Transfected human dendritic cells to induce antitumor immunity - Nature

Gene Therapy (2000) 7, 1458–1466  2000 Macmillan Publishers Ltd All rights reserved 0969-7128/00 $15.00 www.nature.com/gt

CELL-BASED THERAPY

RESEARCH ARTICLE

Transfected human dendritic cells to induce antitumor immunity A Rughetti1, M Biffoni1, M Sabbatucci1, H Rahimi1, I Pellicciotta1, A Fattorossi2, L Pierelli2, G Scambia2, M Lavitrano1, L Frati1 and M Nuti1 1

Department of Experimental Medicine and Pathology, Universita` di Roma ‘La Sapienza’, Rome; and 2Institutes of Haematology and Gynaecology, Universita` Cattolica, Rome, Italy

Dendritic cells are professional antigen-presenting cells able to prime naive T lymphocytes and regulate steadily the delicate balance between tolerance and activation during the immune response. In past years several reports have shown that genetically engineered dendritic cells (DCs) can be a powerful tool for inducing an antigen-specific immune response. The use of such modified antigen-presenting cells is a real working hypothesis in preclinical studies and in clinical vaccination approaches for cancer treatment. The definition of optimal transfection conditions for preserving DC survival and functionality is necessary to design a correct immunotherapeutic protocol. Different lipid-based transfec-

tion compounds were studied for their effects on DC survival, phenotype and functional properties. All the transfection procedures were able to select DCs with a higher expression of activation and costimulatory molecules (ie MHCII-DR, CD83, CD86, CD25) than the untreated DCs. However, only two compounds (LipofectAMINE PLUS and FuGENE 6), preserved or even increased the immunopotency of DCs as antigen-presenting cells. These protocols were applied to modify DCs in order to express an epithelial tumor-associated antigen, MUC1, and such cells were able to induce in vitro a specific immune response in healthy donors. Gene Therapy (2000) 7, 1458–1466.

Keywords: cancer vaccine; human dendritic cells; transfection

Introduction Immunotherapy is an attractive strategy to be used as an alternative or in combination with radiotherapy and chemotherapy for the cure of cancer.1 One of the most widely employed approaches involves the use of specialized antigen-presenting cells such as dendritic cells (DCs) to activate an effective antitumor cytotoxic T lymphocyte (CTL) response.2 CTL responses appear, in fact, to be crucial in tumor cell rejection and have been shown to be capable of eradicating established cancers in experimental models.3 Genetic immunization using naked DNA induces both humoral and cellular responses that are apparently initiated by DCs, either by taking up proteins secreted by other neighboring cells or by direct transfection.4,5 As recently shown, although just a few DCs appear to be effectively transfected during DNA vaccination, this induces a general activation of all DCs found in the draining lymph nodes providing optimal conditions for T cell stimulation.6 DCs are supposed to be activated to optimal differentiation by several ‘danger molecules’ such as inflammatory cytokines or bacterial products.7 The initial up-regulation of surface MHC class I and II molecules is followed by that of costimulatory molecules, indispensable for the priming of naive T cells.8 In vivo Correspondence: M Nuti, Department of Experimental Medicine and Pathology, Universita` di Roma ‘La Sapienza’, Viale Regina Elena 324, 00161 Rome, Italy Received 18 February 2000; accepted 30 May 2000

this process is associated with defined migratory pathways from the site of the initial stimuli to the local lymph node.9–11 The development of relative simple and reproducible methods to in vitro differentiate DCs from blood precursors has, moreover, increased interest in the use of these cells as cell-based vaccine in a clinical setting. Genetically modified DCs can potentially activate a strong immune response against tumor cells and overcome tolerance established towards those antigens that are self proteins overexpressed by the transformed cells.12,13 In the present study we have evaluated the immunological potency of DCs transfected using several lipid formulations. The effects of seven different lipid-based transfection compounds on cell viability, phenotype expression and immunopotency of DCs, were compared and optimal protocol defined. Such an experimental procedure was applied to the cancer self antigen MUC1 that is normally expressed on the luminal surface of epithelial cells.14 During tumor transformation MUC1 is overexpressed and undergoes a deficient glycosylation process leading to the exposure of cryptic epitopes thus resulting as a ‘foreign antigen’.15,16 MUC1-transfected DCs were able to express the protein and activate human PBMCs in vitro.

Results Effect of the transfection compound exposure on DCs morphology and phenotype In order efficiently to mediate cDNA transfection in DCs we assayed several lipid compounds as reported in Table

Human dendritic cells for cancer vaccine A Rughetti et al

1. DCs differentiated from PBMCs after 4 days culture with IL-4 and GM-CSF were exposed for 3 or 4 h to the DNA/lipidic transfection complexes as suggested by the manufacturers using as DNA the pMRS 30 vector (kindly provided by Menarini Ricerche). The percentage of living cells 48 h after treatment was assessed by trypan blue dye exclusion test and the morphology of the cells observed by phase contrast microscopy. The transfection compounds caused cell mortality ranging from 85% (CLONfectin) to 38% (LipofectAMINE). DMRIE-C, LipofectAMINE, LipofectAMINE PLUS and FuGENE 6 treatments resulted in the least cell death, while DOTAP exerted a more pronounced cytotoxic effect; CLONfectin and Lipofectin were highly cytotoxic and were therefore no longer used in this study. The morphology of the DCs is shown in Figure 1: untreated DCs show the typical morphology as single or clustered large cells in suspension with few adherent cells (Figure 1a and b). Parallel cultures of DCs, 48 h after exposure to the transfection complexes, showed a higher percentage of adherent cells displaying marked dendritic morphology (elongated shape) together with loosely adherent cell clumps. DCs treated with the lipid compound FuGENE 6 (Figure 1c and d) showed a less altered morphology as compared with DCs treated with LipofectAMINE PLUS (Figure 1e and f). DOTAP and DMRIE C liposome compounds behaved similarly to the latter (data not shown). The morphological changes were immediately visible within 6 h after transfection. In order to establish optimal timing for the generation of transfected DCs expressing a mature phenotype, experiments were conducted with DC cultures between days 3 and 4 using selected lipidic compounds (DMRIEC, LipofectAMINE PLUS, DOTAP and FuGENE 6) chosen on the basis of the previous viability experiments. As depicted in Figures 2 and 3 (upper row), adherent cells exposed to GM-CSF and IL-4 for 5 or 6 days respectively, expressed the typical DC-associated markers CD1a, CD11b, CD83, CD86 and HLA-DR, while concomitantly down-regulating the monocyte-associated marker CD14. The exposure of DC preparations to the various transfection compounds, induced the expression of the activation marker CD25 and, most importantly, the up-regulation

