Salmonella uptake in sheep exposed to pastures after biosolids application to agricultural land.
Dewatered biosotids derived from waste-water treatment plants are applied to agricultural land in New South Wales to recycle a resource of value to agricultural industries. The pathogen risk to the food chain of such activities has received specific attention (Gerba and Smith 2005). In previous investigations, we examined the survival of Salmonella spp. and E. coli in dewatered biosolids applied to agricultural land at several sites and under a range of seasons and weather conditions (Eamens et al. 1996, 2006; Eamens and Waldron 1999). Those studies have consistently shown that dewatered biosolids applied to agricultural land may contain high concentrations of salmonellae and E. coli for some months after application. Therefore, lengthy periods may be required before treated pastures reach pre-application levels of salmonellae and there is no risk to grazing animals or the human food chain.
Salmonellosis is an increasing issue in meat safety, and farm animals represent an important source for human salmonellosis (Thorns 2000; DuPont 2007). Sheep that are infected with and shed salmonellae in their faeces are a potential risk to the food chain. Cattle and sheep in grazing systems are incriminated less often than animals raised in intensive production systems, but can still act as intestinal carriers of Salmonella spp. without evidence of disease (Jay et al. 1997; Acha and Szyfres 2001; Purvis et al. 2005).
Salmonella disease or intestinal carriage can occur in cattle with access to biosolids (Reilly et al. 1981; Clegg et al. 1986), but cattle or horses grazing sludge-treated contaminated pastures 1 month after application are not easily infected (Jones 1984; Wray and Callow 1985). Little information is available on the impact of biosolids usage on carriage of Salmonella spp. in sheep exposed to land or pastures treated with dewatered biosolids. This study sought to determine the risks associated with grazing sheep on pastures following land treatment with dewatered biosolids by examining young sheep at varying intervals after application of biosolids containing known levels of salmonellae.
Evaluation included examination of faecal shedding of salmonellae and Salmonella infection of predilection tissue sites (lower small intestine, upper large intestine, and draining lymph nodes) at slaughter. The typical application rate of 10 dry t/ha for agricultural land was also compared to a higher rate (15 dry t/ha) that may be applicable in the future if regulatory policies are modified.
Materials and methods
Mechanically dewatered biosolids derived from the Cronulla Waste Water Treatment Plant (WWTP) were applied to agricultural land at Camden, NSW, in winter (June 1999) at 1 of 2 rates (10 or 15 dry t/ha). The soil was a Red Chromosol according to the Australian Soil Classification (Isbell 1996). Stockpiles held at Cronulla WWTP had been tested for salmonellae according to the MPN methods described below, and subsamples from stockpiles selected for this study were found to contain between 1.6 x [10.sup.1] and 1.2 x [10.sup.4] salmonellae/g of dewatered biosolids. The site for application met all criteria recommended for restricted use 2, stabilisation grade B biosolids (Environment Protection Authority NSW 1997) and application was completed over 2 days, 2 weeks apart. Pastures were then sown 1-3 weeks after the biosolids had been applied. The total study site of 16 ha was sown with a total of 52.5kg/ha of the following pasture species: oats (Avena sativa var. Temora; 30.6 kg/ha), phalaris (Phalaris aquatica; 1.75 kg/ha), cocksfoot (Dactylis glomerata; 1.75 kg/ha), subterranean clover (Trifolium subterraneum ssp. subterraneum var. Seaton Park; 3.5 kg/ha), white clover (Trifolium repens var. Haifa; 1.75 kg/ha), and perennial ryegrass (Lolium perenne var. Kangaroo Valley; 13 kg/ha).
Subsequent to pasture sowing, the site was fenced into 8 treatment paddocks of approximately 1.4 ha each (plots A-H) and a 2.73-ha control paddock free from biosolids, which also had not received fertiliser treatment. This layout enabled groups of 10 sheep to be carried per biosolids plot and 2040 sheep on the larger control plot. Drinking water for sheep in each paddock was provided through separate concrete water troughs. The biosolids treatments and fencing were arranged to provide replicate treatment plots for l0 and 15 dry t/ha biosolids treatments and early (3 months after application) or late (6 months after application) sheep entry to both treatments. The plot treatments are indicated in Table 1.
A group of 125 Merino wethers, 8 months old, were obtained from a commercial flock at Dubbo, NSW, confirmed free of ovine Johne's disease by local regulatory authorities, based on herd investigations and history over 10 years. The sheep were transported to Camden and depastured onto untreated pastures. All were individually ear-tagged with different colour codes for each of the 5 treatments (2 biosolids rates x 2 times of introduction, plus controls), and tested for Salmonella excretion in faeces (as described below). Five sheep found to be excreting salmonellae were excluded from the trial. All sheep were treated for intemal parasites with closantel (Seponver, Coopers Animal Health) in August 1999 before entry onto the treatment plots. At 2 and 4 months later, all sheep were drenched with moxidectin (Cydectin, Fort Dodge) in portable yards. At these times, half the sheep were located on the trial plots and half remained on untreated pasture 2 km distant.
Sheep were introduced onto the trial plots at 10 or 23 weeks after the pasture had been sown, at 11-13 weeks and 24-26 weeks after biosolids had been applied. All sheep were slaughtered after a further 10 weeks, 34 36 weeks after biosolids application. For the early introductions, 60 sheep were introduced to the trial site 3 months after biosolids application. At the time of introduction of sheep to the biosolids plots, scattered clumps of biosolids were present on the surface among the pasture, and pasture growth was more lush on the biosolids plots than on unfertilised control plots. Forty sheep were assigned to plots A, C, E, and G (Table 1) as replicates of 2 groups each of 10 sheep for treated plots. The other 20 sheep were introduced on to the control plot at the same time.
The remaining 65 animals were kept on untreated pasture away from the trial site. At 6 months after biosolids application, these remaining sheep were again tested for faecal shedding of Salmonella spp. A final 60 sheep found to be negative for salmonellae on faecal culture were then introduced on to the trial site. Forty of these were assigned in groups of 10 to plots B, D, F, and H, and the remaining 20 were added to the control plot which already contained 20 sheep (Table 1). This trial was undertaken in accordance with animal ethics guidelines and approved by the EMAI animal ethics committee.
Biosolids samples were collected monthly from each plot for bacteriological testing. Faeces (10 15g) was collected from each sheep before introduction onto the trial site, and then monthly after they entered their respective treatment groups. All faecal samples were collected and analysed on an individual sheep basis, and the control group was always sampled first. The 120 trial sheep were slaughtered, necropsied, and samples of intestine and lymph nodes were collected for bacteriological culture. From each animal, 10-20-g portions were collected for separate culture from the ileum, ileal (caudal jejunal) lymph node, caecal lymph node, and from a pool of caecum and proximal colon.
A 3-tube most-probable number (MPN) technique was used to determine the concentration of Salmonella spp. in biosolid samples (Eamens et al. 2006). Each plot that had received biosolids was sampled by collecting visible biosolids lumps from 5 random sites and pooling these lumps to make a minimum pool of 55 g. Duplicate subsamples of 27.5 g were then diluted and mixed in buffered peptone water (BPW), from which 3 replicate subaliquots of 10 mL were taken. The 3 tubes of each sub-aliquot from each duplicate were used to inoculate a 3-tube dilution series of BPW (10-fold from [10.sup.-2] to [10.sup.-5]) for subsequent inoculation of Rappaport-Vassiliadis (RV) broth (Eamens et al. 2006). From each BPW tube, 0.1 mL was inoculated into 10mL of RV broth (Oxoid CM 669 containing Novobiocin 35 mg/L) and incubated for 16-24h at 42[degrees]C. Each RV broth was then subcultured to each of 3 selective solid media: modified Brilliant Green (BG) agar (Oxoid CM 329 with supplement SR 87), bismuth sulfite (BS) agar (Amyl AM 22- I and 22-2), and xylose lysine desoxycholate (XLD) agar (Oxoid CM 469) for Salmonella spp. detemlinations (Bridson 1990; Standards Australia 1991). Typical colonies at 37[degrees]C on BG agar at 24 h (red colonies on reddened medium), BS agar at 48h (metallic silver/grey colonies that blacken the medium), and XLD agar at 24 h (red colonies with or without a black centre) were further selected on cystine-lactoseelectrolyte-deficient (CLED) agar (Oxoid CM 301) by incubation for 16-24h at 37[degrees]C (Standards Australia 1991). Isolates on CLED agar appearing as flat blue colonies were subcultured to nutrient agar tbr 16-24 h at 37[degrees]C and confirmed as Salmonella spp. by their reactivity with both poly O and poly H antisera (Standards Australia 1991).
