Physiological Transport Mechanisms for Peptides and Proteins at the Blood Brain Barrier

Beta Switch Program

Beta Switch Program by Sue Heintze

Get Instant Access

As opposed to the delivery strategies discussed above, which are primarily aimed at short-term application in the treatment regimens of malignant brain tumors, drug treatments of chronic degenerative disorders will require long-

term application of the therapeutic agent. This implies the need to develop a noninvasive approach for brain delivery via the systemic route. To this end, utilization of endothelial transport mechanisms is being explored in preclinical studies. For macromolecules, the uptake mechanism required for transendothe-lial passage is necessarily mediated by vesicular transport, not by a passage through pores. At this point, the analogy to the ''pseudonutrient'' approach, which utilizes nutrient transporters, such as the example with l-Dopa, is limited. In carrier-mediated uptake, the drug or prodrug that is targeted to receptors within the central nervous system is also a substrate for specific transport proteins at the BBB. The structural variations of compounds suitable for delivery are limited to the degree that is tolerated by the corresponding BBB carrier protein. Therefore, only small molecular weight drugs can exploit carrier-mediated transport. In contrast, receptor-mediated endocytosis of a peptide or protein ligand does not have the narrow size restrictions of carrier-mediated uptake through pores in the plasma membrane.

1. Receptor-Mediated Uptake

The concept of saturable, receptor-mediated uptake systems for peptides and proteins at the blood-brain barrier has evolved over the last two decades. There is now combined evidence from in vitro and in vivo studies, at the biochemical and pharmacokinetic levels as well as at the morphological level, that peptides are transported across the endothelial cells. This includes transport of compounds as structurally diverse as insulin and insulin-like growth factors (IGF-I and II) (56), transferrin (57), low density lipoprotein (LDL) (58), and leptin (59). The overall process of transendothelial passage is designated as transcytosis and is composed of binding to a luminal plasma membrane receptor, endocytosis, transfer through the endothelial cytoplasm to the abluminal side, and abluminal exocytosis into brain interstitial space (60).

The binding of insulin at the BBB is mediated by the insulin receptor a-subunit as demonstrated by affinity cross-linking of [125I]insulin to isolated human brain capillaries. Gel electrophoresis (SDS-PAGE) of the solubilized receptor revealed a band corresponding to the 130-135 kDa molecular weight expected for the glycosylated a-subunit (61). This finding fits the results of radioligand binding assays with isolated cerebral microvessels from different species including man (61-64), which showed specific binding and internal-ization of insulin. Endocytosis could be verified by the demonstration that a nonsaturable fraction of approximately 75% of the capillary binding at 37°C was resistant to a mild acid wash (61). Very similar data were obtained in primary cultures of bovine brain microvascular endothelial cells (65). These in vitro results corresponded to the measurable in vivo brain uptake of insulin under intracarotid infusion (66). In these experiments degradation of the tracer was excluded by high performance liquid chromatographic (HPLC) analysis, and evidence of parenchymal uptake beyond the vascular wall was shown by thaw-mount autoradiography of cryosectioned brain slices.

A similar set of in vitro and in vivo data attests to the expression of IGF receptors at the BBB and transport of their ligands (67-69). Apparently, there are species-specific differences in this system, since the human BBB expresses predominantly the type III IGF receptor (68), while the type II IGF receptor, which is identical to the mannose 6-phosphate receptor, is absent. In the rat, presence of IGF type I and type II receptors at the BBB has been described on the basis of receptor binding assays and in situ hybridization (67, 69).

Following the demonstration of high levels of transferrin receptor expression on rat brain microvessels with a specific monoclonal antibody (70), transferrin binding to isolated human brain microvessels was shown in radiore-ceptor studies. A saturable, time-dependent binding with a dissociation constant KD of 5.6 nM was found (71). Subsequently, the transport of transferrin through the BBB was measured in vivo. While it is obvious that transferrin is involved in the delivery of iron to the endothelial cell, there is no agreement yet in the literature on the extent to which the exocytosis of iron into brain interstitial fluid occurs in a transferrin-bound mode. Fishman et al. (57) and Skarlatos et al. (72) have reported experiments in support of significant trans-cytosis of the 80 kDa plasma iron transport protein. These studies employed 125I-labeled transferrin tracer and brain perfusion in the rat; that is, the methods applied allow for the control of transferrin concentration by avoiding admixture of endogenous plasma. This is crucial because of the high concentration of transferrin in plasma of about 25 |J,M. Therefore, the BBB transferrin receptor is saturated under physiological conditions. A 90% inhibition of the uptake was found in the presence of 10% normal rat serum in the perfusate (72), which explains the results after intravenous administration of [59Fe-125I]-trans-ferrin, where only a spurious tracer uptake in brain was found (73). In 1996, however, transcytosis of transferrin from the apical (luminal) to the basolateral (abluminal) surface also was demonstrated in an in vitro BBB model, coculture of bovine brain endothelial cells and rat astrocytes (74). There was saturable and temperature-sensitive transport of [125I]holotransferrin, with measurable transport at 37°C but not at 4°C, and no transport of iron-depleted [125I]apo-transferrin, which has low affinity to transferrin receptors. Moreover, when double-labeled [59Fe-125I]transferrin was used as a tracer, the transport of iron was found to be twice as high on a molar basis as the transport of transferrin.

That ratio corresponds to the two binding sites of transferrin for iron and provides evidence for cotransport. Another concern expressed with regard to transcytosis of transferrin at the BBB is the failure to detect TfR on the ablumi-nal plasma membrane of endothelial cells (75) when a ''preembedding'' approach at the electron microscopy level was employed. However, the detection of abluminal antigens in electron microscopy requires the application of post-embedding techniques, which in turn involves initial tissue fixation, associated with the potential loss of immunoreactivity.

This methodological problem was recently addressed in a confocal microscopy study with freshly isolated rat brain capillaries (76). The method permits omitting of the fixation step. Fluorescent immunoliposomes were syn-thesised by attachment of the anti-TfR antibody OX26 to pegylated liposomes carrying rhodamine-phosphatidylethanolamine. The high fluorescence intensity of the liposomes permitted the full exploitation of the spatial resolution of confocal microscopy, and it was possible to demonstrate the presence of TfR unequivocally on both the luminal and abluminal plasma membranes of endothelial cells. In addition, the pattern of immunofluorescence was compatible with an intracellular accumulation of OX26 liposomes in endosomal structures. These data fit well with the staining pattern seen by confocal microscopy in endothelial cell monolayers after incubation with fluorescein-conjugated holotransferrin (74). Luminal binding of the OX26 TfR antibody, its endocyto-sis, accumulation in endosomes and multivesicular structures, and abluminal exocytosis have been observed at the electron microscopic level after in vivo infusion of the monoclonal antibody conjugated with 5 nm colloidal gold (77). Figure 1 depicts these crucial steps.

Recently, substantial evidence has been accumulated to support the presence at the BBB of a transport system that is involved in the transcytosis of leptin. A short cytoplasmatically truncated leptin receptor isoform was first cloned from the choroid plexus (i.e., the site of the blood-CSF barrier). Subsequently it was shown that leptin in plasma enters brain tissue in mice by a saturable mechanism (59). The specific binding of leptin to a high affinity site (Kd = 5.1 ± 2.8 nM) could be demonstrated with isolated human brain capillaries, which also internalized the ligand at 37°C (78). At the mRNA level, in situ hybridization and evidence acquired by means of the reverse transcriptase polymerase chain reaction showed that brain microvessels express even higher amounts of the short receptor isoform message than choroid plexus (79). These receptors are also subject to regulation, inasmuch as rats on a chronic high fat diet express higher amounts of the corresponding mRNA and protein in their brain capillaries (80).

Fig. 1 Steps in the transcytosis of the TfR antibody OX26 through a brain capillary endothelial cell are shown in the electron micrographs. (From Ref. 77.) The antibody was conjugated to 5 nm colloidal gold particles and was infused into the internal carotid artery of rats. Binding of the antibody to the luminal plasma membrane is indicated by short arrows (top). The long arrows mark clusters of internalized antibodies inside vesicular structures (left). Abluminal exocytosis is indicated by the arrowhead. Scale bar = 100 nm. The scheme on the right depicts transendothelial chimeric peptide delivery. The receptor on the luminal plasma membrane binds the vector moiety and mediates endocytosis. Pharmacological effects have been shown for peptide drugs acting on a cognate plasma membrane receptor on brain cells that is specific for the drug moiety (see Sec. III.C.3). An intracellular drug effect (e.g., by antisense mechanisms) requires release of the drug from endosomal vesicles. Release may occur inside endo-thelial cells or inside brain cells. The latter demands another receptor mediated internal-ization. Abbreviations: vl, vascular lumen; bm, basement membrane; V, vector; D, drug.

Fig. 1 Steps in the transcytosis of the TfR antibody OX26 through a brain capillary endothelial cell are shown in the electron micrographs. (From Ref. 77.) The antibody was conjugated to 5 nm colloidal gold particles and was infused into the internal carotid artery of rats. Binding of the antibody to the luminal plasma membrane is indicated by short arrows (top). The long arrows mark clusters of internalized antibodies inside vesicular structures (left). Abluminal exocytosis is indicated by the arrowhead. Scale bar = 100 nm. The scheme on the right depicts transendothelial chimeric peptide delivery. The receptor on the luminal plasma membrane binds the vector moiety and mediates endocytosis. Pharmacological effects have been shown for peptide drugs acting on a cognate plasma membrane receptor on brain cells that is specific for the drug moiety (see Sec. III.C.3). An intracellular drug effect (e.g., by antisense mechanisms) requires release of the drug from endosomal vesicles. Release may occur inside endo-thelial cells or inside brain cells. The latter demands another receptor mediated internal-ization. Abbreviations: vl, vascular lumen; bm, basement membrane; V, vector; D, drug.

Evidence in support of the notion that the transcytotic pathway can accommodate rather large ''payloads'' comes from in vitro studies with lipoproteins and endothelial monolayers. The size of LDL particles ranges from about 15 to 25 nm. In contrast to chemically modified lipoproteins like acetylated LDL, which is taken up by brain capillary endothelial cells in vitro and in vivo by endocytosis but is not translocated into brain (81), native LDL undergoes apical-to-basolateral transport from the blood to the brain side of bovine brain endothelial cells in primary cultures (58). Also, a modulation of the endothelial transport in this system by cocultered astrocytes was described.

2. Absorptive-Mediated Uptake of Lectins and Cationic Peptides/ Proteins

A mechanism of brain uptake that is related to receptor-mediated transcytosis operates for peptides and proteins with a basic isoelectric point (''cationic'' proteins) and for some lectins (glycoprotein-binding proteins). The initial binding to the luminal plasma membrane is mediated by electrostatic interactions with anionic sites, or by specific interactions with sugar residues, respectively, and the transport is termed ''adsorptive-mediated transcytosis.'' Ultrastructural studies utilizing enzymatic treatment and lectins coupled to colloidal gold revealed that anionic sites and carbohydrate residues exhibit a polarized distribution on the luminal and abluminal membranes (82): negative charges are more abundant on capillaries than on arterioles or venules, and the luminal surface expresses glycoproteins with sialic acid residues, while the abluminal membrane carries heparan sulfates.

Morphologic evidence of transcytosis after intracarotid infusion of cat-ionized polyclonal bovine immunoglobulin was seen by autoradiography at the light microscopic level (83). At the electron microscopic level, conjugates of horseradish peroxidase and wheat germ agglutinin (WGA-HRP) labeled the abluminal subendothelial space of brain microvessels after intravenous administration (84). Uptake of various natural and chemically modified basic proteins through the BBB has been measured in numerous pharmacokinetic studies—forexample,with histories (85), recomninantCD4 (86) ,avidin (87), cationized albumins (88, 89), cationized polyclonal IgG (83, 90, 91), and various cationized monoclonal antibodies (92, 93). As shown in studies by Terasaki et al. (94) for the heptapeptide E-2078, small basic peptides are able to undergo adsorptive-mediated transport, too.

Native or recombinant proteins (e.g., albumin, antibodies, growth factors) can be chemically derivatized by the introduction of amine groups on accessible carboxyl side chains (95). Activation of these carboxyl groups by carbodiimide reagents is followed by coupling of hexamethylenediamine or naturally occurring polyamines like putrescine, spermine, and spermidine (96). Care must be taken not to compromise the biological activity of the cationized proteins. The applicability of site protection was demonstrated in the case of a monoclonal antibody, AMY33, which is directed against a synthetic peptide representing amino acids 1-28 of the 0-amyloid peptide of Alzheimer's disease. Cationization of the antibody in the presence of a molar excess of the specific peptide antigen prevented significant loss of binding affinity (97).

In quantitative terms, the adsorptive mechanism is distinguished from receptor-mediated uptake by lower affinity and higher capacity (98, 99). In theory, this could result in comparable overall transport rates through the BBB. In practice, the measured brain concentrations (percent injected dose per gram: %ID/g) for cationic proteins may be limited by the fact that cationized proteins show profoundly increased uptake into organs other than brain, predominantly liver and kidney (99). Widespread tissue uptake is equivalent to an enhanced systemic clearance and lower AUC, thereby limiting the amount of drug available for BBB transport. When the organ distribution of different cationized proteins is compared, varying degrees of accumulation are found in some organs (e.g., liver uptake of cationized immunoglobulins is much higher than that of cationized albumin) (99, 100).

Structure-activity relationships for the brain uptake of a small tetrapep-tide have been presented (101). The authors found that basicity and C-terminal structure were important determinants for endothelial endocytosis of the synthetic peptide 001-C8 (H-MeTyr-Arg-MeArg-D-Leu-NH(CH2)8NH2). The situation is certainly more complex for large proteins. When the brain uptake of superoxide dismutase (SOD) bearing modifications with either putrescine, spermine, or spermidine was measured, the highest PS product was found for the putrescine derivative, which has the lowest number of cationic charges (96) pointing to additional factors beyond electrostatic interactions. It remains to be seen whether detailed analyses of structural requirements for initiation of adsorptive-mediated transcytosis will identify cationic modifications that allow targeting to the vascular bed of an organ.

Toxicological and immunological consequences of long-term administration of cationized albumin have been addressed (88). It could be shown that under repetitive administration of the homologous protein (i.e., cationized rat albumin used in rats) there was no organ toxicity or deviation in blood chemistry detectable compared to a control group receiving native rat albumin. Apparently, homologous proteins are tolerated after cationization without causing the immunologic reactions and organ damage (e.g., deposition of immune complexes in the glomerulus of rats after treatment with cationized bovine albumin) found with heterologous cationized proteins (102).

The general property of cationic proteins to escape vascular barriers could also be utilized for the delivery of radiopharmaceuticals to tumors or metastasis throughout the body (93, 103) and for the treatment of viral infection with cationized antibodies (91).

Was this article helpful?

0 0
Supplements For Diabetics

Supplements For Diabetics

All you need is a proper diet of fresh fruits and vegetables and get plenty of exercise and you'll be fine. Ever heard those words from your doctor? If that's all heshe recommends then you're missing out an important ingredient for health that he's not telling you. Fact is that you can adhere to the strictest diet, watch everything you eat and get the exercise of amarathon runner and still come down with diabetic complications. Diet, exercise and standard drug treatments simply aren't enough to help keep your diabetes under control.

Get My Free Ebook


Post a comment