A novel pore-forming toxin in type A Clostridium perfringens is associated with both fatal canine hemorrhagic gastroenteritis and fatal foal necrotizing enterocolitis.

A role for type A Clostridium perfringens in acute hemorrhagic and necrotizing gastroenteritis in dogs and in necrotizing enterocolitis of neonatal foals has long been suspected but incompletely characterized. The supernatants of an isolate made from a dog and from a foal that died from these diseases were both found to be highly cytotoxic for an equine ovarian (EO) cell line. Partial genome sequencing of the canine isolate revealed three novel putative toxin genes encoding proteins related to the pore-forming Leukocidin/Hemolysin Superfamily; these were designated netE, netF, and netG. netE and netF were located on one large conjugative plasmid, and netG was located with a cpe enterotoxin gene on a second large conjugative plasmid. Mutation and complementation showed that only netF was associated with the cytotoxicity. Although netE and netG were not associated with cytotoxicity, immunoblotting with specific antisera showed these proteins to be expressed in vitro. There was a highly significant association between the presence of netF with type A strains isolated from cases of canine acute hemorrhagic gastroenteritis and foal necrotizing enterocolitis. netE and netF were found in all cytotoxic isolates, as was cpe, but netG was less consistently present. Pulsed-field gel electrophoresis showed that netF-positive isolates belonged to a clonal population; some canine and equine netF-positive isolates were genetically indistinguishable. Equine antisera to recombinant Net proteins showed that only antiserum to rNetF had high supernatant cytotoxin neutralizing activity. The identifica-tion of this novel necrotizing toxin is an important advance in understanding the virulence of type A C. perfringens in specific enteric disease of animals.

The epidemiology of Clostridium perfringens type A on Ontario swine farms, with special reference to cpb2-positive isolates.

BACKGROUND: There is poor understanding of most aspects of Clostridium perfringens type A as a possible cause of neonatal diarrhea in piglets, and the prevalence and types of C. perfringens present on Ontario swine farms is unknown. To study the prevalence of fecal C. perfringens and selected toxin genes, 48 Ontario swine farms were visited between August 2010 and May 2011, and 354 fecal samples were collected from suckling pigs, lactating sows, weanling pigs, grower-finisher pigs, and gestating sows, as well as from manure pits. The fecal samples were cultured quantitatively, and toxin genes were detected by real-time multiplex polymerase chain reaction (PCR). RESULTS: In mixed multivariable linear analysis, log(10) C. perfringens in fecal samples from suckling pigs were higher than that of weanling pigs, grower-finisher pigs, and manure pit samples (P <0.05). In mixed multivariable logistic analysis, the C. perfringens isolates recovered from lactating sows (OR = 0.069, P <0.001), gestating sows (OR = 0.020, P <0.001), grower-finishers (OR = 0.017, P <0.001), and manure pits (OR = 0.11, P <0.001) were less likely to be positive for the consensus beta2 toxin gene cpb2 compared to the isolates from suckling pigs. The prevalence of cpb2 in the isolates recovered from weanlings did not differ significantly from suckling pigs. C. perfringens isolates that were positive for cpb2 were more likely to carry the atypical cpb2 gene (atyp-cpb2) (OR = 19, P <0.001) compared to isolates that were negative for cpb2. Multivariable analysis did not identify farm factors affecting the presence of consensus cpb2 and atyp-cpb2 genes. CONCLUSIONS: This study provides baseline data on the prevalence of C. perfringens and associated toxin genes in healthy pigs at different stages of production on Ontario swine farms. The study suggests that if C. perfringens type A are involved in neonatal enteritis, there may be strains with specific characteristics that cannot be identified by the existing genotyping system.

Longitudinal study on the seroprevalence of avian influenza, leptospirosis, and tularemia in an urban population of raccoons (Procyon lotor) in Ontario, Canada.

