The sudden cessation in the supply of passive immunity from sow’s milk and an immature immune system make piglets vulnerable to disease (Deprez et al., 1986). In addition, the severe reduction in feed intake associated with weaning affects the integrity of the intestinal mucosa and facilitates the occurrence of pathological disorders (Pluske et al., 1997; Spreeuwenberg et al., 2001). The use of antimicrobial preventive medication at weaning has been advantageous to piglets, but there is a concern that antimicrobial use may cause the microflora of the animals to become antibiotic resistant, and this resistance could be transmitted to human pathogens (Barton, 2000). Banning of use of antibiotics as growth promoters has already begun, as has the research for new alternatives (Brufau, 2000).
The advantages of spray-dried animal plasma (SDAP) in improving the performance of weanling pigs have been demonstrated in many studies over the last decade (reviewed by van Dijk et al., 2001). Spray-dried animal plasma may offer immune protection through its immunoglobulin fraction (Gatnau et al., 1995) or by preventing the adhesion of pathogenic bacteria to the gastrointestinal mucosa due to the presence of its glycoproteins (Nollet et al., 1999). Because piglets kept in challenging environments gave a higher response to SDAP (Coffey and Cromwell, 1995), it may be an alternative to antimicrobial medication. Previously, we observed similar responses in performance to the use of SDAP and to colistin as antimicrobial medication (Torrallardona et al., 2002). The aim of this study was to test SDAP as an alternative to preventive medication with colistin in weanling pigs experimentally infected with Escherichia coli K99 in order to study its effect on performance, morphology of the small intestine, and microbiology of ileal and cecal contents.
Materials and Methods
The experimental procedures with animals described in this study were conducted after approval from the Institut de Recerca i Tecnologia Agroalimentàries (IRTA) Ethical Committee on Animal Experimentation.
Animals and Housing
Forty-eight piglets (Landrace × Landrace; mixed males and females) from IRTA’s closed herd in the “El Prat” experimental station were used. The pigs were weaned at 24 d of age with an average initial live BW of 7.1 kg (SD 1.08). The pigs were housed in a weaning room equipped with automatic heating, forced ventilation and completely slatted floors. The temperature at the start of the trial was set at 28°C and was gradually reduced to 25°C by d 14. Feed was offered in feeding hoppers with four eating spaces, and continuous access to feed and water was allowed throughout the trial.
Experimental Design and Dietary Treatments
At the start of the trial, the animals were distributed by initial BW into three blocks of 16 animals, without taking into account the gender of the animals. The 48 animals were given an oral dose of 5 mL of a solution with 107 cfu/mL of E. coli K99 and were allocated into 12 pens according to a randomized complete block design with three blocks of four pens each. The distribution of the animals took litter origin into account. Thus, within a block, an equal number of piglets from a given litter occurred in the four pens. Each pen within a block was randomly assigned to one of four experimental treatments following a 2 × 2 factorial arrangement with two dietary levels of SDAP (0 or 7%) and two dietary levels of colistin (0 or 300 mg/kg). The diets were formulated to meet or exceed the requirement estimates (NRC, 1998), and they were calculated to contain 15 g of total lysine and 14.1 MJ of ME/kg (Tables 1 and 2). In the diets without SDAP, an isonitrogenous amount of fishmeal was used. The amino acid and energy contents of all diets were adjusted with crystalline amino acids and lard.
The pigs were individually weighed at the start of the trial and on d 7 and 14. Feed intake of each pen was measured between d 0 and 7 and between d 8 and 14.
On d 7 of the trial, the second-heaviest piglet (based on initial weight) from each pen was killed with an i.v. injection of 150 to 200 mg/kg of BW of sodium pentobarbitone (Dolethal, Vétoquinol, S.A., Madrid, Spain), as were the third-heaviest piglets (based on initial weight) from each pen on d 14 (12 piglets each day). Samples of the small intestine were obtained at 2 m distal to the stomach. The samples were cut open longitudinally along the mesenteric attachment and fixed in a 10% (vol/vol) formalin solution until histological measurements were taken. The digestive contents of the ileum and the cecum were also immediately sampled and frozen at −20°C until microbiological analysis. The small intestine was emptied by gentle squeezing, rinsed with water, and dried with paper towels before its length and weight were measured.
The samples of small intestine tissue were dehydrated and embedded in paraffin wax. The samples were cut into 3-μm slices and were stained with hematoxylin and eosin before microscopic examination. From each sample, sections that ran from the tip of the villus to the base of the crypt were identified. For each piglet, the heights of the eight tallest well-oriented villi were measured from tip to crypt mouth. The depths of the adjacent crypts were measured from crypt mouth to base.
The samples of ileal and cecal digesta were cultured in different media for the determination of Lactobacillus spp. (Lactobacilli MRS agar, Difco 0882-17) for 48 h under anaerobiosis; E. coli (MacConkey agar, Merck 1.05465) for 24 h in aerobiosis; Enterococcus spp. (KF Streptococcus agar, Merck 1.10707) with the addition of 10 mL of 1% solution of 2,3,5-triphenyltetrazolium chloride for 48 h in aerobiosis; and finally Clostridium perfringens (Perfringens agar, Oxoid CM543) with selective supplements A (Oxoid SR76) and B (Oxoid SR77) for 24 h in anaerobiosis.
Growth, feed intake, and gain:feed ratio data were analyzed as a randomized complete block design with four treatments and three blocks. Treatment effect comprised the components: SDAP, colistin, and their interaction. Analysis of variance was performed using the general linear models procedure of SAS (SAS Inst., Inc., Cary, NC).
The microbiological results for lactobacilli and enterococci were analyzed as a randomized complete block design (four treatments and three blocks). The treatments were arranged in a 2 × 2 × 2 factorial, with SDAP, colistin, and day postweaning as main effects. The statistical model included the effects due to block and treatment. Treatment effect consisted of the seven following components: SDAP (P), colistin (C), days postweaning (D) and the interactions P × C, P × D, C × D, and P × C × D. The results for E. coli and C. perfringens, however, could not be analyzed by ANOVA since some treatment means were zero (i.e., no microbial counts) and this violated the assumption of equal variance among treatments. For these two microorganisms, the effects of the main factors (P, C, and D) were analyzed using the Kruskall-Wallis test in the one-way, nonparametric procedure of SAS.
Morphology of the small intestine data was analyzed using initial BW as a covariate. The treatments were arranged in a 2 × 2 × 2 factorial, with P, C, and D as main effects. The statistical model included the effects due to initial BW and treatment. As before, treatment effect was divided into seven components: P, C, D, P × C, P × D, C × D, and P × C × D.
Pens were used as the experimental unit for the analysis of growth and feed intake. For the analysis of small intestine morphology and digestive contents microbiology, the individual pigs were used as experimental unit.
The relationship between ADG and villi height was examined by linear regression with villi height as the independent variable using the Prog Reg option of SAS.
Results show that the inclusion of both SDAP and colistin significantly increased the performance of the animals (Table 3). Whereas the positive effects of SDAP appeared to predominate from d 0 to 7 of trial, the effects of colistin were more evident from d 8 to 14. Thus, SDAP improved ADG, ADFI, and gain:feed ratio from d 0 to 7 by 68 g/d (P < 0.05), 41 g/d (P < 0.10), and 0.28 g/g (P < 0.10), respectively. From d 8 to14, SDAP improved ADG and gain:feed by 41 g/d (P < 0.10) and 0.17 g/g (P < 0.01) respectively. On the other hand, colistin improved ADG, ADFI, and gain:feed from d 8 to 14 by 102 g/d (P < 0.01), 62 g/d (P < 0.01), and 0.17 g/g (P < 0.01), respectively. Over the 14 d of the trial, the effects of SDAP and colistin were similar in magnitude; ADG was improved by 54 (P < 0.05) and 75 g/d (P < 0.05) and gain:feed was improved by 0.20 (P < 0.05) and 0.18 g/g (P < 0.05) due to SDAP and colistin, respectively. Additionally, colistin improved ADFI by 45 g/d (P < 0.01) during that period. Significant interactions between colistin and SDAP were observed for gain:feed in wk 1 (P < 0.10) and for ADFI in wk 2 and from d 0 to 14 (P < 0.05), indicating that the positive responses to SDAP and colistin were not additive when presented in combination. Furthermore, colistin and SDAP interactions for ADG in wk 1 and ADG from d 0 to 14 were close to statistical significance (P = 0.11 and P = 0.12, respectively).
Morphology of the Small Intestine
SDAP, colistin, or day postweaning did not significantly affect the length of the small intestine (P > 0.10; Table 4). Its weight, however, was 156 g higher (P < 0.01) at d 14 than at d 7 postweaning and tended to be higher (93 g; P < 0.10) when colistin was present in the diet. The histological observation of the tallest well-oriented villi in the mucosa of the small intestine showed a tendency for villi 100 μm higher (P < 0.10) and crypts 33 μm deeper (P < 0.10) at d 14 than at d 7 postweaning. Inclusion of colistin also tended to increase the height of the villi by 106 μm (P < 0.10), which is in good agreement with the higher weight of the small intestine observed. Although heavier small intestine (61 g) and longer villi (74 μm) were observed with the use of SDAP, these values did not reach statistical significance (P > 0.10). A linear relationship between villi height and ADG of the piglets was observed (P < 0.001; Figure 1).
Microbiology of Digestive Contents
The inclusion of SDAP in the diet resulted in an increase of 1.14 log cfu/g in the number of lactobacilli in the ileal digesta (P < 0.10) and of 0.45 log cfu/g in the cecal digesta (P < 0.05; Table 5). An interaction (P < 0.05) between colistin and SDAP was observed for lactobacilli in cecal digesta, indicating that the effects of SDAP and colistin were not additive when presented in combination. No effect of SDAP was observed on the numbers of enterococci (P = 0.76 and P = 0.14), E. coli (P = 0.67 and P = 0.59), or C. perfringens (P = 0.61 and P = 0.55) in ileal and cecal digesta, respectively. Colistin reduced the number of enterococci in cecal digesta by 0.94 log cfu/g (P < 0.05). This reduction tended to be more pronounced at d 7 than at d 14 (P < 0.10). Colistin also reduced the number of E. coli, in both ileal and cecal digesta by 5.30 and 4.38 log cfu/g, respectively (P < 0.001). No effect of colistin was observed on the numbers of lactobacilli (P = 0.83 and P = 0.67) or C. perfringens (P = 0.22 and P = 0.55) in ileal and cecal digesta, respectively.
The improved performance observed in our study by replacing fishmeal with SDAP is in agreement with previous studies in which the replacement of either fish meal or other protein sources also resulted in positive responses (Hansen et al., 1993; Kats et al., 1994; Angulo and Cubiló, 1998). The high nutritional value of some of the proteins replaced suggests that the positive effects of SDAP cannot be explained solely from its nutrient composition and supports the claim that SDAP has some nonnutritional properties that are beneficial to piglets. If anything, SDAP has been reported to be relatively deficient in methionine (Kats et al., 1994) and to have an apparent ileal digestibility of amino acids lower than that of other protein sources, such as dried skim milk, isolated soy protein, or wheat gluten (Chae et al., 1999). The increased feed intake during the first week postweaning observed with SDAP agrees with previous reports (Hansen et al., 1993; Coffey and Cromwell, 1995; Angulo and Cubiló, 1998). Ermer et al. (1994) observed that piglets preferred diets with SDAP to diets with dried skim milk, and suggested that the increased feed intake found with SDAP diets was due to an improvement in palatability of the feed. This preference for SDAP, however, could also be due to the claimed “health-promoting” properties of SDAP since animals also show a preference for diets that confer health advantages (Forbes, 1999). Since higher feed intakes were observed for both colistin and SDAP, this is more likely to be due to an improvement in the health status of the animals than to an improvement in palatability. The improved gain:feed observed with the use of SDAP and colistin supports this view since a lower activation of the immune system would result in a higher availability of nutrients for weight gain (Stahly, 1996; Williams et al., 1997).
The higher feed intake of the pigs fed on colistin and SDAP coincided with the greater weight of their small intestine and the greater height of the tallest villi, which is in agreement with Pluske et al. (1997), who showed a linear relationship between DMI and villous height, and with Touchette et al. (1997), who found that SDAP improved the morphology of the small intestine independent of feed intake level.
Compared with a previous trial with pigs from the same station and diets of similar composition, but without an experimental challenge with E. coli K99 (Torrallardona et al., 2002), higher responses to SDAP and colistin were obtained in the present trial. This suggests that SDAP and colistin are more effective if the health status of the pigs is compromised and is in agreement with the observations of Coffey and Cromwell (1995), Stahly (1996), and Bergström et al. (1997), who found higher responses to plasma in conventional vs. high-health environments. This supports the claim for a health-promoting mechanism of action of SDAP. The immunoglobulin fraction of SDAP has been suggested to be responsible for these properties. Gatnau et al. (1995), Pierce et al. (1995), and Godfredson-Kisic and Johnson (1997) demonstrated improved performance of pigs with the high-molecular weight fraction of plasma containing immunoglobulins, but not with the medium- and low-molecular weight fractions rich in albumin and peptides, respectively. Additionally, a nonspecific protection mechanism in which the glycoproteins present in SDAP reduce the adhesion of pathogenic bacteria to the gastrointestinal mucosa has also been proposed (Nollet et al., 1999). These authors observed that a particular source of SDAP (without specific antibodies against F18 fimbriae) was effective against an oral challenge with an F18+ E. coli strain.
It is important to note that in the present trial, the diets without SDAP contained fishmeal and a higher proportion of lard in order to balance amino acid and energy content. It is estimated that this resulted in 1.2% higher fat content and a different fatty acids profile. The estimated n-3 PUFA content of the diets with fishmeal was 4.17 g/kg of diet, whereas that of SDAP diets was 2.91 g/kg of diet. As n-3 PUFA have been shown to downregulate immune function in laboratory animals and humans (Miles and Calder, 1998) as well as in pigs (Thies et al., 1999), the question arises as to whether the positive responses are due to a direct effect of SDAP or to an indirect effect of removing fishmeal from diets. From the available evidence, the extent to which the differences in fatty acid composition of the diets could reduce the immune response of the piglets on the fish diet cannot be determined. Nevertheless, fishmeal has been considered for many years to be a very good protein source for piglets (Kim and Easter, 2001), and no adverse effects on immunity have been described. Alternately, positive responses to the use of SDAP have consistently been found when protein sources other than fishmeal were replaced (van Dijk et al., 2001). Additional evidence for a direct health-promoting action of SDAP is given by the studies of Coffey and Cromwell (1995), in which the replacement of dried skimmed milk for SDAP was more effective in poorer health environments. Similarly, Bosi et al. (2001) and van Dijk et al. (2002) found positive responses when SDAP replaced hydrolyzed casein and soybean meal plus whey powder, respectively, in piglets experimentally challenged with E. coli. Therefore, there is evidence to suggest that SDAP by itself has the ability to improve the resistance of the pigs to disease, although further studies must be conducted to assess its relationship with the fatty acid composition of the diets when fishmeal is the protein source being replaced.
Although no pigs died due to the experimental challenge with E. coli K99 in the current study, this may have been due to the relatively small dose administered (5 × 107 cfu/pig). Bosi et al. (2001) report a study of similar design in which SDAP was tested on piglets challenged with 1010 cfu of E. coli K88. They observed that SDAP reduced the mortality and reduced the K88-specific IgA in the saliva of the piglets, suggesting a protective mechanism of SDAP against the experimental challenge, which is in agreement with the current results.
The dramatic reduction in the number of colony-forming units of E. coli in the ileal and cecal contents due to colistin clearly supports the effectiveness of this antimicrobial agent against the experimental challenge used, in agreement with Mateu and Martin (2000), who reported a low degree of antibiotic resistance in porcine enteric E. coli to colistin. On the other hand, SDAP did not significantly reduce the numbers of E. coli, but increased the number of lactobacilli, which indicates that both products must have different modes of action. Whereas colistin has a bactericidal effect against coliforms (Catchpole et al., 1997), SDAP may have acted against E. coli K99 through its immunoglobulin fraction (Gatnau et al., 1995; Pierce et al., 1995; Godfredson-Kisic and Johnson, 1997) or by preventing the adhesion of pathogenic E. coli onto the mucosal surface (Nollet et al., 1999).
The interaction between SDAP and colistin observed for some parameters in the present study suggests that their effects were not additive, and it could be explained by a common principal action of both products against the experimental challenge with E. coli K99. This is in contrast with the results of Coffey and Cromwell (1995) and with our previous study without E. coli K99 challenge (Torrallardona et al., 2002) in which no interaction between SDAP and colistin was observed, suggesting independent modes of action and additive effects. The absence of an experimental challenge in these studies should have resulted in weaker and less specific pathologies than in the current study, and this could explain the independence of the effects of both products. Whereas colistin is effective against gram-negative bacteria (Catchpole et al., 1997), SDAP could have also been effective against other pathogens. The protection of SDAP against rotavirus infection reported by Cain and Zimmerman (1997) supports this view.
Inclusion of spray-dried animal plasma in the diet offers an advantage to newly weaned piglets challenged with an oral dose of Escherichia coli K99. The performance response to spray-dried animal plasma was similar in magnitude to that obtained with the antibiotic colistin, suggesting that spray-dried animal plasma may be a good alternative to medicated feed with antibiotics. Both products contributed to the preservation of small intestine mucosal integrity by maintaining the length of the villi and contributing to a higher small intestinal weight. They also had a direct effect on the microbial population of the gastrointestinal tract. The current situation in which the use of antimicrobials in animal feeding is being questioned should encourage further investigation into the use of spray-dried animal plasma as a means of preventing disease in pigs at weaning.
|Full-fat extruded soybeans||141.7||141.7|
|Soybean meal 48||100||100|
|Whey, dried sweet||150||150|
|Vitamin and mineral premixd||4||4|
|Methionine + cysteine||9||9.4|
|Day 0 to 7|
|Gain:feed, g/g||0.13||0.60||0.68||0.61||0.120||P†, P × C†|
|Day 8 to 14|
|ADG, g/d||179||306||245||321||20.3||P†, C**|
|ADFI, g/d||290||398||324||339||15.2||C**, P × C*|
|Gain:feed, g/g||0.60||0.76||0.76||0.94||0.038||P**, C**|
|Day 0 to 14c|
|ADG, g/d||101||214||193||230||20.7||P*, C*|
|ADFI, g/d||227||298||268||286||10.8||C**, P × C*|
|Gain:feed, g/g||0.43||0.71||0.72||0.81||0.069||P*, C*|
|BW, kgc||6.89||8.13||7.79||8.59||7.66||10.40||9.52||10.30||0.565||P†, C**, D**|
|Weight, kg||0.31||0.42||0.36||0.46||0.44||0.57||0.57||0.60||0.064||C†, D**|
|Mucosa of small intestine|
|Villi height, μm||568||673||535||722||573||737||810||780||76.9||C†, D†|
|Crypt depth, μm||166||179||190||172||202||164||234||241||26.0||D†|
|Ileal digesta, log cfu/g|
|Cecal digesta, log cfu/g|
|Lactobacilli||7.91||8.46||8.69||8.46||8.11||8.46||8.98||8.62||0.242||P*, P × C*|
|Enterococci||6.47||4.64||5.39||3.92||4.53||4.31||4.25||4.01||0.548||C*, D†, C × D†|