This invention relates to new peptides which are useful for carrying drugs through the blood-brain barrier. PRIOR ART

There is considerable interest in the development of new systems able to carry drugs efficiently and specifically to the target tissues in which the substances are to perform their therapeutic action. One of the most difficult challenges is represented by the blood-brain barrier (BBB), which regulates access to the brain by the substances present in the bloodstream in a highly selective manner. The BBB is formed by special endothelial cells joined by tight junctions. This membrane selectively regulates the entry of drugs which could have therapeutic value in the treatment of psychiatric and other disorders if they reached the neurones. One approach to crossing the BBB is represented by the synthesis of suitable prodrugs; recent studies have also investigated the use of liposomes, nanoparticles (NPs), small polymeric colloidal particles (1-1000 nm), or solid lipid nanoparticles into which the drug can be adsorbed or included (1,2).

The use of NPs has the advantage that a large number of molecules of the drug can be carried by each NP, and that they mask the characteristics of the drug to enable it to cross the BBB and protect it against enzymatic degradation. Different strategies have been proposed to enable NPs to cross the BBB, most of which are based on modifications to their hydrophobic surface. NPs mainly consist of polycyanoacrylate (such as poly(butylcyanoacrylate, PBCA), and a hydrophilic polymer such as polysorbate 80 can be adsorbed onto their surface (2); another strategy involves the preparation of NPs from an amphiphilic copolymer wherein the hydrophobic component is able to form the solid phase (the body of the NP),

while the hydrophilic component remains on its surface, facing outwards. This copolymer can be represented by polyethylene glycol (PEG) and n- hexadecylcyanoacrylate (PEG-PHDCA) (3). While it has been postulated on the basis of experimental data that the presence of the surfactant polysorbate 80 may act on the BBB, causing an increase in its permeability, NPs constituted by PHDCA PEGylate are able to cross the BBB without modifying its permeability (3); PEGylation increases the plasma half-life of NPs, and it has been suggested that this may be the reason why they are able to penetrate into the brain to a greater extent than other formulations (3). If the NPs are not coated with surfactant, they remain inside the blood vessels (2). Various mechanisms have been postulated to explain how NPs cross the BBB (1,4). However, although it is important for NPs to be able to cross the BBB, it is also extremely important for them to have the brain as their priority target, so that they can be used with drugs aimed at the CNS, not at other organs or apparatus.

Receptor-mediated endocytosis is responsible for the fact that the BBB is crossed not only by molecules with a low molecular weight such as thiamine, but also by aggregates with a high molecular weight such as low-density lipoproteins (LDL), and by other macromolecules. Specific receptors for insulin, insulin-like growth factors (IGF-I, IGF-2), angiotensin II, atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), interleukin-1 (IL-I) and transferrin have been identified in the brain capillaries. The transcytosis of these molecules through the BBB in vivo has so far only been demonstrated for insulin and transferrin (1). NPs consisting of wax and a surfactant (Brji 78®) conjugated with thiamine (5), or nanogels consisting of flexible hydrophilic polymers of nanometric dimensions (nanogels) composed of polyethyleneimine crosslinked to poly(ethyleneglycol), which in turn are derivatised with transferrin or insulin,

have been created specifically to reach the CNS (6,7).

Peptides are polar compounds, but various proteins are able to cross the lipid cell membranes; there is great interest in studying the possibility of increasing the cell permeability of liposomes, proteins and NPs by using membrane-permeable peptides and proteins as carriers (8). Although the primary structure of the peptides able to cross the cell membranes is highly variable (9), some of their general characteristics are known. It has been postulated that a certain abundance of positive charges due to the presence of

Arg (10-12), and the presence of aminoacids with large, hydrophobic side chains (13) inserted in the peptide sequence, play a crucial role in effective cell uptake.

Recently, various synthetic analogues of opioid peptides have proved able to cross the BBB, and this permeability can be increased by the presence of sugar residues (glucose, lactose, etc.); however, the mechanism whereby these substances cross the BBB is not known. In particular, an opioid peptide able to cross the BBE, known as MMP-2200, with the formula H ₂ N-Tyr-D- Thr-Gly-Phe-Leu-Ser-O-β-D-lactose-CONH ₂ , has been described (14).

The derivatisation of PLGA with short membrane-permeable peptide chains was recently successfully used to prepare NPs able to penetrate into cultured HaCat cells. In this case, PLGA was derivatised with the peptide Tat 49-57, which is the transduction domain of the Tat protein of HIV virus type 1; the NPs thus obtained were carried to the nucleus of the cultured cells without causing cytotoxicity (15). However the need remains for alternative, more effective and/or safer delivery systems. DESCRIPTION OF THE INVENTION

It has now been discovered that the peptide sequence:

H ₂ N-Gly-Phe-D-Thr-Gly-Phe-Leu-Ser-CONH ₂ can be opportunely used to carry through the blood-brain barrier drugs

conjugated to said sequence either directly or via a linear or branched polyvalent spacer, wherein other aminoacids can replace the first two aminoacids of the N-terminal portion, the order of which can be reversed, and their number may be different from two. A first aspect of the invention therefore relates to a peptide having the sequence reported above, wherein the serine residue may be functionalised with sugar residues, with a C- or O-glycoside bond.

The sugar residue is preferably a residue of glucose, galactose, lactose, mannose or xylose, and more preferably an O-β-D-glucoside residue. The invention also relates to a conjugate of said peptides with a pharmaceutically acceptable polyester or polyamide polymer, nanoparticle systems comprising said conjugates, and pharmaceutical compositions comprising said nanoparticle systems.

DESCRIPTION OF FIGURES Fig. 1. ¹ H-NMR spectrum of conjugate PLGA-HN-Gly-Phe-D-Thr-Gly-

Phe-Leu-Ser-CONH ₂ .

Fig. 2. ¹ H-NMR spectrum of conjugate PLGA-HN-Gly-Phe-D-Thr-Gly- Phe-Leu-SerO-β-D-glucose-CONH ₂ .

Fig. 3. ESCA spectrum of NPs prepared with PLGA-HN-Gly-Phe-D- Thr-Gly-Phe-Leu-Ser-CONH ₂ .

Fig. 4. ESCA spectrum of NPs prepared with PLGA-HN-Gly-Phe-D- Thr-Gly-Phe-Leu-SerO-β-D-glucose-CONH ₂ .

Fig. 5. SEM image of NPs prepared with PLGA-HN-Gly-Phe-D-Thr- Gly-Phe-Leu-Ser-CONH ₂ . Fig. 6. SEM image of NPs prepared with PLGA-HN-Gly-Phe-D-Thr-

Gly-Phe-Leu-SerO-β-D-glucose-CONH ₂ .

DETAILED DESCRIPTION OF THE INVENTION

Examples of polymers which can be advantageously used to prepare the

conjugates according to the invention include polymers or copolymers of biodegradable aliphatic hyα"roxyacids, preferably lactic acid and/or glycolic acid. The copolymer poly(D ₅ L-lactide-cø-glycolide) (PLGA) or poly(D,L- lactic) acid is particularly preferred. The copolymer PLGA is a biodegradable polyester, approved by the American FDA, which breaks down without inducing inflammation or immune reactions, and is therefore particularly suitable for the purposes of this invention.

The peptides according to the invention can be conjugated to PLGA or other polymers containing reactive groups (typically carboxyl groups) with the amino-terminal groups of the peptide according to methods already known in themselves, such as by activation of the carboxyl group with N-hydroxysuccinimide in the presence of dicyclohexylcarbodiimide, and subsequent formation of an amide bond with the N-terminal (unprotected) portion of the peptide. Alternatively, the peptide can be conjugated through a suitable all-purpose spacer, such as the aminoacid "Behera's amine". The molecular weight of the polymer used for the conjugation to the peptide is not critical, but typically falls into the interval between 1000 and 50,000 Da; the degree of functionalisation of the polymer with the peptide is between 30 and 80% of the theoretical value. The conjugates thus obtainable can be used to prepare nanoparticle systems according to known techniques, for example as described in (18) and in T. Niwa et al., J. Pharm. Sci. (1994) 83, 5, 727-732; C. S. Cho et al., Biomaterials (1997) 18, 323-326; T. Govender et al., J. Control. ReI. (1999) 57, 171-185; and M. F. Zambaux et al., J. Control. ReI. (1998) 50, 31-40. In addition to the PLGA-peptide conjugate, the nanoparticles (NP) according to the invention will contain a drug, a suitable surfactant, and a pharmaceutical carrier such as water or aqueous saline solutions.

Other micro- or nanoparticle systems (liposomes, microemulsions etc.)

could be considered as an alternative to NPs.

The drugs which can be advantageously used according to this invention are obviously those which are required to perform their pharmacological action in the brain. Examples of these drugs include antibacterial, antiviral, psychotropic, antidepressant, anti-tumoral and antimigraine agents, analgesics, antagonists or agonists of receptors present in the central nervous system, antibodies, antisense oligonucleotides, hormones, narcotic analgesics, and diagnostic agents (paramagnetic complexes for MRI, radionuclides for nuclear medicine or diagnosis, etc.). The drugs can be included, adsorbed or absorbed into the nanoparticles or conjugated directly with the peptide or copolymer by means of a covalent bond which is labile in vivo, such as an ester or amide bond.

The invention therefore also relates to pharmaceutical or diagnostic compositions comprising a peptide or one of its conjugates as defined above, together with any excipients. The dose and administration route of the nanometric systems according to the invention will obviously depend on a number of factors, such as the toxicological and pharmacokinetic characteristics of the drug, the severity of the disorder to be treated, and the patient's condition, especially his/her age, weight and sex. The dose may in any event be determined on the basis of the results of pre-clinical trials and phase I clinical trials, as is usual in drug development.

The invention will be now be described in greater detail by means of the following Examples.

EXAMPLE 1 - Peptide synthesis The peptides were synthesised by solid-phase peptide synthesis (SPPS) using a Rink amide resin and the FMOC method. All the peptides were synthesised manually using a suitable reactor (scale of 0.5 g of resin, 5.00 ml washing volume) and agitation by anhydrous nitrogen flow. Rink amide resin

(4-[[(2',4'-dimethoxy-phenyl)(9-fluorenylmethoxycarbonyl)ami no]methyl] phenoxyl resin) with a degree of functionalisation of 0.63 mmol/g was used. The Ser used to prepare the peptide H ₂ N-Gly-Phe-D-Thr-Gry-Phe-Leu-Ser- CONH ₂ (Peptide 1) was protected as tert-butyl ether; the hydroxyl groups of FMOC-SerO-β-D-glucoside, used for synthesis of the peptide H ₂ N-Gly-Phe- D-Thr-Gly-Phe-Leu-SerO-β-D-glucose-CONH ₂ (peptide 2), were protected by formation of esters with acetic acid. The FMOC was removed with the piperidine/DMF (9:1, 2 x 5 min), followed by washing with DMF (5 x 1 min). FMOC release was monitored by UV spectrophotometry (I = 290nm). The couplings were performed by adding a solution in DMF (4ml) consisting of FMOC aminoacid (3 equiv.), TBTU (3 equiv.), HOBt (3 equiv.) and *Pr ₂ Net (6 equiv.) to the peptide present on the resin. The resin suspension was agitated by bubbling anhydrous nitrogen for 3 h, after which the degree of completeness of coupling was determined with the Kaiser test. When the peptide is entirely synthesised, the FMOC group is removed as reported above; the acetic groups present on the sugar residue are removed by treatment with H ₂ N-NH ₂ .H ₂ O/MeOH 4/1 (2 x 5 mL x 30 min, followed by 1 x 5 mL x 1 h), while the peptide remains bonded to the resin. The excess H ₂ N-NH ₂ -H ₂ O is then removed, and the resin is washed with MeOH (5 x 1 min) followed by CH ₂ Cl ₂ (5 x 1 min). Next, the peptide is released from the dried resin by treatment with CF ₃ COOH (TFA) (84%), anisole (4%) thioanisole (4%), ethanedithiol (4%) and H ₂ O (4%) (4 ml x 2 h); the solution is filtered to remove the depleted resin, and the resulting solution is cooled and precipitated by adding anhydrous ether. The solid thus obtained is filtered and washed several times with anhydrous ether; the crude peptide is then purified by dissolving in MeOH and precipitating with anhydrous ether. The peptides thus obtained are then analysed by mass spectroscopy (Q-TOF Global Ultima, Micromass, MALDI) (m/z): Peptide 1, est. 726.8, found 727.4 (M ⁺ );

peptide 2, est. 889.0, found 889.4 (M ⁺ ), and their purity is determined by HPLC (UV detector, I 210 nm; column Vydac 218, 4.6 x 250 mm, Cl 8, pore diameter 300 A, packing particle size 5 μm; mobile phase Solvent A: H ₂ O/0.1% TFA; solvent B: CH ₃ CNA).1%TFA; elution conditions: linear gradient from 95%A to 35%A in 35 min; flow rate 0.8 ml/min. R _t : peptide 1, 20.5 min, peptide 2, 19.11 min.; purity >90% in both cases).

EXAMPLE 2 - Conjugation of PLGA with peptides 1 and 2 Dicyclohexylcarbodiimide (19.0 mg, 93 μmol) and N-hydroxy- succinimide (11.0 mg, 93 μmol) is added to a solution of 1.00 g (88 μmol) PLGA RG503H (Boehringer Ingelheim) with a molecular weight of 11,000, determined by titration of the carboxyl groups at its ends (4.94 mg KOH/g of polymer) in anhydrous dioxane (5 ml); the solution is then agitated for 4 hours at a temperature ranging from the initial 4 ⁰ C to 20 ⁰ C. The dicyclohexylurea is then filtered, and the solution poured into ethyl ether. The insoluble polymer is separated by decanting and purified by dissolving in CH ₂ Cl ₂ and precipitating with anhydrous ether (twice). The polymer thus obtained is then dissolved in anhydrous DMSO (5 ml), and maintained under agitation at ambient temperature after addition of a solution of the peptide (80 μmol) in DMSO (1 ml) and triethylamine (17 μl, 120 μmol). After 4 hours at ambient temperature, the reaction solution is poured into ethyl ether; the modified polymer is separated by decanting and purified by solubilisation in CH ₂ Cl ₂ and precipitation with methanol (three times).

Confirmation of conjugation of PLGA with the peptide.

Conjugation was confirmed by ¹ H-NMR spectroscopy. Examination of the spectra shown in Figs. 1 and 2 demonstrates the presence of the signal at 7-8 ppm attributable to the presence of aromatic protons of the Phe's present in the peptide. The derivatisation yield of the PLGA polymer was determined by the ratio between the areas of the signals at 7.2-7.5 ppm, corresponding to

the aromatic protons of the Phe's, and the areas of the multiplet at 1.80-1.60 ppm, corresponding to the protons of the methyl groups of the polymer; in both cases (PLGA-peptide 1 and PLGA-peptide 2), the degree of functionalisation was 30 micromoles of peptide/g of polymer. EXAMPLE 3 - preparation of nanoparticles using the conjugates

PLGA-peptide 1 and PLGA-peptide 2

The nanoparticles were prepared by the nanoprecipitation technique (18). The conjugate PLGA-peptide 1 or PLGA-peptide 2, prepared as above (100 mg), and PLGA derivatised with fluorescein (17) (25 mg), were solubilised in acetone (8 ml). The solution thus obtained was poured slowly into deionised water (25 ml) containing Pluronic F68 (100 mg). After agitation at ambient temperature for 10 min, the organic solvent was removed at 30 ⁰ C at low pressure, and the final volume of the suspension adjusted to 10 mL with deionised water; the NPs were then purified by gel-filtration chromatography (Sepharose CL4B gel (160 ml), column 50 x 2 cm), using water as mobile phase, and freeze-dried without the presence of a cryoprotector. Analysis of NPs.

A scanning electron microscope (SEM) (XL-40 Philips, Eindhoven, Netherlands) was used to evaluate the diameter and morphology of the NPs. The samples were coated in an argon atmosphere with a 10 nm thickness of gold and palladium (Emitech K550 Sputter Coated, Emitech Ltd., Ashford, Kent, UK). Electron microphotographs of at least 500 NPs per preparation were evaluated with image analysis (Image Proplus, Media Cybernetics, Silver Spring, MD, USA) to determine the percentage dimensional distribution. Each sample of NP has an average diameter of 120 ± 45 nm, and presents a uniform distribution and an intact surface, both before and after freeze-drying and resuspension, thus confirming that the preparation method and resuspension do not alter the morphological characteristics of NPs.

In order to determine the position of the polymer-modifying hydrophilic groups (peptide chains), X-ray Induced Photoelectronic Spectrophotometry analysis (ESCA) was conducted with an 04-153 X-ray source instrument (PHI, Uvalca-PHI, Tokyo, Japan) and an EAIl hemispheric electron energy analyser (Leybold Optics, Germany), using MgK radiation αl,2 (£=1253.6 eV). The spectra were recorded in FAT (fixed retard ratio) mode with 190 eV of pass energy. The pressure in the chamber containing the samples was evaluated at approx. 10 ^~9 mbars. Data acquisition and processing was conducted with an RBD AugerScan 2. The spectra (reported in Figs. 3 and 4) show the presence of N on the surface of the NPs.

EXAMPLE 4 - Experiments in vivo: in situ brain perfusion technique.

The ability of NPs to cross the BBB was determined by an in vivo test (the in situ brain perfusion technique). The actual crossing of the BBB is usually determined by evaluating the pharmacological effect of a substance present in nanoparticle systems (dalargin, etc.) or using labelled products or polymers, followed by evaluation of the radioactivity present in brain homogenate; however, these methods do not demonstrate that the NPs have actually crossed the BBB. Fluorescent NPs have only been used in two cases (NP of PBCA-polysorbate 80 and NP of PHDCA PEGylate (3,16), which allows their subsequent determination in brain parenchyma by fluorescence microscopy. This detection system has the advantage of allowing effective evaluation of whether the BBB has been crossed, and the method was used in the present study because of this characteristic. The NPs according to the invention were rendered fluorescent by inclusion of 20% in weight of PLGA derivatised with fluorescein (17).

Male albino Wistar Hannover rats weighing 250 ± 30 g (Harlan, San Pietro Natisone) were used; the animals were housed for a period of 15 days at

25°C before the experiments, and fed on a standard diet with water on demand.

Each sample of NP was perfused into the rats after anaesthesia induced with an i.p. injection of an aqueous solution (2 mL/Kg) of xylazine (0.001 mg/Kg) and ketamine (0.125 mg/Kg). The behaviour of the NPs was determined by the in situ brain perfusion technique (5,19,20). The rectal temperature of the rats during the experiment was monitored and maintained at 37°C by a heating device connected to a feedback mechanism.

A suspension of NP in saline (1 mL, containing 5 mg of Np each animal weighting 250 g) was then administered without interrupting the perfusion of plasma-like fluid (19) with a 3 -way valve. The NP suspension (10 mg/0.5 mL of saline solution) was obtained by sonication and agitation in a vortex; the integrity of the NPs and the absence of aggregates were evaluated in advance by SEM analysis. After the rats has been killed, the brain was removed, subjected to preservation procedures by washing with saline solution (0.9% w/v), and frozen with liquid nitrogen. A number of sections were cut with a 5 μm thick cryotome. Some sections were stained with haematoxylin-eosin for observation of the histological morphology of the organ section. Some tissue samples were treated with DAPI (4'-6-diamidino-2- phenylindole), a compound able to form fluorescent blue complexes with double-chain DNA (21), to highlight the presence of the cell nuclei.

After treatment of the sections with DAPI, the samples were observed with a fluorescence microscope with dual-band excitation relating to DAPI and fluorescein isothiocyanate (FITC) and an emission filter for fluorescence images (enlargement 40-10Ox), and with a confocal microscope (SP2-AOBS, Leika, Bannockburn, IL, USA). Green fluorescence, due to the fluorescein- PLGA polymer, constitutes the visual marker for NPs.

Results

Although it is impossible to obtain quantitative data from the experiments conducted, the data analysis obtained from the in vivo study with fluorescence microscopy clearly shows that the NPs obtained with PLGA modified with the peptide H ₂ N-Gly-Phe-D-Thr-Gly-Phe-Leu-Ser-CONH ₂ (peptide 1), and to a greater extent with the peptide H ₂ N-Gly-Phe-D-Thr-Gly- Phe-Leu-SerO-β-D-glucose-CONH ₂ (peptide 2), are able to cross the BBB and penetrate into the brain parenchyma cells. However, NPs only constituted by PLGA are unable to cross the BBB, and remain in the blood vessels (capillaries).

In order to obtain a precise determination of the location of the NPs, studies were conducted with a confocal microscope. This detailed information cannot be obtained with fluorescence microscopy alone, which only provides information in a two-dimensional (2D) mode. The three-dimensional images obtained with confocal fluorescence microscopy enable the position of the NPs to be located (green fluorescence due to the presence of fluorescein) in relation to the nuclei of neuronal cells (blue fluorescence due to staining with DAPI).

There is no doubt that fluorescent NPs consisting only of PLGA are unable to cross the BBB. This result confirms that no damage to the BBB took place during the experiments. Conversely, NPs prepared with PLGA-peptide 1 and PLGA-peptide 2 were observed in the brain parenchyma, in close contact with the nuclear structures, and in some cases in the cytoplasm of nonvascular cells.

References

1. Reddy, J. S.; Venkateswarlu, V. Novel delivery systems for drug targeting to the brain. Drugs Fut. 29, 63 (2004). 2. Lockman, P.R.; Mumper, RJ.; Khan, M.A.; Allen, D.D. Nanoparticle technology for drug delivery across the blood brain barrier. Drug Dev. Ind. Pharm. 28, 1-13 (2002).

3. Calvo, P., Gouritin, B.; Chacun, H.; Desmaele, D.; D'Angelo, J.; Noel, J.P.; Georgin, D.; Fattal, E.; Andreux, J.P.; Couvreur, P. Long-circulating PEGylated polycyanoacrylate nanoparticles as new drug carrier for brain delivery. Pharm. Res. 18, 1157-1166 (2001).

4. Kreuter, J. Nanoparticulate systems for brain delivery of drugs. Adv Drug del Rev. 47, 65-81 (2001).

5. Lockman, P.R.; Oyewumi, M.O.; Koziara, J.M.; Roder, K.E.; Mumper, RJ.; Allen, D.D. Brain uptake of thiamine coated nanoparticles. J. Contr. ReI.

93, 271-282 (2003).

6. A.V. Kabanov, E.V. Batrakova, New technologies for drug delivery across the blood brain barrier. Curr. Pharm. Des. 10, 1355-1363 (2004).

7. Vinogradov, S.V.; Batrakova, E.V.; Kabanov, A.V. Nanogels for oligonucleotide delivery to the brain. Bioconj. Chem. 15, 50-60 (2004).

8. Lindgren, M.; Hallbrink, M.; Prochiantz, A.; Laugel, U. Cell penetrating peptides. Trends in Pharmacol. Sci. 21, 99-102 (2000).

9. Langel, U. Cell penetrating peptides, Processes and applications. CRC Press, Boca Raton (2002). 10. Futaki, S.; Suzuki, T.; Ohashi, W.; Yagami, T.; Tanaka, S.; Ueda, K.; Sogiura, Y. Arginine-rich peptides. An abundant source of membrane- permeable peptides having potential as carrier for intracellular protein delivery. J. Biol. Chem. 276, 5836-5840 (2001).

11. Mitchell, DJ.; Kim, D.T.; Steinman, L.; Fathman, C.G.; Rothbard, J.B. Polyarginine enters cells more efficiently than other polycationic homopolymers. J. Pept. Res. 56, 318-325 (2000).

12. Wender, P.A.; Mitchell, DJ.; Pattabiraman, K.; Pelkey, E.T.; Steinman, L.; Rothbard, J.B. The design, synthesis, and evaluation of molecules that enable or enhance cellular uptake: peptoid molecular transporters. Proc. Nat. Acad. Sci. USA 97, 13003 (2000).

13. Derossi, D.; Chassaing, G.; Prochiantz, A. Trojan peptides: the penetration system for intracellular delivery. Trends Cell Biol. 8, 84-87 (1998).

14. PoIt, R.; Palian, M.M.; Glycopeptide analgesics. Drugs Fut. 26, 561- 576 (2001).

15. Nam, Y.S.; Park, Y.; Han, S.H.; Chang, LS. Intracellular drug delivery using poly(D,L-lactide-co-glycolide) nanoparticles derivatized with a peptide from a transcriptional activator protein of HIV-I. Biotechnol. Lett. 24, 2093-2098 (2003).

16. Kreuter, J.; Alyantdin, R.N.; Kharkevich, D.A.; Ivanov, A.A. Passage of peptides through the blood-brain barrier with colloidal polymer particles (nanoparticles). Brain Res 674, 171-174 (1995). 17. Tosi, G.; Rivasi, F.; Gandolfi, F.; Costantino, L.; Vandelli, M.A.; Forni, F. Conjugated poly-(D,L-lactide-co-glycolide) for the preparation of in-vivo detectable nanoparticles. Biomaterials, in press.

18. Fessi, H.; Puisieux, F.; Devissaguet, J.P.; Ammoury, N.; Benita, S. Nanocapsule formation by interfacial polymer deposition following solvent displacement. Int. J. Pharm. 55, R1-R4 (1989).

19. Takasato, Y.; Rapaport, S.I.; Smith, Q.R. An in situ brain perfusion technique to study cerebrovascular transport in the rat. Am. J. Physiol. 247, H484-93 (1984).

20. Lockman, P.R.; Mumper, R.J.; Allen, D.D. Evaluation of blood-brain barrier thiamine efflux using the in situ rat brain perfusion method. J. Neurochem. 86, 627-634 (2003).

21. Xu, Z.; Pilch, D.S.; Srinivasan, A.R.; Olson, W.K.; Geacintov, N.E.; Breslauer, KJ. Modulation of nucleic acid structure by ligand binding: induction of a DNA-RNA-DNA hybrid triplex by DAPI intercalation. Bioorg. Med. Chem. 5, 1137-1147 (1997).