Bacillus subtilis-Fermented Products Ameliorate the Growth Performance, Alleviate Intestinal Inflammatory Gene Expression, and Modulate Cecal Microbiota Community in Broilers during the Starter Phase under Dextran Sulfate Sodium Challenge.

The aim of this study was to evaluate the effects of B. subtilis-fermented products (BSFP) on growth performance, intestinal inflammatory gene expression, and cecal microbiota community in broilers challenged with dextran sulfate sodium (DSS) in a 14-day experiment. A total of 32, 1-day-old male broiler chickens (Ross 308), were randomly divided into four groups of eight birds per group and reared individually (n=8). The treatments consisted of a control diet without supplementation and DSS challenge, control diet plus 1.5% DSS, control diet plus 1 g/kg BSFP in combination with 1.5% DSS, and control diet plus 3 g/kg of BSFP in combination with 1.5% DSS. The results showed that BSFP supplementation (1 and 3 g/kg) partially improved body weight and average daily gain in broilers under DSS challenge. Relative to DSS treatment alone, BSFP supplementation dose-dependently increased the body weight of broilers at 7 days of age, with the average daily gain being at 1 to 7 days of age. BSFP supplementation (1 and 3 g/kg) alleviated intestinal inflammatory gene expression in broilers under DSS challenge. The richness and evenness of bacterial species in cecal digesta increased in a dose-dependent manner in the groups treated with BSFP (1 and 3 g/kg) in combination with DSS challenge, compared with the control group. Unweighted principal coordinate analysis indicated distinct clusters separating the group treated with 3 g/kg of BSFP in combination with DSS challenge from the other three groups. The abundance of short-chain fatty acid-producing bacteria (genus Ruminococcaceae_unclassified) increased and that of mucindegrading bacteria (genus Ruminococcus torques group) decreased in the cecal digesta of broilers fed 3 g/kg of BSFP, compared with the control group. In conclusion, BSFP supplementation dose-dependently improved growth performance, reduced gut inflammation, and regulated the cecal microbiota of broilers exposed to DSS challenge during the starter phase. 2022, Japan Poultry Science Association.

Introduction

Poultry products are one of the most important sources of protein for human consumption worldwide. However, chronic low-grade intestinal inflammation has a negative impact on poultry productivity by impairing nutrient utilization, resulting in severe economic losses. In the past, the use of antibiotics at low dosages as growth promoters (AGP) was widespread. Supplementation of AGP in feeds also reduces immunological stress and low-level inflammation, thereby improving growth in poultry (Roura ; Niewold, 2007). However, the use of AGP in animal husbandry has been fully banned in the European Union since 2006 (Casewell ). Therefore, there is an urgent need to find alternatives to AGP to prevent intestinal inflammation in broilers.

Chronic intestinal inflammation models for broilers have been proposed, such as nutritional stimulation and dextran sulfate sodium (DSS) (Menconi ; Dal Pont ). Dextran sulfate sodium, a heparin-like polysaccharide, is commonly used to induce enteric inflammation in rodents (Laroui ). Administration of DSS in drinking water has been shown to disrupt the intestinal structure and induce necrotic enteritis in broilers (Menconi ), causing body weight loss, diarrhea, and intestinal bleeding in broilers (Menconi ; Kuttappan ).

Proposed alternatives to AGP in poultry feeds include phytogenic feed additives, antimicrobial peptides, prebiotics, and probiotics (Abudabos ; Abudabos ; Suresh ). The effect of phytogenics or probiotics is similar to that of the AGP in broilers under Salmonella typhimurium or Clostridium perfringens challenge during the starter phase (Abudabos , 2018). Bacillus-based probiotics have been shown to enhance im munity and alleviate lipopolysaccharide-induced (LPS)-induced immunological stress in broilers (Lee ; Li ; Gadde ). Our previous findings dem onstrated that B. subtilis-fermented products (BSFP), con taining B. subtilis spores and the antibacterial cyclic lipo peptides derived from them, can prevent C. perfringens-induced necrotic enteritis and promote growth performance in broilers (Cheng ; Horng ). B. subtilis-derived antibacterial cyclic lipopeptides have inhibitory effects on LPS-induced inflammation (Kim ; Zhang ). Furthermore, BSFP also improve growth performance and modulates the gut microbiota of broilers under LPS challenge (Chen and Yu, 2021).

The gut microbiota regulates several physiological processes in poultry, such as nutrient utilization, gut morphology, and immune response (Diaz Carrasco ). Disruption of intestinal microbiota reduces nutrient utilization and affects the immune response, leading to growth retardation in broilers (Dibner and Richards, 2005). DSS-induced inflammation not only causes intestinal dysfunction, but also disrupts the gut microbial composition (Kozhakhmetov ). It has been demonstrated that DSS can induce microbial dysbiosis, and altered gut microbiota is associated with changes in the expression of inflammation-related genes (Håkansson ; Shen ). DSS has been used to imitate chronic gut inflammation caused by environmental factors, pathogens, and feed ingredients in the field in broilers (Menconi ; Zou , 2019). Our previous study demonstrated that BSFP supplementation increases the abundance of beneficial microorganisms in the cecum of LPS-challenged broilers, and thus improves their growth (Chen and Yu, 2021). However, to the best of our knowledge, no studies have so far examined the effects of BSFP on the growth performance, intestinal inflammation, and cecal microbiota of broilers under DSS challenge.

Therefore, we hypothesized that BSFP supplementation can improve growth, alleviate intestinal inflammation, and normalize the gut microbiota in broilers under DSS challenge. Since the starter phase in broiler production is a crucial stage, the main objective of the present study was to investigate the effects of different levels of BSFP on growth performance, intestinal inflammation, and cecal microbiota of broilers under DSS challenge during the starter phase.

Materials and Methods

Animal Study

BSFP are commercially available feed additives (Life Rainbow Biotech, Yilan, Taiwan), and the concentration of B. subtilis spores and B. subtilis-derived antimicrobial cyclic lipopeptide (surfactin) in fermented products were 2×1013 CFU/g and 1.05 mg/g, respectively (Chen and Yu, 2021). The animal protocol was approved by the Institutional Animal Care and Use Committee of the National Ilan University (108-4). One-day-old healthy male broiler chickens (Ross 308) were obtained from a commercial hatchery. On day 1, 32 chicks with an average body weight of 43.9±2.25 g were randomly assigned to four treatments, with 8 birds per treatment. The birds were reared individually in stainless steel, temperature-controlled cages. Each cage was equipped with a nipple drinker connected to an independent reservoir. The experimental diets were (1) a basal diet with neither BSFP nor DSS as a control (C), (2) a basal diet plus 1.5% DSS in drinking water (D), (3) a basal diet plus 1 g/kg of BSFP in combination with 1.5% DSS in drinking water (LD), and (4) a basal diet plus 3 g/kg of BSFP in combination with 1.5% DSS in drinking water (HD). DSS (molecular weight 40,000; Bioman, New Taipei City, Taiwan) was added to the drinking water from days 3 to 13. The volume of drinking water to be provided to each bird was calculated based on age according to a previous study (Pesti ). The diets were formulated to meet or exceed the requirements of the birds according to the National Research Council recommendations (NRC, Nutrient Requirements for Poultry, 1994, Table 1). No coccidiostats or antibiotics were included in the diets. Feed and water were provided ad libitum throughout the duration of the experiment. Room temperature was maintained between 32 and 34°C from days 1 to 3, at 30°C from days 4 to 7, and at 27°C from days 8 to 14. The birds received continuous light for the first three days, and were then maintained under a 20 h light/4 h dark regime for the remainder of the study. Broilers were vaccinated by nose-drop administration with combined Newcastle disease-infectious bronchitis vaccines on day 4. The individual body weight and feed intake were recorded daily. The growth performance (average body weight, average daily gain, average daily feed intake, and feed conversion ratio) was calculated from 2 phases (days 1 to 7 and days 8 to 14). The mortality of broilers was monitored daily.

Table 1.

Composition of Basal Diets

Item	Day 1 to 14
Ingredients (as fed basis, %)
Corn, yellow	47.08
Soybean meal	44.1
Fish meal	5.0
Limestone	2.0
Monocalcium phosphate	1.0
Salt	0.4
Choline chloride	0.02
DL-methionine	0.2
Vitamin premix1	0.1
Mineral premix2	0.1
Nutrient levels
Metabolizable energy (kcal/kg)	3373.8
Crude protein (%)	23.0
Crude fat (%)	2.67
Lysine (%)	1.17
Methionine + Cystine (%)	0.85
Calcium (%)	1.39
Total phosphorus (%)	0.74

Supplied per kg of diet: 1.8 mg all-trans-retinyl acetate, 0.02 mg cho lecalciferol, 8.3 mg alpha-tocopheryl acetate, 2.2 mg menadione, 2 mg pyridoxine HCl, 8 mg cyanocobalamin, 10 mg nicotinamide, 0.3 mg folic acid, 20 mg D-biotin, and 160 mg choline chloride.

Supplied per kg of diet: 32 mg Mn (MnSO4·H2O), 16 mg Fe (FeSO4·7H2O), 24 mg Zn (ZnO), 2 mg Cu (CuSO4·5H2O), 800 µg I (KI), 200 µg Co (CoSO4), and 60 µg Se.

Small Intestinal Morphology Analysis

At the end of the experiment (day 14), 4 birds per group were randomly chosen (n=4) and euthanized by inhalation of carbon dioxide gas to collect the small intestine, as previously described (Chen and Yu, 2020). Briefly, fixed intestinal samples (duodenum, jejunum, and ileum) were prepared using paraffin embedding techniques. Samples were sectioned at a thickness of 5 µm (three cross-sections from each sample), placed on a glass slide, and stained with hematoxylin and eosin. Villus height and crypt depth were measured randomly for 30 villi per bird, using an Olympus CX43 microscope (Olympus Corporation, Tokyo, Japan).

Gene Expression Analysis

At the end of the experiment, birds chosen for gene expression analysis were the same as those used for the small intestinal morphology analysis (n=4). The expressions of inflammatory genes (cyclooxygenase 2, inducible nitric oxide synthase, interleukin 1β, and interleukin 6) in intestinal samples were determined using quantitative polymerase chain reaction (qPCR). Total RNA from the small intestine was isolated using TRIzol reagent, according to the manufacturer's instructions (Thermo Scientific, Waltham, MA, USA). After quantification by spectrophotometry, RNA was reverse-transcribed using an iScript cDNA Synthesis kit (Bio-Rad, Hercules, CA, USA). qPCR was carried out in triplicates using the iQ SYBR Green Supermix kit and Miniopticon Real-Time PCR Detection System (Bio-Rad). mRNA expression for each gene was calculated after normalization to 18S rRNA expression using the 2-ΔΔCt method. The qPCR primers were as follows: 5-AAC ACA ATA GAG TCT GTG ACG TCT T-3 and 5-TAT TGA ATT CAG CTG CGA TTC GG-3 for cyclooxygenase 2 (cox2); 5-AGG CCA AAC ATC CTG GAG GTC-3 and 5-TCA TAG AGA CGC TGC TGC CAG-3 for inducible nitric oxide synthase (iNOS); 5-CGC TCA CAG TCC TTC GAC-3 and 5-TGA GCC TCA CTT TCT GGC-3 for interleukin 1β (il-1β); 5-AGG ACG AGA TGT GCA AGA AGT TC-3 and 5-TTG GGC AGG TTG AGG TTG TT-3 for interleukin 6 (il-6); 5-ATA ACG AAC GAG ACT CTG GCA-3 and 5-CGG ACA TCT AAG GGC ATC ACA-3 for 18S rRNA.

16S rRNA Sequencing and Analysis

Cecal digesta from broilers was freshly collected at the end of the experiment, and 4 birds per group were used for microbiota analysis (n=4). Total bacterial DNA from the cecum content was extracted using the ZymoBIOMICS DNA Miniprep Kit (Zymo Research, Irvine, CA, USA) according to the manufacturer's instructions. The V3 and V4 hypervariable regions of the 16S rRNA genes were amplified using the 341F-805R primer (5-CCT ACG GGN GGC WGC AG-3 and 5-GAC TAC HVG GGT ATC TAA TCC-3) with the barcode. The PCR products were purified by 2% agarose gel electrophoresis, and the QIAquick Gel Extraction kit (QIAGEN, Germantown, MD, USA) was used to recover DNA from the gels. The library was constructed using TruSeq Nano DNA Library Prep kit (Illumina, San Diego, CA, USA). The constructed library was then sequenced using the paired-end method on the Illumina MiSeq platform, after being subjected to DNA quantification. The sequence reads were merged using Mothur software (version 1.39.5). The sequences were assembled using the QIIME 2 software (version 2017.4) after the chimeric sequences were removed using the USEARCH software (version 7.0.1001). Operational taxonomic units (OTUs) were generated at a 97% similarity threshold, and sequences less than two were removed from all samples. The obtained sequences were aligned to the Genomes Online Database (gold.jgi.doe.gov) to determine the phylogeny of the OTUs. Alpha diversity (richness and evenness) and phylogenetic assignment were assessed using QIIME 2 software and the naïve Bayesian classification method, respectively. Principal component analysis (PCA), principal coordinate analysis (PCoA) based on the unweighted and weighted UniFrac distance matrices, and co-occurrence networks of bacterial associations, were used to visualize the differences in microbiota among the groups using the R package (version 3.5.0 and version 1.7.13). PCA was used to compare the similarities in the relative abundances of bacteria at the phylum level between the groups. PCoA was used to compare the similarities in bacterial diversity based on the distance matrices using UniFrac. Unweighted (qualitative) PCoA is based on the presence or absence of bacterial species, and weighted (quantitative) PCoA accounts for the abundance of the observed bacterial species.

Statistical Analysis

Data were analyzed using one-way ANOVA through the general linear model procedure in the Statistical Analysis System Institute software (version 9.4, 2012; SAS Institute, Cary, NC, USA). Individual birds were considered experimental units. Statistical significance was set at P<0.05. If significant effects were found, individual means were compared using Tukey's honestly significant difference test. Linear and quadratic contrasts were used to determine the effects of different concentrations of BSFP on broilers. PCoA was performed based on UniFrac distances coupled with standard multivariate statistics. The relationship between abun dant genera in broilers from different groups was analyzed using Pearson's correlation coefficient (r).

Results

Effects of BSFP on Growth Performance

The broilers were healthy during the experimental period. The effect of BSFP on the growth performance of broilers under DSS challenge is shown in Table 2. DSS challenge reduced the body weight at 7 and 14 days of age compared to the control group (P<0.05). There were no significant differences in body weight between the D and BSFP (1 and 3 g/kg) groups over the experimental period. Linear improvements in body weight were observed based on the inclusion level of BSFP under DSS challenge at 7 days of age (P<0.05). DSS challenge reduced the average daily gain at 1 to 7 days and 1 to 14 days, compared with the C group (P<0.05). No significant difference in the average daily gain was observed between the D and BSFP (1 and 3 g/kg) groups. Linear improvements in average daily gain were observed based on the inclusion level of BSFP under DSS challenge at 1 to 7 days (P<0.05). DSS challenge reduced the average daily feed intake at 1 to 7 days, compared to the C group (P<0.05). BSFP (1 g/kg) in combination with DSS challenge increased the average daily feed intake at 8 to 14 days and 1 to 14 days, compared with the C and D groups (P<0.05). There was a quadratic effect of BSFP inclusion on average daily feed intake at 8 to 14 days and 1 to 14 days (P<0.05). BSFP supplementation (1 g/kg) in combination with DSS challenge increased the feed conversion ratio at 8 to 14 days, compared with the C and D groups (P<0.05). BSFP supplementation (1 and 3 g/kg) in combination with DSS challenge increased the feed conversion ratio at 1 to 14 days, compared with the C group (P<0.05). The feed conversion ratio at 8 to 14 days and 1 to 14 days also showed quadratic effects as the inclusion level of BSFP increased in response to DSS challenge (P<0.05).

Table 2.

Effects of B. Subtilis-Fermented Products on the Growth Performance of Broilers under Dextran Sulfate Sodium Challenge

							P value2
	C1	D	LD	HD	SEM	P value	Linear	Quadratic
Body weight (g/bird)
1 d	44.3	43.9	43.8	43.7	0.43	0.797	0.876	0.952
7 d	154.5a	117.1b	130.3ab	141.5ab	4.37	0.011	0.032	0.589
14 d	368.6a	278.7b	309.3ab	315.0ab	10.26	0.014	0.225	0.439
Average daily gain (g/d/bird)
1–7 d	15.7a	10.5b	12.4ab	14.0ab	0.62	0.011	0.033	0.588
8–14 d	30.6	23.1	25.6	24.8	1.03	0.054	0.661	0.476
1–14 d	23.2a	16.8b	19.0ab	19.4ab	0.72	0.014	0.217	0.433
Average daily feed intake (g/d/bird)
1–7 d	13.7a	10.0b	12.6ab	12.4ab	0.49	0.044	0.100	0.082
8–14 d	36.1b	30.9b	49.7a	39.6ab	1.89	0.001	0.315	<0.001
1–14 d	26.0b	24.7b	34.3a	29.4ab	0.98	<0.001	0.274	<0.001
Feed conversion ratio
1–7 d	0.9	1.0	1.1	0.9	0.03	0.150	0.449	0.144
8–14 d	1.2b	1.3b	2.0a	1.7ab	0.1	0.009	0.297	0.027
1–14 d	1.1b	1.5ab	1.9a	1.6a	0.1	0.002	0.951	0.026

C=control broilers without BSFP supplementation and DSS challenge; D=DSS-challenged broilers; LD=1 g/kg BSFP-treated DSS-challenged broilers; HD=3 g/kg BSFP-treated DSS-challenged broilers

Data was analyzed using the results of D, LD, and HD groups.

Means of a row with no common superscript are significantly different (P<0.05).

Effects of BSFP on Intestinal Morphology

No significant differences in intestinal morphology were observed among the groups (Table 3). There was a quadratic effect of BSFP inclusion on villus length in the duodenum of DSS-challenged broilers (P<0.05).

Table 3.

Effects of B. Subtilis-Fermented Products on the Small Intestine Morphology of Broilers under Dextran Sulfate Sodium Challenge

								P value2
		C1	D	LD	HD	SEM	P value	Linear	Quadratic
	Villus length (µm)	1083.3	959.4	1234.0	1100.8	40.25	0.105	0.456	0.014
Duodenum	Crypt depth (µm)	116.7	117.8	112.5	125.4	7.50	0.379	0.722	0.757
	Villus length: crypt depth	9.4	8.4	11.4	9.4	0.70	0.588	0.860	0.214
	Villus length (µm)	434.4	473.4	415.8	573.6	32.26	0.945	0.263	0.341
Jejunum	Crypt depth (µm)	80.2	68.8	84.7	75.9	4.85	0.705	0.787	0.355
	Villus length: crypt depth	5.5	7.1	5.6	7.7	0.58	0.593	0.639	0.338
	Villus length (µm)	470.1	395.3	393.3	400.9	19.95	0.228	0.918	0.949
Ileum	Crypt depth (µm)	102.8	81.4	95.5	105.7	6.30	0.345	0.261	0.746
	Villus length: crypt depth	4.6	5.4	4.1	4.0	0.31	0.444	0.209	0.388

C=control broilers without BSFP supplementation and DSS challenge; D=DSS-challenged broilers; LD=1 g/kg BSFP-treated DSS-challenged broilers; HD=3 g/kg BSFP-treated DSS-challenged broilers

Data was analyzed using the results of D, LD, and HD groups.

Effects of BSFP on Intestinal Inflammatory Gene Expression

The effects of BSFP on intestinal inflammatory gene expression in broilers are shown in Table 4. The cox2, il-1β, and il-6 mRNA levels in the duodenum increased (P<0.05) in the D group, whereas BSFP supplementation (1 and 3 g/kg) reduced them. A linear and quadratic decrease in the cox2, il-1β, and il-6 mRNA levels in the duodenum of DSS-challenged broilers was observed as the inclusion level of BSFP increased (P<0.05). DSS challenge increased the cox2, iNOS, il-1β, and il-6 mRNA levels in the jejunum (P<0.05), whereas BSFP supplementation (1 and 3 g/kg) decreased them. A quadratic decrease in cox2 and iNOS mRNA levels in the jejunum of DSS-challenged broilers was observed as the inclusion level of BSFP increased (P<0.05). The il-1β and il-6 mRNA levels in the jejunum of DSS-challenged broilers decreased in the BSFP-supplemented groups (1 and 3 g/kg), showing both linear and quadratic effects (P<0.05). Similar to the jejunum, the expression of inflammatory genes in the ileum also increased (P<0.05) in the D group, whereas BSFP supplementation (1 and 3 g/kg) reduced it. A linear and quadratic decrease in the iNOS and il-1β mRNA levels in the ileum of DSS-challenged broilers was observed as the inclusion level of BSFP increased (P<0.05).

Table 4.

Effects of B. Subtilis-Fermented Products on the Intestinal Inflammatory Gene Expression of Broilers under Dextran Sulfate Sodium Challenge

								P value2
		C1	D	LD	HD	SEM	P value	Linear	Quadratic
Duodenum	cox2	0.9b	8.5a	2.6b	1.9b	0.91	<0.001	0.010	<0.001
	inos	0.8	5.7	1.8	1.3	0.72	0.050	0.063	0.064
	il-1β	0.9c	5.4a	3.1b	0.8c	0.58	<0.001	<0.001	0.008
	il-6	0.9b	3.8a	1.9b	0.6b	0.40	<0.001	0.001	0.020
Jejunum	cox2	0.9b	5.2a	0.9b	1.3b	0.57	0.002	0.059	0.001
	inos	0.9b	4.7a	0.8b	1.4b	0.51	0.002	0.082	0.002
	il-1β	1.2b	6.1a	1.2b	0.8b	0.72	0.003	0.024	0.003
	il-6	1.0c	4.4a	2.4b	1.5c	0.41	<0.001	0.001	0.006
Ileum	cox2	0.9b	4.0a	1.8b	0.5b	0.44	0.002	0.002	0.089
	inos	1.3b	3.6a	1.4b	0.4b	0.38	0.006	0.002	0.032
	il-1β	0.9b	6.5a	2.0b	1.4b	0.71	<0.001	0.014	0.002
	il-6	0.8b	7.2a	2.4ab	1.5ab	0.97	0.031	0.076	0.106

C=control broilers without BSFP supplementation and DSS challenge; D=DSS-challenged broilers; LD=1 g/kg BSFP-treated DSS-challenged broilers; HD=3 g/kg BSFP-treated DSS-challenged broilers

Data was analyzed using the results of D, LD, and HD groups.

Means of a row with no common superscript are significantly different (P<0.05).

Effects of BSFP on Cecal Bacterial Diversity

The effects of BSFP on the cecal microbiota of broilers in response to DSS challenge are shown in Table 5. The cecal species richness (Chao1 and Fisher alpha) was reduced in the D and LD groups (P<0.05). A linear increase in cecal species richness of DSS-challenged broilers was observed as the inclusion level of BSFP increased (P<0.05). The cecal species evenness (Shannon and Enspie alpha) increased in the HD group (P<0.05). A linear increase in cecal species evenness in DSS-challenged broilers was observed as the inclusion level of BSFP increased (P<0.05). Principal component analysis (PCA) revealed that OTU composition among the groups was not well separated (Fig. 1A). Unweighted principal coordinate analysis (PCoA) revealed two major clusters: samples from C, D, and LD groups were clustered together in one group, whereas samples from HD group were clustered in another group (Fig. 1B). Weighted principal coordinate analysis (PCoA) also revealed two major clustering groups: samples from D, LD, and HD groups were clustered together in one group, while samples from C group were clustered in another group (Fig. 1C). A heatmap of the 35 most abundant genera in the cecal digesta is shown in Fig. 2. Differential bacterial community composition was observed between C D, LD, and HD groups, such as with respect to genera Ruminococcaceae_unclassified, Escherichia-Shigella, Terrisporo-bacter, Clostridium sensu stricto 1, and Entero coccus. A unique bacterial cluster, containing Sellimonas, Lactobacillus, and Merdibacter, was more abundant in the D group, whereas another bacterial cluster, containing Butyric icoccus, was less abundant in the D group than in the other groups. Some genera (Oscillibacter, Blautia, and [Rumino coccus] torques group) were more abundant in the C group, whereas their abundance was reduced in the HD group. Similar bacterial community clusters, such as a cluster con taining Lachnospiraceae_unclassified and Ruminiclostridium 5, were observed between the C and LD groups. Out of all four groups, microbial cooccurrence networks in the LD group were the most complicated (Fig. 3). In contrast, mi crobial co-occurrence networks were the simplest in the HD group (Fig. 3).

Table 5.

Effects of B. Subtilis-Fermented Products on Bacterial Alpha Diversity in the Cecal Digesta of Broilers under Dextran Sulfate Sodium Challenge

							P value2
	C1	D	LD	HD	SEM	P value	Linear	Quadratic
Chao1	48.8a	37.5c	41.3bc	46.0ab	1.23	<0.001	<0.001	0.491
Fisher alpha	5.1a	3.8c	4.3bc	4.8ab	0.78	<0.001	<0.001	0.213
Shannon	2.6bc	2.4c	2.7b	3.1a	0.05	<0.001	<0.001	0.402
Enspie	3.4b	3.5b	4.4a	4.8a	0.16	0.001	0.009	0.147

C=control broilers without BSFP supplementation and DSS challenge; D=DSS-challenged broilers; LD=1 g/kg BSFP-treated DSS-challenged broilers; HD=3 g/kg BSFP-treated DSS-challenged broilers

Data was analyzed using the results of D, LD, and HD groups.

