Comparison of alternatives to in-feed antimicrobials for the prevention of clinical necrotic enteritis

Transcription

1 Journal of Applied Microbiology ISSN ORIGINAL ARTICLE Comparison of alternatives to in-feed antimicrobials for the prevention of clinical necrotic enteritis M.S. Geier 1,2, L.L. Mikkelsen 2,3, V.A. Torok 2,4, G.E. Allison 5, C.G. Olnood 2,3, M. Boulianne 1,6, R.J. Hughes 1,2 and M. Choct 2,3 1 Pig and Poultry Production Institute, South Australian Research and Development Institute, The University of Adelaide, Roseworthy Campus, SA, Australia 2 Australian Poultry Co-operative Research Centre, University of New England, Armidale, NSW, Australia 3 School of Environmental and Rural Science, University of New England, Armidale, NSW, Australia 4 Plant and Soil Health, South Australian Research and Development Institute, Plant Research Centre, Waite Campus, Urrbrae, SA, Australia 5 School of Biochemistry and Molecular Biology and ANU Medical School, Australian National University, Canberra, ACT, Australia 6 Faculty of Veterinary Medicine, University of Montreal, Quebec, Canada Keywords antibiotics, microbiota, necrotic enteritis, organic acid, probiotics. Correspondence Mark S. Geier, Pig and Poultry Production Institute, South Australian Research and Development Institute, Roseworthy, SA, Australia Mark.Geier@sa.gov.au : received 9 December 2009, revised 28 March 2010 and accepted 14 April 2010 doi: /j x Abstract Aims: The capacity for Lactobacillus johnsonii and an organic acid (OA) blend to prevent Clostridium perfringens-induced clinical necrotic enteritis (NE) in chickens was studied. Methods and Results: Cobb 500 birds were allocated into six groups (n =25 birds pen, eight pens treatment); Unchallenged, Challenged, Antimicrobial (zinc bacitracin (ZnB) monensin), OA, probiotic Lact. johnsonii and probiotic sham (Phosphatebuffered saline). All birds were challenged with Eimeria spp. and Cl. perfringens except for unchallenged controls. Birds fed antimicrobials were protected from NE development as indicated by maintenance of body weight, low mortality and clostridium levels, and decreased intestinal macroscopic lesion scores compared to challenged controls (P < 0Æ05). Lactobacillus johnsonii-fed birds had reduced lesion scores, whilst OA-fed birds had decreased Cl. perfringens levels. Both Lact. johnsonii and OA-fed birds had improved feed efficiency between days 0 and 28 compared to challenged controls; however, mortality and body weights were not improved by either treatment. Microbial profiling indicated that the challenge procedure significantly altered the jejunal microbiota. The microbiota of antimicrobial-fed birds was significantly different from all other groups. Conclusions: Whilst Lact. johnsonii and OA altered specific intestinal parameters, significant protection against NE was not observed. Significance and Impact of the Study: Lactobacillus johnsonii and OA did not prevent NE; however, some improvements were evident. Other related treatments, or combinations of these two treatments, may provide greater protection. Introduction Necrotic enteritis (NE), caused by Clostridium perfringens, is one of the world s most prominent and severe poultry diseases (Van Immerseel et al. 2004; Cooper and Songer 2009). The economic impact of NE on the world-wide poultry industry is estimated at over US$2 billion per annum (Choct and Kocher 2008). NE is typified by intestinal lesions, diarrhoea, impaired digestive function, reduced nutrient absorption and decreased feed intake (Van Immerseel et al. 2004; Gholamiandehkordi et al. 2007). In its acute clinical form, NE can cause significant flock mortality over a period of several days, whilst subclinical Cl. perfringens infection can significantly impair bird performance (Van Immerseel et al. 2004). Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)

2 Necrotic enteritis in broilers In many countries, NE is controlled by the in-feed supplementation of antibiotics such as ZnB. The European Union has enforced a ban on the use of in-feed antibiotics (Timbermont et al. 2009), and consumer pressure in other regions may force similar restrictions on antibiotic use (Stringfellow et al. 2009). Therefore, alternative strategies for the control of NE are needed which can limit the health and economic impact of the disease. Possible alternatives to in-feed antibiotics include probiotics, prebiotics and organic acids (OAs) (Dahiya et al. 2006), or the development of vaccines targeted towards Cl. perfringensderived toxins, including NetB (Keyburn et al. 2008; Van Immerseel et al. 2009). The aims of this study were to assess the capacity for an OA blend, or a candidate probiotic organism, Lact. johnsonii, to reduce the incidence and severity of clinical NE in broiler chickens fed diets without traditional antimicrobials. In addition, we aimed to assess the effects of these treatments on the intestinal microbiota, by profiling overall microbial communities, Lactobacillus species profiles, and by measuring volatile fatty acid (VFA) levels to assess the metabolic activity of gut microbes. The OA product used was a proprietary blend of formic, acetic, propionic, sorbic, caprylic and capric acids. Individual OAs of this nature have demonstrated antimicrobial activity against poultry pathogens including Escherichia coli (Marounek et al. 2003), Salmonella spp. (Iba and Berchieri 1995; Van Immerseel et al. 2006), Campylobacter jejuni (Solis de los Santos et al. 2008) and Cl. perfringens (Skrivanova et al. 2005). The Lact. johnsonii strain used has been isolated from the gastrointestinal tract of broiler chickens (Vidanarachchi et al. 2006) and has demonstrated anticlostridial properties in vitro (Olnood et al. 2007). Materials and methods Birds, management and diet Twelve hundred male Cobb 500 broiler chickens (Baiada Hatchery, Kootingal, NSW, Australia) were raised in floor pens in a temperature-controlled room at the University of New England (Armidale, NSW, Australia). All procedures were approved by the Animal Ethics Committee of the University of New England. All diets were based on a standard commercial starter diet without any added antibiotics or coccidiostats (Table 1; Ridley Agriproducts, Tamworth, NSW, Australia) and met or exceeded National Research Council guidelines for broiler chickens (NRC 1994). The six experimental groups consisted of an unchallenged control, an Eimeria spp. Cl. perfringens challenged control, and Table 1 Basal diets for necrotic enteritis challenge experiment Component 13 week(starter) gkg )1 Wheat 262Æ0 214Æ0 Sorghum 350Æ25 400Æ2 Mung beans 100Æ0 100Æ0 Tallow in mixer 32Æ5 34Æ0 Sunflower meal 25Æ0 Canola meal 60Æ0 60Æ0 Cottonseed meal 50Æ0 Soya bean meal 157Æ0 81Æ5 Limestone B10 15Æ5 16Æ0 Kynofos Biofos MDCP 11Æ5 11Æ0 Salt 1Æ75 1Æ5 Sodium bicarbonate 2Æ0 2Æ0 Choline chloride 75% 0Æ6 0Æ6 DL-Methionine 2Æ1 1Æ3 L-Lysine scale 3 2Æ1 0Æ4 L-Threonine 0Æ2 Vitamin and mineral premix* 2Æ5 2Æ5 Calculated chemical composition ME, MJ kg )1 12Æ26 12Æ39 Crude protein 200Æ02 190Æ00 Crude fibre 35Æ17 43Æ14 Crude fat 52Æ16 54Æ47 Lysine 11Æ49 8Æ98 Methionine + Cysteine 8Æ32 7Æ37 Calcium 9Æ73 9Æ79 Available phosphorous 6Æ50 6Æ71 Sodium 1Æ62 1Æ65 Chloride 2Æ19 1Æ75 46 week (Finisher) gkg )1 *Vitamin and mineral premix (Ridley Agriproducts Pty Ltd., Tamworth, NSW, Australia) contained the following per kg of diet: vitamin A (as all-trans retinol; 3Æ6 mg), cholecalciferol (87Æ5 lg), vitamin E (as d-atocopherol; 29Æ8 mg), vitamin B12 (0Æ2 mg), biotin (0Æ1 mg), niacin (50 mg), vitamin K3 (2 mg), pantothenic acid (12 mg), folic acid (2 mg), thiamine (2 mg), riboflavin (6 mg), pyridoxine hydrochloride (5 mg), d-calcium pantothenate (12 mg), Mn (80 mg), Fe (60 mg), Cu (8 mg), I (1 mg), Co (0Æ3 mg) and Mo (1 mg). Eimeria spp. Cl. perfringens challenged groups treated with; antimicrobials (45 mg kg )1 ZnB and 100 mg kg )1 monensin), OA (Selacid Green Growth Ò ; Selko, Tilburg, The Netherlands 1 ; 2 kg ton )1 ), probiotic Lact. johnsonii (10 9 CFU ml )1 in phosphate buffered saline, PBS) and probiotic sham (PBS alone; n = 25 birds pen, n = 8 pens treatment). The strain, identified as Lact. johnsonii using 16S rrna sequencing (refer to Genbank accession number DQ ), was isolated from the gastrointestinal tract of broiler chickens (Vidanarachchi et al. 2006; Olnood et al. 2007). The Lact. johnsonii and sham groups were orally administered 0Æ5 ml of solution using a crop needle on day 1, and 1 ml on days 3, 7 and 12. All birds 1 Selacid Green Growth Ò organic acid blend contained formic, acetic, propionic, sorbic, caprylic and capric acids Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)

3 Necrotic enteritis in broilers except for the unchallenged controls were challenged with Eimeria spp. Cl. perfringens to induce gut damage. Unchallenged controls were separated from challenged birds within the holding facility, to minimize the likelihood of cross-contamination. cross-contamination. In addition, staff entering these pens wore disposable over-boots or clean boots washed in a disinfectant bath kept adjacent to the pen door. Feed was withdrawn from all pens for 3 h prior to commencement of inoculation. NE challenge procedure The NE challenge was performed based on the procedures outlined previously (Kocher et al. 2004; Mikkelsen et al. 2009), with minor modifications as stated later. Feed Upon arrival at the facility until day 7, birds were fed starter diets with their corresponding dietary additive included. Between days 8 and 15 inclusive (prior to inoculation with Cl. perfringens), all birds were fed a highprotein diet based on 50% (w w) fish meal and 50% starter diet (with corresponding dietary additives included). After day 15, the original diets were returned until day 21. The starter feeds were replaced by finisher feed on day 21, and remaining birds were kept until day 28. Eimeria inoculation On day 9, all birds (except unchallenged controls) were given a suspension of 2500 oocysts of Eimeria acervulina, Eimeria maxima and Eimeria tenella (Bioproperties Pty Ltd, Glenorie, NSW, Australia) in 1 ml PBS. The three species of Eimeria had been purified by serial passages through 3-week-old Eimeria-free chickens, and the sporulated oocysts were stored in 2% (w v) potassium dichromate at 10 C before inoculation. Unchallenged control birds received sterile PBS in place of Eimeria. Clostridium perfringens inoculation A primary poultry isolate of Cl. perfringens type A (EHE- NE18) (Sheedy et al. 2004) was obtained from CSIRO Australian Animal Health Laboratory (Geelong, Vic., Australia) and maintained in thioglycollate broth (USP alternative, Oxoid, Hampshire, UK) with 30% (v v) glycerol at )20 C. The challenge inocula was prepared fresh by growing the bacterium overnight at 39 C in 1 l of thioglycollate broth with added starch (10 g l )1 ) and casitone (5 g l )1 ). The stock culture of Cl. perfringens was subcultured earlier in cooked meat media (Oxoid) and thioglycollate broth. On day 15, birds in challenged groups were individually inoculated with 1 ml of Cl. perfringens suspended in thioglycollate broth at a concentration of 3Æ CFU ml )1. Unchallenged control birds received 1 ml of sterile thioglycollate broth. Further inoculations on days 16 and 17 were cancelled because of onset of NErelated mortality on days 15 and 16. Unchallenged birds were always serviced first to reduce the likelihood of Bird performance All birds were examined twice daily, and dead chickens were immediately collected for postmortem analysis, and body weight was recorded. Live weight and feed consumption were recorded on days 0, 9, 14, 21 and 28, and feed conversion ratios (FCR) were calculated for each pen (FCR = feed eaten (live bird weight + dead bird weight initial bird weight)). Tissue collection On days 15 (prior to inoculation) and 18, two chickens per pen (n = 16 birds treatment) were randomly selected for lesion scoring and clostridia enumeration. The small intestine from each bird was incised longitudinally and examined for evidence of gross necrotic lesions. Small intestinal lesions were scored according to the previously described criteria (Prescott et al. 1978). On day 18, from two birds per pen (n = 16 birds treatment), a segment (3 cm) of tissue and associated digesta was collected from the midpoint of the jejunum and stored at 4 C until later frozen for microbial profiling. In addition, from 72 birds (n = 12 birds treatment), the contents of one caecum were collected, and H 2 SO 4 (0Æ5 mol l )1 ) was added at a rate of 0Æ5 mlg )1 of caecal content and stored at 4 C until later frozen for VFA analysis. Enumeration of Clostridium perfringens Clostridia were enumerated on days 15 (prior to Cl. perfringens inoculation) and 18. Enumeration was carried out on jejunal digesta collected from the two birds per pen that were selected for macroscopic lesion scoring (n = 16 birds treatment). Fresh digesta samples (1 g) were transferred into 15-ml MacCartney bottles containing 10 ml of a prereduced salt medium (Holdeman et al. 1977). The suspension was homogenized for 2 min in CO 2 -flushed plastic bags using a stomacher laboratory blender (Interscience, St. Norm, France) and serially diluted in 10-fold increments in prereduced salt medium as described previously (Miller and Wolin 1974). A subsample of the homogenized suspension (1 ml) was then transferred into 9 ml of anaerobic broth, serially diluted, and samples (0Æ1 ml) plated on tryptose-sulfite-cycloserine agar (Oxoid) for enumeration of microbial populations. All plates were incubated in an anaerobic cabinet (Model Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)

4 Necrotic enteritis in broilers SJ-3; Kalter Pty. Ltd., Edwardstown, SA, Australia), and bacterial number was counted using a colony counter (Selby, Model SCC100; Biolab Australia, Sydney, NSW, Australia). Microbial profiling Terminal-restriction fragment length polymorphism (T-RFLP) analysis was used to characterize the overall gut bacterial communities following the techniques outlined previously (Torok et al. 2008; Geier et al. 2009). Denaturing gradient gel electrophoresis (Lac PCR-DGGE) was used to identify Lactobacillus species profiles as described previously (Walter et al. 2001; Geier et al. 2009). All gels were analysed using BioNumerics ver software package (Applied Maths, Sint-Martens-Latem, Belgium). The presence absence of bands in individual bird profiles was determined using the BioNumerics band matching function. The data were downloaded into a spreadsheet and analysed statistically as outlined below. VFA analysis VFA profiles of caecal contents were analysed by gas chromatography using the previously described procedures (Storer et al. 1983). Statistical analyses Performance data were analysed with sas for Windows ver. 9.1 (SAS Institute Inc., Cary, NC, USA). Analysis of variance using the general linear model procedure was used to compare parameters with treatment differences determined by Duncan s multiple range test. Intestinal lesion scores, clostridia levels, mortalities and VFA profiles were analysed by KruskalWallis anova. Comparison of microbial communities in the jejunum (T-RFLP and Lac PCR-DGGE) was made following the techniques described previously (Torok et al. 2008; Geier et al. 2009) using the primer-6 software package (PRIMER-E Ltd, Plymouth, UK). BrayCurtis measures of similarity (Bray and Curtis 1957) were calculated to identify similarities between microbial profiles of individual birds from the T-RFLP generated OTU (Operational taxonomic unit) data (n = 28) following standardization and fourth root transformation; and the presence absence data, as scored against the reference Lactobacillus strains, from Lac PCR- DGGE analysis. One-way analysis of similarity (Clarke 1993) was performed to identify treatment differences for both T-RFLP and Lac PCR-DGGE data. Similarity percentages analyses (SIMPER) (Clarke 1993) were carried out to determine which OTU or Lactobacillus species contributed to the dissimilarity between dietary treatments from T-RFLP or Lac PCR-DGGE data, respectively. For all comparisons, P < 0Æ05 was considered significant. Results Bird performance On day 9, prior to Eimeria spp. and Cl. perfringens challenge, birds treated with Lact. johnsonii had a greater Table 2 Growth performance and bird mortality during the experimental period Unchallenged Challenged Antimicrobial Organic acid Sham Lactobacillus johnsonii P-value Weight (g) 0 37 ± 0Æ2 38 ± 0Æ4 38 ± 0Æ3 37 ± 0Æ2 37 ± 0Æ1 38 ± 0Æ3 NS ± 1 bc 192 ± 5 c 201 ± 4 abc 205 ± 2 ab 201 ± 1 abc 208 ± 2 a ** ± ± ± ± ± ± 5 NS ± 10 a 552 ± 26 b 817 ± 12 a 567 ± 20 b 580 ± 14 b 574 ± 13 b *** ± 17 a 1008 ± 46 b 1420 ± 14 a 1048 ± 30 b 1082 ± 15 b 1065 ± 21 b *** FCR (feed gain) 09 1Æ32 ± 0Æ05 1Æ46 ± 0Æ09 1Æ30 ± 0Æ05 1Æ27 ± 0Æ05 1Æ34 ± 0Æ07 1Æ23 ± 0Æ04 NS 014 1Æ18 ± 0Æ03 1Æ24 ± 0Æ05 1Æ12 ± 0Æ01 1Æ16 ± 0Æ01 1Æ20 ± 0Æ04 1Æ15 ± 0Æ01 NS 021 1Æ29 ± 0Æ01 c 1Æ57 ± 0Æ06 a 1Æ25 ± 0Æ01 c 1Æ47 ± 0Æ01 b 1Æ56 ± 0Æ04 ab 1Æ52 ± 0Æ03 ab *** 028 1Æ41 ± 0Æ02 cd 1Æ66 ± 0Æ08 a 1Æ37 ± 0Æ01 d 1Æ51 ± 0Æ01 bc 1Æ59 ± 0Æ04 ab 1Æ51 ± 0Æ03 bc *** Mortality (%) 09 0Æ5 3Æ0 3Æ0 1Æ5 2Æ0 2Æ0 NS 914 2Æ5 1Æ0 0Æ5 3Æ0 1Æ0 1Æ5 NS Æ5 a 36Æ0 b 7Æ5 a 30Æ5 b 35Æ5 b 35Æ5 b *** Æ0 0Æ0 2Æ5 1Æ0 0Æ0 1Æ0 NS Live weight data are expressed as mean (g) ± SEM. Feed conversion ratio (FCR) data are expressed as mean ± SEM. Mortality data are expressed as percentage mortality (%) for all birds within a treatment for each time period. For all data, n = 8 pens per treatment. Values within a row that do not share a common letter are significantly different (P <0Æ05). **P <0Æ01; ***P <0Æ Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)

5 Necrotic enteritis in broilers (P <0Æ05) body weight compared to birds fed basal diet (unchallenged and challenged control groups; Table 2). On days 21 and 28, challenged birds fed antimicrobials had the greatest body weights compared to all other challenged groups (P <0Æ05). The mean body weight of antimicrobial-fed birds was not different from unchallenged birds. No differences in body weight were observed amongst birds in the other challenged groups. Birds fed antimicrobials had a lower (P < 0Æ05) FCR compared to other challenged groups in the periods between days 021 and 028. Challenged birds fed OA had a lower (P <0Æ05) FCR compared to the challenged control group between days 021 and 028. In the period between days 0 and 28, birds treated with Lact. johnsonii also had a lower (P <0Æ05) FCR compared to challenged control birds. In the period following Cl. perfringens challenge (days 1421), bird mortality was significantly elevated in challenged, sham-, OA- and Lact. johnsonii-treated birds compared to unchallenged birds and challenged birds fed antimicrobials (P < 0Æ05; Table 2). Enumeration of Clostridium perfringens No differences in Cl. perfringens numbers were observed amongst treatment groups at day 15 prior to gavage with Cl. perfringens (Table 3). On day 18, challenged control birds had greater (P <0Æ05) Cl. perfringens levels compared to birds fed antimicrobials and OA. Clostridium perfringens levels were reduced (P < 0Æ05) in birds fed OA compared to birds in the Lact. johnsonii and sham groups. Lesion scores Macroscopic lesion scoring was performed on the duodenum, jejunum and ileum. No lesions were observed in the ileum of any bird on day 15 or day 18 (data not shown). On day 15, challenged controls, OA, Lact. johnsonii- and sham-treated birds had elevated (P <0Æ05) duodenal lesion scores compared to unchallenged controls and antimicrobial-fed birds (Table 3). Additionally, in the duodenum on day 15, Lact. johnsonii-treated birds had a significantly lower (P <0Æ05) lesion score compared to birds in the OA and sham groups. On day 18, duodenal lesion scores were greater (P <0Æ05) in challenged controls, sham- and OA-treated birds compared to unchallenged birds. Antimicrobial and Lact. johnsonii treatments reduced (P <0Æ05) duodenal lesion scores compared to other challenged groups, with lesion scores comparable to unchallenged controls. On day 15, jejunal lesion scores were elevated (P <0Æ05) in challenged controls, sham-, OA- and Lact. johnsonii-treated birds compared to unchallenged controls and antimicrobial-fed birds. On day 18, jejunal lesion scores were greater (P <0Æ05) in challenged, sham- and OA-treated birds compared to unchallenged controls; with a trend towards an increase in Lact. johnsonii-treated birds (P = 0Æ07). Antimicrobialfed birds had a lower (P <0Æ05) lesion score compared to challenged, sham- and OA-treated birds, with a score comparable to unchallenged control birds. Lactobacillus johnsonii-treated birds tended (P = 0Æ06) to have a lower lesion score compared to challenged controls. Microbial profiling T-RFLP Unchallenged control birds had a different (P <0Æ05) microbial profile in the jejunum compared to challenged controls (Table 4), indicating that the challenge procedure had a direct impact on the overall microbiota. Unchallenged birds also displayed a different (P <0Æ05) microbial profile compared to Eimeria spp. Cl. perfringens challenged birds treated with antimicrobials, OA and sham, with a strong trend towards a difference compared to Lact. johnsonii-treated birds (P <0Æ1). The Table 3 Intestinal macroscopic lesion scoring and clostridia enumeration prior to (day 15) and following (day 18) Clostridium perfringens challenge Unchallenged Challenged Antimicrobial Organic acid Sham Lactobacillus johnsonii P-value Clostridia (log 10 CFU g )1 digesta) Day 15 3Æ50 ± 0Æ12 3Æ81 ± 0Æ28 3Æ60 ± 0Æ20 5Æ14 ± 0Æ69 3Æ71 ± 0Æ43 3Æ47 ± 0Æ24 NS Day 18 3Æ69 ± 0Æ28 ab 4Æ23 ± 0Æ35 a 3Æ45 ± 0Æ32 b 3Æ42 ± 0Æ16 b 4Æ16 ± 0Æ22 a 4Æ44 ± 0Æ35 a * Duodenal lesions Day 15 0Æ0 ±0Æ0 a 0Æ84 ± 0Æ20 bc 0Æ0 ±0Æ0 a 1Æ13 ± 0Æ17 c 1Æ13 ± 0Æ15 c 0Æ38 ± 0Æ17 b *** Day 18 0Æ0 ±0Æ0 a 0Æ56 ± 0Æ17 b 0Æ06 ± 0Æ04 a 0Æ28 ± 0Æ10 b 0Æ53 ± 0Æ17 b 0Æ09 ± 0Æ09 a *** Jejunal lesions Day 15 0Æ0 ±0Æ0 a 0Æ31 ± 0Æ14 b 0Æ0 ±0Æ0 a 0Æ78 ± 0Æ27 b 0Æ59 ± 0Æ20 b 0Æ28 ± 0Æ13 b ** Day 18 0Æ0 ±0Æ0 a 0Æ75 ± 0Æ27 b 0Æ0 ±0Æ0 a 0Æ50 ± 0Æ18 b 0Æ38 ± 0Æ10 b 0Æ19 ± 0Æ11 ab *** Lesion data are expressed as mean (score) ± SEM. Clostridia enumeration data are expressed as mean (log 10 CFU g )1 digesta) ± SEM. Values within a row that do not share a common letter are significantly different (P <0Æ05). *P <0Æ05; **P <0Æ01; ***P <0Æ001. Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)

6 Necrotic enteritis in broilers Table 4 One-way analysis of similarities (ANOSIM) for jejunal microbial communities associated with diet Unchallenged Challenged Antimicrobial OA Sham Lactobacillus johnsonii Unchallenged 0Æ134 0Æ404 0Æ132 0Æ127 0Æ066 Challenged 0Æ012 0Æ299 0Æ004 0Æ103 )0Æ01 Antimicrobial <0Æ001 <0Æ001 0Æ457 0Æ546 0Æ374 OA 0Æ014 0Æ388 <0Æ001 0Æ060 )0Æ005 Sham 0Æ021 0Æ024 <0Æ001 0Æ084 )0Æ008 Lact. johnsonii 0Æ078 0Æ543 <0Æ001 0Æ483 0Æ513 Data are expressed as the R-statistic (bold), with significance level in italics. For all analyses, a significance level of 0Æ05 was considered significant. For jejunal microbial communities, the global R-value was 0Æ178 at a significance level of 0Æ0001 which is considered significant. OA, organic acid. microbial communities of challenged control birds were different (P <0Æ05) compared to antimicrobial and sham-treated birds; however, they were not different compared to OA- and Lact. johnsonii-treated birds (P > 0Æ05). Challenged birds fed antimicrobials had a significantly different microbial composition compared to all other treatment groups (P < 0Æ001). Birds treated with OA, Lact. johnsonii and sham were not different from one another. SIMPER analysis showed that OTU 178, 180, 188, 566 and 574 contributed significantly to the differences observed among the unchallenged control, challenged control and challenged birds treated with antimicrobials (data not shown). Figure 1 shows the contribution OTU made to the overall jejunal bacterial profile of individual chickens in these three treatment groups. Overall gut microbial community similarity within these treatment groups ranged from 59Æ3 to 62Æ6%. Despite the interbird variation observed statistical analysis showed that OTU 178 was more abundant in the challenged controls when compared to either the unchallenged control or challenged birds treated with antimicrobials, as represented in Fig. 1. OTU 180 was less abundant in the challenged birds treated with antimicrobials than the unchallenged control, while OTU 188 and 576 were more abundant in the challenged birds treated with antimicrobials than either the challenged or unchallenged controls. OTU 188 was also more abundant in the challenged control when compared to the unchallenged control group. Unchallenged control birds had a significantly greater abundance of OTU 566 than the challenged birds treated with antimicrobials. Lac PCR-DGGE The Lactobacillus profiles of the challenged control birds and the antimicrobial-fed birds were different (P < 0Æ05) compared to all other treatments (Table 5). SIMPER analysis showed that Lact. johnsonii, Lactobacillus reuteri and or Lactobacillus salivarius contributed most to the differences observed (data not shown). Of particular interest, the Lact. johnsonii band was detected in only 6 16 birds treated with Lact. johnsonii yet was present in of antimicrobial-fed birds and challenged control birds. This species was not commonly detected (< 7 birds) in each of the other three treatment groups. In contrast, almost every profile contained a band that migrated at the same position as Lactobacillus crispatus, Lactobacillus gallinarum and or Lactobacillus amylovorous. These three species belong to the Group A Lactobacillus acidophilus taxonomic group, which cannot be distinguished using Lac PCR-DGGE (Guan et al. 2003). VFA analysis Iso-valeric acid levels were greater (P <0Æ05) in challenged and sham control birds, and birds treated with OA and Lact. johnsonii compared to unchallenged controls and antimicrobial-fed birds (data not shown). The concentrations of all other measured VFAs did not differ among treatments. Discussion In the current study, an OA blend and Lact. johnsonii did not prevent the onset and reduce the severity of gut damage in chickens with experimentally induced clinical NE. Lactobacillus johnsonii appeared to improve intestinal lesion scores and FCR, whilst OA treatment successfully decreased clostridial load and improved FCR; however, these changes did not translate into improved growth or reduced mortality. Whilst the infectious challenge was quite pronounced in the current study, traditional antimicrobial treatment was able to significantly reduce damage severity and maintain performance. The capacity for OA and Lact. johnsonii to prevent Cl. perfringens-related damage in a subclinical model should also be addressed; however, if these treatments were to be a viable replacement for in-feed antimicrobials, they must be able to perform to the same level in more severe settings, such as those 1334 Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)

7 Necrotic enteritis in broilers (a) C76 C74 C73 C72 C71 C70 C27 C39 C69 C26 C30 C28 C25 C29 C24 C23 C22 C21 0 (b) OTU contribution (%) C88 C87 C78 C77 C60 C59 C50 C49 C40 C12 C02 C04 C11 C01 C03 0 (c) 100 Figure 1 OTU contribution to the overall jejunal microbial profile of individual birds within the control treatment groups. (a) Unchallenged control. (b) Challenged control. (c) Challenged birds treated with antimicrobials. Legend shows OTU (n = 26) identified in these birds ( ) 946; ( ) 944; ( ) 938; ( ).936; ( ) 912; ( ) 886; ( ) 868; ( ) 866; ( ) 576; ( ) 574; ( ) 566; ( ) 544; ( ) 518; ( ) 502; ( ) 492; ( ) 490; ( ) 450; ( ) 188; ( ) 186; ( ) 180; ( ) 178; ( ) 172; ( ) 170; ( ) 152; ( ) 80 and ( ) C90 C89 C80 C79 C62 C61 C52 C51 C42 C41 C31 C14 C13 0 Table 5 One-way analysis of similarities (ANOSIM) for jejunal Lactobacillus species associated with diet Unchallenged Challenged Antimicrobial OA Sham Lact. johnsonii Unchallenged Challenged Antimicrobial OA Sham Lactobacillus johnsonii 0Æ040 0Æ001 0Æ481 0Æ498 0Æ529 0Æ099 0Æ001 0Æ002 0Æ003 0Æ001 0Æ279 0Æ435 0Æ001 0Æ001 0Æ003 )0Æ013 0Æ275 0Æ343 0Æ643 0Æ921 )0Æ012 0Æ186 0Æ375 )0Æ028 0Æ547 )0Æ014 0Æ271 0Æ184 )0Æ052 )0Æ019 Data are expressed as the R-statistic (bold), with significance level in italics. For all analyses, a significance level of 0Æ05 was considered significant. For jejunal microbial communities, the global R-value was 0Æ145 at a significance level of 0Æ001 which is considered significant. OA, organic acid. observed in the current study. The mechanisms underlying the protective capacity of ZnB and monensin and the relative contributions of each compound to the observed protection were not investigated in this study; however, a direct antimicrobial effect may be involved. Recently, nonantimicrobial anti-inflammatory mechanisms of action of Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)

8 Necrotic enteritis in broilers in-feed antimicrobials have also been proposed (Niewold 2007). It should be noted that aspects of the NE challenge procedure, in addition to the Cl. perfringens inoculation, which were employed to predispose birds to NE and improve the infection procedure, may have contributed to any observed intestinal damage and microbial shifts. Inoculation of birds with E. tenella, E. acervulina and E. maxima may have produced significant gut damage prior to Cl. perfringens challenge. Indeed, coccidial lesions were observed in a number of the mortalities throughout the experiment. This is particularly relevant to E. acervulina and E. maxima which are known to damage the duodenum and jejunum (Lillehoj and Trout 1996); the primary sites of gut damage in the current study. Eimeria spp. are also known to influence the microbiota of broilers (Hume et al. 2006); however, the impact of Eimeria spp. alone, in the absence of Cl. perfringens, was not investigated. The observation of elevated lesion scores prior to Cl. perfringens challenge at day 15 may be indicative of Eimeria-induced damage. It is also possible that a subclinical Cl. perfringens challenge was present prior to Cl. perfringens gavage, caused by Cl. perfringens in the environment and affecting birds that were susceptible because of the earlier Eimeria challenge. The mechanism of action of OAs is believed to involve a reduction in intracellular ph via the entry of undissociated acids into the bacterial cell and subsequent dissociation in the cytoplasm (Skrivanova et al. 2005; Van Immerseel et al. 2006). Interestingly, in the current study we observed an OA-induced decrease in intestinal clostridium levels, which is consistent with the reported activity of these compounds (Skrivanova et al. 2005); however, this decrease did not translate to an improvement in performance and survival. Further investigation into the direct effects of OA on Cl. perfringens load and activity is required. Identification of the effects of OA on specific Cl. perfringens strains will be of benefit, as certain Cl. perfringens strains are known to be more pathogenic than others (Timbermont et al. 2009; Keyburn et al. 2010). The use of probiotics to prevent Cl. perfringens colonization and NE has displayed potential as strains including Bacillus subtilis PB6 (Teo and Tan 2005) and Lactobacillus spp. (Olnood et al. 2007) have been found to prevent Cl. perfringens growth in vitro. Few bacterial strains have successfully demonstrated the capacity to prevent the development of NE in vivo; however, Lact. johnsonii FI9785 has been reported to prevent Cl. perfringens colonization in pathogen-free broilers (La Ragione et al. 2004). Interestingly, the Lact. johnsonii strain used in the current study did not demonstrate any in vivo anticlostridial properties. This finding, coupled with the results of La Ragione and colleagues, may indicate strain specificity of Lact. johnsonii against Cl. perfringens. The microbial composition in the jejunum of unchallenged controls was significantly different from challenged birds; whilst birds fed antimicrobials had a unique microbiota, a finding consistent with previous studies investigating ileal communities (Feng et al. 2010). A number of key OTU underlying these shifts in microbial populations were detected. The identities of species within these OTU have not been investigated in the current study. We have used two molecular-based approaches to profile the overall microbiota (T-RFLP) and Lactobacillus species communities (Lac PCR-DGGE). These techniques have been used previously to profile the microbiota of healthy broilers (Torok et al. 2008; Geier et al. 2009); however, few studies have used these or similar techniques to comprehensively profile the microbial communities of Cl. perfringens challenged birds, with culture-based approaches more commonly used (Mikkelsen et al. 2009). The intestinal microbiota of all birds was likely to have been influenced by feeding a 50% fishmeal diet as part of the challenge procedure. High levels of dietary protein are likely to influence the bacterial communities of the intestine (Dahiya et al. 2006; McDevitt et al. 2006). In the current study, any fishmeal-induced shift would be consistent across all birds and therefore minimize any compounding effects; however, further studies in birds fed diets that are more representative of industry are required to completely and accurately understand the impact of NE on the microbiota. The treatment-induced shifts in the Lactobacillus species profiles largely reflected the overall microbial profiles as Eimeria spp. Cl. perfringens challenge and antimicrobial treatment again had a significant effect. However, whilst OA- and Lact. johnsonii-treated birds possessed an overall microbial community comparable to challenged controls, their Lactobacillus profiles were more closely related to unchallenged control birds. Previously, Cl. perfringens levels in the ileum have been found to negatively correlate with overall Lactobacillus communities and Lactobacillus aviarius numbers (Feng et al. 2010). In the current study, we did not quantify Lactobacillus abundance in the jejunum; however Lac PCR-DGGE analysis indicated a significant impact of the challenge procedure on the overall jejunal Lactobacillus community, typically associated with the presence or absence of Lact. johnsonii, Lact. salivarius and or Lact. reuteri. As such, our data support the findings of Feng et al. (2010) which indicate that Lactobacillus species are particularly susceptible to dietary change and Cl. perfringens challenge. Interestingly, Lact. johnsonii was detected in less than half of the birds treated with Lact. johnsonii. This may indicate that the 1336 Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)

9 Necrotic enteritis in broilers particular Lact. johnsonii strain, given at an early stage of the experiment by oral gavage, may not have colonized the intestine long term. The apparent inability of the Lact. johnsonii strain to colonize renders it an ineffective probiotic candidate. The overall gut microbial communities and the Lactobacillus profiles of sham-treated birds did not reflect that of challenged control birds; the reasons underlying these differences are unclear. In summary, treatment with Lact. johnsonii or OA did not prevent the onset of experimentally induced clinical NE in broilers, whilst conventional ZnB and monensin treatment prevented NE development and maintained bird performance. The unique microbial profile produced by ZnB and monensin may be linked with improved health and performance. The potential role of probiotics and OAs in the prevention of NE remains unclear; however, the observations that OA supplementation decreased intestinal clostridial load and Lact. johnsonii decreased lesion scores whilst both partially improved FCR indicate that similar compounds and organisms may indeed have the capacity to reduce the severity of NE. Acknowledgements The authors acknowledge Amit Setia, Shuyu Song, Seng Huan Chee, James Turnell, Adam Sacranie and Mark Porter for assistance with animal handling and tissue collection. We also thank Dr. Brian Siebert and Kylee Swanson for VFA analysis, Dr. Fleur Roberts for Lac PCR-DGGE analysis and Teresa Mammone for technical assistance. This project was supported by the Australian Poultry Co-operative Research Centre (CRC Project 05-2). References Bray, J.R. and Curtis, J.T. (1957) An ordination of the upland forest communities of southern Wisconsin. Ecol Monogr 27, Choct, M. and Kocher, A. (2008) Dietary factors influencing necrotic enteritis outbreaks in broilers. World s Poult Sci J 64 (Suppl.2), 159. Clarke, K.R. (1993) Nonparametric multivariate analyses of changes in community structure. Aust J Ecol 18, Cooper, K.K. and Songer, J.G. (2009) Necrotic enteritis in chickens: a paradigm of enteric infection by Clostridium perfringens type A. Anaerobe 15, Dahiya, J.P., Wilkie, D.C., Van Kessel, A.G. and Drew, M.D. (2006) Potential strategies for controlling necrotic enteritis in broiler chickens in post-antibiotic era. Anim Feed Sci Technol 129, Feng, Y., Gong, J., Yu, H., Jin, Y., Zhu, J. and Han, Y. (2010) Identification of changes in the composition of ileal bacterial microbiota of broiler chickens infected with Clostridium perfringens. Vet Microbiol 140, Geier, M.S., Torok, V.A., Allison, G.E., Ophel-Keller, K. and Hughes, R.J. (2009) Indigestible carbohydrates alter the intestinal microbiota but do not influence the performance of broiler chickens. J Appl Microbiol 106, Gholamiandehkordi, A.R., Timbermont, L., Lanckriet, A., Van Den Broeck, W., Pedersen, K., Dewulf, J., Pasmans, F., Haesebrouck, F. et al. (2007) Quantification of gut lesions in a subclinical necrotic enteritis model. Avian Pathol 36, Guan, L.L., Hagen, K.E., Tannock, G.W., Korver, D.R., Fasenko, G.M. and Allison, G.E. (2003) Detection and identification of Lactobacillus species in crops of broilers of different ages by using PCR-denaturing gradient gel electrophoresis and amplified ribosomal DNA restriction analysis. Appl Environ Microbiol 69, Holdeman, L.V., Cato, E.P. and Moore, E.C. (1977) Anaerobic Laboratory Manual. Blacksburg: Virginia Polytechnic Institute and State University. Hume, M.E., Clemente-Hernandez, S. and Oviedo-Rondnt, E.O. (2006) Effects of feed additives and mixed Eimeria species infection on intestinal microbial ecology of broilers. Poult Sci 85, Iba, A.M. and Berchieri, A. Jr (1995) Studies on the use of a formic acid-propionic acid mixture (Bio-add) to control experimental Salmonella infection in broiler chickens. Avian Pathol 24, Keyburn, A.L., Boyce, J.D., Vaz, P., Bannam, T.L., Ford, M.E., Parker, D., Di Rubbo, A., Rood, J.I. et al. (2008) NetB, a new toxin that is associated with avian necrotic enteritis caused by Clostridium perfringens. PLoS Pathog 4, e26. Keyburn, A.L., Yan, X.X., Bannam, T.L., Van Immerseel, F., Rood, J.I. and Moore, R.J. (2010) Association between avian necrotic enteritis and Clostridium perfringens strains expressing NetB toxin. Vet Res 41, 21. Kocher, A., Choct, M., Teo, A., Tan, H.M. and Carter, R.R. (2004) Effectiveness of alternative feed supplements to broiler diets using a necrotis enteritis challenge model. Aust Poult Sci Symp 16, La Ragione, R.M., Narbad, A., Gasson, M.J. and Woodward, M.J. (2004) In vivo characterization of Lactobacillus johnsonii FI9785 for use as a defined competitive exclusion agent against bacterial pathogens in poultry. Lett Appl Microbiol 38, Lillehoj, H.S. and Trout, J.M. (1996) Avian gut-associated lymphoid tissues and intestinal immune responses to Eimeria parasites. Clin Microbiol Rev 9, Marounek, M., Skrivanova, E. and Rada, V. (2003) Susceptibility of Escherichia coli to C2-C18 fatty acids. Folia Microbiol (Praha) 48, McDevitt, R.M., Brooker, J.D., Acamovic, T. and Sparks, N.H.C. (2006) Necrotic enteritis; a continuing challenge for the poultry industry. World s Poult Sci J 62, Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)

10 Necrotic enteritis in broilers Mikkelsen, L.L., Vidanarachchi, J.K., Olnood, C.G., Bao, Y.M., Selle, P.H. and Choct, M. (2009) Effect of potassium diformate on growth performance and gut microbiota in broiler chickens challenged with necrotic enteritis. Br Poult Sci 50, Miller, T.L. and Wolin, M.J. (1974) A serum bottle modification of the Hungate technique for cultivating obligate anaerobes. Appl Microbiol 27, Niewold, T.A. (2007) The nonantibiotic anti-inflammatory effect of antimicrobial growth promoters, the real mode of action? A hypothesis Poult Sci 86, NRC (1994) Nutrient Requirements of Poultry, 9th rev. edn. Washington DC: Natl. Acad. Press. Olnood, C.G., Mikkelsen, L.L., Choct, M. and Iji, P.A. (2007) Antagonistic activity of novel probiotics and their effect on growth performance of broiler chickens. Aust Poult Sci Symp 19, Prescott, J.F., Sivendra, R. and Barnum, D.A. (1978) The use of bacitracin in the prevention and treatment of experimentally-induced necrotic enteritis in the chicken. Can Vet J 19, Sheedy, S.A., Ingham, A.B., Rood, J.I. and Moore, R.J. (2004) Highly conserved alpha-toxin sequences of avian isolates of Clostridium perfringens. J Clin Microbiol 42, Skrivanova, E., Marounek, M., Dlouha, G. and Kanka, J. (2005) Susceptibility of Clostridium perfringens to C2-C18 fatty acids. Lett Appl Microbiol 41, Solis de los Santos, F., Donoghue, A.M., Venkitanarayanan, K., Reyes-Herrera, I., Metcalf, J.H., Dirain, M.L., Aguiar, V.F., Blore, P.J. et al. (2008) Therapeutic supplementation of caprylic acid in feed reduces Campylobacter jejuni colonization in broiler chicks. Appl Environ Microbiol 74, Storer, G.B., Trimble, R.P., Illman, R.J., Snoswell, A.M. and Topping, D.L. (1983) Effects of dietary oat bran and diabetes on plasma and cecal volatile fatty-acids in the rat. Nutr Res 3, Stringfellow, K., McReynolds, J., Lee, J., Byrd, J., Nisbet, D. and Farnell, M. (2009) Effect of bismuth citrate, lactose, and organic acid on necrotic enteritis in broilers. Poult Sci 88, Teo, A.Y.L. and Tan, H.M. (2005) Inhibition of Clostridium perfringens by a novel strain of Bacilius subtilis isolated from the gastrointestinal tracts of healthy chickens. Appl Environ Microbiol 71, Timbermont, L., Lanckriet, A., Pasmans, F., Haesebrouck, F., Ducatelle, R. and Van Immerseel, F. (2009) Intra-species growth-inhibition by Clostridium perfringens is a possible virulence trait in necrotic enteritis in broilers. Vet Microbiol 137, Torok, V.A., Ophel-Keller, K., Loo, M. and Hughes, R.J. (2008) Application of methods for identifying broiler chicken gut bacterial species linked with increased energy metabolism. Appl Environ Microbiol 74, Van Immerseel, F., De Buck, J., Pasmans, F., Huyghebaert, G., Haesebrouck, F. and Ducatelle, R. (2004) Clostridium perfringens in poultry: an emerging threat for animal and public health. Avian Pathol 33, Van Immerseel, F., Russell, J.B., Flythe, M.D., Gantois, I., Timbermont, L., Pasmans, F., Haesebrouck, F. and Ducatelle, R. (2006) The use of organic acids to combat Salmonella in poultry: a mechanistic explanation of the efficacy. Avian Pathol 35, Van Immerseel, F., Rood, J.I., Moore, R.J. and Titball, R.W. (2009) Rethinking our understanding of the pathogenesis of necrotic enteritis in chickens. Trends Microbiol 17, Vidanarachchi, J.K., Mikkelsen, L.L., Sims, I.M., Iji, P.A. and Choct, M. (2006) Selected plant extracts modulate the gut microflora in broilers. Aust Poult Sci Symp 18, Walter, J., Hertel, C., Tannock, G.W., Lis, C.M., Munro, K. and Hammes, W.P. (2001) Detection of Lactobacillus, Pediococcus, Leuconostoc, and Weissella species in human feces by using group-specific PCR primers and denaturing gradient gel electrophoresis. Appl Environ Microbiol 67, Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 109 (2010)