Co-delivery of cancer-associated antigen and Toll-like receptor 4 ligand in PLGA nanoparticles...

Vaccine 26 (2008) 5046–5057

Contents lists available at ScienceDirect

Vaccine

journa l homepage: www.e lsev ier .com/ locate /vacc ine

Co-delivery of cancer-associated antigen and Toll-like receptor 4 ligand in PLGAnanoparticles induces potent CD8+ T cell-mediated anti-tumor immunity�

Samar Hamdy, Ommoleila Molavi, Zengshuan Ma, Azita Haddadi, Aws Alshamsan,Zahra Gobti, Sara Elhasi, John Samuel, Afsaneh Lavasanifar ∗

Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, Alberta T6G 2N8, Canada

a r t i c l e i n f o

Article history:Received 8 April 2008Received in revised form 14 July 2008Accepted 14 July 2008Available online 3 August 2008

This paper is dedicated to the memory ofProf. John Samuel, who was a distinguishedscientist in the field of cancerimmunotherapy. Prof. Samuel supervisedthis project before he passed away on 17April 2007.

a b s t r a c t

The purpose of this study was to evaluate the efficacy of poly(lactic-co-glycolic acid) (PLGA)-based vaccinesin breaking immunotolerance to cancer-associated self-antigens. Vaccination of mice bearing melanomaB16 tumors with PLGA nanoparticles (NP) co-encapsulating the poorly immunogenic melanoma antigen,tyrosinase-related protein 2 (TRP2), along with Toll-like receptor (TLR) ligand (7-acyl lipid A) was exam-ined. Remarkably, this vaccine was able to induce therapeutic anti-tumor effect. Activated TRP2-specificCD8 T cells were capable of interferon (IFN)-� secretion at lymph nodes and spleens of the vaccinatedmice. More importantly, TRP2/7-acyl lipid A-NP treated group has shown immunostimulatory mileu atthe tumor microenvironment, as evidenced by increased level of pro-inflammatory cytokines comparedto control group. These results support the potential use of PLGA nanoparticles as competent carriers forfuture cancer vaccine formulations.

Keywords:PLGAL

micTmT

Bb“baaMs

ipid A and melanoma

. Introduction

The overall objective response rate to current cancer vaccine for-ulations is only 3.3% [1]. Among the various reasons for this failure

n immune response, two important issues stand out: (1) current

he major goal for improvement of response to cancer vaccines

� BC1-005 (7-acyl lipid A) was recently synthesized by Oncothyreon, Inc. (formerlyiomira, Inc.) as a synthetic analogue of MPLA. Due to the similarity in structureetween the two compounds (BC1-005 and MPLA), we had given BC1-005, the namesynthetic MPLA” in our earlier reports [13,14]. However, to avoid any confusionetween the natural MPLA “produced and purchased from Galaxo” and the syntheticnalogue “synthesized by Biomira”, we have recently decided to use the name “7-cyl lipid A” for BC1-005 (in Ref. [12] and in the current study) instead of syntheticPLA. An erratum to our earlier publications [13,14] with this information will be

ubmitted soon.∗ Corresponding author at: Faculty of Pharmacy and Pharmaceutical Sciences,niversity of Alberta, 4119 Dentistry/Pharmacy Centre, Edmonton, Alberta T6G 2N8,anada. Tel.: +1 780 492 2742; fax: +1 780 492 1217.

E-mail address: alavasanifar@pharmacy.ualberta.ca (A. Lavasanifar).

pigads(sapgiiusdaa

s to develop immunotherapy strategies that can activate robustnd lasting immune responses against cancer antigens and, at theame time, be able to reverse the ‘immunosuppressive milieu’ ofhe tumor microenvironment.

An ideal cancer vaccine formulation comprises three main com-onents; first, an antigen against which the immune responses are

nduced. The second component is an adjuvant that acts as dan-er signal to alert the immune system and activate early as wells long-lasting immune responses. The third component is theelivery system that delivers vaccine antigen and adjuvant to thepecialized antigen presenting cells (APCs), mainly dendritic cellsDCs), in a targeted and prolonged manner [4]. Among differentystems tested, nanoparticles made of poly(d,l-lactic-co-glycoliccid) (PLGA) are of special interest. In addition to their biocom-atibility and biodegradability, PLGA nanoparticles (PLGA-NP) offerreat flexibility with respect to the manipulation of physicochem-cal properties of the polymer and the range of antigens andmmunomodulators that they can accommodate. PLGA-NP are nat-

ine 26

ilumh[tda[dIedo

crtaiemdsswFirPocideTpacrmisteDutootuabmtiite

wcTaTaisbrgAcspt

wEBM(Mnw(f�UpsUp1(r

oedti(oAt(Pw

S. Hamdy et al. / Vacc

o presentation by both MHC class I and class II molecules on theurface of DCs, with the final outcome of simultaneous activationf both CD8+ and CD4+ T cell immune responses, respectively [6].

The discovery of Toll-like receptors (TLRs) and their crucial rolen orchestrating both innate and adaptive immune response haveed to the development of an entire class of potent immunomod-altory adjuvants, namely TLR ligands. One of the most promisingembers in this class is monophosphoryl lipid A (MPLA), which

as been recently approved for Hepatitis B vaccination in Europe7] and has been used extensively in a variety of clinical vaccineesting (reviewed in Refs. [8–10]). MPLA is a chemically modifiederivative of lipopolysaccharide (LPS), that exhibits potent adjuvantctivity, but is up to 10,000-fold less toxic than parent LPS molecule11]. 7-acyl lipid A, the novel synthetic analogue of lipid A, intro-uced in collaboration with Oncothyreon, Inc. (formerly Biomira,

nc., Edmonton, AB, Canada) has comparable immunostimulatoryffects to MPLA, but harbor additional advantages such as repro-ucibility, feasibility for large scale production and better controlver purity of the final products [12].

The long-term objective of this research is to achieve vac-ines based on PLGA-NP that can induce effective T cell immuneesponses against cancer. Our research group has recently shownhat co-delivery of ovalbumin (OVA) as model antigen, and 7-cyl lipid A in PLGA-NP enhances DC maturation and dramaticallynduces robust OVA-specific primary CD4+ and CD8+ T cell prolif-rative responses [13,14]. Moreover, vaccinating healthy C57Bl/6ice with PLGA-NP co-encapsulating OVA and 7-acyl lipid A

ramatically enhanced interferon (IFN)-� secretion by both OVA-pecific CD4+ and CD8+ T cells in both draining lymph nodes andpleens of vaccinated mice. This activation was superior to whate observed in mice immunized with OVA emulsified in complete

reund’s adjuvant (CFA), one of the most powerful adjuvants usedn animal studies (our unpublished results). Consistent with theseesults, in a recent study by Heit et al. [15] a single injection ofLGA microparticles co-encapsulating OVA and TLR9 ligand (CpG-ligodeoxynucleotide (ODN)) efficiently activated OVA-specific Tells and caused complete tumor regression in 80% of animals bear-ng OVA-expressing B16 melanoma (B16-OVA). These results clearlyemonstrate the efficacy of PLGA nano/microparticles in deliveringxogenous antigens and adjuvants into DCs and initiating robustcell responses. However, the vigorous T cell activation can be

artially attributed to the fact that chicken OVA is recognized asforeign antigen (non-self) in mice. As a result, OVA-specific T

ells could be easily activated, as they are not subjected to theegulations of central and peripheral tolerance. By contrast, theajority of cancer antigens belong to self. Accordingly, stimulat-

ng the immune system against those antigens requires breaking ofelf-tolerance mechanisms, which is more challenging and difficulto achieve. An elegant study by Bellone et al. [16], had compared thefficacy of three vaccination strategies (naked DNA, peptide-pulsedCs or mixture of peptide and the Escherichia coli toxin LTR72)sing either the foreign antigen OVA or the naturally expressedyrosine related protein 2 (TRP2) tumor antigen in the B16-OVAr B16 melanoma models, respectively. They found similar levelf cytotoxic T lymphocytes (CTL) activation and tumor protec-ion against B16-OVA melanoma by all the three vaccines thatsed OVA antigen. However, when TRP2, a self-tumor-associatedntigen, was employed, all vaccines elicited B16-specific CTLs,ut only the TRP2-pulsed DCs were able to protect against B16elanoma. Two conclusions can be made from these results: (1)

he induction of in vitro CTL activity does not correlate with then vivo anti-tumor activity; (2) foreign (non-self) antigens usedn evaluating immunotherapeutic strategies can overestimate theherapeutic outcome and lead to bias in the validation of vaccinefficacy.

(2008) 5046–5057 5047

In the light of these findings, the purpose of the present studyas to rigorously evaluate the efficacy of PLGA-based cancer vac-

ines using a realistic and clinically relevant tumor antigen, i.e.,RP2. The results of this study showed that co-delivery of TRP2nd 7-acyl lipid A in PLGA-NP is highly effective in inducingRP2-specific CD8+ T cell responses capable of mediating ther-peutic anti-tumor response in mouse B16 tumor model. Moremportantly, our vaccine strategy led to the reversal of immuneuppressive milieu of the tumor microenvironment, as evidencedy increase in the level of pro-inflammatory T helper 1 (Th1)-elated cytokines and decrease in the level of vascular endothelialrowth factor (VEGF), a key factor required for tumor growth [17].ll together, our results support the potential use of PLGA-NP asompetent antigen delivery systems, and further illustrated theuperior therapeutic outcome of co-delivery 7-acyl lipid, a veryromising immunostimulatory adjuvant, for future cancer vaccinerials.

. Materials and methods

.1. Reagents

Synthetic 7-acyl lipid A (also called BC1-005), M.Wt., 1955.5 Daas kindly provided by Oncothyreon, Inc. (formerly Biomira, Inc.,

dmonton, AB, Canada). TRP2 peptide180–188 was purchased fromioSynthesis, Inc. (Lewisville, TX, USA). Polyvinyl alcohol (PVA),.Wt., 31,000–50,000 Da was obtained from Sigma–Aldrich co.

Oakville, ON, Canada). PLGA co-polymer (monomer ratio 50:50,.Wt., 7000 Da) was purchased from Absorbable Polymers Inter-

ational (Pelham, AL, USA). Chloroform, methanol, acetonitrile,ater (all HPLC grades) were purchased from Fisher Scientific

Fair Lawn, NJ, USA). Murine CD8 isolation kit was purchasedrom StemCell Technologies (Vancouver, BC, Canada). Murine IFN-

ELISPOT kit was purchased from E-Bioscience (San Diego, CA,SA). DMEM, RPMI media, l-glutamine, and gentamicin wereurchased from Gibco-BRL (Burlington, ON, Canada). Fetal calferum (FCS) was obtained from Hyclone Laboratories (Logan,T, USA). Murine IL-6, IL-2, IFN-� and TNF-� ELISA kits wereurchased from E-Bioscience (San Diego, CA, USA). Murine IL-2 and VEGF ELISA kits were purchased from BD BiosciencesMississauga, ON, Canada) and R&D Systems (Minneapolis, USA),espectively.

.2. Preparation of PLGA-NP encapsulating TRP2180–188 with orithout 7-acyl lipid A

Nanoparticles of PLGA containing TRP2 peptide with or with-ut 7-acyl lipid A were prepared by water/oil/water doublemulsion/solvent evaporation method. Briefly, TRP2 peptide wasissolved in 75% acetonitrile in water to make 10 mg/mL solu-ion. From this solution 100 �L was emulsified with PLGA solutionn chloroform (300 �L, 50%, w/v) using a microtip sonicatorHeat systems, Inc., Farmingdale, NY, USA). For the preparationf nanoparticles containing 7-acyl lipid A, 200 �g of 7-acyl lipidin 100 �L of 1:4 methanol–chloroform mixture was added to

he polymer–chloroform solution. The resulting primary emulsionwater/oil) was further emulsified in 2 mL of PVA solution (9%, w/vVA in PBS) by sonication for 45 s at level 4. The secondary emulsionas added drop-wise into 8 mL of stirring PVA solution. Nanopar-

icles were collected after 3 h of stirring by centrifugation of themulsion at 40,000 × g for 10 min at 4 ◦C. The nanoparticles wereashed twice with cold deionized water and lyophilized. The par-

icle size of the nanoparticles was determined by dynamic lightcattering technique using a Zetasizer 3000 (Malvern, UK). Quan-

5048 S. Hamdy et al. / Vaccine 26 (2008) 5046–5057

Table 1Chromatographic gradient program over LC–MS analysis time (15 min)

Time (min) Mobile phase Aa (%) Mobile phase Bb (%) Flow rate (mL/min) Gradient curve

0 0 100 0.2 110 70 30 0.2 511 0 100 0.2 1

4wwihsasa(s(ddsratpnrawtslfil

ifpioii1Loif

Twodws1rdds

msm∼TPbda

FCufwgwEb1Tt

a Mobile phase A is acetonitrile.b Mobile phase B is 0.05% trifluoroacetic acid (TFA)/water.

ification of 7-acyl lipid A content in PLGA-NP was done by liquidhromatography–mass spectrometry (LC–MS) as reported previ-usly [12].

.3. Quantification of encapsulated TRP2180–188 in nanoparticlesy LC–MS

.3.1. Establishment of LC–MS method for the quantification ofRP2180–188

LC–MS analyses were performed using a Waters Micromass ZQ000 spectrometer, coupled to a Waters 2795 separations moduleith an autosampler (Milford, MA, USA). The mass spectrometeras operated in positive ionization mode with selected ion record-

ng (SIR) acquisition. The nebulizer gas was obtained from an inouse high purity nitrogen source. The temperature of source waset at 150 ◦C, and the voltages of the capillary and cone were 3.10 kVnd 9 V, respectively. The gas flow of desolvation and the cone wereet at 550 and 90 L/h, respectively. Chromatographic separation waschieved using a Waters (Milford, MA, USA) XTerraMSC18 3.5 �m2.1 mm × 50 mm) as the stationary phase. The mobile phase con-isted of two solutions; solution A (acetonitrile) and solution B0.05% trifloro acetic acid (TFA)/water). The mobile phase waselivered at a constant flow rate of 0.2 mL/min. The gradient con-itions are shown in Table 1. Lansoprazole was used as the internaltandard (I.S). Single ion recording (SIR) at m/z 1175.2 and 369.9,elated to M + H were selected for quantification of TRP2 peptidend the internal standard, respectively (Fig. 1A). Fig. 1B showshe SIR chromatograms of internal standard and TRP2180–188. Theeak of TRP2180–188 was well separated from that of the inter-al standard in the established chromatographic condition. Theetention times of the internal standard and TRP2180–188 werepproximately 7 and 8 min, respectively. The analytical run timeas 15 min. The regression analysis was constructed by plotting

he peak–area ratio of TRP2180–188 to I.S (response factor) ver-us TRP2180–188 concentration (�g/mL). The calibration curve wasinear within the range of 1.5 �g to 50 �g/mL. The correlation coef-cient (R2) was always greater than 0.99 (Fig. 1C), indicating a good

inearity.

.3.2. Extraction and quantification of TRP2180–188 in PLGA-NPExtraction of TRP2180–188 from PLGA-NP was done by dispers-

ng 5 mg of nanoparticles in 300 �L of 85% acetonitrile in H2O,ollowed by centrifugation at 15,000 × g for 15 min. The PVA wasrecipitated and the supernatant (PLGA + peptide) was transferred

nto eppendorf tube. The solvent (acetonitrile and H2O) was evap-rated using Thermosavant SpeedVac System. Since the peptides soluble in methanol (unlike PLGA), the residue was dissolvedn 500 �L methanol, followed by centrifugation at 15,000 × g for

.3.3. Standard and stock solutionsThe stock solutions were prepared by dissolving 10 mg of

RP2180–188 in 1 mL of 75% acetonitrile in water. The stock solutionas stored at −20 ◦C between experiments. The working solutionf TRP2180–188 was prepared fresh each day by making a 100-foldilution of the stock solution in methanol. The calibration standardsere then prepared by serial dilution of the working solution. The

tock solution of the internal standard was prepared by dissolving0 mg of lansoprazole A in 1 mL methanol, followed by the prepa-ation of a working solution of 100 �g/mL by a further 100-foldilution of the stock solution. The stock solution of the internal stan-ard was stored at −20 ◦C between experiments, and the workingolution was prepared fresh at each experiment.

.4. Animal studies

.4.1. MiceMice (C57Bl/6) were purchased from the Jackson Laboratory (Bar

arbor, ME, USA). All experiments were performed in accordanceo the University of Alberta guidelines for the care and use of labo-atory animals. All experiments were performed using 10–16 weekld male mice.

.4.2. Normal mice vaccination experimentThe efficacy of our vaccine formulations were first tested on nor-

al C57Bl/6 mice (with no tumor). Animals (5 mice/group) wereubcutaneously (s.c.) vaccinated in right flank region with approxi-ately 10 �g TRP2180–188 encapsulated in PLGA-NP with or without17 �g 7-acyl lipid A (abbreviated as TRP2/7-acyl lipid A-NP andRP2-NP, respectively). Control group mice received 10 mg plainLGA-NP (Empty-NP). Eleven days later all mice received similarooster immunization. Seven days after the second immunization,raining lymph nodes and spleens were isolated and the ELISPOTssay was performed (see details below).

.4.3. Tumor therapy experimentB16-F10 cells were grown in DMEM supplemented with 10%

CS, 2 mM l-glutamine and 100 IU/mL penicillin/streptomycin in 5%O2 atmosphere. At day 0, C57Bl/6 mice were injected s.c. at theirpper right flank with 0.1 × 106 B16-F10 melanoma cells obtainedrom 90–95% confluent cultures. Three days later (day 3), animalsere randomly assigned to three treatment groups (8–10 mice per

roup). Similar to normal mice vaccination study, the three groupsere s.c. vaccinated (in the lower right flank region) with either

mpty-NP, TRP2-NP or TRP2/7-acyl lipid A-NP. Animals were givenooster immunization with the same formulations at days 7 and3. Palpable tumors start to appear between day 7 and day 10.umor size was measured with vernier caliper, starting at day 7 andhen every 2–3 days until day 21. The longest length and the length

S. Hamdy et al. / Vaccine 26 (2008) 5046–5057 5049

F P2180

T atogrA e calibt ion (�

ireotm

22EntppopNcsSaScTtPnt

MIetcw1fswInw

ec(tTmdm

ig. 1. Liquid chromatography–mass spectrometry (LC–MS)-based method for TRRP2180–188 solution and TRP2180–188 peptide extracted from PLGA-NP. (B) SIR chromrepresentative standard curve for TRP2180–188 extending from 1.5 to 50 �g/mL. Th

o the internal standard (response factor ± S.D.) on y-axis vs. TRP2180–188 concentrat

solated for further analysis. Weights of individual tumors wereeported and used as a tool to calculate the percentage of mice (forach treatment group) that had controlled tumor growth through-ut the study, using the following formula: (number of the micehat have tumor weight less than 0.3 g at day 21/total number of

ice in that group) × 100.

.4.4. Assessment of vaccine-induced immune stimulation

.4.4.1. Enzyme-linked immunospot (ELISPOT) assay. For theLISPOT assay, spleens and draining lymph nodes (from eitherormal or tumor-bearing mice) were isolated and washedhree times in PBS. Lymphocytes (from lymph nodes) were sus-ended at 1 × 107 cells/mL in complete RPMI media with 1%enicillin–streptomycin, 1% l-glutamine, and 5% FCS. On thether hand, spleenocyte cell suspensions were divided into twoortions; first portion was treated by ACK lysis buffer (156 mMH4Cl, 10 mM KHCO3, 100 �m EDTA) for removal of red bloodells (RBCs). Briefly, 5 mL of ACK lysis buffer was added (perpleen), followed by 1 min incubation at room temperature.plenocytes were then washed twice in cold plain RPMI mediand re-suspended at 1 × 107 splenocytes/mL in complete RPMI.econd part of spleenocyte cell suspension underwent CD8+ Tell isolation, using EasySep® mouse CD8 isolation kit (StemCell

–188 quantification. (A) Mass spectra of lansoprazole, the internal standard (I.S.),ams of the I.S. and TRP2180–188. Gradient conditions are summarized in Table 1. (C)ration curves were constructed by plotting average peak–area ratio of TRP2180–188

g/mL) on x-axis.

Ninety-six-well MultiScreenTM filter plates (Millipore, Bedford,A, USA) were coated overnight at 4 ◦C with 100 �L/well of

FN-�-specific capture antibody (anti-mouse IFN-�, clone AN-18,Bioscience, San Diego, CA, USA), diluted in sterile PBS according tohe manufacturer’s instructions. After the overnight incubation, theoating antibody solution was decanted from the plates. The platesere washed twice with 200 �L/well PBS and then blocked forh at room temperature with 200 �L/well of complete RPMI-1640

or. After blocking, plates were decanted and cells (lymph nodes,plenocytes or isolated CD8+ T cells) were added into individualells (in triplicates) in complete RPMI medium at 1 × 106 cells/well.

n addition, using lymphocytes isolated from normal mice (vacci-ated from TRP2/7-acyl lipid A-NP), two additional cell numbersere plated; 0.5 × 106 and 2 × 106 cells/well.

Cells were then stimulated by 20 �M of either CD8 irrelevantpitope (SIINFEKL) or positive epitope (TRP2180–188). In negativeontrol wells 100 �L/well of complete RPMI media was addednon-stimulated). The cells were incubated at 37 ◦C for 18 h inhe presence of 5% CO2 and then washed three times with 0.05%ween in PBS (PBS-Tween). The detection antibody (Biotin anti-ouse IFN-�, clone R4-6A2, eBioscience, San Diego, CA, USA), was

iluted in 1% bovine serum albumin (BSA) in PBS according to theanufacturer’s instructions. Diluted solution was added at a vol-

me of 100 �L per well. After 2 h incubation at room temperature,lates were washed four times with PBS-Tween and incubated for5 min at room temperature with Streptavidin-HRP (eBioscience,an Diego, CA, USA). Plates were then washed three times with PBS-ween followed by two times wash with PBS. Spots were developed

5 ine 26

bcCbam

2iiattntIkw4df

eSBecvgw

bewfTlwN

Nlst(TalnbleF

iNa0wibucp

lopsTsTI(

tRwmbbpAsawNTsoN(

giaCNtnAnoifed

050 S. Hamdy et al. / Vacc

y adding 100 �L/well of freshly prepared AEC (3-amino-9-ethylarbazole) Substrate Solution (BD Bionsciences, Mississauga, ON,anada). Stopping the substrate reaction was done after 20–30 miny washing three times with distilled water. Plates were then driednd spots were counted in a BioReader 3000 (BioSys, Karben, Ger-any).

.4.4.2. Enzyme-linked immunosorbent assay (ELISA). Vaccine-nduced alteration in the level of pro-inflammatory cytokines andmmuno-suppressive factors in tumor microenvironment weressessed by ELISA. Briefly, isolated tumors were crashed betweenwo slides to form uniform cell suspensions, which were filteredhrough 70 �m cell strainers and then counted. Tumor super-atants obtained after centrifugation of approximately 20 millionumor cells were analyzed for the level of TNF-�, IL-12, IFN-�, IL-2,L-6 and VEGF by ELISA using the commercially available ELISAits in a 96 well microplate using a microplate reader (Powerwaveith KC Junior software; Bio-Tek, Winooski, VT, USA) at OD of

50 nm according to the manufacturer’s directions. The minimumetection levels of the cytokines were: 7, 62, 15, 2, 10 and 7.8 pg/mLor TNF-�, IL-12, IFN-�, IL-2, IL-6 and VEGF, respectively.

.4.5. Statistical analysisThe significance of differences among groups was analyzed

ither by one-way analysis of variance (ANOVA) followed by thetudent–Newman–Keuls post hoc test for multiple comparisons.efore executing the ANOVA, data were tested for normality andqual variance. If neither of the latter criteria were met, data wereompared using a Kruskal–Wallis one-way ANOVA on ranks. P-alue of ≤0.05 was set for the significance of difference amongroups. The statistical analysis was performed with SigmaStat soft-are (Systat Software, Inc. San Jose, CA, USA).

. Results

.1. Characterization of PLGA-NP

The mean hydrodynamic diameter of nanoparticles rangedetween 350 and 410 nm with a polydispersity below 0.2. Thefficiency of the method of TRP2180–188 extraction from PLGA-NPas 87.6%. Based on LC–MS analysis the encapsulation efficiency

or TRP2180–188 was 5.2 ± 0.6% and the loading was 0.94 ± 0.11 �gRP2180–188 entrapped (per 1 mg dry weight of NP). The encapsu-ation efficiency for 7-acyl lipid A was 67.3 ± 6.9% and the loadingas 1.79 ± 0.18 �g 7-acyl lipid A entrapped (per 1 mg dry weight ofP).

.2. Co-delivery of TRP2180–188 and 7-acyl lipid A in PLGA-NPnduces IFN-� secretion by TRP2-specific CD8+ T cells in lymphodes and spleens of the vaccinated mice

Despite the low encapsulated levels of TRP2180–188 in PLGA-P, we were interested to see whether vaccination with such

ow amount of the peptide (∼10 �g) will induce antigen-pecific CD8+ T cell activation in healthy C57BL/6 mice. Inhis experiment, wild-type C57Bl/6 mice were vaccinated twice11 days apart) and the induction of IFN-�-producing CD8+

cells was investigated by a single-cell, ex vivo ELISPOTssay. In the beginning, we performed a titration study usingymphocytes from the test group (OVA/7-acyl lipid A-NP immu-

owest background in wells stimulated with medium or irrel-vant peptide). The result of this titration study is shown inig. 2A. Actual number of spots/well ± standard deviation (S.D.)

(2008) 5046–5057

s presented in the bar graph in the right panel of Fig. 2A.umber of IFN-� secreting antigen-specific T cells increaseds the number of lymphocytes/well increased. When we used.5 × 106 lymphocytes/well, the number of antigen-specific T cellsas relatively low. With 1 × 106 lymphocytes/well, there was clear

ncrease in the number of IFN-� secreting cells with acceptableackground. However, the background was very high when wesed 2 × 106 lymphocytes/well. Based on these observations, weoncluded that the optimum cell number per well is 1 × 106 lym-hocytes.

Fig. 2B shows the number of IFN-� secreting cells/millionymphocytes for the three groups tested. The actual numberf spots/well ± S.D. is presented in the bar graph in the loweranel. Immunization with Empty-NP did not induce any mea-urable amount of IFN-�. Immunization with both TRP2-NP andRP2/7-acyl lipid A-NP induced comparable numbers of INF-�ecreting T cells (113.5 ± 5 and 145.6 ± 10), respectively. However,RP2/7-acyl lipid A-NP immunized group induced higher levels ofFN-� secretion as evidenced by the higher density of the spotsFig. 2B).

The isolated spleens were treated using either ACK lysis bufferhat only removes RBCs or negative selection of only CD8+ T cells.elatively high background was observed in the negative controlells of ACK-lysed spleens of TRP2/7-acyl lipid A-NP immunizedice (Fig. 2C). However, when only CD8+ T cells were plated, the

ackground level dropped almost to zero (Fig. 2D). The actual num-er of spots/well ± S.D. is presented in the bar graph in the loweranels in Fig. 2C and D. Mice immunized with TRP2/7-acyl lipid-NP had significantly high numbers of IFN-� producing TRP2-pecific CD8+ T cells, compared to Empty-NP immunized mice (60-nd 65-fold in Fig. 2C and D, respectively). Unlike lymph nodes,here immunization with either TRP2-NP or TRP2/7-acyl lipid A-P gave comparable number of TRP2-specific IFN-� secreting CD8+

cells, in spleens co-delivery of 7-acyl lipid A along with TRP2 in theame NP formulation led to 3- and 12-fold increase in the numberf TRP2-specific IFN-� secreting CD8+ T cells (compared to TRP2-P) using either ACK lysed spleens (Fig. 2C) or isolated CD8 T cells

Fig. 2D), respectively.

.3. Co-delivery of TRP2180–188 and 7-acyl lipid A in PLGA-NPnduced potent therapeutic anti-tumor immunity

.3.1. Tumor growth and size measurementsThree s.c. injections of TRP2-NP appeared to slow down the

rowth of the tumors compared to what observed in the Empty-NPmmunized group (Fig. 3A), although the difference between aver-ge tumor sizes of the two groups was not statistically different.o-delivery of 7-acyl lipid A together with TRP2180–188 in the sameP formulation also slowed down tumor development compared

o Empty-NP group (p < 0.05 in days 17 and 20). Although there waso statistical difference between TRP2-NP and TRP2/7-acyl lipid-NP immunized groups, the average tumor size of animals immu-ized with TRP2/7-acyl lipid A-NP were almost half the average sizebtained in TRP2-NP immunized animals at all times tested. Largentra-group variability might have impaired the significance of dif-erence for results obtained from these two groups. Conducting thexperiment with higher number of mice is needed to clarify theifference between the treatment groups further.

Number of mice/vaccination group ranged from 8 to 10, with 8,

cyl lipid A-NP, respectively. Whereas all the 10 mice immunizedith TRP2/7-acyl lipid A-NP have survived until the end point of

he study (day 21), one animal from both Empty-NP and TRP2-NProups had to be euthanized due to large tumor growth.

Fig. 2. Vaccination of healthy mice with TRP2180–188 containing NP induces IFN-� secretion by TRP2-specific CD8+ T cells in lymph nodes and spleens of the vaccinatedmice. C57Bl/6 mice were s.c. vaccinated in right flank region with approximately 10 �g TRP2180–188 peptide encapsulated in PLGA-NP with or without ∼17 �g 7-acyl lipidA (TRP2/7-acyl lipid A-NP and TRP2-NP, respectively). Control group mice received 10 mg plain PLGA NP (Empty-NP). Eleven days later all mice received similar boosterimmunization. Seven days after the second immunization, draining lymph nodes and spleens were isolated and the ELISPOT assay was performed (as described in text). IFN-�secretion was measured in the vaccinated groups as evidenced by the number of spots/million cells after overnight stimulation with either media alone (non-stimulated)or 20 �M of either CD8 irrelevant peptide (SIINFEKL) or test peptide (TRP2180–188). (A) Lymphocytes isolated from TRP2/7-acyl lipid A-NP immunized groups were used foro l are sw olatedA kit (Db ata sh

rswtai(

ptimization of cell number/well. Numbers of spots per different cell numbers/welas assessed in (B) lymph nodes (C and D) and spleens of the vaccinated mice. Islternatively, CD8+ T cells were isolated from spleens using CD8+ negative isolationar graphs below. The values in those graphs are averages of triplicate wells ± S.D. D

