[CANCER RESEARCH 57, 2865—2869,July 15. 1997]

Advances in Brief

Genetic Immunization for the Melanoma Antigen MART-lfMelan.A UsingRecombinant Adenovirus-transduced Murine Dendritic Cells1

Antom Ribas, Lisa H. Butterfield, Wffliam H. McBride, Syed M. Jilam, Lynne A. Bui, Charles M. Volimer, Roy Lau,Vivian B. Dissette, Billy Hu, Angela Y. Chen, John A. Glaspy, and James S. Economou2

Divisions of Surgical Oncology [A. R.. L H. B., L A. B., C. M. V., R. L. V. B. D.. B. H.. A. Y. C., J. S. El, Experimental Radiation Oncology 1W. H. MI, and HematologyOncology (S. M. J., J. A. G.J, University of California Los Angeles Medical Center, Los Angeles. California 90024

Abstract

Dendritic cells (DCs) are professional antigen-presenting cells thatprocess and present antigenic peptides and are capable of generatingpotent T-cell immunity. A murine tumor model was developed to evaluatemethods of genetic immunization to the human MART-lIMelan-A(MART-l) melanoma antigen. A poorly immunogenic murine fìbrosarcoma line (NFSA) was stably transfected with the MART-i gene. Thistransfected tumor [NFSA(MART1)] grows progressively in C3Hf/Kam/

Sed (H@2k)mice. Partial protection against a challenge with NFSA(MART1) could be achieved with i.m. injections of a MART-i expressionplasmid or with systemic administration of an adenovirus vector expressing MART-l. However, superior protection was achieved when granulocyte macrophage colony-stimulating factorfinterieukin-4-differentiatedmunne DCs transduced with an adenovirus vector expressing MART-iwere used for immunization. Both partial and complete protection couldbe achieved with i.v. administration of MART-i-engineered DCs. Splenocytes from immunized mice contained MUC class I-restricted CTLs specific for MART-i. This predlinical model of genetic immunization supports a therapeutic strategy for human melanoma.

Introduction

Several groups have described the existence of melanoma antigensand suggested a potential role in tumor immunity (1, 2). TheMART-l3 antigen is one of the best characterized and is expressed bymost melanomas (2, 3). Peptides encoded by this gene can be recognized by melanoma-specific CD8-positive CTLs in the context ofMHC molecules. DCs are the most potent APCs known, and it is nowpossible to isolate, expand, and study their function (4, 5). Recently,investigators have demonstrated that specific antitumor responses canbe generated by pulsing tumor peptide antigens onto DCs (6, 7). Thesetumor antigens may be MHC class I synthetic peptides (8) or autologous acid-eluted tumor peptides (7). The opportunity for DCs toexpress the entire tumor antigen polypeptide has a number of potentialadvantages to peptide-based strategies. These include the expression

of multiple and/or undefined peptide epitopes, possibly in the contextof both MHC class I and II molecules, and with any MHC allele. Wehave previously demonstrated that the most efficient way to expressforeign genes in human DCs is with recombinant AdV vectors (9). In

Received 4/7/97; accepted 5/28/97.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

I Supported in part by NIH/National Cancer Institute Grant P01CA5926—04 and the

Monkarsh, Kesselman, and Minor Funds. A. R. was supported by a fellowship from theFondo de Investigacion Sanitaria 95/5 116. L. A. B. is a Howard Hughes Medical InstituteStudent Research Training Fellow.

2 To whom requests for reprints should be addressed, at Division of Surgical Oncology,

Room 54—140CHS, University of California Los Angeles School of Medicine, 10833 LeConte Avenue, Los Angeles, CA 90024-1782. Phone: (310) 825-2644; Fax: (310) 825-7575; E-mail: jeconomo@surgery.medsch.ucla.edu.

3 The abbreviations used are: MART-I, MART-llMelan-A; AdV, adenovirus: DC,

dendritic cell; AdVMARTI, AdV vector expressing MART-l; IL, interleukin; APC,antigen-presenting cell; C3H, C3Hf/Kam/Sed; RT-PCR, reverse transcription-PCR;CMV, cytomegalovirus; pfu, plaque-forming unit; MO!, multiplicity of infection.

this study, we used a recombinant AdV vector to express the MART-i

gene into munne DCs, and we demonstrate that these engineered DCs

can generate MART-I-specific protective immunity in vivo.

Materials and Methods

Mice and Cell Lines. Female C3H (H-2―)mice, 5—8weeks old, were bredand kept in the Animal Facility ofthe Department of Radiation Oncology at theUniversity of California Los Angeles. All animal studies were conducted in

accordance with the Animal Care Policy of the University of California LosAngeles Chancellor's Animal Research Committee. Animals were kept in acontrolled flora environment and sacrificed according to established guidelines. NFSA is a spontaneous, poorly immunogenic murine fibrosarcoma thatarose in the C3H strain and has been described previously (10). NFSA(MART!) and NFSA(Neo) were developed by transfection of the NFSA linewith a plasmid carrying the MART-i and Neo genes or a plasmid with only the

Neo gene, respectively. A RT-PCR-generated 400-bp product of MART-IcDNA was cloned into the pUC19 plasmid, sequenced to confirm the identityof the PCR product obtained, and inserted into the pRcCMV expressionplasmid (Invitrogen, Carlsbad, CA). MARTpRcCMV plasmid (2.5 @g)was

mixed with 15 p1 of N-[l-(2,3-dioleoyloxyl)propyll-N,N,N-trimethylammoniummethyl sulfate (Boehringer Mannheim, Indianapolis, IN) in 20 mM HEPES

buffer (Sigma, St. Louis, MO) and incubated with 1 X l0@ tumor cells at 37°C

for 6 h, after which the medium was replaced with fresh culture medium.

Forty-eight h later, the transfected population was subjected to selection with0.5 mg/mI G4I8 (Geneticin; Life Technologies, Inc., Gaithersburg, MD) in

RPMI 1640 containing 10% FCS (Gemini Products, Calabasas, CA) and 1%(v/v) penicillin, streptomycin, and fungizone (Gemini) and incubated at 37°Cin 5% CO2. After three to four in vitro passages under G4l8 selection, tumor

cells were injected in the flanks of C3H mice. Developing tumors were passed

in vivo and periodically checked for MART-i and Neo expression by RT-PCR.The number of NFSA(MART1) cells from a single-cell suspension of in vivo

growing tumors required for tumor development in 50% of mice (TD50) was

determined using a previously established protocol (10). SVEC is a murine

endothelial cell line in the H@2kbackground. SVEC(MART1) and SVEC(Neo)were created by the same method used for NFSA sublines.

RNA Analysis. Total cellularRNA was isolatedusing TRIzol (Life Technologies, Inc.). Each RNA sample was treated with DNase (Stratagene, La

Jolla, CA), reverse-transcribed with random hexamer primers, and then subjected to PCR with either MART-I, neomycin resistance, or murine adeninephosphoribosyltransferase-specific PCR primers to confirm the expression ofthe MART-i and neomycin resistance transgenes as well as semiquantify thecDNA yield in each reverse-transcribed sample. The primers used were asfollows: MART-i, 5'-CCTC'ITFCTCTCTAGACCTGTGCCCTGACCCT

ACAA-3' and 5'-TFGGCAGACGTCTAGACTCGCTGGCTCTTAAGGT-3';adenine phosphoribosyltransferase, 5'-ACTCCAGGGGCTTCCTG'ITI'G-3'and 5'-ATCCACAATGACCACTCTCTG-3'; Neo, 5'-GGTGGAGAGGC

TATrCGGCT-3' and 5'-GATAGAAGGCGATGCGCTGC-3'.ConstructIon of Recombinant AdVs. Seed viral stocks of a recombinant

AdV vector (RR5), which does not contain a transgene insert, were generouslyprovided by Drs. Robert Meidell and Michael Barry (University of TexasSouthwestern, Dallas, TX). RR5 is an El-deleted replication-deficient AdV

type-S vector. We constructed the recombinant AdV vector AdVMART1,which contains human MART-I cDNA driven by the CMV enhancer/promoter.

The MART-I cDNA, previously cloned into the pUC19 plasmid, was sub

2865

Research. on February 21, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

GENETIC IMMUNIZATION wtm DCs

cloned into the shuttle vector pAC-CMVpLpA-. The recombinant AdVMART1 vector was made by cotransfection with pJM17, a plasmid containing the El-deleted AdV type-S genomic sequences, and pAC-CMVpLpA/

MART-I into the 293 human embryonic renal cell line. Homologous

recombination between the two plasmids occurs, and recombinant adenoviralcapsids are then released into the media. Clones of AdVMART1 were obtained

by plaque purification on fresh 293 cells. Viral seed stocks were then obtained

by amplification in 293 cells. The titer of each viral stock ranged from

lOb_lOtl pfu/ml as determined by plaque assay on 293 cells. Contamination

with wild-type recombinant AdV was assessed for each viral stock by plaque

assay on HeLa cells and was always found to be negative.

Construction of Plasmid cDNA for i.m. Immunization. MART-i cDNAwas cloned as described above, inserted into the VR-1012 plasmid (Vical, La

Jolla, CA), and prepared for i.m. injection using an endonuclease-free plasmidprep kit (Qiagen, Santa Clarita, CA).

Preparation of DCs. The developmentof DCs frommurinebone marrowprogenitor cells has been described previously (5). Briefly, bone marrow cellswere cultured overnight in RPM! with 10% FCS and antibiotics in a Petri dish.Cells were replated the following day at 1—2X 106 cells/well in 24-well platesin the presence of murine granulocyte macrophage colony-stimulating factor

(100 ng/ml; a generous gift from Amgen, Thousand Oaks, CA) and murineIL-4 (500 units/ml; R&D Systems, Minneapolis, MN). The loosely adherent

cells were harvested on day 8 and found to have a distinctive DC-like

morphology with cytoplasmic processes and to display a strong mixed lym

phocyte reaction-stimulating activity. Flow cytometric analysis of several

preparations showed that DCs comprised more than 90% of the large gated

cells, as judged by high expression of CDI lb. CD1 lc, CD18, CD44, CD8O

(B7.l), CD86 (B7.2), and MHC class I and class II antigens.AdV Transduction of DCs. DCs were incubatedat 37°Cfor 2 h with the

AdV vector at a MO! of 100 in a final volume of 1 ml of infection media

consisting of RPM! 1640 with 2% FCS. Infection was stopped by the addition

of 3 volumes of RPM! 1640 with 10% FCS. The DCs were washed extensively

before the injection of 5 X l0@DCs resuspended in 0.2 ml of PBS (Mediatech,Herndon, VA)/animal. Cell counts were determined using a hemocytometer,

and viability was assessed by trypan blue exclusion. In all cases, viabilityexceeded 95%.

In Vivo Tumor Therapy Models. C3H mice were immunized with DCsadministered iv. via lateral tail vein. For tumor implants, a single-cell suspen

sion of I X 106 cells/animal was isolated from NFSA(MART1) tumors in at

least the third in vivo transplant generation. Tumors were surgically removed,decapsulated, and minced. Minced tumors underwent a brief enzymatic digestion (60 mm) with DNase I (0.1 mg/ml; Sigma) and collagenase D (1 mg/ml;Boehringer Mannheim) in 50 ml of AIM-V media (Life Technologies, Inc.).

Cells were washed three times in PBS and resuspended in 0.2 ml of PBS/animal/injection. Injected cells were >90% viable as determined by trypanblue exclusion. For plasmid MART-i i.m. injection, 50 @gof plasmid in 50 p1

of PBS were injected into the tibialis anterior muscle of mice once a week for3 weeks. For AdVMART1 injection, 5 X 108 pfu were resuspended in a final

volume of 50 p1 (i.m. injection) or 200 p1 (i.p. injection) for each animal andinjected weekly for 3 weeks. The size of tumors was assessed three times

weekly. Tumor volume was approximated by the following calculation:@ ii,.3 (r, radius). All experiments included five mice/group and have been repeated

at least twice.

Cytotoxicity Assays. Splenocytesharvestedfrommice 7 days afterthe lastimmunization were restimulated by coculture with irradiated NFSA(MART1)cells (25: 1 ratio) in RPMI 1640 with 10% FCS and 10 units/mI IL-2 (Hoff

mann-La Roche, Nutley, NJ). Effector cells were harvested 5 days later.

