Protective effect of phytogenic plus short and medium-chain fatty acids-based additives in enterotoxigenic Escherichia coli challenged piglets

Post Weaning Diarrhea (PWD) is the most important multifactorial gastroenteric disease of the weaning in pig livestock. Phytogenic (PHY) natural extracts are largely studied as alternatives to antibiotic treatments in combating the global concern of the antimicrobial resistance. The aim of this study was to evaluate the protective effect of innovative phytogenic premix with or without short and medium chain fatty acids (SCFA and MCFA) in O138 Escherichia coli challenged piglets. Twenty-seven weaned piglets were allotted into four groups fed different diets according to the following dietary treatments: CTRL (n = 13) group fed basal diet, PHY1 (n = 7) fed the basal diet supplemented with 0.2% of phytogenic premix, PHY2 (n = 7) fed the basal diet supplemented with 0.2% of phytogenic premix added with 2000 ppm of SCFA and MCFA. After 6 days of experimental diet feeding, animals were challenged (day 0) with 2 × 109 CFU of E. coli and CTRL group was divided at day 0 into positive (challenged CTRL + ; n = 6) and negative control group (unchallenged CTRL-; n = 7). Body weights were recorded at -14, -6, 0, 4 and 7 days and the feed intake was recorded daily. E. coli shedding was monitored for 4 days post-challenge by plate counting. Fecal consistency was registered daily by a four-point scale (0–3; diarrhea > 1) during the post-challenge period. Tissue samples were obtained for gene expression and histological evaluations at day 7 from four animals per group. Lower average feed intake was observed in CTRL + compared to PHY2 and CTRL during the post-challenge period. Infected groups showed higher E. coli shedding compared to CTRL- during the 4 days post-challenge (p < 0.01). PHY2 showed lower frequency of diarrhea compared to PHY1 and CTRL + from 5 to 7 days post-challenge. No significant alterations among groups were observed in histopathological evaluation. Duodenum expression of occludin tended to be lower in challenged groups compared to CTRL- at 7 days post-challenge (p = 0.066). In conclusion, dietary supplementation of PHY plus SCFA and MCFA revealed encouraging results for diarrhea prevention and growth performance in weaned piglets. Supplementary Information The online version contains supplementary material available at 10.1007/s11259-022-09945-0.


