Weaning is a stressful process for pigs and typically causes a growth check, but the stressors of weaning can also render pigs more susceptible to gastrointestinal tract (GIT) diseases and dysfunction [1
]. Post weaning diarrhoea (PWD) is associated with the proliferation of enterotoxigenic β-haemolytic strains of Escherichia coli
], and can be a major cause of economic loss in a herd [4
]. The commensal bacteria in the intestine, including Lactobacillus
spp., play important roles in preventing the colonization of pathogens through competitive exclusion and excretion of bacteriocins capable of bacterial lysis [5
]. Antibiotics and (or) mineral compounds such as zinc oxide (ZnO) have traditionally been used to control PWD and GIT dysbiosis [6
]. Health and consumer concerns related to the development of antimicrobial resistance [9
], and concerns related to the accumulation of minerals such as zinc in the environment [6
], have resulted in the development of alternative strategies to maintain health and performance in the post-weaning period.
Organic acids in weaner pig diets can control PWD and enhance growth performance [12
]. Acetic, formic and propionic acids have a direct effect on Gram-negative bacteria, impeding the replication of deoxyribonucleic acid [14
]. Formic acid has been found to decrease colonization of enterobacteria in the lower part of the GIT of pigs [15
], increasing growth efficiency when supplemented to market pigs [16
] and having a positive effect on controlling PWD in pigs [17
]. Propionic acid has also displayed positive effects on controlling PWD and a general improvement in performance [17
]. Furthermore, acetic acid, when fed in a combination with formic, phosphoric and citric acid, was found to have positive effects on performance in newly-weaned pigs [18
Essential oils (EO) are phytochemicals acquired from plant material [19
], and are the plants’ natural defence mechanism against predators and pathogens [20
]. There is considerable interest in the antibacterial and (or) antiviral properties of some EO for application in the post-weaning period [21
]. In this regard, cinnamaldehyde was found to decrease faecal E. coli
concentrations, with no effect on the concentration of faecal lactobacilli, in weaned pigs [23
Organic acids and EO have different modes of action to antimicrobial growth promoters (AGP), and it is therefore unlikely that one alone is going to be able to substitute AGP to control PWD. A combination of organic acid blends and EO would give a broader spectrum of activity as organic acids exert their activity in feed and the upper part of the GIT whilst EO exert their activity more in the distal part of the GIT [24
]. Furthermore, Gram-negative bacteria such as E. coli
possess a cell wall in addition to their cell membrane, which prevents compounds such as organic acids from entering the bacteria and destroying vital cellular functions. A permeabilizing complex has been found to disrupt this cell wall, making the bacteria more susceptible to these compounds and contributing to enhanced post-weaning growth [25
The hypothesis tested in this experiment was that a post-weaning diet supplemented with a blend of organic acids (propionic, formic and acetic), cinnamaldehyde and a permeabilizing complex (OACP) will decrease the incidence of PWD in pigs and increase growth performance following infection with enterotoxigenic E. coli. This product was examined against a base diet without any antimicrobial compounds and a diet containing ZnO, which is still commonly used in the post-weaning period to mitigate disease and improve performance.
2. Experimental Section
This study was reviewed and approved by the Animal Ethics Committee of Murdoch University (R2631/14). Animals were handled according to the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes [26
2.1. Animals, Experimental Design, Diets and Housing
A total of 72 entire male pigs (Large White × Landrace) weaned at an average of 22 days (d) of age and weighing 7.2 ± 1.02 kg (mean ± SEM) was used. The pigs were obtained from a commercial pig farm (Yarloop, Western Australia) on the day of weaning and transported to an experimental facility at Murdoch University. Due to supply issues the pigs arrived in two batches 3 days apart, but were placed on test according to the same timeframe. On arrival, the pigs were weighed and faecal rectal swabs were taken and cultured for baseline presence of β-haemolytic E. coli. Pigs were randomly allocated to their experimental diet in 6 replicate pens of 4 pigs per pen according to a completely randomised block distribution and live weight (3 treatments × 6 replicate pens per treatment × 4 pigs per pen; n = 72).
Treatments comprised three different diets: a base diet without any antimicrobial compounds; the base diet with 3 g ZnO/kg added; and the base diet with 1.5 g OACP/kg (Biotronic Top 3®
, Biomin Australia Pty Ltd., Carlingford, Australia) added. The base diet, comprised mainly of wheat, soybean meal, barley and whey, was formulated to meet the animals’ requirements according to National Research Council [27
] (10.4 MJ NE/kg, 0.9 g standardised ileal digestible lysine/MJ DE). Diet compositions and analysed gross energy and nutrient contents are presented in Table 1
. The diets, along with water, were offered on an ad libitum basis for 3 weeks after weaning.
Composition of experimental diets (g/kg, as fed basis).
Composition of experimental diets (g/kg, as fed basis).
|Vitamin/Mineral premix †||1.0||1.0||1.0|
|Choline chloride (60%)||0.4||0.4||0.4|
|Biotronic Top 3® ‡||0.0||0.0||1.5|
|Calculated composition|| || || |
|SID meth + cysteine||8.1||8.1||8.1|
|Analyzed composition|| || || |
|Gross energy, MJ/kg||17.6||17.3||17.6|
Pigs were kept in pens of metal wire-meshed construction with plastic flooring and with a space allowance of at least 0.6 m2 per pig. Each pen was equipped with a nipple water drinker and a plastic feeding trough. Pens were in three different rooms, with 6 pens per room and 2 pens per treatment in each room. The ambient temperature was maintained at 26.3 ± 1.0 °C (mean ± SD). Pigs were monitored twice daily and weighed weekly. Feed disappearance from each pen was recorded weekly and feed wastage was visually assessed (by the same person) and recorded daily to calculate feed intake, body weight gain, and feed conversion ratio.
2.2. Induction of PWD with Enterotoxigenic E. coli and Measurements of PWD
Pigs were infected with an enterotoxigenic E. coli
(ETEC, serotype O149:K98:K88; toxins LT, ST, STb, EAST) on d 4, 5 and 6 after weaning. Inoculation cultures of ETEC were prepared as described by Heo, et al.
]. All pigs were orally dosed, using mild restraint, with the inoculum via a drench gun to provide 9 mL aliquots of 1.03 × 109
colony forming units (CFU)/mL of ETEC per pig. Faecal swabs were taken on d 0, 3, 5, 7, 9 and 11 to assess faecal shedding of ETEC, by inserting a cotton swab into the anus. Swabs were inoculated on sheep blood (50 mL/L) agar plates (Path West Laboratories, Perth, Western Australia). Plates were incubated overnight at 37 °C and assessed based on morphology and haemolysis. Scores were given on a scale from 0 to 5 according to the number of streaked sections containing viable haemolytic E. coli
, where 0 was no growth and 5 was growth out in the fifth section of the plate [28
Faecal consistency and the incidence of diarrhoea of individual pigs were visually assessed daily, by the same person, for 21 d after weaning. A score between 1 and 4 was given, as follows: (1) firm, well-formed faeces; (2) soft formed faeces; (3) soft and loose shape; or (4) watery liquid consistency, with this considered as diarrhoea. To allow for statistical analysis the scores were converted into percentiles (1 = 0%, 2 = 33.3%, 3 = 66.7% and 4 = 100%). The diarrhoea index (DI) was calculated as the mean proportion of days pigs had diarrhoea with respect to 14 d after weaning [29
]. Faecal samples from 8 focus pigs (median weight pigs) were collected per treatment, on d 4 and 11 after weaning, for subsequent volatile fatty acid (VFA) analysis and enumeration of specific bacterial groups by quantitative PCR (qPCR). Samples were stored at −20 °C until analyzed.
2.3. Blood Sampling
Blood samples were collected on d 4 and 11 from 2 pigs per pen (12 samples per treatment). Samples were collected via jugular vein puncture into either a lithium heparin or K3EDTA coated tube. Lithium heparin tubes were immediately placed on ice and K3EDTA tubes were kept cool. The heparin tube was centrifuged at 3000× g for 10 min at room temperature. Plasma was then collected and stored at −20 °C until analyzed for plasma urea nitrogen (PUN), total antioxidant content (TAC), and the acute phase proteins (APP) haptoglobin, albumin, and C-reactive protein (C-RP). Whole blood samples collected in K3EDTA tubes were subjected to blood cell count assessment on the same day.
2.4. Analytical Methods
Diet samples were analyzed for dry matter, gross energy, crude protein, crude fibre, neutral detergent fibre (NDF), acid detergent fibre (ADF), and zinc. Dry matter content was determined using AOAC official method 930.15 [30
]. The N content was determined using combustion method 990.03 [30
] and crude protein content was calculated as N content ×6.25. Crude fibre content was determined using AOAC official method 962.09 [30
]. The NDF and ADF contents were determined using the AOAC official methods 925.10 [30
]. Gross energy content was determined using a ballistic bomb calorimeter (SANYO Gallenkamp, Loughborough, UK). Zinc content was determined using inductively coupled atomic emission spectroscopy. The pH levels in the diets were assessed by preparing 1:9 w
feed in water, which were then measured using a pH meter (ROSS Ultra®
, Thermo Fisher Scientific Inc., Beverly, MA, USA).
Faecal samples were analyzed for volatile fatty acids (VFA) content using gas chromatography. Samples were prepared as described by Kim, et al.
], except that the thawed samples were diluted 1:2 (w
), and expressed as the molar percentage of total VFA. The TAC was determined using the OxiselectTM Total Antioxidant Capacity (TAC) Assay Kit (Cell Biolabs, Inc., catalogue number #STA-360, San Diego, CA, USA). The PUN, haptoglobin and albumin contents were determined using a Beckman Coulter/Olympus Reagent Kit (OSR6134), an in-house method NTM-62 based on [32
], and Randox Ranbut Reagent kit (RB1007), respectively. All kits and methods were performed on an Olympus AU400 Clinical Chemistry Analyser at the Department of Agriculture and Food, Animal Health Laboratories (South Perth, WA, Austrlia). The C-RP levels were determined by the use of a porcine ELISA kit (R & D Systems, catalogue #DY2648). Whole blood cell count was conducted using an automatic haematology analyser (ADIVA 2120, Bayer Healthcare, Siemens, Germany).
2.5. DNA Extraction and Quantitative Real Time Polymerase Chain Reaction
DNA Stool Mini Kit (QIAGEN GmBH, Hilden, Germany) was used to extract DNA from frozen faecal samples with DNA stored at −20 °C until qPCR was performed. Quantitative PCR was used to measure numbers of total E. coli
, E. coli
with F4 fimbria (the specific E. coli
used for oral infection in this experiment), Enterobacteriaceae
spp., and total bacteria by targeting 16S rRNA genes, the 16S–23S rRNA intergenic spacer region and the F4 fimbriae gene (Table 2
). Primers and probes were synthesized by Biosearch Technologies (Novoto, CA, USA). TaqMan probes were labelled with 5′ carboxyfluorescein (FAM) on the 5′ end and Black Hole Quencher (BHQ-1) on the 3′ end (Table 2
). Standard curves for each of the qPCRs were constructed from 10-fold serial dilutions of each DNA extracted using the DNeasy Blood and Tissue Kit (Qiagen, Venlo, Netherlands) from broth cultures of F4 E. coli
, E. coli
, Salmonella typhimurium
and Lactobacillus acidophilus
. Numbers of each bacterial species were enumerated using ISO 16649-2, ISO 21528-2 and ISO 15214 international standards.
Primers and probes used for quantitative real time PCR.
Primers and probes used for quantitative real time PCR.
|Target||Primer/Probe||Reference||PCR Product Length (bp)|
E. coli||Forward||Bartosch, et al. ||195|
|Probe||Designed by Y. Chen, EMAI (unpublished)|
|Enterobacteriaceae||Forward||Bartosch, et al. ||195|
|Probe||Designed by Y. Chen, EMAI (unpublished)|
|Lactobacillus spp.||Forward||Walter, et al. ||190|
|Reverse||Modified from Heilig, et al. |
|Probe||Modified from Delroisse, et al. |
|E. coli with F4 fimbria||Forward||Franklin, et al. ||764|
|Total bacteria||Forward||Suzuki, et al. ||152|
The F4 E. coli qPCR was the only assay to use SYBR Green to detect DNA amplification, and all other assays used the TaqMan probe technology. Each TaqMan qPCR contained 1× AgPath-ID reaction buffer (Applied Biosystems, Foster City, CA, USA), 1 unit AgPath-ID Taq polymerase, 40 nM probe, and 200 nM of the forward and reverse primers. The qPCR amplification protocols for the total E. coli, Enterobacteriaceae and Lactobacillus spp. included an initial denaturation step of 95 °C for 10 min, then 40 cycles of 95 °C for 15 s and 65 °C for 45 s. A reduced annealing temperature was used (58 °C) for the total bacteria qPCR. The F4 E. coli qPCR contained 1× SensiMix SYBR Low-ROX reaction buffer including Taq polymerase (Bioline, London, UK) and 300 nM forward and reverse primers. Cycling conditions for the F4 E. coli qPCR were 95 °C for 10 min, then 40 cycles of 95 °C for 15 s, 65 °C for 30 s and 72 °C for 60 s. Samples were quantified on a ViiA7 PCR machine (Applied Biosystems, Foster City, CA, USA), and numbers of total E. coli, Enterobacteriaceae, Lactobacillus spp. and F4 E. coli were expressed as a percentage of total bacteria to account for differences in faecal water content.
2.6. Statistical Analyses
Statistical analysis of production data, faecal score, faecal ETEC excretion and DI was performed using one-way ANOVA in SPSS (Version 21, IBM Corporation, Armonk, NY, USA) with dietary treatment as the independent variable, and batch as a random factor (to account for a difference in start weight between batches of pigs). Plasma measures, faecal SCFA, faecal bacterial counts and ratios measured on d 4 and 11 were analysed by repeated-measures ANOVA. Faecal bacterial counts for total E. coli, F4 E. coli and Enterobactericeae, and faecal bacterial ratios for Lactobacillus spp.:E. coli and F4 E. coli:Lactobacillus spp., were not normally distributed and therefore were logarithmically transformed before analysis. Means were back-transformed and expressed as least-square means with 95% confidence intervals. Significant interaction means were separated using Tukey’s HSD test. Pen was used as the experimental unit for performance, faecal score, faecal ETEC excretion, and DI. Pig was used as the experimental unit for plasma measurements, blood cell counts, faecal VFA concentration, and bacterial enumeration by qPCR. Chi-squared analysis (SPSS; Version 21, IBM, Armonk, NY, USA) was used to compare the percentage of pigs having PWD between the different diets. Statistical significance was accepted at p < 0.05 and 0.05 < p < 0.10 was considered a trend.
Data from this experiment showed that the incidence of PWD decreased only for the pigs supplemented with ZnO, as both the number of pigs with PWD and the DI were decreased. Despite there being a decrease in the incidence of PWD for the ZnO-treated group, the shedding of ETEC was similar for all groups. These data support other studies [39
] suggesting that supplementation of ZnO suppresses PWD and increases growth performance, but that these effects are not necessarily related to a reduction in ETEC or faecal score but possibly to improved intestinal barrier and (or) epithelial immune functions. Roselli, et al.
], for example, concluded that ZnO protects cells from ETEC by the inhibition of bacterial adhesion and internalization, preventing disruption of barrier integrity. This inhibition reduces levels of endotoxin production and decreases the severity of PWD. This may explain the decrease in DI but not in ETEC shedding for pigs fed ZnO in the present study. Furthermore, pigs may not express the receptors necessary for the E. coli
to attach [42
] and hence not display PWD whilst shedding ETEC. Other pigs may have receptors that are only weakly adhesive, and the presence of receptors also varies between individuals [43
]. These factors will have a strong influence on the occurrence of PWD but not necessarily the faecal shedding of ETEC [43
]. Unfortunately it was not possible to conduct genotype screening for F4 receptors in this experiment.
In the present study and in agreement with numerous other studies [44
], ZnO addition to the diet increased growth rate and feed intake after weaning and reduced PWD. A plethora of reasons have been provided for the mechanism(s) whereby ZnO exerts positive effects in the post-weaning period (e.g., [4
]). However, in the current study we were unable to determine the precise cause(s) for the improved production and reduction in PWD from the data collected. In addition, supplementation with OACP gave a statistically similar production performance as the ZnO-supplemented diet, however and aside from the second week after weaning, performance for the OACP-fed pigs was equivalent to the base-fed pigs. The OACP product contains a mixture of organic acids, cinnamaldehyde and a proprietary permeabilizing complex. An abundance of research has been conducted on organic acids and their effects on performance in weanling pigs, as summarized by Partanen and Mroz [13
] and Mroz [49
], among others. Meta-analyses of the data show that organic acids generally improve growth performance. However large variation exists due to factors such as form and type of the organic acid, inclusion level, production of intraluminal SCFA, differences in the amount of fermentable carbohydrate substrates in the diet for bacterial growth, colonization and activity leading to SCFA production, weaning age, presence or absence of bacterial receptors, and hygiene and welfare [49
Essential oils (EO) such as cinnamaldehyde tend to have ambivalent effects on production and PWD in young pigs [22
]. The EO are believed to exert positive effects through mechanisms including antibacterial actions, alteration of intestinal microbiota, increased digestibility and absorbance of nutrients, and antioxidative and immunomodulatory activities [51
]. In relation to changes in microbiota, Muhl and Liebert [53
] reported that the EO (carvacrol and thymol) did not elicit any change to specific groups of the intestinal microbiota. Jiang, et al.
], however, found a decrease in counts of Lactobacillus
spp. when supplementing with EO (thymol and cinnamaldehyde), whilst Castillo, et al.
] found an increase in bacterial counts of Lactobacillus
spp. when diets were supplemented with carvacrol and cinnamaldehyde. Li, et al.
] found an increase in Lactobacillus
spp. counts and a decrease in the E. coli
count, and suggested this was due to lactobacilli being less sensitive to any antimicrobial effects of EO compared to potentially pathogenic bacteria such as E. coli
In the current study and for the group of pigs supplemented with OACP, a decrease in E. coli
counts and an increase in Lactobacillus
spp. counts before and after ETEC infection was anticipated, as inclusion of organic acids is thought to decrease pH and increase proteolytic enzyme activity in the gastrointestinal tract (GIT), producing a more favourable environment for Lactobacillus
spp. and suppressing E. coli
]. This was not generally supported by our data. No significant decreases were found in the proportions of F4 E. coli
with fimbriae (relative to total bacteria) and Enterobacteriaceae
(relative to total bacteria), nor was an increase seen in the proportion of Lactobacillus
spp. (relative to total bacteria), following infection with ETEC in pigs fed OACP. In contrast, pigs supplemented with ZnO showed increases in the proportions of F4 E. coli
(relative to total bacteria) and Enterobacteriaceae
(relative to total bacteria) following infection, demonstrating further that ZnO had no direct efficacy against F4 E. coli
. Nevertheless Højberg, et al.
] found a decrease in Lactobacillus
spp. when piglets were supplemented with 2.5 g ZnO/kg, and suggested that the influence of ZnO on the GIT microbiota worked in similar ways as some AGP, reducing Gram-positive commensals rather than potentially pathogenic Gram-negative bacteria.
Measuring ratios of commensal to pathogenic bacteria to evaluate GIT health overcomes the potential need to account for differences in faecal dry weight [58
]. Castillo, et al.
] successfully used qPCR to quantify total bacteria, Enterobacteriaceae
spp. numbers in the digestive fluid of pigs, finding that qPCR was more sensitive than traditional microbiological methods, but ratios of lactobacilli to enterobacteria were comparable between methods. In the current study, the trend for an increase in ratios of F4 E. coli
:total E. coli
, F4 E. coli
, and F4 E. coli
for the ZnO-fed pigs on d 11 only suggests that feeding ZnO exacerbated the presence of the F4 fimbriae subtype in the faeces relative to other bacterial populations. Only the ratio of total E. coli:Enterobacteriaceae
was affected by the dietary treatment, this again being higher in ZnO-fed pigs. Numbers and ratios of Enterobacteriaceae
including E. coli
with F4 fimbriae were generally higher after infection in pigs fed ZnO compared to pigs fed the other diets, and were generally not reduced with OACP. These data are in general agreement with the recent work from Starke, et al.
Given the sound growth rates achieved in this study, the decrease in PUN levels from d 4 to d 11 could be attributable to a decrease in microbial fermentation of nitrogenous compounds in the large intestine commensurate with enhanced protein digestion and amino acid absorption. Catabolism of amino acids by microbes produces NH3
, which is converted to urea in the liver. The urea synthesized in the liver is either excreted as urine or diffused back into the caecum and combined into microbial nitrogen [61
]. A decrease in urea could also be linked to more efficient utilization of dietary protein or decrease in protein breakdown [62
]. As pigs overcome the post-weaning growth check, their protein requirements increase from 13.1 g/d (5–7 kg LW) to 23.1 g/d (7–11 kg LW) [27
]. In the present study only one diet was fed for the entire 3-week trial period, thus as pigs became heavier there would be an increase in utilization (i.e.
, less excess N) as the requirements increased.
The APP are proteins that are a part of the acute phase response, the early defence or innate immune system triggered by different stimuli such as trauma, infection, stress and inflammation [63
]. The APP are divided into two categories, positive and negative, which increase or decrease during the acute phase response, respectively [64
]. Haptoglobin, a positive APP, increases in concentration according to deteriorated health status, infection, inflammation or trauma [32
]. No increase in haptoglobin levels was found in the present study. C-RP is another positive APP and would be expected to increase in levels over time after the challenge, which was observed in this study. It has been established that APP have different induction sensitivities, and hence some react to a lesser extent than others to the same infection/inflammation [65
]. When measuring APP levels in pigs infected with ETEC, Houdijk, et al.
] found that C-RP concentrations increased more than 10-fold, whilst haptoglobin concentrations only increased three-fold. Furthermore, the decrease in TAC concentration from d 4 to 11 suggests that the ETEC infection decreased the antioxidative capacity, however and as with the APP, no difference between treatments was observed in the present trial, indicating that they all elicited a similar inflammatory response to the E. coli
Concurrently, the lack of any differences found in white blood cell counts, over time, in the present study confirm that the overall level of infection was most probably insufficient to cause a major health issue for the pigs. The lymphocyte to neutrophil ratio has been used previously as an indicator of the pigs’ responses to stress [67
], and the lack of any difference in the current study supports this notion.
The increase in total VFA concentration in the OACP group compared to the base diet and ZnO groups was expected, as the OACP diet was supplemented with organic acids. The addition of organic acids aids in the acidification of the stomach, thus increasing proteolytic enzyme activity, which then may increase the digestibility of protein and amino acids [68
]. It has also been found that an increase in VFA stimulates GIT epithelial cell proliferation and villous height, hence increasing surface area for absorption [69
], and some studies have shown that increased production of VFA can reduce pathogenic bacterial numbers such as salmonella [70
]. However, the lack of any improvement in growth rate and feed conversion ratio, or any significant decrease in Enterobacteriaceae
numbers in the present study, indicates that the increase in total VFA levels in pigs fed OACP was unrelated to these measurements.