Drug Analogs and Prodrugs
Compound lipophilicity is a fundamental physicochemical property that plays a pivotal role in the absorption, distribution, metabolism, and excretion (ADME) of therapeutic agents. A parabolic relationship often exists between measured lipophilicity and in vivo brain penetration of drugs, where those moderate in lipophilicity often exhibit highest uptake (45). One
Drug analogs and prodrugs
Transdermal drug delivery systems (patch, cream, gel, and
Carrier-mediated CNS drug delivery
Receptor-mediated CNS drug delivery
Enzyme inhibition and absorption enhancer
Gene delivery to CNS
Direct infusion methods
Abbreviations: CDS, chemical delivery systems; CNS, central nervous system.
Abbreviations: CDS, chemical delivery systems; CNS, central nervous system.
way to promote drug permeation through the BBB is to improve the lipophilicity of the parent compound. For example, hydrophobic analogues of small hydrophilic drugs usually can be expected to more readily penetrate the BBB. However, increasing lipophilicity is not always successful. If the lipophilicity is too high, brain extraction of lipophilic compounds will decrease. Moreover, lipophilic compounds are generally more susceptible to hepatic metabolism, leading to increased drug clearance. On the other hand, very polar compounds normally exhibit high water solubility and greater excretion by the kidneys. These compounds also contain ionizable functional groups that limit the BBB penetration at physiologic pH. In other words, a delicate balance between cerebrovascular permeability and plasma solubility is required. Specifically, the optimal log Po/w is ~1.5-2.5. It should be noted that log Po/w alone seems to have limited ability in predicting brain/blood concentration ratios. However, in combination with other parameters, log Po/w can still reasonably predict brain-blood partitioning (46,47).
A prodrug strategy has been tested on various compounds, with the goal of improving CNS uptake. Colinidine analogs have been designed as lipophilic and highly selective alpha 2-adrenergic stimulants. The pharmacologic features of these agents are comparable to clonidine and alpha-methylnorepinephrine, the principal metabolite of methyldopa (48). Liphophilicity modification is not only limited to small molecules, but can be applied to peptides as well (49). One study demonstrated that point modifications of a BBB-impermeable polypeptide, horseradish peroxidase (HRP), with lipophilic (stearoyl) or amphiphilic (pluronic block copolymer) moieties considerably enhanced transport across the BBB, resulting in accumulation of the polypeptide in the brain in vitro and in vivo. The enzymatic activity of HRP was preserved following transport. Modifications of the HRP with amphiphilic block copolymer moieties through degradable disulfide links resulted in the most effective transport of the HRP across in vitro brain microvessel endothelial cell monolayers and efficient delivery of HRP to the brain (50).
Endomorphin II (ENDII), an endogenous ligand for the mu-opioid receptor, was investigated as a possible analgesic agent with fewer side effects than morphine. To improve CNS entry of ENDII, structural modification was also examined to determine whether Pro(4) substitution and cationization affected its physicochemical characteristics, BBB transport, and analgesic profile. The study showed that the structural modifications enhanced analgesic activity after intravenous administration, information that should aid in the future development of successful opioid drugs (51).
Parent drug molecules can also be modified to form a prodrug, which is usually pharmacologically inactive. The chemical change is generally intended to improve physicochemical properties of a drug, such as low membrane permeability or water solubility, and therefore enhance its delivery. After administration, the prodrug is able to readily access the receptor site, where it will be retained there. The prodrug is then converted to its bioactive form, usually via a single activation step. There are several approaches involved in prodrug formation, such as esterification or amidation of hydroxy-, amino-, or carboxylic acid- functional groups. Because these modifications will endow the prodrugs with greater lipid solubility, they will enter into the brain more easily. Once inside the CNS, hydrolysis of the modifying group will generate the compound's active form.
In order to dramatically improve lipophilicity, drug can also be conjugated to a lipid moiety, such as fatty acid, glyceride or phospholipid. For example, several acid-containing drugs, including levodopa, GABA, niflumic acid, valproate or vigabatrin have been coupled to diglycerides or modified diglycerides (52).
As mentioned earlier, while the increased lipophilicity may improve movement across the BBB, it also tends to increase uptake into other tissues, causing increased tissue burden and non-target toxicity. Furthermore, in addition to facilitating drug uptake into the CNS, lipophilicity also enhances efflux processes in the brain. This can result in poor tissue retention and a short biological duration of action. In addition, metabolites of some prodrugs may exert side effects or toxicity. As a result, these effects may decrease the therapeutic index of drugs masked as prodrugs. On the other hand, prodrug approaches that target specific membrane transporters have recently been explored. This involves designing a prodrug as a substrate for a specific membrane transporter, such as the amino acid, peptide or glucose transporters (53).
Non-steroidal anti-inflammatory drugs (NSAIDs) have been proposed to prevent or to cure Alzheimer's disease. Potential prodrugs of several NSAIDs have been synthesized in order to increase their access to the brain. Using a chemical delivery approach (described later), the carboxylic group of an NSAID molecule is attached to 1,4-dihydro-1-methylpyridine-3-carbox-ylate, which acts as a carrier, via an amino alcohol bridge. Experimental measurements show that these prodrugs were more lipophilic compared to their corresponding parent compounds and consequently a better BBB penetration is hypothesized (54). An example is 1,3-diacetyl-2-ketoprofen glyceride, a prodrug of ketoprofen and a model compound that was developed as CNS drug delivery system to treat Alzheimer's disease.(55).
