Identification of hypoxanthine and guanine as the co-repressors for the purine regulon genes of Escherichia coli

Addition of purine compounds to the growth medium of Escherichia coli and Salmonella typhimurium causes repressed synthesis of the purine biosynthetic enzymes. The repression is mediated through a regulatory protein, PurR. To identify the co-repressor(s) of PurR, two approaches were used: (i) mutations were introduced into purine salvage genes and the effects of different purines on pur gene expression were determined; (ii) purine compounds which dictate the binding of the PurR protein to its operator DNA were resolved by gel retardation. Both the in vivo and the in vitro data indicated that guanine and hypoxanthine are co-repressors. The toxic purine analogues 6-mercaptopurine and 6-thioguanine also activated the binding of PurR to its operator DNA.     

Participation of the purine repressor in control of the carbamoylphosphate synthetase operon in Salmonella typhimurium

Previous work has shown that the carAB operon of Salmonella typhimurium is transcribed from tandem promoters, P1 and P2, that are negatively controlled by pyrimidines and arginine, respectively. The results reported here show that purines also negatively control expression of carAB and that this effect is absent in a purR ::Tn 10 derivative. Primer-extension experiments established that the purine effect is exerted at P1, thus redefining this promoter as sensitive to both purines and pyrimidines. The results of gel-retardation experiments as well as DNase I and premethylation footprintings indicate that the purine repressor interacts with a PUR box 85 bp upstream of P1. Modification of this PUR box by site-directed mutagenesis abolishes the repression by purines in a carA :: lacZ fusion, confirming that this box functions in vivo in purine control of carAB expression.     

In vivo effect of mutations at the PRPP binding site of the Bacillus subtilis purine repressor

The Bacillus subtilis PurR mediates adenine repression and guanosine induction of purA. PRPP inhibits binding of PurR to DNA in vitro. Mutations in the PRPP binding motif of PurR caused strong repression regardless of purine exclusions or additions, establishing the role of PRPP as regulator of PurR.     

Structural characterization and corepressor binding of the Escherichia coli purine repressor.

The Escherichia coli purine repressor, PurR, binds to a 16-bp operator sequence and coregulates the genes for de novo synthesis of purine and pyrimidine nucleotides, formation of a one-carbon unit for biosynthesis, and deamination of cytosine. We have characterized the purified repressor. Chemical cross-linking indicates that PurR is dimeric. Each subunit has an N-terminal domain of 52 amino acids for DNA binding and a C-terminal 289-residue domain for corepressor binding. Each domain was isolated after cleavage by trypsin. Sites for dimer formation are present within the corepressor binding domain. The corepressors hypoxanthine and guanine bind cooperatively to distinct sites in each subunit. Competition experiments indicate that binding of one purine abolishes cooperativity and decreases the affinity and the binding of the second corepressor. Binding of each corepressor results in a conformation change in the corepressor binding domain that was detected by intrinsic fluorescence of three tryptophan residues. These experiments characterize PurR as a complex allosteric regulatory protein.     

Purine regulon of gamma-proteobacteria: a detailed description

The structure of the purine regulon was studied by a comparative genomic approach in seven genomes of gamma-proteobacteria: Escherichia coli, Salmonella typhi, Yersinia pestis, Haemophilus influenzae, Pasteurella multocida, Actinobacillus actinomycetemcomitans, and Vibrio cholerae. The palindromic binding site of the purine repressor (consensus ACGCAAACGTTTGCGT) is fairly well retained of genes encoding enzymes that participate in the synthesis of inosinemonophosphate from phosphoribozylpyrophosphate and in transfer of unicarbon groups, and also upstream of some transport protein genes. These genes may be regarded as the main part of the purine regulon. In terms of physiology, the regulation of the purC and gcvTHP/folD genes seems to be especially important, because the PurR site was found upstream of nonorthologous but functionally replaceable genes. However, the PurR site is poorly retained in front of orthologs of some genes belonging to the E. coli purine regulon, such as genes involved in general nitrogen metabolism, biosynthesis of pyrimidines, and synthesis of AMP and GMP from IMP, and also upstream of the purine repressor gene. It is predicted that purine regulons of the examined bacteria include the following genes: upp participating in synthesis of pyrimidines; uraA encoding an uracil transporter gene; serA involved in serine biosynthesis; folD responsible for the conversion of N5,N10-methenyl tetrahydrofolate into N10-formyltetrahydrofolate; rpiA involved in ribose metabolism; and protein genes with an unknown function (yhhQ and ydiK). The PurR site was shown to have different structure in different genomes. Thus, the tendency for a decline of the conservatism of site positions 2 and 15 was observed in genomes of bacteria belonging to the Pasteurellaceae and Vibrionaceae groups.     

Nucleosomal location of the STE6 TATA box and Mat alpha 2p-mediated repression.

It has been proposed that yeast MATa cell-specific genes are repressed in MAT alpha cells by the Mat alpha 2p repressor-directed placement of a nucleosome in a position that incorporates the TATA box of the MATa-specific gene close to the nucleosomal pseudodyad. In this study, we address this proposal directly with a series of plasmids designed to place the MATa-specific STE6 TATA box at different locations in a nucleosome and in the internucleosomal linker. These plasmids contain different lengths of synthetic random DNA between the Mat alpha 2p operator and the TATA box of the STE6 promoter, which is located upstream of a lacZ reporter gene in a multicopy plasmid. We show that in MAT alpha cells, a nucleosome is retained in an identical translational frame relative to the Mat alpha 2p operator in all the constructs investigated, irrespective of the sequence of the DNA wrapped onto the histone octamer. This result shows that the nucleosomal organization of the STE6 promoter in MAT alpha cells is not conferred by the sequence of the promoter itself. No expression of the lacZ reporter gene was detectable in MAT alpha cells in any of the constructs, even with the TATA box located in a short internucleosomal linker. These data indicate that repression of MATa-specific genes in MAT alpha cells does not require the precise translational placement of the TATA box close to the nucleosomal pseudodyad; the gene remains repressed when the TATA box is located within the investigated 250-bp region in the organized chromatin domain abutting the Mat alpha 2p operator in MAT alpha cells and may remain repressed with the TATA box located anywhere within this organized repression domain.