Human Milk Oligosaccharides in Exercise Nutrition: Structure-Specific Evidence, Human Efficacy, and Translational Bottlenecks
Qiao C, Guo S, Yang S, Yang C, Wang Y. · Preprints.org · 2026
Background: /
Objectives: Human milk oligosaccharides (HMOs) are structurally diverse bioactive carbohydrates being investigated beyond early-life nutrition. General gut–muscle and short-chain-fatty-acid frameworks are already well described; the unresolved issue is whether structure-defined HMOs produce reproducible exercise benefits in humans and where translation from model-based mechanisms fails.
Methods: This critical translational review systematically searched PubMed/MEDLINE, Embase, Web of Science Core Collection, Scopus, and SPORTDiscus. Evidence was assigned non-exclusively to Layer A (direct exercise evidence), Layer B (exercise-relevant mechanisms), or Layer C (adult translational evidence). We separately evaluated HMO structure, prespecified human efficacy outcomes, within-model causal support, and distance from healthy human exercise.
Results: The five databases yielded 4,787 records; 40 reports representing 38 independent studies were included. Several HMOs were tolerated at the doses and durations tested and produced structure-, dose-, and host-dependent microbial responses; limited systemic exposure was documented for selected structures. Some intestinal and disease-model pathways had perturbation, knockout, transfer, or rescue support. However, direct human exercise evidence was sparse and inconsistent: primary or overall interaction tests were generally null, whereas favorable
Findings: were mainly secondary, interim, within-group, or exploratory. No study established the complete sequence from oral exposure to target engagement, healthy skeletal-muscle adaptation, and improved performance.
Conclusions: Current evidence supports biological activity and testability of selected HMOs, but not an HMO class effect, a confirmed ergogenic mechanism, or a recommendation for exercise-performance use. Translation is constrained by four unresolved links: structure to exposure, exposure to target engagement, target engagement to muscle phenotype, and muscle phenotype to functional benefit.