Donkey milk study: Lipid profiles of ten milk types (Wei 2025)

In raw milk from eight animal species, 640 lipids were identified. Donkey milk had the lowest fat content at 0.52%; 548 of the 640 lipids were lower than in Holstein milk at the same sample volume.

The study in brief

• Original title: Non-Targeted Lipidomics Analysis of Characteristic Milk Using High-Resolution Mass Spectrometry (UHPLC-HRMS)
• Authors: Wei T., Zhou T., Zhang S. et al.
• Journal: Foods, 14(12), 2068 (2025)
• Country: China · Study type: Laboratory study

How it was investigated

Raw milk from ten milk types from eight animal species was investigated: camel (two dairies in Xinjiang), mare and goat (Xinjiang), donkey and Jersey cow (various dairies in Hebei), buffalo (Guangxi), yak (Qinghai as well as Hongyuan yak milk from Sichuan) and Holstein cow (Inner Mongolia), six replicates each. The total fat content was determined gravimetrically after solvent extraction. For lipidomics, 80 µL of milk with an internal standard was extracted and measured by ultra-high-performance liquid chromatography with high-resolution mass spectrometry in positive and negative ion mode; the assignment was made via a lipid database.

Important figures

• 10 – Milk types from 8 animal species, 6 replicates each
• 640 – Identified lipid species
• 0.52 % – Total fat content of donkey milk (discussion section)
• 3.86 % – Total fat content of Holstein milk
• 8.01 % – Total fat content of buffalo milk (highest value)
• 548 out of 640 – Lipids in donkey milk significantly lower than in Holstein milk

Results

The total fat content ranged from 8.01 % in buffalo milk, through 4.91 and 5.03 % in camel milk, 4.30 % (Jersey) and 3.86 % (Holstein), to 3.75 % (goat) and 3.59–3.91 % (yak). Mare and donkey milk were significantly lower; for donkey milk, the discussion states 0.52 %.

Across all samples, 640 lipid species were identified, including triglycerides, phospholipids, sphingolipids, ceramides and wax esters; triglycerides were the most common class in all milk types. For donkey milk, the authors describe relatively high signals of cholesteryl esters, LBPA and primary fatty acid amides, while ceramides and monoglycerides were barely detectable in some samples.

In the principal component analysis (PC1 59.8 %, PC2 20.3 %), donkey and mare milk were close together, camel milk formed its own group, buffalo milk was furthest away. In the cluster analysis of the 50 strongest differences, donkey and mare milk did not group together, which the authors explain with the low fat content of donkey milk.

Limitations of the study

• Comparison at the same volume: All samples were extracted with the same milk volume (80 µL), missing values replaced with half the detection limit. The many "lower" lipids in donkey milk therefore reflect above all the low fat content and hardly the composition of the milk fat.
• Origin of the donkey milk hardly described: For the donkey milk, only "dairies in Hebei" are mentioned. Breed, number of animals, lactation stage and feeding are missing; whether the six replicates are individual animals, tank milk or repeated measurements remains open.
• Abstract and results contradict each other: The abstract describes donkey milk as rich in cholesteryl esters and polyunsaturated fatty acids. The results show 548 out of 640 lipids lower than in Holstein milk, and fatty acids were not measured separately; statements about fatty acid ratios come from the literature.
• Numbers and information inconsistent: The methods section speaks of ten animal species, the abstract and conclusion of eight. In the comparison mare–donkey, the text mentions three lower lipids, then lists four. The fat content of donkey milk (0.52 %) only appears in the discussion, that of mare milk not at all as a number.
• Application statements not investigated: The authors' recommendations on the suitability of individual milk types for certain diets or products are based on literature and were not tested in the study.

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

Our assessment

The study provides a broad, but only roughly described lipid comparison for donkey milk. Since the samples were compared by volume and not by fat content, the results show above all the low fat content of donkey milk. Statements about effects or individual commercial products cannot be derived from this.

Source: Wei T., Zhou T., Zhang S., Quan Z., Liu Y. (2025). Non-Targeted Lipidomics Analysis of Characteristic Milk Using High-Resolution Mass Spectrometry (UHPLC-HRMS). Foods. DOI: 10.3390/foods14122068 · Summarized by the DON DONKEY'S editorial team, as of: October 2026.

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