Sheep faeces and tissues
Sheep samples were tested for presence or absence of salmonellae. Fresh faecal samples of l g were diluted and mixed l:10in BPW, incubated for 16h at 37[degrees]C, and inoculated into single tubes of RV broth containing 35 mg/L of novobiocin (0.1 mL inoculated into 10mL) and mannitol selenite cystine (MSC) broth (Oxoid CM 399) (1 nat. inoculated into 10 mL) for selective enrichment (Fricker 1987; Bridson 1990; Standards Australia 1991). RV and MSC broths were incubated for 16-24h at 42[degrees]C and 37[degrees]C, respectively. Each broth was subcultured to 3 selective solid media consisting of BG, BS, and XLD agars, and further selected on CLED agar as described above. Multiple enrichment media suited to use in ruminants and humans were chosen to improve the chance of isolation of a wide range of salmonella serovars. MSC enrichment is capable of detecting 1 salmonella inoculated among [10.sup.6] coliforms in ovine faeces (Grau and Smith 1972), while RV medium is widely used in salmonella isolations from many species, effluents, and food products, and novobiocin supplementation is known to increase the selectivity and isolation rates of salmonellae using direct-inoculated RV broth-based media (Fricker 1987). Salmonella spp. isolates were confirmed by reactivity with poly O and poly H antisera as described for biosolids samples.
Fresh tissues were homogeniscd in Ringer's solution in a Colworth 80 Stomacher (A.J. Seward & Co. Ltd, London) at room temperature for 1 rain (1 g tissue in 9mL) and then inoculated into RV and MSC broths as for faeces. Subsequent steps to detemaine and confirm the presence of Sahnonella spp. in tissues were as for biosolids and faecal samples.
Salmonella isolates on BG and CLED plates derived from primary culture in both RV and MSC broths were serotyped at the Australian Salmonella Reference Laboratory, Institute of Medical and Veterinary Science, Adelaide. This was undertaken to detemaine whether strains in biosolids were important human or animal pathogens, and the likelihood of strain transmission to sheep grazing the plots.
Mean concentrations of salmonellae in biosolids were initially [10.sup.4]/g, and varied over time in individual plots between undetectable and [10.sup.4]/g (Figs 1 and 2). Even at 30 weeks post-application, concentrations were generally still between [10.sup.2] and [10.sup.4]/g. Because plots A, B, C, and G received dewatered biosolids 2 weeks earlier than plots D, E, F, and H, the sampling times after application are given in Figs 1 and 2 as different values on the x-axes.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Of 125 sheep originally purchased, 5 were excluded from the study on the basis of Salmonella excretion before introduction to the trial site. In the initial testing of sheep before introduction of 60 onto the plots 3 months after biosolids application, 4 sheep of 125 tested were found to be excreting salmonellae. The additional animal with a positive faecal culture was among the remaining 61 animals initially found culture-negative and was detected before the introduction of these animals to the trial site 6 months after biosolids application.
All monthly samplings of faeces from sheep following their introduction onto treated or control plots at the trial site were negative on Salmonella culture. At slaughter, only 1 animal of the 120 trial sheep yielded Salmonella from tissues. This sheep (3313) was among the 80 sheep grazing biosolids-treated plots, and was located on plot G. It had been introduced onto this plot 3 months after biosolids application of 10 dry t/ha. Salmonella spp. were cultured from the ileal and caecal lymph nodes from this animal, and were detected in all 3 solid media from each of the 2 selective broths used for each of these tissues.
Of 85 salmonella isolates selected for serotyping, l 1 were derived from the single sheep found to be culture-positive from the ileal and the caecal lymph node (sheep 3313), and were chosen from Salmonella growth from the full range of selective broths and solid culture media used for primary isolation. Results for 74 isolates taken at random from plots A-H indicated Mbandaka 14+ (39/74) and Montevideo (26/74) were the most common serovars found in biosolids, while Hessarek (11/11) was the only serovar isolated from sheep 3313 (Table 2).
In previous studies, dewatered biosolids were found to harbour salmonellae many months after soil incorporation (Eamens et al. 1996, 2006; Eamens and Waldron 1999). In those studies, bacterial counts in dewatered biosolids usually declined gradually, but at times erratically, to eventually reach baseline concentrations (no detectable salmonellae). In the present study, the survival of salmonellae was determined for dewatered biosolids derived from an anaerobic digester and applied at Camden. Similar studies at Camden of anaerobically digested dewatered biosolids from other sewage treatment plants (Eamens et al. 1996) have shown Salmonella spp. can reach baseline concentrations after 21-33 weeks, following initial fluctuations in concentrations between 102 and 104/g during the first 20 weeks after application.
