Protective effect of mannan oligosaccharides against early colonization by Salmonella Enteritidis in chicks is improved by higher dietary threonine levels



Dr. Patrícia Emília Naves Givisiez, Department of Animal Science, Universidade Federal da Paraíba, Caixa Postal 13, Centro, Areia, PB 58397-000, Brazil. E-mail:,



To evaluate mannan oligosaccharide (MOS) and threonine effects on performance, small intestine morphology and Salmonella spp. counts in Salmonella Enteritidis-challenged birds.

Methods and Results

One-day-old chicks (1d) were distributed into five treatments: nonchallenged animals fed basal diet (RB-0), animals fed basal diet and infected with Salmonella Enteritidis (RB-I), animals fed high level of threonine and infected (HT-I), birds fed basal diet with MOS and infected (MOS-I), birds fed high level of threonine and MOS and infected (HT+MOS-I). Birds were inoculated at 2d with Salmonella Enteritidis, except RB-0 birds. Chicks fed higher dietary threonine and MOS showed performance similar to RB-0 and intestinal morphology recovery at 8 dpi. Salmonella counts and the number of Salmonella-positive animals were lower in HT+MOS-I compared with other challenged groups.


Mannan oligosaccharides and threonine act synergistically, resulting in improved intestinal environment and recovery after Salmonella inoculation.

Significance and Impact of the Study

Nutritional approaches may be useful to prevent Salmonella infection in the first week and putative carcass contamination at slaughter. This is the first report on the possible synergistic effect of mannan oligosaccharides and threonine, and further studies should be performed including performance, microbiota evaluation, composition of intestinal mucins and immune assessment.


Salmonella enterica is the most important foodborne pathogen transmitted by consumption of contaminated poultry and eggs in the United States of America (Altekruse et al. 2006) and Brazil (Brasil 2008). From 1999 to 2007, there have been 6,000 outbreaks of food poisoning in Brazil and 117,000 cases, from which 42% were caused by Salmonella and resulted in 64 deaths (Brasil 2008). Salmonella enterica Enteritidis (S. Enteritidis) and S. Typhimurium are considered to be responsible for most of the infections in humans (Van Immerseel et al. 2009). Poultry is regarded as the largest reservoir of S. Enteritidis and most risk attributions studies have identified poultry and poultry products as the major source of human infection in the United States of America (Van Immerseel et al. 2009; Shah et al. 2011).

Antimicrobial growth promoters have been extensively used in animal production and have contributed to the great achievements of the poultry industry in regard to the modern poultry production indicators. However, the use of antibiotics in subtherapeutic doses has been associated with the emergence of microbial resistance, which poses a risk to human health (Baurhoo et al. 2009; Van Immerseel et al. 2009).

In view of the increasing demand for safe food by the consumers and considering the fact that the industry needs to keep current performance indicators, alternatives to antimicrobial growth promoters must be evaluated. Thus, there has been increasing interest in the use of probiotics and prebiotics. Prebiotics are nondigestible food ingredients that selectively stimulate the growth of endogenous bacteria such as Lactobacillus sp. and Bifidobacterium, which benefit the host (Van Immerseel et al. 2009). One of the most studied prebiotics is the phosphorylated mannan oligosaccharides (MOS), oligosaccharide complexes of the cell wall of Saccharomyces cerevisiae. There is evidence that MOS can adhere to bacterial fimbriae and thus block the adhesion of bacteria to the intestinal surface (Baurhoo et al. 2009). Therefore, prebiotics can change the intestinal microbiota of the host, which will result in the increased number of Bifidobacterium and Lactobacillus sp. (Sun et al. 2005). There is also evidence that they reduce intestinal turnover and modulate the immune system (Moran 2004). Such properties suggest a great potential to improve performance and decrease mortality in broilers.

Little is known about interactions of MOS with dietary nutrients, which is a paradox considering the enormous advances on poultry nutrition. Recently, a series of studies have reported a positive interaction between MOS and threonine on growth performance, mucosal development and mucin dynamics in broilers, suggesting greater absorptive area and functional ability of the intestine, which would be expected to improve digestion and absorption (Chee et al. 2010a,b,c).

Our hypothesis is that variations in the levels of threonine, the third limiting amino acid in broiler chickens fed corn and soybean meal-based diets, may interfere with the effects of MOS on chicks because it is intimately related to the production of mucins in the small intestine. It is known that mucins form a protective barrier against the action of digestive enzymes and minimize physical damage (Faure et al. 2007). Additionally, the amino acid threonine is present in higher proportion in immunoglobulin molecules (Lemme 2001).

Therefore, the aim of this study was to evaluate the effects of the inclusion of MOS in diets with different threonine levels on chick performance between 1 and 10 days of age (d), the morphology of the small intestine and Salmonella spp. counts after challenge at 2 days of age with Salmonella Enteritidis.

Materials and methods

Animals and experimental design

A hundred and twenty Cobb male chicks with 1 day of age and vaccinated against fowlpox, Marek and Newcastle were used. All management practices as well as slaughter and sampling procedures have been carried out in accordance with EU Directive 2010/63/EU for animal experiments. The birds were weighed (initial body weight) and distributed according to a completely randomized design into five treatments and three repetitions with eight birds per repetition. The treatments were a control group comprised of animals fed basal diet that were not challenged (RB-0), animals that received basal diet and were infected with Salmonella Enteritidis (RB-I), animals fed high level of threonine and infected (HT-I), birds fed basal diet with MOS and infected (MOS-I), birds fed high level of threonine and MOS and infected (HT + MOS-I).

A basal corn and soybean meal-based diet was formulated for the initial phase (1–21 days) in accordance with the recommendations of Rostagno et al. (2005), with 3,100 kcal kg−1 ME, 22% CP, 1·33% lysine, 0·94% digestible Met + Cys, 0·806% of threonine, 2·20% of Gly + Ser. The inert was replaced with 0·198% L-threonine Feed Grade 98·5% (Ajinomoto Animal Nutrition) to reach 0·955% of dietary threonine in the treatment with high threonine level (HT). MOS (0·1%) was added to the diets as recommended by the manufacturer (Bio-Mos®, Alltech, Lexington, KY, USA). The diets did not contain antibiotics or other drugs to prevent interference with the inoculum of Salmonella Enteritidis. Feed and water were provided ad libitum during the experimental period (10 days).

Challenge with Salmonella Enteritidis and slaughter

The challenge with Salmonella was performed using a nalidixic acid-resistant Salmonella Enteritidis (S. EnteritidisNal+). The inoculum was prepared using nutrient broth at 37°C for 4 h in a shaker and contained 1·2 × 108 colony forming units per millilitre (CFU ml−1). All birds were inoculated orally at two days of age (2d) using 0·5 ml of inoculum of SalmonellaNal+, except the birds of the control group, that were given 0·5 ml of nutrient broth without Salmonella. At 8 days postinfection (8 dpi), final body weight was recorded and the animals were killed by cervical dislocation.

Performance data

Initial (IW) and final (BW) body weights were recorded individually at 1 and 10 days of age and were used to calculate body weight gain (BWG). Feed intake (FI) was calculated as the difference between the food provided and the food present in the feeders at 10 days. Feed conversion (FC) was calculated as FI divided by BWG (g g−1).


Three samples with approximately 2 cm were taken from the medial region of duodenum, jejunum and ileum and routinely processed for histology. Tissue was fixed in Bouin's solution, dehydrated, cleared and infiltrated with paraffin wax. After embedding, cuts with 5 μm of thickness were sectioned and mounted on glass microscope slides to be stained with haematoxylin and eosin. The cuts were observed under 10× magnification using a light microscope (Nikon E-100 equipped with a digital camera), and photographs were taken. Villus height and crypt depth were determined using the software Image J (Abramoff et al. 2004); villus height was measured from the top of the villus to the villus-crypt junction. The crypts were measured from the villus-crypt junction until the baseline. Thirty vertically oriented villi and thirty crypts were measured per sample, with a total of 120 readings per variable for each treatment.

Microbiological analysis

For microbiological analysis, caecal contents of the birds of each treatment (n = 24/treatment) were collected and weighed, and a serial dilution was prepared with phosphate buffer solution (PBS) pH 7·2. Aliquots of the dilutions were cultured on Brilliant Green Agar containing nalidixic acid (100 μg ml−1), and the plates were incubated at 37°C for 24 h. Colonies were counted, and values expressed in colony forming units per gram of caecal contents (CFU g−1) and were transformed to log10 before statistical analysis.

Data analysis

Performance data were analysed according to a completely randomized design with five treatments and three repetitions with eight birds per repetition, whereas bacterial counts were analysed considering four treatments, because the control treatment was removed. Histology data were analysed according to a completely randomized design with five treatments and 120 repetitions per treatment (thirty readings for each sample taken). Analysis of variance was performed using statistical software (SAEG 2007), and the variables that showed statistical differences for the F test were compared by Tukey's test at 5% probability.


Performance results are presented in Table 1. There were no differences (P > 0·05) between treatments for IW, FI and FC. However, higher final body weight (≤ 0·05) was observed in the animals challenged with Salmonella Enteritidis and fed high threonine levels and MOS (HT+MOS-I, 254·1 g) compared with inoculated animals fed the basal diet (RB-I, 232·8 g) and only MOS (MOS-I, 227·2 g). Inoculated animals fed high threonine alone (HT-I) or noninoculated animals (RB-0) showed intermediate body weight values that were not different (P > 0·05) from the other treatments. Similarly, higher BWG was observed in the treatment fed high threonine and MOS (204·87 g) compared with treatments RB-I (184·12 g) or MOS-I (177·25 g), and there was no difference in relation to other treatments (Table 1).

Table 1. Means (±SD) of initial individual body weight (IW, g), final individual body weight (BW, g), body weight gain (BWG, g bird−1 day−1), feed intake (FI, g bird−1 day−1) and feed conversion (FC, g g−1) of chicks fed different dietary levels of threonine and mannan oligosaccharide and challenged or not with Salmonella Enteritidis
TreatmentaMean valuesb
IW (g)BW (g)BWG (g bird−1 day−1)FI (g bird−1 day−1)FC (g g−1)
  1. a

    RB-0, noninoculated control group; RB-I, birds fed basal diet and infected with Salmonella Enteritidis; HT-I, high level of threonine and infected; MOS-I, basal diet with MOS and infected; HT + MOS-I, high level of threonine and MOS in diet and infected.

  2. b

    Means followed by different letters in the column are different by Tukey's test at 5% of probability.

RB-049·37 ± 2·21a242·12 ± 4·40ab192·75 ± 4·71ab314·01 ± 71·02a1·66 ± 0·55a
RB-I48·75 ± 1·89a232·87 ± 14·91b184·12 ± 14·40b331·94 ± 70·03a1·86 ± 0·60a
HT-I49·50 ± 2·17a240·12 ± 11·85ab190·50 ± 12·02ab319·32 ± 93·49a1·70 ± 0·30a
MOS-I49·87 ± 2·21a227·25 ± 10·00b177·25 ± 10·07b348·02 ± 27·27a1·99 ± 0·39a
HT+MOS-I49·25 ± 1·17a254·12 ± 11·43a204·87 ± 11·22a271·29 ± 38·85a1·33 ± 0·08a

Histology results from intestinal segments sampled in the present study indicate that the birds challenged and fed basal diet (RB-I) showed greater crypt depth in duodenum. On the other hand, birds fed either higher threonine levels (HT-I) or MOS (MOS-I) had greater (≤ 0·05) villus height than noninoculated birds, whereas birds fed both additives (HT+MOS-I) showed duodenal villus height and crypt depth similar to control birds, indicating that intestinal integrity was improved when MOS or Thr was added (Table 2). Greater and lower villus : crypt ratios (≤ 0·05) in duodenum and ileum were shown by birds fed both additives (HT+MOS-I) and animals fed the basal diet and challenged (RB-I), respectively (Table 2).

Table 2. Villus height (μm), crypt depth (μm) and villus : crypt ratios (μm μm−1) of duodenum, jejunum and ileum of chicks fed different dietary levels of threonine and mannan oligosaccharide and challenged or not with Salmonella Enteritidis. Values represent means ± SD of 120 readings
TreatmentaVillus height (μm)Crypt depth (μm)Villus : crypt ratio (μm μm−1)
  1. a

    RB-0, noninoculated control group; RB-I, birds fed basal diet and infected with Salmonella Enteritidis; HT-I, high level of threonine and infected; MOS-I, basal diet with MOS and infected; HT + MOS-I, high level of threonine and MOS in diet and infected.

  2. b

    Means followed by different letters in the column are different by Tukey's test at 5% of probability.

RB-01097·31 ± 39·35b132·63 ± 11·32d8·36 ± 0·82b
RB-I1177·36 ± 26·65ab212·17 ± 11·83a5·42 ± 0·32d
HT-I1214·93 ± 68·75a178·11 ± 9·71b6·89 ± 0·55c
MOS-I1219·57 ± 31·50a156·02 ± 19·04c7·95 ± 1·92bc
HT + MOS-I1174·45 ± 75·13ab119·69 ± 30·95d13·35 ± 2·47a
RB-0871·24 ± 37·07a149·89 ± 7·75a5·69 ± 0·48d
RB-I854·14 ± 16·01a113·08 ± 6·33c7·56 ± 0·48c
HT-I800·85 ± 22·79b144·11 ± 6·79b5·52 ± 0·22d
MOS-I789·02 ± 50·79b72·04 ± 6·19e10·99 ± 1·46a
HT + MOS-I699·17 ± 12·22c78·86 ± 3·14d8·83 ± 0·64b
RB-0582·48 ± 18·99a78·37 ± 3·77b7·36 ± 0·47b
RB-I428·33 ± 26·66c114·50 ± 13·93a3·82 ± 0·55d
HT-I430·75 ± 7·03c76·90 ± 3·85b5·54 ± 0·28c
MOS-I477·54 ± 14·48b63·46 ± 3·15c7·49 ± 0·42b
HT + MOS-I415·12 ± 12·77d51·26 ± 2·12d8·19 ± 0·33a

Animals fed diets containing MOS and higher levels of threonine had significantly lower mean counts of Salmonella Enteritidis in the caecal contents (4·68) compared with animals fed basal diet (7·20) (Table 3). It can also be observed that Salmonella Enteritidis was recovered from the caecal contents of 23 birds on the basal treatment, while 17 birds were positive when both additives were used.

Table 3. Enumeration of Salmonella Enteritidis (Log10 CFU g−1) in caecal contents of chicks fed different dietary levels of threonine and mannan oligosaccharide and challenged with Salmonella Enteritidis
TreatmentaBirds colonized/Total number of birdsAverage colony forming units (Log10)b
  1. a

    RB-I, birds fed basal diet and infected with Salmonella Enteritidis; HT-I, high level of threonine and infected; MOS-I, basal diet with MOS and infected; HT + MOS-I, high level of threonine and MOS in diet and infected.

  2. b

    Means followed by different letters in the column are different by Tukey's test at 5% of probability.

HT + MOS-I17/244·68b


The results of this study indicate that early intestinal colonization by S. Enteritidis does not necessarily compromise the performance of chicks, because performance was similar between the control group and the challenged group fed the same diet. The improved early performance observed in chicks fed higher threonine, and MOS indicates a synergistic effect, because the results when MOS was used alone were not different from the results observed in inoculated chicks given basal diet. Similarly, Chee et al. (2010a) have reported that MOS significantly prevented reduction in BWG caused by excess threonine in nonchallenged birds up to 35 days. However, the mechanisms responsible for such effect remain to be clarified.

In this sense, some studies have shown changes in intestinal morphology in animals fed diets supplemented with additives, and this is an important parameter to be evaluated. Our histology results indicated greater tissue injury caused by Salmonella and that the organism responded with greater extrusion in the villus apex in the duodenum and ileum. This is a common response during intestinal infections and must be compensated by greater activity in the crypts, which become deeper, to maintain the functional integrity of the intestine (Domeneghini et al. 2006). In such situation, energy is deviated from production to cope with the greater demand on tissue repair, resulting in lower BWG and final body weight. These results are further corroborated by the data of villus : crypt ratio (V : C ratio). The ratio is considered ideal when higher ratios are present as a result of greater villus height and shallower crypt depth, consequently promoting greater absorption of nutrients (Xu et al. 2003). Borsoi et al. (2011) have reported lower V : C ratios in the ceca of birds challenged with Salmonella Enteritidis and Salmonella Heidelberg compared with nonchallenged birds from 0 to 3 dpi, possibly because of injury to the mucosa.

Although being challenged with Salmonella, the birds fed high threonine and MOS have similar performance (Table 1) and duodenal morphology (Table 2) to birds that were not challenged at all. This might have been due to increased immunity (Faure et al. 2006) and improvement in intestinal environment to beneficial bacteria, as shown by Baurhoo et al. (2007), besides greater bacterial adsorption. The final result is improved performance, which probably reflects the tissue recovery that was seen on the morphology of duodenum, evidenced by V : C ratio. The maturation of the gut function in nonchallenged birds was apparently accelerated when MOS and adequate levels of threonine were added to the diet, with higher V : C ratio (Chee et al. 2010b).

It is noteworthy to mention that the treatment effects were very similar if performance and Salmonella colonization are considered, suggesting that the synergistic effect could also be linked to gut microflora reduction. Gut health is not only related to physical development, but also is determined by the microflora that is present in the gut (Choct 2009). Although much has to be discovered about the physiological mechanisms of MOS in the gut, it is known that prebiotics and other feed additives can modulate the gut microflora and performance in broilers (Choct 2009) and that the protective effect of MOS in gut is primary linked to its binding properties to bacteria. Threonine is a key amino acid used for mucin production, which is also known to reduce colonization of bacteria in rats by decreasing adhesion to the mucosa (Faure et al. 2006). Threonine is considered the third limiting amino acid in broiler chickens fed corn and soybean mean-based diets and is an important component of immunoglobulins in birds, rabbits and humans (Wang et al. 2006). The present findings corroborate the importance of threonine, but its role on the improvement of immune response of challenged birds, alone or preferably in combination with MOS, must be further evaluated. Previous studies have also shown reduced Salmonella Enteritidis colonization in chicks fed diets with MOS (Spring et al. 2000; Fernandez et al. 2002). It has been attributed to MOS the ability to adhere to pathogen with Type I fimbriae (Moran 2004; Baurhoo et al. 2009), as occurs in 53% of Salmonella spp. (Sun et al. 2005). Additionally, the main mode of action of oligosaccharide is to modify the intestinal environment and increase in beneficial bacteria counts (Sun et al. 2005), because there is a reduction in competition for starch and sugar between microbiota and host, greater dietary energy availability and reduction in intestinal pH, suppressing bacteria proliferation such as Salmonella (Ferket 2004) and E. coli in broiler chickens (Baurhoo et al. 2007). A previous study (Biggs et al. 2007) showed that when 0·4% yeast cell wall was included in the diets of chicks, metabolizable energy was increased (≤ 0·05) at 21 days of age when compared with the corn-soybean basal diet, indicating that energy utilization may also be improved with the combination of ingredients that may have prebiotic effects.

Besides the effects of MOS on local immunity and integrity, their supplementation to the diet either alone or in combination with Lactobacillus-based probiotic has been shown to improve the humoral immunity of broilers against Newcastle disease virus and infectious bursal disease virus during heat stress (Sohail et al. 2010). On the other hand, the effect of this additive together with increased aminoacid levels has not been reported so far in challenged birds. In nonchallenged birds, it has been suggested that MOS has a positive effect on mucin dynamics, that is, increased mucin output that resulted in greater protection and improved mucosal development in the gut (Chee et al. 2010b). All these factors combined may have eventually resulted in better integrity and improved performance.

These results are corroborated by the bacterial counting results and the number of colonized animals in each treatment (Table 3). There was no significant difference on colonization reduction by S. Enteritidis when diet was supplemented only with MOS (Table 3). Although probiotics and prebiotics alter the intestinal microbiota and immune system to reduce colonization by pathogens in certain conditions, it has been shown that environmental and stress status can influence the efficacy of prebiotics and probiotics (Patterson and Burkholder 2003). Besides, different combinations of prebiotics have shown no consistent effect or synergistic effects on growth performance and caecal microbial populations. However, there is a lack of studies on how nutrient levels in the diet could influence the efficacy of prebiotics and probiotics on animal performance and reduction of colonization by pathogenic bacteria. The present study indicates that higher threonine-MOS supplemented diets improve performance and reduce early colonization by Salmonella Enteritidis in chicks. Decreasing the levels of Salmonella colonization in the animal gut or other tissues can help to reduce Salmonella infection pressure in the environment and/or in the birds (Van Immerseel et al. 2009).

The results presented herein are important in developing new strategies of controlling Salmonella spp. infection in chicks and, eventually, in the poultry chain. This issue is of great importance, because Salmonella spp. are not members of the intestinal microbiota, but chicks are rapidly colonized. Once the birds are exposed to S. Enteritidis, the entire flock can be colonized quickly (Shah et al. 2011) and early infected chicks might remain infected for different periods of time or even until slaughter age (Gast et al. 2005). Therefore, the risk of transmission to humans increases, especially because the majority of older infected birds are asymptomatic. The number of chicken slaughter establishments with Salmonella Enteritidis-positive broilers significantly increased in USA from 2000 to 2005 (Altekruse et al. 2006). Although measures to control Salmonella spp. contamination in slaughter plants have been undertaken (Altekruse et al. 2006), commercial growers have been pressured to prevent and/or eliminate Salmonella spp. in preharvest production facilities (Foley et al. 2011). In Brazil, since 2003, the Government established the regulations of the ‘Program of pathogen reduction –microbiological monitoring and control of Salmonella spp. in broiler and turkey carcasses’ to constantly monitor the contamination level in poultry slaughter facilities (Brasil 2003). From 2003 to March 2008, 128,293 samples have been collected, and it 6·39% of the poultry plants were positive for Salmonella (Brasil 2008). The identification of on-farm risk factors for Salmonella colonization in poultry is necessary to develop strategies for reducing Salmonella load in animals at time of slaughter and processing (Burkholder et al. 2008; Van Immerseel et al. 2009), and nutritional management may be an interesting approach, which include the use of additives such as antibiotics, probiotics, prebiotics, acidic compounds and competitive exclusion products (Van Immerseel et al. 2009).

In summary, data presented herein indicate the existence of a synergistic effect between threonine and MOS when the chicks were inoculated with Salmonella Enteritidis. The addition of MOS and higher threonine levels to the diet has resulted in performance recovery that was similar to noninfected birds (control group), which may have resulted from better intestinal integrity in those birds compared with the other inoculated groups. Finally, there was a decrease in the number of positive animals for Salmonella at 8 dpi and lower Salmonella counts when both additives were used. These findings might, at least in part, respond for the recovery of these birds and the better performance.

Future studies must focus on looking for the possible synergistic effect between the ingredients, besides trying to highlight the mechanism(s) underlying such interaction, and the effects on the microbiota and the intestinal mucosa. Thus, the response of the birds must be further evaluated, including other parameters of performance, composition of intestinal mucins and immune assessment.


The authors thank to Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for scholarship granted to E.G. Santos, to T. S. Santos for carrying statistical analyses, and D. B. Campos for critically reading the manuscript.