Enhanced brain drug delivery and targeting

The blood–brain barrier (BBB) forms an interface between the circulating blood and the brain, and functions as a tremendously effective barrier for the delivery of potential neurotherapeutics into the brain parenchyma. Conversely, the BBB possesses various carrier-mediated transport systems for the uptake of small molecules, such as essential nutrients and vitamins. These transporters have become an attractive target for drug/prodrug design in an attempt to ferry drug molecules across the BBB. Central nervous system (CNS) drug delivery is often limited by poor brain penetration of the potential drug candidate. As a result of its unique barrier properties, the BBB poses a huge challenge for the delivery of potential neurotherapeutics into the brain parenchyma.¹ It is estimated that only 2% of small-molecule drugs and <0.1% of novel protein and peptide pharmaceuticals developed for CNS diseases reach therapeutic concentrations in the brain.^2,3 Many of the pharmacologically active drugs tend to fail early in their development as these molecules lack optimal drug-like properties, such as a relatively small molecular weight (Mw <ca. 500 g/mol); adequate lipophilicity; neutral or uncharged nature; low hydrogen bonding potential; and sufficiently high unbound plasma fraction. These are all essential to achieve passive transcellular diffusion across the tightly joined capillary endothelial cells of the BBB.^2,4

In addition to being a structural diffusion barrier, the BBB constitutes an efficient functional barrier for solutes' attempts to cross the cell membrane. The high metabolic activity of brain capillary endothelial cells and the very effective efflux systems, which actively remove solutes from the brain back to blood circulation, protect the brain from potentially harmful endo- and exogenous agents, as well as making the development of effective neurotherapeutics difficult.^5,6 The adequate brain supply of various essential water-soluble nutrients, such as glucose, amino acids, vitamins and nucleotides, is ensured by a number of specific carrier-mediated inward transport mechanisms expressed in the BBB.⁷ The mechanisms that enable nutrients to cross the BBB are depicted in Figure 1.

Figure 1 Transport mechanisms in BBB for endogenous compounds, such as nutrients, which can also be utilized in drug and prodrug design (modified from Everts S: Brain Barricade, C&EN Online, www.cen-online.org June 4: 33–36, 2007).

Strategies for enhanced brain drug delivery of small molecules

To improve the brain penetration of potential therapeutic small molecules, numerous chemistry- and biology-based drug delivery strategies have been developed and explored (Figure 2). These strategies fall into two main categories — improving physicochemical properties, namely the lipidization of water-soluble molecules, and utilizing the transporter mechanisms in the BBB.

Lipidization. Because passive diffusion remains the preferred route into the brain, early attempts were focused on making a drug lead extremely lipophilic, or deriving from them more lipophilic prodrugs. This was encouraged, from a technological point of view, by the highly successful example of the diacetylated form of morphine, i.e., heroin. Heroin penetrates the BBB 100 times more effectively than morphine because it is more lipophilic. In many cases, however, the brain entry of more lipophilic compounds remains unchanged as the plasma concentrations of the drug or prodrug decrease because of increased distribution to the peripheral tissues and enhanced plasma protein binding. This was illustrated with lipophilic chlorambucil prodrugs. Despite increased brain-to-plasma ratios, the chlorambucil prodrugs did not demonstrate superior anticancer activity in disease models when compared with equimolar parent chlorambucil administration.⁹

Figure 2 Chemistry- and biology-based approaches to increase brain delivery of CNS-targeted therapeutic agents (modified from Pardridge 2003).8

Overcoming efflux transport. Evading the efflux transporter is a tremendous challenge in CNS drug delivery. Several different efflux transporters (e.g., P-glycoprotein, P-gp) are present in the BBB and function as clearance systems for both metabolic and catabolic compounds produced in the brain. Moreover, these efflux transporters recognize a wide diversity of xenobiotics, which contributes to their restricted BBB permeability. Some empirical rules to avoid P-gp expulsion have been devised. One such rule is that the sum of nitrogen and oxygen atoms in a structure should be eight or less. However, these rules are not hard and fast. Second-generation, nonsedating antihistamines, which are serendipitously effluxed by P-gp, nicely demonstrate the complexity involved in the function of P-gp. In one example, conversion of a CH₂OH group into a carboxylic acid group makes the compound P-gp substrate. Therefore, chemical modification of a drug with the hope of preventing its efflux transporter recognition is very challenging.

Figure 3 The chemical structures of L-phenylalanine, melphalan and DL-NAM.

Carrier-mediated drug delivery

Recently, more efforts have been made to develop novel drug or prodrug candidates that can exploit the endogenous inward nutrient transport systems at the BBB in their brain penetration. The drug or prodrug is designed to structurally resemble the endogenous ligands of a specific transport system, which recognizes the compound as its own substrate and transports it across the BBB.^10–12 More than 20 carrier-mediated transporter proteins have been identified in cerebral capillaries of the BBB including transporters for glucose, amino acids, choline, vitamins, low-density lipoproteins and nucleosides.^3,12

Figure 4 Decarboxylation of L-dopa to its active form dopamine.

The determinant of brain permeability for a particular transport system is the V_max/K_m ratio. While V_max indicates the overall capacity of a transporter, K_m represents affinity. Whereas a higher V_max indicates a higher capacity, a lower K_m indicates a higher affinity for a transporter. In general, when the affinity of the transporter increases, the capacity decreases. Table 1 shows estimates of the maximum permeability attainable (i.e., V_max/K_m), also known as the BBB permeability surface (PS) area product, for particular substrates of various nutrient transporters. Therefore, of the transporter systems present at the BBB, the carriers especially for the large amino acids (e.g., LAT1) and glucose (e.g., GLUT1) have a sufficiently high transport capacity to hold promise for significant drug delivery to the brain.

Table 1 BBB nutrient carriers.

Because many drug molecules have similar structural properties to endogenous substrates, some membrane transporters can take part in drug transport as well. Modifying chemical drugs so as to allow the drug to be recognised by specific transporters, but still maintain therapeutic efficacy, has proven to be very challenging. One attractive approach is to conjugate an endogenous transporter substrate to the active drug molecule in a bioreversible manner; that is, to utilize the prodrug approach.

Figure 5 Chemical structure of L-tyrosine prodrug of ketoprofen and its rat brain uptake. Km and Vmax are 22.49 ± 9.18 μM and 1.41 ± 0.15 pmol/mg/min (mean ± SD, n=53), respectively.

LAT1. The LAT1 is expressed both on the luminal and abluminal membrane of the capillary endothelial cells. It efficiently transports neutral L-amino acids that contain a relatively large and lipophilic R-substituents (e.g., L-phenylalanine and L-leucine) into the brain.^12–14 The brain uptake via LAT1 is sodium independent and stereospecific, and structural requirements for compound binding to LAT1 are fairly simple. Similar to L-phenylalanine and L-leucine, the potential substrate of LAT1 should have a positively charged amino group, a negatively charged carboxyl group and a hydrophobic side-chain. This was demonstrated by extensive studies of structure–transport relationships for LAT1 by Smith et al.¹⁵ , which showed a linear free energy relationship between affinity for the transporter and amino acid side-chain lipophilicity, determined by octanol–water partition coefficients.

This structure–transport relationship was further utilized to improve brain uptake of melphalan, a nitrogen mustard alkylating agent, which is transported into the brain predominantly via cerebrovascular LAT1-mediated transport.¹⁶ In contrast to melphalan, which showed affinity for LAT1 of only one-tenth of that for L-phenylalanine, an analog of melphalan (DL-NAM) with increased side-chain lipophilicity exhibited 100-fold greater affinity for BBB LAT1 than L-phenylalanine, and more than 1000-fold greater than melphalan itself. Additionally, the in situ brain perfusion method showed 40-fold greater unidirectional BBB influx rate for DL-NAM brain uptake than that for melphalan. Both melphalan and DL-NAM bear a very close structural resemblance to endogenous LAT1-substrates (Figure 3).

Figure 6 Structures of glucose prodrugs of chlorambucil and dopamine. Pro-moieties are indicated in red.

The most well-known example of drugs entering the brain predominantly via LAT1-mediated transport is L-dopa (Figure 4). Neurotransmitter dopamine is not able to cross the BBB because of its hydrophilic nature. However, the conversion of dopamine into α-amino acid, L-dopa, enables the uptake of dopamine into the brain via LAT1. L-Dopa is decarboxylated into dopamine in the brain tissue and in peripheral circulation, therefore, functioning as a prodrug of dopamine. Although approximately 95% of L-dopa is metabolized to dopamine in the peripheral tissues, the percentage of the remaining L-dopa has been therapeutic enough to apply this approach in clinics for more than 30 years.

Key points

Another approach to utilize LAT1 for BBB transport is to conjugate a small molecular drug with a LAT1 substrate, typically an amino acid. An amino acid L-tyrosine is a LAT1 substrate that has a phenolic hydroxyl group suitable for the conjugation of various structurally different drug molecules with a biodegradable linkage. In our studies, the L-tyrosine prodrug of ketoprofen demonstrated significant reversible inhibition of uptake into the brain of the radiotracer [¹⁴C]L-eucine in the in situ rat brain perfusion model, indicating that the prodrug binds to LAT1.¹⁷ More importantly, the prodrug entered the brain parenchyma, having both concentration-dependent and saturable uptake (Figure 5). In addition, the LAT1 inhibitor 2-aminobicyclo-(2, 2, 1)-heptane-2carboxylic acid significantly decreased the brain uptake of the prodrug, which further confirms that the drug-substrate conjugate is not only recognized, but also transported across the rat BBB by LAT1 — results that have not been earlier reported for any drug-substrate prodrug.

GLUT1. There are several glucose transporter genes that have been found to be localized to the BBB. GLUT1, which constitutes more than 90% of all BBB glucose transporters, is present both on the luminal and the abluminal membrane of the endothelial cells forming the BBB.¹⁸ GLUT1 transports glucose and other hexoses, and has the highest transport capacity of the carrier-mediated transporters present at the BBB, making it an attractive transporter for prodrug delivery.¹⁹

Several in vitro studies have showed that glycosyl derivatives of various drugs can bind to GLUT1 (Figure 6).^20,21 For example, a glucose-chlorambucil prodrug was able to inhibit the uptake of [¹⁴C]D-glucose into human erythrocytes.²¹ However, in these in vitro uptake studies the prodrug was an inhibitor rather than a substrate of GLUT1. In another study, glycosyl derivatives of dopamine substituted at the C-6 positions of glucose showed better affinity for GLUT1 in human erythrocytes than those substituted at the C-1 or C-3 positions (Figure 6).²⁰ These studies are consistent with a proposed model of the binding site of the GLUT1, which involves a hydrophobic site near the C6-position of bound glucose that can accommodate relatively bulky and lipophilic substituents.^22,23

Bonina et al.²⁴ tested both L-dopa and dopamine glycoside prodrugs conjugated with glucose at the C-3 position and galactose at the C-6 position by a succinyl linker with classic dopaminergic models, morphine-induced locomotion in mice and reserpine-induced hypolocomotion in rats. Both of the dopamine glycosidic prodrugs were more active in reversing the reserpine-induced hypolocomotion in rats than L-dopa or the L-dopa prodrugs. While conjugating dopamine with glycosides increased the pharmacological efficacy, the mechanism of brain uptake remained unclear.

Our own preliminary studies in the in situ rat brain perfusion model have shown that ketoprofen attached to glucose at the C-6 position significantly decreased the brain uptake of [¹⁴C]D-glucose indicating binding to the GLUT1 and, more importantly, was able to cross the BBB. The brain uptake of ketoprofen glucose prodrug was further decreased with the co-administration of D-glucose indicating carrier-mediated transport.

Conclusion

Multiple endogenous transport systems at the BBB offer an attractive approach to deliver drugs into the brain via carrier-mediated delivery. Of the transporter systems, the carriers for the large amino acids (LAT1) and glucose (GLUT1) have a sufficiently high transport capacity to hold promise for significant drug delivery. To enable transport into the brain via carrier-mediated transport, drugs need to be engineered to resemble endogenous substrates or, alternatively, conjugate drugs with substrates in a bioreversible manner to form brain-permeable prodrugs.

Jarkko Rautio is a professor at the department of pharmaceutical chemistry.

Krista Laine is a postdoctoral researcher at the department of pharmaceutical chemistry.

Mikko Gynther is a postgraduare student at the department of pharmaceutical chemistry. All at University of Kuopio (Finland).

References

1. M.W. Bradbury, Fed. Proc. 43(2), 186–190 (1984).

2. W.M. Pardridge, Drug Discov. Today 7(1), 5–7 (2002).

3. W.M. Pardridge, NeuroRx 2(1), 3–14 (2005).

4. W.M. Pardridge, Pharm. Res. 24(9),1733–1744 (2007).

5. D.J. Begley, Curr. Pharm. Des. 10(12), 1295–1312 (2004).

6. W. Loscher and H. Potschka, NeuroRx 2(1), 86–98 (2005).

7. S. Ohtsuki and T. Terasaki, Pharm. Res. 24(9), 1745–1758 (2007).

8. W.M. Pardridge, Curr. Opin. Drug Discov. Devel. 6(5), 683–691 (2003).

9. N.H. Greig et al., Cancer Chemother. Pharmacol. 25(5), 311–319 (1990).

10. T. Halmos et al., STP Pharma. Sci., 7(1), 37–42 (1997).

11. C. Yang, G.S. Tirucherai and A.K. Mitra, Expert Opin. Biol. Ther. 1(2), 159–175 (2001).

12. Q.R. Smith, International Congress Series 1277, 63–74 (2005).

13. R.J. Boado et al., Proc. Natl. Acad. Sci. USA 96(21), 12079–12084 (1999).

14. R. Duelli et al., J. Cereb. Blood Flow Metab. 20(11), 1557–1562 (2000).

15. Q.R.Smith et al., J. Neurochem 49(5), 1651–1658 (1987).

16. Y. Takada et al., Cancer Res., 52(8), 2191–2196 (1992).

17. M. Gynther et al., J. Med. Chem. 51(4), 932–936 (2008).

18. C.L. Farrell and W.M. Pardridge, Proc. Natl. Acad. Sci. USA 88(13), 5779–5783 (1991).

19. B.D. Anderson, Advanced Drug Delivery Reviews 19(2), 171–202 (1996).

20. C. Fernandez et al., Org. Biomol. Chem. 1(5), 767–771 (2003).

21. T. Halmos et al., Eur. J. Pharmacol. 318(2–3), 477–484 (1996).

22. J.E.G. Barnett et al., Biochemistry Journal 145(3), 417–429 (1975).

23. J.E.G. Barnett, G.D. Holman and K.A. Munday, Biochemistry Journal 131(2), 211–221 (1973).

24. F. Bonina et al., J. Drug Target 11(1), 25–36 (2003).