Several lines of research support a role for human milk oligosaccharides in the defense of breast-fed infants against pathogens. Some ofthese oligosaccharides contain at least one moiety of sialic acid and are, thus, termed sialyloligosaccharides. These constitute a significant component (>1 g/L) of human milk. It is well established that milk composition varies among species, and previous reports have indicated that one ofthe differences between human and bovine milk is precisely their contents of sialyloligosaccharides. Because most infant formulas are manufactured with bovine milk components, it follows that formula-fed and breast-fed infants ingest dissimilar quantities of these carbohydrate structures. To ascertain these differences and their impact along lactation, the contents of oligosaccharide-bound sialic acids and major sialyloligosaccharides in samples of human and bovine milk (obtained at different lactation stages) were determined. In addition, infant formulas were assayed for their sialyloligosaccharide contents. Seven sialyloligosaccharides were identified in human milk; namely, 3'-sialyl-3-fucosyllactose and sialyllacto-N-tetraoses (a and b+c), the predominant structures at all lactation stages. Five sialyloligosaccharides were identified in bovine milk, of which 6'-sialyllactosamine and 3'-sialyllactose were the most abundant. In addition, sialyloligosaccharides in human and bovine milk decreased along lactation, and infant formulas did not contain significant amounts of sialyloligosaccharides. The results point to the general conclusion that regarding both qualitative and quantitative aspects, milk from humans and cows and infant formulas have different oligosaccharide contents. In this sense, bottle-fed infants are subject to reduced sialyloligosaccharide intake as compared to breast-fed infants.

Bovine milk undergoes changes in its ganglioside contents during the different stages of lactation. These contents are higher in colostrum (7.5 mg of lipid-bound Neu Ac/kg) than in transitional (2.3 mg) or mature (1.4 mg) milk. The sialic acid content of milk follows a similar profile to that of gangliosides with the highest content during the first few days post partum followed by a gradual decrease towards the end of the period studied. When the individual distribution of gangliosides was examined throughout the course of lactation, several changes were also found. G D3 is the major ganglioside (about 60-70%) found; its content decreases from the first to the fith day, increasing towards the end of the period considered. G M3 , G D 3 and GTS, sialyllactosylceramide-containing gangliosides account for 80-90% of the total lipid-bound Neu Ac content. The most striking change in the ganglioside pattern was the gradual increase in G 3 . Ganglioside in der Kuhmilch: Änderungen des Gehaltes und der Verteilung einzelner Ganglioside während der LaktationZusammenfassung: Der Gangliosidgehalt von Kuhmilch ändert sich während der verschiedenen Phasen der Laktation. Die Gehalte sind im Kolostrum (7.5 mg lipidgebundene Neu Ac/kg) höher als in Über-gangsmilch (2.3 mg) oder in reifer Milch (1.4 mg). Der Sialinsäuregehalt der Milch geht dem der Ganglioside etwa parallel, mit dem höchsten Gehalt wäh-rend der ersten Tage post partum, gefolgt von einem langsamen Abfall bis zum Ende der Versuchszeit. Bei Untersuchung der Verteilung einzelner Ganglioside während der Laktation wurden verschiedene Verän-derungen beobachtet. Das vorherrschende Gangliosid ist G D3 (60-70%); sein Gehalt nimmt vom 1. bis zum 5. Tag ab und steigt gegen Ende der Versuchszeit wieder an. Die Sialyllactosylceramid enthaltenden Ganglioside G M3 , G D3 und Gj 3 sind für 80-90% aller lipidgebundenen Neu Ac verantwortlich. Die auf fäl-ligste Veränderung war der graduelle Anstieg von G 3 .

We studied the presence and numbers of macrophages in the different compartments of the human menstrual corpus luteum (CL) in relation to the proliferative activity and apoptosis in luteal cells. Macrophages were recognized by immunohistochemical demonstration of the lysosome-associated glycoprotein CD68, and proliferating cells by immunohistochemical detection of the cell cycle-related protein Ki67 and by counting mitotic cells. In general, changes in the number of macrophages were parallel to the functional activity of the CL. Macrophage numbers increased up to the end of the early luteal phase, remained relatively unchanged during the midluteal phase, and decreased at the late luteal phase. Furthermore, macrophages showed prominent morphological changes during the cycle. They showed round or elongated cytoplasm during the early and late luteal phases, and dendritic features in the midluteal phase. Proliferating cells were very abundant on Days 15-16 and showed a significant decrease thereafter. Most proliferating cells corresponded to stromal (mainly vascular) cells. However, about 5% of granulosa-lutein cells and about 15% of theca-lutein cells were proliferating during the early and midluteal phases. Regression of the CL at the late luteal phase was associated with both a decrease in the number of proliferating cells and an increase in the number of apoptotic cells, which were highly increased on Days 25-27 of the cycle. The number of macrophages was not related to cell proliferation nor to cell death during the luteal phase. The observed changes in both macrophage number and morphology suggest the existence of a bidirectional communication between macrophages and steroidogenic cells in the human CL, or regulation of both cell populations by similar mechanisms.

Endothelial cells are the most abundant cell type in the corpus luteum (CL), and changes in blood vessels have been proposed to play a pivotal role in CL regression. We have studied quantitatively the changes in the human granulosa-luteal microvasculature in CL of various ages: young (Days 17-19 of the cycle), mature (Days 20-24), old (Days 25-27), early regressing (follicular phase of the following cycle), and late regressing (luteal phase of the following cycle). Blood vessels were identified by immunohistochemical staining for the endothelial cell marker CD34. Because of the anisotropy of blood vessels, both vertical and transverse sections of the granulosa-lutein layer (GLL) were used to estimate relative (volume, surface, and length densities) and absolute (mean cross-sectional area) vascular variables. Full luteinization from young to mature CL was accompanied by a 61% increase in the mean cross-sectional area of vascular profiles and a 52% increase in the mean volume of granulosa-lutein cells, as an estimator of changes in the volume of the GLL. In old and early regressing CL, there was a progressive increase in relative structural vascular variables, due to the shrinkage of the GLL, whereas the mean cross-sectional area of capillaries showed a 53% decrease from mature to old CL. Finally, in late regressing CL, there was a decrease in most relative structural variables, in spite of the increasingly shrunken GLL. The decrease in the capillary diameter found at the late luteal phase most likely leads to a decreased blood flow, and early changes in blood vessels could initiate and/or accelerate CL regression.

In human milk from Spanish women we observed slightly different values than those previously reported. This could be a result of population differences but nutritional or methodological aspects can not be discarded.