Introduction
Post weaning diarrhea (PWD) is a gastrointestinal multifactorial disease that generally occurs during the first two weeks after the weaning phase. It is one of the most economically-relevant diseases in swine husbandry due to the costs of treatments, reduced growth, and increased of mortality (Bonetti et al. 2021). Although many factors are involved in the development of this disease, PWD is often exacerbated by many enterotoxigenic Escherichia coli pathotypes characterized by the presence of virulence factors such as toxins and adhesive fimbriae (Sun and Kim 2017). Bacterial resistance to a wide range of commonly used antibiotics is a global concern and a recent increase in prevalence and severity of PWD required alternative measures for their control (Renzhammer et al. 2020;Dell'Anno et al. 2021a, b). Reducing and replacing antimicrobials in animal farming is a crucial aim of the European policies, even if the mechanisms of cross-species transmission of resistant bacteria and their genetic elements spread from livestock to humans has not been fully understood (Rossi et al. 2014a;Cormican et al. 2017;Tang et al. 2017).
The aim of nutrition is no longer simply to satisfy the nutritional requirements, but also play a key role in the health and welfare of humans and animals (Domínguez Díaz et al. 2020;Grossi et al. 2021). Functional feed additives, which sustain the health status and reduce the risk of pathologies, have thus become fundamental in replacing or reducing antimicrobials in food-producing animals. The dietary inclusion of phytogenics (PHYs), represented by plant secondary metabolites, are largely studied as alternative growth promoters because of their biological properties which include antimicrobial, antioxidant, and nutrigenomic effects on the development of animal (Durmic and Blache 2012;Yang et al. 2015;Lillehoj et al. 2018;Reyes-Camacho et al. 2020). In particular Yan and Kim (2012) observed a significant reduction in fecal E. coli count after 1 g/kg of eugenol supplementation in pigs. A blend of oregano, anise, and citrus peel (40 mg/kg diet) supplementation to piglets' diet has been demonstrated to evolve anti-inflammatory effect by reducing the gene expression of NF-kB and TNFα (Upadhaya et al. 2016). The dietary supplementation of thymol, cinnamaldehyde and menthol have been reported to positively affect the feed digestibility in swine (Maenner et al. 2011;Li et al. 2012).The in vivo effects, resulting from the various biological activities of the PHYs, depend on their structure, dosage, and pharmaco-kinetics, as well as the animal species, productive phase and administration period. For this reason, several combinations of natural extracts are currently studied in order to promote their possible synergistic or complementary effect on animal health. Although PHYs show antimicrobial activity in the gastrointestinal tract against specific pathogens such as Escherichia coli, Clostridium perfringens and Salmonella spp. (Thacker 2013;Mohammadi Gheisar and Kim 2018), their effectiveness can vary due to the presence and the location of functional hydroxyl and phenolic terpenoids (Dubreuil 2013). Rational combinations of PHYs have been studied in order to increase the spectrum of beneficial activities. In addition, the synergistic or complementary effect of PHYs with other compounds leads to various beneficial activities of several compounds, especially organic acid (OA). Amongst feed additives with antimicrobial activities, organic acids, in particular short-chain fatty acids (SCFAs) and medium-chain fatty acids (MCFAs), have a strong antimicrobial activity and are key to modulating intestinal health and improving animal performance (Ferronato and Prandini 2020;Jackman et al. 2020). SCFAs and MCFAs regulate the growth and virulence of enteric pathogens, such as enterohemorrhagic E. coli, Klebsiella and Salmonella . They damage the bacterial structure and in some cases separate the inner and outer membranes (Hanczakowska 2017) and thus increase the concentration of IgG and IgM in piglets challenged with enterotoxigenic Escherichia coli (ETEC) strains (Han et al. 2020). A synergistic antimicrobial effect has been observed in the combination of PHYs and organic acids in vitro (Costa et al. 2013). However, the effect of their dietary supplementation on pigs' growth and the optimization of the inclusion level for diarrhea prevention against major pathogens of weaned piglets has not been fully investigated. Therefore, it was hypothesized that the dietary supplementation of phytogenic additive with or without organic acids could prevent or limit the detrimental effects of enterotoxigenic Escherichia coli infection improving animal health status.
The aim of this study was thus to evaluate the protective effect against O138 E. coli F18 + infection of an innovative phytogenic premix composed by caraway oil, lemon oil, clove, cinnamon, nutmeg, onion, pimento, orange peel, peppermint and chamomile powder with and without short and medium chain fatty acids in weaned piglets' diet.

Animal selection criteria
The trial was performed at the Experimental Animal Research and Application Centre of University of Milan and was authorized by the Italian Health Ministry (authorization n° 711/-PR) in accordance with EU regulations (Directive 2010/63/EU 2010).
Animals enrolled in the experimental trial were selected from a conventional herd free from contagious diseases (Ex A-list International Office of Epizootic, porcine reproductive and respiratory syndrome, atrophic rhinitis, Aujeszky's disease, transmissible gastroenteritis, salmonellosis) and without a history of PWD or oedema disease. Sows were assessed for genetic susceptibility to Escherichia coli carrying F18 adhesive fimbriae (F18 E. coli) by screening the fucosyltransferase 1 (FUT1) genotypes using polymerase chain reaction (PCR) reaction according to Luise et al. (2019a, b). Briefly, genomic DNA was extracted from hair samples of sows and genotyped to identify polymorphic variants. Sows carrying the GG genotypes at FUT1 gene were considered for piglet enrolment. A further selection criterion was the absence of hemolytic E. coli in piglets feces. Microbiological analyses of selective mediums (Agar MacConkey) (Hayer et al. 2020;Li et al. 2020;Remfry et al. 2020) were thus carried out before transport and upon arrival on fecal samples collected from enrolled piglets.

