Formulation and the transport mechanisms of a glutathione-modified liposomal system for brain-targeted delivery

Abstract

Background: Brain disorders are the leading cause of disability-adjusted life years and the second leading cause of death globally. Yet, effective management is limited due to the protective nature of the blood brain barrier (BBB), limiting the transport of medicines into the brain. Glutathione PEGylated liposomes (GSH-PEG liposomes) can follow the transcellular transport pathway by exploiting GSH receptors of brain endothelial cells (BECs) of the BBB for active transport of drug molecules into the brain were developed. As a result, the overall clinical benefit of model drugs could be improved by encapsulating them in liposomes. However, while the GSH-PEG liposomes could improve brain delivery efficiency in animals, the detailed formulation process, such as GSH conjugation, and in particular, the whole view of the transport process of the liposomes or cargo across the BECs to the brain, in brain-targeted delivery, is still largely unknown. Emerging evidence shows that transcellular mechanisms involving the transport machinery of BECs, such as the sorting endosomes and multivesicular bodies (MVB), can lead to exocytosis of extracellular vesicles. Hence, it is essential to understand how liposome utilises the transcellular transport pathway and how some selected formulation factors affect this pathway in developing more effective brain-targeted liposomal formulations. Aim: The overall aim of this Ph.D. thesis was to develop GSH-PEGylated liposomes and investigate their transcellular transport mechanisms for brain-targeted delivery of medicines. The effect of glutathione (GSH) modification and fusogenic/pH-sensitivity of the liposomes on their transcellular mechanisms in human brain endothelial cells are investigated by comparing GSH-PEGylated-liposomes (GSH-PEG-L, non pH-sensitive) with a non GSH modified PEGylated liposome (PEG-L), or GSH- PEGylated pH-sensitive liposomes (GSHPEG- pSL). The in vitro transcellular transport process (using fluorescent dyes), in vivo pharmacokinetics and brain distribution, were studied with gemcitabine used /as a model drug. Finally, dual drug-loaded (methylprednisolone and fingolimod) PEG-L and GSH-PEG-L were developed potentially to manage multiple sclerosis, an autoimmune disease destroying the myelin sheath that insulates nerve fibres in the brain and spinal cord. Methods: Firstly, a systematic approach was followed to develop PEG-L and GSHPEG- liposomes with optimised GSH density in Chapter 3. To support the formulation development, an isocratic stability-indicating high-performance liquid chromatography (HPLC) method for indirect quantification of GSH following its derivatisation with 5,5'-dithiobis-(2-nitrobenzoic acid) or DTNB was initially developed. The method was then applied in monitoring the formation of the glutathione-polyethylene glycol-1,2-distearoyl-sn-glycero- 3-phosphoethanolamine (GSH-PEG-DSPE) conjugate via Michael’s addition reaction. The synthesis of the GSH-PEG-DSPE conjugate was confirmed by proton nuclear magnetic resonance (1H NMR) spectroscopy. Subsequently, the optimum conjugation conditions such as GSH concentration, GSH-lipid ratio, reaction temperature and time were defined. PEG-L was prepared by thin-film hydration-extrusion method using the lipids, DPPC/cholesterol/mPEGDPPE2000 at molar ratios of 6:4:0.5. For GSH conjugation, mPEG-DPPE 2000 (4%) was replaced with DSPE-PEG2000 maleimide to form PEG-L-maleimide, while pH-sensitive liposomes (pSL)-maleimide were formed with DSPC/DOPE/CHEMS/cholesterol/mPEGDPPE2000/ DSPE-PEG2000 maleimide (at 2:4:2:2:0.1:0.4 molar ratios). Using the optimised conditions, GSH-PEG-liposomes were prepared by two methods: the post-insertion of GSHPEG- DSPE into preformed liposomes without DSPE-PEG maleimide (6:4:0.1) or direct conjugation of GSH to PEG-L-maleimide or pSL-maleimide. Next, the effect of ligand conjugation (Chapter 4) and fusogenic/pH-sensitivity (Chapter 5) of the liposomes on their whole transcellular transport process across the BBB were evaluated by comparing the developed formulations (GSH-PEG-L and PEG-L or GSHPEG- pSL). Fluorescent dyes (calcein or rhodamine-DOPE representing the liposomes core and membrane, respectively) or gemcitabine were used as a model drug. The physicochemical properties of the gemcitabine liposomal formulations were analysed, including particle size, zeta potential, and morphology under the cryo-transmission electron microscope (TEM), drug loading (DL%), and drug release profile. Cellular uptake of fluorescent-labelled liposomes was investigated in human brain endothelial cells (hBMECs) and rat astrocyte cells using fluorescence and confocal laser scanning microscopy. Following endocytosis studies, the intracellular trafficking, including exocytosis of liposomes, was evaluated by confocal live cellimaging with the aid of colocalisation analysis and fluorescence microscopy. Exocytosis was further assessed by isolation of vesicles released from BECs following liposome treatment and characterisation in terms of their particle size, number, morphology, and content. The role of Rab11A positive endosomes in exocytosis of GSH-PEG-liposomes was explored since it directs the sorting of lipid molecules for exocytosis or recycling depending on the endocytosis pathway. Furthermore, an in vitro BBB model was developed and used to evaluate the transcellular transport efficiency of GSH-PEG-L compared with PEG-L. The pharmacokinetics and brain tissue distribution of gemcitabine PEG-L, GSH-PEG-L and GSH-PEG-pSL formulations were studied in healthy Sprague-Dawley rats. Lastly, in Chapter 6, PEG-L and GSH-PEG-L were selected for a preliminary study involving dual drug (methylprednisolone hemisuccinate, MPS and fingolimod) loading into the liposomes, potentially for managing multiple sclerosis. To support the formulation development, a validated HPLC method for simultaneous assay of methylprednisolone and fingolimod was developed with the aid of multiple regression analysis. PEG-L and GSHconjugated PEG-L (GSH-PEG-L, 4% GSH density) were prepared. Methylprednisolone hemisuccinate (MPS) was actively loaded into liposomes using a calcium acetate gradient, and fingolimod was passively loaded in the bilayers. Drug-loaded liposomes were characterised. Results and discussion: The successful development of an isocratic-stability indicating HPLC assay for quantification of GSH provided a rapid analysis time (10 min) of GSH-lipid conjugation efficiency and validated to be highly reproducible and reliable (Chapter 3). The optimal conjugation conditions, such as GSH concentration (≥ 2 mg/ml), GSH-lipid ratio of 1:0.67 in HEPES buffer at 22 ºC for 15 h, led to complete lipid conjugation in GSHPEG- liposomes with a GSH density of 4% either by post-insertion or direct conjugation methods. The GSH-PEG-liposomes were < 120 nm in size and negatively charged, which were ideal for enhanced in vivo stability and transport across BBB. The gemcitabine PEG-L and GSH-PEG-L formulations had comparable size (114 nm) and zeta potential with non-drug loaded formulations (Chapter 4). The in vitro drug release study revealed a sustained drug release property of GSH-PEG-L and PEG-L with approximately 40-50% of the gemcitabine remaining within 24 h, which is favourable for targeted drug delivery. The presence of GSH increased the cellular uptake of the liposomes up to 3-fold in hBMECs but not in astrocytes in a dose-dependent manner, confirming preferential GSHtransporters in hBMECs. Internalised liposomes are extruded from hBMECs via release of extracellular vesicles (EVs). While, more GSH-PEG-L and payload could exploit small EVs (sEVs) and medium EVs (mEVs, microvesicles) for exocytosis, more payload of PEG-L was released in sEVs. The transport efficiency of GSH-PEG-L across the in vitro BBB model was approximately 3% in 24 h, which was 1.7-fold higher than that of PEG-L (p< 0.05). This completes the whole transport process for liposomes in human brain endothelial cells. In rats, PEG-L and GSH-PEG-L showed comparable long circulation properties due to similar PEGylation (5%) and reduced gemcitabine clearance, thus extending the drug availability for transport into the brain tissue. At 8 h post-injection, 3.8% of the total injected dose (ID) of gemcitabine (4 mg/kg) was found in the brain of the GSH-PEG-L group, and 2.8% ID in the PEG-L group. A brain: blood concentration ratio of 1.27 ± 0.12 suggested that an active transport mechanism to cross the BBB for GSH-PEG-L ) in healthy rats. On the other hand, GSH-PEG-pSL showed significantly (p <0.05) higher cellular uptake (1.35 to 1.6-fold) compared to GSH-PEG-L due to the fusogenic property of DOPE lipid (Chapter 5). After endocytosis, GSH-PEG-pSL showed the ability for endosome escape and cytosolic delivery of its payload while GSH-PEG-L retained in endo-lysosomes initially and subsequently extruded via extracellular vesicles (EVs). Interestingly, GSH-PEG-Ltreatment increased the release of mEVs and sEVs from hBMECs 2.3- and 7.9-fold, respectively, while GSH-PEG-pSL slightly reduced the number of mEVs due to possible interaction with the plasma membrane. Regardless of pH sensitivity of the GSH-PEGliposomes, these vesicles were positive for CD144 proteins which are enriched on endothelial cell EVs, confirming their origin from hBMECs. In addition, cryo-TEM showed that GSHPEG- L were exocytosed as whole liposomes sheathed in a biomembrane. Consistently, much more GSH-PEG-L membrane (Rh-PE labelled, 2.7-fold) and payload (calcein, 2.5-fold) were found in the EV lysates (more in mEVs) than those of GSH-PEG-pSL. More exocytosis of GSH-PEG-L compared to GSH-PEG-pSL involved Rab11A positive endosomes. Lastly, the successful development and validation of an HPLC assay for simultaneous analysis of MPS and fingolimod supported the loading of the two drug molecules into liposomes. MPS-PEG-L and GSH-PEG-L were negatively charged (-19.3 ± 0.9 mV or -30.5 ± 2.0 mV) with an average particle size of 113 ± 1.2 nm or 109 ± 1.5 nm, respectively. The loading of fingolimod with MPS increased the particle size and reduced the negative surface charge of the liposomes (8.46 ± 0.2 mV and -9.96 ± 0.3 mV in PEG-L and GSH-PEG-L, respectively). The EE (%) and DL (%) of MPS in dual drug-loaded PEG-L was 2.2 and 2.1- fold higher (49.3 ± 0.8% and 10.6 ± 0.1%, respectively) than GSH-PEG-L (21.4 ± 1.8% and 4.9 ± 0.4%, respectively). In contrast, similar DL (%) and EE (%) for fingolimod (approximately 9% and 99%, respectively) were found in both formulations. The loading of fingolimod in MPS-fingolimod PEG-L and GSH-PEG-L improved the stability of MPS. Conclusion: This Ph.D. research first optimised the method for GSH conjugation to liposomes for brain-targeted drug delivery. Overall, the transcellular transport mechanisms of the GSH-modified liposomal system across the brain endothelial cells of the BBB to the brain involved endocytosis, endosomal sorting, and exocytosis. This research highlighted the exocytosis pathway of GSH-modified non pH-sensitive liposomes via the form of vesicles from the BECs, which could be an effective mechanism for drug delivery to the brain in vivo. GSH conjugation on liposomes enhances the whole transcellular transport process, to improve drug delivery efficiency across the BBB.While, pH sensitivity of GSH-PEG-liposomes can increase cellular uptake and cytosolic delivery, it may suppress their transcellular transport process through the brain endothelial cells, therefore reducing drug delivery to the brain. In addition, the dual drug- (MPS and fingolimod) loaded PEG-L and GSH-PEG-L for brain-targeted delivery may potentially be useful in managing of multiple sclerosis and therefore further in vitro and in vivo studies would be of great interest.