Transformed cells have been separated from nontransformed cells by centrifugation on a density gradient of Renografin-76. Separation was achieved both on a linear gradient and on a discontinuous gradient. Under optimal conditions, all of the cells in one band (median density, 1.110 g/ml) were transformants, whereas virtually all cells in the other (median density, 1.131) were nontransformants. In some instances, recentrifugation of the transformant band further enriched the transformant population. The transformed population can also be enriched by zonal centrifugation in a linear gradient of Ficoll. However, this technique is far less efficient than centrifugation in Renografin-76. Sincethe densityof competent cells is identical to that of transformants, we conclude that the low density is a property of competent cells. The significance of this low density to the physiology of competent cells is discussed. MATERIALS AND METHODS Bacterial strains. Strains of B. subtilis useJ were 168 (trpC168), SBI (hisAl trpCJ68, markers unlinked, previously designated his,-, try27), and WB746 (prototrophic). SB1 and WB746 were derived from strain 168. Chemicals. Ficoll (obtained from Pharmacia, Uppsala, Sweden) was dialyzei ovemight against two changes of distilled water, lyophilized, dissolved in Davis minimal medium (5), and sterilized by autoclaving. Renografin-76 (Squibb Methylglucamine Diatrizoate Injection USP, E. R. Squibb and Sons, New York, N.Y.) was stored at room temperature after dilution with sterile Davis minimal medium. The commercial preparation, a 76% solution of methylglucamine diatrizoate, was designated 100% in concentration to simplify calculations (21). DNA was prepared by the method of Marmur (10), in some cases with the omission of ribonuclease treatment and isopropanol precipitation. Pancreatic deoxyribonuclease (five times crystallized) was obtained from Calbio-876

Rates of deoxyribonucleic acid, ribonucleic acid, and protein synthesis were examined in purified competent cells of Bacillus subtilis during the development of the transformable state. To become competent, a cell must depart from the normal course of vegetative growth and pass through a precompetent phase beginning as early as 90 to 180 min before the appearance of transformability. While in the precompetent state, the cell decreases its rate of deoxyribonucleic acid synthesis and lowers its ratio of ribonucleic acid synthesis to protein synthesis. This altered pattern of synthesis eventually leads to a decreased buoyant density of precompetent cells. Once a cell has become both precompetent and low in density, it can be converted to a competent (transformable) cell. The early alterations in macromolecular synthesis were found in two competence regimens, one utilizing a nutritional step-down and one free of such a shift. The data imply that the precompetent state is a generalized characteristic of the B. subtilis transformation system and is not specific to the procedure used to allow competence development. Since precompetence-specific events occur very early in a competence regimen, we conclude that the induction of precompetence is unrelated to sporulation or a nutritional shift.

Human health and ecological risks must be balanced at hazardous waste sites in order to ensure that remedial actions prevent unacceptable risks of either type. Actions that are designed to protect humans may fail to protect nonhuman populations and ecosystems or may damage ecosystems. However, there is no common scale of health and ecological risk that would allow comparisons to be performed. This paper presents an approach to addressing this problem based on classifying all risks (i.e., health and ecological risks due contaminants and remediation) as insignificant (de minimis), highly significant (de manifestis), or intermediate. For health risks the classification is based on standard criteria. However, in the absence of national guidance concerning the acceptability of ecological risks, new ecological criteria are proposed based on an analysis of regulatory precedents. Matrices and flow charts are presented to guide the use of these risk categories in remedial decision making. The assessment of mercury contamination of the East Fork Poplar Creek is presented as an example of the implementation of the approach.

A mutant (uvr-1) of Bacillus subtilis that is deficient in excision of ultraviolet (UV)-induced pyrimidine dimers from deoxyribonucleic acid (DNA) shows a marked increase in ability to survive UV irradiation when plated on amino acid-supplemented agar medium compared with its survival ability when plated on nutrient plating medium, the effect is considered to be one of growth-dependent lethality. Irradiated stationary phase uvr-1 cells, incubated in liquid medium lacking amino acids required for growth, recover from this sensitivity to rich medium within 3 to 4 h after irradiation. Recovery is greatly reduced in the absence of glucose oiminated. Exponentially growing cells have a limited ability to recover from sensitivity to rich medium. Growth-dependent lethality can also occur in liquid medium. In nutrient broth the ability of irradiated stationary-phase uvr-1 cells to form colonies on defined agar medium decreases during postirradiation incubation, but treatmeth with chloramphenicol inhibits the loss of colony-forming ability. Recovery from sensitivity to rich media is inhibited by caffeine but not by 6-(p-hydroxyphenylazo)-uracil, and inhibitor of DNA replication. Alkaline sucrose gradient profiles show that conditions allowing recovery also favor maintaining intact DNA strands, whereas DNA strand breakage or degradation is associated with loss of viability. Recovery from sensitivity to rich medium has not been observed in the Ur+ parent or in strains carrying the mutations uvs-42 (another deficiency in dimer excision), recA1, or polA59. A uvr-1 recA1 mutants shows a higher level of recovery than does the recA1 single mutant, but a much lower level than the uvr-1 single mutant. Apparently, both the uvr-1 defect and Rec+ and PoII+ functions are essential for recovery from sensitivity to rich medium. For optimal recovery, growth immediately after irradiation must be delayed. The process requires energy, apparently involves recombination, and probably results in rejoining of DNA strands in which incision but not excision has occurred.

Deoxyribonucleic acid repair synthesis was studied in one wild-type and two mutant strains of Bacillus subtilis that are defective in excision of pyrimidine dimers. The cells were irradiated with ultraviolet light, and 6-(p-hydroxyphenylazo)-uracil was used to block replicative synthesis, allowing only repair synthesis. One of the mutations (uvs-42) resulted in a severe inhibition of incision, diiner excision, and repair synthesis. In contrast, the other mutant (uvr-1 ) slowly incised and excised dimers and did repair synthesis in patches which appear to be severalfold longer than those in the wild-type strain, apparently because large gaps are produced at excision sites. The results indicate that the primary defect in uvs-42 cells is in initiation of dimer excision, whereas the uvr-1 mutation appears to be a defect in the exonuclease normally used to complete dimer excision.