Animals and experimental design
Twenty-seven weaned piglets (28 ± 2 days) balanced per weight (9.79 ± 1.25 kg) and sex, were randomly allotted in four experimental groups in randomized complete block design and, after 7 days of adaptation period, fed ad libitum for the entire experimental period according to the following dietary treatments: control group (CTRL, n = 13) fed basal diet, phytogenic additive group 1 (PHY1, n = 7) fed basal diet supplemented with 200 g/100 kg phytogenic additive, phytogenic additive group 2 (PHY2, n = 7) fed basal diet supplemented with phytogenic additive plus 2000 ppm of short and medium chain fatty acids premix.
In order to achieve the same nutrient concentrations, the control group received basal diet supplemented with the same premix carrier used for treatment groups (95% wheat meal and 5% of coconut oil) without phytogenic compounds. The iso-energetic and iso-proteic diets (Table S1) were formulated (Plurimix; Fabermatica, CR, Italy) according to animal requirements for the post weaning phase defined by the US National Research Council (NRC 2012). The phytogenic feed additive (FRESTA®F, Delacon Biotechnik GmBH), approved by EU regulation (Reg. CE 1831/2003), as zootechnical additive, was composed of essential oil from caraway oil (d-carvone 3.5-6.0 mg/g) and lemon (limonene: 2.3-9.0 mg/g), dried herbs and spices (1.5% clove powder, 10% cinnamon powder, 1.5% nutmeg powder, 5% onion powder, 2% pimento powder, 5% orange peel powder, 12.5% peppermint powder and 12.5% chamomile powder). The SCFA and MCFA premix was composed by butyric (C4), caprylic (C8), capric (C10) and lauric acid (C12). The phytogenic products (with or without SCFA and MCFA) or the premix carrier were mixed with the compound diets for 30 min in order to ensure a homogeneous distribution. Diets have been provided in meal without any technological treatments, except for mixing procedure. During the mixing process the temperature was monitored in order to do not overcome 30 °C.
Piglets were housed in two environmentally controlled rooms, in individual pens, with a plastic slatted floor and constant temperature (27° C) and humidity (60%) for the entire experimental period. The trial was divided into a pre-and post-challenge, considering the challenge as day 0 (Fig. 1).

Chemical analysis of experimental diets
Diets were analyzed for proximate analysis, including moisture, crude protein (CP), crude fibre (CF), ether extract (EE), and ash.

Experimental challenge
E. coli challenger strain was genetically characterized by polymerase chain reaction (PCR) (Applied Biosystem 7500) in order to detect the presence of the two important virulence profile: subunit B of verocytotoxin type 2 and F18 adhesive fimbriae (Table 1).
Twenty piglets, except for piglets in CTRL-group (n = 7), on day 0 (challenge day) were orally infected with O138 Escherichia coli F18 + strain obtained from a permanent collection of the University of Milan and previously characterized (Rossi et al. 2014b. Sixty minutes before the challenge, the piglets were sedated with azaperone (2 mL/head, Stresnil®, Janssen Cilag Spa, Milan, Italy), thereafter 30 mL of a 10% bicarbonate solution was orally administered to neutralize gastric acid and to increase the survival rate of the challenger strain in the stomach. After 10-15 min, the inoculum was given orally in a single dose of 5 mL of bacterial medium with 2 × 10 9 colony-forming unit (CFU) of challenger strain, using a 16G catheter . Animals were fasted 3 h before and 3 h after the challenge. At the same time, piglets in CTRL-were orally inoculated with 5 mL of Luria Bertani (LB) medium to balance the level of stress associated with the oral challenge.

Zootechnical performance, clinical and fecal score
Average daily feed intake (ADFI) was recorded daily from day -6 to day 7 by measuring the refusals. Body weight (BW) was recorded on day -6 (first day of experimental diets), day 0 (challenge day), day 4 and day 7 (sacrifice day). Average daily gain (ADG) and feed efficiency were also calculated.
Piglets were individually evaluated throughout the trial by clinical examination, including observation of behavioral disturbances. In particular, oedema, epiphora, respiratory and hair scores were evaluated through three-point scales (oedema score: 0 = normal, 1 = mild, 2 = severe; epiphora score: 0 = normal, 1 = mild, 2 = severe; respiratory score: 0 = normal, 1 = slightly quick, 2 = quick; hair/bristles score: 0 = smooth, 1 = lightly brushy, 2 = highly brushy) ). In addition, cyanosis, a blue or red discoloration of the skin, which may or may not be localized to small areas, was considered not as a specific skin condition but as a symptom of disease. From day -6 to day 7, all piglets were evaluated for the fecal score. Clinical signs of the disease were identified according to the point scale score described by Rossi et al. (Rossi et al. 2014b). A four-point scale was adopted to score fecal consistency: 0 = normal, 1 = soft consistency, 2 = mild diarrhea, 3 = severe diarrhea; considering > 1 as an indicative of diarrhea. Fecal color was evaluated using a three-point scale: 1 = yellow, 2 = green; 3 = brown.

Microbiological evaluation of fecal samples
Individual fecal samples were collected from rectal ampulla from each piglet, on days -1, 1, 2, 3 and 4 to perform microbiological analysis and evaluate the challenger strain shedding. For each sample, 1 g of feces was homogenized with 1 ml of saline solution and incubated overnight at 37 °C on sheep blood agar plates 5% (Blood Agar Base No. 2-Oxoid) in order to examine the presence of hemolytic colonies. The total hemolytic bacteria count was performed by counting the number of colonies cultured from serial dilutions of each fecal sample in order to evaluate the presence of hemolytic E. coli in relation to the total bacteria population.

Necropsy, intestinal samples, and histopathology
At day 7 post-challenge, sixteen animals (n = 4/treatment), were randomly selected and euthanized and tissue samples were collected for histopathological and molecular analyses of intestinal tissues.
Animal care and euthanasia procedures were conducted in accordance with the European Union guidelines (86/609/EEC) and approved by the Italian Ministry of Health. Briefly, selected piglets were sedated with 2 mL/head of azaperone (Stresnil®, Janssen Cilag SpA, Milan, Italy) intramuscularly. After 20 min, animals received a bolus injection of propofol intravenously in the right and left lateral auricular vein. Anesthesia was maintained with 40 mg/kg of tiletamine/zolazepam intramuscularly (Zoletil 100, Virbac UK, Bury Saint Edmund, England). Finally, unconscious animals were euthanized by the intracardiac administration of a solution with embutramide, mebezonium iodide and tetracaine hydrochloride (0.3 mg/kg, Tanax, MSD Animal Health, Boxmeer, Netherlands). The intestine of each animal was weighed, and intestinal samples of ileum at 1 cm from ileocecal valve, mesenteric lymph nodes were harvested. For the histological evaluation, samples were diluted in 10% neutral formalin buffer and stored at 4 °C. Tissues were rinsed with sterile saline solution and transferred into 2 mL cryotubes, snap-frozen in liquid nitrogen and stored at -80 °C until further analysis.
Histological examinations of collected intestinal and lymph nodes samples for each piglet were carried out. The fixed samples were embedded in paraffin, and 5 µm thick histological sections were performed with a microtome. Cross sections were stained with hematoxylin and eosin and were blind evaluated by light microscopy. A four-point scale was adopted for inf lammatory infiltrates, epithelial regeneration, fusion of villi, oedema, hyperemia, necrosis of mucosa, T atrophy, stroma, and follicular hyperplasia; considering: 0 = no evidence; 1 = slight presence; 2 = moderate; 3 = severe. Samples of duodenum were collected and frozen in liquid nitrogen for gene expression analysis.

Duodenum gene expression
Total RNA was extracted from the duodenum using Fast-Gene Scriptase Basic (Nippon genetics) according to the manufacturer's instructions. The integrity of total RNA was assessed by gel electrophoresis to detect the 18S and 28S rRNA bands. A combination of oligo-dT and random primers was used to reverse transcribe 100 ng of total duodenal RNA to cDNA (cDNA synthesis kit, FastGene Scriptase Basic, Nippon Genetics). Primer pairs were first tested for their specificity in qualitative PCR, using the pooled cDNA as a template. The cycling profile for the assay consisted of initial denaturation of RNA (65 °C × 5'), then the annealing of random primers (25 °C × 10'), followed by the annealing of oligo-dT and transcription (42 °C × 60'). At the end of the cycle, the enzyme deactivation (90 °C × 5') was performed.
The mean values for the transcripts were normalized to the arithmetic mean of mRNA abundance of βactin as the reference gene within each sample. The comparative CT method was used to determine fold changes in gene expression, calculated as 2 −∆∆CT . The final results were presented as the fold changes of target gene expression in a target sample relative to a reference sample, normalized to βactin rRNA (Livak and Schmittgen, 2001). The βactin rRNA was used to calculate the threshold cycles, since it previously showed constant values under all the conditions adopted.

Statistical analysis
Zootechnical performance and fecal microbiological analysis were analyzed using a linear model after testing the normality of data through Shapiro-Wilk test using JMP Pro 15® (SAS Inst. Inc., Cary, NC, USA). The model included the fixed effect of treatments (Trt), the effect of time (Time), and the interaction between treatment and time (Trt x Time). Serum metabolites were evaluated performing analysis of covariance (ANCOVA) to adjust the initial variability of the pre-challenge period after testing the normality of data through Shapiro-Wilk test using JMP Pro 15® (SAS Inst. Inc., Cary, NC, USA).
Clinical score data were converted into a dichotomous variable (normal/pathological), and observed frequencies were assessed using the Chi-squared Test. Histological scores, intestinal weight and relative gene expression were analyzed using Kruskal-Wallis test (PROC NPAR1WAY of SAS 9.4 software) for non-parametric data due to the small sample size of euthanized animals at day 7. Multiple comparisons for parametric statistics were evaluated with the Tukey's Honestly Significant Difference test (Tukey's HSD) or Tukey-Kramer test and Steel-Dwass test was used for non-parametric multiple comparisons. The results were presented as least square means (LSMEANS) ± standard error (SE) for parametric data and as medians and range (minimum-maximum) for non-parametric results. Means or medians were considered statistically different when p ≤ 0.050 and statistical tendency was considered when p < 0.100.

Chemical composition of the experimental diets
Proximate analysis of the experimental diets showed comparable contents of the principal nutrients. The inclusion of phytogenic based additives with or without MCFA and SCFA did not affect the nutrient balance of feed (Table S1).

Zootechnical performance
During the pre-challenge period, no statistically significant differences among experimental groups were observed. Considering the entire post-challenge period, ADFI of CTRL + was lower than PHY2 and CTRL-(p < 0.005; Table 3).

Influence of phytogenic treatments on clinical score, fecal consistency and color
During the pre-challenge period and at day 0, the piglets did not show significant differences among clinical scores, indicating a general good health status. Although statistical differences among treatments were not identified, several altered scores were registered from 1 to 4 days post-challenge. After experimental infection, considering the numerical differences of clinical score frequencies (considered as altered clinical conditions for a score of ≥ 1) revealed that the experimental procedures influenced the clinical status of piglets (Table 4). However, from 5 to 7 days post challenge, a non-normal hair score frequency tended to increase in CTRL + compared to the other experimental groups (9.52% for PHY1, 14.81% for PHY2, 33.33% for CTRL + and 14.29% for CTRL-; p = 0.071).
Fecal score and color were the most informative indicators during the post-challenge period (Table 5). Significant higher frequencies of altered fecal color were recorded in challenged groups compared to CTRLfrom 1 to 4 days post-challenge (p < 0.050). Significant differences in the manifestations of diarrhea (fecal consistency ≥ 2) were observed from 5 to 7 days after the challenge. In particular, PHY1 had higher number of diarrhea cases compared to PHY2, CTRL + and CTRL-, and PHY2 had a lower incidence compared to CTRL + and PHY1 (p < 0.010).

Microbiological evaluation of feces and challenger strain shedding
Weaned piglets did not show the presence of challenger E. coli in feces during the adaptation period and on day 0. Total bacterial count did not show statistically significant differences among groups from 1 to 4 days after the challenge (8.37 ± 0.47 log 10 CFU/g for PHY1, 8.04 ± 0.47 log 10 CFU/g for PHY2, 7.73 ± 0.51 log 10 CFU/g for CTRL + and 7.71 ± 0.47 log 10 CFU/g for CTRL-). Also after the challenge, all the experimental groups (except for negative control, CTRL-) registered fecal shedding of challenger E. coli strain (Fig. 2). Statistically significant increased fecal shedding of hemolytic E. coli was observed in challenged groups compared to CTRL-from day 1 to day 4 post-challenge (4.09 ± 0.01 log 10 CFU/g for PHY1, 5.25 ± 1.10 log 10 CFU/g for PHY2, 5.95 ± 1.09 log 10 CFU/g for CTRL + and 0.00 ± 1.01 log 10 CFU/g for CTRL-; p < 0.001).

Histological evaluation and gene expression
Samples were examined for the presence of inflammation both in villi and in lamina propria, epithelial regeneration, fusion of villi, oedema in deep lamina propria, T atrophy, stroma (fibroconnective and histiocytes), and follicular hyperplasia.
Intestinal weight results did not reveal significant differences between treatment groups after 7 days post-challenge (Table S2). Phytogenic dietary treatments did not significantly affect ileum inflammatory infiltrates, epithelial regeneration, oedema and hyperemia after 7 days (Table 6). Relative expressions of IL-10, IL-6, LOX, GPX2, NQ01 and CLDND1 were not affected by phytogenic dietary treatments (Table 7). The relative expression of occludin was downregulated at day 7 post-challenge (p < 0.012). Pairwise comparisons revealed only a tendency to increase in challenged groups compared to CTRL-(p = 0.066).

Influence of phytogenic treatments on hematological and serum metabolites
The serum metabolic profile did not show statistically significant differences between the experimental groups at day 3 after the challenge (Table S3). After 7 days postchallenge, a significantly higher level of total protein content was observed in CTRL + compared to CTRL-(p = 0.050) ( Table 8). Globulin content tended to be higher in CTRL + than CTRL-at 7 days post-challenge (p = 0.055). PHY2 had a higher level of AST-GOT at day 7 compared to the other challenged groups (p < 0.050). Acute phase proteins were not affected by dietary treatments and experimental challenge and showed no statistically significant differences after 3-and 7-days post-challenge (Table S4).

Discussion
Weaning is a critical period where piglets need to adapt to a new diet, environment and to develop their own immunity (Tretola et al. 2019). During this phase, PWD is one of the major causes of gastrointestinal disorders leading to high morbidity, antibiotic use and economic losses. Several natural extracts have been investigated for their functional proprieties to decrease diarrhea occurrence in piglets, with discordant results. The general aim of this study was to evaluate the protective effect of innovative phytogenic premix with or without MCFA and SCFA against O138 E. coli in weaned piglets. Genetic characterization of the sows led to the enrollment of piglets that were potentially susceptible to F18 fimbriae. In fact, the presence of F18 receptor (F18R) on porcine intestinal epithelium is crucial for the development of E. coli infections. During the pre-challenge period (day -6 to day 0), the piglets showed comparable growth performance, demonstrating that the supplementation of additives did not influence their growth and feed consumption or feed palatability. Even if the effect on zootechnical performance was limited by the short experimental period (EFSA 2018), ADFI was affected by the treatment.
However, ADFI of CTRL + was significantly lower compared to PHY2 and CTRL-groups. The observed decrease in feed intake of CTRL + suggests that the challenged group without any supplementation reduced the feed consumption probably due to the detrimental effect of experimental infection. In addition, higher dietary intake is often related to a better health status (Czech et al. 2021), indicating that the treatment with PHY and organic acids could have supported animals' health resulting in increased feed intake during the entire post-challenge period. PHY2 group showed a similar performance to CTRL-(uninfected), suggesting that dietary supplementation with the phytogenic premix, MCFA and SCFA was very effective in dealing with O138 E. coli infection, thus supporting intestinal health of animals. The addition of MCFA and SCFA may enhance animal growth by several mechanisms as previous studies described (e.g. inhibitory activity, mucosal epithelium integrity support) (Royce et al. 2013;Ferrara et al. 2017;Diao et al. 2019). In addition, phytogenic feed additives derived from spices and herbs are commonly used in animal nutrition as an alternative to in-feed antibiotics due to their antibacterial, antiviral and antioxidant properties. These effects are generally due to the presence of different bioactive compounds such as  alkaloids, flavonoids, glycosides, mucilage, saponins, tannins, phenolics, polyphenols, terpenoids, and polypeptides (Upadhaya et al. 2016;Nowak et al. 2017;Caprarulo et al. 2020a;Reggi et al. 2020). Our results are in line with other studies demonstrating the antibacterial activity of PHYs, MCFA and SCFA on a wide range of pathogens (Dibner and Buttin 2002;Salsali et al. 2008).
In terms of clinical examination, from day 1, clinical scores were affected by experimental infection, confirming that disease development impaired the clinical status of challenged animals compared to the pre-challenge period. Moreover, significant differences in pathological hair, respiratory, oedema and epiphora scores were not detected in infected groups. This was probably due to the individual variability and the small sample size that could prevented to observe differences among groups. The O138 Escherichia coli challenger strain can impair gut health due to its capacity to adhere to the intestinal epithelium by specific fimbriae which could be followed by verocytotoxin production (Rossi et al. 2012(Rossi et al. , 2013 and in consequence may show systemic symptoms. A slightly different situation was found during the evaluation of the fecal score and incidence of diarrhea. Experimental challenge affected transitory the fecal color and consistency during the 7 days post-challenge. Firstly, from day 1 to day 4 post-challenge, feces of yellowish color were registered more frequently in challenged group compared to CTRL-typically related to gastrointestinal disorders (Rossi et al. 2012). Considering total diarrhea cases recorded among experimental groups, from day 5 to day 7 the highest diarrhea frequency was registered, suggesting a late effect of challenge on fecal consistency compared to fecal color. These data are confirmed by a previous study by Rossi et al. (Rossi et al. 2021) showing that O138 E. coli experimental infection increased the sum of fecal score from 3 to 9 days post-challenge. Particularly, the highest diarrhea occurrence was observed in PHY1 compared to other groups, while PHY2 showed a fecal consistency comparable to CTRLsuggesting the counteracting activity of the phytogenic additives, SCFA and MCFA against experimental infection. Even if antibacterial activity of phytogenic additives was reported (Namkung et al. 2004), the observed effect on diarrhea incidence was probably related to their combined effect with SCFA and MCFA. It has been demonstrated, that SCFA and MCFA can exert an inhibitory activity (Lei et al. 2017;Swanson et al. 2018;Zhang et al. 2020) or enhance the functional properties of phytogenic additives (McKnight et al. 2019).
In addition, dietary supplementation of organic acids can modulate the intestinal environment, creating undesirable environmental conditions for pathogenic bacteria, thus also influencing the intestinal microbiota (Verstegen and Williams 2002). Even if is difficult to establish the exact mechanisms for the enhancing antimicrobial effect by the combination of PHYs with organic acids (SCFA and MCFA) in pigs, we can suppose that PHYs can act as a permeabilizing complex and modify pores of the bacterial wall, thus facilitating the entrance of organic acids with antimicrobial activity (Tugnoli et al. 2020). In addition, the reduction in undigested feed protein by organic acids reduces the negative fermentative processes, increases growth performance and repairment of damaged intestinal tissues (Jia et al. 2020). Our results suggest that the addition of MCFA and SCFA to the phytogenic premix significantly inhibited enterotoxigenic E. coli diarrhea, thus supporting intestinal health of animals. Considering the challenger strain shedding, the proliferation started gradually from the day of challenge in line with clinical observations. Compared with the uninfected control group infected animals showed hemolytic E. coli shedding from day 1 post-challenge, thus confirming the success of the experimental infection.
Histopathological examination of the ileum, jejunum and large intestine is thus used to highlight clinical signs of E. coli infection (Luppi 2017). In our study, histological evaluation of the ileum and lymphoid of intestinal tissues did not reveal significant lesions. The animals in the experimental trial thus did not show severe signs of intestinal lesions. Although more frequent lesions were registered in the PHY2 group, these did not impair animal performance and there was a comparable growth curve to CTRL-. This was probably due to the supplementation of phytogenic with SCFA and MCFA which could have supported intestinal health.
Gene expressions showed a high individual variability in terms of inflammatory parameters and tight junctions (TJs), probably due to the limited number of animals. We thus analyzed the expression of the TJ transmembrane protein (occludin) and the observed data were in line with morphological analyses. Our findings suggested that tight junction integrity tended to be disrupted seven days after infection in challenged groups compared to the CTRL-. Intestinal permeability is regulated by the tight junctions which are a primary determinant of epithelial paracellular permeability (Zhang et al. 2021). Disruption of occludin regulation is related to many diseases. During the inflammation process, specific domains of occludin are in fact thought to mediate the transepithelial migration of neutrophils across the TJ (Feldman et al. 2005). Inflammation produces effects on epithelial barriers, increasing the leakiness of occludin, and decreasing the barrier function of this protein. Occludin responds earlier to oxidative stress than claudin, which responds later to reactive oxygen species (ROS) (Blasig et al. 2011). Intestinal bacterial infection is associated with intestinal epithelial and crypt architectural irregularity and with barrier dysfunction, leading to an increase in intestinal mucosal permeability. The observed slight downregulation of occludin after seven days in challenged groups could be due to the harmful activity of the challenger strain. Further investigations are required to better understand the effect of PHYs, SCFA and MCFA on the modulation of genes involved in inflammation and intestinal integrity.
Considering the biochemical parameters of the experimental groups (PHY1, PHY2, CTRL + and CTRL-), the values were within the reference range of weaned pigs (Klem et al. 2010;IZSLER 2017), thus confirming that phytogenic additive supplementation had no detrimental effect on serum metabolism. The metabolite profile showed an increased level of total protein and a higher globulin content in CTRL + compared to CTRL-. However, globulin together with albumin are the two major constituents of serum proteins, which play a crucial role in the inflammatory process (Balan et al. 2020;Wang et al. 2020). The increase in globulin could be associated with an inflammatory process probably due to the experimental E. coli infection leading to an increased concentration of total serum proteins. The serum AST-GOT level is a specific marker for liver tissue and represents a valuable indicator for acute hepatocyte injury or cell membranes damage (Kim 2020;Amirabagya et al. 2021). Although our results are in line with the proper range of pig physiology parameters (Klem et al. 2010;IZSLER 2017;Caprarulo et al. 2020b), AST-GOT was probably higher in the PHY2 group due to the presence of SCFA and MCFA which are immediately available for hepatic metabolism. In fact, short-chain fatty acids can activate lipid and glucose metabolism independently of the pig gut microbiota (Zhou et al. 2021).

Conclusions
Our study showed that phytogenic additive dietary supplementation limited the detrimental effect of experimental challenge. Phytogenic premix plus SCFA and MCFA revealed a positive effect on animal performance and health improving ADFI and fecal consistency during the post-challenge period compared to infected control group, suggesting that the combination of PHYs and organic acids can be considered as effective against pathogenic E. coli strains of weaned piglets. Due to the lack of studies on the argument, at this stage is too early to state that phytogenics are effective. Future studies will be necessary to confirm our results and extensively investigate how phytogenic additives and organic acids affect gene expression over time.
Data availability All data generated or analyzed in this study are available from the corresponding author upon reasonable request. Data are stored on Google Drive at the following link: https:// drive. google. com/ drive/ folde rs/ 1fpkk uaELR 6a5ZI eTh3g szVoM rQJcG cMF? usp= shari ng.

Declarations
Ethics approval This study was performed in line with the principles of the Declaration of Helsinki.
The procedures and protocols used in this study were designed in accordance with the guidelines for animal welfare and the use of animals regulated under Directive 2010/63/EU on the protection of animals used for scientific purposes. The protocol was approved by the Animal Welfare Organization of the University of Milan and by the Italian Ministry for project (authorization number: 711/2017-PR, 28/09/2017).

Consent to participate
All authors agreed to participate in this work. They also approved the content of the research and the submission of the manuscript to Veterinary Research Communication journal.

Consent to publish
All authors agree to the content of the paper for publication.

Conflicts of interest The authors declare no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.