.3.2. Tumor weight measurementsAfter 21 days from tumor cell injections, the mice were sac-

ificed, tumors were isolated from each mouse and weightedeparately. Results are presented in Fig. 3B as individual tumor

he study (7, 8 and 10 mice for Empty-NP, TRP2-NP and TRP2/7-cyl lipid A-NP immunized groups, respectively). In the groupmmunized with Empty-NP, all the animals developed large tumors>0.3 g), with the exception of only one mouse that had a smaller

hown in the bar graph to the right side. Using 1 million cells/well, IFN-� secretionsplenocytes were either treated with ACK lysis buffer to lyse red blood cells (C).). Numbers of spots/million cells for different treatment groups are shown in the

own are representative of three independent experiments that gave similar results.

umor (∼0.1 g). On the other hand, among the 8 mice in theRP2-NP immunized group, only 3 mice developed tumors >0.3 g.o-delivery of 7-acyl lipid A with TRP2180–188 in PLGA-NP decreasedhe number of mice that develop >0.3 g tumors to only one (out of

5052 S. Hamdy et al. / Vaccine 26 (2008) 5046–5057

Fig. 3. Therapeutic anti-tumor immunity in mice vaccinated with TRP2180–188 containing NP. C57Bl/6 mice were injected s.c. at their upper right flank with 0.1 × 106 B16-F10melanoma cells (day 0). Three days later, animals were randomly assigned to three treatment groups (8–10 mice per group). Similar to normal mice vaccination study (Fig. 2),the three groups were s.c. vaccinated (in the lower right flank region) with either Empty-NP, TRP2-NP or TRP2/7-acyl lipid A-NP. Animals were given booster immunizationwith the same formulations at days 7 and 13. Tumor size was measured with vernier caliper every 2–3 days. The longest length and the length perpendicular to it weremultiplied to obtain the tumor area in mm2. The mean tumor area ± standard error (S.E.) for each group was plotted vs. time (A). *Indicates significant differences (p < 0.05)in tumor area for mice immunized with TRP2/7-acyl lipid A-NP, compared with Empty-NP immunized mice. On day 21 animals were sacrificed and tumors were isolated andw re shT each0 sent tho dpoin

diaIdi

hNnp1l

eighted separately. Tumor weights of individual mice for each vaccination group ahe experiment was repeated one more time, and similar results were obtained. For.3 g at the endpoint of the study (day 21) is shown in (C). Numbers below (C) repref mice used for each group. Pictures of representative mice in each group at the en

mice representative from each treatment group are shown inig. 3D.

.3.3. Ex vivo detection of TRP2-specific CD8+ T cellsTo examine the underlying mechanism behind controlled tumor

evelopment in TRP2/7-acyl lipid A-NP immunized mice, we have

ng lymph nodes (Fig. 4A) or spleens (Fig. 4B and C). On the other

own as scatter plot (B). Averages of the tumor weights from each group are shown.treatment group, the average percentage of mice that had tumor weights less thane actual numbers of mice that had had tumor weights less than 0.3 g/total number

t of the study are shown in (D). Arrows indicate the position of tumors.

and, among different treatments, mice immunized with TRP2-P develop strong antigen-specific response in the draining lymphodes, as evidenced by higher number of INF-� secreting cells com-ared to TRP2/7-acyl lipid A-NP immunized group (141.5 ± 10.8 vs.07.25 ± 8.5, p < 0.05). In contrast, analysis of spleens either by ACKysis (Fig. 4B) or by CD8+ T cell isolation (Fig. 4C) revealed superior

Fig. 4. Ex vivo detection of IFN-� secretion by TRP2180–188-specific CD8+ T cells in lymph nodes and spleen of the tumor-bearing vaccinated mice. C57Bl/6 mice were challengedwith melanoma cells and vaccinated as described in the text and in Fig. 3. At day 21, animals were sacrificed and ELISPOT assay was performed (as described in text and inF een. Sie lls wec te wels

oftlciT

ig. 2) to assess IFN-� secretion in the (A) draining lymph nodes, and (B and C) splither treated with ACK lysis buffer to lyse red blood cells. (C) Alternatively, CD8+ T ceells are presented in bar graphs. The values in those graphs are averages of triplicaimilar results.

ysis buffer (Fig. 4B). Plating pure CD8+ T cell population had elim-nated any non-specific INF-� secretion (Fig. 4C). Average numberf spots ± S.D. (n = 3) are presented in the bar graphs on the rightanel of Fig. 4A (for draining lymph node) and in the lower panelsf Fig. 4B and C (for spleens).

.3.4. Vaccine-induced alteration in the level of pro-inflammatory

Activation of antigen-specific CD8+ T cells within lymph nodesnd/or spleens may be not enough for complete tumor immunity,s the tumor suppressive microenvironment may hinder migration

milar to Fig. 2, spleens were handled by two different ways; splenocytes were (B)re isolated from spleens using CD8+ negative isolation kit. Numbers of spots/millionls ± S.D. Data shown are representative of two independent experiments that gave

f antigen-specific activated T cells into the tumor site or inter-ere with their anti-tumor activity. So in order to carefully evaluatehe effect of our vaccine formulations, it was crucial to detect theevel of various immuno-stimulatory versus immuno-suppressiveytokines in the tumor microenvironmentl. Our results showed thatmmunization with TRP2-NP slightly increased the level of IFN-�,NF-�, IL-6 and IL-2 (p < 0.05 compared to Empty-NP). Co-delivery

5 ine 26

wmoumMtcloriBSttgTeacv(siBhiu

melccpLicattabLi

PlrnTv21asiOtac

td(cliatiT(lT

cmhTaICcoa(mNwtttatIbspcC

npptaletIMttgliBl

. Discussion

Induction of potent, specific and lasting T cell responses asell as reversal of immunosuppressive network in the tumoricroenvironment are two major challenges in the development

f efficacious cancer vaccine strategies. Our objective was to eval-ate the potential of PLGA-based cancer vaccine formulations foreeting those challenges in a reliable and realistic tumor model.urine melanoma B16 is a very aggressive and rapidly growing

umor. These B16 cells possess extensive defects in their MHClass I antigen-processing pathway. As a result, they express veryow amounts of MHC I molecules on their surface [18]. Becausef their poor immunogenicity and aggressiveness, B16 melanomaepresents a challenging tumor model for developing T cell-basedmmunotherapeutic responses. Despite its poor immunogenicity,16 contains antigens capable of activating a specific CTL response.ome of melanoma differentiation antigens include MART-1, gp100,yrosinase, TRP1, and TRP2 [19]. Among those antigens, one epi-ope of TRP2 (TRP2180–188) is of special interest. Several researchroups have shown strong correlation between the presence ofRP2180–188-specific T cells and tumor regression [20–24]. Forxample, a combination therapy of anti-cytotoxic T lymphocytentigen (CTLA)-4 monoclonal antibody and granulocyte monocyte-olony stimulating factor (GM-CSF) producing B16 melanomaaccine revealed that successfully treated mice had elevated level1.7%) of TRP2180–188-specific T cells circulating in blood [25]. Othertudies showed that adoptive transfer of TRP2180–188-specific T cellsnto C57BL/6 mice reduced the number of experimentally induced16 lung metastases [26]. Interestingly, TRP2180–188 can bind bothuman HLA-A*0201 and murine MHC class I molecule H2-Kb mak-

ng it an attractive model antigen that is very relevant to humanse.

TRP2180–188 was combined with TLR ligands in a wide variety ofelanoma immunotherapeutic strategies [27–30]. Development of

fficient methods for quantitative analysis of TRP2180–188 and TLRigands in a given vaccine is an essential requirement for betterharacterization and optimization of different aspects of the vac-ine formulation, e.g., encapsulation efficiency, loading and releaseattern. We have recently developed a quick, sensitive and reliableC–MS-based method for the quantification of lipid A analoguesn PLGA-NP. Such method had overcome all the problems asso-iated with HPLC analysis of lipid A compounds (e.g., poor UVbsorption and the need for pre-column derivatization) [12]. Inhe current study, we described another LC–MS-based method forhe quantification of TRP2180–188 encapsulated in PLGA-NP (Fig. 1nd Table 1). Conventional HPLC analysis of TRP2180–188 is feasi-le and not as problematic as that of lipid A compounds. However,C–MS is preferred due to quick analysis time and high sensitiv-ty.

Our results showed that immunization of normal mice withLGA-NP encapsulating 10 �g of TRP2180–188 with or without 7-acylipid A (TRP2/7-acyl lipid A-NP and TRP2-NP, respectively) activatedobust TRP2-specific CD8+ T cell responses in the draining lymphodes and spleen of immunized mice and break self-tolerance toRP2 peptide. Our next challenge was to evaluate the efficacy of ouraccination strategy in tumor-bearing mice. As described in Section, therapeutic immunization started 3 days after s.c. inoculation of05 B16 melanoma cells. Although the tumors were not palpablet this stage (day 3), previous studies have shown that 24 h after.c. tumor implantation, B16 cells were clearly visible at the site of

he mice immunized with either TRP2-NP or TRP2/7-acyl lipid A-NPs evidenced by decreased tumor area (Fig. 3A) and weight (Fig. 3B),ompared to the control group (Empty-NP immunized mice). Due

(2008) 5046–5057

o large intra-group variability, we could not observe significantifferences in tumor size/weights between the two test groupsTRP2-NP and TRP2/7-acyl lipid A-NP). However, there were severallues demonstrating the superior therapeutic effect of TRP2/7-acylipid A-NP over TRP2-NP: (1) the average tumor size of animalsmmunized with TRP2/7-acyl lipid A-NP were almost half the aver-ge tumor size obtained in TRP2-NP immunized animals at all timesested (Fig. 3A); (2) none of the mice in the TRP2/7-acyl lipid A NPmmunized group had reached the morbid state, where 11% of theRP2-NP group had to be euthanized before the end of the study;3) more importantly, 85% of animals immunized with TRP2/7-acylipid A-NP had controlled tumor growth compared to 40% in theRP2-NP immunized group (Fig. 3C).

Data obtained from ex vivo analysis of antigen-specific CD8+ Tell activation in lymph nodes and spleens of the tumor bearingice (Fig. 4) was consistent to what we previously observed in

ealthy mice study (Fig. 2). For instance, immunization with eitherRP2-NP or TRP2/7-acyl lipid A-NP overcame self-tolerance mech-nisms and activated robust TRP2-specific CD8+ T cell responses.n the TRP2-NP immunized group, higher number of the activatedD8+ T cells were found in the draining lymph nodes (Fig. 4A),ompared to spleens (Fig. 4B and C). The opposite pattern wasbserved in the TRP2/7-acyl lipid A-NP immunized group, wherectivated CD8+ T cells were found in higher numbers in the spleensFig. 4B and C), compared to lymph node (Fig. 4A). This observation

ay imply that activated CD8+ T cells in the TRP2/7-acyl lipid A-P immunized group have acquired better migratory capacity andere able to leave the lymph node and enter the spleen through

he blood stream. Such T cells may also have better accessibilityo the tumor site. Finally, we have noticed that s.c. immuniza-ion with PLGA-NP co-encapsulating antigens and 7-acyl lipid Always induced strong inflammatory response at the site of injec-ion and in the draining lymph nodes (unpublished observation).n the ELISPOT assay, this inflammation could cause relatively highackground readings in the wells that have no stimulation or welltimulated with the irrelevant CD8 peptide (Fig. 2C and Fig. 4B),robably due to the non-specific IFN-� secretion by innate immuneells in those wells. This background was eliminated when pureD8+ T cells were plated (Fig. 2D and Fig. 4C).

B16 melanoma exhibit severe impairment in multiple compo-ents of MHC class I antigen processing machinery, including: theeptide transporter associated with antigen processing (TAP), theroteasome subunits LMP2, LMP7, and LMP10, PA28-� and -�, andhe chaperone tapasin [18]. Down regulations or loss of expressionnd/or functions of those components result in the reduction oross of MHC class I surface expression, which often leads to immunescape of the tumor cells. Interestingly, earlier studies have shownhat all these defects could be corrected by the administration ofFN-�, which induces the expression of multiple components of the

HC class I antigen processing machinery, and ultimately enhanceshe surface expression of MHC class I [18]. In the present study,he induction of IFN-� secretion by host immune cells may havereat impact on the success of our vaccination strategy. The elevatedevel of IFN-� secretion (either by antigen-specific CD8+ T cells ornnate cells) could up-regulate MHC class I surface expression on16 tumor cells, resulting in immune recognition and increased

ysis of tumor cells by CTLs.It is worth mentioning that no anti-tumor therapeutic effect was

bserved in tumor-bearing mice immunized with 7-acyl lipid A-NPwithout antigen) (data not shown). Furthermore, immunization

ctivation of CD8+ T cell responses in either normal or tumor bear-ng mice as measured by IFN-� ELISPOT assay (data not shown).alf of the mice in group receiving 7-acyl lipid A-NP (4 out of) had to be euthanized before the end point of the study due

Fig. 5. Assessment of the level of proinflammatory cytokines and immuno-suppressive factors in tumor microenvironment. C57Bl/6 mice were challenged with melanomac sacrifis superE ls. Eacf TRP2-

tit�ATiioai

Daol2sTtTs

ells and vaccinated as described in the text and in Fig. 3. At day 21, animals wereuspension, which were counted and filtered through 70 �m cell strainers. TumorLISA. Results are presented as amount of cytokine (pg/mL) per 20 million tumor celrom Empty-NP immunized group (P < 0.05). **Indicates significant difference from

o big tumor size and/or ulceration of the tumor. In the exper-ment described in Fig. 5, we observed up to 2-fold increase inhe level of pro-inflammtory cytokines (e.g., IL-6, IL-12 and TNF-) in the tumors isolated from mice immunized with 7-acyl lipid-NP compared to Empty-NP immunized mice (data is not shown).his observation may imply the ability of nanoparticles contain-ng 7-acyl lipid A to induce DC maturation and activation (evenn the absence of antigen). However, the lack of therapeutic effectbserved in this group highlights the importance of activatingntigen-specific CD8+ T cells to provide the full-scale anti-tumor

The capability of TRP2/7-acyl lipid A-NP vaccine strategy toreak self-tolerance and to induce superior anti-tumor effect coulde further explained through numerous mechanisms. Co-deliveryf TLR ligand (7-acyl lipid A) along with TRP2180–188 to the same

ced; tumors were isolated and crushed between two slides to form uniform cellnatants were analyzed for the level of TNF-�, IL-12, IFN-�, IL-2, IL-6 and VEGF byh bar represents the mean of triplicate wells ± S.D. *Indicates significant differenceNP immunized group (P < 0.05).

C poulation provides the three signals required for optimum CTLctivation. DC stimulated with TLR ligand increase the expressionf peptide/MHC I complex on the cell surface (signal 1), upregu-ate costimulatory molecules, e.g., CD40, CD80 and CD86 (signal), and secrete various cytokines, e.g., IL-12 (signal 3). The threeignals combined lead to enhanced activation and proliferation ofRP2-specific CD8+ T cell [32]. Another avenue for breaking self-olerance is through the ability of TLR activated DCs to reverse theregulatory (Treg) suppressive effects. In fact, it has been recently

hown that IL-6 secreted by TLR4-activated DCs renders antigen-

5 ine 26

iflotiri

orimit[issiIVii

tTiIIgcbrc

caaCpcnwimoicctP

ncrease in the amount of IL-6 secreted by DCs (relative to solubleorm) [13]. We have also shown that co-delivery of OVA and 7-acylipid A in PLGA-NP to DCs have dramatically enhanced the extentf in vitro primary CD4+ T cell activation by 1000-fold (comparedo soluble formulation) [13]. Vaccine delivery system capable ofnducing IL-6 production by DCs, and activation of primary T cellesponses to this extent, may assure overcoming Treg-mediatedmmunosuppression through involvement of activated T cells.

