The blood-brain barrier exerts a strong diffusional restriction on the exchange of solutes between blood and brain [1]. It is also believed to secrete a large proportion of the brain interstitial fluid [2] which requires the regulated transport of ions such as Na⁺, K⁺, H⁺, and HCO₃^– across the membranes of the endothelial cells of the brain microvessels. A useful model for the role of acid-base transporters in this secretion has been proposed based on functional data for rat brain endothelial cells in primary culture [3]. This present work investigates functional evidence for the existence of an AE-like transporter and evidence for the molecular expression of AE2 at both the mRNA and protein level. Cells on coverslips were initially exposed to HCO₃^– buffered NaCl solution. When Cl^– was replaced by gluconate^– an alkaline shift followed [3]. This shift was absent in the presence of 250 μM DIDS. If 20 mM trimethylamine was introduced in partial replacement of Na⁺, there was a rapid alkaline shift followed by a recovery towards the original pH (rate 0.042 ± 0.006 mM s^-1, n = 3). This recovery was much slower in the absence of HCO₃^– (0.007 ± 0.001 mM s^-1, n = 3, p = 0.02), in the absence of Cl^– but presence of HCO₃^– (0.013 ± 0.002 mM s^-1, n = 3, p = 0.03) or in the presence of 250 μM DIDS (0.017 ± 0.001 mM s^-1, n = 3, p < 0.05). These are the properties expected of an AE-like transporter. Results of real-time PCR showed that AE2 mRNA was expressed at a level comparable to that of NHE1 while AE1 expression was 100 fold lower [3]. Immunocytochemistry in cultured cells using SA6 polyclonal antibody against AE1/AE2 showed a predominantly perinuclear localization reminiscent of Golgi distribution[4]. The presence of the transporter was further supported from the results obtained using western blotting. Immunohistochemistry (SA6 with fluorescent secondary) using frozen rat brain sections revealed prominent fluorescence in brain microvessels after 1% SDS antigen unmasking treatment [5] but not before. AE1 was seen in red blood cells with or without unmasking. It has been shown in other tissues [5] that SA6 detects AE2 only after unmasking. Thus the pattern observed here is evidence for AE2 in the microvessels. These results provide a molecular candidate, AE2, with properties consistent with those required to explain the functional data.