Effects of Dietary Protein Source and Level on Intestinal Populations of Clostridium perfringens in Broiler Chickens

Transcription

1 Effects of Dietary Protein Source and Level on Intestinal Populations of Clostridium perfringens in Broiler Chickens M. D. Drew, 1 N. A. Syed, B. G. Goldade, B. Laarveld, and A. G. Van Kessel Department of Animal and Poultry Science, University of Saskatchewan, Saskatoon, SK, Canada S7N 5A8 ABSTRACT Two experiments were conducted to examine SPC-based diets. In experiment 2, the dietary treatments the effect of the level of dietary crude protein and protein source on intestinal populations of Clostridium perfringens in broilers. In experiment 1, 6 groups of 12 birds were fed diets containing 230, 315 or 400 g/kg crude protein with soy protein concentrate (SPC) or low-temperature-dried fishmeal as the major protein sources in a2 3 factorial arrangement of treatments. A significant interaction between protein source and level was observed where the number of C. perfringens present in the ileum and cecum increased as the level of crude protein in the diets increased from 230 to 400 g/kg in the birds fed fishmeal-based diets (P < 0.05) but not in the birds fed used were arranged in a 2 2 factorial arrangement with 2 levels of crude protein (230 and 400 g/kg) and 2 protein sources (SPC or fishmeal). The main effects of protein source and protein level significantly (P < 0.05) affected numbers of C. perfringens without interaction. Amino acid analysis of the diets showed that the glycine and methionine contents of the fishmeal diets were elevated compared with the SPC diets. This suggests that the level of crude protein, protein source, and amino acid content of diets affect the growth of C. perfringens in the lower intestinal tract of the broiler chicken and might be predisposing factors to outbreaks of clinical necrotic enteritis. (Key words: broiler, Clostridium perfringens, dietary protein, necrotic enteritis) 2004 Poultry Science 83: INTRODUCTION Necrotic enteritis (NE) is a common disease in broiler chickens throughout the world. The disease usually occurs in broiler chickens 2 to 6 wk after hatching and is caused by the overgrowth of Clostridium perfringens in the small intestine (Dykstra and Reid, 1977; Fukata et al., 1991). Normally the number of C. perfringens in the intestine is low (~10 4 cfu/g of digesta) but under certain circumstances C. perfringens may multiply, increasing bacterial numbers to 10 7 to 10 9 cfu/g of digesta, and cause clinical disease. Although C. perfringens is recognized as the etiologic agent of NE, other co-factors are usually required to precipitate an outbreak of the disease including: coccidiosis infection, environment, climate, management of hygiene, and diet (Elwinger et al., 1992, Kalhusdal and Skjerve, 1996; Estrada and Wilkie, 2000; Kalhusdal, 2000). Although NE has been reproduced by the oral inoculation of C. perfringens alone (Truscott and Al-Sheikhly, 1977), consistent results have been difficult to obtain. Successful 2004 Poultry Science Association, Inc. Received for publication July 29, Accepted for publication October 13, To whom correspondence should be addressed: drew@sask. usask.ca. induction of the disease usually requires previous intestinal damage, particularly sloughing of intestinal epithelial cells and leakage of protein into the intestinal lumen. This initial step of intestinal damage sets up a cascade of physiological changes which result in an intestinal environment conducive to the overgrowth of C. perfringens and increased production of α-toxin in the intestine. In commercial production, coccidiosis is an important predisposing factor for triggering outbreaks of NE, and feeding of coccidiostats has been shown to reduce the severity of NE in broiler flocks (Elwinger et al., 1994). Experimental induction of intestinal damage to cause NE in broilers has been successfully accomplished by co-infection with Eimeria spp. Shane et al. (1985) showed that infection of broiler chickens with Eimeria acervulina prior to inoculation with C. perfringens increased mortality from 8% in Eimeria-free controls to 35% in Eimeria-infected birds. This was associated with destruction and sloughing of the intestinal epithelium, a 39% increase in intestinal passage time, reduced intestinal ph, and depressed serum protein. There are 2 major dietary factors that predispose broiler chickens to NE. The first factor is cereal grains that increase the viscosity of digesta, including wheat and bar- Abbreviation Key: NE = necrotic enteritis; SPC = soy protein concentrate. 414

2 PROTEIN EFFECT ON C. PERFRINGENS IN BROILERS 415 TABLE 1. Diet formulations used in experiment 1 Ingredient (g/kg) Fishmeal Corn Wheat Soy protein concentrate Limestone Canola oil Dicalcium phosphate Choline chloride Vitamin/mineral premix Met Supplied per kilogram of diet: vitamin A, 11,000 IU; cholecalciferol, 2,200 IU; vitamin E, 30 IU; vitamin K, 0.5 mg; vitamin B 12 ; 0.02 mg; thiamine, 1.5 mg; riboflavin, 6 mg; folic acid, 0.6 mg; biotin, 0.15 mg; niacin, 60 mg; pyridoxine, 5 mg; chloride, 788 mg; sodium, 511 mg; iron, 80 mg; manganese, 21.8 mg; selenium, 0.1 mg; iodine, 0.35 mg; zinc, 100 mg. ley. Kaldhusdal and Skjerve (1996) studied the association of the cereal content of diets fed to broiler chickens in Norway from 1969 to 1989 and demonstrated that 2 major outbreaks of NE were associated with increased use of barley and wheat in broiler diets. The use of corn was negatively associated with the incidence of NE. The second dietary factor predisposing broilers to NE is diets with high percentages of proteins from animal sources (Kaldhusdal and Skjerve, 1996). Recent studies (Baker and Han, 1994; Mack et al., 1999; Hoehler et al. 2001) suggest that for diets which are balanced according to ideal protein ratios, the CP and amino acid requirements for broiler chickens to maximize growth performance and efficiency may be higher than present NRC requirements (NRC, 1994). Thus, an understanding of the effects of both dietary protein source and level on intestinal populations of C. perfringens is important to developing dietary formulations for broiler chickens that may reduce the risk of an outbreak of clinical NE, particularly in light of the decreased acceptability of antibiotics in poultry feed. The purpose of the current study was to examine the effects of protein source (animal vs. plant protein) and protein concentration on intestinal populations of C. perfringens in broiler chickens and improve our understanding of which factors present in these proteins are associated with increased numbers of C. perfringens. Fishmeal and soy protein concentrate (SPC) were the animal and plant protein sources, respectively, used in the study at CP levels ranging from 230 to 400 g/kg. Experiment 1 MATERIALS AND METHODS Experimental protocols were approved by the Animal Care Committee of the University of Saskatchewan and were performed in accordance with recommendations of 2 Soycomil K, ADM Protein Specialty Division, Decatur, IL. 3 EWOS Canada Ltd, Surrey, BC, Canada. 4 Degussa, Allendale, NJ. the Canadian Council on Animal Care as specified in the Guide to the Care and Use of Experimental Animals (1993). Seventy-two, 1-d-old, mixed-sex broiler chickens (Hubbard Hubbard) were placed in 6 floor pens (12 birds per pen) with clean straw flooring and were fed and watered ad libitum throughout the experiment. The birds were wing-banded for individual identification and were not vaccinated. The birds received an unmedicated commercial broiler diet for the first 14 d of the experiment. From d 14 to 28, the birds were fed the experimental diets (Table 1). The diets were formulated to contain 230, 315, or 400 g/kg CP and were based on SPC 2 or low-temperaturedried aquaculture grade fishmeal. 3 The diets were isocaloric and met or exceeded the NRC requirements for broiler chickens (NRC, 1994). The experimental diets did not contain antibiotics or coccidiostats and were not pelleted. The feeds were analyzed for their amino acid content. 4 Experiment 2 Forty-eight, 1-d-old, mixed-sex broiler chickens (Hubbard Hubbard) were placed in 4 floor pens (12 birds per pen) with clean straw flooring and were fed and watered ad libitum throughout the experiment. The birds were wing-banded for individual identification and were not vaccinated. The birds received an unmedicated commercial broiler diet for the first 14 d of the experiment. From d 14 to 28, the birds were fed the experimental diets shown in Table 2. The dietary treatments were arranged in a 2 2 factorial arrangement with 2 levels of CP (230 and 400 g/kg) and 2 protein sources (SPC 2 or fishmeal 3 ). The diets were isocaloric and met or exceeded the NRC requirements for broiler chickens (NRC, 1994). The experimental diets did not contain antibiotics or coccidiostats and were not pelleted. The feeds were analyzed for their amino acid content. 4 C. perfringens Inoculation In both experiments, the feed was inoculated with a lyophilized clinical isolate of C. perfringens. Previous stud-

3 416 DREW ET AL. TABLE 2. Diet formulations used in experiment 2 Ingredient (g/kg) Fishmeal Corn Wheat Soy protein concentrate Limestone Canola oil Dicalcium phosphate Choline chloride Vitamin/mineral premix Methionine Supplied per kilogram of diet: vitamin A, 11,000 IU; cholecalciferol, 2,200 IU; vitamin E, 30 IU; vitamin K, 0.5 mg; vitamin B 12, 0.02 mg; thiamine, 1.5 mg; riboflavin, 6 mg; folic acid, 0.6 mg; biotin, 0.15 mg; niacin, 60 mg; pyridoxine, 5 mg; chloride, 788 mg; sodium, 511 mg; iron, 80 mg; manganese, 21.8 mg; selenium, 0.1 mg; iodine, 0.35 mg; zinc, 100 mg. ies in our laboratory showed that only a small percentage of broiler chickens in our experimental facility had measurable C. perfringens present in their intestinal tracts unless inoculated with a viable culture of these bacteria. However, the purpose of the inoculation was only to establish populations of C. perfringens and then allow dietary effects to affect the size of these populations in the week following inoculation. The recovery rate of viable bacteria from the lyophilized culture used for inoculation was approximately cfu/g. The dried culture was added daily to fresh feed and mixed by hand to provide a final inoculum of cfu/g of feed. The inoculated feed was fed ad libitum from d 14 to 21 to ensure intestinal colonization by C. perfringens. After this period the feeders were cleaned and uninoculated feed was provided from d21to28. Enumeration of Intestinal Bacteria At 28 d of age, birds were killed by cervical dislocation and weighed, and their intestinal tracts were removed. Samples of fresh digesta (0.1 to 0.2 g) from the cecum and ileum (1 cm proximal to the ileocecal junction) were taken aseptically by cutting the ileal or cecal wall and using a sterile spatula to transfer pea-sized samples into preweighed 15-mL sterile plastic tubes containing 5 g/l cysteine hydrochloride. 5 The samples consisted of luminal contents only. The samples were kept on ice until plated within 3 h of collection. The samples were weighed and diluted in peptone water to an initial 10 1 dilution. Ten-fold dilutions were spread in duplicate using an automated spiral plater 6 on BBL blood agar base 7 containing 5% sheep blood and 100 mg/l neomycin. 8 The plates were incubated anaerobically at 37 C for 48 h. α- and β- hemolytic colonies, microscopically confirmed as gram- 5 Sigma Chemical Co., St. Louis, MO. 6 Autoplate, Spiral Biotech Inc., Bethesda, MD. 7 VWR International, Mississauga, ON, Canada. 8 The Upjohn Company, Orangeville, ON, Canada. 9 Version , SPSS Inc., Chicago, IL. negative rods, were counted as C. perfringens. Counts were expressed as the Log 10 cfu/g of intestinal contents. Statistical Analysis Individual bird weights and C. perfringens counts on d 28 were analyzed using the general linear models procedure of SPSS. 9 Experiment 1 was analyzed as a 3 2 factorial arrangement with 2 protein sources and 3 protein levels. A significant (P < 0.05) interaction between the main effects was observed in this experiment, so individual treatment means were compared using the Ryan-Einot-Gabriel-Welsch multiple F test; differences between the means were considered significant when P < Experiment 2 was analyzed as a 2 2 factorial arrangement with 2 levels of CP (230 and 400 g/kg) and 2 protein sources (SPC or fishmeal). Individual treatment means were compared using the Ryan-Einot-Gabriel-Welsch multiple F test and differences between the means were considered significant when P < Feed intakes were not measured because, with only one pen per treatment, no statistical analysis could be performed on these data. RESULTS During experiment 1, no clinical signs of NE were observed following challenge with C. perfringens, and only 1 of the 72 birds in this trial died. The d 28 BW of the chickens are shown in Table 3. The birds fed diets based on SPC had significantly lower 28 d BW at the 230 and 315 g/kg CP levels than the fishmeal-fed birds. C. perfringens counts in ileum and cecum are shown in Table 3. No C. perfringens were enumerated in the ileum from birds on the 230 and 315 g/kg CP fishmeal diets. Although all samples were handled in a similar manner, it is possible that laboratory error was the cause of this because it is unlikely that all of the birds in these 2 treatment groups had no C. perfringens present in their intestinal tracts. C. perfringens numbers increased sharply in the birds fed the 400 g/kg CP fishmeal diet. Low levels of C. perfringens were observed in the ileum of birds fed SPCbased diets at all levels of CP resulting in a significant

4 PROTEIN EFFECT ON C. PERFRINGENS IN BROILERS 417 TABLE 3. Mean Clostridium perfringens populations 1 in the ileum and cecum and body weight of birds on d 28 of experiment 1 Protein level Body weight Ileum Cecum Protein source (g/kg) (g) (cfu/g) (cfu/g) Soy protein concentrate * Soy protein concentrate * Soy protein concentrate * 4.56* Fishmeal 230 1,152* Fishmeal 315 1,088* Fishmeal * 7.20* SEM Effects P-value Protein source < < 0.01 Protein level < 0.01 Source level 0.04 < 0.01 < Means are log 10 colony-forming units counted on blood agar containing 100 mg of neomycin/l. *Protein sources at the same protein concentration are significantly different (P < 0.05). Crude protein level within the same protein source are significantly different (P < 0.05). protein source protein level interaction. Because the interaction between protein source and protein level was significant for C. perfringens numbers in both the ileum and the cecum, individual treatment comparisons were made to characterize the interaction. The number of C. perfringens present in the ileum and cecum of the birds was significantly affected by protein source only at the 400 g/kg CP level. Also, C. perfringens counts in the ileum and cecum increased significantly as CP level increased from 230 to 400 g/kg in the birds fed fishmeal-based diets but not in the birds fed SPC diets. The CP and amino acid analysis of the diets as fed is shown in Table 4. Also shown is a ratio calculated by dividing the percentage CP or amino acid present in the 400 g/kg fishmeal diet by the percentage CP or amino acid present in the 400 g/kg SPC diet. The ratio is high (greater than 1.3) for methionine, histidine, glycine, and alanine suggesting that these amino acids are in relative excess in the 400 g/kg fishmeal diet compared with the 400 g/kg SPC diet. Since the number of C. perfringens present in the ileum and cecum of the birds was significantly affected by protein source only at the 400 g/kg CP level, the 315 g/kg CP level was eliminated in experiment 2. During experiment 2, no clinical signs of NE were observed following challenge with C. perfringens and no birds died during the experiment. The 28 d BW of the birds are shown in Table 5. There was a significant interaction between protein source and protein level. As with experiment 1, the fishmeal-fed birds were heavier than the SPC-fed birds and this difference increased as the level of CP increased. The numbers of C. perfringens in the ileum and cecum of the birds on d 28 of experiment 2 are shown in Table 5. The main effects of protein source and protein level were significant without interaction. C. perfringens counts in the ileum and cecum were significantly higher in the TABLE 4. Crude protein and amino acid analysis of diets used in experiment 1 Amino acid Ratio 1 (g/kg as is) Crude Protein Met Cys Lys Thr Trp Arg Ile Leu Val His Phe Gly Ser Pro Ala Asp Glu The ratio column shown was calculated by dividing the percentage of CP or amino acid present in the 400 g/kg fishmeal diet by the percentage of CP or amino acid present in the 400 g/kg soy protein concentrate diet.

5 418 DREW ET AL. TABLE 5. Mean Clostridium perfringens populations 1 in the ileum and cecum and body weight of birds on d 28 of experiment 2 Protein level Body weight Ileum Cecum Protein source (g/kg) (g) (cfu/g) (cfu/g) Fishmeal 230 g/kg 1,064* 3.93* 4.57* Fishmeal 400 g/kg 1,125* 6.98* 7.55* Soy protein concentrate 230 g/kg 794* 1.69* 3.25* Soy protein concentrate 400 g/kg 689* 5.28* 6.36* SEM Effects P-value Protein source < 0.01 < 0.01 < 0.01 Protein level 0.37 < 0.01 < 0.01 Source level < Means are log 10 colony-forming units counted on blood agar containing 100 mg of neomycin/l. *Protein sources at the same protein concentration are significantly different (P < 0.05). Crude protein level within the same protein source are significantly different (P < 0.05). fishmeal-fed birds than in the SPC-fed birds at both CP levels. The CP and amino acid analysis of the diets as fed are shown in Table 6. Also shown is a ratio calculated by dividing the percentage CP or amino acid present in the 400 g/kg fishmeal diet by the percentage CP or amino acid present in the 400 g/kg SPC diet. The CP levels in the fishmeal diets were lower than the target values of 230 and 400 g/kg. This was due to a much lower level of CP in the fishmeal used in this experiment compared with that used in the first experiment (763 and 642 g/kg CP for experiments 1 and 2, respectively). However, the diets were formulated on the assumption that the CP of the fishmeal was the same in each experiment. The amino acid ratios for glycine (1.65) and methionine (1.35) were elevated (greater than 1.3) compared with other amino acids present in diets as fed. DISCUSSION In a longitudinal study of NE and feeding practices over a 20-year period, Kaldhusdal and Skjerve (1996) reported that the use of animal proteins in broiler diets in Norway was associated with an increased incidence of NE. This observation is based on feeding diets at or near NRC requirements. However, several recent reports suggest that when diets are balanced according to ideal protein ratios (Baker and Han, 1994; Mack et al., 1999), the CP and amino acid requirements for broiler chickens to maximize growth performance and efficiency may be higher than present NRC requirements. Hoehler et al. (2001) tested the effect of increasing dietary CP from 174 to 275 g/kg with an optimized amino acid profile in broiler chickens from 28 to 41 d of age. They showed that weight gain and feed conversion ratio improved linearly with increasing CP up to 268 g/kg dietary CP. This suggests that high protein diets balanced according to ideal protein may be commercially feasible. However, there is a potential for negative interaction between the high protein diets and NE. Thus, there is an urgent need to better understand the relationship between high CP diets and overgrowth of C. perfringens. Increasing dietary protein did not consistently affect bird growth performance in the present studies. How- TABLE 6. Crude protein and amino acid analysis of diets used in experiment 2 Amino acid Ratio 1 (g/kg as is) Crude protein Met Cys Lys Thr Trp Arg Ile Leu Val His Phe Gly Ser Pro Ala Asp Glu The ratio column shown was calculated by dividing the percentage CP or amino acid present in the 400 g/ kg fishmeal diet by the percentage CP or amino acid present in the 400 g/kg soy protein concentrate diet.

6 PROTEIN EFFECT ON C. PERFRINGENS IN BROILERS 419 ever, the diets used in these studies were not formulated to ideal protein ratios, small bird numbers were used, and the protein levels were increased to very high levels in an attempt to maximize overgrowth of C. perfringens. Initial physiological changes seen during clinical NE include decreased intestinal motility, decreased intestinal ph, and an increase in intestinal protein concentration due to malabsorption or leakage of serum proteins into the lumen of the intestine or both (Shane et al., 1985). This local increase in protein concentrations in the intestine may initiate the overgrowth of C. perfringens because the growth of this organism and the production of α- toxin, its major pathogenic factor, are influenced by the presence of amino acids (Titball et al., 1999). Methionine, while not a required nutrient for C. perfringens, is highly stimulatory to growth and is required for sporulation (Muhammed et al., 1975). Another study reported that glycine also accelerated the growth of C. perfringens (Ispolatovskaya, 1971). The α-toxin produced by C. perfringens is the major pathogenic factor in NE, and α-toxin alone is able to cause the lesions and mortality typical of NE (Al-Sheikhly and Truscott, 1977; Fukata et al., 1988). In defined media, α-toxin production requires the presence of glycine-containing peptides (Nakamura et al., 1978). Taken together, these reports suggest that an increase in the concentration of particular amino acids in the lower small intestine may be a triggering event for the overgrowth of C. perfringens and clinical NE. An increase in level of CP and the amino acid balance of the diet might therefore be contributing factors in NE. Although there are no published studies that directly examined the effect of high levels of dietary CP on the numbers of C. perfringens in the intestinal tract in chickens, several such studies have been performed using pigs and dogs. Mansson and Smith (1962) placed 8-wk-old pigs on a diet high in CP and reported that there was an increase in the number of C. perfringens type A from about cfu/g in feces from pigs on a low protein diet (200 g/kg CP) to cfu/g in the feces of pigs on a high protein diet (300 g/kg CP). In another study, Mansson and Olhagen (1967) demonstrated that when pigs were fed on a high protein diet the number of C. perfringens type A was significantly increased in the feces 5 to 10 d after initiation of the diet. More recent studies have demonstrated that feeding meat-based diets to dogs results in increased numbers of C. perfringens. Zentek et al. (1998) compared the effect of feeding a low protein diet (220 g/kg CP) to a high protein diet (630 g/kg CP) containing 510 g/kg greaves meal (beef/pork by-product) to dogs. The numbers of C. perfringens increased from 10 6 to 10 9 cfu/g of digesta in dogs fed the high protein diet while streptococci and lactobacilli were unaffected. Zentek (2000) compared the effect of feeding greaves meal, soy protein hydrolysate, or corn gluten meal to dogs in high protein diets. The numbers of C. perfringens were significantly increased in the ileal chyme of dogs fed the greaves meal diets compared with those fed the plantbased diets. The counts of other bacterial species were not significantly affected by diet. In experiments 1 and 2 we found a significant effect of both protein level and protein source on the number of C. perfringens in the cecum. This demonstrates that CP alone does not adequately explain the changes in C. perfringens populations seen in this study. Possible factors causing this effect include lipids, carbohydrates, antinutritional factors, and amino acid balance. Although further studies will be required to definitively identify which of these factors are important, the amino acid balance of fishmeal relative to soy protein concentrate provides some interesting possibilities for further investigations. In both Experiments 1 and 2, the ratio of methionine and glycine in the fishmeal diets relative to the SPC diets at the 400 g/kg CP level was greater than 1.3. Alanine and histidine had ratios greater than 1.3 in experiment 1 only. Interestingly, both methionine and glycine have been reported to stimulate the growth of C. perfringens in vitro (Ispolatovskaya, 1971; Muhammed et al., 1975; Nakamura et al., 1978). The association of methionine with high numbers of intestinal C. perfringens is of particular interest to commercial broiler production. Methionine is the first limiting amino acid in most broiler diets. It is supplemented by the addition of DL-methionine or the hydroxy-analogue of methionine, 2-hydroxy-4-methylthiobutanoic acid. Several studies have shown that 10 to 20% analogue present in feed is not absorbed by broilers and is present in the distal sections of the small intestine compared to 4 to 5% for DL-methionine (Lingens and Molnar, 1996; Maenz and Engele-Schaan, 1996; Drew et al., 2003). Since the analogue enters the lower intestinal tract, it may have an effect on C. perfringens populations. Control of NE in commercial broiler production is well managed by the use of antibiotic growth promoters such as bacitracin, lincomycin, and virginiamycin in the feed (Feed Additive Compendium, 1996). Despite the undeniable benefits of antibiotics in animal feed, concerns by consumers, researchers, and governments are growing over their use. The major concern of using antibiotics in animal feed is the potential for generating antibioticresistant strains of bacteria that may cause human disease. At present there is a heated debate over the actual danger that feed antibiotics pose to human health but the potential danger has already started sweeping changes in the use of antibiotics particularly in Europe. The European Union has already banned the use of many antibiotics in animal feeds and may completely ban their use by 2005 or earlier. This has had a significant effect on the incidence of NE in Europe. For example, the incidence of NE in France increased from 4% in 1995 to 12.4% of reported diseases in 1999 (Drouin, 1999), and similar increases have been reported in other countries in Europe. Although an outright ban of antibiotics has not occurred in North America, consumer concern is creating pressure on producers to limit their use. As a result there is a pressing need to better understand factors that predispose birds to NE and to develop alternative management and dietary strategies to control the incidence and severity of infection.

7 420 DREW ET AL. The present study has demonstrated that the level and source of dietary protein have significant effects on intestinal populations of C. perfringens in broiler chickens. Further studies in this area will allow identification of specific dietary components contributing to C. perfringens overgrowth and will improve our ability to formulate diets that decrease the likelihood of clinical outbreaks of NE in commercial broiler production. ACKNOWLEDGMENTS The authors acknowledge the support of the Agricultural Development Fund of Saskatchewan. REFERENCES Al-Sheikhly, F., and R. Truscott The pathology of necrotic enteritis of chickens following infusion of crude toxins of Clostridium perfringens into the duodenum. Avian Dis. 21: Baker, D. H., and Y. Han Ideal amino acid profile for broiler chickens during the first 3 weeks posthatching. Poult. Sci. 73: Canadian Council on Animal Care Guide to the Care and Use of Experimental Animals. Vol. 1. Canadian Council on Animal Care, Ottawa, ON. Drew, M. D., A. G. Van Kessel, and D. D. Maenz Absorption of methionine and 2-hydroxy-4-methylthiobutoanic acid in conventional and germ-free chickens. Poult. Sci 82: Drouin, P Retour en force de l enterite necrotique. Filieres Avicoles 1999: Dykstra, D. D., and W. M. Reid Monensin, Eimeria tenella infection and effects on the bacterial populations in the ceca of gnotobiotic chickens. Poult. Sci. 57: Elwinger, K., B. Engstrom, O. Fossum, S. Hassan, and B. Teglof Effect of coccidiostats on necrotic enteritis and performance in broiler chickens. Swed. J. Agric. Res. 24: Elwinger, K., C. Schneitz, E. Berndtson, O. Fossum, B. Teglof, and B. Engstrom Factors affecting the incidence of necrotic enteritis, caecal carriage of Clostridium perfringens and bird performance in broiler chicks. Acta Vet. Scand. 33: Estrada, A., and D. Wilkie Influence of diet, cold stress, immunosuppression and intestinal microflora on the development of NE in chickens. Page 125 in Proceedings of the XXI World s Poultry Congress, Montreal, QC. Feed Additive Compendium Vol. 34, no. 12. Miller Publishing, Minnetonka, MN. Fukata, T., Y. Hadate, E. Baba, and A. Arakawa Influence of bacteria on Clostridium perfringens infection in young chickens. Avian Dis. 35: Fukata, T., Y. Hadate, E. Baba, T. Uemura, and A. Arakawa Influence of Clostridium perfringens and its toxin in germ-free chickens. Res. Vet. Sci. 44: Hoehler, D., M. Pack, P. J. A. Wijtten, R. Prak, and D. J. Langhout Effects of varying dietary protein levels based on the ideal amino acid profile on broiler performance. Page 4 in Proceedings of the International Poultry Scientific Forum, Atlanta, GA. Ispolatovskaya, M. V Type A Clostridium perfringens toxin. Pages in Microbial Toxins. Vol. 2. A. S. Kadis, T. C. Montie and S. J. Ajl, ed. Academic Press, New York. Kaldhusdal, M Necrotic enteritis as affected by dietary ingredients. World s Poult. 16: Kaldhusdal, M., and E. Skjerve Association between cereal contents in the diet and incidence of NE in broiler chickens in Norway. Prev. Vet. Med. 28:1 16. Lingens, G., and S. Molnar Studies on metabolism of broilers by using 14 C-labelled DL-methionine and DL-methionine hydroxy analogue Ca-salt. Archiv Tierernahrung 49: Mack, S., D. Bercovici, G. De Groote, B. LeClercq, M. Lippens, M. Pack, J. B. Schutte, and S. Van Cauwenberghe Ideal amino acid profile and dietary lysine specification for broiler chickens of 20 to 40 days of age. Br. Poult. Sci. 40: Maenz, D. D., and C. M. Engele-Schaan Methionine and 2-hydroxy-4-methylthiobutanoic acid are transported by distinct Na+-dependent and H+-dependent systems in the brush border membrane of the chick intestinal epithelium. J. Nutr. 126: Mansson, I., and B. Olhagen Intestinal Clostridium perfringens in arthritis and parakeratosis induced by dietary factors. Experimental studies in pigs. Bull. Off. Int. Epiz. 67: Mansson, I., and L. D. S. Smith Atypical strains of Clostridium perfringens from swine. Acta Pathol. Microbiol. Scand. 55: Muhammed, S. I., S. M. Morrison, and W. L. Boyd Nutritional requirements for growth and sporulation of Clostridium perfringens. J. Appl. Bacteriol. 3: Nakamura, M., J. Cook, and W. Cross Lecithinase production by Clostridium perfringens in chemically defined media. Appl. Microbiol. 16: National Research Council Nutrient Requirements of Poultry. 9th rev. ed. National Academy Press, Washington, DC. Shane, S. M., J. E. Gyimah, K. S. Harrington, and T. G. Snider Etiology and pathogenesis of necrotic enteritis. Vet. Res. Comm. 4: Titball, R. W., C. E. Naylor, and A. K. Basak The Clostridium perfringens α-toxin. Anaerobe 5: Truscott, R. B., and F. Al-Sheikhly Reproduction and treatment of NE in broilers. Am. J. Vet. Res. 38: Zentek, J Bakterienflora des caninen intestinaltrakts: Physiologie, futterungseinflusse und diatetische konsequenzen. Kleintierpraxis 45: Zentek, J., I. van der Steen, J. Rohde, and G. Amtsberg Dietary effects on the occurrence and enterotoxin production of Clostridium perfringens in the canine gastrointestinal tract. J. Anim. Physiol. 80: