidence for the Differential Distribution of the Capsule at the Poles and the Equator of the Cell

The production of Escherichia coli K1 serotype capsule was investigated using direct stochastic optical reconstruction microscopy with live bacteria and graphene oxide-coated coverslips, overcoming many morphological artifacts found in other high-resolution imaging techniques. Super-resolution fluorescence images showed that the K1 capsular polysaccharide is not uniformly distributed on the cell surface, as previously thought. These studies demonstrated that on the cell surfaces the K1 capsule at the poles had bimodal thicknesses of 238 ± 41 and 323 ± 62 nm, whereas at the equator, there was a monomodal thickness of 217 ± 29 nm. This bimodal variation was also observed in high-pressure light-scattering chromatography measurements of purified K1 capsular polysaccharide. Particle tracking demonstrated that the formation of the capsule was dominated by the expansion of lyso-phosphatidylglycerol (lyso-PG) rafts that anchor the capsular polysaccharide in the outer membrane, and the expansion of these rafts across the cell surface was driven by new material transported through the capsular biosynthesis channels. The discovery of thicker capsules at the poles of the cell will have implications in mediating interactions between the bacterium and its immediate environment.


Super-resolution images of encapsulated K1 E. coli, EV36 strain
. Examples of wide-field super-resolution images of live E. coli bacteria, The capsular structures were labelled with anti-K1 conjugated with anti-mouse-AF647 placed in imaging buffer and M9 media. The images were created using dSTORM 1 (50 nm resolution) with PLLGO coated coverslips at different focal planes. a) Shows example data used for the analysis of capsular brush thicknesses and their polydispersity. b) Shows bacterial cross-sectioning at a level above their mid-point and we omitted this type of image for capsule thickness measurement to avoid projection artefacts. The scale bars are 5 m.
Examples of wide-field super-resolution images are shown in figure S1. High resolution images (50 nm) could be obtained of the capsular structures of a large number of bacteria. The bacteria used for the quantitative thickness calculations were imaged at their mid-points (figure S1a) and bacteria were analysed in their more common side on orientation. Other projections in the images, such as end on orientations and slices significantly beyond their midpoints (figure S1b) were omitted from the analysis. The depth of focus (DOF) of our super-resolution microscope was analysed using a virtual light sheet software package 2 and it was found to be nm. We used 347 ± 22 this DOF to quantify its effects on the bacterial capsule thickness measurements and it was found to be negligible e.g. on the results shown in figure 3.

The calculation of steric forces of the K1 capsule
Consider the osmotic brush model of Pincus for the conformation of polyelectrolyte brushs 3 . When considered at low salt concentration, e.g. less than 10 mM, it is found that: Using the assumption that the highly negative charged polymers experienced strong charge screening, we can characterise the conformation and thickness of capsular polysaccharide using the osmotic pressure model to describe the thicknesses and forces experienced by the lyso-PG brushes, in which the chains are highly strecthed. Hence the thickness of capsular polysaccharide brush can be approximated, by using a grafting length nm 4 , a degree of polymerization 5,6 , a Kuhn length ( ) of 1 nm 7 , and a salt concentration of around a few 10 mM due to monovalent salt a in the M9 media. The brush thickness is found using equation (S1) in the range nm, which is in reasonable agreement with the measurements from 300 200   H STORM images. Figure S3. a) Force as a function of separation distance from AFM for live encapsulated K1 E.coli under hydrated conditions, with filtered miliQ water. The force was fitted with an exponential decay to determine the contact points of the probe with the bacterium (coloured curves). b) and c) are 3D and 2D images respectively of encapsulated bacteria using AFM scanning with a 2.5 µm spherical probe. is a nonspecific imaging technique, so the presence of the capsules could only be determined from the force curves (figure S3a). Figure S4. Growth curve of E. coli bacteria (EV36-wild type and EV136-capsule mutant strains) showing OD600 as a function of incubation time with different carbon sources in the growth media. a) and b) are the same data plotted over different times. The media abbreviations are GG is 0.4% Glucose and 0.4% Glycerol, GGIMB is 0.4% Glucose and 0.4% Glycerol with imaging buffer, and blank is a control media without bacteria. The imaging buffer (IMB) has a significant impact on bacterial proliferation after 2 hours. Figure S4 shows that the imaging buffer with the oxygen scavenger did perturb the growth of bacterial cells and reduced the population of the bacteria cells over time. However, to circumvent this problem in super-resolution microscopy, we first deposited a high density of bacteria cells on the PLGO coated substrate and then changed the media including the imaging buffer and the M9 media with the carbon source supplement every 1 hour 8 . This allowed us to observe the expansion of capsular lyso-PG raft over 3 hours (the time scale of capsule growth) with no significant artefacts. S6 6. Effect of the poly-L-lysine coated graphene oxide coverslips on bacterial proliferation. Figure S5. The effect of positive charge substrate (poly-L-lysine) on bacterial growth. Bacteria were grown on the substrate without poly-L-lysine coated (red) and with poly-L-lysine coated (blue) from 0 to 6hrs. Then the deposited cells were lifted off using 0.1% triton X100 and grown on a culture plate to perform a viable count.

Effect of the oxygen scavenger system with the imaging buffer on bacterial proliferation
The proliferation of bacteria was also examined in contact with poly-L-lysine coatings (figure S5). No significant changes were observed over the time scales of our experiments (3 hours) compared with the glass substrate control. The function of capsular proteins expressed from the gene clusters ( Figure S6a) in region 1 and region 3 is to support the translocation of K1 polysaccharides through the ABC transporter (the KpsMT complex [9][10][11] ). However, the function of other capsular proteins involved in this activity is not yet clear, such as KpsS and NeuE. Enzymes encoded in region 2 behave as polymerases for the creation of the polysaccharides 12 . Although there are only a few essential enzymes in the polysialic synthesis process, e.g. NeuCBAS, the clarification of their roles and the interactions of these enzymes with others are still questioned. A better understanding of the role of NeuS would provide a much better picture of the activity of the whole biosynthesis pathway.

The protein machines involved in the production
In figure S6b, the number of proteins per cell is shown: KpsD and KpsE after up regulation by the shift in temperature to 37 C. The number of these complexes relates to the probability of the capsular brush rafts emerging on the bacterial surface. a)