ORIGINAL RESEARCH published: 16 April 2021 doi: 10.3389/fmicb.2021.642549 Effects of the Administration of Bifidobacterium longum subsp. infantis CECT 7210 and Lactobacillus rhamnosus HN001 and Their Synbiotic Combination With Galacto- Oligosaccharides Against Enterotoxigenic Escherichia coli F4 in an Early Weaned Piglet Model Agustina Rodríguez-Sorrento 1, Lorena Castillejos 1*, Paola López-Colom 1, Gloria Cifuentes-Orjuela 2, María Rodríguez-Palmero 2, José Antonio Moreno-Muñoz 2, Edited by: Diana Luise 3, Paolo Trevisi 3 and Susana María Martín-Orúe 1 Hang Xiao, University of Massachusetts Amherst, 1 Servicio de Nutrición y Bienestar Animal, Departament de Ciència Animal i dels Aliments, Universitat Autònoma de United States Barcelona, Bellaterra, Spain, 2 Laboratorios Ordesa S.L., Parc Científic de Barcelona, Barcelona, Spain, 3 Department of Reviewed by: Agricultural and Food Science, University of Bologna, Bologna, Italy Eric Cox, Ghent University, Belgium We evaluated the potential of multi-strain probiotic (Bifidobacterium longum subsp. infantis Junqiu Luo, Sichuan Agricultural University, China CECT 7210 and Lactobacillus rhamnosus HN001) with or without galacto-oligosaccharides *Correspondence: against enterotoxigenic Escherichia coli (ETEC) F4 infection in post-weaning pigs. Ninety-six Lorena Castillejos piglets were distributed into 32 pens assigned to five treatments: one non-challenged (CTR+) lorena.castillejos@uab.cat and four challenged: control diet (CTR−), with probiotics (>3 × 1010 CFU/kg body weight each, Specialty section: PRO), prebiotic (5%, PRE), or their combination (SYN). After 1 week, animals were orally This article was submitted to inoculated with ETEC F4. Feed intake, weight, and clinical signs were recorded. On days 4 and Food Microbiology, a section of the journal 8 post-inoculation (PI), one animal per pen was euthanized and samples from blood, digesta, Frontiers in Microbiology and tissues collected. Microbiological counts, ETEC F4 real-time PCR (qPCR) quantification, Received: 16 December 2020 fermentation products, serum biomarkers, ileal histomorphometry, and genotype for mucin 4 Accepted: 23 March 2021 Published: 16 April 2021 (MUC4) polymorphism were determined. Animals in the PRO group had similar enterobacteria Citation: and coliform numbers to the CTR+ group, and the ETEC F4 prevalence, the number of mitotic Rodríguez-Sorrento A, Castillejos L, cells at day 4 PI, and villus height at day 8 PI were between that observed in the CTR+ and López-Colom P, Cifuentes-Orjuela G, CTR− groups. The PRO group exhibited reduced pig major acute-phase protein (Pig-MAP) Rodríguez-Palmero M, Moreno-Muñoz JA, Luise D, levels on day 4 PI. The PRE diet group presented similar reductions in ETEC F4 and Pig-MAP, Trevisi P and Martín-Orúe SM (2021) but there was no effect on microbial groups. The SYN group showed reduced fecal enterobacteria Effects of the Administration of Bifidobacterium longum subsp. and coliform counts after the adaptation week but, after the inoculation, the SYN group showed infantis CECT 7210 and Lactobacillus lower performance and more animals with high ETEC F4 counts at day 8 PI. SYN treatment rhamnosus HN001 and Their modified the colonic fermentation differently depending on the MUC4 polymorphism. These Synbiotic Combination With Galacto- Oligosaccharides Against results confirm the potential of the probiotic strains and the prebiotic to fight ETEC F4, but do Enterotoxigenic Escherichia coli F4 in not show any synergy when administered together, at least in this animal model. an Early Weaned Piglet Model. Front. Microbiol. 12:642549. Keywords: synbiotic, enterotoxigenic Escherichia coli, piglet, probiotic, prebiotic, galacto-oligosaccharides, doi: 10.3389/fmicb.2021.642549 Bifidobacterium, Lactobacillus Frontiers in Microbiology | www.frontiersin.org 1 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 INTRODUCTION in vitro as well as in human clinical trials (Grimaldi et al., 2016; Canfora et al., 2017; Paganini et al., 2017). Moreover, GOS Diarrhea is the second leading infectious cause of death – after are considered to be highly similar to oligosaccharides from pneumonia – among children younger than 5 years (Liu et al., human milk (HMO), a factor that makes it an attractive 2015). The main pathogenic agents responsible for these diseases prebiotic when designing synbiotics with B. longum subsp. are viruses – namely rotavirus and calicivirus, which are infantis or other probiotic strains isolated from the infant responsible for 38 and 13% of the cases, respectively – followed intestine. Notably, genomic adaptations for HMO utilization by enteropathogenic and enterotoxigenic Escherichia coli (EPEC have been described for B. longum subsp. infantis (Sela et al., and ETEC, respectively), which contribute to 12 and 8%, 2008). Furthermore, GOS have been shown to interfere with respectively, of the deaths (Lanata et al., 2013). ETEC is also E. coli adhesion to tissue culture cells (Shoaf et al., 2006). one of most common pathogens in pigs; it is responsible for Moreover, recent studies have demonstrated the ability of the post-weaning diarrhea syndrome (Luppi, 2017). ETEC carrying probiotic strain B. longum subsp. infantis CECT 7210 to fimbriae F4 and F18 is considered as the most common pathogen metabolize GOS as substrate, promoting a higher growth than and, thus, these pathogens are used to perform standardized the other prebiotics. This combination has also shown significant challenges with in vivo trials (Luise et al., 2019a). Although antimicrobial properties against E. coli (Ruiz et al., 2020). E. coli is part of the normal intestinal microbiota, some strains, Based on this information, we hypothesized that the efficacy like the above mentioned, have developed pathogenic mechanisms of combined probiotic strains of Bifidobacterium and Lactobacillus to cause intestinal or extraintestinal disease (Clements et al., will be increased by the addition of fermentable carbohydrates 2012). Due to the increasing development of antibiotic resistance that may promote their growth and activity in the gut. Hence, in this organism (Poirel et al., 2018), new strategies for prevention the objective of this work was to evaluate whether combining and therapy must be implemented. Probiotics are defined by multi-strain probiotic, composed of B. longum subsp. infantis the Food and Agriculture Organization (FAO) as “live CECT 7210 and L. rhamnosus HN001, with GOS could improve microorganisms which, when administered in adequate amounts, activity against ETEC F4 in post-weaning piglets. confer a health benefit on the host.” These advantageous effects are strain dependent and can include gastrointestinal disorder prophylaxis and treatment, immune system enhancement, cancer MATERIALS AND METHODS prevention, and cholesterol normalization, among others (Kechagia et al., 2013). Regarding the potential of probiotics to ameliorate The trial was performed at the Experimental Unit of the diarrhea caused by ETEC, different strains of bifidobacteria Universitat Autònoma de Barcelona (UAB) and received prior and lactobacilli have proven antagonistic activity against E. coli approval (Permit No. CEEAH: 4026 DMAH: 10118) from the in vitro (Vazquez-Gutierrez et al., 2016; Fijan et al., 2018; Song Animal and Human Experimental Ethical Committee of this et al., 2019). There have also been positive results when tested Institution. The treatment, management, housing, husbandry, in vivo (Romond et al., 1997; Kumar et al., 2016). Our previous and slaughtering conditions conformed to European Union work has shown that the particular strain Bifidobacterium longum Guidelines (Directive 2010/63/EU; European Commission, 2010). subsp. infantis CECT 7210 can reduce ileal colonization by All efforts were made to minimize animal suffering. ETEC, a phenomenon that improves the local immune response in a piglet model (Barba-Vidal et al., 2017). Regarding lactobacilli Animals, Housing, and Experimental Design strains, Lactobacillus rhamnosus has been reported to have This trial was carried out as level 2 high-risk biosecurity positive effects against pathogenic E. coli, although the benefits procedures and involved personnel who had received appropriate could depend on the probiotic strain. In this sense, L. rhamnosus training. A total of 96 male piglets (Landrace × Large HN001 has been demonstrated to reduce enterohemorrhagic White) × Pietrain, 21 (±2) days of age and weighting E. coli translocation and to increase immunoglobulin A (IgA) 5.04 ± 0.32 kg, were used. Only males were selected to reduce concentration and blood leukocyte phagocytic activity in mice possible residual variability associated to sex as sample size is (Shu and Gill, 2001). However, administration of L. rhamnosus limited in trials involving pathogen challenges. All animals ATCC 53103 to post-weaning piglet infected with ETEC F4 came from a high-sanitary-status farm in which mothers were does not confer protection (Trevisi et al., 2009). A combination not vaccinated against E. coli. of B. longum and L. rhamnosus HN001 has also been reported Piglets were transferred to the experimental unit located in as an effective way to modulate the intestinal environment in the UAB, which consisted of four boxes of eight pens each (32 humans, with the reductions of potential harmful bacteria and pens, three animals per pen). Each 2 m2 pen comprised a feeder an increase of beneficial ones (Toscano et al., 2017). and water nipple to provide feed and water ad libitum. All On the other hand, prebiotics can also be useful against weaning rooms were equipped with an automatic heater and pathogenic bacteria (de Vrese and Marteau, 2007). Galacto- forced ventilation, and each pen had an individual heating light. oligosaccharides (GOSs) are obtained from lactose by At arrival, animals were distributed among treatment groups transgalactosylation reactions catalyzed by β-galactosidases. according to their initial body weight. The trial was a completely These reactions result in a chain of galactose units with a randomized design that included five experimental groups: terminal glucose unit (Tzortzis, 2009). This prebiotic has been (i) positive control (CTR+), pigs not challenged with ETEC associated with increases in lactobacilli and bifidobacteria but orally inoculated with a sterile placebo solution; (ii) negative Frontiers in Microbiology | www.frontiersin.org 2 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 control (CTR−), pigs orally challenged with ETEC F4; (iii) orbital incubator. A total volume of 6 ml from the culture probiotic (PRO), pigs receiving a combination of B. longum was given directly to the animals; the final concentration was subsp. infantis CECT 7210 and L. rhamnosus HN001 and orally 1 × 109 CFU/ml. Inoculum concentrations were also determined dosed with ETEC F4; (iv) prebiotic (PRE), pigs receiving before the inoculation by McFarland standards and were plated galacto-oligosaccharides and orally dosed with ETEC F4; and in Luria Agar (LA; made in house: tryptase, yeast extract, (v) synbiotic (SYN), pigs receiving the combination of the NaCl, and agar, Oxoid; Hampshire, United Kingdom) for the products supplied in the PRO and PRE groups and orally same day manual plate counting. dosed with ETEC F4. These challenged groups were equally distributed in three of the four rooms, while one full room Experimental Procedure was kept for non-inoculated control pigs; hence, the design The experiment lasted 15 days. Animals were offered the was unbalanced. Each experimental group had six replicates, experimental diets from the first moment after arrival. After except for the non-challenged groups, which had eight replicates. an adaptation period of 7 days (day 7), animals were orally In the challenged rooms, probiotic, and synbiotic treatments challenged with the pathogen. The inoculum was given to the were distributed within four pens on one side of the room, challenged groups via the oral route as a single dose of 6 ml and the control and prebiotic pens were on the other side of of ETEC F4 bacterial culture containing 6 × 109 CFU. The the room, separated by a corridor in between to avoid same amount of sterile broth was administered to non-challenged cross-contamination. piglets. To ensure that the animals had a full stomach at the Probiotic Strains, Prebiotic, and Diets TABLE 1 | Ingredient and nutritional composition of the diets. The tested probiotics were B. longum subsp. infantis CECT Ingredients (g/kg FM) CTR/PRO PRE/SYN 7210, supplied by Laboratorios Ordesa S.L., and L.s rhamnosus HN001 (Danisco USA Inc.). Both strains were provided Maize 207.4 196.3 lyophilized, containing 5 × 1010 and 3 × 1010 colony forming Wheat 180.0 170.0 units (CFU) per gram of product, respectively, in a maltodextrin Barley 2 row 170.0 160.9 base. The lyophilized probiotics were mixed daily into the feed Extruded soybean 149.1 141.1 Sweet whey-powder 100.0 94.6 for a final dosage of 5.5 × 107 and 3.3 × 107 CFU/g, respectively. (cattle) The feed was totally replaced daily. Before the trial, we confirmed Fish meal 60.0 56.8 the viability of the probiotics in the dry feed during the day. Soybean meal 44 80.0 75.7 The experimental prebiotic containing galacto-oligosaccharides Whey-powder 50% fat 25.0 23.6 (GOS) was in a syrup form and was thoroughly mixed with the Mono-calcium phosphate 6.8 6.4 Calcium carbonate 3.9 3.6 diet by hand every day to guarantee a homogenous blend (no (CaCO3) visual lumps). The final GOS concentration was 5% (w/w) trying L-Lysine HCL 4.5 5.0 to reach the amount of oligosaccharides provided by human milk Vit-Min Premix* 4.0 3.7 (5–10 g/L; Kunz et al., 2017) and also based in previous studies Sodium chloride (marine 2.5 2.3 supplementing GOS to neonatal and weaned piglets (Alizadeh salt) DL-Methionine 99 2.6 3.6 et al., 2016; Tian et al., 2019; Wang et al., 2019). When mixed L-Threonine 2.3 2.6 in the SYN diet, GOS was mixed prior to the probiotics. L-Tryptophan 0.6 0.7 Pre-starter diets were formulated in concordance with the L-Valine 1.5 1.9 nutrient requirement standards for pigs [National Research Prebiotic 0 5 Analyzed composition CTR/PRO PRE/SYN Council (NRC), 2012] and given in a mash form. The possible (g/kg FM) amino acid dilution in the PRE and SYN diets due to the Dry matter 920.2 916.1 incorporation of the prebiotic was compensated by the addition Ashes 49.6 48.2 of synthetic amino acids: 0.5 g L-valine, 0.9 g L-lysine HCl, Crude fat 64.1 61.6 1.2 g DL-methionine, 0.5 g L-threonine, and 0.2 g L-tryptophan Crude protein 203.2 191.1 Neutral detergent fiber 92.7 91.0 per kg of feed. Details for their ingredient and chemical Acid detergent fiber 32.7 32.4 composition are given in Table 1. Calculated composition (g/kg FM except ME) Metabolizable energy 3,378 3,213 (kcal/kg FM) Escherichia coli F4 Strain Crude protein 214 205 The bacterial strain of ETEC F4 used was isolated from feces Lysine 15 15 of 14-week old pigs and provided by the Infectious Diseases Methionine 6 7 Laboratory (Ref. 30/14) of the UAB. This strain presented the Threonine 10 10 Tryptophan 3 3 following virulence factors: F4ab, F4ac, LT, STb, and EAST1, Valine 11 10 and it was negative for K99, F6, F18, F41, STa, VT1, VT2, Cysteine 3 3 and EAE. The oral inoculum was prepared by a 12-h overnight Leucine 15 15 incubation at 37°C in brain heart infusion broth (Oxoid; Isoleucine 9 8 Hampshire, England) with slow agitation (250 rpm) in an *indicates that values of p are <0.05 and statistical difference is present. Frontiers in Microbiology | www.frontiersin.org 3 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 moment of the oral challenge, pigs were starved for 12 h analysis for short-chain fatty acids (SCFA) and lactic (night period from 8:00 pm to 8:00 am) and feed was reintroduced acid determination. 30 min before the inoculation. From each animal, 5 cm of distal ileum were collected, From the challenge onward, animals were checked daily to washed thoroughly with sterile phosphate-buffered saline (PBS), determine clinical signs and evaluate their post-inoculation opened longitudinally, and scraped with a glass microscope status – always by the same person. The fecal score was slide to obtain the mucosa sample to determine the number measured using a scale: 1 = solid and cloddy; 2 = soft with of enterobacteria and coliforms attached to the intestinal mucosa. shape; 3 = very soft or viscous liquid; and 4 = watery or with A subsample was also stored at −80°C for ETEC F4 quantification blood. The rectal temperature was assessed with a digital by qPCR. thermometer (Accuvet, Sanchung City, Taiwan) on days 1 For the histological study, 1 cm of ileum was removed, and 2 PI. opened longitudinally, and washed thoroughly with 4% For microbiological analysis, fecal samples were collected formaldehyde solution (Panreac; Castellar del Vallès, Spain) aseptically after spontaneous defecation or by digital stimulation before fixing them by immersion in the same solution. Blood at arrival and on the day of the inoculation (day 0 PI). Fecal samples were centrifuged (3,000 g for 15 min) after blood samples were always obtained from the largest animal of each coagulation; the obtained serum was stored at −20°C. pen (N = 32), except at day 4 PI, when the average piglet was sampled. Analytical Procedures Animal performance was monitored. Individual body weight Feed Analysis was registered at arrival (day 7) and on days 0, 4, and 8 PI, Chemical analyses of the diets, including dry matter (DM), and feed intake was monitored daily, associated with the regular ash, crude protein, and diethyl ether extract, were performed feed change to maintain probiotic viability. The average daily according to the Association of Official Agricultural Chemists gain (ADG) and average daily feed intake (ADFI) were calculated standard procedures (AOAC International, 1995). Neutral by pen for the adaptation period (days 0–6), days 0–4 PI detergent fiber and acid-detergent fiber were determined (days 7–11), and days 4–8 PI (days 12–15). The G:F ratio was according to the method of Van Soest et al. (1991). calculated by pen for the adaptation period (days 0–6), post inoculation period (days 7–15), and total length of the trial (days 0–15). The animals received no antibiotic treatment. Microbiological Analysis On days 4 and 8 PI, one pig per pen was euthanized. On For enterobacteria and coliform counts, samples were serially day 4 PI, the selected animal was the one with an intermediate diluted in Ringer’s lactate solution (Sigma-Aldrich, Madrid, initial BW, while on day 8 PI, the chosen piglet was the heaviest Spain) and proper dilutions seeded in MacConkey agar (Oxoid; from each pen. Animals were euthanized and sequentially Madrid, Spain) and eosin methylene blue agar (Scharlab; sampled during the morning of each day (between 8:00 and Barcelona, Spain). Plaques were incubated for 24 h at 37°C 13:00 h). An intramuscular injection containing 20 mg/kg and colonies were manually counted. From colon content and ketamine (Ketamidor; Wels, Austria) and 2 mg/kg of xylazine ileal mucosal scrapings, the total bacterial DNA was extracted (Xilagesic; Les Franqueses del Vallès, Spain) was given to the using QIAmp DNA Stool Mini Kit (Qiagen; West Sussex, animals to induce deep sedation. Prior to injection of the United Kingdom) following the manufacturer’s instructions. euthanasia drug, 10 ml sample blood was taken from each The ETEC F4 concentration was than assessed by qPCR targeting animal via venipuncture of the cranial cava vein using 10 ml the gene coding the F4 fimbria of E. coli, according to the blood collection tubes without anticoagulant (Aquisel; Madrid, procedure described by Hermes et al. (2013), using SYBR green Spain). Right after blood sampling, pigs were euthanized with dye with the ABI 7900 HT Sequence Detection System (PE an intravenous injection of sodium pentobarbital (140 mg/kg Biosystems, Warrington, United Kingdom) with optical grade BW, Euthasol, Le Vet B.V., Oudewater, Netherlands). Once 96-well plates. The results were scored in five levels according dead, animals were bled, the abdomen was immediately open, to the number of gene copies per gram of fresh matter (FM). and the gastrointestinal tract extracted. Scores were defined as following: negative = less than 4 A fecal sample taken from rectum was kept in ice and logarithmic units of gene copies per gram FM; low = 4–5.5 used for culture-based microbiology; it was analyzed within logarithmic units of gene copies per gram FM; medium = 5.5–7 4 h of collection. Subsequently, content from the ileum and logarithmic units of gene copies per gram FM; high = 7–8.5 proximal colon was collected and homogenized prior to pH logarithmic units of gene copies per gram FM; and very determination with a pH meter calibrated on each day of use high = more than 8.5 logarithmic units of gene copies per (Crison 52–32 electrode, Net Interlab; Barcelona, Spain). gram FM. Subsamples of colonic and ileal contents were collected for different analyses. One aliquot of colonic content was kept Short-Chain Fatty Acids, Lactic Acid, and frozen at −80°C for ETEC F4 quantification by real-time PCR Ammonia Analyses (qPCR). A set of ileal and colonic digesta aliquots were stored Short-chain fatty acids and lactic acid analyses were performed at −20°C in H2SO4 solution (3 ml of content plus 3 ml of on ileal and colonic digesta samples by gas liquid chromatography. 0.2 N H2SO4) for ammonia (NH3) determination and an Samples were subjected to acid-base treatment prior an ether additional sample (~10 g) was also frozen at −20°C until extraction and derivatization with N-(tertbutyldimethylsilyl)- Frontiers in Microbiology | www.frontiersin.org 4 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 N-methyl-trifluoroacetamide (MBTSTFA) plus 1% tert- (ANOVA) was used to examine the effect of the five experimental butyldimethylchlorosilane (TBDMCS) agent, using the method treatments and the MUC4 (MUC4+ and MUC4−) with the of Richardson et al. (1989), modified by Jensen et al. (1995). following model: Ammonia concentrations were assessed on ileal and colonic Yijk = m + treat i + MUC4 j + (treat ∗ MUC4ij ) + eij , digesta samples using a gas-sensitive electrode (Hatch Co.; CO, United States) combined with a digital voltmeter (Crison GLP 22, where Yijk relates to each observation of the outcome variable, Crison Instruments, S.A.; Barcelona, Spain) and following a m is the global mean, treati is the main effect of treatment, procedure described by Hermes et al. (2009). Samples were MUC4j is the main effect of MUC4 polymorphism, and diluted (1:2) in 0.16 M NaOH and, after homogenization, were treat*MUC4ij corresponds to the interaction between treatment centrifuged at 1500 g for 10 min. Once the ammonia was and MUC4. Finally, eij is the experimental error term. Regarding released, it was measured in the supernatants as a change in MUC4 effect and interaction term, they were removed from voltage (in mV). the model when found not to be significant. The effects on the post-slaughter measurements were examined using the R v3.4 (R Core Team, 2013) lm function for two-way Serum Analysis ANOVA, with the treatment and MUC4 effects as main effects. Serum samples were analyzed for Tumor Necrosis Factor-α When the MUC4 effect was not observed (p > 0.05), one-way (TNF-α) and pig major acute-phase protein (Pig-MAP). The ANOVA was performed with only the treatment effect. For concentrations of TNF-α were determined by enzyme-linked E. coli F4 prevalence values, data were subjected to frequency immunosorbent assay (ELISA) using the Quantikine Porcine analysis using the fisher.test function in the same package. TNF-α kit (R&D Systems; Minneapolis, United States), using ADFI and daily fecal scores were also analyzed using the lme4 a 4-parameter logistic regression (4PL) fit for calibration curve package (Bates et al., 2015) lmer function for a generalized adjustment. Pig-MAP concentrations were determined by a linear mixed model (GLMM) with a treatment-by-time turbidimetric method using the TURBOVET Pig-MAP (Acuvet interaction term and considering the animal as the random effect. Biotech; Zaragoza, Spain). Samples were not diluted in any of The experimental unit was the pen. For all analyses, the analyses. significance was set at p < 0.05 and p ≥ 0.05, but ≤ 0.10 were considered to indicate a statistical trend. When treatment Histological Analysis effects were established, the mean comparison was adjusted For histological study, ileal samples were dehydrated and with the Tukey-Kramer test. Data are presented as means and embedded in paraffin wax, sectioned at 4 μm, and stained residual standard error (RSE). with hematoxylin and eosin. Measurements of 10 different villus-crypt complexes per sample were considered including counting of intraepithelial lymphocytes (IEL), Goblet cells RESULTS (GC), and the number of mitotic cells in each complex. Analyses were performed with a light microscope (BHS, Following the oral challenge with ETEC, animals developed Olympus, Barcelona, Spain) following the procedure described moderate clinical signs of diarrhea that began to resolve by Nofrarías et al. (2006). spontaneously at the end of the study. Eight spontaneous casualties occurred (by group: 3 CTR−, 3 PRE, 1 PRO, and Piglet Genotyping for ETEC Susceptibility 1 SYN), and no humane euthanasia was required. Table 2 The mucin 4 (MUC4) gene has been suggested as one of the presents the analysis of mucin 4 (MUC4) polymorphism for candidate gene associated with piglet susceptibility to ETEC each treatment group. F4 (Jørgensen et al., 2003; Luise et al., 2019a). Thus, in the present study, the animals were genotyped for the polymorphism described by Jørgensen et al. (2003). For MUC4 genotype Performance Parameters determination, hair follicles were collected from 81 pigs. DNA Table 3 presents the results obtained for body weight (BW), was extracted following the procedure described by Luise et al. average daily feed intake (ADFI), average daily weight gain (2019b). A restriction fragment length polymorphism PCR (ADG), and gain to feed (G:F) ratio. During the 1st week, (PCR-RFLP) was performed following the guidelines described the animals from all experimental groups had a similar feed by Jørgensen et al. (2003). Pigs were classified into two groups: intake. However, after pathogen inoculation, all challenged susceptible pigs having the MUC4GG or MUC4CG genotype groups showed numerical reductions in intake, although the (MUC4+) or resistant pigs having MUC4CC homozygotes differences only reached statistical significance for the CTR− (MUC4−). and SYN groups for the days 0–4 post-inoculation (PI) period (p = 0.002) and for the SYN group for the days 4–8 PI period (p < 0.001). Statistical Analysis Figure 1 shows the evolution of feed intake per day during The results are expressed as means with their standard errors the experimental period. There were statistical differences unless otherwise stated. Microbiological counts were log between groups beginning on day 3. The SYN group was the transformed for analysis. A two-way analysis of variance most affected: this group consistently showed the lowest feed Frontiers in Microbiology | www.frontiersin.org 5 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 TABLE 2 | Distribution of resistant and susceptible animals for MUC4 gene in (1.60, 1.87, 1.95, 2.07, and 2.10 for CTR+, PRO, PRE, SYN, each experimental group. and CTR−, respectively). There was no treatment × MUC4 Treatment MUC4 genotype interaction. Rectal temperature was within the normal range and was not affected by the challenge, experimental diets, or Resistant Susceptible MUC4 status. Total animals Microbiological Analysis CTR+ 14 6 Table 4 shows enterobacteria and coliform plate counts from PRO 10 7 PRE 8 7 fecal samples taken at the arrival of the animals, before SYN 7 10 the oral challenge, and at days 4 and 8 PI. The table also CTR− 7 5 includes plate counts from ileal mucosa scrapings at day 4 p = 0.529 and 8 PI. Animals sampled at day 4 PI There were no differences between experimental groups on CTR+ 4 4 PRO 5 1 the day of arrival. However, at the end of the adaptation phase, PRE 3 3 the SYN group showed lower values of fecal enterobacteria SYN 1 5 (p = 0.001) and coliforms (p < 0.001) compared to the PRE CTR− 3 3 and PRO groups. On day 4 PI, the SYN group was the only p = 0.276 one with lower fecal coliforms counts compared to the CTR− Animals sampled at day 8 PI CTR+ 7 1 group (p = 0.023). On day 8 PI, only the PRE and CTR PRO 2 4 groups maintained higher plate counts compared to PRE 3 3 non-challenged pigs (p < 0.001 for both groups). SYN 4 2 MUC4 also had an impact on fecal counts of enterobacteria CTR− 4 2 and coliforms on day 4 PI: resistant animals had lower counts p = 0.294 compared to susceptible carriers (enterobacteria 6.27 vs. CTR+: non-inoculated animals receiving placebo; PRO: inoculated animals receiving the 7.44 CFU/g; P MUC4 = 0.015; coliforms 6.23 vs. 7.22 CFU/g; probiotics; PRE: inoculated animals receiving the prebiotic; SYN: inoculated animals P MUC4 = 0.010). The interaction between MUC4 and receiving the synbiotic; CTR−: Inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). p values were obtained using treatment was not significant, except for fecal enterobacteria Fisher’s exact test in R software. MUC4, mucin 4; PI, post-inoculation. before the challenge, when there were differences in the SYN group compared to the other treatment groups only in non-susceptible pigs who had a lower number of intakes; being significantly different from CTR+ on days 3, 4, enterobacteria (9.17, 10.86, 10.77, 5.92, and 10.05 CGU/g 5, 7, and 8 PI. for CTR+, PRO, PRE, SYN, and CTR−, respectively; P Similarly, ADG did not differ between treatments during interaction = 0.003). the 1st week of the trial, but it decreased after the challenge. Regarding ileal scrapings, only on day 8 PI were there During the first period after the inoculation (days 0–4 PI), differences between treatments (p < 0.001). Specifically, the all the challenged groups showed significant decreases in PRO group was the only one not that different from the growth, except for the PRO group, in which numerical CTR+ group. reductions in weight gains did not reach statistical significance. The ETEC F4 quantification results by qPCR in colonic In the second phase (days 4–8 PI), the PRO and PRE groups digesta and ileal scrapings showed that MUC4 status significantly showed a fast recovery of weight gain. Indeed, they reached affected the prevalence of ETEC F4 in colonic digesta at day similar levels (even numerically higher) to non-inoculated 4 PI. The challenge effect was much clearer in carrier animals piglets. However, the SYN group showed a tendency toward (P MUC4 < 0.001), with no significant difference among lower weight gain compared to the CTR+ group (p = 0.083). treatments (Figure 3). At day 8 PI, there was an interaction There were no differences among the groups for G:F ratio between MUC4 and treatment, with a higher prevalence of during the trial. ETEC F4 in the SYN group – but only in the MUC4 non-carrier animals. MUC4 did not significantly impact the ileal scraping Clinical Signs counts at days 4 or 8 PI. Figure 2 shows the evolution of fecal consistency along with When considering all animals, regardless of MUC4 status the post-infection period. ETEC F4-challenged animals showed (Figure 4), there were differences between treatments in the higher fecal scores (lower fecal consistency) immediately after colonic prevalence of ETEC F4. On day 4 PI, the SYN group the inoculation; these differences were significant on days 1 showed a higher prevalence compared to the CTR+ and similar and 2 PI and recovered from day 3 PI onward. MUC4+ animals to the CTR− group; the PRO and PRE groups showed had worse scale numbers compared to MUC4− animals on intermediate levels (p = 0.01). On day 8 PI, the SYN group days 1 and 2 PI (2.03 vs. 2.47; p = 0.009 and 1.95 vs. 2.34; maintained high excretion levels (SYN vs. CTR+; p = 0.002), p = 0.013 for not susceptible and susceptible on days 1 and while animals from the CTR− group recovered from the 2 PI, respectively). Among the challenged groups, the PRO challenge and the values decreased; the PRO and PRE groups group had the best consistency and the SYN group the worst showed intermediate values. Frontiers in Microbiology | www.frontiersin.org 6 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 TABLE 3 | Effects of experimental treatments on feed intake and weight gain. Treatments RSD p CTR+ PRO PRE SYN CTR− Body weight (kg) Initial 5.02 5.09 5.01 5.05 5.03 0.142 0.880 Final 7.24 6.89 7.25 6.48 6.58 0.726 0.223 Average daily feed intake (g/day) Adaptation 105.1 111.8 118.8 102.3 83.9 27.86 0.285 0–4 PI 213.0a 185.3ab 179.0ab 124.1b 130.1b 42.35 0.002* 4–8 PI 404.8a 350.5a 330.9ab 207.7b 306.2ab 70.83 <0.001* Average daily gain (g/day) Adaptation 40.7 46.2 64.6 52.0 34.4 38.93 0.707 0–4 PI 155.4a 74.2ab 26.7b −13.7b −44.7b 71.50 <0.001* 4–8 PI 317.6a 329.3a 333.0a 198.5b 282.5b 82.15 0.047* Gain:Feed ratio Adaptation 0.34 0.04 0.36 0.54 0.50 0.665 0.727 PI 0.76 0.52 0.75 0.54 0.44 0.328 0.300 Total 0.68 0.50 0.65 0.59 0.51 0.185 0.298 CTR+: non-inoculated animals receiving placebo; PRO: inoculated animals receiving the probiotics; PRE: inoculated animals receiving the prebiotic; SYN: inoculated animals receiving the synbiotic; CTR−: Inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). Values of p were obtained by an ANOVA using the generalized linear procedure in R software. a,bIndicate statistically significant differences between groups. PI, post-inoculation; RSD, relative standard deviation. To differentiate RSD and values of p from the results numbers, they are in italics. *indicates that values of p are <0.05 and statistical difference is present. At the ileal level, the PRE group had the highest pH values TABLE 4 | Effects of experimental treatments on enterobacteria and coliform at day 4 PI; they were different from the CTR+ and PRO counts in fecal samples and ileal scrapings. groups (p < 0.05), but not from the other challenged groups. Treatments RSD p At day 8 PI, the PRE group had the lowest pH values (P treatment = 0.044). Ammonia concentrations diminished CTR+ PRO PRE SYN CTR− at day 8 PI in the PRO, PRE, and SYN groups compared to the CTR+ and CTR− groups (p = 0.044). There was no treatment Enterobacteria (log CFU/g MF) Feces effect in lactic acid or total SCFA concentrations. Arrival 9.35 9.76 8.10 8.91 7.87 1.259 0.242 In the colon, the pH was numerically increased at day 4 Day 0 PI 8.82ab 10.93a 10.76a 7.55b 10.12ab 1.685 0.001* PI in all challenged groups, although the differences compared Day 4 PI 6.41 6.63 6.92 6.29 8.52 1.501 0.043* to the CTR+ group were only significant for the SYN and Day 8 PI 5.40b 5.34b 6.85a 5.79b 6.75a 0.441 <0.001* CTR− groups (P treatment = 0.041). The PRE group was Ileal scrapings Day 4 PI 7.59 7.61 7.54 7.46 7.41 0.201 0.393 the only treatment with reduced ammonia values at day 4 Day 8 PI 6.62b 6.62b 7.62a 7.59a 7.24ab 0.423 <0.001* PI compared to the CTR+ group (P treatment = 0.007). There Total coliforms (log CFU/g FM) were no differences in lactic acid concentration at any time. Feces Regarding total SCFA, the SYN and CTR− groups showed a Arrival 9.18 9.70 8.09 8.81 7.87 1.325 0.335 Day 0 PI 8.34b 10.83a 10.66a 5.86b 9.72ab 1.248 <0.001* trend for a decreased concentration compared to the other groups Day 4 PI 6.35ab 6.31ab 6.82ab 5.90b 8.48a 1.380 0.008* (P treatment = 0.085) at day 4 PI but not at day 8 PI. At day 8 Day 8 PI 5.31b 5.22b 6.83a 5.71b 6.62a 0.441 <0.001* PI, there was a marked increase in lactic acid concentration in Ileal scrapings the CTR− group, although differences with other treatments did Day 4 PI 7.48 7.46 7.46 7.42 7.31 0.222 0.642 not reach statistical significance (p = 0.119). The molar ratios of Day 8 PI 5.90c 6.55bc 7.18ab 7.65a 7.29ab 0.576 <0.001* different SCFA showed some changes related to the experimental CTR+: non-inoculated animals receiving placebo; PRO: inoculated animals receiving the treatments. On day 4 PI, there was a numerical trend for an probiotics; PRE: inoculated animals receiving the prebiotic; SYN: inoculated animals increased acetic acid percentage in the challenged groups (P receiving the synbiotic; CTR−: inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). Values of p were obtained by an ANOVA using treatment = 0.057), and at day 8 PI, the PRO group showed a the generalized linear procedure in R software. a,bIndicate statistically significant higher acetic acid concentration compared to other inoculated piglets differences between groups. CFU, colony forming units; PI, post-inoculation; RSD, relative (p = 0.004). The PRO group showed lower percentages of branched standard deviation. To differentiate RSD and values of p from the results numbers, they chain fatty acids (BCFA) compared to the SYN group (p = 0.049), are in italics. *indicates that values of p are <0.05 and statistical difference is present. and valeric acid was reduced in the PRO, PRE, and SYN groups compared to the CTR+ group at day 4 PI (P treatment = 0.047). Intestinal Fermentation Notably, these treatment-mediated differences with regard Table 5 shows values of intestinal pH, ammonia, lactic acid, to the fermentation profile were influenced by MUC4 and short-chain fatty acid (SCFA) concentrations in ileal and polymorphism when analyzed at day 8 PI. The identified colonic content for the different treatments. interactions are shown in Figure 5. The SYN group animals Frontiers in Microbiology | www.frontiersin.org 7 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 FIGURE 1 | Evolution of feed consumption for the experimental groups during the entire experimental period. CTR+: non-inoculated animals receiving placebo; PRO: inoculated animals receiving the probiotics; PRE: inoculated animals receiving the prebiotic; SYN−: inoculated animals receiving the synbiotic; CTR−: inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). a and b indicate statistically significant differences between groups. Bars correspond to the standard error. ADFI, average daily feed intake; PreI, pre-inoculation; PI, post-inoculation. FIGURE 2 | Evolution of average fecal scores for the different experimental groups during the post-inoculation (PI) period. CTR+: non-inoculated animals receiving placebo; PRO−: inoculated animals receiving the probiotics; PRE−: inoculated animals receiving the prebiotic; SYN−: inoculated animals receiving the synbiotic; CTR−: inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). MUC4 represents the effect of polymorphism of the MUC4 gene. a,b, and c indicate statistically significant differences between groups. x and y indicate statistical trends among groups. Bars correspond to the standard error. PI, post-inoculation. showed a lower colonic pH and a higher amount of total had a sharp increase in the molar ratio, but only in susceptible SCFA in the colon, but only in the carrier pigs pigs (P interaction = 0.019). The BCFA molar ratio was (P interaction < 0.001). Regarding valerate, the SYN group increased in resistant animals in the SYN group and in susceptible Frontiers in Microbiology | www.frontiersin.org 8 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 FIGURE 3 | Effect of MUC4 polymorphism on the percentage of animals with different levels of ETEC F4 counts in colon content on days 4 and 8 post- inoculation (PI). Different animals were sampled on days 4 and 8 post-inoculation (PI). CTR+: non-inoculated animals receiving placebo; PRO−: inoculated animals receiving the probiotics; PRE: inoculated animals receiving the prebiotic; SYN: inoculated animals receiving the synbiotic; CTR−: inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). Values of p were obtained using Fisher’s exact test in R software. a and b indicate statistically significant differences between groups. MUC4+, Mucin 4 susceptible genotype; MUC4−, Mucin 4 resistant genotype; PI, post- inoculation. animals in the PRE group. Finally, piglets with a MUC4 resistant (P treatment = 0.023); any differences did not reach significance phenotype showed increased colon ammonia levels on day 4 with the rest of the groups. At day 8 PI, the SYN group PI (13.31 vs. 9.79 mmol/L; P MUC4 = 0.006); there was no showed the lowest values (p = 0.050). interaction with treatments. On day 4 PI, the Pig-MAP value was higher in the CTR− compared to the CTR+ group. The PRO and PRE groups had Immune Response levels closer to the CTR+ group, and the SYN group showed Table 6 presents values corresponding to serum levels of the intermediate levels (P treatment = 0.013). On day 8 PI, the pro-inflammatory cytokine tumor necrosis factor alpha (TNF-α) Pig-MAP concentration was clearly higher in the SYN group, and the acute-phase protein pig major acute-phase protein (Pig-MAP). whereas the rest of challenged groups resembled the CTR+ The TNF-α concentration on day 4 PI was higher in the group level (P treatment = 0.003). SYN group compared to the CTR− and PRE groups There were no effects due to MUC4 gene polymorphism. Frontiers in Microbiology | www.frontiersin.org 9 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 FIGURE 4 | Effect of experimental treatments on the percentage of animals with different levels of ETEC F4 counts in colonic digesta and ileal scrapings on days 4 and 8 post-inoculation (PI). Different animals were sampled on day 4 PI and 8 PI. CTR+: non-inoculated animals receiving placebo; PRO−: inoculated animals receiving the probiotics; PRE: inoculated animals receiving the prebiotic; SYN: inoculated animals receiving the synbiotic; CTR−: inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). Values of p were obtained using Fisher’s exact test in R software. a and b indicate statistically significant differences between groups. FM, fresh matter; PI, post-inoculation. Intestinal Histological Structure The challenge was also associated with a higher number The effects of the experimental treatments on villi height, crypt of mitotic cells (P treatment = 0.036), although this increase depth, and mitosis number are shown in Table 7. was not significant in the PRO group when compared to the The impact of the ETEC F4 oral challenge was evidenced CTR+ group (p = 0.649) at day 4 PI. Regarding goblet cells by the ileal epithelium structure. On day 4 PI, all the challenged and interepithelial lymphocytes (IELs), there were no significant groups showed a reduction in the villi height (p < 0.001), a differences related to the experimental treatments. phenomenon which was still significant on day 8 PI, except for MUC4 significantly affected the number of mitotic cells the PRO group (P treatment < 0.001). When analyzing crypt on day 4 PI: susceptible animals presented reduced mitosis depth on day 4 PI, the CTR− group had the lowest values, compared to resistant animals (0.23 vs. 0.29 cell whereas the PRE group had the highest; the difference between number/100 μm, P MUC4 = 0.049). Furthermore, the villous the two treatments was statistically significant (p = 0.028). height:crypt depth ratio on day 8 PI showed an interaction According to this result, the villous height:crypt depth ratio effect (p = 0.004). Specifically, the SYN group exhibited was modified by the oral challenge on both sampling days a higher villous:crypt ratio compared to the challenged (p < 0.001), with reductions in the ratio for all the challenged groups, but only in the susceptible animals (1.33, 1.07, groups except SYN. In that group, the reductions did not reach 1.01, 1.32, and 0.95 for CTR+, PRO, PRE, SYN and CTR−, statistical significance compared to the CTR+ group (p = 0.064). respectively). Frontiers in Microbiology | www.frontiersin.org 10 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 TABLE 5 | Effects of experimental treatments on ileal and colonic fermentation. Treatment RSD p Day PI CTR+ PRO PRE SYN CTR− Ileum 4 6.49c 6.58bc 6.94a 6.70abc 6.86ab 0.169 <0.001* pH 8 6.53 6.52 6.41 6.56 6.57 0.110 0.044* 4 1.99 1.86 0.99 1.84 1.53 0.973 0.390 NH3 (mmol/L) 8 2.69a 1.38b 1.52b 1.59b 2.04ab 0.835 0.044* Lactic acid 4 35.8 18.1 5.6 8.3 16.7 27.48 0.368 (mmol/kg) 8 19.8 24.5 27.6 23.1 12.5 19.60 0.742 4 3.85 3.59 2.33 4.71 1.99 2.306 0.288 SCFA (mmol/kg) 8 2.67 3.07 2.59 3.37 3.47 1.662 0.849 Colon 4 6.04c 6.11bc 6.35abc 6.66a 6.54ab 0.389 0.041* pH 8 6.02 5.92 6.04 5.95 5.79 0.307 0.149 4 16.4a 10.6ab 7.2b 10.6ab 12.7ab 4.86 0.007* NH3 (mmol/L) 8 8.84 6.63 7.15 9.02 9.16 3.01 0.477 Lactic acid 4 2.57 0.56 2.71 0.26 0.27 3.542 0.550 (mmol/kg) 8 0.85 0.43 0.10 0.00 7.27 4.877 0.119 4 124.1 105.7 100.6 75.5 78.1 33.72 0.085 SCFA(mmol/kg) 8 130.9 136.2 137.3 127.4 145.4 32.07 0.442 Molar ratio of SCFA (%) 4 50.7 58.6 59.6 52.8 55.4 5.99 0.057 Acetic 8 50.3ab 55.3a 48.7b 46.4b 48.7b 3.44 0.004* 4 26.6 25.1 26.7 28.5 28.1 3.68 0.518 Propionic 8 26.1 25.9 29.7 28.2 25.8 3.76 0.317 4 16.1 12.9 10.2 12.6 11.1 5.48 0.332 Butyric 8 17.6 14.2 14.1 16.7 18.2 3.33 0.118 4 4.20a 1.70b 1.72b 2.30b 2.37ab 1.665 0.047* Valeric 8 3.85ab 2.97b 5.28ab 6.23a 5.49ab 2.018 0.011* 4 2.30ab 1.70b 1.74ab 3.73a 2.95ab 1.119 0.031* BCFA 8 2.12 1.65 2.24 2.51 1.77 0.743 0.097 This table includes values corresponding to pH, ammonia concentration (NH3; mmol/L of FM), lactic acid (mmol/L of FM), total short-chain fatty acids (SCFA; mmol/L of FM), and molar ratio of these SFCA. CTR+: non-inoculated animals receiving placebo; PRO: inoculated animals receiving the probiotics; PRE: inoculated animals receiving the prebiotic; SYN: inoculated animals receiving the synbiotic; CTR−: inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). Values of p were obtained by an ANOVA using the generalized linear procedure in R software. a,b,cIndicate statistically significant differences between groups. BCFA, branched-chain fatty acids; NH3, ammonia concentration; PI, post-inoculation; RSD, relative standard deviation; SFCA, short-chain fatty acids. To differentiate RSD and values of p from the results numbers, they are in italics. *indicates that values of p are <0.05 and statistical difference is present. DISCUSSION F4 on some of the sampling days, as well as a worsened fecal consistency. These results support the usefulness of this candidate In this study, a piglet model of ETEC F4 colibacillosis was gene as marker for the genetic selection of farmed pig toward used to evaluate the efficacy of treatment with B. longum subsp. increased resistance to diarrhea (Jørgensen et al., 2003). However, infantis CECT 7210 and L. rhamnosus HN001, galacto- it is also fair to remark that nor MUC4 polymorphism or any oligosaccharides, and their combination. The current experimental other candidate marker has yet been confirmed as the univocal trial clearly demonstrated the effects caused by the pathogen causative gene for ETEC F4 susceptibility. Actually Rasschaert challenge. Specifically, in all the challenged groups, there was et al. (2007) showed that 30% of the genotypic resistant pigs a decrease in feed intake, followed by a reduction in weight are positive in the in vitro villous adhesion assay, suggesting gain. Piglets also showed impaired fecal consistency immediately that polymorphisms other than MUC4 could also been involved. after the inoculation, with evident effects on the intestinal Between those other candidate markers there has been described epithelium structure and the Pig-MAP response. single nucleotide polymorphisms (SNPs) located on MUC4, The work also analyses the possible role of the MUC4 gene MUC13, MUC20, the transferrin receptor (TFRC), tyrosine kinase in the development of the disease. Actually the presence of non-receptor 2 (ACK1), or the UDP-GlcNAc:betaGalbeta-1,3- ETEC F4-specific receptors on the intestinal brush border of N-acetylglucosaminyltransferase 5 (B3GNT5) gene (reviewed by pigs has been associated with MUC4 polymorphism (Peng et al., Luise et al., 2019a). Moreover, Goetstouwers et al. (2014) found 2007; Trevisi et al., 2009) and suggested as criteria for the a refined candidate region for F4ab/ac ETEC susceptibility situated inclusion of animals in ETEC F4 challenge models (Luise et al., proximal to MUC13 gene, however, it should be noted that this 2019a). As expected, in the present study, animals that were marker maps on a non-coding region. No protocols, in addition heterozygous or homozygous for the susceptible MUC4 allele to an Illumina chip or a next-generation sequencing (NGS) showed the elevated levels of enterobacteria, coliforms, and ETEC technique, are available for genotyping the pigs for this marker, Frontiers in Microbiology | www.frontiersin.org 11 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 TABLE 6 | Effect of experimental treatments in serum levels of acute-phase group, we could also demonstrate the positive effects of this protein Pig-MAP and TNF-α. probiotic combination against a Salmonella challenge, leading Treatment RSD p to a faster clearance of the pathogen and recovery from the intestinal damage (Rodríguez-Sorrento et al., 2020). This putative CTR + PRO PRE SYN CTR− reduction in the enteropathogen challenge promoted by the probiotic treatment could have also explained the intermediate TNF-α (pg/ml) level of mitosis observed between the CTR+ and CTR− groups Day 4 PI 85.7ab 97.6ab 75.3b 118.2a 76.7b 23.83 0.023* at day 4 PI and the faster recovery of villus height at day 8 Day 8 PI 83.6 74.3 83.1 70.2 93.8 13.68 0.050 PI. These effects could also be associated with the reduced Pig-MAP (mg/ml) Day 4 PI 0.59b 0.76b 0.72b 1.49ab 2.42a 0.986 0.013* response observed in Pig-MAP at day 4 PI. This treatment Day 8 PI 0.51b 0.57b 0.56b 2.43a 0.62b 0.752 0.003* showed values between 0.43 and 1.50 mg/ml, which are clearly below the 2 mg/ml considered to be normal in weanlings CTR+: non-inoculated animals receiving placebo; PRO: inoculated animals receiving the probiotics; PRE: inoculated animals receiving the prebiotic; SYN: inoculated animals [Piñeiro et al., 2009; normal (≤1 mg/ml), borderline (1–2 mg/ml) receiving the synbiotic; CTR−: inoculated animals receiving placebo. N = 6 for all and high levels (>2 mg/ml)]. Pig-MAP is commonly induced groups except for non-challenged animals (N = 8). Values of p were obtained by an by interleukin 6 (IL-6; González-Ramón et al., 2000) that is ANOVA using the generalized linear procedure in R software. a,bIndicate statistically simultaneously stimulated by nuclear factor kappa B (NF-κB) significant differences between groups. TNF-α, tumor necrosis factor alpha; Pig-MAP, pig major acute-phase protein; PI, post-inoculation; RSD, relative standard deviation. To activation (Brasier, 2010). NF-κB is a protein complex, which differentiate RSD and values of p from the results numbers, they are in italics. *indicates controls the expression of genes implicated in inflammation that values of p are <0.05 and statistical difference is present. process (Baker et al., 2011). Bifidobacterium infantis and L. rhamnosus can also modify the expression of IL-6 or NF-kB making nowadays difficult to routinely use for animal selection. (Khailova et al., 2014; Gamallat et al., 2016; Ishizuka et al., To better dilucidated the role of MUC4 gen in this study, it 2016), an eventuality that is consistent with our results. A had been interesting to determine the phenotypical presence of possible modulation of the immune response by this probiotic F4 receptors on brush borders, enterocytes, or villi of the small should also be considered. In this regard, previous works with intestine by in vitro test (Van den Broeck et al., 1999). However, B. infantis CECT 7210 have shown consistent increases in ileal as characterization of the F4ab/ac ETEC susceptibility was not IEL (13), and L. rhamnosus HN001 is attributed immune- initially planned, unfortunately the receptor expression in modulating properties (Gill et al., 2001). However, we could susceptible pigs could not be confirmed. not find such an effect in the present study when combined Regarding the potential of the probiotic to fight the disease, with L. rhamnosus HN001. the probiotic combination of B. infantis CECT 7210 and Galacto-oligosaccharides supplementation alone (the PRE L. rhamnosus HN001 ameliorated the impairment in weight group) in the piglets’ diet also promoted favorable outcomes. gain immediately after ETEC F4 inoculation (day 0–4 PI); the Like the probiotics alone, the PRE group presented improved only challenged group that was not significantly different from weight gain for the 4–8 PI period, reaching similar values to the CTR+ group. Moreover, in the 4–8 PI period, the PRO the CTR+ group (and even numerically higher). Supplementing group also showed improved weight gain, reaching levels similar diets with the GOS alone affected neither enterobacteria nor to the CTR+ group. This enhanced response could be the coliform populations nor ileal histomorphometry, although, like result of a competitive exclusion of the pathogen by the the PRO group, the prevalence of colonic ETEC F4 was between probiotics. In fact, at day 4 PI this group showed an ETEC the levels of the CTR+ and CTR− groups. At day 4 PI, the F4 colonic prevalence with values intermediate to those in PRE group showed the lowest Pig-MAP level compared to the the CTR+ and CTR− groups. In addition, there was a lower other challenged groups; only the PRE group was significantly number of fecal enterobacteria and coliform plate counts different from the CTR− group and similar to the CTR+ group. compared to the CTR− group as well as a reduced number In fact, GOS may alleviate inflammation, as shown in several of attached bacteria to the ileal mucus – only the experimental studies (Vulevic et al., 2013; Verheijden et al., 2015). Wang diet group was not different from the CTR+ group. Other et al. (2019) attributed the modulatory effects of inflammatory authors have reported the beneficial effects of these two probiotic state to the ability of GOS to increase anti-inflammatory cytokine strains against pathogenic agents, albeit when used separately. IL-10 while decreasing IL-8 by modulating, once again, NF-κB Barba-Vidal et al. (2017) showed, in a similar piglet model, protein complex. Regarding fermentation, the prebiotic encouraged that B. longum subsp. infantis CECT 7210 tends to reduce the a decrease in colonic and ileal ammonia, data that suggest a percentage of animals with countable ileal coliforms and decreases possible shift toward a less proteolytic and beneficial microbiota fecal Salmonella excretion after an ETEC F4 or a Salmonella due to its inclusion, as, for example, ammonia may buffer SCFA Typhimurium oral challenge, respectively. Moreover, this strain and block their activity (Davila et al., 2013; Shen et al., 2015). has antiviral activity when tested against rotavirus in mice The interest of mixing the tested probiotic strains with GOS (Muñoz et al., 2011) as well as antidiarrhoeal properties in is based on the reported ability of Bifidobacteria and Lactobacillus healthy infants (Escribano et al., 2018). Similarly, L. rhamnosus to degrade and use GOS as energy source. β-glycosidic linkages HN001 is successful against pathogens such as ETEC and connecting saccharides that comprise GOS are hydrolyzed in Staphylococcus aureus (Gopal et al., 2001; Inturri et al., 2016; the colon by these two genera of bacteria that bear β-galactosidases Eggers et al., 2018). More recently, in previous studies of our (Andersen et al., 2012; Garrido et al., 2013) and, specifically, Frontiers in Microbiology | www.frontiersin.org 12 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 FIGURE 5 | Interactions between MUC4 gene polymorphism and diets on pH values, total short-chain fatty acids (SCFA), valerate molar ratio, and branched-chain fatty acids molar ratio (BCFA) on day 8 PI. CTR+: non-inoculated animals receiving placebo; PRO−: inoculated animals receiving the probiotics; PRE−: inoculated animals receiving the prebiotic; SYN−: inoculated animals receiving the synbiotic; CTR−: inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). MUC4 represents the effect of polymorphism of the MUC4 gene. a and b indicate statistically significant differences between groups. Bars correspond to standard error. BCFA, branched-chain fatty acids; MUC4+, Mucin 4 susceptible genotype; MUC4−, Mucin 4 resistant genotype; SCFA, short-chain fatty acids. TABLE 7 | Effects of treatments on ileal histomorphological parameters on days 4 and 8 post-inoculation (PI). Day PI Treatment RSD p CTR+ PRO PRE SYN CTR− 4 311.8a 241.4b 246.8b 245.3b 220.9b 34.76 <0.001* Villi height (μm) 8 336.8a 291.8ab 272.1b 269.0b 266.5b 28.59 <0.001* Crypth depth 4 272.4ab 271.9ab 282.5a 251.6ab 241.8b 22.19 0.020* (μm) 8 271.3 274.0 271.3 250.5 278.3 36.17 0.710 Villi height: crypt 4 1.14a 0.88b 0.87b 0.97ab 0.91b 0.114 <0.001* depth ratio 8 1.24a 1.06b 1.00b 1.09ab 0.96b 0.099 <0.001* IEL (cell 4 0.37 0.50 0.52 0.59 0.38 0.207 0.285 number/100 μm) 8 0.69 0.56 0.66 0.60 0.71 0.293 0.885 GC (cell 4 2.38 1.96 2.39 2.95 2.08 1.031 0.513 number/100 μm) 8 1.94 1.77 2.28 2.07 2.76 0.790 0.250 Mitosis (cell 4 0.16a 0.24ab 0.32b 0.30b 0.31b 0.103 0.036* number/100 μm) 8 0.23 0.22 0.27 0.22 0.26 0.098 0.896 CTR+: non-inoculated animals receiving placebo; PRO: inoculated animals receiving the probiotics; PRE: inoculated animals receiving the prebiotic; SYN: inoculated animals receiving the synbiotic; CTR−: inoculated animals receiving placebo. N = 6 for all groups except for non-challenged animals (N = 8). Values of p were obtained by an ANOVA using the generalized linear procedure in R software. a,bIndicate statistically significant differences between groups. GC, Goblet cells; IEL, intraepithelial lymphocytes, PI, post-inoculation; RSD, relative standard deviation. To differentiate RSD and values of p from the results numbers, they are in italics. *indicates that values of p are <0.05 and statistical difference is present. B. infantis CECT 7210 prefers GOS as a growth substrate rather together with different types of GOS, whether or not it is very than other prebiotics (Ruiz et al., 2020). Several works have pure (Hong et al., 2016; Monteagudo-Mera et al., 2016; reported a rise in Bifidobacteria and Lactobacillus when tested Kittibunchakul et al., 2018). However, GOS utilization varies Frontiers in Microbiology | www.frontiersin.org 13 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 depending on the bacterial strain and the oligosaccharide consistent with previous findings (Luise et al., 2019b). Moreover, composition (Thongaram et al., 2017). Hence, considering the the concentration of inflammatory serum markers was higher previously mentioned characteristics, administering these in the SYN group. On day 4 PI, the TNF-α value was higher probiotics with GOS potentially would lead to synergistic activity in the SYN group compared to the CTR− group, although it enhancing the benefits that each component produces on its was not different from the PRO or CTR+ groups. Regarding own, however, in the present study, we were not able to Pig-MAP, whereas on day 4 PI the SYN group showed a level demonstrate any additive o synergic effect. In the literature, between the CTR+ and CTR− groups, at the end of the trial works evaluating the synbiotic strategy are limited, and the (day 8 PI), the SYN group maintained markedly high values, results have not always been consistent. For example, Tanner in contrast to the rest of the challenged groups that were able et al. (2014) and Abrahamse-Berkeveld et al. (2016) observed to normalize Pig-MAP levels. A higher energy expenditure positive outcomes, including enhanced growth and pathogen associated with an inflammatory response could have also inhibition, when mixing a bifidobacteria strain with GOS in contributed to explain the lower performance of the SYN animals. both in vitro and in vivo studies. By contrast, Krumbeck et al. Based on the above findings, it is evident that the synbiotic (2018) did not find any synergy in the capacity of improvement combination did not show any additive or beneficial effect on intestinal barrier function when combining bifidobacteria and the challenged animals but could even have provided a better GOS in humans. Previous studies of our group also failed to opportunity for ETEC F4+ to thrive. Whereas before the demonstrate the synergic activities of combination of the strains inoculation the SYN treatment showed a positive outcome with CECT 7210 and HN001 with FOS in Samonella orally challenged reductions in the number of enterobacteria and coliforms, after piglets (Rodríguez-Sorrento et al., 2020). In the present study, the ETEC challenge the combination of the probiotic and we observed some synergy before the challenge, particularly prebiotic could have reduced the ability of the animal to fight in the number of enterobacteria and coliforms, which were the pathogen. An explanation for that phenomenon could lower compared to the rest of supplemented groups. This effect be the complex interactions between members of the intestinal could suggest a positive impact of the synbiotic treatment on microbiota that could have been potentially disturbed by the the resident microbiota, specifically promoting the growth of synbiotic. It is known that a well-established and developed specific microorganisms, which together with the probiotics, microbiota is characterized by an equilibrium between different would have displaced enterobacteria. These results are particularly microbial groups through cross-feeding, competitive exclusion, relevant since the 1st week after weaning is one of the most and quorum sensing mechanisms that provide to the microbial critical periods in the pig’s life, in which they have to cope community with a high resilience against external inputs (Walter with numerous stressors and dysbiosis is commonly present et al., 2018). The administration of SYN could have determined (Gresse et al., 2017; Motta et al., 2019). Shifts of microbiota changes in the sequence of gut colonization of the young pig toward bifidobacteria via the administration of a synbiotic that during the 1st week post-weaning leading to a microbial contains GOS [±oligofructose (OF)] and strains of Bifidobacterium ecosystem more susceptible to be colonized by opportunistic have been reported in trials with healthy newborn babies and pathogens like ETEC F4. The fermentation of GOS by the infants (Simeoni et al., 2016; Chua et al., 2017). In particular, probiotic along the small intestine, could have increased the the latter authors attributed this outcome to the increase of amount of probiotic reaching the hindgut and also have reduced endogenous bifidobacteria and also in piglets by OF (Trevisi the amount of GOS arriving to the colon as intact prebiotic. et al., 2008; Modesto et al., 2009). However, the outcomes in This hypothesis could explain why the ammonia concentration the SYN group were different after the challenge. During the was only decreased in colon at day 4 PI in the PRE but not acute period of the infection (days 0–4 PI), the SYN group in the SYN group as less GOS would have arrived to the showed a larger decrease in feed intake compared to the other colon. Moreover, the reduction observed in enterobacteria and treatments, with similar values to challenged pigs that were coliforms with SYN, although generally regarded as positive, not supplemented. This feed intake depression was even more could have been a consequence of a reduced biodiversity and pronounced in the 4–8 PI period. According to this reduced resilience of the ecosystem. Not necessarily a reduction in feed intake, SYN and CTR− were the only two groups that number of enterobacteria in the microbial community is always lost weight in the 0–4 PI period, and the SYN group showed positive. Previous authors have described how commensal a trend for lower gains compared to the CTR− group in the Enterobacteriaceae can protect neonates against Salmonella 4–8 PI period. Together with reduced intake, impaired nutrient colonization through oxygen competition (Litvak and Bäumler, utilization associated with diarrhea could also explain weight 2019) as well as producing bacteriocins able to eliminate close- loss (Clements et al., 2012). Although we were unable to find related competitors (Stecher, 2015). Seems therefore plausible differences in the fecal score due to the dietary supplementation, to hypothesize that the biodiversity of the intestinal community the highest fecal scores on day 1 PI occurred in the SYN could have been reduced by the synbiotic treatment, making group, with values that were significantly higher than those the ecosystem more susceptible to dysbiosis with more niches found in the PRO group. This trend for a more acute peak becoming available for opportunistic/pathogenic bacteria. of diarrhea in the SYN group is supported by the higher Interestingly, the impact of the synbiotic treatment on prevalence of ETEC F4 found in the colon on day 8 PI compared microbial metabolism seemed to be MUC4 genotype dependent. to the CTR+ group. This outcome was particularly evident in The SCFA concentration was increased, and also the molar the groups of animals with the MUC4 susceptible genotype, ratio of valeric acid at day 8 PI, but only in pigs with the Frontiers in Microbiology | www.frontiersin.org 14 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 MUC4 susceptible genotype. A more acute course of diarrhea, Nevertheless, not all the evaluated parameters were modified determined by the susceptible genotype, could have promoted by MUC4 status. Specifically, neither the number of enterobacteria a differential effect on microbial activity, suggesting that the and coliforms attached to the ileal epithelium nor ileal effect of the SYN was somehow challenge dependent. In support histomorphology were altered. This outcome could be due to of this hypothesis, it is known that the MUC4 genotype can other genes and receptors involved in the susceptibility to ETEC influence the small intestinal glycomic patterns (Luise et al., F4 infection (Nguyen et al., 2012). In this regard, Rasschaert 2019b) and the intestinal microbiota (Massacci et al., 2020), et al. (2007) confirmed that the in vitro adhesion of F4ac and facts that could explain the differential response of the susceptible F4ab ETEC to the villous brush borders is not always associated animals to the SYN treatment. The increase observed in the with the MUC4 gene, suggesting the existence of at least one molar proportion of valeric acid with SYN supplementation other receptor for F4ab/ac fimbria. Indeed, other genes like could be attributed to favored valeric-producing bacteria. In MUC13 or TNRC (transferrin receptor gene) have also been this vain, Van Nevel et al. (2005) observed that the inclusion associated with ETEC F4 susceptibility (Zhang et al., 2008; of galactomannans from locust bean gum increases valeric Jacobsen et al., 2010; Ren et al., 2012). acid and decreases the mitotic index in crypts, similarly to In our experiment, we did not select animals based on the present study, where we also found a lower mitosis index MUC4 susceptibility, but we determined this characteristic at and shorter crypts in the susceptible animals of the SYN group. the end of the trial. Although many studies with ETEC challenges It is also fair to remark that results from challenge studies do not apply any control a priori or a posteriori on the genetic like this one are always conditioned by the experimental model susceptibility of the animals, we found that this could be of of disease and need to be approached with caution. An oral interest to better understand the variability of the data and challenge with a single high dose of ETEC F4 is undoubtedly to elucidate the role of this candidate gene in the piglet response far from what would naturally occur, namely repeated lower in front of an ETEC F4 challenge. The MUC4 analysis showed doses. Factors, as the inoculated bacterial strains, doses, timing certain unbalanced distribution of susceptible animals between of the inoculum, post-weaning age, and immunity status of treatments. This could have had an impact on the results, the animals, undoubtedly are variables that could determine particularly on those parameters that were influenced by MUC4 differential responses (Luise et al., 2019a). One of the factors polymorphism (performance, fecal consistency, numbers of fecal that could deserve some discussion is timing for inoculation. coliforms, and colonic ETEC F4). When the whole group of We inoculated animals 7 days post-weaning (around 28 days animals are considered, the ratio between resistant and susceptible of age) aiming to test the potential prophylactic effect of the animals varied between 0.7 for SYN to 2.3 for CTR+ (see symbiotic. Inoculating animals in shorter times (i.e., 1–2 days Table 2). Although it was a reasonable difference, it could post-weaning) could had better resembled the situation in the partially explain the lower ADFI and ADG observed for the farm, tackling the time of the highest risk for intestinal dysbiosis SYN treatment. When contemplating only the sampled animals and colonization by ETEC. However, challenging animals only after euthanasia, the distribution of resistant and susceptible 1–2 days after weaning had limited the possibilities for the animals was more unbalanced between treatments causing a symbiotic to exert a prophylactic effect. Considering this, different scenario between treatments at days 4 and 8 PI. we decided to inoculate animals after 7 days post-weaning, Whereas at day 4 PI, PRO treatment presented a favorable although undoubtedly this could had determined differences ratio of 5 and SYN of 0.7, at day 8 PI PRO showed a ratio with what is found in the field. of 0.5 and SYN of 2. Nonetheless, SYN treatment always Regarding the possible role of the MUC4 gene on results exhibited higher ETEC F4 counts than PRO, evidencing that, of this trial, it is known that ETEC attaches to the jejunal despite the uneven distribution, SYN treatment consistently brush border through the binding of its fimbriae to many brought about the same effects both sampling days. Also, it putative receptors. These receptors are carbohydrates of is interesting to remark that in defiance of the association glycoproteins in the intestinal epithelial cells and intestinal between the MUC4 polymorphism and fecal counts of coliforms, mucus, which have been shown to differ among pigs (Van particularly relevant at day 4 PI (7.43 vs. 6.27 log CFU/g for den Broeck et al., 2000; Rasschaert et al., 2007). Different MUC4+ and MUC4−, respectively; p = 0.027), the SYN treatment antigenic variants of F4 fimbriae have been identified (Bakker was always associated with lower plate counts compared to et al., 1991), including the F4ab and F4ac variants, in the CTR−, regardless the dissimilar distribution of resistant and strain used in this study. MUC4 codes for a membrane-bound susceptible animals (5.90 vs. 8.48 log CFU/g, p < 0.01 at day O-glycoprotein in the mucus layer. A mutation in this gene 4 PI and 5.71 vs. 6.62; p < 0.001 at day 8 PI). Altogether, it has been proposed as a useful genetic marker to identify appears that, although the unbalanced distribution of MUC4 susceptible genotypes (Jørgensen et al., 2003; Luise et al., 2019a). polymorphism could have increased residual variability and In our study, as expected, MUC4 susceptible piglets showed a have had an impact on the magnitude of the response, the worsened fecal consistency, as well as a lower weight daily direction and meaning of the observed differences would gain and higher numbers of enterobacteria, coliforms, and ETEC be consistent. F4 in colon. Other authors have also described similar associations In conclusion, the multi-strain probiotic B. longum subsp. in challenge and non-challenge conditions (Casini et al., 2003; infantis CECT 7210 and L. rhamnosus HN001 reduced growth Trevisi et al., 2015; Luise et al., 2019b; Sterndale et al., 2019); impairment after oral ETEC F4 challenge. Furthermore, this these data support the usefulness of this candidate gene. supplementation seems to provide competitive exclusion for Frontiers in Microbiology | www.frontiersin.org 15 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 ETEC F4, because there were fewer colonic enterobacteria ETHICS STATEMENT and coliforms in the gut and a trend to diminish the pathogen. This outcome could explain the lower Pig-MAP The animal study was reviewed and approved by Animal and levels and improved villus height found after 1 week of the Human Experimental Ethical Committee of the Autonomous challenge with this treatment. The supplementation of the University of Barcelona (Permit No. CEEAH: 4026 diets with galacto-oligosaccharides also diminished the growth DMAH: 10118). impairment induced by the challenge and is associated with lower levels of plasmatic Pig-MAP. This phenomenon suggests a modulation of the inflammatory response by this prebiotic. AUTHOR CONTRIBUTIONS These beneficial effects were not synergistic when the probiotics and the prebiotic were administered together. The AR-S participated in the experimental design and was responsible synbiotic treatment had a differential impact on colonic for the animal trial, laboratory analysis, data analysis, and fermentation depending on MUC4 susceptibility. The writing. LC participated in the experimental design, animal differential response in animals with more acute diarrhea trials, data analysis, and writing. PL-C participated in animal suggests that the absence of synergistic effect could be related trials and data analysis. GC-O, JM-M, and MR-P participated to the limitations of experimental model that rely on a in the experimental design and contributed to data analysis single exposure to high doses of the pathogen rather than and writing. DL and PT contributed to laboratory analysis, the low-dose exposure that would occur naturally. More data analysis, and writing. SM-O participated in the experimental research should be performed in this field to understand design, animal trials, laboratory analysis, data analysis, and the complex interactions produced in the gastrointestinal writing. All authors contributed to the article and approved tract, with an emphasis on the microbiota establishment at the submitted version. early ages. FUNDING DATA AVAILABILITY STATEMENT This project has been supported by the CDTI (Ministerio de The original contributions presented in the study are included Ciencia e Innovación de España) and the Fondo Europeo de in the article/supplementary material, further inquiries can Desarrollo Regional (FEDER; Acronym: SMARTFOODS, project be directed to the corresponding author. number IDI-20141206). REFERENCES Brasier, A. R. (2010). The nuclear factor-κB–interleukin-6 signalling pathway mediating vascular inflammation. Cardiovasc. Res. 86, 211–218. doi: 10.1093/ Abrahamse-Berkeveld, M., Alles, M., Franke-Beckmann, E., Helm, K., Knecht, R., cvr/cvq076 Köllges, R., et al. (2016). Infant formula containing galacto-and fructo- Canfora, E. E., van der Beek, C. M., Hermes, G. D. A., Goossens, G. H., oligosaccharides and Bifidobacterium breve M-16V supports adequate growth Jocken, J. W. E., Holst, J. J., et al. (2017). Supplementation of diet with and tolerance in healthy infants in a randomised, controlled, double-blind, galacto-oligosaccharides increases bifidobacteria, but not insulin sensitivity, prospective, multicentre study. J. Nutr. Sci. 5:e42. doi: 10.1017/jns. in obese prediabetic individuals. Gastroenterology 153, 87–97.e3. doi: 10.1053/j. 2016.35 gastro.2017.03.051 Alizadeh, A., Akbari, P., Difilippo, E., Schols, H. A., Ulfman, L. H., Casini, L., Bosi, P., Gremokolini, C., Trevisi, P., Mazzoni, M., and Bonilauri, P. Schoterman, M. H., et al. (2016). The piglet as a model for studying dietary (2003). Oral challenge with E.coli K88 as a tool to assess growth and health components in infant diets: effects of galacto-oligosaccharides on intestinal performance in feeding trials of-weaned pigs. Ital. J. Anim. Sci. 2(Suppl. 1), functions. Br. J. Nutr. 115, 605–618. doi: 10.1017/S0007114515004997 358–360. doi: 10.4081/ijas.2003.11676010 Andersen, J. M., Barrangou, R., Hachem, M. A., Lahtinen, S. J., Goh, Y. J., Chua, M. C., Ben-Amor, K., Lay, C., Neo, A. G. E., Chiang, W. C., Rao, R., Svensson, B., et al. (2012). Transcriptional analysis of prebiotic uptake and et al. (2017). Effect of synbiotic on the gut microbiota of cesarean delivered catabolism by Lactobacillus acidophilus NCFM. PLoS One 7:e44409. doi: infants: a randomized, double-blind, multicenter study. J. Pediatr. Gastroenterol. 10.1371/journal.pone.0044409 Nutr. 65, 102–106. doi: 10.1097/MPG.0000000000001623 AOAC International (1995). Official methods of analysis of AOAC international. Clements, A., Young, J. C., Constantinou, N., and Frankel, G. (2012). Infection Washington, DC: AOAC. strategies of enteric pathogenic Escherichia coli. Gut Microbes 3, 71–87. doi: Baker, R. G., Hayden, M. S., and Ghosh, S. (2011). NF-κB, inflammation, and 10.4161/gmic.19182 metabolic disease. Cell Metab. 13, 11–22. doi: 10.1016/j.cmet.2010.12.008 Davila, A. M., Blachier, F., Gotteland, M., Andriamihaja, M., Benetti, P. H., Bakker, D., Willemsen, P. T., Willems, R. H., Huisman, T. T., Mooi, F. R., Sanz, Y., et al. (2013). Intestinal luminal nitrogen metabolism: role of the Oudega, B., et al. (1991). Identification of minor fimbrial subunits involved gut microbiota and consequences for the host. Pharmacol. Res. 68, 95–107. in biosynthesis of K88 fimbriae. J. Bacteriol. 174, 6350–6358. doi: 10.1128/ doi: 10.1016/j.phrs.2012.11.005 jb.174.20.6350-6358.1992 de Vrese, M., and Marteau, P. R. (2007). Probiotics and prebiotics: effects on Barba-Vidal, E., Castillejos, L., López-Colom, P., Rivero Urgell, M., Moreno diarrhea. J. Nutr. 137(3 Suppl. 2), 803S–811S. doi: 10.1093/jn/137.3. Muñoz, J. A., and Martín-Orúe, S. M. (2017). Evaluation of the probiotic 803S strain Bifidobacterium longum subsp. infantis CECT 7210 capacities to improve Eggers, S., Barker, A. K., Valentine, S., Hess, T., Duster, M., and Safdar, N. health status and fight digestive pathogens in a piglet model. Front. Microbiol. (2018). Effect of Lactobacillus rhamnosus HN001 on carriage of 8:533. doi: 10.3389/fmicb.2017.00533 Staphylococcus aureus: results of the impact of probiotics for reducing Bates, D., Mächler, M., Bolker, B., and Walker, S. (2014). Package Lme4: linear infections in veterans (IMPROVE) study. BMC Infect. Dis. 18:129. doi: mixed-effects models using Eigen and S4. R package version. 67. 10.1186/s12879-018-3028-6 Frontiers in Microbiology | www.frontiersin.org 16 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 Escribano, J., Ferré, N., Gispert-Llaurado, M., Luque, V., Rubio-Torrents, C., Jørgensen, C. B., Cirera, S., Anderson, S. I., Archibald, A. L., Raudsepp, T., Zaragoza-Jordana, M., et al. (2018). Bifidobacterium longum subsp infantis Chowdhary, B., et al. (2003). Linkage and comparative mapping of the CECT7210-supplemented formula reduces diarrhoea in healthy infants: a locus controlling susceptibility towards E. coli F4ab/ac diarrhoea in pigs. randomized controlled trial. Pediatr. Res. 83, 1120–1128. doi: 10.1038/pr.2018.34 Cytogenet. Genome Res. 102, 157–162. doi: 10.1159/000075742 European Commission (2010). Directive 2010/63/EU. Kechagia, M., Basoulis, D., Konstantopoulou, S., Dimitriadi, D., Gyftopoulou, K., Fijan, S., Šulc, D., and Steyer, A. (2018). Study of the in vitro antagonistic Skarmoutsou, N., et al. (2013). Health benefits of probiotics: a review. ISRN activity of various single-strain and multi-strain probiotics against Escherichia Nutr. 2013:481651. doi: 10.5402/2013/481651 coli. Int. J. Environ. Res. Public Health 15:E1539. doi: 10.3390/ijerph15071539 Khailova, L., Petrie, B., Baird, C. H., Dominguez Rieg, J. A., and Wischmeyer, P. E. Gamallat, Y., Meyiah, A., Kuugbee, E. D., Hago, A. M., Chiwala, G., Awadasseid, A., (2014). Lactobacillus rhamnosus GG and Bifidobacterium longum attenuate et al. (2016). Lactobacillus rhamnosus induced epithelial cell apoptosis, ameliorates lung injury and inflammatory response in experimental sepsis. PLoS One inflammation and prevents colon cancer development in an animal model. 9:e97861. doi: 10.1371/journal.pone.0097861 Biomed. Pharmacother. 83, 536–541. doi: 10.1016/j.biopha.2016.07.001 Kittibunchakul, S., Maischberger, T., Domig, K. J., Kneifel, W., Nguyen, H. M., Garrido, D., Ruiz-Moyano, S., Jimenez-Espinoza, R., Eom, H. J., Block, D. E., Haltrich, D., et al. (2018). Fermentability of a novel galacto-oligosaccharide and Mills, D. A. (2013). Utilization of galactooligosaccharides by Bifidobacterium mixture by Lactobacillus spp. and Bifidobacterium spp. Molecules 23:E3352. longum subsp. infantis isolates. Food Microbiol. 33, 262–270. doi: 10.1016/j. doi: 10.3390/molecules23123352 fm.2012.10.003 Krumbeck, J. A., Rasmussen, H. E., Hutkins, R. W., Clarke, J., Shawron, K., Gill, H. S., Shu, Q., Lin, H., Rutherfurd, K. J., and Cross, M. L. (2001). Protection Keshavarzian, A., et al. (2018). Probiotic Bifidobacterium strains and against translocating Salmonella typhimurium infection in mice by feeding galactooligosaccharides improve intestinal barrier function in obese adults the immuno-enhancing probiotic Lactobacillus rhamnosus strain HN001. but show no synergism when used together as synbiotics. Microbiome 6:121. Med. Microbiol. Immunol. 190, 97–104. doi: 10.1007/s004300100095 doi: 10.1186/s40168-018-0494-4 Goetstouwers, T., Van Poucke, M., Coppieters, W., Nguyen, V. U., Melkebeek, V., Kumar, M., Dhaka, P., Vijay, D., Vergis, J., Mohan, V., Kumar, A., et al. (2016). Coddens, A., et al. (2014). Refined candidate region for F4ab/ac enterotoxigenic Antimicrobial effects of Lactobacillus plantarum and Lactobacillus acidophilus Escherichia coli susceptibility situated proximal to MUC13 in pigs. PLoS against multidrug-resistant enteroaggregative Escherichia coli. Int. J. Antimicrob. One 9:e105013. doi: 10.1371/journal.pone.0105013 Agents 48, 265–270. doi: 10.1016/j.ijantimicag.2016.05.014 González-Ramón, N., Hoebe, K., Alava, M. A., Van Leengoed, L., Piñeiro, M., Kunz, C., Meyer, C., Collado, M. C., Geiger, L., García-Mantrana, I., Bertua-Ríos, B., Carmona, S., et al. (2000). Pig MAP/ITIH4 and haptoglobin are interleukin-6- et al. (2017). Influence of gestational age, secretor, and Lewis blood group dependent acute-phase plasma proteins in porcine primary cultured hepatocytes. status on the oligosaccharide content of human Milk. J. Pediatr. Gastroenterol. Eur. J. Biochem. 267, 1878–1885. doi: 10.1046/j.1432-1327.2000.01195.x Nutr. 64, 789–798. doi: 10.1097/MPG.0000000000001402 Gopal, P. K., Prasad, J., Smart, J., and Gill, H. S. (2001). In vitro adherence Lanata, C. F., Fischer-Walker, C. L., Olascoaga, A. C., Torres, C. X., Aryee, M. J., properties of Lactobacillus rhamnosus DR20 and Bifidobacterium lactis DR10 Black, R. E., et al. (2013). Global causes of diarrheal disease mortality in strains and their antagonistic activity against an enterotoxigenic Escherichia children <5 years of age: a systematic review. PLoS One 8:e72788. doi: coli. Int. J. Food Microbiol. 67, 207–216. doi: 10.1016/S0168-1605(01)00440-8 10.1371/journal.pone.0072788 Gresse, R., Chaucheyras-Durand, F., Fleury, M. A., Van de Wiele, T., Forano, E., Litvak, Y., and Bäumler, A. J. (2019). The founder hypothesis: A basis for and Blanquet-Diot, S. (2017). Gut microbiota dysbiosis in postweaning microbiota resistance, diversity in taxa carriage, and colonization resistance piglets: understanding the keys to health. Trends Microbiol. 25, 851–873. against pathogens. PLoS Pathog. 15:e1007563. doi: 10.1371/journal.ppat.1007563 doi: 10.1016/j.tim.2017.05.004 Liu, L., Oza, S., Hogan, D., Perin, J., Rudan, I., Lawn, J. E., et al. (2015). Grimaldi, R., Swann, J. R., Vulevic, J., Gibson, G. R., and Costabile, A. Global, regional, and national causes of child mortality in 2000-13, with (2016). Fermentation properties and potential prebiotic activity of Bimuno galacto-oligosaccharide (65% galacto-oligosaccharide content) on in vitro ® projections to inform post-2015 priorities: an updated systematic analysis. Lancet 385, 430–440. doi: 10.1016/S0140-6736(14)61698-6 gut microbiota parameters. Br. J. Nutr. 116, 480–486. doi: 10.1017/ Luise, D., Lauridsen, C., Bosi, B., and Trevisi, P. (2019a). Methodology and S0007114516002269 application of Escherichia coli F4 and F18 encoding infection models in Hermes, R. G., Molist, F., Pérez, J. F., Gómez de Segura, A., Ywazaki, M., post-weaning pigs. J. Anim. Sci. Biotechnol. 10:53. doi: 10.1186/s40104- Davin, R., et al. (2013). Casein glycomacropeptide in the diet may reduce 019-0352-7 Escherichia coli attachment to the intestinal mucosa and increase the Luise, D., Motta, V., Bertocchi, M., Salvarani, C., Clavenzani, P., Fanelli, F., intestinal lactobacilli of early weaned piglets after an enterotoxigenic E. coli et al. (2019b). Effect of Mucine 4 and Fucosyltransferase 1 genetic variants K88 challenge. Br. J. Nutr. 109, 1001–1012. doi: 10.1017/S0007114512002978 on gut homoeostasis of growing healthy pigs. J. Anim. Physiol. Anim. Nutr. Hermes, R. G., Molist, F., Ywazaki, M., Nofrarías, M., Gomez de Segura, A., 103, 801–812. doi: 10.1111/jpn.13063 Gasa, J., et al. (2009). Effect of dietary level of protein and fiber on the Luppi, A. (2017). Swine enteric colibacillosis: diagnosis, therapy and antimicrobial productive performance and health status of piglets. J. Anim. Sci. 87, 3569–3577. resistance. Porcine Health Manag. 3:16. doi: 10.1186/s40813-017-0063-4 doi: 10.2527/jas.2008-1241 Massacci, F. R., Tofani, S., Forte, C., Bertocchi, M., Lovito, C., Orsini, S., et al. Hong, K. B., Kim, J. H., Kwon, H. K., Han, S. H., Park, Y., and Suh, H. J. (2020). Host genotype and amoxicillin administration affect the incidence (2016). Evaluation of prebiotic effects of high-purity galactooligosaccharides of diarrhoea and faecal microbiota of weaned piglets during a natural in vitro and in vivo. Food Technol. Biotechnol. 54, 156–163. doi: 10.17113/ multiresistant ETEC infection. J. Anim. Breed. Genet. 137, 60–72. doi: 10.1111/ ftb.54.02.16.4292 jbg.12432 Inturri, R., Stivala, A., Furneri, P. M., and Blandino, G. (2016). Growth and Modesto, M., D'Aimmo, M., Stefanini, I., Trevisi, P., Filippi, S., Casini, L., adhesion to HT-29 cells inhibition of gram-negatives by Bifidobacterium et al. (2009). A novel strategy to select Bifidobacterium strains and prebiotics longum BB536 e Lactobacillus rhamnosus HN001 alone and in combination. as natural growth promoters in newly weaned pigs. Livest. Sci. 122, 248–258. Eur. Rev. Med. Pharmacol. Sci. 20, 4943–4949. doi: 10.1016/j.livsci.2008.08.017 Ishizuka, T., Kanmani, P., Kobayashi, H., Miyazaki, A., Soma, J., Suda, Y., et al. Monteagudo-Mera, A., Arthur, J. C., Jobin, C., Keku, T., Bruno-Barcena, J. M., (2016). Immunobiotic bifidobacteria strains modulate rotavirus immune and Azcarate-Peril, M. (2016). High purity galacto-oligosaccharides response in porcine intestinal epitheliocytes via pattern recognition receptor enhance specific Bifidobacterium species and their metabolic activity signaling. PLoS One 11:e0152416. doi: 10.1371/journal.pone.0152416 in the mouse gut microbiome. Benefic. Microbes 7, 247–264. doi: 10.3920/ Jacobsen, M., Kracht, S. S., Esteso, G., Cirera, S., Edfors, I., Archibald, A. L., BM2015.0114 et al. (2010). Refined candidate region specified by haplotype sharing for Motta, V., Luise, D., Bosi, P., and Trevisi, P. (2019). Faecal microbiota shift Escherichia coli F4ab/F4ac susceptibility alleles in pigs. Anim. Genet. 41, during weaning transition in piglets and evaluation of AO blood types as 21–25. doi: 10.1111/j.1365-2052.2009.01952.x shaping factor for the bacterial community profile. PLoS One 14:e0217001. Jensen, M. T., Cox, R. P., and Jensen, B. B. (1995). Microbial production of doi: 10.1371/journal.pone.0217001 skatole in the hind gut of pigs given different diets and its relation to skatole Muñoz, J. A., Chenoll, E., Casinos, B., Bataller, E., Ramón, D., Genovés, S., deposition in backfat. Anim. Sci. 61, 293–304. doi: 10.1017/S1357729800013837 et al. (2011). Novel probiotic Bifidobacterium longum subsp. infantis CECT Frontiers in Microbiology | www.frontiersin.org 17 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 7210 strain active against rotavirus infections. Appl. Environ. Microbiol. 77, a starter infant formula containing a synbiotic of bovine milk-derived 8775–8783. doi: 10.1128/AEM.05548-11 oligosaccharides and Bifidobacterium animalis subsp. lactis CNCM I-3446. National Research Council (NRC) (2012). Nutrient requirements of swine. Environ. Microbiol. 18, 2185–2195. doi: 10.1111/1462-2920.13144 Washington, DC: National Academy Press. Song, H., Zhang, J., Qu, J., Liu, J., Yin, P., Zhang, G., et al. (2019). Lactobacillus Nguyen, V. U., Goetstouwers, T., Coddens, A., Van Poucke, M., Peelman, L., rhamnosus GG microcapsules inhibit Escherichia coli biofilm formation in Deforce, D., et al. (2012). Differentiation of F4 receptor profiles in pigs coculture. Biotechnol. Lett. 41, 1007–1014. doi: 10.1007/s10529-019-02694-2 based on their mucin 4 polymorphism, responsiveness to oral F4 immunization Stecher, B. (2015). The roles of inflammation, nutrient availability and the and in vitro binding of F4 to villi. Vet. Immunol. Immunopathol. 152, commensal microbiota in enteric pathogen infection. Microbiol. Spectr. 3. 93–100. doi: 10.1016/j.vetimm.2012.09.015 doi: 10.1128/microbiolspec.MBP-0008-2014 Nofrarías, M., Manzanilla, E. G., Pujols, J., Gibert, X., Majó, N., Segalés, J., Sterndale, S. O., Evans, D. J., Mansfield, J. P., Clarke, J., Sahibzada, S., Abraham, S., et al. (2006). Effects of spray-dried porcine plasma and plant extracts on et al. (2019). Effect of mucin 4 allele on susceptibility to experimental intestinal morphology and on leukocyte cell subsets of weaned pigs. J. Anim. infection with enterotoxigenic F4 Escherichia coli in pigs fed experimental Sci. 84, 2735–2742. doi: 10.2527/jas.2005-414 diets. J. Anim. Sci. Biotechnol. 10:56. doi: 10.1186/s40104-019-0366-1 Paganini, D., Uyoga, M. A., Kortman, G. A. M., Cercamondi, C. I., Moretti, D., Tanner, S. A., Chassard, C., Zihler Berner, A., and Lacroix, C. (2014). Synergistic Barth-Jaeggi, T., et al. (2017). Prebiotic galacto-oligosaccharides mitigate the effects of Bifidobacterium thermophilum RBL67 and selected prebiotics on adverse effects of iron fortification on the gut microbiome: a randomised inhibition of Salmonella colonization in the swine proximal colon PolyFermS controlled study in Kenyan infants. Gut 66, 1956–1967. doi: 10.1136/ model. Gut Pathog. 6:44. doi: 10.1186/s13099-014-0044-y gutjnl-2017-314418 Thongaram, T., Hoeflinger, J. L., Chow, J., and Miller, M. J. (2017). Prebiotic Peng, Q. L., Ren, J., Yan, X. M., Huang, X., Tang, H., Wang, Y. Z., et al. galactooligosaccharide metabolism by probiotic lactobacilli and bifidobacteria. (2007). The g.243A>G mutation in intron 17 of MUC4 is significantly J. Agric. Food Chem. 65, 4184–4192. doi: 10.1021/acs.jafc.7b00851 associated with susceptibility/resistance to ETEC F4ab/ac infection in pigs. Tian, S., Wang, J., Yu, H., Wang, J., and Zhu, W. (2019). Changes in Ileal Anim. Genet. 38, 397–400. doi: 10.1111/j.1365-2052.2007.01608.x microbial composition and microbial metabolism by an early-life Galacto- Piñeiro, C., Piñeiro, M., Morales, J., Andrés, M., Lorenzo, E., Pozo, M. D., oligosaccharides intervention in a neonatal porcine model. Nutrients 11:1753. et al. (2009). Pig-MAP and haptoglobin concentration reference values in doi: 10.3390/nu11081753 swine from commercial farms. Vet. J. 179, 78–84. doi: 10.1016/j.tvjl.2007.08.010 Toscano, M., De Grandi, R., Stronati, L., De Vecchi, E., and Drago, L. (2017). Poirel, L., Madec, J. Y., Lupo, A., Schink, A. K., Kieffer, N., Nordmann, P., Effect of Lactobacillus rhamnosus HN001 and Bifidobacterium longum BB536 et al. (2018). Antimicrobial resistance in Escherichia coli. Microbiol. Spectr. on the healthy gut microbiota composition at phyla and species level: A 6. doi: 10.1128/microbiolspec.ARBA-0026-2017 preliminary study. World J. Gastroenterol. 23, 2696–2704. doi: 10.3748/wjg. Rasschaert, K., Verdonck, F., Goddeeris, B. M., Duchateau, L., and Cox, E. v23.i15.2696 (2007). Screening of pigs resistant to F4 enterotoxigenic Escherichia coli Trevisi, P., Corrent, E., Mazzoni, M., Messori, S., Priori, D., Gherpelli, Y., et al. (ETEC) infection. Vet. Microbiol. 123, 249–253. doi: 10.1016/j. (2015). Effect of added dietary threonine on growth performance, health, vetmic.2007.02.017 immunity and gastrointestinal function of weaning pigs with differing genetic Ren, J., Yan, X., Ai, H., Zhang, Z., Huang, X., Ouyang, J., et al. (2012). susceptibility to Escherichia coli infection and challenged with E. coli K88ac. Susceptibility towards enterotoxigenic Escherichia coli F4ac diarrhea is governed J. Anim. Physiol. Anim. Nutr. 99, 511–520. doi: 10.1111/jpn.12216 by the MUC13 gene in pigs. PLoS One 7:e44573. doi: 10.1371/journal. Trevisi, P., De Filippi, S., Minieri, L., Mazzoni, M., Modesto, M., Biavati, B., pone.0044573 et al. (2008). Effect of fructo-oligosaccharides and different doses of Richardson, A. J., Calder, A. G., Stewart, C. S., and Smith, A. (1989). Simultaneous Bifidobacterium animalis in a weaning diet on bacterial translocation and determination of volatile and non-volatile acidic fermentation products of toll-like receptor gene expression in pigs. Nutrition 24, 1023–1029. doi: anaerobes by capillary gas chromatography. Lett. Appl. Microbiol. 9, 5–8. 10.1016/j.nut.2008.04.008 doi: 10.1111/j.1472-65X.1989.tb00278.x Trevisi, P., Melchior, D., Mazzoni, M., Casini, L., De Filippi, S., Minieri, L., Rodríguez-Sorrento, A., Castillejos, L., López-Colom, P., Cifuentes-Orjuela, G., et al. (2009). A tryptophan-enriched diet improves feed intake and growth Rodríguez-Palmero, M., Moreno-Muñoz, J. A., et al. (2020). Effects of performance of susceptible weanling pigs orally challenged with E. coli K88. Bifidobacterium longum Subsp. infantis CECT 7210 and lactobacillus J. Anim. Sci. 87, 148–156. doi: 10.2527/jas.2007-0732 rhamnosus HN001, combined or not With Oligofructose-enriched inulin, Tzortzis, G. (2009). Development and functional properties of Bimuno: a second- on weaned pigs orally challenged With Salmonella Typhimurium. Front. generation prebiotic mixture. Food Sci. Technol. Bull. 6, 81–89. doi: Microbiol. 11:2012. doi: 10.3389/fmicb.2020.02012 10.1616/1476-2137.15818 Romond, M. B., Haddou, Z., Mielcareck, C., and Romond, C. (1997). Bifidobacteria Van den Broeck, W., Cox, E., and Goddeeris, B. M. (1999). Receptor-specific and human health: regulatory effect of indigenous bifidobacteria on Escherichia binding of purified F4 to isolated villi. Vet. Microbiol. 68, 255–263. doi: coli intestinal colonization. Anaerobe 3, 131–136. doi: 10.1006/anae.1997.0089 10.1016/S0378-1135(99)00076-0 Ruiz, L., Flórez, A. B., Sánchez, B., Moreno-Muñoz, J. A., Rodriguez-Palmero, M., Van den Broeck, W., Cox, E., Oudega, B., and Goddeeris, B. (2000). The F4 Jiménez, J., et al. (2020). Bifidobacteriumlongum subsp. infantis CECT7210 fimbrial antigen of Escherichia coli and its receptors. Vet. Microbiol. 71, ® (B. infantis IM-1 ) displays In vitro activity against Some intestinal pathogens. Nutrients 12:3259. doi: 10.3390/nu12113259 223–244. doi: 10.1016/S0378-1135(99)00174-1 Van Nevel, C. J., Decuypere, J. A., Dierick, N. A., and Molly, K. (2005). Sela, D. A., Chapman, J., Adeuya, A., Kim, J. H., Chen, F., Whitehead, T. R., Incorporation of galactomannans in the diet of newly weaned piglets: effect et al. (2008). The genome sequence of Bifidobacterium longum subsp. infantis on bacteriological and some morphological characteristics of the small reveals adaptations for milk utilization within the infant microbiome. Proc. intestine. Arch. Anim. Nutr. 59, 123–138. doi: 10.1080/1745039051233 Natl. Acad. Sci. U. S. A. 105, 18964–18969. doi: 10.1073/pnas.0809584105 1387936 Shen, T. C., Albenberg, L., Bittinger, K., Chehoud, C., Chen, Y. Y., Judge, C. A., Van Soest, P. J., Robertson, J. B., and Lewis, B. A. (1991). Methods for dietary et al. (2015). Engineering the gut microbiota to treat hyperammonemia. J. fiber, neutral detergent fiber, and non-starch polysaccharides in relation to Clin. Invest. 125, 2841–2850. doi: 10.1172/JCI79214 animal nutrition. J. Dairy Sci. 74, 3583–3597. doi: 10.3168/jds.S0022-0302(91) Shoaf, K., Mulvey, G. L., Armstrong, G. D., and Hutkins, R. W. (2006). Prebiotic 78551-2 galactooligosaccharides reduce adherence of enteropathogenic Escherichia Vazquez-Gutierrez, P., de Wouters, T., Werder, J., Chassard, C., and Lacroix, C. coli to tissue culture cells. Infect. Immun. 74, 6920–6928. doi: 10.1128/ (2016). High iron-sequestrating bifidobacteria inhibit enteropathogen growth IAI.01030-06 and adhesion to intestinal epithelial cells in vitro. Front. Microbiol. 7:1480. Shu, Q., and Gill, H. S. (2001). A dietary probiotic (Bifidobacterium lactis doi: 10.3389/fmicb.2016.01480 HN019) reduces the severity of Escherichia coli O157:H7 infection in mice. Verheijden, K. A., Akbari, P., Willemsen, L. E., Kraneveld, A. D., Folkerts, G., Med. Microbiol. Immunol. 189, 147–152. doi: 10.1007/s430-001-8021-9 Garssen, J., et al. (2015). Inflammation-induced expression of the alarmin Simeoni, U., Berger, B., Junick, J., Blaut, M., Pecquet, S., Rezzonico, E., et al. interleukin 33 can be suppressed by galacto-oligosaccharides. Int. Arch. Allergy (2016). Gut microbiota analysis reveals a marked shift to bifidobacteria by Immunol. 167, 127–136. doi: 10.1159/000437327 Frontiers in Microbiology | www.frontiersin.org 18 April 2021 | Volume 12 | Article 642549 Rodríguez-Sorrento et al. Synbiotic Efficacy Against ETEC F4 Vulevic, J., Juric, A., Tzortzis, G., and Gibson, G. R. (2013). A mixture of Conflict of Interest: GC-O, JM-M, and MR-P were employed by the company trans-galactooligosaccharides reduces markers of metabolic syndrome and Laboratorios Ordesa S.L. modulates the fecal microbiota and immune function of overweight adults. J. Nutr. 143, 324–331. doi: 10.3945/jn.112.166132 The remaining authors declare that the research was conducted in the absence Walter, J., Maldonado-Gómez, M. X., and Martínez, I. (2018). To engraft or not of any commercial or financial relationships that could be construed as a potential to engraft: an ecological framework for gut microbiome modulation with live conflict of interest. microbes. Curr. Opin. Biotechnol. 49, 129–139. doi: 10.1016/j.copbio.2017.08.008 Wang, J., Tian, S., Yu, H., Wang, J., and Zhu, W. (2019). The response of colonic Copyright © 2021 Rodríguez-Sorrento, Castillejos, López-Colom, Cifuentes-Orjuela, mucosa-associated microbiota composition, mucosal immune homeostasis, and Rodríguez-Palmero, Moreno-Muñoz, Luise, Trevisi and Martín-Orúe. This is an barrier function to early-life galactooligosaccharides intervention in suckling open-access article distributed under the terms of the Creative Commons Attribution piglets. J. Agric. Food Chem. 67, 578–588. doi: 10.1021/acs.jafc.8b05679 License (CC BY). The use, distribution or reproduction in other forums is permitted, Zhang, B., Ren, J., Yan, X., Huang, X., Ji, H., Peng, Q., et al. (2008). Investigation provided the original author(s) and the copyright owner(s) are credited and that of the porcine MUC13 gene: isolation, expression, polymorphisms and strong the original publication in this journal is cited, in accordance with accepted academic association with susceptibility to enterotoxigenic Escherichia coli F4ab/ac. practice. No use, distribution or reproduction is permitted which does not comply Anim. Genet. 39, 258–266. doi: 10.1111/j.1365-2052.2008.01721.x with these terms. Frontiers in Microbiology | www.frontiersin.org 19 April 2021 | Volume 12 | Article 642549