Structure of the Bacterial Cellulose Ribbon and Its Assembly-Guiding Cytoskeleton by Electron Cryotomography

Interactions between the bacterial envelope and the cellulose ribbon: the loose configuration. (A) A 9-nm-thick tomographic slice showing a cell where aggregates of disorganized cellulose (asterisks) occur between the ribbon (yellow arrows) and the OM. Notethat the cortical belt (CB) cannot be seen in this slice. Black-outlined orange arrows indicate points of contact between the cellulose sheet and the disorganized aggregates. Red arrows point to vesicles. (B) Manual segmentation of the tomogram in panel A showing these disorganized aggregates in 3D. (C and D) Transverse 9-nm-thick tomographic slices through the envelope of the cell shown in panel A at the levels indicated by the blue and pink dashed lines highlighting the distance between the two cellulose sheets (arrows) and the OM and the presence of the disorganized clusters (dashed brackets). (E) Plot showing the OM-to-closest-sheet distance in the two types of configuration. n = 3 and 23 for the loose and tight configuration, respectively. Two-tailed P value = 0.0008 (Mann-Whitney test).

Cellulose sheet dimensions. (A and B) Longitudinal and transverse schematic depictions defining the different dimensions measured, namely, OM-to-sheet distance, sheet width, and intersheet distance. Identical terminology is used for the measurements of the cortical belt. (C) Transverse 12-nm-thick slice of the bacterial envelope of the cell shown in panel D at the level indicated by the blue dashed line. The arrows highlight the two stacked sheets. On the right, the average density profile along the red line demonstrates how the cellulose sheet widths were estimated. Vertical axis is length in nanometers along the red line and horizontal axis is the normalized electron density. (D) A 12-nm-thick tomographic slice showing the typical organization of the bacterial envelope on the side where cellulose sheets (arrows) are being synthesized. The average density profile on the right taken along the red line shows the CB-IM, IM-OM, OM-sheet and intersheet distances (green dashed lines).

The cellulose ribbon is a composite structure made of stacked sheets. (A) Percentages of cells exhibiting disorganized aggregates (blue) and cellulose ribbons (red) at 13-, 20- and 300 min postseparation. While disorganized aggregate occurrence is steady, there is an increase in the occurrence of cellulose ribbons over time. n = 6, 15, and 33 for 13, 20, and 300 min, respectively. (B) Number of cellulose sheets composing the ribbons as a function of time after cell separation. n = 6 and n = 21 tomograms for 20 and 300 min postseparation,respectively. Two-tailed P < 0.0001 (one-sample Wilcoxon signed rank test against a theoretical value of 1 [the number of sheets observed at 20 min postseparation]). (C) Composite image, composed of 10-nm thick tomographic slices spaced by 24 nm in Z, of a cell 20 min postseparation in the tight configuration. The cellulose ribbon is thin (arrows), being composed of one sheet immediately adjacent to the OM. Limits of the two original images are indicated by the dashed line. (D) An 11-nm-thick tomographic slice of a cell 300 min postseparation. The cellulose ribbon (arrows) is large and composed of multiple sheets. (E) Nascent cellulose sheet 20 min postseparation (arrow). Putative microfibrils can be seen coming out perpendicularly from the outer membrane (arrowheads). (F) Corresponding manual segmentation of the image in panel E. (G) Enlarged view of the boxed region in panel E. Below is the average density profile showing the estimation of the diameter of one putative microfibril (red line). (H) Estimated diameters of microfibrils observed at 20 min postseparation in the two cells where they are visible (left vertical axis) as in panel E and the intersheet distances measured in the 300-min-postseparation cellulose ribbons (right vertical axis). Twelve and four microfibril thickness measurements were performed on two separate tomograms (cells 1 and 2). Forty-seven measurements for intersheet distances were performed on 23 tomograms. ANOVA followed by Tukey’s multiple-comparison test was performed. Cell 1 versus cell 2, cell 1 versus 300-min intersheet distances, and cell 2 versus 300-min intersheet distances showed adjusted P values of 0.073, 0.15, and 0.0015, respectively. (I) Sheet width estimations at 20 and 300 min postseparation. Six and 45 sheets were measured at 20 and 300 minutes postseparation. Welch’s t test (parametric t test without equal SD assumption) showed a P value of 0.23.

