1. Introduction
The weaning of piglets is a critical period due to the dietary challenges for the pigs, and the social and environmental changes to which they are subjected [
1]. Antimicrobial compounds have been widely used in feed for weaned piglets to improve performance and to prevent diarrhea [
2]. However, concerns regarding the use of antibiotics have grown, due to the possible consequences that antibiotic resistance can have on human and animal health [
3]. The European Union (EU) implemented a ban on using antibiotics as growth-promoting additives in non-medicated feeds for livestock from January 2006; they currently have a limited use even in medicated feeds. In addition, other products used to prevent digestive problems in piglets, based on zinc or copper, have been restricted due to the environmental problems that might be resulted [
4]. Furthermore, feed supplementation with high levels of zinc oxide (ZnO) has been related to the development of antibiotic-resistant microorganisms in piglets [
5]. However, ZnO has been widely used at high levels (2500 ppm of Zn) as a strategy to prevent diarrhea in piglets, but the European legislation has set a target of zero ZnO usage in pharmacological doses in piglet feed by 2022 [
6]. In this context, searching for alternative strategies to modulate the intestinal microorganisms of weaned piglets is a research goal. Among these alternatives are probiotics [
7], which are harmless live microorganisms that balance intestinal microbiota for the benefit of the animal [
8]. An example is the preparation
Clostridium butyricum FERM-BP 2789, which was authorized as a zootechnical feed additive for weaned piglets (Regulation (EU) No 373/2011) [
9]. This bacterium can produce butyric acid, which is an energetic substrate for intestinal epithelium cells and a regulator of intestinal pH; consequently, it can preserve an optimal intestinal environment [
10]. In piglets, some studies have shown that diet supplementation with
Clostridium butyricum can improve the immune status, the structure of the intestinal mucosa, and the microbial profile of the intestine [
11], resulting in an improvement of the feed conversion ratio [
12]. This additive has also had positive effects during the weaning of piglets in combination with other probiotics [
13,
14]. However, there is little information on the use of
Clostridium butyricum in combination with other substances with an additive or synergistic action that could improve the potential results of its use. In this way, the combination of probiotics with prebiotics is a strategy that is currently being studied [
15]. A prebiotic is a food ingredient, totally or partially non-digestible, which can ferment in the digestive tract increasing the growth of the bacteria that could improve the health of the host [
16]. Some fiber sources have shown prebiotic effects [
17]. In post-weaning piglets, ingredients such as wheat bran and pea fibers have been able to modulate the intestinal microorganisms by stimulating the beneficial bacterial species, although others such as maize fiber and soyabean fiber did not show any positive effects [
18].
Carob meal (originally from the fruit of
Ceratonia siliqua) and citrus pulp are two fiber-rich by-products typical of the Mediterranean area having different fiber profile compositions [
19], with carob meal having a higher neutral detergent fiber (NDF) and lignin content. In addition, carob pods are sweet with high levels of sucrose [
20], giving it a more palatable characteristic, and carob is rich in tannins that confer astringent properties (preventing diarrhea), being a natural source of antioxidants [
21]. Tannins (polyphenolic compounds), present in carob and its derivatives, are considered powerful antioxidants and anti-inflammatories, having bacteriostatic and antidiarrheal properties that could be of interest for piglet diets [
22]. In addition, Andrés-Elias et al. [
23] observed that the incorporation of carob in diets for piglets could affect the intestinal microbiota. Lizardo et al. [
24] studied the incorporation of carob meal in the feed of weaned piglets and found that it did not affect the productive parameters, and the prevalence of post-weaning diarrhea dropped by 20–33% when carob meal was included at 3% and 6%, respectively.
Citrus by-products are rich in pectins (fermentable fiber) and, in addition, they have available biologically active compounds, such as polyphenols, carotenoids, and essential oils [
25]. Furthermore, Hotchkiss et al. [
26] concluded that pectic oligosaccharides from citrus fruits had prebiotic bifidogenic properties. In piglets, Pascoal et al. [
27] indicated that a 9% citrus pulp inclusion decreased the occurrence of
E. coli in the small intestines of piglets. Moreover, Collier et al. [
28] showed that in newly weaned pigs challenged by
E. coli F18, a 10% citrus pulp inclusion suppressed the ileal and cecal recovery of this pathogen. However, some studies on citrus pulp incorporation showed negative effects on nutrient utilization (at 4.5% of inclusion) [
29], and on incidences of diarrhea [
30] or negative effects on performance [
27] (at 9% of inclusion); although, others authors did not find negative effects on performance when they included 7.5% citrus pulp in the diets of weaned piglets [
31]. Therefore, there is some controversy regarding the use of this ingredient in piglets.
Thus, beyond the traditional concept of using a prebiotic to enhance the action of a specific probiotic, the hypothesis of the present research is that the inclusion of a probiotic such as Clostridium butyricum, in combination with sources of fiber that could have a positive complementary effect, could generate a healthy digestive environment and improving the digestibility and physiological status of weaned piglets, but this effect could be different depending on the type of fiber ingredient. Therefore, the objective of the present work is to evaluate the effect of the inclusion of Clostridium butyricum, alone or with carob meal or citrus pulp, on the digestive and metabolic status of post-weaning piglets in optimal nursery conditions.
2. Materials and Methods
All experimental procedures performed in this work were in compliance with the protection of animals used for scientific purposes regulated by the European Union (2010/63/EU Directive) [
32]. The administrative authorities and the Ethics Committee of the University of Murcia (Murcia, Spain) approved the protocol (code A13170502).
2.1. Animal and Experimental Design
The study was conducted for 27 days in the Animal Nutrition Experimental Unit at the veterinary farm of the University of Murcia (Guadalupe, Murcia, Spain) in optimal nursery conditions. A total of 30 piglets (non-castrated male, 100% Large White breed) were weaned at 21 days old, individually identified, and weighed. The piglets had an average body weight (BW) of 5.06 ± 0.64 kg. They were housed in a controlled environment nursery, and were randomly allotted to one of the five dietary treatments, with six piglets per treatment. Each dietary treatment group was housed in a pen with a plastic slat floor and a space of 0.5 m2 per animal, equipped with feeders and nipple drinkers, offering ad libitum access to feed and water throughout the experiment.
The diets consisted of two control treatments, one negative without ZnO at high doses (C−), and one positive control (C+) consisting of a basal diet with ZnO added at a high level (2500 ppm of Zn), and another three experimental treatments, one only supplemented with
Clostridium butyricum (
Clostridium butyricum Miyairi 588 (FERM BP-2789), MIYA-GOLD
® S, Huvepharma
®, Antwerp, Belgium) to provide 2.5 × 10
8 CFU per kg feed (PRO), and another two with the same probiotic plus 5% carob meal (PROC) or 5% citrus pulp (PROP). The commercial additive with
Clostridium butyricum was included at 0.05%, during the mixing of ingredients, to provide the feeds with this probiotic, at the established dose. The carob meal or citrus pulp were incorporated by substituting all of the sugar beet pulp and part of the barley of a basal diet, and carefully balancing the other minor ingredients until all diets were iso-energetic, iso-aminoacidic, and iso-neutral detergent fiber (iso-NDF). All experimental feeds were formulated to meet or exceed the requirements of piglets as indicated by the Spanish Foundation for the Development of Animal Nutrition (Fundación Española para el Desarrollo de la Nutrición Animal, FEDNA) [
33]. In addition, 0.5% TiO
2 was included in all diets as an indigestible marker to determine the apparent fecal and ileal digestibility of the feeds (
Table 1). The feeds were manufactured by Agrarian Transformation Society number 2439 (La Hoya, Spain) from the same ingredient batches. The diets were presented in mash form. In addition, the body weight (BW) of the piglets was controlled during the experiment on days 0, 14, and 26.
2.2. Sampling Collection
On day 25 of the experiment, after a period of 8 h of fasting, blood samples were obtained from the jugular vein of the piglets by venipuncture, using a 4 mL vacuum tube per animal (Z Serum Clot Activator, VACUETTE®, Greiner Bio-One GmbH, Kremsmünster, Austria). Next, in the laboratory, they were centrifuged at
2500× g for 10 min to obtain serum, which was stored in aliquots at −80 °C for the subsequent determination of the blood biochemical profile and cytokines. In addition, individual fecal samples were collected on day 26 of the trial directly from the anus of each piglet and placed in a sterile bottle for fecal microbiology analysis.
On the last day of the trial (day 27), prior to animal slaughtering, individual fecal samples were collected for chemical analysis, then the piglets were tranquilized with an intramuscular injection of azaperone, and later euthanized with an intravenous overdose of pentobarbital. They were bled, and their abdomens were immediately opened by an incision from sternum to pubis to remove the entire digestive tract. Samples were taken from the content of the stomach, duodenum, jejunum, ileum, cecum, and colon. The pH of these samples was immediately determined from the different sections of the gastrointestinal tract by insertion of a pH meter electrode (pH-Meter GLP 21, Crison Instruments, S.A., Alella, Barcelona, Spain), except for the colon where the pH determination was recorded according to the procedure described by Peters et al. [
34]. Additionally, the ileal content was collected for chemical analysis.
The fecal and ileal content were lyophilized and stored at −20 °C in airtight containers until TiO2 analysis in the laboratory. In addition, sub-samples of fecal and cecum content (1 g) were acidified with 0.032 mL of H2SO4:H2O (50:50) dilution and stored at –20 °C until volatile fatty acid (VFA) analysis.
For histomorphometrical and immunohistochemical analysis, samples of the piglets’ middle section of jejunum and ileum were obtained (2 cm) and fixed in 10% buffered formaldehyde and, subsequently, embedded in paraffin wax.
2.3. Analysis of Feed, Digesta, and Feces
The feed samples were ground to pass through a 1 mm sieve in a laboratory mill (RETSCH ZM 200 Ultra Centrifugal Mill; RETSCH, Hann, Germany). These samples were analyzed using the Association of Official Analytical Chemists (AOAC) procedures [
35]: dry matter (DM) by the 934.01 method; crude protein (CP) by the 2001.11 method. Van Soest et al. [
36] procedures were used to determine the neutral detergent fiber (NDF) and acid detergent fiber (ADF), analyzing the acid detergent lignin (ADL) through the solubilization of cellulose with 72% H
2SOH
4.
Sub-samples of the feeds were ground to pass a 0.5 mm sieve (in the same mill indicated above) to analyze the total, insoluble, and soluble dietary fiber using the AOAC 991.43 enzymatic–gravimetric method with the Megazyme K-TDFR-100A/K-TDFR-200A 04/17 kit (Megazyme Ltd., County Wicklow, Ireland).
Furthermore, the marker (TiO
2) used to calculate digestibility was analyzed in the feeds, feces, and ileum contents using the method described by Myers et al. [
37]. For this analysis, the samples were also ground to pass through a 0.5 mm sieve. Feces and ileal contents were also analyzed for CP, as it was indicated previously.
The VFA concentration in the cecum and feces was determined by capillary gas chromatography using an adaptation of the method described by Madrid et al. [
38]. The gas chromatograph equipment used was a TRACE GC Ultra (Thermo Finnigan Italia SpA, Milan, Italy) with a flame ionization detector, and the capillary column was fused silica (30 m × 0.25 mm × 0.25 μm ID) coated with FFAP-TR as the stationary phase (Teknokroma, Barcelona, Spain). Standard solutions of acetic, propionic, butyric, isobutyric, isovaleric, and valeric acids were prepared for calibration, using 3-methyl-n-valeric acid as the internal standard as indicated by Oliveira et al. [
39].
One gram of each collected fecal sample for microbial study was diluted with a sterile saline solution in a 1:10 dilution and homogenized on a mixer-homogenizer for 2 min. Subsequently, aliquots of the ten-fold serial dilutions were spread-plated onto selective media. The dilutions were used for counting
Enterobacteriaceae, coliforms, and lactic acid bacteria. Thus, the
Enterobacteriaceae counts were determined by the ISO 21528-2:2004 [
40] adapted method (RAPID’ Enterobacteriaceae/Agar, Bio-Rad Laboratories, S.A., Madrid, Spain) (at 37 °C for 24 h). Coliform bacteria were enumerated using violet red bile lactose (VRBL) agar (at 37 °C for 24 h) according to ISO 4832:2006 [
41]. Mesophilic lactic acid bacteria were enumerated using De Man, Rogosa, and Sharpe (MRS) agar (at 30 °C for 72 h) following ISO 15214:1998 [
42] and the incubation was carried out with a double-layer MRS medium to provide anaerobic conditions. The results were expressed as log
10 CFU (colony-forming units)/g of feces.
2.4. Histomorphometrical and Immunohistochemical Procedures
Sections (4 µm) of each intestinal tissue sample were stained with hematoxylin and eosin for the morphometric study. For this determination, a ZEISS Axioskop 40 microscope (Carl Zeiss, Oberkochen, Germany) was used with a Spot Insight camera, and the Spot Advanced software (Spot Imaging Solution, MI, USA). In each slide, the height and crypt depth of 10 villi were measured and the results were expressed in μm. In addition, the villus height/crypt depth ratio was determined. The number of intraepithelial lymphocytes and goblet cells was quantified by counting in 10 fields of epithelium of 25,000 μm2.
For immunohistochemical analyses, the detection of IgA-secretory cells in the jejunum and ileum was performed by the avidin–biotin–peroxidase complex technique, according to Oliveira et al. [
39]. In the intestinal lamina propria, the number of IgA-positive cells was checked with a ZEISS Axioskop 40 microscope (Carl Zeiss, Oberkochen, Germany) using a Spot Insight camera and the Spot Advanced software (Spot Imaging Solution, MI, USA). Immunolabeled cells were recorded in 10 non-overlapping consecutive fields of 25,000 μm
2.
The same researcher, blinded to the treatments, performed the morphometric and immunohistochemical determinations.
2.5. Serum Analysis
The serum metabolic profile was analyzed for glucose, urea, total cholesterol, triglycerides, total bilirubin, total protein, albumin, aspartate aminotransferase (AST), and alanine aminotransferase (ALT) using commercially available kits (Beckman Coulter Inc., Fullerton, CA, USA). The total antioxidant capacity (TAC) and total oxidative status (TOS) were measured according to Erel [
43,
44], using colorimetric assays. All analyses were executed in an automated chemistry analyzer (Olympus AU600, Olympus Europe GmbH, Hamburg, Germany).
Serum cytokine levels were assayed after centrifugation of the samples at 20,000× g to eliminate the lipid phase. Interleukin (IL)-1β, IL-6, IL-8, IL-10, IL-12, and tumor necrosis factor-α (TNF-α) were determined using a porcine-specific multiplex cytokine/chemokine assay (cat. no. PCYTMAG-23K; Millipore MILLIPLEX, Billerica, MA, USA) on a MAGPIX instrument (Luminex; Luminex Technologies, Austin, TX, USA), as indicated by the manufacturer.
2.6. Calculations and Statistical Analyses
The apparent fecal and ileal digestibility of DM and CP were determined by the digestibility equation as follows:
All statistical analyses were performed using the SPSS Statistics software (IBM Corporation, Armonk, NY, USA), with the piglet set as the experimental unit. Data were analyzed by one-way ANOVA. The Shapiro–Wilks test was performed to assess the normality of data from the serum study; a logarithmic scale transformation was performed when the data did not follow normal distribution. Pairwise comparisons of means were performed using the Tukey test. Significance was determined at p < 0.05, and a trend was assumed at 0.05 ≤ p < 0.10.
4. Discussion
Restrictions on high-level ZnO usage in piglet feed is a major challenge for the sustainability of the current levels of intensive pig production [
4]. There are many changes related to the handling, environmental, and feeding conditions that need to be addressed. Therefore, with a view to optimizing piglet feed that is free from restricted substances, exhaustive studies need to be performed.
In our experiment, the inclusion of
Clostridium butyricum in feeds without high levels of ZnO, alone or in combination with carob meal or citrus pulp, did not affect the body weight or weigh gain of piglets after weaning, compared to both control feeds (with or without ZnO). Chen et al. [
11] found similar results when weaned piglets were fed
Clostridium butyricum in increasing doses, compared to a negative control diet. Regarding the inclusion of carob meal, authors such as Špoljarić et al. [
45] observed an improvement in the body weight of piglets at 42 days post-weaning when the feed was supplemented with 4% carob whole meal, although this effect was not found at 28 days post-weaning by these authors. The effect of including citrus pulp in feed on the body weight or weight gain of weaned piglets showed varying results, depending on the percentage of the ingredient that was included. Almeida et al. [
31] found no negative effect on body weight when they incorporated 7.5% citrus pulp in the feed of weaned piglets, although Pascoal et al. [
27] showed negative results on weight gain when this ingredient was incorporated at 9%. It should be highlighted that despite of the fact that our trial diets did not show negative effects on the growth of the animals, they could affect the intake or the feed conversion ratio, so performance tests should be carried out to evaluate these potential effects.
The apparent ileal digestibility (of DM or CP) was not affected by the treatments, although the apparent fecal digestibility of DM improved by 5% with the C+ (high level of ZnO) treatment when compared to C−; in addition, in spite of fecal CP digestibility not being significantly altered by treatments, C+ reached a quantitatively higher value than C−. Dębski [
46] indicated that dietary Zn supplementation is used to reduce the fermentation of nutrients in the intestine, improving nutrient digestibility. The inclusion of
Clostridium butyricum, alone or in combination with carob meal or citrus pulp, did not result in a difference in C+, and neither in C−, where fecal DM digestibility reached intermediate levels. Han et al. [
47] suggested that supplementation with
Clostridium butyricum increased total tract digestibility, as it increased the concentrations of the VFAs that reduce gut pH, achieving an anti-bacterial effect and increasing butyrate, which provides energy for intestinal epithelial cells. In addition, these authors indicated that this effect was more marked when the diet was supplemented with a higher level of probiotic (2.5 × 10
9 CFU/kg of
Clostridium butyricum).
We hypothesized that the inclusion of a combination of
Clostridium butyricum plus carob meal or citrus pulp could generate a positive complementary effect on the intestinal environment. The probiotic could help to improve digestive tract health; and the fiber ingredients could provide an additional favorable environment for beneficial bacteria. In this way, it can be said that, depending on the physicochemical properties of the fiber diet, the effects could be different [
48]. Soluble fiber is more fermentable, and it may produce an increase in VFAs that affects the intestinal environment; insoluble fiber may reduce the digesta transit time, preventing the proliferation of pathogenic bacteria [
49]. In addition, different effects resulting from the amount, type of fiber, and post-weaning period that were applied have been noted [
50]. In our trial, different fiber ingredients were used, but all diets were iso-NDF formulated, resulting in few analytical differences in the feeds, both in the Van Soest fractions and in dietary fiber. No additional improvements in digestibility nor harmful effects were found with the carob meal or citrus pulp at 5% inclusion. On the other hand, when Zhang et al. [
51] evaluated the effects of
Clostridium butyricum and corn bran at 5% on weaned piglets, they observed a decrease in digestibility, and no positive interaction between the corn bran and
Clostridium butyricum. However, Chen et al. [
50] evaluated the effects of dietary soluble and insoluble fiber inclusion (alone or in different combinations) on weaning piglets, and found that all treatments supplemented with fiber presented a higher apparent fecal digestibility of DM than the control group; and the effects on the ileum
Lactobacillus content and cecum digesta VFAs depended on the type of dietary treatment.
In our study, despite finding that the treatments had no effect on the general population of
Enterobacteriaceae, coliforms, or lactic acid bacteria in feces, the concentrations of VFAs were affected by the treatments. Therefore, a more exhaustive study of the microbiota is desirable for improving knowledge in this area. We found that the concentration of VFAs in feces with C+ was lower than that in the PROC treatment. O’Shea et al. [
52] also found that ZnO in piglet feed decreased the VFA content in feces. They justified this fact with the implication that ZnO could have decreased the secretion of chloride from the colon mucosa; therefore, reducing the secretion of fluids, and contributing to a reduction in water in the digesta and an alteration of the microbial activity. On the other hand, although changes in the VFA profile in the cecum content were not observed, they were found in the feces. Thus, the diets that included the probiotic showed a higher molar percentage of butyric acid, in relation to the diet with high levels of ZnO.
Clostridium butyricum is a Gram-positive anaerobe that produces butyric acid, which could provide nutrients for the regeneration of intestinal epithelial cells, contributing to intestinal health [
53].
It is known that the inclusion of ZnO at high levels increases the height of villi and decreases crypt depth, improving the villus height/crypt depth ratio [
54,
55], which could be related to a greater capacity for nutrient absorption in the small intestine. However, in our case, any dietary treatment, including the ZnO, did not affect the histomorphology of the intestinal mucosa. Moreover, the number of goblet cells, lymphocytes, and IgA cells in the intestinal mucosa were also not affected. This could be related to the fact that the piglets were not subjected to challenging aggressions that could affect the intestinal mucosa, and they did not manifest diarrhea problems regardless. In this way, Liu et al. [
56] indicated that the effect on the intestinal morphology of adding Zn into piglet diets, even at high doses, is limited under optimal physiological conditions.
In general, the dietary treatments in our trial did not affect the serum biochemistry profile of the piglets. In addition, the levels of glucose, urea, cholesterol, triglycerides, total bilirubin, total proteins, albumin, and the albumin/globulin ratio were within the ranges indicated by Perri et al. [
57] and Ventrella et al. [
58] for piglets close to weaning. The mean levels of ALT and AST enzymatic activity were within the reference ranges indicated by Caprarulo et al. [
59], although these were close to the maximum levels observed by Klem et al. [
60] for growing pigs. It should be noted that the ALT, a liver enzyme that, at high levels, is considered to be a biomarker of liver damage [
61], was different for the PROC and C+ treatments. Neither the TOS or TAC values were affected by the treatment, which was evidenced by the absence of dietary effects on the oxidative status of the piglets. Despite the PROC treatment containing carob, an ingredient with high levels of phenolic compounds, potentially with an antioxidant effect, most of those are in the form of condensed tannins [
62], a chemical form which is poorly bioavailable [
63,
64].
Many studies have determined that, during the inflammation of the gastrointestinal tract, there is an imbalance of the inflammatory cytokine profile [
65]. In our experiment, the serum cytokines values were not altered by dietary treatments, except for the concentration of IL-8 in the PROC diet. Zinc at high levels in feeding piglets has shown different results on cytokines. Zhu et al. [
55] found differences in the gene expression of certain cytokines from the jejunum mucosa when comparing piglets supplemented with ZnO (3000 ppm) with piglets that were fed a basal diet; specifically, a downregulated IL-1β expression (a pro-inflammatory cytokine) and an upregulated TGF-β expression (an anti-inflammatory cytokine). Similarly, other authors observed a decrease in the expression of other pro-inflammatory cytokines, such as TNF-α, IL-6, and IFN-γ, at day 7 post-weaning in piglets supplemented with high levels of ZnO; however, no differences at day 14 post-weaning were found [
66]. In addition, Kloubert et al. [
67] studied Zn supplementation at 0, 100, and 2500 ppm in the form of ZnO in weaned piglets, finding that the concentrations of IL-1β, IL-6, or TNF-α in diluted whole blood cultures (incubated with or without substance stimulates of cytokine production such as lipopolysaccharide or phytohemagglutinin) from piglets with dietary treatments were not affected by Zn supplementation. However, IL-2 concentration (a pro-inflammatory cytokine) increased in peripheral blood mononuclear cell cultures (incubated with substance stimulates of cytokine production) from piglets supplemented with 2500 ppm of Zn [
67].
Regarding feed supplementation with probiotics, one study showed a decrease in the concentration of IL-1β in the serum of weaned piglets that were supplemented with
Clostridium butyricum and challenged with enterotoxigenic
Escherichia coli K88 [
68]. In addition, other authors have found changes in the cytokine profile when probiotics were used in the feed of piglets challenged with
Escherichia coli lipopolysaccharide [
69], or affected by post-weaning colibacillosis [
70]. Nevertheless, in our study, the general biochemical serum profile and other parameters that were evaluated indicated the absence of nutritional imbalances or harmful processes on the animals’ health. In this sense, as cytokines are produced by the action of a stimulus, it could be hypothesized that, in favorable environmental situations, changes in their concentrations could be difficult to detect.
However, the PROC dietary treatment had the lowest IL-8 cytokine value, which is a type of pro-inflammatory cytokine [
65], being lower than that in the C−, C+, and PRO treatments. These results could be related to the inclusion of carob meal as an ingredient, since this feedstuff, or its derivates, has shown anti-inflammatory activity in in vitro assays and animal models [
71,
72]. Moreover, the advantageous effects of carob products on the other immunity parameters have been suggested. Špoljarić et al. [
45] observed that supplementation with carob whole meal did not affect the amount of red blood cells or leucocytes in weaned pigs, but an increase in the proportions of various types of lymphoid cells in the peripheral blood was observed.