Skip to main content
NCBI home page
Cancer Immunology, Immunotherapy : CII logoLink to Cancer Immunology, Immunotherapy : CII
. 2003 Oct 18;53(2):92–99. doi: 10.1007/s00262-003-0428-1

Immunization with a plant-produced colorectal cancer antigen

Thorsten Verch 1,2, D Craig Hooper 1, Anatoly Kiyatkin 1,3, Zenon Steplewski 1, Hilary Koprowski 1,✉
PMCID: PMC11032813  PMID: 14566428

Abstract

Cancer vaccination has become an important focus of oncology in recent years. Active immunization with tumor-associated antigens such as colorectal cancer antigen GA733-2 is thought to potentially overcome the reoccurrence of metastasis. As recombinant protein production in bioreactors is costly and subject to growing safety concerns, we tested plants as an alternative for the expression of a potential colorectal cancer vaccine. Comparing colorectal cancer antigen GA733-2 produced in tobacco plants with the same antigen produced in insect cell culture, we found a similar humoral immune response to injection of either of the two antigen preparations into mice. Some minor differences were observed in the cellular response that might be due to impurities. Our studies compare for the first time, immunization with the same antigen expressed in either plants or insect cell culture. This will provide important data for use of plants as production systems of therapeutics.

Keywords: Cancer vaccination, Plant-produced therapeutics, Antigen GA733-2, Colorectal cancer, Virus vector

Introduction

Colorectal cancer is the second leading cause of cancer-related death in Western societies. Unless the disease is recognized early, the survival rate after surgical removal of the main tumor is only 50% due to outgrowth of metastases. More efficient treatment options with lower side effects than currently employed therapies are needed in order to reduce metastatic spread and thus prolong survival.

Immunotherapy is a promising alternative to the use of radio- and chemotherapy in the treatment of metastatic cancer [14, 24, 50, 51]. Tumor-associated antigens such as c-erbB2, EGF-receptor, CEA, and GA733-2 are widely being used as targets in a number of tumors [8, 16, 31, 32, 46, 47, 48] including colorectal cancer [1, 19, 40]. Active immunotherapy relies on the patient’s own immune system to fight cancer cells after stimulation by cancer-associated antigens. This so-called cancer vaccination overcomes immune tolerance and leads to a cytotoxic response that is directed primarily against cancer cells [33, 34]. Rather than targeting only a single epitope with a monoclonal antibody, as is typical for passive immunotherapy, active immunization can potentially generate an immune response against multiple epitopes. Moreover, in addition to a humoral response, a strong cellular response may be generated by immunization resulting in more efficient tumor cell killing and a better chance of preventing the metastatic spread of the tumor.

The colorectal cancer–associated antigen GA733-2 has been studied for more than 20 years [3, 7, 9, 18, 20, 22, 26, 28, 35, 36]. It is a transmembrane glycoprotein with functions in cell adhesion [4, 7, 17, 28, 39]. As it is expressed on adenocarcinomas in significantly higher copy numbers than on healthy tissue [17], the extracellular domain of antigen GA733-2 was used as a target of cancer vaccination in several clinical trials [6, 21, 41]. When applied as recombinant protein in cancer patients, a tumor cell–specific humoral and cellular immune response was observed [41]. Currently, vaccinia virus is used to deliver GA733-2 to cancer patients in ongoing clinical trials [49].

Recombinant antigen GA733-2 used in research and early clinical trials has been produced in the baculovirus expression vector system (BEVS) [43]. Despite several advantages of this system, such as well-controlled and reproducible expression rates, there are some drawbacks. Fermentor technology and specialized culture media can make scaled-up protein production in BEVS cost-intensive. When using media containing animal products such as serum, there is a risk of the introduction of pathogens into the product. FDA-required pathogen testing also adds to the final product costs. Plant-based expression could be an alternative to BVES and is being investigated particularly for the production of biomedicals [2, 5, 11, 13, 27]. Biomass accumulation is relatively inexpensive employing only conventional farming methods [10]. Furthermore, plants are not known to be host to any human pathogen [44], which is a very important benefit for product safety. Despite successful applications of plant-produced biomedicals in research, there is only limited information available on the comparability of proteins expressed in plants with proteins from established expression systems when used for immunization. In order to evaluate the potential of anticancer therapeutics produced with a plant-virus expression system in relation to BEVS, we have used purified antigen GA733-2 expressed in either system to immunize mice.