Means of a row with no common superscript are significantly different (P<0.05).

Fig. 1.

Analysis of the bacterial communities of cecal digesta. (A) Principal component analysis of the cecal digesta of basal diet without treatment (C), basal diet plus dextran sodium sulfate (DSS) challenge (D), basal diet plus 1 g/kg of Bacillus subtilis-fermented products (BSFP) and DSS (LD), and basal diet plus 3 g/kg of BSFP and DSS (HD) (n=4). Principal coordinate analysis of (B) quantitative traits (unweighted UniFrac distances) and (C) qualitative traits (weighted UniFrac distances) of the cecal bacterial communities from C, D, LD, and HD (n=4).

Fig. 2.

Taxonomic composition analysis of cecal digesta. Heatmap showing the dominant 35 genera (y-axis) across different treatment groups (x-axis, basal diet without treatment (C), basal diet plus dextran sodium sulfate (DSS) challenge (D), basal diet plus 1 g/kg of Bacillus subtilis-fermented products (BSFP) and DSS challenge (LD), and basal diet plus 3 g/kg of BSFP and DSS challenge (HD), n=4).

Fig. 3.

Microbial co-occurrence networks in the cecal digesta. Co-occurrence networks showing the dominant 50 genera across different treatment groups (basal diet without treatment (C), basal diet plus dextran sodium sulfate (DSS) challenge (D), basal diet plus 1 g/kg of Bacillus subtilis-fermented products (BSFP) and DSS challenge (LD), and basal diet plus 3 g/kg of BSFP and DSS challenge (HD), n=4). The solid and dotted lines represent positive and negative correlations, respectively.

Effects of BSFP on Cecal Bacterial Taxonomic Composition

The effects of BSFP on bacterial taxonomy in the cecal digesta of broilers in response to DSS challenge is shown in Table 6. Relative to the C group, at the phylum level, the abundance of the phylum Firmicutes was reduced in the D, LD, and HD groups (P<0.05). The abundance of the phylum Proteobacteria increased in the D, LD, and HD groups, compared with the C group (P<0.05). At the genus level, the abundance of the genus Ruminococcus torques group in the D, LD, and HD groups was lower than that in the C group (P<0.05). BSFP supplementation at 3 g/kg further reduced the abundance of the genus Ruminococcus torques group in DSS challenged broilers, compared with the D and LD groups (P<0.05). A linear decrease in the abundance of the genus Ruminococcus torques in DSS-challenged broilers was observed as the inclusion level of BSFP increased (P<0.05). Relative to the C group, the abundance of the genus Escherichia-Shigella in the D, LD, and HD groups was higher than that in the C group (P<0.05). The abundance of the genera Lachnospiraceae_unclassified and Butyricicoccus in the D and HD groups decreased compared with other groups (P<0.05). In contrast, the abundance of the genus Clostridium sensu stricto 1 in the D and HD groups was higher than in the other groups (P<0.05). The abundance of the genus Rumino coccaceae_unclassified in the HD group was higher than that in the other groups (P<0.05). A linear and quadratic increase in the abundance of the genus Ruminococcaceae_unclassified in DSS-challenged broilers was observed as the inclusion level of BSFP increased (P<0.05). Relative to the C group, the abundance of the genus Lactobacillus in the D group increased (P<0.05). BSFP supplementation (1 and 3 g/kg) in combination with DSS challenge reduced the abundance of the genus Lactobacillus compared with the other groups (P<0.05). A linear decrease in the abundance of the genus Lactobacillus in DSS-challenged broilers was observed as the inclusion level of BSFP increased (P<0.05). The abundances of the genera Enterococcus and Sellimonas in the D group were higher than those in the other groups (P<0.05). A linear and quadratic decrease in the abundance of the genus Sellimonas in DSS-challenged broilers was observed as the inclusion level of BSFP increased (P<0.05).

Table 6.

Bacterial Taxonomy within the Cecal Digesta Of Broilers

	Relative abundance (%)						P value2
	C1	D	LD	HD	SEM	P value	Linear	Quadratic
Phylum
Firmicutes	95.8a	83.0b	88.3b	83.6b	1.47	<0.001	0.870	0.751
Proteobacteria	1.5b	17.0a	11.7a	16.4a	1.72	<0.001	0.858	0.764
Genus
Lachnospiraceae_unclassified	46.0a	25.5b	37.7a	21.3b	2.73	<0.001	0.261	0.370
Ruminococcaceae_unclassified	3.3b	12.0b	15.2b	30.4a	2.76	<0.001	<0.001	0.019
Ruminococcus torques group	23.6a	12.4b	14.7b	6.4c	1.65	<0.001	0.008	0.098
Escherichia-Shigella	1.4b	15.7a	11.2a	15.1a	1.60	<0.001	0.895	0.696
Clostridium sensu stricto 1	0.5b	8.8a	2.3b	10.8a	1.21	<0.001	0.262	0.293
Lactobacillus	3.7b	8.2a	0.7c	0.8c	0.80	<0.001	0.007	0.053
Enterococcus	0.4b	3.9a	2.3ab	1.7ab	0.45	0.041	0.087	0.075
Butyricicoccus	2.3ab	0.5c	3.6a	1.4bc	0.32	<0.001	0.731	0.545
Sellimonas	1.3b	2.7a	1.7b	1.4b	0.16	<0.001	0.002	0.009
Erysipelatoclostridium	0.8	1.1	1.5	2.0	0.24	0.329	0.251	0.390

C=control broilers without BSFP supplementation and DSS challenge; D=DSS-challenged broilers; LD=1 g/kg BSFP-treated DSS-challenged broilers; HD=3 g/kg BSFP-treated DSS-challenged broilers

Data was analyzed using the results of D, LD, and HD groups.

Means of a row with no common superscript are significantly different (P<0.05).

Correlation Analysis of Bacterial Abundance Among the Dominant 10 Genera

The abundances of the genera Butyricicoccus, Lachnospiraceae_unclassified, and Ruminococcus torques group were positively correlated with each other (Fig. 4). The abundance of the genus Butyricicoccus was negatively correlated with the abundances of the genera Enterococcus, Lactobacillus, and Sellimonas (Fig. 4). The abundances of the genera Enterococcus, Lactobacillus, and Sellimonas were positively correlated with each other (Fig. 4).

Fig. 4.

Correlation analysis of bacterial abundance among the dominant 10 genera. Positive correlations are shown in blue, while negative correlations are shown in red. The values from +1 to −1 indicate the strength of the association.

Discussion

The gut microbiota is strongly shaped by the host environment, and developing a healthy gut microbiota can prevent inflammation and improve the growth of broilers (Pourabedin and Zhao, 2015). Intestinal inflammation disturbs microbial communities, resulting in intestinal microbiota dysbiosis (Lupp ; Lobionda ). DSS challenge can cause intestinal inflammation, body weight loss, diarrhea, and necrotic enteritis in poultry (Menconi ; Kuttappan ; Nii ). In this study, the body weight and average daily gain were reduced in the D group, which is in agreement with a previous study (Nii ). Dietary supplementation with B. subtilis has been shown to alleviate LPS-induced intestinal immunological stress, and improve intestinal barrier gene expression in broilers (Gadde ). Furthermore, our recent study demonstrated that BSFP can improve the growth performance of broilers under LPS challenge (Chen and Yu, 2021). The body weight and average daily gain of broilers were partially improved by supplementation with BSFP in the present study. Interestingly, BSFP supplementation at 1 g/kg increased the average daily feed intake of DSS-challenged broilers at 8 to 14 days and 1 to 14 days of age, resulting in an increased feed conversion ratio. There is no clear explanation for this increase in average daily feed intake in the LD group. In this study, BSFP supplementation alleviated the expression of inflammatory genes in the small intestine of DSS-challenged broilers during the starter phase, which is in agreement with previous studies (Gadde ; Chen and Yu, 2021). Taken together, the results indicate that BSFP supplementation can partially improve body weight and average daily gain of DSS-challenged broilers. DSS-induced intestinal inflammation in broilers was normalized by BSFP supplementation. Whether the beneficial effect of BSFP on anti-inflammation and growth performance of broilers under DSS challenge is still observed after a prolonged feeding period (market age of 35 days) remains to be confirmed in the future.

The gut morphology was not improved by supplementation with BSFP in DSS-challenged broilers in the present study. The gut microbiota may play a critical role in improving the growth performance of DSS-challenged broilers. Our previous study also demonstrated that BSFP can modulate the gut microflora composition of broilers under LPS challenge (Chen and Yu, 2021). The cecal species richness and evenness were not altered in response to LPS challenge in broilers, whereas LPS challenge in combination with BSFP supplementation dose-dependently reduced cecal species richness (Chen and Yu, 2021). In contrast, cecal species richness was reduced in response to DSS challenge in broilers, whereas BSFP supplementation at 3 g/kg increased the cecal species richness of DSS-challenged broilers. Similar to the LPS challenge, the cecal species evenness was not altered in response to DSS challenge in broilers. However, BSFP supplementation (1 and 3 g/kg) increased cecal species evenness in DSS-challenged broilers. Interestingly, BSFP supplementation at 3 g/kg simplified microbial co-occurrence networks in the cecal digesta of DSS-challenged broilers. Simplified microbial occurrence networks generally indicate a low degree of interaction between beneficial and harmful bacteria. The beneficial bacterial population may be greater than the harmful bacterial population, and vice versa. In the present study, BSFP supplementation at 3 g/kg simplified microbial cooccurrence networks, and partially improved the growth performance of broilers under DSS challenge. Therefore, the simplified microbial co-occurrence networks in this group may benefit the health of broilers. Unweighted principal coordinate analysis also demonstrated distinct clusters separating the HD group from the other three groups, indicating that BSFP supplementation at 3 g/kg can significantly modify the cecal microbiota of broilers under DSS challenge. Collectively, these findings demonstrate that LPS and DSS challenge is unable to change the cecal species evenness, but differentially regulates the cecal species richness in broilers. Furthermore, BSFP supplementation differentially modulates cecal bacterial diversity in two in vivo inflammatory models in broilers. However, the detailed mechanism by which BSFP supplementation differentially modulates cecal bacterial diversity in response to LPS or DSS challenges remains to be elucidated.

The Firmicutes phylum is predominant in the gut of poultry, and is associated with the efficiency of energy harvesting in broilers (Hou ). Decreased abundance of the phylum Firmicutes has been associated with inflammatory bowel disease (Magne ). The phylum Proteobacteria contains many opportunistic pathogens, and an increase in its abundance has been found to be correlated to gut dysbiosis (Shin ). In this study, the abundance of the phylum Firmicutes was decreased, and the abundance of the phylum Proteobacteria was increased, in the D, LD, and HD groups. These results indicate that DSS challenge can change the balance of cecal microbiota composition at the phylum level in broilers. At the genus level, the abundance of the genus Escherichia-Shigella increased in the D, LD, and HD groups. The genus Escherichia-Shigella consists of gram-negative pathogenic bacteria that have been reported to be negatively associated with health and growth performance in broilers (Shi ; Han ). In contrast, the abundance of the genus Ruminococcus torques group decreased in the D, LD, and HD groups. The bacteria of genus Ruminococcus torques group are able to degrade mucin in the gastrointestinal tract, decreasing gut barrier integrity (Malinen ; De Cesare ). The abundance of the genus Escherichia-Shigella in cecal digesta of broilers was negatively correlated with the abundance of the genus Ruminococcus torques group in this study, indicating that bacteria of genus Escherichia-Shigella may inhibit the growth of bacteria of genus Ruminococcus torques group. It has been demonstrated that bacteria of the family Ruminococcaceae are involved in short-chain fatty acid synthesis, and short-chain fatty acids can be used as an energy and carbon source for broilers (van Der Wielen et al., 2000; Xie ). In this study, the abundance of the genus Ruminococcaceae_unclassified specifically increased in the HD group. Whether BSFP supplementation at 3 g/kg can promote short-chain fatty acid production in cecal digesta remains to be confirmed in the future. Bacteria of the genus Sellimonas are considered a potential biomarker of homeostasis gut recovery (Muñoz ). In the present study, the abundance of the genus Sellimonas specifically increased in the D group, but fell back to normal levels when BSFP (1 and 3 g/kg) was supplied in the diet. Bacteria of the genus Enterococcus are opportunistic pathogens, and are resistant to several antimicrobials in poultry (d'Azevedo ; Medina Fernández ). Furthermore, Bacillus strains have been shown to be effective at inhibiting pathogenic Enterococcus species in vitro (Medina Fernández ). In this study, the abundance of genus Enterococcus was partially reduced under DSS challenge in combination with BSFP supplementation, which is in agreement with a previous study (Medina Fernández ). The genus Butyricicoccus consists of butyric acid-producing bacteria with probiotic potential in broilers (Eeckhaut ). Butyric acid produced by bacteria of the genus Butyricicoccus has been reported to inhibit the growth of pathogenic bacteria such as Salmonella and Clostridium perfringens in broilers (Timbermont ). The abundance of the genus Butyricicoccus was specifically reduced in the D group, but returned back to normal levels when BSFP (1 g/kg) was supplied in the diet in the present study. This finding is in agreement with the results of Keerqin , who observed that Bacillus subtilis supplementation increased the abundance of the genus Butyricicoccus in the cecal digesta, and improved the health of broiler chickens under necrotic enteritis conditions. Taken together, these results indicate that BSFP supplementation can restore gut microbiota dysbiosis in broilers during the starter phase under DSS challenge.

It has been reported that B. subtilis-derived antimicrobial cyclic lipopeptides are effective at inhibiting LPS-induced inflammation in vitro (Zhang ). B. subtilis-derived antimicrobial cyclic lipopeptides also exhibit antibacterial activity against common enteric pathogens, such as C. perfringens, Brachyspira hyodysenteriae, and Eimeria tenella (Horng ; Yu ). B. subtilis spores are able to germinate in the gastrointestinal tract of broilers, exerting immunomodulatory and health-promoting effects through the production of antimicrobial cyclic lipopeptides, and competitive exclusion of pathogens (Latorre ; Gadde ). BSFP contain both B. subtilis spores and its antimicrobial cyclic lipopeptides, which may contribute to the protective effects on broilers under DSS challenge. Thus, the potential mechanisms proposed on the basis of this study are as follows: (1) B. subtilis-derived antimicrobial cyclic lipopeptides promote immunomodulation in the gut mucosal immune system, (2) B. subtilis-derived antimicrobial cyclic lipopeptides inhibit enteric pathogen growth, and (3) B. subtilis spores regulate gut microbiota by competitive exclusion or production of antimicrobial lipopeptides.

In conclusion, we demonstrated for the first time that BSFP supplementation partially improved the body weight of DSS-challenged broilers at 7 and 14 days of age. Broiler body weight at 7 days of age and average daily gain at 1 to 7 days of age increased linearly with increasing BSFP levels. BSFP supplementation dose-dependently normalized intestinal inflammation and cecal bacterial diversity in broilers under DSS challenge. Whether altered gut microbiota caused by BSFP has a direct positive effect on health and growth remains to be confirmed in the future.