Alteration of lysine 178 in the hinge region of the Escherichia coli ada protein interferes with activation of ada, but not alkA, transcription

Abstract

The ada gene of Escherichia coli K-12 encodes the 39-kDa Ada protein, which consists of two domains joined by a hinge region that is sensitive to proteolytic cleavage in vitro. The amino-terminal domain has a DNA methyltransferase activity that repairs the S-diastereoisomer of methylphosphotriesters while the carboxyl-terminal domain has a DNA methyltransferase activity that repairs O6-methylguanine and O4-methylthymine lesions. Transfer of a methyl group to Cys-69 by repair of a methylphosphotriester lesion converts Ada into a transcriptional activator of the ada and alkA genes. Activation of ada, but not alkA, requires elements contained within the carboxyl-terminal domain of Ada. In addition, physiologically relevant concentrations of the unmethylated form of Ada specifically inhibit methylated Ada-promoted ada transcription both in vitro and in vivo and it has been suggested that this phenomenon plays a pivotal role in the down-regulation of the adaptive response. A set of site-directed mutations were generated within the hinge region, changing the lysine residue at position 178 to leucine, valine, glycine, tyrosine, arginine, cysteine, proline, and serine. All eight mutant proteins have deficiencies in their ability to activate ada transcription in the presence or absence of a methylating agent but are proficient in alkA activation. AdaK178P (lysine 178 changed to proline) is completely defective for the transcriptional activation function of ada while it is completely proficient for transcriptional activation of alkA. In addition, AdaK178P possesses both classes of DNA repair activities both in vitro and in vivo. Transcriptional activation of ada does not occur if both the amino- and carboxyl-terminal domains are produced separately within the same cell. The mutation at position 178 might interfere with activation of ada transcription by changing a critical contact with RNA polymerase, by causing a conformational change of Ada, or by interfering with the communication of conformational information between the amino- and the carboxyl-terminal domains. These results indicate that the hinge region of Ada is important for ada but not alkA transcription and further support the notion that the mechanism(s) by which Ada activates ada transcription differs from that by which it activates transcription at alkA.