• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. Sep 2006; 188(18): 6652–6660.
PMCID: PMC1595480

Granular Layer in the Periplasmic Space of Gram-Positive Bacteria and Fine Structures of Enterococcus gallinarum and Streptococcus gordonii Septa Revealed by Cryo-Electron Microscopy of Vitreous Sections


High-resolution structural information on optimally preserved bacterial cells can be obtained with cryo-electron microscopy of vitreous sections. With the help of this technique, the existence of a periplasmic space between the plasma membrane and the thick peptidoglycan layer of the gram-positive bacteria Bacillus subtilis and Staphylococcus aureus was recently shown. This raises questions about the mode of polymerization of peptidoglycan. In the present study, we report the structure of the cell envelope of three gram-positive bacteria (B. subtilis, Streptococcus gordonii, and Enterococcus gallinarum). In the three cases, a previously undescribed granular layer adjacent to the plasma membrane is found in the periplasmic space. In order to better understand how nascent peptidoglycan is incorporated into the mature peptidoglycan, we investigated cellular regions known to represent the sites of cell wall production. Each of these sites possesses a specific structure. We propose a hypothetic model of peptidoglycan polymerization that accommodates these differences: peptidoglycan precursors could be exported from the cytoplasm to the periplasmic space, where they could diffuse until they would interact with the interface between the granular layer and the thick peptidoglycan layer. They could then polymerize with mature peptidoglycan. We report cytoplasmic structures at the E. gallinarum septum that could be interpreted as cytoskeletal elements driving cell division (FtsZ ring). Although immunoelectron microscopy and fluorescence microscopy studies have demonstrated the septal and cytoplasmic localization of FtsZ, direct visualization of in situ FtsZ filaments has not been obtained in any electron microscopy study of fixed and dehydrated bacteria.

Gram-positive bacteria possess a cell envelope composed of a plasma membrane and a thick network of peptidoglycan, secondary acids, and proteins, which maintains cell shape and provides resistance to osmotic pressure (39). Electron microscopy studies performed on fixed (either chemically or by freezing) and dehydrated specimens favored the widely accepted model of the cell envelope consisting of the peptidoglycan network in direct contact with the plasma membrane (4). Nevertheless, the existence of a periplasmic space separated from the cytoplasm and the outer medium by a plasma membrane and a thick cell wall has been suggested by previous biochemical studies (3, 14, 25). Recent data obtained with cryo-electron microscopy of vitreous sections (CEMOVIS), a method that allows the observation of biological specimens in the closest-to-native state, challenged the classical model (23, 24). CEMOVIS consists of cooling the specimen at high pressure so rapidly that water cannot crystallize and instead becomes vitreous. Water stays liquid, but its viscosity dramatically increases (10). It is then possible to make thin sections at low temperatures and observe them using a cryo-electron microscope (1). Consequently, water does not need to be removed from the sample, and molecular aggregation is suppressed (11). Moreover, CEMOVIS is performed in the absence of any stain so that contrast in the image is proportional to density variations in the sample. Using this technique, Matias and Beveridge previously demonstrated that Bacillus subtilis 168 and Staphylococcus aureus possess a periplasmic space (termed the inner wall zone [IWZ]) lying between the plasma membrane and the mature peptidoglycan (termed the outer wall zone [OWZ]) in which molecules should be able to freely diffuse (23, 24). Those authors proposed that the low-density periplasmic space seen with CEMOVIS corresponds to the electron-dense innermost layer observed in stained sections of freeze-substituted gram-positive bacteria (15). Accordingly, they suggested that the periplasmic space is filled with a low-density material that has a high affinity for heavy metal stains (23).

The general pathway of peptidoglycan synthesis is as follows: (i) generation of nucleotide sugar-linked precursors UDP-N-acetylmuramyl-pentapeptide and UDP-N-acetylglucosamine in the cytoplasm, (ii) dimerization and translocation through the plasma membrane, and (iii) cross-linking of the periplasmic precursors with mature peptidoglycan via transglycosylation and transpeptidation reactions. This last step is performed by a set of membrane-bound proteins, termed the penicillin-binding proteins (PBPs) (37, 40). Several studies have shown that translocated peptidoglycan precursors and PBPs are confined to specific areas of the bacterial cell (8, 27, 31, 32, 36). In particular, most gram-positive bacteria have them concentrated at the septum, indicating that polymerization of the mature peptidoglycan happens at least partly in this area (37). The septum is a cellular structure consisting of a ring of invaginating cell envelope. In the vast majority of bacterial species, its development, eventually leading to cell division, is orchestrated by FtsZ, a protein homologous to eukaryotic tubulin (22, 44). FtsZ is expected to form a ring beneath the septal plasma membrane, although such structures have never been observed by any electron microscopy study (5, 45).

Because PBPs are bound to the plasma membrane, it is generally considered that polymerization takes place at the contact between the plasma membrane and the mature peptidoglycan. Evidence for the presence of a wide periplasmic space (22.3 nm in B. subtilis and 15.8 nm in S. aureus) raises questions about the spatial organization and the mechanism of assembly of peptidoglycan. In their recent publication, Matias and Beveridge reported that at least part of the septum of S. aureus possesses a periplasmic space (24). The deeper part of the septum where the plasma membrane is curved, which we define as the constriction ring, was not investigated at high resolution. Whether mature peptidoglycan is in contact with the plasma membrane at this place thus remained an open question, which we have investigated.

We report the structure of the cell envelope of three gram-positive bacteria (B. subtilis, Enterococcus gallinarum, and Streptococcus gordonii) revealed by CEMOVIS. In the three cases, we observed a previously undescribed granular layer (GL) adjacent to the plasma membrane in the periplasmic space. The structure of peptidoglycan assembly sites in the coccus species (E. gallinarum and S. gordonii) suggests a common mechanism of peptidoglycan cross-linking. We propose that nascent peptidoglycan diffuses into the periplasmic space and is cross-linked by components of the granular layer to form mature peptidoglycan. Finally, CEMOVIS revealed cytoplasmic structures associated with the constriction ring, which could be interpreted as being FtsZ complexes.


Culture conditions.

B. subtilis 168 and B. subtilis W23 were cultured in TS medium [1.4% K2HPO4, 0.6% KH2PO4, 0.2% (NH4)2SO4, 0.1% Na3-citrate · 2H2O, 0.02% MgSO4 · 7H2O, 5 μM MnSO4, 0.5% glucose, 0.5% Na-glutamate, and 0.001% yeast extract prepared in bidistilled water] (41) at 37°C with aeration. Cells were harvested after the end of the exponential growth phase (optical density at 600 nm [OD600] of about 2.0). S. gordonii Challis (33) and E. gallinarum DSMZ 20628 (DSMZ, Braunschweig, Germany) cells were cultured in brain heart infusion (BHI) broth (Oxoid Ltd., Hampshire, United Kingdom) at 37°C without aeration. S. gordonii cells were harvested in the middle of the exponential growth phase (OD620 of about 0.5). E. gallinarum cells were harvested after the end of the exponential growth phase (OD600 of about 1.6). In control experiments, S. gordonii cells were grown in BHI broth supplemented with 10% glycerol (10% glycerol-BHI) and harvested in the middle of the exponential growth phase (OD620 of about 0.5).

Vitrification and cryosectioning.

B. subtilis and S. gordonii cells were washed twice in phosphate-buffered saline (PBS) supplemented with 20% dextran (20% dextran-PBS) (average mass, 40 kDa; Sigma-Aldrich, Buchs, Switzerland). E. gallinarum cells were washed twice in water supplemented with 20% dextran. Cells were then introduced into copper tubes and high-pressure frozen with an EM PACT high-pressure freezer (Leica, Vienna, Austria). The tubes were trimmed at −140°C in an FCS Ultracut S cryomicrotome (Leica). Ultrathin sections were produced with a 45° diamond cryo-knife (Diatome, Biel, Switzerland) at a nominal thickness of 50 nm and at a nominal cutting feed of 1 mm/s.

For control experiments, S. gordonii cells were washed twice in 10 mM HEPES (pH 7.0) supplemented with 10% glycerol (10% glycerol-HEPES). After the last centrifugation, the excess medium was absorbed with filter paper. The concentrated pellet was introduced into a sandwich aluminum carrier (sample cavity depth, 200 μm; Bal-Tec, Balzers, Liechtenstein). Empty space was filled with 1-hexadecene. Cells were then vitrified using an HPM 010 high-pressure freezer (Bal-Tec). At −140°C, the cells were mounted onto a specimen holder covered with cryoglue (mixture of ethanol-isopropanol [ratio, 1:3]) (34). Blocks were trimmed, and ultrathin sections were produced at −160°C.

Cryo-electron microscopy.

Sections were transferred onto carbon-covered 1,000-mesh copper grids (Agar Scientific, Essex, United Kingdom). Grids were transferred to a Gatan cryoholder (Gatan, Warrendale, PA) kept at a temperature below −170°C and inserted into a CM100 cryo-electron microscope (Philips, Eindhoven, The Netherlands) equipped with an LaB6 cathode. The accelerating voltage was either 80 or 100 kV. Specimens were irradiated with a low electron dose. Electron diffraction was used to check whether water was vitreous or crystalline. Crystalline sections were discarded. Images were recorded with a TemCam-F224HD charge-coupled-device camera (Tietz Video and Image Processing Systems, Munich, Germany) at magnifications ranging from ×1,350 to ×22,500. No image processing other than that described in the figure legends was performed.

Dimension measurements.

Pixel size and post magnification factor were calibrated by using a two-dimensional catalase crystal (Agar Scientific). Dimensions were measured using images recorded at a magnification of ×22,500. At this magnification, the post magnification factor is 1.69 and the pixel size is 0.63 nm. The horizontal resolution is therefore approximately 1.3 nm. During cryosectioning, material was compressed along the cutting direction. We could not precisely determine the average compression rate because the studied specimens were not perfectly spherical. By analogy to our previous study, where we used similar cutting conditions, we estimate the average compression rate to be 0.2 to 0.3 (46). Nevertheless, it has been shown that compression does not affect dimensions measured perpendicularly to the cutting direction (11). Dimensions were measured accordingly with ImageJ software (NIH, Bethesda, MD). Average-density profiles were recorded along rectangular selections with ImageJ. The width of selection rectangles is specified in the figure legends.


Aspect of vitreous sections.

A typical section of S. gordonii cells vitrified in 20% dextran-PBS is shown in Fig. Fig.1.1. It is devoid of crevasses, chatter, and cracks, which are the most severe artifacts induced by cryosectioning and which complicate image interpretation because they are not homogenous in space. Only homogenous deformation, namely, compression along the cutting direction and knife marks, cannot be avoided. A low level of ice contaminations lies on the surface of the section.

FIG. 1.
Low-magnification view of a cryosection of S. gordonii. Arrows, knife marks; asterisks, ice contamination; arrowhead, lysed cell. Bar, 1 μm.

Granular layer in the periplasmic space.

The structures of the envelopes of three representative species of gram-positive bacteria (S. gordonii, B. subtilis, and E. gallinarum) are shown in Fig. Fig.2.2. Images are two-dimensional projections of three-dimensional sections, and for geometrical reasons, the envelope is clearly visible, and membranes look sharp only when they are aligned with the electron beam (see reference 46 for a schematic explanation). Therefore, an apparent discontinuity of a membrane (e.g., bottom of Fig. Fig.2A)2A) is most likely due to a change in the orientation of the plasma membrane. This statement is supported by stereo-pair images (data not shown). As described previously by Matias and Beveridge, who used CEMOVIS, the envelope consists of a plasma membrane, a low-density IWZ identified as a periplasmic space, and a high-density OWZ corresponding to the network of cell wall polymers (23, 24). In addition, a previously unreported dense GL in the IWZ is visible in most of our images and in each of the investigated species. It is more contrasted and therefore denser in S. gordonii (Fig. 2A and B) and B. subtilis (strains W23 [Fig. [Fig.2C]2C] and 168 [data not shown]) than in E. gallinarum (Fig. 2D and E). Images were recorded with defocus values ranging from −1.5 μm to −5.7 μm, which corresponds to the first minima of the contrast transfer function in the range from 2.3 nm to 4.7 nm. A GL is visible over this whole range, indicating that it is not an optical artifact. The GL is 3 to 4 nm thick and is 8 to 9 nm distant from the plasma membrane. Measurements of thickness and the positions of the envelope structures are summarized in Table Table1.1. In S. gordonii cultures, a small proportion of lysed cells are present (Fig. (Fig.1,1, arrowhead). Similarly to observations in osmotically lysed cells described previously by Matias and Beveridge, the IWZ thickness is larger and highly variable in every single lysed cell, suggesting that the plasma membrane and OWZ are not bound anymore (Fig. (Fig.2F)2F) (23, 24). Also, in agreement with the observations reported previously by those authors, the OWZ is thicker in lysed cells than in normal cells (Table (Table1).1). This increase is consistent with lysis-induced release of osmotic pressure acting on the OWZ, allowing the relaxation of cell wall polymers (23, 24). In some of the lysed cells, a GL is parallel to the plasma membrane (Fig. (Fig.2F).2F). The GL is slightly closer to the plasma membrane in lysed cells than in normal cells, but its thickness and granular aspect are identical in both kinds of cell (Table (Table11).

FIG. 2.
Structure of the cell envelope of three gram-positive species. (A) S. gordonii. (B) Higher magnification of the square in A. (C) B. subtilis W23. (D and E) E. gallinarum. (F) S. gordonii lysed cell. (G) S. gordonii cell grown and vitrified in the presence ...
Dimensions of cell envelope structures

Dextran does not induce the granular layer.

Dextran was used as a cryoprotectant because it produces a minimal osmotic effect due to its high molecular weight, and it provides good cryosectioning properties (2). In order to rule out the possibility that the previously undescribed GL is an artifact due to the sudden addition of dextran to the extracellular fluid, S. gordonii cells were grown in medium containing 10% glycerol as described previously (23, 24) and high-pressure frozen in a buffered solution containing the same amount of this cryoprotectant. S. gordonii cells grew similarly with or without glycerol. The only difference was a slight decrease in the growth rate (doubling time, 37 min without glycerol; doubling time, 48 min with glycerol). Cryosections of good quality were rarely obtained when glycerol was used instead of dextran. However, in the regions devoid of the severe artifacts that we could occasionally obtain, the cell envelope is the same as that in S. gordonii cells cryoprotected with 20% dextran (Fig. (Fig.2G2G).

Structure of the outer wall zone.

B. subtilis 168 IWZ and OWZ thicknesses are similar to previously published values (Table (Table1)1) (23). The density of the OWZ fades from the inner face to the outer face in B. subtilis (168 and W23) (Fig. (Fig.2I)2I) and, less steeply, in S. gordonii (Fig. (Fig.2H).2H). The structure of the E. gallinarum OWZ is different: it is subdivided into three layers of different densities (Fig. (Fig.2J).2J). This feature is best visible in Fig. Fig.2E.2E. The first layer (OWZ1) is close to the inner face, the second layer (OWZ2) corresponds to the central portion, and the third layer (OWZ3) is close to the outer face of the OWZ. The central portion is less dense than the first and the third portions. The thickness of OWZ1 is smaller and more constant among individuals than the thicknesses of OWZ2 and OWZ3 (Table (Table11).

Structure of the E. gallinarum septum.