Table 1 Effect of different liposomes on DC viability Composition

DMRIE-C

DNA/Liposome % Cell survivala (w/w)

DMRIE/Cholesterol 1:1 (M/M) LipofectAMINE DOSPA/DOPE 3:1 (w/w) LipofectAMINE Lipofectamine/ PLUS Plus Reagent 3:2 (v/v) Lipofectin DOTMA/DOPE 1:1 (w/w) DOTAP DOTAP FuGENE 6 Lipids, non-liposomal formulation CLONfectin CLONfectin a

1:12

61.93 ± 4.9

1:8

62.2 ± 7.1

1:8

58.6 ± 3.56

1:8

20 ± 4.2

1:5 1:3

33.35 ± 11.65 54.63 ± 6.84

1:8

15 ± 3.52

Cell survival was assessed by trypan blue dye exclusion test. % refers to the mean of five distinct experiments.

1459

Figure 1 Morphological changes occurring in DCs after exposure to lipid transfection reagents. Untreated DCs (a, b), FuGENE 6 (c, d) and LipofectAMINE PLUS (e, f)-treated dendritic cells at day 6 of culture 48 h from transfection. The pictures were taken at 20× original magnification (a, c, e) and 50× (b, d, f). Transfected DC cultures show different degrees of activation in the morphology as a result of the stressful effect exerted by the lipid reagents.

of the level of CD83, a marker of DC lineage,17 and of the levels of those molecules involved in antigen presentation (HLA-DR) and T cell costimulation (CD86). This phenomenon was mostly evident with DMRIE-C and LipofectAMINE Plus, albeit other compounds had similar effects.

Effect of DNA/lipid exposure on stimulatory DC capacity To evaluate whether the liposome treatment would affect the immunostimulatory potential of DCs as a consequence of surface marker expression changes observed, DCs exposed to transfection reagent were compared with untreated DCs for their ability to stimulate alloreactive T cells. Fixed number of allogeneic T cells were cultured with different number of DCs treated or not treated with FuGENE 6 or LipofectAMINE PLUS. A number of DCs ranging from 5 × 103 to 10 × 103 was sufficient to induce the maximum of the allogeneic response as shown for two different samples in Figure 4a and b. Moreover, liposome-treated DCs appeared to be equivalent to (Figure Gene Therapy

Human dendritic cells for cancer vaccine A Rughetti et al

1460

Figure 2 DC phenotypic profile after exposure to different transfection compounds. DCs were generated from blood-derived adherent leukocytes, transfected at different times of culture (72 and 96 h for experiments 1 and 2, respectively) and different marker expression tested after 48 h from transfection. Dot plots representing changes occurring in the coordinate expression of CD1a and HLA-DR are on the left side of the figure. On the right side are dot plots representing the behavior of CD14 and CD11b in one representative experiment (experiment 2). Cursor lines indicate staining level of isotype control preparations.

4a) or even better antigen-presenting cells as compared with untreated DCs (Figure 4b). In order to characterize the variability observed among the donors, several MLR assays were conducted. Figure 4c shows the average of five different MLR experiments where the allogeneic responses was measured as 3H-thymidine incorporation and reported as a percentage considering as 100% the allogeneic response of the untreated DCs used as stimulators. Results indicate that the transfection reagents LipofectAMINE PLUS and FuGENE 6 induce a stronger allogeneic response than untreated DCs, whereas DOTAP and DMRIE-C reagents maintain or decrease, respectively, the APC capability of DCs. The ability of transfected DCs to induce a specific immune response to the transfected antigen in autologous system was also studied (Figure 5). Since bacterial DNA sequences are known to be strongly immunogenic in humans,18–20 we have used the mammalian expression vector pMEP4 containing the hygromicine resistance gene to transfect DCs and assess the resulting immunogenicity in an autologous system. DCs at day 4 of culture were transfected with pMEP4 using FuGENE 6. Fortyeight hours after transfection a fixed number of autologous T cells were cultured with different numbers of transfected DCs and the proliferative response evaluated as 3H-thymidine incorporation after 7 days. As control cultures, T cells were cultured with different numbers of untransfected DCs or DCs exposed to the transfection lipid reagent alone. A significant proliferatGene Therapy

Figure 3 Modifications induced by the different transfection compounds on the expression of the DC-associated markers (CD25, CD83 and CD86. In all plots, empty and filled histograms represent the staining level of the isotype control or the indicated monoclonal antibody, respectively. The amount of the rightward shift of the filled histograms on the x-axis reflects the intensity of the staining, while on the y-axis the number of cells are reported. Data refer to experiment 2 in Figure 2.

ive response can be observed when pMEP4-transfected DCs were used as stimulators in numbers ranging from 2 × 103 to 10 × 103 cells. The FuGENE 6 protocol seems to give a better performance combining transfection efficiency, morphology, phenotype and functional efficiency of DCs.

Expression of MUC1 tumor-associated epitopes in DC transfected with MUC1 cDNA sequences We applied transfection procedures employing LipofectAMINE PLUS and FuGENE 6 reagents to transfect the tumor-associated antigen MUC1 cDNA in DCs for evaluating the immune potential of such genetically engineered DCs. We used the episomal vector pMRS 169 (kindly provided by Menarini Ricerche) containing a nucleotide sequence coding for a stretch of the MUC1 extracellular domain encompassing one tandem repeat. The leader sequence is deleted from the construct, thus the miniprotein is not targeted to membrane expression. The efficiency of transfection was assayed by RT-PCR as shown in Figure 6. The specific amplification product (lane 2) was present only in the samples where the template was cDNA from pMRS 169-transfected DCs (lanes 6, 10). Amplification reaction did not occur in the absence of reverse transcriptase (lanes 5, 7, 9) or when the template was mRNA extracted from DCs transfected with the mock expression vector pMRS 30 (lanes 4, 8). Expression of the transfected MUC1 in DCs was

Human dendritic cells for cancer vaccine A Rughetti et al

1461

Figure 6 RT-PCR analysis of transfected MUC1 mRNA in DCs from PBMCs. Two cultures of DCs from two different healthy donors (DCs A and DCs B) were transfected with the mock vector pMRS 30 and with MUC1 expressing vector pMRS 169 RNA was extracted and first strand cDNA synthesized. RNA and cDNA samples were then used as template for PCR amplification as described in Materials and methods. Amplification products were then visualized after electrophoresis in agarose gel 1.4%. Lanes 1, 11: 100 bps ladder marker (Gibco); lane 2: amplification reaction positive control using pMRS 169 as template; lanes 3, 7: RNA from pMRS 30 transfected DCs; lanes 4, 8: cDNA from pMRS transfected DCs; lanes 5, 9: RNA from pMRS 169 transfected DCs; lanes 6, 10: cDNA from pMRS 169 transfected DCs. Specific amplification product size was 355 bps.

Figure 4 Immunogenicity of transfected DC. The ability of untreated or lipid-treated DCs to induce a proliferative response in MLR was tested. (a, b) Different numbers of untreated or transfected dendritic cells from PBMCs of healthy donors were used as antigen-presenting cells and cultured with a fixed number of monocyte-depleted PBMCs (105). Lipofect AMINE PLUS and FuGENE 6-treated DCs were able to induce a similar (a) or even stronger proliferative response (b) than untreated DCs. (c) The median values of the percentage of the proliferative response of 105 T cells to 104 DMRIE-C, LiofectAMIN E PLUS, DOTAP and FuGENE 6treated dendritic cells in five different experiments were reported compared with the proliferative response to allogeneic untreated DCs considered as 100%. FuGene 6 or LipofectAMINE PLUS-transfected DCs showed an increased immunogenicity.

Figure 5 Transfected DCs as antigen-presenting cells in an autologous system. 5 × 104 monocyte-depleted PBMCs were used as responder in autologous MLR and co-cultured for 7 days with different numbers of DCs transfected with pMEP4 vector/FuGENE 6 complexes as stimulators. As negative control untransfected or FuGENE 6-transfected DCs were used. As few as 5 × 103 pMEP4-transfected DCs were able to induce a strong and significant proliferative response.

observed by indirect immunofluorescence staining with MoAbs directed against the TR sequence as shown in Figure 7. DC transfection levels could be estimated by counting MUC1-positive DCs by immunofluorescence staining. Variability was observed among different

experiments ranging from 8 to 15% of positive transfected DCs. Immunofluorescence staining was weakly diffuse in the cytoplasm and bright when localized in vacuoles within the periphery of the transfected cells (Figure 7a and b). Only cells with a size and morphology compatible with DCs were stained by anti-MUC1 MoAbs, whereas contaminating lymphocytes were not. The fine mapping of the possible MUC1 peptides originating from processing was achieved by using a panel of several MoAbs directed against MUC1 core protein and recognizing distinct epitopes within the TR sequence (Table 2). Three of six MoAbs (MoAb SM3, HMFG1, HMFG2) directed against the core protein of the MUC1 gene failed to react with the MUC1-transfected protein suggesting that the epitope recognized by the MoAbs was not available for recognition by the specific monoclonal antibodies.

MUC1-transfected DCs can induce a specific immune response in healthy donors The ability to induce a MUC1-specific in vitro immune response in healthy donors was tested in autologous proliferation assays using MUC1 DCs as APCs. In Figure 8 the results of six different experiments are reported. DCs were transfected at day 4 of culture and used in autologous MLR at day 6, after 48 h from transfection using the transfection protocol FuGENE 6. Untransfected DCs or mock-transfected DCs were used as control. Autologous cultures were carried out in the presence or in the absence of low doses of IL-2 (20 IU/ml). Primary proliferative responses were observed in two of six donors and in those cases the response was amplified by the addition of IL-2.

Discussion A specific and efficacious antitumor response can be achieved only by activating the different arms of the immune Gene Therapy

Human dendritic cells for cancer vaccine A Rughetti et al

1462

Figure 8 Primary proliferative response to MUC1-transfected DCs in healthy donors. The immunogenicity of MUC1-transfected DCs was tested in autologous MLR and evaluated as 3H-thymidine incorporation after 16 h at day 6 of culture. 105T cells were cultured with 104-transfected DCs (pMRS 30-DCs or pMRS 169 DCs, respectively) with or without IL-2. Results were plotted as proliferation index, obtained as ratio between c.p.m. of pMRS 169 DCs/T cell culture and c.p.m. of pMRS 30/T cell culture. As control, the proliferation of PBMCs alone or DCs alone was monitored. The presence of IL-2 in the cultures did not increase background proliferation level that was always beyond 1.5 × 103.

Figure 7 Immunofluorescence staining of MUC1-transfected DCs. DCs from PBMCs were transfected with the MUC1 expressing vector pMRS 169 after 4 days of culture and indirect immunofluorescence staining performed after 48 h. Acetone/methanol fixed dendritic cells were stained with MoAbs Ma552 and BC2 directed against the TR core protein sequence (a and b, respectively). In positive cells, immunofluorescence staining is mainly localized in vesicles at the periphery of the cells.

Table 2 MoAb Ma552 436 BC2 HMFG1 SM3 HMFG2 a

Expression pattern of transfected MUC1 in DCs Epitopea

Reactivity

GVTSA PDTRPAP RPAP APDTR A PDTR APDTRP DTR

+ + + − − −

In plain text: amino acid residues involved in antibody binding; bold text: amino acid sequence of the linear epitope; underlined text: antibody anchor amino acid residues.17

Gene Therapy

response, namely cellular and humoral immunity. In cancers, T cell activation appears to be crucial for tumor cell rejection and DCs as potent antigen-presenting cells perform a central role.21 Recently, DCs have become popular as potential vaccine components since it is now possible to culture them in large numbers.5,12 Genetically engineered DCs as well as antigen-pulsed DCs represent an interesting and attractive tool in immunotherapy.13 In this study, we demonstrate that DCs can be successfully transfected in vitro and that the transfection procedure itself may contribute to enhancing immunogenicity. Several liposome and non-liposome compounds were compared for the possible effects on DC viability, morphology, phenotype and function. Among the compounds tested, the optimal transfection reagent appears to be FuGENE 6, while DOTAP and LipofectAMINE PLUS, showed a more stressful effect on the cells, although giving satisfactory results (respectively, higher cell toxicity and morphological changes) (Table 1, Figure 1e and f). Phenotypic analysis of the transfected DCs performed by cytofluorimetry revealed that liposometreated DCs showed an increased expression of CD83, specific differentiation marker of the dendritic cell lineage, CD86, costimulator molecule B7.2, and DR antigens (Figures 2 and 3). The up-regulation of these molecules is a common feature of DCs in response to a potential danger signal such as bacterial products and/or inflammatory stimuli. During this process, DCs acquire maximum competence for presenting antigens and for the triggering of specific T cell response.11 Thus, lipid compounds may mimic a danger signal for DCs inducing maturation in vitro. Lipid-treated DCs (Figure 4) indeed showed an increased potency in inducing alloreactive response of T cells, particularly after treatment with LipofectAMINE PLUS or FuGENE 6. Despite the high up-regulation of the maturation markers, DMRIE-C treatment appears to alter cell functionality as shown by the decreased immunostimula-

Human dendritic cells for cancer vaccine A Rughetti et al

tory effect on PBMCs (Figure 4c). The up-regulation of the DC maturation markers after lipid treatment is, therefore, not necessarily indicative of optimal antigen presenting capability. The choice of transfection protocol must consider the result of an overall balance between direct cytotoxic effects and modulatory signals leading to DC maturation, possible consequences of the liposome treatment. So far, different approaches have been applied to obtain genetically modified DCs, including electroporation, calcium phosphate precipitation, gene gun, liposome complex treatment and recombinant viral vectors.22–24 In some cases the origin of DCs has been described as influencing susceptibility to the different transfection protocols.25 Dendritic cells have been genetically modified by using retroviral or adenoviral vectors to induce specific immune response against tumor antigens,26–28 although concern still remains regarding the use of such vectors in vaccination trials in humans. The use of liposomes has the advantage of being simple, safe and suitable for wide applications.29,30 Other investigators have addressed the issue of performing optimal transfection of DCs with liposomes. DCs were shown to be phenotypically and functionally unaffected by liposome treatment and able to induce MART1-specific CTLs.31 The use of genetically modified DCs for cancer treatment implies that several requirements must be considered: the source of DCs (monocyte or CD34+ progenitors); the simplicity of culture and transfection methods and at the same time efficiency in transgene expression and maintenance of DC functionality.32 Also, the antigen expression should be sufficient to allow antigen presentation to occur, but limited to avoid possible autoimmune side-effects.33 Transient transfection, in fact, appears to be sufficient to activate a specific long-term T cell response.6 In our protocol, genetically modified DCs after a 4 day culture, can be efficiently used as APC 48 h from transfection and are able to induce significant allogeneic (Figure 4) and autologous responses (Figure 5) even when used in a ratio of 1:200. The finding that liposome treatment may modulate APC phenotype (ie induction of an activated dendritic cell phenotype) can be viewed as a ‘natural adjuvant’ effect for the induction of a specific and effective antitumor response. We tested this experimental hypothesis using the human MUC1 gene as a prototype of tumorassociated self antigen. MUC1 is able to induce humoral and cellular immune responses when overexpressed and aberrantly glycosylated as in tumor epithelial cells and is considered an attractive target for the immunotherapy of epithelial cancers.34 Several experimental models and clinical trials have proposed the use of different MUC1 antigen formulations.35,36 We used the episomal vector pMRS 169 (kindly provided by Menarini Ricerche), containing a partial MUC1 cDNA coding for one tandem repeat deleted from the leader sequence, thus preventing membrane expression. Transient antigen expression using such an episomal vector is probably sufficient to prime the immune response, avoiding at the same time a possible autoimmune response induced by prolonged expression of the self antigen. Moreover, the use of such vectors can be

