ABSTRACT: I have reviewed recent (March 1995–December
1997) papers on human milk lipids including many on fatty acid
(FA) composition. The effects of maternal diets on the proﬁles are
apparent. However, more data on the composition of milk lipids
are needed. It is noteworthy that so few papers on milk FA com-
position have reported analyses using high-resolution gas–liquid
chromatography columns. Two of these were on milk from
women in North America. The diets in North America are varied
and the number of analyses few. We do not have a reliable data
base showing the ranges of biologically important acids. Except
for the gangliosides, few new data on the other lipids appeared
during this period.
Paper no. L8263 in Lipids 34, 1243–1271 (December 1999).
In this paper, I review and evaluate papers on human milk
lipids that were published from March 1995 through Decem-
ber 1997. Some relevant earlier papers are included. It is the
latest in a series of my reviews on the subject, which began in
1978. The last was published in 1996 (1).
THE NATURE OF MILK
Milk is a very complex fluid, containing carbohydrates and
salts in true solution, caseins in colloidal dispersion, cells and
cellular debris, and lipids mostly in emulsiﬁed globules.
The lipids are triacylglycerols (TG, 98%), phospholipids
(PL, 0.8%), cholesterol (C, 0.5%), and many others. The
lipids (3–5%) occur as globules emulsified in the aqueous
phase (87%) of milk. Nonpolar lipids, e.g., TG, cholesteryl
esters, and retinyl esters, are found in the core of the globules
(1,2). The globules are covered with bipolar materials—PL
proteins, mucopolysaccharides, C, enzymes—organized into
a loose layer called the milk lipid globule membrane
(MLGM) (2). The MLGM acts as an emulsion stabilizer and
represents the membrane of the secretory mammary gland
cell. The globules range in size from 1 to 10 µM, with most
of the globules less than 1 µM, but those around 4 µM ac-
counting for the most weight. The globules present a large
surface area (500 cm
/mL) to lipolytic enzymes and other ad-
Effects of milk lipids. Based on my earlier reviews, I have
listed in Table 1 effects of milk lipids, both beneficial and
detrimental. Further discussion will be found throughout the
review. The unperturbed system delivers energy, nutrients,
protective components, and metabolic messages to the infant.
The emulsion is thermodynamically unstable, maintaining its
original compartmentation for the few minutes of nursing.
Thereafter, some constituents may change compartments, but
the original value of milk for the infants apparently remains.
However, off-ﬂavors may develop to the extent that the infant
will not consume stored milk.
Fatty acids (FA) in milk lipids (90% of TG as esters) are
the only components that can be altered to any extent by ma-
ternal dietary manipulation. This observation suggests that
the FA composition of milk may not be usable as a standard
for the preparation of formulas because the range in contents
is broad. However, profiles of milk FA related to the optimal
growth and development of the infant could be used. As of
December 1997, more analyses of milk FA by the best meth-
ods available were needed.
SAMPLING FOR AND DETERMINATION OF LIPIDS
Sampling. It is difficult to sample milk and obtain an aliquot
that contains the true amount of lipid. The heterogeneously
sized globules, which have a density of 0.9, rise at different
rates and must be shaken at ≥38°C, to achieve a random dis-
tribution. The globules must be liquid. The globules also
clump and these must be dispersed. In addition to the prob-
lems caused by the physical nature of the globules, the lacta-
tion process causes more difficulties, e.g., the lipid content
rises as a nursing proceeds. Detailed sampling protocols are
available and must be followed to obtain a representative
sample (3–5). However, not discussed in References 3–5 are
the cyclic changes in lipid content that occur in milks from
malnourished women on low-fat, high-carbohydrate diets,
relative to ingestion of meals. The lipid content changes diur-
nally, reaching maxima about 8 h after a meal. Several sam-
ples that bracket meal time should be taken and averaged to
ascertain the true lipid content.
Copyright © 1999 by AOCS Press 1243 Lipids, Vol. 34, no. 12 (1999)
*Address correspondence at 186 Chaffeeville Rd., Storrs, CT 06268-2637.
Abbreviations: BSSL, bile salt-stimulated lipase; C, cholesterol; CE, choles-
teryl ester; CLA, conjugated linoleic acid; DHA, docosahexaenoic acid; DG:
diglyceride, diacylglycerol; FA, fatty acid; FAME, fatty acid methyl ester;
FFA, free fatty acid; GLC, gas–liquid chromatography; HODE, hydroxyoc-
tadecadienoic; HPLC, high-performance liquid chromatography; LBSA,
lipid-bound sialic acid; LCPUFA, long-chain polyunsaturated fatty acids;
LDL, low density lipoproteins; MG, monoglyceride, monoacylglycerol;
MLGM, milk lipid globule membrane; MP, melting point; MMP, mean melt-
ing point; MS, mass spectrometry; PL, phospholipid; PUFA, polyunsaturated
fatty acid; TG: triglyceride, triacylglycerol; TLC, thin-layer chromatogra-
Lipids in Human Milk
Robert G. Jensen*
Department of Nutritional Sciences, University of Connecticut, Storrs, Connecticut 06269-4017