Also, since the composition of Gibco hESFM used in the initial method is not reported in the literature, it appeared worthwhile to test simpler growth media (DMEM) supplemented with hydrocortisone. To optimise the isolation of brain microvessels, special attention was given during initial isolation to removing all meninges (including inside sulci) and most of
white matter, and this led to increased culture purity, with fewer contaminating cells growing out from the isolated vessel fragments. The extra time taken over the preparation, while slightly reducing yield, resulted in purer cultures. To reduce the ‘edge effects’ caused by leak of current around the edges of the monolayer at the circumference of the insert, larger inserts were used (12 mm diameter, hence smaller circumference:surface area ratio).
In the first series of experiments (Fig. 3A), TEER of cells grown in normal PBEC medium peaked at ∼100 Ω cm2 Ion Channel Ligand Library high throughput at 2 days and then declined. A similar pattern was seen in cells grown in PBEC medium or medium without serum, but supplementation at 48 h by adding hydrocortisone and increasing cAMPi increased peak resistance to ∼400 Ω cm2 in serum-free medium and to ∼530 Ω cm2 in serum-containing medium; in supplemented medium, especially medium containing serum, the high resistance phase lasted longer than in normal PBEC medium (Fig. 3B). Puromycin treat-ment was introduced to kill pericytes (Perrière et al., 2005). Addition of 4 µg/mL Proteases inhibitor puromycin in the first three days of growth led to a significant further improvement in purity triclocarban of the PBEC culture and a significant increase in TEER. In addition, using BPDS rather than foetal calf serum (FCS) in the culture medium also increased TEER (Fig. 4). To reduce variability of TEER observed with the STX2 chopstick electrodes, the
WPI Endohm chamber system was used, with large concentric plate electrodes above and below the insert. TEER of 485–1300 Ω cm2 (Fig. 5) was typically obtained (mean TEER=789±18 Ω cm2; n=91 inserts), with good reproducibility between vials ( Fig. 6) and batches. Furthermore, the corresponding TEER and Papp values from each batch confirm the reliability of the model, showing high TEER correlated with low [14C]suc-rose permeability ( Fig. 7). Mean Papp for [14C]sucrose was 5.7±0.7×10–6 cm/s (n=7 experiments, 3 inserts each). Further functional characterisation of this phase of the porcine BBB model is described in detail elsewhere ( Patabendige et al., this issue). Pericyte contamination was reduced by differential trypsinisation during passaging the cells before seeding onto inserts and DMEM was used with ACM (i.e. DMEM/ACM). Confluent monocultures of PBECs had an elongated cobblestone-shaped morphology, although not generally so clearly spindle-shaped as reported for rat and bovine brain endothelial cell cultures.