Polymeric Nanoparticles for Drug Delivery to the Central Nervous System

Abstract

The central nervous system (CNS) poses a unique challenge for drug delivery. The blood-brain barrier significantly hinders the passage of systemically-delivered therapeutics and the brain extracellular matrix limits the distribution and longevity of locally-delivered agents. Polymeric nanoparticles represent a promising solution to these problems. Over the past 40 years, substantial research efforts have demonstrated that polymeric nanoparticles can be engineered for effective systemic and local delivery of therapeutics to the CNS. Moreover, many of the polymers used in nanoparticle fabrication are both biodegradable and biocompatible, thereby increasing the clinical utility of this strategy. Here, we review the major advances in the development of polymeric nanoparticles for drug delivery to the CNS.

1. Introduction

Neurological disorders contribute to 6.3% of the global burden of disease and this number is projected to rise in the coming years due to an aging population [1]. Further, neurologic diseases are often associated with chronic disability, resulting in significant suffering for both the patient and their caregivers. The economic costs of these conditions are immense: in 2010, the global economic impact of dementias alone was > $600 billion [2]. Given the substantial public health burden, significant research efforts have been directed towards the development of improved therapies for central nervous system (CNS) diseases. However, despite these efforts, treatments for CNS diseases remain limited due to the inability of therapeutic agents to adequately cross the blood-brain barrier (BBB) [3].

The BBB serves to restrict the movement of substances between the peripheral circulation and the CNS. In doing so, the BBB plays a critical role in regulating the brain microenvironment [3–6]. The BBB is formed primarily by endothelial cells that line the cerebral microvasculature and surrounding perivascular elements. Adjacent endothelial cells form complex tight junctions with substantial transendothelial electrical resistance (TEER), creating a physical barrier which severely limits paracellular transport across the BBB. This is in stark contrast to the peripheral circulation, which exhibits a low TEER and efficient paracellular transport. In addition to the physical barrier created by the tight junctions, the BBB also contains several metabolic barriers to drug delivery. Specifically, the endothelial cells of the BBB are designed to allow minimal pinocytosis, thereby largely eliminating nonspecific transcellular trafficking. Additionally, each of the cellular constituents of the BBB expresses an array of intra- and extracellular enzymes which inactivate many compounds that attempt passage through the BBB. Finally, the capillary endothelium contains a large number of efflux transporters. Thus, the main routes for transport across the BBB are: 1) the paracellular aqueous pathway, which is restricted by the tight junctions; 2) the transcellular lipophilic pathway; 3) substrate-specific transport proteins; 4) receptor-mediated transcytosis; and 5) adsorptive-mediated transcytosis [3–6]. Due to the specificity and restrictive nature of the BBB, only lipophilic drugs with a molecular weight <500 Daltons cross the BBB in pharmacologically significant amounts; the vast majority of traditional candidate drugs do not meet this requirement [7].

In an attempt to overcome these limitations, nanocarriers have been investigated as drug delivery vehicles for CNS therapeutics [8–13]. The term nanocarriers applies to a wide variety of drug delivery vehicles including dendrimers, micelles, liposomes, nanoscale ceramics, and polymer nanoparticles. These nanocarriers have been proposed as a means to improve delivery efficiency, reduce off-target effects, improve drug kinetics, and allow delivery of a chemically diverse range of therapeutic agents. For the purposes of CNS delivery, it should be possible to optimize nanocarriers for either systemic (through the BBB) or local (behind the BBB) delivery. For systemically-delivered nanoparticles, the particles must be optimized to cross the BBB by exploiting receptor-mediated and adsorptive-mediated transcytosis pathways (Figure 1a) [3, 8]. In contrast, local delivery of nanoparticles completely bypasses the BBB and relies on convection-enhanced delivery (CED) to achieve clinically-relevant volumes of distribution (Figure 1b) [13, 14]. In this case, the delivery systems must be optimized for delivery by CED and transport through the brain interstitium. In both cases, the rate of agent release from the nanocarrier can also be optimized for each clinical application. This review summarizes the development of a particular class of nanocarriers—polymeric nanoparticles—which are promising vehicles for both systemic and local drug delivery to the CNS.

Figure 1.

a) Systemically-delivered nanoparticles cross the BBB via either receptor-mediated transcytosis, which requires the presence of specific ligands on the nanoparticle surface, or adsorptive-mediated transcytosis, which utilizes charge-based interactions. b) Locally-delivered nanoparticles bypass the BBB altogether and rely on CED to achieve adequate distribution. With CED of free drugs (top row), there is adequate initial distribution, but the drugs disappear quickly after the infusion stops. However, with CED of nanoparticles (bottom row), widely-distributed nanoparticles provide long-lasting drug release.

2. Polymeric Nanoparticles

Nanocarriers are colloidal systems that range in size from 1 to 300 nm and contain a therapeutic agent. They can be fabricated from a variety of substances, including polymers, lipids, ceramics, and carbon nanotubes. In contrast to other compositions, polymer materials have the best combination of characteristics (Figure 2): they are stable and allow high loading of many agents, they provide control over drug release kinetics, they can be readily modified to display a variety of surface-attached ligands, and many polymers have a long history of safe use in humans [15].

Figure 2.

Schematic of a functionalized polymer nanoparticle. This type of vehicle can be readily customized to suit the needs of a particular drug delivery application. In particular, the polymer, surface ligands, and encapsulated therapeutic agents can all be modified. This flexibility allows for control over release kinetics, cell targeting, and treatment strategy, respectively. Reproduced, with permission from ref. 62.

2.1 Pre-Clinical Studies

Many natural and synthetic polymers have been used to create nanoparticles for CNS delivery, including polysaccharides, proteins, amino acids, poly(ethylenimines), poly(alkylcyanoacrylates), poly(methylidene malonates), and polyesters. In general, polymer choice is determined by the therapeutic goals of the nanoparticle system. Here, we highlight the major pre-clinical, in vivo studies which have investigated the use of polymer nanoparticles for CNS drug delivery.

2.11 Systemic Delivery

As stated earlier, systemic delivery of polymeric nanoparticles to the CNS is based largely on their potential for receptor-mediated transcytosis and adsorptive-mediated transcytosis through the BBB. This process can be enhanced by the addition of cell-penetrating peptides and/or targeting ligands to the nanoparticle surface. In studies to date, the nanoparticle systems described in this section have shown the most promise for bypassing the BBB.

Poly(butylcyanoacrylate) (PBCA) nanoparticles were the first polymer-based nanoparticle system used to deliver drugs to the CNS [16]. In these first studies, PBCA nanoparticles were loaded with dalargin (a compound with opioid activity), coated with polysorbate 80, and delivered intravenously, with the goal of achieving therapeutic drug levels within the CNS. In vivo studies demonstrated that dalargin-loaded PBCA particles had an antinociceptive effect [16]. Follow-up studies, using particles that were radiolabeled for sensitive detection, demonstrated that in the absence of polysorbate 80 coating, there was a significant decrease in the number of PBCA nanoparticles that crossed the BBB [17]. As a result of this and other work, polysorbate 80 (also known as Tween 80 or polyoxyethylene-20 sorbitan monooleate) appears to enhance the CNS penetration of systemically-delivered polymer nanoparticles. Polysorbate 80, which is a commonly used emulsifier and surfactant, appears to act by a) decreasing nanoparticle clearance by the reticuloendothelial system (RES), b) interacting with BBB endothelial receptors in a manner similar to low density lipoproteins, and c) possibly modulating tight junctions and efflux transporters [18–20].

Polyesters such as poly(lactic acid) (PLA) and poly(glycolic acid) (PGA), and their copolymer poly(lactic-co-glycolic acid) (PLGA) have also been widely studied because of their history of safe use in medicine. Systemically-administered PLA nanoparticles, loaded with breviscapine (a flavonoid), have been shown to penetrate the BBB in a size-dependent manner, with larger particles (~300 nm) delivering more drug to the brain than smaller ones (~200 nm) [21]. In other work, drug-loaded PLA nanoparticles that had trans-activating transcriptor (TAT) peptide attached to the particle surface leads to increased nanoparticle transport across the BBB via bypass of efflux transporters [22]. Similar to the initial PBCA studies, surface coating of PLGA nanoparticles with polysorbate 80 and poloaxmer 188 has also been proven to improve CNS penetration [23]. Recent studies with paclitaxel-loaded PLGA nanoparticles demonstrates that surface modification with glutathione may also improve BBB bypass [24].

Several types of polymer nanoparticles with high density positive charge have been reported to cross the BBB. Chitosan is a naturally occurring biodegradable, biocompatible polysaccharide which has the ability to efficiently form nanoparticles [25]. Early work demonstrated that intranasal delivery of estradiol-loaded chitosan nanoparticles leads to significant amounts of estradiol within the CNS [26]. More recent work has demonstrated that chitosan nanoparticles can be used to deliver peptides (amyloid-beta subfragments), dopamine, and caspase inhibitors to the CNS following systemic administration [27–29]. Additionally, chitosan particles can be surface modified to display a variety of ligands for BBB bypass, including transferrin receptor antibodies [30]. Of particular interest for future efforts, at pH < 6, the amino groups of chitosan are protonated and the polymer is positively charged, making it an attractive for nucleic acid delivery [25].

Similar to chitosan, poly(ethylenimines) (PEIs) are cationic polymers that are well-suited for nucleic acid delivery. Recently, systemically-delivered, disulfide-linked PEI nanoparticles (surface modified with peptides derived from a rabies virus glycoprotein that is known to facilitate BBB passage [31]) have been shown to deliver microRNAs to the CNS [32]. Although there were size-related inefficiencies, the addition of mannitol assisted in blood-brain barrier disruption and improved delivery.

Our review of the literature—and our own experience—suggests that, in the absence of surface modification, polymeric nanoparticles have limited capacity to cross the BBB. Perhaps future research efforts will succeed in finding strategies to enhance particle transit through the BBB. Surface modification with surfactants or ligands can enhance receptor-mediated endocytosis, while surface display of positive charges can enhance adsorptive-mediated endocytosis. Transient physical disruption of the BBB via intravenous administration of hyperosmolar agents or deposition of ultrasound energy also shows some promise, although the risk of side effects of physical disruption is high. A further problem with evaluating BBB penetration is the difficulty in quantifying BBB transit, and the risk of artifacts with most experimental techniques that employed. We believe that radiolabeling techniques are best-suited to quantify the delivery efficiency of systemically administered polymeric nanoparticles. Still, these techniques will need to be carefully employed to demonstrate that systemically delivered nanoparticles can accumulate to therapeutic amounts in the brain.

2.12 Local Delivery

Local delivery of therapeutic agents in the CNS has a long history of clinical success [13, 33–36]. Local delivery of therapeutics bypasses the BBB altogether. Initial work in this field focused on the implantation of drug-loaded biodegradable polymer wafers (Gliadel®), which are able to release drugs in a controlled fashion over a prolonged period of time and resulted in modest improvements in patient survival [37, 38]. Although capable of delivering large doses of drug to the site of tumor resection over a sustained period, the drug released from the implants had limited penetration beyond the tumor margin, which could limit overall efficacy [39]. CED of drug-loaded polymer nanoparticles offers a solution to this problem. In CED, an external pressure gradient is established, typically through a syringe pump, and agents are infused continuously into the brain tissue via bulk fluid flow [40]. This can lead to distribution of therapeutics over large volumes in the brain. Although the investigation of nanoparticles for local CNS delivery has, thus far, focused largely on liposomal preparations [41–43], it is now possible to design polymer nanoparticles that can be delivered by CED [44, 45].

