Co-Cultures of Lactobacillus acidophilus and Bacillus subtilis Enhance Mucosal Barrier by Modulating Gut Microbiota-Derived Short-Chain Fatty Acids
嗜酸乳杆菌和枯草芽孢杆菌的共培养通过调节肠道微生物群衍生的短链脂肪酸来增强粘膜屏障

Abstract 抽象

Weaning stress induces intestinal barrier dysfunction and immune dysregulation in mammals. Various interventions based on the modulation of intestinal microbiota have been proposed. Our study aims to explore the effects of co-cultures from Lactobacillus acidophilus and Bacillus subtilis (FAM^®) on intestinal mucosal barrier from the perspective of metabolic function of gut microbiota. A total of 180 piglets were allocated to three groups, i.e., a control group (C, basal diet), a FAM group (F, basal diet supplemented with 0.1% FAM), and an antibiotic group (A, basal diet supplemented with antibiotic mixtures). Here, we showed FAM supplementation significantly increased body weight and reduced diarrhea incidence, accompanied by attenuated mucosal damage, increased levels of tight junction proteins, serum diamine oxidase (DAO) and antimicrobial peptides. In addition, 16S rRNA sequencing and metabolomic analysis revealed an increase in relative abundance of Clostridiales, Ruminococcaceae, Firmicutes and Muribaculaceae and a significant increase in the total short-chain fatty acids (SCFAs) and butyric acid in FAM-treated piglets. FAM also increased CD4⁺ T cells and SIgA⁺ cells in intestinal mucosa and SIgA production in colon contents. Furthermore, FAM upregulated the expression of IL-22, short-chain fatty acid receptors GPR43 and GPR41, aryl hydrocarbon receptor (AhR), and hypoxia-inducible factor 1α (HIF-1α). FAM shows great application prospect in gut health and provides a reference for infant weaning.
脱机应激会诱导哺乳动物的肠道屏障功能障碍和免疫失调。已经提出了基于肠道微生物群调节的各种干预措施。本研究旨在从肠道菌群代谢功能的角度探讨嗜酸乳杆菌和枯草芽孢杆菌 （FAM^®） 共培养对肠粘膜屏障的影响。共将 180 头仔猪分为 3 组，即对照组（C，基础日粮）、FAM 组（F，补充 0.1% FAM 的基础日粮）和抗生素组（A，补充抗生素混合物的基础日粮）。在这里，我们显示 FAM 补充剂显着增加了体重并降低了腹泻的发生率，并伴有粘膜损伤减轻、紧密连接蛋白、血清二胺氧化酶 （DAO） 和抗菌肽水平增加。此外，16S rRNA 测序和代谢组学分析显示，在 FAM 处理的仔猪中，梭菌门、瘤胃球菌科、厚壁菌门和毛霉菌门的相对丰度增加，总短链脂肪酸 （SCFAs） 和丁酸显著增加。FAM 还增加了肠粘膜中的 CD4 ⁺ T 细胞和 SIgA ⁺ 细胞以及结肠内容物中 SIgA 的产生。此外，FAM 上调了 IL-22 、短链脂肪酸受体 GPR43 和 GPR41 、芳烃受体 （AhR） 和缺氧诱导因子 1α （HIF-1α） 的表达。FAM在肠道健康中显示出良好的应用前景，为婴儿断奶提供了参考。

1. Introduction 1. 引言

Intestinal barrier functions as a selective construction to prevent environmental antigens invasion, which is critical for immune resistance and host survival [1]. The development of intestinal barrier occurs rapidly after birth and is characterized by decreased gut permeability [2]. Epithelial permeability barrier is mainly regulated by the tight junctions which consist of intracellular and apical intercellular membrane proteins, including zonula occludens (ZO), claudins and occluding [3,4,5], which regulate epithelial leakiness by selectively modulating ion and pore size of the intestinal epithelium [5]. Specialized epithelial cell types such as goblet cells and Paneth cells could also support intestinal barrier function by providing protective mucous layer and secreting antimicrobial peptides [6,7]. Weaning stress induces intestinal barrier dysfunctions, including defects of intestinal epithelial junction, decreased thickness of mucosal layer, and defective production of antimicrobial peptides [8].
肠道屏障作为一种选择性结构来防止环境抗原侵袭，这对免疫抵抗和宿主存活至关重要 [1]。肠道屏障的发育在出生后迅速发生，其特征是肠道通透性降低 [2]。上皮通透性屏障主要受紧密连接调节，紧密连接由细胞内和顶端细胞间膜蛋白组成，包括闭合小带 （ZO）、密蛋白和闭塞 [3,4,5]，它们通过选择性调节肠上皮的离子和孔径来调节上皮渗漏 [5].杯状细胞和潘氏细胞等特殊上皮细胞类型也可以通过提供保护性粘液层和分泌抗菌肽来支持肠道屏障功能[6,7]。脱机应激会诱发肠道屏障功能障碍，包括肠上皮交界处缺陷、粘膜层厚度减少和抗菌肽产生缺陷 [8]。

Intestinal microbiota composition and activity co-evolve with the host from birth and is influenced by nutrition and lifestyle [9]. A million years of coevolution have led to a symbiotic relationship between microbiota and host in which gut microbiota likely mediate host physiology and metabolism [9,10,11]. Specifically, microbiota-induced cell signaling lead to changes in the mucosa barrier function, immune response and metabolic pathway, thereby affecting host physiology and pathophysiology [12,13]. Most of human diseases, such as diarrhea, Crohn’s disease (CD), irritable bowel syndrome (IBS) and obesity, are associated with dysbiosis of gut microbiota and loss of microbial diversity [14,15]. A wide use of antibiotics and other factors may reduce the number of bacterial predators, leading to a decrease in intestinal microbial diversity [16,17]. Considering the complex interplay between gut microbiota and host health, interventions based on the modulation of gut microbiota have been considered one of the most potential. In recent years, increasing studies have demonstrated that probiotic microorganisms confer a health benefit in humans and animals, including interaction with resident microbiota [18,19], modulation of immune function [20], production of antimicrobial compounds [21] and organic acids [22,23], and improving gut barrier integrity [24].
肠道微生物群的组成和活动从出生起就与宿主共同进化，并受营养和生活方式的影响 [9]。一百万年的共同进化导致了微生物群和宿主之间的共生关系，其中肠道微生物群可能介导宿主的生理和代谢[9,10,11]。具体来说，微生物群诱导的细胞信号传导导致粘膜屏障功能、免疫反应和代谢途径发生变化，从而影响宿主生理学和病理生理学[12,13]。大多数人类疾病，如腹泻、克罗恩病 （CD）、肠易激综合征 （IBS） 和肥胖，都与肠道菌群失调和微生物多样性丧失有关 [14,15]。广泛使用抗生素和其他因素可能会减少细菌捕食者的数量，从而导致肠道微生物多样性减少[16,17]。考虑到肠道微生物群与宿主健康之间复杂的相互作用，基于肠道微生物群调节的干预措施被认为是最具潜力的干预措施之一。近年来，越来越多的研究表明，益生菌微生物对人类和动物的健康有益，包括与常驻微生物群的相互作用[18,19]、调节免疫功能[20]、产生抗菌化合物[21]和有机酸[22,23]，以及改善肠道屏障的完整性[24]。

Pigs have many similarities with humans in the intestinal composition and function, making the pig an ideal animal model [25]. With regard to fermenting dietary fibrous composition, pigs and humans both use the colon rather than the cecum as the main site [25]. Furthermore, porcine and humans share 96% similarity in gastro-intestinal microbiota functional pathways [26]. Here, we use a piglet model to study the in vivo mechanism of FAM^® (co-cultures of Lactobacillus acidophilus and Bacillus subtilis) on intestinal barrier function from the perspective of gut microbiota. The effect of FAM on gut barrier was associated with alteration in gut microbiota and butyrate metabolic processes. Our study may give new insights in helping to understand the modulation of probiotic ferment on gut health and their potential mechanisms underlying in animals or humans.
猪在肠道组成和功能上与人类有许多相似之处，使猪成为理想的动物模型[25]。在发酵膳食纤维成分方面，猪和人都使用结肠而不是盲肠作为主要部位[25]。此外，猪和人类在胃肠道微生物群功能途径方面具有 96% 的相似性 [26]。在这里，我们使用仔猪模型从肠道菌群的角度研究 FAM^® （嗜酸乳杆菌和枯草芽孢杆菌的共培养） 对肠道屏障功能的体内机制。FAM 对肠道屏障的影响与肠道菌群和丁酸盐代谢过程的改变有关。我们的研究可能为帮助了解益生菌发酵对肠道健康的调节及其在动物或人类中的潜在机制提供新的见解。

2. Materials and Methods 2. 材料和方法

2.1. Animals and Experimental Design
2.1. 动物和实验设计

The animal experiment was approved by the Animal Care and Use Committee of Zhejiang University (SYXK 2012- 0178) and all experimental procedures conformed to the institutional guideline for animal study. A total of 180 piglets (Duroc ×Landrace × Yorkshire hybrid) weaned at 28 days old with similar initial body weight were randomly allocated to three groups, i.e., a control group (C, basal diet), a FAM group (F, basal diet supplemented with 0.1% FAM), and an antibiotic group (A, basal diet supplemented with antibiotic mixtures). Each group has four replicates (i.e., pens) and fifteen piglets per replicate. All piglets were housed in pens and had free access to feed and water. The three groups were independently housed in separated region and all environmental conditions of the three regions were kept consistently. The basal diet was designed to meet the nutrient requirements of the National Research Council (NRC, 2016) for weaned piglets. FAM^® (provided by Zhejiang Kangwandechuan Technology Co., Ltd., Shaoxing, China) is co-fermented by Lactobacillus acidophilus (≥1 × 106 CFU/g) and Bacillus subtilis (≥1 × 106 CFU/g). Antibiotic mixtures contained 50 mg/kg quinocetone, 55 mg/kg kitasamycin, 75 mg/kg chlortetracycline, and 200 mg/kg oxytetracycline. After a 7 days adaptation period, piglets were fed their respective diet for a 30-day experimental period. The body weight, feed consumption, and diarrhea incidence of each piglet were observed and recorded. The schematic diagram for the animals and experimental design was seen as below (Scheme 1).
动物实验经浙江大学动物护理与使用专业委员会 （SYXK 2012- 0178） 批准，所有实验程序均符合动物研究的机构指南。将 180 头在 28 日龄断奶的仔猪 （杜洛克 ×长白 × 约克郡杂交种） 以相似的初始体重随机分为三组，即对照组 （C，基础日粮）、FAM 组 （F，补充 0.1% FAM 的基础日粮） 和抗生素组 （A，补充抗生素混合物的基础日粮）。每组有四个重复（即猪栏），每个重复有 15 头仔猪。所有仔猪都被饲养在围栏里，可以免费获得饲料和水。3 个组独立安置在不同的区域，并且 3 个区域的所有环境条件保持一致。基础日粮旨在满足国家研究委员会 （NRC， 2016） 对断奶仔猪的营养需求。FAM^®（由浙江康万德川科技有限公司提供，绍兴，中国）由嗜酸乳杆菌 （≥1 × 106 CFU/g） 和枯草芽孢杆菌 （≥1 × 106 CFU/g） 共发酵。抗生素混合物含有 50 mg/kg 喹唑酮、55 mg/kg 基他霉素、75 mg/kg 金霉素和 200 mg/kg 土霉素。经过 7 天的适应期后，仔猪饲喂各自的饮食进行 30 天的实验期。观察并记录每头仔猪的体重、饲料消耗和腹泻发生率。动物和实验设计的示意图如下（方案 1）。

Scheme 1. Animal and experimental design.

2.2. Sample Collection 2.2. 样本采集

At the end of the feeding trial, eight piglets per group (two pig from each replicate with average body weight) were randomly selected and humanely killed after 12 h fasting. Blood samples were collected and centrifuged at 3000× g at 4 °C for 15 min to obtain the serum. Mucosa samples from ileum and colon were collected and snap-frozen in liquid nitrogen and then stored at −80 °C. The intestinal tissues (1 × 1 cm²) from jejunum and ileum were collected and fixed in 4% paraformaldehyde for histomorphology and immunohistochemistry analysis. The jejunal tissues (0.5 × 0.5 cm²) were collected and fixed in 2.5% glutaraldehyde fixative for electron microscopy. The colonic contents (middle part) were collected in 1.5 mL sterile centrifuge tubes and snap-frozen in liquid nitrogen and stored at −80 °C for microbiome and metabolite analyses.
在饲喂试验结束时，随机选择每组 8 头仔猪（每份复制 2 头猪，平均体重）并在禁食 12 小时后人道杀死。收集血样并在 4 °C 下以 3000 × g 离心 15 分钟，得到血清。收集来自回肠和结肠的粘膜样品，并在液氮中快速冷冻，然后储存在 -80 °C 下。 收集来自空肠和回肠的肠道组织 （1 × 1 cm²） 并固定在 4% 多聚甲醛中用于组织形态学和免疫组化分析。收集空肠组织 （0.5 × 0.5 cm²） 并固定在 2.5% 戊二醛固定剂中用于电子显微镜检查。将结肠内容物（中间部分）收集在 1.5 mL 无菌离心管中，并在液氮中快速冷冻并储存在 -80 °C 用于微生物组和代谢物分析。

2.3. Intestinal Histomorphology
2.3. 肠道组织形态学

After dehydration, jejunal and ileal samples were imbedded in paraffin wax and cut into 5 μm sections using a microtome. The sections were further stained with hematoxylin and eosin (H&E) and Periodic Acid-Schiff for morphological analysis and images were acquired using an Olympus BX 51 microscope (Olympus Corporation, Tokyo, Japan). The number of goblet cells (PAS) was measured with computer-assisted microscopy (Micrometrics TM; Nikon ECLIPSE E200, Tokyo, Japan). Transmission electron microscopy and scanning electron microscopy visualization were conducted according to the previous study [27].
脱水后，将空肠和回肠样品包埋在石蜡中，并使用切片机切成 5 μm 的切片。进一步用苏木精和伊红 （H&E） 和过碘酸-希夫染色切片进行形态学分析，并使用奥林巴斯 BX 51 显微镜（奥林巴斯公司，日本东京）获取图像。用计算机辅助显微镜 （Micrometrics TM;Nikon ECLIPSE E200，日本东京）。根据之前的研究，进行了透射电子显微镜和扫描电子显微镜可视化 [27]。

2.4. Gene Expression Determined by Real-Time Quantitative PCR (qRT-PCR)
2.4. 实时定量 PCR （qRT-PCR） 测定基因表达

Relative mRNA expressions of Porcine beta defensin-2 (PBD-2), Porcine beta defensin-3 (PBD-3) and Regenerating islet-derived IIIγ (RegⅢ γ) in the ileum mucosa were determined by qRT-PCR using the designed primers (shown in Supplementary Table S1. Quantitative analysis was performed with the Power SYBR Green PCR Master Mix (Applied Biosystems) on the CFX384 real-time fluorescence quantitative PCR system. GAPDH was used as a housekeeping gene and the relative expression of target gene was analyzed based on the 2−ΔΔCt method.
使用设计的引物（如补充表 S1 所示），通过 qRT-PCR 测定回肠粘膜中猪 β 防御素-2 （PBD-2） 、猪 β 防御素-3 （PBD-3） 和再生胰岛衍生的 IIIγ （RegIII. γ） 的相对 mRNA 表达。在 CFX384 实时荧光定量 PCR 系统上使用 Power SYBR Green PCR 预混液 （Applied Biosystems） 进行定量分析。以 GAPDH 为管家基因，基于 2−ΔΔCt 法分析靶基因的相对表达。

2.5. Western Blot (WB) 2.5. 蛋白质印迹 （WB）

Approximately 100 mg of tissues was dissolved in T-PER Tissue Protein Extraction Reagent (including Protease Inhibitor Cocktail) for protein extraction. The concentrations of total protein were measured with BCA protein assay kit. Relative protein expressions were determined by WB according to the previous study. The primary antibodies used in this study were as follows: MUC2 (Novus NB120-11197, Littleton, CO, USA), ZO-1 (Thermo Fisher 40-2200, Waltham, MA, USA), Claudin1 (Abcam ab129119, Cambridge, UK), Occludin (Abcam ab222691, Cambridge, UK), GPR43 (Proteintech 19952-1-AP, Wuhan, China), GPR41 (Abcam ab103718, Cambridge, UK), HIF-1α (Thermo Fisher MA1-516, Waltham, MA, USA), AhR (Novus Biologicals NB100-2289, Littleton, CO, USA), IL-22 (Affinity DF8343, Changzhou, China) and GADPH (Abcam ab181602, Cambridge, UK). Relative levels of target protein were normalized against GAPDH.
将大约 100 mg 组织溶解在 T-PER 组织蛋白提取试剂（包括蛋白酶抑制剂混合物）中进行蛋白质提取。用 BCA 蛋白测定试剂盒测定总蛋白的浓度。根据以前的研究，通过 WB 确定相对蛋白质表达。本研究中使用的一抗如下：MUC2（Novus NB120-11197，美国科罗拉多州利特尔顿）、ZO-1（Thermo Fisher 40-2200，美国马萨诸塞州沃尔瑟姆）、Claudin1（Abcam ab129119，英国剑桥）、Occludin（Abcam ab222691，英国剑桥）、GPR43（Proteintech 19952-1-AP，中国武汉）、GPR41（Abcam ab103718，英国剑桥）、HIF-1α（Thermo Fisher MA1-516，美国马萨诸塞州沃尔瑟姆）、AhR（Novus Biologicals NB100-2289，Littleton、 CO，美国）、IL-22（Affinity DF8343，中国常州）和 GADPH（Abcam ab181602，英国剑桥）。靶蛋白的相对水平针对 GAPDH 进行归一化。

2.6. Enzyme-Linked Immunosorbent Assay (ELISA)
2.6. 酶联免疫吸附测定 （ELISA）

Protein levels of diamine oxidase (DAO), endotoxin (ET) and D-lactate in serum and SIgA in ileum content were detected by pig ELISA kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) following the manufacturer’s recommendations. The concentrations of total protein were measured by BCA protein assay kit (Beyotime Institute of Biotechnology, Shanghai, China) to normalize the target protein concentration.
按照制造商的建议，用猪 ELISA 试剂盒 （南京建成生物工程研究所） 检测血清中二胺氧化酶 （DAO） 、内毒素 （ET） 和 D-乳酸的蛋白质水平以及回肠内容物中的 SIgA。通过 BCA 蛋白测定试剂盒 （Beyotime Institute of Biotechnology ， Shanghai， China） 测量总蛋白的浓度以使目标蛋白浓度正常化。

