Immediate transfection of patient-derived leukemia: a novel source for generating cell-based vaccines
© Gershan et al; licensee BioMed Central Ltd. 2005
Received: 26 March 2005
Accepted: 21 June 2005
Published: 21 June 2005
The production of cell-based cancer vaccines by gene vectors encoding proteins that stimulate the immune system has advanced rapidly in model systems. We sought to develop non-viral transfection methods that could transform patient tumor cells into cancer vaccines, paving the way for rapid production of autologous cell-based vaccines.
As the extended culture and expansion of most patient tumor cells is not possible, we sought to first evaluate a new technology that combines electroporation and chemical transfection in order to determine if plasmid-based gene vectors could be instantaneously delivered to the nucleus, and to determine if gene expression was possible in a cell-cycle independent manner. We tested cultured cell lines, a primary murine tumor, and primary human leukemia cells from diagnostic work-up for transgene expression, using both RFP and CD137L expression vectors.
Combined electroporation-transfection directly delivered plasmid DNA to the nucleus of transfected cells, as demonstrated by confocal microscopy and real-time PCR analysis of isolated nuclei. Expression of protein from plasmid vectors could be detected as early as two hours post transfection. However, the kinetics of gene expression from plasmid-based vectors in tumor cell lines indicated that optimal gene expression was still dependent on cell division. We then tested to see if pediatric acute lymphocytic leukemia (ALL) would also display the rapid gene expression kinetics of tumor cells lines, determining gene expression 24 hours after transfection. Six of 12 specimens showed greater than 17% transgene expression, and all samples showed at least some transgene expression.
Given that transgene expression could be detected in a majority of primary tumor samples analyzed within hours, direct electroporation-based transfection of primary leukemia holds the potential to generate patient-specific cancer vaccines. Plasmid-based gene therapy represents a simple means to generate cell-based cancer vaccines and does not require the extensive infrastructure of a virus-based vector system.
The efficacy of cell-based tumor vaccines in murine models of malignancy is well established. Using tumor cells lines transfected with soluble immune stimulatory molecules such as IL-2 or IL-12, or cell surface co-stimulatory antigens including CD80, and CD137L, or even allogeneic MHC results in profound immune activation [1–5]. The advantage of working in model systems is that unlimited amounts of tumor are available to produce cell-based vaccines. The ability to produce cell-based vaccines from clinic-derived material, however, remains a challenge.
Cell-based vaccines from tumor-derived material have been prepared and administered in either an allogeneic or autologous fashion, recently reviewed by Mocellin, et al. . An allogeneic vaccine usually features the expansion of a single tumor cell line that can grow well in culture, genetic transduction by the desired vector, and preparation of large vaccine stocks that can be qualified for clinical use. A vaccine for neuroblastoma featuring the expression of both a cytokine and a chemokine transgene (IL-2 and lymphotactin) by a single human neuroblastoma cell line is a recent example of this strategy . The disadvantage of a single cell line approach is that each malignancy is to some degree unique, and perhaps the most immunogenic antigens, or the most relevant ones for a given patient, will fail to be expressed by the allogeneic vaccine.
Give these limitations, we propose that a cell-based vaccine could be produced in an autologous manner for patients with a high disease burden, such as those who present with significant bone marrow involvement. For example, the large amount of tumor material typically available from leukemia patients makes these cells accessible for autologous patient-derived vaccine production.
A major hurdle to be overcome in using primary cells is the need to culture tumor cells in vitro in order for transduction or transfection procedures to be carried out. Most malignancies will not survive in culture in large enough numbers to be utilized. However, if the time required for in vitro manipulation was minimized, for example to 8–24 hours, patient-derived leukemia cells could be isolated from blood or bone marrow, transfected, and then upon irradiation used as a cell-based vaccine. Here we report the application of a novel electroporation-based transfection methodology that holds the potential to immediately transform a patient tumor sample into a cell-based cancer vaccine. This process, termed nucleofection, was pursued in our laboratory because it is the most rapid method of gene vector introduction available. We demonstrate that even though delivery of a plasmid gene vector to the nucleus is immediate, short-term culture is still required, and that a single-round of cell division may be needed to reach optimal gene expression levels. Importantly the time for tumor vaccine preparation is now measured in terms of hours instead of days. Our findings confirm studies carried out by Schakowski et al., where 3 samples from acute myeloid leukemia (AML) patients were transfected with a GFP expression vector . The large degree of transgene expression in the majority of patient-derived acute lymphoblastic leukemia (ALL) specimens that we present here suggests that a clinical trial using these procedures should be pursued.
The mouse neuroblastoma cell line AGN2a, an aggressive subclone of Neuro-2a, was cultured in Dulbecco's modified Eagle's medium (DMEM), 100 U/ml penicillin, 100 μg/ml streptomycin, 2 mM L-glutamine and 10% heat inactivated fetal bovine serum (FBS), 1 mM MEM sodium pyruvate, 100 U/ml penicillin, 100 μg/ml streptomycin, 0.01 M HEPES buffer, 2 mM L-glutamine, 0.05 M 2-mercaptoethanol, and 0.069 M L-arginine HCl . Primary murine tumor was generated by subcutaneous injection of 1 × 106 cultured AGN2a cells into strain A/J mice (Jackson Labs, Bar Harbor, ME). The human osteosarcoma cell line U2OS, kindly provided by Dr. Kent Wilcox, Medical College of Wisconsin, and the mouse squamous cell carcinoma cell line SCCVII, kindly provided by Dr. Scott Strome, Mayo Clinic, Rochester, MN, were cultured in DMEM as above. Mouse primary tumors were processed into single-cell suspensions by injection of 1–2 ml of 1 mg/ml collagenase D into the excised tumor mass (1 mg/ml in 10 mM HEPES, 150 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2) and incubated at 37°C for 45 min followed by mechanical disruption through a sterile screen. Tumor cells were washed in DMEM and PBS and viable cells were separated by centrifugation over a Ficoll-Paque™ (Amersham Biosciences, Piscataway, NJ) density gradient.
Transfection of tumor cell lines
Tumor cells were transfected with either pcDNA3.1/Hygro(-) (Invitrogen, Carlsbad, CA) or pDSRed2-C1 (BD Biosciences, San Diego, CA) plasmid vectors using a cationic lipid-based transfection methodology (Novafection, VennNova, Inc., Pompano Beach, FL) or a proprietary electroporation method (Nucleofection, Amaxa, Köln, Germany). Cells were nucleofected with 0.5 μg plasmid per 106 cells or lipid transfected with 0.5 μg plasmid and 2 μg of NovaFECTOR reagent per 106 cells. Similarly, U2OS, SCCVII and AGN2a cells were nucleofected with 0.5 μg per 106 cells pDSRed2-C1. To determine expression levels, cells were stained with 7AAD (BD Biosciences) and the expression of red fluorescent protein (RFP) in live gated cells was analyzed by flow cytometry (FACScan, Becton Dickinson, Franklin Lakes, NJ) at designated time points. U2OS and SCCVII cells were also nucleofected with pCI-neo (Promega, Madison, WI) encoding CD137L (4-1BBL) cDNA at a concentration of 1.5 μg per 106 cells . The CD137L transfected cells were stained with phycoerythrin (PE) conjugated mouse anti-human CDw137 Ligand (BD Biosciences Pharmingen, San Diego, CA).
Patient leukemia and lymphoma samples were obtained in accordance to the Helsinki Declaration from excess bone marrow or peripheral blood specimens submitted to the Cell Marker Lab, Children's Hospital of Wisconsin, for routine leukemia screening. Studies involving these samples were approved by the Medical College of Wisconsin and Children's Hospital of Wisconsin Institutional Review Boards. Informed consent was obtained from the parents or guardians of each child and each sample was assigned a unique identifier number to ensure confidentiality.
Transfection of primary acute lymphocytic leukemia cells
Leukocytes from bone marrow or peripheral blood patient samples were separated by centrifugation over a Ficoll-Paque™ density gradient. Cells were nucleofected with 1 μg pDSRed2C-1 (red fluorescent protein, RFP, expression vector) plasmid per 106 cells using a variety of Amaxa solutions and program parameters, cultured in RPMI-1640, 100 U/ml penicillin, 100 μg/ml streptomycin and 10% heat inactivated FBS for 24 hours then analyzed for RFP expression by flow cytometry (FACScan, Becton Dickinson). FACS acquisition and analysis was done using either propidium iodide (PI) or 7AAD to exclude dead cells. The leukemic blast population phenotype was determined by the flow cytometric and CD antigen expression profile as compared to normal cell populations. Both CD45+ and CD45- leukemic blasts could be gated when stained with anti-CD45 antibody and analyzed by flow cytometry for CD45 expression and side scatter properties. All antibodies utilized were clinical grade direct fluorochrome conjugates (Becton Dickinson).
U2OS cells were nucleofected with 3 μg fluorescein labeled (Mirus Label IT® Tracker, Madison, WI) pUC19 plasmid or pDSRed2C-1 plasmid per 106 cells. Immediately, or 3 days following nucleofection, cells were washed in cDMEM and counted. Cells were fixed on a glass slide with 3.7% buffered formalin, washed, permeabilized with 0.5% Triton X-100 (Surfact-amp, Pierce, Rockford, IL) and washed again. Pearmeabilized cells were incubated with 2.4 nM TOTO3 (Molecular Probes, Eugene, OR) and washed. Vectashield (Vector Laboratories, Inc., Burlingame, CA) was added to the cells prior to sealing with a coverslip. Optical sectioning of cells was taken sequentially using argon (488 nm excitation) and helium/neon (633 nm excitation) lasers on a Leica SPT S2 confocal microscope with a 100x oil immersion lens.
Quantitative real-time PCR
U2OS cells were nucleofected with 0.5 μg pDSRed2C-1 plasmid per 106 cells. Cells were harvested and used for nuclear DNA isolation. Prior to DNA isolation, nuclei were washed in PBS and incubated with 0.5U DNase I (Ambion, Austin, TX) at 37° for 10 min and washed again twice in PBS. Nuclear DNA was isolated (Nuclei EZ prep, Sigma, Saint Louis, MO) from transfected cells at designated time-points. The plasmid encoded neomycin phosphotransferase gene (neo) was amplified with primers and TaqMan hydrolysis probe as described by Sanburn, et al. . Nuclear DNA from each of three experimental and three parallel control samples (U2OS cells Nucleofected without plasmid) was amplified in triplicate in an Opticon™ 2 Cycler (MJ Research™, Inc., Waltham, MA) with the following cycling protocol: 50°C 2 min, 95°C 10 min, with 40 cycles of 95°C for 15 sec., and 62°C for 1 min. To normalize the number of cells/nuclei, human RNase P was amplified using the TaqMan® RNase P reagents kit (Applied Biosystems, Foster City, CA), or for mouse cells, mouse Apo B was amplified using the primers 5' CACGTGGGCTCCAGCATT 3'and 5' TCACCAGTCATTTCTGCCTTTG 3' and the TaqMan hydrolysis probe 5'(FAM) CCAATGGTCGGGCACTGCTCAATA (TAMRA) 3' (courtesy Renee Horner, qpcrlistserver, yahoo groups, yahoo.com). The neo gene copy number per cell was determined using a plasmid-based standard curve.
Analysis of cell division
Cells were suspended in PBS and incubated with CFDA SE (5-(and -6)-carboxyfluorescein diacetate succinimidyl ester, CFSE (Molecular Probes, Eugene, OR) at a final concentration of 0.35 μM per 4 × 106 cells, incubated for 10 min at 37°C, and washed x3 in DMEM, 10% FBS. CFSE expression was analyzed by flow cytometry to assess cell division.
Cell cycle blockade
At 50–60% confluency, 0.6 mM mimosine (Sigma, Saint Louis, MO) was added to U2OS cells (2). Both U2OS and U2OS cells treated with mimosine were incubated at 37°C for 48 hours at which time the cells were harvested, counted and nucleofected with 1.5 μg per 106 cells pCI-neo vector encoding human 4IBBL (CD137L) cDNA . As a control, cells were also nucleofected without plasmid. Four hours post-nucleofection cells were harvested, counted, stained with phycoerytherin (PE) labeled anti-human CD137L (BD Biosciences) and 7AAD (BD Biosciences), and analyzed for CD137L-expression by flow cytometry. To determine DNA content prior to nucleofection, cells were washed in phosphate buffered saline (PBS), fixed with 4% paraformaldehyde, washed again in PBS, and 1 μl propidium iodide (BD Bioscience) at 50 ug/ml was added. The propidium iodide labeled cells were then analyzed by flow cytometry .
Optimization of Transfection by Electroporation
The primary limitation of electroporation-based transfection is cell death. Preliminary experiments confirmed that increasing the strength of the electric field corresponded to both a higher transfection rate, and increased cell death. The nucleofection setting that we found optimal resulted in 70% cell death, Figure 1A. Cell numbers gradually recovered post-nucleofection, beginning at 24 hours. In contrast, there was no decrease in cell number following novafection, Figure 1B.
Delivery of plasmid DNA to the nucleus by electroporation is rapid and short-lived
The inability to culture most primary human tumors led us to search for methods of transfection that would require minimal culture and processing time while allowing for efficient gene transfection. Given the rapid kinetics of expression using nucleofection, we sought to determine if this was due to direct delivery of plasmid DNA in to the nucleus. Confocal microscopy was used to visualize individual z-plane sections that represent internal nuclear layers of U2OS cells that had been nucleofected with 3 μg FITC-labeled pUC19 plasmid per 106 cells, immediately cytospun onto glass slides, and then prepared for microscopy. The nuclear and cytoplasmic boundaries of nucleofected cells were visualized by phase contrast microscopy, Figure 3A, panel b, or by staining with the nuclear dye TOTO3, Figure 3A, panel c. The nuclei are stained dark blue, with a lighter blue staining in the cytoplasmic compartment. The plasmid-associated fluorescein signal was present in both the cytoplasmic and nuclear compartments immediately following nucleofection, Figure 3A, panel d. Visual inspection reveals that most cells contained nuclear plasmid, Figure 3A, d (an overlay of the plasmid signal with the TOTO3 stain).
Using the same technique, we then sought to determine how long after nucleofection the plasmid vector remained in the nucleus. Three days after nucleofection of U2OS cells with 3 μg FITC-labeled pDSRed2C-1 (RFP) plasmid per 106 cells, the presence of plasmid vector DNA, was greatly diminished, Figure 3B, panel a. The presence of plasmid vector DNA, as detected by FITC fluorescence, was seen in a small minority of cells, and when present on day 3 it appeared to associate more with a punctate fluorescence in the cytoplasm, Figure 3B, a and d. Despite the loss of plasmid vector from the nucleus, intense red fluorescence was seen in many of the cells at this time, indicating the continued presence of red fluorescent protein, Figure 3B, panels b and d.
To further confirm the presence of plasmid in the nucleus, the copy number of plasmid vector per cell was calculated by real-time PCR amplification of the pDSRed2C-1 encoded neomycin phosphotransferase gene (neo). U2OS cells were nucleofected with 0.5 μg pDSRed2C-1 per 106 cells and immediately, or at day 3, nuclear DNA was isolated from the nuclear fraction of cell lysates and PCR amplified using neo primers and a neo-specific TaqMan probe. The total number of plasmid neo copies was calculated based on comparison to a standard curve generated with the same plasmid vector. The number of cells (or nuclei) analyzed was determined using a standard curve calibrated to genomic DNA mass and signal from the single copy gene RNAseP. Nuclei were purified on a sucrose cushion, washed with PBS, digested with DNAse in order to remove contaminating cytoplasmic plasmid DNA, and total DNA extracted. Immediately following nucleofection, there were 200 to 400 copies of plasmid per cell, Figure 3C. In agreement with microscopy data, copy number in U2OS cell nuclei decreased to 50 copies or less per cell by day 3 post-nucleofection, Figure 3D. Immediate localization of plasmid to the nucleus following nucleofection was also observed by real-time PCR in the AGN2a and SCCVII cell lines (data not shown).
Delivery of plasmid gene vectors to the nucleus requires cell division for optimal gene expression
To further explore this finding, we used an alternate plasmid-encoded transgene and compared the kinetics of expression in rapidly and slower dividing cell lines. Our laboratory has produced a number of immune co-stimulatory expression vectors, and we chose a human CD137L (4IBBL) expression vector for further study. Expression of this cell-surface antigen can be directly measured by flow cytometry using a labeled CD137L-specific antibody. One of our concerns with using RFP was that the DsRed2 protein we utilized requires approximately 6 hours to reach full fluorescence intensity due to a requirement for intracellular oxidation . Therefore a CD137L-encoding plasmid, 1.5 μg per 106 cells, was nucleofected into the U2OS and SCCVII cells and expression compared. In the rapidly dividing U2OS, 40% of the cells expressed CD137L as early as 2 hours post-nucleofection. At 6 hours post-nucleofection, 80% of the U2OS cells expressed CD137L, Figure 4E. In contrast, only 10% of SCCVII cells expressed CD137L at 6 hours post-nucleofection. The nucleofection process also induced significant cell death, demonstrating that cell death was not an RFP-associated phenomenon, Figure 4F. All cells experienced 60 to 80% cell death upon nucleofection, however the SCCVII cells recovered much more rapidly than either the rapidly dividing U2OS cells or the less rapidly dividing AGN2a cells, indicating that factors other than cell division are involved in cellular recovery from the electroporation and transfection processes.
Application of electroporation-based transfection to human leukemias
Comparison of nucleofection parameters in ALL patient samples. Nucleofection using an RFP expression vector was carried out using 6 different patient samples (patients 4,13,8,11,19,12) from one of two potential tissue sources, PB, peripheral blood, or BM, bone marrow. 1No NF, not nucleofected autofluorescence control. 2Nucleofection was carried out with one of three solutions R, T, or V, and the following electrical settings on the Amaxa nucleofection device: T20, U15, T17, T27, S04, O17, T16, T01, O17. 3The table lists the percentage of viable cells expressing the plasmid-encoded RFP at 24 hours post-nucleofection.
Cell-based autologous cancer vaccines hold great promise in the effort to shift the adaptive immune system from ignorance or anergy toward cell-mediated immune recognition of cancer. The generation of a Th1-type immune response with cell-based vaccines, and the resultant CTL-mediated killing of tumor, have offered the most effective anti-tumor responses in pre-clinical models, and also initiated clinical responses, as reviewed by Nitti et al., . Ex vivo introduction of gene vectors encoding immune activating signals (such as co-stimulatory antigens, cytokines and adhesion molecules) into tumor cells with subsequent re-introduction of irradiated modified tumor cells into patients is currently being pursued with the aim of initiating a clinically-relevant immune response against tumor [14, 15]. Because this type of therapy involves cell processing, culture procedures, as well as genetic transfer, any steps that streamline the process will make a significant contribution toward tumor vaccine development.
Newer generation transfection procedures based on electroporation or branched polyethyleneamine (PEI) have been reported to be independent of cell cycle effects . Nevertheless, we found that cell cycle progression does enhance transgene expression and that certain tumor cell lines (such as U2OS) thought to be highly "transfectable" may be highly amenable to vector-derived transgene expression precisely because they have rapid rates of cell division. Most reports do not analyze transgene expression at the early times (2–8 hours) we report here, and unless proven otherwise, the contribution of cell cycle progression within the first 8 hours of transfection cannot be excluded. We examined this question in detail because our goal is to transfect primary human leukemias that do not persist in tissue culture and thus have limited in vitro proliferative capacity.
Having thus defined the problem, we examined in detail a transfection procedure that appears to be the most rapid. Nucleofection delivers plasmid vector DNA in to the nucleus immediately, Figure 3. However, as we demonstrate here, nucleofection still benefits from cell cycle progression and cell division, Figures 4,5. Using nucleofection we found that tumor cells harvested directly from a mouse bearing the AGN2a neuroblastoma could be nucleofected with high efficiency. Thus, the process of tumor disaggregation and processing did not render the neuroblastoma cells we studied resistant to nucleofection. Confident of these results, we initiated analysis of primary human tumors.
Transfection of Patient Leukemia. Patient leukemia cells were nucleofected with solution R, setting T20, as described in Methods. Patient diagnosis and specimen type are listed (PB, peripheral blood, BM, bone marrow). Relevant diagnostic criteria with respect to phenotype is also listed. Transfection efficiency is based upon the percentage of viable cells expressing RFP at 24 hours post-nucleofection.
The process of nucloefection delivers plasmid DNA directly to the nucleus. Even though delivery to the nucleus is thought to circumvent dependence on cell division, we found that the highest and earliest levels of transgene expression from plasmid-based vectors occurred in rapidly dividing cells. We were also able to demonstrate that primary acute lymphocytic leukemia cells (ALL) from pediatric patients could also be nucleofected with plasmid-based vectors, thus opening the door to patient-specific cell manipulation. In light of the laboratory studies we present, transfection rates of clinical samples may be increased even further if some degree of cell division could be induced during the in vitro culture of these specimens. Even though this culture time is less than 24 hours, a single cell division could potently increase transgene expression levels, and thus immunogenicity of the vaccine preparation. Future work will include analysis of the correlation between the ability to be nucleofected, markers of cell cycle progression, and the induction of cell cycle progression post-nucleofection.
We would like to thank Elizabeth C. Glaser for expert technical assistance, and Dr. James Casper and Dr. Bruce Camitta for helpful discussions and clinical leadership. This work was supported by grants from the Midwest Athletes Against Childhood Cancer, MACC Fund, Inc., and by an Internal Research Grant from the American Cancer Society, Medical College of Wisconsin Cancer Center.