a suitable tool for clinical application. Transfection efficiency of pMRS 169-DCs was assessed by RT-PCR (Figure 6) and by immunofluorescence staining of transfected DCs (Figure 7a, b). The staining is diffuse in the cytoplasm and localized in vesicles (Figure 7). The absence of the leader sequence together with the shortened length of the MUC1 construct should increase cytoplasmic protein cleavage mediated by the proteasome and other peptidases present in the cytoplasm. Protein degradation could explain the different reactivity displayed by the several MoAbs utilised for the characterization of MUC1 expression (Table 2). The newly synthesized MUC1 protein might be rapidly degraded and the resulting peptides can be recognized by MoAbs Ma 552, BC2 (Figure 7) and 436 (Table 2) only. The unreactive MoAbs may need important amino acid residues for binding that are missed in the cleaved peptides produced (within the DTR region).37 All the MUC1 peptides recognised by the MoAbs in the transfected DCs appear either to maintain the RPAP sequence at the N terminus allowing binding of MoAbs 436 and Ma552 or to contain other antibody anchor amino acids (P residue for binding of MoAb BC2). This hypothesis is also in agreement with the requirements described for peptide transport in the E.R. lumen. In fact the presence of specific amino acid residues such as R, K, N, W in the first positions seem to be crucial for peptide selection by TAP transporter.38,39 MUC1-transfected DCs were able to induce a specific proliferative T cell response when used as antigenpresenting cells in 7 day autologous MLR assays. Such a primary response can be enforced when cytokine stimuli is subsequently added to the culture. Immunological response to the MUC1 antigen in healthy donors has been shown and correlated to possible previous expositions to the antigen as in pregnancy.40 It is interesting to notice that both the responders to MUC1DCs were females. These results, although preliminary, are suggestive of the efficacy of such MUC1-presenting APCs in inducing a specific response. They may also be indicative of a possible use of MUC1-engineered DCs for preclinical studies and eventually their transfer to the clinic for the formulation of immunotherapeutic approaches to cancer treatment.

1463

Materials and methods Dendritic cell generation DCs were generated from peripheral mononuclear cells (PBMCs) as previously described.41 Briefly, PBMCs from buffy coats of healthy donors, were isolated through a Ficoll–Hypaque gradient (1077g ml; Pharmacia, LKB, Uppsala, Sweden), washed three times in phosphate buffer saline (PBS) without Ca++ and Mg++ (Hyclone, Logan, UT, USA) and resuspended at a concentration of 6 × 106/ml in RPMI 1600 (Hyclone) supplemented with 5% fetal bovine serum (FBS) (Hyclone). Cells were seeded in T25 flasks (Falcon, Becton Dickinson, San Jose´, CA, USA) and left to adhere for 2 h at 37°C. Non-adherent cells were then removed by washing gently twice using PBS with Ca++ and Mg++ (Hyclone) and then frozen in Cryoserv (Research Industries Corporation, Salt Lake City, UT, USA) and FBS (1:9) at 25 × 106 cells/ml. AdherGene Therapy

Human dendritic cells for cancer vaccine A Rughetti et al

1464

ent cells were cultured in RPMI 1600 supplemented with 2 mm glutamine (Sigma Chemical Company, St Louis, MO, USA), 100 U/ml penicillin, 100 ␮g/ml streptomycin, 5% FBS with a cytokine cocktail of 1000 U/ml IL-4 kindly provided by Dr A Lanzavecchia (Roche Institute, Basel, Switzerland) and 50 ng/ml granulocyte-macrophage colony-stimulating factor (GM-CSF) (Molgrastim; ScheringPlough, Milan, Italy) at 37°C, 5% CO2. In these culture conditions the adherent cell fraction gives rise to a dendritic cell population within 7 days. Alternatively, monocytes were purified from PBMCs through a Percoll gradient (Pharmacia) and plated at 30 × 103 cells/ml. Trypan blue dye exclusion test was routinely used to assess cellular viability when cells were thawed or counted. Pictures were taken with a Zeiss microscope acquired as digital images with a CCD camera Sensys (Photometrics, Tucson, AZ, USA) using the IPLab software (Scanalytic, Firefax, VA, USA). Buffy coats were obtained after informed consent and the study was approved by the hospital investigational review board.

(200 ␮l/106 cells) and left for 1 min at RT. Optimem medium (Gibco-Life Technologies) was then added (800 ␮l/106 cells) and cells were left at 37°C for 3–4 h, ie the minimum time as suggested by the manufacturer’s recommendation. For transfection experiments the following plasmids were used: pMEP4 (Invitrogen, San Diego, CA, USA), pMRS 30 and pMRS 169 both kindly provided by Menarini Ricerche (Pomezia, Italy). The latter two were eukaryotic expression vectors based on the pMRS 79,42 ie pMRS 30 as mock vector, while the pMRS 169 contains 500 bp insert coding for a portion of the MUC1 extracellular domain with one TR and without a leader sequence (manuscript in preparation). The plasmids used in transfection experiments were extracted and purified from bacterial cultures by Endotoxin Free Qiagen Maxi Prep kit (Qiagen Ltd, Crawley, UK).

