Liposomal formulation of polyacrylate-peptide conjugate as a new vaccine candidate against cervical cancer
Mattaka Khongkowa,b, Tzu-Yu Liuc, Stacey Bartlettb, Waleed M. Husseinb,d, Reshma Nevagib, Zhongfan Jiae, Michael J. Monteiroe, James Wellsc, Uracha Rungsardthong Ruktanonchaia, Mariusz Skwarczynskib, Istvan Tothb,f,g,
aNational Nanotechnology Center (NANOTEC), National Science and Technology Development Agency, 111 Thailand Science Park, Phahonyothin Rd., Khlong Luang, Pathumthani 12120, Thailand.
bSchool of Chemistry and Molecular Biosciences, The University of Queensland, St Lucia, Brisbane, QLD 4072, Australia
cThe University of Queensland Diamantina Institute, The University of Queensland, Translational Research Institute, Brisbane, Australia.
dHelwan University, Pharmaceutical Organic Chemistry Department, Faculty of Pharmacy, Ein Helwan, Helwan, Egypt
eAustralian Institute for Bioengineering and Nanotechnology, The University of Queensland, Brisbane, QLD 4072, Australia
fSchool of Pharmacy, The University of Queensland, Brisbane, QLD 4072, Australia
gInstitute for Molecular Bioscience, The University of Queensland, Brisbane, QLD 4072 Australia
Submitted: October 3, 2018; Accepted: October 25, 2018
We demonstrate that a polymer-based delivery system for peptide-based vaccines and liposomes can be incorporated together to greatly improve therapeutic efficacy of the anticancer vaccine.
Peptide-based vaccines have been proposed as a therapeutic strategy for many infectious diseases, including human papilloma virus (HPV)-related cervical cancer. Peptide-based vaccines are a better treatment option than traditional chemotherapeutic agents and surgery, as they rely on the use of the body’s immune system to fight cancer cells, resulting in minimal risk of side effects. However, to increase the efficacy of peptide-based vaccines, the application of potent adjuvant and a suitable delivery system is essential. In this study, we developed a self-adjuvating delivery system based on a combination of polymer and liposomes, for a therapeutic vaccine against cervical cancer. Peptide epitope (8Qm) derived from HPV-16 E7 protein was conjugated to dendritic poly(tert-butyl acrylate) as a primary delivery system and incorporated into cationic liposomes, which served as a secondary delivery system. Our vaccine candidate was able to kill established HPV-16 E7-positive tumor (TC-1) cells in mice following a single immunization. The immunized mice had 80% survival rate after two months. In contrast, both polymer-8Qm conjugate and liposomes bearing 8Qm failed to eradicate TC-1 tumors. The survival rate of mice was only 20% when immunized with 8Qm formulated with standard incomplete Freund’s adjuvant.
- peptide-based vaccine
- polymer-peptide conjugate
- anticancer vaccine
- human papilloma virus
- CTL: cytotoxic T lymphocyte (CTL)
- CuAAC: copper-catalyzed alkyne-azide cycloaddition
- DIPEA: N,N’-diisopropylethylamine
- DDAB: didodecyldimethyl-ammonium bromide
- DPPC: dipalmitoylphosphatidylcholine
- HATU: (dimethylamino)-N,N-dimethyl(3H-[1,2,3]triazolo[4,5-b]pyridin-3-yloxy)-methanim-inium hexafluorophosphate
- HPV: human papilloma virus
- PDI: polydispersity index
Purpose and Rationale
The purpose of this study was to improve potency of the polymer-peptide conjugate vaccine against HPV associated cancers. The combination of the conjugate with liposomes was expected to not only reduce the size of the anticancer vaccine construct, from micrometers to nanometers range, but to also improve therapeutic efficacy of the vaccine.
Cervical cancer is caused by human papilloma virus (HPV) infection. Approximately half a million cases of cervical cancer are reported annually with the disease’s death rate reaching a quarter of a million worldwide. While treatments for cervical cancer are available, including chemotherapy and surgery, most treatments fail due to high disease recurrence and the development of drug resistance.1 Prophylactic vaccines were recently developed against HPV infections2 however, a large proportion of women worldwide are currently infected with the virus and are consequently still at risk of developing cervical cancer. Therefore, the development of a therapeutic vaccine that targets HPV-infected cells is in high demand.3
In contrast to standard vaccines, which induce humoral immune responses (mainly via antibody production), therapeutic anticancer vaccines are designed to elicit cytotoxic T lymphocyte (CTL) responses to eliminate cells bearing specific tumor antigens. As whole HPV or oncoprotein-based vaccines can induce oncogenic changes, a peptide-based approach has been proposed as a safe alternative.4 E7 HPV oncoprotein is unique to cells infected by HPV and is required to maintain HPV-associated tumor cell growth. Therefore, CD8+ peptide epitopes from E7 capable of activating CTLs are commonly used to develop vaccines against cervical cancer.5-9 In general, peptides alone are unable to stimulate the immune system because of their poor immunogenic properties; therefore, their integration with appropriate adjuvants and/or a delivery system is required.10 Unfortunately, many adjuvants are toxic, unsuitable for human use, or too weak to stimulate immune responses against weak peptide antigens.11
To overcome this problem, the authors developed a self-adjuvating polymer-based delivery system for prophylactic peptide-based vaccines.12-15 The system was then adapted for the development of a therapeutic vaccine against cervical cancer.16 A variety of branched and linear poly(tert-butyl acrylate) were conjugated to 8Qm (E744-57, QAEPDRAHYNIVTF) peptide derived from HPV-16 E7 oncoprotein (Fig. 1) bearing both CD4+ and CD8+ epitopes.17, 18 Of the tested hydrophobic polymers, dendritic polymer (D) conjugated to 8Qm via copper (wires) catalyzed by copper-catalyzed alkyne-azide cyclo-addition (CuAAC) eradicated 3-day-old E7-positive TC-1 tumors in mice after a single immunization. The survival rate for mice immunized with this vaccine reached 90%, which represented a significantly improvement over immunization via antigen mixed with classic incomplete Freund’s adjuvant (IFA), ISA51. It was also shown that 8Qm-D was promptly taken up by antigen presenting cells, and stimulated strong CD8+ cell responses, without activation of TLR2, which is usually responsible for the recognition of hydrophobic ligands.19, 20 In addition, depletion experiments with anti-CD4+, anti-CD8+, anti-NK antibodies demonstrated, that vaccine efficacy was strongly dependent on CD8.19 It was also confirmed that mice did not produce antibodies against 8Qm.17 However, the survival rate of mice immunized 7 days post tumor implantation dropped to 20-50% after 90 days, even when additional boosts were given.21 Therefore, while progressing efforts, this vaccine candidate still required further optimization to improve its efficacy against late stage cervical cancer; an issue we suggest can be addressed via improvement to the vaccine delivery system.
In this study, we developed a new vaccine strategy by integrating a polymer-based delivery system with cationic liposomes. We tested 8Qm-D anchored to liposomes (8Qm-D-L) as a therapeutic vaccine against a model of cervical cancer in mice. 8Qm-D-L greatly reduced the growth of 7-day-old E7-expressing TC-1 tumors and significantly improved mouse survival (4/5) compared to 8Qm adjuvanted with IFA (1/5).
Figure 1. Synthesis of 8Qm-D conjugate by copper-catalyzed alkyne-azide cycloaddition reaction.
Materials and methods
pMBHA resin was purchased from Peptides International (Kentucky, USA). Rink amide MBHA resin, N,N’-dimethylformamide (DMF) dichloromethane (DCM), methanol, N,N’-diisopropylethylamine (DIPEA), piperidine and trifluoroacetic acid were obtained from Merck (Hohenbrunn, Germany). (Dimethylamino)-N,N-dimethyl(3H-[1,2,3]tri-azolo[4,5-b]pyridin-3-yloxy)-methan-iminium hexafluorophosphate (HATU) was purchased from Mimotopes (Melbourne, Australia). Protected L-amino acids were purchased from Novabiochem (Läufelfingen, Switzerland) and Mimotopes (Melbourne, Australia). HPLC grade acetonitrile was obtained from Labscan (Bangkok, Thailand). Copper wires were purchased from Aldrich (Steinheim, Germany). All other reagents were obtained at the highest available purity from Sigma-Aldrich (Castle Hill, Australia). Anhydrous hydrofluoric acid (HF) was supplied by BOC gases (Sydney, Australia). A Kel-F HF apparatus (Peptide Institute, Osaka, Japan) was used for HF cleavage. Electrospray ionization mass spectrometry (ESI-MS) was performed using a Perkin-Elmer-Sciex API3000 instrument with Analyst 1.4 software (Applied Biosys-tems/MDS Sciex, Toronto, Canada). Analytical RP-HPLC was performed using Shimadzu (Kyoto, Japan) instrumentation (DGU-20A5, LC-20AB, SIL-20ACHT, SPD-M10AVP) with a 1 mL/min flow rate and detection at 214 nm and/or with an evaporative light scattering detector. Separation was achieved using a 0-100% linear gradient of solvent B over 40 minutes with 0.1% TFA/H2O as solvent A and 90% MeCN/0.1% TFA/H2O as solvent B on either a Vydac analytical C4 column (214TP54; 5 m, 4.6 mm x 250 mm) or a Vydac analytical C18 column (218TP54; 5 m, 4.6 mm x 250 mm). Preparative RP-HPLC was performed on Shimadzu (Kyoto, Japan) instrumentation (either LC-20AT, SIL-10A, CBM-20A, SPD-20AV, FRC-10A or LC-20AP x 2, CBM-20A, SPD-20A, FRC-10A) in linear gradient mode using a 5-20 mL/min flow rate, with detection at 230 nm. Separations were performed with solvent A and solvent B on a Vydac preparative C18 column (218TP1022; 10 m, 22 mm x 250 mm), Vydac semi-preparative C18 column (218TP510; 5 m, 10 mm x 250 mm) or Vydac semi-preparative C4 column (214TP510; 5 m, 10 mm x 250 mm). Dipalmitoyl-phosphatidylcholine (DPPC) and cholesterol (>98%) were purchased from Aventi Polar Lipid, Inc (USA). Didodecyldimethyl-ammonium bromide (98%) (DDAB) was purchased from Sigma Aldrich, USA. Particle size distribution and measurement of the average particle size were taken using a laser particle size analyzer Mastersizer 2000 (Malvern Instruments, England, UK). Multiplicate measurements were performed and the average particle size was recorded.
Synthesis of 8Qm.
8Qm epitope (QAEPDRAHYNIVTF;E744−57) was synthesized using the previously reported method.18 Briefly, the peptide was synthesized on pMBHA resin (substitution ratio: 0.59 mmol/g, 0.2 mmol scale, 0.34 g) using HBTU/DIPEA Boc-chemistry by microwave-assisted solid-phase peptide synthesis (MW-SPPS). The temperature was set at 70°C (at 20 W, 10 minute) for amino acid coupling except for His and Asp, which were coupled at 50°C (at 20 W, 15 minute). Each amino acid coupling cycle consisted of Boc-deprotection with 100% TFA (2 × 1 minute, at room temperature (RT)), a 1 min DMF flow wash, followed by two 10 min couplings with the pre-activated amino acid. Amino acid activation was achieved by dissolving Boc-amino acid (0.84 mmol, 4.2 equivalent), in a 0.5 M HBTU/DMF solution (1.6 mL, 0.8 mmol, 4.0 equivalent) followed by the addition of DIPEA (0.22 mL, 1.24 mmol, 6.2 equivalent). Amino acids were pre-activated for 1 min prior to their addition to the resin. Synthesis of 8Qm was finalized with Boc removal, N-acetylation and DNP (2,4-dinitrophenyl) group removal from His by treating the resin with 20% (v/v) β-mercaptoethanol and 10% (v/v) DIPEA in DMF (2 × 1 h) prior to peptide cleavage. Upon synthesis completion, the resin was washed with DMF, DCM, and MeOH, then dried under vacuum. The peptide was cleaved from the resin using HF, with p-cresol as scavengers at -5 °C. The cleaved peptide was precipitated, washed thoroughly with ice-cold Et2O and dissolved in 50% MeCN/0.1% TFA/H2O. After lyophilization, the crude peptide was obtained as an amorphous powder. The product was purified by preparative RP-HPLC on a C18 column with a solvent gradient of 30−50% solvent B over 40 min. Pure 8Qm peptide took the form of a white powder after lyophilization. HPLC analysis (C18 column): tR = 20.92 min, purity >95%. Yield: 41.6%, ESI-MS: m/z 1702.2 (calc 1702.9) [M+H]+; 851.9 (calc 851.9) [M+2H]2+; 635.2 (calc 568.3) [M+3H]3+; MW 1701.9.
Synthesis of N3CH2C(O)-8Qm.
N3CH2C(O)-8Qm was synthesized by manual stepwise SPPS on rink amide MBHA resin (substitution ratio: 0.79 mmol/g, 0.2 mmol scale, 0.25 g) using HBTU/DIPEA Fmoc-chemistry. Amino acid activation was achieved by dissolving Fmoc-amino acid (0.84 mmol, 4.2 equivalent) in 0.5 M HBTU/DMF solution (1.6 mL, 0.8 mmol, 4.0 equivalent.) followed by the addition of DIPEA (146 μL, 0.84 mmol, 4.2 equivalent.). The coupling cycle consisted of Fmoc deprotection with 20% piperidine in DMF (for 10 min, then again for 20 min), a 1 min DMF flow-wash, followed by coupling with 4.2 equivalent of pre-activated Fmoc-amino acid (for 30 minutes, then again for 1 hour). The attachment of azido acetic acid (4.2 equivalent) was achieved using HBTU (3 equivalent)/DIPEA (4.2 equivalent) at RT for 4 h in the dark. Upon synthesis completion, the resin was washed with DMF, DCM and MeOH, then dried under vacuum. Cleavage of N3CH2C(O)-8Qm was carried out by stirring the resin in a solution of TFA (99%)/triisopropylsilane/water (95:2.5:2.5) for 4 h. The cleaved peptide was precipitated, filtered and washed with ice-cold Et2O. After lyophilization the crude peptide was obtained as an amorphous powder. The product was purified by preparative RP-HPLC on a C18 column with a solvent gradient of 10−35% solvent B over 10-40 minute. Pure N3CH2C(O)-8Qm was a white powder after lyophilization. HPLC analysis (C18 column): tR = 22.5 min, purity >95%. Yield: 80% ESI-MS: m/z 1742.9 (calc. 1743.9) [M+H]+; 872.2 (calc. 872.5) [M+2H]2+; MW 1742.9.
Synthesis of 8Qm-D Conjugate.
8Qm-D (4.8 mg, 0.26 mmol, 10 equivalent) and dendrimer D (5 mg, 0.26 mmol, 1.0 equivalent) were dissolved in DMF (1 mL). Copper wires (55 mg) were added into the mixture. The mixture was bubbled with nitrogen gas to remove the air for 30s. Reaction mixtures were covered and protected from light with aluminum foil and stirred at 45-50 °C in a temperature-controlled oil bath under nitrogen atmosphere for 12 h. The wires were filtered off from the warm solution and washed with 1 mL of DMF. Particles were self-assembled by slow addition of Millipore endotoxin-free water (7 mL) into the DMF reaction solution (flow rate: 0.05 mL/min) and then exhaustively dialyzed against endotoxin-free water using presoaked and rinsed dialysis bags (Pierce Snakeskin, MWCO 3K) for 3 days. The self-assembled particles were then lyophilized to obtain a white powder form of 8Qm-D conjugate.
All liposomes were formulated using thin-film formation followed by sonication. Dipalmitoylphosphatidylcholine, di-dodecyldimethylammonium bromide and cholesterol were dissolved in chloroform to final concentrations of 10 mg/mL, 5 mg/mL and 5 mg/mL, respectively. All components were mixed in a round bottom flask with a 5:2:1 weight ratio. One mg of lyophilized 8Qm peptide or 8Qm-D conjugate was dissolved in 1 mL of 1:1 chloroform:MeOH and transferred into the lipid mixture flask. For thin film formation, the organic solvents were gradually evaporated using a rotary evaporator and lyophilized overnight to complete solvent removal. Liposome thin films were rehydrated with 900 L of Millipore endotoxin-free water. The suspensions were then sonicated four times with a micro-sonicate probe (40% of power, 20s of pulsing for 2 minutes) to obtain homogenous liposomes. Prior to injection, 100 L of 10x PBS was added to the formulation producing 1 mg/mL concentration.
The characterization of liposome containing 8Qm and 8Qm-D.
The hydrodynamic diameter and zeta potential of prepared blank liposome (8Qm-L and 8Qm-D-L) were measured by dynamic light scattering (DLS) using a zetasizer (Nano ZX, Malvern, England). The particle solution was diluted in sterile 1x PBS before being measured. Measurements were performed five times at 25 °C. The particle morphology of 8Qm-L and 8Qm-D-L were examined using transmission electron microscopy (HT7700 Exalens, HITACHI Ltd., Japan) after vacuum-drying. Briefly, samples were diluted in pure distilled water (1:100) and dropped directly on a glow-discharged carbon coated copper grid and then stained with 2% uranyl acetate. Samples were observed at a magnification of 200,000.
Encapsulation efficacy (%E.E.) of 8Qm into liposomes was determined by addition onto Amicon membrane filter (Amicon Ultra-15, Merch Millipore Ltd., Darmstadt, Germany). Filled amicon tubes were centrifuged at 8000 g for 1 hour. The concentration of un-encapsulated 8Qm in the aqueous phase (supernatant) was measured using a microplate reader and nanodrop with UV-Vis absorption at 280 nm wavelength.
%E.E. was calculated as:
where Ci represents the initial concentration of peptide added to liposomes and Cf represents the concentration of encapsulated peptide.
Mice and cell lines TC-1.
TC-1 cells (murine C57BL/6 lung epithelial cells transformed with HPV-16 E6/E7 and ras oncogenes) were obtained from TC Wu. TC-1 cells were cultured and maintained at 37°C/5% CO2 in RPMI 1640 medium (Gibco) supplemented with 10% heat inactivated fetal bovine serum (Gibco). Female C57BL/6 (6−8-week-old) mice purchased from the Animal Resources Centre (Perth, Western Australia) were used. Animal experiments were approved by The University of Queensland Animal Ethics Committee (UQDI/TRI/351/15) in accordance with National Health and Medical Research Council (NHMRC) of Australia guidelines.
In Vivo Tumor Treatment Experiments.
C57BL/6 mice (5 per group) were first challenged subcutaneously in the right flank with 2 × 105 TC-1 tumor cells/mouse. After 7 days, the mice were injected subcutaneously on each side of the tail base (2 injections) with 8Qm-D, 8Qm-D-L or 8Qm-L (each bearing of 50 μg of 8Qm-D or 15 g of 8Qm in 50 μL of PBS). Positive control mice received 30 μg of 8Qm (equivalent to the 8Qm content in 100 μg of 8Qm-D) emulsified in a total volume of 100 μL of IFA (Seppic, France)/PBS (1:1, v/v). The negative control group was administered 100 µL of PBS. All mice received a single dose of vaccine. Tumor size was measured by palpation and calipers every second day and reported as the average tumor size across the group of five mice or as tumor size in individual mice. Tumor volume was calculated using the formula: V(cm3)=3.14×[largest diameter×(perpendicular diameter)2]/6. Mice were euthanized when tumors reached 1 cm3 or if they started bleeding to avoid unnecessary suffering.
All data were analyzed using GraphPad Prism 7 software. Kaplan−Meier survival curves for tumor treatment experiments were applied. Differences in survival treatments were determined using the log-rank (Mantel-Cox) test, with p < 0.05 considered statistically significant.
Results and discussion
Synthesis and characterization of 8Qm-D and 8Qm-D-L
8Qm-D conjugate was synthesized by copper-catalyzed alkyne−azide cycloaddition reaction (Fig. 1) and self-assembled into microparticles (10 µm) via solvent replacement, as reported previously.21 Unreacted peptides, copper and organic solvents were removed by extensive dialysis against water for 3 days. The degree of substitution was 85%; thus 7 arms of the 8-arm-dendrimer were substituted with peptide epitopes. The substitution was quantified by elemental analysis based on the observed nitrogen to carbon ratio for 8Qm-D (N/C = 0.134), as compared with that of the polymer (N/C = 0.017),15 and was slightly higher than reported previously (N/C = 0.124).17
Table 1. Physicochemical Properties of 8Qm-L and 8Qm-D-L Measured by Dynamic Light scattering.
PDI (polydispersity index)
Zeta potential (mV)
145 ± 2
39 ± 2
120 ± 2
0.16 ± 0.01
26 ± 1
136 ± 2
0.30 ± 0.02
23 ± 1
8Qm-D was anchored to liposomes during thin film formation, followed by hydration with sterile PBS and sonication. The 8Qm peptide was encapsulated into liposomes in a similar manner to form 8Qm-L. Liposome hydro-dynamic size and zeta potential was examined using dynamic light scattering (DLS, Table 1). 8Qm-L and 8Qm-D-L were 120-140 nm in diameter and had similar positive surface charge (23-26 mV). As expected, the incorporation of 8Qm-D in liposomes resulted in increased PDI (0.30) compared to blank liposomes (0.21). TEM photographs demonstrated spherical liposomes with similar size to those measured by DLS. 8Qm-D-L showed a higher tendency to collapse during sample preparation (drying) for TEM analysis (Fig. 2), similar to what was reported by other groups.22, 23 The encapsulation efficacy was indirectly measured using UV-Vis absorption at 280 nm and was found to be approximately 70% for 8Qm-L. As 8Qm-D formed microparticles by itself, the measurement of encapsulation efficacy was not reliable. However, we did not visually observe such particles in the 8Qm-D liposomal formulation through DLS or TEM, suggesting quantitative incorporation of 8Qm-D into liposomes due to the highly hydrophobic nature of poly(tert-butyl acrylate). A similar effect was observed when lipids were used to anchor peptides to liposomes.24, 25 Taken together, the physicochemical properties of liposomes 8Qm-L and 8Qm-D-L were only slightly different.