- Hock RA, Reynolds BD, Tucker-McClung CL, Kwok WW: Human class II major histocompatibility complex gene transfer into murine neuroblastoma leads to loss of tumorigenicity, immunity against subsequent tumor challenge, and elimination of microscopic preestablished tumors. J Immunother. 1995, 17: 12-8.View ArticleGoogle Scholar
- Bausero MA, Panoskaltsis-Mortari A, Blazar BR, Katsanis E: Effective immunization against neuroblastoma using double-transduced tumor cells secreting GM-CSF and interferon-γ. J Immunother. 1996, 19: 113-24.View ArticleGoogle Scholar
- Katsanis E, Orchard PJ, Bausero MA, Gorden KB, McIvor RS, Blazar BR: Interleukin-2 gene transfer into murine neuroblastoma decreases tumorigenicity and enhances systemic immunity causing regression of preestablished retroperitoneal tumors. J Immunother. 1997, 15: 81-90.View ArticleGoogle Scholar
- Johnson BD, Yan X, Schauer DW, Orentas RJ: Dual expression of CD80 and CD86 produces a tumor vaccine superior to single expression of either molecule. Cell Immunol. 2003, 222: 15-26. 10.1016/S0008-8749(03)00079-0.View ArticlePubMedGoogle Scholar
- Yan X, Johnson BD, Orentas RJ: Murine CD8 lymphocyte expansion in vitro by artificial antigen-presenting cells expressing CD137L (4-1BBL) is superior to CD28, and CD137L expressed on neuroblastoma expands CD8 tumor-reactive effector cells in vivo. Immunol. 2004, 112: 105-116. 10.1111/j.1365-2567.2004.01853.x.View ArticleGoogle Scholar
- Mocellin S, Mandruzzato S, Bronte V, Lise M, Nitti D: Part I: Vaccines for solid tumors. Lancet Oncol. 2004, 5: 681-689. 10.1016/S1470-2045(04)01610-9.View ArticlePubMedGoogle Scholar
- Rousseau RF, Haight AE, Hirschmann-Jax C, et al: Local and systemic effects of an allogeneic tumor cell vaccine combining human lymphotactin with interleukin-2 in patients with advanced or refractory neuroblastoma. Blood. 2003, 101: 1718-1726. 10.1182/blood-2002-08-2493.View ArticlePubMedGoogle Scholar
- Schakowski F, Buttgereit P, Mazur M, et al: Novel non-viral method for transfection of primary leukemia cells and cell lines. Genet Vaccines Ther. 2004, 2: 1-11. 10.1186/1479-0556-2-1.PubMed CentralView ArticlePubMedGoogle Scholar
- Sanburn N, Cornetta K: Rapid titer determination using quantitative real-time PCR. Gene Ther. 1999, 6: 1340-1345. 10.1038/sj.gt.3300948.View ArticlePubMedGoogle Scholar
- Perry C, Sastry R, Masrallah IM, Stover PJ: Mimosine attenuates serine hydroxymethyltransferase transcription by chelating zinc. J Biol Chem. 2005, 280: 396-400. 10.1074/jbc.M412914200.View ArticlePubMedGoogle Scholar
- Tlsty T, Briot A, Poulose B: Analysis of cell cycle checkpoint status in mammalian cells. Methods Enzymol. 1995, 254: 125-133.View ArticlePubMedGoogle Scholar
- Bevis BJ, Glick BS: Rapidly maturing variants of the Discosoma red fluorescent protein (DsRed). Nat Biotechnol. 2002, 20: 83-87. 10.1038/nbt0102-83.View ArticlePubMedGoogle Scholar
- Mocellin S, Rossi CR, Nitti D: Cancer vaccine development: on the way to break immune tolerance to malignant cells. Exp Cell Res. 2004, 299: 267-78. 10.1016/j.yexcr.2004.06.017.View ArticlePubMedGoogle Scholar
- Brunner S, Fürtbauer E, Sauer T, Kursa M, Wagner E: Overcoming the nuclear barrier: cell cycle independent nonviral gene transfer with linear polyethylenimine or electroporation. Mol Ther. 2002, 5: 80-86. 10.1006/mthe.2001.0509.View ArticlePubMedGoogle Scholar
- Soiffer R, Lynch T, Mihm M, Jung K, Rhuda C, Schmollinger JC, Hodi FS, Liebster L, Lam P, Mentzer S, Singer S, Tanabe KK, Cosimi AB, Duda R, Sober A, Bhan A, Daley J, Neuberg D, Parry G, Rokovich J, Richards L, Drayer J, Berns A, Clift S, Cohen LK, Mulligan RC, Dranoff G: Vaccination with irradiated autologous melanoma cells engineered to secrete human granulocyte-macrophage colony-stimulating factor generates potent antitumor immunity in patients with metastatic melanoma. Proc Natl Acad Sci USA. 1998, 95: 13141-13146. 10.1073/pnas.95.22.13141.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.