HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Arroyo, L. G Articles by Weese, J S. Search for Related Content PubMed PubMed Citation Articles by Arroyo, L. G Articles by Weese, J S. Agricola Articles by Arroyo, L. G Articles by Weese, J S.

PCR ribotyping of Clostridium difficile isolates originating from human and animal sources

¹Department of Clinical Studies, Ontario Veterinary College, University of Guelph, Guelph, ON, Canada N1G 2W1 ²Department of Microbiology, Mount Sinai Hospital, Toronto, Ontario, Canada

Correspondence J. Scott Weese jsweese{at}uoguelph.ca

Received July 1, 2004
Accepted October 21, 2004

Molecular typing of Clostridium difficile isolates from animals and humans may be useful for evaluation of the possibility for interspecies transmission. The objective of this study was to evaluate C. difficile isolates from domestic animals and humans using PCR ribotyping. Isolates were also tested using PCR for the presence of genes encoding toxins A and B. One hundred and thirty-three isolates of C. difficile from dogs (n = 92), horses (n = 21) and humans (n = 20), plus one each from a cat and a calf, were evaluated. Overall, 23 ribotypes were identified. Of these, nine were identified from dogs, 12 from horses, seven from humans and one each from the cat and calf. In dogs, humans and horses, one or two different ribotypes predominated. Overall, 25 % of isolates from humans were indistinguishable from isolates from one or more animal species. Genes encoding C. difficile toxins A and B were detected in all human, equine and bovine isolates, and in 69 % of canine isolates. While different ribotypes appear to predominate in different mammalian species, several indistinguishable strains may be found in multiple species. This suggests that there is a potential for interspecies transmission of C. difficile and epidemiological studies are warranted.

	INTRODUCTION

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Clostridium difficile is an anaerobic, Gram-positive to Gram-variable, spore-forming bacillus that has been associated with a wide spectrum of disease in humans and several animal species including horses, dogs, ostriches, rabbits, cats and pigs (Frazier et al., 1993; Perkins et al., 1995; Songer et al., 2000; Jones, 2000; Barbut & Petit, 2001; Weese et al., 2001a; Marks et al., 2002). Reports of the prevalence of C. difficile in the faeces of dogs and cats range from 6 to 40 %. In horses, faecal carriage rates of 2–29 % have been reported (Riley et al., 1991; Perrin et al., 1993; Weese et al., 2001b; Cave et al., 2002).

Despite the isolation of C. difficile from a variety of domestic animals and the suggestion that household pets may be a potential reservoir of infection for humans (Borriello et al., 1983), there has been little investigation regarding the potential for interspecies transmission of this organism. O'Neill et al. (1993) found no correlation between the types recovered from pets and isolates of human origin using restriction enzyme analysis. However, they suggested that there might still be a risk of humans acquiring C. difficile from domestic pets.

Risk factors for the development of C. difficile-associated disease in humans include antimicrobial treatment, age (>65 years), anti-neoplastic chemotherapy, duration of hospitalization and underlying disease (Beaugerie et al., 2003). Considering that there are approximately 132.4 million dogs and cats in households and over 5 million domestic horses in the USA (http://www.AAHAnet.org) and the frequent, close contact between many humans and animals, the potential for transmission of C. difficile could be high.

PCR ribotyping is increasingly being employed for molecular epidemiological analysis of C. difficile. This technique is considered to be highly discriminative, reproducible and can be performed relatively easily and rapidly (O'Neill et al., 1996; Stubbs et al., 1999). It can be used for interlaboratory comparison and a reference library has been created (Brazier, 2001). The objective of this study was to investigate the ribotype patterns of C. difficile isolated from different animal species and human clinical cases from a regional human hospital. In addition, this study looked at a comparison of ribotype patterns present in different isolates from animals and from human clinical cases.

	METHODS

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Bacterial strains and culture conditions.

