Developmental defects and behavioral changes in a diet-induced inflammation model of zebrafish

Abstract

Soybean meal evokes diet-induced intestinal inflammation in certain fishes. Although the molecular aspects of soybean-induced intestinal inflammation in zebrafish are known, the impact of the inflammatory diet on fish behavior remain largely underexplored. We fed zebrafish larvae with three diets - control, soybean meal and soybean meal with β-glucan to gain deeper insight into the behavioral changes associated with the soybean meal-induced inflammation model. We assessed the effect of the diets on the locomotor behavior, morphological development, oxygen consumption and larval transcriptome. Our study revealed that dietary soybean meal can reduce the locomotor activity, induce developmental defects and increase the oxygen demand in zebrafish larvae. Transcriptomic analysis pointed to the suppression of genes linked to visual perception, organ development, phototransduction pathway and activation of genes linked to the steroid biosynthesis pathway. On the contrary, β-glucan, an anti-inflammatory feed additive, counteracted the behavioral and phenotypic changes linked to dietary soybean. Although we did not identify any differentially expressed genes from the soybean meal alone fed group vs soybean meal + β-glucan-fed group comparison, the unique genes from the comparisons of the two groups with the control likely indicate reduction in inflammatory cytokine signaling, inhibition of proteolysis and induction of epigenetic modifications by the dietary glucan. Furthermore, we found that feeding an inflammatory diet at the larval stage can lead to long-lasting developmental defects. In conclusion, our study reveals the extra-intestinal manifestations associated with soybean meal-induced inflammation model.

Keywords: behavior; inflammation; soybean meal; zebrafish; β-glucan.

Figure 1

Diet-induced morphological changes in zebrafish larvae. Representative images of the zebrafish larvae fed the (A) CT (B) SBM and (C) BG diets (D). Measurement strategy that was adopted to assess the morphological changes in zebrafish larvae. The measured parameters include (E) Standard length (F) Snout to vent length (G) Eye area (H) Swim bladder area (I) Head to trunk angle. Asterisks *** indicate p < 0.001, ** indicate p < 0.01, * indicates p < 0.05. Larvae were assessed at 15 dpf (n = 9-10 per group). Control - CT, soybean - SBM and β-glucan - BG. Scale bar = 500 μm.

Figure 2

Changes in parameters linked to locomotor activity of zebrafish larvae. The measured parameters included (A) Distance travelled (B) Average velocity (C) Movement (D) Angular velocity and (E) Heading angle. Effect of alternating light-dark phases on the (F) velocity of zebrafish larvae. *** p < 0.001 and (.) p < 0.1. Larvae assessed at 15 dpf (n = 18-20 per group). Control - CT, soybean - SBM and β-glucan - BG.

Figure 3

Localization of granulocytes in the intestine of zebrafish larvae. (A) Representative images of the intestine region of zebrafish larvae stained with Sudan Black to reveal the presence of granulocytes, yellow arrows indicate the granulocytes in the intestine region. (B) Quantification of Sudan-Black⁺ cells in the intestine region of zebrafish larvae. *** p < 0.001 and * p < 0.05. n = 24-25 per group. Control - CT, soybean - SBM and β-glucan – BG. Scale bar = 200 μm.

Figure 4

Oxygen consumption of zebrafish larvae from the 3 study groups. A generalized additive model using R package mgcv indicated significant differences (p < 0.05) in oxygen saturation. Larvae assessed at 15 dpf (n = 12 per group). Control - CT, soybean - SBM and β-glucan - BG.

Figure 5

Transcriptome-based differences in zebrafish larvae from the soybean group compared to the control group. Principal component analyses (A) and heatmap (B) of the differentially expressed genes (DEGs) in the soybean (SBM) group compared to the control (CT) group. Transcripts with an adjusted p-value below 0.05 and |Log2 fold change| ≥ 1 were considered significantly differentially expressed. There are six biological replicates in each study group.

Figure 6

Network plot showing the link between the enriched GO terms. DEGs (downregulated; SBM vs CT) that were considered for the enrichment are indicated using red circles and only the non-redundant GO terms are shown in the cluster. The gradient color bar intensity varies with the p value and the sizes of the nodes of the GO terms increase with the associated fold change. There are six biological replicates in each study group. Control - CT, soybean - SBM.

Figure 7

KEGG pathways that were enriched in the zebrafish larvae from the soybean group. Differentially expressed genes in the soybean (SBM) group compared to the control (CT) group were employed for the pathway enrichment analysis. The size of the circles is proportional to the gene count and gradient color bar intensity of circle correlates with the p value. There are six biological replicates in each study group.

Figure 8

Network plot showing the link between enriched GO terms. DEGs (upregulated; SBM vs CT) that were considered for the enrichment are indicated using red circles and only the non-redundant GO terms are shown in the cluster. The gradient color bar intensity varies with the p value and the sizes of nodes of the GO terms increase with the associated fold change. There are six biological replicates in each study group. Control - CT, soybean - SBM.

Figure 9

Transcriptome-based differences in the zebrafish larvae from the β-glucan group compared to the control group. Principal component analyses (A) and heatmap (B) of differentially expressed genes (DEGs) in the β-glucan (BG) group compared to the control (CT) group. Transcripts with an adjusted p below 0.05 and |Log2 fold change| ≥ 1 were considered as significantly differentially expressed. There are six biological replicates in each study group.

Figure 10

Network plot showing the link between the enriched GO terms. The enriched GO terms that were unique (from the BG and SBM vs CT comparisons) to the zebrafish larvae fed with the BG diet are shown in the figure, and 298 and 95 genes were significantly (A) upregulated and (B) downregulated, respectively, in BG vs CT comparison. Control - CT, soybean - SBM and β-glucan - BG.