Vol. 266, No. 3 Printed in U.S.A. 1649 0022-3565/93/2663-1649$03.OO/O THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Copyright C 1993 by The American Society for Pharmacology and Experimental Therapeutics Assessment of an In Vitro Blood-Brain Barrier Model Using Several [Met5]Enkephalin Opioid Analogs1 STEVEN J. WEBER, THOMAS J. ABBRUSCATO, E. A. BROWNSON, ANDRZEJ W. LIPKOWSKI, ROBIN POLT, ALEKSANDRA MISICKA, RONALD C. HAASETH, HUBERT BARTOSZ, VICTOR J. HRUBY and THOMAS P. DAVIS OREAD Laboratories, Inc., 1501 Wakarusa Drive, Lawrence, Kansas (S.J.W.), Department of Pharmacology, University of Arizona College of Medicine, Tucson, Arizona (T.J.A., E.A.B., T.P.D.), Department of Chemistry, University of Arizona, Tucson, Arizona (AWL., A.M., R.C.H., H.B., R.P., V.J.H.) Accepted for publication April 5, 1993 ABSTRACT Confluent monolayers of primary and continuous passaged cul- tures of bovine brain microvessel endothelial cells (BMEC) have been suggested to model the blood-brain barrier (BBB). In- creased lipophilicity has been previously suggested to increase BBB penetration. The intent of this study was to examine the effect that structural modifications of the [Met5]enkephalin analog DPDPE had on lipophilicity and passage across the BMEC. The BMEC consisted of a monolayer of confluent primary BMEC grown on polycarbonate (10 m) filters. Permeability coefficients were calculated on the basis of the diffusion of peptides across the BMEC in a Side-Bi-Side5 diffusion chamber. Lipophilicity of the peptides examined was determined by using reversed-phase HPLC and calculating the capacity factor (k). Diffusion across the BMEC (for all peptides examined) was linear from 15 to 120 mm; therefore, these time points were used to calculate perme- ability coefficients. Permeability coefficients ranged from 14.34 to 92.00 cm/mm (x I 0), with [p-ClPhe4’4’biphaIin the highest. Analysis of variance coupled with the Newman-Keuls test showed significantly greater (P < .01) passage of select peptide analogs across the BMEC, including [p-ClPhe44’]biphalin, [p- CIPhe4]DPDPE and reduced DPDPE. Interestingly, upon pas- sage across the confluent monolayer, reduced DPDPE was converted to cyclized DPDPE. Calculated HPLC capacity factors ranged from 3.82 to 1 2.50. The most lipophilic peptide (highest) examined was acetylated Phe#{176}-DPDPE. Analysis of the regres- sion line of permeability coefficients plotted against capacity factors yielded a correlation coefficient of 0.745 (P < .01). The data provided in this study offer strong evidence that increasing peptide lipophilicity enhances passage across the BMEC. The greatest BMEC permeability coefficients, though not the greatest capacity factors, were obtained with peptides having a chloro- halogenation at the Phe4 residue, suggesting that factors other than lipophilicity may play a role in BMEC passage. Comparison of the permeability coefficients obtained from the BMEC system with those obtained from in vivo BBB studies suggest that the BMEC system may be very useful in predicting peptide (analog) passage across the in vivo BBB. The blood-brain barrier (BBB) acts as a selective partition between the and central peripheral nervous systems and can limit the passage of blood-borne substances into the central nervous system on the basis of molecular size, charge, and/or solubility. However, certain nutrients essential for the proper functioning of the central nervous system, such as glucose and some amino acids, are selectively transported across the BBB by facilitated carrier transport mechanisms. The selective bar- rier known as the BBB is the result of the brain microvessel endothelial cells, which have tight intercellular junctions, ex- Received for publication November 11, 1992 This work was supported by a U.S.P.H.S. grant from NIDA #DA-06284 and NIMH #MH-42600. hibit a distinct decrease in pinocytotic vesicles and lack of fenestra (Brightman and Reese, 1969; Reese and Karnovsky, 1967). Numerous endogenous factors also play an important role in the development of the BBB (Rubin et at., 1991). Several in vivo methods have been developed to characterize transport across the BBB, such as the brain uptake index (BUI) method, intravenous administration method and the brain per- fusion method (Gjedde, 1981; Oldendorf, 1970; Takesato et at., 1984). The advantage of these in vivo methods is that all physiological conditions are close to the norm; however, because of experimental protocol, several factors may be altered (i.e., cerebral blood flow and hormonal levels) and may skew the results. Another common disadvantage of studying the BBB by ABBREVIATIONS: [Met5]-Enk, Tyr-Gly-Gly-Phe-Met; DPDPE (where Pen is penicillamine), Tyr-D-PeAly-Phe-Den; ac. Phe#{176}-DPDPE,acetylated Phe#{176}-DPDPE;DPADPE, [L-Ala3]DPDPE; descarboxy DPLCE, Tyr-D-PLGly-Phe-NHCH2H2; DPLCE, Tyr-D-PLGIy-Phe-L-(ls-Phe-OH; biphalin, (Tyr-o-Ala-GIy-Phe-NH)2; [Glu4jDeltorphin, Tyr-D-Ala-Phe-Glu-Val-Val-Gly-NH2; DCDCE, Tyr-D-Cs-Gly-Phe-D-Cys-Ser-Gly-NH2.