Molecular genetic analysis of the Escherichia coli phn(psiD) gene cluster and its role in phosphonate metabolism

William Walter Metcalf, Purdue University

Abstract

The gram-negative bacterium Escherichia coli can use phosphonates (Pn), a class of compounds containing extremely stable carbon-phosphorus (C-P) bonds, as a sole phosphorus (P) source. Pn are metabolized in E. coli by the enzyme C-P lyase, which is poorly understood due to the inability to detect its activity in vitro. This difficulty has been overcome using a molecular genetic approach in the study of Pn metabolism. Earlier studies showed that Pn degradation required the phn gene cluster which, based on its DNA sequence, was proposed to contain seventeen genes, phnA through phnQ. However, the functions of individual genes in Pn utilization were undefined. In this study a detailed analysis of the phn gene cluster is presented. Phenotypic analysis of phn mutants revealed three distinct functions for the phn locus. These are: (i) C-P lyase activity, (ii) phosphite oxidation, a previously unsuspected but related activity, and (iii) the nonspecific transport of phosphorus compounds including Pn, phosphite, phosphate and phosphate esters. Using molecular genetic techniques the roles for each of the seventeen phn genes were defined. First, three genes (phnA, phnB and phnQ) were shown to have no role in Pn metabolism because plasmid subclones lacking these genes can still complement phn deletion mutants. The remaining fourteen genes were assigned a role(s) based on mutant analysis. A total of 53 independent transposon-induced mutants were characterized; these include 3 Mu d1, 7 Tn5 and 43 Tn$phoA\sp\prime$ prime insertions. The mutations were localized to an individual phn gene by DNA sequencing of the transposon insertion site. All fourteen genes were mutated at least once. Both polar and non-polar insertion mutants were studied at 20 sites. Each Tn$phoA\sp\prime$ prime insertion was characterized with respect to transcription. Eight insertions were also characterized with respect to translation and protein localization after conversion of the insertion to a lacZ or phoA protein fusion. The results showed: (i) three genes (phnC, phnD, and phnE) comprise a binding-protein dependent Pn transport system; (ii) eight genes (phnG, phnH, phnI, phnJ, phnK, phnL, phnM and phnP) are required for the utilization of both Pn and phosphite, suggesting that their products form a multi-subunit enzyme complex needed for both activities; (iii) one gene (phnN) is not absolutely required for catalysis, although it seems to modulate enzyme activity; and (iv) two genes (phnF and phnO) have no biochemical functions and, based on protein homologies, are probably involved in gene regulation. Lastly, the regulation of the phn gene cluster was examined. The phn gene cluster was shown to comprise an operon of 10.9 kb, which is the largest yet reported in E. coli. An internal site for down-regulation of phn operon transcription was also revealed, which lies between the phnE and phnF genes.

Degree

Ph.D.

Advisors

Wanner, Purdue University.

Subject Area

Microbiology|Genetics|Molecular biology

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS