| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
, and interleukin-1ß in weaned pigs1Department of Animal Sciences and Industry, Weber Hall, Kansas State University, Manhattan 66506-0201
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
, and IL-1ß. There was no change in the rectal temperature of the control or the Typhimurium-challenged pigs (compared with d 0) or when comparing Typhimurium-challenged pigs with control animals. In contrast, pigs fed Choleraesuis had increased rectal temperatures beginning on d 2 and continuing through d 7 (P < 0.05), with the greatest elevation on d 3 (P < 0.001) compared with the control pigs. Average daily gain and ADFI of pigs challenged with Typhimurium did not differ from those of the control animals. Pigs fed Choleraesuis had a 25% reduction in ADG (P < 0.0001) and ADFI (P < 0.002) compared with the control pigs. On d 7, pigs fed Choleraesuis had reduced serum IGF-I compared with control (P < 0.01) or Typhimurium-challenged pigs (P = 0.01). Bacterial feeding did not affect serum tumor necrosis factor- or IL-1ß compared with control pigs at any time throughout the experiment. We conclude that repeated exposure of weaned pigs to Choleraesuis reduced growth performance in the absence of changes in systemic inflammatory cytokines.
Key Words: insulin-like growth factor-I • interleukin-1ß • Salmonella enterica • swine • tumor necrosis factor-
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). These salmonellae serovars produce different patterns of disease in growing pigs.
Pigs infected with Typhimurium are more likely to develop mild enteritis and self-limiting diarrhea, whereas pigs infected with Choleraesuis, a so-called swine host-adapted serovar, usually develop systemic disease such as septicemia (Schwartz, 1999). Although oral exposure of weaned pigs to Typhimurium caused fever and growth suppression (Balaji et al., 2000), Typhimurium did not stimulate changes in systemic concentrations of the inflammatory cytokines tumor necrosis factor- (TNF-; Balaji et al., 2000) or IL-6 (Burkey et al., 2004). Based on these results, we concluded that Typhimurium was largely contained by the mucosal immune system without provoking systemic inflammatory cytokine secretion. This contrasts with swine lipopolysaccharide (LPS) models of bacterial infection, which generally result in large elevations in inflammatory cytokines (summarized by Johnson et al., 2005).
Given that the swine host-adapted serovar Choleraesuis was reported to cause systemic disease, which is a different disease outcome from that of Typhimurium, our working hypothesis for the current study was that this serovar would likely engage immune-inflammatory mechanisms that would result in elevated systemic cytokine secretion.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Bacteria were cultured for feeding, as needed, using Typhimurium or Choleraesuis that had been transformed with green or red fluorescent protein, respectively, as described previously by our laboratory (Burkey et al., 2006; Skjolaas, 2006). Importantly, these transformed bacteria were confirmed to retain their inflammatory signaling in swine gastrointestinal epithelial cells and to affect the relative expression of toll-like receptors and selected chemoattractive cytokines and chemokines (Burkey et al., 2006; Skjolaas, 2006). The fluorescence and kanamycin resistance conferred by the transformed plasmids provided 2 phenotypic markers from which to isolate and distinguish the serovars from potential environmental salmonellae. On the days of bacterial feeding, the bacteria were washed and diluted in PBS to deliver 108 cfu/100 µL. A small, disposable pipette tip was used to make a depression in the dough ball, and the 100 µL containing the desired bacteria was pipetted into the depression. The depression was then pinched closed.
Bacteria for the initial 2 bacterial feedings were from colonies grown in the laboratory after transformation. Subsequently, the bacteria were obtained after passage through the pigs. For this, fecal samples were collected and pooled among all pens within day and treatment. Fecal samples were cultured on d 1 to 8 and again on d 11. All pigs within a treatment were then subsequently fed bacteria from a fecally isolated colony.
Fecal samples were preenriched in tetrathionate broth (cat. no. T-1938, Sigma, St. Louis, MO) for 24 h at 37°C. Selective enrichment was performed by transferring 1% of the tetrathionate broth/fecal culture to Rappaport’s medium (cat. no. 218581, BD Biosciences, Sparks, MD) for an additional 24 h at 37°C. Selective agar plating was then performed by streaking 100 µL of the Rappaport’s medium onto Luria-Bertani plates (Bertani, 1951) containing 50 µg/mL of kanamycin (catalog no. 60615, Sigma), followed by overnight incubation at 37°C. Bacteria were isolated and deemed the appropriate salmonellae serovar based on growth in the presence of kanamycin as well as by the presence of either green (Typhimurium) or red (Choleraesuis) fluorescence.
This level of bacterial feeding was arrived at empirically from our previous experience with this same Typhimurium isolate (Balaji et al., 2000; Turner et al., 2002a; Burkey et al., 2004) and from published work that suggested this level of oral exposure would be expected to produce only mild clinical effects (Schwartz, 1999). We had not used this isolate of Choleraesuis in our previous research, but like the Typhimurium isolate, Choleraesuis was derived from a swine clinical case and was a gift from Jerome Nietfeld (Department of Diagnostic Medicine/Pathobiology, Kansas State University). The identities of both the wild type and transformed isolates were confirmed by the National Veterinary Services Laboratory (Ames, IA).
All pigs were housed in temperature-controlled rooms under constant lighting. Pigs fed the bacteria were housed in one room, and the control pigs were housed in another room. Each pen contained a single nipple waterer and a single self-feeder to allow ad libitum access to water and feed. The pigs were fed a standard corn-soybean meal-based nursery diet with 3% added choice white grease without any other specialty ingredients. The diet was formulated to contain 3,434 kcal of ME/kg, 1.51% total lysine, 0.78% calcium, and 0.38% available phosphorus on an as-fed basis. All other nutrients were formulated to exceed NRC requirements for growth (NRC, 1998). To ensure that the diet itself was not antimicrobial, it was formulated to be free of growth-promoting antibiotics, zinc oxide, or copper sulfate. To ensure that the pigs began the study free of clinical salmonellosis, fecal samples were cultured before beginning the study and were confirmed to be negative for salmonellae organisms.
Pigs were from a single farrowing group (PIC 327 x PIC 1050) and averaged 48 d of age and 18.2 kg of BW on d 0 of the study. Pig BW and feed disappearance were recorded initially and at the conclusion of the study to determine ADG, ADFI, and G:F. Rectal temperatures were obtained daily from 1 pig/pen beginning 2 d before the first bacterial feeding through d 7 using rapid-response digital thermometers (model 679, Welch Allyn Inc., San Diego, CA). Blood sampling (detailed below) and rectal temperature measures were consistently obtained early in the morning, generally between 0500 and 0700, and were obtained from the same pig within a pen across days.
Serum Analysis
Serum was collected from a single pig/pen for analysis of TNF- and IL-1ß on d 0, 7, and 14. Blood was collected into glass tubes containing no anticoagulant, and was allowed to clot at room temperature and stored overnight at 4°C before harvest of the serum by centrifugation. An RIA, described previously for use in pigs (Balaji et al., 2000), was used to analyze serum IGF-I concentrations. A swine-specific ELISA was used for determination of serum TNF- concentrations (Quantikine Porcine TNF-/TNFSF2 immunoassay, cat. no. PTA00, R& D Systems, Minneapolis, MN). A swine-specific ELISA was also used for determination of IL-1ß (Quantikine Porcine IL-1ß immunoassay, cat. no. PLB00, R&D Systems).
Concentrations of IGF-I were quantified in a single assay that was sensitive to 6.0 ng/mL, with an intraassay CV of 6.3%. Concentrations of TNF- were evaluated in 2 assays. The sensitivity was 5.6 pg/mL. The intraassay CV averaged 7.6%, and the interassay CV was 21.1%. Concentrations of IL-1ß were determined in 2 assays that were sensitive to 39.1 pg/mL. Different control samples were inadvertently included in the 2 assays, and this error precluded calculation of an interassay CV. The intraassay CV averaged 4.0%. When the concentrations of IL-1ß were below the limit of detection of the assay, the assay sensitivity was assigned as the concentration for that sample for the sake of statistical analysis.
Statistical Analyses
Data were analyzed with PROC MIXED (SAS Inst. Inc., Cary, NC) as a randomized complete block design with repeated measures over time for each experimental unit (individual pens). The model included terms for the fixed effects of disease challenge, time, and their interaction, and block and pen were considered random effects. Comparisons between bacterial challenges or sampling times, or both, were made only when a significant (P < 0.05) F-test for the main effect or interaction was found using the least significant differences procedure. All means reported are least squares means.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Rectal temperatures were monitored daily beginning 2 d prior to challenge with bacteria and until 7 d after the first bacterial feeding (Figure 1). There was no change in rectal temperature of either the control or the Typhimurium-challenged pigs (compared with d 0) and no change in the Typhimurium-challenged pigs compared with the controls. In contrast, pigs fed Choleraesuis had increased rectal temperatures beginning on d 2 and continuing through d 7 (P < 0.05), with the greatest elevation on d 3 (P < 0.001) compared with controls.
|
). The ADG and feed intake of pigs challenged with Typhimurium did not differ from those of the control pigs. Pigs fed Choleraesuis had about a 25% reduction in ADG (P < 0.0001) and ADFI (P < 0.002) compared with control pigs. There were no differences in G:F among the treatment groups.
|
). Concentrations of IGF-I were similar between the treatment groups on d 0 and 14. On d 7, pigs fed Choleraesuis had reduced IGF-I compared with both control (P < 0.01) and Typhimurium-challenged pigs (P = 0.01). Serum TNF- and IL-1ß concentrations are depicted in Figure 4. Bacterial feeding did not affect either cytokine compared with control pigs.
|
|
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
,b; Jenkins et al., 2004). In those experiments, we used a single oral dose of approximately 109 to 1010 bacteria. In general, in those experiments, we used rectal temperature, feed intake, and circulating IGF-I as indicators of the clinical effects of the enteric pathogen. The current experiment differed in 2 important ways. First, the current experiment included the so-called swine host-adapted serovar Choleraesuis (discussed in more detail below). Second, here we elected to use a lower level of bacterial exposure but to provide it repeatedly in order to more closely mimic the oral-fecal transmission that likely occurs within nursery pens under production conditions. Because we had never evaluated the low-dose repeated exposure to Typhimurium in nursery pigs, nor had we previously used the Choleraesuis serovar in this disease model, we considered the rectal temperature, feed intake, and IGF-I data to be vital to the thorough characterization of this experimental model. Because all pigs infected with Typhimurium and Choleraesuis were housed in one room and control pigs in a separate (identical) room, the authors acknowledge that the effects of the enteric pathogens technically were confounded with the housing location. This confounding is essentially always a reality when infected animals are compared with uninfected controls when containment is required to ensure biosecurity and prevent unintended infection of control animals. However, the facts that Choleraesuis and Typhimurium pigs were housed in the same animal room, and that Choleraesuis produced marked effects whereas Typhimurium did not (detailed below) indicates that the observed pathophysiological effects were indeed attributable to the pathogen and not to the room itself.
Within 2 d of Choleraesuis exposure, rectal temperature was elevated in pigs fed this serovar. It peaked at d 3 and remained elevated above that of control pigs through d 7. We elected a priori not to continue daily rectal temperature measurements through the entire study to reduce undue stress of continued animal handling and temperature measurement. Hence, we are not certain whether pigs fed Choleraesuis remained febrile through the conclusion of the study. We were somewhat surprised, however, that Typhimurium failed to elevate rectal temperatures above those of control pigs. On the other hand, the lack of a rectal temperature response to Typhimurium may be related to the lower numbers of bacteria provided in the current experiment. We favor this interpretation, because even though Typhimurium stimulated a strong febrile response to a single oral dose of 109 to 1010 Typhimurium (Balaji et al., 2000; Burkey et al., 2004; Jenkins et al., 2004), the response varied among studies with pigs given 1010 cfu of Typhimurium and resulted in only a single day of elevated rectal temperature in one study (Jenkins et al., 2004). Moreover, 108 oral salmonellae are reported to be at the low end of bacteria-producing clinical signs (Schwartz, 1999). However, it is clear from the current experiment that the febrile response to small oral doses of salmonellae organisms is very much serovar dependent, making it difficult to predict a minimum oral exposure to produce clinical symptoms across all important swine serovars.
Oral exposure to Choleraesuis resulted in approximately 25% reduction in growth that likely is explained by a reduction in feed intake of similar magnitude. However, oral exposure to an identical oral dose of Typhimurium did not result in erosion of growth performance. Although the reduction in growth in response to Choleraesuis is generally consistent with other published reports of young growing pigs carrying Choleraesuis (Gray et al., 1995), it differs from our previous findings with a single exposure to Typhimurium. In our other studies (Balaji et al., 2000; Burkey et al., 2004), Typhimurium challenge resulted in reduced feed intake. In the current study, the Typhimurium-challenged pigs did not have a significant reduction in feed intake compared with control animals. Again, we suspect that the lack of effect of Typhimurium on intake likely reflects the reduced oral dose compared with that used in our previous studies, yet this observation may have more far-reaching implications. Namely, the current dogmatic view is that pigs respond to low-level antibiotic feeding with improved growth performance because doing so controls pathogens in the gastrointestinal tract (Dritz et al., 2002). Our data here with Typhimurium-challenged pigs, compared with control animals, suggest that the mere presence of invasive enteric pathogens is not in itself sufficient to slow growth and that it appears to be a dose-dependent effect.
The association between level of intake and circulating IGF-I is unmistakably coupled in young pigs, and circulating IGF-I declines rapidly after feed deprivation (Salfen et al., 2003). We have previously evaluated circulating IGF-I as an ancillary marker of the inappetence associated with carrying an enteric pathogen (Johnson et al., 2005). Indeed, pigs fed Choleraesuis in the current study demonstrated the expected reduction in circulating IGF-I, and this reduction was likely the result of Choleraesuis-induced reduced feed intake. However, IGF-I did not remain reduced in pigs fed Choleraesuis in that the growth factor was similar among all treatments by the conclusion of the study. The return of IGF-I in pigs fed Choleraesuis compared with those of control pigs and pigs fed Typhimurium by the conclusion of the study likely indicates that the majority of the reduction in intake (Figure 2
) probably occurred within the first week of Choleraesuis feeding.
Models of immune or inflammatory challenge based on injection of pigs with LPS, without exception, have demonstrated unmistakable elevations in the inflammatory cytokine trio TNF-, IL-1ß, and IL-6 after LPS treatment. These models of LPS have generally helped to shape the dogmatic view that circulating inflammatory cytokines associated with systemic inflammatory processes participate in slowed growth in sick animals (Johnson, 1997; Spurlock, 1997; Fossum et al., 1998). However, in our previous studies with Typhimurium, the bacteria failed to affect circulating TNF- (Balaji et al., 2000) or IL-6 (Burkey et al., 2004). Therefore, that neither TNF- nor IL-1ß was affected by Typhimurium in the current study is generally consistent with those reports. In this regard, it is important to again point out that, although the dose of Typhimurium was less than we had used previously, the transformed bacteria produced changes in tissue expression of toll-like receptors, IL-8, macrophage migration inhibitory factor, osteopontin (Burkey et al., 2006), and CC chemokine ligand 20 (Skjolaas, 2006). Nevertheless, this is our first evaluation of IL-1ß in response to Typhimurium. Here, we evaluated peripheral cytokines 7 and 14 d after the first exposure to the bacteria. Moreover, in the current study, we reexposed pigs throughout the study. We cannot rule out the possibility that both TNF- and IL-1ß may have been affected before d 7, although we feel this is not likely. Of greater interest, however, relative to a primary objective of the current study, is that, although Choleraesuis reduced growth and produced fever, these effects occurred in the absence of changes in TNF- and IL-1ß. Thus, despite the documented likelihood of Choleraesuis producing systemic disease in pigs, we failed to gather data to support our working hypothesis that Choleraesuis, in contrast to Typhimurium, would result in elevated peripheral inflammatory cytokines.
In conclusion, we report a model of refeeding transformed Typhimurium or Choleraesuis that may mimic fecal-oral exposure in production settings and perhaps offer advantages over that of our previously used model of a single intragastic inoculation of bacteria. Although Choleraesuis reduced growth by approximately 25% and transiently decreased IGF-I, neither swine salmonellae serovar produced changes in systemic TNF- or IL-1ß.
|
Footnotes |
|---|
2 Current address: 206f Animal Sciences, University of Nebraska, PO Box 830908, Lincoln, NE 68583-0908.
3 Corresponding author email: eminton{at}ksu.edu
Received for publication July 19, 2006. Accepted for publication December 17, 2006.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. A. Carroll Bidirectional communication: Growth and immunity in domestic livestock J Anim Sci, April 1, 2008; 86(14_suppl): E126 - E137. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