Based on the physicochemical and pharmacological properties of drugs possessing an adamantane skeleton, an adamantane-based moiety was evaluated as a drug carrier for poorly absorbed compounds, including peptides that are active towards the CNS. In one investigation, various [D-Ala2]Leu-enkephalin derivatives were prepared through conjugation with an adamantane-based moiety at the C-terminus or N-terminus, and their biological activities were examined. The results suggest that introduction of the lipophilic adamantane moiety into [D-Ala2]Leu-enkephalin would improve the permeation of the poorly absorbed parent peptide through the BBB without loss of antinociceptive effect (56).
Several azidothymidine (AZT) prodrugs (conjugated with the 1-ada-mantane moiety via an ester bond) were synthesized to improve the transport of AZT into the CNS. Inclusion of a 1-adamantane moiety to AZT resulted in enhancement of the BBB penetration. This pharmaceutical approach may be beneficial for the efficient treatment of the CNS infection by human immunodeficiency virus (HIV) (57).
Tenofovir disoproxil fumarate, a prodrug of tenofovir (9-[9(R)-2-(phosphonomethoxy)propyl]adenine, PMPA), has been approved for use in the combination therapy of HIV-1 infection. PMPA can cross the choroid plexus and access the CSF, and thus may be effective towards HIV-infected perivascular and meningeal macrophages. However, PMPA is unlikely to reach the infected microglia of deep brain sites (58).
Fosphenytoin sodium (Cerebyx®), a phosphate ester prodrug of phenytoin, was developed as an alternative to parenteral formulations that utilize phenytoin sodium. Fosphenytoin has fewer local adverse effects (e.g., pain, burning, and itching at the injection site) after intramuscular and intravenous administration than parenteral phenytoin. Therapeutic effects are similar for both preparations, although transient paresthesias are more common with fosphenytoin (59,60).
Nelarabine is a water-soluble prodrug of the cytotoxic deoxyguanosine analog ara-G, which is rapidly converted to ara-G in vivo by adenosine deaminase. Nelarabine has shown activity in the treatment of T-cell malignancies, especially T-cell acute lymphoblastic leukemia. The excellent CSF penetration of nelarabine and ara-G supports further investigation into the contribution of nelarabine to the prevention and treatment of CNS leukemia (61).
A metabolically stable and centrally acting analog of pGlu-Glu-Pro-NH2 thyrotropin-releasing hormone ([Glu2]TRH), a tripeptide structurally related to thyrotropin-releasing hormone (TRH), was designed by replacing the amino-terminal pyroglutamyl residue with a pyridinium moiety. The novel analog maintained its antidepressant potency, but showed reduced analeptic action compared to [Glu2]TRH. Thus, an increase in the selectivity of CNS-action was obtained by the incorporation of the pyridinium moiety (62).
A promising prodrug has been developed to treat Parkinson's disease. SPD148903 represents a new type of prodrug. The compound has an enone structure that undergoes oxidative bioactivation to the catecholamine S-5,6-diOH-DPAT, which is delivered enantioselectively into the CNS. S-5,6-diOH-DPAT displays mixed dopamine D(1)/D(2) receptor agonist properties similar to apomorphine. This concept has the potential to improve therapeutic outcomes in Parkinson's disease by complimenting L-dopa therapy, the current treatment of choice (63).
A prodrug approach was used to enhance CNS delivery of the antiviral medications 2',3'-dideoxyinosine (ddI) and nelfinavir. In addition, inhibition of P-gp efflux was initiated to further increase the CNS delivery of the medications. Overall, preclinical studies in rats showed that this therapeutic strategy increased the brain/plasma ratios of both ddI and nelfinavir (64). Enhanced CNS delivery of certain poorly penetrating 2' ,3'-dideoxynucleo-sides has been achieved by designing prodrugs that are substrates for enzymes, such as adenosine deaminase (ADA), that are present at high activities in brain tissue (65).
It should be noted that not all prodrugs work better than their parent compounds. For example, the esters of chlorambucil did not exhibit superior anticancer activity than equimolar chlorambucil in a rat model of brain-sequestered carcinosarcoma (71). An aza-analogue of furamidine, 6-[5-(4-amidinophenyl)-furan-2-yl] nicotinamidine (DB820), has potent in vitro antitrypanosomal activity. However, the prodrug suffers from poor oral activity because of its positively charged amidine groups (72). In addition, increased lipophilicity alone does not ensure that a given prodrug will deliver higher levels of a parent compound to the CNS. Both the selectivity and absolute rate of bioconversion in the brain are important factors (73).
Transdermal Drug Delivery Systems (Patch, Cream,
Gel, and Microemulsion)
Transdermal drug delivery systems (TDDS) or transdermal therapeutic systems (TTS) are convenient dosage forms in terms of application, patient compliance and readily withdrawal of drug (if desired). Several CNS medications have been formulated into various transdermal systems for local and prolonged delivery. Examples of these approaches are detailed below.
Hyoscine (scopolamine), a competitive inhibitor of the muscarinic receptors of acetylcholine, has been shown to be one of the most effective agents for preventing motion sickness. However, a relatively high incidence of side effects and a short duration of action restrict the usefulness of this agent when administered orally or parenterally. To address these drawbacks, a novel transdermal preparation of hyoscine was developed. Pharmacokinetic studies indicate that transdermal administration delivers the drug into the systemic circulation at a controlled rate over an extended period (72 h), providing a means of delivery that is similar to a slow intravenous infusion. Therapeutic trials demonstrated that a single transdermal hyoscine patch is significantly superior to placebo and oral meclizine in preventing motion sickness. Thus, evidence suggests transdermal hyoscine may offer an effective and conveniently administered alternative for the prevention of motion-induced nausea and vomiting in certain situations (70).
Another transdermal therapeutic system for scopolamine (TTS-S) was developed to overcome the adverse effects and short duration of action seen when scopolamine when is administered orally or parenterally. The system contains a drug reservoir (1.5mg) that is programmed to deliver 0.5mg over a 3-day period. The TTS provides the approximate functional equivalent of a 72-h slow intravenous infusion of drug (71,72).
Rotigotine (Neupro) is formulated as a transdermal delivery system designed to provide a selective, non-ergot D3/D2/D1 agonist to the systemic circulation over a 24-h period in patients with Parkinson's disease. This formulation is described in detail later in this chapter.
A TDDS containing fentanyl (Duragesic®) is available to treat moderate to severe pain. Transdermally-administered fentanyl exhibits a favorable pharmacokinetic profile, an is a standard therapy for chronic pain associated with cancer and other diseases (73).
Nicotinic receptor dysfunction and impaired semantic memory occur early in patients suffering from Alzheimer's disease. Previous research indicated that the ability of nicotine to enhance alertness, arousal, and cognition in a number of nonclinical populations was a function of its ability to stimulate CNS nicotinic cholinergic receptors. Studies showed that transdermal administration of nicotine increased both regional cerebral glucose metabolism (rCMRglc) and semantic memory (as assessed by verbal fluency) (74). The results point to a possible alternative indication for transdermal nicotine therapy.
A submicron emulsion of the benzodiazepine diazepam, a lipophilic molecule with CNS activity, was investigated for efficacy as a novel transdermal formulation. Diazepam was formulated in various topical regular creams and submicron emulsion creams of different compositions. The different formulations were applied topically and protection against pentamethylenetetrazole-induced convulsive effects in mice was monitored. The efficacy of diazepam applied topically in emulsion creams strongly depended on the oil droplet size and to a lesser degree on formulation composition. Preparing the emulsion with a high-pressure homogenizer caused a drastic reduction in the droplet size, significantly increasing the activity of diazepam following transdermal application. Combined with high-pressure homogenization, the presence of lecithin (an efficient dispersant) into the formulation resulted in effective droplet size reduction to below 1 ^m (100-300 nm). Overall, submicron emulsions were found to be effective vehicles for transdermal delivery of diazepam, generating significant systemic activity of the drug compared with regular creams or ointments (75).
Two 5-nitroimidazoles compounds, MK-436 and fexinidazole, were formulated as gels by the addition of hydroxypropylcellulose. When used in combination with melarsoprol, the formulation was able to cure experimental murine CNS-trypanosomiasis with one-day therapy. Likewise, combined melarsoprol/MK-436 gel successfully cured experimental CNS-trypanosomiasis with a single treatment. Topical application of melarso-prol/MK-436 gel also eliminated hind leg paralysis that associated with post-treatment reactive encephalopathy caused by non-curative treatment of CNS-trypanosomiasis (76,77).
Basic fibroblast growth factor (bFGF) is a morphogenic, chemotactic, mitogenic, and angiogenic peptide found within the CNS with potent neurotrophic effects. A potential role in ischemia-induced vascular growth has been suggested for bFGF. Research showed that single, topical administration of human recombinant basic fibroblast growth factor (rbFGF) in the rat cerebral cortex promoted capillary overgrowth, an effect that might mimic the angiogenic response observed after brain ischemia. This finding has potential implications in angiogenic therapy (78).
Another approach used to increase lipophilicity of a CNS-active drug without modifying its molecular structure is liposomes. Liposomes are spherical vesicles with membranes composed of a phospholipid bilayer. For drug delivery, liposomes are able to encapsulate a hydrophilic drug molecule inside its double layer-structure. The transport capacity of liposomes could be enhanced by subsequent conjugation of the liposome to a BBB drug delivery vector. It is well established that liposomes, even small unilamellar vesicles, generally do not undergo significant transport through the BBB in the absence of vector-mediated drug delivery (79). Another disadvantage of liposomes as a drug delivery system is that these structures are rapidly removed from the bloodstream following intravenous administration, owing to uptake by cells lining the reticulo-endothelial system or the mononuclear phagocytic system. Incorporation of PEGylation technology and chimeric peptide technology into the design of these systems helps to mediate BBB transport and inhibit peripheral clearance of liposomes (Fig. 3).
A bi-functional PEG 2000 derivative was synthesized that contained thiolated murine monoclonal antibody (MAb) and a distearoylphosphatidy-lethanolamine moiety, to incorporate into the liposome surface. This combined technology resulted in the construction of PEGylated immuno-liposomes that are capable of receptor-mediated transport through the BBB in vivo (79). After administration, MAb binds to the BBB transferrin receptor, which has been successfully used as a vector in delivery of other large molecules across the BBB. The immuno-liposome delivery system has the ability to dramatically increase brain drug delivery by up to four orders of magnitude. This delivery system may be of significance to brain drug delivery because it permits brain targeting of the liposomally encapsulated drug, and may consequently offer a significant reduction in side effects. Compounds with excellent neuro-pharmacologic potential in vitro that may have been rejected for clinical use because of low brain delivery (or some minor side-effects) may now be reevaluated for potential use in conjunction with this delivery system.
Liposomes can also overcome some disadvantages of prodrugs. Since the liposome capsule undergoes degradation to release its contents, drug is delivered without the use of disulfide or ester linkages, which may significantly
Figure 3 Schematic representation of a liposome. Liposomes are spherical vesicles with membranes composed of a phospholipid bilayer. Liposomes can be used as drug carriers and loaded with a great variety of molecules, such as small drug molecules, proteins, and nucleotides. As shown in the figure, liposomes are extremely versatile and can be used in number of applications, including CNS drug delivery.
Figure 3 Schematic representation of a liposome. Liposomes are spherical vesicles with membranes composed of a phospholipid bilayer. Liposomes can be used as drug carriers and loaded with a great variety of molecules, such as small drug molecules, proteins, and nucleotides. As shown in the figure, liposomes are extremely versatile and can be used in number of applications, including CNS drug delivery.
affect pharmacological actions (80). The surgical delivery of therapeutic agents into the parenchyma of the brain is problematic because it has been difficult to establish the biological fate of the material after infusion, or to determine a suitable dose. Gadoteridol (GDL) loaded liposomes have been used as tracer agents that allow one to track material in real-time using MRI. MRI allows precise tracking and measurement of liposomes loaded with markers and therapeutics in the CNS (81).
Convection-enhanced delivery (CED) is a recently developed technique for local delivery of agents to a large area of tissue in the CNS. CED was combined with a highly stable nanoparticle/liposome containing CPT-11 (nanoliposomal CPT-11) to provide a dual drug delivery strategy for brain tumor treatment. CED of nanoliposomal CPT-11 greatly prolonged tissue residence while also substantially reducing toxicity, resulting in a highly effective treatment strategy in preclinical brain tumor models (82,83).
Liposomes loaded with GDL, in combination with CED, offer an excellent option to monitor CNS delivery of therapeutic compounds using MRI (84,85). Saposin C is one of four small lipid-binding proteins that derive from a single precursor protein, named prosaposin (PSAP). PSAP has several neuronal effects, including neurite outgrowth stimulation, neuron preservation, and nerve regeneration enhancement. Delivery of saposin C ex vivo into cultured neurons and in vivo into brain neuronal cells in mice across the BBB was accomplished with intravenously administered dioleoylphosphatidylserine liposomes. These studies may yield a new therapeutic approach for neuron protection, preservation, and regeneration (86).
High-dose glucocorticosteroid hormones are a mainstay in the treatment of relapses in multiple sclerosis. Ultrahigh doses of glucocorti-costeroids were delivered to the CNS of rats with experimental autoimmune encephalomyelitis (EAE) using a novel formulation of polyethylene glycol (PEG)-coated long-circulating liposomes encapsulating prednisolone (pre-dnisolone liposomes, PL). PL is a highly effective drug therapy for EAE, and it was found to be superior to a 5-fold higher dose of free methylprednisolone, possibly due to drug targeting by PL. These findings may have implications for future therapy of autoimmune disorders such as multiple sclerosis. (87).
Mirfentanil hydrochloride, a novel CNS analgesic with a short duration of action, was successfully encapsulated in liposomes of varying composition. The lipid composition of the formulation was varied to optimize the stabilization of liposomes and the encapsulation of solute. Only 35% of encapsulated drug was released when liposome formulations containing monosialoganglioside (GM1) were incubated with human plasma over a 24-h period, suggesting that these systems could be used for controlled drug delivery in vivo (88).
In addition to liposomes, the use of other nanocarriers, such as solid polymeric and lipid nanoparticles, may be advantageous over existing CNS delivery strategies. Not only can these nanocarriers mask the BBB uptake-limiting characteristics of the therapeutic drug molecule, but they also protect the drug from chemical/enzymatic degradation, and provide opportunity for sustained release. Reduction of toxicity to peripheral organs can also be achieved with these nanocarriers (89). Polymeric nanoparticles have been used to enhance therapeutic efficacy and reduce toxicity of a variety of medications (98), as well as to enhance delivery of imaging agents (90). Drugs that have successfully been transported into the brain using this carrier include dalargin (a hexapeptide), kytorphin (a dipeptide), loperamide, tubocurarine, MRZ 2/576 (an NMDA receptor antagonist), and doxorubicin (91). The nanoparticles may be especially helpful for the treatment of the disseminated and very aggressive brain tumors. Intravenously injected doxorubicin-loaded polysorbate 80-coated nano-particles were associated with a 40% cure rate in rats with intracranially transplanted glioblastomas (92).
The mechanism of the nanoparticle-mediated transport of the drugs across the BBB has not been completely elucidated. The most likely mechanism is endocytosis by the endothelial cells lining the brain blood capillaries. Nanoparticle-mediated drug transport to the brain depends on the overcoating of the particles with polysorbate molecules such as polysorbate 80. Overcoating with these materials presumably results in adsorption of apolipoprotein E from blood plasma onto the nanoparticle surface. As a result, the particles seem to mimic low density lipoprotein (LDL) particles that can interact with the LDL receptor, leading to their uptake by the endothelial cells. Subsequently, the drug may be released in these cells and diffuse into the brain interior, or the particles may be transcytosed. Other processes such as tight junction modulation or P-gp inhibition also may occur. Moreover, these mechanisms may run in parallel or may be cooperative, thus enabling drug delivery to the brain (93-97). Research shows that PEGylated n-hexadecylcyanoacrylate nanoparticles, made by PEGylated amphiphilic copolymer, penetrate into the brain to a larger extent without inducing any modification of BBB permeability (98). Cationic bovine serum albumin (CBSA) conjugated with poly(ethyleneglycol)-poly(lactide) (PEG-PLA) nanoparticle (CBSA-NP) has been studied for brain drug delivery. CBSA-NP was shown to be preferentially transported across the BBB with minimal toxicity, suggesting the possibility to deliver therapeutic agents to the CNS using this carrier (99).
Nanoparticle drug delivery systems have also been studied for delivery of HIV/AIDS medications, to facilitate complete eradication of viral load from reservoir sites in the body (100). Moreover, iron oxide based contrast agents have been investigated as more specific MRI agents for diagnosing CNS inflammation. Ferumoxtran-10, a virus-size nanoparticle, is taken up by reactive cells and is used to visualize phagocytic components of CNS lesions, a useful tool for monitoring therapy in patients (101). Iron oxide nanoparticles can therefore be used as a marker for the long-term noninvasive MR tracking of implanted stem cells (102).
Metal ions accumulate in the brain with aging and in several neurodegenerative diseases. Aside from Wilson's disease (the copper storage disease), recent attention has focused on the accumulation of zinc, copper, and iron in the brain in Alzheimer's disease, and the accumulation of iron in Parkinson's disease. Nanoparticles have been demonstrated to deliver D-penicillamine to the brain, as well as to reduce metal ion accumulation in Alzheimer's disease and other CNS disorders (103).
Nanoparticles have also been studied as transport vectors for peptides. Among these is the enkephalin analog dilargin. Nanoparticles were formulated with colloidal polymer particles of poly-butylcyanoacrylate with dilargin absorbed onto the surface. The particles were then coated with polysorbate 80. Intravenous injections of the vector-dalargin produce analgesia, while dalargin alone does not (104).
Overall, nanoparticle formulations appear to have no effect on primary BBB parameters in established in vitro and in vivo BBB models (105). Additionally, there is little in vivo or in vitro evidence to suggest that a generalized toxic effect on the BBB is the primary mechanism for drug delivery to the brain (106). Using these systems, however, the possibility of a general toxicity is still a serious concern for nanoparticle systems, and this requires further study (107).
Chemical delivery systems (CDS) represent a method to target drug to specific sites or organs in the body based on predictable enzymatic activation. After certain chemical modifications, compounds are transformed into inactive chemical derivatives. The newly attached moieties usually form monomolecular units, and provide site-specific or site-enhanced delivery of the drug through multi-step enzymatic and/or chemical transformations. CDS is similar to a prodrug approach, differing by the fact that multi-step activation and targetor moieties are involved. Using the general CDS concept, successful deliveries have been achieved not only to the brain, but also to the eye, and to the lung (108,109).
By converting a lipophilic drug into a lipid-insoluble molecule in the brain, permeation of the molecule out of the brain is diminished. If the same conversion also takes place in the rest of the body, this accelerates peripheral elimination and improves targeting. It should be emphasized that CDS not only achieves delivery to the brain, but it also provides preferential delivery, which means brain targeting. Ultimately, this should allow administration of smaller doses, reduction of peripheral side effects, and prolonged therapeutic response. Therefore, CDSs can be used not only to deliver compounds that otherwise have no access to the brain (e.g., steroid hormones and peptides), but also to retain lipophilic compounds within the brain (110).
Anionic chemical delivery systems have been developed to deliver testosterone and zidovudine to the brain via phosphonate derivatives (111). When incorporated into a bulky molecule, a peptide unit can be delivered by CDS to provide direct BBB penetration, and avoid recognition by peptidases in the plasma (112-115). To achieve delivery and sustained activity with such complex systems, it is important that the targeted enzymatic reactions take place in a specific sequence. Upon delivery, the first step must be the conversion of the targetor to allow for "lock-in." This must be followed by second step (removal of the L function) to form a direct precursor of the peptide that is still attached to the charged targetor. Subsequent cleavage of the targetor-spacer moiety finally leads to the active peptide. Several CDS have been synthesized for the cholinesterase inhibitor 9-amino-1,2,3,4-tetrahydroacridine (THA). In vivo distribution studies indicate that elevated and sustained levels of the pyridinium quaternary ion derivative were present in the CNS. In addition, THA was generated in the CNS from the quaternary salt precursor at low concentrations, indicating a slow but sustained release. The CDS for THA were found to be less acutely toxic compared to THA alone (116).
A CDS system based on the redox conversion of a lipophilic dihydropyridine to an ionic, lipid-insoluble pyridinium salt has been developed to improve the access of therapeutic agents to the CNS. A dihydropyridinium-type CDS or a redox analog of a drug is sufficiently lipophilic to enter the brain by passive transport, where it then undergoes enzymatic oxidation to an ionic pyridinium compound, which promotes retention in the CNS. At the same time, peripheral elimination of the entity is accelerated due to facile conversion of the CDS in the body (117). This methodology has been extended to deliver neuroactive peptides such as enkephalin to the brain, and has demonstrated promise in laboratory models (118).
There are several other CDS-based approaches of note. A zidovudine-CDS complex is capable of delivering higher zidovudine doses to lymphocytes and neural cells, resulting in improved antiretroviral activity. This represents a potential therapy for AIDS dementia (119). CDSs based on a dihydropyridine-quaternary pyridinium ion redox system, analogous to the naturally occurring NADH-NAD+ system, were synthesized for a group of staphylococcal penicillinase resistant penicillins (e.g., methicillin, oxacillin, cloxacillin, dicloxacillin) in order to improve their penetration of the CNS. In vivo distribution studied in rats and rabbits demonstrated BBB penetration of the compounds by CDS, whereas no drug was detected in the brain following direct drug administration (120).
Given the nature of the BBB and BCB, drug transport to the brain is a transcellular process, with lipophilicity being a primary determinant of the process. For a number of compounds, however, lipophilicity does not correlate with CNS penetration; that is, drug uptake across the BBB is lower than expected (121). This discrepancy can be explained by the presence of transport systems that may function in CNS uptake and efflux of xenobiotics (Fig. 2). Understanding the key features of these pathways may allow for improved treatment of diseases of the CNS though enhanced uptake of neuropharmaceuticals. Furthermore, CNS-related side effects of medications could be avoided by blocking these mechanisms.
Various transport systems can be targeted to promote drug uptake into the brain. Among these are the moncarboxylic acid transporters (MCT), of which at least eight isoforms have been identified (122). MCTs are involved in bidirectional membrane transport across the BBB. MCT1 has been identified on the basolateral membrane of brain capillary endothelial cells, where it transports lactic acid and other short chain acids. The pH dependence of this acidic system suggests a proton cotransport mechanism is involved. Medications bearing a carboxylic acid moiety appear to be substrates for MCT1. One such class of medications is HMG-CoA reductase inhibitors (e.g., lovastatin and simvastatin). CNS distribution of the less lipophilic drug pravastatin suggests a lower affinity for the transporter, resulting in a reduced incidence of CNS side effects. Overall, an understanding of the role of MCTs in the BBB is still evolving.
Another family of transporters implicated in CNS uptake is the OATP family. Two isozymes that have been identified are OATP1 and OATP2. OATP1 is expressed in the luminal membrane of the BCB, while OATP2 is expressed in the basolateral membrane of the BCB and the brush border membrane of the BBB. Transport across OATP2 is bi-directional (123).
In addition to passive transport (a function of lipophilicity and molecular weight), at least two distinct carrier mediated transport mechanisms have been identified for CNS transport of organic cations. Analogous to MCT1, structural differences among a class of compounds such as the H1-antagonists alters the affinity for these systems and, consequently, the side effect profile of individual medications (122).
Receptor mediated transport (RMT) is a potential pathway for delivery of large molecule peptides or proteins to the brain. During RMT, the molecule is shuttled across the BBB into brain interstitial fluid by specific receptors. Among the receptors that have been identified is the endothelial transferrin receptor (TfR). TfR is a bidirectional transporter, mediating the transport of halo-transferrin (blood/brain) and apo-transferrin (brain/blood). Additionally, endocytotic mechanisms exist on the BBB that mediate uptake into the CNS. Among these, the type I scavenger receptor (SR-BI) is involved with uptake of low density lipoprotein into the brain (124).
Both absorptive-mediated and receptor medicated endocytosis represent promising routes for peptide delivery. For example, complex-ation of neuroactive peptide with cationized albumin could facilitate translocation into the CNS through absorptive endocytosis (125). Alternatively, monoclonal antibodies to a specific receptor could be used to enhance CNS uptake via chimerization (126,127). Covalent attachment of a poorly permeable compound to a suitable vector could enhance brain uptake.
Chimeric peptide technology, wherein a non-transportable drug is conjugated to a BBB transport vector, has been proposed. A BBB transport vector is a modified protein or receptor-specific monoclonal antibody that undergoes receptor-mediated transcytosis through the BBB in vivo. Different approaches for linking drugs to transport vectors may be broadly classified as belonging to one of three classes: chemical, avidin-biotin, or genetic engineering (128). Multiple classes of therapeutics have been delivered to the brain with the chimeric peptide technology, including peptide-based pharmaceuticals, such as a vasoactive intestinal peptide analog or neurotrophins such as brain-derived neurotrophic factor, antisense therapeutics including peptide nucleic acids, and small molecules encapsulated within liposomes.
This strategy has been called Molecular Trojan Horse Technology, and it offers potential for ferrying molecules across the BBB (129). Specific substances can be delivered to the brain by attaching them to a protein that is normally able to cross the barrier. Conjugation of a peptide or antisense therapeutic to a BBB molecular Trojan horse has been shown to generate CNS effects in vivo after intravenous dosing compared with control (administration of therapeutic agent alone) (124).
The successful delivery of a drug through the BBB in vivo requires special molecular formulation of the drug. Therefore, it is important to merge CNS drug discovery and delivery as early as possible in drug development process (130). A disulfide linker can be used to deliver the drug after disulfide cleavage in brain (131). A non-cleavable linkage such as an amide bond can also be used to attach the drug to the transport vector. Cleavage is achieved by reduction of the disulfide bond, and all the bonds including amide bonds are ultimately hydrolyzed in the lysosomal compartment. PEGylation technology is used with a longer spacer arm comprised of a PEG moiety. The placement of this long spacer arm between the transport vector and the drug removes any steric hindrance caused by attachment of the drug to the transport vector, and drug binding to the cognate receptor is not impaired (128).
As noted above, lipophilicity remains an important determinant of drug permeation through the BBB. However, there a number of compounds that display poor CNS distribution despite high lipophilicity. While plasma protein binding and molecular size of these compounds are possible explanations for these findings, accumulating evidence (e.g., transgenic animal studies) points to the presence of efflux mechanisms in the BBB and BCB that restrict access of these compounds to the CNS. Most notable among these transport mechanisms is P-gp (17).
The identification of P-gp on the apical surface of brain capillary endothelial cells suggests a role of this transport system in limiting access to the CNS. P-gp is also expressed in the apical side of the BCB. Indeed, a number of in vitro and in vivo studies support the general concept that P-gp pumps xenobiotics from the brain interstitial fluid into the blood. Much of the information regarding P-gp-mediated CNS efflux has been generated using transgenic mice lacking P-gp. The various P-gp "knockout" mouse models are mutant strains of mice that are genetically deficient in MDRla, MDRlb (or both genes). Knockout mice do not display physiologic abnormalities and have life span comparable to normal mice. However, CNS distribution of a number of compounds is increased dramatically in knockout mice. This is perhaps best illustrated by the pesticide ivermectin. While ivermectin has a good safety profile in normal mice, it causes lethal neurotoxicity in knockout strains (122). Studies with knockout mice have demonstrated a role of P-gp in limiting the CNS penetration of medications used to treat HIV (e.g., indinavir, ritonavir), cancer (e.g., doxorubicin, vincristine), cardiovascular disorders (e.g., digoxin, quinidine) and pain (e.g., morphine).
In addition to P-gp, other efflux transporter systems exist in the CNS. Like P-gp, BCRP is drug efflux transporter belonging to the ABC family. Substrate specificity of BCRP shows considerable overlap with P-gp, and this indicates a similar role of this transporter on drug pharmacokinetics (17).
Inhibition of P-gp and/or BCRP is being investigated as a potential strategy to improve CNS penetration and delivery of drugs to the brain. This approach may be particularly useful in the treatment of brain tumors and CNS metastases. Preclinical studies have demonstrated that CNS penetration of anticancer drugs (e.g., paclitaxel, docetaxil) can be improved by co-administrating with a P-gp inhibitor (e.g., valspodar, elacridar). Similar results may be expected for BCRP (132).
In terms of oral drug delivery, new insights regarding the role of the intestine as a selective barrier to drug absorption have emerged. Numerous membrane transport systems are present in the intestine to facilitate the absorption of essential nutrients, systems that may also be responsible for oral absorption of certain classes of medications. Conversely, transporters in the enterocyte also serve as detoxification mechanisms in the body, which limit bioavail-ability through intestinal exsorption. By understanding the membrane transport mechanisms involved in oral drug absorption, strategies can be developed to enhance drug delivery of poorly bioavailable compounds.
The efflux transporters P-gp and BCRP are known to be expressed in the intestine, where they actively extrude a variety of compounds. Inhibition of these transporters, therefore, is a logical strategy to improve oral bioavailability. Preclinical studies in mice found that the bioavailability of topotecan increased from 40% to 97% upon co-administration with a P-gp inhibitor (132).
Intestinal enzyme inhibition may also be an effective tool to increase the oral bioavailability of compounds that undergo first-pass intestinal metabolism. However, simultaneous systemic enzyme inhibition may be undesirable, and thus should be minimized. For example, 2-Beta-fluoro-2',
3'-dideoxyadenosine (F-ddA) is an ADA activated prodrug of 2-beta-fluoro-2',3'-dideoxyinosine (F-ddI) that provides enhanced delivery to the CNS. F-ddA has been tested clinically for the treatment of AIDS. Unfortunately, intestinally localized ADA constitutes a formidable enzymatic barrier to the oral absorption of F-ddA. Through careful selection of an enzyme inhibitor, dosage regimen design, and by considering the inhibition vs. drug absorption profiles, local enzyme inhibition can be optimized to achieve local ADA inhibition, with minimal systemic inhibition (133,134).
DeltaG, the 12 kDa active fragment of zonula occludens toxin, has been used as absorption enhancer to increase the brain distribution of MTX and paclitaxel, two commonly used anticancer agents with poor distribution into the brain. DeltaG significantly enhances the brain distribution of MTX (hydrophilic) and paclitaxel (lipophilic) and has the potential to be further developed as adjunct therapy to increase delivery of poorly permeable chemotherapeutic and other CNS targeted compounds (135).
Gene transfer offers the potential to explore basic physiological processes and to intervene in human disease. The extension of gene therapy to the CNS, however, faces the delivery obstacles of a target region that is postmitotic and isolated behind the BBB. Approaches to this problem have included grafting genetically modified cells to deliver novel proteins, or introducing genes by viral or synthetic vectors geared toward the CNS cell population. Invasive approaches such as direct inoculation and bulk flow, as well as osmotic and pharmacological disruption, have also been used to circumvent the BBB's exclusionary role. Once the gene is delivered, a myriad of strategies have been tested to affect a therapeutic result. Gene-activating prodrugs are the most common antitumor approach. Other approaches focus on activating immune responses, targeting angiogenesis, and influencing apoptosis and tumor suppression. At this time, therapy directed at neurodegenerative diseases has centered on ex vivo gene therapy for supply of trophic factors to promote neuronal survival, axonal outgrowth, and target tissue function. Despite early promise, gene therapy for CNS disorders will require advancements in methods for delivery and long-term expression before becoming feasible for human disease (136).
Gene therapy has already shown promise as a tool for brain protection and repair from neuronal insults and degeneration in several animal models, and is currently being tested in clinical trials. The choice of an appropriate vector system for transferring the desired gene into the affected area of the brain is an important issue for developing a safe and efficient gene therapy approach for the CNS. Both viral and non-viral vectors have been used in gene delivery to treat brain disorders (137,138).
In non-viral vector gene delivery, lipid-based vectors such as liposomes are commonly used vectors in mediating the delivery of therapeutic genes to the CNS (139). Promising results have been obtained in terms of the level and duration of gene expression, particularly in cortical neurons that were transfected with the Tf-associated lipoplexes. These findings highlight the potential utility of these lipid-based carriers for delivering genes within the CNS (140,141). A simple and highly efficient lipofection method was used for primary embryonic rat hippocampal neurons (up to 25% transfection) that exploits the M9 sequence of the non-classical nuclear localization signal of heterogeneous nuclear ribonucleoprotein A1 for targeting beta(2)-karyopherin (transportin-1). This technique can facilitate the implementation of promoter construct experiments in post-mitotic cells, stable transformant generation, and dominant-negative mutant expression techniques in CNS cells (142). Multi-lipofection provides a mild and efficient means of delivering foreign genes into astrocytes in a primary culture, making astrocytes good candidate vehicle cells for gene/cell therapy in the CNS (143). DC-Chol (dimethylaminoethane-carbamoyl-cholesterol) liposome-mediated NGF gene transfection may have therapeutic potential for treatment of brain injury (144).
Sufficient gene transfer into CNS-derived cells is the critical step for developing strategies for gene therapy. Research shows that liposome-mediated gene transfer is an efficient method for gene transfer into CNS cells in vitro, but the transfection efficiency into the rat brain in vivo is low (145). Alternative gene transfer techniques, such as using cationic liposomes to achieve therapeutically useful levels of expression of neurotrophins in the CNS, could provide new strategies for treating a traumatically injured CNS (146). Likewise, antisense oligonucleotides (AS-ODNs) offers a precise and specific means of knocking down expression of a target gene, and is a major focus of research in neuroscience and other areas. It has become increasingly obvious, however, that there are a number of hurdles to overcome before antisense can be used effectively in the CNS, most notably finding suitable nucleic acid chemistries and an effective delivery vehicle to transport AS-ODNs across the BBB to their site of action. Despite these problems, a number of potential applications of AS-ODNs in CNS therapeutics have been validated in vitro and, in some cases, in vivo (147).
The efficient and targeted transfer of genes is the goal of gene therapy. In the CNS, this is challenging due in part to the exquisite anatomy of the brain. Viral vectors have better transfection rates but a higher incidence of deleterious effects than non-viral vectors. Herpes simplex virus (HSV) vectors are particularly amenable to CNS therapies as they are capable of transducing a variety of cells, have a large transgene capacity and can exist as either oncolytic or non-immunogenic vectors. The versatility and therapeutic use of this vector platform has been used in two CNS disorders, Alzheimer's disease and malignant brain tumors (148). Herpes simplex virus type 1-thymidine kinase (HSV1-TK) in combination with the prodrug ganciclovir (GCV) represents an efficient suicide gene approach in brain tumor gene therapy. The effectiveness of HSV1-DeltaTK in preventing brain tumor growth in vivo, combined with its reduced cytotoxicity, both in vivo and in primary cultures of CNS cells, could represent an advantage for treatment of brain tumors using gene therapy (149,150). A hemagglutinating virus of Japan (HVJ)-liposome vector has been used to deliver oligodeoxy-nucleotides in the CNS in vivo and in vitro. Thus it is an efficient method for ODN transfer and holds promise as a gene delivery method in the CNS (151). (HVJ)-AVE liposome, an anionic type liposome with a lipid composition similar to that of HIV envelopes and coated by the fusogenic envelope proteins of inactivated HVJ, has been used for gene delivery into the CSF space (152).
Gene delivery could be optimized by a combination of drug-loaded liposomes, polymeric nanoparticles, non-viral DNA complexes, viruses and therapeutic CED to overcome particle binding and clearance by cells within the CNS (153). CNS gene transfer could provide new approaches for modeling neurodegenerative diseases and devising potential therapies for these diseases. One such disorder is Parkinson's disease, in which dysfunction of several different metabolic processes has been implicated (154).
Drug delivery through the nose, also known as the olfactory pathway, has also been explored as an alternative strategy to deliver CNS drugs. Drugs delivered intranasally are transported along olfactory sensory neurons, and can yield significant concentrations in the CSF and olfactory bulb. Hydroxyzine and triprolidine have both been reported to reach the CNS following nasal administration. In vitro experiments have been conducted to evaluate the effect of directionality, donor concentration and pH on the permeation of hydroxyzine and triprolidine across excised bovine olfactory mucosa. The studies demonstrated that bidirectional flux (mucosal! submucosal and submucosal!mucosal flux) of hydroxyzine and triprolidine across the olfactory mucosa was linearly dependent upon the donor concentration, without any evidence of saturable transport. The lipophilicity of these compounds, coupled with their ability to inhibit P-gp, enable them to freely permeate across the olfactory mucosa. Despite the presence of a number of protective barriers such as efflux transporters and metabolizing enzymes in the olfactory system, lipophilic compounds such as hydroxyzine and triprolidine can access the CNS primarily by passive diffusion when administered via the nasal cavity (155).
In a previous study, the plasma pharmacokinetics of nipecotic acid and its n-butyl ester were compared after intranasal and intravenous administration. Intranasal administration of an ester formulation was crucial for delivery of nipecotic acid to the brain, as the evidence suggested that ester hydrolysis is rate limiting to nipecotic acid brain delivery. The formed nipeoctic acid displayed tissue trapping in brain. The study also demonstrated that parenteral dosing of nipecotic acid esters is unnecessary for systemic or brain delivery of nipecotic acid, the finding may apply to other CNS active zwitterion esters (156).
The nose has also been studied as a possible route for the systemic delivery of water soluble prodrugs of L-dopa. L-dopa prodrugs had improved solubility and lipophilicity, with relatively rapid in vitro conversion in rat plasma. Following intranasal administration, prodrug absorption was rapid and complete, with a bioavailability of approximately 90%. Thus, utilization of water soluble prodrugs of L-dopa via the nasal route in the treatment of Parkinson's disease may have therapeutic advantages such as improved bioavailability, decreased side effects, and potentially enhanced CNS delivery (157).
In general, there are several obstacles that must be overcome nasal delivery to be successful. These include enzyme activity in nasal epithelium, low pH, mucosal irritation, and regional variability in absorption caused by nasal pathology. However, this method is convenient and relatively noninvasive compared to some other strategies. Nevertheless, more research is needed in order to intranasal delivery systems and to achieve therapeutic drug concentrations in CNS using this pathway.
Was this article helpful?