The plant expression system that we employed was based on tobacco mosaic virus (TMV)–based gene vector 30B [38]. It has been successfully used to produce a number of proteins in tobacco plants (Nicotiana benthamiana) including cancer biomedicals such as the therapeutic lymphoma single chain antibody 38C13 [30] and colorectal cancer antibody CO17-1A [45]. Recombinant proteins are expressed by TMV 30B under the control of an additional viral subgenomic RNA promoter rather than as coat protein fusion which strategy is often employed in other plant virus expression systems. We compared the capacity of plant- or BEVS-expressed cancer antigen to elicit antigen- and tumor cell- specific immune responses.

Material and methods

Materials

All chemicals were purchased from Sigma (St. Louis, MO). Restriction enzymes were obtained from Promega (Madison, WI) or from New England Biolabs (Beverly, MA). The High Fidelity PCR system was purchased from Roche (Indianapolis, IN). All cancer cell lines were obtained through the American Type Culture Collection (Manassas, VA). Antibody mAb733 purified from hybridoma cells and the hybridoma cell line expressing mAb733 were the kind gifts of Dr Dorothee Herlyn (Wistar Insitute, Philadelphia, PA). Animal experiments were performed following the “Principles of laboratory animal care” as published by the NIH.

Cloning

All DNA cloning steps were performed as described [37]. A cDNA of antigen GA733-2 (kindly provided by Dr A. Linnenbach, Wistar Institute, Philadelphia, PA) was PCR-modified by introducing a PacI-site at the 5’-end and an XhoI-site at the 3’-end of the gene. Furthermore, sequences encoding a C-terminal His6-tag and a KDEL peptide sequence for ER-retention were added to the GA733-2 gene. Cancer antigen PL-733FK2 was cloned into bacterial vector pGEMT (Promega, Madison, WI) and contained only the extracellular domain of antigen GA733-2 (amino acids 1–264). After sequence verification (Nucleic Acid Facility, Thomas Jefferson University, Philadelphia, PA), the antigen gene was excised by PacI and XhoI for ligation into similarly prepared TMV vector p30B resulting in p30B-733FK2 and 30B-733F, respectively.

Expression and purification of the cancer antigens

Plant virus vector p30B-733FK2 was transcribed in vitro, and N. benthamiana plants were infected with in vitro transcript as described [45]. Infected plants were ground in phosphate buffer (50 mM Na2HPO4, 50 mM NaCl, 10 mM EDTA, pH 7) and recombinant virus was concentrated by precipitation with polyethylene glycol 15–20,000 (PEG). The PEG pellet was resuspended in phosphate buffered saline (PBS) and used for further plant infection. The second generation of infected plants was harvested at 21 dpi for extraction of recombinant antigen PL-733FK2. Plants were ground in phosphate buffer with added polyvinylpolypyrrolidone, celite, 1-mM Leupeptin, and 1-mM PMSF. The sap was centrifuged (12,000 g, 30 min), filtered through Miracloth, and adjusted to pH 7.4 with 1-M Tris-HCl, pH 9 before precipitation of contaminating proteins with 20% (NH4)2SO4 (3 h, 4°C). After removal of precipitated proteins by centrifugation (12,000 g, 60 min), the supernatant was adjusted to a final concentration 70% (NH4)2SO4 (14 h,4°C). Proteins were pelleted as described above and resuspended in PBS. Undissolved particulates were removed by centrifugation and filtration through 0.22 μm. The supernatant was subjected to immunoaffinity chromatography using antibody mAb733 to capture the plant-expressed antigen PL-733FK2 using similar conditions as published for the purification of antigen BV-733 [43].

Antigen BV-733 was expressed in BVES and isolated from supernatants (kindly provided by Dr William Wunner, Wistar Institute, Philadelphia, PA) by immunoaffinity chromatography as described [43].

Immunization of animals

BalbC mice were immunized three times on a biweekly schedule with either antigen BV-733 or antigen PL-733FK2. Twenty micrograms of antigen was mixed with 10 μg of adjuvant QS21 (a kind gift of Aquila Pharmaceuticals, Framingham, MA) as recommended by the manufacturer and injected subcutaneously, followed by two further injections intraperitoneal without adjuvant. Blood samples were taken by orbital bleeding 2 days before each immunization for analysis of sera by ELISA.

ELISA

To assess tumor cell–specific antibody expression, SW948 colorectal cancer cells were split into 96-well plates at 100,000 cells/well. After incubation for 20 h, the plates were washed with 200-μl PBS, fixed with 0.05% glutaraldehyde for 10–15 min and blocked with 100-mM glycine in PBS at 4°C until use. Sera from immunized mice were applied at different concentrations, bound antibody was detected by antimouse IgG horse raddish peroxidase conjugate (Sigma, St. Louis, MO) developed with OPD (Agdia, Elkhart, IN). Upon color development, the reaction was stopped by the addition of 1 volume (v/v) of 2-M H2SO4, and plates were read at OD490nm. The plates were washed between each step three times with PBS + 0.05% Tween 20.

Complement-dependent cytotoxicity

Complement-dependent cytotoxicity (CDC) was tested using mouse sera 2 weeks after the last immunization. For experimental purposes, the breast cancer cell line SK-BR3 that also carries the antigen GA733-2 was used as a target because it was found to be more sensitive than the colorectal cancer line SW948. Monoclonal antibody mAb733 against the antigen GA733-2 was used as a positive control, and monoclonal antibody mAbMe3.61 against a melanoma cell antigen as a negative control. Sera of each group were pooled for use in the assay. A single cell suspension of the breast cancer cell line SK-BR3 was labeled with 51Cr (Perkin Elmer, Boston, MA) (75 μCi/2×106 cells in 100 μl) for 2 h at 37°C. Nonincorporated isotope was removed by three washes with CDC buffer (RPMI media, 5% FCS, 4-mM HEPES, 0.6% NaOH). The cell pellet was finally resuspended in RPMI at a concentration of 4×105 cells/ml. Fifty microliters of cell suspension was mixed with 50 μl of antibody or sera, respectively. After 1.5 h of incubation at 37°C, 50 μl of ice-cold LowTox rabbit complement (Cedarlane, Hornby, Canada) was added at a dilution of 1:5. The complement reaction was stopped after a further 45 min incubation by the addition of 100-μl CDC buffer. The cells were pelleted by centrifugation and the CPM from 50 μl of media supernatant were analyzed in a γ-counter (Perkin Elmer).

Antibody-dependent cellular cytotoxicity

Antibody-dependent cellular cytotoxicity (ADCC) was tested using the same sera and target cell line as used for CDC assays (see above). The SK-BR3 target cells were labeled with 10 μCi of 111In-Oxine (Mallinckrodt, Folcroft, PA) per 1×106 cells for 10 min at room temperature. Nonincorporated isotope was removed by three washes with RPMI media. Labeled cells were preincubated with antibody for 1 h at 37°C before adding Ficoll-Hypaque purified human peripheral blood cells that had been demonstrated to be suitable effectors for control antibody mAb733 [29, 42]. After incubation for 20 h at 37°C, all cells were pelleted by centrifugation (10 min at 500 g) and 100 μl of supernatant was analyzed for released isotope in a γ-counter. Outliers were not considered for analysis.

T-cell proliferation

The T-cell proliferation assay was performed essentially as described elsewhere [23, 25]. Briefly, total cells from spleen and lymph nodes were washed in PBS. After hypotonic lysis of red blood cells, B cells were removed from the cell mixture by panning on antimouse IgG–coated Petri dishes. Unbound lymph node and spleen cells were washed with PBS and resuspended in α-MEM media + 0.6% mouse sera. A separate set of spleen cells that was not subjected to panning was used as antigen-presenting cells (APCs). Lymph node cells were mixed with APCs in a ratio of 1:1 and T cells from spleen were mixed with APCs in a ratio of 2:1. Immunoaffinity-purified recombinant antigens were added as stimulants at a concentration of 10 μg/ml. T-cell proliferation was measured at 48 h, 72 h, 96 h, and 120 h by the incorporation of 3H-thymidine into newly synthesized DNA. Cell proliferation was assessed using a response index RI = CPM (stimulated) / CPM (nonstimulated).

Results

Production of GA733-2 in Nicotiana benthamiana

Production of recombinant protein by the plant virus expression system is dependent on the extent of plant infection. In order to visualize the spread of the virus vector in infected plants under our greenhouse conditions, we used a construct-expressing green fluorescent protein (GFP) [38]. At 21 days post infection (dpi), most of the plant tissue was expressing GFP as assessed by fluorescence under UV light. Using the cancer antigen–expressing virus vector, PL-733FK2 was also found to be expressed up to at least 18 dpi as verified by Western blots of samples from uppermost leaves (data not shown). Therefore, total plant tissue of expression vector 30B-733FK2–infected plants was harvested at 18–21 dpi. Recombinant antigen PL-733FK2 was purified by immunoaffinity chromatography and verified by SDS-PAGE (Fig. 1A) and Western blot (Fig. 1B). Due to the additional H6KDEL C-terminal amino acid sequence on PL-733FK2, the protein was slightly larger than BV-733.

Fig. 1A, B.

Fig. 1A, B

Open in a new tab

Western blot (A) and SDS-PAGE (B) analysis of purified antigens BV-733 (lane 1) and PL-733 (lane 2). Antigen GA733-2-specific antibody mAb733 was used for detection in Western blot. The SDS-PAGE was stained with GelCode to detect separated proteins. Separation by SDS-PAGE was performed in parallel on two different gels for subsequent Western blotting or staining, respectively

Humoral immune response to antigen injection in mice

Two versions of antigen GA733-2 were compared for their ability to elicit a cancer cell–specific immune response: PL-733FK2 which was expressed in the plant virus expression system and BV-733 which was expressed in a baculovirus insect cell culture system. The amino sequences of both antigens encoding the extracellular domain of the cancer antigen were identical. PL-733FK2 also contained an additional C-terminal His6 and KDEL-sequence in PL-733FK2.

Mice were immunized with adjuvant QS21 only or a mixture of adjuvant with cancer antigen. Immunization with PL-733FK2 or BV-733 resulted in the generation of antibodies recognizing cancer cells expressing the antigen (Fig. 2A). Antibodies generated were predominantly of subtype IgG1 with minor IgG2a and IgG2b populations (Fig. 2B, C). IgM, IgA, and IgG3 were not detected above background levels. Serum IgG from immunized mice was found to bind to colorectal cancer cell line SW948 (Fig 2A), breast cancer cell line SK-BR3 (CDC and ADCC assays), and to both recombinant antigens (Fig. 3). The antibody titers were virtually the same in sera from mice immunized with antigen from either source.

Fig. 2A–C.

Fig. 2A–C

Open in a new tab

Analysis of the antibody response of cancer antigen immunized mice by ELISA. SW948 colorectal cancer cells grown for 24 h in 96-well plates (1×105 cells/well) were fixed with 0.05% glutaraldehyde and used for ELISA. Sera from immunized mice were applied at the dilutions indicated, followed by antimouse HRP-conjugated antibodies specific for the immunoglobulin subtypes indicated. A Total cancer antigen–specific antibody response in serum from immunized mice; B Immunoglobulin subtype in mice immunized with plant-produced antigen PL-733FK2; C Immunoglobulin subtype in mice immunized with antigen BV-733

Fig. 3.

Fig. 3

Open in a new tab

Determination of cross-recognition of the plant-virus and the baculovirus-produced antigen by sera from mice immunized with either one of these. ELISA plates were coated with plant-produced antigen PL-733FK2 and alternatively with BV-733. Sera from mice immunized with BV-733 were applied on both differently coated plates. Also, sera from mice immunized with PL-733FK2 were applied on both of the differently coated plates. Bound antibody was detected using an antimouse HRP-conjugated antibody

To determine if either of the two cancer antigen preparations contained immunogenic contaminants such as carrier proteins or copurified polysaccharides, we tested if the serum IgG titers were different when binding to both PL-733 and BV-733 was compared. ELISA plates were coated with either recombinant antigen PL-733FK2 or BV-733 and assessed for reactivity with both sera from mice immunized with PL-733FK2 and BV-733. Cross-reactivity was virtually equivalent (Fig. 3) indicating that there are unlikely to be any differences between the two antigen preparations contributing to the antibody response. A similar test of cross-reactivity using Western blots confirmed these results as no additional protein bands were observed (data not shown) indicating the absence of immunostimulatory protein contaminants. Thus, potential contaminants in the antigen preparations did not generate detectable antibody responses in the mice. Adjuvant-like effects cannot be entirely excluded but are unlikely given the similar titers of the antibody response against either antigen in ELISA.

Cancer cell specific serum cytotoxicity

A potential cancer vaccine should generate an immune response that ultimately leads to the killing of the tumor. As an initial test of this aspect of the humoral response, sera from immunized mice were tested for antibody-mediated cytotoxicity against cancer cells. Sera from both PL-733FK2- and BV-733-immunized mice, exhibited complement-dependent cytotoxicity (CDC) against antigen GA733-2 positive breast cancer cells but not against melanoma cells that lack this antigen (Fig. 4). At a high serum concentration, no significant difference was observed in the percentage of cytotoxicity between the two types of sera. However, within the range of dilutions tested, the cytotoxicity of the PL-733FK2 serum increased at higher dilutions. This effect could be due to an inhibitor that is present in the serum and that is ultimately diluted out. Alternatively, this could also be due to the polyclonal nature of the immune response. Some of the antigen-specific serum antibodies detected (Fig. 2) may not fix complement and, before being diluted out, compete with those active in CDC. As expected, the cytotoxicity of the BV-733 serum declined with decreasing serum concentration.

Fig. 4A, B.

Fig. 4A, B

Open in a new tab

Complement-dependent cytotoxicity assay. 51Cr-labeled target cells were incubated with sera from mice immunized with either cancer antigen BV-733, with antigen PL-733FK2, or with adjuvant only (QS21). SK-Br3 breast cancer cells carrying the antigen GA733-2 (A) or GA733-2 negative WM115 melanoma cells (B) were used. A monoclonal antibody specific for the cancer antigen GA733-2 (Ab733) and a monoclonal antibody specific for melanoma cells (AbMe3.61) were chosen as controls (concentrations 10 μg/ml, 5 μg/ml, 1 μg/ml, and 0.1 μg/ml). Released isotope was measured after 45 min in a γ-counter and the percentage cytotoxicity was calculated

The capacity of the sera to exhibit antibody-dependent cellular cytotoxicity (ADCC) was tested using peripheral blood effector cells. The ADCC effect of sera from mice immunized with either of the two antigen preparations was similar (Fig. 5). However, the cytotoxicity resulting from ADCC was found to be lower than that for CDC at a given serum concentration, which might be due to the prevalence of IgG1 subtype in the sera. Immunization with adjuvant QS21 alone did not result in specific antibody-mediated cellular cytotoxicity.

Fig. 5.

Fig. 5

Open in a new tab

Antibody-dependent cellular cytotoxicity assay. 111In-Oxine-labeled SK-BR3 breast cancer cells were incubated with sera from mice immunized with either plant virus–produced antigen PL-733FK2, with baculovirus-produced BV-733 or with adjuvant only. Monoclonal control antibody (Ab733) was used as a control. After adding human peripheral blood mononuclear cells, released isotope was measured after 24 h. Outliers were not considered for analysis. WM115 cells only released similar background amounts of label (data not shown)

Antigen-specific T-cell responses in immunized mice

To determine if the recombinant-expressed cancer antigens stimulated specific T-cell responses in a distinct or similar manner, a T-cell proliferation assay was performed. T cells were obtained from spleen and lymph nodes of mice immunized subcutaneously with the different antigen preparations and mixed with APCs. The cells were stimulated separately with either recombinant PL-733FK2 or BV-733 from the same batch as the immunization antigen. Proliferation of T cells from immunized mice was cross-stimulated by either antigen (Fig. 6). While there were differences in the degree of the response, T-cell proliferation was stimulated by the recombinant cancer antigen regardless of the expression system used for its production. Antigen-specific T-cell responses were consistently obtained with cells from lymph nodes and spleen following the third immunization. The first and the second immunization resulted in varying T-cell responses with little or no proliferation (data not shown).

Fig. 6A–C.

Fig. 6A–C

Open in a new tab

Proliferation assay of T cells isolated from spleens (A) and lymph nodes (B) of mice immunized with either plant virus–produced antigen PL-733FK2 or with baculovirus-produced BV-733. Pooled total spleen cells from all groups were used as antigen-presenting cells. T cells were isolated 2 weeks after the third immunization from spleen cells (A) by antibody-mediated removal of B cells prior to the proliferation assay. Total cells from lymph nodes were used (B). Either plant-produced antigen PL-733FK2 or baculovirus insect cell culture–produced antigen BV-733 were used as indicated in the graph legend at a concentration of 10 μg/ml to stimulate T-cell proliferation. The proliferative response was graphed using the response index response index RI = CPM (stimulated) / CPM (nonstimulated). C shows the T-cell response from mice immunized one time with QS21 without antigen subcutanously followed 2 weeks later by one time intraperitoneal immunization with antigen PL-733FK2. T cells isolated from spleen and lymph nodes were stimulated with either BV-733 or with PL-733FK2 as indicated in the graph