Overall, the change in Salmonella concentrations in biosolids during this study (Figs 1 and 2) showed many similarities to those of plots in the earlier study except that 7 of the 8 plots in the current study showed a low concentration 4-7 weeks after application. The latter finding was similar to results for incorporated dewatered biosolids from the Cronulla WWTP applied at Goulburn NSW (Eamens et al. 2006), where biosolids applied to agricultural land in spring revealed low salmonella concentrations 4-20 weeks after application, but subsequently rose to [10.sup.3]-[10.sup.4]/g by 25 weeks. In those earlier studies using stockpiled dewatered biosolids from the Cronulla WWTP, Salmonella concentrations ranged from 10z to [10.sup.5]-[10.sup.6]/g at the time of land application (Eamens et al. 2006). Salmonella concentrations in the present study of approximately 104/g are thus consistent with those expected for this product, but did not represent maximum concentrations.
The dewatered biosolids used in this study would be classified as Stabilisation Grade B, Contaminant Grade C biosolids according to current NSW Environment Protection Authority guidelines (Environment Protection Authority NSW 1997), and required soil incorporation within 36h of land application according to those guidelines. These requirements would therefore always necessitate application and incorporation before pasture sowing and would prevent direct access of grazing livestock to biosolids at least for the period needed for pasture establishment. The procedures adopted in this trial attempted to cover the minimum possible time between land treatment and grazing stock access (3 months), to maximise the risk of salmonella uptake. This was compared to a more conservative period (6 months) between pasture sowing and stock access.
Salmonella spp. represent the potential major bacterial risk factor in biosolids. Other important bacterial pathogens can occur in faeces, but Campylobacter jejuni and related campylobacters, which can cause human gastrointestinal disease, are thought not to survive well in sewage sludge (Jones et al. 1990; Stelzer and Jacob 1991). Our earlier studies have shown that E. coli also survives well in dewatered biosolids applied to land (Eamens et al. 2006), but the survival rate of the important shiga-toxin producing E. coli (STEC) subset is unknown. Other microorganisms in biosolids may include protozoa, viruses, parasite ova, yeasts, and fungi. However, protozoa such as Giardia can survive in dewatered biosolids on storage, but do not survive land application, and both Giardia and enteroviruses are not considered a high potential hazard for sewage sludge applied to agricultural land (Gibbs et al. 1995; Hu et al. 1996).
The food chain is an important source of human salmonellosis (Jay et al. 1997), and the current study relates to the indirect risks of biosolids for humans from infection via the food chain. Salmonella from the intestinal contents or faeces is a major source of contamination of carcasses in the abattoir environment, and thus the food chain (Taylor and McCoy1969; Jay et al. 1997; Purvis et al. 2005). Excretion of salmonellae from livestock as a source of human infection is influenced by the animal's environment, which can be both a source of bacteria and a stress factor that increases shedding rates of bacteria already present in the intestine. From biosolids on pasture, the animal's age and level of soil uptake are likely risk factors for infection by salmonellae, since this applies in humans, where age is also known to influence the minimum infectious dose (Skanavis and Yanko 1994). Prior transport, feed stress and the level of environmental contamination, particularly from high stocking rates and holding in yards (McAuliffe et al. 1978), are important environmental factors that can influence the outcome of Salmonella carriage. Adverse animal handling, particularly starvation and extended lairage, are known to increase Salmonella shedding and carcass contamination rates in sheep and cattle (McCaughey et al. 1971; Grau and Smith 1972; Grau 1974).
The initial prevalence of Salmonella excretion in the experimental sheep before the current trial began (4/125 or 3.2%) was estimated within a week of their transport to the study site and may reflect an increased shedding rate due to that recent stress. Subsequent testing of experimental sheep on plots and at slaughter used procedures that were unlikely to contribute to increased excretion of salmonellae by either prolonged stress or starvation, and thereby gave a more accurate estimate of the ongoing Salmonella shedding status of each animal. Animals grazing biosolids plots thus appeared to represent a lower risk of Salmonella carriage or spread than those which had been recently transported from a commercial property.
Previous studies (Eamens et al. 1996) concluded that salmonellae within incorporated dewatered biosolids are relatively well protected from environmental conditions, but mowing or slashing of pastures reduces their survival. Based on this, introduction of sheep may have reduced some of the grass cover, but data presented in Figs 1 and 2 indicate this was unlikely to have affected bacterial survival in the incorporated biosolids. While the introduction of sheep at week 12 preceded a subsequent fall in Salmonella concentrations in the incorporated biosolids 5-7 weeks later, this fall was not confined to plots containing sheep (Figs 1 and 2, plots A, C, D, E), as it also extended to 3 of the 4 plots without sheep (Figs 1 and 2, plots G, F, H). In addition, later entry of sheep (at 24-25 weeks) appeared to have no effect on Salmonella concentrations (Figs 1 and 2, plots B, G, F, H) in the incorporated biosolids, at least over the ensuing 5-7 weeks.
