Enhanced antigen-specific primary CD4+ and CD8+ responses by codelivery of ovalbumin and toll-like...

Enhanced antigen-specific primary CD4+ and CD8+

responses by codelivery of ovalbumin and toll-likereceptor ligand monophosphoryl lipid A inpoly(D,L-lactic-co-glycolic acid) nanoparticles

Samar Hamdy, Praveen Elamanchili, Aws Alshamsan, Ommoleila Molavi, Tadaaki Satou, John SamuelFaculty of Pharmacy and Pharmaceutical Sciences, 3133 Dentistry/Pharmacy Centre, University of Alberta,Edmonton, Alberta, Canada, T6G 2N8

Received 23 March 2006; revised 11 July 2006; accepted 26 July 2006Published online 22 December 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.31019

Abstract: The purpose of this research was to investigatethe use of biodegradable poly(D,L-lactic-co-glycolic acid)nanoparticles (PLGA-NP) as a vaccine delivery system tocodeliver antigen, ovalbumin (OVA) along with mono-phosphoryl lipid A (MPLA) as adjuvant for induction ofpotent CD4þ and CD8þ T cell responses. The primaryCD4þ T responses to OVA/MPLA NP were investigatedusing OVA-specific T cells from DO11.10 transgenic mice.Following adoptive transfer of these cells, mice wereimmunized s.c. by NP formulations. For assessing theCD8þ responses, bone marrow derived dendritic cells(DCs) were pulsed with different OVA formulations, then,cocultured with CD8þ T cells from OT-1 mice. T cell prolif-eration/activation and IFN-g secretion profile have been

examined. Particulate delivery of OVA and MPLA to theDCs lead to markedly increase in in vitro CD8þ T cell Tcell proliferative responses (stimulation index >3000) and>13-folds increase in in vivo clonal expanded CD4þ T cells.The expanded T cells were capable of cytokine secretionand expressed an activation and memory surface pheno-type (CD62Llo, CD11ahi, and CD44hi). Codelivery of anti-gen and MPLA in PLGA-NP offers an effective method forinduction of potent antigen specific CD4þ and CD8þ T cellresponses. � 2006 Wiley Periodicals, Inc. J Biomed MaterRes 81A: 652–662, 2007

Key words: antigen delivery; adjuvant; clonal expansion;activation/memory; vaccination

INTRODUCTION

The ultimate goal of therapeutic vaccination is thestimulation of a specific immune response and theinduction of a long lasting immunologic memory toprotect against subsequent disease. For a successfulvaccination, the vaccine antigen (Ag)s have to bepresented by antigen presenting cells (APCs) thatactivate na€�ve T cells and initiate Ag-specific im-mune responses. Among the APCs, the dendriticcells (DCs) are efficient stimulators of primary im-mune responses and a subsequent establishment ofimmunologic memory.1 A rational way of improvingthe potency and safety of new and already existing

vaccines could, therefore, be to direct vaccines spe-cifically to DCs.

Successful design of therapeutic cancer vaccinesrequires definition of the effective immunologicalmechanisms of cancer rejection. Numerous animalmodels and several clinical studies highlight the im-portant roles of T cells in mediating cancer rejec-tion.2–6 The T cell-mediated arm, or cellular immuneresponse, of the adoptive immune system, is com-prised of both, T helper (Th) lymphocytes (CD4þ)and cytolytic T lymphocytes (CTL) or killer T lym-phocytes (CD8þ). Th cells are responsible for orches-trating and directing an immune response, whereasCTLs are the killer cells that traffic to the sites ofinfection and lyse infected cells. Although CTLs wereinitially considered as primary mediators of antitu-mor activity, more recent studies highlight the signifi-cance of CD4þ Th cells in controlling the antitumorresponses.2,7 Activation of Th responses would beexpected to augment CTL responses, antibody re-sponses that can mediate cytotoxicity against tumor,and provide stimulus to the innate immune mecha-nisms involving natural killer cells and macrophages.

No benefit of any kind will be received either directly orindirectly by the author(s).Correspondence to: J. Samuel; e-mail: jsamuel@pharmacy.

ualberta.caContract grant sponsor: Canadian Institutes for Health

Research (CIHR)

' 2006 Wiley Periodicals, Inc.

Here we are investigating a vaccine deliveryapproach that can target the DC both in vitro andin vivo in order to activate potent Th and CTLresponses. This approach is based on the inclusionof peptides or proteins in biodegradable poly(D,L-lactic-co-glycolic acid) nanoparticles (PLGA-NP).PLGA-NP of appropriate size are efficiently takenup by mouse and human DC, and they are slowlyhydrolyzed so that prolonged Ag presentationensues.8 The simultaneous induction of robust Th1CD4þ and CTL T cell responses is an important aimof vaccination. Remarkably, PLGA-NP-encapsulatedproteins reach the processing pathways for MHCclass I and class II molecules, so that the stimulationof Ag-specific CTL and Th cells occurs.8 Shen et al.have demonstrated that Ag encapsulation into PLGAparticles increased the efficiency of MHC class I anti-gen presentation in both cells of normal competencefor crosspresentation (DC) and those of none (Bcells). This lead to markedly improved T cell stimu-lation in vitro. Compared to free soluble Ag, inclusionof Ag into PLGA particles enhanced class I Ag pre-sentation by greater than twofolds. Consistently, over1000-fold less Ag was required to achieve the samemagnitude of class I Ag presentation.9 In addition,class I Ag presentation mediated by PLGA particlespersisted much longer (>4 days) than soluble Ag orAg on nondegradable beads indicating that thismethod of Ag delivery could significantly improveand prolong T-cell stimulation in vitro and in vivo.9

Codelivery of Ag and immunostimulatory mole-cules (adjuvant), especially ligands for toll like recep-tors (TLR) such as monophosphoryl lipid A (MPLA),would result in concurrent Ag processing and pre-sentation as well as signaling of TLR pathway lead-ing to generation of mature and activated DCs capa-ble of inducing primary T cell responses.10–12 Previ-ous study in our lab showed that PLGA delivery ofovalbumin (OVA) induced an increase in the pri-mary CD4þ T cell proliferative response in vitro by500–1000 folds, and incorporation of MPLA in thenanoparticle formulation further increased the T cellactivation potential of DCs and furthermore reducedthe Ag dose necessary for efficient T cell priming(Elamanchili et al., manuscript in preparation). How-ever, there has been no information available on thein vivo CD4þ primary T cell responses associatedwith particulate delivery of OVA and MPLA coen-capsulated in PLGA-NP. Simultaneous induction ofAg-specific CD4þ as well as CD8þ T cell responses isan important aim of vaccination, so we haveextended our studies to characterize the in vivo pri-mary CD4þ T cell immune responses following NPvaccination, as well as the in vitro primary CD8þ Tcell activation, quantitatively through determiningthe increase in the percentage of Ag-specific T cellsand functionally by detecting the increase in expres-

sion of activation markers and cytokine secretion.Our results show that particulate delivery of OVAand MPLA to the DCs lead to markedly increase inin vitro CD8þ T cell proliferative responses (stimula-tion index (S.I) >3000) and >13-fold increase inin vivo clonal expanded CD4þ T cells. The expandedT cells were capable of cytokine secretion andexpressed an activation and memory surface pheno-type (CD62Llo, CD11ahi, and CD44hi).

