Maternal fiber-rich diet promotes early-life intestinal development in offspring through milk-derived extracellular vesicles carrying miR-146a-5p

Abstract

Backgrounds: The intestinal development in early life is profoundly influenced by multiple biological components of breast milk, in which milk-derived extracellular vesicles (mEVs) contain a large amount of vertically transmitted signal from the mother. However, little is known about how maternal fiber-rich diet regulates offspring intestinal development by influencing the mEVs.

Results: In this study, we found that maternal resistant starch (RS) consumption during late gestation and lactation improved the growth and intestinal health of offspring. The mEVs in breast milk are the primary factor driving these beneficial effects, especially enhancing intestinal cell proliferation and migration. To be specific, administration of mEVs after maternal RS intake enhanced intestinal cell proliferation and migration in vivo (performed in mice model and indicated by intestinal histological observation, EdU assay, and the quantification of cyclin proteins) and in vitro (indicated by CCK8, MTT, EdU, and wound healing experiments). Noteworthily, miR-146a-5p was found to be highly expressed in the mEVs from maternal RS group, which also promotes intestinal cell proliferation in cells and mice models. Mechanically, miR-146a-5p target to silence the expression of ubiquitin ligase 3 gene NEDD4L, thereby inhibiting DVL2 ubiquitination, activating the Wnt pathway, and promoting intestinal development.

Conclusion: These findings demonstrated the beneficial role of mEVs in the connection between maternal fiber rich diet and offspring intestinal growth. In addition, we identified a novel miRNA-146a-5p-NEDD4L-β-catenin/Wnt signaling axis in regulating early intestinal development. This work provided a new perspective for studying the influence of maternal diet on offspring development.

Keywords: Intestinal development; Maternal diet; Milk-derived extracellular vesicles; Offspring; Resistant starch; miR-146a-5p.

Fig. 1

Maternal RS supplementation enhances growth performance and reduces diarrhea in weanling piglets (n = 20). (A) Animal design and maternal RS diet treatment. (B) Diarrhea rate of piglets during weaning. Pearson chi-square value = 11.38 (p < 0.001). (C) Diarrhea index of piglets during weaning. (D) Serum growth hormone (GH) level in piglets at 21 d of age. (E) Serum insulin-like growth factor 1 (IGF-1) level in piglets at 21 d of age. (F) Immunoglobulins G, A, and M in sow milk on day 21 of lactation. Data are expressed as means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant

Fig. 2

RmEVs improve intestinal development in vivo during early life (n = 7). (A) Animal designs and mEVs administration schedule in young C57BL/6 mice. (B) The body weight of mice. the spherical test results were not satisfied (p = 0.01), the difference between groups was significant (p = 0.03), and interaction is not existed (time*group = 0.08). The body weight at each time point was analyzed separately and the superscript means the difference degree between RmEVs and CmEVs/Blank. (C) The length of the whole intestine of mice. (D) The images of HE dye in the jejunum, scale bar: 200 μm. (E) The crypt depth of jejunum. (F) The villi length of jejunum. (G, H) EdU positive cells observation and assay in the jejunum of mice. Sections were stained with EdU and DAPI, scale bar: 100 μm. The quantitative analysis of EdU positive cells by Image J. (I) The positive CCND-1 protein in the jejunum, scale bars: 50 μm. (J, K) The relative mRNA expression of IGF-1R and PCNA in the jejunum. Continuous data of body weight were expressed as mean ± SD, others are expressed as means ± SEM. *p < 0.05, **p < 0.01; ns, not significant

Fig. 3

RmEVs promote cell proliferation and migration of IEPC-J2 (n = 3). (A-C) The assays of the proliferation of IPEC-J2 by CCK8 assay (A), MTT assay (B), and EdU assay (C). (D-F) Wound-healing assay. IPEC-J2 cells were treated with RmEVs, CmEVs, and Blank. Images were taken on the 0, 12, and 24 h after the scratch, scale bar: 400 μm. The spherical test results were not satisfied (p = 0.02), the difference between groups was significant (p = 0.04), and interaction did not exist (time*group = 0.22). The percentage of scratched area recovery of IPEC-J2 at 12 h (E) and 24 h (F) were analyzed separately. (G-I) The relative mRNA expression of CDX2, PCNA, and CCND-1. (J) Images of PKH26-labeled mEVs absorbed by IPEC-J2 after 6 h of culturing, scale bars = 50 μm. (K) The PKH-26 positive cell rates of IPEC-J2. Continuous data of recovery rate were expressed as mean ± SD, others are expressed as means ± SEM. *p < 0.05, **p < 0.01; ns, not significant

Fig. 4

miR-146a-5p increases the proliferation and migration of IPEC-J2 (n = 3). (A) The images of EdU, scale bar: 50 μm. (B) The percentages of EdU in IPEC-J2. (C) Images of scratched recovery of IPEC-J2 cells after being transfected by miR-146a-5p mimics, inhibitors, mimics-NC, inhibitors-NC, scale bar: 400 μm. (D, E) The percentage of scratched area recovery of IPEC-J2. Data were analyzed by repeated measures analysis of variance, followed by within-group comparisons. The spherical test results were not satisfied (p = 0.04), the difference between groups was significant (p = 0.001), and interaction did not exist (time*group = 0.86). The scratched area recovery of IPEC-J2 at 12 h (D) and 24 h (E) were analyzed separately. (F, G) The cell proliferation of IPEC-J2 treated with RmEVs + inhibitors-NC, RmEVs + miR-146a-5p inhibitors, or Inhibitors-NC for 24 h followed by CCK (F) and MTT assays (G). Continuous data of recovery rate were expressed as mean ± SD, others are expressed as means ± SEM. *p < 0.05, **p < 0.01

Fig. 5

miR-146a-5p enhances the activation of Wnt pathway via targeting NEDD4L (n = 3). (A) The potential binding sites of miR-146a-5p in NEDD4L 3’UTR and their wild (WT) or mutant (MUT) sequence designs. (B) Luciferase reporter assays. HEK239 cells were co-transfected with pirGLO vector-WT/MUT and miR-146a-5p mimics or negative control. WT, wild type; MUT, mutant type; NC, negative control. (C-I) The relative mRNA expression of NEDD4L, DVL2, AXIN2, c-MYC, GSK-3β, β-catenin, and CCND-1 in IPEC-J2 cells after transfected with miR-146a-5p. (J-M) The protein levels of NEDD4L, DVL2, and β-catenin after transfected with miR-146a-5p. Data are expressed as means ± SEM. *p < 0.05, **p < 0.01; ns, not significant

Fig. 6

miR-146a-5p reduces proteasomal degradation of DVL2 via silencing the NEDD4L expression and modulates Wnt/β-catenin signaling (n = 3). (A-C) The protein expression levels of NEDD4L and DVL2 in IPEC-J2 cells were transfected with pcDNA3.1-NEDD4L or pcDNA3.1-control, then treated with MG132 or DMSO, and followed by Western blot analysis. (D-G) The protein expression levels of NEDD4L, DVL2, and β-catenin in IPEC-J2 cells were transfected with miR-146a-5p mimics or NC, then treated with MG132 or DMSO, and followed by Western blot analysis. (H-J) The relative TOP/FOP Flash reporter activities of IPEC-J2 cells after transfected with miR-146a-5p mimics or NC (H), pcDNA3.1-NEDD4L or pcDNA3.1-control (I), pcDNA3.1-DVL2 or pcDNA3.1-control (J). (K) Western blot images of HEK293 cells were transfected with constructs expressing Flag-NEDD4L, HA-Ubiquitin (HA-UB), MYC-DVL2, and miR-146-5p mimics or mimics-NC, then followed by anti-MYC IP under denaturing conditions. Data are expressed as means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001

Fig. 7

Overexpression of miR-146a-5p improves renewal of intestinal cells in young mice models (n = 8). (A) Animal designs and schedule. (B) The changes in body weight during the experiment. The spherical test results were satisfied (p = 0.144), the difference between groups was significant (p = 0.01), and interaction did not exist (time*group = 0.06). The body weight at each time point was analyzed separately and the superscript means the difference degree between agomiR-146a-5p and NC/Control. (C) The representative images of the intestine of mice from different treatments. (D-G) Quantitative analysis of intestinal length and intestinal weight of mice from different treatments. (H) Images of Cy3 and DAPI fluorescence in the jejunum of mice at 6 h after the injection of Cy3-labeled agomiR-NC, Cy3-labeled agomiR-146a-5p, or PBS. Scale bar: 50 μm. (I) The expression of miR-146a-5p in the jejunum. (J, K) Images of NEDD4L protein expression in the jejunum, presented by IHC method and quantitative analysis, scale bar: 50 μm. Continuous data of body weight were expressed as mean ± SD, others are expressed as means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant