Changes of Colon Flora and Antitoxic Effect Induced by Zearalenone in Mice

According to previous reports, zearalenone (ZEA) is one of the most threatening environmental toxins in the world and it was associated with many fungal diseases in farm animals (Fink-Gremmels and Malekinejad, 2007). Nowadays, we often concern that ZEA induces reproductive toxicity in mammals. There are few studies on the toxic effects of ZEA on intestinal mucosal barriers in animal or human. Studies have revealed that ZEA impairs the small intestinal mucosal barrier (Liu et al., 2014), and we have previously demonstrated that ZEA not only disrupts the jejunal microecological balance but also causes inflammation of the jejunal mucosa (Wang et al., 2018). But we cannot ignore that colon contains large number of intestinal microbes and its microecological stability is more vulnerable to interference from external substances. Increasing evidence suggests that the incidence of colitis and colon cancer is closely related to environmental pollutants, especially mycotoxins (Wild and Gong, 2010; Seonghwan et al., 2010; Maresca and Fantini, 2010). Therefore, it is necessary to investigate the effects of ZEA exposure on the colonic mucosal barrier and microecological stability. As we all know, the intestinal mucosa is the first line of defense of the intestinal barrier. Notably, the intestinal epithelial cells, intestinal flora, mucus, sIgA and antimicrobial peptides (AMPs) in the intestinal mucosa constitute the “firewalls” of intestinal mucosal immunity, which protect the body from pathogenic bacteria, virus and environmental pollutants (Macpherson et al., 2009). Recent studies have shown that most mycotoxins can induce intestinal epithelial cell inflammation by destroying intestinal mucosal barrier (Seonghwan et al., 2010). Indeed, The intestinal flora plays a role in regulating intestinal immunity, and it acts as an adjuvant to the entire immune system (Molloy et al., 2012; Belkaid and Hand, 2014). However, there are few studies on the toxic effects of ZEA on the intestinal microflora and colonic mucosal barrier in mammals, and its mechanism is still unclear (Gajęcka et al., 2016).

Therefore, we systematically investigated the effects of ZEA on colonic microorganisms and mucosal barriers in mice by 16S rRNA sequencing technology, and explored the potential molecular mechanisms of reduced immunotoxicity, which provided theoretical basis for prevention and treatment of ZEA exposure.

ZEA (purity > 98%) was purchased from Sigma-Aldrich (St Louis, MO, USA). It was dissolved in ethanol and made into 100 mg/ml. Then the storage solution was dissolved in olive oil at 10mg/ml products before use.

Twenty-four immature male BALB/C mice (Laboratory Animal Center of Jilin University, China), were individually divided into two groups, each group had three replicates, each containing four mice. After a weekend for a trial period, all mice were reared under standard test conditions, and were fed by basal diet (Liaoning Changsheng Biological Company) and sterile drinking water. The recommended dose of ZEA induces a physiological disorder in mice of 20 mg/(kg.bw) (~1% LD50) (Zinedine et al., 2007). We adopted this recommended dose and conducted many tests on mice, we found it was indeed effective. Therefore, the experimental group (ZEA group) was administered ZEA by oral gavage at a dosage of 20 mg/kg body weight (~1% LD50) once daily for one week, and the control group (CON group) was given an equal volume of vector. All experiments were approved by the Institutional Animal Care and Use Committee of Jilin University.

At the end of the experiment, the mice were sacrificed by cervical dislocation. The colon was separated, and the intermediate segment (1-2 cm long) was taken and washed away with saline, and the tissue was fixed with 4% polyoxymethylene. The remaining colon and feces were carefully collected, quickly placed in liquid nitrogen, and then transferred to -80°C.

The intestinal tissues were fixed with 4% polyoxymethylene and embedded in petrolin, then sectioned into 2-3 µm sections. The sections were stained with hematoxylin first, counterstained with eosin, and finally observed under an optical microscope.

Total genomic bacterial DNA was extracted from the feces of colon in BALB/c mice. Then the qualified genomic DNA was used for 16S-rDNA V3-V4 region PCR amplification. The primer was 341F (5’ ACTCCTACGGGAGGCAGCAG-3’) and 806R (5’ GGACTACHVGGGTWTCTAAT-3’). All the procedures were performed according to the manufacturer’s protocol (BGI, Shengzhen, China). Two hypervariable regions of 16S rDNA, the V3 and V4 regions, were used to identify the vast majority of the bacteria based on the 16S rDNA sequencing. The qualified PCR products were purified by 16S V3-V4 amplification, and the DNA library was constructed. The qualified DNA library was sequenced using an Illumina MiSeq 2*300, and information was acquired for the bioinformatic analysis.

The raw data were filtered to eliminate adapter pollution and low quality to obtain clean reads. Then, the paired-end reads with overlapping regions were merged to raw tags with FLASH software (v1.2.11) (Magoč and Salzberg, 2011). Paired-end reads without overlaps were removed. Following removal of the primer sequences, the forward and reverse amplification primers were mapped to the two ends of the tags. The tags were clustered to the operational taxonomic unit (OUT) at 97% sequence similarity by scripts using the software USEARCH (v7.0.1090) (RC, 2013). The PCR-generated chimaera from the OTU representation sequence was removed, and effective tags were obtained by UCHIME software (v4.2.40) (Edgar et al., 2011). The database used for species annotation was Greengene (V201305) (Desantis et al., 2006).

Total RNA was extracted from colon using TriPure (Roche, Basel, Switzerland) and the cDNA was synthesized using Reverse Transcription System (Promega, Madison, USA) and SYBR Green Mix Kit (Trans, Beijing, China) was used for real-time PCR. Primer design as shown in Supplementary Table I.

Equal amounts of colonic feces were dissolved with 0.01 M PBS, and broken down at low temperature, centrifuged for 10 min at 3000 r, then the supernatant was collected. The total protein concentration of the feces was detected by BCA. Then the absorbance was detected at 562 nm according to the instructions of mouse sIgA ELISA kit (eBioscience, California, USA), and the sIgA content of feces was calculated (ng/ml).

Colon was homogenized with 0.01 M PBS to make 10% tissue homogenate. Then at low temperature, the homogenate was centrifuged for 10 min at 3000 r. The supernatant was used for estimation of total protein according to the instructions of IL-1β, IL-10, and TNF (Naniing Jiancheng Bioengineering Institute, Nanjing, China). The cytokine levels were calculated as ng/g prot.

All the data were analysed using SPSS statistical software (version 20.0). The variation between the CON group and the ZEA group was analysed by a t-test, and the results are expressed as the mean ± SEM. The differences were judged as statistically significant at P values < 0.05.

To objectively assess the effects of ZEA on intestinal mucosa, the histological sections of colon mucosal morphology was observed, as shown in Figures 1A, 1B. We found that there was no significant pathological change between ZEA group and control group, but the intestinal cavity of the ZEA group became significantly larger (Fig. 1C).

In order to study the effect of ZEA on intestine microflora in the mice, a total of 148209 effective tags were obtained by 16S-rRNA sequencing, with an average of 24702 effective tags for each sample and an average length of 416 bp (Table I). Using the ACE index (Table I) to assess bacterial richness, short-term ZEA exposure led to a tendency to increase the richness of the colonic microflora. At the same time, the Simpson index was used to evaluate changes in the intestinal microbial diversity (Table I). In the ZEA group, the bacterial diversity was almost similar to that of the CON group, but the diversity of the colonic microflora showed a tendency to increase. At 97% similarity, the colonic bacterial community of the CON and ZEA groups shared 325 OTUs (Fig. 2A). The flattening of the rarefaction curve based on the values of observed species (Fig. 2B) indicated that our data volume covered all species of the community in the colon samples. At phylum level (Fig. 2C), compared with the CON group, the abundance of Proteobacteria in the ZEA group was reduced more than 2-fold (P < 0.001). The abundance of Actinobacteria in the ZEA group was increased (P < 0.01). At genus levels (Fig. 2C), Lachnoclostridium and Alloprevotella were highly abundant. Compared to the CON group, ZEA reduced Lachnoclostridium from 16.7% to 9.8% (P > 0.05) and reduced Alloprevotella from 9.6% to 4.9% (P > 0.05). The abundance of Desulfovibrio was reduced from 3.1% to 1.5% (P > 0.05). Compared with CON group, the abundance of Anaeroplasma, Flavonifractor, Helicobacter, Prevotella and Ruminiclostridium in the ZEA group was decreased (P > 0.05). Interestingly, the Lactobacillus was upregulated more than 4-fold (P < 0.05). In addition, we found that short-term ZEA exposure led the emergence of Akkermansia in colon.

We further detected the effect of ZEA on colonic mucosal barrier. We examined the mRNA expression of key genes that play the first line of defense in the colonic mucosa (Fig. 3). The mRNA expression of Mucin 1, Mucin 2, and Reg3γ were significantly downregulated by ZEA exposure. Interestingly, ZEA increased mRNA expression of β-defensin, Reg3α, and Reg3β. In addition, the mRNA expression of inflammatory cytokines IL-1β and TNF were remarkably downregulated after ZEA exposure. There was no significant change in the mRNA level of IL-10.

To assess the effect of ZEA on intestinal mucosal immunity, we firstly detect the faecal sIgA levels in the colon. The results showed that short-term ZEA exposure induced the mucosal sIgA (P < 0.001) and the faecal IgA in the ZEA group increased more than 3-fold compared to that in the CON group (Fig. 4A). At the same time, we investigated the effects of ZEA on cytokine secretion involved in the inflammatory response in the colon. Interestingly, ZEA inhibited the secretion of IL-1β in the colon, but there was no change in TNF and IL-10 levels (Fig. 4B).

Mycotoxin is a class of environmental pollutant that easily causes an intestinal ecological imbalance and induces some intestinal diseases (Maresca and Fantini, 2010; Choung and Iii, 2011; Compare and Nardone, 2014). We found that the structure of colonic microbial communities was altered by short-term ZEA exposure. Changes in some of the dominant bacteria in the gut indicated that our ZEA exposure induced disturbances in the intestinal microecology. We found that Proteobacteria which are associated with inflammation was decreased (Carvalho et al., 2012). At the same time, Lachnoclostridium and Alloprevotella, Desulfovibrio,

Anaeroplasma, Flavonifractor, Helicobacter, Prevotella and Ruminiclostridium were decreased. Previous studies have shown that Clostridium leptum, Lachnospiraceae, and Prevotellaceae were reduced in inflammatory bowel disease patients (Swidsinski et al., 2002). Changes in these micrbiotes suggest instability of the intestinal flora or sensitivity of the intestines. It is worth noting that ZEA increased the Lactobacillus. Piotrowska et al. (2014) argued that long-term exposure to low doses of ZEA caused an increase in the Lactobacillus in the colon, but the change was not significant. Reddy et al. (2018) found that the ZEA-contaminated feed significantly affected the colon microbiota, especially Lactobacillus, which was increased by 2.7%. They speculate that Lactobacillus plays a major role in detoxification of these mycotoxins. Indeed, Lactobacillus have been evaluated in studies in animals and humans with respect to antibiotic-associated diarrhoea, travellers’ diarrhoea, pediatric diarrhoea, inflammatory bowel disease, and irritable bowel syndrome. Lactobacillus is not only beneficial to intestinal health (Sonnenburg et al., 2005) but also combined with mycotoxins to remove mycotoxins (Elnezami et al., 2002). The decrease of Proteobacteria and the increase of Lactobacillus seem to indicate that damage to the colon by ZEA has been controlled, as evidenced by normal colon morphology. We suspect that the increase of the intestinal cavity area may also be due to the changes in intestinal flora.

Further studies found that ZEA exposure significantly reduced the mRNA expression of Mucin 1, Mucin 2 and Reg3γ in colonic mucosa. Current studies suggest that ZEA increases the secretion of mucin in Caco-2/HT29-MTX cells (Wan et al., 2014). Reg3γ and mucins are used to prevent microbial contact with the epithelial layer (Turner, 2009; Sanchez et al., 2014; Stefania et al., 2015). The decrease of these mRNA expressions indicates the decrease of mucosal defense. Interestingly, ZEA increased the mRNA expression of colonic mucosal β-defensin, Reg3α and Reg3β. In addition, the sIgA in the colon was also increased. Indeed, β-defensin Reg3α/β and sIgA play an essential role in the immune defense of the intestinal mucosa (Wang et al., 2015).

ZEA has been described as either an inductor or a suppressor of pro-inflammatory cytokines (Salah-Abbès et al., 2010; Marin et al., 2011). Our study found that ZEA exposure inhibited the IL-1β and the mRNA expression of TNF was downregulated. Fan et al. (2017) suggested that ZEA leads to increased levels of IL-1β in colon tissue of mice. It contradicts our findings. However, Liu et al. (2014) revealed that different doses of ZEA showed a downward trend of IL-1β and TNF in the jejunum of rats. Braicu et al. (2016) found that exposure to ZEA caused a decrease in the IL-1β and TNF in animal which is still waiting for us to further study.

Current study suggests that ZEA is absorbed through the small intestine and some of it is degraded through the circulation in the liver and intestine (Fan et al., 2017) while some combines with intestinal microbes. Eventually, only a small amount of ZEA and its degradation products act on the large intestine (Kollarczik et al., 2010). On one hand, a low dose of ZEA is considered to be immunosuppressant (Bondy and Pestka, 2000) and may inhibit the inflammatory response. On the other hand, the increase of colonic Lactobacillus only remove mycotoxins by combining with mycotoxins but also regulate intestinal mucosal immunity by producing short chain fatty acid (SCFA) (Elnezami et al., 2002; Martín et al., 2013). Thereby mucosal damage is alleviated to some extent. But the specific reasons are still waiting to be explored.

Our study found that ZEA enhanced mucosal immunity in the colon and decreased the colonic Proteobacteria. It is worth noting that the intestinal Lactobacillus was increased. We believe that low doses of ZEA enter the colon and stimulate the growth of beneficial bacteria in the colon (Fig. 5), which indirectly activates the mucosal defense system. Ultimately the toxicity of ZEA to the intestinal mucosa was alleviated. All of this credit comes from the special role of Lactobacillus, which play a key role in regulating intestinal mucosal immunity and promoting colon health. Thus, the colonic mucosa can protect against a certain degree of toxins such as ZEA. This appears to provide us with a new way of thinking about controlling mycotoxins. We believe that intestinal Lactobacillus play an essential role in the intestinal mucosal immune defense against environmental toxins such as mycotoxins.

This study was financially supported by the Jilin Scientific and Technological Development Program (20170307018NY, 20190301034NY); the Special Project of the Province-University Co-constructing Program of Jilin Province (SXGJXX2017-4); the central finance forestry science and technology promotion demonstration fund project and the National Key Research and Development Program of China Stem Cell and Translational Research (2017YFA0105101); the Open Projects of Northeast Agricultural University/Scientific Observing and Experimental Station of Animal Nutrition and Feed Science in Northeast, Ministry of Agriculture, P.R. China; and the Jilin Modern Agricultural Technology Demonstration and Extension Project.

There is supplementary material associated with this article. Access the material online at: https://dx.doi.org/10.17582/journal.pjz/20190115060153

The authors declare that there are no conflicts of interest.