Recent Advances in Medicinal Chemistry and

Current Topics in Medicinal Chemistry, 2009, 9, 182-196

182

Recent Advances in Medicinal Chemistry and Pharmaceutical TechnologyStrategies for Drug Delivery to the Brain Nunzio Denora, Adriana Trapani, Valentino Laquintana, Angela Lopedota and Giuseppe Trapani* Dipartimento Farmaco-Chimico, Facoltà di Farmacia, Università degli Studi di Bari, Via Orabona 4, 70125 Bari, Italy Abstract: This paper provides a mini-review of some recent approaches for the treatment of brain pathologies examining both medicinal chemistry and pharmaceutical technology contributions. Medicinal chemistry-based strategies are essentially aimed at the chemical modification of low molecular weight drugs in order to increase their lipophilicity or the design of appropriate prodrugs, although this review will focus primarily on the use of prodrugs and not analog development. Recently, interest has been focused on the design and evaluation of prodrugs that are capable of exploiting one or more of the various endogenous transport systems at the level of the blood brain barrier (BBB). The technological strategies are essentially non-invasive methods of drug delivery to malignancies of the central nervous system (CNS) and are based on the use of nanosystems (colloidal carriers) such as liposomes, polymeric nanoparticles, solid lipid nanoparticles, polymeric micelles and dendrimers. The biodistribution of these nanocarriers can be manipulated by modifying their surface physico-chemical properties or by coating them with surfactants and polyethylene-glycols (PEGs). Liposomes, surfactant coated polymeric nanoparticles, and solid lipid nanoparticles are promising systems for delivery of drugs to tumors of the CNS. This mini-review discusses issues concerning the scope and limitations of both the medicinal chemistry and technological approaches. Based on the current findings, it can be concluded that crossing of the BBB and drug delivery to CNS is extremely complex and requires a multidisciplinary approach such as a close collaboration and common efforts among researchers of several scientific areas, particularly medicinal chemists, biologists and pharmaceutical technologists.

Keywords: Blood-brain barrier, Drug targeting, Prodrugs, Carrier mediated transport, Liposomes, Nanoparticles. 1. INTRODUCTION Diseases of the central nervous system (CNS) are numerous and affect a large part of the world’s population. Stroke, ischemia, human immunodeficiency virus-1 (HIV-1) infection, epilepsy, and other psychiatric disorders such as anxiety, depression and schizophrenia are debilitating conditions that markedly affect the morbidity and mortality in modern society. The neurodegenerative diseases, such as Alzheimer’s (AD), Parkinson’s diseases (PD) and multiple sclerosis are characterized by symptoms related to movement, memory, and dementia due to the gradual loss of neurons. Brain tumors, including gliomas, astrocytomas and glioblastomas, constitute a relevant and unsolved clinical problem and the treatment of brain cancers are major challenges [1]. Unfortunately, few safe and effective methods are known for diagnosis and treatment of CNS disorders and this is mainly due to the anatomical characteristics of the CNS (see later discussion). The blood-brain barrier (BBB) represents an effective obstacle for the delivery of neuroactive agents to the central nervous system (CNS). The presence of the BBB makes treatment of many CNS diseases difficult to achieve, because the required therapeutic agents cannot be delivered across the barrier in sufficient amounts. It is estimated that more than 98% of small molecular weight drugs and practically *Address correspondence to this author at the Dipartimento FarmacoChimico, Facoltà di Farmacia, Università degli Studi di Bari, Via Orabona 4, 70125 Bari, Italy; Tel: (039) 080-5442764; Fax: (039) 0805442754; E-mail: [email protected] 1568-0266/09 $55.00+.00

1015

100% of large molecular weight drugs (mainly peptides and proteins) developed for CNS pathologies do not readily cross the BBB [2]. To improve the brain penetration of potential therapeutic agents numerous medicinal chemistry- and pharmaceutical technology-based strategies have been explored and developed. The present mini-review deals with the various approaches, which have been recently established for the treatment of brain pathologies utilizing both the medicinal chemistry and technological contributions. Several reviews on specific topics have appeared in the literature and summarize the progress made in this area [3-12]. Here we build on the review of Ricci et al. in 2006 [13]. The aim of the present paper is to review the latest developments, evaluating both the scope and limitations of some strategies as well as the evidence supporting the importance of the therapeutic molecule features such as molecular weight and lipophilicity. The various barriers that impede the delivery of the drugs to the brain are reviewed. This is followed by a discussion of the use of both chemical modifications (i.e., medicinal chemistry approach) and nanocarriers (i.e., technological approach) for overcoming these barriers in order to effect delivery of drugs to sites in the CNS. 2. PHYSICAL BARRIERS TO THE PASSAGE OF MOLECULES FROM THE BLOOD COMPARTMENT TO THE BRAIN There are two physical barriers that separate the brain extracellular fluid from the blood. The first is constituted by © 2009 Bentham Science Publishers Ltd.

Methodologies to Assess Brain Drug Delivery in Lead Optimization

Current Topics in Medicinal Chemistry, 2009, Vol. 9, No. 2

the brain capillary endothelial cells that form the BBB. This physical barrier is characterized by tight junctions between endothelial cells, by the absence of fenestrations and low occurrence of pinocytic activity [14]. These features restrict the movement of compounds from the blood into the extracellular environment of the brain (Fig. (1)). The second barrier, located at the choroid plexus, is represented by the blood-cerebrospinal fluid barrier (BCSFB) that separates the blood from the cerebrospinal fluid (CSF) which, in turn, runs in the subarachnoid space surrounding the brain. Unlike the capillaries that form the BBB, the capillaries in the choroid plexus allow free movement of molecules via intracellular gaps and fenestrations [15]. The epithelial cells in the choroid plexus that form the BCSFB have complex tight junctions on the CSF (apical) side of the cells. These tight junctions of the epithelial cells in the choroid plexus are slightly more permeable than those found in the endothelial cells of the BBB [16] (Fig. (1)). In this context, it should be recognized that the BBB is interrupted in the case of brain tumors and some methods for exploiting it for therapeutic purposes have been suggested [17]. It is now well established that a tumor must develop its own vascular network to grow and the neovasculature within tumors consists of vessels with increased permeability due to the presence of large endothelial cell gaps compared with normal vessels [17]. Furthermore, recent studies highlighted

Fig. (1). Schematic representation of the two main barriers in the CNS.

1016

183

the possibility to reach the brain following the nasal route of administration. In fact, it has been shown that, by using this pathway, the transport of drug across the olfactory region in the nasal cavity occurs thus reaching directly the brain tissue or the CSF. It is based on the connection existing between the nose and the brain, that is, the olfactory bulb. In fact, the olfactory epithelium is situated between the nasal septum and the lateral wall of each side of the two nasal cavities and just below the cribriform plate of the ethmoid bone separating the nasal cavity from the cranial cavity [18]. 2.1. Approaches for Increasing Brain Penetration Generally, there are three approaches for increasing the penetration of drugs into the brain. The first is an invasive route that circumvents the obstacle of the BBB and/or BCSFB by direct administration of the drug into the brain. An alternative approach consists in generating a transient disruption of the BBB, allowing the therapeutic agents to enter into the brain from the blood through a more permeable BBB. The third approach concerns the chemical modification of the drug improving its penetration into the CNS. In addition to the previous mentioned approaches, a fourth option has been increasingly investigated in the last decades and consists in the use of formulation approaches. These technological strategies are essentially non-invasive methods and are based on the use of colloidal carriers

184 Current Topics in Medicinal Chemistry, 2009, Vol. 9, No. 2

Trapani et al.

including mainly liposomes, polymeric- and solid lipidnanoparticles (see Section 4). However, work in this area has been primary limited to drug delivery to tumors of the brain. To bypass the BBB, direct injection of drugs into the brain involves intracerebral and intrathecal administration. This approach is invasive and requires a craniotomy. An advantage of this pathway is that a wide range of compounds including large-and small-molecules can be administered. As mentioned above, another option to by-pass the BBB is the intranasal administration based on the connection existing between the nose and the brain, that is the olfactory bulb [18]. This pathway provides a mean for the administration of various compounds into the CNS including toxic agents such as pathogens, viruses, and toxic metals as well as various therapeutic agents including small molecules and proteins. Small molecules such as cocaine and cephalexin as well as a number of protein therapeutic agents, such as insulin have been successfully delivered to the CNS using intranasal delivery [19-21] although for polar drug molecules this route is questionable with respect to the quantity of drug that can be delivered. This administration route has been found as a promising approach for rapid-onset delivery of some medications to the CNS bypassing the BBB. The strategy consisting in generating a transient disruption of the BBB includes the systemic administration of hyperosmotic solutions or vasoactive compounds such as bradykinin and related analogs, or various alkylglycerols. The use of osmotic agents such as mannitol or arabinose involves expansion of the blood volume caused by the addition of the hyperosmotic agent and disruption of the BBB. This barrier resumes its normal integrity and function returning the osmolarity of the blood to normal value. During this period when the tight cellular junctions between the brain capillary endothelial cells have been compromised, paracellular diffusion [Fig. (2)] of water-soluble drugs and solutes into the brain is increased [22]. The BBB can also be disrupted by pharmacological means. Several endogenous proinflammatory vasoactive agents, such as bradykinin, histamine, nitric oxide are known to induce increases in BBB permeability in a concentrationand time-dependent manner. However, although capable of producing increases in BBB permeability, these endogenous agents cannot be applied safely for CNS drug delivery. Thus, the effects of bradykinin on BBB permeability are shortlived, requiring carotid artery infusion. Better results were obtained with bradykynin analogs such as labradimil. The effect of labradimil is characterized by a significant greater plasma and tissue stability than seen with bradykinin [23]. The transient disruption of the BBB has also been observed by the systemic administration of various alkylglycerols. The extent of BBB disruption is seen to depend on the length of the alkyl group and the number of glycerols present in the structure. The exact mechanism(s) for the transient BBB disruption observed with the alkylglycerols is unknown [24]. Moreover, as with the bradykinin analogs, the disruption of the BBB by alkylglycerols is very short lived. The third strategy for improving the brain penetration of therapeutic agents utilizes the chemical modification of the drug to improve transcellular migration. As shown in Fig.

1017

(2), the transcellular routes available include passive diffusion, specific transport systems, and endocytic processes in brain capillary endothelial cells. Overall, this strategy should lead to lower neurotoxicity compared to that associated with BBB disruption. 2.2 Physiological Factors Affecting Drug Delivery to CNS Many transport mechanisms for the uptake of nutrients into CNS exist in the brain (Fig. (2)). These transport mechanisms may be exploited for brain drug delivery. These include: a) Passive Diffusion. The main factors affecting the passive diffusion of drugs across the BBB involve an adequate lipophilicity, neutral or uncharged nature, low hydrogen bonding potential and small molecular size (< 500 g/mol). Thus, the improvement of the passive diffusion of drugs across the BBB can often be achieved by either increasing lipophilicity or reducing molecular size. As lipophilicity is dependent on polarity and ionization, modification of functional groups on drugs provides a method for improving passive diffusion across the BBB. b) Carrier-mediated (Active) Transport. More than 20 carrier-mediated transporter proteins have been identified in cerebral capillaries of the BBB including transporters for glucose, amino acids, vitamins, and nucleotides. The carriers for the large amino acids (e.g., amino acid transporter of type 1, LAT 1) and glucose (e.g., glucose transporter of type 1 GLUT 1), especially, have a sufficiently high transport capacity [25]. Therefore, an approach for increasing the transcellular passage of drugs across BBB is to design drugs that structurally resemble or can be linked to endogenous compounds that are transported into the brain by the carriers or transporters expressed in the brain microvessel endothelial cells [26]. Transporters that have received the greatest attention are: i)

Amino Acid Transporters. Together with the large neutral amino acid transporters, LA transporters, cationic-, anionic- and neutral-amino acid transporters have also been identified. LA transporters have been most exploited for drug delivery purposes [27]. L-Dopa is the most well-known example of a drug that is transported by LA transporters in the BBB. L-Dopa is an endogenous large amino acid and is a precursor of the neurotransmitter dopamine. LA transporters are also involved in the transport of other drugs such as L-melphalan, baclofen, and gabapentin across the BBB [28].

ii) Glucose Transporters. The most important glucose transporter present in the brain capillary endothelial cells is the type 1, glucose transporter, GLUT 1. Compared to other nutrient transport/carrier systems in the BBB, GLUT 1 has the highest transport capacity (more than 10-50 times greater than that of amino acid and carboxylic acid transporters) and therefore represents an attractive target for drug delivery to the CNS. Glycosylated analogs of various opioid compounds have shown increased CNS analgesic properties compared to the non-glycosylated compounds [29].



185

Fig. (2). Blood-brain barrier transport mechanisms.

iii) Monocarboxylic Acid Transporter. The best-characterized organic acid transporter in the BBB is the monocarboxylic acid transporter (MCT). Examples of drugs entering the CNS through the MCT are salicylic acid and various cholesterol-lowering 3-hydroxy-3methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors [30]. iv) Nucleoside Transporters. There are two general types of nucleoside transporter expressed in the brain capillary endothelial cells: the facilitative nucleoside transporters that carry selective nucleosides either into or out of the cell and active and the sodium-dependent transporters that can move selective nueleosides into the cell against a concentration gradient. There are several examples of drugs that are substrates for nucleoside transport systems such as the anticancer agent, gemcitabine, the antiviral agents, 3'-azidodeoxythymidine (AZT) and 2',3'-dideoxycytidine (ddC) [31, 32]. v) Peptide Transport Systems. Peptide transport systems are present in the brain capillary endothelial cells forming the BBB. The exact molecular nature of these peptide transporters remains to be determined. However, specific saturable transport systems have been identified in the BBB for glutathione [33], peptide hormones [34] and growth factors [35].

1018

c)

Vesicular Transport. Two types of vesicular transport processes are known: the fluid-phase endocytosis and the adsorptive endocytosis. However, only adsorptive endocytosis involves an initial binding or interaction with the plasma membrane of the cell. Vesicular transport due to adsorptive endocytosis is a saturable, ligand selective process. Several large macromolecules are transported from the blood into the brain through receptor-mediated endocytosis. Thus, these specific receptor-mediated transport processes represent another approach for enhancing transcellular permeability across the BBB. The most well-known processes are the transferrin- and insulin receptor-mediated vesicular transport.

i)

Transferrin is a glycoprotein that controls the transport of iron throughout the body. The brain capillary endothelial cells have a high density of transferrin receptors on their surface. Iron enters the cell as a complex with transferrin through an endocytic process that is initiated by the binding of transferrin to its receptor. Inside the brain endothelial cell, the iron is removed from the transferrin in the endosome. However, transferrin shows a limitation as a brain delivery vector because it is recycled back to the luminal surface of the brain capillary endothelial cell. As an alternative, a murine monoclonal antibody (MAb), OX-26 has been iden-


Trapani et al.

tified, which appears to be suitable for use as a drug carrier for this transport system. First, this MAb binds to the receptor and triggers endocytosis. Second, the OX26 MAb binds to an extracellular epitope on the transferrin receptor that is distinct from the transferrin ligand-binding site; thus, the OX-26 MAb does not interfere with transferrin binding to its receptor on the brain endothelial cells. The OX-26 antibody has proven to be an effective brain delivery vector, as it has been conjugated to a variety of drugs including methotrexate and nerve growth factor [36, 37]. ii) Insulin is a peptide hormone involved in glucose metabolism. The presence of insulin receptors in the CNS, suggest that insulin may have important functions also within the brain. Studies demonstrating the presence of high-affinity insulin receptors on the luminal plasma membrane of brain microvessel endothelial cells indicate that the peptide penetrates the BBB through a receptor-mediated transport process. Other studies support the potential use of insulin as brain delivery vector of therapeutic agents and macromolecules to the brain. For example, insulin has been used as a BBB transport vector for proteins [38]. Many other receptors such as insulin-like growth factor receptor, and a leptin receptor may be used for the same purpose. The corresponding ligands could be used as brain drug delivery vectors. In general, the strategy of using receptor-mediated transport consists in covalently binding drugs to peptides or MAbs, allowing the receptor recognition and transport into the brain of the drug-MAb complex. 2.3. Drug Efflux Transporter Systems in the BBB While it may seem simple to improve drug or prodrug transport to the CNS by changing lipophilicity or improving affinity for a transporter, attempts often fail. Several drugs are effluxed from brain to blood and the impact of this outwardly directed transport systems on CNS drug delivery is of great interest. In particular, several different efflux transporters are present in the BBB such as the Pglycoprotein (P-gp) and the multidrug resistance protein (MRP). These transporters are part of the larger ATP binding cassette (ABC) family of proteins that remove a wide variety of compounds from the cell through an ATP-dependent active transport process. In the past, these proteins were found to be over-expressed in various drug-resistant cancer cells; later on, they were also evidenced in normal cells such as intestinal and renal epithelial cells, hepatocytes, and brain capillary endothelial cells. Consequently, they influence the absorption, distribution, and elimination of many drugs. P-gp is a well-known drug efflux transporter. It has been demonstrated that the brain levels of drugs such as vincristine and ivermectin increased by 80-100-fold in the mice lacking P-gp. P-gp drug efflux in the BBB has been invoked to account for the reduced brain penetration of a number of structurally different drugs including digoxin, cyclosporin A (CSA), itraconazole, and opioid analgesics. To increase the CNS delivery of drugs that undergo active efflux, the co-administration of an inhibitor of the

1019

transporter or another substrate that can competitively saturate the efflux transporter has been explored. Thus, P-gp modulators such as verapamil diltiazem and CSA, have been used to increase the brain penetration of a number of compounds with low BBB permeability such as antiviral protease inhibitors, the anticancer agent paclitaxel, and the antifungal agent itraconazole. In addition, some polymers have been shown to inhibit drug efflux transporters. The mechanism by which polymers modulate drug efflux transporter activity involves alterations in membrane fluidity. For example, Pluronic block copolymer (P85) has been shown to enhance the permeability of a wide variety of drugs in an in vitro model of the BBB. These studies suggest that P85 can be used for improving drug permeability of the BBB through inhibition of drug efflux transporter system [39]. The practicability of such an approach is unknown. 3. MEDICINAL CHEMISTRY-BASED STRATEGIES Medicinal chemistry-based strategies are essentially aimed at the chemical modification of low molecular weight drugs in order to increase their lipophilicity [8]. This includes both an analog approach as well as a prodrug approach. Only the prodrug approach will be discussed here. For instance, the formation of an inactive prodrug is a way to increase the lipophilicity of a drug by attaching a lipophilic promoiety that can be cleaved to the parent drug on entering the CNS. Many of these examples involve ester-based prodrugs since by appropriate esterification of molecules containing -COOH, -OH, or -SH groups, it is possible to obtain derivatives with the desired lipophilicity. The classical example of such approach is heroin (i.e., the diacetyl ester of morphine) that rapidly crosses the BBB due to its high lipophilicity. Once in the brain, it is presumed to be hydrolyzed to morphine. The same approach has been employed with other therapeutic agents such as the anticancer agent chlorambucil and the neurotransmitters dopamine and gamma-aminobutyric acid (GABA) [3, 5, 8]. In recent years, notable interest has been focused on the design and evaluation of prodrugs that are capable of exploiting one or more of the various endogenous transport systems in the BBB. This is an attractive approach because it provides a more targeted approach to CNS drug delivery. The prodrug is designed to structurally resemble the endogeneous ligand of a specific transport system, which recognizes the prodrug as a substrate and transports it across the BBB. Thus, for example, the neurotransmitter dopamine is not able to cross the BBB due to its hydrophilic nature. However, the conversion of dopamine into its -amino acid, L-Dopa 1, enables the brain to uptake 1 via the large aminoacid transporter of type 1, LAT 1. L-Dopa 1 is then decarboxylated to dopamine by L-amino acid decarboxylase in the brain. This decarboxylation also occurs in peripheral tissues. Although approximately 95% of L-Dopa, 1, is metabolized to dopamine in the peripheral tissues, the percentage of remaining L-Dopa allows sufficient brain therapeutic activity [4, 5] (Fig. (3)). Also utilizing LAT 1 are 4-chlorokynurenine, a prodrug of 7-chlorokynurenic acid 2 [40], an N-methyl-D-aspartate antagonist, the anticonvulsant gabapentin, 3, and the anticancer agent melphalan, 4 [5] (Fig. (3)). A recent example of this strategy have been published by Gynther et al. who described the conjugation of a



187

O O HO

NH2 OH

NH 2 OH

HO

Cl

NH2

OH

NH 2 OH

O

Cl

O

O

1

H2N

2

N

3

4

Cl

O

O O

O

NH 3+ClOH

O

S

HOOC NH2

O

N H

N H

5

S O

HN

N H

O COOH

OH OH

6

HO

HO NH

NH

HO

O

HO OH

O O

7

HO

O O O

O OH

HO

O

HO

8

OH

HO

OH CH3

O O

O NH 2

N H H 3C

O

EtO

Cl O

O

O HCl

n

N

F

CH 3

9

10 R2

R2

COOR N H

11 R = H 12 R = n-butyl

N R1

N

R3

N

OH

O

R1

N

R3

O HN

OH R4

R 1 = R2 = R 3 = H, Cl R 4 = H, COOH, COOEt

HN (CH2 )n COOR4 R 1 = R 2 = R3 = H, Cl; R4 = H, Et

14 n = 1 15 n = 3

13

Fig. (3). Chemical structures of low molecular weight drugs and prodrugs studied for delivery to the brain.

hydrophilic drug, ketoprofen, to L-tyrosine [41]. Unlike ketoprofen itself, the amino acid L-tyrosine is a LAT 1substrate and, interestingly, is characterized by a phenolic hydroxyl group suitable for the conjugation with various drug molecules. The mechanism and the kinetics of the brain uptake of the ketoprofen-tyrosine prodrug 5 was studied using the in situ rat brain uptake model. The uptake of the prodrug was found to be concentration-dependent. In addition, a specific

1020

LAT 1 inhibitor significantly decreased the brain uptake of the prodrug. Therefore, the results indicate that a drugsubstrate conjugate is able to transport drugs into the brain LAT 1. Other transporters such as the peptide transport systems have been exploited for enhanced BBB penetration of dopamine. Thus, More and Vince [42] prepared and evaluated the compound 6 that can be considered a dopamine prodrug containing the peptide glutathione (–glutamyl-


Trapani et al.

cysteinylglycine) as promoiety and mercaptopyruvate as a linker. The prodrug design rationale is based on the ability of 6 to cross the BBB through recognition by glutathione transporters and on its ability to release the active drug once inside the brain. These glutathione transporters are located on the luminal side of the BBB and show a broad substrate specificity. In vitro directional transport experiments using the Madin Darby canine kidney (MDCK II) cell line as BBB model, showed the ability of prodrug 6 to cross this model cell line. This data supports the possible use of glutathione as a carrier for drug targeting. A number of investigations have been performed with different drug molecules in order to determine the possibility of exploiting the GLUT 1 transporter system. Fernandez et al. [43] synthesized several glycosyl derivatives of dopamine and tested the affinity of the prodrugs to GLUT 1 in human erythrocytes. Dopamine was linked to glucose with different linkers at the C-1, C-3 and C-6 positions of glucose (compounds 7 and 8, respectively, in (Fig. (3))). The results of glucose uptake inhibition showed that the glucose derivatives that were conjugated at position C-6 showed better affinity for GLUT 1 in human erythrocytes than those substituted at C-1 or C-3 positions. The relevancy of these findings to BBB transport needs to be explored. Limitations for Carrier-Mediated Transport in the CNS Undoubtedly, the design of drugs that utilize endogenous transport systems at the BBB is an attractive approach for increasing drug delivery to the brain. By exploiting specific transport systems it is possible to provide a more targeted approach to CNS drug delivery than physicochemical alterations aimed at creating a more lipophilic therapeutic agent or prodrug. However, some criticism has been raised to this approach. The most relevant is that the transport systems are relatively selective systems and devoted to the passage of essential nutrients and metabolites inside and outside the brain. Thus, chemical modifications made to therapeutic agents to target specific transport systems in the BBB are much more restricted than those used to enhance lipophilicity. Furthermore, the resulting modifications often result in compounds with lower affinity for the transporter than the endogenous ligand. For example, the nucleosidebased antiviral agents, AZT and ddC are transported by the concentrative nucleoside transporter of type 1, CNT 1. However, the affinity of these agents for the transporter is approximately 25-fold lower than that of endogenous pyrimidine-based nucleosides. An additional often-overlooked obstacle is that one would prefer to give drugs via the oral route. If oral dosing is desired, conjugates or prodrugs such as those described above must be capable of being absorbed from the GIT and remain intact until they reach the brain. Designing prodrugs that can be efficiently absorbed, remain intact, cross the BBB and then be cleaved in brain tissue remains one of the great challenges both medicinal chemists and their drug delivery colleagues. New Trends to the Prodrug Approach The prodrugs described above essentially aimed to enhance the lipophilicity or to target specific transport

1021

systems. Recently new trends in the prodrug approach have been explored. An interesting application has been reported by Pignatello et al. [44] who studied amphiphilic prodrugs of the flurbiprofen containing lipoamino acids (LAA) as promoieties. LAA are -amino acids bearing alkyl side chains, whose length and structure can be modified to achieve the desired physicochemical properties. Because of the presence of an alkyl chain and a polar amino acid head, LAA conjugation yields amphiphilic derivatives, with a membrane-like character that can favor interaction with and penetration through biological membranes and barriers. Due to its anti-inflammatory and analgesic activities, flurbiprofen can be considered as a potential neuroprotective agent in AD therapy. This drug, indeed, reduced the secretion of amyloid protein A42, the major component of senile plaques of AD brain, both in Neuro-2a cells and rat primary cortical neurons, as well as in the brain of tg2576 -amyloid transgenic mice. Flurbiprofen was modified with LAA residues to give compounds of general formula 9 (Fig. (3)) with the aim of increasing its availability in the brain. Biodistribution and pharmacokinetic studies showed that i.v. injection of the parent drug led to constant brain levels for 3 h; thereafter, the drug rapidly disappeared from this area. After injection of 9 (n = 9) the released flurbiprofen appeared in the brain after 1 h, and the intracerebral levels continued to rise up to the ninth hour, reaching higher values than those observed in the first 3 h after the injection of the parent drug. The findings suggest that using 9 (n = 9), the parent drug, flurbiprofen, is released and accumulated preferentially in the brain tissue. In addition, the results highlight the role played by the LAA residue on the biopharmaceutical profiles of prodrugs. In particular, relatively small differences in the chain length of the side alkyl chain of the LAA promoiety can led to a considerably different distribution profile. Further amphiphilic prodrugs containing LAA as promoieties were studied by the same authors using cloricromene, a coumarin derivative that possesses antithrombotic, antiplatlet actions and causes vasodilatation. It was found that the intraperitoneal administration of compound 10 (CLOR-C4) to rats was able to provide a slight but statistically significant higher concentration of the active metabolite (cloricromene acid) in the brain compared with the parent drug administered by the same way [45]. Based on the available data, it can be stated that these amphiphilic LAA containing prodrugs need to be further evaluated to establish their potential for improved brain delivery. Another interesting application of the prodrug approach is the chemical modification of drugs to enhance delivery across the nasal mucosa and to prevent their metabolic degradation. Nasal administration route allows rapid-onset delivery of drugs to the CNS bypassing the BBB. An example of such application has been provided by Wang et al.[46] who studied the nasal administration of the nipecotic acid and n-butyl nipecotate (i.e., compounds 11 and 12, respectively, in (Fig. (3)) to rats. It was found that nasal dosing of the n-butyl ester of nipecotic acid provided a viable approach for delivering of nipecotic acid, a zwitterion, to the brain. A pharmacological response was observed only after dosing the ester, but not after i.v. nipecotic acid administration. There was no significant differences in



189

nipecotic acid total brain levels after intravenous (i.v.) or nasal ester administration, suggesting the nasal route is as effective as the i.v. route for delivery of the acid to the brain thanks to the cleavable ester functional group. Rat brain disposition studies showed strong evidence that the ester hydrolysis was the rate limiting to nipecotic acid brain accumulation. This brain-targeted delivery system via nasal prodrugs seems a very promising approach for obtaining brain uptake of drugs.

13 is also that to reduce side effects. It should be noted that compounds 13 cannot be considered as simple co-drugs, i.e., prodrugs consisting of two pharmacologically active drugs that are coupled together in a single molecule so that each drug acts as a promoiety for the other; in fact, in compounds 13 the phenyl-imidazopyridine portion does not only play a pharmacodynamic role (i.e., the high affinity and selectivity for the GABA-BZR complex) but also a pharmacokinetic one (i.e., the BBB crossing ability).

Prodrugs that utilize the chemical delivery systems (CDS) involving the conjugation between low weight drugs and molecules having specific targeting properties have been extensively evaluated [8]. The most studied CDS for brain delivery exploits linking a drug molecule to 1,4-dihydro-Nmethyl-nicotinic acid (i.e., dihydrotrigonelline). EstradiolCDS (Estredox) can be considered the most promising and practical application example to date and is undergoing Phase II clinical testing for the treatment of postmenopausal symptoms [47]. Derivatization of estradiol with the targetor, 1,4-dihydrotrigonelline, leads to an increase in lipophilicity thus enabling better transport across the BBB. By using estradiol-CDS, the concentration of estradiol in rat brain was elevated four to five times longer than after estradiol treatment alone.

Evaluation of compounds such as 13 showed that all the prepared compounds were adequately stable to the chemical hydrolysis but unstable in physiological medium [54]. Receptor binding studies demonstrated that the examined compounds are essentially devoid of affinity for dopaminergic and benzodiazepine receptors. Bi-directional transport experiments across Madin–Darby Canine Kidney retrovirally transfected with the human MDR1 cDNA (MDCKII-MDR1) cell monolayers indicated that the compounds tested are not substrates of P-gp. Transport studies involving co-culture of bovine brain microvessel endothelial cell (BBMECs) and astrocytes monolayers (a well established in vitro method to estimate the BBB permeability) indicated that some of compounds 13 were able to cross the BBB. Interestingly, brain micro-dialysis experiments in rat showed that intraperitoneal acute administration of compounds 13 (R1 = R2 = R3 = Cl; R4 = COOEt; R5 = R6 = OH and R1 = R2 = R3 = Cl; R4 = H, R5 = R6 = OH) induced a dose- and time-dependent increase in the dopamine levels (up to 197%) in the rat medial prefrontal cortex. Based on these results, these compounds can be proposed as novel LDopa and dopamine prodrugs [54].

In the design of novel brain delivery systems by chemical modification, drug conjugation with receptor ligands represents another explored approach. In recent years, indeed, new useful potential cellular targets have been identified and characterized. Thus, for example, the peripheral benzodiazepine receptors (PBRs) have been identified in various peripheral tissues as well as in glial cells in the brain [48]. They are pharmacologically distinct from the central benzodiazepine receptors (CBRs) which are associated with GABA A receptors and mediate classical sedative, anxiolytic, and anticonvulsant properties of benzodiazepines. Among the different functions associated with the PBRs those associated with neurosteroid synthesis and the involvement in apoptosis processes [49, 50] are of particular interest. Evidence also indicates that PBRs are over-expressed in a number of tumor types, especially in the brain, and PBR expression appears to be related to the tumor malignancy grade. Based on all these observations, there are many potential clinical applications of PBR modulation, such as in oncologic, endocrine, neuropsychiatric and neurodegenerative diseases [51, 52]. Our research group showed that some 2-phenylimidazo[1,2-a]pyridine-3-acetamides are potent and selective ligands for PBRs [53]. Therefore, it seemed of interest to prepare compounds such as 13, which are characterized by an L-Dopa or dopamine moiety linked to appropriately substituted 2-phenyl-imidazopyridine-3-acetic acids. These conjugates were prepared in order to take advantage of i) the high affinity and selectivity for the GABA-benzodiazepine receptor (GABA-BZR) complex shown by most phenylimidazopyridine compounds; and ii) the high BBB crossing ability and lipophilicity shown by most phenyl-imidazopyridine derivatives. Moreover, since it is known that selective gabaergic agonists could be useful in patients who have complications associated with long-term L-Dopa treatment, a further expected advantage of using compounds

1022

A similar approach has been also applied to prepare compounds 14 and 15 which are characterized by a GABA or glycine moiety, respectively, linked to appropriately substituted 2-phenyl-imidazopyridine-3-acetic acids [55]. Again, stability, receptor binding, microdialysis and pharmacological studies were carried out on the compounds. The results demonstrated the feasibility of synthesizing useful anticonvulsants by coupling the amine function of the GABA with a phenylimidazopyridine portion [55,56]. Taking into account the functions of PBRs discussed above, it is clear that these receptors could also be the target to selectively increase anticancer drug delivery at brain level by using an appropriate PBR ligand-anticancer drug conjugate. Another possibility is PBR selective ligands that can serve as diagnostic imaging agents. The treatment of brain cancer is a formidable challenge in oncology. The failure of chemotherapy for brain tumors is due to the inability of intravenously administered anticancer agents to reach the brain parenchyma for the presence of the BBB although the capillaries that serve brain tumors are thought to be leakier. The first proof-of-concept of the potential of PBR ligandanticancer drug conjugate for brain delivery was reported by Guo et al. [57] who administered gemcitabine (GEM) and a PK11195 (a known PBR ligand)-GEM conjugate to nude rats bearing intracerebral tumors over-expressing PBRs. The pharmacokinetic and tissue distribution results showed a greater tumor selectivity and brain uptake by PK11195-GEM compared with GEM, attributable to receptor-mediated drug delivery and greater lipophilicity of the conjugate. These


Trapani et al.

findings prompted us to evaluate for the same purpose phenylimidazopyridine acetyl-melphalan conjugates in melphalan-sensitive human and rat glioma cell lines [58] as well as a cisplatin-like complex involving an imidazopyridin acetyl PBR ligand [59]. Moreover, we successfully used phenylimidazopyridine acetyl-fluorescein conjugates as a neurodiagnostic agent for activated microglia [60], an index of disease progression of several neurodegenerative disorders such as AD, PD, multiple sclerosis and HIV-associated dementia [61]. 4. PHARMACEUTICAL STRATEGIES

TECHNOLOGY-BASED

The technological strategies are essentially non-invasive methods of drug delivery to the CNS and represent valuable approaches for enhancing transcellular permeability of therapeutic agents and biomacromolecules across the BBB. They are based on the use of nanosystems (colloidal carriers), mainly liposomes and polymeric nanoparticles even though other systems such as solid lipid nanoparticles, polymeric micelles and dendrimers are also being tried. Following intravenous administration, the colloidal systems can extravasate only in tissues with a discontinuous capillary endothelium, such as the liver, spleen, and bone marrow, as well as into solid tumors and inflammed tissues where the endothelial cells are not closely joined together and increased vascular permeability occurs. An important requirement of the systemic intravenous use of these nanocarriers is their ability to circulate in the bloodstream for a prolonged period of time. However, after intravenous administration, they interact with the reticuloendothelial system (RES) which removes them from the blood stream [62]. This process mainly depends on particle size, charge and surface

Fig. (4). Capture mechanisms of nanocarriers by cells.

1023

properties of the nanocarrier [63]. To prevent the uptake by the RES, poly(ethylene glycol) (PEG) coating or direct chemical linking of PEG to the particle surface provides relatively long plasma residence times. However, PEGylated carriers are characterized by a low affinity for brain tissue as they are not transported through the BBB. Nevertheless, the nanosystems may represent useful tools for non-invasive drug delivery to brain utilizing active targeting. In fact, these nanocarriers can be taken up actively by carrier–mediated transport (CMT), receptor-mediated endocytosis (RME) and adsorptive-endocytosis (AME) (Fig. (4)) and hence reach the cerebral parenchyma, or are degradated within lysosomes leading to the drug being released into the brain tissue. Liposomes Liposomes have long been used as carrier systems for the delivery of therapeutic agents because of their easy preparation, good biocompatibility, low toxicity and commercial availability. They are vesicles composed of lipid bilayers surrounding internal aqueous compartments. Relatively large amounts of drug molecules can be incorporated into the aqueous compartment (water soluble compounds) or lipid bilayers (lipophilic compounds). Conventional liposomes are rapidly cleared from circulation by macrophages of the RES and this limits their usefulness as drug delivery systems. Extended circulation time can be accomplished by decreasing the particle size (