Surprisingly, T cells from mice immunized with adjuvant QS21 alone responded very well to stimulation with the plant-produced antigen PL-733 but not to BV-733. As QS21 is derived from plants, there may be similar structures present in the plant antigen preparation. Although these putative contaminants did not raise an antibody response (Fig. 3), a stimulatory effect on T-cell proliferation which might also account for the differences observed in the groups from immunized mice (see above) cannot be excluded at the present time.

Interestingly, a single intraperitoneal immunization with PL-733FK2, as opposed to subcutaneous administration, resulted in a strong T-cell response in spleen as well as in lymph node cells (Fig. 6C). These findings highlight the importance of the immunization route and schedule in raising a response to this antigen.

Discussion

In recent years, plants have been increasingly used in research as a source of recombinant proteins and potential therapeutics [10, 12, 13, 44]. Some of the more important reasons for developing plant expression systems are low-cost, low-tech biomass scale-up, protein processing, and safety from human pathogens. Despite many similarities between proteins expressed in plants and in other systems, some distinct differences exist, for example in protein glycosylation. Therefore plant-expressed proteins with potential therapeutic value for humans have to be carefully examined for extraneous effects of plant-specific modifications on immunogenicity. We have compared the immune responses of mice following immunization with virtually the same cancer antigen produced either in plants using a viral expression vector or in baculovirus-infected insect cell culture.

Both antigens evoked a similar antibody response upon immunization of mice. The antibodies of both groups had the same subtypes and recognized both recombinant antigens as well as the natural antigen present on the surface of colorectal and breast cancer cells. Furthermore, sera from groups of mice immunized with the different antigen preparations displayed similar antibody titers when cross-reacting in ELISA or Western blot with antigen produced in either expression system. This indicates that PL-733FK2 and BV-733 are antigenically indistinguishable.

Antibodies from both immunization groups exhibited with similar cancer-specific CDC and ADCC functions. Some differences in CDC at higher serum dilutions may be due to undetected impurities in the antigen preparations that did not generate an antibody response and/or adjuvant effects. However, both of the CDC and ADCC effects of the sera raised in this study were rather low and may not be sufficient for effective cancer therapy. To increase the efficacy of humoral and cellular immune responses against the cancer cells, optimization of immunization doses, adjuvant, and application routes is a likely requirement. In previous publications [41], higher CDC and ADCC values have been demonstrated to be achievable using antigen GA733-2, and we showed that plant-based production of this antigen does not have an inhibitory effect.

Both PL-733FK2 and BV-733 were able to stimulate the proliferation of T cells from mice immunized with either recombinant antigen. The kinetics of the response were similar indicating that the dominant antigenic determinants of PL-733FK2 and BV-733 are likely to be the same. T cells from both groups, however, proliferated to a significantly greater extent in presence of PL-733FK2 than BV-733. Since the antigen content was similar, this might indicate the presence of some undetected stimulant in the plant-derived antigen preparation. Notably, when mice were given QS21 adjuvant alone, T cells could be stimulated with plant-produced antigen, but not with BV-733. In this regard, the contribution of the QS21 adjuvant used needs to be considered. QS21 is a saponin that was originally derived from plants and despite the clinical grade degree of purity may contain determinants that enhanced responses to the plant-produced cancer antigen. On the other hand, QS21-like substances in the plant-produced antigen preparation might also have some cross-reactivity with QS21 and possibly an adjuvant-like effect. This non–cancer antigen-specific response could account for the differences observed when comparing the PL-733FK2 and the BV-733 groups. Further tests using different adjuvants and highly purified antigen preparations are required to resolve this issue. Detailed in vivo studies will be necessary to evaluate the potential of vaccination with this colorectal cancer antigen against established tumors. Recent data indicate that this approach may be feasible even if a crude plant extract is used for immunization [15] rather than purified antigen as in our experiments.

Our studies demonstrate that plants are a potential vehicle for large-scale production of a putative cancer vaccine. A colorectal cancer antigen expressed in plants had similar immunogenic properties to virtually the same antigen produced in conventional BVES. As plants and the supplies used for their growth are inexpensive and not host to any known human pathogen, the plant expression system tested here could represent an attractive alternative for the production of therapeutics for human use at potentially low production costs.

