These studies demonstrate an unexpected high degree of transcytosis of a monoclonal antibody through the primate BBB in vivo.

Polyamide ("peptide") nucleic acids (PNAs) are molecules with antigene and antisense effects that may prove to be effective neuropharmaceuticals if these molecules are enabled to undergo transport through the brain capillary endothelial wall, which makes up the blood-brain barrier in vivo. The model PNA used in the present studies is an 18-mer that is antisense to the rev gene of human immunodeficiency virus type 1 and is biotinylated at the amino terminus and iodinated at a tyrosine residue near the carboxyl terminus. The biotinylated PNA was linked to a conjugate of streptavidin (SA) and the 0X26 murine monoclonal antibody to the rat transferrin receptor. The blood-brain barrier is endowed with high transferrin receptor concentrations, enabling the 0X26-SA conjugate to deliver the biotinylated PNA to the brain. Although the brain uptake of the free PNA was negligible following intravenous administration, the brain uptake of the PNA was increased at least 28-fold when the PNA was bound to the 0X26-SA vector. The brain uptake of the PNA bound to the 0X26-SA vector was 0.1% of the injected dose per gram of brain at 60 min after an intravenous injection, approximating the brain uptake of intravenously injected morphine. The PNA bound to the 0X26-SA vector retained the ability to bind to synthetic rev mRNA as shown by RNase protection assays. In summary, the present studies show that while the transport of PNAs across the blood-brain barrier is negligible, delivery of these potential neuropharmaceutical drugs to the brain may be achieved by coupling them to vector-mediated peptide-drug delivery systems.

No abstract

The hypothesis that P-glycoprotein plays a functional role at the brain capillary endothelium, which makes up the blood-brain barrier in vivo, is based largely on immunocytochemical studies showing immunoreactive P-glycoprotein localized to either isolated brain microvessels or microvessels within tissue sections. The present studies use the MRK1 6 monoclonal antibody to human P-glycoprotein to demonstrate that the pattern of immunolocalization of P-glycoprotein in microvessels of human or primate brain is similar to the pattern of immunolocalization of an astrocyte protein, glial fibrillary acidic protein. In contrast, the discontinuous staining pattern of MRK16 is not colocalized with the continuous immunostaining of the brain endothelial GLUT1 glucose transporter. The MRK16 antibody was radiolabeled with [125I]ĩ odine, and 1251-MRK16 avidly bound isolated human brain capillaries via a saturable mechanism. However, the 1251MRK16 antibody was not taken up by primate brain capillaries in vivo following intravenous injection. In conclusion, these studies provide evidence that P-glycoprotein does not play a functional role at the luminal membrane of the brain capillary endothelium in vivo, and that a principal site of immunoreactive P-glycoprotein in brain microvasculature is localized to astrocyte foot processes.Abbreviations used: BBB, blood-brain barrier; ED, effective dose; GFAP, glial fibrillary acidic protein; ID, injected dose; MDR, multidrug resistance; PBS, phosphate-buffered saline; TCA, trichloroacetic acid; V 0, organ volume of distribution.