PR_b-Targeted PEGylated Liposomes for Prostate Cancer Therapy

Oct 28, 2008 - the bicinchoninic acid protein assay. Fibronectin is the natural ligand for R5β1, and a promising design for a fibronectin- mimetic pe...
3 downloads 0 Views 1MB Size
Download PDF
13518

Langmuir 2008, 24, 13518-13524

PR_b-Targeted PEGylated Liposomes for Prostate Cancer Therapy Do¨ne Demirgo¨z, Ashish Garg, and Efrosini Kokkoli* Department of Chemical Engineering and Materials Science, UniVersity of Minnesota, Minneapolis, Minnesota 55455 ReceiVed June 22, 2008. ReVised Manuscript ReceiVed September 11, 2008 In recent years, there has been considerable effort in designing improved delivery systems by including site-directed surface ligands to further enhance their selective targeting. The goal of this study is to engineer R5β1-targeted stealth liposomes (nanoparticles covered with poly(ethylene glycol) (PEG)) that will bind to R5β1-expressing LNCaP human prostate cancer cells and efficiently release the encapsulated load intracellularly. For this purpose, liposomes (with and without PEG2000) were functionalized with a fibronectin-mimetic peptide (PR_b) and delivered to LNCaPs. The amount of PEG2000 and other liposomal components were characterized by 1H NMR, and the amount of peptide by the bicinchoninic acid protein assay. Fibronectin is the natural ligand for R5β1, and a promising design for a fibronectinmimetic peptide includes both the primary binding site (RGD) and the synergy site (PHSRN) connected by a linker and extended off a surface by a spacer. We have previously designed a peptide-amphiphile, PR_b, that employed a hydrophobic tail, connected to the N-terminus of a peptide headgroup composed of a spacer, the synergy site sequence, a linker mimicking both the distance and hydrophobicity/hydrophilicity present in the native protein fibronectin (thus presenting an overall “neutral” linker), and finally the primary binding sequence. We have examined different liposomal formulations, functionalized only with PR_b or with PR_b and PEG2000. For PR_b-targeted PEGylated liposomes, efficient cell binding was observed for peptide concentrations of 2 mol % and higher. When compared to GRGDSPtargeted stealth liposomes, PR_b functionalization was superior to that of GRGDSP as shown by increased LNCaP binding, internalization efficiency, as well as cytotoxicity after incubation of LNCaPs with tumor necrosis factor-R (TNFR)-encapsulated liposomes. More importantly, PR_b is R5β1-specific, whereas many integrins bind to small RGD peptides. Thus, the proposed PR_b-targeted delivery system has the potential to deliver a therapeutic payload to prostate cancer cells in an efficient and specific manner.

Introduction Over the past decade, we have witnessed an explosion in research aimed at creating new and improved drug delivery systems driven by a strong need in clinical practice. Advances in materials science and biotechnology are permitting the development of new physical and chemical methods of drug delivery. A variety of delivery systems includes microspheres, nanoparticles, lipoproteins, soluble polymers, cells, micelles, and liposomes. Currently, the main problems associated with systemic drug administration are the necessity of a large drug dose to achieve high local concentration, nonspecific toxicity, other adverse side effects due to high drug doses, even biodistribution throughout the body, and lack of specific affinity for the pathological site.1 Drug targeting can bring a solution to all these problems. In order to improve upon different therapies in the future, drug delivery systems need to include site-directed surface ligands to further enhance their selective delivery. The next generation of active targeting carriers is based on delivery systems that appear to mimic the local bioadhesion, rolling, and stopping of activated leukocytes.2 Therefore, attention has been directed to cell adhesion molecules, such as integrins, for specific targeting. Integrins, and in particular the R5β1 integrin, have an impact on several dynamic processes such as adenovirus infection mediation, wound healing acceleration, providing a protection mechanism against Alzheimer’s disease, and acting as a promising target for breast, colon, prostate, and rectal cancer.3-13 * To whom correspondence should be addressed. Telephone: +16126261185. Fax: +16126267246. E-mail: [email protected]. (1) Torchilin, V. P. Eur. J. Pharm. Sci. 2000, 11, S81–S91. (2) Ranney, D. F. Biochem. Pharmacol. 2000, 59, 105–114. (3) Varner, J. A.; Cheresh, D. A. Curr. Opin. Cell Biol. 1996, 8, 724–730.

The R5β1 integrin expression is significantly elevated in vivo on tumor vasculature in spontaneous as well as experimentally induced human and murine tumors compared to normal tissues.8 There is strong evidence that R5β1 and its ligand fibronectin are up-regulated on blood vessels in human tumor biopsies and, moreover, that antibody and peptide antagonists of R5β1 integrin, such as peptides that mimic the cell adhesion domain of fibronectin and contain the Arg-Gly-Asp (RGD) sequence, are potent inhibitors of tumor-induced angiogenesis, tumor growth, and tumor metastasis.8,12,14-19 The R5β1 integrin is therefore a (4) Davison, E.; Diaz, R. M.; Hart, I. R.; Santis, G.; Marshall, J. F. J. Virol. 1997, 71, 6204–6207. (5) Matter, M. L.; Zhang, Z.; Nordstedt, C.; Ruoslahti, E. J. Cell Biol. 1998, 141, 1019–1030. (6) Livant, D. L.; Brabec, R. K.; Kurachi, K.; Allen, D. L.; Wu, Y.; Haaseth, R.; rews, P.; Ethier, S. P.; Markwart, S. J. Clin. InVest. 2000, 105, 1537–1545. (7) van Golen, K. L.; Bao, L. W.; Brewer, G. J.; Pienta, K. L.; Kamradt, J. M.; Livant, D. L.; Merajver, S. D. Neoplasia 2002, 4, 373–379. (8) Kim, S.; Bell, K.; Mousa, S. A.; Varner, J. Am. J. Path. 2000, 156, 1345– 1362. (9) Gong, J.; Wang, D.; Sun, L.; Zborowska, E.; Willson, J. K.; Brattain, M. G. Cell Growth Differ. 1997, 8, 83–90. (10) Jayne, D. G.; Heath, R. M.; Dewhurst, O.; Scott, N.; Guillou, P. J. Eur. J. Surg. Oncol. 2002, 28, 30–36. (11) Ellis, L. M. Am. Surg. 2003, 69, 3–10. (12) Jia, Y. F.; Zeng, Z.-Z.; Markwart, S. M.; Rockwood, K. F.; Ignatoski, K. M. W.; Ethier, S. P.; Livant, D. L. Cancer Res. 2004, 64, 8674–8681. (13) Chen, J. H.; De, S.; Brainard, J.; Byzova, T. V. Cell Commun. Adhes. 2004, 11, 1–11. (14) Livant, D. L.; Brabec, R. K.; Pienta, K. J.; Allen, D. L.; Kurachi, K.; Markwart, S.; Upadhyaya, A. Cancer Res. 2000, 60, 309–320. (15) White, E. S.; Livant, D. L.; Markwart, S.; Arenberg, D. A. J. Immunol. 2001, 167, 5362–5366. (16) Stoeltzing, O.; Liu, W. B.; Reinmuth, N.; Fan, F.; Parry, G. C.; Parikh, A. A.; McCarthy, M. F.; Corazon, D. B.; Mazar, A. P.; Ellis, L. M. Int. J. Cancer 2003, 104, 496–503. (17) Meerovitch, K.; Bergeron, F.; Leblond, L.; Grouix, B.; Poirier, C.; Bubenik, M.; Chan, L.; Gourdeau, H.; Bowlin, T.; Attardo, G. Vasc. Pharmacol. 2003, 40, 77–89. (18) Yokoyama, Y.; Ramakrishnan, S. Int. J. Cancer 2004, 111, 839–848.

10.1021/la801961r CCC: $40.75  2008 American Chemical Society Published on Web 10/28/2008

PR_b-Targeted PEGylated Liposomes

promising target for breast, colon, prostate, and rectal cancer, since amounts of this integrin and its ligand are significantly increased on tumor vessels and in some cases also expressed on tumor cells..8-14 Functionalized liposomes for targeted cancer therapy, and in particular RGD-functionalized liposomes encapsulating fluorophores, doxorubicin, 5-fluorouracil, or dodecahydrododecaborate, have been developed for selective targeting of angiogenic endothelial cells or tumor cells overexpressing different integrins.20-28 The therapeutic use of RGD peptides though has been limited for targeting the R5β1 integrin, since they lack synergistic effects that come from the Pro-His-Ser-Arg-Asn (PHSRN) site.29-32 Efforts in the past to construct a single peptide sequence incorporating both RGD and PHSRN domains have also met with limited success.33-37 Here, we focus on peptide-amphiphiles as ligands for the integrin receptor, as they can assemble with other lipids and poly(ethylene glycol) (PEG), and represent a promising way of functionalizing interfaces and incorporating cell binding ability and recognition into biomaterials.38-39 A peptide-amphiphile, PR_b, previously designed in our laboratory, employed a hydrophobic tail (C16 dialkyl ester tail with a glutamic acid (Glu) tail connector and a -(CH2)2- tail spacer, (C16)2-Glu-C2-) connected to the N-terminus of a peptide headgroup that was composed of a spacer (KSS), the synergy site sequence (PHSRN), a linker ((SG)5) mimicking both the distance and hydrophobicity/ hydrophilicity present in the native protein fibronectin (thus presenting an overall “neutral” linker), and finally the primary binding sequence (RGDSP).40 Our original hypothesis was that the degree of hydrophobicity/hydrophilicity between the two sequences (RGD and PHSRN) in fibronectin is an important parameter in designing a fibronectin-mimetic peptide.41 Previous (19) Shannon, K. E.; Keene, J. L.; Settle, S. L.; Duffin, T. D.; Nickols, M. A.; Westlin, M.; Schroeter, S.; Ruminski, P. G.; Griggs, D. W. Clin. Exp. Metastasis 2004, 21, 129–138. (20) Oku, N.; Koike, C.; Tokudome, Y.; Okada, S.; Nishikawa, N.; Tsukada, H.; Kiso, M.; Hasegawa, A.; Fujii, H.; Murata, J.; Saiki, I. AdV. Drug DeliVery ReV. 1997, 24, 215–223. (21) Kurohane, K.; Namba, Y.; Oku, N. Life Sci. 2000, 68, 273–281. (22) Schiffelers, R. M.; Molema, G.; ten Hagen, T. L. M.; Janssen, A. P. C. A.; Schraa, A. J.; Kok, R. J.; Koning, G. A.; Storm, G. J. Liposome Res. 2002, 12, 129–135. (23) Schiffelers, R. M.; Koning, G. A.; ten Hagen, T. L. M.; Fens, M. H. A. M.; Schraa, A. J.; Janssen, A. N. P. C. A.; Kok, R. J.; Molema, G.; Storm, G. J. Controlled Release 2003, 91, 115–122. (24) Koning, G. A.; Fretz, M. M.; Woroniecka, U.; Storm, G.; Krijger, G. C. Appl. Radiat. Isot. 2004, 61, 963–967. (25) Ho¨lig, P.; Bach, M.; Vo¨lkel, T.; Nahde, T.; Hoffmann, S.; Mu¨ller, R.; Kontermann, R. E. Protein Eng., Des. Sel. 2004, 17, 433–441. (26) Dubey, P. K.; Mishra, V.; Jain, S.; Mahor, S.; Vyas, S. P. J. Drug Targeting 2004, 12, 257–264. (27) Xiong, X.-B.; Huang, Y.; Lu, W.-L.; Zhang, X.; Zhang, H.; Nagai, T.; Zhang, Q. J. Pharm. Sci. 2005, 94, 1782–1793. (28) Sofou, S. Nanomedicine 2007, 2(5), 711–724. (29) Pierschabacher, M. D.; Hayman, E. G.; Ruoslahti, E. Proc. Natl. Acad. Sci. U.S.A. 1983, 80, 1224–1227. (30) Yang, X. B.; Roach, H. I.; Clarke, N. M.; Howdle, S. M.; Quirk, R.; Shakesheff, K. M.; Oreffo, R. O. Bone 2001, 29, 523–531. (31) Akiyama, S. K.; Aota, S.; Yamada, K. M. Cell Adhes. Commun. 1995, 3, 13–25. (32) Ochsenhirt, S. E.; Kokkoli, E.; McCarthy, J. B.; Tirrell, M. Biomaterials 2006, 27(20), 3863–3874. (33) Aucoin, L.; Griffith, C. M.; Pleizier, G.; Deslandes, Y.; Sheardown, H. J. Biomater. Sci., Polym. Ed. 2002, 13, 447–462. (34) Benoit, D. S. W.; Anseth, K. S. Biomaterials 2005, 26, 5209–5220. (35) Kao, W. J. Biomaterials 1999, 20(23-24), 2213–2221. (36) Suzuki, Y.; Hojo, K.; Okazaki, I.; Kamada, H.; Sazaki, M.; Maeda, M.; Nomizu, M.; Yamamoto, Y.; Nakagawa, S.; Mayumi, T.; Kawasaki, K. Chem. Pharm. Bull. 2002, 50, 1229–1232. (37) Kim, T.-I.; Jang, J.-H.; Lee, Y.-M.; Ryu, I.-C.; Chung, C.-P.; Han, S.-B.; Choi, S.-M.; Ku, Y. Biotechnol. Lett. 2002, 24, 2029–2033. (38) Mardilovich, A.; Kokkoli, E. Langmuir 2005, 21, 7468–7475. (39) Kokkoli, E.; Mardilovich, A.; Wedekind, A.; Rexeisen, E. L.; Garg, A.; Craig, J. A. Soft Matter 2006, 2(12), 1015–1024. (40) Mardilovich, A.; Craig, J. A.; McCammon, M. Q.; Garg, A.; Kokkoli, E. Langmuir 2006, 22, 3259–3264.

Langmuir, Vol. 24, No. 23, 2008 13519

work in our laboratory demonstrated that PR_b is a promising sequence compared to fibronectin, and for the first time in the literature a surface functionalized with a peptide (PR_b) was shown to outperform a fibronectin-functionalized surface.40 We established specificity of the PR_b peptide for the integrin R5β1 by blocking cell adhesion, and eliminating liposomal binding to cancer cells, with antibodies and free peptides.40,42 Recently, we tested our original hypothesis by comparing PR_b to other peptides with “hydrophobic” or “hydrophilic” linkers. Cell adhesion studies validated our hypothesis and demonstrated that a “neutral” linker, that more closely mimics the cell adhesion domain of fibronectin, supports higher levels of adhesion compared to other peptide designs with a “hydrophobic” or “hydrophilic” linker and fibronectin.43 In this study, we incorporated the PR_b peptide-amphiphile into PEGylated or stealth liposomes (liposomes covered with PEG) with the goal of targeting the integrin R5β1 that is expressed on prostate cancer cells. The peptide-functionalized nanoparticles were characterized with NMR and in vitro studies with LNCaP prostate cancer cells. Our work shows that by varying the amount of PEG and PR_b on the liposomal interface we can engineer nanovectors that bind to human prostate cancer cells, that express the R5β1 integrin, and subsequently internalize. GRGDSP-targeted stealth liposomes bind to prostate cancer cells and internalize as well but to a lesser extent. We have further shown that the PR_b PEGylated liposomes upon internalization are capable of releasing their therapeutic payload (TNFR) intracellularly. Tumor necrosis factor-R (TNFR) is a cytokine involved in systemic inflammation and causes apoptotic cell death. Apoptosis is the carefully regulated process of cell death. LNCaP is a cancer cell line that is sensitive to the cytotoxic effects of TNFR.44,45 PR_b-targeted stealth liposomes are more cytotoxic on LNCaP cells compared to GRGDSP-targeted stealth liposomes and nontargeted stealth liposomes.

Materials and Methods Materials. Lipids, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), cholesterol, and 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-(methoxy(polyethylene glycol)-2000)-(ammonium salt) (PEG2000) were purchased from Avanti Polar Lipids Inc. (Alabaster, AL). The extruder and 100 nm polycarbonate membranes for liposome extrusion were purchased from Avestin Inc. (Ottawa, Canada). 1,2,3-Trichlorobenzene was purchased from Supelco (Bellefonte, PA). The peptide headgroups PR_b (KSSPHSRN(SG)5RGDSP) and GRGDSP were purchased in crude form from the Microchemical Facility at the University of Minnesota. The PR_b peptide-amphiphile ((C16)2-Glu-C2-KSSPHSRN(SG)5RGDSP) and GRGDSP peptide-amphiphile ((C16)2-GluC2-GRGDSP) were synthesized as described previously.41 LNCaP cells (human prostate cancer cell line) and alpha-dihydrotestosterone (DHT) were a gift from Prof. Bischof at the University of Minnesota. Hoechst 33342 nucleic stain, Alexa Fluor 594 wheat germ aggutin (WGA) cell membrane stain, cell culture media (D-MEM/F-12), Vybrant apoptosis kit #9, and ProLong Gold antifade reagent were purchased from Invitrogen (Carlsbad, CA). Primary antibody antiintegrin R5β1 (AB1950) and secondary antibody anti-goat IgG (AP180F) were purchased from Milipore-Chemicon International Inc. (Temecula, CA). Goat IgG isotype control was purchased from (41) Mardilovich, A.; Kokkoli, E. Biomacromolecules 2004, 5, 950–957. (42) Garg, A.; Tisdale, A. W.; Haidari, E.; Kokkoli, E. Int. J. Pharm.,. published online Sept 19, http://dx.doi.org/10.1016/j.ijpharm.2008.09.016. (43) Craig, J. A.; Rexeisen, E. L.; Mardilovich, A.; Shroff, K.; Kokkoli, E. Langmuir 2008, 24(18), 10282–10292. (44) Sensibar, J. A.; Sutkowski, D. M.; Raffo, A.; Buttyan, R.; Griswold, M. D.; Sylvester, S. R.; Kozlowski, J. M.; Lee, C. Cancer Res. 1995, 55, 2431– 2437. (45) Sherwood, E.; Ford, T.; Lee, C.; Kozlowski, J. J. Biol. Response Modif. 1990, 9, 44–52.

13520 Langmuir, Vol. 24, No. 23, 2008 Sigma Aldrich Corporation (St. Louis, MO). Fetal bovine serum (FBS) was purchased from Atlas Biologicals (Fort Collins, CO), and human fibronectin-coated round coverslips were purchased from BD Biosciences (San Jose, CA). TNFR was purchased from R&D Systems (Minneapolis, MN). The BCA (bicinchoninic acid) protein assay kit was purchased from Pierce (Rockford, IL). All other reagents were purchased from Sigma Aldrich Corporation (St. Louis, MO) and were of biotechnology performance certified grade. Liposome Preparation and Characterization. Liposomes were prepared as described elsewhere.46 Briefly, lipids were dissolved in chloroform, and peptide-amphiphiles were dissolved in methanol and water. Lipids (DPPC), cholesterol, PEG2000, and peptideamphiphiles were combined at concentrations of x mol % PEG2000, y mol % peptide-amphiphile, approximately 35 mol % cholesterol, and (65-x-y) mol % DPPC. Considering that in designing a targeted drug delivery system the focus in this work is at the interface, in the majority of the text only the mol % of the peptide-amphiphile and PEG2000 are mentioned. Solvents were removed by evaporating under a gentle stream of argon at 65 °C, and lipids were dissolved again in chloroform to form a homogeneous mixture. The lipid mixture was finally dried under a gentle stream of argon at 65 °C until a uniform lipid film was formed, followed by drying under vacuum overnight. Fluorescently labeled liposomes were prepared by hydrating the lipid film with HBSE buffer (10 mM Hepes, 150 mM NaCl, 0.1 mM EDTA, pH 7.4) containing 2 mM calcein at 65 °C and at a concentration of 10 mM total lipids. Tumor necrosis factor-R (TNFR) encapsulated liposomes were prepared similarly by hydrating the lipid film with HBSE buffer containing 1 µg/µL TNFR in DI water at 40 °C. Hydrated lipids were freeze-thawed five times and then extruded for 21 cycles through two stacks of 100 nm polycarbonate membranes using the hand-held extruder. Liposomes were filtered over a Sepharose CL-4B gel filtration column to remove unencapsulated fluorescent dye or TNFR and other molecules that do not get incorporated in the liposomes. Liposomes were stored at 4-8 °C and were used within 2 weeks. The liposome diameter was determined by dynamic light scattering and ranged from 80-150 nm. The peptide concentration for the purified liposome formulations was determined using the BCA assay according to the manufacturer’s protocol, and the reported values were usually slightly different from the ones initially mixed in solution.42 Similarly, since PEG concentrations in the purified liposomes may differ slightly from the concentrations mixed in solution during preparation, it is therefore important to accurately characterize both the peptide and PEG concentrations on the liposomal formulations. Recently, we evaluated different PEG assays such as complexation with barium chloride and iodine, the ferrothiocyanate method, the picric acid method, and the Bradford assay, but we were not able to characterize the PEG concentration in the liposomes due to experimental limitations.42 PEG and PEO (poly(ethylene oxide)) quantification with 1H NMR studies has been reported in the literature.47,48 In the previous studies, PEG was quantified by the standard addition method, where known aliquots of PEG were added into the sample, which gave rise to a linear increase in the signal, and the slope was used to determine the initial mass of PEG in the sample.48 In our study, an internal standard, 1,2,3-trichlorobenzene, was used for direct quantification. We used 1H NMR spectroscopy for the quantification of PEG2000 concentration, as well as the concentration of other components such as DPPC, and cholesterol for all the liposomes used in this study in the presence or absence of PR_b or TNFR depending on the formulation. Quantitative analysis was done by integrating peaks characteristic for each component in the 1H NMR spectra, collected with a Varian Inova 800 magnet. An amount of 150 µL of different liposomal samples was lyophilized and redissolved in 1 mL of d-chloroform (PR_b and TNFR were not soluble in chloroform, so they formed a solid at the bottom of the NMR tube along with other (46) Fenske, D. B.; Maurer, N.; Cullis, P. R. In Liposomes, 2nd ed.; Torchilin, V. P., Weissig, V., Eds.; Oxford University Press: Oxford, 2003; pp 167-191. (47) Csaba, N.; Gonzalez, L.; Sanchez, A.; Alonso, M. J. J. Biomater. Sci., Polym. Ed. 2004, 15(9), 1137–1151. (48) Duncanson, W. J.; Figa, M. A.; Hallock, K.; Zalipsky, S.; Hamilton, J. A.; Wong, J. Y. Biomaterials 2007, 28(33), 4991–4999.

Demirgo¨z et al. undissolved compounds from the buffer), and 30 µL of the standard was spiked into each one of the sample NMR tubes. The results from the NMR study were validated by comparing the NMR lipid concentration for every liposome formulation used in this study with the lipid concentration calculated from the phosphorus colorimetric assay.49,50 Cell Culture. LNCaP cells were grown in LNCaP media (modified D-MEM/F-12 media, supplemented with 10% FBS, 1% penicillin/ streptomycin, 10-9 M DHT, and 1.2 g of sodium bicarbonate). Cells were grown in T-75 flasks with a feeding cycle of 2 days. After cells became 80% confluent (usually after 1 week), they were washed with HBSS buffer, trypsinized (0.25% Trypsin + 0.1% EDTA), and suspended in LNCaP media. For subsequent passages, cells were seeded in fresh T-75 flasks at a density 40 000 cells/cm2 and were cultured in LNCaP media. Flow Cytometry. LNCaP confluent cell monolayers were trypsinized (0.25% trypsin + 0.1% EDTA) and resuspended in growth media containing liposomes at a lipid concentration of 250 µM and a cell concentration of 1 × 106/mL in 15 mL centrifuge tubes. Tubes were incubated at 37 °C over a rotary shaker for the specified duration of time. Cells were then pelleted and washed twice in ice-cold buffer (phosphate buffered saline (PBS) supplemented with 0.02% sodium azide and 2.5% fetal bovine serum). Flow cytometric analysis was carried out immediately. For integrin R5β1 expression studies, LNCaP confluent cell monolayers were trypsinized (0.25% trypsin + 0.1% EDTA) and resuspended in ice-cold buffer at a cell concentration of 1 × 106/mL in 15 mL centrifuge tubes. Tubes were incubated at 4 °C with primary antibody (anti-integrin R5β1) or goat isotype control (goat IgG) over a rotary shaker for 35 min. Cells were pelleted and washed twice in ice-cold buffer and then incubated again with the secondary antibody (anti-goat IgG FITC conjugated) for 35 min. Finally, cells were pelleted and washed twice in ice-cold buffer. Flow cytometric analysis was carried out immediately. The FACS Calibur instrument located at the Flow Cytometry Core facility in the Cancer Research Center of the University of Minnesota was used. All experiments were repeated twice, but results are presented from a single experiment. Confocal Microscopy. LNCaP confluent cell monolayers grown on fibronectin coverslips were incubated with liposomes at a lipid concentration of 250 µM in a 5% CO2 incubator at 37 °C for the specified duration in LNCaP media. Cell monolayers were then washed with ice-cold buffer twice. Cells were later fixed with a fixation buffer (4% paraformaldhyde in PBS, pH ∼ 7.4) for 15 min at 37 °C. Nuclear staining was carried out using a cell membrane permeable blue-fluorescent Hoechst 33342 dye at a concentration of 2.0 µmol/mL, and the cell membrane was stained with a cell impermeable red-fluorescent Alexa Fluor 594 wheat germ aggutin (WGA) at 5.0 µg/mL in ice-cold buffer for 10 min. Cells were washed three times with ice-cold buffer, and coverslips were mounted on glass slides over ProLong Gold antifade reagent. For every sample, 15 z-scans (horizontal cross section of a cell at a particular z height) were taken at 0.25 µm z-step height to cover the entire height of the cell. On the confocal images, liposomes were labeled with green, cell membrane with red, and nucleus with blue. The Olympus Fluoview 1000 confocal laser scanning microscope at the Biomedical Image Processing Laboratory in the Department of Neuroscience at the University of Minnesota was used. Apoptosis Assay. The amount of encapsulated TNFR in different liposomal formulations was evaluated with the Quantikine Human TNF-R/TNFSF1A immunoassay kit (R&D Systems, Minneapolis, MN). The TNFR-encapsulated liposomes were lysed with 90% methanol/10% DI water prior to the experiment, and the assay was performed according to the manufacturer’s protocol. LNCaP confluent cell monolayers grown on fibronectin coverslips were incubated with equal volumes of TNFR-encapsulated liposomal formulations in LNCaP media, at a lipid concentration of 250 µM, for 3 and 24 h at 37 °C. Cells were washed and incubated with the Vybrant apoptosis assay kit #9 (with the allophycocyanin conjugate of annexin V that (49) Chen, P. S.; Toribara, T. Y.; Warner, H. Anal. Chem. 1956, 28, 1756– 1758. (50) Fiske, C. H.; Subbarow, Y. J. Biol. Chem. 1925, 66, 375–400.

PR_b-Targeted PEGylated Liposomes

Langmuir, Vol. 24, No. 23, 2008 13521

Figure 1. 1H NMR spectra of PEG2000, DPPC, cholesterol (CHOL), and a DPPC/CHOL/PEG2000 liposome in d-chloroform. The arrows show the protons (or the carbons attached to the protons) that were used for the identification of each compound. The shaded areas show the peaks for those protons, and highlight the peaks that are characteristic for each species and absent from the spectra of other compounds. Table 1. Pure Component Quantification through 1H NMR and Phosphorus Assay

component

added in solution (µmol)

measured by 1 H NMR (µmol)

PEG2000 cholesterol DPPC

2.5 2.9 3.3

3.0 3.4 3.5

Figure 2. Integrin expression on LNCaP cells evaluated via flow cytometry. Cells were incubated with antibodies to integrin R5β1 and Rvβ3. Appropriate isotype control is included. The gray area is the autofluorescence of the cells.

measured by phosphorus assay (µmol) 2.4 3.6

stains early apoptotic cells and the SYTOX Green dye that stains late apoptotic (dead) cells) according to the manufacturer’s protocol. At the end of the staining protocol, cells were fixed with a fixation buffer (4% paraformaldhyde in PBS, pH ∼ 7.4) for 15 min at 37 °C, and coverslips were mounted on glass slides over ProLong Gold antifade reagent. The fluorescence intensity of each dye was measured with an inverted microscope equipped with a PixCell II laser capture microdissection (LCM) system at the Biomedical Image Processing Laboratory at the University of Minnesota.

Results and Discussion In this study, NMR spectroscopy was used to accurately quantify the concentration of PEG2000 in the liposomes as well as the concentration of other individual components such as cholesterol and DPPC. NMR quantification was done by integrating peaks characteristic for each component in the 1H NMR spectra. The chemical shifts of the peaks used for integration of DPPC, cholesterol, and PEG2000 were at 3.7, 5.32, and 3.63 ppm, respectively (Figure 1). First, known amounts of pure molecules (DPPC, PEG2000, and cholesterol) were dissolved in d-chloroform, and their 1H NMR spectra were collected. The results from the 1H NMR study were validated by comparing the 1H NMR lipid concentration with the results from the phosphorus colorimetric assay. As shown in Table 1, the results obtained from both 1H NMR and the phosphorus assay for the pure molecules (i.e., not in a liposome) are in close agreement with the amount that was added originally in the solution. For the liposomal characterization, the phosphorus assay allows for the quantification of the total lipid concentration within the bilayer (DPPC and PEG2000 combined), whereas NMR permits for the accurate characterization of individual components such as DPPC, PEG2000, and cholesterol. For example, different liposomes that were prepared by adding 35 mol % cholesterol in solution were shown by 1H NMR to have 38.5-41 mol % cholesterol (always higher percentage than the initial solution concentration). The PEG2000 concentration also varied between initial concentration

Figure 3. Liposomes functionalized with different concentrations (mol %) of PR_b were incubated with the LNCaP cells for 3 h at 37 °C. Liposomal binding increases with increased PR_b concentration.

in solution and final concentration incorporated in the liposome. For example, formulations shown in Figure 5 had 1.9-2.3 mol % PEG2000 which varied slightly from the 2.0 mol % PEG2000 initial solution concentration used for preparation of these formulations. Integrin expression was characterized next. Figure 2 shows the histogram for the expression of R5β1 on LNCaP cells. Rvβ3 (an integrin that also binds to fibronectin) expression was also characterized for comparison, as well as isotype control binding, and both were found to be minimal. This result confirms that integrin R5β1 is present on human prostate cancer cells and that R5β1 can serve as a target for the delivery of liposomes. To test the effect of PR_b on LNCaP binding, different liposomes were initially prepared with increasing amounts of PR_b and without PEG2000 and incubated with cells for 3 h at 37 °C. Figure 3 shows the effect of peptide concentration on liposome binding to integrin R5β1-expressing LNCaP cells. Liposomes without PR_b showed minimal binding to cells. However, small concentrations of PR_b (2.2 mol %) gave sufficient binding to the LNCaP cells, and cell adhesion increased

13522 Langmuir, Vol. 24, No. 23, 2008

Demirgo¨z et al.

Figure 4. Binding of PR_b-targeted stealth liposomes to LNCaP cells. Different liposomal formulations (concentrations shown in mol %) with just PEG2000 or PEG2000 and PR_b were incubated with the LNCaP cells at 37 °C for 3 h. Cell binding was increased for PEGylated liposomes functionalized with PR_b.

Figure 6. Intracellular uptake of different liposomes to LNCaP cells characterized with confocal microscopy. First row shows liposomes functionalized with 2 mol % PEG2000, second row shows 6.1 mol % GRGDSP and 2.1 mol % PEG2000, and third row shows formulations functionalized with 6.5 mol % PR_b and 2.2 mol % PEG2000. Liposomes (shown with green) were incubated with LNCaP at 37 °C for 3 and 24 h. Fifteen scans were taken (0.25 µm step) across the body of the cell. Images shown are approximately 1-1.25 µm above the coverslip and 2 µm below the surface of the cells, and they were merged with the nucleus (shown in blue) and cell membrane (shown in red). The scale bar is 50 µm for all images. Internalization of liposomes into endosomes can be identified by the green liposome signal in between the red cell membrane and blue nucleus. These images illustrate that PR_b-targeted stealth liposomes can be internalized by the LNCaP cells after binding to the integrin R5β1. GRGDSP-targeted stealth liposomes also show internalization but to a lesser extent compared to the PR_b-targeted liposomes. Figure 5. Binding of liposomes functionalized with PEG2000 (mol %) and either PR_b or GRGDSP (mol %). LNCaP cells were incubated with different liposomal formulations for 3 h at 37 °C. Results show that PEGylated liposomes functionalized with PR_b demonstrate increased binding compared to the GRGDSP-functionalized stealth liposomes.

further with increased peptide functionalization of the liposomes. PR_b concentrations greater than 6.4 mol % were not examined. The next step was the incorporation of PEG2000, since balancing the concentrations of the peptide and PEG is critical for the efficient targeted delivery of liposomes to prostate cancer cells. It is important to mention that both the PEG2000 and peptide concentrations reported here were accurately determined via 1H NMR and the BCA assay, respectively, for the purified liposomes, and show the actual amounts incorporated in the formulations. As discussed earlier, the reported values for both the PEG and peptide were slightly different from the ones initially mixed in solution during preparation. PEG2000 and PR_b were studied at low (2-2.3 mol % for PEG2000, and 1-2.5 mol % for PR_b) and high (5.8-6 mol % PEG2000 and 5.3 mol % PR_b) concentrations. Figure 4 shows flow cytometry results for PEGylated liposomes functionalized with PR_b and targeted to LNCaPs for 3 h at 37 °C. PEGylated

liposomes with low or high PEG2000 concentrations and without peptide showed no binding. Small concentrations of PR_b (1, 1.3 mol %) had no significant effect on binding in the presence of low (2.3 mol %) or high (5.9 mol %) PEG2000. We speculate that about 1 mol % PR_b functionalized PEGylated liposomes lack the required surface density of targeting peptides needed for significant improvements in targeted delivery. However, increased liposomal binding was observed for higher amounts of peptide (2.5 mol % and 5.3 mol %) in the presence of low or high PEG2000. As a result, functionalization of PEGylated liposomes with the PR_b peptide, designed to specifically target the integrin R5β1, at a level of 2 mol % or higher is desired to help achieve increased binding efficiencies. PR_b-functionalized stealth liposomes were compared to GRGDSP-functionalized stealth liposomes. Figure 5 shows that PR_b-functionalized stealth liposomes outperform the GRGDSPfunctionalized stealth liposomes, since PR_b liposomes gave higher levels of binding compared to the GRGDSP formulations. For example, liposomes with 1 mol % PR_b/2.3 mol % PEG2000 gave the same adhesion as a system that has the same PEG2000 concentration (2.1 mol %) and 6 times more GRGDSP.

PR_b-Targeted PEGylated Liposomes

Langmuir, Vol. 24, No. 23, 2008 13523

Figure 7. Results for LNCaP cells that were early apoptotic or late apoptotic (dead) after incubation with equal volumes of different TNFRencapsulated liposomes for 3 and 24 h at 37 °C. The amount of TNFR encapsulated is shown for each formulation. The signal from early and late apoptotic cells is stacked to give an overall picture of cytotoxicity.

Internalization of PEGylated liposomes and PEGylated liposomes functionalized with GRGDSP or PR_b was investigated with a confocal laser scanning microscope. Figure 6 shows confocal images for a horizontal cell section about 1 µm above the coverslip and 2 µm below the cell surface. Internalization is identified by localization of green fluorescent dots (liposomes) between the blue nuclear region and the red cell membrane. After internalization, nanoparticles are initially confined inside endosomes, which are submicrometric vesicles of the endocytotic pathway.51-53 Therefore, the green dots in Figure 6 show liposomes that have internalized into endosomal compartments. Figure 6 demonstrates that at the times examined there is no internalization of stealth liposomes with 2 mol % PEG2000. GRGDSP stealth liposomes show evidence of internalization at 37 °C after 3 h, and at 24 h internalization is greater. One possible explanation for this is that 24 h of incubation allows more time for the recycling of integrins and thereby increasing the amount of internalized liposomes. Similar trends were observed for the PR_b-targeted stealth liposomes (with similar peptide and PEG2000 concentrations), except there are significantly higher levels of internalization. At 3 h of incubation, the amount of green fluorescence inside the red cell membrane is higher, and at 24 h this effect is even more pronounced. Thus, Figure 6 demonstrates that when PEGylated liposomes are functionalized with the PR_b peptide, more liposomes are internalized compared to GRGDSP formulations. These results also support the binding studies from the flow cytometry experiments (Figure 5), and show that PR_b targeting can significantly improve the performance of stealth liposomes as compared to conventional RGD targeting techniques. (51) Colin, M.; Maurice, M.; Trugnan, G.; Kornprobst, M.; Harbottle, R. P.; Knight, A.; Cooper, R. G.; Miller, A. D.; Capeau, J.; Coutelle, C.; Brahimi-Horn, M. C. Gene Ther. 2000, 7(2), 139–152. (52) Bertorelle, F.; Wilhelm, C.; Roger, J.; Gazeau, F.; Menager, C.; Cabuil, V. Langmuir 2006, 22(12), 5385–5391. (53) Richardson, S. C. W.; Wallom, K.-L.; Ferguson, E. L.; Deacon, S. P. E.; Davies, M. W.; Powell, A. J.; Piper, R. C.; Duncan, R. J. Controlled Release 2008, 127, 1–11.

In this study, we have also examined the cytotoxicity of different formulations that encapsulated TNFR. Figure 7 clearly shows that, compared to GRGDSP-targeted stealth liposomes or to nontargeted stealth liposomes, PR_b-targeted stealth liposomes show improved cytotoxicity as shown by the total higher fluorescent intensity for LNCaP cells at the early and late stages of apoptosis. Recent studies reported that RGD-targeted stealth liposomes show increased cytotoxicity over nontargeted liposomes. For example, RGD-targeted liposomes containing the anticancer drug doxorubicin showed a 30-40% increase in efficacy over nontargeted stealth liposomes delivered to murine B16 and human A375 melanoma cells after incubation for 8 h.27,54 Our studies, however, showed that GRGDSP-functionalized stealth liposomes did not show any clear advantage over the nontargeted TNFRencapsulated formulations, since, for example, after 24 h of incubation, stealth liposomes with 2.8 mol % PEG2000 and GRGDSP-functionalized stealth liposomes with 2.2 mol % GRGDSP/2.6 mol % PEG2000 or 3.2 mol % GRGDSP/2.7 mol % PEG 2000 all gave similar levels of fluorescent intensity for apoptotic LNCaP cells. On the other hand, TNFR-encapsulated liposomes with 1.8 mol % PR_b/2.6 mol % PEG 2000 and 3.8 mol % PR_b/2.9 mol % PEG 2000 gave a 37% and 72% increase in efficacy, respectively (as measured by the total fluorescence for early and late apoptotic cells), over the nontargeted stealth liposomes.

Conclusions Liposomes with varying PEG2000 and peptide concentrations were targeted to LNCaP human prostate cancer cells that express the integrin R5β1. The nanoparticles were first characterized with 1H NMR for the evaluation of PEG2000 and other components in the liposomes. Conventional liposomes (without PEG2000 and PR_b) and stealth liposomes (54) Xiong, X. B.; Huang, Y.; Lu, W. L.; Zhang, X.; Zhang, H.; Nagai, T.; Zhang, Q. J. Controlled Release 2005, 107, 262–275.

13524 Langmuir, Vol. 24, No. 23, 2008

without any peptide showed minimal binding to LNCaP cells. On the other hand, liposomes functionalized with only PR_b showed cell adhesion that was peptide concentration dependent. When both PEG2000 and PR_b were present at the interface, cell binding was observed for peptide concentrations of 2 mol % and higher. PR_b targeting of PEGylated liposomes is superior to that of GRGDSP stealth liposomes, as shown by improved cell binding, internalization, and cytotoxicity associated with the delivery of encapsulated TNFR. More importantly, PR_b is R5β1-specific,40,42 whereas many integrins bind to simple RGD peptides. This work demonstrated that PR_b targeting can significantly improve the performance of PEGylated liposomes as compared to nontargeted formulations or conventional RGD targeting techniques.

Demirgo¨z et al.

Acknowledgment. This work was supported in part by the MRSEC Program of the National Science Foundation under Award No. DMR-0212302, the Center for Nanostructured Applications at the University of Minnesota, the National Science Foundation (CBET-0553682), and the National Institute of Biomedical Imaging and Bioengineering (R03EB006125). We would like to acknowledge the assistance of the Flow Cytometry Core Facility at the University of Minnesota Cancer Center, a comprehensive cancer center designated by the National Cancer Institute, supported in part by P30 CA77598. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Biomedical Imaging and Bioengineering or the National Institutes of Health. LA801961R