2.7. Immunohistochemistry
2.7. 免疫组化

The density of SIgA+ cells and CD4+ T cells were analyzed by immunohistochemistry according to the previous study. Briefly, paraffin-embedded ileum sections were deparaffinized and rehydrated. After unmasking antigens, the sections were stained with SIgA (Pig IgA Antibody, 1:1000; Bethyl Laboratories, Montgomery, TX, USA) and CD4 (Rabbit CD4 Antibody, 1:500; Servicebio, GB11064) primary antibodies overnight at 4 °C and then treated by secondary antibody with streptavidin-horseradish peroxidase. Finally, signals were detected by diaminobenzidine.
根据既往研究，通过免疫组化分析 SIgA+ 细胞和 CD4+ T 细胞的密度。简而言之，石蜡包埋的回肠切片被脱蜡和再水化。揭开抗原后，用 SIgA（猪 IgA 抗体，1：1000;Bethyl Laboratories，美国德克萨斯州蒙哥马利）和 CD4（兔 CD4 抗体，1：500;Servicebio，GB11064） 一抗在 4 °C 下过夜，然后用链霉亲和素-辣根过氧化物酶处理二抗。最后，用二氨基联苯胺检测信号。

2.8. Colonic Content Microbiome Analysis
2.8. 结肠内容物微生物组分析

The microbiome genomic DNA of colon contents was extracted using QIAamp Stool DNA mini kits (Qiagen, New York, NY, USA) following the manufacturer’s instructions, and then variable region 4 (V4) of 16s rRNA gene was amplified by polymerase chain reaction (PCR). The PCR products were mixed and subsequently purified with Qiagen Gel Extraction Kits (Qiagen). Sequencing library was generated using TruSeq DNA PCR-free Sample Preparation Kits (Illumina) according to the manufacturer’s recommendations and sequenced on the Illumina HiSeq2500 platform to obtain 250 bp paired-end reads, which further merged with FLASH (Version 1.2.7, http://ccb.jhu.edu/software/FLASH/) to generate raw tags. Effective tags were generated for further analysis after data filtration and chimera removal. Sequences with > 97% similarity were classified to the same operational taxonomic unit (OTU) by the Uparse software (Version 7.0.1001, http://drive5.com/uparse/), and the taxonomic information for representative sequence was annotated using GreenGene Database. Multiple sequence alignment was conducted on the MUSCLE software (Version 3.8.31, http://www.drive5.com/muscle/) to determine the phylogenetic relationship of different OTUs and the dominant species of different groups. The selected predictive functions were conducted using Spearman’s correlation analysis (IBM SPSS Inc., Chicago, IL, USA).
按照制造商的说明，使用 QIAamp Stool DNA mini kits （Qiagen， New York， NY， USA） 提取结肠内容物的微生物组基因组 DNA，然后通过聚合酶链反应 （PCR） 扩增 16s rRNA 基因的可变区 4 （V4）。将 PCR 产物混合，随后用 Qiagen 凝胶提取试剂盒 （Qiagen） 纯化。根据制造商的建议，使用 TruSeq DNA PCR-free Sample Preparation Kits （Illumina） 生成测序文库，并在 Illumina HiSeq2500 平台上测序以获得 250 bp 的双端读数，进一步与 FLASH （版本 1.2.7， http://ccb.jhu.edu/software/FLASH/） 合并以生成原始标签。在数据过滤和嵌合体删除后，生成了有效的标签以供进一步分析。通过 Uparse 软件 （Version 7.0.1001， http://drive5.com/uparse/） 将相似度> 97% 的序列归类到相同的业务分类单元 （OTU），并使用 GreenGene 数据库对代表性序列的分类信息进行注释。在 MUSCLE 软件 （Version 3.8.31， http://www.drive5.com/muscle/） 上进行多序列比对，以确定不同 OTUs 与不同类群优势种的系统发育关系。使用 Spearman 的相关性分析（IBM SPSS Inc.，芝加哥，伊利诺伊州，美国）进行选定的预测功能。

2.9. Untargeted Metabolomic Analysis
2.9. 非靶向代谢组学分析

Metabolomic analysis of colonic contents was conducted by multiple mass spectrometry (MS) platforms, including gas chromatography mass spectrometry/time-of-flight (GC-MS/TOF) and ultrahigh-performance liquid chromatography/mass spectrometry (UHPLC/MS). GC-MS/TOF analysis was performed on the Agilent 7890A gas chromatograph system coupled with the Pegasus HT TOF MS (Leco) while UHPLC/MS analysis was conducted on the Agilent 1290 UHPLC system coupled to TripleTOF 6600 system (Q-TOF, AB Sciex, Concord, ON, Canada). Further multivariate statistical analysis was performed on the SIMCA software (Version 14.1, MKS Data Analytics Solutions, Concord, ON, Canada). Group differences and group separation variables were analyzed using Orthogonal projections to latent structures-discriminate analysis (OPLS-DA). The predictive ability parameter Q2 and goodness-of-fit parameter R2Y were obtained for estimating the model quality after seven-fold cross validation. The metabolite set enrichment analysis and pathway analysis were carried out to generate the related Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway of each differential metabolite and biomarker metabolic pathways, separately, on the web-based tool MetaboAnalyst (http://www.meta-boanalyst.ca, accessed on 17 March 2021).
通过多种质谱 （MS） 平台对结肠内容物进行代谢组学分析，包括气相色谱质谱/飞行时间 （GC-MS/TOF） 和超高效液相色谱/质谱 （UHPLC/MS）。在Agilent 7890A气相色谱仪系统与Pegasus HT TOF MS （Leco）联用系统上进行GC-MS/TOF分析，在Agilent 1290 UHPLC系统与TripleTOF 6600系统联用（Q-TOF，AB Sciex，Concord，ON，Canada）联用上进行UHPLC/MS分析。在 SIMCA 软件（版本 14.1，MKS Data Analytics Solutions，Concord，ON，Canada）上进行进一步的多变量统计分析。使用潜在结构判别分析的正交投影 （OPLS-DA） 分析组差异和组分离变量。获得预测能力参数 Q2 和拟合优度参数 R2Y，用于估计 7 倍交叉验证后的模型质量。进行代谢物集富集分析和通路分析，以分别在基于网络的工具 MetaboAnalyst 上生成每个差异代谢物和生物标志物代谢通路的相关京都基因和基因组百科全书 （KEGG） 通路（http://www.meta-boanalyst.ca，于 2021 年 3 月 17 日访问）。

2.10. Targeted Metabonomic Analysis
2.10. 靶向代谢组学分析

Quantification of short-chain fatty acid (Acetic acid, propionic acid, butyric acid, isobutyric acid, valeric acid, isovaleric acid and caproic acid) in the colonic contents was analyzed using the following procedures. Briefly, MS analysis of targeted metabolite was performed using the Agilent 1290 Infinity series UHPLC System, equipped with a Waters ACQUITY UPLC BEH amide column or a Waters ACQUITY UPLC BEH C18 column. The standard curve was obtained by subjecting the standard solution with UPLC-parallel reaction monitoring (PRM)-MS/MS analysis. Finally, the extracted ion chromatographs (EICs) of targeted analyte from the samples and the standard solution were obtained and the metabolite concentration of colonic content in each sample was calculated (nmol/g).
使用以下程序定量结肠内容物中的短链脂肪酸（乙酸、丙酸、丁酸、异丁酸、戊酸、异戊酸和己酸）。简而言之，使用配备Waters ACQUITY UPLC BEH酰胺柱或Waters ACQUITY UPLC BEH C18色谱柱的Agilent 1290 Infinity系列UHPLC系统对目标代谢物进行MS分析。对标准溶液进行 UPLC-平行反应监测 （PRM）-MS/MS 分析，获得标准曲线。最后，从样品和标准溶液中获得目标分析物的提取离子色谱仪 （EIC），并计算每个样品中结肠内容物的代谢物浓度 （nmol/g）。

2.11. Statistical Analysis
2.11. 统计分析

Data in this article are expressed as means ± standard error of the means (SEM). Statistical differences between two groups were analyzed by Mann–Whitney U-test or by unpaired Student’s t-test, and data among three groups were evaluated by one-way ANOVA, followed by Tukey’s multiple comparisons (SPSS 23.0). Probability values p < 0.05 were considered statistically significant and 0.05 < p < 0.10 were considered a trend.
本文中的数据表示为均值±均值的标准误差 （SEM）。通过 Mann-Whitney U 检验或未配对的学生检验分析两组之间的统计差异，通过单因素方差分析评估三组之间的数据，然后进行 Tukey 多重比较 （SPSS 23.0）。概率值 p < 0.05 被认为具有统计学意义，0.05 < p < 0.10 被认为是一种趋势。

Microbial richness was measured based on richness indices (Chao1, observed species, ACE) and microbial diversity was accessed by diversity indices (Simpson and Shannon). Anosim analysis was performed using the Bray–Curtis methods. β-diversity was determined based on the principal coordinate analysis (PCoA) and non-metric multidimensional scaling (NMDS). Relative abundance of the microbial flora was expressed as median percentages. Linear discriminant analysis effect size (LEfSe) analysis was accessed on online LEfSe tool (http://huttenhower.sph.harvard.edu/galaxy, accessed on 21 September 2021).
微生物丰富度根据丰富度指数 （Chao1、观察物种、ACE） 测量，并通过多样性指数 （Simpson 和 Shannon） 获取微生物多样性。使用 Bray-Curtis 方法进行 Anosim 分析。β集是根据主坐标分析 （PCoA） 和非度量多维刻度 （NMDS） 确定的。微生物菌群的相对丰度以中位数百分比表示。线性判别分析效应量 （LEfSe） 分析在在线 LEfSe 工具（http://huttenhower.sph.harvard.edu/galaxy，2021 年 9 月 21 日访问）上访问。

3. Results 3. 结果

3.1. FAM Supplementation Promoted Intestinal Barrier Function
3.1. 补充 FAM 促进肠道屏障功能

As shown in Table 1, FAM and antibiotics significantly increased body weight and decreased diarrhea incidence of weaned piglets. In addition, increased number of microvilli and mitochondria, and restored epithelial junctions were observed in FAM and antibiotic group, indicating FAM or antibiotics improved intestinal morphology (Figure 1A). Furthermore, intestinal barrier function in weaned piglets was measured. As shown in Figure 1B,D, compared to control and antibiotic-treated piglets, FAM-treated piglets had higher expression levels of tight junction proteins (ZO-1, and Occludin) and MUC2 protein as well as higher number of goblet cells. Relative mRNA expression of antimicrobial peptides (AMPs) was analyzed by qPCR, involving porcine beta defensin-2 (PBD-2), porcine beta defensin-3 (PBD-3) and regenerating islet-derived IIIγ (RegIIIγ) (Figure 1E). The results showed that FAM or antibiotics significantly upregulated the expression of PBD-2, PBD-3 and RegIIIγ, and PBD-2 expression in FAM group was much higher than that of in antibiotics. To further clarify intestinal epithelial functions were improved following FAM supplementation, the levels of serum endotoxin, DAO and D-lactate were detected. As shown in Figure 1F, the level of DAO was significantly decreased following FAM treatment (Figure 1G), demonstrating FAM improved intestinal barrier function and intestinal permeability.
如表 1 所示，FAM 和抗生素显着增加了断奶仔猪的体重并降低了腹泻发生率。此外，在 FAM 和抗生素组中观察到微绒毛和线粒体数量增加，上皮连接恢复，表明 FAM 或抗生素改善了肠道形态（图 1A）。此外，还测量了断奶仔猪的肠道屏障功能。如图 1B、D 所示，与对照和抗生素处理的仔猪相比，FAM 处理的仔猪具有更高的紧密连接蛋白（ZO-1 和 Occludin）和 MUC2 蛋白表达水平以及更高的杯状细胞数量。通过 qPCR 分析抗菌肽 （AMPs） 的相对 mRNA 表达，涉及猪 β 防御素-2 （PBD-2）、猪 β 防御素-3 （PBD-3） 和再生胰岛衍生的 IIIγ （RegIIIγ） （图 1E）。结果显示，FAM 或抗生素显著上调 PBD-2 、 PBD-3 和 RegIIIγ 的表达，且 FAM 组 PBD-2 表达远高于抗生素。为了进一步阐明补充 FAM 后肠上皮功能得到改善，检测血清内毒素、 DAO 和 D-乳酸水平。如图 1F 所示，FAM 治疗后 DAO 水平显著降低（图 1G），表明 FAM 改善了肠道屏障功能和肠道通透性。

Figure 1. The intestinal barrier homeostasis in piglets. (A) Representative electron microscopy images of cross-sections in the piglet jejunum. From top to bottom: 1–3 Scanning electron microscopy images of villi morphology, scale bars = 300 µm; 4–6 SEM images of microvilli morphology, Scale bars = 100 µm; 7–9 Transmission electron microscope images of microvillous structure, including microvilli morphology: red, epithelial cell junctions: blue, mitochondria: yellow, Scale bars = 1 µm. (B) Up, relative expression of tight junction protein ZO-1, Claudin-1 and Occludin determined by WB assay. Down, quantitative analysis of tight junction protein levels. (C) Left, PAS-stained images of goblet cells in ileum and colon. Right, quantitative analysis of goblets cells. Scale bars = 40μm. (D) Left, WB assay of MUC2 in the ileum. Right, quantitative analysis of MUC2 protein expression. (E) qPCR was performed to detect relative mRNA expression of innate immune factors PBD-2, PBD-3 and RegIIIγ. (F) Contents of serum DAO, endotoxin and D-lactate measured by ELISA assay (n = 8). ** p < 0.01, * p < 0.05.
图 1.仔猪的肠道屏障稳态。（A） 仔猪空肠横截面的代表性电子显微镜图像。从上到下：1-3 绒毛形态的扫描电子显微镜图像，比例尺 = 300 μm;微绒毛形态的 4-6 张 SEM 图像，比例尺 = 100 μm;7-9 微绒毛结构的透射电子显微镜图像，包括微绒毛形态：红色，上皮细胞连接：蓝色，线粒体：黄色，比例尺 = 1 μm。（B） 向上，通过 WB 测定测定紧密连接蛋白 ZO-1、Claudin-1 和 Occludin 的相对表达。下，对紧密连接蛋白水平进行定量分析。（C） 左图，回肠和结肠杯状细胞的 PAS 染色图像。右图，杯状细胞的定量分析。比例尺 = 40μm。（D） 左图，回肠中 MUC2 的 WB 测定。右图为 MUC2 蛋白表达的定量分析。（E） qPCR 检测先天免疫因子 PBD-2 、 PBD-3 和 RegIIIγ 的相对 mRNA 表达。（F） 通过 ELISA 测定法测量的血清 DAO、内毒素和 D-乳酸含量 （n = 8）。** p < 0.01， * p < 0.05.

Table 1. Growth performance of different experimental groups.

Item	Control	FAM	Antibiotic
Initial BW, kg	10.34 ± 0.45	10.02 ± 0.46	9.96 ± 0.51
Final BW, kg	23.75 ± 2.99	25.98 ± 3.31	24.71 ± 2.85
ADG, g/d	447 ± 22.3 b	532 ± 28.5 a	492 ± 19.7 a
ADFI, g/d	688 ± 45.7 ab	736 ± 47.8 a	686 ± 38.6 ab
F/G	1.53 ± 0.05 a	1.36 ± 0.02 b	1.39 ± 0.03 b
Diarrhea incidence	8.00 ± 0.50 a	4.14 ± 0.25 b	4.47 ± 0.32 b

BW: body weight; ADG: average daily gain; ADFI: average daily feed intake; F/G: feed-to-gain ratio. ^{a, b} Within a row, means without a common superscript differ at p < 0.05.

3.2. FAM Regulated Microbial Diversity and Structure in the Colon

To detect the direct influence of FAM on the richness and diversity of the gut microbiome, α-diversity analysis was performed. As shown in Figure 2A, observed species, Chao1, ACE and PD whole tree in FAM group were significantly increased compared with those in antibiotic group, suggesting FAM could increase the richness and diversity of intestinal microbiome. To examine the composition alteration of gut microbiota, β-diversity analysis was conducted. The cluster tree of UPGMA Cluster Analysis indicated a significant dissimilarity in the composition of intestinal microbiota among three groups at the phylum level (Supplementary Figure S1A). The difference of the colonic microbiome among three groups was also confirmed by separately clustered gut microbiota shown in PCoA and NMDS analysis (Figure 2B).
为了检测 FAM 对肠道微生物组丰富度和多样性的直接影响，进行了α多样性分析。如图 2A 所示，FAM 组观察到的物种、Chao1、ACE 和 PD 整棵树与抗生素组相比显著增加，表明 FAM 可以增加肠道微生物组的丰富度和多样性。为了检查肠道微生物群的组成改变，进行了β多样性分析。UPGMA 聚类分析的聚类树表明，在门水平上，三组肠道微生物群的组成存在显著差异（补充图 S1A）。PCoA 和 NMDS 分析中分别聚集的肠道微生物群也证实了三组结肠微生物组的差异（图 2B）。

Figure 2. The diversity and structural changes of intestinal microbiota in piglets. (A) Alpha diversity of the microbial community estimated by observed species, Chao1, ACE, and PD whole tree in the colon. (B) Scatter plot of PCoA based on weighted Unifrac distance. NMDS plot of bacterial communities based on BrayCurtis distance separated the three groups. (C) LDA score plot of colonic microbiota with significant differences among groups were obtained from LDA Effect Size (LEfSe) analysis. Microbiota with LDA score higher than 2 were considered as biomarker microbiota (p: phylum level, c: class level, o: order level, f: family level, g: genus level). C: control group, F: FAM group, A: antibiotic group. ** p < 0.01, * p < 0.05.
图 2.仔猪肠道菌群的多样性和结构变化。（A） 通过观察到的物种、结肠中的 Chao1、ACE 和 PD 整棵树估计的微生物群落的 α 多样性。（B） 基于加权 Unifrac 距离的 PCoA 散点图。基于 BrayCurtis 距离的细菌群落 NMDS 图将三组分开。（C） 通过 LDA 效应大小 （LEfSe） 分析获得组间差异显著的结肠微生物群的 LDA 评分图。LDA 评分高于 2 的微生物群被视为生物标志物微生物群 （p： 门水平，c： 类水平， o： 目水平， f： 科水平， g： 属水平）。C： 对照组， F： FAM 组， A： 抗生素组。** p < 0.01， * p < 0.05.

