|
|
||||||||
1,2Department of Poultry1 and Department of Bacteriology2, National Veterinary Institute, SE-75189 Uppsala, Sweden 3,5Department of Veterinary Microbiology3 and Department of Large Animal Clinical Sciences5, Swedish University of Agricultural Sciences, Uppsala, Sweden 4Department of Food Technology, Lund University, Lund, Sweden 6Department of Disease Control and Biosecurity, Zoonosis Centre, National Veterinary Institute, Uppsala, Sweden 7Department of Biotechnology, The Royal Institute of Technology, Stockholm, Sweden
Correspondence Désirée S. Jansson desiree.s.jansson{at}sva.se
Received September 30, 2003
Accepted January 12, 2004
The aims of the current study were to collect intestinal spirochaetes (genus Brachyspira) from farmed and wild mallards (Anas platyrhynchos) and to identify and classify those isolates that phenotypically resembled Brachyspira hyodysenteriae, an enteric pathogen of pigs. The isolation rate of Brachyspira spp. was high from both farmed (93 %) and wild mallards (78 %). In wild mallards, it appeared that Brachyspira spp. were more likely to be found in migratory birds (multivariate analysis: RR = 1.8, 95 % CI 1.13.1) than in mallards sampled in a public park. Pure cultures of putative B. hyodysenteriae were obtained from 22 birds. All five isolates from farmed mallards and ten randomly selected isolates with this phenotype were used for further studies. All isolates from farmed mallards and two of the isolates from wild mallards were PCR-positive for the tlyA gene of B. hyodysenteriae. Two isolates from farmed mallards were selected for pulsed field gel electrophoresis (PFGE) and randomly amplified polymorphic DNA (RAPD) analysis. These isolates clustered with the type and reference strains of B. hyodysenteriae. 16S rDNA sequence analysis performed on 11 of the strains showed that they were all closely related to each other and to the B. hyodysenteriaeBrachyspira intermedia cluster. Three of the mallard isolates had 16S rDNA sequences that were identical to those of B. hyodysenteriae strains R1 and NIV-1 previously isolated from common rheas (Rhea americana). To conclude, the isolates from farmed mallards and two isolates from wild mallards were classified as B. hyodysenteriae based on the fact that they could not be differentiated by any of the applied methods from type, reference and field strains of B. hyodysenteriae. The remaining isolates could not be assigned irrefutably to any of the presently recognized Brachyspira species. These results point to a broader host spectrum of B. hyodysenteriae than is generally recognized, and to the presence in mallards of strongly ß-haemolytic and indole-producing spirochaetes that possess many, but not all, of the currently recognized characteristics of B. hyodysenteriae.
Abbreviation: RAPD, randomly amplified polymorphic DNA.
The GenBank accession numbers for the new 16S rDNA sequences reported in this study are AY352281AY352291.
| INTRODUCTION |
|---|
|
|
|---|
Brachyspira hyodysenteriae is recognized as the aetiological agent of swine dysentery, a mucohaemorrhagic diarrhoeic disease of pigs (Ochiai et al., 1997; Harris et al., 1999). This agent has also been suggested as the cause of necrotizing typhlocolitis in common rheas (Rhea americana) (Sagartz et al., 1992; Buckles et al., 1997). Furthermore, B. hyodysenteriae may be present in wild rodents on farms with infected pigs (Duhamel, 2001). The natural occurrence of B. hyodysenteriae in other mammalian and avian species has not been verified. Interestingly, neither B. hyodysenteriae nor any other intestinal spirochaete has been isolated from wild boars in Sweden (Fellström & Jacobsson, 2002).
B. hyodysenteriae is routinely diagnosed by culture, phenotypic traits, which consist of strong ß-haemolysis, indole production, the inability to cleave hippurate and
-galactosidase activity (Fellström et al., 1999), and by specific PCR assays (Leser et al., 1997; Ateyo et al., 1999; Fellström et al., 2001). Several atypical B. hyodysenteriae isolates of porcine origin have recently been described. These include isolates that do not produce indole (Fellström et al., 1999) and those that are negative for a B. hyodysenteriae-specific 23S rDNA-based PCR test (Murray et al., 2003). PFGE, randomly amplified polymorphic DNA (RAPD) and phylogenetic analysis based on 16S rRNA sequence analysis showed that these atypical variants are genetically closely related to typical porcine B. hyodysenteriae (Fellström et al., 1999; Murray et al., 2003).