Splenocytes from normal unimmunized mice served as controls and werestimulated in vitro in the same manner. For cytotoxicity assays, 2 X 106 target

cells were labeled by incubation in 2.5 ml of RPMI 1640 plus 10% FCS with

5tCr (100 p@Ci;Amersham Life Science, Arlington Heights, IL) for 18 h at37°C, washed extensively, and cocultured at 1 X l0@ cells/well with effector

cells (at the ratios given in the figures) in 96-well round-bottomed plates (200

p1/well) for 4 h at 37°C.The supernatants (150 p.1) from triplicate cultureswere collected and counted. In monoclonal antibody MHC class I blocking

experiments, 10 pA of purified hybridoma supernatant (16—l-2N hybridoma,

HB-l4; American Type Culture Collection, Rockville, MD) were added tochromated target cells before mixing with effector cells. Data points areexpressed as the mean percentage of release of 5tCr from target cells ±SE.

Statistical Analysis. Student's t test was performed to interpret the signif

icance of differences between final tumor volumes of animals immunized with

DCs and other immunization strategies (presented as mean ± SE). The

comparisons were performed excluding the animals that were completely

protected from tumor challenge to give a reliable assessment of the effect ofimmunization in the tumor growth curves. Two-sided P values are presentedfor individual comparisons.

Results

Characterization of NFSA(MART1). NFSA, a poorly immunogenic murine fibrosarcoma syngeneic in C3H mice (H@2k),was stablytransfected with human MART-i cDNA. Expression of MART-i wasstable over a 7-month period of multiple in vitro and in vivo passagesas determined by RT-PCR (Fig. 1). The phenotype of tumor line

NFSA(MART1) was unchanged compared with that of the wild-type

tumor with respect to in vitro doubling time (20 h), expression ofMHC class I, and lack of expression of MHC class II. The in vivoTD50 for NFSA(MART1) is approximately 3 X l0@ tumor cells,which is only slightly higher than that of NFSA (6 X iO@tumor cells;Ref. 10).

Transduction of Murine DCs by AdVMART1. We previouslyreported that El -deleted AdV vectors are excellent gene transfer

vectors for DCs, superior to any physical method tested (7). MurineDCs were differentiated for 8 days in vitro from normal C3H femoralbone marrow progenitor cells in the presence of murine granulocytemacrophage colony-stimulating factor and IL-4. We constructed anAdV vector in which the human MART-I cDNA was driven by theCMV enhancer/promoter. Transduction of munne DCs with AdVMARTI resulted in MART-i expression that persisted for at least 5additional days of cell culture as determined by RT-PCR (Fig. 1).Semiquantitative RT-PCR assessment of the transduction of murineDCs by AdVMART1 indicates a viral dose effect, with higherMART-i mRNA expression at increasing MOI (Fig. 1).

Genetic Immunization Using AdVMART1-transduced DCs. Invivo studies were performed to determine the optimal immunization

schedule using murine DCs transduced with AdVMART1.4 We haveused our best immunization protocol to directly compare AdVMART1/DC immunization with two other genetic immunization

strategies: systemic or i.m. AdVMART1 administration and i.m.MART-i plasmid injection. Groups of naive C3H mice received twoi.v. AdVMART1IDC injections (MOl = 100; 5 X iO@ DCs/animal),

three weekly i.m. injections of 50 p.g of MART-i plasmid cDNA, orthree weekly i.m. or i.p. injections with 5 X 108 pfu of AdVMART1.All animals were challenged 1 week after the last immunization with106 NFSA(MART1) cells. A representative experiment is shown inFig. 2. The DC-based strategy was superior in eliciting protectiveimmunity (i.v. AdVMART1/DC versus i.m. MART-i cDNA,P = 0.002; i.v. AdVMARTIIDC versus i.m. AdVMART1, P = 0.01;i.v. AdVMART1/DC versus i.p. AdVMART1, P < 0.001).

Immunization with AdVMART1/DCs Generates MART-1-specific CTLs. In vitro restimulated splenocytes from mice immunizedtwice with AdVMART1IDCs were able to lyse target cells thatexpressed MART-l but were unable to lyse the parental NFSA line.Splenocytes from control unimmunized mice were unable to lyseNFSA(MART1) or the untransfected parental cell line, and theMART-i-specific lysis was blocked with an anti-MHC class I anti

body (HB14; Fig. 3). We have performed several additional 51Crrelease assays in which splenocytes from AdVMART1IDC-immu

nized mice were unable to lyse the NFSA(Neo) cell line, which is theNFSA line transfected with an empty plasmid, but lysed another C3H(H@2k) MART-i-transfected target, SVEC(MARTI). Furthermore,

4 A. Ribas, manuscript in preparation.

2866

Research. on February 21, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

GENETIC IMMUNIZATION WITH DCs

A immunization was effective in generating protective immunity, suggesting rapid CTL induction. We therefore attempted to treat micro

scopically established tumors with either one or two iv. injections ofAdVMART1/DCs.AdVMART1IDCswere injectedi.v. 3 days (fiveanimals) or 3 and 7 days (five animals) after s.c. inoculation of 106NFSA(MART1) cells. Empty DCs were injected twice, at 3 and 7days after tumor implantation. Significant protection was observedwith both one and two i.v. treatments with AdVMARTI/DCs. Ani

mals treated twice with empty DCs showed a slight initial delay in the

tumor growth curve, but final tumor volumes were equal to those of

the control group (Fig. 4).

Discussion

—‘S. 4• MART-i

——@— 4 APRT

MART- 1 is a well-characterized human melanoma-associated antigen (2, 3). A nonapeptide derived from the MART-i gene presentedin the context of the HLA-A2.l is recognized by the majority of

tumor-infiltrating lymphocyte cultures tested (2). We recently clonedand analyzed the MART-i promoter, and we have shown that it

confers MART- 1-specific expression on a reporter gene (I 1). A

murine model of genetic immunization against MART- I was developed to define an optimal genetic immunization strategy. Our modeluses a human gene inserted into a highly aggressive and poorly

immunogenic murine fibrosarcoma cell line, NFSA (10).DCs are the most effective APCs for the induction of primary

immune responses (4, 5). Their high level of costimulatory molecule

expression makes them well suited for antigen presentation and less

likely to induce tolerance. Transduced with an AdVMART1 vector,murine DCs express high levels of MART-i mRNA in a viral dosedependent fashion, and this high level of expression is maintained for

—.-- i.m. pMART-1

—0—i.p.kjti-MA.RT-1

—.-. isa Mv-MART-i

-1@s-- kiv-MART-tDC

..“o...U@ün@n@ized

I I I I I I I I I

0 2 4 6 8 10 12 14 16 18

B

C

———*.. 4. APRT

.//@

.........@f/

5500

500o@

4500

@4000E

E.@3000>

@ 2500E

I- 2000CCe

@ 1500

1000

500

0

Fig. 2. Comparison of genetic immunization strategies. Animals (five animals/group)were immunized with: (a) three i.m. injections of 50 @sgof pRcCMVMART-l (•);(b)5 x l0@pfu of AdVMART1 administered i.m. (volume: 50 @l;U) or i.p. (volume, 200p3;U) in threeweekly injections;or (c) murineDCs (5 X lO@DCs/animal)transducedwith AdVMARTI at a MO! of 100 administered iv. in two weekly doses (is). One weekafter the last immunization, all immunized mice and a control unimmunized group (0)were challenged s.c. with 1 X 106NFSA(MART1) cells. Subsequent tumor appearanceand growth were recorded three times weekly. Mean tumor volumes ±SE of mice thatdeveloped tumors in each group are presented. Mice not developing tumors are presentedseparately. This experiment was repeated three times with comparable results. *, numberof animals with tumor/total number of animals in the group. **, number of animalswithout tumor/total number of animals in the group.

2867

— —— — _4

I-@

.0 .@ .@@

1!

#@%I-@

4. MART—i

m @- 4. MART-i

@ @——-. 4. APRT

Fig. 1. A, expression of MART-l mRNA in AdVMART1-transduced DCs. MurineDCs (1 X 106) were transduced with AdVMART1 or RR5 at different MOI. After a 72-hincubation, MART-l mRNA was assayed by RT-PCR. M202, a MART-l-positive humanmelanoma cell line, served as a positive control. B, DCs (I X 10'@)transduced withAdVMART1 at a MO! of 100 were analyzed at different time points to assess the durationof MART-l expression. C. the expression of MART-i was assessed in the parental NFSAcell line, the empty plasmid transfectant NFSA(Neo), and the stably transfected lineNFSA(MARTI) from both tissue culture samples and single-cell suspensions of freshtumors harvested from mice.

splenocytes from animals immunized twice with untransduced DCs orDCs transduced with the empty AdV vector RR5 were also unable tolyse NFSA(MART1) (data not shown).

Treatment of Established Microscopic Tumors. In a series ofexperiments, we have observed that a single AdVMART1IDC

Research. on February 21, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

4500

4000

3500It

EE 3000U)a,

@ 2500

@ 2000E

l@ 1500

1000

500

0

0 5 10 15 20 25

GENETIC IMMUNIZATION WITH DCs

at least 5 days in vitro (Fig. 1). Immunization of mice with thesegenetically engineered cells has the theoretical advantages of allowingthe processing and presentation of the MART-l antigen by the appropriate restriction elements and the continuous exposure of themurine immune system to endogenously processed MART-l-MHCclass I and possibly class II complexes on the surface of DCs.

Our results suggest that the presentation of MART-l-derived peptides by DCs elicits a better protective antitumoral response comparedto immunization with plasmid cDNA or the in vivo delivery of theAdVMART1 vector. Both the i.m. immunization with naked DNAand the systemic administration of AdV vector rely on host APCs topresent MART-l-denved epitopes (12—14).In this model, direct presentation of antigens by genetically engineered DCs seems to besuperior for the generation of tumor-specific immunity. The use of invivo passaged tumors for tumor challenge avoids the theoreticalconcern of eliciting culture media or FCS-derived immune responsesusing cultured DCs. The in vitro cytotoxicity studies provide confirm

atory evidence that the immunity generated was MART-l specific.Splenocytes from AdVMARTIIDC-immunized animals were able tolyse only MART-l-expressing tumor cells in a MHC class I-restricted

manner.We were able to significantly protect mice with previously im

planted MART-I-expressing tumor cells using AdVMART1IDCs.The s.c. injection of 1 X 106 NFSA(MART1) cells into naive miceinduces the appearance of a palpable tumor 5—I1 days after injection,

and this inoculum of tumor cells will kill all of the mice in less than1 month.This result impliesthat the AdVMART1IDCtreatmentcanelicit very rapid induction of cytotoxic immune effectors. Recentstudies have shown that freshly isolated DCs from both tumor-bearing

—.-- Adv-MART-1/DC x 2

—0—Adv-MART-1/DC xl-—-- Empty/DC

—0--Unimmunized

Days

(4/5)*

(l/5)**

Tumor DC DC

Fig. 4. Genetic immunization of mice with microscopically established tumors. Mice(five animals/group) were inoculated with 1 X 106 NFSA(MART1) cells on day 0 andtreated once (0) or twice (•)with 5 X l0@ MART-I-transduced DCs or twice with5 x l0@untransduced (empty) DCs (U), at day 3 and/or day 7. Mice in the control group(0) did not receive any treatment after the injection of 1 X 106 NFSA(MART1) cells.Mean tumor volumes ±SE of mice that developed tumors in each group are presented.Mice not developing tumors are presented separately. This experiment was repeated twicewith comparable results. *, number of animals with tumor/total number of animals in thegroup. **, number of animals without tumor/total number of animals in the group.

animals (15) and cancer patients (16) are unable to adequately presentantigens to the immune system, whereas DCs cultured in vitro from

bone marrow progenitor cells from the same subjects with malignanttumors allow for the restoration of the antigen-presenting function ofDCs (15, 16).

Several groups have attempted to use human DCs to stimulate theproduction of antigen-specific CTLs in vitro. Cultured DCs seem to besuperior to other APCs for the generation of specific immune effectors, and genetically engineered DCs are able to induce specific CTLswith fewer in vitro stimulations than when DCs are exogenouslypulsed with antigenic peptides (17, 18). The transduction of ourAdVMART1 construct into HLA.A2.l human DCs cultured fromperipheral blood mononuclear cells was able to present appropriateMART-i peptides that were recognized and specifically lysed bypreviously established CTL lines from MART-i-expressing melanoma patients.5

Results of this murine model cannot be directly extrapolated intohuman melanoma patients: MART-i is a differentiation antigen forcells of melanocyte lineage, and tolerance to this self-antigen must be

overcome to be able to elicit an effective immune response in melanoma patients, who are likely to have a higher tumor burden than themice protected in these murine studies. However, there is a large body

of evidence that T-cell responses can be generated to this antigen inmelanoma patients (14, 17). In summary, we report the successful useof genetically engineered DCs to generate protective antitumor immunity in vivo to a defined human melanoma antigen. This preclinicalmodel supports a gene therapy strategy for human melanoma.