To explore the specific changes in the structure of gut microbiota, relative abundance of microbial community was analyzed at three different taxonomic levels (Supplementary Figure S1B). At the phylum level, Firmicutes (75.16%) and Bacteroidetes (22.67%) represented the two most dominant microbiota. The relative abundance of Proteobacteria in FAM group (0.32%) and antibiotic group (0.35%) was lower than that in control group (0.52%), indicating FAM and antibiotics could reduce harmful bacteria in colon. At the family level, Ruminococcaceae, Prevotellaceae, Lachnospiraceae, and unidentified_clostridiales were the predominant taxon. Notably, FAM group displayed increased Muribaculaceae abundance, indicating FAM might promote intestinal functions for degrading complex carbohydrates. At the genus level, Terrisporobacter, Bacteroides, Blautia, Faecalibacterium, and Lactobacillus are predominant taxon. LEfSe analysis further revealed FAM markedly increased the relative abundance of 18 bacterial biomarkers including Clostridiales, Ruminococcaceae, Firmicutes and Muribaculaceae and decreased relative abundance of 22 microbial biomarkers involving swine Streptococcus, Bacteroides, and Megamonas (Figure 2C).
为了探索肠道微生物群结构的具体变化，在三个不同的分类水平上分析了微生物群落的相对丰度（补充图 S1B）。在门水平上，厚壁菌门 （75.16%） 和拟杆菌门 （22.67%） 是两种最主要的菌群。FAM 组 （0.32%） 和抗生素组 （0.35%） 变形菌门的相对丰度低于对照组 （0.52%），表明 FAM 和抗生素可以减少结肠有害细菌。在科水平上，瘤胃球菌科、Prevotellaceae、Lachnospiraceae 和 unidentified_clostridiales 是主要的分类单元。值得注意的是，FAM 组显示 Muribaculaceae 丰度增加，表明 FAM 可能促进肠道功能降解复合碳水化合物。在属水平上，Terrisporobacter、Bacteroides、Blautia、Faecalibacterium 和 Lactobacillus 是主要分类单元。LEfSe 分析进一步显示，FAM 显着增加了 18 种细菌生物标志物的相对丰度，包括梭菌门、瘤胃球菌科、厚壁菌门和毛霉菌门，并降低了涉及猪链球菌属、拟杆菌门和巨单胞菌属的 22 种微生物生物标志物的相对丰度（图 2C）。

3.3. Functional Metagenomics Prediction and Spearman’s Correlation Analysis of Gut Microbiota
3.3. 肠道微生物群的功能宏基因组学预测和 Spearman 相关性分析

To explore whether FAM-induced microbial changes modulate the metabolic function of gut microbiota, functional metagenomics prediction of gut microbiota was conducted based on 16S rRNA gene sequencing (Figure 3A–C). Our results showed 18 pathways at the third level of KEGG pathway were significantly altered following FAM supplementation, including significantly increased proportions of ‘butanoate metabolism’, ‘pyruvate metabolism’, and ‘propanoate metabolism’ and significantly decreased proportions of ‘peptidases’, ‘sphingolipid metabolism’, and ‘chaperones and folding catalysts’. Moreover, compared to antibiotics-treated piglets, the proportions of ‘butanoate metabolism’, ‘pyruvate metabolism’, ‘ABC transporters’, ‘propanoate metabolism’, and ‘tryptophan metabolism’ were markedly increased and proportions of ‘lipopolysaccharide biosynthesis’ and ‘lipopolysaccharide biosynthesis proteins’ were decreased in FAM-treated piglets. To further analyze the correlation between significantly different genera and specific predictive functions, spearman’s correlation analysis was performed. As shown in Figure 3D, phyla Firmicutes and genera Clostridium were positively correlated with ‘butanoate metabolism’, ‘pyruvate metabolism’, ‘tryptophan metabolism’, and ‘fatty acid metabolism’ and negatively correlated with ‘lipopolysaccharide biosynthesis’. Meanwhile, phyla Bacteroidetes, Tenericutes, and Helotiales were negatively correlated with ‘lipopolysaccharide biosynthesis’ and positively correlated with ‘butanoate metabolism’ and ‘pyruvate metabolism’.
为了探讨 FAM 诱导的微生物变化是否调节肠道微生物群的代谢功能，基于 16S rRNA 基因测序对肠道微生物群进行了功能宏基因组学预测（图 3A-C）。我们的结果显示，补充 FAM 后，KEGG 通路第三级的 18 条通路发生显著改变，包括 “丁酸代谢” 、 “丙酮酸代谢” 和 “丙酸代谢” 的比例显著增加，以及 “肽酶” 、 “鞘脂代谢” 和 “伴侣和折叠催化剂” 的比例显着降低。此外，与抗生素处理的仔猪相比，FAM 处理仔猪的“丁酸代谢”、“丙酮酸代谢”、“ABC 转运蛋白”、“丙酸代谢”和“色氨酸代谢”的比例显著增加，而“脂多糖生物合成”和“脂多糖生物合成蛋白”的比例降低。为了进一步分析显著不同的属与特定预测功能之间的相关性，进行了 spearman 相关性分析。如图 3D 所示，厚壁菌门和梭状芽胞杆菌属与“丁酸代谢”、“丙酮酸代谢”、“色氨酸代谢”和“脂肪酸代谢”呈正相关，与“脂多糖生物合成”呈负相关。同时，拟杆菌门、特内里菌门和拟杆菌门与“脂多糖生物合成”呈负相关，与“丁酸代谢”和“丙酮酸代谢”呈正相关。

Figure 3. Functional metagenomics prediction and spearman’s correlation analysis of gut microbiota. (A) Scatter plot of Principal Component Analysis (PCA) based on Euclidean distances at level 2 and level 3. (B) Predictive functions of gut microbiota at the third level of the KEGG pathways between F group and C group. (C) Predictive functions of gut microbiota at the third level of the KEGG pathways between F group and A group. (D) Association analysis of specified predictive functions and microbiota with significant differences by Spearman correlation test. Red: positive correlation; Blue: negative correlation. C: control group, F: FAM group, A: antibiotic group. ** p < 0.01, * p < 0.05, blank represents no significant differences.
图 3.肠道微生物群的功能宏基因组学预测和 Spearman 相关性分析。（A） 基于水平 2 和水平 3 的欧几里得距离的主成分分析 （PCA） 散点图。（B） F 组和 C 组之间 KEGG 通路第三级肠道菌群的预测功能。（C） F 组和 A 组之间 KEGG 通路第三级肠道菌群的预测功能。（D） 通过 Spearman 相关检验对特定预测功能与微生物群进行显著差异的关联分析。红色：正相关;蓝色：负相关。C： 对照组， F： FAM 组， A： 抗生素组。** p < 0.01，* p < 0.05，空白表示无显著差异。

3.4. FAM Induced Functional Changes of Intestinal Metabolome in the Colon Contents
3.4. FAM 诱导的结肠内容物肠道代谢组功能变化

The changes of microbiota metabolic function usually result in the changes in metabolites. To exam functional changes of intestinal metabolome in piglets, the colon contents were analyzed by GC-MS. OPLS-DA score plots showed that intestinal metabolome clustered separately among three groups and volcano plots further revealed the significantly different metabolites between groups (Figure 4A,B). As demonstrated in Figure 4C, the levels of palmitoleic acid, lithocholic acid and squalene in FAM group was much higher than those in control and antibiotic group. Moreover, we observed significantly higher concentration of amines (cadaverine, maleimide and o-phosphorylethanolamine) in antibiotic group compared to those in FAM group (Figure 4D). To further identify biomarker KEGG pathways, metabolite set enrichment analysis was performed. The results showed that amino acid metabolism, aminoacyl-tRNA biosynthesis and short chain fatty acid metabolism were significantly affected following FAM treatment (Figure 4E), which was inconsistent with the function prediction of the gut microbiota (Figure 3A,B) that butanoate metabolism was significantly increased in FAM group. Therefore, we subsequently explored the modulatory role of FAM in butanoate metabolism of gut microbiota.
微生物群代谢功能的变化通常会导致代谢物的变化。为了检查仔猪肠道代谢组的功能变化，通过 GC-MS 分析结肠内容物。OPLS-DA 评分图显示肠道代谢组分别聚集在三组之间，火山图进一步揭示了组间代谢物的显着差异（图 4A、B）。如图 4C 所示，FAM 组的棕榈油酸、石胆酸和角鲨烯水平远高于对照组和抗生素组。此外，与 FAM 组相比，我们观察到抗生素组的胺 （尸胺、马来酰亚胺和邻磷酸乙醇胺） 浓度显着升高 （图 4D）。为了进一步鉴定生物标志物 KEGG 通路，进行了代谢物集富集分析。结果显示，FAM 处理后氨基酸代谢、氨酰基-tRNA 生物合成和短链脂肪酸代谢受到显著影响 （图 4E），这与肠道菌群的功能预测 （图 3A、B） 不一致，即 FAM 组丁酸代谢显着增加。因此，我们随后探讨了 FAM 在肠道菌群丁酸酯代谢中的调节作用。

Figure 4. Selected metabolites that significantly changed and the enrichment analysis among groups. (A) OPLS-DA score plots of metabolomics were obtained by GC/MS, from left to right: the comparison of F and C, A and C, F and A, also, F and A and C. All individuals were separated as clusters that conforming the experimental deign. (B) Volcano plots showed the relationship between log2 (Fold Change) and -log10 (p.value). (C) Several metabolites with beneficial effects markedly increased in F group. (D) Relative concentration of amines between F and A groups that showed differences. (E) The top 25 KEGG metabolic pathways with differences between F and C groups, F and A groups. The bottom scale showed the enrichment ratio of each pathway and the color depth of each bar indicated the degree of the difference. C: control group, F: FAM group, A: antibiotic group. ** p < 0.01, * p < 0.05, blank represents no significant differences.
图 4.选择显着变化的代谢物和组间富集分析。（A） 通过 GC/MS 获得代谢组学的 OPLS-DA 评分图，从左到右：F 和 C、A 和 C、F 和 A 的比较，以及 F 和 A 和 C。所有个体都被分成符合实验设计的集群。（B） 火山图显示了 log2 （倍数变化） 和 -log10 （p 值） 之间的关系。（C） F 组几种具有有益作用的代谢物显著增加。（D） F 组和 A 组之间表现出差异的胺的相对浓度。（E） F 组和 C 组、F 和 A 组之间差异的前 25 个 KEGG 代谢通路。底部量表显示每条通路的富集率，每个条形的颜色深度表示差异的程度。C： 对照组， F： FAM 组， A： 抗生素组。** p < 0.01，* p < 0.05，空白表示无显著差异。

3.5. FAM Promoted Butyric Acid Production and Enhanced Mucosal Immune Function
3.5. FAM 促进丁酸生成并增强粘膜免疫功能

To further confirm the alteration in butanoate metabolism detected by the untargeted metabolomic analysis, LC-MS/MS-based targeted metabolomic approach was used to quantify the precise concentration of short-chain fatty acids (SCFAs), including acetic acid, propionic acid, iso-butyric acid, butyric acid, isovaleric acid and valeric acid. As shown in Figure 5A, the levels of butyric acid and total SCFAs in the FAM group were much higher than other groups. Compared to the control and FAM group, the antibiotic group had lower content of total SCFAs, acetic acid and butyric acid. SCFAs are the major metabolic products of microbiota from dietary fiber and have been recognized as important mediators in regulating mucosal immunity and intestinal homeostasis [28]. Therefore, we detected mucosal immune functions and inflammatory responses. The results showed that FAM increased the integrated optical density (IOD) of CD4+ T cells and SIgA+ cells in intestinal mucosa as well as SIgA production in colon contents, suggesting FAM enhanced mucosal immunity (Figure 5B–D). Compared to the antibiotic group, the FAM group had lower expression level of proinflammatory gene NF-κB and TNF-α and higher level of anti-inflammatory gene IL-22 (Figure 5E), which was in accordance with previous studies that intestinal microbiota-derived SCFAs could regulate CD4+ T cell and gut immunity [28]. Compared to the control group, antibiotic group had higher expression of proinflammatory gene TNF-α and lower level of anti-inflammatory gene IL-22, indicating the adverse effects of antibiotics on piglets.
为了进一步确认非靶向代谢组学分析检测到的丁酸酯代谢变化，采用基于 LC-MS/MS 的靶向代谢组学方法定量短链脂肪酸 （SCFA） 的精确浓度，包括乙酸、丙酸、异丁酸、丁酸、异戊酸和戊酸。如图 5A 所示，FAM 组的丁酸和总 SCFA 水平远高于其他组。与对照组和 FAM 组相比，抗生素组总 SCFAs 、 乙酸 和 丁酸 含量较低。SCFA 是膳食纤维中微生物群的主要代谢产物，已被公认为调节粘膜免疫和肠道稳态的重要介质 [28]。因此，我们检测到粘膜免疫功能和炎症反应。结果表明，FAM 增加了肠粘膜中 CD4+ T 细胞和 SIgA+ 细胞的积分光密度 （IOD） 以及结肠内容物中 SIgA 的产生，表明 FAM 增强了粘膜免疫力（图 5B-D）。与抗生素组相比，FAM 组促炎基因 NF-κB 和 TNF-α 表达水平较低，抗炎基因 IL-22 表达水平较高（图 5E），这与既往研究一致，即肠道菌群衍生的 SCFA 可以调节 CD4+ T 细胞和肠道免疫 [28]。与对照组相比，抗生素组促炎基因 TNF-α 表达较高，抗炎基因 IL-22 水平较低，表明抗生素对仔猪有不良影响。

Figure 5. FAM enhanced mucosal immunity and inhibited inflammatory responses. (A) Targeted metabolomic analysis of short-chain fatty acids in colon contents: acetic acid, propionic acid, isobutyric acid, butyric acid, isovaleric acid and valeric acid. (B) Immunohistochemistry staining of SIgA+ cells and CD4+ T cells among three groups. Scale bars = 40 μm. (C) The integrated optical density (IOD) of SIgA+ cells and CD4+ T cells. (D) Secretory immunoglobulin A (SIgA) concentration in colon contents determined by ELISA analysis. (E) Relative mRNA expression of inflammatory-related genes NF-κB, TNF-α, IL-22, and IL-10 determined by qPCR. ** p < 0.01, * p < 0.05.
图 5.FAM 增强粘膜免疫并抑制炎症反应。（A） 结肠内容物中短链脂肪酸的靶向代谢组学分析：乙酸、丙酸、异丁酸、丁酸、异戊酸和戊酸。（B） 三组 SIgA+ 细胞和 CD4+ T 细胞的免疫组化染色。比例尺 = 40 μm。（C） SIgA+ 细胞和 CD4+ T 细胞的积分光密度 （IOD）。（D） 通过 ELISA 分析确定结肠内容物中的分泌型免疫球蛋白 A （SIgA） 浓度。（E） 通过 qPCR 测定炎症相关基因 NF-κB 、 TNF-α 、 IL-22 和 IL-10 的相对 mRNA 表达。** p < 0.01， * p < 0.05.

3.6. FAM Upregulated IL-22 Production and Enhanced GPR41 and GPR43 Activation
3.6. FAM 上调 IL-22 产生并增强 GPR41 和 GPR43 激活

From the above results, we concluded that FAM enhanced intestinal barrier function and upregulated intestinal microbiota derived SCFAs production. To further demonstrate how FAM regulated intestinal barrier, we evaluated the expression short-chain fatty acid receptors including GPR43, GPR41and GPR109A. The results demonstrated that the relative expression of GPR43 and GPR41 in the FAM group was much higher than the control and antibiotic group (Figure 6A). Based on the changes of GPR43, GPR41 and IL-22, we further determined their relative protein expression. Consistently, FAM significantly increased the protein expression of GPR43, GPR41 and IL-22 (Figure 6B). The hydrocarbon receptor (AhR) and hypoxia-inducible factor 1α (HIF-1α) are considered as the essential factors involved in SCFAs-GPR41-IL-22 pathway in CD4+ T cells. We therefore determined the relative protein expression of intestinal mucosal AhR and HIF-1α. Compared to the control and antibiotic group, the FAM group had higher protein expressions of AhR and HIF-1α (Figure 6B), indicating FAM might upregulate of intestinal microbiota-derived SCFA and modulate IL-22 production in CD4+ T cell in two pathways, SCFAs-GPR41-AhR/HIF1α-IL-22 pathway and SCFAs-GPR43-IL-22 pathway. Moreover, the protein levels of GPR43, GPR41, AhR, HIF-1α and IL-22 in the antibiotic group were significantly lower than those in the control and FAM group.
从上述结果，我们得出结论，FAM 增强了肠道屏障功能并上调了肠道微生物群衍生的 SCFA 的产生。为了进一步证明 FAM 如何调节肠道屏障，我们评估了包括 GPR43 、 GPR41 和 GPR109A 在内的短链脂肪酸受体的表达。结果表明，FAM 组中 GPR43 和 GPR41 的相对表达远高于对照组和抗生素组（图 6A）。基于 GPR43 、 GPR41 和 IL-22 的变化，我们进一步确定了它们的相对蛋白表达。一致地，FAM 显着增加了 GPR43 、 GPR41 和 IL-22 的蛋白表达 （图 6B）。碳氢化合物受体 （AhR） 和缺氧诱导因子 1α （HIF-1α） 被认为是 CD4+ T 细胞中 SCFAs-GPR41-IL-22 通路的重要因子。因此，我们确定了肠粘膜 AhR 和 HIF-1α 的相对蛋白表达。与对照组和抗生素组相比，FAM 组 AhR 和 HIF-1α 蛋白表达较高 （图 6B），表明 FAM 可能在 SCFAs-GPR41-AhR/HIF1α-IL-22 通路和 SCFAs-GPR43-IL-22 通路两个途径上调肠道菌群来源的 SCFA 并调节 CD4+ T 细胞中 IL-22 的产生。此外，抗生素组 GPR43 、 GPR41 、 AhR 、 HIF-1α 和 IL-22 蛋白水平显著低于对照组和 FAM 组。

Figure 6. The variation of factors on SCFAs-IL-22 pathway in the colonic mucosa. (A) Relative mRNA expression of GPR43, GPR41 and GPR109A in colonic mucosa. (B) Relative protein levels of GPR43, GPR41, AhR, HIF-1α and IL-22 in colonic mucosa. ** p < 0.01, * p < 0.05.