Is the structure of the cell envelope different in cellular domains involved in peptidoglycan cross-linking? In E. gallinarum, peptidoglycan precursors are translocated at the septum (T. Ribeiro, M. Ruivo, and F. Lopes, unpublished data). A cell with septa at two different steps of formation is shown in Fig. Fig.3A.3A. High-magnification views of septa of this cell and other cells are shown in the rest of Fig. Fig.3.3. The OWZ enters the septum already at the initial step of its formation (Fig. (Fig.3B)3B) and develops during constriction of the septum (Fig. 3C, D, and E). The GL and IWZ also enter the septum (best seen in Fig. 3D and E). The OWZ is in contact with the plasma membrane at the constriction ring, and no space remains in between for the IWZ and GL (Fig. (Fig.3,3, arrowheads). In some septa, OWZ density is uniform (Fig. (Fig.3D);3D); in others, the central layer of the OWZ is less dense than the layers closer to the surfaces (Fig. 3A and E). The septal IWZ is thinner than the regular IWZ by 9.5% (P < 0.01). In contrast, the septal OWZ is thicker than the regular OWZ by 25.1% (P < 0.01).

FIG. 3.
E. gallinarum septum. (A) Dividing coccus. Arrow, newly forming septum. (B to D) Progressive formation of septum. Arrowhead, constriction ring. (E) Higher magnification of the rectangular area in A. Asterisks, ice contamination. See the legend to Fig. ...

Structure of the S. gordonii septum.

In Streptococcus pneumoniae, PBPs responsible for the cross-linking of peptidoglycan are concentrated at the equatorial ring and at the septum (26, 27). We investigated their structure in S. gordonii, a close nonpathogenic relative of S. pneumoniae (18). S. pneumoniae and S. gordonii are ovoid (43). The equatorial ring is identified by an increased curvature of the plasma membrane and is approximately halfway between the septum and the cell pole (6, 43). The GL, IWZ, and OWZ are not different at the equatorial ring (Fig. (Fig.4A).4A). Contrary to the situation in E. gallinarum and similarly to S. pneumoniae, the cell envelope is not deeply invaginated at the level of the S. gordonii septum (Fig. (Fig.4)4) (38, 43). Accordingly, the septal OWZ is most probably split in two simultaneously with the addition of new components on its inner surface. The GL does not penetrate the septum but is slightly concave (Fig. 4B and C). The OWZ is also slightly concave, but it does not cross the GL and is excluded from the septum. Only the IWZ penetrates the septum, where its density is higher.

FIG. 4.
S. gordonii equatorial ring and septum. (A) Dividing coccus. A well-defined equatorial ring is shown between arrows. (B and C) Septa of other cocci. See the legend to Fig. Fig.22 and the text for definitions of the abbreviations. A model of the ...

Cytoplasmic structures at the constriction ring of E. gallinarum.

At the level of the constriction ring of the E. gallinarum septum, dense elements are found below the plasma membrane. In some cases, they have the aspect of concentrated dots, without higher-order organization (Fig. (Fig.5A).5A). In others, the dots are assembled in one or two layers parallel to the plasma membrane (Fig. 5B, C, and D, arrowheads). The density profile of such a structure is shown in Fig. Fig.5G.5G. The dots are 3 to 5 nm in diameter. The distance between the plasma membrane and the first cytoplasmic layer as well as the distance between the first and the second cytoplasmic layers is between 7 and 12 nm. Inside the layers, the distance between the center of mass of the dots lies between 5 and 7 nm. Finally, in some images, weakly contrasted cytoplasmic filaments are situated between the two edges of the septum (Fig. 5E and F). Figure Figure5G5G shows a density profile recorded across four filaments of Fig. Fig.5E.5E. These are 3 to 6 nm in diameter, i.e., similar to the diameter of the dots.

FIG. 5.
E. gallinarum cytoplasmic structures associated with the constriction ring. (A to D) Arrows indicate concentrated dots without higher-order organization; white arrowheads indicate dots assembled in parallel layers. (E and F) Black and white arrowheads ...


In every species investigated in the present study, the previously reported IWZ and OWZ were observed. This confirms the new model of the structure of the gram-positive bacterial cell envelope (23). Besides differences in IWZ and OWZ thicknesses, subtle differences in mass distribution within the OWZ exist between species. In B. subtilis, the density of the OWZ decreases from the inner to the outer face, which is consistent with a centrifuge turnover of the cell wall (4, 20). According to this model, nascent polymers are inserted on the inner face of the cell wall, where they are not stretched. Due to a continuous addition of new material, they are gradually displaced towards the exterior of the wall. During the first step of this displacement, the polymers become stretched. Consequently, they provide resistance against osmotic pressure but also become more prone to cleavage by autolysins. This leads to reduced cross-linking and eventually to the release of peptidoglycan fragments in the extracellular medium. This mechanism, in combination with the fact that the cell wall is inert (i.e., not synthesized or degraded) at cell poles, is thought to be at the origin of the rod shape (19, 20). In S. gordonii, the OWZ also fades but in a less pronounced manner, reflecting the probably reduced centrifuge turnover due to the incorporation of wall polymers in very restricted sites. A similar difference between B. subtilis and S. aureus was previously noticed and interpreted likewise by Matias and Beveridge (23). Indeed, in S. aureus, the cell wall is polymerized exclusively at the septum (31). In E. gallinarum, whose peptidoglycan polymers are also mostly assembled at the septum (Ribeiro et al., unpublished), the center of the OWZ is less dense than its surfaces, suggesting yet another mode for the regulation of wall polymer degradation.

We report a dense GL located in the IWZ, which was not seen previously. The GL is observed over a wide range of defocus values, including close to focus (Fig. (Fig.2C).2C). This demonstrates that the GL is not an optical artifact of phase contrast. To confirm that the GL was not induced by osmotic shock due to the sudden change in dextran concentrations, we adapted the protocol used previously by Matias and Beveridge (23, 24): S. gordonii cells were grown in 10% glycerol-BHI and vitrified in 10% glycerol-HEPES. The GL is also visible under these conditions, indicating that the GL is certainly a native feature of the cell envelope. The reason why the GL was not discovered by Matias and Beveridge (23, 24) is perhaps related to the use of glycerol, which prevents the routine production of crevasse-free sections. Lysed cells of S. gordonii reveal that the GL is retained by the plasma membrane, suggesting that GL components are anchored in the plasma membrane (Fig. (Fig.2F).2F). The only exception is found at the septum of (nonlysed) S. gordonii, where the GL is aligned not with the plasma membrane but with the OWZ (Fig. 4B and C and and6B).6B). A possible interpretation is that under normal conditions, the GL bridges the plasma membrane to the mature peptidoglycan; under special conditions, one of the connections can be broken.

FIG. 6.
Model septa of (A) E. gallinarum and (B) S. gordonii. The position of peptidoglycan precursors is hypothetic and based on considerations explained in the Discussion. C, cytoplasm. For definitions of other abbreviations, see the legend to Fig. ...

In order to better understand how nascent polymers are incorporated into the OWZ, we investigated areas of cell wall assembly. In B. subtilis, these areas have been identified as the septum and in a helical structure along the rod (7, 36). With CEMOVIS, it was not possible to routinely observe B. subtilis assembly sites because on the one hand, rod cells align longitudinally along the long axis of the freezing tube when they are introduced into the tube (23). Consequently, only cross-sections of the cells and almost no longitudinal sections are obtained, making the observation of septa in rod cells virtually impossible. On the other hand, the helical pattern of assembly is not well defined morphologically.

Nevertheless, CEMOVIS revealed the close-to-native structure of peptidoglycan assembly sites of E. gallinarum in the stationary state (septum) and of S. gordonii in the exponential state (septum and equatorial ring), whose shape is closer to a sphere, allowing the random orientation of cells in the freezing tube. The structure of the cell envelope at the S. gordonii equatorial ring is not different than that in the rest of the cell. The structures of the E. gallinarum and S. gordonii septa are modeled in Fig. Fig.66 according to the observations shown in Fig. Fig.33 and and4.4. It is unlikely that these two cartoons represent two successive steps of one septum maturation process, since we have never observed the typical wide OWZ-containing septum of E. gallinarum in S. gordonii or the characteristic narrow OWZ-free septum of S. gordonii in E. gallinarum. The structures of the three peptidoglycan assembly sites studied are not identical but can be interpreted with a common mechanism. In the three cases, the OWZ does not cross the GL. The OWZ is therefore always exposed to the outer face of the GL. Moreover, the GL is always in the vicinity of the OWZ. Together, these data suggest that after translocation through the plasma membrane in the periplasmic space (IWZ), peptidoglycan precursors could diffuse relatively freely without polymerizing until they interact with the interface between the GL and the OWZ. Precursors could then polymerize with the OWZ. In the case of the S. gordonii septum, the high density of the septal IWZ situated below the GL might represent a high concentration of unpolymerized precursors (Fig. 4B and C). In the E. gallinarum septum, the OWZ reaches the plasma membrane in the constriction ring (Fig. (Fig.6A).6A). At this site, peptidoglycan precursors are exposed to the outer face of the GL immediately after translocation. On the basis of these considerations, the GL could contain PBPs, the enzymes responsible for peptidoglycan cross-linking. The X-ray structure indicates that PBPs protrude 9 to 13 nm into the periplasmic space, which is consistent with the distance between the centers of mass of the GL and of the plasma membrane (21, 29). PBPs are not distributed homogenously over the plasma membrane; different members of the PBP family are concentrated in specific areas (8, 26, 27, 32). The GL is, however, homogenously distributed over the whole cell (except in the situations described above). It is therefore most probable that the GL is not only composed of the various PBPs. Other molecules interacting with the plasma membrane, such as lipoproteins and lipoteichoic acids, could compose the GL. It may also contain molecules that are not bound to the plasma membrane, as suggested by the structure of the GL at the S. gordonii septum (Fig. (Fig.6B).6B). Interestingly, electron microscopy experiments using labeled penicillin in S. aureus cells indicated that a fraction of the PBPs may not be bound to the plasma membrane (30). Since vitreous sections cannot be immunolabeled, targeted gene deletions or selective chemical extractions might contribute to the characterization of the molecular composition of the GL.

The layered structure of the gram-positive envelope raises questions. It is well accepted that the plasma membrane is the diffusion barrier between the cell interior and the outer world. It is subjected to the effect of the osmotic pressure, but it does not have the mechanical strength to hold it. The peptidoglycan layer of the cell wall (OWZ) probably confers the mechanical resistance (39). However, the low-density zone of the IWZ, in which we propose that envelope precursors can diffuse, separates the plasma membrane from the OWZ. How is the force that is generated at the plasma membrane reported to the OWZ through the IWZ? One possibility, proposed previously by Matias and Beveridge, is that rigid molecules keep the OWZ distant from the plasma membrane by forming a scaffold in between (24). Alternatively, pressure could be transferred from the plasma membrane to the base of the OWZ by a gel composed of osmotically active polymers freely floating in the IWZ. Their osmotic effect would preclude that the IWZ volume decreases beyond a threshold value reached when the periplasmic osmotic pressure equals the cytoplasmic osmotic pressure. The osmotically active polymers should be large enough not to diffuse through the plasma membrane or through the OWZ, and they should be flexible enough (short persistence length) to be osmotically active at relatively low concentrations, which would be consistent with the low density of the IWZ (9). Regarding this hypothesis, it is noteworthy that the cytoplasm and periplasm of gram-negative bacteria are isosmotic (42). Obviously, the cell envelope still has many facets to reveal until its structure and functions are fully understood.

CEMOVIS revealed cytoplasmic structures at the constriction ring of the E. gallinarum septum. They are strongly contrasted dots (Fig. 5A to D) or weakly contrasted lines (E and F) of 3 to 6 nm in diameter. The signal is far above the noise level, as is demonstrated by two successive images recorded in the same area (data not shown). Dots of similar sizes are found everywhere in the cell but not at this high concentration nor with this layered organization. Weak lines were not found anywhere other than in the septal area. These structures are thus specific for the septum. In S. gordonii, the occurrence of such structures was rarely observed.

The contrasted dots are present in images where the plasma membrane in the constriction ring (i.e., the bottom of the septum) is sharp or relatively sharp, indicating that these sections cross the center of the septum (shown in yellow in Fig. Fig.5I).5I). The weak lines are found in images where the plasma membrane is not defined in the constriction ring, implying that these sections are cut tangentially to the constriction ring (shown in blue in Fig. Fig.5I).5I). Note that membranes and cell wall layers that are close to parallel to the section plane are expected to slightly darken the corresponding area of the image but are not expected to produce any sharp structure (46). The blurred aspect of the lateral invaginated membranes indicates that they are slightly tilted, but this tilt cannot account for the blurred aspect of the membrane in the constriction ring. Together, these data are consistent with a ring of cytoplasmic filaments localized near the plasma membrane (red ring in Fig. Fig.5I).5I). Filaments are indeed seen with CEMOVIS as contrasted dots in top views and as weak lines in longitudinal views (12, 35). From this assumption, we speculate that the observed structure may represent polymerized FtsZ proteins. FtsZ is a prokaryotic cytoskeletal protein that is homologous to eukaryotic tubulin and is thought to be organized as a cytoplasmic ring of filaments beneath the plasma membrane (5). It is essential for septation and serves as a scaffold for proteins involved in the division process (22, 37). Although immunoelectron microscopy and fluorescence microscopy studies have proven its septal localization and suggested a ring organization, FtsZ filaments have never been observed in thin sections of fixed and dehydrated bacteria (45). Most probably, FtsZ collapses during dehydration steps and becomes indiscernible from the plasma membrane. In spite of that, an electron microscopy study of isolated FtsZ has revealed that it can polymerize in vitro as (i) straight single protofilaments, (ii) two-dimensional sheets of protofilaments, or (iii) curved protofilaments (13). The cytoplasmic dots concentrated near the constriction ring observed with CEMOVIS (Fig. (Fig.5A,5A, arrows) can be interpreted as cross-sectioned protofilaments; the layers of dots parallel to the plasma membrane may represent cross-sectioned two-dimensional sheets of protofilaments (Fig. 5B, C, and D). Supporting this hypothesis is the agreement of the 5.3-nm distance between protofilaments in vitro and the 5- to 7-nm distance between dots within a single layer in close-to-native cells (13). A three-dimensional representation of the cytoplasmic structures is necessary to confirm our interpretation (by determining the curvature of the filaments and their relationship with the plasma membrane, for example), and the developing technique of cryo-electron tomography of vitreous sections should make it possible to solve the structures in the next few years (16, 17, 28). Obtaining tomograms of vitreous sections at 3-nm resolution will certainly require the thinnest possible sections. Therefore, the probability of having a complete septal ring in a section will be very low. Moreover, it might be very difficult to detect such a configuration, because sections must be screened at a low magnification to minimize the electron dose. Such a cell slice might lack highly contrasted characteristic features visible at low magnifications. However, tomograms of sections containing an important part of septal ring, as shown in Fig. 5E and F, should be sufficient to measure the curvature of the filaments and their relationship with the plasma membrane. If this prediction comes true, it will be possible to gain structural insights into the cell division macromolecular machinery at work, and this will most likely help to solve its mechanism.


This work was supported by the 3D-Electron Microscopy Network of European Research Framework Program 6 (to J.D.), the Fundação para a Ciência e Tecnologia through project POCI/CVT/59636/2004, FEDER, and the Fundação Caluste Gulbenkian (to T.R. and F.L.).