The immunosuppressive environment around the tumor isne of the major elements of compromised anti-tumor immuneesponses (reviewed in Refs. [36–38]). Several lines of evidencendicate that dominant immunosuppressive cytokines in the tumor

icroenvironment, e.g., IL-10, TGF-�, and VEGF, significantlynhibit DC maturation/activation, leading to preferential activa-ion of Treg cells on the expense of anti-tumor effector T cells39–41]. One of the most remarkable finding of our current studys the ability of our vaccine formulation (TRP2/7-acyl lipid A-NP) tohift the balance at the tumor microenvironment towards immunetimulation, as evidenced by the increase in the level of pro-nflammatory/Th1-biased cytokines (IL-2, IL-6, IL-12, TNF-� andFN-�) and the decrease in the level of the immunosuppressantEGF (Fig. 5). These results imply that our vaccine is able to provide

mmune stimulation and rescue impaired DCs from tumor-inducedmmune suppression.

TRP2/7-acyl lipid A-NP may also activate innate immune cellshrough an indirect way. In fact, recent studies have shown thatLR-activated DCs constitute a source of numerous cytokines thatnduce natural killer (NK) cell activation. In particular, DCs-derivedL-15, IL-12/IL-18 and IFN-�/� could induce NK cell proliferation,FN-� secretion and cytotoxic function, respectively [42–44]. A sin-le vaccine strategy capable of activating both CTL and NK cellsan act as a double-edged sword, capable of targeting and killingoth MHC class I positive and negative tumor cells. One of our nextesearch goals is to directly assess NK cell activation and relativeontribution of NK cells to the anti-tumor effects of this vaccine.

. Conclusion

Our results validated the potential of PLGA-based cancer vac-ines to break self-tolerance against caner antigen, induce potentnd specific anti-tumor T cell responses with no concomitantutoimmunity, and activate IFN-� secretion by antigen-specificD8+ T cells. Remarkably, this vaccine was able to induce thera-eutic anti-tumor effect. Activated TRP2-specific CD8 T cells wereapable of IFN-� secretion at lymph nodes and spleens of the vacci-ated mice. More importantly, co-delivery of cancer antigen alongith 7-acyl lipid A in PLGA-NP have proven to be effective strategy

n inducing immunostimulatory mileu at the tumor microenviron-ent, as evidenced by decreased level of VEGF and elevated levels

f IL-2, IL-6, IL-12, IFN-� and TNF-�. Our results also highlight themmuostimulatory properties of 7-acyl lipid A as a novel adjuvantan potentially serve as a powerful companion to antigens in vac-ine formulation. To our knowledge, this is the first report showinghat co-delivery of a real cancer antigen along with TLR ligand inLGA-NP could induce such effects.

cknowledgments

This project was supported by research grants from Canadian

ndustrial partner for their incessant co-operation and provid-ng the synthetic lipid A analogue used in this study. Thanks

(2008) 5046–5057

re extended to Dr. Damayanthi Yalamati and Dr. Rao KogantyOncothyreon, Inc., Edmonton) for providing outstanding assis-ance and advice during the project.

eferences

[1] Rosenberg SA, Yang JC, Restifo NP. Cancer immunotherapy: moving beyondcurrent vaccines. Nat Med 2004;10(9):909–15.

[2] Zou W. Immunosuppressive networks in the tumour environment and theirtherapeutic relevance. Nat Rev Cancer 2005;5(4):263–74.

[3] Kaufman HL, Disis ML. Immune system versus tumor: shifting the balance infavor of DCs and effective immunity. J Clin Invest 2004;113(5):664–7.

[4] Pashine A, Valiante NM, Ulmer JB. Targeting the innate immune response withimproved vaccine adjuvants. Nat Med 2005;11(4 Suppl):S63–8.

[5] Foged C, Sundblad A, Hovgaard L. Targeting vaccines to dendritic cells. PharmRes 2002;19(3):229–38.

[6] Waeckerle-Men Y, Groettrup M. PLGA microspheres for improved anti-gen delivery to dendritic cells as cellular vaccines. Adv Drug Deliv Rev2005;57(3):475–82.

[7] Kundi M. New hepatitis B vaccine formulated with an improved adjuvant sys-tem. Expert Rev Vaccines 2007;6(2):133–40.

[8] Baldrick P, Richardson D, Elliott G, Wheeler AW. Safety evaluation of monophos-phoryl lipid A (MPL): an immunostimulatory adjuvant. Regul Toxicol Pharmacol2002;35(3):398–413.

[9] Baldridge JR, McGowan P, Evans JT, Cluff C, Mossman S, Johnson D, et al. Takinga toll on human disease: toll-like receptor 4 agonists as vaccine adjuvants andmonotherapeutic agents. Expert Opin Biol Ther 2004;4(7):1129–38.

10] Persing DH, Coler RN, Lacy MJ, Johnson DA, Baldridge JR, Hershberg RM, etal. Taking toll: lipid A mimetics as adjuvants and immunomodulators. TrendsMicrobiol 2002;10(10 Suppl):S32–7.

[11] Takayama K, Ribi E, Cantrell JL. Isolation of a nontoxic lipid A fraction containingtumor regression activity. Cancer Res 1981;41(7):2654–7.

12] Hamdy S, Haddadi A, Somayaji V, Ruan D, Samuel J. Pharmaceutical analysis ofsynthetic lipid A-based vaccine adjuvants in poly (d,l-lactic-co-glycolic acid)nanoparticle formulations. J Pharm Biomed Anal 2007;44(4):914–23.

13] Elamanchili P, Lutsiak CM, Hamdy S, Diwan M, Samuel J. “Pathogen-mimicking”nanoparticles for vaccine delivery to dendritic cells. J Immunother (1997)2007;30(4):378–95.

14] Hamdy S, Elamanchili P, Alshamsan A, Molavi O, Satou T, Samuel J, etal. Enhanced antigen-specific primary CD4+ and CD8+ responses by code-livery of ovalbumin and toll-like receptor ligand monophosphoryl lipidA in poly(d,l-lactic-co-glycolic acid) nanoparticles. J Biomed Mater Res A2007;81(3):652–62.

15] Heit A, Schmitz F, Haas T, Busch DH, Wagner H. Antigen co-encapsulatedwith adjuvants efficiently drive protective T cell immunity. Eur J Immunol2007;37(8):2063–74.

16] Bellone M, Cantarella D, Castiglioni P, Crosti MC, Ronchetti A, Moro M, et al.Relevance of the tumor antigen in the validation of three vaccination strategiesfor melanoma. J Immunol 2000;165(5):2651–6.

17] Ohm JE, Carbone DP. VEGF as a mediator of tumor-associated immunodefi-ciency. Immunol Res 2001;23(2–3):263–72.

18] Seliger B, Wollscheid U, Momburg F, Blankenstein T, Huber C. Characterizationof the major histocompatibility complex class I deficiencies in B16 melanomacells. Cancer Res 2001;61(3):1095–9.

19] Kirkin AF, Dzhandzhugazyan K, Zeuthen J. Melanoma-associated antigens rec-ognized by cytotoxic T lymphocytes. APMIS 1998;106(7):665–79.

