Olive Extract Ameliorates Oxidative Stress and Inammation, and Protects Intestinal Villus and Microbiota in Piglets Induced by Lipopolysaccharides

Background: The olive extract contains compounds with antioxidant and anti-inammatory properties. This study was designed to investigate whether olive cake extract, enriched with maslinic acid and hydroxytyrosol, alleviates the lipopolysaccharides (LPS)-induced oxidative stress, inammation and intestinal villus damage in piglets. Methods: Thirty weaned piglets (6.9±0.9 kg) were assigned to ve groups using a randomized complete block design. Piglets were fed a basal diet before intraperitoneal (i.p.) injection of physiological saline (C); fed a basal diet alone (CL) or fed a basal diet plus olive extract (OL), antibiotics (AL), or olive extract and antibiotics (OAL) before i.p. injection of LPS. The feeding lasted for 2 weeks. Piglets were euthanized 4h after LPS injection. Systemic anti-oxidant and inammation levels were measured and villus morphology in the intestine was examined. Results: Compared with those in the C group, piglets in the CL group had signicantly lower GSH-Px, SOD, ALB levels and higher MDA, NO, LDH, ALT and AST levels in the serum (P<0.05). Compared with the CL group, piglets in OL, AL, and OAL groups had signicantly higher serum GSH-Px, SOD and ALB levels and lower MDA, NO, LDH, ALT and AST levels (P<0.05). LPS administration signicantly increased the serum concentration of TNF-α, IL-6, DAO and D-xylose in the CL group compared with the control group (P<0.05). Piglets in OL, AL, and OAL groups had signicantly lower serum TNF-α, IL-6, DAO and D-xylose levels and higher IL-10 level (P<0.05). In the duodenum and ileum of piglets, LPS challenge led to signicantly lower villus height (VH), higher crypt depth (CD) and lower VH/CD compared with the control group (P<0.05),


Background
Modern livestock production has become highly intensive and large scaled to increase production e ciency. However, this production environment could add stressors affecting the health and growth of animals (1). Oxidative stress occurs when there is an imbalance between free radical production and antioxidant capacity (2). Weaned piglets are subjected to lots of stress, including changes in nutrition, separation from their mothers and littermates, and a new environment, which cause reduced growth and compromised small intestine function (3). In piglets, weaning disrupts the oxidative balance, cause oxidative stress and leads to free radical-mediated oxidative injury (4). The maintenance of redox homeostasis, which is very important for normal cellular processes and organic function, highly depends on the balance between pro-oxidative and anti-oxidative systems (5). The intestinal tract is the primary location of the digestion and absorption of nutrients in the animal body and it acts as the largest immune organ (6). The intestinal tract is the rst line of defense against bacterial and harmful endogenous and exogenous substances (7). Oxidative stress induces intestinal cell cycle arrest and apoptosis (8) and decreases the expression of intestinal tight junction proteins and mucosal barrier function (9). Together this results in the deterioration of gastrointestinal digestive and absorptive functions in piglets (5).
The Olive tree (Olea europaea L.), a native of the Mediterranean basin and parts of Asia, is widely cultivated in many other parts of the world for the production of olive oil and table olives. Olive fruit contains an appreciable concentration, 1-3% of fresh pulp weight, of hydrophilic and lipophilic (cresols) phenolic compounds that possess antioxidant, anticarcinogenic, anti-in ammatory, antimicrobial, and antihypertensive biological activities (10). The phenolic compounds in olives have been described as oleuropein, hydroxytyrosol and verbascoside (11). Previous studies have shown that olives contain substances similar to ibuprofen, anon-steroidal anti-in ammatory drug with anti-in ammatory and antioxidant functions (12). The Mediterranean diet has attracted worldwide attention, mainly because it can reduce the incidence of cardiovascular disease and cancer on the premise of meeting human dietary needs (13). Olive leaves have been used as folk medicines to remedy diseases including fever and malaria (14). Olive leaf extract, containing polyphenols such as oleuropein and hydroxytyrosol, reverses the chronic in ammation and oxidative stress observed in the rat model of diet-induced obesity (15).
Although olive cake is an economical biomass present in large quantities, it causes some environmental problems for Mediterranean countries (16). Olive cake is considered a rich source of phenolic compounds with a wide array of biological activities. Antioxidant attributes have been investigated in olive cakes for anti-radical activities (17). The recovery of phenolic compounds from olive cake has a potential use in the pharmaceutical and nutraceutical industries.
In this study, induced in ammation in piglets by challenging them with LPS (18), and investigated the protective effect of olive extract supplementation on lipopolysaccharide (LPS) induced-damage to the intestine of piglets. We hypothesize that dietary inclusion of olive extract maslinic acid and hydroxytyrosol would enhance antioxidant and anti-in ammatory capacity in weaned piglets and improve the intestinal health and performance of weaning piglets. Therefore, this study was designed to examine the effects of olive extract addition on intestinal function and growth performance in piglets.

