Introduction

The enormous amount utilization of antibiotics leads to heightened possibilities for the existence of residual antibiotics in fishery commodities [1,2,3]. Consequently, the conventional aquaculture approach reliant on antibiotics is poised to be eradicated as the contemporary aquatic industry progresses. Probiotics have been chosen and tapped as the preferred eco-friendly prophylactic approach as an alternative to antibiotics for the improvement of disease control and aquatic production [4,5,6]. However, the challenge of active quality control during processing, transportation, and preservation of probiotics has emerged as a major obstacle to their development [7].

Postbiotics, defined as “reagent of inanimate microorganisms and/or their components that is beneficial to the health of host” by The International Scientific Association for Probiotics and Prebiotics, receive extensive attention recently [8]. Postbiotics have been found to play a crucial role in preserving intestinal homeostasis, mitigating intestinal inflammation, and boosting immunity in aquatic animals [9]. Following the dietary administration of peptidoglycan in Oncorhynchus mykiss, Paralichthys olivaceus, and Apostichopus japonicus [10,11,12] and lipoteichoic acid in Epinephelus coioides and Pampus argenteus, enhanced immune function of these reagents has been demonstrated [13]. It is generally acknowledged that the microbial-associated molecular patterns (MAMPs), such as peptidoglycan and lipoteichoic acid from commensal microorganisms/probiotics, can be recognized by pattern recognition receptor (PPR), which would facilitate to maintain the homeostasis of commensal microbial community while limiting the invasion of pathogenic bacteria [14].

Bacillus pumilus SE5, which was isolated from the intestine of the grouper, showed impressively potentials in improving growth performance, feed utilization, regulating the microbiota homeostasis, and maintaining fish health [15]. In addition, dietary administration of B. pumilus SE5-derived peptidoglycan (PG) and lipoteichoic acid (LTA) alone can effectively improve the growth performance, nutrient utilization, intestinal immunity, and microbial balance of the grouper [16]. However, it is unclear whether the two cellular components (PG and LTA) have synergistic effects in improving the growth and health of fish. Therefore, we evaluated the effects of PG and LTA alone or in combination on the growth, immunity, and intestinal health in E. coioides.

Materials and Methods

Animal Ethics Statement

The experimental design and procedures in this study were reviewed and approved by the Animal Ethics Committee of Jimei University, Xiamen, China (approval number: 2011–58).

Probiotic Strain and Cell Wall Component Preparation

Probiotic B. pumilus SE5, which was isolated from the intestine of juvenile E. coioides [17], was prepared and cultured as previously described [18]. The 95 °C water bath was used to inactive the live bacterial suspension after the re-suspended cells in PBS were harvested and counted by plating on tryptone soya agar (TSA).

A TCA-based purification method was used to extract PG from B. pumilus SE5 as previously described with slightly modification [18]. Briefly, cell sediment was suspended in TCA and treated with a solvent, chloroform, and methanol. The undissolved material was agitated in Tris–HCL containing trypsin. Finally, the sediment was lyophilized after washing in sterile water.

LTA was extracted from B. pumilus SE5 as previously described [18]. Briefly, the suspension of bacteria was mixed with n-butanoland; then, the sediment was lyophilized followed by being mixed with chromatography buffer. The target product was obtained by chromatography with octyl agarose and confirmed as previously described [19].

Diet Preparation and Experimental Design

The ingredients listed in Table 1 was used to formulate the basal diet, and its proximate composition was determined in accordance with the protocols of AOAC (2005). Samples were burned in a muffle furnace to a constant weight, and the ash composition was calculated. The crude protein content was determined by Kjeldahl (N × 6.25) (Kjeltec 2200, FOSS, Denmark). The crude lipid content was determined based on the chloroform–methanol extraction method [20]. A generally recognized dose (1.0 × 108 CFU/g) of B. pumilus SE5 has verified the potential in improving the feed efficiency and immune response of E. coioides [21, 22]. The heat-inactivated SE5 with the dose of 1.0 × 108 CFU/g, the PG (21.30 mg/kg, extracted from 1.0 × 108 CFU/g SE5), LTA (6.70 mg/kg, extracted from 1.0 × 108 CFU/g SE5), the mixture of PG (10.65 mg/kg) and LTA (3.35 mg/kg), and the mixture of PG (21.30 mg/kg) and LTA (6.70 mg/kg) were supplemented to the basal diet (CG), which was named as HIB, PG, LTA, PL1, and PL2, respectively. The heat-inactivated SE5, PG, and LTA were added to diets with demand doses followed by being mixed in a dimensional drum mixer (SYH-100, Punaier Drying Equipment Co., Ltd., Changzhou, China) and then extruded to produce 5-mm pellets (CD4XITS extruder, South China University of Technology, Guangzhou, China).

Table 1 Formulation and proximate composition of the experimental diets (g/100 g dry matter)
Full size table

The animal trial was carried out in Haikang Aquaculture Research Base of Dabeinong group (Zhaoan, China). The feeding experiment was carried out in 18 recirculating tanks that held 450-L saltwater with a salinity of 30 g/L at 26 ± 2 °C. Thirty fish (14.69 ± 0.05 g) were randomly deposited into each tank. Fish in triplicate tanks fed to apparent satiation twice daily (08:30 and 18:30) for 60 days. After 24 h of fasting, fish were anesthetized with diluted eugenol (1:10,000) for sampling. To access the survival rate and growth performance (final weight, weight gain rate, specific growth rate, feed intake, and feed conversion rate) [16], fish number and weight in each tank were recorded, and then, the blood of caudal vein was collected. Thereafter, the fish was dissected under sterile conditions and tissue samples were collected, and visceral index (VSI), hepatic index (HSI), intestinal somatic index (ISI), and intestinal length index (ILI) were determined. Four intestinal samples per treatment were stored in − 80 °C refrigerator for microbial analysis, and the remaining samples were kept in − 80 °C refrigerator for digestive enzyme activity and gene expression analysis.

Digestive Enzymatic Analysis

Digestive amylase, trypsin, and lipase activities of the total intestine (n = 5) from each group was analyzed using corresponding kits (BioVision, USA) according to the manufacturer’s instructions.

Hematological Analysis

The kits of Nanjing Jiancheng Bioengineering Institute (Nanjing, China) were used to test the serum protein concentration, glutamate–oxaloacetate transaminase (GOT), glutamate-pyruvate transaminase (GPT), and alkaline phosphatase (AKP) activities. The serum complements 3 (C3) and immunoglobulin M (IgM) concentrations were determined using the enzyme-linked immunosorbent assay kits (Shanghai Jianglai Biotechnology Co., Ltd., Shanghai, China).

Intestinal Morphology Analysis

Tissues were stained with hematoxylin and eosin (HE) staining to assess morphological of the intestine samples [23]. The foregut samples were treated with ethanol and xylene treatment followed by embedded in paraffin to cutting into a 6-µm slice and then stained with HE. The morphology of the intestine was micrographed using a Leica DM5500B microscope (Germany). Measurement of muscle thickness (MT) and mucosal folding height (MFH) was accomplished by Lecia Application Suite version 4.7.0 (Leica Microsystems, Germany).

Intestinal Microbiota Analysis

Total DNA of intestinal samples was extracted and checked; then, the V3–V4 region of the 16S rRNA gene was amplified by polymerase chain reaction [24]. The llumina HiSeq platform (Beijing Biomarker Biotechnology Co., Ltd., Beijing, China) was used to perform high-throughput sequencing. Gut microbiota abundance and diversity were analyzed using BMKCloud1 as reported in our previous study [18].

Intestinal Gene Expression Analysis

Total RNA of the intestinal samples was extracted as described previously [21, 25]. Intestinal barrier-related genes (Claudin3, Caspase8, Caspase9, Occludin, and ZO-1) and immune-related genes including pattern recognition receptors (TLR1, TLR2, TLR3, TLR5, TLR22, and NOD2), NF-κB pathway-related genes (MyD88, TAK1, IKKα, IKKβ, NF-κB, BCL2, and BCLXL), MAPK pathway-related genes (ERK1, ERK2, MKK4, MKK6, and P38), antibacterial peptides (hepcidin-1, β-defensin, and epinecidin-1), inflammatory cytokine genes (IL-8, IL-12, IL-1β, TNFα, IL-10, and TGF-β1), and immune effector genes (IgM, CD4, and MHCIIα) were determined using RT-qPCR with specific primers (Table 2). The ABI 7500 real-time PCR Detection system (Applied Biosystems, California, USA) was used to perform RT-qPCR in a 20-μL reaction system utilizing SYBR® Green Pro Taq HS. β-Actin was used as the housekeeping gene. The gene expression levels were calculated using 2−ΔΔCT method.

Table 2 Real-time PCR primers used for intestinal genes of E. coioides
Full size table

Data Statistical Analysis

The results were expressed as mean ± standard error of the mean (SEM). Data were examined by one-way analysis of variance (ANOVA) using Statistical Package for Social Science (SPSS), release 19.0 (SPSS Inc., Chicago, IL, USA). Significant differences were indicated at P ≤ 0.05.

Results

Growth Performance

The significant increment of final weight, weight gain rate, and specific growth rate was observed in treatments as compared with the control (P = 0.01, P = 0.01, and P < 0.01). The feed intake in groups PL1 and PL2 were slightly higher than that of control (P > 0.05), and the feed conversion rate in groups PG, PL1, and PL2 were reduced significantly (P < 0.01). The survival rate in the groups PG, PL1, and PL2 were significantly higher than that of the control (P = 0.03, Table 3).

Table 3 Effects of B. pumilus-derived peptidoglycan and lipoteichoic acid on growth performance, feed utilization and body index of E. coioides
Full size table

The CF, HIS, and VSI were not impacted by the PG and LTA alone or in combination (P = 0.10, P = 0.35, and P = 0.28). The ISI in groups LTA, PL1, and PL2 were significantly lower than that of the control (P < 0.01).

Digestive Enzyme Activities

As shown in Table 4, the highest amylase activity was observed in group HIB, which was significantly higher than other groups (P = 0.01). The lipase activity in groups LTA and PL2 was significantly higher than that of groups HIB and control (P < 0.01). The increased protease activity was observed in groups LTA and PL1 compared with groups CG, HIB, and PL2, and the increased protease activity was observed in group PL1 as compared with group PG (P < 0.01).

Table 4 Effects of B. pumilus-derived peptidoglycan and lipoteichoic acid on intestinal digestive enzyme activity of E. coioides
Full size table

Hematological Analysis

Significantly higher alkaline phosphatase (AKP) activity was observed in group PL1 compared with group PG (P < 0.01). The decreased glutamic oxaloacetic transaminase (GOT) activity in group PL2 was observed compared with group PG (P < 0.01). The immunoglobulin M (IgM) and complement C3 contents showed no variation among groups PG, LTA, PL1, and PL2, but significantly higher than the control (P < 0.01), and the maximum values were recorded in group PL1 (Table 5).

Table 5 Effects of B. pumilus-derived peptidoglycan and lipoteichoic acid on serum biochemical indices of E. coioides
Full size table

Intestinal Morphology

The improved villus length and width were observed in groups PG, LTA, and PL1 (P < 0.01, Fig. 1 and Table 6). Increased muscle thickness was observed in each treatment in comparison to the control (P < 0.01). The slightly increased number of microvilli goblet cells was observed in all treatments in comparison to the control (P = 0.22, Fig. 2).

Fig. 1
figure 1

The HE-stained section of intestinal tract of E. coioides (4x). Note: CG group (A), HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg PG + 6.70 mg/kg LTA

Full size image
Table 6 Effects of B. pumilus-derived peptidoglycan and lipoteichoic acid on intestinal morphology of E. coioides
Full size table
Fig. 2
figure 2

The PAS-stained section of intestinal tract of E. coioides (10x). Blue bidirectional arrows indicate sarcomere thickness; black bidirectional arrows villi length; black unidirectional arrows indicate goblet cell after staining. Note: CG group (A), HIB group (B), PG group (C), LTA group (D), PL1 group (E), PL2 group (F). CG = control group; HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg PG + 6.70 mg/kg LTA

Full size image

Intestinal Microbiota

The PG and LTA treatments did not influence the OUT diversity indices (Shannon and Simpson) (P = 0.88 and P = 0.60) and the richness indexes (ACE and Chao1) (P = 0.34 and P = 0.74, Table 7). The slightly improved richness indexes (ACE and Chao1) were observed in groups PL1 and PL2 compared with groups PG and LTA. On the other hand, the three-dimensional analysis of principal component analysis (PCA) showed a considerable similarity between treatments and control (Fig. 3).

Table 7 Effects of B. pumilus-derived peptidoglycan and lipoteichoic acid on alpha diversity of intestinal microbiota of E. coioides
Full size table
Fig. 3
figure 3

Beta diversity of intestinal microbiota of E. coioides. CG = control group; HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg

Full size image

At phylum level, the most prominent bacterial phyla detected in all treatments were Firmicutes, Proteobacteria, Bacteroidetes, and Actinobacteriota (Fig. 4). Slightly increased relative abundances of Firmicutes were observed in groups LTA, PL1, and PL2 compared with that of the control (P = 0.29), while slightly decreased Proteobacteria and Actinobacteriota (P = 0.99 and P = 0.78) in groups PG, LTA, PL1, and PL2 compared with those of the control (Table 8). There was no significant variation between all groups at subject level (P > 0.05, Table 9). At genus level, Paenalcaligenes, Muribaculaceae, Corynebacterium, Bacteroides, and Staphylococcus constituted the most dominant genera. Although there was no significant alteration, relatively elevated Ligilactobacillus and Lactococcus abundances were observed in all treatments compared with the control (P = 0.46 and P = 0.90). Moreover, the relative abundances of Muribaculaceae, Corynebacterium, Lachnospiraceae, Ligilactobacillus, and Lactococcus in group PL1 were slightly higher than those of group LTA (P > 0.05, Table 10).

Fig. 4
figure 4

Taxonomy classification of reads from 16S RNA V3–V4 regions of intestinal bacteria at phylum (A), family (B), and genus (C) taxonomic levels. Only top 15 most abundant phyla and genera (based on relative abundance) were shown in the figures. Other phyla and genera were all assigned as ‘Others’. CG = control group; HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg PG + 6.70 mg/kg LTA

Full size image
Table 8 Effects of B. pumilus-derived peptidoglycan and lipoteichoic acid on relative abundance of intestinal bacteria at phylum level of E. coioides
Full size table
Table 9 Effects of B. pumilus-derived peptidoglycan and lipoteichoic acid on relative abundance of intestinal bacteria at subject level of E. coioides
Full size table
Table 10 Effects of B. pumilus-derived peptidoglycan and lipoteichoic acid on relative abundance of intestinal bacteria at genus level of E. coioides
Full size table

Relative Expression of Intestinal Physical Barrier-Related Genes

The PG and LTA alone or in combination led to a slight increment in the expression of intestinal Occludin, with the highest level observed in group PL1 (P = 0.93, Fig. 5). The expression of Claudin-3 and ZO-1 was slightly elevated in treatments (P = 0.81 and P = 0.63).

Fig. 5
figure 5

The expression of Claudin3, Occludin, and ZO-1 in the intestine of E. coioides. ZO-1 = zonula occludens protein 1; CG = control group; HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg PG + 6.70 mg/kg LTA. Bars with the same superscripts are not significantly different (P > 0.05)

Full size image

Relative Expression of Intestinal Immune-Related Genes

As shown in Fig. 6, the significant upregulation of intestinal TLR2 expression was observed in groups PL2 and PG compared to the control (P = 0.01). The expression of intestinal NOD2 in group PL1 was higher than that in groups PG and LTA and that in group PL2 was higher than that in group PG (P < 0.01). There was no significant difference in the expression of other pattern recognition receptors (TLR1, TLR3, TLR5, and TLR22) among all groups (P > 0.05). The downregulated expression of the NF-κB pathway-related genes (MyD88, TAK1, IKKa, IKKb, NF-κB, BCL-2, and BCL-XL) was observed in treatments (P > 0.05, Fig. 7). There was no significant difference in the MAPK pathway-related genes (ERK2, MKK4, MKK6, and P38) in all groups (P > 0.05, Fig. 8). The downregulated expression of ERK1 was observed in groups LTA and PL2 (P < 0.01).

Fig. 6
figure 6

The expression of TLRs (TLR1, TLR2, TLR3, TLR5, TLR22), NOD2, and MYD88 in the intestine of E. coioides. TLR1 = toll-like receptor 1; TLR2 = toll-like receptor 2; TLR5 = toll-like receptor 5; TLR22 = toll-like receptor 22; NOD2 = nucleotide-binding oligomerization domain 2; MyD88 = myeloid differentiation factor 88. CG = control group; HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg PG + 6.70 mg/kg LTA. Bars with the same superscripts are not significantly different (P > 0.05)

Full size image
Fig. 7
figure 7

The expression of NF-κB pathway (TAK1, IKKα, IKKβ, NF-κB, BCL-2, and BCL-XL) in the intestine of E. coioides. TAK1 = transforming growth factor-β-activated kinase 1; IKKα = inhibitor of κB kinase α; IKKβ = inhibitor of κB kinase β; NF-κB = nuclear factor kappa-B; Bcl-2 = B-cell lymphoma-2; Bcl-xL = B-cell lymphoma-extra large. CG = control group; HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg PG + 6.70 mg/kg LTA. Bars with the same superscripts are not significantly different (P > 0.05)

Full size image
Fig. 8
figure 8

The expression of MAPK pathway (ERK1, MKK4, MKK6, and P38) in the intestine of E. coioides. ERK1 = extracellular regulated protein kinases 1; ERK2 = extracellular regulated protein kinases 1; MKK4 = mitogen-activated protein kinase kinase4; MKK6 = mitogen-activated protein kinase kinase4. CG = control group; HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg PG + 6.70 mg/kg LTA. Bars with the same superscripts are not significantly different (P > 0.05)

Full size image

The upregulated expressions of intestinal antimicrobial peptides, which include β-defensin, epinecidin-1, and hepcidin-1, were observed in treatments (Fig. 9). The expressions of β-defensin and epinecidin-1 in group PL1 were significantly higher than those of groups PG and LTA (P < 0.01). The slightly upregulated expressions of the intestinal immunity effectors, such as IgM, CD4, and MHCIIα, were observed in treatments (P > 0.05).

Fig. 9
figure 9

The expression of antibacterial peptides (Epinecidin-1, Hepcidin-1, and β-defensin) and immunity effectors in the intestine of E. coioides. IgM = Immunoglobin M; CD4 = Cluster of differentiation 4; MHC-II = Major histocompatibility complex class II. CG = control group; HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg PG + 6.70 mg/kg LTA. Bars with the same superscripts are not significantly different (P > 0.05)

Full size image

No significant variation in the pro-inflammatory cytokine gene expression was observed among groups PG, LTA, PL1, and PL2 (P > 0.05, Fig. 10). The upregulated expressions of IL-10 and TGF-β1 were observed in group PL1 compared with groups PG and LTA (P < 0.01 and P = 0.04).

Fig. 10
figure 10

The expression of pro-inflammatory cytokines (IL-1β, IL-8, IL-12 and TNF-α) and anti-inflammatory cytokines (IL-10 and TGF-β1) in the intestine of E. coioides. IL-1β = interleukin-1β; TNF-α = tumor necrosis factor-α; IL-8 = interleukin-8; IL-12 = interleukin-12; IL-10 = interleukin-10; TGF-β1 = transforming growth factor β 1. CG = control group; HIB = control diet + 1.0 × 108 CFU/g heat-inactivated B. pumilus SE5; PG = control diet + 21.30 mg/kg PG; LTA = control diet + 6.70 mg/kg LTA; PL1 = control diet + 10.65 mg/kg PG + 3.35 mg/kg LTA; PL2 = control diet + 21.30 mg/kg PG + 6.70 mg/kg LTA. Bars with the same superscripts are not significantly different (P > 0.05)

Full size image

Discussion

As the most critical MAMPs of Gram-positive bacteria, LTA and PG have been extensively reported to have positive functions in improving the growth and general health of aquatic animals [11]. However, no information is available about their combined effects in aquaculture. The present study evaluated the effects of the mixture of PG and LTA derived from probiotic B. pumilus SE5 on the growth performance, feed utilization, immune function, and intestinal health in grouper E. coioides. Results indicated that PG and LTA alone or in combination significantly improved the SGR, FCR, and survival rate. In accordance with the present results, our previous study has shown that PG and LTA improved the growth performance and digestive enzyme activities and intestinal function in E. coioides [18]. Therefore, the improved growth performance probably attributed to better nutrient absorption caused by improved digestive enzyme activity and intestinal development [26,27,28,29,30]. In fact, the amylase and lipase activities in group PL2 were superior to that in groups PG and LTA, and the protease activity in group PL1 were superior to that in group PG. On the other hand, the microvilli length, microvilli width, muscle layer thickness, and goblet cells in group PL1 was superior to those in groups PG and LTA. Furthermore, the highest expression of tight junction (TJ) proteins Claudin3 and Occludin was observed in group PL1, facilitating the establishment of tight junctions among intestinal epithelial cells and ensuring the stability of the physical barrier [31]. Taken together, the B. pumilus SE5-derived PG and LTA demonstrated synergistic efficacy in enhancing the intestinal structural integrity and digestive enzyme activity, which may contribute to the improved growth performance and feed utilization in the grouper.

Intestinal microbiota plays a crucial role in regulating the immune system of fish, thereby boosting its resilience against pathogenic microorganisms. Concurrently, the host’s immune system actively surveils the dynamics of the microbial community to maintain the intestinal microbiota homeostasis [32]. Probiotic and postbiotics supplementation has been suggested as an effective approach to shape the intestinal microbial composition, thereby promoting the growth and overall health of fish [33]. Our previous study demonstrated that B. pumilus SE5-derived PG and LTA could effectively modulate the overall structure of intestinal microbiota of the grouper [18]. Similarly, the Lactobacillus plantarum-derived LTA has been identified to be effective in preventing diseases associated with pathogenic Staphylococcus aureus by inhibiting the expression of the ica-operon and thereby inhibiting the production of poly-N-acetylglucosamine, which is required for biofilm development in S. aureus [34]. In the present study, the mixture groups (PL1 and PL2) increased the relative abundance of common beneficial Firmicutes while pathogenic Proteobacteria were decreased [35, 36]. At genus level, the mixture increased the relative abundance of Lactobacillus and Lactococcus (both belonging to Firmicutes) while decreasing the deleterious Bacteroides and Staphylococcus. Therefore, B. pumilus SE5-derived PG and LTA mixtures presented more effective impact on the improvement of gut microbial homeostasis, but the mechanisms are largely not clear, which deserve further study.

The remodeling intestinal microbiota by B. pumilus SE5-derived PG and LTA prompted us to inquire about any potential alterations in the functioning of the immune system. In fact, multiple studies have substantiated the impact of postbiotic stimulation of the intestinal immunity [37], which relies on the presence of T-lymphocytes in the epithelium and lamina propria, alongside B-lymphocytes, plasma cells, dendritic cells, macrophages, and granulocytes [38]. There are intricate intercellular interactions involving the initiation of the response by T lymphocytes, the effects of which are amplified by the binding of MHC (MHC class I and II) to T cell co-receptors (CD4 and CD8), which are considered to protect the fish from pathogen invasion [39]. In the present study, a slightly elevated expression of MHCII and CD4 was found with administration of PG and LTA alone or in combination. IgM is a type of immunoglobulin primarily secreted by B lymphocytes [40]; its upregulation expression was observed in the grouper fed with diets supplemented with B. pumilus SE5-derived PG or LTA diets in our previous study [18]. In this study, a higher relative expression of IgM was observed in group PL1 compared to that in groups PG and LTA. On the other hand, antimicrobial peptides (AMPs) represent a category of small molecular peptide compounds endowed with dual functionality: they can directly eliminate pathogenic microorganisms, regulate immune responses, and facilitate mucosal repair [41]. The upregulated AMPs, such as epinecidin-1, hepcidin-1, and β-defensin, were observed in fish fed diets supplemented with PG and LTA alone, and the most pronounced effect was observed in group PL1. Therefore, the combination of PG and LTA could potentially activate the expression of antimicrobial effectors and IgM and thus maintain the general health of the grouper, which in turn may benefit their growth performance.

Previous research has demonstrated that pattern-recognition receptors (PPRs), such as toll-like receptors (TLRs) and nucleotide oligomerization domain (NOD)-like receptors (NLRs), are capable of recognizing microbial-associated molecular patterns (MAMPs) released by microbes/probiotics [42], and these receptors subsequently trigger signaling cascades that intricately regulate the production of antimicrobial peptides (AMPs) and other effectors [43]. Peptidoglycan of potentially pathogenic bacteria is recognized by NOD2, which in consequence activates MAPK signaling thus inducing the secretion of antimicrobial peptides in human Paneth cells [44]. Our previous study observed that B. pumilus SE5-originated PG and LTA could enhance TLR2 and NOD2 expressions in E. coioides [16]. In this study, the mixture of PG and LTA is more effective in inducing the expression of NOD2 and TLR2 in the grouper. In addition, the expression of ERK1 was downregulated in groups LTA and PL2, which in line with a previous study indicated that Lactobacillus paracasei-derived LTA could inhibit ERK signaling pathways, thereby suppressing the release of cytokines [45]. Previous studies have found that Lactobacillus plantarum and Staphylococcal-derived lipopeptide LP01 could be recognized by TLR2 receptors and stimulate MAPK signaling pathway to regulate antimicrobial peptide synthesis, respectively [46, 47]. Therefore, we postulated that either PG and LTA alone or in combination can activate the expression of TLR2 and NOD2 and stimulate the P38MAPK pathway to promote the secretion of antimicrobial peptides thus shaping the intestinal microbiota and further maintaining intestinal health in grouper, and the mixture of PG and LTA showed better potential compared to PG or LTA alone.

The regulation of intestinal microbiota and reinforcement of immune function are crucial for the prevention and treatment of intestinal inflammation [48], which is closely related to the activation of the NF-κB pathway [49]. The inhibition of NF-κB kinase and the alleviation of Shigella fowleri-induced inflammation were demonstrated with Lactobacillus plantarum-derived LTA [50]. Our results showed that PG and LTA alone or in combination likely acts to inhibit the NF-κB pathway, with downregulation of the TLR1, TLR3, TLR22, MyD88, TAK1, IKKα, IKKβ, and NF-κB genes, and the most pronounced suppression efficacy occurred in group PL2. Similarly, the L. paracasei D35-derived PG and LTA could inhibit NF-κB pathways thus alleviating inflammation [51, 52]. Pro-inflammatory cytokines such as IL-8 and TNF-α have also been described to be downregulated by Lactobacillus plantarum-derived LTA in human intestinal epithelial Caco-2 cells [53]. Furthermore, Lactobacillus acidophilus-derived LTA has been reported to downregulate IL-12 and TNF-α expressions in mice while upregulating IL-10 expression [54]. In the present study, the combination of PG and LTA could upregulate the expression of IL-10 and TGF-β1 while inhibiting IL-1β, and the highest expression of IL-10 and TGF-β1 was observed in group PL1. Taken together, the administration of the PG and LTA mixture in diet could ameliorate inflammation by inhibiting the NF-κB pathway and thus protecting the intestinal mucosal barrier and ultimately enhancing intestinal health of grouper.

In summary, compared to the individual addition of PG and LTA to feed, the mixture (10.65 mg/kg PG and 3.35 mg/kg LTA) demonstrates better efficacy in enhancing intestinal microbiota homeostasis and immune function, preventing enteritis and improving growth performance and feed utilization in E. coioides. These results provide valuable insights into the interactions between commensals and host fish and also lay the necessary foundation for the application of postbiotics in grouper aquaculture.