4. Discussion

Weaning stress induces intestinal barrier dysfunction and immune dysregulation in mammals. Here, we use a piglet model to explore the effects of FAM (a probiotic cocultured by Lactobacillus acidophilus and Bacillus subtilis) on intestinal mucosal barrier from the perspective of metabolic function of gut microbiota. Lactic acid bacteria and bacillus spores have been used as probiotics for human and animal consumption due to their immunostimulatory properties on intestinal immune system [29]. Using single Lactobacillus acidophilus, CRL 1014 was reported to influence lactobacilli community composition and microbial metabolism, thereby improving gut health in humans [30]. Single Bacillus subtilis reduced the Escherichia coli infection and diarrhea incidence in weaned pigs via modulating intestinal microbiota, immune response and blood metabolomics [31]. However, combination of bacterial strains that complement each other and occupy different niches within the gut microbiota environment could lead to an enhancement of the effects on the host immune response and health [32]. Coculture of Lactobacillus with Bacillus cereus could stimulate the biosynthetic capacities of lactic acid strains [33]. Combination of Bacillus coagulans BC198 and Lactobacillus paracasei S38 appeared to show greater efficacy in reducing body fat accumulation and modulating gut microbiota than single strains even at a relatively low dose [34]. Another study reported that the effects of oral administration of Lactobacillus delbrüeckii and Bacillus subtilis combined is better than single on gilthead seabream cellular innate immune responses [32]. Our results showed that co-cultures of Lactobacillus acidophilus and Bacillus subtilis (FAM) increased body weight and decreased diarrhea incidence in piglets, indicating FAM exerted protective roles in weaning stress.
脱机应激会诱导哺乳动物的肠道屏障功能障碍和免疫失调。在这里，我们使用仔猪模型从肠道菌群代谢功能的角度探讨 FAM （一种由嗜酸乳杆菌和枯草芽孢杆菌共培养的益生菌） 对肠道粘膜屏障的影响。乳酸菌和芽孢杆菌孢子因其对肠道免疫系统的免疫刺激特性而已被用作人类和动物食用的益生菌[29]。据报道，使用单个嗜酸乳杆菌，CRL 1014 会影响乳酸杆菌群落组成和微生物代谢，从而改善人类的肠道健康 [30]。单一枯草芽孢杆菌通过调节肠道菌群、免疫反应和血液代谢组学来降低断奶仔猪的大肠杆菌感染和腹泻发生率[31]。然而，在肠道微生物群环境中相互补充并占据不同生态位的细菌菌株的组合可能会导致对宿主免疫反应和健康的影响增强[32]。乳酸菌与蜡样芽孢杆菌共培养可以刺激乳酸菌株的生物合成能力 [33]。凝结芽孢杆菌 BC198 和副干酪乳杆菌 S38 的组合似乎在减少体内脂肪堆积和调节肠道微生物群方面显示出比单一菌株更强的功效，即使在相对较低的剂量下也是如此 [34]。 另一项研究报道，口服 Lactobacillus delbrüeckii 和 Bacillus subtilis 联合给药对金头鲷鱼细胞先天免疫反应的影响优于单独给药 [32]。我们的结果表明，嗜酸乳杆菌和枯草芽孢杆菌 （FAM） 的共培养增加了仔猪的体重并降低了腹泻的发生率，表明 FAM 在断奶应激中发挥了保护作用。

Intestinal barrier functions as a selective construction to prevent environmental antigens invasion, which is critical for immune resistance and host survival [1]. The development of intestinal barrier occurs rapidly after birth and is characterized by decreased gut permeability [2]. Weaning stress induces intestinal barrier dysfunctions, including defects of the intestinal epithelial junction, decreased thickness of the mucosal layer, and defective production of antimicrobial peptides [8]. In the present study, villi injury and impaired epithelial junctions were observed in weaned piglets. FAM treatment received an improvement in intestinal morphological structure and expression of tight junction protein (Occludin and ZO-1) and AMPs. Occludin and ZO-1 are the main transmembrane proteins in the apical junctional complex (AJC), which regulate the paracellular diffusion of ions and small molecules across epithelial barriers [35,36]. It has been reported that disorder of the AJC contributes to the impaired epithelial barrier function and is a common feature of numerous inflammatory diseases [37]. The increased expression of Occludin and ZO-1 in the present study indicated that epithelial permeability barrier was improved following FAM treatment. Goblet cells and their secretion products mucus are essential portions in mucosal barrier to defense environmental antigens [38]. It has been reported that a Bacillus licheniformis—B. subtilis mixture could increase the number of goblet cells in the ileum. Our study showed that FAM increased goblet cells and MUC2 levels, evidencing an enhanced mucosal barrier in FAM piglets. PBD-2, PBD-3 and RegⅢ γ belong to a family of antimicrobial peptides, which stimulate innate immunity [39]. PBD-2 could alleviate inflammation through interacting with Toll-like receptor 4 and suppressing the downstream NF-κB signaling pathway [7]. RegIIIγ has canonical C-type lectin domains that bind to the peptidoglycan of the bacterial cell wall and has direct antimicrobial activity against Gram-positive bacteria, thus protecting epithelial barrier function of intestinal mucosa [28]. PBD-2, PBD-3 and RegⅢ γ levels in the FAM group were higher than the control group, indicating the protective effects of FAM on mucosal barrier. Moreover, FAM reduced the serum level of endotoxin, DAO and D-lactate, indicating that FAM could promote the development of mucosal barrier function and decrease intestinal permeability.
肠道屏障作为一种选择性结构来防止环境抗原侵袭，这对免疫抵抗和宿主存活至关重要 [1]。肠道屏障的发育在出生后迅速发生，其特征是肠道通透性降低 [2]。脱机应激会诱发肠道屏障功能障碍，包括肠上皮连接处缺陷、粘膜层厚度减少和抗菌肽产生缺陷 [8]。在本研究中，在断奶仔猪中观察到绒毛损伤和上皮连接受损。FAM 治疗改善了肠道形态结构以及紧密连接蛋白 （Occludin 和 ZO-1） 和 AMPs 的表达。Occludin 和 ZO-1 是顶端连接复合体 （AJC） 中的主要跨膜蛋白，它们调节离子和小分子跨上皮屏障的细胞旁扩散 [35,36]。据报道，AJC 疾病会导致上皮屏障功能受损，是许多炎症性疾病的常见特征 [37]。本研究中 Occludin 和 ZO-1 表达的增加表明 FAM 治疗后上皮通透性屏障得到改善。杯状细胞及其分泌产物粘液是粘膜屏障中防御环境抗原的重要组成部分 [38]。据报道，地衣芽孢杆菌—枯草芽孢杆菌混合物可以增加回肠中杯状细胞的数量。我们的研究表明，FAM 增加了杯状细胞和 MUC2 水平，证明了 FAM 仔猪的粘膜屏障增强。 PBD-2、PBD-3 和 RegIII. γ属于抗菌肽家族，可刺激先天免疫 [39]。PBD-2 可以通过与 Toll 样受体 4 相互作用并抑制下游 NF-κB 信号通路来缓解炎症 [7]。RegIIIγ 具有经典的 C 型凝集素结构域，可与细菌细胞壁的肽聚糖结合，对革兰氏阳性菌具有直接的抗菌活性，从而保护肠粘膜的上皮屏障功能 [28]。FAM 组 PBD-2 、 PBD-3 和 RegIII. γ 水平高于对照组，表明 FAM 对黏膜屏障有保护作用。此外，FAM 降低了血清内毒素、 DAO 和 D-乳酸水平，表明 FAM 可以促进粘膜屏障功能的发育，降低肠道通透性。

Studies have demonstrated that probiotics can regulate the structure and diversity of gut microbiota in the host [40]. Our present study showed that observed species, Chao1, ACE and PD whole tree of gut microbiota in FAM-treated piglets were significantly increased, suggesting FAM could improve the α-diversity of gut microbiota. Regarding structural alterations of the microbial community, a study showed that probiotics could increase the Firmicutes: Bacteroidetes ratio and reduce diarrhea in a piglet model [41]. In the present experiment, the Firmicutes was increased while the Bacteroidetes was decreased following FAM treatment, consistent with the previous study. In addition, LEfSe analysis further revealed FAM markedly increased the relative abundance of 18 bacterial biomarkers including Clostridiales, Ruminococcaceae, Firmicutes and Muribaculaceae and decreased relative abundance of 22 microbial biomarkers involving swine Streptococcus, Bacteroides, and Megamonas. Ruminococcaceae is the main bacteria that convert primary bile acids into secondary bile acids and further facilitated intestinal absorption and innate defense [42,43]. The Genomic analysis revealed that Muribaculaceae (S24-7) had functional diversity in degrading complex carbohydrates [44]. The results suggested that FAM modulated microbial diversity and promoted the relative abundance of beneficial microbiota.
研究表明，益生菌可以调节宿主肠道菌群的结构和多样性 [40]。我们目前的研究表明，在 FAM 处理的仔猪中观察到的肠道微生物群物种 Chao1 、 ACE 和 PD 全树显著增加，表明 FAM 可以改善肠道菌群的α多样性。关于微生物群落的结构改变，一项研究表明，益生菌可以增加厚壁菌门：拟杆菌门的比率并减少仔猪模型中的腹泻 [41]。在本实验中，FAM 处理后厚壁菌门增加，而拟杆菌门减少，与之前的研究一致。此外，LEfSe 分析进一步揭示 FAM 显着增加了 18 种细菌生物标志物的相对丰度，包括梭菌门、瘤胃球菌科、厚壁菌门和毛霉菌门，降低了涉及猪链球菌属、拟杆菌门和巨单胞菌属的 22 种微生物生物标志物的相对丰度。瘤胃球菌科是将初级胆汁酸转化为次级胆汁酸并进一步促进肠道吸收和先天防御的主要细菌[42,43]。基因组分析显示，Muribaculaceae （S24-7） 在降解复合碳水化合物方面具有功能多样性 [44]。结果表明，FAM 调节微生物多样性并促进有益微生物群的相对丰度。

