Development and Characterization of a Human Mammary Epithelial Cell Culture Model for the Blood-Milk Barrier-A Contribution from the ConcePTION Project

Abstract

It is currently impossible to perform an evidence-based risk assessment for medication use during breastfeeding. The ConcePTION project aims to provide information about the use of medicines during lactation. The study aimed to develop and characterize an in vitro model of the blood-milk barrier to determine the extent of the milk transfer of xenobiotics, relying on either on human mammary epithelial cells (hMECs) or immortalized cell lines derived from breast tissue. The hMECs were cultured and characterized for epithelial markers; further, the ability to form an epithelial barrier was investigated. Drug transporter functionality in the cultured hMECs was analyzed with specific probe substrates. The hMECs showed an epithelial morphology and the expression of epithelial markers and tight junctions. They formed a reproducible tight barrier with a transepithelial electrical resistance greater than 400 Ωcm², unlike immortalized cell lines. Different levels of mRNA expression were detected for 81 genes of membrane transporters. Functional assays showed no evidence for the transporter-mediated secretion of medicines across the hMECs. Nevertheless, the hMEC-based in vitro model covered a 50-fold range of permeability values, differentiating between passive transcellular and paracellular-mediated transport. The cultured hMECs proved to be a promising in vitro model for biorelevance; the wide characterization of hMECs makes them useful for studying medicine partitioning in milk.

Keywords: blood–milk barrier; breastfeeding; cell membrane permeability; human mammary epithelium; in vitro barrier model; membrane transport proteins; primary cell culture; transepithelial electrical resistance.

Figure 1

Transepithelial electrical resistance (TEER) of immortalized cell lines under explored culture conditions measured over time: MCF-10A (a), PMC42-LA (b), MCF-7 (c,d) MCF-7 cells (independent replication represented in red and blue) were cultured in full growth medium without beta-estradiol containing 5% fetal bovine serum seeded at 2 × 10⁵ cells/cm² on polyethylene terephthalate (PET) transwell inserts of 0.4 µm membrane pore size (d). Points represent mean values; error bars represent standard deviation; different colors represent explored culture conditions (details are shown in Table S1).

Figure 2

Representative images of human mammary epithelial cells (hMECs) at 30% confluence from seeding showing the typical epithelial morphology and growth in cluster, 20× objective (a’), 10× objective (a’’), cells at around 80% confluence reached in one week, 20× objective (a’’’). Scale bar 100 μm. Representative images of flow cytometric analysis of E-cadherin (E-CAD) (b) and cytokeratin-18 (CK-18) (b,c) in hMECs when reaching 80% confluence. Overlay between unmarked sample (in blue) and samples marked with E-CAD (in violet) and with CK-18 (in green).

Figure 3

Monolayer integrity of primary human mammary epithelial cells (hMECs). hMECs grow compact on transwell inserts (a). Transepithelial electrical resistance (TEER, Ohm × cm²) graph in black and sodium fluorescein transport (%) in red (b). Points represent mean values; error bars represent standard deviations. Immunofluorescence staining of hMECs showing the expression of epithelial markers: cytokeratin-18 (CK18) and E-cadherin (E-CAD) (c,d), and the expression of tight junctions: zonula occludens-1 (ZO-1) and occludin (OCL) (e,f) after 35 days of culture on transwell inserts. Scale bar 50 μm.

Figure 4

mRNA array analysis of transporters present in human mammary epithelial cells (hMECs). The genes analyzed are represented as relative expression calculated as ∆Ct (mean Ct reference genes (RG)—Ct gene of interest (GI)) ± SD (n = 3).

Figure 5

Permeability of probe substrates for transport routes in human mammary epithelial cells (hMECs). Apparent permeability (Papp, 10⁻⁶ cm/s) of probe substrates for functionality of transport routes in hMECs. Three independent experiments with each three or six technical replicates are shown in red (n = 6), green (n = 6), and blue (n = 3). Boxplots represent median and interquartile range; lines represent the range and points represent potential outliers. 5(6)-Carboxy-2,7-dichlorofluorescein diacetate (CDFDA) is hydrolyzed to 5-(and-6)-carboxy-2′,7′-dichlorofluorescein (CDF) via intracellular esterases. The apparent permeability was calculated using the cumulative amount of CDF transported in function of time and the donor concentration of CDFDA. The apical-to-basolateral apparent permeability for CDFDA-CDF (green) was estimated from only two time points (i.e., linearity could not be evaluated), as the earlier time points were below the limit of quantification.