Biochemical and Biophysical Research Communications 370 (2008) 279–284

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications

journal homepage: www.elsevier .com/locate /ybbrc

Anti-tumor immunity induced by CDR3-based DNA vaccination in a murineB-cell lymphoma model

Monica Rinaldi a,*, Daniela Fioretti a,1, Sandra Iurescia a,1, Emanuela Signori a,b, Pasquale Pierimarchi a,Davide Seripa c, Giancarlo Tonon d, Vito Michele Fazio b,e

a Institute of Neurobiology and Molecular Medicine, CNR, Via Fosso del Cavaliere 100, 00133 Rome, Italyb CIR, Section of Molecular Medicine and Biotechnology, University ‘‘Campus Bio-Medico”, Via E. Longoni 83, 00155 Rome, Italyc Laboratory of Gerontology & Geriatrics, Research Department, IRCCS ‘‘Casa Sollievo della Sofferenza”, Viale Cappuccini, 71013 San Giovanni Rotondo (FG), Italyd Bio-Ker s.r.l., Consorzio 21-Polaris, 09010 Pula (CA), Italye Laboratory of Oncology, Research Department, IRCCS ‘‘Casa Sollievo della Sofferenza”, Viale Padre Pio, 71013 San Giovanni Rotondo (FG), Italy

a r t i c l e i n f o a b s t r a c t

Article history:Received 17 March 2008Available online 24 March 2008

Keywords:DNA vaccineCancer vaccineLymphomaProphylactic naked DNA vaccination

0006-291X/$ - see front matter � 2008 Elsevier Inc. Adoi:10.1016/j.bbrc.2008.03.076

* Corresponding author. Fax: +39 0649934257.E-mail address: monica.rinaldi@artov.inmm.cnr.it

1 Both the authors contributed equally to this work.

The idiotypic structure present on B-cell neoplasms is a tumor-specific antigen and an attractive tar-get for immunotherapy. Here, the tumor protective effects recruited by CDR3-based DNA vaccines inthe poorly immunogenic, highly aggressive 38C13 murine B-cell lymphoma model were evaluated.The regions belonging to the idiotypic VH and VL CDR3 sequences were chosen for the design oftwo synthetic mini-genes and arranged in high-level expression plasmids. Syngeneic C3H/HeN micewere immunized by intramuscular electroporation with pVHCDR3-IL-2 and pVLCDR3-IL-2 naked DNAs.This approach provided protection in about 60% of animals challenged with a 2-fold lethal dose oftumor cells, as opposed to non-survivors in control groups. Furthermore, a long-term survival wasinduced in these mice since they were still alive and tumor-free 4 months following tumor challenge.Analysis of the humoral immunity revealed the presence of antibodies reactive with the peptidesencompassing the CDR3 sequences in the sera of vaccinated mice. Moreover, immune sera specificallyreacted with the parental 38C13 tumor cells in flow cytometry assays, indicating that such immuni-zation elicited anti-idiotypic antibodies.These findings provide a basis for exploring the use of CDR3-based DNA vaccines against B-cell lymphoma.

� 2008 Elsevier Inc. All rights reserved.

In spite of advances in chemotherapy and radiation, low gradeB-cell lymphomas remain incurable and Non-Hodgkin’s lympho-mas (NHL) are responsible for more than 19,000 deaths in the Uni-ted States annually [1].

B-cell lymphomas express tumor-specific immunoglobulin (Ig),the variable (V) regions of which contain unique antigenic determi-nants, termed Idiotype (Id). The Id can serve as tumor-specific antigen(Ag) and therefore is a suitable target for vaccine immunotherapy.

Several preclinical studies as well as clinical studies demon-strated the effectiveness of idiotypic vaccination against B-celllymphoma (reviewed in Refs. [2,3]). However, the need to produceand certify high amounts of purified idiotypic protein for each pa-tient makes this approach time consuming and expensive, thuslimiting the clinical relevance of a patient-specific vaccine.

An alternative approach to Id protein vaccination is to use DNAvaccines.

ll rights reserved.

(M. Rinaldi).

The potential advantages for this vaccination system includeease of patient-specific vaccine generation by gene amplificationof tumor hypervariable domains, simple scale-up and safety [4–6].

Plasmid DNA vaccination has been successfully tested in ani-mal models against a wide range of infectious and tumor anti-gens. Currently, several clinical trials are ongoing [7]. Thisapproach has been applied to therapy of experimental murinelymphomas and has proved effective in eliciting anti-idiotypicimmune response [2,3]. Idiotypic antigenic determinants arisefrom the variable regions of the heavy (VH) and light (VL) chainsof immunoglobulin. Several reports have indicated that theimmunodominant epitopes of the clone-specific Ig lie withinthe hypervariable regions and mainly within the complementar-ity-determining regions (CDR)s 3 [8–10]. The CDR3s are consid-ered a clonal signature of an expanded B-cell populationfollowing neoplastic transformation and a ‘‘hot spot” of particu-lar interest for construction of subunit vaccines [6,11,12]. Vac-cines including only this minimal antigenic domain wereproved to induce antibody response [5,13].

Our group has previously demonstrated the possibility of usingdifferent patient-derived VHCDR3 peptides as a target for eliciting a

280 M. Rinaldi et al. / Biochemical and Biophysical Research Communications 370 (2008) 279–284

tumor specific immune response via DNA-based vaccination, sug-gesting that the CDR3 sequence may be suitable as an anti-lym-phoma vaccine [5].

However, such an approach requires the demonstration that theCDR3-based DNA vaccine could generate an immune response ableto protect against a lethal tumor challenge.

In the present study, we assessed the possibility that geneticimmunization with two plasmids encoding the VLCDR3 andVHCDR3 region could induce humoral response and provide pro-tection against a lethal tumor challenge in the 38C13 B-cell lym-phoma as tumor model.

Materials and methods

Animal and cell lines. C3H/HeN mice, 8- to 9-week-old, were obtained fromCharles River Italia S.p.A. (Calco, CO, Italy). Animal experimental protocols were ap-proved by the Italian Ministry of Health. Serum samples were collected by tailbleeding.

Murine 38C13 cells (kindly provided by Dr. Arya Biragyn) are a carcinogen-in-duced B-cell lymphoma tumor line that expresses an IgM/j surface antigen [14].All experiments were performed from a working cell bank of frozen 38C13 cells.Prior to tumor challenge, a fresh aliquot of frozen 38C13 cells were grown for72 h in RPMI 1640, 10% heat-inactivated FBS, 2 mM L-glutamine, 100 U/ml penicil-lin, 100 U/ml streptomycin, and 50 lM b-Mercaptoethanol at 37 �C, 5% CO2, in ahumidified incubator.

COS-7 cells (ATCC CRL-1651) were transfected with LipofectamineTM 2000(Invitrogen S.r.l., Italy) and 5 lg of plasmid DNAs.

A20 B lymphoma cell line, derived from Balb/c mice, was purchased from theAmerican Type Culture Collection (ATCC, LGC Promochem, Italy).

Peptide synthesis. The native peptides NH2-DPNYYDGSYEGYFDYWAQGTTL-COOH (IgM 38C13 VH 101–122), NH2-YSFSISNLEPEDIATYYCLQYDNLYTFGGGTKLEIK-COOH (IgM 38C13 VL 71–106), NH2-MHTAVYYCAKGAQGASLGKAYFFDCWGQGTQVTVSS-COOH (VH CDR3-PA; [5]) were synthesized by Primm (Primm s.r.l., Italy)and dissolved in the suggested buffer prior to use.

DNA vaccines. The DNA vaccines were constructed in the plasmid pRC110 back-bone previously described [5]. A DNA nuclear targeting sequence (NTS), consistingof two complete, contiguous copies of the functional 72-bp element of the SV40 en-hancer was amplified from the pIRES vector (Clontech Laboratories Inc., MountainView, CA, U.S.A) by means of two primers (upstream NTS: 50-CTAGTCTAGAGGTACCTTCTGAGG-30 and downstream NTS: 50-CTAGTCTAGAGCATGCTTTGCATA-30 , which show in bold the XbaI restriction site) and cloned into the pRC110.PCR amplification was carried out with Pfu DNA Polymerase (Promega Italia S.r.l.,Italy) for 30 cycles (45 s at 94 �C, 1 min at 65 �C, 2 min at 72 �C). The resulting plas-mid was named pRC110-NTS-IL-2. The NTS region has been reported to facilitatenuclear plasmid uptake in murine skeletal muscle following naked DNA electropor-ation (reviewed in [15]).

The plasmids pVHCDR3-IL-2 and pVLCDR3-IL-2 were constructed using twopairs of complementary oligonucleotides, synthesized to assemble the double-stranded mini-fragments.

Sense VHCDR3(W8I)109-116 primer.CTAGCTAGCGCCACCATGTACGAAGGGTACTTTGACTACATTTAAGCGGCCGCTAAA

CTATAntisense VHCDR3(W8I)109–116 primer:

Fig. 1. Schematic representation of the constructs used for genetic immunization. (A) pVthe amino acid sequences of the antigenic peptides.

ATAGTTTAGCGGCCGCTTAAATGTAGTCAAAGTAGCCTTCGTACATGGTGGCGCTAGCTAG

Sense VLCDR388–98 primer:ATAGCTAGCCCCACCATGTGTCTACAGTATGATAATCTGTACACGTTCGGATAAGCG

GCCGCAAAAGGAAATAntisense VLCDR388–98 primer:ATTTCCTTTTGCGGCCGCTTATCCGAACGTGTACAGATTATCATACTGTAGACACATG

GTGGGGCTAGCTATEach primers incorporated a NheI site (50 end) and a NotI sites (30 end), respec-

tively, (underlined sequence). The double stranded fragments were generated byannealing in 50 ll of 10 mM Tris–HCl (pH 7.5), 5 mM MgCl2, and 7.5 mM DTT buffer.The VH synthetic CDR3 nucleic acid sequence refers to murine idiotypic IgM 38C13B-cell lymphoma heavy chain (l-chain) variable region (GenBank Accession No.X02676) while the VL nucleotide sequence encompassing the CDR3 belongs to light(j) chain variable region of the same IgM (GenBank Accession No. M17782).

Plasmid DNA was purified using the Qiagen EndoFreeTMPlasmid Mega Kit (QiagenS.p.A., Italy) and resuspended in sterile, endotoxin-free 150 mM sodium phosphate,pH 7.0.

Intramuscular DNA electrotransfer and tumor challenge. Anesthetized mice wereinjected in both posterior muscle legs with plasmid solution (40–50 lg of each plas-mid) and subjected to electrotransfer with BTX ECM 830 Pulse Generator (Gene-tronics, San Diego, CA) at 175 V/cm, 10 ms square-wave pulses, 1 Hz. Muscleswere pre-treated with bovine hyaluronidase as reported elsewhere [16].

Animals were challenged intraperitoneally (i.p.) with 2 � 102 38C13 tumorcells. The survival of challenged mice was followed and results were analyzed forsignificance by statistical analysis.

IL-2 elisa assay. Murine IL-2 ELISA development kit (PeproTech EC Ltd., UK) wasused according to the manufacturer’s instructions.

Samples and the murine recombinant IL-2 standard were serially diluted in 1�PBS and added to the microplates. IL-2 binding was detected by biotin–avidindetection step, followed by chromogen ABTS incubation (Sigma–Aldrich S.r.l., Italy).Color development was monitored at 405 nm. The concentration of cytokine in thesamples was determined from the standard curve.

Antibody detection. ELISA plates were coated with 50 lg/ml of VH 101–122 peptide,VL 71–106 peptide or VH CDR3-PA as irrelevant peptide. Mice sera (diluted 1:100 in1 � PBS/0.1% BSA/0.05% Tween 20) were added and incubated for 2 h at r.t.; thiswas followed by sheep anti-mouse Ig-HRP (1:5000 diluted, Amersham Biosciences,Italy) 2 h incubation. Enzyme substrate ABTS was added and the absorption mea-sured at 405 nm.

FACS analysis. 38C13 or A20 cells (5 � 105 cells) were incubated with individualserum (diluted 1:10) at 4 �C in 3% BSA/1 � PBS, containing 0.02% NaN3. Cells werestained with FITC-conjugated goat anti-mouse IgG (c-chain specific) (Sigma–Al-drich, Italy). Analysis was performed with a FACS scan flow cytometer (Becton Dick-inson, Mountain View, CA).

Statistical analysis. ELISA data were analyzed by two-tailed Student’s t test. Sur-vival analyses were performed using the Kaplan-Meier product-limit method andthe log-rank test. Tumor-bearing animal proportions were compared by X2 analysis.

Results

Generation of DNA-minigene vaccines

The bicistronic expression vector pRC110 has been previouslydeveloped. This plasmid bears two expression cassettes, which

LCDR3-IL-2 and (B) pVHCDR3-IL-2 constructs. Enlargements show the minigene and

M. Rinaldi et al. / Biochemical and Biophysical Research Communications 370 (2008) 279–284 281

transcriptional activity is driven by the immediate-early cytomeg-alovirus (i/e CMV) promoter-enhancer and the long terminal re-peat of Rous sarcoma virus (LTR RSV) promoter, respectively. Themurine interleukin-2 (IL-2) sequence is cloned downstream theCMV promoter [5]. A DNA nuclear targeting sequence was includedinto vector backbone (see Materials and methods) and the result-ing plasmid was named pRC110-NTS-IL-2.

The nucleic acid sequence of the idiotypic IgM (38C13-Id) lightand heavy chains was analyzed and the regions belonging to theVLCDR3 and VHCDR3 were chosen for the production of two syn-thetic mini-genes. Actually, the VLCDR3 sequence expressed the11-mer peptide starting with the Cysteine88 (i.e. Cys104 in theIMGT unique numbering) [17] and encompassing the CDR3 andthe conserved Phenylalanine and Glycine residues of framework(FR)4. The VHCDR3 sequence specifies the 8-mer H-2Kk ‘‘anchor-modified” YEGYFDYI109–116 epitope. [18]. The substitution con-cerned the FR4-Tryptophan116 which has been replaced withIsoleucine.

Two vaccines containing the VLCDR3 or VHCDR3 region of38C13-Id were constructed in pRC110-NTS-IL-2 and namedpVLCDR3-IL-2 and pVHCDR3-IL-2, respectively (Fig. 1A and B).The minigenes included a Kozak consensus sequence beside an

Fig. 2. (A) In vitro IL-2 production after COS-7 cell line transfection with pRC100, pRcollected at the indicated time after transfection and tested at two dilution points. pRC11(B) In vivo IL-2 secretion following electrotransfer with pVHCDR3-IL-2/pVLCDR3-IL-2 or pRa value from a single mouse; bars indicated the mean values at the indicated serum dil

in frame ATG. pVLCDR3-IL-2 and pVHCDR3-IL-2 vaccines werecombined in a single vaccine formulation. The constructpRC110-NTS-IL-2 was used as nonspecifically immunizingcontrol.

Plasmids expression evaluation

To verify the plasmids expression we evaluated the IL-2 in vitroand in vivo secretion.

COS-7 cells were transfected with pRC110-NTS-IL-2, pVHCDR3-IL-2 and pVLCDR3-IL-2. pRC100 (pRC110-NTS lacking IL-2) wasused as negative control. With the exception of the pRC100 group,comparable levels of IL-2 were measured in culture supernatantsof all experimental groups at 1 day and 2 days after transfection.The same results were obtained by using different supernatantdilution (Fig. 2A). As expected, pRC100 transfected culture super-natant showed undetectable levels of IL-2.

The in vivo secretion was evaluated 7 days following electro-transfer with pVHCDR3-IL-2 and pVLCDR3-IL-2 used in combina-tion or pRC110-NTS-IL-2. As shown in Fig. 2B, IL-2 serum levelswere clearly detectable in 100% of samples. No different levels ofstatistical significance were observed between the pVHCDR3-IL-2/

C110-NTS-IL-2, pVHCDR3-IL-2, and pVLCDR3-IL-2. The culture supernatants were0-NTS-IL-2, pVHCDR3-IL-2 and pVLCDR3-IL-2 contain IL-2 whereas pRC100 does not.C110-NTS-IL-2. Uninjected mice represent the control group. Each marker indicates

ution. NS, not significant.

Table 1Reactivity of mouse sera against a CDR3 control peptide as assessed by ELISA

Mice groups Vaccine formulation Absorbance (nm)a PA vs B

A pVHCDR3-IL-2/pVLCDR3-IL-2 0.192±0.255 NS

B pRC110-NTS-IL-2 0.086±0.238

a Negative control�subtracted values.

282 M. Rinaldi et al. / Biochemical and Biophysical Research Communications 370 (2008) 279–284

pVLCDR3-IL-2 and pRC110-NTS-IL-2 groups, at two serum dilu-tions. Our findings are consistent with previous published dataon dose–response expression of plasmid vectors for i.m. injection,confirming that the different amounts of plasmids we used were inthe plateau range of gene expression [19].