Augmentation of antitumor immune response canbe achieved through in vivo vaccination or adoptivetransfer therapy (ATC). ATC involves the transfer ofex vivo expanded effector cells as a means of aug-menting the antitumor immune response.13 Tumor-eradicating therapy in murine models suggests thata frequency of Ag-specific T cells of at least 1–10%of CD8þ T cells is required for eradication of estab-lished tumors.13 In patients, this translates to a doseof 2–20 � 109 cells. The nonspecific expansion meth-ods (cytokines, TCR, and costimulatory moleculetriggering) have been used to successfully expandunselected peripheral blood mononuclear cells to>1010 cells in vitro over a period of 2–4 weeks.14,15 Inthis case, however, the frequency of tumor-reactiveT cells in the final infused product is often notknown. Besides, these methods are time and laborintensive, requires infrastructure support to cultivateand qualify T cell products and can be prohibitivelyexpensive in its current experimental phase.13 Partic-ulate delivery of Ag plus MPLA to DCs could be anoptimum strategy to generate a population of T cellsof desired magnitude with defined phenotypic andfunctional properties in much shorter time. Suchapproach could reduce many of these cost-related,and logistical issues.

MATERIALS AND METHODS

Reagents

MPLA, MW, 1955.5 Da was kindly provided by Biomira(Edmonton, AB, Canada). Chicken OVA protein (Grade-V)and polyvinyl alcohol (PVA), MW 21–50,000 Da, FreundAdjuvant Complete (CFA) were obtained from SigmaAldrich (Oakville, ON, Canada). PLGA copolymer,monomer ratio 50:50, MW, 7000 Da, was purchased fromAbsorbable Polymers International, (Pelham, AL). Re-combinant murine granulocyte monocyte colony stimulat-ing factor (GM-CSF) was purchased from Peprotech(Rocky Hill, NJ). Murine CD4 and CD8 isolation kits werepurchased from StemCell Technologies (Vancouver, BC,Canada). Murine IFN-g ELISA kit was purchased fromE-Bioscience (San Diego, CA). RPMI-1640, L-glutamine,and gentamicin were purchased from Gibco-BRL (Bur-lington, ON, Canada). Fetal calf serum (FCS) was obtainedfrom Hyclone Laboratories (Logan, UT). Micro bicincho-

CODELIVERY OF OVA AND MPLA IN PLGA-NP 653

Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

ninic acidTM (BCA) Protein Assay kit was purchased fromBioLynx (Brockville, ON, Canada).

Preparation and characterization of biodegradablenanoparticles using PLGA copolymer

PLGA nanoparticles containing OVA protein with orwithout MPLA were prepared as water/oil/water doubleemulsion formulation by the solvent evaporation method.16

Briefly, OVA protein (either 10 or 25 mg in 50 mL PBS) wasemulsified with PLGA solution in chloroform (300 mL, 50%w/v) using a microtip sonicator (Heat systems, Farming-dale, NY). For preparation of nanoparticle containingMPLA, 200 mg of MPLA in 100 mL of 1:4 methanol–chloro-form mixture was added to the polymer–chloroform solu-tion. The resulting primary emulsion (water/oil) was fur-ther emulsified in 2 mL of PVA solution (9% w/v PVA inPBS) by sonication for 45 s at level 4. The secondary emul-sion was added drop-wise into 8 mL of stirring PVA solu-tion. Nanoparticles were collected after 3 h of stirring bycentrifugation of the emulsion at 40,000 � g for 10 min at48C. The nanoparticles were washed twice with cold deion-ized water and lyophilized. The mean size diameter of thenanoparticles was determined by dynamic light scatteringtechnique using a Zetasizer 3000 (Malvern, UK).

OVA content in the nanoparticles was determined by BCAprotein assay. Briefly, 5 mg of nanoparticles was redispersedin 3.0 mL of 0.1M NaOH containing 5% (w/v) SDS. The mix-ture was incubated overnight in an orbital shaker. The super-natant was then collected by centrifugation at 15,000 � g for10 min. Finally, a BCA protein assay was used to determinethe protein concentration in the supernatant.

Quantification of MPLA in nanoparticles was done by anewly developed liquid chromatography/mass spectrome-try (LC-MS) method (Hamdy et al., manuscript in prepara-tion). Since MPLA is not soluble in acetonitrile unlikePLGA, extraction was done by first dispersing 10 mg ofnanoparticles in 400 mL of Acetonitrile, followed by centrif-ugation at 15,000 � g for 15 min. The supernatant wasremoved. The residue was then extracted by adding500 mL of 1:4 methanol–chloroform mixture. The samplewas centrifuged at 15,000 � g for 15 min and the superna-tant was assayed for MPLA by LC/MS. Development andvalidation of the LC-MS method for MPLA quantification,will be described in detail by Hamdy et al. (manuscript inpreparation).

Based on the measured protein (OVA) and MPLA con-centration, the encapsulation efficiency was calculated asthe amount of encapsulated OVA or MPLA relative to thetotal OVA or MPLA used during nanoparticle preparation,respectively. The loading was calculated from the weightof the nanoparticles and the amount of OVA or MPLAincorporated.

Mice

Balb/c mice were bred and maintained at the Universityof Alberta’s Health Science’s Laboratory Animal Facility.C57Bl/6 mice and both DO11.10 and OT-1 TCR transgenicmice were purchased from the Jackson Laboratory (Bar

Harbor, ME). All experiments were performed in accord-ance to the University of Alberta guidelines for the careand use of laboratory animals. All experiments were per-formed using 6–12-week-old male mice.

Preparation of murine bone marrow derived DCs

DC primary cultures were generated from murine bonemarrow precursors from femurs of C57BL/6 mice in com-plete media in presence of GM-CSF as described earlier.17

Briefly, femurs were removed and purified from the sur-rounding muscle tissue. Thereafter intact bones were left in70% ethanol for 2 min for disinfection and washed withPBS. Then both ends were cut with scissors and the marrowflushed with PBS using a syringe with a 0.45-mm diameterneedle. After one wash in PBS, about 1–1.5 � 107 leukocyteswere obtained per femur. Cell culture medium was RPMI-1640 supplemented with gentamicin (80 mg/mL), L-gluta-mine (2 mM), and 10% heat-inactivated FCS. At day 0, bonemarrow leukocytes were seeded at 2 � 106 per 100 mm dishin 10 mL complete medium containing 20 ng/mL GM-CSF.At day 3, another 10 mL complete medium containing20 ng/mL GM-CSF were added to the plates. At day 6, halfof the culture supernatant was collected, centrifuged, andthe cell pellet resuspended in 10 mL fresh medium contain-ing 20 ng/mL GM-CSF, and given back into the originalplate. At day 7 cells can be used already. The purity of theDC population on day 7 was found to be between 70 and75% based on the expression of CD11c on the semiadherentand nonadherent cell populations.

Flow cytometric analysis

For CD4þ and CD8þ T cell activation studies, 2.5 � 105

cells were suspended in FACS buffer (PBS with 5% FCS and0.09% sodium azide) and incubated with anti-mouse CD16/CD32 mAb to block Fc receptors and then stained appropri-ate fluorescence conjugated antibodies. Anti-mouse CD16/CD32 mAb (2.4G2) and phycoerythrin (PE)-Cy5 conjugatedanti-mouse CD8 mAb (53-6.7) were purchased from BD Bio-sciences (Mississauga, ON, Canada). FITC conjugated antimouse CD90.2 mAb (53-2.1) and PE conjugated anti-mouseCD4 (GK1.5), CD11a (M17/4), CD25 (PC61.5), CD44 (1M7),CD62L (MEL-14), and CD69 (H1.2F3) mAbs and their re-spective isotype controls, were purchased from E-Bioscience(San Diego, CA). Clonotype specific DO11.10 TCR trans-genic mAb (KJ1-26) and its isotype control were purchasedfrom Caltag Laboratories (Burlingame, CA). All sampleswere finally acquired on a Becton-Dickinson Facsort and an-alyzed by Cell-Quest software.

CD4+ T cell adoptive transfer and immunization

CD4þ T cells from the spleens of DO11.10 mice wereisolated by negative selection using the EasySep1 mouseCD4 isolation kit (StemCell Technologies, Vancouver, BC,Canada), according to the manufacturer’s instructions. Thepurity of the isolated CD4þ T cells was >95% and greaterthan 85% of the CD4þ T cells stained positive for the

654 HAMDY ET AL.


clonotype specific KJ1-26 mAb. BALB/c mice were adop-tively transferred with 3.5 � 106 DO11.10 CD4þ T cells.One day later; groups of mice were immunized i.p. with20 mg OVA, either soluble or encapsulated in PLGA-NP,with or without MPLA. The dosage of soluble MPLA (inthe Sol OVA/MPLA immunized group) was 2 mg /mouse(this corresponds to the theoretical maximum amount ofMPLA that could be encapsulated in 1 mg nanoparticlesbased on 100% encapsulation efficiency). However, theactual amount loaded of MPLA in the nanoparticles was0.1% (w/w). Therefore, in the OVA/MPLA NP immunizedgroup, each mouse received 1 mg of MPLA entrapped in1 mg PLGA-NP. After 72 h, the spleens of the vaccinatedmice were harvested and single cell suspensions preparedand blocked with CD16/32 mAb. For antigen specific clo-nal expansion studies the spleenocytes were stained withCD4 and KJ1-26 mAbs. For T cell memory responses, thespleenocytes were stained with CD90, KJ1-26 and eitherCD11a, CD44, or CD62L mAbs.

In other set of experiments, BALB/c mice were adop-tively transferred with 3.5 � 106 DO11.10 TCR transgenicCD4þ T cells. One day later; groups of mice were immu-nized s.c. in the right flank region with 50 mg OVA encap-sulated in 1 mg PLGA-NP, with or without MPLA. Theapproximate dosage of MPLA was 1 mg /mouse. Negativecontrol mice received PBS. Positive control mice received200 mg OVA emulsified in CFA. To characterize the anti-gen-specific clonal expansion, the draining lymph node ofthe vaccinated mice were harvested at 3, 5, and 7 days af-ter vaccination and stained with CD4 and KJ1-26 mAbs.For T cell memory responses, 5 days after vaccinations, thelymphocytes were stained with CD90, KJ1-26 and eitherCD11a, CD44, or CD62L mAbs.

OT-1 CD8+ T cell isolation, in vitro T cellactivation/proliferation, and cytokine analysis

Day 7 bone marrow derived DCs were incubated with1 mg/mL OVA (test antigen), either soluble or encapsulatedin PLGA-NP, both in the presence or absence of MPLA.Negative controls includes untreated (unpulsed) DCs, DCsincubated with either empty NP, MPLA NP, or nanopar-toicles containing MUC1 lipopeptide as an irrelevant anti-gen, with or without MPLA (MUC1/MPLA NP and MUC1NP, respectively). The concentration of MPLA in the solublegroup (Sol OVA/MPLA) was 0.1 mg/mL (this correspondsto the theoretical maximum amount of MPLA that could beencapsulated in 1 mg nanoparticles based on 100% encapsu-lation efficiency). However, for the nanoparticles treatedgroups the approximate concentration of MPLA was0.05 mg/mL (based on 0.1% w/w loading in the PLGA-NP).After 12 h, Ag treated or untreated DCs were harvested,irradiated, washed, and cocultured with CD8þ T cells iso-lated from the spleens of OT-1 mice (the purity of the iso-lated CD8þ T cells was >95%). T cell proliferation wasassessed after adding 3H-thymidine during the last 18 h of a60 h coculture. Stimulation index (SI) for any group was cal-culated as the ratio of counts per minute (cpm) of T cellcocultred with DCs pulsed with any formulation (testgroup)/cpm of T cells cocultured with unpulsed DCs (con-trol group). The expression of CD11a, CD25, CD44, CD62L,

and CD69 was analyzed after 90 h by flow cytometry. Super-natants from Ag treated DCs- OT-1 CD8þ T cell cocultrewere collected over period of 24–96 h and analyzed for IFN-g production by ELISA kit in a 96-well microplate using amicroplate reader (Powerwave with KC Junior software;Bio-Tek, Winooski, VT) at OD of 450 nm with reference setat 570 nm according to the manufacturer’s directions. Allsamples were analyzed in duplicates. The minimum levelfor IFN-g detection was: 15 pg/mL).

RESULTS

Characterization of PLGA nanoparticles

The mean hydrodynamic diameter of the nanopar-ticles ranged between 350 and 450 nm with a poly-dispersity below 0.1. Based on BCA assay, the encap-sulation efficiency for OVA was 14% and the loadingwas either 2 or 5% (w/w) of OVA entrapped perdry weight of nanoparticles. The formulation withlower OVA content (2% w/w), has been used inmost of our studies. However, for the kinetic studyof clonal expansion and activation of Ag-specificCD4þ T cells (Fig. 2), the other formulation withhigher OVA content (5% w/w) has been used. Basedon LC-MS analysis, the encapsulation efficiency ofMPLA was 38% and the loading was 0.1% (w/w) ofMPLA entrapped per dry weight of nanoparticles.

Clonal expansion and activation of Ag-specificCD4+/ KJ1-26+ cells after i.p. vaccination of OVA(soluble versus encapsulated in PLGA-NP)

For studying Th responses, a transgenic mousemodel (DO11.10; T cell receptor specific for ovalbuminpeptide, OVA323–339) was used. This model as previ-ously described by Kearney et al.18 takes advantage ofthe ability to detect small numbers of adoptively trans-ferred CD4þ cells from DO11.10 TCR transgenic miceusing a clonotype-specific monoclonal antibody KJ1-26,that specifically recognizes this transgenic TCR. Adop-tive transfer of OVA-specific T cells from these trans-genic mice to Balb/c recipient allows examination ofthe qualitative (e.g., activation markers, cytokine pro-file) and quantitative (T cell clonal expansion) aspectsof the Th responses induced after vaccination with thenanoparticle formulations of OVA. Our results[Fig. 1(A)] show that, after i.p. vaccination, the nano-particulate formulations were far more potent than thesoluble formulations for inducing Th responses as evi-denced by about 12- and 17-fold increase in the num-ber of antigen-specific CD4þ/ KJ1-26þ cells/spleen inOVA NP (280 � 103) and OVA/MPLA NP (540 � 103)immunized mice, respectively. Only a trace populationof CD4þ/ KJ1-26þ was observed in our control groups,PBS and Empty NP (28 � 103 and 11 �103, respec-



tively) and groups immunized with soluble formula-tions, Sol OVA and Sol OVA/MPLA (19 � 103 and 14 �103, respectively). In a separate set of experiment moretreatment groups (soluble MPLA and MPLA in NP)have been examined and show the same back groundlevel of CD4þ/ KJ1-26þ population as in the PBS controlgroup (data not shown).

Following encounter with Ag, T lymphocytesmodulate their surface expression of various activa-tion markers, facilitating their migration in and outof immune tissue and conferring specific effectorfunctions on them. So by using the same adoptivetransfer model, we determined whether Ag-stimu-lated CD4þ T cells would display an effector/mem-ory phenotype following NP vaccination. Examina-tion of the spleenocytes from recipient mice 3 days

after i.p. immunization with PBS showed that theDO11.10 donor T cells displayed a CD62Lhi, CD11alo,and CD44lo ‘‘naı̈ve’’ surface phenotype. The DO11.10T cells in the mice immunized with OVA NP orOVA/MPLA NP possessed a CD62Llo, CD11ahi, andCD44hi surface phenotype characteristic of effector/memory T cells [Fig. 1(B)].

Kinetics of clonal expansion and activationof Ag-specific CD4+/KJ1-26+ cells in thedraining lymph nodes after s.c. vaccinationof OVA NP ± MPLA

To further characterize the adjuvant function ofMPLA and the effect of particulate delivery of OVA

Figure 1. Vaccination with nanoparticles containing antigen and MPLA induces antigen specific T cell clonal expansionand activation in vivo. BALB/c mice were adoptively transferred with 3.5 � 106 DO11.10 CD4þ T cells. One day later;groups of mice were immunized i.p. with 20 mg OVA, either soluble or encapsulated in PLGA-NP, with or without MPLA.Soluble vaccinations included: PBS, Sol OVA, and Sol OVA/MPLA. Nanoparticle vaccinations included 1 mg of NP in100 mL PBS. The nanoparticle formulations used are; empty NP, OVA NP, and OVA/MPLA NP. The approximate dosagesof soluble MPLA and MPLA encapsulated in the nanoparticles were 2 mg, and 1 mg/mice, respectively. As described inMaterials and Methods, 3 days following the vaccination, the absolute numbers of CD4þ/KJ1-26þ T cells in the spleen (A),and activation markers on the expanded T cells (B) were determined. For T cell activation/memory responses, spleenocytesfrom mice vaccinated with PBS (shaded histogram), OVA NP (thin line), and OVA/MPLA NP (thick line) were stained withCD90, KJ1-26 and either CD11a, CD44, or CD62L mAbs. Histograms of the activation markers represent CD90þ/KJ1-26þ

gated population. Data shown are representative of three independent experiments (each with three mice per group).

656 HAMDY ET AL.


on the primary Th responses, 3.5 � 106 DO11.10CD4þ T cells were adoptively transferred to Balb/cmice. Twenty-four hours later, mice were immu-nized s.c. with OVA NP in the presence or absenceof MPLA. The percentages of accumulated CD4þ/KJ1-26þ cells in the draining lymph nodes wasassessed after 3, 5, and 7 days postimmunization[Fig. 2(A)]. The injection of OVA NP alone caused9–10-folds increase in the percentage of OVA specificCD4þ/KJ1-26þ cells relative to PBS immunizedgroup at day 3 (2.9%) and start to decline at day 5(1.54%) and finally reached 1.27% at day 7. On theother hand, the injection of OVA/MPLA NP showedits peak at day 5, where 13-fold increase in the per-centage of CD4þ/KJ1-26þ cells was observed(4.32%). At day 7, three-fold reduction of the per-centage and absolute numbers of DO11.10 cells wasdetected (1.41%). Interestingly, similar pattern wasobserved for the OVA/CFA positive control group,while the DO11.10 cells peaked at day 5 (12.70%)and contracted at day 7 (3.96%) with three-foldreduction in their percentage. Taking into accountthat the concentration of OVA in the OVA/CFAgroup was four-fold higher than the OVA encapsu-lated in the PLGA-NP (200 and 50 mg, respectively),we can conclude that, at least in terms of CD4þ Tcell expansion, coencapsulation of protein andMPLA in PLGA-NP can induce a response which iscomparable to that induced by protein emulsified inCFA, one of the strongest adjuvant combinationavailable. These results also demonstrate that deliv-ery of the Ag, its acquisition by APCs, and its pre-sentation to naı̈ve CD4þ T cells all had occurredwithin this 7-day period. Furthermore, the presenceof MPLA during T cell priming resulted in increasein Ag- specific T cell numbers at the peak of clonalexpansion and after clonal contraction had occurred.

We have also determined whether these Ag-stimu-lated CD4þ T cells in the draining lymph node woulddisplay an effector/memory phenotype. Lymphocytesfrom recipient mice 5 days after s.c. immunizationwith OVA NP or OVA/MPLA NP showed that theDO11.10 donor T cells express CD62Llo, CD11ahi, andCD44hi surface phenotype [Fig. 2(B)], similar to thephenotype observed in the spleenocytes of the i.p.vaccinated mice [Fig. 1(B)]. Such a phenotype is char-acteristic to the effector/memory T cells.

In vitro OT-1 CD8+ T cell proliferation

A recent study in our lab showed that PLGAdelivery of OVA to the DCs induced increase in theprimary CD4þ T cell proliferative response in vitroby 500–1000 folds. Incorporation of MPLA in thenanoparticle formulation further increased the T cellactivation potential of DCs and furthermore reduced

the Ag dose necessary for efficient T cell priming(Elamanchili et al., manuscript in preparation). Inthe current report, we extended those studies toinvestigate the effect of nanoparticulate delivery ofOVA with or without MPLA to DCs on the induc-tion of in vitro primary CD8þ T cell responses. Forassessing the CD8þ responses, another transgenicmouse model was used, (OT-I). OT-I T cells (H-2b)recognize the OVA peptide257–264 (SIINFEKL) inassociation with MHC class I. As illustrated in Fig-ure 3, DCs incubated with soluble OVA protein1 mg/mL, either alone (Sol OVA) or together withMPLA (Sol OVA/MPLA) did not result in any de-tectable priming of OVA specific T cells. However,delivery of OVA protein encapsulated in PLGA-NP(OVA NP) to DCs efficiently primed naı̈ve T cells asevidenced by stimulation index (SI) of 600. Codeliv-ery of MPLA along with OVA protein in the PLGA-NP (OVA/MPLA NP) markedly increased the extentof T cell stimulation (SI > 3000). This markedlyimproved CD8þ T- cell in accordance on what weobserved earlier on CD4þ T cell activation (Elaman-chili et al., manuscript in preparation). These resultsare also consistent with the findings of Datta et al.,19

which suggest that certain TLR ligands promoteshuttling of Ag from early in the endosomal path-way to the cystosol for processing by the MHCIpathway. The molecular mechanisms involved inthis Ag escape from endosome to cytosol remain elu-sive and require further investigation. Shen et al.9

have recently shown that, the MHC class I presenta-tion of PLGA-encapsulated OVA stimulated T cellinterleukin-2 secretion at 1000-fold lower concentra-tion than soluble Ag and 10-fold lower than Ag-coated latex beads. Taken together, these resultsemphasize that PLGA-NP-encapsulated proteinsreach the processing pathways for MHC class I andclass II molecules, so that the stimulation of Ag-spe-cific CD4þ and CD8þ T cell responses occur. Also,incorporation of TLR ligands like MPLA in the nano-particle formulation could directly activate the DCsto cross-present Ag to CD8þ T cells in the absence ofCD4þ T cell help.

In vitro OT-1 CD8+ T cell activationand cytokine secretion

As shown in Figure 3, particulate delivery of OVAtogether with MPLA to the DCs generated a largepopulation of OVA specific CD8þ T cells. Then, wefocused our attention on the functionality of the Tcells generated by this method. As shown in Figure4, CD8þT cells harvested from spleen of OT-1 micedisplayed the canonical naı̈ve phenotype of CD8þ Tcells: CD11aloCD25�CD69�CD62LhiCD44lo.20 Uponcoculture with OVA/MPLA NP pulsed DCs, these



Figure 2. In vivo kinetics of clonal expansion of antigen specific T cells after particulate delivery of OVA and MPLA.BALB/c mice were adoptively transferred with 3.5 � 106 DO11.10 CD4þ T cells. One day later; groups of mice wereimmunized s.c. in the right flank region with 50 mg OVA encapsulated in PLGA-NP, with or without MPLA (OVA/MPLANP or OVA NP, respectively). The approximate dosages of MPLA was 1 mg/mice. Control mice received PBS. Positivecontrol mice received 200 mg OVA emulsified in CFA. To characterize the antigen specific clonal expansion, the draininglymph node of the vaccinated mice were harvested at 3, 5, and 7 days after vaccination and stained with CD4 and KJ1-26mAbs (A). For T cell activation/memory responses (B), lymphocytes from mice vaccinated with PBS (shaded histogram),OVA NP (thin line), and OVA/MPLA NP (thick line) were harvested 5 days after vaccination and then, stained withCD90, KJ1-26 and either CD11a, CD44, or CD62L mAbs. Histograms of the activation markers represent CD90þ/KJ1-26þ

gated population.


OT-1 T cells exhibited the typical effector cytotoxic Tcells phenotype: CD11ahiCD25þCD69þCD62LloCD44hi.Coculture of CD8þ T cells with DCs pulsed withOVA NP alone (without MPLA) resulted in similarlevel of expression of CD11a, CD44, CD62L as ineffector T cell phenotype, however, lower expression

of CD25 was observed. Also the expression of CD69was similar to that of naı̈ve CD8þ T cells (Fig. 4).The amount of proliferation (Fig. 3) and activation(Fig. 4) correlated also with IFN-g production (Fig.5). OT-1 T cells cocultured with unpulsed DCs didn’tproduce any detectable level of IFN-g at all time

Figure 4. In vitro activation/memory phenotype of expanded OT-1 CD8þ T cells. Day 7 bone marrow derived DCs wereincubated with either OVA NP or OVA/MPLA NP. After 12 h, nonadherent cells were harvested and cocultured withCD8þ T cells from OT-1 mice as described in Materials and Methods. Shaded histograms represent naı̈ve CD8þ T cells iso-lated from OT-1 spleenocytes. Thin lines and thick lines represent CD8þ OT-1 cocultured with DCs pulsed with OVA NPand OVA/MPLA NP, respectively. After 90 h, the cells were stained with CD8, CD90 and either CD11a, CD25, CD44,CD62L, or CD69 mAbs, and analyzed by flow cytometry. Histograms represent CD90þ/CD8þ gated population. Asexplained in the text, the expression pattern of CD11a, CD44, and CD62L was similar for both OVA NP and OVA/MPLANP group (the thin line is overlapping with the thick line). However, the CD69 expression of OVA NP group was similarto that of naı̈ve CD8þ T cells (the thin line is overlapping with the shaded histogram). [Color figure can be viewed in theonline issue, which is available at www.interscience.wiley.com.]

Figure 3. In vitro OT-1 CD8þ T cell proliferation. Day 7 bone marrow derived DCs were incubated with 1 mg/mL OVA(test antigen) either soluble or encapsulated in PLGA-NP, both in the presence or absence of MPLA. Negative controlsincludes untreated (unpulsed) DCs, DCs incubated with either empty NP, MPLA NP, or nanopartoicles containing MUC1lipopeptide as an irrelevant antigen, with or without MPLA (MUC1/MPLA NP and MUC1 NP, respectively). The approx-imate concentration of soluble MPLA and MPLA encapsulated in the nanoparticles were 0.1 and 0.05 mg/mL, respectively.After 12 h, nonadherent cells were harvested and cocultured with CD8þ T cells from OT-1 mice as described in Materialsand Methods. T cell proliferation was assessed after adding 3H-thymidine during the last 18 h of a 60 h coculture. The val-ues are mean of triplicates 6 SE. Stimulation index (SI) for any group was calculated as the ratio of counts per minute(cpm) of T cell cocultred with DCs pulsed with formulations (test group)/cpm of T cells cocultured with unpulsed DCs(control group). Data shown are representative of three independent experiments.



points. OVA NP induced moderate secretion of IFN-g, which appeared after 24 h from coculture and sus-tained at the same level up to 96 h. Codelivery ofOVA and MPLA (OVA/MPLA NP) resulted inmarked increase in IFN-g production, especially after72–96 h from coculture, which is consistent withthe extent of CD8þ T cell proliferation observed inFigure 3.

DISCUSSION

It is becoming increasingly apparent that, for can-cer and chronic viral infections, a strong cellularresponse is required for protective immunity. There-fore, a fundamental understanding of how robustcellular immune responses can be initiated andmaintained following vaccination will facilitate thedevelopment of effective vaccines against these dis-eases. The simultaneous induction of robust Th1 andCTL T cell responses is an important aim of vaccina-tion. Targeting of DCs by PLGA-NP can lead to theenhancement of MHC II-restricted presentation ofdelivered epitopes and to the induction of Th1-polar-ized T cell responses in parallel to an efficient MHCI-restricted presentation of epitopes delivered simul-taneously.8 Another primary focus for the design ofeffective vaccines is the development of effective andsafe adjuvants, which instruct and control the selec-tive induction of the appropriate type of Ag-specificimmune response. In our study, we used MPLAbecause of its established safety and efficacy inhumans, as well as its ability to induce strong cellu-

lar immune responses.21–24 In the current study, wehypothesized that; these ‘‘pathogen-mimicking’’ NPwill be internalized efficiently by DC both in vitroand in vivo, leading to DC maturation/activationand induction of potent T cell responses. A funda-mental feature of T cell-dependent immune re-sponses is the necessity for a very small populationof naı̈ve CD4þ T cells to undergo clonal expansionand activation following encounter with a specificAg. While several previous studies have attemptedto evaluate the influence of PLGA NP and/or MPLAon CD4þ T cell functions, there have not been anyreports that directly monitored their effects on theactivation of Ag-specific CD4þ Th cells in vivo. In thestudies reported here, we have utilized the DO11.10adoptive transfer model to directly monitor theeffects of particulate delivery of OVA together withMPLA on the primary antigen specific CD4þ T cellresponses in vivo. Using that model, evidence ofT cell encounter with the antigen have beenobserved, as reflected by the activated surface phe-notype (increased CD11a and CD44 and decrease inCD62L expression) and the clonal expansion ofOVA-specific CD4þ T cell following either i.p. or s.c.route of immunization.

Although adoptive transfer of TCR transgenic Tcells has already been used for the study of in vivo Tcell activation using protein antigen,25,26 this is thefirst time such a system has been applied to a clini-cally acceptable vaccine delivery system (PLGA).The ability of MPLA to induce rapid activation andclonal expansion of DO11.10 T cells is analogous tothat reported earlier using CFA, lipopolysaccharide,

Figure 5. In vitro IFN-g secretion pattern by the activated OT-1 CD8þ T cells. Day 7 bone marrow derived DCs were ei-ther untreated (unpulsed) (white bars) or incubated with either of OVA NP (gray bars), or OVA/MPLA NP (black bars).After 12 h, nonadherent cells were harvested and cocultured with CD8þ T cells from OT-1 mice as described in Materialsand Methods. White, gray, and black bars represent OT-1 CD8þ T cells cocultured with unpulsed DCs or DCs pulsed witheither OVA NP or OVA/MPLA NP, respectively. IFN-g secretion was detected in the supernatant after 24, 48, 72, and96 h from coculture using routine ELISA technique per the manufacturer’s instruction. The values are mean of triplicates6 SE. Data shown are representative of three independent experiments.

660 HAMDY ET AL.


inflammatory cytokines, DNA microparticles, bacte-rial flagellin as adjuvants in the adoptive transfermodel.27–31 A number of TLR agonists function asimmunological adjuvants.10,32 MPLA might be a par-ticularly attractive candidate as a chemically definedsynthetic adjuvant in vaccine formulations designedto activate T cell mediated immune responses.MPLA has been shown to promote Th1-type antigenspecific responses in particular, and has been usedextensively in clinical trials as a component in pro-phylactic and therapeutic vaccines targeting infec-tious disease, cancer, and allergies. With over 33,000doses administered to date, MPLA has emerged as asafe and effective vaccine adjuvant.23

The ability of MPLA-activated DCs to activate CD4þTcell response and shape a Th1-biased immune responseshas been well described.21,33 However, the role of partic-ulate delivery of MPLA in priming CD8þ T cells has notbeen thoroughly investigated. Codelivery of MPLAalong with OVA protein to bone marrow-derived DCsmarkedly increased the extent of OT-1 CD8þ T cell stim-ulation as measured by proliferative response and IFN-gproduction. The activated OT-1 CD8þ T cells also exhib-ited the typical effector cytotoxic T cells phenotype:CD11ahiCD25þCD69þCD62LloCD44hi.

Stimulation of CD8þ T cells by class I MHC-associ-ated peptides from exogenous Ag requires transportof the Ag to the cytosol of the APC prior to its trans-location to the endoplasmic reticulum for associationwith nascent MHC class I molecules.34 Consequently,agents which augment delivery of exogenous Ag intothe cytoplasm of APC and thereby into the classicalMHC class I route could be effective in induction ofcytotoxic T lymphocyte (CTL) responses.35,36 Com-pared to free soluble Ag, inclusion of antigen intoPLGA-NP increased the amount of protein thatescaped from endosomes into cytoplasm, thereby in-creasing the access of exogenous Ag to the classicMHC class I loading pathway.9 However, the exactmechanisms that lead to MPLA-induced crosspresen-tation are yet to be delineated. MPLA activation likelyleads to multiple effects on DCs that promote cross-presentation. MPLA might upregulate costimulatorymolecules and induce cytokine secretion, both ofwhich play a role in T cell activation. In addition,other TLR ligands such as immunostimulatory se-quence DNA induces components of the Ag-process-ing machinery, including MHC molecules,37 andcould also reroute endocytosed Ag to the cytosol.19

Further investigation of the ability of MPLA to inducecrosspresentation will lead to better understandingand uncover vaccination strategies that elicit im-proved cell-mediated immunity.

The magnitude of the antitumor immune re-sponses has been demonstrated in murine models ofimmunotherapy to be a critical factor in tumor eradi-cation.38 Moreover, the combination of vaccination

and adoptive T-cell transfer led to a more robustantitumor response than the use of each treatmentindividually.39 Particulate delivery of Ag plus MPLAto DCs could be an optimum strategy to generate apopulation of T cells of desired magnitude withdefined phenotypic and functional properties in rela-tively short time. Those expanded T cells could beoptimized to be used in adoptive transfer studiesalong with normal vaccination protocol.

In conclusion, the experiments presented here arethe first to study the ability of combining a clinicallyacceptable vaccine delivery system (PLGA) with clin-ically proven adjuvant (MPLA) to prime Ag-specificT cells at the cellular level. Our results demonstratethat, PLGA-NP represents an effective means todeliver protein Ag along with immunomodulators toDCs for the induction of CD4þ and, most notably,CD8þ Ag-specific T cell responses. In combinationwith their properties of biodegradability and bio-compatibility, PLGA-NP is a promising delivery sys-tem for future vaccines. Our next goals are to deter-mine whether the activated T cells have the potentialto exert protective immunity and to further optimizetherapeutic immunization through stepwise vaccinemodifications in parallel with careful investigation ofinduced immune responses.

References

1. Foged C, Sundblad A, Hovgaard L. Targeting vaccines todendritic cells. Pharm Res 2002;19:229–238.

2. Zeng G. MHC Class II-restricted tumor antigens recognizedby CD4þ T cells: New strategies for cancer vaccine design.J Immunother 2001;24:195–204.

3. Parmiani G. Melanoma antigens and their recognition by Tcells. Keio J Med 2001;50:86–90.

4. Foss FM. Immunologic mechanisms of antitumor activity.Semin Oncol 2002;29(Suppl 7):5–11.

5. Kirk CJ, Hartigan-O’Connor D, Nickoloff BJ, Chamberlain JS,Giedlin M, Aukerman L, Mule JJ. T cell-dependent antitumorimmunity mediated by secondary lymphoid tissue chemo-kine: Augmentation of dendritic cell-based immunotherapy.Cancer Res 2001;61:2062–2070.

6. Panelli MC, Wang E, Monsurro V, Jin P, Zavaglia K, SmithK, Ngalame Y, Marincola FM. Vaccination with T cell-definedantigens. Expert Opin Biol Ther 2004;4:697–707.

7. Velders MP, Markiewicz MA, Eiben GL, Kast WM. CD4þ T cellmatters in tumor immunity. Int Rev Immunol 2003;22:113–140.

8. Waeckerle-Men Y, Groettrup M. PLGA microspheres forimproved antigen delivery to dendritic cells as cellular vac-cines. Adv Drug Deliv Rev 2005;57:475–482.

9. Shen H, Ackerman AL, Cody V, Giodini A, Hinson ER, Cress-well P, Edelson RL, Saltzman WM, Hanlon DJ. Enhanced andprolonged cross-presentation following endosomal escape ofexogenous antigens encapsulated in biodegradable nanopar-ticles. Immunology 2006;117:78–88.

10. Kaisho T, Akira S. Toll-like receptors as adjuvant receptors.Biochim Biophys Acta 2002;1589:1–13.

11. Kaisho T, Akira S. Regulation of dendritic cell functionthrough toll-like receptors. Curr Mol Med 2003;3:759–771.

12. Schwarz K, Storni T, Manolova V, Didierlaurent A, Sirard JC,Rothlisberger P, Bachmann MF. Role of toll-like receptors in



costimulating cytotoxic T cell responses. Eur J Immunol 2003;33:1465–1470.

13. Yee C. Adoptive T cell therapy: Addressing challenges incancer immunotherapy. J Transl Med 2005;3:17.

14. Laport GG, Levine BL, Stadtmauer EA, Schuster SJ, LugerSM, Grupp S, Bunin N, Strobl FJ, Cotte J, Zheng Z, GregsonB, Rivers P, Vonderheide RH, Liebowitz DN, Porter DL, JuneCH. Adoptive transfer of costimulated T cells induces lym-phocytosis in patients with relapsed/refractory non-Hodgkinlymphoma following CD34þ-selected hematopoietic celltransplantation. Blood 2003;102:2004–2013.

15. Maus MV, Thomas AK, Leonard DG, Allman D, Addya K,Schlienger K, Riley JL, June CH. Ex vivo expansion of poly-clonal and antigen-specific cytotoxic T lymphocytes by artifi-cial APCs expressing ligands for the T-cell receptor, CD28and 4-1BB. Nat Biotechnol 2002;20:143–148.

