Milking miRNAs for All Their Worth

See corresponding article on page 94.

Mother's milk is the original, and unparalleled, superfood (1). It's loaded with nutrients and other benefits for babies’ health (1, 2). Decades of research have associated breastfeeding with lower risks of infection, obesity, and diabetes in infants (3). Breast milk is a complex assortment of bioactive molecules, and the role of many of these components remains unclear. In 2010, scientists found that breast milk contains microRNAs (miRNAs) (4); short RNAs approximately 22 bases long (5). To explain their presence, the nutrition community has come up with 2 theories (3). The first, known as the nutrition hypothesis, states that miRNAs are bundles of nutrients—similar to one of breast milk's major proteins, serum albumin—that are broken down in the gut (1). The second, termed the functional hypothesis, states the miRNAs in breast milk function like those within a cell, and have a regulatory role; they survive digestion and impact the consuming infant's gene expression (6). In this issue of the Journal, Raymond et al. (7) perform a tour de force addressing milk miRNA composition among a large group of women (n = 44), while also providing critical information regarding the longitudinal dynamic changes in miRNA composition during lactation. Meanwhile, their work regarding functional relevance will require follow-up studies.

A strength of the Raymond et al. study is the sampling and measuring of the miRNAs (7). Given their size and low abundance, measuring and quantitation of dietary miRNAs can be an arduous task, fraught with experimental complications (5). Raymond et al. used an elegant extraction-free chemistry that reduces sample input requirements; in fact, the group could profile breast-milk miRNAs from <30 µL of sample (7). After 17, 60, and 90 d of lactation, 44 mothers donated their milk. Among the approximately 2000 miRNAs measurable with this technology, 685 were commonly found in all 3 time points. This represents about 30% of the miRNome (the full spectrum of miRNAs expressed in a specific genome) and helps validate their experimental approach, as it is consistent with previous studies using other technologies (2). Among the miRNAs, 35 appeared to be abundant at each time point; 15 of these abundant miRNA had not been previously characterized, highlighting the robust nature of their analytical approach (7).

Raymond et al. also present the most thorough analysis to date on the longitudinal dynamics of miRNA abundance in breast milk (7); previous studies have been limited (2). Applying computer modeling, they identify 11 dynamic miRNAs, among which 7 had significant stage-specific upregulation and 4 had significant stage-specific downregulation. This dataset supports observations that demonstrate that miRNA content in breast milk adapts to the infant's nutrient requirement (2). Previous work has shown premature delivery results in a unique breast-milk RNA profile with metabolic targets (8). This suggests that preterm miRNAs in milk may have specialized functions for growth in premature infants. Raymond et al. now show that this dynamic fluctuation occurs even among healthy mothers delivering term infants (7); one can't help but ponder the intricate signaling that could be occurring through these regulatory RNAs.

Establishing a definitive connection between these miRNAs and infant health is an enormous challenge. In terms of addressing functionality of these exciting miRNAs, this work could have addressed how these miRNAs are packaged. In milk, miRNAs may be protected by membrane-derived vesicles (exosomes and microparticles), lipoproteins, and other ribonucleoprotein complexes (5, 9). Protected from ribonucleases, functional miRNAs are delivered to recipient cells. Studies have shown that protected miRNAs in cow's milk can reach the liver and brain in mice (10). In fact, studies suggest that vesicle properties could be dynamic, and that vesicle properties rather than the quantity of a given miRNA could impact function (11). It remains an open question if the miRNAs identified in this study are packaged in a way that facilitates survival during digestion.

Raymond et al. do use in silico predictions and in vitro studies to demonstrate that some of their identified miRNAs may be involved in gene regulation (7). While these studies are rigorous and utilize the best technologies available, they remain imperfect; the use of computer programs has shown limited overlap between predicted and observed targets of miRNAs (12). The nutritional relevance of their in vitro work must also be cautiously interpreted; their studies incubate cells with exogenous miRNAs that are at levels probably a hundred times higher than those found in milk (13).

The ultimate experiment for establishing the functional consequences of the miRNAs in mother's milk has not yet been performed. Experiments in mice offer some insights into future approaches; one refined study focused on examining the uptake of maternal milk–derived miRNAs into the intestines of newborn mice (14). The pups were genetically manipulated to lack 2 miRNAs and received milk from wild‐type foster mothers whose milk contained the missing miRNAs. Analysis of intestinal epithelium, blood, liver, and spleen in the pups revealed no evidence of miRNA uptake (14). While these data do not support functionality, they have several limitations. These miRNAs may not have been packaged correctly; alternatively, a failure to detect a postprandial increase in plasma miRNAs or tissue miRNAs could be caused by rapid metabolism in intestinal cells or the liver. Advances in our understanding of miRNA signaling between mother and infant will require incremental advances using model systems like pigs, mice, or rats. Regardless of the biological system, it is imperative to understand the packaging of these RNAs as this directly relates to potential function (6). To resolve miRNA uptake in humans will require well-designed clinical studies of both mothers and breastfed infants. However, demonstrating functionality in infants will require even further methodological innovations.

Discerning the biological properties of breast-milk miRNAs is a complicated pursuit. Raymond et al. have firmly demonstrated there are dynamic and consistent miRNA fluctuations during lactation (7). This work has framed the problem; we now work toward the solution.

Notes

This material is based on work that is supported by USDA/ARIS 6250-51000-051-00D and National Institutes of Health (NIH) 1R03AI149201.

Author disclosures: The author reports no conflicts of interest.