Investigation of Medium Chain Fatty Acid Feed Supplementation for Reducing Salmonella Typhimurium Colonization in Turkey Poults.

Studies indicate that persistent Salmonella colonization occurs in poultry that are infected early in life, leading to both food safety and public health concerns. Development of improved preharvest Salmonella management strategies is needed to reduce poultry product contamination. The objective of this study was to evaluate the efficacy of a product containing medium chain fatty acids (MCFA) for reducing early Salmonella colonization in turkey poults. Day-of-hatch turkeys were provided a standard starter diet supplemented with MCFA at 0 (negative and positive controls), 1.5, 3, 4.5, or 6 lbs/ton of feed. Positive control and MCFA treated birds were also crop-gavaged with 108 colony forming units (CFU) of bioluminescent Salmonella Typhimurium. Gastrointestinal tissue samples were collected at 3 days postinoculation for bioluminescence imaging (Meckel's diverticulum to the cloaca) and selective enumeration (cecal contents). Quantification of bioluminescence indicated that the 4.5 and 6 lbs/ton MCFA groups had significantly less colonization than the positive control group (p = 0.0412 and p < 0.0001, respectively). Similarly, significantly lower numbers (1-log10 CFU/g reduction) of Salmonella were observed in the ceca of the 6 lbs/ton MCFA group compared to the positive control group (p = 0.0153). These findings indicate that incorporation of MCFA in turkey diets can significantly reduce early Salmonella colonization. In addition, this study highlights the utility of bioluminescence imaging as a screening methodology for assessing the efficacy of treatments that may reduce Salmonella in poultry.

Introduction

In the United States, an estimated 1.2 million human salmonellosis cases occur annually and are predominately associated with the consumption of contaminated food products (Scallan et al., 2011; Jackson et al., 2013). Foodborne illness surveillance has linked poultry products to multiple Salmonella outbreaks in humans (Gould et al. 2013). Although contamination of poultry meat with Salmonella can be attributed to a number of factors, there remains a strong relationship between the gastrointestinal colonization of poultry and product contamination risk (Rasschaert et al., 2008; Evans et al., 2015).

Poultry are susceptible to developing persistent Salmonella gastrointestinal colonization if they become infected early in life (Gast and Beard, 1989; Gast and Holt, 1998; Van Immerseel et al., 2004b). These birds remain colonized for the duration of the production cycle and provide a reservoir for the continued spread of Salmonella throughout the flock (Gast and Beard, 1989; Gast and Holt, 1998; Van Immerseel et al., 2004b). The stress associated with transport to the processing plant can then induce fecal shedding and lead to raw product contamination (Mulder, 1995; Marin and Lainez, 2009; Evans et al., 2015). Therefore, treatments that reduce early Salmonella colonization may have an impact on overall flock carriage levels and may help reduce product contamination.

Medium chain fatty acids (MCFA; C6-C12) have recently been investigated as a preharvest treatment for controlling Salmonella during poultry production. These molecules are more bactericidal than short chain fatty acids (≤C4) which are routinely used to combat Salmonella (Kwon and Ricke, 1998; Nakai and Siebert, 2003; Van Immerseel et al., 2003, 2006; Skrivanova et al., 2006). Interestingly, MCFA also suppress the expression of the invasion genes, hilA and hilD, in Salmonella and decrease intestinal epithelial cell invasion (Van Immerseel et al., 2004a; Kollanoor-Johny et al., 2012). Studies in both layers and broilers show that MCFA can reduce Salmonella colonization in the gastrointestinal tract and in other tissues (Van Immerseel et al., 2004a; Kollanoor-Johny et al., 2009, 2012; Upadhyaya et al., 2015). Potentially, preharvest MCFA treatment could have an impact on product contamination by reducing Salmonella colonization during production. However, additional studies are needed to further evaluate the efficacy of MCFA for combating Salmonella in not only chickens but also in turkeys.

Our laboratory has developed new methods for evaluating Salmonella gastrointestinal colonization. By inoculating birds with a bioluminescent Salmonella Typhimurium LT2 isolate and imaging with the In Vivo Imaging System (IVIS®; PerkinElmer), we can view colonization in the gastrointestinal tract and investigate various treatment strategies. One advantage of the IVIS is the ability to quantify bioluminescence which is directly related to the levels of the target bacteria present (Karsi et al., 2008; Foucault et al., 2009). The objective of this study was to evaluate the efficacy of a product containing a combination of MCFA (C6-C12) for reducing early Salmonella colonization in turkey poults.

Materials and Methods

Animals

All animal experiments were approved by the Virginia Tech Institutional Animal Care and Use Committee. Upon arrival, 36 day-of-hatch female turkey poults (Aviagen Turkeys, Inc., Lewisburg, WV) were randomly assigned into six groups and placed into individual isolator cages (n = 5–7, with 340 cm2/bird). To ensure birds were Salmonella negative at the start of the study, pooled cecal droppings were collected from each group and tested with the DNAble Molecular Detection Kit for Salmonella (DF-026) using Extraction Set 1 (EnviroLogix, Inc., Portland, ME) as described previously (Evans et al., 2015).

Bioluminescent Salmonella culture and inoculum preparation

A bioluminescent Salmonella enterica serovar Typhimurium LT2 isolate was used for this study. The isolate contains the expression vector pNSTrcD with bioluminescent luciferase cassette (CDABE) and chloramphenicol resistance gene (Seleem et al., 2007a, b; Seleem et al., 2008a, b). The transformed isolate was streaked onto tryptic soy agar (TSA; BD Difco™, Sparks, MD) and incubated overnight at 37°C. For preparation of frozen inoculum stocks, a single bright colony was selected from a TSA plate and used to inoculate 20 mL of tryptic soy broth (TSB; BD Difco) in 50 mL conical tubes (Fisher Scientific, Hampton, NH). Tubes were then placed in a shaking (175 rpm) incubator overnight at 37°C. Cultures were pelleted by centrifugation at 2000 × g for 15 min. Pellets were washed twice in sterile phosphate-buffered saline (PBS; Fisher Scientific), pooled, and frozen at −80°C with a final glycerol concentration of 20% v/v. One week later, an aliquot of the frozen stock was thawed and streaked onto TSA and TSA +5% sheep blood (Remel™, San Diego, CA). Plates were incubated overnight at 37°C to confirm purity. A plate count was performed to determine the concentration of Salmonella in the prepared stocks. In addition, a minimum inhibitory concentration (MIC) assay (Sensititre™; Thermo Scientific, Waltham, MA) was carried out by the Clinical Bacteriology Lab at the Virginia-Maryland College of Veterinary Medicine. The isolate was found to be resistant to 0.25 μg/mL of ceftiofur (chloramphenicol was not included in the panel). Chloramphenicol resistance was confirmed by growing the isolate overnight at 37°C on TSA plates supplemented with 30 μg/mL chloramphenicol (Fisher Scientific).

Salmonella challenge

Before challenge, each group of birds (day-of-hatch) was provided access to 82.5 g of HydroGel™ (ClearH2O, Westbrook, ME) mixed with 18 g of standard starter diet (Table 1) containing a MCFA product (Nuscience Group, Ghent, Belgium) at 0 (negative and positive controls), 1.5, 3, 4.5, or 6 lbs/ton of feed (0, 0.75, 1.5, 2.25, 3 g/kg) for 1 h. Birds were then fasted for 2 h during which time a frozen stock of bioluminescent Salmonella was thawed on ice and diluted in sterile PBS. Birds were either crop-gavaged with 1 mL of 1 × 108 colony forming units (CFU)/mL of bioluminescent Salmonella (positive control and MCFA treated groups) or 1 mL of sterile PBS (negative control). Birds were then provided a standard starter feed (Table 1) containing the corresponding MCFA treatment as described above and water ad libitum for the duration of the study.

1.

Diet Composition of the Standard Starter Diet Fed to Turkey Poults

Ingredient	(%) Composition
Corn	44.23
Soybean meal (48%)	43.86
Poultry byproduct meal	5.00
Soy oil	0.99
Salt (NaCl)	0.19
Sodium bicarbonate	0.20
dl-Methionine	0.30
Lysine•HCl	0.26
Limestone	1.09
Dicalcium phosphate	3.02
Choline chloride (60%)	0.10
Vitamin/mineral premix[a]	0.75
Calculated composition
ME (kcal/kg)	2850
Crude protein (%)	28.12
Calcium (%)	1.40
Available P (%)	0.75
Dig Lys (%)	1.62
Dig Met (%)	0.66
Dig Thr (%)	0.97
Dig Met+Cys (%)	1.05

Provided the following per kg vitamin/mineral premix: cobalt 34 ppm; copper 540 ppm; iodine 134 ppm; iron 6750 ppm; manganese 8580 ppm; zinc 6500 ppm; vitamin A 881,848 IU/kg; vitamin D3 295,419 ICU/kg; vitamin E 220 IU/kg; vitamin B12 0.9 mg/kg; menadione 154 mg/kg; riboflavin 551 mg/kg; d-pantothenic acid 811 mg/kg; niacin 2646 mg/kg; and choline 51,030 mg/kg.

P, Phosphorus; Lys, lysine; Met, methionine; Thr, threonine; Met+Cys, methionine and cystine.

Imaging of gastrointestinal samples

Turkeys were humanely euthanized with CO2 at 3 days postinoculation (PI) for bioluminescence imaging. Gastrointestinal tissue samples were collected from Meckel's diverticulum to the cloaca. Samples were placed into containers premoistened with sterile PBS to prevent drying and kept warm in a 37°C incubator. The samples were opened longitudinally just before imaging and placed onto a black low-background Lexan™ sheet (PerkinElmer, Waltham, MA). An IVIS SpectrumCT (PerkinElmer) was used to quantify bioluminescence. After a 1-min incubation on the imaging stage at 37°C, images were taken with a specimen height of 1.5 cm. Exposure and binning were set accordingly to keep the bioluminescence values within 600–60,000 counts (relative luminescence units) as specified by the manufacturer. Regions of interest (ROI) were overlaid (22 × 22 cm) on each image, and total flux (photons/s) was recorded for each sample.

Salmonella enumeration

Salmonella levels in the ceca were enumerated for each bird using a modified plate count procedure and then compared by treatment (Evans et al., 2015). Briefly, cecal contents were collected and weighed after imaging. Samples were diluted 1:10 w/v in sterile buffered peptone water (BPW; BD Difco) and then serially diluted (1:10) in PBS. From each dilution, 10 μL drops were plated in quadruplicate onto TSA plates supplemented with 30 μg/mL chloramphenicol and 0.25 μg/mL ceftiofur (Fisher Scientific). Plates were incubated for 10 h (colony diameter: 1–2 mm) at 41°C to approximate turkey body temperature (40.6–41.5°C). Following incubation, all plates were inspected with the IVIS SpectrumCT to confirm that only bioluminescent colonies were present, and CFU/g were calculated for each sample.

Statistical analysis

Analysis of both bioluminescence and enumeration data was conducted using JMP Pro 11 (SAS Institute, Inc., Cary, NC). Bioluminescence (photons/s) and enumeration (CFU/g) data were reported as the log10 mean for each treatment ± standard error of the log10 mean. Data were analyzed as a one-way analysis of variance (ANOVA). When the model was significant (p ≤ 0.05), a post-hoc analysis (Least Squares [LS] Means Student's t-test) was performed.

Results

Bioluminescent Salmonella imaging

Imaging of bioluminescent Salmonella was conducted on intestinal tissues from Meckel's diverticulum to the cloaca at 3 days PI. A pseudocolor overlay representing the bioluminescent light produced by Salmonella colonizing the intestinal tract is shown in Figure 1. Visually, bioluminescence appeared to be lowest in the negative control group and highest in the positive control group. Light levels also appeared to be progressively reduced as the MCFA concentration increased. The highest levels of bioluminescence were typically observed in the ceca of infected birds, but bioluminescence was also present throughout the gastrointestinal samples. Total flux (photons/s) was quantified for each sample ROI and compared by treatment group (Fig. 2). The ANOVA analysis was significant (p < 0.0001), and the post-hoc LS Means Student's t-test indicated that both the 4.5 lbs/ton (7.9 log10 photons/s ± 0.10) and 6 lbs/ton (7.5 log10 photons/s ± 0.11) MCFA groups had significantly less (p = 0.0412 and p < 0.0001, respectively) bioluminescence than the positive control group (8.4 log10 photons/s ± 0.15). The 6 lbs/ton MCFA group was also significantly lower (p = 0.0276) than the 4 lbs/ton MCFA group. There were no significant differences among the 1.5, 3, and 4 lbs/ton MCFA groups.

Images of gastrointestinal samples (Meckel's diverticulum to cloaca) with pseudocolor overlays representing the amount of light produced by bioluminescent Salmonella Typhimurium (3 days PI). The color bar indicates the amount of bioluminescence (surface radiance: p/s/cm2/sr) detected during imaging. ROI were overlaid on each image to quantify the total flux (photons/s) for each sample. MCFA, medium chain fatty acids; PI, postinoculation; ROI, regions of interest.

Salmonella bioluminescence in gastrointestinal samples (Meckel's diverticulum to cloaca) from control and MCFA treated poults (3 days PI). Error bars represent standard error of the log10 mean for each treatment group. Treatments not connected by the same letter are significantly different (p ≤ 0.05). MCFA, medium chain fatty acids; PI, postinoculation.

Samples collected from the ceca were quantified by plate count (Fig. 3). As expected, no Salmonella was recovered from birds in the negative control group. However, Salmonella was recovered from all other birds and ranged between 6.0 and 9.3 log10 CFU/g. The ANOVA analysis was significant (p < 0.0001), and the post-hoc test (LS Means Student's t-test) indicated that the 6 lbs/ton MCFA (7.0 log10 CFU/g ± 0.37) group had a significant reduction (p = 0.0153) in Salmonella (1-log10 CFU/g reduction) compared to the positive control group (8.0 log10 CFU/g ± 0.17).

Salmonella enumeration of cecal contents from control and MCFA treated poults (3 days PI). Error bars represent standard error of the log10 mean for each treatment group. Treatments not connected by the same letter are significantly different (p ≤ 0.05). MCFA, medium chain fatty acids; PI, postinoculation.

Discussion

Poultry are susceptible to Salmonella infections during the first few days of life and begin developing greater resistance thereafter (Gast and Beard, 1989). Birds infected immediately after hatch have more persistent colonization and higher levels of fecal shedding than birds infected at 1 week posthatch (Gast and Beard, 1989; Van Immerseel et al., 2004b). Preharvest management strategies that effectively decrease early carriage levels may reduce the spread of Salmonella within a flock (Gast and Beard, 1989). Presumably, this would also help reduce product contamination. Therefore, the objective of this study was to evaluate the efficacy of a MCFA product for reducing early Salmonella colonization in turkey poults.

A novel bioluminescent Salmonella infection model was used to assess MCFA efficacy. One benefit of this model is the ability to screen treatments by comparing both visual and quantitative changes. The images from the gastrointestinal samples suggest that Salmonella colonization was reduced as the MCFA concentration increased. Quantification of bioluminescence (photons/s) revealed a significant decrease for both the 4.5 and 6 lbs/ton MCFA groups compared to the positive control group. In addition, the 6 lbs/ton MCFA group had significantly lower bioluminescence than the 4.5 lbs/ton MCFA group, but there were no differences among the 1.5, 3, and 4.5 lbs/ton MCFA groups. These data indicate that there was a dose-dependent decrease in Salmonella (Meckel's diverticulum to the cloaca) resulting from MCFA supplementation. Furthermore, when Salmonella was enumerated in the ceca, there was a significant decrease in Salmonella numbers for the 6 lbs/ton MCFA group (1-log10 CFU/g reduction) compared to the positive control group. This study shows that MCFA can reduce early Salmonella colonization in turkeys and highlights the utility of bioluminescence imaging for evaluating treatment efficacy.

Several studies have evaluated the effect of MCFA on Salmonella colonization in poultry. Broilers fed diets containing caprylic acid (C8) at 0.7% (14 lbs/ton) and 1% (20 lbs/ton) had reduced Salmonella numbers in the gastrointestinal tract, liver, and spleen at 5–10 days PI (Kollanoor-Johny et al., 2009, 2012). Similarly, layers supplemented with caproic acid (C6) at 3 g/kg of feed (6 lbs/ton) had lower Salmonella numbers in the ceca, liver, and spleen at 3 days PI (Van Immerseel et al., 2004a). In this study, we showed that turkeys provided a combination of MCFA (C6-C12) at 6 lbs/ton of feed had reduced Salmonella colonization in the gastrointestinal tract at 3 days PI. All of the above treatments resulted in a 1-log10 CFU/g, or greater, reduction of Salmonella in the ceca. Collectively, these results indicate that MCFA decrease Salmonella colonization in poultry. Feed inclusion rates as low as 6 lbs/ton appear to be effective and may further enhance the practicality of using such products in commercial poultry production.

Conclusion

For this study a bioluminescent Salmonella infection model was used for in vivo screening of a MCFA product in turkeys. Early colonization was evaluated because previous studies have indicated that treatment during this time may be particularly effective for controlling Salmonella within a flock. Birds were infected with bioluminescent Salmonella Typhimurium on day-of-hatch and provided feed containing MCFA ranging from 0 to 6 lbs/ton. Significant decreases in both bioluminescence and Salmonella numbers (1-log10 CFU/g reduction) were detected in the 6 lbs/ton MCFA treated poults at 3 days PI. This was the first study to evaluate the effect of MCFA on Salmonella in turkeys and these results indicate that MCFA supplementation can reduce early colonization. Further studies are needed to determine what impact MCFA treatments have on Salmonella colonization throughout the production cycle.