The cortical belt lies below the cellulose ribbon in the cytoplasm. (A) A 9-nm-thick tomographic slice showing a representative cortical belt (purple arrows) just inside the IM and proximal to the cellulose ribbon on the outside of the cell (yellow arrows). (B) Manual segmentation of the tomogram shown in panel A highlighting the cellulose ribbon and the cortical belt. (C) Same segmentation rotated 90° about the long axis of the cell shows how the cortical belt and the cellulose ribbon follow the same trajectory. (D) A 9-nm-thick tomographic slice taken from the same tomogram as in Fig. 2, showing one of several cases where the cortical belt presented stacked layers (dashed box). (E) Enlarged view of the boxed region in panel D showing the arrangement of the stacked layers. On the right is a density profile displayed normal to the cortical belt to measure the interlayer distance (15 nm). (F) Transverse 9-nm-thick tomographic slice of the cell region shown in panel D, at the level indicated by the blue dashed line, highlighting stacked layers of the cortical belt. The cellulose ribbon can be seen at a distance (yellow arrowheads) with disorganized aggregates in between (dashed bracket and asterisk).

FIB milling through native G. hansenii biofilms. (A) Cryo-SEM overview of a 6-h biofilm (outlined in red) grown on a gold Quantifoil grid. (B) Cryo-SEM view of a thick biofilm area (boxed in blue in panel A). Wrinkles in the biofilm are typical of a biofilm a few micrometers thick. (C) Milled lamella (boxed in yellow) from the green boxed region in panel A. (D) A 23-nm-thick tomographic slice of a low-magnification tomogram of the lamella shown in panel C. Live (when frozen) and dead cells are visible (green and red asterisks, respectively), and large cellulose arrays can be seen filling the gaps between the cells (arrowheads). (E) Manual segmentation of the tomogram shown in panel D. (F) The fraction of the lamella volume occupied by the cells was assessed for each lamella. Six and 4 biofilms were grown for 3 h and 6 h, respectively. An unpaired t test showed a two-tailed P value of 0.0011. (G) Live cell ratio in 3 h and 6 h biofilms. Six and 4 biofilms were grown for 3 h and 6 h, respectively. An unpaired t test showed a two-tailed P value of 0.2720. (H) Violin box plots reporting the absolute depth of the live and dead cells within the biofilms grown for 3 and 6 h. The dashed red lines indicate the first and third quartiles, and solid red lines represent medians. This shows that while the biofilms get thicker with time, the ratio of live to dead cells appears constant through depth and time. The method of calculation is detailed on the left of the panel and in Materials and Methods. The lamella is drawn in blue, with the platinum-coated leading edge is in gray. n = 49, 46, 4, and 11 for live and dead cells in 3-h and 6-h biofilms, respectively. Mann-Whitney tests were performed on live versus dead cells in 3-h and 6-h biofilms, showing two-tailed P values of 0.82 and 0.54, respectively.

Lamellae of native biofilms also reveal numerous vesicles and the cortical belt. (A and B) Two tomographic slices of a G. hansenii cell from a biofilm grown for 6 h surrounded by cellulose ribbons (yellow arrowheads). The cortical belt is visible in panel B (arrow) and seems to follow the trajectory of the cellulose sheet proximal to the OM (dark-outlined yellow arrowhead). (C) Manual segmentation of the tomogram displayed in panels A and B showing the juxtaposition of the cortical belt (purple to red) and the nascent cellulose ribbon (yellow). (D) Enlargement of the boxed region in panel B showing the layered cortical belt. (E) Tomographic slice of a cell surrounded by cellulose ribbons (yellow arrowheads) from a biofilm grown for 3 h and harboring numerous vesicles in its cytosol (white arrowheads). Disorganized aggregates (dashed lines) are visible at this time point. (F and G) Tomographic slices showing additional examples of disorganized cellulose aggregates (dashed lines) surrounded by cellulose ribbons (arrowheads) visible in 3-h biofilms. Bars, 100 nm. All tomographic slices are 11 nm thick.

The cortical belt is not found in other cellulose-synthesizing species. (A) Maximum projection of A. tumefaciens cells synthesizing cellulose. Cells are stained with MitoTracker Deep Red (red) and cellulose with calcofluor white (cyan). (B) A 10-nm-thick tomographic slice of a typical A. tumefaciens cell with cellulose microfibrils around (arrowheads). No cortical belt can be seen in the cells. A polar flattening can be seen at the lower pole (cyan arrow) with an amorphous aggregate (dashed lines). These aggregates are most probably the unipolar polysaccharide (UPP) synthesized specifically by A. tumefaciens. (C) Manual segmentation of the tomogram in panel B showing the organization of the cellulose microfibrils around the cell, the absence of the cortical belt, and the putative UPP. (D) A 50-nm optical slice of an induced E. coli 1094 cellulose biofilm. Cells are stained with MitoTracker Deep Red (red) and cellulose with calcofluor white (cyan). (E) A 6-nm tomographic slice of a lamellar tomogram of a bacterial mat showing three E. coli 1094 cells and an amorphous cellulose aggregate between them (asterisk). (F) A 6-nm tomographic slice of a lamella through a bacterial mat treated with cellulase, showing multiple cells. No cellulose was visible under this condition. No cortical belt can be seen in the cells under either condition.