Humoral immunity induced by CDR3-based DNA vaccines

To investigate the immune response in the mice vaccinatedwith pVLCDR3-IL-2 and pVHCDR3-IL-2, on the basis of our previousdata (manuscript in preparation) animals were immunized twiceat 3-week intervals by intramuscular injection and electroporation(Fig. 3A). Serum samples were collected 2 weeks after boosting,and tested in ELISA for their reactivity with either VH peptide(D101–L122) or VL peptide (Y71–K106). The results are presented inFig. 3. As shown, pVHCDR3-IL-2/pVLCDR3-IL-2 vaccine elicited anti-bodies with reactivity against VH peptide (P = 0.0305 vs. pRC110-NTS-IL-2 group) (Fig. 3B) as well as a significant response againstVL peptide (P = 0.0464 vs. pRC110-NTS-IL-2 group) (Fig. 3C). Theantigen specificity was assessed by using a control CDR3 peptide

Fig. 3. (A) Outline of the experimental procedure. Two weeks after boosting mice were in(B) and VL (Y71–K106) (C) peptides determined by ELISA. Unimmunized mice represent therepresented by a horizontal bar. (D, E) Assessment of ability of CDR3-based DNA vaccinindividual serum of pVHCDR3-IL-2/pVLCDR3-IL-2 group is shown. Immune serum was reirrelevant Id was used as negative control. Shaded regions represent fluorescence from

as negative control. No significant response was detected to theirrelevant peptide (Table 1).

We next examined the reactivity of the antibodies induced bythe pVHCDR3-IL-2/pVLCDR3-IL-2 association towards the surfaceIgM on 38C13 lymphoma cells. Flow cytometry data shown inFig. 3D demonstrated that antibodies raised against thepVHCDR3-IL-2/pVLCDR3-IL-2 reacted with the native idiotypicIgM displayed on the tumor cell surface, indicating the presenceof anti-Id Abs in immune sera. None of the sera reacted with a B-cell lymphoma expressing an irrelevant Id (A20), demonstratingthat the humoral response elicited by pVHCDR3-IL-2/pVLCDR3-IL-2 combination vaccine is specific for the parental Id (Fig. 3E).

jected i.p. with 2 � 102 38C13 cells. (B, C) Antibody response against VH (D101–L122)control group. Each marker indicates a value from a single mouse; group means are

es to induce antibody against native 38C13-Id. A representative FACS analysis withacted with 38C13 cells (D) or A20 cells (E). The A20 B-cell lymphoma expressing ancells reacted with serum from mock immunized mouse.

M. Rinaldi et al. / Biochemical and Biophysical Research Communications 370 (2008) 279–284 283

Anti-tumor effect of vaccination elicited by CDR3-based DNA vaccines

To address the protective tumor immunity of pVLCDR3-IL-2 andpVHCDR3-IL-2 vaccines, we evaluated their efficacy in inducingprotection against a lethal tumor challenge. Two weeks after boos-ter injection, animals were inoculated i.p. with a 2-fold lethal doseof syngeneic 38C13 tumor.

The results of a representative experiment are shown in Fig. 4Aand B. All unimmunized animals challenged with a lethal dose of38C13 tumor cells showed tumor formation and died within 4 weeksof the tumor challenge, with a median survival time of 21 days. Incontrast, mice immunized with pVHCDR3-IL-2/pVLCDR3-IL-2 plas-mids significantly suppressed tumor growth and resulted in 57%long-term survivors. These mice remained disease-free for an obser-vation period of 120 days. Otherwise, mice immunized withpRC110-NTS-IL-2 did not show any protection, with median survivaltime of 19 days. This value is significantly different from that ob-tained in the pVHCDR3-IL-2/pVLCDR3-IL-2 group (P = 0.0273).

Vaccination with pVLCDR3-IL-2 or pVHCDR3-IL-2 failed to con-fer tumor protection (data not shown).

Discussion

In a previous study we demonstrated the possibility of usingdifferent patient-derived VHCDR3 peptides as a target for elicitinga tumor specific humoral response via DNA-based vaccination [5].

Fig. 4. Tumor resistance (A) and survival analysis (B) following immunization with CDsignificant value (P = 0.0183) by X2 analysis when comparing the pVHCDR3-IL-2/pVLCDR

Here, we used the murine B-cell tumor 38C13 model system totest CDR3-based DNA vaccination as a possible approach for thetreatment of B-cell lymphomas.

Several investigators have demonstrated the utility of idiotypicdeterminants present on B-cell neoplasms as tumor-specific anti-gens for active immunotherapy [9,20,21]. Immunization of C3H/HeN mice with molecular fragments of syngeneic 38C13-Id in-duced immune responses and provided protection against a lethaltumor challenge. Peptide fragments of the idiotype might be moreimmunogenic for induction of idiotype-specific immunity than thecomplete intact idiotypic Ig [22]. Moreover, restriction to the indi-vidual CDR3 region has the advantage of excluding xenogenic orallogenic epitopes contained in the variable and constant regionsof the idiotypic Ig, enhancing the safety margin of this approach.

In this report, we analyzed the idiotypic IgM 38C13 B-cell lym-phoma to produce two minigene constructs independentlyexpressing VH and VL determinants.

We first demonstrated that vaccination of syngeneic mice withboth DNA constructs elicited humoral response and the antibodiesdetected were reactive in vitro with the peptides encompassing theCDR3 sequences. Moreover, the immune sera recognized the nativeidiotypic Ig exposed on 38C13 malignant cells, as demonstrated byimmunofluorescence analysis.

We further investigated the in vivo efficacy of co-administrationof the two plasmids to mediate the protective anti-tumor effect inthe syngeneic mouse model. Despite the aggressive nature of the

R3-based DNA vaccines and 38C13 lymphoma challenge. (�) denotes a statistically3-IL-2 group (n = 7 mice) to all other groups (n = 6 mice).

284 M. Rinaldi et al. / Biochemical and Biophysical Research Communications 370 (2008) 279–284

38C13 B-cell lymphoma almost 57% of the animals receivingpVHCDR3-IL-2/pVLCDR3-IL-2 showed resistance to tumor forma-tion. These animals were reported as long-term survivors since theywere still alive and tumor free 4 months following tumor challenge.

The combined contribution achieved by the separate expressionof the 38C13 VH and VL regions in the induction of Id-specific im-mune response and protection from tumor challenge has beendemonstrated elsewhere [23–25].

Earlier work by Campbell et al. suggested that anti-idiotypicantibodies play the major role in the destruction of 38C13 tumor[26]. Using naked DNA as the immunogen, the role of anti-Id anti-bodies in resistance to the 38C13 tumor was demonstrated in pas-sive transfer experiments in which sera from Id-immunized miceconferred resistance to tumor growth in non-immunized mice,whereas protection of naïve recipients against tumor challengecould not be demonstrated by adoptive transfer of spleen andlymph node cells [27].

The results presented herein demonstrated that the CDR3-based immunization strategy was adequate to induce immune re-sponse and to provide protection against a lethal tumor challengeeven for a very aggressive B-cell lymphoma as 38C13. Little isknown of the structural basis of the idiotype/antibody interactionfollowing DNA vaccination. Investigation of the mechanism forthe effect of VHCDR3 and VLCDR3 DNA constructs is currentlyongoing, including experiments to assess antitumor T-cell re-sponse. In the present study, we have also explored complemen-tary strategies to enhance the efficacy of a CDR3-based DNAvaccine in terms of in vivo antigen expression and immunogenic-ity. We combined i.m. injection with electroporation since in-creased DNA uptake, enhanced transcriptional or translationalactivity, as well as the adjuvant-like properties have been demon-strated [28,29]. Different cytokines have been used as adjuvantsto enhance the efficacy of DNA immunization in mice [30,31].An appropriate cytokine environment might be helpful and evennecessary to foster activation of antigen presenting cells forDNA immunization when mini-genes that do not contain T helpercell epitopes are used. In these experiments the adjuvant was theIL-2 molecule which in our previous work was found to enhancethe percentage of responding animals [5].

In conclusion, our data indicate that CDR3-based DNA vaccinescan be explored to induce protective immune response against B-cell lymphoma. A goal of our ongoing studies is the developmentof effective immunization strategies against previously establishedtumors. Further improvements of the potency of CDR3-based DNAvaccines will be required to develop them as a genetic vaccinationstrategy with potential uses in the clinical settings.

Acknowledgments

We are grateful to the technical staff of the Animal Facility at the‘‘Sacro Cuore” Catholic University of Rome, Italy, for their assistance.This work was supported by MUR Grant FIRB 2001 (V.M. Fazio) andby Bio-Ker S.r.l. contract/Grant No. 07007/R3 (M. Rinaldi).

References

[1] A. Jemal, T. Murray, E. Ward, A. Samuels, R.C. Tiwari, A. Ghafoor, E.J. Feuer, M.J.Thun, Cancer statistics, 2005, CA Cancer J. Clin. 55 (2005) 10–30.

[2] S.A. Hurvitz, J.M. Timmerman, Current status of therapeutic vaccines for non-Hodgkin’s lymphoma, Curr. Opin. Oncol. 17 (2005) 432–440.

[3] S.S. Neelapu, S.T. Lee, H. Qin, S.C. Cha, A.F. Woo, L.W. Kwak, Therapeuticlymphoma vaccines: importance of T-cell immunity, Expert Rev. Vaccines 5(2006) 381–394.

[4] M. Rinaldi, E. Signori, P. Rosati, G. Cannelli, P. Parrella, E. Iannace, G. Monego,S.A. Ciafre, M.G. Farace, S. Iurescia, D. Fioretti, G. Rasi, V.M. Fazio, Feasibilty ofin utero DNA vaccination following naked gene transfer into pig fetal muscle:transgene expression, immunity and safety, Vaccine 24 (2006) 4586–4591.

[5] M. Rinaldi, F. Ria, P. Parrella, E. Signori, A. Serra, S.A. Ciafre, I. Vespignani, M.Lazzari, M.G. Farace, G. Saglio, V.M. Fazio, Antibodies elicited by naked DNA

vaccination against the complementary-determining region 3 hypervariableregion of immunoglobulin heavy chain idiotypic determinants of B-lymphoproliferative disorders specifically react with patients’ tumor cells,Cancer Res. 61 (2001) 1555–1562.

[6] A. Pession, R. Tonelli, G. Canderan, M. Bendandi, A. Rosolen, K. Basso, G. Basso, F.Locatelli, N. Zabalegui, G. Paolucci, Non-radiolabelled PCR consensus primers andautomatic sequencing enable rapid identification of tumor-specific V(H) CDR3 inaggressive B-cell malignancies, Leuk. Lymphoma 44 (2003) 1597–1601.

[7] M.A. Liu, J.B. Ulmer, Human clinical trials of plasmid DNA vaccines, Adv. Genet.55 (2005) 25–40.

[8] S. Baskar, C.B. Kobrin, L.W. Kwak, Autologous lymphoma vaccines inducehuman T cell responses against multiple, unique epitopes, J. Clin. Invest. 113(2004) 1498–1510.

[9] M.J. Campbell, W. Carroll, S. Kon, K. Thielemans, J.B. Rothbard, S. Levy, R. Levy,Idiotype vaccination against murine B cell lymphoma. Humoral and cellularresponses elicited by tumor-derived immunoglobulin M and its molecularsubunits, J. Immunol. 139 (1987) 2825–2833.

[10] A. Watanabe, E. Raz, H. Kohsaka, H. Tighe, S.M. Baird, T.J. Kipps, D.A. Carson,Induction of antibodies to a kappa V region by gene immunization, J. Immunol.151 (1993) 2871–2876.

[11] S.S. Sahota, R. Garand, R. Bataille, A.J. Smith, F.K. Stevenson, VH gene analysis ofclonally related IgM and IgG from human lymphoplasmacytoid B-cell tumorswith chronic lymphocytic leukemia features and high serum monoclonal IgG,Blood 91 (1998) 238–243.

[12] L. Hansson, H. Rabbani, J. Fagerberg, A. Osterborg, H. Mellstedt, T-cell epitopeswithin the complementarity-determining and framework regions of thetumor-derived immunoglobulin heavy chain in multiple myeloma, Blood101 (2003) 4930–4936.

[13] S.Y. Lim, S. Laxmanan, G. Stuart, S.K. Ghosh, Anti-lymphoma immunity:relative efficacy of peptide and recombinant DNA vaccine, Cancer Detect. Prev.25 (2001) 470–478.

[14] Y. Bergman, J. Haimovich, Characterization of a carcinogen-induced murine Blymphocyte cell line of C3H/eB origin, Eur. J. Immunol. 7 (1977) 413–417.

[15] D.A. Dean, D.D. Strong, W.E. Zimmer, Nuclear entry of nonviral vectors, GeneTher. 12 (2005) 881–890.

[16] J.M. McMahon, E. Signori, K.E. Wells, V.M. Fazio, D.J. Wells, Optimisation ofelectrotransfer of plasmid into skeletal muscle by pretreatment withhyaluronidase – increased expression with reduced muscle damage, GeneTher. 8 (2001) 1264–1270.

[17] M.P. Lefranc, C. Pommie, M. Ruiz, V. Giudicelli, E. Foulquier, L. Truong, V.Thouvenin-Contet, G. Lefranc, IMGT unique numbering for immunoglobulinand T cell receptor variable domains and Ig superfamily V-like domains, Dev.Comp. Immunol. 27 (2003) 55–77.

[18] H. Rammensee, J. Bachmann, N.P. Emmerich, O.A. Bachor, S. Stevanovic,SYFPEITHI: database for MHC ligands and peptide motifs, Immunogenetics 50(1999) 213–219.

[19] J. Hartikka, M. Sawdey, F. Cornefert-Jensen, M. Margalith, K. Barnhart, M.Nolasco, H.L. Vahlsing, J. Meek, M. Marquet, P. Hobart, J. Norman, M.Manthorpe, An improved plasmid DNA expression vector for direct injectioninto skeletal muscle, Hum. Gene Ther. 7 (1996) 1205–1217.

[20] M.J. Campbell, L. Esserman, R. Levy, Immunotherapy of established murine Bcell lymphoma. Combination of idiotype immunization andcyclophosphamide, J. Immunol. 141 (1988) 3227–3233.

[21] M.S. Kaminski, K. Kitamura, D.G. Maloney, R. Levy, Idiotype vaccination againstmurine B cell lymphoma. Inhibition of tumor immunity by free idiotypeprotein, J. Immunol. 138 (1987) 1289–1296.

[22] S. Dabadghao, S. Bergenbrant, D. Anton, W. He, G. Holm, Q. Yi, Anti-idiotypic T-cell activation in multiple myeloma induced by M-component fragmentspresented by dendritic cells, Br. J. Haematol. 100 (1998) 647–654.

[23] S. Muraro, A. Bondanza, M. Bellone, P.D. Greenberg, C. Bonini, Molecularmodification of idiotypes from B-cell lymphomas for expression in maturedendritic cells as a strategy to induce tumor-reactive CD4+ and CD8+ T-cellresponses, Blood 105 (2005) 3596–3604.

[24] G. Singh, S. Parker, P. Hobart, The development of a bicistronic plasmid DNAvaccine for B-cell lymphoma, Vaccine 20 (2002) 1400–1411.

[25] A.D. Syrengelas, T.T. Chen, R. Levy, DNA immunization induces protectiveimmunity against B-cell lymphoma, Nat. Med. 2 (1996) 1038–1041.

[26] M.J. Campbell, L. Esserman, N.E. Byars, A.C. Allison, R. Levy, Idiotype vaccinationagainst murine B cell lymphoma. Humoral and cellular requirements for the fullexpression of antitumor immunity, J. Immunol. 145 (1990) 1029–1036.

[27] A.D. Syrengelas, R. Levy, DNA vaccination against the idiotype of a murine Bcell lymphoma: mechanism of tumor protection, J. Immunol. 162 (1999)4790–4795.

[28] M. Miller, G. Rekas, K. Dayball, Y.H. Wan, J. Bramson, The efficacy ofelectroporated plasmid vaccines correlates with long-term antigenproduction in vivo, Vaccine 22 (2004) 2517–2523.

[29] S. Tollefsen, T. Tjelle, J. Schneider, M. Harboe, H. Wiker, G. Hewinson, K. Huygen, I.Mathiesen, Improved cellular and humoral immune responses againstMycobacterium tuberculosis antigens after intramuscular DNA immunisationcombined with muscle electroporation, Vaccine 20 (2002) 3370–3378.

[30] K.R. Irvine, J.B. Rao, S.A. Rosenberg, N.P. Restifo, Cytokine enhancement of DNAimmunization leads to effective treatment of established pulmonarymetastases, J. Immunol. 156 (1996) 238–245.

[31] I. Hakim, S. Levy, R. Levy, A nine-amino acid peptide from IL-1beta augmentsantitumor immune responses induced by protein and DNA vaccines, J.Immunol. 157 (1996) 5503–5511.