16. Ogawa Y, Yamamoto M, Okada H, Yashiki T, Shimamoto T.A new technique to efficiently entrap leuprolide acetate intomicrocapsules of polylactic acid or copoly(lactic/glycolic)acid. Chem Pharm Bull (Tokyo) 1988;36:1095–1103.

17. Lutz MB, Kukutsch N, Ogilvie AL, Rossner S, Koch F,Romani N, Schuler G. An advanced culture method for gen-erating large quantities of highly pure dendritic cells frommouse bone marrow. J Immunol Methods 1999;223: 77–92.

18. Kearney ER, Pape KA, Loh DY, Jenkins MK. Visualization ofpeptide-specific T cell immunity and peripheral toleranceinduction in vivo. Immunity 1994;1:327–339.

19. Datta SK, Redecke V, Prilliman KR, Takabayashi K, Corr M,Tallant T, DiDonato J, Dziarski R, Akira S, Schoenberger SP,Raz E. A subset of Toll-like receptor ligands induces cross-pre-sentation by bone marrow-derived dendritic cells. J Immunol2003;170:4102–4110.

20. Yang L, Baltimore D. Long-term in vivo provision of antigen-specific T cell immunity by programming hematopoietic stemcells. Proc Natl Acad Sci USA 2005;102:4518–4523.

21. Chong CS, Cao M, Wong WW, Fischer KP, Addison WR,Kwon GS, Tyrrell DL, Samuel J. Enhancement of T helpertype 1 immune responses against hepatitis B virus core antigenby PLGA nanoparticle vaccine delivery. J Controlled Release2005;102:85–99.

22. Ulrich JT, Myers KR. Monophosphoryl lipid A as an adju-vant. Past experiences and new directions. Pharm Biotechnol1995;6:495–524.

23. Evans JT, Cluff CW, Johnson DA, Lacy MJ, Persing DH, Bal-dridge JR. Enhancement of antigen-specific immunity via theTLR4 ligands MPL adjuvant and Ribi. 529. Expert Rev Vac-cines 2003;2:219–229.

24. Elamanchili P, Diwan M, Cao M, Samuel J. Characterizationof poly(D,L-lactic-co-glycolic acid) based nanoparticulate sys-tem for enhanced delivery of antigens to dendritic cells. Vac-cine 2004;22:2406–2412.

25. Pape KA, Kearney ER, Khoruts A, Mondino A, Merica R,Chen ZM, Ingulli E, White J, Johnson JG, Jenkins MK. Use ofadoptive transfer of T-cell-antigen-receptor-transgenic T cell

for the study of T-cell activation in vivo. Immunol Rev 1997;156:67–78.

26. Gudmundsdottir H, Wells AD, Turka LA. Dynamics andrequirements of T cell clonal expansion in vivo at the single-cell level: Effector function is linked to proliferative capacity.J Immunol 1999;162:5212–5223.

27. Pape KA, Khoruts A, Mondino A, Jenkins MK. Inflammatorycytokines enhance the in vivo clonal expansion and differentiationof antigen-activated CD4þ T cells. J Immunol 1997;159:591–598.

28. Merica R, Khoruts A, Pape KA, Reinhardt RL, Jenkins MK.Antigen-experienced CD4 T cells display a reduced capacityfor clonal expansion in vivo that is imposed by factors pres-ent in the immune host. J Immunol 2000;164:4551–4557.

29. Sun J, Dirden-Kramer B, Ito K, Ernst PB, Van Houten N.Antigen-specific T cell activation and proliferation duringoral tolerance induction. J Immunol 1999;162:5868–5875.

30. Creusot RJ, Thomsen LL, van Wely CA, Topley P, Tite JP, ChainBM. Early commitment of adoptively transferred CD4þ T cellsfollowing particle-mediated DNA vaccination: Implications forthe study of immunomodulation. Vaccine 2001;19:1678–1687.

31. McSorley SJ, Ehst BD, Yu Y, Gewirtz AT. Bacterial flagellin isan effective adjuvant for CD4þ T cells in vivo. J Immunol2002;169:3914–3919.

32. Seya T, Akazawa T, Uehori J, Matsumoto M, Azuma I, Toyosh-ima K. Role of toll-like receptors and their adaptors in adjuvantimmunotherapy for cancer. Anticancer Res 2003;23:4369–4376.

33. Newman KD, Samuel J, Kwon G. Ovalbumin peptide encap-sulated in poly(D,L-lactic-co-glycolic acid) microspheres is ca-pable of inducing a T helper type 1 immune response. J Con-trolled Release 1998;54:49–59.

34. Moore MW, Carbone FR, Bevan MJ. Introduction of solubleprotein into the class I pathway of antigen processing andpresentation. Cell 1988;54:777–785.

35. Heeg K, Kuon W, Wagner H. Vaccination of class I major his-tocompatibility complex (MHC)-restricted murine CD8þ cy-totoxic T lymphocytes towards soluble antigens: Immunosti-mulating-ovalbumin complexes enter the class I MHC-re-stricted antigen pathway and allow sensitization against theimmunodominant peptide. Eur J Immunol 1991;21:1523–1527.

36. Tartour E, Ciree A, Haicheur N, Benchetrit F, Fridman WH.Development of non-live vectors and procedures (liposomes,pseudo-viral particles, toxin, beads, adjuvantsellipsis) as toolsfor cancer vaccines. Immunol Lett 2000;74:45–50.

37. Cho HJ, Hayashi T, Datta SK, Takabayashi K, Van Uden JH,Horner A, Corr M, Raz E. IFN-ab promote priming of antigen-specific CD8þ and CD4þ T lymphocytes by immunostimula-tory DNA-based vaccines. J Immunol 2002;168:4907–4913.

38. Greenberg PD. Adoptive T cell therapy of tumors: Mecha-nisms operative in the recognition and elimination of tumorcells. Adv Immunol 1991;49:281–355.

39. Lou Y, Wang G, Lizee G, Kim GJ, Finkelstein SE, Feng C,Restifo NP, Hwu P. Dendritic cells strongly boost the antitu-mor activity of adoptively transferred T cells in vivo. CancerRes 2004;64:6783–6790.

662 HAMDY ET AL.


Enhanced antigen-specific primary CD4+ and CD8+ responses by codelivery of ovalbumin and toll-like...

Documents

Transcript of Enhanced antigen-specific primary CD4+ and CD8+ responses by codelivery of ovalbumin and toll-like...