Genetic characterization and virulence of enterotoxigenic Escherichia coli mutants which have lost virulence genes in vivo

Abstract

Loss of K99 and STaP genes from enterotoxigenic Escherichia coli 431 during infection occurred by either plasmid curing or plasmid deletion. These mutants expressed the F41 adhesin and colonized neonatal pigs, but only those mutants that retained STaP caused diarrhea with significant weight loss.In vitro and in vivo loss of enterotoxigenic Escherichia coli virulence genes has been reported to occur by plasmid curing (5,6,8,9,11,(16)(17)(18). Pigs infected with enterotoxigenic E. coli 431 (O101:K30:NM, K99+ F41+ STaP+) shed mutants which had lost both K99 and STaP or only K99 (11). Mutants which had lost STaP and retained K99 were not found. We examined these mutants and found that a single plasmid was lost in K99-STaPmutants and that in K99-STaP+ mutants the K99 genes were deleted from the plasmid. We found that both classes of mutants expressed F41 and colonized experimentally inoculated neonatal pigs. One of the K99-F41+ STaP+ mutants caused severe diarrhea comparable with that caused by wild-type strain 431.Agarose gel electrophoresis of plasmid DNA isolated from strain 431 showed two large plasmid bands of approximately 115 and 85 kb and two smaller plasmid bands (Fig. 1A, lane 2). A K99-STaPmutant (strain 518M) had a plasmid profile indistinguishable from the wild-type parent (Fig. 1A, lane 1). The 115-kb plasmid band from strain 431 hybridized with DNA probes for both K99 and STaP ( Fig. 1B and C, lane 2). However, neither probe hybridized with the 115-kb plasmid band in the mutant ( Fig. 1B and C, lane 1). Similar plasmid patterns for additional K99-STaPmutants are shown in Fig. 2. Comparison of BglII restriction endonuclease fragment patterns of plasmids from strain 431 and a K99-STaPmutant (Fig. 3, lanes 2 and 3) showed that several fragments were missing from the K99-STaPmutants, suggesting the loss of a single plasmid coding for both of the traits.The location of the STaP gene was determined by comparison of plasmid DNA isolated from four K99-STaP+ mutants with plasmids isolated from strain 431 and from five K99-STaPmutants (Fig. 2). The K99-STaP+ mutants had an additional, 105-kb plasmid band which hybridized with the STaP probe (Fig. 2B). The same size BglII fragment of plasmid DNA from strain 431 and three different K99-STaP+ mutants hybridized with the STaP probe ( Fig. 3B, lanes 2, 4, and 5). These results suggest that the K99-STaP+ mutants occurred by deletion of K99 genes from one of the 115-kb plasmids, leaving a 105-kb plasmid carrying STaP.The additional plasmid bands in each of the K99-STaP+ mutants shown in Fig. 2 were indistinguishable in size. Although the K99-STaP+ mutants were from different animals, they may not be independent isolates because the animals were all from the same litter and were inoculated together in the same experiment. We identified three inde-* Corresponding author. pendent K99-STaP+ mutants isolated from different animals inoculated in separate experiments to determine if a specific deletion of the K99 genes had occurred during infection. Figure 3 shows BglII fragme...

Mapping the neutralizing epitopes of F18 fimbrial adhesin subunit FedF of enterotoxigenic Escherichia coli (ETEC)

“…T ETEC fimbriae and enterotoxins are immunologically heterogeneous (Gaastra and de Graaf, 1982). Fimbriae of ETEC causing PWD include K88 (F4) and F18, and occasionally K99 (F5), 987 P (F6) and F41 (F7) (Awad-Masalmeh et al, 1982;Casey and Moon, 1990;Frydendahl, 2002;Moseley et al, 1986;Nagy et al, 1977;Zhang et al, 2007). Enterotoxins produced by ETEC are heat-labile toxin (LT), heat-stable toxin type I (STa), heat-stable toxin type II (STb), Shiga toxin 2e (Stx2e) and enteroaggregative heat-stable toxin type 1 (EAST1) (Frydendahl, 2002;Lee et al, 1983;Moon et al, 1980;Nakazawa et al, 1987;Osek, 1999b;Zhang et al, 2007).…”

Passive protection of suckling infant mice against F41-positive enterotoxigenic Escherichia coli strains by intravenous inoculation of the dams with monoclonal antibodies against F41

“…There is, however, strong suggestive evidence that F41 can mediate colonization by adhesion. Variants of a K99-, F41-positive porcine ETEC strain that have lost the K99 gene (19) and still carry the gene for and produce F41 are still virulent for newborn pigs (6).…”

Replicon typing of virulence plasmids of enterotoxigenic Escherichia coli isolates from cattle

“…The presence of two plasmids of identical size, one being STaP+ and the other being F5+, cannot be excluded absolutely (10), but association of the STaP enterotoxin and F5 adhesin genes on the same plasmid molecule seems to be widespread in STa+ F5+ ETEC isolates from cattle and from pigs (5,13,30; for a review, see reference 32). Associations of genes coding for other adhesins (F2, F3, and F6) with genes coding for enterotoxins (STa, STh, and LT1) are also common in human ETEC and other porcine ETEC (9,11,22,27; for a review, see reference 32).…”