Diabetes-preventive molecular mechanisms of breast versus formula feeding: new insights into the impact of milk on stem cell Wnt signaling

Abstract

Human milk serves as a transmitter for epigenetic programming involved in postnatal tissue development and organ maturation of the infant. In contrast to formula feeding (FF), prolonged breastfeeding (BF) has been associated with diabetes-preventive effects. Polymorphisms of the transcription factor 7-like 2 (TCF7L2), the key downstream effector of Wingless (Wnt) signaling, increase the risk of diabetes mellitus. Wnt signaling is crucial for β-cell development and proliferation. However, there is limited information regarding Wnt/β-catenin/TCF7L2-dependent effects of BF versus FF on postnatal β-cell progenitor cell development, β-cell proliferation and β-cell mass expansion. The objective of our literature review is to collect and analyze data to provide translational evidence that different components of human milk promote Wnt signaling. We will specifically focus on the variations in Wnt signaling in enteroendocrine L-cells and pancreatic β-cells in response to either FF or BF. FF-induced overstimulation of mTORC1 may suppress Wnt gene expression through S6K1-mediated histone H3K27 trimethylation (H3K27me3). Moreover, the absence of milk exosomal miRNAs in formula that target mRNAs of crucial Wnt inhibitors, as well as reduced levels of eicosapentaenoic acid and glutamine in formula, may further hinder appropriate Wnt signaling, negatively impacting intestinal stem cells, enteroendocrine L-cells and potentially β-cell progenitor cells. Overall, the evidence presented supports the conclusion that FF has a detrimental impact on the Wnt/β-catenin/TCF7L2-regulated enteroendocrine-islet axis, disrupting proper β-cell maturation and proliferation. We propose that human milk, compared to formula, offers optimized conditions for physiological Wnt signaling promoting adequate neonatal β-cell mass expansion, which could explain the early diabetes-preventive effects of prolonged BF.

Keywords: breastfeeding; diabetes mellitus; diabetes prevention; formula feeding; microRNA; milk exosome; wingless signaling; β-cell.

Figure 1

Simplified overview of the canonical Wnt signaling pathway. (A) In the absence of Wnt ligands, cytoplasmic β-catenin is phosphorylated by the destruction complex (DC), which includes Axin, adenomatosis polyposis coli (APC), glycogen synthase kinase 3 (GSK3) and casein kinase 1 (CK1). Phosphorylation of β-catenin within this complex by CK1 and GSK3 targets it for ubiquitination (Ub) and subsequent proteolytic destruction. Without nuclear β-catenin, TCF7L2 engages with Groucho (TLE3 in β-cells), preventing the transcription of Wnt target genes. (B) When Wnt protein binds, it leads to the heterodimerization of the Frizzled receptor (FZD) with low-density lipoprotein receptor-related protein 5/6 (LRP5/6) followed by conformational changes resulting in the phosphorylation of LRP5/6 intracellular domain recruiting Axin and the DC to the cell membrane thereby inhibiting its activity. Subsequently, stable, non-phosphorylated β-catenin accumulates in the cytoplasm and translocates into the nucleus. The β-catenin/TCF7L2 complex regulates the expression of various Wnt target genes. Dickkopf (DKK) is a specific Wnt inhibitor that antagonizes Wnt signaling through direct interaction with the LRP5/6 receptor. Adapted from Napolitano et al. (92), licensed under CC BY 4.0.

Figure 2

TCF7L2-dependent gene regulation of the entero-insular axis. In enteroendocrine L-cells, TCF7L2 upregulates the expression of the proglucagon gene (GCG). After proteolytic cleavage of proglucagon by proprotein convertase, subtilisin/kexin-type 1 (PCSK1), glucagon-like peptide 1 (GLP-1) is generated. GLP-1 binds to GLP-1 receptor (GLP1R) of pancreatic β-cells and via cAMP/protein kinase A activation stabilizes β-catenin (β-cat). The activated β-catenin/TCF7L2 complex promotes the expression of critical target genes including MYC, PITX2 (β-cell proliferation), ISL1 (β-cell identity and function), GCG (GLP-1 generation), LGR5 (stem cell marker, augmentation of Wnt signaling), GLP1R (β-cell responsiveness to GLP-1) and NR2F2 (activation of GLP-1 signaling). TCF7L2 thus integrates and maintains the molecular cross talk between enteroendocrine L-cells and pancreatic β-cells. For further explanation of gene symbols, see Glossary.

Figure 3

Potential mTORC1-mediated mechanisms of reduced Wnt signaling caused by formula feeding. Increased formula feeding leads to elevated levels of branched-chain amino acids (BCAAs), insulin, and insulin-like growth factor 1 (IGF-1), which in turn over-activate mechanistic target of rapamycin complex 1 (mTORC1) and its downstream kinase S6K1. Once phosphorylated S6K1 enters the nucleus, triggers enhancer of zeste homolog 2 (EZH2) to trimethylate histone 3 at lysine 27, creating a negative regulatory complex that suppresses WNT gene expression. In addition, overactivation of mTORC1 suppresses the availability of frizzled (FZD) receptors on the cell membrane thereby reducing Wnt signaling and Wnt target gene expression. β-cat: β-catenin, LAT: L-type amino acid transporter, Leu: leucine, RAGD: Ras-related GTP binding protein D, IR: insulin receptor, IGF1R: insulin-like growth factor 1 receptor, PI3K: phosphatidylinositol 3-kinase, and Akt: Akt kinase (protein kinase B).

Figure 4

Amplification of Wnt signaling by the LGR5 receptor. The Wnt target gene and stem cell marker LGR5 enhances Wnt signaling, leading to the upregulation of genes that promote the expansion of LGR5⁺ β-cell progenitors and β-cell proliferation. For gene symbols, see Glossary.

Figure 5

Milk-controlled Wnt crosstalk between enteroendocrine L-cells and pancreatic β-cells. Human breastmilk exosomes (HBME) are taken up by intestinal stem cells and further differentiated L-cells. The most abundant miRNAs of early lactation (miR-148a and miR-22) directly or indirectly (via suppression of FOSB) target the Wnt inhibitor Dickkopf 1 (DKK1) resulting in enhanced Wnt signaling. In addition, miR-22 targets peptidyl arginine deiminase 2 (PADI2). This attenuates β-catenin (β-cat) citrullination, which increases β-catenin stability. Compared to infant formula, human colostrum and human milk provide higher quantities of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), which via binding to free fatty acid receptor 4 (FFAR4) results in increased expressions of lysine-specific demethylase 1 (LSD1). LSD1-mediated demethylation of β-catenin enhances its nuclear stability thus augmenting Wnt signaling. In contrast to formula, human milk provides higher amounts of free glutamine (Gln), which promotes intestinal stem cell proliferation and secretion of glucagon-like peptide 1 (GLP-1). Human milk also contains GLP-1 and the chemokine CXCL12, which additionally stabilizes β-catenin. These pathways converge in activating TCF7L2 and GLP-1 generation, which interacts with TCF7L2 signaling of the β-cell. Whether or not and how long HBME reach β-cell precursors or differentiated β-cell cells is an open research question. However, systemic nutrigenomic effects of Gln and EPA/DHA may also affect Wnt-regulated β-cell development. Upregulated TCF7L2-mediated gene expression by various convergent milk factors - all deficient in formula - may promote sustained β-cell proliferation and mass expansion, explaining the potential diabetes-preventive effect of breastfeeding.

Figure 6

Model illustrating deviated Wnt stem signaling by artificial formula feeding versus physiological breastfeeding. Breastfeeding maintains the appropriate Wnt signaling magnitude and kinetics required for the regulated development of β-cell progenitor cells and adipocyte stem cells. In contrast, formula feeding impairs the amplitude of Wnt signaling diminishing the pool of β-cell progenitor cells and reducing the postnatal pool of β-cells. Furthermore, reduced Wnt signaling enhances the commitment of adipocyte stem cells to adipogenesis. Formula feeding is thus a priming factor for both diabetes and obesity.