Fumonisins affect the intestinal microbial homeostasis in broiler chickens, predisposing to necrotic enteritis

Abstract

Fumonisins (FBs) are mycotoxins produced by Fusarium fungi. This study aimed to investigate the effect of these feed contaminants on the intestinal morphology and microbiota composition, and to evaluate whether FBs predispose broilers to necrotic enteritis. One-day-old broiler chicks were divided into a group fed a control diet, and a group fed a FBs contaminated diet (18.6 mg FB1+FB2/kg feed). A significant increase in the plasma sphinganine/sphingosine ratio in the FBs-treated group (0.21 ± 0.016) compared to the control (0.14 ± 0.014) indicated disturbance of the sphingolipid biosynthesis. Furthermore, villus height and crypt depth of the ileum was significantly reduced by FBs. Denaturing gradient gel electrophoresis showed a shift in the microbiota composition in the ileum in the FBs group compared to the control. A reduced presence of low-GC containing operational taxonomic units in ileal digesta of birds exposed to FBs was demonstrated, and identified as a reduced abundance of Candidatus Savagella and Lactobaccilus spp. Quantification of total Clostridium perfringens in these ileal samples, previous to experimental infection, using cpa gene (alpha toxin) quantification by qPCR showed an increase in C. perfringens in chickens fed a FBs contaminated diet compared to control (7.5 ± 0.30 versus 6.3 ± 0.24 log10 copies/g intestinal content). After C. perfringens challenge, a higher percentage of birds developed subclinical necrotic enteritis in the group fed a FBs contaminated diet as compared to the control (44.9 ± 2.22% versus 29.8 ± 5.46%).

Figure 1

Denaturing gradient gel electrophoresis (DGGE) fingerprint of DNA samples of ileal content applying community PCR with universal bacterial primers targeting the variable V3 region of the 16S ribosomal RNA (18 animals per group (3 pens/group, 6 animals/pen)). Percentage of similarity between DGGE profiles was analyzed using the Dice similarity coefficient, derived from presence or absence of bands. On the basis of a distance matrix, which was generated from the similarity values, dendrograms were constructed using the unweighted pair group method with arithmetic means (UPGMA) as clustering-method. The microbial richness (R) was assessed as the number of OTUs within a profile. Within ileum-samples a separation according to the treatment is visible, ^(D)although exceptions occur like samples of chicken 20, 31, 33 and 34 of FBs group which show high similarity (69.6 – 81%) with control group. ^(A)The majority of the control-samples contains bands in the lower GC-range which ascribes them to one clade. ^(B)Among the FBs group a clade of clearly reduced diversity is formed by 8 samples. ^(C)The remaining of this treatment group shows a similar diversity of OTUs of medium GC-content like the control-group, however the low-GC-OTUs were absent.

Figure 2

NE lesion score of individual broiler chickens challenged with C. perfringens. Chickens were fed either a control diet or a FBs contaminated diet. Subsequently, birds were orally inoculated with C. perfringens strain 56. Macroscopic intestinal NE lesions in the small intestine (duodenum to ileum) were scored as follow; 0 no gross lesions; 1 small focal necrosis or ulceration (one to five foci); 2 focal necrosis or ulceration (six to 15 foci); 3 focal necrosis or ulceration (16 or more); 4 patches of necrosis of 2 to 3 cm long; 5 diffuse necrosis typical field cases. Chickens with NE lesions scores of 1 or more were categorized as NE positive. No effect was observed on the mean lesion scores of NE positive chickens.