In a previous study, we cultured a putative B. hyodysenteriae isolate from a mallard (Anas platyrhynchos) on a gamebird farm (Jansson et al., 2001). The present study was designed to confirm this finding, to investigate whether the same spirochaete phenotype was present in wild-living mallards and to identify these isolates. Finally, we wanted to examine whether age, season, sex or migratory status was associated with the presence of Brachyspira spp. in wild mallards. We chose the mallard as our target species because it is the most numerous and widespread of all ducks. The mallard is an omnivorous dabbling duck that lives on wetlands, sometimes in close association to human populations. In addition, it is a common game species and is also the ancestor of domestic ducks.
| METHODS |
|---|
|
|
|---|
|
PCR analysis.
PCR for detection of the tlyA haemolysin gene of B. hyodysenteriae was used according to previously described protocols (Fellström et al., 2001), with some modifications. Each sample was tested twice. In total, ten type, reference and field strains of porcine origin, one chicken isolate, one isolate from a dog and 15 mallard isolates were analysed by PCR.
PFGE.
PFGE was performed as described previously (Fellström et al., 1999). A dendrogram was created with the GelCompar program based on combined gels of MluI- and SalI-digested DNA, using a matrix of band-matching coefficients by the unweighted pair group method with arithmetic mean (UPGMA) clustering fusion strategy. Ten type, reference and field strains of porcine origin, one chicken isolate, one isolate from a dog and two mallard isolates with a biochemical phenotype consistent with group I spirochaetes were studied by PFGE.
RAPD.
Two mallard isolates with a biochemical phenotype consistent with group I spirochaetes, one chicken isolate and ten type, reference and field strains of porcine origin were studied by RAPD, as described previously (Quednau et al., 1998), with a few modifications. The cells were lysed by boiling and the isolates were typed with primers P1254 (5'-CCGCAGCCAA-3') and P73 (5'-ACGCGCCCT-3'), separately, which resulted in two sets of banding patterns. Those patterns were combined in the GelCompar program with the combined gels option.
Sequence analysis of the 16S rRNA gene.
16S rDNA sequences of 11 of the selected Brachyspira isolates from farmed and wild-living mallards were determined as described by Pettersson et al. (1996), with some modifications. The 16S rRNA gene was first amplified with the PCR primers listed in Table 2. The sequencing system (Pettersson et al., 1996) was adapted to the ABI Prism 3100 Genetic Analyzer (Applied Biosystems) with the primers listed in Table 2. Labelled terminators (Big Dye; Applied Biosystems) were used in the cycle sequencing reactions. The primer sequences are given in Table 2. Contigs were created using the ContigExpress program included in the Vector NTI suite (InforMax). The 16S rDNA sequences were aligned manually using Genetic Data Environment software (Smith, 1992), to a prealigned set of Brachyspira spp. sequences retrieved from the Ribosomal Database Project (Maidak et al., 2001).
|
Statistical analyses.
Relative risks between isolation rate of intestinal spirochaetes and sampling sites, seasons, age groups and sexes were calculated, and the confidence intervals were precision based (Kleinbaum et al., 1982). To examine which of the potential risk factors (age, sex, migratory status and season) were most important for finding Brachyspira spp., a log linear model specified as a multivariate generalized linear model, with a Poisson distribution of the error term and log link function, was used. For inclusion in the final model, a P value < 0.01 was required.
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
|
Isolates with a particular phenotype characteristic of B. hyodysenteriae were obtained in pure culture from five of 27 (19 %) farmed mallards and from 18 of 217 (8 %) wild mallards. Birds with strongly ß-haemolytic Brachyspira spp. were distributed throughout the sampling period. This spirochaetal phenotype was cultured from birds of both sexes and from both juvenile and adult birds. In most cases, spirochaetes with strong ß-haemolysis were accompanied by growth of weakly haemolytic spirochaetes on the same agar plate.
The overall isolation rate of Brachyspira spp. (78 %) in the wild mallards in this study was similar to previous studies (7488 %) where waterbirds of various species were analysed in Australia and the United States by culture or indirect fluorescent antibody test (Swayne & McLaren, 1997; Oxberry et al., 1998). Taken together, these results and those of previous studies strongly support the conclusion that intestinal spirochaetes are commonly found in at least some wild-living waterbird species. The isolation rate from farmed mallards was even higher than in the wild mallards. The population density is often higher on game farms than in natural habitats, and wild birds frequently fly in and out of game farms. The level of faecal contamination in ponds, lakes and on the ground may be significant.
In selective sampling from autumn-migrating mallards, the aim was to sample birds flying in from different areas in order to avoid the effect of local variations on prevalence. High population densities may be present in some breeding, resting and feeding areas. There was no way of determining whether the migrating mallards that were sampled in this study originated from different areas, or if they had lingered together to breed, feed or rest before being caught at the bird observatory. The isolation rate from these migrating mallards during autumn was almost twice as high as in mallards sampled during spring in the public park. However, of the putative risk factors for Brachyspira spp. in wild mallards, it appeared that migratory status was the most important. As migration may temporarily suppress immunocompetence in birds through increased corticosteroid levels (Holberton et al., 1996; Råberg et al., 1998), it cannot be excluded that their susceptibility to spirochaetal infections is increased during migratory seasons, or that the shedding of spirochaetes is promoted by migration. However, young age, sex (females) and season (autumn) could also be important risk factors for finding Brachyspira spp. or simply reduce the prominence of migration as a risk factor. The current evidence does not enable a firm conclusion on this issue. Moreover, there were several missing observations concerning age and sex that could distort the results of the statistical analyses. Thus, the inference that Brachyspira spp. is associated with migration should be considered indicative rather than conclusive.
PCR assay
The results of the PCR assay are shown in Table 1. All selected isolates from farmed mallards and two isolates from wild mallards (AN1409 : 2/01 and AN3907 : 2/02) were PCR-positive for the tlyA gene and were therefore classified as B. hyodysenteriae. The remaining isolates from wild mallards were negative for the tlyA gene and could not be classified by PCR.
PFGE and RAPD
The two isolates that were analysed by PFGE and RAPD originated from the mallard farm. The dendrogram based on RAPD analysis is shown in Fig. 1. The similarity values for both assays showed that the avian isolates were closely related to the reference strains B78T and B204 of B. hyodysenteriae, thus confirming their classification based on PCR as B. hyodysenteriae.
|
Phylogenetic analysis
The phylogenetic tree based on 16S rRNA sequence analysis is shown in Fig. 2. Accession numbers of the 16S rRNA gene of mallard isolates are given in Table 1. Phylogenetic analysis based on 16S rRNA sequences showed that all of the 11 studied Brachyspira strains grouped in the B. hyodysenteriaeBrachyspira intermedia cluster. The similarity values of the 16S rRNA sequences in this cluster varied from 99.4 to 100 %. This observation, and the fact that there were no coherent subclusters comprising only representatives of the respective species, shows that it is not possible to distinguish between B. hyodysenteriae and B. intermedia by sequence analysis of 16S rDNA, in accordance with previously published results (Pettersson et al., 1996). Three of the mallard strains (AN383 : 2/00, AN1409 : 2/01 and AN3907 : 2/02) had identical sequences, that were also identical to two strains of B. hyodysenteriae (R1 and NIV-1) that have been isolated from common rheas from farms in Ohio and Iowa in the United States (Sagartz et al., 1992; Jensen et al., 1996). The mallard isolates possessing this sequence were from a farmed bird, and from the two wild birds. These two wild birds had been sampled during different seasons, years and at separate locations. We believe that these three mallard isolates and the two previously described isolates from common rheas should be considered as B. hyodysenteriae. This specific B. hyodysenteriae genotype seems to be genetically rather stable, present on two continents in both farmed and wild-living avian hosts and possibly adapted to birds, as it has not been found in pigs. Two additional strains (AN3930 : 2/02 and AN3931 : 2/02) were also identical with respect to their 16S rRNA sequences, and differed from strains R1 and NIV-1 by only one nucleotide. These two latter isolates both originated from wild-living migrating mallards. Three strains from wild mallards (AN3599 : 2/02, AN3624 : 2/02 and AN3886 : 2/02) were identical and related to the sequence of strain AN3902 : 2/02 and the reference strain PWS/A of B. intermedia. Finally, two strains (AN1418 : 2/01 and AN3949 : 2/02) formed a cluster with a relatively stable node (82.2 %) basal to the B. hyodysenteriaeB. intermedia cluster. However, this cluster has no unique nucleotides that differentiate it from all other Brachyspira strains. Consequently, the negative PCR analysis and phylogenetic analyses of these isolates could not assign them to any Brachyspira species; however, they seem to be most closely related to B. hyodysenteriae and B. intermedia. The phylogenetic tree (Fig. 2) should therefore be interpreted with caution, because it is based on few sequence differences and many of the nodes are not stable (bootstrap values lower than 67 %).
|
Clinical significance
The clinical significance of B. hyodysenteriae in non-porcine species remains somewhat unclear. In common rheas, the infection has been associated with necrotizing typhlocolitis and high mortality (Sagartz et al., 1992; Buckles et al., 1997). On the other hand, challenge infection of juvenile common rheas with the R1 rhea strain of B. hyodysenteriae was inconclusive as to the development of lesions, although spirochaetes were seen in histological sections in 75 % of the infected birds (Buckles, 1996). In the same study, mallard ducklings were infected by strain R1. Histological evaluation revealed spirochaetes in only 7 % of the inoculated ducks and none of the birds developed any lesions (Buckles, 1996). Furthermore, in inoculation experiments of pigs, B. hyodysenteriae strains R1 and R358 from common rheas also failed to colonize the animals and they did not develop any lesions (Stanton et al., 1997). In a preliminary challenge infection trial approved by the Swedish Ethical Committee for Scientific Experiments (protocol C17/2), we used one of the mallard isolates of B. hyodysenteriae (AN3052 : 1/00) to inoculate three grower pigs twice orally (data not shown). We were not able to reisolate the bacterium from rectal swabs obtained daily from each pig, except from one pig within 24 h of the inoculation. Spirochaetes did not grow from caecal or colonic samples obtained at necropsy 4 weeks post- inoculation. All pigs remained healthy. In the case of mallards, we have at present no reason to believe that B. hyodysenteriae or B. hyodysenteriae-like variants are pathogenic. None of the sampled birds in this study showed any signs of disease or diarrhoea. However, we were not able to investigate any of the birds for histological evidence of lesions. This does not necessarily mean that B. hyodysenteriae and B. hyodysenteriae-like variants are avirulent in mallards; they seem more likely to be opportunistic or apathogenic to the host. Taken together, the current lack of evidence of pathogenicity of avian isolates of B. hyodysenteriae in pigs indicates avirulence. Additional studies are needed to confirm this conclusion. Moreover, if these extended challenge trials should confirm the apathogenic nature of avian B. hyodysenteriae, these isolates may serve as useful comparative tools in identifying pathogenicity mechanisms in other strains of B. hyodysenteriae.
| Conclusions |
|---|
|
|
|---|
| ACKNOWLEDGEMENTS |
|---|
|
|
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Feberwee, D. J. Hampson, N. D. Phillips, T. La, H. M. J. F. van der Heijden, G. J. Wellenberg, R. M. Dwars, and W. J. M. Landman Identification of Brachyspira hyodysenteriae and Other Pathogenic Brachyspira Species in Chickens from Laying Flocks with Diarrhea or Reduced Production or Both J. Clin. Microbiol., February 1, 2008; 46(2): 593 - 600. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Rasback, K.-E. Johansson, D. S. Jansson, C. Fellstrom, M. Y. Alikhani, T. La, D. S. Dunn, and D. J. Hampson Development of a multilocus sequence typing scheme for intestinal spirochaetes within the genus Brachyspira Microbiology, December 1, 2007; 153(12): 4074 - 4087. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |