DOI: https://doi.org/10.21203/rs.3.rs-2158303/v1
Necrotizing enterocolitis (NEC) is a devastating gastrointestinal disease with high morbidity and mortality, affecting preterm infants especially those with very low and extremely low birth weight. β-glucan has manifested of multiple biological effects including anti-inflammatory, regulating gut microbiota and immunomodulatory activities. At present, there is no relevant study considering the effects of β-glucan on NEC. This study aimed to investigate the effects of β-glucan on NEC.
Neonatal C57BL/6 mice were randomly divided into three groups: control group, NEC group and β-glucan group. Newborn 3-day-old mice were gavage with either 1mg/ml β-glucan or PBS at 0.03 ml/g for consecutive 7 days before NEC induction and a NEC model was established with hypoxia combined with cold exposure and formula feeding. All the pups were killed after 72-hour modeling. HE staining was performed to assess the pathological injury of the intestine. The mRNA expression levels of inflammatory factors in intestinal tissues were determined by quantitative real-time PCR. The protein levels of TLR4, NF-κB and tight junctions proteins in intestinal tissues were evaluated using western blotting and immunohistochemistry. 16S rRNA sequencing was performed to determine the structure of gut microbiota.
β-glucan administration ameliorated intestinal injury of NEC mice; reduced the intestinal expression of TLR4, NF-κB, interleukin- (IL-) 1β, IL-6 and TNF-α; and increased the intestinal expression of IL-10 (P < 0.05); improved the expression of ZO-1, Occludin and Claudin-1 within intestinal barrier. Pre-treatment with β-glucan also increased the proportion of Actinobacteria, Clostridium_butyricum, Lactobacillus_johnsonii,Lactobacillus_murinus and Lachnospiraceae_bacterium_mt14 and reduced the proportion of Klebsiella_oxytoca_g__Klebsiella in the NEC model.
β-glucan intervention can prevent necrotizing enterocolitis in neonatal mice, possibly by suppressing TLR4-NF-κB signaling pathway, improving intestinal barrier function and partially regulating intestinal microbiota.
Necrotizing enterocolitis (NEC) is a devastating gastrointestinal disease with high morbidity and mortality, affecting preterm infants especially those with very low and extremely low birth weight (1, 2). NEC is characterized by acute intestinal ischemia and necrosis and a mortality of approximately 25% in most cases, however in the most severe cases it can reach as high as 80% in 48 hours after diagnosis (3). At present, therapies for NEC are limited to abrosia, haemodynamic resuscitation and the administration broad-spectrum antibiotics (4). However, for infants who continue to deteriorate rapidly, urgent surgery is required to excise necrotic bowel (5). Moreover, in recent years, the complications of NEC beyond intestine including the lung and the brain have been noticed in NEC survivors (6, 7).
Although the definite pathogenesis of NEC remains not fully understood (8), it is generally acknowledged that lipopolysaccharide receptor Toll-like receptor 4 (TLR4) plays a key role in triggering the mucosal inflammation. Activated TLR4 interacts with microbial dysbiosis and ultimately leads to NEC (9, 10). Exaggerated TLR4 signalling triggers a typical cascade that activates nuclear factor-κB (NF-κB) and induces the accumulation of proinflammatory cytokines (11). The progression of inflammatory cascade above leads to intestinal epithelial cell death by apoptosis and necroptosis, weakened mucosal reconstruction and gut barrier disruption. The subsequent increased permeability, allows more toxic substances and pathogenic microorganisms to translocate into the underlying vascular network, ultimately resulting in progression to severe NEC (12).
Currently, there is no effective treatment for NEC. Preventive strategies to reduce the severity of the disease and elucidating underlying processes that lead to the long- term complications of NEC have been gradually emphasized (13).
β-glucan is a kind of bioactive polysaccharides obtained from yeast, mushrooms, algae and cereals, such as oat β-glucan and lentinus edodes β-glucan, which is believed to be beneficial for health and is edible to be taken orally for food supplements and daily diet (14–17). It consists of β-D-glucose monomeric units linked together by glycosidic bonds at different positions, e.g (1, 3), (1, 4), or (1, 6). β-glucans and β-1, 3-D-glucans as well as β-1, 6-D-glucans derived from advanced fungi, mushrooms, molds and yeasts differ from β-1, 3-D-glucans and β-1, 4-D-glucans, which are primarily obtained from the cell walls and the seeds of some cereals (18). Currently, β-glucan has manifested of multiple biological effects including anti-tumor (19), anti-obesity (20), anti-allergy (21), anti-osteoporosis (22), anti-inflammatory (23, 24), regulating gut microbiota (25–27) and immunomodulatory activities (28, 29). Among them, fungal β-glucans have more significant immunomodulatory effects (30). However, the precise mechanism by which β-glucans inhibit the release of inflammatory cytokines and induce anti-inflammatory immune cells is complex and not fully understood. It was reported that fungal β-glucan inhibited the release of lipopolysaccharide (LPS)-induced nitric oxide and TNF-α in vitro and reduced the secretion of tumor necrosis factor (TNF)-α and interleukin (IL)-6 in vivo (31, 32). Monocytes isolated from β-glucan-treated mice released less TNF-α and IL-6 after stimulation (33). In addition, β-(1, 3)/(1, 6)-glucan attenuates the increased expression of proinflammatory cytokines TNF-α and IL-1β via TLR4/MyD88/NF-κB signaling pathway (34, 35).
At present, there is no relevant study considering the effects of β-glucan on NEC. We hypothesized that oral β-glucan supplementation could inhibit inflammation, improve intestinal barrier and modulate gut microbiota to protect against NEC by inhibiting TLR4-NF-κB signaling pathway. In order to elucidate the preventive effect of β-glucan on NEC, we established a neonatal mouse model of NEC and gavage with β-glucan before the establishment of NEC.
β-glucan and its preparation
The β-glucan used in this study was β-glucan peptide, a high molecular weight polysaccharide extracted from the fungus Trametes versicolor (Invivogen, category code: tlrl-bgp). The β-glucan consists of a highly ramified glucan portion, including a β-(1, 4) main chain and β-(1, 3) side chain, with β-(1, 6) side chains covalently linked to a polypeptide portion rich in aspartic, glutamic and other amino acids. The main structure of β-glucan is shown in Fig. 1B. β-glucan was diluted in sterile PBS to a concentration of 1 mg/ml.
NEC induction and drug treatment
All operations performed in our experiment were approved by the Animal Ethics Committee at Chongqing Medical University. Following protocol previously described (36), 10-day-old C57BL/6 mice were separated from their mothers and were fed by gavage with hyperosmolar formula (Similac Advance (Abbott Nutrition, USA)/Esbilac puppy milk replacer (PetAg, USA) = 1.7) every 4 h, subjected to hypoxia and hypothermia (100% N2 for 90 s subsequently with 4 ℃ for 10 minutes, 3 times per day) for 3 days to induce NEC. In our study, newborn 3-day-old mice were gavage with either 1mg/ml β-glucan or PBS at 0.03 ml/g for consecutive 7 days before NEC induction. In addition, age-matched and untreated mice were left with their mothers as control group. Body weight and survival condition were recorded daily throughout the establishment of NEC. On postnatal day 13, mice were sacrificed by cervical decapitation and intestine were harvested for further analysis.
Gut histology
The intestines were completely removed and a 1-cm portion of the distal ileum was fixed in 4% paraformaldehyde solution overnight. Then, the samples were dehydrated, embedded in paraffin and cut into 4µm sections. Subsequently, 4-µm tissue sections were stained with hematoxylin and eosin, and the histological injury was assessed by an established scoring criteria (37) in a double-blinded manner as follow: 0: no damage; 1: epithelial cell lifting or separation; 2: necrosis to the mid-villous level; 3: necrosis of the entire villus; 4: transmural necrosis. Animals with a histologic tissue injury score ≥ 2 were considered positive for NEC.
Immunohistochemistry
Paraffin sections from three groups were deparaffinized in xylene and rehydrated with decreasing concentrations of ethanol, followed by antigen retrieval with citric acid antigen repair buffer (PH6.0) in a microwave oven. The slides were placed in a wet box and incubated at 4°C overnight with the following primary antibodies diluted in PBS (PH7.4) respectively against primary antibodies. After slightly shaken dry, the slides were incubated with anti-rabbit antibody (HRP labeled) at room temperature for 50min. Finally, slides were stained with DAB, and nuclei were stained using hematoxylin. After incubating with the appropriate secondary antibody for 60 min, slides were stained with DAB, and nuclei were stained using hematoxylin for about 1min. Images were collected under microscope after dehydration and sealing.
Real-time PCR
Total RNA was extracted from the intestine tissue using TRIzol (Life Technologies CA, USA). The purity of RNA was quantified using a nanodrop spectrophotometer (Thermo Fisher Scientific, CA, USA) and eligible RNA samples (OD260/280 = 1.8–2.2, OD260/230 ≥ 2.0) were used. cDNAs were synthesized using a Prime Script RT Reagent (Takara, Japan) and were used for qRT-PCR assay using TB Green Premix Ex Taq II (Tli RNase H Plus) Kit (Takara, Japan). β-actin was used as an internal control, and the relative expression of mRNA (TLR4, IL-1β, IL-6, IL-10 and TNF-α) in intestine tissue was determined using ΔΔCT method. The detailed information of RT-PCR primer sequences is shown in Table 1.
Western blotting
Intestine tissue soaked in RIPA lysis buffer (Beyotime, China) supplemented with protease inhibitor (Beyotime, China) was homogenized using an electric homogenizer and centrifuged to obtain the supernatant. Protein concentrations were measured by Pierce BCA Protein Assay Kit (Beyotime, China). The protein supernatant was mixed with sodium dodecyl sulfate sample buffer (Beyotime, China) at a ratio of 4:1 and denatured in 100℃ for 10 minutes. Protein samples were separated in 10% polyacrylamide gels and transferred to 0.45 µm PVDF membranes, and measured using anti-TLR4 (Servicebio, China), anti-NF-κB P65 (Servicebio, China), anti-ZO-1, anti-Occludin, Claudin-1 (Proteintech, China) and β-actin (ZENBIO Biotechnology, China) at 4℃ overinight. Signals were detected using chemiluminescence (ECL Western Blotting Substrate, Bio-Rad). The relative intensity of target bands was quantified by Image J analysis system (Bio-Rad).
Fecal Sample Collection and Microbiota Analysis
The ileum and colon feces of mice were collected in 1.5 ml sterile tubes and immediately placed in liquid nitrogen and transferred to -80℃ refrigerator for fecal sample microbiota analysis. Total genomic DNA was extracted from fecal samples using the E.Z.N.A.® soil DNA Kit (Omega Bio-tek, Norcross, GA, U.S.) according to manufacturer’s instructions. The quality and concentration of DNA were measured using 1% agarose gel electrophoresis and a NanoDrop® ND-2000 spectrophotometer (Thermo Scientific Inc., USA). The hypervariable region V3-V4 of the bacterial 16S rRNA gene were amplified with primer pairs 338F (5'-ACTCCTACGGGAGGCAGCAG-3') and 806R(5'-GGACTACHVGGGTWTCTAAT-3') by an ABI GeneAmp® 9700 PCR thermocycler (ABI, CA, USA). PCR amplification cycling conditions were as follows: initial denaturation at 95 ℃ (3 min), 27 cycles of denaturing at 95 ℃ (30 s), annealing at 55 ℃ (30 s) and extension at 72 ℃ (45 s), and single extension at 72 ℃ (10 min), and end at 4 ℃. All samples were amplified in triplicate. The PCR product was extracted from 2% agarose gel, purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA) according to manufacturer’s instructions and quantified using Quantus™ Fluorometer (Promega, USA). Purified amplicons were sequenced on an Illumina MiSeq PE300 platform/NovaSeq PE250 platform (Illumina, San Diego,USA) and the data were deposited into database. The original data were analyzed. Briefly, reads containing bases with a quality score < 20 were truncated, and sequences longer than 10 bp were combined together with their overlapped sequence. Reads that exceed the maximum mismatch ratio of 0.2 were discarded. Using UPARSE 7.1, the optimal sequences were clustered into operational taxonomic units (OTUs) which 97% sequence were in the similarity level. The most abundant sequence was selected as a representative sequence from each OUT. To minimize the effects of sequencing depth on alpha and beta diversity measure, the number of 16S rRNA gene sequences from each sample were rarefied to 27555. Bioinformatic analysis of the gut microbiota was conducted using the Majorbio Cloud platform (https://cloud.majorbio.com). Using Mothur software (http://www.mothur.org/wiki/Calculators), Alpha diversity, Chao1 richness and Shannon index were figured out. Differences of Alpha diversity among groups were analyzed using Wilcoxon rank sum test. The similarity among the microbial communities in different samples was measured by principal coordinate analysis (PCoA) based on Bray-curtis dissimilarity and the PERMANOVA test was applied to assess whether the variation could be explained by the treatment accompanied with its statistical significance. The linear discriminant analysis (LDA) effect size (LEfSe) (LDA score > 2, P < 0.05)was used to identify the significantly different taxa (phylum to genera) of bacteria among the different groups.
Molecular docking
Molecular docking was used to identify the binding mode between the β-glucan and the TLR4 using AutoDock4.2. The structure of TLR4 (PDB ID: 3FXI) was downloaded from Research Collaboratory for Structural Bioinformatics Protein Data Bank (38). (RCSB PDB, RRID:SCR_012820) (http://www.rcsb.org/pdb/). The 3D structure of β-glucan peptide was drawn by RDKit. The protein Amber14SB charge and the protonation state were allocated respectively using UCSF Chimera software and H++ (39, 40), and the structure was optimized using the classical MMFF94 force field. The optimized molecules were employed for AM1-BCC local charge calculation with UCSF Chimera software. The geometric center of the binding site that was predicted by SiteMap was applied as the docking center. The docking centre of the TLR4 was identified as center_x: -7.88, center_y: −13.00, and center_z: 45.55. The docking calculation was limited to the rectangular box with the center of each protein docking and the side length was 22.5 Å, and the Spacing step was set to 0.375 Å. The maximum number of search conformations was set to 10000. Amino acids in the docking center as well as ligands were regarded as flexible objects, and the outside amino acids were regarded as rigid objects, allowing amino acid side chains, such as aspartic acid and tryptophan, to flip over. Semi-flexible docking method was carried out to docking and genetic algorithm was used for conformational sampling and scoring.
Statistical analysis
Data analysis was performed by the GraphPad Prism (version 9.3.0). Normally distributed data were expressed as the mean ± SD and significance was identified by One-way ANOVA. Median and interquartile range (IQR) were used to describe nonnormally distributed data, and differences were determined by the Kruskal-Wallis test. P < 0.05 was considered statistically significant.
General conditions such as vitality, weight gain, hair luster, subcutaneous fat differ no obvious differences among three groups before modeling. However, mice in the NEC group began to develop abdominal distension, low vitality, obvious weight loss, diarrhea and hematochezia during modeling, whereas β-glucan group showed abdominal distension, decreased activity but without diarrhea and hematochezia (Fig. 1A). The body weight of the NEC group mice decreased more than that of β-glucan group mice on the 2nd day of modeling (P < 0.0001). Before sacrifice, body weight was significantly higher in the β-glucan group than in the NEC group (P < 0.0001) (Fig. 1D). No deaths were observed in the three groups before modeling. As shown in Fig. 1C, we did not observe deaths in the Control group during the 3 days of modeling. In the NEC group, three deaths occurred on the 1st day, and five deaths occurred on the 2nd and four on the 3rd days. In the β-glucan group, three deaths occurred on the 2nd day, and zero on the 3rd day. The final survival rate differs significantly among the three groups: 70% (28/40) in NEC group, 92.5% (37/40) in β-glucan group, and 100% in the Control group (P < 0.001).
No obvious damage was observed in the Control group (Fig. 2A) but gas accumulation and droplet-like changes in the intestinal tissue in the NEC group (Fig. 2B). Only slight gas accumulation and edema were observed in the β-glucan group (Fig. 2C). Based on the observation under a light microscope, a complete intestinal tissue structure, with orderly arranged villi without edema and patchy necrosis, and a thick muscle layer without separation from lamina propria, was observed in the control group (Fig. 2D). In the NEC group, the villi were disrupted, patchy necrotic, exfoliated, or even disappeared, and the muscle layer was thin or even fractured (Fig. 2E). In the β-glucan group, the villi were relatively complete with slight edema, and the muscle layer was thicker than in the NEC group. No obvious necrosis or shedding of villi was observed (Fig. 2F). The median intestinal histological pathological score showed that there were statistically significant differences among three groups (P < 0.001), with score in the β-glucan group was significantly lower than that in the NEC group (P = 0.0226) (Fig. 2G).
To investigate the effect of β-glucan on inflammatory cytokines in the intestines of NEC, qRT-PCR was employed to detect the mRNA expression of these cytokines. We found that the secretion of proinflammatory cytokines IL-1β, IL-6 and TNF-α were significantly increased in the NEC group compared with the control and β-glucan group (Figs. 3A,B,D). Meanwhile, the secretion of anti-inflammatory cytokine IL-10 in the NEC group was markedly reduced compared with the control and β-glucan group (Fig. 3C).
TLR4, widely expressed in intestinal epithelial cells and lymphocytes, plays a key role in the pathogenesis of NEC (10). When activated by pathogenic microorganisms, TLR4 triggers the innate immune response and subsequently the downstream NF-κB signaling pathway, ultimately mediates the expression of proinflammatory factors IL-1, IL-6, and TNF-α (41). To determine the impact of β-glucan on the TLR4-NF-κB intestinal signaling pathway in NEC mice, we assessed TLR4 mRNA expression by QPCR and TLR4 and NF-κB protein expression by western blotting and immunohistochemistry. The mRNA and protein expression levels of TLR4 (Figs. 3E,F,G) and the protein expression of NF-κB (Figs. 3F,H) were increased in the NEC group. However, TLR4 and NF-κB expression was decreased in the β-glucan group (P < 0.05) (Figs. 3F,G,H) In addition, data exhibited much more protein expression of TLR4 (Fig. 4A) and NF-κB (Fig. 4B) compared with the control and β-glucan group, as evidenced by enhanced TLR4 and NF-κB immunoreactivity. These results indicated that the amplification of a series of inflammatory cascades involving IL-1β, IL-6 and TNF-α is suppressed by β-glucan treatment, thereby alleviating intestinal inflammation.
To further demonstrate that β-glucan interacted with the active site of TLR4, molecular docking was carried out. As shown in Figs. 4C-G), β-glucan bound to TLR4 mainly through hydrogen bonds and hydrophobic interactions, forming a total of 7 hydrogen bonds whose hydrogen atoms came from the hydroxyl group of glycoside and oxygen atoms came from the skeletal carbonyl group or side chain of the amino acid inTLR4. β-glucan formed 7 hydrogen bonds with the amino acid GLN, HIS 199, GLU 225, ARG227, PRO202 of TLR4. β-glucan also formed hydrophobic effect with amino acids ILE 226, LEU 228, LEU 204, LEU203, PRO 202, MET 201, LEU 198. The best configuration was selected as the most likely binding mode from 30 conformations, with the score of -60.77 kcal/mol. The highest value of binding energy was estimated − 19.76 kcal/mol (Ki value 3.25fM), indicating that β-glucan displayed a relatively strong binding ability to TLR4. These results suggested that β-glucan exhibited significant affinities to the binding sites of TLR4, subsequently impact the TLR4-NF-κB signaling pathway, which was consistent with the results in vivo.
To identify the effect of β-glucan on intestinal barrier integrity, tight junction proteins expression such as ZO-1, Occludin and Claudin-1 were analyzed by immunohistochemistry and western blotting. Immunohistochemistry staining exhibited much less expression of ZO-1 (Fig. 5A) as well as Occludin (Fig. 5B) in NEC group compared with control and β-glucan group. Western blotting showed that the protein expression levels of ZO-1, Occludin and Claudin-1 were reduced in the NEC group. By contrast, ZO-1, Occludin and Claudin-1 expression was increased in the β-glucan group (Fig. 5C). These results suggested that β-glucan might prevent NEC-induced disruption of intestinal integrity by enhancing the expression of tight junction proteins.
To understand the influence of β-glucan on the gut microbiota in NEC mice, 16sRNA sequencing was performed. The sparse curve based on the OTU level of bacterial community gradually reached a saturation plateau along with the increase of sampling readings, indicating that the sequencing depth was sufficient to represent most microbial species (Fig. 6A). The α-diversity of gut microbiota, including Chao1, Ace, Shannon, and Simpson indices, did not differ significantly between NEC and β-glucan group (Figs. 6C-F), suggesting that β-glucan could not affect the overall bacterial richness and community diversity of mice with NEC. However, PCoA showed that the β-glucan and NEC group differed from the control group at OTU levels, but there was some overlap between the NEC and β-glucan group (Fig. 6G). Venn plots suggested that the β-glucan and NEC group, respectively, shared 7 and 11 OTUs with the control group (Fig. 6B). Circos presented that NEC group was mainly composed of Firmicutes (32%) and Proteobacteria (40%), while β-glucan group consisted of Firmicutes (37%), Proteobacteria (32%) and Bacteroides (about 0.072%) (Fig. 6H). In terms of phyla, the average relative abundance of Actinomycetes was lower in the NEC group than in the β-glucan group, but higher in the β-glucan group (Fig. 7A). In terms of genera (Fig. 7B), the relative abundance of Klebsiella_oxytoca_g__Klebsiella was obviously higher in NEC group than in the control and β-glucan group. The abundance of Clostridium_butyricum, Lactobacillus_johnsonii, Lactobacillus_murinus and Lachnospiraceae_bacterium_mt14 tended to increase in the β-glucan group compared to the NEC group. In addition, LEfSe analysis is shown in Fig. 7E, and the LDA score based on LEfSe analysis is shown in Fig. 7C,D. These data suggested that β-glucan could partially alter the gut microbiota structure in mice with NEC.
Numerous studies have suggested that the pathogenesis of NEC is multifactorial, but the etiology remains unclear. The immoderate inflammatory response, dysbiosis of gut microbiota and bacterial translocation are reported as important in NEC pathogenesis (4). Our study indicated that early intervention with β-glucan could help alleviate intestinal inflammation, promote gut barrier function and partially correct the dysbiosis of gut microbiota in mouse model with NEC.
The incidence of NEC continues to increase as improvements in early neonatal survival, but the mortality rate has not changed due to limitations in prevention and treatment. Although human milk has been reported an effective preventive measure against NEC (42), most premature babies do not obtain enough milk. Therefore, it is of great significance to explore other effective prevention and treatment to NEC.
β-glucan possesses various biological activities mentioned above. Previous studies also have demonstrated that oat and lentinus edodes β-glucan can suppresses DSS-induced colitis as well as the expression of pro-inflammatory factors in colonic tissues (26, 43). Lentinus edodes β-glucan was identified to perform an anti-inflammatory effect by impacting MAPK-Elk-1 and MAPK-PPARγ signaling pathways and subsequently inhibiting NF-κB activation in vitro and in vivo studies. NF-κB is one of the most important transcript factors controlling inflammatory cytokines expressions which most of inflammatory cytokine genes exist NF-κB binding sites in the promoter area. It is well known that TLR4-NF-κB signaling pathway plays very critical role in the development of NEC (44). It has been shown that TLR4 expression in intestinal epithelial cells is increased in response to intestinal inflammation in humans and mice (4, 5, 8). TLR4 activation induces nuclear translocation of NF-κB, which ultimately promotes excessive expression of proinflammatory cytokines, leading to the development of NEC (44). The proinflammatory cytokines TNF-α, IL-1β, and IL-6 were elevated in infants and animal models with NEC (45, 46), which can be suppressed by the use of TLR4 inhibitors (8, 47). Our experiment showed that early supplement with β-glucan reduced TLR4 and NF-κB protein expression levels in NEC mice. In addition, β-glucan also effectively inhibited the expression of NF-κB downstream proinflammatory cytokines TNF-α, IL-6 and IL-1β, which was in accordance with the results reported by Minmin Hu1 (48). Moreover, the results of molecular docking also demonstrated that β-glucan had high affinity with TLR4 binding site, which might impact the activation of TLR4-NF-κB signaling pathway. Upon these evidences, we speculate that β-glucan may alleviate intestinal inflammation through the TLR4/NF-κB pathway, however further study is required to confirm this hypothesis.
Intestinal barrier dysfunction mainly refers to the abnormal elevation of permeability, allowing pathogens to cross the barrier (32), which is regulated by tight junctions (TJs) formed between intestinal epithelial cells (IEC) in the apical region. Functional TJS is essential for maintaining intestinal permeability and intestinal barrier function (49). Transmembrane proteins Occludin, Claudins and ZO-1 are reported to be critical for regulating intestinal permeability (50). Gut barrier injury is implicated to play a key role in NEC pathogenesis (51). Gut barrier injury leads to the elevation of permeability, which subsequently facilitates translocation of pathogens into the underlying vascular circulation (52). Meanwhile, elevated proinflammatory cytokines (e.g., TNF-α, IL-1β, and TNF-α) also can disrupt TJs, thus forming a vicious cycle of inflammation and intestinal barrier impairment, leading to further deterioration of gut barrier function and aggravating intestinal inflammation (53). Hence, restoration of the gut barrier function may contribute to the alleviation of NEC. Previous study has implicated that oat β-glucan attenuated barrier function disruption in DSS-induced colitis mice. Similarly, in this experiment, the down-regulated protein expressions of Occludin, Claudins and ZO-1 in NEC mice were significantly inhibited by β-glucan. These results suggested that the preventive effect of β-glucan against NEC might be tightly related with the improvement of the epithelial TJs.
Gut microbiota is also an important part of the intestinal barrier, which is essential for the maturation of intestinal mucosal barrier function, immune system development, nutrient absorption and energy metabolism (54, 55). The increased permeability of gut barrier caused by microbial dysbiosis can trigger mucosal inflammation and promote NEC development (56). Therefore, modulating gut microbiota may provide new insights into therapeutic strategies for infants with NEC. In this study, we found that β-glucan did not change the diversity of intestinal microbiota in the NEC mice at the OTU level. In terms of phyla, Firmicutes and Proteobacteria showed no significant differences among three groups, but the relative abundance of actinobacteria in the β-glucan group was higher than that of in the NEC group. Early studies have revealed that microbial dysbiosis prior to NEC in preterm infants is characterized by increased relative abundance of Proteobacteria and decreased abundance of firmicutes, Bacteroidetes and actinobacteria (57). In terms of genera, we found that the proportion of Clostridium_butyricum, Lactobacillus_johnsonii, Lactobacillus_murinus and Lachnospiraceae_bacterium_mt14 increased after β-glucan intervention. Previous studies have found that after birth, the newborn gut is gradually colonized with facultatively and strictly anaerobic bacteria, including Clostridium butyricum (58). Clostridium butyricum is a species that embraces a variety of known strains, some of which encompass genes contributing them to release toxins. However, genomic analyses indicate that other strains do not encompass these genes, and that these nonpathogenic strains even encompass beneficial potential to promote host health through various mechanisms. In addition, Certain strains of Clostridium butyricum have been used as a probiotic for decades (58, 59). We also found that Klebsiella_oxytoca_g__Klebsiella was much more abundant in NEC mice than in control and β-glucan group. There exists study that confirm the tight association between the increase of Klebsiella and NEC pathogenesis (60). To sum up, β-glucan partially altered the structure of gut microbiota in NEC mice, increasing the abundance of beneficial bacteria while reducing the portion of harmful bacteria, but its effects to modulate the intestinal microbiota were limited. The possible reason may be that the β-glucan we used is from fungi, whose ability of modulating the intestinal microbiota is not as excellent as that of β-glucan from plants.
In conclusion, important findings of this current study were that β-glucan intervention alleviated intestinal inflammation through the TLR4/NF-κB pathway, improved gut barrier and partially modulated gut microbiota, ultimately protected against in the NEC mouse model. These results suggest that β-glucan has the potential beneficial preventive impacts on NEC. However, this experiment is based on an animal study and the exact mechanism and safety of β-glucan need to be completely elucidated in further cellular studies. In addition, due to the limited sample size, the dose of β-glucan was selected according to the results of pre-experiment. Further researches need to explore whether different concentrations of β-glucan have different preventive effects on NEC in cell experiments.
Acknowledgements
We are grateful to all the study participants and their families for their cooperation.
Author Contributions
X-DZ developed ideas and designed experiments. X-DZ, Y-NZ, and X-WZ led the experiments and collected data. X-DZ and Y-NZ performed the formal analysis and investigation. QA, YH, and YS provided resources and equipment and supervised the process. X-DZ and Y-NZ wrote the original draft. YS and YH reviewed and edited the article. All authors contributed to the study and approved the final submitted manuscript. Xingdao Zhang and Yuni Zhang contributed equally to this study.
Funding
This work was supported by 2020 General Project of Clinical Medical Research, National Clinical Medical Research Center for Children's Health and Disease (NCRCCHD-2020-GP-03), the special key project of technology innovation and application development of Chongqing Science and Technology Bureau (CSTC2021jscx-gksb-N0015), National Natural Science Foundation of China (NO.82001602), Science and health project of Chongqing Health Commission (NO.2020FYYX217).
Availability of data and materials
All data generated or analysed during this study are included in this published article.
Ethics approval and consent to participate
The studies were approved by the Animal Care and Use Ethics Committee of the Chongqing Medical University (Chongqing, China).
Consent for publication
Not applicable.
Competing interests
We confirm that none of the authors has any conflict of interest.