Development of PEG-sheddable pH-sensitive liposomes for cytoplasmic drug delivery to pancreatic cancer

Abstract

Background: Conventional chemotherapy is still the main stay treatment for cancer, but its efficacy is limited by its nonspecific drug distribution and lack of selectivity. Compared to small molecular anticancer drugs, liposomes have many advantages including high drug loading, prolonged circulation time and enhanced tumor accumulation via the enhanced permeation and retention (EPR) effect. To further improve their specificity, the design of liposomes has been refined to exploit the three main features of tumor microenvironment, i.e. low extracellular pH, ‘leaky’ vasculature and hypoxia. Because of their abilities to undergo acid triggered destabilization, PEGylated pH-sensitive liposomes (pPSL) have been investigated as an effective drug carrier to confer preferential pH-sensitive tumor-targeted drug delivery. However, PEGylation, required for long circulation, also brings with it the ‘PEG dilemma’, which sterically hinders cellular uptake and subsequent endosomal escape at the target. As a result, drug activity is inevitably reduced after PEG modification. To achieve successful drug delivery for effective treatment, crucial issues associated with the PEG dilemma must be addressed. Aim: The overall aim of this thesis is to investigate a dual pH-responsive strategy, which involves the design and development of a pSL modified with cleavable PEGylation to overcome the PEG dilemma. Gemcitabine, a first-line chemotherapeutic drug for pancreatic cancer, is used as the drug model, and long circulation, intracellular delivery, endosome escape and tumor biodistribution abilities of the drug-loaded liposomes are systemically investigated. Methods: First two pH-cleavable PEG polymers containing acid labile hydrazone bonds between long PEG chains (PEG2000) and lipid anchors, (CHEMS in Chapters 3 and 4 and DPPE in Chapter 5) were designed and synthesized. The polymer synthesis was carried out via the reductive amination of Schiff’s base between the aromatic benzaldehyde group of PEG (PEGB) and amino group of hydrazide activated CHEMS / DPPE to form PEGB-Hz-CHEMS and PEGBHz- DPPE respectively. The structural characterization was performed by nuclear magnetic resonance spectroscopy (1H NMR) and mass spectrometric techniques. An isocratic highperformance liquid chromatographic (HPLC) method was developed to facilitate the characterisation of PEGB-Hz-CHEMS modified pPSL (CL-pPSL1). The composition of the mobile phase was optimised by a multiple linear regression model. Due to the lack of UVchromophore in the structure of DPPE, fluorescence spectroscopy was used for characterization and the development of PEGB-Hz-DPPE polymer modified liposomes (CL-pPSL2). An optimised pH-sensitive liposome system comprised of DOPE, DSPC, CHEMS and cholesterol (at a molar ratio of 4:2:2:2) with enhanced stability and pH-responsiveness was used based on our previous research. Thereafter, the cleavable PEG polymers were utilised to modify the surface of the pH-sensitive liposome membrane (pSL) using a post insertion technique with the aim of enhancing the intracellular delivery of the liposomes. The insertion efficiencies of the two cleavable polymers into liposomes at various concentrations (equivalent to 3, 5, and 10 mol% to lipids) were systematically investigated using the HPLC-UV method for CL-pPSL1 (see Chapter 3) and the Nile Red spectrometric method for CL-pPSL2 (see Chapter 5). Furthermore, the PEG conformation of the polymers (brush or mushroom) on the surface of the liposomes was determined and optimised to estimate their long-circulation property (see Chapter 3). Conventional pPSL coated with DSPE-PEG2000 were used as the control for both CL-pPSL (see Chapters 4 and 5). To increase gemcitabine loading (DL) and entrapment efficiency (EE), a small volume incubation method was used for drug loading. In vitro cell uptake and the endosomal escape abilities of the dual fluorescent labelled CL-pPSL in comparison to pPSL were observed using confocal laser scanning microscopy (CLSM). An investigation of the pathways of endosome escape was performed by the live cell imaging of lysotracker red labelled cells incubated with CL-pPSL2 and pPSL using CLSM (see Chapter 5). The HPLC method was used to measure the intracellular drug concentrations (Ci) of Mia PaCa- 2 cells following their exposure to free drug, pPSL and CL-pPSL1 for 2 h. The cytotoxicity of gemcitabine-loaded CL-pPSL1 and CL-pPSL2 was investigated in Mia PaCa-2 pancreatic cell lines using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. IC50 values were compared with pPSL and the free drug solution. The pharmacokinetics of CLpPSL1 and CL-pPSL2 with respect to pPSL and the free drug solution was studied in Sprague Dawley (SD) rats. With some successful tumor growth in animal models, Mia PaCa-2 pancreatic tumor-bearing CD-1 nude mice were used to investigate the biodistribution (n = 3) of CL-pPSL in comparison with pPSL and the free drug solution. Results and discussion: The PEGB-Hz-CHEMS conjugate was synthesized with a yield of 90% and the structure was confirmed by 1H NMR and mass spectrometry. An HPLC method for the simultaneous analysis of PEGB and PEGB-Hz-CHEMS was established using a multiple linear regression model. Retention times for both the analytes were found to be functions of mobile phase composition as represented by regression equations. The established HPLC method provided rapid analysis for both analytes in 15 min and was validated to be highly reproducible and reliable. Using this simultaneous HPLC assay, the pH-sensitive PEG detachment rate of PEGB-Hz-CHEMS was determined as 50% at pH 6.5, and 80% at pH 5.5 in 1 h, while it was found to be relatively stable at pH 7.4 with a half-life of 24 h. The stabilityindicating HPLC assay also revealed that at acidic pH, two degradation products, PEGB-HZ and PEGB were formed via the cleavage of the hydrazide and the hydrazone bond respectively. The presence of the hydrazide bond was found to enhance the pH-sensitivity of the polymer without affecting its stability at pH 7.4 (discussed in Chapter 3). Liposomes consisting of DOPE, DSPC, CHEMS and cholesterol (4:2:2:2) with desirable sizes (100-120 nm) were obtained. This formulation was then post-inserted with PEGB-Hz-CHEMS for CL-pPSL1 and DSPE-PEG2000 for pPSL using a post-insertion technique. The maximum insertion efficiency of the PEGB-Hz-CHEMS into the liposomes was achieved by incubating with a polymer solution equivalent to 5 mol% to total lipids for 24 h at 4 0C. The grafting densities of PEGB-Hz-CHEMS on the liposomes was determined as 1.7 mol% by HPLC, achieving the ‘mushroom’ conformation on the liposomal surface, which is essential for their long circulation in vivo. Chapter 4 discusses the development and complete characterisation of CL-pPSL1 compared to pPSL1. Drug-loaded polymer-coated liposomes with an average size of 135-140 nm showed accelerated pH-triggered drug release at endosomal pH 5.0 compared to pPSL. Nevertheless, the results of the endosome escape study on Mia PaCa-2 cells suggested that the PEGylation of pPSL significantly decreased their endosome escape, with CL-pPSL1 achieving substantially enhanced calcein release to cytosol compared to pPSL. This correlated with cell uptake results, where CL-pPSL1 showed a 2-fold enhanced intracellular concentration compared to pPSL (P < 0.01). MTT cytotoxicity assays revealed that gemcitabine containing CL-pPSL1 showed a significantly higher cell killing effect compared to that of pPSL (52.4 nM vs 79.5 nM, P ≤ 0.05). In SD rats, the plasma pharmacokinetics study revealed that the clearance of both pPSL and CL-pPSL1 were 7 times slower compared to free gemcitabine, indicating their better long circulation abilities. However, at 12 h post intravenous injection, the gemcitabine-loaded CLpPSL1 resulted in a 2.5 times lower plasma concentration and 1.7 times lower AUC than pPSL, and simultaneously a 3-times higher liver accumulation compared to pPSL. This could be due to the poor insertion efficiency of the PEGB-Hz-CHEMS polymer, resulting in a lower degree of PEGylation (1.7% vs 5% of pPSL), leading to their more rapid clearance by the host reticuloendothelial system. The reduced insertion efficiency into the liposomes was due to the shorter lipid chain of the CHEMS, and therefore a lower critical micellar concentration of the polymer, which facilitated the majority of the polymer to exist as micelles. To overcome the issues encountered with the CL-pPSL1 polymer, we synthesized a second pHsensitive PEG polymer, PEGB-Hz-DPPE, by replacing the lipid anchor CHEMS with DPPE, a phospholipid (see Chapter 5). PEGB-Hz-DPPE was synthesized and confirmed by 1H NMR spectroscopy. pH-dependent degradation studies of polymeric micelles using DLS showed that, after 2 h of incubation, polymeric micelles at pH 5.0 displayed a rapid increase in particle size compared with pH 7.4. By monitoring the color of the supernatant post incubation of the liposomes with Nile Red labelled micelles, the insertion efficiency of polymeric micelles of PEGB-Hz-DPPE into the liposomes was found to be > 95%. In addition, CL-pPSL2 showed a rapid pH-triggered gemcitabine release compared to pPSL. In Mia PaCa-2 cells, the CL-pPSL2 showed rapid cellular uptake and endosome escape abilities, compared to pPSL. Interestingly, live cell imaging results clearly showed a fusion mechanism of endosome escape for CL-pPSL, and that CL-pPSL2 could rapidly escape into cytoplasm compared to pPSL. The plasma pharmacokinetics study revealed that free gemcitabine cleared 7-times faster compared with both CL-pPSL2 and pPSL. No significant difference in Vd, CL or the terminal T1/2 was observed between CL-pPSL2 and pPSL, indicating that cleavable PEGylation did not influenced the long circulation of CL-pPSL2. Because of its dual pHresponsiveness, increased cell uptake, rapid endosome escape and higher intracellular drug delivery abilities, CL-pPSL2 showed higher accumulation in tumors compared to pPSL (1.5- times) and the free drug solution (6-times), suggesting its potential for improved therapeutic efficacy. Conclusion: Dual pH-responsive liposomal tumor targeting successfully improved the cytotoxicity of gemcitabine in cancer cells by enhancing intracellular cytoplasmic delivery, due to its rapid pH-sensitive PEG detachment and endosomal escape abilities. Furthermore, the drug-loaded CL-pPSL2 was able to improve the tumor distribution of gemcitabine without compromising its stealth properties. In fact, for the first time, a dual pH-responsive strategy was reported with long circulation and pH-responsive tumor targeting abilities. However, future anti-tumor studies are still required to confirm if the enhanced tumor targeting abilities of these cleavable liposomes can result in greater therapeutic potential.