Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Borgnia, M. J. Articles by Milne, J. L. S. Search for Related Content PubMed PubMed Citation Articles by Borgnia, M. J. Articles by Milne, J. L. S.

 Previous Article  |  Next Article 

Journal of Bacteriology, April 2008, p. 2588-2596, Vol. 190, No. 7
0021-9193/08/$08.00+0     doi:10.1128/JB.01538-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Three-Dimensional Imaging of the Highly Bent Architecture of Bdellovibrio bacteriovorus by Using Cryo-Electron Tomography ^,

Mario J. Borgnia, Sriram Subramaniam, and Jacqueline L. S. Milne^*

Laboratory of Cell Biology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892

Received 24 September 2007/ Accepted 7 January 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bdellovibrio bacteriovorus cells are small deltaproteobacterial cells that feed on other gram-negative bacteria, including human pathogens. Using cryo-electron tomography, we demonstrated that B. bacteriovorus cells are capable of substantial flexibility and local deformation of the outer and inner membranes without loss of cell integrity. These shape changes can occur in less than 2 min, and analysis of the internal architecture of highly bent cells showed that the overall distribution of molecular machines and the nucleoid is similar to that in moderately bent cells. B. bacteriovorus cells appear to contain an extensive internal network of short and long filamentous structures. We propose that rearrangements of these structures, in combination with the unique properties of the cell envelope, may underlie the remarkable ability of B. bacteriovorus cells to find and enter bacterial prey.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacteria of the genus Bdellovibrio are highly motile predators that reside in diverse aquatic and terrestrial environments, as well as in the mammalian digestive tract (28; for a review, see reference 33). They prey on a variety of gram-negative bacteria, including several human pathogens, and could become an important tool in the probiotic treatment of disease (32). Within this genus, Bdellovibrio bacteriovorus is the species that has been best characterized biochemically and genetically (27). Investigations over the last four decades (20, 27, 33) have helped elucidate several aspects of the complex life cycle of this organism, which includes a free-swimming or attack phase and a prey-bound phase. Attack-phase B. bacteriovorus is propelled by a single long sheathed flagellum and is capable of attaining speeds up to 160 µm/s. The predator swims and turns in apparently random directions until it finds and attaches to a prey cell. It then digests a region of the host cell outer membrane to make a small entry pore, penetrates into the host periplasm, and forms a growth chamber by reshaping and resealing the host outer membrane. The established parasite depletes the host cytoplasm and undergoes growth to form a single elongated spiral cell and segmented division to generate multiple progeny, and it develops a flagellum to gain motility. Ultimately, the chamber ruptures, releasing the nascent bacteria to reinitiate the life cycle.

Conventional electron microscopic (EM) studies have provided valuable insights into morphological changes that occur during the life cycle of B. bacteriovorus (1, 2, 6, 7). However, room temperature EM imaging typically requires specimen preparation steps that include chemical fixation, removal of the aqueous phase by treatment with organic solvents, and embedding in plastic resins. These procedures can introduce artifacts, such as distortion of the cell shape, deterioration of membrane structures, and aggregation of soluble multiprotein complexes. Moreover, the contrast in the images originates primarily from the stain and not from the intrinsic density of target structures, limiting the resolution attainable by this approach. Atomic force microscopic studies have also been used to probe the bacterial surface, but detailed internal structures are not visible and the samples are also subjected to freeze-drying and staining (24, 25). Cryo-EM and cryo-electron tomography are powerful alternatives for visualization of the global architecture of bacterial cells that have been preserved in their native state (4, 19, 29, 39). In this method, thin hydrated specimens are rapidly cooled to –180°C in liquid ethane to produce vitrified samples without chemical fixation and dehydration. Further, the specimens can be imaged using minimal electron doses and without additives to enhance contrast, so the resultant images closely reflect native cellular conditions.

Here we used cryo-EM and cryo-electron tomography to visualize attack-phase B. bacteriovorus and the spatial distribution of key molecular complexes in situ. We found that even though this organism has internal architectural elements that are very similar to those of other gram-negative bacteria, it can undergo dramatic changes in shape, as demonstrated by molding of the bacterial shape to the topography of a carbon substrate in less than 1 to 2 min.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains and culture conditions. The host-dependent B. bacteriovorus strain used in this study was HD100 (= DSM50701). Predatory B. bacteriovorus was cocultured at 22 to 25°C with Escherichia coli strain RP3098, a mutant with all flagellar and chemotaxis proteins deleted (31). Liquid cocultures were shaken at 250 rpm. To prepare prey stock cultures, individual RP3098 colonies selected from YPD agar plates (1.0% yeast extract, 2.0% peptone, 2.0% agar, 2.0% glucose; Teknova, Hollister, CA) were grown overnight at 37°C and 250 rpm in 3 ml of YT broth (1.6% tryptone, 1.0% yeast extract, 0.5% NaCl; Teknova, Hollister, CA) supplemented with 10 mM Tris-HCl (pH 7.5), harvested by centrifugation (Eppendorf 5415 D microcentrifuge; 16,000 x g, 30 s, 23°C), and resuspended in 3 ml of NB500 medium (25 mM Na-HEPES, 2 mM CaCl₂, 16 mg/liter WL Difco nutrient broth [pH 7.6]). Clonal B. bacteriovorus isolates were obtained initially and then at periodic intervals to prevent enrichment of spontaneous host-independent variants, using a two-layer agar selection procedure. Liquid cultures passaged through 0.45-µm filters were diluted, mixed with 1 optical density at 600 nm (OD₆₀₀) unit of prey cells in 5 ml of 42°C top agar (0.7% agar in NB500), and plated onto 10 ml of NB500 medium containing 1.5% agar at 22°C in a 10-cm-diameter round petri dish. Plaques that appeared within 1 week were transferred to 2 ml of NB500 medium containing 0.2 OD₆₀₀ unit of an E. coli prey stock culture, most of which cleared within 24 h. Cleared cultures were filtered through 0.45-µm filters and passaged by 20-fold dilution into 2 ml of NB500 medium containing 0.2 OD₆₀₀ unit of a prey culture. Maintenance cultures were passaged at least biweekly.

Specimen preparation. Enriched B. bacteriovorus cultures were prepared for EM by adding prey cells at a final concentration of 1 to 2 OD₆₀₀ units to an overnight maintenance culture and monitoring the preparation by optical microscopy (magnification, x400) until most prey cells were cleared from the culture. The clearing times were variable, especially with high prey densities. Aliquots of cleared culture (1 ml) were centrifuged (Eppendorf 5415 D microcentrifuge; 5,000 x g, 30 s, 23°C) to remove the prey cells and immature cells trapped in bdelloplasts. All culture transfers were done carefully to prevent disruption of the remaining bdelloplasts and to avoid the release of immature cells. Supernatants were transferred to a fresh tube and examined with a light microscope. Highly motile B. bacteriovorus was the dominant species in this fraction. A 3- to 5-µl aliquot of the supernatant was deposited onto the carbon side of a holey carbon grid (Quantifoil MultiA; Micro Tools GmbH, Germany) prepared as indicated below and held with tweezers. After 1 min, the tweezers and grid were positioned in a plunge-freeze apparatus (Vitrobot; FEI Corp., Oregon), blotted from both sides (2 to 10 s, 23°C, 90% chamber humidity), and rapidly vitrified in liquid ethane. To aid in the alignment of tomographic tilt series, all holey carbon grids were glow discharged, incubated for 2 min with 3 µl of 15-nm protein A-gold conjugate (BBInternational, Cardiff, United Kingdom) on the carbon side, which was followed by two 30-s washes with MilliQ water, air dried, and glow discharged a second time immediately prior to application of specimens and plunge freezing.

Cryo-EM and cryo-tomography. Specimens were transferred and imaged at liquid nitrogen temperatures using a Polara microscope (FEI Corp., Oregon) equipped with a field emission gun operating at 300 kV. Projection images were recorded using a 2K by 2K charge-coupled device camera located at the end of a postcolumn GIF 2000 (Gatan Inc., Pleasanton, CA) energy filter. Low-dose tomographic tilt series (1 to 2 e^–/Ų per image) were collected over an angular range of ±69° in 1.5° to 5° intervals (linear or Saxton collection schemes), using an effective magnification of x18,000, x22,500, or x27,500 and underfocus values ranging from 5 to 15 µm. Gold fiducial alignment was performed for full-resolution images. Three-dimensional (3D) reconstructions were obtained by weighted back projection of aligned images that were binned 4x4. Tomographic reconstructions obtained in this way were processed by 3D nonlinear anisotropic diffusion prior to segmentation. Alignment, reconstruction, denoising, and segmentation were all done using IMOD (18).

Elemental mapping. Electron energy loss spectroscopy (EELS) for elemental mapping was performed with an FEI Tecnai F20 S-Twin electron microscope with a probe size of 1 nm, using previously described procedures (21). Hyperspectral images were recorded using a Gatan energy filter (model 2001), incorporating a cooled 1K by 1K charge-coupled device camera. For each pixel in the scanned image, spectral channel intensities were summed in the direction perpendicular to the energy dispersion and were corrected for dark-current and channel gain variations. Vitrified specimens of B. bacteriovorus cells were allowed to sublime in the vacuum of the microscope column for 12 h. Data were collected from these dried samples after they were transferred to a room temperature specimen holder. To determine the presence of different elements in a specimen, a high-angle angular dark-field scanning transmission EM image was first obtained. EELS spectra of regions containing electron-dense granules showed the presence of distinct phosphorus, oxygen, and calcium peaks higher than those for control regions not containing the electron-dense granules. The relative proportions of various elements present were estimated using software in DigitalMicrograph 3.9.4 GMS1.4.4 and a Hartree-Slater model for estimating cross sections.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Representative projection images of frozen hydrated B. bacteriovorus cells (Fig. 1) demonstrated the qualitative improvement obtained by use of rapid vitrification and cryo-EM imaging in a nearly native state compared to conventional EM imaging. The cellular architecture was well preserved, as shown by the integrity of the outer and inner membranes and the considerable internal detail visible in the cytoplasm.

View larger version (194K): [in this window] [in a new window]

FIG. 1. Intracellular structure and variation in the shape of B. bacteriovorus. (a) Low-dose, energy-filtered, projection EM images of vitrified cells allow detection of key intracellular components, as indicated by arrows. A, outer membrane; B, inner membrane; C, sheathed flagellum; D, rotor complex; E, chemotaxis receptor array; F, dense granule; G, nucleoid; H, macromolecular complexes; I, peptidoglycan layer; J, needlelike structures at the anterior pole. (a to e) Examples illustrating the flexibility of B. bacteriovorus cells, which range from cells with moderate bends (a) to U-shaped bends (b) to cells distorted by virtue of intercellular contact (c) and cell edges flattened by contact with the carbon substrate. The large black dots at the top and bottom right of the images are 15-nm gold particles deposited on the carbon film and used as fiducial markers to align multiple images in a tilt series. (a and c to e) Scale bar = 200 nm. (b) Scale bar = 1 µm.

Attack-phase B. bacteriovorus cells are typically shaped like curved rods with a length of 1.02 ± 0.15 µm and a width of 0.3 ± 0.02 µm (means ± standard deviations; n = 50). The two poles of a cell are distinct; the posterior end containing the flagellum is rounder, whereas the anterior end, which is involved in prey penetration, is usually flatter. Single cells and clusters of stacked cells having different degrees and directions of curvature, ranging from U- to comma- to S-shaped cells, can be captured under physiologically relevant conditions in vitreous ice, as shown in the low-magnification image in Fig. 1b. Some cells bend so extensively that they exhibit apparent surface interactions between different membrane regions of the same cell (Fig. 1c). Occasionally, constrictions in the cell width were observed at areas of contact with other cells (Fig. 1d) or with the substrate (see Fig. 6). In rare cases, membrane extensions induced by contact with the carbon were found to occur both along the length of the cell (Fig. 1e) and at the poles. In contrast, these types of bends and distortions have not been observed in any of hundreds of tomographically imaged E. coli cells examined (39; Cezar Khursigara and Sriram Subramaniam, unpublished observations).

View larger version (43K): [in this window] [in a new window]

FIG. 6. Cryo-electron tomography of vitrified B. bacteriovorus cells shaped around the edges of the carbon film-vitreous ice interface. Panels a to d and panels e to h show two examples. (a and e) Low-dose projection images; (b, c, f, and g) 6-nm-thick tomographic slices from different depths of the cellular tomograms; (d and h) segmented rendering of portions of the bent cells, showing the wrapping of the cells around the edge of the carbon film. Scale bars = 200 nm. The inset in panel c shows that despite extensive changes in curvature, the spacing between the inner and outer membranes is maintained along the length of the bacterium.

Cryo-electron tomographic methods (34) allow visualization of structure in the third dimension and significantly extend the information that can be obtained from projection images such as those shown in Fig. 1. An example of the improvement in visualization of cellular detail is shown in Fig. 2, in which the type of information obtained from a two-dimensional projection (Fig. 2a) is compared with the type of information obtained from 3D tomographic analysis (Fig. 2b and 2c) of the same cell. The space between the inner and outer membranes, measured using tomographic analysis, is 25 nm along the body of the cell and increases to 40 nm at the anterior pole. The prominent granules in the cytoplasm appear to be quasi-spherical and have different diameters. In some cases, a thin boundary layer of density is discernible at the interface between the granules and the cytoplasm. EELS demonstrated that the prominent dense granules are enriched in phosphorus, oxygen, and calcium compared to other regions of the same cell or regions corresponding to the grid, both of which have distinct elemental profiles (Fig. 3). These granules have no preferential cellular localization, and there is no obvious correlation between granule size and cell length (Fig. 4).

View larger version (42K): [in this window] [in a new window]

FIG. 2. Cellular interior revealed by cryo-electron tomography. (a) Projection image recorded as part of a tilt series for a specimen of plunge-frozen B. bacteriovorus cells. (b) Central 8-nm slice through the reconstructed 3D volume obtained from the same tilt series. (c) 3D rendering of key cellular structures segmented from the tomographic volume. The flagellum, inner and outer membranes, and filamentlike structures in the cytoplasm are green, putative ribosomes are red, and dense granules are blue. The inset in panel a shows a projection image with periodic, 26-Å-spaced striations in the nucleoid region, consistent with the expected packing arrangement of DNA. The inset in panel b shows the anterior end of the cell, including structures (indicated by arrowheads) that protrude outward and may be relevant for making contact with prey. (a and b) Scale bars = 150 nm. (Panel a inset) Scale bar = 10 nm. (Panel b inset) Scale bar = 25 nm.

View larger version (48K): [in this window] [in a new window]

FIG. 3. Elemental analysis of Bdellovibrio dense granules. (a) Projection image of a cell recorded using scanning transmission EM. Circles 1, 2, and 3 indicate the centers of square regions (20 by 20 nm) analyzed by EELS to determine the elemental composition of the dense granules, the surrounding nucleoid, and the extracellular medium, respectively. (b) EELS spectra of the three regions, indicating that the dense granules are enriched in phosphorus, oxygen, and calcium. See Materials and Methods for additional details.

View larger version (20K): [in this window] [in a new window]

FIG. 4. Distribution of the locations and sizes of the dense granules along the length of the cell. (a) The data show that the dense granules are not present at specific positions relative to the pole of the cell and can be various sizes. (b) Correlation between the size of the dense granules and the size of the cell, expressed in granule size/cell length, indicating that there is no obvious correlation between the size of the cells and the overall size of the granules.

The elemental composition of the granules suggests that they are similar to the acidocalcisomes of other cells (12) but not to the carbon-enriched, phosphorus-depleted bodies seen in Caulobacter crescentus, which are thought to consist of butyric acid in association with electron-dense salts (9). B. bacteriovorus cells are both highly enriched in cellular polyphosphates compared to E. coli (3) and contain inwardly directed phosphate transporters, as well as the genes encoding the key enzymes required for polyphosphate metabolism, including polyphosphate kinase 1 and several polyphosphate phosphohydrolases (27). Whether any of these enzymes localize at the outer edges of the dense granules remains to be determined, but the dense granules likely serve as critical energy reserves to ensure survival under starvation conditions (5), among other potential functions attributed to polyphosphates (17).

Detailed inspection of the interior of the cell revealed several other features whose probable identities can be surmised based on their shape, size, and location. In the outer reaches of the cytoplasm are scattered dense macromolecular complexes, likely to be ribosomes. These complexes are not present in the central region of the cell, which is diffuse and has a different texture than the rest of the cytoplasm. Based on its central location, we believe that this region corresponds to the nucleoid. Further, two-dimensional images of this region in other cells revealed a stacked pattern with a periodicity of 26 Å consistent with higher-order structures of DNA (Fig. 2a, inset), such as those previously observed in thin vitreous sections of Deinococcus radiodurans (13). Distinct structures were also visible on the outside of the cell. In particular, protruding dense structures perpendicular to the outer membrane were observed at the anterior pole of the cell (Fig. 2b, inset). It is possible that these structures correspond to the structures thought to be responsible for prey attachment or penetration (27). At the other end of the cell, the base of the flagellum is clearly visible, and tomographic imaging showed that the membrane-embedded rotor complex is always offset from the longitudinal axis of the cell (see Movies S1, S2, and 3 in the supplemental material). Remarkably, the major components of the rotor complex are discernible even at the extremely low signal-to-noise ratios corresponding to the 3D density map for a single complex. The nearby polar chemotaxis receptor array (39) is also evident in tomograms of individual cells (Fig. 5).

View larger version (78K): [in this window] [in a new window]

FIG. 5. Visualization of the motility and chemotaxis machinery of B. bacteriovorus. (a) (Top left panel) Eight-nanometer slice extracted from a tomographic volume in an orientation coplanar with the axis of the rotor complex. Sections of the individual elements of the ring are visible. (Top right panel) Schematic diagram of the tomogram corresponding to the slice shown in the top left panel showing components expected to be found in a single rotor complex. (Bottom panels) Four orthogonal slices across the rotor complex shown in the top left panel at places indicated by the tick marks. The density resulting from the flagellum is visible as a small dark ring in the fourth bottom panel and is white in the schematic diagram in the top right panel; a faintly visible density corresponding to the flagellar sheath is green in the top right panel. The density from the flagellum is also visible at the center of the third bottom panel along with contributions from the density indicated by yellow in the top right panel and likely results from the P-ring. The much darker density in the outermost regions of the third panel is due to the outer membrane, and the level is roughly the same as the level of density indicated by yellow in the top right panel. The densities in the first and second panels indicated by red, blue, and orange in the schematic diagram likely correspond to contributions from the C-ring, the MS-ring, and the MotAB complex, respectively. (b) Eight-nanometer tomographic slice through the center of the same cell near the flagellar pole, showing the spatial arrangement of the chemotaxis receptor array (visible as a band in the cytoplasm [dark line of density inside the inner membrane]) relative to the pole. Scale bar = 50 nm.

While most cells were completely embedded in vitreous ice, the interaction with occasional thin layers of carbon extending from the rim or narrow strips separating two closely spaced holes resulted in cell membranes bending around the carbon film (Fig. 6). These cells presented an opportunity to examine the interaction with the substrate. Tomographic reconstruction allowed us to determine that both the inner and outer membranes wrap around the substrate without any apparent loss of cellular integrity. Comparison of highly bent and moderately bent cells revealed no discernible differences in the cellular features noted above, including the nucleoid, which remained undistorted and closely followed the curvature of the bend (see Movies S2 and S3 in the supplemental material). There was also no appreciable difference in the 25-nm spacing between the inner and outer membranes at regions with the greatest curvature relative to other areas of the cell (Fig. 6c, inset).

Another interesting structural feature of the B. bacteriovorus cellular interior is the extended network of internal filaments (Fig. 7; see Movie S1 in the supplemental material). Some filaments are parallel to the longitudinal axis of the cell (Fig. 7a), while others are distributed in the transverse direction (Fig. 7b) or are bundled in clusters perpendicular to the membrane plane and arranged in a periodic pattern with a spacing of 12 nm (Fig. 7c). These filaments appear to be similar to those previously reported for C. crescentus (4), and although their identity is not known at present, they are highly suggestive of the structures formed in vitro by bacterial actin-like protein homologs (for a review, see reference 23).

View larger version (72K): [in this window] [in a new window]

FIG. 7. Visualization of cytoskeletal elements in the cytoplasm of B. bacteriovorus: tomographic slices showing the presence of bundles of filaments oriented parallel (a and b) and transverse (indicated by the circle) (c) to the plane of the plasma membrane. Scale bars = 100 nm.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The name Bdellovibrio reflects two characteristics of the genus: (i) the behavior of the cells resembles that of a feeding leech (bdello) because the cells usually attach by one of the poles to the side of the prey and (ii) the shape of the cell body is approximately the shape of a comma (vibrio). However, the most striking feature of attack-phase B. bacteriovorus cells observed with a light microscope is their motility (33). Swimming cells resemble elongated wiggling rods, but precise determination of their shape and its influence on internal cellular architecture is difficult using light microscopy since the cells are very small, move rapidly through the field of focus, and turn abruptly in random directions. Adherent cells appear to be primarily comma shaped, but detailed morphological characterization is hindered by the limited resolution of light microscopy. Conventional EM pictures have revealed pronounced deviations from the comma shape (1, 2, 6, 7). However, the cell membranes and internal structures of these cells are prone to damage resulting from fixation, desiccation, and staining procedures, which precludes visualization of detailed cellular architecture for the range of observed shapes in the population. The analysis that we report here of intact frozen hydrated cells using cryo-EM and cryo-electron tomography indicated that the cells can bend with extreme curvature without suffering any obvious structural damage. Cell bending is likely to occur prior to rather than during vitrification because the rapid rate of cooling (10⁵°C/s) precludes formation of crystalline ice and is thus expected to block changes in cellular shape, which cannot occur more rapidly than the rate of ice crystal formation.

Although B. bacteriovorus reproduces by multiple sequential segmentation, the bent shapes that we observed likely do not represent cells that were prematurely released from bdelloplasts. First, cells released prior to division would be expected to show evidence of segmentation, which was not apparent in any of the hundreds of cellular images that we recorded. Second, a population of prematurely released cells is expected to be heterogeneous in terms of length, contrary to the narrow distribution that we observed in a given culture. Third, early studies on B. bacteriovorus cell division by use of conventional EM also showed that the flagellum starts to develop during cell division and continues to mature after septation has been completed (7), but here all B. bacteriovorus cells that were bent also had fully developed flagella. Finally, the nucleoid occupies a relatively large and well-defined volume in the cell, which must nevertheless be highly compact, because the size of the B. bacteriovorus genome (3.78 Mb) is similar to the size of the genome of the much larger organism E. coli (27). In dividing cells, genetic material is expected to double in size and to be segregated to the daughter cells. Several projection images of bent cells and the 3D tomographic reconstruction of an L-shaped cell shown in Fig. 6a show a single uninterrupted volume in which ribosomes are not observed; this is again consistent with the presence of a single nucleoid instead of the two separate volumes that would be consistent with segregation of genetic material prior to segmentation. Thus, we concluded that the vast majority of cells observed in our cultures must be mature, attack-phase B. bacteriovorus cells.

The next question concerns whether the shape of individual cells in a culture is fixed or changes dynamically before cells are locked into a particular shape by vitrification. Our results indicate that cells could alter both their overall curvature and the shape of their membrane upon interaction with the substrate within the 2 min that elapsed from deposition of the sample on the holey carbon grid to freezing (Fig. 1). Importantly, these changes in shape occurred without compromising the integrity of the membrane, suggesting that an underlying active mechanism mediates the changes, although the possibility that there are simpler alternative mechanisms that cause changes in adhesion of outer membrane components cannot be excluded. Control of the degree of curvature could provide the cell with a mechanism to modulate rotation about its longitudinal axis and, combined with the off-axis location of the flagellum, may enable a motile B. bacteriovorus cell to modify its speed and direction. Flexibility may also play a fundamental role during prey penetration, as B. bacteriovorus enters the host periplasm very rapidly through a restricted pore (2), possibly using a mechanism involving twitching mobility mediated by type IV pili (14). An intriguing possibility is that the protrusions observed at the anterior pole (Fig. 2, inset) may represent these type IV pili. To our knowledge, dynamic shape variations of the kind that we are proposing here have generally not been observed in prokaryotes, although it has been suggested that alterations in filament length may be involved in cellular motility and alterations in cell shape in the cell wall-free organism Spiroplasma (19).

The ability of B. bacteriovorus to undergo alterations in shape along its length leads to the pleomorphic shape diversity seen in culture and during entry into prey cells. This is in contrast to findings for other bacteria, which show that typically, well-defined shapes, such as rods, cocci, or spirals, occur within a population of a particular species (38). Bacterial morphology appears to be maintained by the ordered insertion of the peptidoglycan, using preexisiting strands as templates, and regulated peptidoglycan bond turnover (36), which is higher in zones of active growth and cell division and does not occur in the polar area (10, 15). The exact mechanisms of spatially regulated peptidoglycan synthesis are uncertain. They are likely to involve not only the penicillin binding enzymes involved in cell wall synthetic complexes but also the cytoskeletal elements MreB and FtsZ and cell shape-determining proteins, such as MreC and RodA, which may position the peptidoglycan synthesis machinery in the cell (8, 11, 36). Only in certain instances, such as the imposition of external templates, genetic mutation, or starvation, is the normal shape typically disrupted (11, 38). Intracellular turgor pressure in bacteria, estimated to be as high as 3 to 5 atm (16), could also play a key role in maintaining established shapes by keeping the peptidoglycan layer in a more extended conformation. Interaction of outer membrane lipoproteins and membrane proteins with the peptidoglycan layer and the presence of complexes that span the inner and outer membranes or relatively weak electrostatic interactions may also stabilize cell shape, although the relative contributions may vary among species (37).

The mechanisms which B. bacteriovorus employs to establish and maintain its vibrioid shape likely parallel those of other bacteria since B. bacteriovorus has a peptidoglycan composition similar to that of other bacteria (35) and possesses penicillin-like binding proteins (26, 27) and cytoskeletal elements homologous to MreB, MreC, RodA, FtsZ, and MltA (27), which are known to modulate shape in other bacteria (8, 30, 38), as discussed above. The discovery of fiber-like structures in the cytoplasm in the bent regions of the cell suggests that these proteins may also have unanticipated roles in mediating the observed relatively rapid changes in shape coupled with changes in membrane curvature. In addition to the cytoskeletal components, we also propose that there must be a high degree of plasticity of the inner and outer membranes and possibly the peptidoglycan layer in order for these bacteria to achieve rapid changes in shape along the cell surface. At least in the case of the U-shaped cells and the cells that curve around carbon film, such changes may require rapid expansion of peptidoglycan in one region coupled with degradation or contraction in another region. The bending occurs along the longitudinal axis of the cell, consistent with atomic force microscopic measurements that show that the primary elasticity follows the longitudinal axis in isolated sacculi from E. coli (37). Our findings indicate that the density profiles of inner and outer monolayers of the B. bacteriovorus outer membrane are comparable (see Fig. S1 in the supplemental material). This is in contrast to findings reported by workers in our laboratory and by other workers for the corresponding E. coli outer membrane, in which the outer monolayer has been observed to be denser than the inner monolayer (22, 39). This morphological difference suggests that there may be biochemical differences between the membranes of B. bacteriovorus and the membranes of other bacteria. B. bacteriovorus spheroplasts have been reported to be more resistant to osmotic shock than E. coli spheroplasts (35), the membranes are sensitive to detergent extraction (35), and cells are more susceptible to sonication or freeze-thaw treatment (1). Further, cells released from a bdelloplast have unusual levels of peptidoglycan turnover (35), supporting the idea that the B. bacteriovorus cell envelope may have a unique composition and/or that there may be unique regulation of its murein layer and its interactions with the inner and outer cell membranes. The electron tomographic studies that we describe here provided a foundation for further understanding the structural origins of the unique biology of B. bacteriovorus cells and for assessing the connections between cellular shape, cytoskeletal organization, and membrane structure.

 
We thank Elizabeth Sockett, University of Nottingham, Nottingham, United Kingdom, for providing strain HD100 and Y. C. Wang of FEI Company, Hillsboro, OR, for assistance with EELS analysis.

This work was supported by funds from the Center for Cancer Research, National Cancer Institute, National Institutes of Health, to J.L.S.M. and S.S.

 
* Corresponding author. Mailing address: Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Building 50, Room 4306, 50 South Drive, Bethesda, MD 20892. Phone: (301) 594-2063. Fax: (301) 480-3834. E-mail: jmilne{at}nih.gov

Published ahead of print on 18 January 2008.

Supplemental material for this article may be found at http://jb.asm.org/.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Abram, D., and B. K. Davis. 1970. Structural properties and features of parasitic Bdellovibrio bacteriovorus. J. Bacteriol. 104:948-965.[Abstract/Free Full Text]
2
2. Abram, D., J. Castro e Melo, and D. Chou. 1974. Penetration of Bdellovibrio bacteriovorus into host cells. J. Bacteriol. 118:663-680.[Abstract/Free Full Text]
3
3. Bobyk, M. A., A. V. Afinogenova, M. V. Dudinskaya, V. A. Lambina, and I. S. Kulaev. 1980. Detection of polyphosphates and enzymes of polyphosphate metabolism in Bdellovibrio bacteriovorus. Zentralbl. Bakteriol. Naturwiss. 135:461-466.[Medline]
4
4. Briegel, A., D. P. Dias, Z. Li, R. B. Jensen, A. S. Frangakis, and G. J. Jensen. 2006. Multiple large filament bundles observed in Caulobacter crescentus by electron cryotomography. Mol. Microbiol. 62:5-14.[CrossRef][Medline]
5
5. Brown, M. R., and A. Kornberg. 2004. Inorganic polyphosphate in the origin and survival of species. Proc. Natl. Acad. Sci. USA 101:16085-16087.[Abstract/Free Full Text]
6
6. Burnham, J. C., T. Hashimoto, and C. F. Conti. 1968. Electron microscopic observations on the penetration of Bdellovibrio bacteriovorus into gram-negative bacterial hosts J. Bacteriol. 96:1366-1381.[Abstract/Free Full Text]
7
7. Burnham, J. C., T. Hashimoto, and S. F. Conti. 1970. Ultrastructure and cell division of a facultatively parasitic strain of Bdellovibrio bacteriovorus. J. Bacteriol. 101:997-1004.[Abstract/Free Full Text]
8
8. Cabeen, M. T., and C. Jacobs-Wagner. 2005. Bacterial cell shape. Nat. Rev. Microbiol. 3:601-610.[CrossRef][Medline]
9
9. Comolli, L. R., M. Kundmann, and K. H. Downing. 2006. Characterization of intact subcellular bodies in whole bacteria by cryo-electron tomography and spectroscopic imaging. J. Microsc. 223:40-52.[CrossRef][Medline]
10
10. de Pedro, M. A., J. C. Quintela, J.-V. Höltje, and H. Schwarz. 1997. Murein segregation in Escherichia coli. J. Bacteriol. 179:2823-2834.[Abstract/Free Full Text]
11
11. Divakaruni, A. V., C. Baida, C. White, and J. W. Gober. 2007. The cell shape proteins MreB and MreC control cell morphogenesis by positioning cell wall synthetic complexes. Mol. Microbiol. 66:174-188.[CrossRef][Medline]
12
12. Docampo, R., W. de Souza, K. Miranda, P. Rohloff, and S. N. Moreno. 2005. Acidocalcisomes—conserved from bacteria to man. Nat. Rev. Microbiol. 3:251-261.[CrossRef][Medline]
13
13. 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.[Abstract/Free Full Text]
14
14. Evans, K. J., C. Lambert, and R. E. Sockett. 2007. Predation by Bdellovibrio bacteriovorus HD100 requires type IV pili. J. Bacteriol. 189:4850-4859.[Abstract/Free Full Text]
15
15. Goodell, E. W., and U. Schwarz. 1983. Cleavage and resynthesis of peptide cross bridges in Escherichia coli murein. J. Bacteriol. 156:136-140.[Abstract/Free Full Text]
16
16. Koch, A. L. 1983. The surface stress theory of microbial morphogenesis. Adv. Microb. Physiol. 24:301-366.[Medline]
17
17. Kornberg, A., N. N. Rao, and D. Ault-Riche. 1999. Inorganic polyphosphate: a molecule of many functions. Annu. Rev. Biochem. 68:89-125.[CrossRef][Medline]
18
18. Kremer, J. R., D. N. Mastronarde, and J. R. McIntosh. 1996. Computer visualization of three-dimensional image data using IMOD. J. Struct. Biol. 116:71-76.[CrossRef][Medline]
19
19. Kürner, J., A. S. Frangakis, and W. Baumeister. 2005. Cryo-electron tomography reveals the cytoskeletal structure of Spiroplasma melliferum. Science 307:436-438.[Abstract/Free Full Text]
20
20. Lambert, C., K. A. Morehouse, C. Y. Chang, and R. E. Sockett. 2006. Bdellovibrio: growth and development during the predatory cycle. Curr. Opin. Microbiol. 9:639-644.[CrossRef][Medline]
21
21. Leapman, R. D., and R. L. Ornberg. 1988. Quantitative electron-energy loss spectroscopy in biology. Ultramicroscopy 24:251-268.[CrossRef][Medline]
22
22. Matias, V. R. F., A. Al-Amoudi, J. Dubochet, and T. J. Beveridge. 2003. Cryo-transmission electron microscopy of frozen-hydrated sections of Escherichia coli and Pseudomonas aeruginosa. J. Bacteriol. 185:6112-6118.[Abstract/Free Full Text]
23
23. Michie, K. A., and J. Lowe. 2006. Dynamic filaments of the bacterial cytoskeleton. Annu. Rev. Biochem. 75:467-492.[CrossRef][Medline]
24
24. Núñez, M. E., M. O. Martin, L. K. Duong, E. Ly, and E. M. Spain. 2003. Investigations into the life cycle of the bacterial predator Bdellovibrio bacteriovorus 109J at an interface by atomic force microscopy. Biophys. J. 84:3379-3388.[Medline]
25
25. Núñez, M. E., M. O. Martin, P. H. Chan, and E. M. Spain. 2005. Predation, death, and survival in a biofilm: Bdellovibrio investigated by atomic force microscopy. Colloids Surf. B Biointerfaces 42:263-271.[CrossRef][Medline]
26
26. Park, J. T., and S. Mahadevan. 1988. Penicillin-binding proteins of Bdellovibrios. J. Bacteriol. 170:3750-3751.[Abstract/Free Full Text]
27
27. Rendulic, S., P. Jagtap, A. Rosinus, M. Eppinger, C. Baar, C. Lanz, H. Keller, C. Lambert, K. J. Evans, A. Goesmann, F. Meyer, R. E. Sockett, and S. C. Schuster. 2004. A predator unmasked: life cycle of Bdellovibrio bacteriovorus from a genomic perspective. Science 303:689-692.[Abstract/Free Full Text]
28
28. Schwudke, D., E. Strauch, M. Krueger, and B. Appel. 2001. Taxonomic studies of predatory bdellovibrios based on 16S rRNA analysis, ribotyping and the hit locus and characterization of isolates from the gut of animals. Syst. Appl. Microbiol. 24:385-394.[CrossRef][Medline]
29
29. Seybert, A., R. Herrmann, and A. S. Frangakis. 2006. Structural analysis of Mycoplasma pneumoniae by cryo-electron tomography. J. Struct. Biol. 156:342-354.[CrossRef][Medline]
30
30. Shih, Y. L., and L. Rothfield. 2006. The bacterial cytoskeleton. Microbiol. Mol. Biol. Rev. 70:729-754.[Abstract/Free Full Text]
31
31. Smith, R. A., and J. S. Parkinson. 1980. Overlapping genes at the cheA locus of Escherichia coli. Proc. Natl. Acad. Sci. USA 77:5370-5374.[Abstract/Free Full Text]
32
32. Sockett, R. E., and C. Lambert. 2004. Bdellovibrio as therapeutic agents: a predatory renaissnance? Nat. Rev. Microbiol. 2:669-675.[CrossRef][Medline]
33
33. Starr, M. P., and R. J. Seidler. 1971. The bdellovibros. Annu. Rev. Microbiol. 25:649-678.[CrossRef][Medline]
34
34. Subramaniam, S. 2005. Bridging the imaging gap: visualizing subcellular architecture with electron tomography. Curr. Opin. Microbiol. 8:316-322.[CrossRef][Medline]
35
35. Thomashow, M. F., and S. C. Rittenberg. 1978. Penicillin-induced formation of osmotically stable spheroplasts in nongrowing Bdellovibrio bacteriovorus. J. Bacteriol. 133:1484-1491.[Abstract/Free Full Text]
36
36. Vollmer, W., and J. V. Höltje. 2004. Architecture of the murein (peptidoglycan) in gram-negative bacteria: vertical scaffold or horizontal layer(s)? J. Bacteriol. 186:5978-5987.[Free Full Text]
37
37. Yao, X., M. Jericho, D. Pink, and T. Beveridge. 1999. Thickness and elasticity of gram-negative murein sacculi measured by atomic force microscopy. J. Bacteriol. 181:6865-6875.[Abstract/Free Full Text]
38
38. Young, K. D. 2006. The selective value of bacterial shape. Microbiol. Mol. Biol. Rev. 70:660-703.[Abstract/Free Full Text]
39
39. Zhang, P., C. M. Khursigara, L. M. Hartnell, and S. Subramaniam. 2007. Direct visualization of Escherichia coli chemotaxis receptor arrays using cryo-electron microscopy. Proc. Natl. Acad. Sci. USA 104:3777-3781.[Abstract/Free Full Text]

---

Journal of Bacteriology, April 2008, p. 2588-2596, Vol. 190, No. 7
0021-9193/08/$08.00+0     doi:10.1128/JB.01538-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Wu, Y., Kaiser, A. D., Jiang, Y., Alber, M. S. (2009). Periodic reversal of direction allows Myxobacteria to swarm. Proc. Natl. Acad. Sci. USA 106: 1222-1227 [Abstract] [Full Text]  
• Dori-Bachash, M., Dassa, B., Pietrokovski, S., Jurkevitch, E. (2008). Proteome-Based Comparative Analyses of Growth Stages Reveal New Cell Cycle-Dependent Functions in the Predatory Bacterium Bdellovibrio bacteriovorus. Appl. Environ. Microbiol. 74: 7152-7162 [Abstract] [Full Text]  
• Khursigara, C. M., Wu, X., Subramaniam, S. (2008). Chemoreceptors in Caulobacter crescentus: Trimers of Receptor Dimers in a Partially Ordered Hexagonally Packed Array. J. Bacteriol. 190: 6805-6810 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Borgnia, M. J. Articles by Milne, J. L. S. Search for Related Content PubMed PubMed Citation Articles by Borgnia, M. J. Articles by Milne, J. L. S.