Donkey Milk Study: miRNA in Milk Vesicles from Donkey, Human, and Cow (Yan 2026)

In vesicles from the milk of three Dezhou donkey mares, 292 miRNAs were detected, including 48 potentially new ones. 73 mature miRNA sequences were also found in datasets of human and cow milk.

The study in brief

• Original title: Comparative small RNA profiling of donkey, human and bovine milk-derived extracellular vesicles reveals conserved and distinct miRNA abundance patterns
• Authors: Yan M., Gomchok D., Niangji C. et al.
• Journal: Food Chemistry: Molecular Sciences, 13, 100457 (2026)
• Country: China · Study type: Laboratory study

How it was investigated

Milk from three Dezhou donkey mares (5 years, 2nd lactation, 28–45 days after birth) from a breeding farm in Qinghai, China, was processed separately for each animal. Extracellular vesicles (EV) were enriched by centrifugation, enzymatic casein removal, precipitation, and size-exclusion chromatography; electron microscopy, particle size measurement, and marker proteins were determined only for the sample from animal D2. The small RNAs were sequenced and analyzed against the donkey genome and a miRNA database.

Important figures

• 3 – Dezhou donkey mares, sequenced separately for each animal (D1–D3)
• 84.76 million – raw sequences in total, of which 56.15 million after quality filtering
• 292 – miRNAs in the three donkey samples together
• 244 + 48 – known and potentially new miRNAs
• 203–257 – miRNAs per individual sample
• 261 / 428 / 441 – unique mature miRNA sequences in donkey, human, and cow

Results

Sequencing of the three donkey samples yielded 84.76 million raw sequences, of which 56.15 million remained after filtering; 67.84–72.02% could be assigned to the donkey genome. In total, 292 miRNAs were found (244 known, 48 potentially new), with 203 to 257 per sample.

For sample D2, electron microscopy, particle measurement (main peak 127.5 nm, median 130.0 nm, 1.30 × 10¹² particles/mL), and marker proteins showed a vesicle-enriched material. These characterizations are not available for D1 and D3.

After uniform re-analysis, 261, 428, and 441 unique mature miRNA sequences remained for donkey, human, and cow. 73 occurred in all three datasets, 127 were donkey and human together, 114 donkey and cow.

Limitations of the study

• Type and study not separable: Donkey, human, and cow milk come from three independent studies with different sampling, vesicle isolation, library preparation, and sequencing. The authors themselves emphasize that differences therefore cannot be attributed solely to the animal species; lactation stages also differed (donkey 28–45 days, human milk around day 50).
• Only three animals, one characterized sample: The donkey data are based on three mares from one farm. Whether samples D1 and D3 were comparably enriched with vesicles remains unclear, as microscopy, particle measurement, and marker proteins are only available for D2; purity and yield were not determined.
• Unequal RNA input: Due to a fixed pipetting volume, approximately twice as much RNA entered the library in D1 as in D2 and D3 (2.67 vs. 1.35 ng). RNA integrity was not checked, and the authors no longer had documentation on library concentrations.
• Figures and processes not entirely consistent: After re-analysis, 261 unique mature sequences are stated for the donkey, although previously only 244 known miRNAs were identified. The methods section only mentions a common batch for libraries and sequencing; that the vesicles were isolated in two runs is only mentioned in the limitations section.
• Only computational target genes: Target genes and metabolic pathways are based on computer predictions, not experiments. Whether miRNAs from milk vesicles survive processing and digestion or cause anything in the body was not investigated; the authors expressly draw no conclusion about an effect from this.

Level of evidence: Basic research (laboratory) – describes properties of milk, no statement about effects in humans.

Our assessment

The study describes which small RNA molecules are detectable in vesicles from raw donkey milk and cautiously compares them with published data from other types of milk. Neither an effect in humans nor a statement about individual commercial products can be derived from the detection.

Source: Yan M., Gomchok D., Niangji C., Wuren T. (2026). Comparative small RNA profiling of donkey, human and bovine milk-derived extracellular vesicles reveals conserved and distinct miRNA abundance patterns. Food Chemistry: Molecular Sciences. DOI: 10.1016/j.fochms.2026.100457 · Summarized by the DON DONKEY'S editorial team, as of October 2026.

End of the study summary. The study was conducted independently of DON DONKEY'S® and does not refer to our products. Our donkey milk is a food.