Escherichia albertii in wild and domestic birds

Abstract

Escherichia albertii has been associated with diarrhea in humans but not with disease or infection in animals. However, in December 2004, E. albertii was found, by biochemical and genetic methods, to be the probable cause of death for redpoll finches (Carduelis flammea) in Alaska. Subsequent investigation found this organism in dead and subclinically infected birds of other species from North America and Australia. Isolates from dead finches in Scotland, previously identified as Escherichia coli O86:K61, also were shown to be E. albertii. Similar to the isolates from humans, E. albertii isolates from birds possessed intimin (eae) and cytolethal distending toxin (cdtB) genes but lacked Shiga toxin (stx) genes. Genetic analysis of eae and cdtB sequences, multilocus sequence typing, and pulsed-field gel electrophoresis patterns showed that the E. albertii strains from birds are heterogeneous but similar to isolates that cause disease in humans.

Figure 1

Neighbor-joining dendrogram of bird and human isolates, based on nucleotide variation at 6 conserved housekeeping loci in the Escherichia albertii genome (multilocus sequence typing), supporting identification of all isolates as E. albertii. Three distinct clades are designated EA 1–EA 3. No clustering of isolates is apparent on the basis of their host type, geographic origin, or association with disease. Inset is a SplitsTree phylogenetic network of the genus Escherichia, showing the E. albertii/Shigella boydii serotype 7/13 lineage with respect to the other named species. S. bongori, Salmonella bongori; S. enterica, Salmonella enterica. Scale bars indicate genetic distance.

Figure 2

Neighbor-joining dendrogram based on the predicted amino acid sequences of the intimin (eae) and cytolethal distending toxin (cdtB) loci of Escherichia spp. Isolates analyzed in this study are designated by their isolate identification number; reference alleles are designated by their GenBank accession numbers. All Escherichia albertii isolates are indicated by circles or squares; all other isolates are E. coli. Scale bars indicate genetic distance. A) eae alleles carried by E. albertii isolates from birds (filled circles) represent diverse allelic subtypes and do not form separate clusters from allelic subtypes carried by isolates from humans (open circles). The eae alleles from North America and 1 from Australia (B1086) are novel but most similar to γ intimins. The eae alleles from Scotland and 1 from Australia (B101) are also novel but most similar to μ intimins. B) The E. albertii cdtB alleles from bird isolates, amplified by the s1/as1 primers, are most similar to cdtB types II, III, and V. E. albertii alleles from human isolates are marked with open circles. The bird E. albertii cdtB alleles amplified by the s2/as2 primers are distantly related to the other bird and human E. albertii alleles and are most similar to cdtB type I in E. coli. Because all the avian alleles amplified by the s2/as2 primers in each group are identical, only 1 sequence for each is shown.

Figure 3

Dendrogram (unweighted pair group method with arithmetic mean) of Escherichia albertii isolates from birds, based on pulsed-field gel electrophoresis band profiles. Isolates from disease outbreaks in Alaska and Scotland form clonal groups, indicating that these outbreaks are associated with expansion of a single clone. The profiles for the other isolates indicate that the bird and human E. albertii isolates are heterogeneous and do not segregate on the basis of host, geographic origin, or disease status. Scale bar indicates percent similarity.