FACS analysis Transfected or untransfected DCs were routinely tested for phenotype in flow cytometry 48 h from transfection, evaluating the expression of dendritic–monocyte differentiation markers such as CD1a, CD83, HLA-DR, CD14, CD11b and activation markers such as CD25 and CD86. Cell staining was performed using mouse monoclonal antibodies (MoAbs) conjugated with fluorescein isothiocyanate (FITC) or phycoerythrin (PE). The FITC-conjugated MoAbs used were: HB15A (IgG2b, anti-CD83), BL 6 (IgG1␬, anti-CD1a), Bear 1 (IgG1, anti-CD11b) from Immunotech (Marseille, France); M-A251 (IgG1␬, antiCD25), 2331(FUN-1) (IgG␬, anti-CD86) from Pharmingen (San Diego, CA, USA). The PE-conjugated MoAbs used were: B-F1 (IgG␬, anti-HLA class II DR) and B-A8 (IgG1␬, anti-CD14) (Biosource International, Camarillo, CA, USA). PE/FITC isotype-matched mouse immunoglobulins were used as negative control (Immunotech). Briefly, DCs were harvested, washed twice and incubated with the fluorochrome-conjugated monoclonal antibodies for 30 min at 4°C in the dark. Then, cells were washed twice, samples were acquired using a FACScan (Becton Dickinson) and analysis was performed using the Lysis II software.

MLRs and autologous MLRs Responding T cells (105) from allogeneic PBMCs purified by Percoll gradient were cultured in 96-well U-bottom microplates (Costar, Cambridge, MA, USA) in the presence of different numbers of irradiated (3000 rads from a 137Cs source) transfected or untransfected DCs as stimulator (from 0 to 5 × 104 DCs per well). 105 Lymphocytes per well were used in all the experiments where a fixed number of APCs was used. Thymidine incorporation was measured on day 4 or 5 by a 16 h pulse with 3H-thymidine (5 ␮Ci/ml, sp. act. 247.9 GBq/mmol, 37 MBq/ml; NEN, Zaventem, Belgium). The cells were then harvested and the incorporated radioactivity measured by using a ␤-counter (TopCount Packard Instruments BioSciences, Milan, Italy). A similar procedure was used when autologous MLRs were set up. In the titration experiment autologous DCs were transfected with pMEP4 plasmid following the FuGENE 6 transfection protocol and used as stimulators for autologous T cells (5 × 104 per well). Autologous DCs untreated or treated with FuGENE 6 alone were used as negative control. The ability of DCs to induce a proliferative response to MUC1 antigen was assayed in autologous MLRs where pMRS 169 transfected DCs were compared with pMRS 30 transfected DCs as APCs. In the autologous MLR assay, IL-2 (Genzyme, Cambridge, MA, USA) was also added at 20 IU/ml after 72 h.

Transfection of DCs Several lipidic formulations were used in this study as reported in Table 1. The liposome compounds used were: CLONfectin (Clontech, Palo Alto, CA, USA), DOTAP (Roche Diagnostic, Milan, Italy); Lipofectin, LipofectAMINE, LipofectAMINE PLUS and DMRIE-C (Gibco-Life Technologies, Paisley, UK). The FuGENE 6 lipidic transfection reagent (Roche Diagnostic) was also used. The DNA/lipid compound ratio and the incubation time were selected according to the manufacturer’s recommendation. Optimem Medium (Gibco-Life Technologies) was used as transfection medium. For Dotap and FuGENE 6 the transfection medium used was 20 mm Hepes and 12.5 mm Hepes Optimem Medium, respectively. Briefly, DCs were harvested, washed twice in PBS without Ca++ and Mg++ (Hyclone) and then the pellet was gently resuspended in the DNA–liposomal complexes

Immunofluorescence staining DCs were harvested 48 h after transfection, washed twice in PBS without Ca++ and Mg++ (Hyclone) and resuspended in 2 × 106/ml. The cells were then spread on to slides (50 ␮l cell suspension per slide) and left to dry overnight at RT. The samples were fixed in acetone/methanol (1:1, pre-cooled to −20°C) for 3 min at −20°C. The samples were then processed for immunofluorescence staining or stored at −20°C until use. Briefly, after blocking with 10% FBS/PBS for 20 min, cells were incubated with the appropriate first MoAb and then mouse immunoglobulins were detected with FITC-conjugated goat anti-mouse F(ab)2 Ig (Jackson ImmunoResearch, West Grove, PA, USA). Slides were then washed and mounted with cover slip in Vectashield (Vector Laboratories, Burlingame, CA, USA). The MoAbs used for the immunostaining of MUC1 protein after transfection

Gene Therapy

Human dendritic cells for cancer vaccine A Rughetti et al

in DCs were directed against the tandem repeat core sequences. The MoAbs HMFG1,43 HMFG243 and SM344 were a kind gift of Dr J Taylor-Papadimitriou (Imperial Cancer Research Fund, London, UK); the MoAb BC245 was kindly provided by Dr IFC McKenzie (Austin Research Institute, Hidelberg, Australia); the MoAb Ma 55246 was provided by Dr O Nilsson (CanAg Diagnostics AB, Gothenburg, Sweden). The MoAb 43647 was generated in our laboratory. As negative control the MoAb 9E10,48 a kind gift of Dr G Evan (ICRF), directed against a cryptic peptide sequence of c-myc was used.

RT-PCR RNA was obtained from dendritic cells transfected with FuGENE 6 and the mock expression vector pMRS 30 or the pMRS 169 containing the cDNA of the TR sequence. DCs were harvested 36 h after transfection and washed three times in cold PBS without Ca++ and Mg++ (Hyclone). The cell pellet was then resuspended in Trizol (Gibco BRL-Life Technologies) (106cells/200 ␮l). RNA was extracted following the procedure suggested by the manufacturer. The RNA obtained was resuspended in distillate sterile water and RNA concentration and purity was evaluated as OD at 260 A and as a ratio of 260/280 A OD, respectively. The first strand of cDNA was synthesised from 2 ␮g of RNA. Briefly, the RNA was incubated with 50 pm Oligo dT and 200 pm Random Primers (Gibco BRL-Life Technologies) in water and kept at 70°C for 10 min, samples were then cooled in ice, centrifuged and left at RT for 10 min to allow annealing of the primers to the RNA strands to occur. The reaction buffer was then added to the samples for a final concentration of 50 mm Tris-HCl (pH 8.3), 75 mm KCl, 3 mm MgCl2, 10 mm DDT, 0.025 U/␮l RNaseOUT (Gibco BRL-Life Technologies) in a final volume of 20 ␮l. Samples were heated at 42°C for 2 min, RNAse H Reverse Transcriptase (SuperScript II; Gibco BRL-Life Technologies) was added (10 U/␮l) and the cDNA synthesis was left at 42°C for 1 h. The RNAse H Reverse transcriptase was then inactivated, heating the samples at 70°C for 15 min. Two ␮l of the RT reaction were used as template for a polymerase chain reaction (PCR) performed in a final volume of 50 ␮l 2.5 mm MgCl2, 0.25 mm dNTPs, 20 mm Tris-HCl (pH 8.4), 50 mm KCl, specific primers at 20 pm/␮l and 2.5 U per sample of DNA Taq Polymerase AmpliTaq Gold (Perkin Elmer, Roche Molecular System, Branchburg, NJ, USA). The nucleotide sequences of the primers used were: 5⬘CAGGATGTCACTCTG3⬘ and 5⬘GACCAGAGTAGAAGC3⬘, the expected size of the amplified product was 355 bps. The PCR was carried out with the following protocol: 1 cycle at 94°C for 10 min, and then 30 cycles 1 min at 94°C for, 45 s at 50°C and 90 s at 72°C and the last cycle at 72°C for 15 min. The same procedure was used to amplify the specific cDNA from the control plasmid pMRS 30. The PCR products were monitored in 1.5% agarose gel using as marker 100 bps (Gibco BRL-Life Technologies). All the reactions were carried out in Gene Amp PCR 2400 (Perkin-Elmer, Beaconsfield, UK).

Acknowledgements We thank Dr A Lanzavecchia for the kind gift of the IL4.6 cell line and Dr J Taylor-Papadimitriou for providing the MoAbs HMFG1, HMFG2 and SM3. The MoAbs BC2 and

Ma552 were generous gifts of Dr IFC McKenzie and Dr O Nilsson (CanDiagnostic), respectively. We thank Menarini Ricerche for providing us with the expression vectors pMRS 30 and pMRS 169. This work was supported by Associazione Italiana per la Ricerca sul Cancro (AIRC), European Communities Fifth Program QLRT-1999-00217 and Ministero per l’Universita e Ricerca Scientifica (MURST 40% and 60%).

1465

References 1 Sogn JA. Tumor immunology: the glass is half full. Immunity 1998; 9: 757–763. 2 Colaco CA. Why are dendritic cells central to cancer immunotherapy? Mol Med Today 1999; 5: 14–17. 3 Gilboa E, Nair SK, Lyerly HK. Immunotherapy of cancer with dendritic-cell-based vaccines. Cancer Immunol Immunother 1998; 46: 82–87. 4 Cella M, Sallusto F, Lanzavecchia A. Origin, maturation and antigen presenting function of dendritic cells. Curr Opin Immunol 1997; 9: 10–16. 5 Tarte K, Klein B. Dendritic cell-based vaccine: a promising approach for cancer immunotherapy. Leukemia 1999; 13: 653– 663. 6 Akbari O et al. DNA vaccination: transfection and activation of dendritic cells as key events for immunity. J Exp Med 1999; 189: 169–178. 7 Rescigno M et al. Coordinated events during bacteria-induced DC maturation. Immunol Today 1999; 20: 200–203. 8 Lanzavecchia A. From antigen presentation to T-cell activation. Res Immunol 1998; 149: 626. 9 Dieu-Nosjean MC, Vicari A, Lebecque S, Caux C. Regulation of dendritic cell trafficking: a process that involves the participation of selective chemokines. J Leukoc Biol 1999; 66: 252–262. 10 Sallusto F et al. Distinct patterns and kinetics of chemokine production regulate dendritic cell function. Eur J Immunol 1999; 29: 1617–1625. 11 Rescigno M, Granucci F, Ricciardi-Castagnoli P. Dendritic cells at the end of the millennium. Immunol Cell Biol 1999; 77: 404–410. 12 Liu M. Transfected human dendritic cells as cancer vaccines. Nat Biotechnol 1998; 16: 335–336. 13 Nair SK. Immunotherapy of cancer with dendritic cell-based vaccines. Gene Therapy 1998; 5: 1445–1446. 14 Patton S, Gendler SJ, Spicer AP. The epithelial mucin, MUC1, of milk, mammary gland and other tissues. Biochim Biophys Acta 1995; 1241: 407–423. 15 Taylor-Papadimitriou J, Finn OJ. Biology, biochemistry and immunology of carcinoma-associated mucins. Immunol Today 1997; 18: 105–107. 16 Agrawal B, Gendler SJ, Longenecker BM. The biological role of mucins in cellular interactions and immune regulation: prospects for cancer immunotherapy. Mol Med Today 1998; 4: 397– 403. 17 Zhou LJ, Tedder TF. Human blood dendritic cells selectively express CD83, a member of the immunoglobulin superfamily. J Immunol 1995; 154: 3821–3835. 18 Kumar V, Sercarz E. Genetic vaccination: the advantages of going naked. Nature Med 1996; 2: 857–859. 19 Tighe H, Corr M, Roman M, Raz E. Gene vaccination: plasmid DNA is more than just a blueprint. Immunol Today 1998; 19: 89–97. 20 Jakob T et al. Activation of cutaneous dendritic cells by CpGcontaining oligodeoxynucleotides: a role for dendritic cells in the augmentation of Th1 responses by immunostimulatory DNA. J Immunol 1998; 161: 3042–3049. 21 Timmerman JM, Levy R. Dendritic cell vaccines for cancer immunotherapy. Annu Rev Med 1999; 50: 507–529. 22 Arthur JF et al. A comparison of gene transfer methods in human dendritic cells. Cancer Gene Ther 1997; 4: 17–25. Gene Therapy

Human dendritic cells for cancer vaccine A Rughetti et al

1466

23 Porter CD et al. Cationic liposomes enhance the rate of transduction by a recombinant retroviral vector in vitro and in vivo. J Virol 1998; 72: 4832–4840. 24 Timares L, Takashima A, Johnston SA. Quantitative analysis of the immunopotency of genetically transfected dendritic cells. Proc Natl Acad Sci USA 1998; 95: 13147–13152. 25 Van Tendeloo VF et al. Nonviral transfection of distinct types of human dendritic cells: high-efficiency gene transfer by electroporation into hematopoietic progenitor, but not monocytederived dendritic cells. Gene Therapy 1998; 5: 700–707. 26 Kaplan JM et al. Induction of antitumor immunity with dendritic cells transduced with adenovirus vector-encoding endogenous tumor-associated antigens. J Immunol 1999; 163: 699–707. 27 Perez-Diez A et al. Generation of CD8+ and CD4+ T-cell response to dendritic cells genetically engineered to express the MART-1/Melan-A gene. Cancer Res 1998; 58: 5305–5309. 28 Butterfield LH et al. Generation of melanoma-specific cytotoxic T lymphocytes by dendritic cells transduced with a MART-1 adenovirus. J Immunol 1998; 161: 5607–5613. 29 Lasic DD. Novel applications of liposomes. Trends Biotechnol 1998; 16: 307–321. 30 Templeton NS, Lasic DD. New directions in liposome gene delivery. Mol Biotechnol 1999; 11: 175–180. 31 Philip R et al. Transgene expression in dendritic cells to induce antigen-specific cytotoxic T cells in healthy donors. Cancer Gene Ther 1998; 5: 236–246. 32 Citterio S et al. Dendritic cells as natural adjuvants. Methods 1999; 19: 142–147. 33 Sallusto F, Lanzavecchia A. Mobilizing dendritic cells for tolerance, priming, and chronic inflammation. J Exp Med 1999; 189: 611–614. 34 Taylor-Papadimitriou J, Burchell J, Miles DW, Dalziel M. MUC1 and cancer. Biochim Biophys Acta 1999; 1455: 301–313. 35 Graham RA, Burchell JM, Beverley P, Taylor-Papadimitriou J. Intramuscular immunisation with MUC1 cDNA can protect C57 mice challenged with MUC1-expressing syngeneic mouse tumour cells. Int J Cancer 1996; 65: 664–670. 36 Acres B et al. MUC1-specific immune responses in human MUC1 transgenic mice immunized with various human MUC1 vaccines. Cancer Immunol Immunother 2000; 48: 588–594.

Gene Therapy

37 Schol DJ et al. ‘Epitope fingerprinting’ using overlapping 20-mer peptides of the MUC1 tandem repeat sequence. Tumour Biol 1998; 19 (Suppl. 1): 35–45. 38 Uebel S, Tampe R. Specificity of the proteasome and the TAP transporter. Curr Opin Immunol 1999; 11: 203–208. 39 Rock KL, Goldberg AL. Degradation of cell proteins and the generation of MHC class I-presented peptides. Annu Rev Immunol 1999; 17: 739–779. 40 Agrawal B, Reddish MA, Longenecker BM. In vitro induction of MUC-1 peptide-specific type 1 T lymphocyte and cytotoxic T lymphocyte responses from healthy multiparous donors. J Immunol 1996; 157: 2089–2095. 41 Sallusto F, Lanzavecchia A. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha. J Exp Med 1994; 179: 1109–1118. 42 Rotondaro L, Mele A, Rovera G. Efficiency of different viral promoters in directing gene expression in mammalian cells: effect of 3⬘-untranslated sequences. Gene 1996; 168: 195–198. 43 Taylor-Papadimitriou J et al. Monoclonal antibodies to epithelium-specific components of the human milk fat globule membrane: production and reaction with cells in culture. Int J Cancer 1981; 28: 17–21. 44 Burchell J et al. Development and characterization of breast cancer reactive monoclonal antibodies directed to the core protein of the human milk mucin. Cancer Res 1987; 47: 5476–5482. 45 Xing PX et al. Second generation anti-MUC1 peptide monoclonal antibodies. Cancer Res 1992; 52: 2310–2321. 46 Baeckstrom D et al. Discrimination of MUC1 mucins from other sialyl-Le(a)-carrying glycoproteins produced by colon carcinoma cells using a novel monoclonal antibody. Cancer Res 1993; 53: 755–761. 47 Nuti M et al. Characterization of monoclonal antibody 436 recognizing the Arg-Pro-Ala-Pro sequence of the polymorphic epithelial mucin (PEM) protein core in breast carcinoma cells. Int J Biol Markers 1992; 7: 71–79. 48 Evan GI, Lewis GK, Ramsay G, Bishop JM. Isolation of monoclonal antibodies specific for human c-myc proto-oncogene product. Mol Cell Biol 1985; 5: 3610–3616.