Raccoons (Procyon lotor) live at high densities, often in close association with people, in urban areas in Ontario and have been implicated as potential reservoirs of numerous zoonotic disease agents. We collected 137 blood samples from 61 apparently healthy raccoons in a small area of Toronto, Ontario, from June to October 2007 as part of a longitudinal study to determine the seasonal patterns of seroprevalence of Francisella tularensis, avian influenza, and Leptospira. In addition, we collected 35 urine samples by cystocentesis from 23 animals to look for evidence of urinary shedding of Leptospira. All samples were serologically negative for F. tularensis and avian influenza. Nineteen of 61 animals (31%) were positive for Leptospira antibodies in one or more trapping periods. The seroprevalence of Leptospira increased from 5% in June to 38% in October. Of the 19 positive animals, 14 were seropositive for serogroup Grippotyphosa, 4 for serogroup Pomona, and 1 for both serogroups Australis and Grippotyphosa. Raccoons were seronegative to serovars representative of serogroups Autumnalis, Canicola, Icterohaemorrhagiae, and Sejroe. Only one urine sample was culture positive for Leptospira (2.9%). Although we found evidence that raccoons in this study were exposed to leptospires belonging to serogroup Grippotyphosa, likely serovar Grippotyphosa, during the summer and able to shed leptospires in urine, further work is required to determine the importance of raccoons as reservoirs of Leptospira in Ontario.

A mycolyl transferase mutant of Rhodococcus equi lacking capsule integrity is fully virulent.

Rhodococcus equi is a mucoid Gram-positive facultative intracellular pathogen which can cause severe bronchopneumonia in foals and AIDS patients. A polysaccharide capsule which gives R. equi a mucoid appearance has long been suspected to be a virulence factor. Here, we describe a transposome mutant in the gene fbpA of strain R. equi 103 causing absence of a capsular structure. FbpA is a chromosomal gene homologous to antigen 85 (Ag85) mycolyl chain transferase gene of Mycobacterium tuberculosis. The mutant multiplied normally in isolated macrophages, was able to establish the unusual R. equi-containing vacuole in macrophages, was cytotoxic for macrophages, and was virulent in a mouse model. Colonies had a dry appearance on nutrient agar and defective capsule structure. Surprisingly, fbpA mutants cured of the virulence-associated plasmid were found in a phagosome that was more alkaline than that of the corresponding wild-type bacteria, were more cytotoxic and even multiplied to some extent. This study suggests that the capsule is not an important virulence factor of R. equi and that it may even counteract virulence traits.

Mutation and virulence assessment of chromosomal genes of Rhodococcus equi 103.

Rhodococcus equi can cause severe or fatal pneumonia in foals as well as in immunocompromised animals and humans. Its ability to persist in macrophages is fundamental to how it causes disease, but the basis of this is poorly understood. To examine further the general application of a recently developed system of targeted gene mutation and to assess the importance of different genes in resistance to innate immune defenses, we disrupted the genes encoding high-temperature requirement A (htrA), nitrate reductase (narG), peptidase D (pepD), phosphoribosylaminoimidazole-succinocarboxamide synthase (purC), and superoxide dismutase (sodC) in strain 103 of R. equi using a double-crossover homologous recombination approach. Virulence testing by clearance after intravenous injection in mice showed that the htrA and narG mutants were fully attenuated, the purC and sodC mutants were unchanged, and the pepD mutant was slightly attenuated. Complementation with the pREM shuttle plasmid restored the virulence of the htrA and pepD mutants but not that of the narG mutant. A single-crossover mutation approach was simpler and faster than the double-crossover homologous recombination technique and was used to obtain mutations in 6 other genes potentially involved in virulence (clpB, fadD8, fbpB, glnA1, regX3, and sigF). These mutants were not attenuated in the mouse clearance assay. We were not able to obtain mutants for genesfurA, galE, and sigE using the single-crossover mutation approach. In summary, the targeted-mutation system had general applicability but was not always completely successful, perhaps because some genes are essential under the growth conditions used or because the success of mutation depends on the target genes.