Fluctuation Analysis of Redox Potential to Distinguish Microbial Fe(II) Oxidation.

We developed a novel method for distinguishing abiotic and biological iron oxidation in liquid media using oxidation-reduction (redox) potential time series data. The instrument and processing algorithm were tested by immersing the tip of a Pt electrode with an Ag-AgCl reference electrode into an active iron-oxidizing biofilm in a groundwater discharge zone, as well as in two abiotic systems: a killed sample and a chemical control from the same site. We used detrended fluctuation analysis to characterize average root mean square fluctuation behavior, which was distinct in the live system. The calculated α value scaling exponents determined by detrended fluctuation analysis were significantly different at p < 0.001. This indicates that time series of electrode response data may be used to distinguish live and abiotic chemical reaction pathways. Due to the simplicity, portability, and small size, it may be suitable for characterization of extraterrestrial environments where water has been observed, such as Mars and Europa. Key Words: Oxidation-reduction potential-Detrended fluctuation analysis-Iron-oxidizing bacteria. Astrobiology 16, 846-852.

Thermal intensification of microbial Fe(II)/Fe(III) redox cycling in a pristine shallow sand aquifer on the Canadian Shield.

This investigation evaluates spatial relationships between summer (July) groundwater temperatures and Fe(II)/Fe(III) biogeochemical cycling over a five year period in a shallow pristine sand aquifer at Meilleurs Bay near Deep River, Ontario, Canada. A warm subsurface thermal island of 12.5-16.1 °C, compared to background conditions of 10-11 °C, was manifest in contour maps of average groundwater temperature over the study period. The warm zone coincided with an area of convergent groundwater flow, implicating horizontal heat transfer by advective convection as the reason for elevated temperatures. Additionally, high concentrations of dissolved Fe(II) and Fe(III) overlapped the warm thermal island, indicative of increased rates of bacterial Fe(II)-oxidation and Fe(III)-reduction. A depletion in the modal abundance of Fe(II)-bearing minerals, notably amphibole and biotite, inside the area of the warm thermal island was also observed, suggesting enhanced mineral dissolution owing to chemoautotrophic Fe(II)-oxidation coupled to the reduction and fixation of dissolved inorganic carbon as biomass. Throughout the aquifer, redox conditions were poised in terms of Eh and pH close to equilibrium with respect to the Fe(II)/Fe(OH)3 couple, feasibly enabling simultaneous bacterial Fe(II)-oxidation and Fe(III)-reduction with an adequate supply of electron acceptors and donors, respectively. The significance of higher groundwater temperature as a determinant of elevated dissolved Fe(II) and Fe(III) concentrations induced by thermal intensification of microbial biogeochemical activities yielded Pearson product-moment correlations in which temperature alone, as a single independent variable, explains almost 30 to nearly 60 percent of the variation in the measured dissolved Fe(II) and Fe(III) concentrations in the groundwater. These results emphasize the important influence of thermal conditions on biogeochemical processes in aquifers coupled to the development of steep gradients in groundwater quality over short distances in shallow unconfined groundwater systems.

Influence of lipopolysaccharide on the surface proton-binding behavior of Shewanella spp.

This study investigates the potentiometric properties of several strains of Shewanella spp. and determines whether these properties can be correlated with lipopolysaccharide (LPS) type. The LPS of eight Shewanella strains was characterized using silver-stained sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and their potentiometric properties determined using high-resolution acid-base titrations. Titrations showed that total ligand concentrations (L(T)) ranged from 0.903 +/- 0.007 micromol/mg (S. baltica 63) to 1.387 +/- 0.007 micromol/mg (S. amazonensis SB2B). Smooth strains (possessing O-side chains) exhibited higher mean L(T) values than rough strains (no O-side chain). A Tukey's honestly significantly different (HSD) test revealed, smooth strains exhibited significantly higher L(T) values than rough strains in 69% of comparisons. Comparison of individual pK(a) concentrations revealed that smooth LPS strains of Shewanella were relatively enriched in reactive groups at pK(a) 5, suggesting their LPS O-side chains contained detectable carboxyl groups. Combined pKa spectra from all eight Shewanella strains produced a common trend indicating that the way in which the surface proton-buffering capacity changes with pH is similar for the species studied here.

Ultrastructure and potential sub-seafloor evidence of bacteriogenic iron oxides from Axial Volcano, Juan de Fuca Ridge, north-east Pacific Ocean.

Iron oxides from the caldera of Axial Volcano, a site of hydrothermal vent activity along the Juan de Fuca Ridge, were found to consist predominantly of microbial structures in hydrated whole mounts examined using an environmental scanning electron microscope. Novel observations were made of the iron oxides revealing the spatial relationships of the bacteria within to be more consistent with microbial mats than mineral precipitates. The bacterial structures are attributed to the sheaths of Leptothrix ochracea, the stalks of Gallionella ferruginea, and the filaments of a novel iron oxidizing PV-1 strain, based on the distinctive morphological characteristics of these three bacteria. Energy dispersive X-ray spectroscopy revealed the presence and distribution of Fe, Si, and Cl on the bacterial sheaths, stalks and filaments. The iron oxides were identified by X-ray diffraction to be two-line ferrihydrite, a poorly ordered iron oxyhydroxide. Adsorption of Si in particular to two-line ferrihydrite likely contributes to its stability on the seafloor, and might also be a preservation mechanism creating microfossils of the bacterial structures encrusted with ferrihydrite. Presumptive evidence of the sub-seafloor presence of L. ochracea, G. ferruginea and PV-1 at Axial Volcano was obtained from the presence of these bacteria on a trap that had been placed within an active vent, and also in a vent fluid sample. If indeed these bacteria are present in the sub-seafloor, it may be an indication that the surface expression of iron oxide deposits at Axial Volcano is minimal in comparison to what exists beneath the seafloor.

Characterization and implications of the cell surface reactivity of Calothrix sp. strain KC97.

The cell surface reactivity of the cyanobacterium Calothrix sp. strain KC97, an isolate from the Krisuvik hot spring, Iceland, was investigated in terms of its proton binding behavior and charge characteristics by using acid-base titrations, electrophoretic mobility analysis, and transmission electron microscopy. Analysis of titration data with the linear programming optimization method showed that intact filaments were dominated by surface proton binding sites inferred to be carboxyl groups (acid dissociation constants [pK(a)] between 5.0 and 6.2) and amine groups (mean pK(a) of 8.9). Sheath material isolated by using lysozyme and sodium dodecyl sulfate generated pK(a) spectra similarly dominated by carboxyls (pK(a) of 4.6 to 6.1) and amines (pK(a) of 8.1 to 9.2). In both intact filaments and isolated sheath material, the lower ligand concentrations at mid-pK(a) values were ascribed to phosphoryl groups. Whole filaments and isolated sheath material displayed total reactive-site densities of 80.3 x 10(-5) and 12.3 x 10(-5) mol/g (dry mass) of cyanobacteria, respectively, implying that much of the surface reactivity of this microorganism is located on the cell wall and not the sheath. This is corroborated by electrophoretic mobility measurements that showed that the sheath has a net neutral charge at mid-pHs. In contrast, unsheathed cells exhibited a stronger negative-charge characteristic. Additionally, transmission electron microscopy analysis of ultrathin sections stained with heavy metals further demonstrated that most of the reactive binding sites are located upon the cell wall. Thus, the cell surface reactivity of Calothrix sp. strain KC97 can be described as a dual layer composed of a highly reactive cell wall enclosed within a poorly reactive sheath.

Computational and experimental approaches to studying metal interactions with microbial biofilms.

The structural and compositional heterogeneity of biofilms poses unique problems in metal fate and transport. A starting point for quantitative understanding of biofilm-metal interactions is surface complexation theory, with roots in chemical equilibria and thermodynamics. This approach permits fitting of experimental data to a variety of mathematical models from which predictive parameters, such as K, may be extracted. Applications of more sophisticated fitting routines such as tableau (as in FITEQL) or spectra pK methods provide a better measure of the heterogeneity. There remain large theoretical and computational challenges, as there is ample evidence to suggest that the principle of additivity is problematic, owing to chemical interactions between individual sorbent phases within biofilms. And finally, the question of how bacterial metabolic activity is likely to influence metal uptake by biofilms adds yet another layer of complexity for future investigations.

Cell surface electrochemical heterogeneity of the Fe(III)-reducing bacteria Shewanella putrefaciens.

Acid-base titration experiments and electrostatic force microscopy (EFM) were used to investigate the cell surface electrochemical heterogeneity of the Fe(III)-reducing bacteria, Shewanella putrefaciens. The acid-base titrations extended from pH 4 to 10, and the titration data were fit using a linear programming pKa spectrum approach. Overall, a five-site model accounted for the observed titration behavior with the most acidic sites corresponding to carboxylic groups and phosphodiester groups, intermediate sites phosphoryl groups, and two basic sites equivalent to amine or hydroxyl groups. The pH for the point of zero charge on the bacteria was 5.4. In EFM images of cells rinsed in solutions at pH 4.0, 7.0, and 8.0, a pronounced increase in small (< or = 100 nm diameter) high contrast patches was observed on the cells with increasing pH. The pH dependence of EFM image contrast paralleled the pattern of cell surface charge development inferred from the titration experiments; however, quantitative analysis of high contrast regions in the EFM images yielded lower surface charge values than those anticipated from the titration data. For example at pH 7, the calculated surface charge of high contrast regions in EFM images of the bacterial cells was -0.23 microC/cm2 versus -20.0 microC/cm2 based on the titration curve. The differences in surface charge estimates between the EFM images and titration data are consistent not only with charge development throughout the entire volume of the bacterial cell wall (i.e., in association with functional groups that are not directly exposed at the cell surface) but also with the presence of a thin structural layer of water containing charge-compensating counterions. In combination, the pKa spectra and EFM data demonstrate that a particularly high degree of electrochemical heterogeneity exists within the cell wall and at the cell surface of S. putrefaciens.

Proton binding by hydrous ferric oxide and aluminum oxide surfaces interpreted using fully optimized continuous pKa spectra.

A modified regularized least squares pKa spectrum approach is proposed to determine proton stability constants and concentrations for binding sites on hydrous ferric oxide (HFO) and aluminum oxide surfaces. Acid-base titration data are fit to a continuous binding site model for the system represented as a pKa spectrum. The modified parameter fitting method optimizes simultaneously for both smoothness of the pKa spectrum and goodness-of-fit, whereas other methods optimize for goodness-of-fit given a fixed smoothness factor. The modified method is tested with aluminum oxide and recovers values consistent with theoretical values. The regularized pKa spectrum method optimized for smoothness is applied to prepared samples of two types of HFO. The prepared HFO samples differ only in the total iron concentration of the parent solution. The resultant pKa distributions are compared to proton binding constants from MUSIC model results for crystalline iron oxides. The types of binding sites in the HFO sample are consistent with theoretical binding site stability constants for crystalline iron oxides. Overall, the prepared HFO samples have binding constants most consistent with values for lepidocrocite and goethite.

Mineral precipitation by epilithic biofilms in the speed river, ontario, Canada.

Epilithic microbial communities, ubiquitously found in biofilms on submerged granite, limestone, and sandstone, as well as on the concrete support pillars of bridges, were examined in the Speed River, Ontario, Canada. Transmission electron microscopy showed that attached bacteria (on all substrata) were highly mineralized, ranging from Fe-rich capsular material to fine-grained (<1 mum) authigenic (primary) mineral precipitates. The authigenic grains exhibited a wide range of morphologies, from amorphous gel-like phases to crystalline structures. Energy-dispersive X-ray spectroscopy indicated that the most abundant mineral associated with epilithic bacteria was a complex (Fe, Al) silicate of variable composition. The gel-like phases were similar in composition to a chamositic clay, whereas the crystalline structures were more siliceous and had compositions between those of glauconite and kaolinite. The consistent formation of (Fe, Al) silicates by all bacterial populations, regardless of substratum lithology, implies that biomineralization was a surface process associated with the anionic nature of the cell wall. The adsorption of dissolved constituents from the aqueous environment contributed significantly to the mineral formation process. In this regard, it appears that epilithic microbial biofilms dominate the reactivity of the rock-water interface and may determine the type of minerals formed, which will ultimately become part of the riverbed sediment. Because rivers typically contain high concentrations of dissolved iron, silicon, and aluminum, these findings provide a unique insight into biogeochemical activities that are potentially widespread in natural waters.

Distribution of Hydrogenase Genes in Desulfovibrio spp. and Their Use in Identification of Species from the Oil Field Environment.

The distribution of genes for [Fe], [NiFe], and [NiFeSe] hydrogenases was determined for 22 Desulfovibrio species. The genes for [NiFe] hydrogenase were present in all species, whereas those for the [Fe] and [NiFeSe] hydrogenases had a more limited distribution. Sulfate-reducing bacteria from 16S rRNA groups other than the genus Desulfovibrio (R. Devereux, M. Delaney, F. Widdel, and D. A. Stahl, J. Bacteriol. 171:6689-6695, 1989) did not react with the [NiFe] hydrogenase gene probe, which could be used to identify different Desulfovibrio species in oil field samples following growth on lactate-sulfate medium.

Remobilization of toxic heavy metals adsorbed to bacterial wall-clay composites.

Significant quantities of Ag(I), Cu(II), and Cr(III) were bound to isolated Bacillus subtilis 168 walls, Escherichia coli K-12 envelopes, kaolinite and smectite clays, and the corresponding organic material-clay aggregates (1:1, wt/wt). These sorbed metals were leached with HNO3, Ca(NO3)2, EDTA, fulvic acid, and lysozyme at several concentrations over 48 h at room temperature. The remobilization of the sorbed metals depended on the physical properties of the organic and clay surfaces and on the character and concentration of the leaching agents. In general, the order of remobilization of metals was Cr much less than Ag less than Cu. Cr was very stable in the wall, clay, and composite systems; pH 3.0, 500 microM EDTA, 120-ppm [mg liter-1] fulvic acid, and 160-ppm Ca remobilized less than 32% (wt/wt) of sorbed Cr. Ag (45 to 87%) and Cu (up to 100%) were readily removed by these agents. Although each leaching agent was effective at mobilizing certain metals, elevated Ca or acidic pH produced the greatest overall mobility. The organic chelators were less effective. Lysozyme digestion of Bacillus walls remobilized Cu from walls and Cu-wall-kaolinite composites, but Ag, Cr, and smectite partially inhibited enzyme activity, and the metals remained insoluble. The extent of metal remobilization was not always dependent on increasing concentrations of leaching agents; for example, Ag mobility decreased with some clays and some composites treated with high fulvic acid, EDTA, and lysozyme concentrations. Sometimes the organic material-clay composites reacted in a manner distinctly different from that of their individual counterparts; e.g., 25% less Cu was remobilized from wall- and envelope-smectite composites than from walls, envelopes, or smectite individually in 500 microM EDTA. Alternatively, treatment with 160-ppm Ca removed 1.5 to 10 times more Ag from envelope-kaolinite composites than from the individual components. The particle size of the deposited metal may account for some of the stability changes; those metals that formed large, compact aggregates (Cr and Ag) as seen by transmission electron microscopy were less likely to be remobilized. In summary, it is apparent that remobilization of toxic heavy metals in sediments, soils, and the vadose zone is a complicated issue. Predictions based on single inorganic or organic component systems are too simplistic.

Bacterial sorption of heavy metals.

Four bacteria, Bacillus cereus, B. subtilis, Escherichia coli, and Pseudomonas aeruginosa, were examined for the ability to remove Ag+, Cd2+, Cu2+, and La3+ from solution by batch equilibration methods. Cd and Cu sorption over the concentration range 0.001 to 1 mM was described by Freundlich isotherms. At 1 mM concentrations of both Cd2+ and Cu2+, P. aeruginosa and B. cereus were the most and least efficient at metal removal, respectively. Freundlich K constants indicated that E. coli was most efficient at Cd2+ removal and B. subtilis removed the most Cu2+. Removal of Ag+ from solution by bacteria was very efficient; an average of 89% of the total Ag+ was removed from the 1 mM solution, while only 12, 29, and 27% of the total Cd2+, Cu2+, and La3+, respectively, were sorbed from 1 mM solutions. Electron microscopy indicated that La3+ accumulated at the cell surface as needlelike, crystalline precipitates. Silver precipitated as discrete colloidal aggregates at the cell surface and occasionally in the cytoplasm. Neither Cd2+ nor Cu2+ provided enough electron scattering to identify the location of sorption. The affinity series for bacterial removal of these metals decreased in the order Ag greater than La greater than Cu greater than Cd. The results indicate that bacterial cells are capable of binding large quantities of different metals. Adsorption equations may be useful for describing bacterium-metal interactions with metals such as Cd and Cu; however, this approach may not be adequate when precipitation of metals occurs.

Physicochemical interaction of Escherichia coli cell envelopes and Bacillus subtilis cell walls with two clays and ability of the composite to immobilize heavy metals from solution.

Isolated Escherichia coli K-12 cell envelopes or Bacillus subtilis 168 cell walls were reacted with smectite or kaolinite clay in distilled deionized water (pH 6.0); unbound envelopes or walls were separated by sucrose density gradient centrifugation, and the extent of adsorption was calculated. At saturation, both clays adsorbed approximately 1.0 mg (dry weight) of envelopes or walls per mg (dry weight) of clay. Clays showed a preference for edge-on orientation with both walls and envelopes, which was indicative of an aluminum polynuclear bridging mechanism between the wall or envelope surface and the clay edge. The addition of heavy metals increased the incidence of planar surface orientations, which suggested that multivalent metal cation bridging was coming into play and was of increasing importance. The metal-binding capacity of isolated envelopes, walls, clays, and envelope-clay or wall-clay mixtures was determined by atomic absorption spectroscopy after exposure to aqueous 5.0 mM Ag+, Cu2+, Cd2+, Ni2+, Pb2+, Zn2+, and Cr3+ nitrate salt solutions at pHs determined by the buffering capacity of wall, envelope, clay, or composite system. The order of metal uptake was walls greater than envelopes greater than smectite clay greater than kaolinite clay for the individual components, and walls plus smectite greater than walls plus kaolinite greater than envelopes plus smectite greater than envelopes plus kaolinite for the mixtures. On a dry-weight basis, the envelope-clay and wall-clay mixtures bound 20 to 90% less metal than equal amounts of the individual components did.(ABSTRACT TRUNCATED AT 250 WORDS)

Metal Interactions with Microbial Biofilms in Acidic and Neutral pH Environments.

Microbial biofilms were grown on strips of epoxy-impregnated filter paper submerged at four sites in water contaminated with metals from mine wastes. At two sample stations, the water was acidic (pH 3.1); the other sites were in a lake restored to a near neutral pH level by application of a crushed limestone slurry. During a 17-week study period, planktonic bacterial counts increased from 10 to 10 CFU/ml at all sites. Biofilm counts increased rapidly over the first 5 weeks and then leveled to 10 CFU/cm in the neutral pH system and 10 CFU/cm at the acidic sites. In each case, the biofilms bound Mn, Fe, Ni, and Cu in excess of the amounts adsorbed by control strips covered with nylon filters (pore size, 0.22 mum) to exclude microbial growth; Co bound under neutral conditions but not under acidic conditions. Conditional adsorption capacity constants, obtained graphically from the data, showed that biofilm metal uptake at a neutral pH level was enhanced by up to 12 orders of magnitude over acidic conditions. Similarly, adsorption strength values were usually higher at elevated pH levels. In thin sections of the biofilms, encapsulated bacterial cells were commonly found enmeshed together in microcolonies. The extracellular polymers often contained iron oxide precipitates which generated weak electron diffraction patterns with characteristic reflections for ferrihydrite (Fe(2)O(3) . H(2)O) at d equaling 0.15 and 0.25 nm. At neutral pH levels, these deposits incorporated trace amounts of Si and exhibited a granular morphology, whereas acicular crystalloids containing S developed under acidic conditions.

Structural and biochemical analyses of a surface array protein of Campylobacter fetus.

Electron microscopic examination of ultrathin sections and freeze-etched and shadow cast preparations of a bovine prepuce isolate of Campylobacter fetus VC119 showed an S layer with subunits in an apparent linear arrangement. Surface radioiodination, enzyme digestion, low-pH extraction, and Western immunoblotting showed that the layer was composed mainly of one protein which is the predominant protein antigen of C. fetus. This protein was purified to homogeneity by gel filtration, ion-exchange chromatography, and high-performance liquid chromatography. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis showed an apparent molecular weight of 131,000 for this protein with a pI of 6.35, and no carbohydrate could be detected by a variety of techniques. Amino acid composition analysis showed that the protein contained approximately 1,304 residues per molecule, 41.2% of which were hydrophobic and approximately 22% of which were acidic. Cysteine and histidine were absent. Circular dichroism spectra showed that the prominent structure of the S layer protein was a beta-pleated sheet (36%) with aperiodic foldings (31%); a moderate amount of alpha-helix (28%) and a low amount of beta-turn (5%) were also present. The N-terminal amino acid sequence was determined for the first 18 residues. No sequence homology with other S layer proteins was found.

Physicochemical roles of soluble metal cations in the outer membrane of Escherichia coli K-12.

Atomic absorption spectroscopy of isolated native and EDTA-modified (lipopolysaccharide-depleted) outer membrane revealed trace amounts of potassium, manganese, and iron (1.0-7.0 nmol/mg dry weight outer membrane). Sodium, magnesium, and calcium were approximately one order of magnitude more plentiful, but EDTA-modified outer membrane was deficient in calcium. When metal-binding assays were conducted to find the binding capacity of native and EDTA-modified outer membrane, potassium bound poorly compared with sodium. However, there was no difference in the binding of these ions between the OM preparations. In contrast, reduced amounts of magnesium, calcium, manganese, and iron III bound to the EDTA-modified OM. Partitioning of intact cells in a biphasic dextran-polyethyleneglycol system indicated that the reduced lipopolysaccharide content of the EDTA-modified outer membrane increased the hydrophobicity of the cell surface. Exposure of control and EDTA-treated cells to divalent metal salt solutions before phase partitioning also increased cell surface hydrophobicity. Freeze-etching showed that sodium ions had no effect on the membrane fractures observed in control cells, but with EDTA-treated cells, this cation increased the occurrence of small outer membrane fractures (plateaus) which are characteristic of EDTA treatment. Both magnesium and manganese increased the frequency of outer membrane cleavage in control cells, whereas calcium did not. In contrast, all three divalent metallic ions increased the frequency and extent of cleavage in the outer membrane of EDTA-treated cells.

Site specificity of metallic ion binding in Escherichia coli K-12 lipopolysaccharide.

The site specificity of metallic ion binding in Escherichia coli K-12 lipopolysaccharide was assessed by collecting high-resolution phosphorus nuclear magnetic resonance spectra in the presence of manganese, a paramagnetic divalent cation. This technique revealed high-affinity interactions between the cation and all of the lipopolysaccharide phosphoryl groups. To ascertain whether the carboxyl groups of 2-keto-3-deoxyoctonate contributed to the metal cation binding, lipopolysaccharide was chemically modified using a glycine ethyl ester - carbodiimide reaction. Of the three available carboxyl groups, only one was neutralized by the exogenously added ligand; the others appeared to be cross-linked within the molecule. By analogy, only one carboxyl group should be freely available for binding metallic ions, while the others are probably neutralized by the close proximity of endogenous amino substituents. Although high-resolution phosphorus nuclear magnetic resonance showed that an intermolecular conformational change had occurred after the carboxyl groups were neutralized, titration with manganese revealed no differences in the apparent strength of the interactions between the cation and the phosphoryl groups. Together, these data suggest that the high affinity of lipopolysaccharide for divalent metallic ions can be attributed primarily to the phosphoryl substituents and not free carboxyl groups.

Structure and cell envelope associations of flagellar basal complexes of Vibrio cholerae and Campylobacter fetus.

To isolate intact flagella with basal complexes from Vibrio cholerae, a rhamnolipid hemolysin from Pseudomonas aeruginosa was used to disrupt the cell envelope and flagellar sheath. The nonionic detergent, Triton X-100, provided similar results for Campylobacter fetus. Each of these basal complexes possessed, in addition to the four classical rings, concentric membrane rings (CMR's) similar to those found in Aquaspirillum serpens. Through the use of stereo imaging (which allows structures to be visualized in three dimensions) of thin sections of cells which had been sequentially treated with a number of envelope perturbants (i.e., ethylenediaminetetraacetate, lysozyme, Triton X-100, rhamnolipid hemolysin, and sodium dodecyl sulfate), we have progressively exposed the component parts of the basal organelles in V. cholerae and C. fetus. Since the action of these envelope perturbants has been well documented, we have been able to determine the associations of the exposed portions of the flagellar basal complex and the layer of the cell envelope in which they would normally reside. From our observations we have concluded that in both V. cholerae and C. fetus the L ring is embedded in the outer membrane and the P ring is associated with the peptidoglycan. The CMR's are bracketed by the L and P rings and are sandwiched between the outer membrane and the peptidoglycan. Elements of both the S and M rings appear to be associated with the plasma membrane.