Animal and diets
Thirty piglets (Duroc × Landrace × Large White, including males and female, 6.9 ± 0.9 kg) aged 28 ± 1 d were assigned to ve groups (n = 6/group). All animal procedures were performed in full accordance with the Regulation for the Use of Experimental Animals in Zhejiang Province, China. This work was speci cally approved by the Animal Care and Use Committee of Zhejiang University (ethics code permit no. ZJU20170529). Piglets were fed a basal diet before intraperitoneal (i.p.) injection of physiological saline (C). Piglets in the other four groups were fed the basal diet alone (CL) or fed a basal diet plus olive extract (OL), antibiotics (AL), or olive cake extract and antibiotics (OAL) before i.p. injection of LPS (20 µg/kg body weight, Sigma-Aldrich, Saint Louis, MO, USA). The 0.1% olive cake extract in diet mainly included maslinic acid and hydroxytyrosol. Antibiotics included 120 mg/kg oxytetracycline-calcium and 16 mg/kg enduracidin in diet. The basal feed composition is shown in Table 1. Animals were allowed free access to water and feed. The feeding lasted two weeks. Four hours after LPS challenge, piglets were euthanized via an intravenous injection of sodium pentobarbital (50 mg/kg body weight) (19). Blood was collected and centrifuged at 4000 × g for 10 min to collect serum. Serum samples were stored at -80℃ for later analysis. Segments (1 cm × 1 cm) of the duodenum, jejunum and ileum were gently ushed with 5 mL 0.9% saline twice, and then xed in 10% formalin xative solution for use in histopathological (20). The luminal contents of the ileum were collected, frozen with liquid nitrogen and stored at -80℃ for further analysis.

Gastrointestinal Ph Value And Organ Weight
Rectal temperature and gastrointestinal pH value were determined in stomach, duodenum, jejunum, ileum, and colon. Spleen weight was determined.

Histomorphological Analysis
In brief, the intestine segments xed in 10% formalin solution were dehydrated with different alcohol concentrations, and soaked in para n as previously described (21). Cross-sections were cut into 5-8 micron thin slices, stained with hematoxylin and eosin (22). Histological assays were performed using Scion image software (Scion Image; Scion Corporation, Frederick, MD) to determine villus height and crypt depth in duodenum, jejunum, and ileum sections.

Sequencing Of Bacterial 16s Rrna In Feces
The microbial community in the pig intestine was detected and analyzed using high-throughput sequencing. Primers 515F (5'-ACTCCTACGGGAGGCAGCA-3') and 806R (5'-GGACTACHVGGGTWTCTAAT-3') were used to amplify the V4 variable region of the 16S rDNA gene. The constructed amplicon library was sequenced using the Illumina Hiseq2500 platform (Guangdong Magigene Biotechnology Co., Ltd. Guangdong, China). Paired sequence reads were assembled using Flash (V1.2.11, http://ccb.jhu.edu/software/FLASH/). The default QIIME (1.9.1) was used for quality ltering (23). The trimmed sequence was chimeric ltered, single sequences is discarded, and the obtained sequences were allocated to operational taxonomic units (OTUs) using the USEARCH pipeline (http://www.drive5.com/usearch/) (24). The Mothur algorithm was used to compare representative sequence of each OTU with the non-redundant SILVA database. Alpha and beta diversity analyses were performed using the determined OTUs and UniFrac distances, respectively, and implemented in QIIME.

Statistical analysis
Each piglet was used as a statistical unit. The signi cance of differences was analyzed by one-way ANOVA using General Linear Model procedures in SAS (SAS Institute, Cary, NC, USA), followed by Duncan's multiple range tests. Signi cance was considered at P < 0.05.

Rectal temperature and intestinal pH value
Piglets challenged with LPS in the CL, OL, AL, and OAL groups trended toward higher rectal temperatures than did piglets without LPS challenge in the C group (Fig. 1A). Piglets in the CL, OL, AL, and OAL groups had signi cantly higher stomach pH values than did piglets in the C group (P < 0.05) (Fig. 1B). Piglets in the AL and OAL groups had signi cantly lower duodenum, jejunum, and ileum pH values than did piglets in the C groups (P < 0.05). Inclusion of antibiotics or antibiotics plus olive extract in the diet signi cantly lowered the pH value in the jejunum and ileum compared with the CL groups (P < 0.05) (Fig. 1D, E). There was no signi cant difference in colon pH values in piglets among different treatment groups (P > 0.05) (Fig. 1F).

Serum Biochemical Characteristics
The LPS challenge signi cantly lowered serum glucose levels in the CL group compared with the C group (P < 0.05). Diet inclusion of olive extract signi cantly increased the serum glucose in OL piglets when compared with CL group piglets (P < 0.05). AL group piglets did not signi cantly differ from CL group piglets (P > 0.05). However, OAL group piglets, fed a diet with antibiotics plus olive extract, had signi cantly higher glucose than did the CL group piglets (P < 0.05)( Fig. 2A). The serum total protein content was signi cantly lower in CL, OL, and AL group piglets than in C group piglets (P < 0.05) (Fig. 2B). Serum total protein content was signi cantly lower in AL group piglets than in OAL group piglets (P < 0.05) (Fig. 2B). The piglets in OL groups had signi cantly lower total serum cholesterol than did piglets in the CL groups (P < 0.05) (Fig. 2C).

Serum Antioxidant Capacity
Compared with those in the C group, after i.p. injection of LPS, piglets in the CL group had signi cantly lower GSH-Px (P < 0.05) and SOD (P < 0.05) levels and higher MDA (P < 0.05) and NO (P < 0.05) levels in the serum (Fig. 3A-D). However, compared with the CL group, the piglets in OL, AL, and OAL groups had signi cantly higher GSH-Px (P < 0.05) and SOD (P < 0.05) levels and lower MDA (P < 0.05) and NO (P < 0.05) levels in the serum (Fig. 3A-D). There was no signi cant difference in these four parameters in serum of piglets from OL, AL, and OAL groups (P > 0.05) (Fig. 3A-D).
Compared with the C group, the piglets in the CL group had signi cantly higher levels of LDH (P < 0.05), ALT (P < 0.05), and AST (P < 0.05) and lower ALB levels (P < 0.05) in the serum (Fig. 3E-H). Compared with the CL group, the piglets in the OL, AL, and OAL groups had signi cantly lower serum LDH (P < 0.05), ALT (P < 0.05), and AST (P < 0.05) levels and higher ALB (P < 0.05) levels in the serum (Fig. 3E-H). The piglets in the OL and OAL groups were fed diets containing olive extract and had signi cantly lower serum LDH (P < 0.05) levels than did those in the AL groups (Fig. 3E).
Serum In ammatory Factor And Spleen Index LPS administration signi cantly increased the serum concentration of TNF-α (P < 0.05) and IL-6 (P < 0.05) in the CL group compared with the control group (Fig. 4A, B). Compared with those in the CL group, piglets in the OL, AL, and OAL groups had signi cantly lower serum TNF-α (P < 0.05) and IL-6 (P < 0.05) levels and signi cantly higher IL-10 (P < 0.05) levels ( Fig. 4A-C). There were no signi cant differences in serum TNF-α, IL-6, and IL-10 levels in piglets from OL, AL, and OAL groups (P > 0.05) (Fig. 4A-C). Piglets challenged with LPS in the CL group had signi cantly higher spleen index than did C group piglets (P < 0.05, Fig. 4D). Spleen index values of piglets in OL, AL, and OAL groups did not signi cantly differ from those in CL group piglets (P > 0.05, Fig. 4D).

