There is increasing evidence that the immortalized human brain capillary endothelial cell line hCMEC/D3 is a good model to predict BBB permeability in the human brain [5, 19]. In the present study, we further investigated the utility of this model by examining whether H+/OC antiporter is functionally expressed in the cells. Although this putative H+/OC antiporter still remains to be identified at the molecular level, it is known to transport several CNS-acting drugs with secondary or tertiary amine moieties, including diphenhydramine, oxycodone, pyrilamine and clonidine [9, 10, 12].
Diphenhydramine, pyrilamine and oxycodone each form a cation at physiological pH because they are weak bases having a tertiary amine moiety. We previously showed that diphenhydramine, pyrilamine and oxycodone are taken up via a pH-sensitive, energy-dependent, proton-coupled antiport system in TR-BBB13 cells, which are an in vitro rat BBB model [9, 10]. In addition, both diphenhydramine and oxycodone have been reported to be actively taken up by the brain across the BBB with a Kp,uu (unbound concentration ratio of brain interstitial fluid to plasma) value of more than 3 in rats [10, 11]. Consequently, we have suggested that H+/OC antiporter works in the rat BBB as an active influx transporter. In the present study, diphenhydramine and pyrilamine were used as substrates to investigate this activity.
The uptakes of diphenhydramine and 3H]pyrilamine were time- and concentration-dependent. Kinetic analyses revealed that the calculated uptake clearances (Vmax/Km) for diphenhydramine (220 μL/min/mg protein) and 3H]pyrilamine (184 μL/min/mg protein) were in good agreement with those of diphenhydramine (440 μL/min/mg protein) and 3H]pyrilamine (140 μL/min/mg protein) in conditionally immortalized rat brain capillary endothelial cells (TR-BBB13) [9, 10]. These results suggest that a transporter with similar transport activity and transport clearance to those observed in the case of TR-BBB13 cells is involved in uptake of diphenhydramine and 3H]pyrilamine by hCMEC/D3 cells. Furthermore, the transporter seems to show no marked species difference between TR-BBB13 cells and hCMEC/D3 cells.
The uptake of diphenhydramine was significantly inhibited by pretreatment with metabolic inhibitors, but was insensitive to extracellular sodium and membrane potential in hCMEC/D3 cells (Table 2), suggesting the involvement of a transporter having similar energy and membrane potential dependencies to those of TR-BBB13 cells. Furthermore, the diphenhydramine uptake by hCMEC/D3 cells showed pH-dependency characteristic of a proton-coupled antiporter. The uptake was increased at higher extracellular pH (pH 8.4), and decreased in the presence of FCCP. Intracellular acidification induced with NH4Cl, stimulated the uptake (Figure 3 and Table 2). As the pK a value of diphenhydramine is 8.98, the proportion of uncharged diphenhydramine can be estimated to be 20.8% at pH 8.4, 2.6% at pH 7.4 and 0.1% at pH 6.0. Compared to the large change of the uncharged fraction (one twenty-sixth at pH 6.0 and eightfold at pH 8.4 compared with pH 7.4), the acidification (pH 6.0) or alkalization (pH 8.4) caused a small change in diphenhydramine uptake (two-fifths at pH6.0 and twofold at pH 8.4 of control uptake at pH7.4), suggesting that passive diffusion according to the pH-partition theory could not be solely responsible for diphenhydramine uptake by hCMEC/D3 cells. This view is further supported by the result that an outward proton gradient from intracellular fluid to extracellular medium effectively enhanced diphenhydramine uptake by hCMEC/D3 cells.
The results of the inhibition study (Figure 4 and Table 3) also indicate that H+/OC antiporter is functionally expressed in hCMEC/D3 cells. Diphenhydramine and pyrilamine each mutually inhibited the uptake of the other, suggesting the occurrence of competition between oxycodone and pyrilamine for a common transporter. Oxycodone also competitively inhibited diphenhydramine transport in hCMEC/D3 cells (Figure 4). A variety of organic cations with widely differing molecular structures (type II cations), such as pyrilamine, oxycodone, quinidine and amantadine, markedly inhibited diphenhydramine uptake by hCMEC/D3 cells (Figure 4 and Table 3). These results are consistent with those obtained in TR-BBB13 cells [9, 10, 20]. In contrast, TEA and serotonin, which are prototypical substrates/inhibitors of OCT1-3 and PMAT, respectively, had no significant effect. Given that the transport activities of OCT1-3 and PMAT are reduced by membrane depolarization, it is unlikely that these cation transporters are the molecular entity of the H+/OC antiporter. Low or negligible expression of hOCT1-2 mRNA in hCMEC/D3 cells also supports this idea (Table 4).
Quantitative RT-PCR analysis showed that the expression level of hOCTN2 was the highest in hCMEC/D3 cells, followed by hOCTN1, hPMAT, hOCT3 and hOCT1. On the other hand, the expression levels of hOCT2 and hMATE1-2 were negligible or low in hCMEC/D3 cells. Expression levels of these mRNAs in hCMEC/D3 cells were similar to those reported in TR-BBB13 cells  and rat brain capillary endothelial cells (RBEC1) . The H+/OC antiporter at the BBB remains molecularly unidentified even in rodents. Because hCMEC/D3 cells possess a H+/OC antiporter, like rodent BBB, and show similar mRNA expression of identified organic cation transporters to those in rat BBB model cells, hCMEC/D3 cells should be a good in vitro model for further studies on the H+/OC antiporter. Kooijmans et al. have reported that amino acid transporter B0,+ (SLC6A14) is involved in Na+- and Cl--dependent amantadine transport in hCMEC/D3 cells . Although amantadine inhibited diphenhydramine transport in hCMEC/D3 cells, diphenhydramine transport was insensitive to extracellular Na+. Thus, an unidentified transport system different from SLC6A14 seems to be a candidate for the H+/OC antiporter.
An aim of this study was to investigate whether or not the results of in vitro uptake study using hCMEC/D3 cells can be extrapolated to the human BBB. As a first step, the influx BBB permeability-surface area product (PSBBB,inf) for diphenhydramine and 3H]pyrilamine in rats were compared with those measured by the in vitro uptake study using TR-BBB13 cells. The value of PSBBB,inf measured by the in situ brain perfusion is mainly reflected in the unidirectional clearance from perfusate to brain across the BBB, as far as the BBB transport process is the rate-limiting step . The values of PSBBB,inf was estimated to be 44 and 13 μL/min/cm2 for diphenhydramine and 3H]pyrilamine, respectively, assuming that the rat brain capillary surface area is 100 cm2/g of brain . These values approximate to the in vitro uptake clearances in TR-BBB13 cells for diphenhydramine (21 μL/min/cm2)  and 3H]pyrilamine (6.3 μL/min/cm2) . These results indicate the possibility that the in vivo BBB permeability can be roughly predicted from the in vitro uptake clearance estimated by BBB model cells, as far as diphenhydramine and pyrilamine. In hCMEC/D3 cells, in vitro uptake clearance for 3H]pyrilamine is estimated to be 8.39 μL/min/cm2, which is in fairly good agreement with the in vivo human BBB of 11C]pyrilamine (15 μL/min/cm2) estimated from a positron emission tomography (PET) study . Extensive further studies and human data will be needed to allow reliable prediction of human BBB permeability from in vitro uptake studies using hCMEC/D3 cells.