Uptake of salmonellae would depend on ingestion of surface biosolids clumps and/or soil by sheep. While this is possible, especially as not all of the applied biosolids are fully incorporated below the soil surface, there was no evidence that availability of such material resulted in Salmonella carriage in sheep. This may be in part due to the nature of the Salmonella serovars in biosolids being unable to infect sheep. Cattle are resistant to naturally occurring salmonellae in raw sewage sludge containing [10.sup.5] salmonellae/L, fed orally at up to 1 L/day (Hall and Jones 1978). However, they are susceptible to sterile sludge supplemented with the bovine-pathogenic serovar Dublin at the same feeding rate, with resultant isolation from faeces and tissues (Hall and Jones 1978). These findings support the view that the serovars present in biosolids may determine whether ruminants develop continuing infection. Sheep, like cattle, may therefore be resistant to continuing infection from the salmonella serovars predominant in the biosolids used in this trial. Previous work (G. J. Eamens, unpublished data) has shown that important ruminant serovars such as Typhimurium and Dublin are infrequent or at low concentrations in biosolids from the Sydney area. However, serovar Bovismorbificans, which can cause disease in Australian cattle (Anon. 1999, 2000), and Infantis, which is a significant ovine pathogen but not frequently associated with ovine disease in Australia (Anon. 1999, 2000), have been prominent serovars in some of those dewatered biosolids.
Selection of tissues from slaughter sheep was based on earlier reports on ovine salmonellosis. These indicated the lower small intestine and upper large intestine, particularly the ileum, caecum and proximal colon, and draining lymph nodes, are the predilection sites for subacute and chronic disease as it occurs in Australia (Moo et al. 1980; Samuel et al. 1981; Richards et al. 1993), and the UK. (Brown etal. 1977; Daniel et al. 1997). The inclusion of mannitol selenite cystine (MSC) broth in the bacteriological methods as an adjunct to Rappaport-Vassiliadis (RV) broth was based on earlier work (Grau and Smith 1972) showing MSC broth useful for ovine faeces, and superior to tetrathionate and mannitol-selenite broths. In addition, since Oxoid RV broth may be inhibitory to the ovine/bovine serovar Dublin (Peterz et al. 1989; June et al. 1995), this additional selective broth gave coverage of a wider range of serovars that may have been present in sheep samples. Previous studies (G. J. Eamens, unpublished data), in which both RV and MSC broths were used for salmonella isolations from a range of biosolids, have not found serovar Dublin in Sydney biosolids (0/918 isolates examined), and serotyping statistics indicate this serovar is rarely detected in humans in Australia (Anon. 2004c, 20066).
Salmonella serotyping has been used to study the risk of biosolids to animal populations. One study (Linklater et al. 1985) described 317 isolates from biosolids representing 34 different serotypes, but only 5 of which occurred in animals. That report confirmed the presence of S. Typhimurium of the same phage type to that recovered from incidents on 11 farms, 4 of which had received biosolids. These workers concluded that human populations are likely to carry a higher Salmonella background than farm animal populations.
The current study identified one serovar not previously found (Albany) in biosolids in earlier projects where 918 and 114 isolates were serotyped and 31 and 19 serovars, respectively, were identified (G. J. Eamens, unpublished data). Of the 5 other serovars detected, Montevideo (0.9-12%), Virchow (4.4-8.7%), Ohio (1.1-17.5%), and Subsp 1 (0.9-17.1%) have been relatively common in biosolids in our earlier studies, while Mbandaka was infrequent (0-0.2%). In contrast, serovar Hessarek, isolated from 1 sheep in this study, has not been detected among a total of 1105 isolates from biosolids in this or the 2 prior studies of biosolids derived from sewage treatment plants in the Sydney region. This provides strong evidence that the Salmonella serovar detected in sheep 3313 was unlikely to have originated from biosolids.
The 6 serovars found in biosolids in this study represent 7% of all isolates typed from Australian humans annually in 2003 and 2005 (Anon. 2004a, 20066), and only one of these serovars (Virchow) was among the top 10 serovars responsible for human disease in Australia, being associated with 5-6% of confirmed human cases. The Salmonella serovar isolated from sheep 3133 tissues (Hessarek) is infrequently found in animals, with reports of its occurrence in limited numbers from horses, cattle, poultry and poultry meat, cats, and wild birds or animals (Anon. 2000, 2003, 2004a, 2004b, 2005, 2006a). In humans, this serovar was found in 1 and 15 human submissions among 7000-8000 strains received annually in Australia in 2003 and 2005 (Anon. 2004c, 2006b).
Earlier studies of liquid sludge or slurries applied onto pasture or into plastic bins containing soil, and sometimes using insensitive techniques, have reported that salmonellae can die out on herbage within 30 days following application (Jones 1984; Dickson and Tribe 1986). Since bacterial survival is reported to be greater in slurries with >5% solids (Strauch 1991) and dewatered biosolids contain 20-25% solids, figures for liquid slurries will underestimate the survival times in dewatered biosolids. Guidelines should therefore be specific for the biosolids product in use, rather than based on reports for effluents containing only 1-2% solids. Current NSW guidelines (Environment Protection Authority NSW 1997) restrict grazing animal access for 30-90 days after biosolids application and incorporation. This time-frame is not relevant for dewatered biosolids in which a period of many months may elapse before baseline counts for bacteria are reached (Eamens et al. 2006), and when pastures are sown after biosolids applications. Withholding periods of 6-12 months, consistent with bacterial survival times (Gerba and Smith 2005; Eamens et al. 2006), and normal practice under current pasture management systems to enable pasture establishment after biosolids incorporation, are more meaningful.
As demonstrated in this study, dewatered biosolids applications may pose little threat of salmonella uptake to grazing sheep as young as 8 months, even when grazed 3 months after soil incorporation. However, as the biosolids were found to contain moderate to low Salmonella concentrations (at times up to [10.sup.4]/g) in this study, further studies may be required if biosolids are used that are shown to contain very high concentrations of salmonellae (e.g. [10.sup.6]/g), as found in our earlier studies (Eamens et al. 2006). It is possible that only those biosolids which contain moderate to high concentrations of ruminant-pathogenic serovars may pose a risk of grazing-animal transfer to the food chain. Our current and previous findings indicate these serovars are not common in biosolids from the Sydney region.
This work was supported by funds from the Sydney Water Corporation. We thank Jocelyn Gonsalves who provided excellent backup technical support, Stephen Pitt for assistance in the regular mustering of experimental sheep for monthly samplings, and Dr Trevor Gibson for valuable comments on the manuscript. The assistance of Dorothy Thompkins, Kim Koeford and Joe Kormos in initial sheep handling procedures and fencing required for the experiment was also invaluable.
Manuscript received 4 October 2007, accepted 19 March 2008
Acha PN, Szyfres B (2001) Salmonellosis. In 'Zoonoses and communicable diseases common to man and animals.' pp. 233-246. (Pan American Health Organisation: Washington, DC)
Anon. (1999) National Enteric Pathogens Surveillance Scheme--Non-human Annual Report 1998. (Ed. J Powling) p. 14.
Anon. (2000)National Enteric Pathogens Surveillance Scheme--Non-human Annual Report 1999. (Ed. J Powling) p. 19.
Anon. (2003) National Enteric Pathogens Surveillance Scheme--Non-Human Annual Report 2002. (Ed. J Powling) p. 24.
Anon. (2004a) National Enteric Pathogens Surveillance Scheme--Non-Human Annual Report 2003. (Ed. J Powling) p. 25.
Anon. (20046) Australian Salmonella Reference Centre--Annual Report 2004. Institute of Medical and Veterinary Science, Infectious Diseases Laboratories, Adelaide, S. Aust. pp. 23-32.
Anon. (2004c) National Enteric Pathogens Surveillance Scheme--Human Annual Report 2003. (Ed. J Powling) p. 18.
Anon. (2005) National Enteric Pathogens Surveillance Scheme--Non-Human Annual Report 2004. (Ed. J Powling) p. 24.
Anon. (2006a) Australian Salmonella Reference Centre--Annual Report 2006. Institute of Medical and Veterinary Science, Infectious Diseases Laboratories: Adelaide, S. Aust. pp. 25-37.
Anon. (20066) National Enteric Pathogens Surveillance Scheme--Human Annual Report 2005. (Ed. J Powling) p. 41.
Bridson EY (1990) Rappaport Vassiliadis Enrichment Broth. In 'The Oxoid manual', pp. 2.181-2.182. (Unipath Ltd: Basingstoke, Hants, UK)
Brown DD, Ross JG, Smith AFG (1977) Experimental infection of sheep with Salmonella infantis. The British Veterinary Journal 133, 435-441.
Clegg FG, Wray C, Dancan AL, Appleyard WT (1986) Salmonellosis in two dairy herds associated with a sewage farm and water reclamation. Journal of Hygiene 97, 237-246.
Daniel RG, Barrow P, Szmolleny G, Wood M (1997) Salmonella hindmarsh infection in sheep and ponies. The Veterinary Record 141, 203.
Dickson PH, Tribe HT (1986) 'The fate of salmonellae, presumptive coliform bacteria and faecal streptococci in raw sewage sludge buried in soils.' (WRC Environment Medmenham: Bucks, UK)
DuPont HL (2007) The growing threat of foodborne bacterial enteropathogens of animal origin. Clinical Infectious Diseases 45, 1353-1361. doi: 10.1086/522662
Eamens G, Waldron A (1999) Survival and growth of pathogenic and indicator bacteria in dewatered sewage sludges after agricultural application. In 'IX International Congress of Bacteriology and Applied Microbiology', p. 20. (International Union of Microbiological Societies: Utrecht, The Netherlands)
Eamens GJ, Lavis AM, Ross AD (1996) Survival of pathogenic and indicator bacteria in biosolids applied to agricultural land. In 'Proceedings of the Biosolids Summit--Biosolids Research in NSW'. Richmond, NSW. (Eds GJ Osborne, RL Parkin, DL Michalk, AM Grieve) pp. 129-138. (NSW Agriculture Waste Recycling Unit: Sydney)
Eamens GJ, Lavis AM, Nicholls PJ (2006) Survival of pathogenic and indicator bacteria in biosolids in agricultural applications. Australian Journal of Soil Research 44, 647-659. doi: 10.1071/SR06015
Environment Protection Authority NSW (1997) Section 3.4-3.6. In 'Environmental guidelines: use and disposal of biosolids products'. pp. 11-18. (Environment Protection Authority: Chatswood, NSW)
Fricker CR (1987) The isolation of salmonellas and campylobacters. Journal of Applied Bacteriology 63, 99-116.
Gerba CP, Smith JEJ (2005) Sources of pathogenic microorganisms and their fate during land application of wastes. Journal of Environmental Quality 34, 42-48.
Gibbs RA, Hu CJ, Ho GE, Philips PA, Unkovitch 1 (1995) Die-off of human pathogens in sludge amended soil. In 'Proceedings of the Australian Water and Wastewater Association 16th Federal Convention'. pp. 667-674. (Australian Water and Wastewater Association: Artarmon, NSW)
Grau FH, Smith MG (1972) A comparison of methods for the enrichment of salmonellae from sheep faeces. Journal of Applied Bacteriology 35, 531-536.
Grau FH (1974) Salmonella contamination of sheep and mutton carcasses related to pre-slaughter holding conditions. Journal of Applied Bacteriology 37, 111-116.
Hall GA, Jones PW (1978) A study of the susceptibility of cattle to oral infection by salmonellas contained in raw sewage sludges. Journal of Hygiene 80, 409-414.
Hu C, Gibbs R, Mort N, Hofstede H, Ho G, Unkovitch I (1996) Organisms a threat in stored sludge. Waste Management and Environment 7, 21-23.
Isbell RF (1996) 'The Australian Soil Classification. Vol 4.' (CSIRO: Collingwood, Vic.)
Jay S, Grau FH, Smith K, Lightfoot D, Murray C, Davey GR (1997) Salmonella. In 'Foodborne microorganisms of public health significance'. (Eds AD Hocking, G Arnold, I Jenson, K Newton, P Sutherland) pp. 169-192. (CSIRO Food Science and Technology: North Ryde, NSW)
Jones K, Betaieb M, Telford DR (1990) Seasonal variation of thermophilic campylobacters in sewage sludge. The Journal of Applied Bacteriology 69, 185-189.
Jones PW (1984) The survival and infectivity for cattle of salmonellas on grassland. In 'Processing and use of sewage sludge'. (Eds P L'Hetmite, H Ott) pp. 178-190. (D. Reidel Publishing Co.: Dortrecht, The Netherlands)
June GA, Sherrod PS, Hammack TS, Amaguana RM, Andrews RH (1995) Relative effectiveness of selenite cystine broth, tetrathionate broth, and Rappaport-Vasiliadis medium for the recovery of Salmonella from raw flesh and other highly contaminated foods. Journal of AOAC International 78, 375-380.
Linklater KA, Graham MM, Sharp JCM (1985) Salmonellae in sewage sludge and abattoir effluent in south-east Scotland. Journal of Hygiene 94,301-307.
McAuliffe PR, Hardefeldt KW, Hucker DA (1978) Health problems associated with export of live sheep. Australian Veterinary Journal 54, 594-595. doi: 10.111 l/j.1751-0813.1978.tb02427.x
McCaughey WJ, McClelland TG, Hanna J (1971) Some observations on Salmonella dublin infection in clinically healthy beef cattle. The British Veterinary Journal 127, 549-556.
Moo D, O' Boyle D, Mathers W, Frost AJ (1980) The isolation of salmonella from jejunal and caecal lymph nodes of slaughtered animals. Australian Veterinary Journal 56, 181-183. doi: 10.111 l/j.1751-0813.1980. tb05675.x
Peterz M, Wilberg C, Norberg P (1989) The effect of incubation temperature and magnesium chloride concentration on growth of salmonella in homemade and in commercially available dehydrated Rappaport-Vassiliadis broths. Journal of Applied Bacteriology 66, 523-528.
Purvis GM, Hullah K, Pascoe SJS, Evans SJ, Davies RH (2005) Persistence of Salmonella Typhimurium DTI 20 in abattoir paddocks holding sheep. The Veterinary Record 157, 165-167.
Reilly WJ, Forbes GI, Paterson GM, Sharp JCM (1981) Human and animal salmonellosis in Scotland associated with environmental contamination 1973-1979. The Veterinary Record 108, 553-555.
Richards RB, Norris RT, Higgs ARB (1993) Distribution of lesions in ovine salmonellosis. Australian Veterinary Journal 70, 326-330. doi: 10.1111/ j.1751-0813.1993.tb00873.x
Samuel JL, Eccles J A, Francis J (1981) Salmonella in the intestinal tract and associated lymph nodes of sheep and cattle. Journal of Hygiene 87, 225-232.
Skanavis C, Yanko WA (1994) Evaluation of composted sewage sludge based soil amendments for potential risks of salmonellosis. Journal of Environmental Health 56, 19-23.
Standards Australia (1991) Method 2.5. Examination for specific organisms--Salmonellae. In 'Australian Standard 1766.2.5. Food microbiology', pp. 1-8. (Standards Australia: Homebush, NSW)
Stelzer W, Jacob J (1991) A study of Campylobacter in sewage, sewage sludge and in river water. Water Science and Technology 24, 117-120.
Strauch D (1991) Survival of pathogenic micro-organisms and parasites in excreta, manure and sewage sludge. Revue Scientifique et Technique Office International des Epizooties 10, 813-846.
Taylor J, McCoy JH (1969) Salmonella and Arizona infections. In 'Food borne infections and intoxications'. (Ed. H Riemann) pp. 4-5. (Academic Press: New York)
Thorns CJ (2000) Bacterial food-borne zoonoses. Revue Scientifique et Technique Office International des Epizooties 19, 226-239.
Wray C, Callow RJ (1985) A note on potential hazards to animals grazing on pasture improperly treated with sewage sludge. Journal of Applied Bacteriology 58, 257-258.
G. J. Eamens (A,C) and A. M. Waldron (A,B)
(A) NSW Department of Primary Industries, Elizabeth Macarthur Agricultural Institute, PMB 8, Camden, NSW 2570, Australia.
(B)Present address: University of Sydney, PMB 3, Camden, NSW 2570, Australia.
(C) Corresponding author. Email: firstname.lastname@example.org
Table 1. Plot designations for experimental treatments Early, 3 months after biosolids application; late, 6 months after biosolids application Biosolids rate No. of Plot (dry t/ha) sheep Sheep entry A 10 10 Early B 10 10 Late C 15 10 Early D 15 10 Late E 10 10 Early F 10 10 Late G 15 10 Early H 15 10 Late Control 0 20,20 Early and late Table 2. Frequency of Salmonella serovars among 85 isolates tested from biosolids (n=74) and sheep (n=11, all from sheep 3313) Source Serovar No. of isolates Biosolids Mbandaka var. 14+ 39 Biosolids Montevideo 26 Biosolids Ohio 4 Biosolids Virchow 3 Biosolids Albany 1 Biosolids Subsp I serovar 16:1, v:- 1 Sheep 3313 Hessarek 11