Lectin histochemical characteristics of the canine female mammary gland

Abstract

SUMMARY Twelve biotinylated lectins and an avidin-biotin-peroxidase method were used to detect and localize specific carbohydrate residues on formalin-fixed, paraffin-embedded female canine mammary gland sections. Histologic sections from 3 lactating and 7 nonlactating mixed-breed dogs (age 5.6 ± 0.35 years) were incubated with Arachis hypogea agglutinin (peanut agglutunin; pna), Concanavalia ensiformis agglutinin (conA), Dolichos biflorus agglutinin (dba), Glycine max agglutinin (sba), Griffonia simplicifolia agglutinin-I (gs-I), Lens culinaris agglutinin (lca), Lycopersicon esculentum agglutinin (lea), Phytolacca americana mitogen (pokeweed mitogen; pwm), Ricinus communis agglutinin-I and-II (rca-I and -II), Triticum vulgaris (wga), and Ulex europaeus agglutinin-I (uea-I). Each lectin had a specific binding pattern, except sba and dba. In nonlactating glands, pna, conA, lea, and uea-I stained duct cells in a linear-binding pattern, with a mean percentage of positive ducts per section of 28.7 (± 0.6), 65.7 (± 0.3), 100 (± 0), and 8.4 (± 0.2), respectively. Strong apical, lateral, basal, and cytoplasmic positivity on duct cells was seen after incubation of the sections with rca-I, rca-II, and wga in all ducts. In acinar cells, the binding pattern and the staining distribution of all the lectins studied were similar to those in duct cells. However, for pna, conA, and uea-I, the mean percentage of positive lobules per section was 33.7 (± 0.9), 62 (± 0.5), and 10.5 (± 0.2), respectively. In glands from lactating dogs, conA and uea-I did not stain. The cytoplasm of all myoepithelial cells was moderately stained with rca-I, rca-II, and wga. Endothelial cells stained with gs-I, pwm, rca-I, rca-II, wga, conA, and lca. The extracellular matrix, especially the periacinar and periduct regions, and the interstitial fibroblasts were positive for lca, rca-I, rca-II, and wga. Peripheral unmyelinated nerve fibers of the nipple were strongly positive for GS-I, pwm, rca-I, rca-II, and wga. Some of the lectins used (ie, pna, conA, uea-I, gs-I, pwm, and lea) appear to have selective staining of mammary gland structures that seems to be correlated with various physiologic functions. The contrasting binding pattern of lectins specific for the same sugar indicates a lack of knowledge of interactions between lectins and carbohydrate residues in tissue sections.