To explore the connection between FAM-induced microbial alterations and modulation of intestinal barrier function, functional metagenomics prediction of gut microbiota was conducted based on 16S rRNA gene sequencing. We found FAM altered 18 pathways at the third level of KEGG pathway including ‘butanoate metabolism’, ‘pyruvate metabolism’ and ‘propanoate metabolism’. Spearman’s correlation analysis further showed higher proportions of bacteria (phyla Firmicutes and genera Clostridium) related to ‘butanoate metabolism’, ‘pyruvate metabolism’, ‘propanoate metabolism’ and ‘fatty acid metabolism’, indicating that FAM might have rebuilt intestinal barrier homeostasis through modulating the metabolic function of gut microbiota. To further confirm the changes in metabolic function of gut microbiota, we performed multiple MS platform-based untargeted metabolomic analysis. The metabolome of colon contents in FAM group was associated with a marked increase in palmitoleic acid, lithocholic acid and squalene, which showed beneficial effects on host immunity and inflammation [45,46,47]. Furthermore, metabolite set enrichment analysis demonstrated amino acid metabolism, aminoacyl-tRNA biosynthesis and short chain fatty acid (SCFA) metabolism were significantly affected following FAM. SCFAs are major metabolic products of intestinal microbiota from dietary fiber and have been recognized as important mediators in regulating mucosal barrier function [40]. The alteration of SCFA metabolism in untargeted metabolomic analysis was in consistent with the function prediction of the gut microbiota, suggesting FAM regulated the SCFA metabolism of gut microbiota in weaning piglets.
为探讨 FAM 诱导的微生物改变与肠道屏障功能调节之间的联系，基于 16S rRNA 基因测序对肠道菌群进行了功能性宏基因组学预测。我们发现 FAM 改变了 KEGG 通路第三级的 18 条通路，包括 'butanoate metabolism' 、 'pyruvate metabolism' 和 'propanoate metabolism'。Spearman 的相关性分析进一步显示，与“丁酸代谢”、“丙酮酸代谢”、“丙酸代谢”和“脂肪酸代谢”相关的细菌 （厚壁菌门和梭菌属） 比例较高，表明 FAM 可能通过调节肠道菌群的代谢功能重建了肠道屏障稳态。为了进一步确认肠道菌群代谢功能的变化，我们进行了多个基于 MS 平台的非靶向代谢组学分析。FAM 组结肠内容物代谢组与棕榈油酸、石胆酸和角鲨烯的显著增加有关，对宿主免疫和炎症有有益作用 [45,46,47]。此外，代谢物集富集分析表明，FAM 后氨基酸代谢、氨酰基-tRNA 生物合成和短链脂肪酸 （SCFA） 代谢受到显着影响。SCFA 是膳食纤维中肠道微生物群的主要代谢产物，已被公认为调节粘膜屏障功能的重要介质 [40].非靶向代谢组学分析中 SCFA 代谢的改变与肠道菌群的功能预测一致，表明 FAM 调节断奶仔猪肠道菌群的 SCFA 代谢。

To further confirm the changes of SCFA metabolism detected by untargeted metabolomic analysis, we performed a targeted metabolomic analysis to quantify SCFA metabolism. Previous studies have shown that probiotics exerted protective effects on gut health in humans and animals through modulating SCFAs production by bacteria [48,49,50]. In the present study, the production of butyric acid and total SCFAs in colon content was significantly increased following FAM. SCFAs are regarded as important mediators in the communication between the intestinal microbiome and the immune system [51]. Microbiota-derived SCFAs have been demonstrated to protect against several diseases by regulating different immune cells [52,53,54]. Therefore, we detected mucosal immune functions and inflammatory responses in colon tissues. Intestinal-associated lymphoid tissues and antibodies were the important components of mucosal immunity. The differentiation of B cells into SIgA+ cells needed the cytokines secreted by CD4+ T cells and the SIgA+ cells secreted SIgA to the mucus. In the previous study, formula-fed infants receiving a probiotic supplementation maintained higher fecal SIgA levels at the treatment period [55,56]. In the present study, we observed higher levels of CD4+ T cells and SIgA+ cells in intestinal mucosa and SIgA in colon contents following FAM treatment, which was consistent with previous study. A recent study reported that microbiota- derived SCFAs could regulate CD4+ T cell IL-22 production and gut immunity [54]. In our study, FAM decreased the expression of proinflammatory genes and increased the expression of anti-inflammatory gene Il22, indicating FAM strengthened mucosal immunity by modulating SCFAs metabolism and IL-22 production. SCFAs have been shown to reduce intestinal inflammation through regulating host immunity, in which the SCFAs-GPR pathways is considered as one of the most vital mechanisms. Free fatty acids receptors (FFARs) belong to G-protein coupled receptors (GPCRs) including GPR41, GPR43 and GPR109A, which are mainly expressed in intestinal epithelial cells and immune cells and activated by SCFAs [57]. A recent study found that butyrate could promote IL-22 production in CD4+ T cells and ILCs through GPR41 and HDAC inhibition, where AhR and HIF-1α participated in IL-22 production [58]. In our study, we evaluated the expression of IL-22 and short-chain fatty acid receptors including GPR43, GPR41and GPR109A. The results showed that GPR41 and GPR43 expression as well as AHR and HIF1α expression in FAM-treated piglets had a significant increase, suggesting the SCFAs-GPR41 pathway might participate in IL-22 upregulation. The interaction of mucosal IL-22 and intestinal microbiota was pivotal in keeping intestinal homeostasis. On one hand, IL-22 adjusted the gut microbiota composition to improve intestinal barrier. On the other hand, the metabolites of gut microbiota promoted the IL-22 production [58,59].
为了进一步确认非靶向代谢组学分析检测到的 SCFA 代谢变化，我们进行了靶向代谢组学分析以量化 SCFA 代谢。先前的研究表明，益生菌通过调节细菌产生SCFAs，对人类和动物的肠道健康产生保护作用[48,49,50]。在本研究中，FAM 后结肠内容物中丁酸和总 SCFA 的产生显着增加。SCFA 被认为是肠道微生物组和免疫系统之间通讯的重要介质 [51]。微生物群衍生的 SCFA 已被证明可以通过调节不同的免疫细胞来预防多种疾病 [52,53,54]。因此，我们检测到结肠组织中的粘膜免疫功能和炎症反应。肠道相关淋巴组织和抗体是黏膜免疫的重要组成部分。B 细胞分化为 SIgA+ 细胞需要 CD4+ T 细胞分泌的细胞因子，而 SIgA+ 细胞将 SIgA 分泌到粘液中。在之前的研究中，接受益生菌补充剂的配方奶粉喂养婴儿在治疗期间保持了较高的粪便 SIgA 水平 [55,56]。在本研究中，我们观察到 FAM 治疗后肠粘膜中 CD4 + T 细胞和 SIgA + 细胞以及结肠内容物中 SIgA 水平较高，这与之前的研究一致。最近的一项研究报道，微生物群衍生的 SCFA 可以调节 CD4+ T 细胞 IL-22 的产生和肠道免疫 [54]。 在我们的研究中，FAM 降低了促炎基因的表达并增加了抗炎基因 Il22 的表达，表明 FAM 通过调节 SCFAs 代谢和 IL-22 的产生来增强粘膜免疫力。SCFA 已被证明可以通过调节宿主免疫来减少肠道炎症，其中 SCFAs-GPR 通路被认为是最重要的机制之一。游离脂肪酸受体 （FFARs） 属于 G 蛋白偶联受体 （GPCRs），包括 GPR41、GPR43 和 GPR109A，主要在肠上皮细胞和免疫细胞中表达，并被 SCFA 激活 [57]。最近的一项研究发现，丁酸盐可以通过 GPR41 和 HDAC 抑制促进 CD4+ T 细胞和 ILC 中 IL-22 的产生，其中 AhR 和 HIF-1α 参与 IL-22 的产生 [58]。在我们的研究中，我们评估了 IL-22 和短链脂肪酸受体（包括 GPR43、GPR41 和 GPR109A）的表达。结果显示，FAM 处理仔猪中 GPR41 和 GPR43 表达以及 AHR 和 HIF1α 表达显著增加，提示 SCFAs-GPR41 通路可能参与 IL-22 上调。粘膜 IL-22 和肠道菌群的相互作用在维持肠道稳态方面起着关键作用。一方面，IL-22 调整肠道菌群组成以改善肠道屏障。另一方面，肠道微生物群的代谢物促进了 IL-22 的产生 [58,59]。

5. Conclusions 5. 结论

In conclusion, our results demonstrated that FAM modulated intestinal barrier homeostasis and immune function though regulation of gut microbiota and SCFAs. GPR41, GPR43, AhR and HIF1α might mediate in the butyric acid induction of IL-22. Thus, our study provided FAM as a novel probiotic intervention in the regulation of gut homeostasis and may give new insights in helping to understand their potential mechanisms underlying in animals or humans.
总之，我们的结果表明，FAM 通过调节肠道菌群和 SCFA 调节肠道屏障稳态和免疫功能。GPR41 、 GPR43 、 AhR 和 HIF1α 可能介导丁酸诱导 IL-22。因此，我们的研究将 FAM 作为调节肠道稳态的新型益生菌干预措施，并可能为帮助了解它们在动物或人类中的潜在机制提供新的见解。