A secreted phosphodiesterase/aIkaline phosphatase, APaseD, was purified from a culture of Bacillus subfilis JH646MS. Its phosphodiesterase activity was reminiscent of an APase isolated and characterized previously. lmmunoassay and N-terminal sequencing showed the two proteins to be identical. Using the first 20 amino acids of the mature protein, a BLAST search of GenBank was used to find an homologous sequence. An exact match was found but in a putative non-coding region. It was hypothesized that there was a base pair deletion in the phoD gene. A DNA fragment internal to the coding region was generated by PCR using template DNA from a strain which produced APaseD. The PCR fragment was cloned and used to interrupt the gene. Western blot analysis of the parent and the mutated strains showed that APaseD was missing in the mutant. Resequencing of the gene revealed a larger ORF encoding a protein similar in size to the 49 kDa APaseD estimated by SDS-PAGE. The promoter was then cloned, sequenced and used in phoD-lacZ promoter fusions which showed that the gene was phosphate-starvation-induced and dependent on PhoP and PhoR for expression.

Two secreted alkaline phosphatase proteins were purified from cultures of Bacillus subtilis JH646MS. The two proteins showed slight differences in subunit molecular weight, substrate specificity, and charge characteristics. A total of 62% of the first 22 amino-terminal amino acids were identical. Both sequences showed conservation of structural features identified in Escherichia coli and human alkaline phosphatases. One alkaline phosphatase was a monomer and the other was a dimer. Culture conditions for APase production. The culture conditions used to maximize vegetative APase production in B. subtilis were a modification of the procedure developed for B. licheniformis (17). The liquid defined medium in which each culture was grown and assayed was composed of ammonium sulfate (3.03 mM), sodium citrate (0.68 mM), ferric chloride (3.04 mM), manganese sulfate (1.00 mM), magnesium sulfate (35 mM), potassium phosphate (dibasic) (0.42 mM), Trizma (50 mM), and zinc chloride (0.01 mM). Before the medium was sterilized, the pH was adjusted to 7.12 by the addition of glacial acetic acid. Before inoculation, the following sterile solutions were added to the medium to achieve the indicated concentrations: fructose (135 mM) and amino acids to supplement auxotrophic requirements (50 ,ug/ml

Alkaline phosphatase (APase) is induced as a culture enters stationary phase because of limiting phosphate.The results presented here show that expression of APase is regulated both negatively and positively. PhoP, a homolog of a family of bacterial transcription factors, and PhoR, a homolog of bacterial histidine protein kinases, are required for induction of APases when phosphate becomes limiting. The induction period lasts 2 to 3 h, after which the rate of APase accumulation is decreased. Mutant strains defective in the SpoOA transcription factor failed to decrease APase production. The consequent hyperinduction of APase in a spoOA strain was dependent on phoP and phoR. spoOB and spoOF strains also overexpressed APase, suggesting that phosphorylated SpoiA is required for repression of APase. An abrB mutant allele in the presence of the mutant spoOA allele in these strains did not significantly change the APase hyperinduction phenotype, demonstrating that SpoOA repression ofabrB expression is not the mechanism by which SpoOA-P regulates APase expression. Our previous report that spoOA mutants do not express APases is in conflict with the present data. We show here that the previously used mutants and a number of commonly used spo0 strains, all of which have an APase deficiency phenotype, contain a previously unrecognized mutation in phoR.

RNA transcripts derived from recombinant chimeras between the highly virulent GDVII virus and the less virulent BeAn virus were constructed to study the molecular pathogenesis of Theiler's murine encephalomyelitis virus infection. The presence of the BeAn 5' noncoding sequences in chimera 2 (BeAn 5' noncoding sequences joined with the GDVII nucleotides encoding the polyprotein and present in the 3' end) resulted in dramatic attenuation of GDVII neurovirulence and development of poliomyelitis in mice. This reduced neurovirulence was associated with slower virus growth and lower peak titers in the brain and spinal cord than with parental GDVII virus replication. On the other hand, the sites of replication following chimera 2 infection were the same as those seen in GDVII-infected mice; the distribution of virus antigen and histopathological changes indicated that chimera 2 replicates in neurons in the brain, e.g., in the neocortex, hippocampus, caudate putamen, and brain stem, as well as in anterior-horn cells in the spinal cord. Chimera 2 was efficiently cleared from the mouse central nervous system by day 30 postinfection, in marked contrast to the persistence of the BeAn parent in the central nervous system. This suggests that elements in the BeAn sequences that encode the polyprotein or are present in the 3' noncoding region are necessary for viral persistence. It is of interest that chimera 2-infected mice developed localized inflammatory, demyelinating lesions which were detected at day 28 postinfection but these lesions did not become larger with time. Thus, virus persistence appears to be required for maintenance and progression of immune-mediated demyelination. If the demyelinating lesions become sufficiently large, clinical signs and disease may develop.

Growth conditions established to optimize vegetative alkaline phosphatase production and stability in Bacillus subtilis were used to compare alkaline phosphatase synthesis and secretion in isogenic strains JH646 (spoOA12) and JH646MS (spoOA12 abrB15). A mutation in spoOA blocked vegetative alkaline phosphatase production, and a second mutation at the abrB locus resulted in hyperinduction of vegetative alkaline phosphatase. Phosphate regulation of vegetative alkaline phosphatase synthesis was unaffected in the double mutant. spoOH, on a multicopy plasmid, partially overcame the spoOA effect.