1. Al-Amoudi, A., J. J. Chang, A. Leforestier, A. McDowall, L. M. Salamin, L. P. Norlen, K. Richter, N. Sartori Blanc, D. Studer, and J. Dubochet. 2004. Cryo-electron microscopy of vitreous sections. EMBO J. 23:3583-3588. [PMC free article] [PubMed]
2. Al-Amoudi, A., L. P. Norlen, and J. Dubochet. 2004. Cryo-electron microscopy of vitreous sections of native biological cells and tissues. J. Struct. Biol. 148:131-135. [PubMed]
3. Beveridge, T. J. 1995. The periplasmic space and the periplasm in gram-positive and gram-negative bacteria. ASM News 61:125-130.
4. Beveridge, T. J. 2000. Ultrastructure of gram-positive cell walls, p. 3-10. In V. A. Fischetti (ed.), Gram-positive pathogens. ASM Press, Washington, D.C.
5. Bi, E. F., and J. Lutkenhaus. 1991. FtsZ ring structure associated with division in Escherichia coli. Nature 354:161-164. [PubMed]
6. Briles, E. B., and A. Tomasz. 1970. Radioautographic evidence for equatorial wall growth in a gram-positive bacterium. Segregation of choline-3H-labeled teichoic acid. Cell Biol. 47:786-790. [PMC free article] [PubMed]
7. Daniel, R. A., and J. Errington. 2003. Control of cell morphogenesis in bacteria: two distinct ways to make a rod-shaped cell. Cell 113:767-776. [PubMed]
8. Daniel, R. A., E. J. Harry, and J. Errington. 2000. Role of penicillin-binding protein PBP 2B in assembly and functioning of the division machinery of Bacillus subtilis. Mol. Microbiol. 35:299-311. [PubMed]
9. de Gennes, P. G. 1979. Scaling concepts in polymer physics. Cornell University Press, Ithaca, N.Y.
10. Dubochet, J., M. Adrian, J. J. Chang, J. C. Homo, J. Lepault, A. W. McDowall, and P. Schultz. 1988. Cryo-electron microscopy of vitrified specimens. Q. Rev. Biophys. 21:129-228. [PubMed]
11. Dubochet, J., and N. Sartori Blanc. 2001. The cell in absence of aggregation artifacts. Micron 32:91-99. [PubMed]
12. Eltsov, M., and J. Dubochet. 2005. Fine structure of the Deinococcus radiodurans nucleoid revealed by cryoelectron microscopy of vitreous sections. J. Bacteriol. 187:8047-8054. [PMC free article] [PubMed]
13. Erickson, H. P., D. W. Taylor, K. A. Taylor, and D. Bramhill. 1996. Bacterial cell division protein FtsZ assembles into protofilament sheets and minirings, structural homologs of tubulin polymers. Proc. Natl. Acad. Sci. USA 93:519-523. [PMC free article] [PubMed]
14. Gould, A. R., B. K. May, and W. H. Elliott. 1975. Release of extracellular enzymes from Bacillus amyloliquefaciens. J. Bacteriol. 122:34-40. [PMC free article] [PubMed]
15. Graham, L. L., and T. J. Beveridge. 1994. Structural differentiation of the Bacillus subtilis 168 cell wall. J. Bacteriol. 176:1413-1421. [PMC free article] [PubMed]
16. Hsieh, C. E., A. Leith, C. A. Mannella, J. Frank, and M. Marko. 2006. Towards high-resolution three-dimensional imaging of native mammalian tissue: electron tomography of frozen-hydrated rat liver sections. J. Struct. Biol. 153:1-13. [PubMed]
17. Hsieh, C. E., M. Marko, J. Frank, and C. A. Mannella. 2002. Electron tomographic analysis of frozen-hydrated tissue sections. J. Struct. Biol. 138:63-73. [PubMed]
18. Kawamura, Y., X. G. Hou, F. Sultana, H. Miura, and T. Ezaki. 1995. Determination of 16S rRNA sequences of Streptococcus mitis and Streptococcus gordonii and phylogenetic relationships among members of the genus Streptococcus. Int. J. Syst. Bacteriol. 45:406-408. [PubMed]
19. Koch, A. L. 2003. Were gram-positive rods the first bacteria? Trends Microbiol. 11:166-170. [PubMed]
20. Koch, A. L., and R. J. Doyle. 1985. Inside-to-outside growth and turnover of the wall of gram-positive rods. J. Theor. Biol. 117:137-157. [PubMed]
21. Lim, D., and N. C. Strynadka. 2002. Structural basis for the beta lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus. Nat. Struct. Biol. 9:870-876. [PubMed]
22. Margolin, W. 2005. FtsZ and the division of prokaryotic cells and organelles. Nat. Rev. Mol. Cell Biol. 6:862-871. [PubMed]
23. Matias, V. R., and T. J. Beveridge. 2005. Cryo-electron microscopy reveals native polymeric cell wall structure in Bacillus subtilis 168 and the existence of a periplasmic space. Mol. Microbiol. 56:240-251. [PubMed]
24. Matias, V. R., and T. J. Beveridge. 2006. Native cell wall organization shown by cryo-electron microscopy confirms the existence of a periplasmic space in Staphylococcus aureus. J. Bacteriol. 188:1011-1021. [PMC free article] [PubMed]
25. Merchante, R., H. M. Pooley, and D. Karamata. 1995. A periplasm in Bacillus subtilis. J. Bacteriol. 177:6176-6183. [PMC free article] [PubMed]
26. Morlot, C., M. Noirclerc-Savoye, A. Zapun, O. Dideberg, and T. Vernet. 2004. The D,D-carboxypeptidase PBP3 organizes the division process of Streptococcus pneumoniae. Mol. Microbiol. 51:1641-1648. [PubMed]
27. Morlot, C., A. Zapun, O. Dideberg, and T. Vernet. 2003. Growth and division of Streptococcus pneumoniae: localization of the high molecular weight penicillin-binding proteins during the cell cycle. Mol. Microbiol. 50:845-855. [PubMed]
28. Nickell, S., C. Kofler, A. P. Leis, and W. Baumeister. 2006. A visual approach to proteomics. Nat. Rev. Mol. Cell Biol. 7:225-230. [PubMed]
29. Pares, S., N. Mouz, Y. Petillot, R. Hakenbeck, and O. Dideberg. 1996. X-ray structure of Streptococcus pneumoniae PBP2x, a primary penicillin target enzyme. Nat. Struct. Biol. 3:284-289. [PubMed]
30. Paul, T. R., A. Venter, L. C. Blaszczak, T. R. Parr, Jr., H. Labischinski, and T. J. Beveridge. 1995. Localization of penicillin-binding proteins to the splitting system of Staphylococcus aureus septa by using a mercury-penicillin V derivative. J. Bacteriol. 177:3631-3640. [PMC free article] [PubMed]
31. Pinho, M. G., and J. Errington. 2003. Dispersed mode of Staphylococcus aureus cell wall synthesis in the absence of the division machinery. Mol. Microbiol. 50:871-881. [PubMed]
32. Pinho, M. G., and J. Errington. 2005. Recruitment of penicillin-binding protein PBP2 to the division site of Staphylococcus aureus is dependent on its transpeptidation substrates. Mol. Microbiol. 55:799-807. [PubMed]
33. Pozzi, G., R. A. Musmanno, P. M.-J. Lievens, M. R. Oggioni, P. Plevani, and R. Manganelli. 1990. Method and parameters for genetic transformation of Streptococcus sanguis Challis. Res. Microbiol. 141:659-670. [PubMed]
34. Richter, K. 1994. A cryoglue to mount vitreous biological specimens for cryoultramicrotomy at 110K. J. Microsc. 173:143-147. [PubMed]
35. Sartori Blanc, N. , A. Senn, A. Leforestier, F. Livolant, and J. Dubochet. 2001. DNA in human and stallion spermatozoa forms local hexagonal packing with twist and many defects. J. Struct. Biol. 134:76-81. [PubMed]
36. Scheffers, D. J., L. J. Jones, and J. Errington. 2004. Several distinct localization patterns for penicillin-binding proteins in Bacillus subtilis. Mol. Microbiol. 51:749-764. [PubMed]
37. Scheffers, D. J., and M. G. Pinho. 2005. Bacterial cell wall synthesis: new insights from localization studies. Microbiol. Mol. Biol. Rev. 69:585-607. [PMC free article] [PubMed]
38. Schuster, C., B. Dobrinski, and R. Hakenbeck. 1990. Unusual septum formation in Streptococcus pneumoniae mutants with an alteration in the d,d-carboxypeptidase penicillin-binding protein 3. J. Bacteriol. 172:6499-6505. [PMC free article] [PubMed]
39. Seltmann, G., and O. Holst. 2001. The bacterial cell wall. Springer, Berlin, Germany.
40. Shockman, G. D., and J. F. Barrett. 1983. Structure, function, and assembly of cell walls of gram-positive bacteria. Annu. Rev. Microbiol. 37:501-527. [PubMed]
41. Spizizen, J. 1958. Transformation of biochemically deficient strains of Bacillus subtilis by deoxyribonucleate. Proc. Natl. Acad. Sci. USA 44:1072-1078. [PMC free article] [PubMed]
42. Stock, J. B., B. Rauch, and S. Roseman. 1977. Periplasmic space in Salmonella typhimurium and Escherichia coli. J. Biol. Chem. 252:7850-7861. [PubMed]
43. Tomasz, A., J. D. Jamieson, and E. Ottolenghi. 1964. The fine structure of Diplococcus pneumoniae. J. Cell Biol. 22:453-467. [PMC free article] [PubMed]
44. Vaughan, S., B. Wickstead, K. Gull, and S. G. Addinall. 2004. Molecular evolution of FtsZ protein sequences encoded within the genomes of archaea, bacteria, and eukaryota. J. Mol. Evol. 58:19-29. [PubMed]
45. Weiss, D. S. 2004. Bacterial cell division and the septal ring. Mol. Microbiol. 54:588-597. [PubMed]
46. Zuber, B., I. Nikonenko, P. Klauser, D. Muller, and J. Dubochet. 2005. The mammalian central nervous synaptic cleft contains a high density of periodically organized complexes. Proc. Natl. Acad. Sci. USA 102:19192-19197. [PMC free article] [PubMed]

Articles from Journal of Bacteriology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...