The first particulate systems that were used for direct drug delivery to the brain were microspheres. Polymer microspheres have been fabricated from a variety of materials for the purposes of local delivery including PLGA, poly(methylidene malonate) (PMM), poly(epsilon-caprolactone), and chitosan. These systems have been used to deliver a range of therapeutics, including cyclosporine, paclitaxel, imatinib, mitoxantrone, phenytoin, and nerve growth factor [46–51]. One advantage of microparticles, over earlier implant systems such as Gliadel®, is that the particles can be introduced without surgery. But, because particles larger than 1 micron in diameter do not move readily through the BBB or the brain interstitium [52, 53], it is difficult for microparticles to distribute through large volumes of brain tissue.

In contrast, when nanoparticles are used to deliver agents instead of microparticles, particularly nanoparticles that are less than 100 nm in diameter, CED can be used to transport the particles over clinically relevant volumes of distribution [44, 54]. For most intracranial applications, this implies achieving a volume of distribution that is >3 times the volume of infused therapeutic agent [54–56]. A recent study showed, for the first time, that drug-loaded polymer nanoparticles can be used effectively to treat disease in the brain when delivered via CED. Specifically, camptothecin-loaded PLGA nanoparticles were delivered locally and demonstrated to be effective for treatment of an intracranial tumor model [45]. Further optimization of polymer nanoparticle design, by control of nanoparticle size, charge, and surface coatings, promise to make this delivery strategy even more effective [57].

CED of polymeric nanoparticles is an attractive therapeutic strategy. Large clinical trials have already demonstrated the feasibility of intracranial CED with free drug [35]. In contrast to systemic delivery, CED has limited off target effects and no significant systemic toxicity. Moreover, with the currently available polymer nanoparticle systems, CED allows for delivery of larger amounts of therapeutic agents than systemic nanoparticle administration. In our experience, CED of polymeric nanoparticles can be achieved in large animal models, to volumes which are relevant for human clinical use. Despite this promise, intracranial CED is an invasive neurosurgical procedure, which may be impractical for some patients.

2.2 Features of an Ideal Polymer Nanoparticle Delivery System

Based on the existing data, the key features of an ideal nanoparticle system can be deduced. In general, for optimal CNS delivery, a nanoparticle should be: 1) scalable and cost-effective, 2) biocompatible/biodegradable, 3) non-toxic, 4) non-immunogenic, 5) < 100 nm in diameter, and 6) amenable to robust surface modification. For systemic delivery, nanoparticles should also: 1) be stable in blood, 2) avoid the RES, and 3) have prolonged circulation times. For local delivery, nanoparticles should also: 1) be neutral or negatively charged and 2) be infused in a slightly viscous, slightly hyperosmolar solution.

3. Current Clinical Applications

Although several clinical trials have investigated the role of nanoliposomal vehicles for CNS drug delivery, there have not yet been similar studies for polymeric nanoparticles [58]. Currently, there are two polymeric nanoparticle drug delivery systems on the market for non-CNS applications: 1) Abraxane®, an albumin-based nanoparticle loaded with paclitaxel and used in the treatment for breast cancer and 2) Abdoscan®, an iron oxide and dextran-based nanoparticle used for diagnostic imaging of the liver and spleen.

4. Conclusions

Polymeric nanoparticles have significant potential for drug delivery to the CNS. Over the past 40 years, this technology has undergone rapid expansion and is now poised for clinical translation. With the array of polymers and surface modification techniques currently available, polymeric nanoparticles have the potential to deliver not only traditional small molecule drugs, but also nucleic acids [59], proteins [60], and diagnostic agents [61]. Moreover, in comparison to other nanocarrier systems, polymeric nanoparticles are generally safer and more stable; they can also be easier to prepare and offer better control over agent release. As this technology moves forward, some of the major challenges to clinical translation will be the ability to scale-up this system in a cost-effective manner. Nonetheless, given the aging population and increasing prevalence of neurological disorders, the demand for improved CNS therapeutics is only going to increase with time. In particular, the application of polymeric nanoparticles to CNS malignancies, neurodegenerative disorders, and ischemic disease will be of interest.