Overproduction of flotillin influences cell differentiation and shape in Bacillus subtilis

Abstract

Bacteria organize many membrane-related signaling processes in functional microdomains that are structurally and functionally similar to the lipid rafts of eukaryotic cells. An important structural component of these microdomains is the protein flotillin, which seems to act as a chaperone in recruiting other proteins to lipid rafts to facilitate their interaction. In eukaryotic cells, the occurrence of severe diseases is often observed in combination with an overproduction of flotillin, but a functional link between these two phenomena is yet to be demonstrated. In this work, we used the bacterial model Bacillus subtilis as a tractable system to study the physiological alterations that occur in cells that overproduce flotillin. We discovered that an excess of flotillin altered specific signal transduction pathways that are associated with the membrane microdomains of bacteria. As a consequence of this, we detected significant defects in cell division and cell differentiation. These physiological alterations were in part caused by an unusual stabilization of the raft-associated protease FtsH. This report opens the possibility of using bacteria as a working model to better understand fundamental questions related to the functionality of lipid rafts.

Importance: The identification of signaling platforms in the membrane of bacteria that are functionally and structurally equivalent to eukaryotic lipid rafts reveals a level of sophistication in signal transduction and membrane organization unexpected in bacteria. It opens new and promising venues to address intricate questions related to the functionality of lipid rafts by using bacteria as a more tractable system. This is the first report that uses bacteria as a working model to investigate a fundamental question that was previously raised while studying the role of eukaryotic lipid rafts. It also provides evidence of the critical role of these signaling platforms in orchestrating diverse physiological processes in prokaryotic cells.

FIG 1

Saturation of FloA and FloT in functional membrane microdomains of B. subtilis. Shown are levels of expression in cells expressing FloA-GFP (A) or FloT-GFP (B) under the control of an IPTG-inducible promoter (P_hp) (bottom row) compared to natural expression levels obtained with their native promoters (upper row). Cells were grown in MSgg medium until they reached the stationary phase. IPTG was added to a final concentration of 15 mM. Membrane staining (left column) was performed with FM4-64 membrane dye. The center column shows the fluorescence signal emitted by the GFP. The right column shows the merged signals. The FM4-64 signal is false colored in red, and GFP is false colored in green. Scale bars are 2 µm. (C and D) Quantification of the fluorescence signal in single cells expressing FloA-GFP and FloT-GFP. The upper row shows the fluorescence signal obtained when expressed under the control of their native promoters. The bottom row shows the fluorescence signal obtained when expressed under the control of an IPTG-inducible promoter. The fluorescent micrographs show membrane staining false colored in red and the GFP fluorescence signal false colored in green. Additionally, the relative GFP fluorescence signal is quantified in relation to the background fluorescence, using a color spectrum logarithmic scale. (The spectrum scale is presented on the bottom.) The relative fluorescence intensity values of each micrograph are represented in a graph (on the left of each micrograph). The x axis represents the cell length (in pixels), and the y axis represents the value of relative fluorescence intensity detected (in arbitrary units [AU]). The midcell is marked with a red dot. The scale bar is 1 μm.

FIG 2

Overexpression of the flotillin-like proteins FloA and FloT promotes biofilm formation via FtsH. (A) Schematic representation of how FloA and FloT influence the signaling pathway leading to matrix production. Spo0A is activated by phosphorylation (Spo0A~P) and indirectly derepresses the expression of the eps and tapA operons. Phosphorylation of Spo0A~P is driven by the action of the membrane-bound kinase KinC, which is stabilized by FloA and FloT. Spo0A~P is subjected to dephosphorylation by a pool of phosphates that are degraded by the protease FtsH. The activity of FtsH is dependent on FloA and FloT. Solid arrows denote direct regulation, and dashed arrows denote indirect regulation. The asterisk denotes mutations in PY79. (B) Colonies of diverse mutants in the PY79 strain grown in MSgg agar medium for 72 h at 30°C. (C) Effects of flotillin overexpression in colony morphology of PY79 grown on solid MSgg agar medium for 72 h at 30°C with 15 mM IPTG. The scale bar represents 0.5 cm. (D) Flow cytometry analysis to monitor the subpopulation of matrix-producing cells in different genetic backgrounds using the P_tapA-yfp reporter system. Cells were grown for 72 h on solid MSgg medium in the presence of 15 mM IPTG. Approximately 50,000 ungated cells were analyzed. Fluorescence is represented in arbitrary units (AU). ctrl, control; wt, wild type. (E) Diverse mutants in the PY79 genetic background grown on MSgg medium and incubated for 72 h at 30°C supplemented with 15 mM IPTG. The scale bar represents 0.5 cm. (F) Flow cytometry analysis to monitor the subpopulation of matrix-producing cells in different genetic backgrounds that lack FtsH using the P_tapA-yfp reporter system. Cells were grown for 72 h on solid MSgg medium in the presence of 15 mM IPTG. Approximately 50,000 ungated cells were analyzed.

FIG 3

Overproduction of FloA and FloT increase the levels of FtsH and TasA proteins. (A) Immunoblot analysis to detect FtsH in different mutants using polyclonal antibodies against FtsH. The wild-type strain is used as a positive control, while the ∆ftsH mutant is used as a negative control. The immunoblot signal is presented in the upper panel, and the respective SDS-PAGE result is presented in the lower panel as a loading control. (B) Detection of TasA production in different mutants by immunoblot analysis using polyclonal antibodies against TasA. The wild-type strain (PY79) is used as a positive control, while the ∆tasA mutant is used as a negative control. The immunoblot signal is presented in the upper panel, and the respective SDS-PAGE result is presented in the lower panel. Samples were obtained from the extracellular protein fraction of pellicles that were grown in liquid cultures for 24 h at 30°C. Protein levels were normalized relative to cell number.

FIG 4

Overexpression of FloA and FloT affects cell shape. (A) Fluorescence micrographs of strains overexpressing FloA, FloT, or both FloA and FloT compared to wild-type cells. The upper row presents membrane staining using FM4-64. The center row shows DNA staining using Hoechst 33342. The bottom row shows the merge of the previous images, with the membrane staining false colored in red and the DNA staining false colored in blue. Cells were grown in liquid MSgg medium at 37°C until they reached the late exponential growth phase. IPTG was added to a final concentration of 15 mM. The scale bar represents 2 µm. (B) Histogram representing the variation in cell length in the different mutants. The number of the cells considered for this analysis was 500. The cell count is represented in the y axis. Calculation of cell length was performed using Leica Application Suite Advance Fluorescence software. (C) Detailed view of cell shape aberrations in cells that simultaneously overexpressed FloA and FloT. Spherical cells are shown in panel I. Spherical anucleate cells are shown in panel II. Elipsoid cells are shown in panel III, and cells with a failure in septation are shown in panels IV and V. The scale bars represent 2 µm.

FIG 5

Simultaneous overproduction of FloA and FloT affects formation of the division septum and formation of the Z-ring. (A) Comparison between wild-type cells and the P _hpFloA P _hpFloT strain during the early exponential growth phase. Cells were grown in liquid MSgg medium at 37°C until they reached an OD₆₀₀ of 0.1. For overexpression, the medium was supplemented with 15 mM IPTG. The membrane (membr.) was stained with FM4-64 (left panel), and the DNA was visualized with Hoechst 33342 (middle panel). The right panel shows merges of the two signals, with the membrane false colored in red and DNA false colored in blue. The scale bar represents 2 µm. (B) Visualization of Z-ring formation in a wild-type strain and the P_hpFloA P_hpFloT strain bearing an FtsZ-GFP fusion. The first column shows the membrane stained with FM4-64, the second column shows the DNA stained with Hoechst 33342, and the third column shows the FtsZ-GFP signal. In the fourth column, all signals were merged, with the membrane false colored in red, the DNA false colored in blue, and the GFP false colored with green. The cells were grown in liquid MSgg medium at 37°C and harvested in the exponential growth phase. IPTG was added to the growth medium at a final concentration of 15 mM, and basal induction of FtsZ-GFP was achieved with 1% (wt/vol) xylose. The scale bar is 2 µm. (C) Graphical illustration of the length of cells with a visible Z-ring. Approximately 500 cells were used for counting. (D) Graphical illustration of the percentage of cells with a Z-ring; approximately 500 cells were used for counting.

FIG 6

Levels of EzrA are influenced by FtsH. (A and B) Western blot analysis of whole-cell extracts of different mutants to detect the level of EzrA-GFP using polyclonal antibodies against GFP. The extra protein band is shown at 25 to 30 kDa in panel B, which was only detected in the extract of cells that overproduced FtsH. SDS-PAGE results are shown as loading controls. The protein amount was normalized relative to cell number.

FIG 7

Hypothetical model for the dual role of FtsH activity in B. subtilis membrane microdomains. In the membrane microdomains (green), the flotillins FloA and FloT regulate the protease activity of FtsH. The upper panel represents a situation with FtsH degrading regulatory phosphatases, which ultimately leads to biofilm formation via activation of the TasA operon. During cell division (lower panel), FtsH might also degrade the negative regulator for Z-ring formation, EzrA, by a yet unknown mechanism (referred to as dashed lines) and thus contribute indirectly to formation of the division septum.