The aim of this study was to develop and evaluate a primary culture of choroid plexus (CP) epithelial cells as an in vitro model for studying transport processes between the plasma and the CSF. Epithelial cells were disseminated from sheep (Clun Forest) CPs by enzymatic digestion. The animals were anaesthetised with thiopenthone-Na⁺ (20-25 mg kg^-1 I.V.), killed humanely and then fourth ventricle and lateral ventricle CPs removed and kept in warm (37 °C) CO₂-independent medium. The CPs were washed and incubated with DMEM (w/o fetal calf serum (FCS), Ca²⁺, Mg²⁺) containing proteolytic enzymes and released cells were finally resuspended in DMEM/F12 medium supplemented with FCS, antibiotic/antimicotic solution and hormones and 2 X 10⁵ cells cm^-2 seeded on polyester inserts (0.4 mm pore) which were uncoated or coated with different basal lamina components. The cells were left in the incubator at 37 °C and 5 % CO₂ for 4-72 h for the attachment to the surface, then the medium was changed first time and subsequently each 2-3 days. The results showed that the digestion of tissue with 0.25 % trypsin (30 min) yielded the maximal number of the cells (3.4 ± 0.9 X 10⁶ cells (100 mg)^-1, mean ± S.D.) but < 5 % plating efficiency (PE), while mild and short digestion with pronase and trypsin released less cells but the PE was 12.2 ± 4.3 % (mean ± S.D.) (although no difference in the viability of the cells was observed). Cellular attachment and formation of the monolayer depended also on the coating of inserts. Although the lowest PE and longest time for the initial attachment (more than 24 h) was observed on laminin-coated inserts, the cells spread subsequently more rapidly on laminin, with the population doubling time 3-4 days and they made optical confluence at day 4 after seeding showing typical cobblestone-like arrangement. The plated cells maintained epithelial-like morphology and showed positive staining with a mixture of anti-cytokeratin antibodies. The electrical resistance across the monolayer increased with time and reached 87 ± 6Ω cm^-2 (mean ± S.D.) at day 8, after which no further increase was observed, while the permeability for ¹⁴C sucrose and ³H mannitol decreased from 15.0 ± 2.5 and 15.9 ± 3.0 to 3.5 ± 1.0 and 4.9 ± 1.4 X 10^-4 cm min^-1 (means ± S.E.M.), respectively, indicating that the tight junctions were developed, which was also proved by positive staining with anti-occludin antibodies. These cells seem to be highly differentiated since the immunocytochemical study revealed strong positive stain of transthyretin in the cytoplasma (mostly around the nucleus). Another sign of differentiation of these cells was the CSF secretion from the apical side that was detected by measuring the dilution of ¹²⁵I-albumin or blue dextran over 18 h incubation. When natural sheep CSF was present in the apical (CSF) compartment 12 h before and during the incubation CSF secretion was 2.32 ± 0.98 ml cm^-2 h^-1 (¹²⁵I-albumin, n = 4) and 1.46 ± 0.66 ml cm^-2 h^-1 (blue dextran, n = 5) (means ± S.E.M.), respectively, and the presence of medium with serum in the apical chamber affected that secretion. These results suggest that this primary cell culture system possesses typical choroidal epithelial characteristics and appears to be a suitable model for in vitro studies.

The Wellcome Trust supported this research.

All procedures accord with current UK legislation.