References

Immunization

—. AdVMART1/DC—0-- AdVMART1/DC—0---Unimmunized.“.&.. AdVMART1/DC

—.—. Unimmunized

Target

NFSA(MART-1)NFSA(MART-1) + Anti-Class I mAbNFSA(MART-l)NFSANFSA

70 -

60-

50 -

.—Ca40-U)

-J

@ 30-

20 -

10 -

0-

E:T

Fig. 3. Generation ofCl'Ls by AdVMART1IDC administration. Mice were immunizedwith two weekly iv. administrations of AdVMARTIIDCs. One week after the lastimmunization, splenocytes were harvested and cultured for 5 days in the presence ofNFSA(MART1) irradiated tumor cells at a 25:1 ratio with 10 units/mi !L-2 and thenassayed against indicated “Crtargets. Splenocytes from AdVMART1/DC-immunizedmice were incubated for 4 h with either NFSA(MART1 ) (U), NFSA (is), or NFSA(MART1) previously treated in vitro with the anti-MHC class I antibody HB-14 (0).Splenocytes from unimmunized naive mice were restimulated in vitro using the sameprotocol and exposed to NFSA(MARTI) (0) or NFSA (@) as target cells. Chromiumrelease from three separate assays has been pooled, and the results are plotted as the meanpercentage of lysis ±SE.

I@ I

11 33 100

1. Traversari, C., van der Bruggen, P., Lulescher, I. F., Lurquin, C., Chomez, P., VanPd, A., Dc Plaen, E., Amar-Costesec, A., and Boon, T. A nonapeptide encoded by

5 L. H. Butterfield and B. Mukherji, unpublished observations.

2868

Research. on February 21, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

GENETIC IMMUNIZATION WITH DCs

human gene MAGE-I is recognized on HLA-Al by cytotoxic T lymphocytes directedagainst tumor antigen MZ2-E. J. Exp. Med., 176: 1453—1457,1992.

2. Kawakami, Y., Eliyahu, S., Delgado, C. H., Robbins, P. F., Rivoltini, L., Topalian,S. L., Mild, T., and Rosenberg, S. A. Cloning of the gene coding for a shared humanmelanoma antigen recognized by autologous T cells infiltrating into tumor. Proc.Nat!.Acad.Sci.USA, 91:3515—3519,1994.

3. Coulie, P. G., Brichard, V., Van Pd, A., Wolfel, T., Schneider, J., Traversari, C.,Mattei, S., Dc Plaen, E., Lurquin, C., Szikora, J. P., Renauld, J. C., and Boon, T. Anew gene coding for a differentiation antigen recognized by autologous cytotoxic Tlymphocytes on HLA-A2 melanomas. J. Exp. Med., 180: 35—42,1994.

4. Steinman, R. M. The dendritic cell system and its role in immunogenicity. Annu. Rev.Immunol., 9: 271—296,1991.

5. Inaba, K., Inaba, M., Romani, N., Aya, H., Deguchi, M., Ikehara, S., Muramatsu, S.,and Steinman, R. M. Generation of large numbers of dendritic cells from mouse bonemat-row cultures supplemented with granulocyte/macrophage colony-stimulating fac

tor. J. Exp. Med., 176: 1693—1702,1992.6. Mayordomo, J. I., Zorina, T., Storkus, W. J., Zitvogel, L.. Celluzzi, C., Falo, L. D.,

Meief, C. J., fldstad, S. T., Martin Kast, W., Delco, A. B., and Lotze, M. T. Bonemarrow-derived dendritic cells pulsed with synthetic tumor peptides elicit protectiveand therapeutic antitumor immunity. Nat. Med., 1: 1297—1302,1995.

