Bacillus subtilis W23 was infected with a clear-plaque variant of SP-10 phage, namely, SP-lOc. Exogenous thymidine was not incorporated into phage DNA (even in the presence of deoxyadenosine), nor was there any transfer of thymidine nucleotides from bacterial to viral DNA. The lytic program was unaffected by concentrations of 5-fluorodeoxyuridine sufficient to reduce bacterial DNA synthesis by >95%. Although these data are consistent with the interpretation that thymidine nucleotides are excluded from phage DNA, formic acid digests of SP-lOc DNA contained what appeared to be the four conventional bases; however, adenine and thymine were not recovered in equimolar yields. DNA-RNA hybridization and hybridization competition experiments were done. Synthesis of host RNA started to wane moments postinfection and stopped completely by 36 min. SP-lOc coded for discrete classes ofearly and late RNA. The possibility of discrete subclasses of early RNA exists. Replication of the bacterial genome appeared to terminate 12 min postinfection. Degradation of the host DNA to acid-soluble material started at 36 min and, by the end of the latent period, >90% of the host chromosome was hydrolyzed. Four apparent phage-coded enzymes have been identified. A di-and triphosphatase degraded dUTP, dUDP, dTTP, and dTDP (and, to a lesser extent, dCDP and dCTP) to the corresponding monophosphates; the enzyme had no apparent activity on dATP and dGTP. SP-lOc also coded for a DNA-dependent DNA polymerase, lysozyme, and a nuclease that degrades native bacterial DNA. Judging from the dependence of enzyme synthesis on the time of addition of rifampin (an inhibitor of the initiation of RNA synthesis), messengers for the di-and triphosphatase, as well as the nuclease, are transcribed from promoters that start to function 6 min postinfection. Promoters for polymerase and lysozyme did not become functional until 8 and 16 min postinfection, respectively. MATERIALS AND METHODS Phage and bacteria. B. subtilis W23 (ATCC 23059) was used as host. SP-lOc is a clear-plaque variant present in ATCC 23059B. SP-lOc was readily inactivated by antiserum prepared against SP-10c and provided by K. Bott (Table 1). The eclipse and latent periods of SP-10c at 37°C were 25 and 55 min, 84

A 300C, functional messengers for dCMP hydroxymethylase first appeared 3 to 6 min postinfection and reached their maximum levels at 12 min. Chloramphenicol, added before the phage, reduced the rate of mRNA accumulation. When the antibiotic was added 6 min postinfection, mRNA levels increased at their normal rate but there was no obvious repression of messenger accumulation. Delaying the addition of drug until 8 or 12 min had progressively less effect on the pattern of hydroxymethylase mRNA metabolism. When chloramphenicol was present from preinfection times or from 6 min postinfection, all hydroxymethylase mRNA's synthesized were stable; at later times, however, the ability of the drug to stabilize mRNA decreased with its ability to delay the turnoff of mRNA production. An overaccumulation of hydroxymethylase mRNA was also seen when phage-specific DNA synthesis was inhibited either by mutational lesion in an essential viral gene or by 5-fluorodeoxyuridine. By min 20 of a DNA-negative program, hydroxymethylase mRNA synthesis was repressed to the point where it no longer compensated for decay. However, a finite level of hydroxymethylase mRNA synthesis was maintained at later times of a DNA-negative infection. Such results indicate that replication of the phage chromosome is necessary but not sufficient for a complete turnoff of hydroxymethylase mRNA production. Functions controlled by the maturation-defective proteins (the products of genes 55 and 33) played only a minor role in the regulation of hydroxymethylase mRNA metabolism. Thus, we favor the hypothesis that a complete turnoff of hydroxymethylase messenger production requires one or more new proteins as well as an interval of DNA replication. The absence of DNA synthesis had no particular effect upon dihydrofolate reductase messenger production. The preinfection addition of chloramphenicol likewise had little effect on dihydrofolate reductase messenger metabolism. These latter data imply that prior synthesis of a phage-coded protein synthesis may not be required for the turnoff of reductase messenger production.