Downloaded from www.microbiologyresearch.org by IP: 54.90.45.205 On: Wed, 27 Jan 2016 15:16:37 Microbiology (1 997), 143, 2039-2046 Printed in Great Britain Division of Botany and Zoology, Australian National University, Canberra, ACT 0200, Australia The genetic structure of Escherichia coli populations in feral house mice David M. Gordon Tel: +61 6 249 3552. Fax: +6l 6 249 5573. e-mail: David.Gordon(u anu.edu.au Escherichia coli was isolated from feral house mice (Mus domesticus) during the course of a mouse plague in the state of Victoria, Australia. Two farms were sampled over a period of 7 months and a total of 447 isolates were collected. The isolates were characterized using the techniques of randomly amplified polymorphic DNA and multi-locus enzyme electrophoresis. The mean genetic diversity of this E. coli population (H = 024) was found to be substantially lower than the diversity of an E. coli population reported elsewhere for a single human host. Analysis of the allozyme data revealed that there were significant differences in the relative abundance of genotypes between the two localities sampled and among sample dates. Overall, however, spatial and temporal effects accounted for less than 5% of the genotypic diversity. Allele frequencies and the relative abundance of the more common genotypes did not differ between male and female hosts. The number of genotypes and genotype diversity increased as the age of the host increased, suggesting that the mice are continuing to acquire new E. coli clones throughout their life. The frequency of some alleles changed with respect to host age, which indicates that clone acquisition may not be a random process. It is argued that the low level of genetic diversity observed in this population of E. coli reflects the boom and bust nature of mouse population density in this region of Australia. Keywords : Escherichia coli, allozymes, genotypic diversity, population genetics, mouse r- ~ INTRODUCTION Studies concerning Escherichia coli have formed the empirical basis of our understanding of the genetic structure of bacterial populations (Milkman, 1973 ; Selander & Levin, 1980; Whittam et al., 1983a, b; Miller & Hartl, 1986; Maynard Smith et al., 1993 ; Guttman & Dykuizen, 1994). Many of these studies have focused on clinical isolates responsible for a variety of diseases (Caugant et al., 1983; Selander et al., 1986b; Whittam et al., 1993). Some have examined issues related to clonal diversity and turnover in a single host (Caugant et al., 1981) or the degree to which strains are shared between hosts (Caugant et al., 1984).Virtually all of these studies have been restricted to E. coli isolated from humans. Notable exceptions to the observed bias towards human isolates do exist but they are few in number (Routman et af., 1985; Whittam, 1989). An additional major bias present in many previous studies has been their basis on Abbreviations: ETs, electrophoretic types; MLEE, multi-locus enzyme electrophoresis; RAPD, randomly amplified polymorphic DNA. ad hoc collections of isolates of diverse geographic origin. Few studies have examined the structure of E. coli populations from a single host species or geographic locality. The purpose of this study was to examine the genetic structure of E. coli from feral house mouse (Mus domesticus) populations and to investigate the role that host, spatial and temporal factors play in determining the structure of these populations. Collections were made in a region of Australia where house mouse population numbers regularly attain plague status, an event which occurred during this study. Over 440 strains of E. coli were isolated from two localities over a 7 month period and characterized using the techniques of randomly amplified polymorphic DNA (RAPD) and multi-locus enzyme electrophoresis (MLEE) . METHODS Study site and host collection. Mice were collected from two wheat-growing properties (Stone and Symes) 15 km apart near the town of Walpeup in the Mallee district of north western 0002-1271 0 1997SGM 2039