C. difficile isolates obtained from horses, dogs, a cat and a calf as part of previous studies and isolates collected prospectively from human faecal samples from individuals diagnosed with C. difficile-associated diarrhoea at a local hospital were evaluated. Isolates were inoculated on to cefoxitin–cycloserine– fructose agar (CCFA; Oxoid) and incubated for 48 h at 37 °C in an anaerobic chamber.

DNA extraction.

DNA was extracted using a Chelex resin-based DNA extraction commercial kit (InstaGene Matrix; Bio-Rad). Briefly, after obtaining a pure culture on CCFA, a single colony was subcultured on blood agar and incubated for 24–36 h at 37 °C in an anaerobic chamber. DNA was extracted from a suspension of three to five colonies in 200 µl 5 % Chelex 100 by heating at 56 °C for 30 min and vortexing for 10 s, followed by boiling at 100 °C for 8 min. Cell debris was removed from the suspension by centrifugation at 10 000 r.p.m. for 3 min. The supernatant that was obtained was then used as the DNA template for PCR ribotyping and toxin gene detection.

PCR ribotyping.

All isolates were typed using the PCR ribotyping method described by Bidet et al. (1999). Briefly, amplification reactions were performed in 25 µl containing 10 mM Tris/HCl (pH 8.8), 50 mM KCl, 1.5 mM MgCl₂, 200 mM each dNTP (Invitrogen), 50 pmol each primer, 2.5 U Taq polymerase (Invitrogen) and 5 µl DNA template (or sterile water as a negative control). Amplifications were carried out in a thermal cycler (Touchgene Gradient; Techne) for 1 cycle of 6 min at 94 °C for denaturation, followed by 35 cycles (1 min at 94 °C, 1 min at 57 °C and 1 min at 72 °C) and a final extension of 7 min at 72 °C. Amplification products were fractionated by electrophoresis in 1.5 % agarose gel (Invitrogen) for 6 h at 8 °C, 150 V, 0.5x TBE and analysed under UV light after staining with 0.5 µg ethidium bromide ml^–1. Gel images were acquired by CCD camera with the GeneSnap acquisition software. Reproducibility was studied by repeated PCR assays of the same DNA extract. Ribotype groups were arbitrarily designated by consecutive upper-case letters.

Determination of toxin genes.

All C. difficile isolates were screened for toxigenicity by PCR gene detection as previously described (Kato et al., 1998). Briefly, two primer sets were used to detect the toxin A gene: primers NK-3 and NK-2 derived from the non-repeating portion of the C. difficile toxin A gene, and primers NK-11 and NK-9 derived from the repeating portion of the C. difficile toxin A gene. A segment of the toxin B gene was amplified using primers NK-104 and NK-105 derived from the non-repeating sequence of the C. difficile toxin B gene.

The thermal profile for primer pairs NK-3/NK-2 and NK-104/NK-105 was 35 cycles comprising 95 °C for 20 s and 55 °C for 120 s, and amplification with primers NK-11/NK-9 was performed for 35 cycles consisting of 95 °C for 20 s and 62 °C for 120 s.

	RESULTS

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
A total of 133 samples from dogs (n = 92), horses (n = 21) and humans (n = 20), plus one each from a cat and a calf, was analysed. Twenty-three distinct ribotypes were identified (Table 1, Fig. 1). Between 1 and 12 ribotypes were identified per species; however, in dogs, humans and horses, one or two ribotypes predominated. The bovine strain possessed an indistinguishable ribotype pattern from the equine type W, and the feline strain was classified within the non-toxigenic canine type G.

View larger version (116K): [in this window] [in a new window]   Fig. 1. Representative PCR ribotype profiles from C. difficile isolated from human and domestic animals. Lanes: 1, sporadic canine ribotype D; 7 and 11, sporadic canine ribotype G; 2–4, ribotype A from human, horse and dog, respectively; 5, ribotype O isolated from horses and humans; 6 and 12, sporadic canine ribotype I; 8–10, ribotype B of canine and equine origin; 13, negative control. MW, 100 bp molecular mass marker.

Of the seven ribotypes identified in the human isolates, two were also found in veterinary isolates (canine and equine). Overall, 5/20 (25 %) isolates from humans were indistinguishable from isolates from one or more animal species.

All equine isolates were from horses with community-associated disease; therefore, the ribotype distribution suggested marked diversity of C. difficile in the horse population. In contrast to horses, one ribotype accounted for the majority of canine isolates. Thirty-one canine isolates were from dogs suspected of having acquired C. difficile during hospitalization in the Ontario Veterinary College Intensive Care Unit. Of these, 24 (77 %) were one ribotype. However, even if these cases were excluded, one ribotype still accounted for 34/61 (56 %) isolates. Considering that there was no known contact between any infected dogs and that the dogs were from a variety of locations in Ontario, it appears that this ribotype is disseminated in dogs across the province. There was no reported contact between any of the animals or humans in this study.

Genes encoding C. difficile toxins A and B were detected in all human, equine and bovine isolates and in 69 % of canine isolates. Of the 26 non-toxigenic canine isolates, 20 (77 %) were type B, while six (23 %) belonged to five different types. Toxin A^– toxin B⁺ strains were not identified.

	DISCUSSION

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
This study identified a large number of discernible ribotypes among C. difficile isolates from different mammalian species. Interestingly, there was some homology between animal and human isolates. While the main human PCR ribotype was not identified in other species, some human isolates were indistinguishable from those found in animals. It was interesting and of potential concern that the predominant canine PCR ribotype, a toxigenic strain, was indistinguishable from that found in 20 % of humans. While this does not confirm interspecies transmission of C. difficile, it supports the idea that this could occur. To our knowledge, this percentage of overlapping among types from animal and human sources has not been reported.

The presence of two predominant C. difficile PCR ribotypes among the canine isolates is similar to results reported by O'Neill et al. (1993). That study described six restriction enzyme analysis types from C. difficile strains isolated from dogs and cats and also found that two clones accounted for 84.6 % of all isolates. This is also consistent with studies of human isolates that have reported a diverse variety of types with a few predominating types (Kato et al., 2001; Fawley et al., 2003; Barbut et al., 2002; Svenungsson et al., 2003).

The fact that all of the non-canine isolates were toxigenic should be interpreted with care because those isolates were obtained from individuals with C. difficile-associated disease, while the canine isolates were collected as part of a surveillance study and were from diarrhoeic and non-diarrhoeic dogs (J. K. Clooten, S. A. Kruth & J. S. Weese, unpublished data). Previous studies have reported the prevalence of non-toxigenic C. difficile strains isolated from dogs and cats to be as high as 50 %, sometimes represented by a single type (O'Neill et al., 1993; Struble et al., 1994; Madewell et al., 1999). Toxin A^– toxin B⁺ strains have been reported in isolates from humans (van den Berg et al., 2004), but were not detected in this study. The prevalence of toxin A^– toxin B⁺ strains among human isolated strains has been reported to range from 3 to 12 % (Barbut et al., 2002). The role of this variant of C. difficile in animal disease has not been adequately explored.

This study has demonstrated that, while different PCR ribotypes appear to predominate in different mammalian species, certain indistinguishable types may be found in multiple species. Because there was no reported contact between the infected animals and humans in this study, the lack of identifiable variation among some isolates of animal and human origin suggests that there is the potential for interspecies transmission of C. difficile. Further epidemiological studies of the potential for transmission of C. difficile between animals and humans are warranted.

	ACKNOWLEDGEMENTS

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
The authors would like to acknowledge the technical assistance of J. Rousseau in this project. Reference strains were kindly provided by Dr J. S. Brazier. All work on this project was performed at the Ontario Veterinary College. This project was supported by the Ontario Veterinary College Pet Trust.

This information was presented in part as an oral abstract at the 22nd Annual American College of Veterinary Internal Medicine Conference, 9–12 June, 2004, Minneapolis, MN, USA, and as a poster (P-7) at the First International Clostridium difficile Symposium, 5–7 May, 2004, in Kranjska Gora, Slovenia.

	REFERENCES

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES