Approaches to CNS Drug Delivery with a Focus on Transporter-Mediated Transcytosis

Abstract

Drug delivery to the central nervous system (CNS) conferred by brain barriers is a major obstacle in the development of effective neurotherapeutics. In this review, a classification of current approaches of clinical or investigational importance for the delivery of therapeutics to the CNS is presented. This classification includes the use of formulations administered systemically that can elicit transcytosis-mediated transport by interacting with transporters expressed by transvascular endothelial cells. Neurotherapeutics can also be delivered to the CNS by means of surgical intervention using specialized catheters or implantable reservoirs. Strategies for delivering drugs to the CNS have evolved tremendously during the last two decades, yet, some factors can affect the quality of data generated in preclinical investigation, which can hamper the extension of the applications of these strategies into clinically useful tools. Here, we disclose some of these factors and propose some solutions that may prove valuable at bridging the gap between preclinical findings and clinical trials.

Keywords: Blood-brain barrier; CNS-targeted drug delivery; Efflux-pump inhibition; Receptor-mediated transcytosis; Ring-opening metathesis polymerization; Transient BBB disruption.

Figure 1

The three main barriers in the central nervous system (CNS), namely the meningeal or arachnoid barrier, the choroid plexus barrier and the blood brain barrier (BBB). The arachnoid and choroid plexus barriers separate the blood from the cerebrospinal fluid (CSF), and the BBB separates the blood from the interstitial fluid (ISF). At each site, the barrier is mainly formed by tight junctions that seal off the paracellular space. The blood brain barrier possesses an intricate architecture of basement membrane, mural and glial cells that work in synergy to maintain the barrier’s integrity and regulate its permeability in response to neuronal needs.

Figure 2

Major transporter proteins expressed by the endothelial cells (ECs) of the blood-brain barrier. Depicted in green are the solute-carrier bi-directional transport proteins that translocate glucose (GLUTs), large neutral amino acids (LATs), monocarboxylates, such as lactate (MCTs), and organic anion polypeptide (OATPs). Protein and peptide transporters depicted in red are the transferrin receptor (TfR), insulin receptor (IR) and low-density lipoprotein receptor-related protein (LPR). Multidrug-resistance-associated proteins and P-glycoprotein, depicted in yellow and purple, respectively, are efflux pumps that transport xenobiotics against their concentration gradient. Electrolytes are regulated by ion exchangers such as Na⁺/K⁺-ATPase pump.

Figure 3

Direct delivery of therapeutic material into the brain interstitial space using a specialized catheter. The interstitial space is occupied with interstitial fluid (ISF). Directly injected therapeutic material diffuses through the ISF and accesses different regions of brain parenchyma.

Figure 4

Direct delivery of the therapeutic material to the central nervous system (CSF) in the subarachnoid space of the spinal canal (intrathecal delivery) or brain ventricles (ventricular delivery) as time-separated boluses via an implantable reservoir. The injected therapeutic material traverses the CSF, accessing different regions of the CNS. The reservoirs are toped up by pump replacement or direct injection.

Figure 5

Direct nose-to-brain access via the olfactory pathway. The therapeutic material injected into the nasal cavity diffuses through the olfactory sensory nerves and olfactory glomeruli via passive or active transport mechanisms. It then reaches brain tissues via projection neurons, tufted or mitral cells.

Figure 6

Biological barriers that the therapeutic material encounters upon intravenous injection that can reduce the blood-circulation lifetime of the therapeutic material. This can compromise the pharmacokinetic of the therapeutic material and ultimately its effectiveness. Biological barriers include water-solubility in the blood stream, drug absorption, inadequate biodistribution, biotransformation by enzymatic degradation, and clearance by immune system and excretion by glomerular filtration.

Figure 7

Regulating drug efflux by co-administration of efflux-inhibitors with the therapeutic material. Efflux inhibitors may compete for efflux pump-mediated extrusion, which favors intraparenchymal accumulation of substrates (A), and may bind to a site distinct from the primary substrate binding site causing the protein to undergo a conformational change that blocks the substrate binding site (B) or may interact with the ATPase binding site in the efflux pump, obstructing production of energy required to transport substances against their concentration gradient.

Figure 8

Transient BBB disruption by treatment with a hyperosmotic agent. The hyperosmotic solution, typically co-injected with the therapeutic material, causes water to be withdrawn out of the ECs, resulting in their shrinkage, contraction of their cytoskeleton and disruption of their TJs, allowing the paracellular diffusion of the co-administered therapeutic material.

Figure 9

Principle of BBB disruption using MRI-guided microbubble-assisted focused-ultrasound technique (MB-FUS). (A) The patient’s head is rested in a semi-spherical ultrasound transducer integrated into an MRI scanner. The transducer is attached to a mechanical positioning system. The focused ultrasound and the magnetic resonance parameters are remotely controlled by electronic interfaces. The patient’s head is immobilized by a stereotactic frame. Overheating of the scalp, skull and brain tissue is minimized by the use of a water interface, which also acts as an acoustic coupler. (B) Pretreatment of the patient with microbubbles harnesses the acoustic power and concentrates it to the blood vessel, which attenuates acoustic power levels. Microbubbles move in the direction of the FUS wave propagation and under the influence of the FUS waves they oscillate, micro-streaming the medium surrounding them, inducing mechanical stress that disrupts the TJs between ECs. Figure 9 (A) adapted from Martin et al. [57].

Figure 10

Postulated mechanism of carrier-mediated transcytosis (CMT). Solute carrier transporter proteins (SLC) undergo conformational change from outward to inward-facing orientation, translocating their substrates across luminal and abluminal membranes from high intravascular to low intracerebral.

Figure 11

Structure of sugar-conjugates designed to highlight the structural requirements of GLUT-mediated transport of therapeutic material. Compounds 1–3 are glucose and galactose conjugates of the BBB-impermeable antidepressant 7-chlorokynurenic acid. Compounds 4–7 are glucose and galactose conjugates of l-DOPA, a dopamine precursor used for the management of Parkinson’s disease symptoms. Compounds 8 and 9 are glucose conjugates of chlorambucil, a BBB-impermeable chemotherapeutic reagent. The three studies revealed the sensitivity of this approach to the conjugate’s size and the ligation regiochemistry of its cargo.

Figure 12

Structure of large amino acid substrates of large neutral amino acid transporters (LATs). l-DOPA, a dopamine precursor, gabapentin, an anticonvulsant, and melphalan, a chemotherapeutic reagent, are the pharmacologically relevant LAT substrates. Compounds 13–15, derivatives of melphalan, were designed to highlight the structural requirements of LAT-mediated transport of therapeutic material. The study revealed the sensitivity of this approach to the conformational rigidity and regiochemistry of substituents of LATs substrates.

Figure 13

Structure of oligopeptides transported by organic anion transporting polypeptide carriers (OTAPs). Identification of DPDPE and deltorphin as OTAPs substrates can inspire the design of neurotherapeutic material or OTAPs-targeting ligands that mimic their structure.

Figure 14

Postulated mechanism of receptor-mediated transcytosis (RMT). The mechanism proceeds with a receptor–vector recognition event at the luminal membrane, followed by endosome formation and transport to the opposite abluminal membrane from which the cargo is exocytosed. Endosomal escape is the limiting step, the obstruction of which requires further understanding of receptor binding dynamics that can control intracellular transport mechanisms.

Figure 15

Structure of anti-human insulin receptor monoclonal antibody (HIRMAb) engineered to deliver erythropoietin (EPO) to the brain by exploiting insulin receptor-mediated transcytosis. In this example, the HIRMAb-EPO fusion protein was engineered by fusing the EPO-encoding gene to a HIRMAb expression plasmid. The HIRMAb tandem expression vector encodes the EPO protein linked to the C_H³ region of the HIRMAb through its amino terminus.

Figure 16

Structure of anti-rat transferrin receptor monoclonal antibody (TfRMAb) engineered to deliver methotrexate (MTX) to the brain by exploiting transferrin receptor-mediated transcytosis. The TfRMAb-MTX conjugate in this example was prepared by oxidation of the carbohydrate groups located on the Fc portion of the TfRMAb to form an aldehyde derivative, which was treated with a hydrazide-functionalized MTX forming a hydrazone-linker between MTX and TfRMAb.

Figure 17

Structure of the paclitaxel-Angiopep-2 conjugate (ANG1005) engineered to deliver paclitaxel to the brain by exploiting low-density lipoprotein receptor-related protein (LRP)-mediated transcytosis. Angiopep-2 was discovered by sequence alignment of LRP substrate proteins, such as amyloid precursor protein (APP) and aprotinin. Paclitaxel is released upon hydrolysis of ester linkages by hydrolyzing enzymes.

Figure 18

Schematic representation of the structure of rabies virus glycoprotein-derived peptide (RVG) in complex with GFP-siRNA. The conjugate was engineered to deliver GFP-siRNA to the brain by exploiting nicotinic acetylcholine receptor-mediated transcytosis. To enable the complexation of RVG to GFP-siRNA, RVG was coupled to a nona-d-arginine peptide (9dR) through a tri-l-glycine spacer.

Figure 19

Delivery of therapeutics to the brain by exploiting adsorptive-mediated transcytosis (AMT) of the delivery vectors with anionic proteoglycans, such as syndecan-4. (A) Structure of syndecan-4. Syndecan-4 is a proteoglycan that is a heavily glycosylated transmembrane protein. The protein consists of an extracellular domain that carry one or more covalently attached glycosaminoglycan (GAG) chain(s). GAG chains are long, linear carbohydrate polymers that are negatively charged under physiological conditions due to the presence of sulfate and uronic acid groups. The GAG chain allows for the interaction with a large variety of ligands, such as growth factors and enzymes. GAG chains are attached to the extracellular domain through a tetra-saccharide bridge (GlcA-Gal-Gal-Xyl). (B) Structure of SynB₁ drug conjugates of cysteamide-modified dalargin, an analgesic drug, doxorubicin, a chemotherapeutic agent, and benzylpenicillin, an antibiotic that exploits AMT to trigger brain uptake of its therapeutic cargo.

Figure 20

The pharmacokinetics and pharmacodynamics aspects that determine the fate of the therapeutic material and ability to accumulate in the diseased site at its therapeutically effective concentration. Pharmacokinetics of a therapeutic material can be improved by altering the physicochemical properties to reduce the rate of drug disposition by excretion or metabolism. The pharmacodynamics can be improved by optimizing the mechanism of action or the cellular targeting capacity. Altering pharmacokinetics or pharmacodynamics without negatively effecting the other is a challenging aspect of drug and drug-formulation design. An optimal drug or drug-formulation is a design with balanced pharmacokinetics and dynamics.

Figure 21

Nanoscopic formulations for the delivery of therapeutics to the brain. NU-0129 is a gold nanoparticle formulation whose surface is conjugated with of polyethylene glycol chains and small interfering RNAs (siRNAs) that target the Bcl-2-like protein. NU-0129 is currently under clinical trial for its antineoplastic activity and potential in the treatment of recurrent glioblastoma multiforme. SGT-53 is a cationic liposome in complex with its cargo, a genetic therapeutic material. The liposomal surface is conjugated with an anti-transferrin receptor single-chain antibody fragment that serves as the targeting vector. SGT-53 is currently under clinical trial for its potential to induce apoptosis in patients with recurrent glioblastoma. Onzeald ^® (or etirinotecan pegol) is a pegylated irinotecan formulation with a long-acting effect. Onzeald ^TM is currently under clinical trial to evaluate its efficacy at the control of brain metastases. Onivyde^® and Doxil^® are liposomal formulations that encapsulate irinotecan and doxorubicin, respectively, in an aqueous space. The liposomal surface of Onivyde^® and Doxil^® is conjugated with chains of polyethylene glycol to improve their colloidal stability and non-immunogenicity. Because of their long-acting effect, Onivyde^® and Doxil^® are under clinical trials to evaluate their efficacy at treating patients with glioblastoma. Doxil^® is also under pre-clinical investigation to assess its efficiency at the treatment glioblastoma after transient disruption of the BBB induced by microbubble-assisted focused ultrasound (MB-FUS).

Scheme 1

General design of co-polymers bearing dodeca-ethylene glycol as the hydrophilic segment, 7-(diethylamino)coumarin-3-carboxylic amide as the optical probe and triphenylphosphonium iodide as the mitochondrial targeting ligand. The block-co-polymers, poly(NorbEO₁₂Coumarin-co-NorbEO₁₂PPh₃I), were synthesized from their corresponding monomers at a coumarin:PPh₃I ratio of 4:6, 4:10 and 4:14, by ROMP mediated by Grubbs III catalyst. Mn is the number average molecular weight and measured in kDa and Ð is the ratio of the weight average molecular weight (Mw) to Mn. ^{(a, b)} Determined using GPC; samples were run in LiBr in DMF and Mn were assigned relative to a standard curve fitted using polyethylene glycol standards.