20] Khong HT, Rosenberg SA. Pre-existing immunity to tyrosinase-related protein(TRP)-2, a new TRP-2 isoform, and the NY-ESO-1 melanoma antigen in a patientwith a dramatic response to immunotherapy. J Immunol 2002;168(2):951–6.

21] Liu G, Khong HT, Wheeler CJ, Yu JS, Black KL, Ying H, et al. Molecular and func-tional analysis of tyrosinase-related protein (TRP)-2 as a cytotoxic T lymphocytetarget in patients with malignant glioma. J Immunother 2003;26(4):301–12.

22] Rosenberg SA, Yang JC, Schwartzentruber DJ, Hwu P, Marincola FM, TopalianSL, et al. Immunologic and therapeutic evaluation of a synthetic peptidevaccine for the treatment of patients with metastatic melanoma. Nat Med1998;4(3):321–7.

23] Wang J, Saffold S, Cao X, Krauss J, Chen W. Eliciting T cell immunity againstpoorly immunogenic tumors by immunization with dendritic cell–tumorfusion vaccines. J Immunol 1998;161(10):5516–24.

24] Wang RF, Johnston SL, Southwood S, Sette A, Rosenberg SA. Recognition ofan antigenic peptide derived from tyrosinase-related protein-2 by CTL in thecontext of HLA-A31 and -A33. J Immunol 1998;160(2):890–7.

25] van Elsas A, Sutmuller RP, Hurwitz AA, Ziskin J, Villasenor J, Medema JP, etal. Elucidating the autoimmune and antitumor effector mechanisms of a treat-ment based on cytotoxic T lymphocyte antigen-4 blockade in combination witha B16 melanoma vaccine: comparison of prophylaxis and therapy. J Exp Med2001;194(4):481–9.

26] Wang RF, Appella E, Kawakami Y, Kang X, Rosenberg SA. Identification of TRP-2as a human tumor antigen recognized by cytotoxic T lymphocytes. J Exp Med1996;184(6):2207–16.

27] Kochenderfer JN, Chien CD, Simpson JL, Gress RE. Synergism between CpG-containing oligodeoxynucleotides and IL-2 causes dramatic enhancement ofvaccine-elicited CD8+ T cell responses. J Immunol 2006;177(12):8860–73.

ine 26

S. Hamdy et al. / Vacc

28] Mansour M, Pohajdak B, Kast WM, Fuentes-Ortega A, Korets-Smith E, Weir GM,et al. Therapy of established B16-F10 melanoma tumors by a single vaccinationof CTL/T helper peptides in VacciMax. J Transl Med 2007;5:20.

29] Mahnke K, Qian Y, Fondel S, Brueck J, Becker C, Enk AH, et al. Targeting ofantigens to activated dendritic cells in vivo cures metastatic melanoma in mice.Cancer Res 2005;65(15):7007–12.

30] Davila E, Kennedy R, Celis E. Generation of antitumor immunity by cytotoxic Tlymphocyte epitope peptide vaccination, CpG-oligodeoxynucleotide adjuvant,and CTLA-4 blockade. Cancer Res 2003;63(12):3281–8.

31] Bellone M, Iezzi G, Martin-Fontecha A, Rivolta L, Manfredi AA, Protti MP,et al. Rejection of a nonimmunogenic melanoma by vaccination with nat-ural melanoma peptides on engineered antigen-presenting cells. J Immunol1997;158(2):783–9.

32] Celis E. Toll-like receptor ligands energize peptide vaccines through multiplepaths. Cancer Res 2007;67(17):7945–7.

33] Pasare C, Medzhitov R. Toll pathway-dependent blockade of CD4+CD25+ T cell-mediated suppression by dendritic cells. Science 2003;299(5609):1033–6.

34] Fehervari Z, Sakaguchi S. Control of Foxp3+ CD25+CD4+ regulatory cell activa-tion and function by dendritic cells. Int Immunol 2004;16(12):1769–80.

35] Kubo T, Hatton RD, Oliver J, Liu X, Elson CO, Weaver CT, et al. RegulatoryT cell suppression and anergy are differentially regulated by proinflam-matory cytokines produced by TLR-activated dendritic cells. J Immunol2004;173(12):7249–58.

36] Rabinovich GA, Gabrilovich D, Sotomayor EM. Immunosuppressive strategiesthat are mediated by tumor cells. Annu Rev Immunol 2007;25:267–96.

(2008) 5046–5057 5057

37] Gajewski TF, Meng Y, Blank C, Brown I, Kacha A, Kline J, et al. Immune resis-tance orchestrated by the tumor microenvironment. Immunol Rev 2006;213:131–45.

38] Gajewski TF. Failure at the effector phase: immune barriers at the levelof the melanoma tumor microenvironment. Clin Cancer Res 2007;13(18 Pt1):5256–61.

39] Gabrilovich D, Ishida T, Oyama T, Ran S, Kravtsov V, Nadaf S, et al. Vascularendothelial growth factor inhibits the development of dendritic cells and dra-matically affects the differentiation of multiple hematopoietic lineages in vivo.Blood 1998;92(11):4150–66.

40] Mesa C, Fernandez LE. Challenges facing adjuvants for cancer immunotherapy.Immunol Cell Biol 2004;82(6):644–50.

41] Jonuleit H, Schmitt E, Schuler G, Knop J, Enk AH. Induction of interleukin10-producing, nonproliferating CD4(+) T cells with regulatory properties byrepetitive stimulation with allogeneic immature human dendritic cells. J ExpMed 2000;192(9):1213–22.

42] Kamath AT, Sheasby CE, Tough DF. Dendritic cells and NK cells stim-ulate bystander T cell activation in response to TLR agonists throughsecretion of IFN-alpha beta and IFN-gamma. J Immunol 2005;174(2):

43] Zanoni I, Foti M, Ricciardi-Castagnoli P, Granucci F. TLR-dependent activationstimuli associated with Th1 responses confer NK cell stimulatory capacity tomouse dendritic cells. J Immunol 2005;175(1):286–92.

44] Walzer T, Dalod M, Robbins SH, Zitvogel L, Vivier E. Natural-killer cells anddendritic cells: “l’union fait la force”. Blood 2005;106(7):2252–8.

Co-delivery of cancer-associated antigen and Toll-like receptor 4 ligand in PLGA nanoparticles...

Documents

Transcript of Co-delivery of cancer-associated antigen and Toll-like receptor 4 ligand in PLGA nanoparticles...

Informal Hospice Caregiving: The Toll on Quality of Life

Toll Road Financing - PPIAF

Development of Risperidone PLGA Microspheres

Risk Averse Second Best Toll Pricing

CD8 Controls T Cell Cross-Reactivity

Aligned PLGA/HA nanofibrous nanocomposite scaffolds for bone tissue engineering

Toll and Bridge: Risk Assessment Framework update - UBS

CD8 � Endows CD8 with Efficient Coreceptor Function by Coupling T Cell Receptor/CD3 to Raft-associated CD8/p56 lck Complexes

Continuous Effector CD8+ T Cell Production in a Controlled ...

Optimization of in vitro test conditions for PLGA-based drug ...

Efficient Chemotherapy of Rat Glioblastoma Using Doxorubicin-Loaded PLGA Nanoparticles with Different Stabilizers

CD8 T Cell Vaccines and a Cytomegalovirus-Based Vector ...

The Illinois State Toll Highway Authority

Comparison of Surface Functionalization of PLGA Composite ...

Hollow polymeric (PLGA) nano capsules synthesized using ...

Privatization and Regulatory Reform of Toll Motorways in Europe

The Chillicothe Telephone Company Catalog of Toll ... - Horizon

Protein-loaded PLGA-PEG-PLGA microspheres: a tool for cell therapy

Craftsman Tools - LH Toll Free.indd

Chlorambucil Encapsulation into PLGA Nanoparticles and Cytotoxic Effects in Breast Cancer Cell