7. Zitvogel, L., Mayordomo, J. I., Tjandrawan. T., DeLco, A. B., Clarke, M. R., Lotze,M. T., and Storkus, W. J. Therapy of munne tumors with tumor peptide-pulseddendritic cells: dependence on T cells, B7 costimulation, and T helper 1-associatedcytokines. J. Exp. Med., 183: 287—297, 1996.

8. Mayordomo, J. I., Loftus, D. J., Sakamoto, H., Dc Cesare, C. M., Appasamy, P. M.,Lotze, M. T., Storkus, W. J., Appella, E., and DeLco, A. B. Therapy of murine tumorswith p53 wild-type and mutant sequence peptide-based vaccines. J. Exp. Med., 183:1357—1365,1996.

9. Arthur, J. F., Butterfield, L. H., Kiertscher, S. M., Roth, M. D., Bui, L. A., Lau, R.,Dubinett, S., Glaspy, J., and Economou, J. S. A comparison of gene transfer methodsin human dendritic cells. Cancer Gene Ther., 4: 17—25,1997.

10. MilIas, L., Wike, J., Hunter, N., Volpe, J., and Basic, I. Macrophage content of murinesarcomas and carcinomas: associations with tumor growth parameters and tumorradiocurability. Cancer Res., 47: 1069—1075, 1987.

11. Butterfield, L. H., Stoll, T., Lau, R., and Economou, J. S. Cloning and analysis ofMART-l/Melan-A human melanoma antigen promoter regions. Gene (Amst.), inpress,1997.

12. Doe, B., Selby. M., Barnett, S., Baenziger, J., and Walker, C. H. Induction ofcytotoxic T lymphocytes by intramuscular immunization with plasmid DNA isfacilitated by bone marrow-derived cells. Proc. Natl. Acad. Sci. USA, 93: 8578—8583, 1996.

13. Con, M., Lee, D. J., Carson, D. A., and Tighe, H. Gene vaccination with nakedplasmid DNA: mechanism of CTL priming. J. Exp. Med., 184: 1555—1560,1996.

14. Thai, Y., Yang, J. C.. Kawakami, Y., Spiess, P., Wadsworth, S. C., Cardoza, L. M.,Couture, L. A., Smith, A. E., and Rosenbert, S. A. Antigen-specific tumor vaccines:development and characterization of recombinant adenoviruses encoding MARTI orgpl0O for cancer therapy. J. Immunol., 156: 700—710, 1996.

15. Gabrilovich, D. I., Chen, H. L., Girgis, K. R., Cunningham. T., Meny, G. M., Nadaf,S., Kavanaugh, D., and Carbone, D. P. Production of vascular endothelial growthfactor by human tumors inhibits the functional maturation of dendritic cells. Nat.Med., 2: 1096—1103,1996.

16. Gabrilovich, D. I., Kavanaugh, D., Corak, J., Nadaf-Rahrov, S., Cunningham, T., andCarbone, D. P. Defective function of dendritic cells in patients with breast cancer canbe overcome by generation of these cells from precursors, a new approach to cancerimmunotherapy. Proc. Am. Soc. Clin. Oncol.. 15: 354, 1996.

17. Reeves, M. E., Royalo, R. E., Lam, J. S., Rosenberg, S. A., and Hwo, P. Retroviraltransduction of human dendritic cells with a tumor-associated antigen gene. CancerRes., 56: 5672—5677, 1996.

18. Marincola, F. M., Rivoltini, L., Salgaller, M. L., Player, M., and Rosenberg, S. A.Differential anti-MART-lfMelanA CTL activity in peripheral blood of HLA-A2melanoma patients in comparison to healthy donors: evidence of in vito priming bytumor cells. J. Immunother., 19: 266—277, 1996.

2869

Research. on February 21, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

1997;57:2865-2869. Cancer Res   Antoni Ribas, Lisa H. Butterfield, William H. McBride, et al.   Murine Dendritic CellsMART-1/Melan-A Using Recombinant Adenovirus-transduced Genetic Immunization for the Melanoma Antigen

  Updated version

  http://cancerres.aacrjournals.org/content/57/14/2865

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/57/14/2865To request permission to re-use all or part of this article, use this link

Research. on February 21, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from