Search

A multiple mutant of Bacillus subtilis that grows in an unusual double-helix morphology was studied. Construction of models led to the assumption that cell surface elongation must proceed in a helical path in this mutant. The observation that all newly formed double-helix clones propagated, after spore outgrowth in fluid culture, consisted of closed-circular structures suggested that double-helix structures are tension-registered forms.

Purified minicells produced by Bacillus subtilis div IV-B1 mutants were used to immunize rabbits. Immune serum was obtained that agglutinated minicells and was able to form precipitin lines when reacted with minicell antigens in double-diffusion or immunoelectrophoretic procedures. Antiminicell serum agglutinated rod-shaped B. subtilis cells to long filaments produced by growth of a cell division-defective mutant at restrictive temperature. These findings indicate that minicells are immunogens capable of eliciting the production of antibodies that cross-react with the lateral, cylindrical regions of B. subtilis rods. It appears, therefore, that poles share a common antigen(s) with cylindrical regions of the cell.

Cells of Bacillus subtilis heated in high concentrations of sodium dodecyl sulfate (5%) and then washed free of detergent with a hot salt solution (80 C) become structurally reorganized into regions of densely compacted cytoplasm (termed zebras) and regions of sparsely filled material (termed spaces). Size distribution studies of zebras indicate that division-suppressed mutants and wild-type cells both yield zebras of comparable length. Similarly the lengths of zebras found in populations emerging from spores are uniform in one-, two-, three-, and four-zebra-containing cells. In contrast, the length of spaces is slightly larger than that of zebras and is unusually large in two-zebra-containing cells. The locations of zebras and spaces along cell length have been studied in spore out-growth populations. A statistical procedure developed previously in genome location investigations was used to analyze the location of zebras along cell length. The data indicate that as cells elongate, new sites arise where the cell contents are strongly bound to the cell surface. Within filament populations produced by division-suppressed mutants there is a linear relationship of mean filament length and zebra number per filament. These data indicate that cytoplasm in filaments with no obvious structural compartmentalizations may be organized into units associated with particular regions of cell surface. The attachment of cell contents to the cell surface may involve deoxyribonucleic acid. Zebra-containing cells digested with proteolytic enzyme and ribonuclease are converted to cells that contain a crystalline-like granule fixed at the location of each zebra. Exposure to deoxyribonuclease mobilizes these granules within the cell wall.

Organized bacterial superstructures have been used as 3-D templates for the fabrication of ordered inorganic−organic fibrous composites. Preformed magnetic (Fe3O4) and semiconducting (CdS) inorganic nanoparticles were incorporated into macroscopic threads of Bacillus subtilis by reversible swelling of the superstructure in colloidal sols. The air-dried mineralized fibers consisted of a closely packed array of 0.5 μm diameter multicellular bacterial filaments, each of which was coated with a 30−70 nm thick layer of aggregated colloidal particles. Inorganic patterning of the interfilament spaces was influenced by the surface charge of the nanoparticles used. Whereas negatively charged magnetite colloids gave good infiltration and replication of the bacterial superstructure, the neutral-ligand-capped CdS colloid, although internalized to some extent, was preferentially localized at the surface of the thread. Positively charged sols of TiO2, in contrast, did not penetrate the swollen fiber but produced coherent surface coatings of uniform thickness. Attempts to pattern the deposition of CdS using molecular precursors by exposing a Cd(II)-containing bacterial fiber to H2S gas produced an uneven surface coat of CdS particles. Removal of the bacterial component from the magnetic composite by heating at elevated temperatures resulted in structural collapse.

Bacterial threads of up to 1 m in length have been produced from filaments of separation-suppressed mutants of Bacillus subtilis. Individual threads may contain 20,000 cellular filaments in parallel alignment. The tensile properties of bacterial threads have been examined by using conventional textile engineering techniques. The kinetics of elongation at constant load are indicative of a viscoelastic material. Both Young's modulus and breaking stress are highly dependent upon relative humidity. By extrapolation to 100% relative humidity, it appears that cell walls may be able to bear only internal osmotic pressures of about 2 atmospheres (2.03 X 105(5) Pa) in living cells. Similarly, the strength of wall material limits the amount of cell-surface charge permissible to only a small fraction of that known to be carried by the negatively charged wall polymers.

Engineering approaches used in the study of textile fibers have been applied to the measurement of mechanical properties of bacterial cell walls by using the Bacillus subtilis bacterial thread system. Improved methods have been developed for the production of thread and for measuring its mechanical properties. The best specimens of thread produced from cultures of strain FJ7 grown in TB medium at 20 degrees C varied in diameter by a factor of 1.09 over a 30-mm thread length. The stress-strain behavior of cell walls was determined over the range of relative humidities between 11 and 98%. Measurements of over 125 specimens indicated that cell wall behaved like other viscoelastic polymers, both natural and man-made, exhibiting relaxation under constant elongation and recovery upon load removal. This kinetic behavior and also the cell wall strength depended greatly on humidity. The recovery from extension observed after loading even up to a substantial fraction of the breaking load indicated that the properties measured were those of cell wall material rather than of behavior of the thread assemblage. Control experiments showed that neither drying of thread nor the length of time it remained dry before testing influenced the mechanical properties of the cell walls. Specimens drawn from TB medium and then washed in water and redrawn were found to be stiffer and stronger than controls not washed. However, tensile properties were not changed by exposure of cells to lysozyme before thread production. This suggests that glycan backbones are not arranged along the length of the cell cylinder. The strength of the cell wall in vivo was estimated by extrapolation to 100% relative humidity to be about 3 N/mm2. Walls of this strength would be able to bear a turgor pressure of 6 atm (ca. 607.8 kPa), but if the increase in strength of water-washed threads was appropriate, the figure could be 24 atm (ca. 2,431.2 kPa).

The effect of D-cycloserine on the establishment of twist states in Bacillus subtilis macrofibers was examined. Macrofibers produced in the presence of the drug differed in twist compared with those produced in its absence. The degree of twist alteration was dependent on the concentration of D-cycloserine in the growth medium. Macrofibers of different twist states representative of the entire twist spectrum from tight left-handedness to tight right-handedness were produced in strains FJ7 and C6D in four different ways: by control of the concentration of D-alanine, magnesium sulfate, or ammonium sulfate in the growth medium or by control of the growth temperature. The structures so produced were used to determine the effect of D-cycloserine on twist establishment starting from different twist states throughout the twist spectrum. In all but one case, twist resulting from growth in the presence of D-cycloserine was further towards the left-hand end of the twist spectrum than that produced in its absence, the exception being the unusual left-handed twist states produced in strains C6D and the closely related RHX 11S at high D-alanine concentrations described here. Studies of the interaction between D-cycloserine and D-alanine both used alone and used independently with the other twist-modifying systems (temperature, magnesium sulfate, and ammonium sulfate) revealed that changes in twist resulting from D-cycloserine were always in the opposite direction from those resulting from D-alanine. This antagonism suggests that the biochemical mechanism of twist regulation involves the metabolism of peptidoglycan, particularly reactions involving D-alanine or the dipeptide D-alanyl-D-alanine. This antagonism suggests that the biochemical mechanism of twist regulation involves the metabolism of peptidoglycan, particularly reactions involving D-alanine or the dipeptide D-alanyl-D-alanine. The possibility that peptidoglycan cross-linking is involved is discussed.

The steady-state twist of Bacillus subtilis macrofibers produced by growth in complex medium was found to vary as a function of the magnesium and ammonium concentrations. Four categories of macrofiber-producing strains that differed in their response to temperature regulation of twist were studied. Macrofibers were cultured in the complex medium TB used in previous experiments and in two derivative media, T (consisting of Bacto Tryptose), in which most strains produced left-handed structures, and Be (consisting of Bacto Beef Extract), in which right-handed macrofibers arose. In nearly all cases, increasing concentrations of magnesium led to the production of macrofibers with greater right-handed twist. Some strains unable to form right-handed structures as a function of temperature could be made to do so by the addition of magnesium. Inversion from right- to left-handedness in strain FJ7 induced by temperature shift-up was blocked by the addition of magnesium. The presence of magnesium during a high-temperature pulse did not block the establishment of "memory," although it delayed the initiation of the transient inversion following return to low temperature. The twist state of macrofibers grown without a magnesium supplement was not instantaneously affected by the addition of magnesium. Such fibers were, however, protected from lysozyme attack and associated relaxation motions. Lysozyme degradation of purified cell walls (both intact and lacking teichoic acid) was also blocked by the addition of magnesium. Ammonium ions influenced macrofiber twist development towards the left-hand end of the twist spectrum. Macrofiber twist produced in mixtures of magnesium and ammonium was strain and medium dependent.(ABSTRACT TRUNCATED AT 250 WORDS)

The folding process required for helical macrofiber formation after the outgrowth of Bacillus subtilis spores was found to be blocked by very low concentrations of penicillin G (1 to 3 ng/ml). Under such conditions, growth and septation without cell separation resulted in characteristic disorganized multicellular structures. Higher concentrations (4 and 10 ng/ml) were needed to inhibit spore outgrowth and vegetative growth, respectively.

The kinetics of Bacillus subtilis macrofiber helix hand inversion was examined. Inversion was induced by transfer of structures produced in one medium to another medium. When cultured at 20 degrees C in either medium, the doubling time was approximately 100 min. To establish a baseline, the macrofiber twist state produced in one medium was measured over the same time course during which other macrofibers underwent inversion after transfer to a second medium. The baseline was used to identify the time of inversion initiation: the point at which curves representing changes of twist as a function of time after transfer to the new medium intersected the baseline. Right- and left-handed macrofibers of different twists were produced by growth in mixtures of TB and S1 media. These were used to determine the influence of initial twist on the time course of inversion initiation. In the right to left inversion, a positive correlation was found between initial twist and the time of inversion initiation. The left to right inversion differed, however, in that a constant time was required for inversion initiation regardless of the starting left-handed twist. When a nutritional pulse was administered by transferring fibers from TB to S1 to TB medium, the time to initiation of inversion was found to decrease with incubation of increasing duration in S1 medium. A similar pulse protocol was used in conjunction with inhibitors to examine the protein and peptidoglycan synthesis requirements for the establishment of nutrition-induced memory that leads to initiation of inversion. Nutritionally induced right to left inversion but not left to right inversion required protein synthesis. The addition of trypsin to left-handed macrofibers apparently required, as described previously for the temperature-regulated twist system (D. Favre, D. Karamata, and N. H. Mendelson, J. Bacteriol. 164:1141-1145, 1985), for the production of left-handed twist states in the nutrition system.

Bacillus subtilis C6 phi R4 is an SPO1-resistant derivative of strain C6D, a left-hand macrofiber-producing strain described previously (N. H. Mendelson, Proc. Natl. Acad. Sci. U.S.A. 75:2478-2482, 1978). In addition to the phage resistance property, strain C6 phi R4 differs from its parent in macrofiber organization and formation of aggregates in liquid shake cultures. The phage resistance mutation was located in the gtaC gene. The macrofiber organization and aggregation phenotypes also appear to be controlled by the gtaC locus. Strains constructed by introduction of the gtaC mutation into C6D appear to be identical to the original C6 phi R4 strain in all phenotypic properties. In contrast, other constructs carrying either gtaA or gtaB that are resistant to SPO1 do not display the characteristic C6 phi R4 morphological phenotypes.

Growth and division patterns of Bacillus subtilis wild-type (div IV-A1+) and minicell-producing mutant (div IV-A1) clones were studied after spore germination during microcolony development in chambers that facilitate continuous observation with a phase contrast microscope. Data obtained from 13 div IV-A1+ clones were used to derive the equation DE equals [(mum minus 17.6)/8.8], which expresses the relationship of cell divisions present in clones of various lengths. This equation was used to determine the number of divisions expected in div IV-A1 clones if the mutant clones were able to divide as often as wild-type clones. The observed number of divisions present in mutant clones was found to be only 25.27% of the number expected on the basis of this equation. Although individual div IV-A1 clones varied in the percentage of division equivalents expressed, there appeared to be no correlation between the overall clone growth rate and the number of divisions expressed. Culturing div IV-A1+ and div IV-A1 clones together in the same growth chamber revealed that there were no diffusible interactions influencing the division phenotypes of either mutant or wild-type cells. At later stages of growth, mixed microcolonies containing cells of both genotypes were formed. A length analysis of individual cells in these populations indicated that the relative division suppression of mutant compared with wild-type cells characteristic of the initial stages of clone development was maintained. It is likely, therefore, that the excessive length of minicell-producing cells (div IV-A1) is a reflection primarily of division suppression in the mutant and not simply of mislocation of division along cell length.

In this review, we describe four approaches to the materials synthesis of organized inorganic matter. These include the use of self-assembled organic templates (transcriptive synthesis), cooperative assemblies of templates and building blocks (synergistic synthesis), spatially restricted reaction fields (morphosynthesis), and combinations of these approaches (integrative synthesis) in the area of sol−gel chemistry. We illustrate these strategies, respectively, by describing recent work on the formation of silica-based organized materials, viz. the preparation of ordered silica macrostructures using bacterial templates, template-directed synthesis of ordered hybrid mesophases and organoclays, synthesis of microskeletal frameworks of silica and other metal oxides in compartmentalized liquids, and use of bacterial superstructures in the fabrication of hierarchical macrostructures of mesoscopically ordered silica.

Static and dynamic studies of helical Bacillus subtilis macrofibers reveal that a spectrum of twisted states exists ranging from tight left-handed structures with twist equal to approximately equal to 40 left turns per mm to tight right-handed structures with twist equal to 57 right turns per mm. In the lytic-deficient strain FJ7 , twist varies as a function of growth temperature above or below 39 degrees C, where there is zero twist. The relationship between the temperature (below 39 degrees C) at which right-hand structures are produced to the time it takes for them to begin the inversion process in which they become left-handed following transfer to 48 degrees C reveals that structures with less twist are more rapidly converted to left-handedness than are those with higher values of twist. The initial response of live macrofibers to digestion by lysozyme consists of "relaxation" motions in which the twist of both left- and right-handed structures changes towards the right-hand end of the spectrum. The rate of relaxation is approximately equal to 5-fold higher at the left-hand end than at the right-hand end. These findings suggest that cell wall polymers can assume a range of structural states during helical growth and that these determine the quantitative aspects of macrofiber shape as well as the sensitivity of walls to attack by lysozyme.