Ian Carmichael

Longitudinal prevalence, oocyst shedding and molecular characterisation of Cryptosporidium species in sheep across four states in Australia.

The prevalence of Cryptosporidium in sheep in the eastern states of Australia has not been well described, therefore a study of the prevalence, oocyst concentration, species and subtypes of Cryptosporidium were assessed from lamb faecal samples at three sampling periods (weaning, post-weaning and pre-slaughter) from eight farms across South Australia, New South Wales, Victoria and Western Australia. A total of 3412 faecal samples were collected from approximately 1182 lambs across the four states and screened for the presence of Cryptosporidium using a quantitative PCR (qPCR) at the actin locus. Positives were typed at the 18S locus and at a second locus using C. parvum and C. hominis specific qPCR primers. The overall prevalence was 16.9% (95% CI: 15.6-18.1%) and of the 576 positives, 500 were successfully genotyped. In general, the prevalence of Cryptosporidium was higher in WA than the eastern states. Cryptosporidium prevalence peaked at 43.9% and 37.1% at Pingelly (WA2) and Arthur River (WA1), respectively during weaning and at Pingelly (WA2) during pre-slaughter (36.4%). The range of oocyst shedding at weaning overall across all states was 63-7.9×10(6) and the median was 3.2 × 10(4) oocysts g(-1). The following species were identified; C. xiaoi (69%-345/500), C. ubiquitum (17.6%-88/500), C. parvum (9.8%-49/500), C. scrofarum (0.8%-4/500), mixed C. parvum and C. xiaoi (2.4%-12/500), C. andersoni (0.2%-1/500) and sheep genotype 1 (0.2%-1/500). Subtyping of C. parvum and C. ubiquitum isolates identified IIa and IId subtype families within C. parvum (with IId as the dominant subtype) and XIIa within C. ubiquitum. This is the first published description of C. parvum subtypes detected in lambs in Australia.

Development of a quantitative PCR (qPCR) for Giardia and analysis of the prevalence, cyst shedding and genotypes of Giardia present in sheep across four states in Australia

A novel quantitative PCR (qPCR) for Giardia at the glutamate dehydrogenase (gdh) locus was developed and validated. The qPCR was used to screen a total of 3412 lamb faecal samples collected from approximately 1189 lambs at three sampling periods (weaning, post-weaning and pre-slaughter) from eight farms across South Australia (SA), New South Wales (NSW), Victoria (Vic) and Western Australia (WA). The overall prevalence was 20.2% (95% CI 18.9-21.6) and of the 690 positives, 473 were successfully typed. In general, the prevalence of Giardia varied widely across the different farms with the highest prevalence in one WA farm (42.1%) at pre-slaughter sampling and the lowest prevalence in one Victorian farm (7.2%) at weaning. The range of cyst shedding at weaning, post-weaning and pre-slaughter overall across all states was 63-1.3×10(9) cysts g(-1) (median=1.7×10(4)), 63-1.1×10(9) cysts g(-1) (median=9.6×10(3)), 63-4.7×10(9) cysts g(-1) (median=8.1×10(4)) respectively. Assemblage specific primers at the triose phosphate isomerase (tpi) locus identified assemblage A in 22.4% (106/473) of positive samples typed, assemblage E in 75.9% (359/473) and mixed A and E assemblages in 1.7% (8/473) of samples. A subset of representative samples from the 8 farms (n=32) were typed at both the gdh and beta-giardin loci and confirmed these results and identified sub-assemblage AII in 16 representative assemblage A isolates across the 8 farms. This demonstrates a prevalence of Giardia previously not recognised in Australian sheep, highlighting a need for further research to quantify the production impacts of this protozoan parasite.

Variation among Merino sheep in susceptibilty to lice (Bovicola ovis) and association with susceptibility to trichostrongylid gastrointestinal parasites

Sheep of two bloodlines of Merino were artificially infested with equal numbers of lice (Bovicola ovis) and the resulting louse populations were monitored over the following 20 months. The sheep were shorn 6 and 17 months after infestation and, for analysis, the louse counts considered in 3 years separated by shearings. Nematode faecal egg counts (FECs) were assessed on faecal samples collected on five occasions, three times following natural challenge and twice after artificial challenge with 40,000 trichostrongyloid larvae (84% Trichostrongylus vitrinus). In addition, blood samples were collected and measured for B. ovis-specific immunoglobulins (predominantly IgG), B. ovis-specific IgE and serum total IgE. Bloodlines differed significantly in the size of louse populations at the end of year 2, FEC after both natural and artificial challenge and in serum levels of all three antibodies (p<0.05). There were also large variations in louse counts and FEC among sheep within bloodlines. Louse counts at inspections after louse populations had been allowed to build up were highly repeatable, both between and within years. However, correlations with counts at inspections soon after initial infestation and following shearing were lower. FEC after natural challenge was correlated with louse counts in year 2 (r=0.45, p<0.01) and year 3 (r=0.38, p<0.05), but the correlation with counts in year 1 was not significant (r=0.25, p>0.05). FEC following artificial challenge was significantly correlated with louse counts in year 3 (r=0.36, p<0.05), but not in year 2 (r=0.25, p>0.05) or year 1 (r=0.04, p>0.05). Louse counts in the 3 years were significantly correlated with anti-B. ovis antibody concentration (r=0.60, 0.48, 0.36), but not with levels of either anti-B. ovis or total serum IgE. These results suggest that sheep with greater resistance to gastrointestinal parasites also tend to be less susceptible to lice. Whether this is due to interaction of the effects of the parasites or to correlation in underlying resistance mechanisms requires clarification.

Longitudinal prevalence and faecal shedding of Chlamydia pecorum in sheep

The prevalence and faecal shedding of Chlamydia spp. in sheep in Australia has not been well described. Two species-specific quantitative PCRs (qPCRs) targeting the chlamydial outer membrane protein cell surface antigen gene (ompA) were validated and used to determine the prevalence and faecal shedding of C. abortus and C. pecorum from faecal samples of lambs at three sampling times (weaning, post-weaning and pre-slaughter) from eight farms in South Australia, New South Wales, Victoria and Western Australia. A total of 3412 faecal samples were collected and screened from approximately 1189 lambs across the four states. C. abortus was not detected in any of the samples screened. The overall prevalence of C. pecorum was 1027/3412 (30.1%) and median bacterial concentrations at weaning, post-weaning and pre-slaughter were 1.8 × 10(7), 1.2 × 10(7) and 9.6 × 10(5)/g faeces, respectively. A subset of C. pecorum positive samples from each farm, (n = 48) was sequenced to confirm their identity. The present study demonstrates that C. pecorum is prevalent in Australian sheep, highlighting a need for further research on the impact of this bacterium on production.

Greater intensity and frequency of Cryptosporidium and Giardia oocyst shedding beyond the neonatal period is associated with reductions in growth, carcase weight and dressing efficiency in sheep

Associations between intensity and frequency of Cryptosporidium and Giardia shedding with growth, carcase weight and dressing% were investigated using a longitudinal study of 1182 lambs on eight Australian farms. Live weight was recorded and faecal samples were collected on three sampling occasions; weaning (approximately 12 weeks of age), post-weaning (approximately 19 weeks) and pre-slaughter (approximately 29 weeks). Hot standard carcase weight (HSCW) and dressing% were measured at slaughter. Faecal samples were screened for presence and concentration of Cryptosporidium, Giardia and Haemonchus oocysts using a quantitative PCR. Trichostrongylid eggs were quantified with modified McMaster faecal worm egg count (WEC). Protozoan shedding intensity was categorised as high (above median oocyst concentration in positive sheep), low (below median oocyst concentration in positive sheep) or not detected. Shedding was also categorised for shedding type (no shedding, single Giardia infection, single Cryptosporidium infection, concurrent Giardia and Cryptosporidium infection) and lambs were categorised for frequency of shedding (shedding identified on 0, 1, 2 or 3 occasions). Associations of parasite shedding intensity category, shedding type, shedding frequency, WEC and Haemonchus status (positive or negative) with lamb production were assessed using general linear models (HSCW and dressing%) and linear mixed effects models (live weight). High Cryptosporidium parvum shedding was associated with lower live weight, ranging 2.31-4.52kg over the 3 sampling occasions. Cryptosporidium parvum shedding was associated with less HSCW in high (3.22kg less) and low (3.22kg less) shedding lambs post-weaning, and high (2.21kg less) and low (2.60kg less) shedding lambs pre-slaughter as well as lower dressing% (2.7% lower in high shedding lambs post-weaning). Cryptosporidium (all species) shedding pre-slaughter was associated with reduced dressing% in both high (1.25% lower) and low (1.21% lower) shedding lambs. Giardia shedding pre-slaughter was associated with 0.59kg less HSCW in high shedding lambs. Increased frequency of C. parvum and Giardia shedding in a specific animal (repeated detection) were associated with reduced HSCW and dressing%. Concurrent Giardia and Cryptosporidium shedding pre-slaughter was associated with reduced dressing%. No statistically significant main effects for either WEC (P>0.05) or Haemonchus status (P>0.05) were identified for any of the sheep meat productivity measures (live weight, HSCW and dressing%). The findings suggest naturally acquired Cryptosporidium and Giardia infections in grazing sheep are associated with depressed growth, carcase weight and dressing efficiency beyond the neonatal period in sheep representing a range of genetic backgrounds and different sheep production environments.

Prevalence, faecal shedding and genetic characterisation of Yersinia spp. in sheep across four states of Australia

OBJECTIVES To develop molecular tools for the investigation of the prevalence, species and faecal shedding of Yersinia in sheep. METHODS A quantitative PCR (qPCR) targeting the β subunit of the Yersinia spp. RNA polymerase gene was developed and validated. The prevalence of pathogenic Y. enterocolitica was determined by screening for the virulent yst gene. These qPCR assays were used to determine Yersinia spp. prevalence and faecal shedding concentration from 3412 faecal samples collected from approximately 1189 lambs (100-180 lambs/flock) on eight farms across Australia. This was a longitudinal study, with sheep sampled on three occasions (weaning, post-weaning and pre-slaughter). A subset of up to five positive samples from each sampling on each farm (n = 111) was sequenced. RESULTS Yersinia spp. (including both pathogenic and non-pathogenic species) were identified in all flocks, with 60.7% of lambs shedding Yersinia spp. on at least one sampling occasion. Point prevalence ranged from 4% to 91% across farms and sampling occasions. Median Yersinia spp. bacterial concentration was 1.1 × 10(6) , 2.8 × 10(6) and 5.6 × 10(5) organisms/g faeces at weaning, post-weaning and pre-slaughter, respectively, across all farms. Pathogenic Y. enterocolitica was identified in all eight flocks sampled, with 14.8% of lambs shedding pathogenic Y. enterocolitica on at least one sampling occasion. CONCLUSION Yersinia spp. and pathogenic Y. enterocolitica in particular were commonly identified in a sample of Australian sheep flocks using molecular techniques. Further studies into associations between faecal shedding of pathogenic Yersinia spp. and sheep productivity or clinical disease may utilise qPCR in conjunction with other diagnostic tools.

Longitudinal prevalence, faecal shedding and molecular characterisation of Campylobacter spp. and Salmonella enterica in sheep.

Faecal excretion of Campylobacter spp. and Salmonella enterica in sheep in Australia was determined using a quantitative multiplex PCR (qPCR) targeting the Campylobacter spp. purine biosynthesis gene (PurA) and the S. enterica outer membrane protein (ompF). The mutiplex qPCR was specific and Campylobacter spp. and S. enterica were each detected with a sensitivity of 5 organisms/µL faecal DNA extract. This multiplex qPCR was used to determine the prevalence and concentration of Campylobacter spp. and S. enterica in 3412 faecal samples collected from 1189 lambs on eight farms across South Australia (n = 2 farms), New South Wales (n = 1), Victoria (n = 2) and Western Australia (n = 3) at three sampling periods (weaning, post-weaning and pre-slaughter). The overall prevalences of Campylobacter spp. and S. enterica were 13.3% and 5.0%, respectively, with the highest prevalence for Campylobacter spp. in South Australia and the highest prevalence for S. enterica in New South Wales. Campylobacter jejuni was the only Campylobacter sp. identified from a subset of 120 positive samples sequenced at the 16S locus. S. enterica serovar Typhimurium was the only serovar of S. enterica identified from a subset of 120 positive samples sequenced at the ompF locus. Across all states, Campylobacter spp. had the highest median bacterial concentration in faeces at weaning and post-weaning (medians of 3.4 × 10(6) and 1.1 × 10(5), respectively), whereas S. enterica had the highest median bacterial concentration at pre-slaughter (1.8 × 10(5)/g faeces).

Longitudinal prevalence, oocyst shedding and molecular characterisation of Eimeria species in sheep across four states in Australia

The prevalence of Eimeria in sheep in Australia has not been well described, therefore a quantitative PCR (qPCR) was developed, validated and used to study the prevalence and oocyst concentration in lamb faecal samples at three sampling periods (weaning, post-weaning and pre-slaughter) from eight farms across South Australia, New South Wales, Victoria and Western Australia. A total of 3412 faecal samples were collected from approximately 1182 lambs across the 4 states and screened for the presence of Eimeria using this qPCR at the 18S ribosomal RNA (rRNA) locus. A subset of positives was typed by sequence analysis at the 18S locus. The overall prevalence was 18.1% (95% CI 16.8-19.3%) and of the 616 positives, 118 were successfully genotyped. The prevalence of Eimeria was highest in NSW and peaked at 70% during the post-weaning period. The range of oocyst shedding per gram of faeces (g(-1)) at weaning, post-weaning and pre-slaughter overall across all states was 23-2.1×10(7), 23-1.3×10(7) and 23-2.1×10(5), respectively. Median Eimeria shedding g(-1) was higher during post-weaning (1.1×10(3)) and pre-slaughter (1.1×10(3)) than during weaning (206). The following species were identified: Eimeria crandallis, Eimeria ahsata, Eimeria ovinoidalis, Eimeria weybridgensis and Eimeria cylindrica. Of these, E. crandallis and E. ovinoidalis, the most pathogenic species in sheep were responsible for 58.5% of infections typed. This highlights a need for further research to quantify the production impacts of Eimeria in sheep.

Faecal shedding of pathogenic Yersinia enterocolitica determined by qPCR for yst virulence gene associated with reduced live weight but not diarrhoea in prime lambs

Associations between faecal shedding of pathogenic Yersinia enterocolitica (based on the yst virulence gene) with growth, carcass weight and diarrhoea were investigated using an observational longitudinal study of 1200 crossbred prime (meat) lambs on eight Australian farms. Live weight, breech faecal soiling score (scale 1-5) and faecal consistency score (FCS; scale 1-5) were recorded, and faecal samples collected from each lamb on three sampling occasions; weaning (≈12 weeks of age), post-weaning (≈19 weeks) and pre-slaughter (≈29 weeks). Hot standard carcass weight was measured at slaughter. Faecal samples were screened for presence and concentration of pathogenic Y. enterocolitica using quantitative PCR. Associations of pathogenic Y. enterocolitica detection and shedding intensity with lamb health and production were assessed using general linear models (carcass weight), linear mixed effects models (live weight, FCS and breech soiling score) and non-parametric tests (FCS and breech soiling score). Prevalence for non-pelleted faeces (FCS ≥ 3.0) and diarrhoea (FCS ≥ 4.0) were compared with the two-tailed z-test, odds ratios and relative risk. Lambs shedding pathogenic Y. enterocolitica were 3.78 kg lighter post-weaning (P < 0.001) and 2.61 kg lighter pre-slaughter (P = 0.035) compared to lambs in which pathogenic Y. enterocolitica was not detected. Higher faecal concentration of pathogenic Y. enterocolitica was associated with lower live weight (P < 0.001). There was no association between pathogenic Y. enterocolitica detection and carcass weight. Overall, there was no evidence of association between pathogenic Y. enterocolitica detection and diarrhoea (higher FCS, higher risk for non-pelleted faeces or diarrhoea, or higher breech soiling score). Only one flock had increased relative risk for non-pelleted faeces associated with pathogenic Y. enterocolitica detection, and one other flock had increased relative risk for diarrhoea associated with pathogenic Y. enterocolitica detection. This is the first report of an association between reduced sheep live weight and pathogenic Y. enterocolitica based on the presence of the yst gene for heat stable enterotoxin determined by qPCR in sheep. Notably, impacts on live weight were observed in the absence of diarrhoea.