Acknowledgements

This research was supported by the Commonwealth of Pennsylvania. Special thanks to Rhonda Keen for invaluable technical help and to Dr William Wunner for BVES supernatants and critical reading of the manuscript.

References

  • 1.Abicht Curr Opin Mol Ther. 2000;2:593. [PubMed] [Google Scholar]
  • 2.Arntzen Nat Biotechnol. 1997;15:221. doi: 10.1038/nbt0397-221. [DOI] [PubMed] [Google Scholar]
  • 3.Balzar J Mol Med. 1999;77:699. doi: 10.1007/s001099900038. [DOI] [PubMed] [Google Scholar]
  • 4.Balzar Mol Cell Biol. 2001;21:2570. doi: 10.1128/MCB.21.7.2570-2580.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Belanger Faseb J. 2000;14:2323. doi: 10.1096/fj.00-0144com. [DOI] [PubMed] [Google Scholar]
  • 6.Birebent Crit Rev Oncol Hematol. 2001;39:107. doi: 10.1016/S1040-8428(01)00125-1. [DOI] [PubMed] [Google Scholar]
  • 7.Chong J Biol Chem. 2001;276:5804. doi: 10.1074/jbc.M008839200. [DOI] [PubMed] [Google Scholar]
  • 8.Crombet Hybridoma. 2001;20:131. doi: 10.1089/02724570152057634. [DOI] [PubMed] [Google Scholar]
  • 9.Fernsten Cancer Res. 1990;50:4656. [PubMed] [Google Scholar]
  • 10.Fischer Biotechnol Appl Biochem. 1999;30:101. [PubMed] [Google Scholar]
  • 11.Fischer Biol Chem. 1999;380:825. doi: 10.1515/BC.1999.102. [DOI] [PubMed] [Google Scholar]
  • 12.Fischer Biotechnol Appl Biochem. 1999;30:113. [PubMed] [Google Scholar]
  • 13.Fischer J Biol Regul Homeost Agents. 2000;14:83. [PubMed] [Google Scholar]
  • 14.Foon Curr Oncol Rep. 2001;3:116. doi: 10.1007/s11912-001-0010-y. [DOI] [PubMed] [Google Scholar]
  • 15.Franconi Cancer Res. 2002;62:3654. [PubMed] [Google Scholar]
  • 16.Gilewski Cancer Chemother Pharmacol. 2000;46:S23. doi: 10.1007/pl00014044. [DOI] [PubMed] [Google Scholar]
  • 17.Göttlinger Int J Cancer. 1986;38:47. doi: 10.1002/ijc.2910380109. [DOI] [PubMed] [Google Scholar]
  • 18.Göttlinger Hybridoma. 1986;5:S29. [PubMed] [Google Scholar]
  • 19.Haller Semin Oncol. 2001;28:25. doi: 10.1053/sonc.2001.19719. [DOI] [PubMed] [Google Scholar]
  • 20.Helfrich W, Koning PW, The HT, Lej LD. Epitope mapping of SCLC-Cluster-2-MAbs and generation of antibodies directed against new EGP-2 epitopes. Int J Cancer [Suppl. 1994;]:8–69. doi: 10.1002/ijc.2910570714. [DOI] [PubMed] [Google Scholar]
  • 21.Herlyn Cell Biophys. 1994;25:143. doi: 10.1007/BF02789225. [DOI] [PubMed] [Google Scholar]
  • 22.Herlyn Proc Natl Acad Sci U S A. 1979;76:1438. [Google Scholar]
  • 23.Hooper Int Immunol. 1994;6:873. doi: 10.1093/intimm/6.6.873. [DOI] [PubMed] [Google Scholar]
  • 24.Im Curr Oncol Rep. 2001;3:322. doi: 10.1007/s11912-001-0084-6. [DOI] [PubMed] [Google Scholar]
  • 25.Kean J Immunol. 2000;165:6511. doi: 10.4049/jimmunol.165.11.6511. [DOI] [PubMed] [Google Scholar]
  • 26.Koprowski Somatic Cell Genetics. 1979;5:957. doi: 10.1007/BF01542654. [DOI] [PubMed] [Google Scholar]
  • 27.Koprowski Vaccine. 2001;19:2735. doi: 10.1016/S0264-410X(00)00511-9. [DOI] [PubMed] [Google Scholar]
  • 28.Litvinov J Cell Biol. 1994;125:437. doi: 10.1083/jcb.125.2.437. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Lubeck Cell Immunol. 1988;111:107. doi: 10.1016/0008-8749(88)90055-x. [DOI] [PubMed] [Google Scholar]
  • 30.McCormick Proc Natl Acad Sci U S A. 1999;96:703. doi: 10.1073/pnas.96.2.703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Mendelsohn Semin Cancer Biol. 1990;1:339. [PubMed] [Google Scholar]
  • 32.Modjtahedi Cell Biophys. 1993;22:129. doi: 10.1007/BF03033871. [DOI] [PubMed] [Google Scholar]
  • 33.Moingeon Vaccine. 2001;19:1305. doi: 10.1016/S0264-410X(00)00372-8. [DOI] [PubMed] [Google Scholar]
  • 34.Monzavi-Karbassi Biotechniques. 2001;30:170. doi: 10.2144/01301dd02. [DOI] [PubMed] [Google Scholar]
  • 35.Mosolits Cancer Immunol Immunother. 2002;51:209. doi: 10.1007/s00262-002-0272-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Salem J Surg Res. 1993;55:249. doi: 10.1006/jsre.1993.1136. [DOI] [PubMed] [Google Scholar]
  • 37.Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning, 2nd edn. Cold Spring Harbor Laboratory Press, New York
  • 38.Shivprasad S, Pogue GP, Lewandowski DJ, Hidalgo J, Donson J, Grill LK, Dawson WO. Heterologous sequences greatly affect foreign gene expression in tobacco mosaic virus-based vectors. Virology. 1999;255:312–323. doi: 10.1006/viro.1998.9579. [DOI] [PubMed] [Google Scholar]
  • 39.Simon Proc Natl Acad Sci U S A. 1990;87:2755. [Google Scholar]
  • 40.Sindelar WF, Maher MM, Herlyn D, Sears HF, Steplewski Z, Koprowski H (1986) Trial of therapy with monoclonal antibody 17-1A in pancreatic carcinoma: preliminary results. Hybridoma 5[Suppl 1]:S125–S132 [PubMed]
  • 41.Staib Int J Cancer. 2001;92:79. doi: 10.1002/1097-0215(20010401)92:1<79::AID-IJC1164>3.0.CO;2-M. [DOI] [PubMed] [Google Scholar]
  • 42.Steplewski Science. 1983;221:865. doi: 10.1126/science.6879183. [DOI] [PubMed] [Google Scholar]
  • 43.Strassburg Cancer Res. 1992;52:815. [PubMed] [Google Scholar]
  • 44.Turpen TH, Reinl SJ (1998) Tobamovirus vectors for expression of recombinant genes in plants. In: Cunnigham C, Porter AJR (eds) Recombinant proteins from plants: production and isolation of clinically useful compounds, vol 3. Humana Press, Totowa, NJ, pp 89–101
  • 45.Verch J Immunol Methods. 1998;220:69. doi: 10.1016/S0022-1759(98)00149-5. [DOI] [PubMed] [Google Scholar]
  • 46.Vogel C, Cobleigh MA, Tripathy D, Gutheil JC, Harris LN, Fehrenbacher L, Slamon DJ, Murphy M, Novotny WF, Burchmore M, Shak S, Stewart SJ (2001) First-line, single-agent Herceptin(trastuzumab) in metastatic breast cancer: a preliminary report. Eur J Cancer 37[Suppl 1]:S25–S29 [PubMed]
  • 47.Vuillez Clin Cancer Res. 1999;5:3259s. [PubMed] [Google Scholar]
  • 48.Weiner Semin Oncol. 1999;26:41. [Google Scholar]
  • 49.Zaloudik Cancer Gene Ther. 2002;9:382. doi: 10.1038/sj.cgt.7700452. [DOI] [PubMed] [Google Scholar]
  • 50.Zeh Trends Mol Med. 2001;7:307. doi: 10.1016/S1471-4914(01)01992-X. [DOI] [PubMed] [Google Scholar]
  • 51.Zinkernagel Int J Cancer. 2001;93:1. doi: 10.1002/ijc.1305. [DOI] [PubMed] [Google Scholar]

Articles from Cancer Immunology, Immunotherapy : CII are provided here courtesy of Springer

RESOURCES