Intestinal Integrity And Histomorphology
The DAO and D-xylose was determined as indicators of the intestinal integrity. CL group piglets challenged with LPS had signi cantly higher serum DAO and D-xylose levels than did group C piglets (P < 0.05, Fig. 5A, B). The OL and OAL group piglets had signi cantly lower DAO and D-xylose concentrations than did CL group piglets (P < 0.05, Fig. 5A, B). The villus height and crypt depth, based on H&E staining, are shown in Fig. 6. The duodenum and ileum of piglets in control, OL, AL and OAL groups showed regular structures, while LPS stimulation destroyed the intestinal villus structure in the duodenum and ileum of piglets in the CL group (Fig. 6A, B). The duodenum and ileum of piglets in the CL group had signi cantly lower villus height (P < 0.05), higher crypt depth (P < 0.05) and lower villus height/crypt depth ratios (P < 0.05) than those of piglets in the control group (Fig. 6C, D). However, compared with those of the CL group, the duodenum and ileum of piglets in OL, AL, and OAL groups had signi cantly higher villi (P < 0.05), lower crypt depth (P < 0.05), and higher villus height/crypt depth ratios (P < 0.05) with the exception of ileum villus height in AL and OAL groups (P > 0.05) (Fig. 6C, D).
Chao 1 and Shannon and Simpson index values, which re ect species richness and evenness, did not differ among the ve treatment groups (Fig. 8A-C). At the family level (Fig. 8D), there was no signi cant difference in relative abundance of Muribaculaceae, Tannerellaceae, Tannerellaceae, and Ruminococcaceae between the C and CL groups, while the CL group had a signi cantly higher relative abundance of Veillonellaceae and T34 (P < 0.05) and a lower relative abundance of Spirochaetaceae (P < 0.05) than did the C group. The relative abundance of Muribaculaceae was signi cantly lower in CL group piglets than in OAL group piglets (P < 0.05). The relative abundance of Tannerellaceae was signi cantly lower in CL group piglets than in OL, AL, and OAL group piglets (P < 0.05). The relative abundance of Lactobacillaceae was signi cantly lower in CL group piglets than in OL and AL group piglets (P < 0.05). The relative abundance of Ruminococcaceae was signi cantly lower in OL group piglets than in CL group piglets (P < 0.05). There was no signi cant difference in the relative abundance of Veillonellaceae in the intestines among the CL, OL, AL, and OAL group piglets (P > 0.05). The relative abundance of T34 was signi cantly lower in the OL, AL, and OAL group piglets than in the CL group piglets (P < 0.05). The relative abundance of Spirochaetaceae was signi cantly lower in CL group piglets than in OL and AL group piglets (P < 0.05).
At the genus level (Fig. 8E), LPS challenge induced a signi cantly higher relative abundance of Alloprevotella (P < 0.05) and a lower abundance of Clostridium_sensu_stricto_1, Lachnospiraceae_NK4A136_group, and Succinivibrio (P < 0.05) in CL group piglets. The relative abundance of Alloprevotella was signi cantly lower in OL, AL and OAL group piglets than in CL group piglets (P < 0.05). Dietary inclusion of olive extract in the OL group had signi cantly higher relative abundance of Clostridium_sensu_stricto_1 than did the CL group (P < 0.05). The relative abundance of Lactobacillus was signi cantly lower in CL group piglets than in OL and AL group piglets (P < 0.05). There was no signi cant difference in the relative abundance of Lachnospiraceae_NK4A136_group in the intestine of CL, OL, AL, and OAL group piglets (P > 0.05). The OL and OAL group piglets had signi cantly higher relative Succinivibrio abundance than did the CL group piglets (P < 0.05). There was no signi cant difference in relative Sutterella abundance between C and CL group piglets (P > 0.05). The relative abundance of Sutterella was signi cantly lower in CL group piglets than in the OL and OAL groups (P < 0.05). The relative Terrisporobacter abundance was signi cantly lower in the CL group piglets than in OL group piglets (P < 0.05). The OL and AL group piglets had signi cantly higher relative Treponema_2 abundance than did the CL groups (P < 0.05).

Discussion
In this study, weaned piglet that underwent LPS challenge were used as an experimental model of in ammation. The body temperature of healthy piglets is between 38.6℃ and 39.7℃ (25), and in ammation increases the body temperature. Our test results show that the body temperature trended upwards after LPS treatment (P = 0.0965), and that all temperatures were higher than 39.7℃ after LPS treatment. This result indicates that LPS had activated the immune response and in ammation to increase body temperature. Lower gastrointestinal pH is bene cial for the persistence and colonization of bene cial intestinal bacteria. Our results show that the pH of gastric contents in four groups induced by LPS was signi cantly increased compared with that in the control group, indicating that LPS may induce secretion of acid in the stomach. Compared with the CL group, the dietary inclusion of olive extract, antibiotics, or olive extract and antibiotics reduced the intestinal tract pH to a certain extent, which also may lay the foundation for the growth environment of bene cial bacteria.
Blood glucose is the main energy source for body activities. LPS challenge induced an in ammatory response and oxidant stress, which may lead to the consumption of glucose as energy and cause signi cantly lower blood glucose in the CL group. Blood glucose levels in OL and OAL groups were restored to levels similar to those observed in the control without LPS challenge. This may contribute to the anti-in ammatory and anti-oxidant effects of the olive extract, which may allow for resist of LPS stimulation to maintain stable blood glucose concentration. Serum total protein is signi cantly reduced in piglets after stimulation with LPS, which may be attributed to the acute in ammation induced by LPS in CL, OL, and AL groups. The OAL group did not signi cantly differ from the C group, indicating that in the olive and antibiotic group in ammation induced by LPS could be alleviated and in uence the level of total protein in serum. Ampelopsin, a plant extract from ampelopsis grossedentata, has antioxidant activity, alleviates the in ammatory response induced by LPS, and maintains glucose and ALB levels in serum (26).
Olive oil was shown to signi cantly reduce the cholesterol content in serum of rabbits fed high cholesterol diet in an antithrombotic study (27). An in vitro and in vivo study of lipid metabolism regulation showed that Chinese olive extract effectively reduces the level of serum cholesterol (28). Olive extract signi cantly decreased total cholesterol in plasma from 2.1 to 1.3 mmol/L in mice fed highcarbohydrate and high-fat diets (15). Consistently, in our study, the diet containing olive extract, in the OL and OAL groups, had lower total serum cholesterol levels than in the C and CL groups. Taken together, these results indicate that olive extract has the capacity to reduce serum cholesterol.
Olive leaf and its constituents have been previously described to have antioxidant and free radical scavenging properties (29). Weaning of mammals involves many complicated events, including the in uence of environmental and dietary pressures on intestinal development and adaptability, which can lead to in ammation (30). MDA is a reliable indicator of oxidative stress, and can re ect the degree of damage caused by reactive oxygen metabolites in cells (31). SOD, GSH-Px, and NO play key roles in the defense against reactive oxygen species activated by oxidative stress injury (32). Our results show that OL, AL, and OAL signi cantly increase the GSH-Px and SOD levels and reduce the MDA and NO levels in serum after LPS induction. Previous results also showed that olive extract signi cantly decreases plasma MDA, a marker of lipid peroxidation, from 32.2 to 23.6 µmol/L in mice fed a high-carbohydrate and highfat diet (15). These results are consistent with previous results which showed that the levels of serum MDA and NO are increased, and SOD and GSH-Px levels are decreased after LPS injection, while the carnosic acid treatment alleviated the in uence of LPS (33). Therefore, olive extract can exert certain role in alleviating the oxidant stress induced by LPS.
ALT and AST are important indicators of liver function (34). LDH is an important enzyme in sugar metabolism and is a sensitive indicator of cell membrane damage (35). The signi cantly higher serum levels of ALT, AST, and LDH observed in CL group piglets indicate that LPS challenge damages the liver. However, the dietary inclusion of olive extract, antibiotics, or olive extract plus antibiotics alleviates LPSinduced liver damage as observed by signi cantly lower levels of ALT, AST, and LDH in serum from OL, AL, and OAL group piglets. Consistent with our result, it was shown that LPS signi cantly increases ALT and AST levels in serum, while hesperidin treatment reduces the elevated serum AST and ALT (36). Limited data are available on the effects of olive leaf, or its major constituents, on the liver. Olive extract signi cantly decreases plasma ALT, AST, and LDH levels in mice fed a high-carbohydrate and high-fat diet, indicating the hepatoprotective effects of OLE (15). Due to liver damage, ALB content in serum is reduced, while OL, AL, and OAL weaken the degree of ALB reduction in serum (37), indicating that olive extract can relieve liver damage caused by LPS to a certain extent.
IL-10 has immunosuppressive and anti-in ammatory effects, which are opposite to the effects of IL-6 and TNF-α (38). Our results indicate that olive extract alleviates the in ammatory reaction induced by LPS. Consistently, a previous study found that hypericum triquetrifolium extract signi cantly inhibited TNF-α and IL-6 expression and secretion and signi cantly elevated IL-10 secretion (39). Olive leaf-derived polyphenols exert an improtant role in preventing in ammation and oxidative damage-induced primary myocardial insult (15). Similarly, the olive leaf and olive oil treatments decreased oxidative stress and in ammation, TNF-α levels, and increased IL-10 levels (40). These results show that olive extract has antiin ammatory properties.
Villus height, crypt depth, and VCR (villus height to crypt depth ratio) are criteria that re ect the general morphology of the intestinal tract (20,41). LPS treatment led to a rise in the villus end epithelium, reduced the villus height in the duodenum, jejunum, and ileum mucosa, and increased the depth of the ileum mucosa crypt (20,42). Our results show that LPS stimulation decreased the villus height of the duodenum, jejunum, and ileum, increased the crypt depth and reduced the VCR. However, supplementation with olive extract (OL) alleviated LPS-induced in ammation, thus restoring villus height and reducing crypt depth.
The observed increase in DAO and D-xylose levels in piglets challenged with LPS alone suggests that increased intestinal permeability was achieved using this experimental model (43). This abnormal "leaky" gut was associated with destroyed villus structure. Dietary olive extract attenuated the LPS-induced rise in DAO and D-xylose levels in the sera of piglets in the OL and OAL groups, suggesting that olive extract may restore intestinal integrity, preventing DAO and D-xylose from crossing the intestinal barrier and entering the circulation. Previously, LPS injection increased the crypt depth in the ileum and jejunum, and reduced the VCR. However, treatment with procyanidin alleviated LPS-mediated destruction of small intestine morphology (44). Dietary supplementation with resveratrol signi cantly increased villus height and villus height/crypt depth in piglets (45). Morphological damage induced by LPS stimulation in the small intestine includes hemorrhage caused by ulcer and erosion, lumen stenosis, tube wall perforation, and long villi tip damage and plaques (46). In our study, the OL group showed less breakage in the morphology of the small intestine, indicating that, to a certain extent, olive extract could alleviate the damage induced by LPS to ensure the integrity of intestinal morphology and structure.
Venn diagram analysis revealed that LPS treatment reduced the relative OUTs in piglets, while, in the OL group, olive extract supplementation maintains OUTs numbers similar to those of the control group.
Compared with the CL group, olive extract restored the diversity of intestinal microbiota in OL group weaned piglets. Clostridium and Treponema are key participants in nutrition metabolism including in carbohydrate fermentation and polysaccharide and steroid metabolism, and are the key to maintaining the normal physiological function of the intestinal tract (47,48). Several members of the Clostridium family are involved in carbohydrate degradation and metabolic e ciency improvement, which are enriched in feces samples from pigs with higher feed utilization capacity (49,50). Our results indicate that olive extract could increase the relative abundance of Sutterella and Clostridium_sensu_stricto_1 in OL group piglets. Clostridium bacteria involved in butyric acid production and mucin degradation are related to improved gastrointestinal tract health in pigs (51). Therefore, the relative abundance of Clostridium_sensu_stricto_1 in the OL group may enhance epithelial barrier function (52).
Lactobacillus is considered a carbohydrate-utilizing bacterium, and has many genes encoding various functional capabilities related to carbohydrate transportation and utilization (53). Our results suggest that there is a signi cantly higher abundance of Lactobacillus in piglets treated with olive extract than in CL group piglets. The reduced small intestine pH in the group with olive extract supplementation may aid in