Effect of a High Proportion of Rye in Compound Feed for Reduction of Salmonella Typhimurium in Experimentally Infected Young Pigs

Public health concerns and the potential for food-borne zoonotic transmission have made Salmonella a subject of surveillance programs in food-producing animals. Forty-two piglets (25 d of age and initially 7.48 kg) were used in a 28 d infection period to evaluate the effects of a high proportion of rye on reducing Salmonella Typhimurium. Piglets were divided into two diet groups: control diet (wheat 69%) and experimental diet (rye 69%). After a one-week adaptation period, all piglets were orally infected with Salmonella Typhimurium (107 log CFU/mL; 2mL/pig). Salmonella in fecal shedding were evaluated at day 1, 3, 5, 7 and then weekly after infection. At the end of the experimental period (at day 28 after infection), the piglets were euthanized to sample feces, cecal digesta contents and ileocecal lymph nodes to determine the bacterial counts of Salmonella. The results suggest that the bacterial counts in the experimental group fed rye diets showed evidence of reducing Salmonella fecal shedding from day 14 onwards and decreasing the number of Salmonella in cecal digesta. However, the translocation of Salmonella in ileocecal lymph nodes was not affected. Furthermore, feed intake, weight gain and feed conversion did not differ between the groups (p > 0.05).


Introduction
Salmonellosis is primarily a zoonosis and interventions are possible at any stage from farm to fork [1]. Salmonella enterica serovar Typhimurium is strongly associated with pigs, with infections in humans caused by contaminated pork and meat products, causing an impact on public health [2,3]. In Germany, 33% of acquired infections or 13,529 cases of salmonellosis were reported in 2018 [4], making salmonellosis the second most commonly reported bacterial gastrointestinal disease in Germany after Campylobacter enteritis [4].
With regard to reducing the number of Salmonella cases in the feed-to-food chain, it has frequently been demonstrated that the use of antimicrobial agents in food animals favors the development of resistance among foodborne pathogens like Salmonella spp. [5]. Since antimicrobial resistance is on Germany Sales GmbH & Co. KG, Weilheim, Germany) during the respective trial period (24.5 ± 0.8 • C) and the lighting program in the stable was set at a twelve-hour day and night rhythm, so that the light was switched on from 07:00 to 19:00.
Before beginning the trials, the stable and all materials were disinfected and tests were carried out to confirm that they were free of Salmonella species contamination. To rule out Salmonella contamination of the feed and thus entry thereof via the feed into the infection stable, the feed batches were examined in advance for Salmonella. To prevent cross-contamination between the individual pens, each pen was equipped with its own utensils (brooms, trays, spatulas). Protective clothing, disposable boot covers and gloves could be changed before entering the corresponding area of the stable. The positions of the groups in the stable were changed between the trials.

Diets
The composition of diets is described in Table 1. The piglets were allocated to two groups (control: N = 21, n = 7 and experimental: N = 21, n = 7) and fed ad libitum with the complete pelleted feed depending on whether wheat served as the control group (contained 69% wheat) or rye (experimental group; contained 69% rye, Table 1), according to the previous result from Wilke [14]. Feed chemical composition as well as particle size distribution of the diets are summarized in Table 2. Diets were analyzed by standard procedures in accordance with the official methods of the Verband Landwirtschaftlicher Untersuchungs-und Forschungsanstalten (VDLUFA [23]). The standardized methods of the Department of Animal Science, Aarhus University, Denmark were used to evaluate the NSP and arabinoxylans levels. Feed particle size distribution was assessed by a wet-sieve method of Wolf et al. [24].

Experimental Design
The Salmonella Typhimurium strain in this study was obtained from the field study [25]: S. Typhimurium (antigenic formula: 1,4,5,12:i:1,2, Phage type DT 193). Three consecutive trials of the infection experiment followed in accordance with the same investigation scheme (Figure 1). For the experiment carried out after a one-week adaptation period, all piglets were orally infected (day 0) with 2 mL of a broth containing~1 × 10 7 colony-forming units (CFU) S. Typhimurium/animal directly into the throat using a drencher as already described [26]. After the infection, animals were used in a 28-day experiment to determine the effects of control (69% wheat) or experimental group (69% rye) in complete pelleted feed, and to determine a possible effect of the diets on the duration of and level of Salmonella excretion. Fecal samples of the piglets were examined microbiologically for Salmonella species at defined time points (at 1, 3, 5, 7, 14, 21 and 28 days post-infection (dpi), as shown in Figure 1) to evaluate the status of the Salmonella infection. At the end of the experimental period (28 dpi), cecal chyme and lymph nodes were tested for Salmonella.
During the experimental phase, performance parameters, i.e., feed intake, were recorded on a weekly basis (Figure 1). To prevent cross-contamination between the animals during the infection, animals were weighed before the infection (0 dpi) and again at the end of the experiment (28 dpi; Figure 1). Corresponding parameters, average body weight gain (BWG) and feed conversion ratio (FCR) were determined.
In accordance with standardized Salmonella diagnostics, a blood sample was taken from each animal on the day of the experimental infection and at the end of the experiment (Figure 1). Serum was examined serologically for specific antibodies against Salmonella; Salmonella antibody ELISA (IDEXX Swine Salmonella antibody Test, IDEXX Europe B.V., Hoofddorp, the Netherlands) was used. The cut-off value was ≥10% optical density (OD).

Bacteriological Analyses and Salmonella Detection
Samples were taken for bacteriological analysis, and bacterial counts were determined for Salmonella detection. All test sections were carried out following the DIN EN ISO 6579-1 and 6579-2 guidelines [27,28]. Briefly, all collected samples were first placed for pre-enrichment in buffered peptone water (BPW; Oxoid Deutschland GmbH, Wesel, Germany), the ratio of which was 1:10, based on the volume of fecal or cecal chyme sample or the mass of the lymph node. For the next process of qualitative analysis, an incubation overnight at 37 • C took place. Each BPW-inoculated sample was placed with three drops (100 µL/drop) on modified semisolid Rappaport-Vassiliadis agar (MSRV; Oxoid Deutschland GmbH, Wesel, Germany) and incubated for a further 24 h (h) at 41 • C. After the incubation period, the evaluation was carried out macroscopically. If the result was positive, a white-grayish cloudy swarming zone spread over the entire agar resulting from the drops. Suspected Salmonella samples were also spread on the second selective nutrient media, xylose lysine deoxycholate agar (XLD; Oxoid Deutschland GmbH, Wesel, Germany) and Brilliance Salmonella agar (Oxoid Deutschland GmbH, Wesel, Germany) so that a confirmed result could be read after a 24 h incubation period.
In addition, Salmonella quantitative analysis of the fecal and cecal contents was performed as previously described [28,29]. Due to the high number of samples, a most probable number (MPN) method was used. One gram of the homogenized fecal or cecal content and 9 mL of BPW were vortexed. The bacterial count of the sample material was determined by a serial dilution with BPW in a deep well block (Sarstedt AG & Co, Nümbrecht, Germany). Quantitative proof of the dilution steps in a microtiter plate (Sarstedt AG & Co, Nümbrecht, Germany) was carried out in triplicate, as already described in detail [26]. After a 24 h incubation at 37 • C, the total volume of each well was transferred to another deep well block filled with MSRV agar and incubated for 24 h at 41 • C. The results were confirmed by cultural cultivation on Brilliance Salmonella agar, and subsequently the number of bacteria was calculated by an MPN software program [30]. To confirm the experimentally used Salmonella strain, salmonella-like colonies on Brilliance Salmonella agar were finally excluded by serotyping.

Statistical Analysis
The SAS software package version 7.1 (SAS Institute, Cary, NC, USA) was used for the statistical evaluation. The evaluation was carried out in cooperation with the Institute for Biometry, Epidemiology and Information Processing of the University of Veterinary Medicine Hannover, Foundation. Measurements such as mean values and standard deviations were calculated for the descriptive statistics. Salmonella results and performance parameters, i.e., body weight and feed intake, were analyzed at the level of the individual animal.
For evaluating the quantitative parameters between the control and experimental groups, two-way analysis of variance (ANOVA) with group and trial as independent factors was conducted. For analysis differences concerning distribution in qualitative parameters (Salmonella detection in cecal content, etc.), the chi-squared homogeneity test was used. Differences with a significant level of p < 0.05 were considered significant.

Salmonella Prevalence
The results of this study demonstrated that at the beginning of the post-infection period (1, 3, 5, 7 dpi), all the pigs were colonized. There were no significant differences in mean bacterial counts (log 10 ± SD) of Salmonella in the feces between the wheat and rye groups. A peak in Salmonella shedding occurred at day 5 after infection in our study. However, at day 14 onwards after infection, feeding a pellet diet containing 69% rye was associated with a significantly lower bacterial count of Salmonella in fecal samples than in those taken from the group fed with 69% wheat in the diet, as shown in Figure 2 and in detail in Table S1 (control and experimental, 14 dpi: 3.30 a ± 0.50 and 2.62 b ± 0.18 (p < 0.001), 21 dpi: 3.11 a ± 0.36 and 2.40 b ± 0.65 (p < 0.001), 28 dpi: 3.02 a ± 0.45 and 2.36 b ± 0.57 (p = 0.001), respectively). content, etc.), the chi-squared homogeneity test was used. Differences with a significant level of p < 0.05 were considered significant.

Salmonella Prevalence
The results of this study demonstrated that at the beginning of the post-infection period (1, 3, 5, 7 dpi), all the pigs were colonized. There were no significant differences in mean bacterial counts (log10 ± SD) of Salmonella in the feces between the wheat and rye groups. A peak in Salmonella shedding occurred at day 5 after infection in our study. However, at day 14 onwards after infection, feeding a pellet diet containing 69% rye was associated with a significantly lower bacterial count of Salmonella in fecal samples than in those taken from the group fed with 69% wheat in the diet, as shown in Figure 2 and in detail in Table S1 (control and experimental, 14 dpi: 3.30 a ± 0.50 and 2.62 b ± 0.18 (p < 0.001), 21 dpi: 3.11 a ± 0.36 and 2.40 b ± 0.65 (p < 0.001), 28 dpi: 3.02 a ± 0.45 and 2.36 b ± 0.57 (p = 0.001), respectively). After oral infection with S. Typhimurium, different diets had significant influence on the counts of Salmonella in the cecal content ( Table 3). The bacterial counts in the cecal content also differed significantly (log10 CFU/g; control group: 3.34 a ± 0.50, experimental: 3.08 b ± 0.56; p = 0.038; Table 3).
However, cecal content and ileocecal lymph node samples were qualitatively Salmonella positive (control: 100% and 61.9% and experimental: 100% and 66.7%, respectively; Table 3), but no differences could be seen between the two groups.  After oral infection with S. Typhimurium, different diets had significant influence on the counts of Salmonella in the cecal content ( Table 3). The bacterial counts in the cecal content also differed significantly (log 10 CFU/g; control group: 3.34 a ± 0.50, experimental: 3.08 b ± 0.56; p = 0.038; Table 3).
However, cecal content and ileocecal lymph node samples were qualitatively Salmonella positive (control: 100% and 61.9% and experimental: 100% and 66.7%, respectively; Table 3), but no differences could be seen between the two groups. Table 3. Number of Salmonella-positive samples from cecal content and ileocecal lymph nodes at dissection after oral infection with S. Typhimurium.

Salmonella Test Control Experimental
Cecal content Quantitative log 10 CFU/g 3.34 a ± 0. Three consecutive experiments were conducted using piglets from different sows. The results of the counts of Salmonella (log 10 CFU/g fecal sample) for different trials and different diets are shown in Table 4. After initial statistical evaluations, a marked effect for the factor trial was shown, with significant effects at 1 dpi (p < 0.001), 7 dpi (p = 0.036), 21 dpi (p = 0.001) and 28 dpi (p < 0.001). An effect of diet seems to be present for days 14 (p < 0.001), 21 (p < 0.001) and 28 (p < 0.001) after infection. In addition, a bacterial count for Salmonella in cecal content was significantly affected by the diet (p = 0.035). Interestingly, the two-way analysis of variance (ANOVA) showed significant interactions between the factors diet and trial, which were observed for days 5 and 7 (p = 0.011 and p = 0.041, respectively; Table 4). On the other hand, there was not a significant effect of the type of diet by the trial interaction at day 3 after infection (p = 0.188; Table 4).

Serological Test for Specific Antibodies Against Salmonella
Optical density (OD)% was used to determine the Salmonella antibody status in blood samples from piglets. In both groups, on the day before infection, all of the pigs were seronegative. At day 28 after infection, regarding the number of positive blood samples from piglets, there was no significant difference in seroprevalence detected between the groups (p > 0.05; Table S2).

Animal Performance
As a result of the evaluations carried out, there were no significant differences in the performance parameters between the groups. During the experimental period, before and after infection, there were no significant differences in the body weight (BW), as shown in Figure 3 and in detail in Table 5 and Supplementary Table S3. Furthermore, feed intake was measured individually for the entire trial period. Differences between the groups regarding the mean weekly feed intake (FI) were not statistically significant (control and experimental; mean FI at week 5 (21-28 dpi, in kg): 9.78 a ± 1.52 and 9.40 a ± 1.48; p = 0.136; Figure 3 and Table S3).

Animal Performance
As a result of the evaluations carried out, there were no significant differences in the performance parameters between the groups. During the experimental period, before and after infection, there were no significant differences in the body weight (BW), as shown in Figure 3 and in detail in Table 5 and Supplementary Table S3. Furthermore, feed intake was measured individually for the entire trial period. Differences between the groups regarding the mean weekly feed intake (FI) were not statistically significant (control and experimental; mean FI at week 5 (21-28 dpi, in kg): 9.78 a ± 1.52 and 9.40 a ± 1.48; p = 0.136; Figure 3 and Table S3).      (28 dpi). No significant differences were noted between the groups for BW at each time point (0 dpi (p = 0.085) and 28 dpi (p = 0.980)) and body weight gain (BWG; Table 5). The average daily weight gain (ADG) in the control group was about 639 g and in the experimental group, 629 g (Table 5). Similarly, the feed conversion ratio (FCR = feed requirement in kg per kg BWG) was not significantly affected between the wheat and rye diet groups (1.50 and 1.51, respectively).

Discussion
In recent years, the interest in rye has increased in terms of its good sustainability [31]. However, the use of rye in large amounts in diets for pig production is not typical because it can be associated with ergot poisoning [32]. On the other hand, rye has a high dietary fiber content [17], the fraction of NSPs in rye being more fermentable for monogastric animals than is the case for other cereals [13]. Nowadays, varieties of rye have been developed which reduce its susceptibility to ergot contamination; therefore, we can include rye in diets for swine [11,33].
The present study shows results of a feeding concept in young pigs fed with a high proportion of rye, and its effects on Salmonella prevalence, along with analyzing its effects on performance. In terms of Salmonella shedding, a significant reduction in the number of bacterial counts of Salmonella was seen in feces from 14 day onwards when considering the entire infection period for the experimental group. Feeding a diet containing 69% rye to young pigs may play a significant role in reducing Salmonella shedding after infection. In addition, the prevalence of Salmonella was found to be significantly lower in the cecal chyme of pigs fed with rye diets. Hence, if there was a reduction in the incidence of Salmonella in feces of pigs, which is the most likely source of contamination in the pig production chain, then it may help to reduce the risk of human salmonellosis.
To our knowledge, fermented rye has been previously studied to control Salmonella in nursery pigs [34]. Fabà et al. [34] reported that pigs fed with organic acids combined with fermented rye had reduced S. Typhimurium shedding compared to pigs fed with organic acids plus coated butyrate (3.11 and 3.87 log 10 CFU/g, respectively) over the 21 d period post-challenge with 10 9 CFU/mL. However, the pigs in our study were infected with 10 7 CFU of S. Typhimurium, so comparison across infection studies is difficult. In this respect, an infection dosage in the present study could provide more opportunities to determine the effects of diet, which corroborates with other studies [8,35,36]. The S. Typhimurium oral challenge resulted in an effective infection in our study. At the beginning of the trial, all the pigs in the study showed a Salmonella-negative result. Then, after oral infection, nearly all infected animals excreted Salmonella until the end of the study. In addition, it is assumed that the challenge dose in practice or natural infection is often established with low or moderate numbers of Salmonella when compared with our study [36]. However, the scientific literature on the usefulness of a high proportion of rye in diets in a dry and pelleted form for Salmonella reduction in livestock animals, specifically young pigs, is scarce.
A reduction in Salmonella shedding in pigs in this study was possibly related to the components in rye such as dietary fiber, including NSPs. Compared to the total concentration of NSPs and arabinoxylans between groups, the mean concentration was higher in rye in the present study. The concentration of total NSPs was 123 g/kg dry matter (DM) in wheat and 140 g/kg DM in rye. In addition, the mean concentration of arabinoxylans was 63 g/kg DM and 74 g/kg DM in wheat and rye diets, respectively ( Table 2). The concentrations of carbohydrate fractions in wheat and rye were in general agreement with those values of previous studies [17,18]. The NSP content of rye has traditionally been put forward to explain the effects on gut health benefits for pigs, especially arabinoxylans (8-9%) and fructans (approximately 3% and range up to about 6%) [33,37]. Rye has greater concentrations of arabinoxylans and fructans than other cereal grains [38,39] that can be changed by microorganisms to butyrate in the digestive tract [10,17].
According to Wilke [14], the effects in terms of nutritional physiology have been evaluated. In the group with 69% rye in the diet, significantly higher concentrations of lactate in the anterior digestive tract were found, as well as an increased entry of fermentable organic substances in the large intestine (approx. 35% higher butyrate concentrations) in the ingesta of cecum and colon, the preferred colonization area of Salmonella, in comparison to those animals given a wheat-based diet [22]. How rye affects Salmonella colonization is unclear. Consequently, the antimicrobial effects of lactate and butyrate seem to play an important role against Salmonella [14]. High butyrate concentrations in the hindgut have been found when a high amount of rye in diets is given to young pigs [14]. The focus on the mechanism of action of butyrate in reducing Salmonella has been explained in various studies [7,9,19,40,41]; similar mechanisms possibly play a role in our study. Further investigations are needed to better elucidate the Salmonella shedding reduction and other health-promoting mechanisms potentially associated with rye on the physiological processes in the gastrointestinal tract and the stability of the intestinal microbiota of pigs.
Less fecal Salmonella excretion indicates less colonization of Salmonella in the intestine and reduced severity of the infection [42,43]. In addition, in our study, the lymph node was used to confirm colonization. A translocation of Salmonella from the intestine into the adjacent ileocecal lymph nodes was investigated in all experimental runs. The piglets were totally Salmonella positive at the time of dissection (in cecal chyme), but positive evidence in lymph nodes was found in twenty-seven of a total of forty-two animals (control: n = 13; experimental: n = 14). Less is known about the role of rye in the translocation effect of Salmonella in the pig model. However, in previous experiments with poultry, this was determined with other pathogens such as Escherichia coli and Clostridium perfringens [44,45]. When observing the elevated bacteria translocation while feeding a diet rich in NSPs [44], the prolonged transit time of the intestinal contents possibly induces proliferation of pathogenic bacteria. Moreover, Tellez et al. [45] also reported that a rye diet negatively influences bacterial translocation and intestinal viscosity in poultry. Nevertheless, the viscosity does not pose a large problem in swine as much as in poultry [46].
The outcome of the current study showed that the use of compound feeds with higher dietary proportions of rye in young pigs caused no significant differences in performance, i.e., average body weight gain, feed intake and feed conversion ratio compared with those of animals fed with generally high amounts of wheat. Similar results were obtained by Grone [15] and Wilke [14], who observed no negative impact on growth performance when young pigs were fed diets containing a high proportion of rye. Previous studies have shown positive effects of feeding rye on performance data [46][47][48]. Among the beneficial effects of the dietary fiber components in rye, it can provide energy to the pig mostly via the digestion of starch and fermentation of fiber in the hindgut. Moreover, the NSPs in rye may promote greater butyrate production and improve intestinal health [33]. Furthermore, scientific studies have evaluated the feed digestibility in pigs concerning diets containing rye [48]. The rye-based diets had higher digestibility coefficients for dry matter, crude protein and gross energy than barley when rye was included in a young growing pig diet at 60% [49]. Thacker et al. [46] suggest that rye may have much more potential in pig performance when administered in pellet form. Furthermore, to improve feed efficiency and average daily gain, the NSP enzyme would be recommended for diets containing high rye levels [47]. Contrary to the present results, a previous study on the use of rye in pig feeding reported that if more than 50% of the barley or wheat in a swine diet was replaced with rye, a significant reduction in pig performance was observed [50]. The reason for rye being less palatable is that it is more bitter than other small grains such as wheat [51]. Alkylresorcinols in rye may be responsible for the unpleasant taste, but reduced levels of alkylresorcinols are found in hybrid rye [11].

Conclusions
The results of this study support the view that a high proportion of rye might contribute to reducing Salmonella shedding via feces in young pigs. In addition, overall, the animal performance was good, so that no relevant differences could be observed between the rye and wheat diets. It can be suggested that up to 70% rye in compound pelleted feed poses no problem for animal health and welfare, additionally decreasing possible contamination with Salmonella. Another advantage that puts rye in a better light today than in the past is its sustainable benefits, namely, it needs less fertilizer, demands substantially less water, can reduce CO 2 emissions in pork production in comparison with wheat [52] and its price, as it is an inexpensive raw feed material compared to other cereals [11].
Nonetheless, to our knowledge, this is the first time that high amounts of rye were used in a Salmonella challenge in a young pig model. Furthermore, our results can provide useful information, prompting further studies of a field trial on a commercial pig farm or large-scale farming study over a longer period, not only in young pigs, but also in all phases of pig production. However, we cannot consider all factors when only a small group of pigs are observed. Therefore, research is needed to better understand the gut and fecal microbiota composition underlying the effects observed when a high proportion of rye is fed to pigs.
Supplementary Materials: The following are available online at http://www.mdpi.com/2076-2607/8/11/1629/s1, Table S1: Counts of Salmonella (log 10 CFU/g fecal sample) of piglets after an experimental oral infection with Salmonella Typhimurium for different diets. Table S2: Salmonella antibody status in serum of piglets in control and experimental groups. Table S3: Performance data during the entire experimental period.